","Title: Application of fully recurrent neural networks for speech recognition\nYear: 1991\n\nAbstract: The authors describe an extended backpropagation algorithm for fully connected recurrent neural networks applied to speech recognition. The extended delta rule is approximated by excluding some of the past activities of the dynamic neurons to reduce computational complexity without performance degradation. In speaker-dependent recognition of a confusable syllable set, the fully recurrent neural network with the approximated backpropagation algorithm showed better performance than the multilayer perceptron and the self-recurrent network with comparable time complexity. In addition, it is found that most self-recurrent connections become excitatory and most mutual recurrent connections become inhibitory","Title: Design and analysis of a fair scheduling algorithm for QoS guarantees in high-speed packet-switched networks\nYear: 1998\n\nAbstract: B-ISDNs are required to support a variety of services such as audio, data, and video, so that the guarantee of quality-of-service (QoS) has become an increasingly important problem. An effective fair scheduling algorithm permits high-speed switches to divide link bandwidth fairly among competing connections. Together with the connection admission control, it can guarantee the QoS of connections. We propose a novel fair scheduling algorithm, called “virtual-time-based round robin (VTRR)”. Our scheme maps the priorities of packets into classes and provides service to the first non-empty class in each round. Also, it uses an estimation method of the virtual time necessary to this service discipline. To find the first non-empty class, the VTRR adopts a priority queueing system of O(loglog c) which decreases the number of instructions which need to be carried out in one packet transmission time segment. These policies help the VTRR implementation in software, which presents flexibility for upgrades. Our analysis has demonstrated that the VTRR provides bounded unfairness and its performance is close to that of weighted fair queuing. Therefore, the VTRR has a good performance as well as simplicity, so that it is suitable for high-speed B-ISDN","Title: Two-phase multicast in wormhole-switched bidirectional multistage Banyan networks\nYear: 1998\n\nAbstract: A multistage interconnection network is a suitable class of interconnection architecture for constructing large-scale multicomputers. Broadcast and multicast communication are fundamental in supporting collective communication operations such as reduction and barrier synchronization. In this paper, we propose a new multicast technique in wormhole-switched bidirectional multistage banyan networks for constructing large-scale multicomputers. To efficiently support broadcast and multicast with simple additional hardware without deadlock, we propose a two-phase multicast algorithm which takes only two transmissions to perform a broadcast and a multicast to an arbitrary number of desired destinations. We encode a header as a cube and adopt the most upper input link first scheme with periodic priority rotation as arbitration mechanism on contented output links. We coalesce the desired destination addresses into multiple number of cubes. And then, we evaluate the performance of the proposed algorithm by simulation. The proposed two-phase multicast algorithm makes a significant improvement in terms of latency. It is noticeable that the two-phase algorithm keeps broadcast latency as efficient as the multicast latency of fanout $2^m$ where $m$ is the minimum integer satisfying $2^m \\geq \\sqrt{N}$($N$ is a network size).","Title: Improved base-φ expansion method for Koblitz curves over optimal extension fields\nYear: 2007\n\nAbstract: An improved base-phi expansion method is proposed, in which the bit-length of coefficients is shorter and the number of coefficients is smaller than in Kobayashi's expansion method. The proposed method meshes well with efficient multi-exponentiation algorithms. In addition, two efficient algorithms based on the proposed expansion method, named phi-wNAF and phi-SJSF, are presented which significantly reduce the computational effort involved in online precomputation by using the property of Frobenius endomorphism. The proposed algorithms noticeably accelerate computation of a scalar multiplication on Koblitz curves over optimal extension fields (OEFs). In particular, for OEFs where the characteristic is close to 32 bits or 64 bits, the required number of additions is reduced up to 50% in comparison with Kobayashi's base-phi scalar multiplication algorithm. Finally, a method that significantly reduces the memory usage of the precomputation table at the expense of slightly more computation is presented","Title: The Design of a Robust Intrusion Tolerance System through Advanced Adaptive Cluster Transformation and Vulnerability-Based VM Selection\nYear: 2013\n\nAbstract: In this paper, we suggested novel schemes which use advanced adaptive cluster transformation and VM selection policy for intrusion tolerant systems (ITSs). The cluster size is transformed adaptively in order to maintain a certain level of services by using the adaptive cluster expansion scheme. Also, all the servers in clusters can be substituted in case serious threat such as massive packets incoming is expected by using the adaptive cluster substitution method. If there exists reliable historical data, more fast transformation is possible. In addition, the less-vulnerable virtual machines (VMs) are chosen using evaluation policies to reduce data leakage occured from system's vulnerabilities. Simulation results done with CSIM 20 prove that the suggested schemes improve intrusion tolerance efficiently compared to other conventional methods.","Title: Analysis of Cell Sojourn Time in Heterogeneous Networks with Small Cells\nYear: 2016\n\nAbstract: Recently, heterogeneous networks with small cells have been widely used to increase the capacity of mobile systems. In such environments, accurate estimation of the mean cell sojourn time is critical for evaluating the performance of the network and its applications. It is especially important to analyze the cell sojourn times of mobile users as they reside in different network tiers: either macro-cell-only or small-cell-covered areas. But because small cells are deployed in an irregular manner, it is difficult to derive the analytical mean cell sojourn time in a macro-cell-only area. In this letter, we propose a novel approach to resolve this difficulty. We developed a simple but effective trick that approximates the heterogeneous network to a discrete grid so that it becomes tractable, making it possible to derive the analytical mean sojourn time in the macro-cell-only area. Simulation results confirm that the proposed method has excellent accuracy for general random walk mobility models including random waypoint, Brownian motion, tailored Brownian motion, and truncated Levy walk.","Title: Efficient Packet Routing in Highly Mobile Wireless Networks\nYear: 2015\n\nAbstract: Traditional mobile ad-hoc network (MANET) and delay tolerant network (DTN) routing schemes work properly only under the limited assumptions. MANET routing schemes are designed for densely populated networks while DTN routing schemes were developed for dealing with intermittently connected topologies. However, in a real environment, the assumption cannot be hold: the density of nodes constantly changes and a network can be separated due to node mobility even when it is composed of large number of nodes. We propose a routing scheme for maximizing the delivery rate in a network with varying density. In the proposed scheme, data packets are used to search an available path while they are delivered by DTN-based routing. If a path exists, subsequent packets are relayed along the path. Otherwise, packets are delivered continuously using DTN-based routing. Using simulation we show that the proposed scheme achieves high throughput and a low overhead in a dense part of the network, while it outperforms existing DTN routing schemes in an intermittently connected situation.","Title: Hybrid Recovery-Based Intrusion Tolerant System For Practical Cyber-Defense\nYear: 2016\n\nAbstract: Due to the periodic recovery of virtual machines regardless of whether malicious intrusions exist, proactive recovery-based Intrusion Tolerant Systems (ITSs) are being considered for mission-critical applications. However, the virtual replicas can easily be exposed to attacks during their working period, and additionally, proactive recovery based ITSs are ineffective in eliminating the vulnerability of exposure time, which is closely related to service availability. To address these problems, we propose a novel hybrid recovery-based ITS in this paper. The proposed method utilizes availability-driven recovery and dynamic cluster resizing. The availability-driven recovery method operates the recovery process by both proactive and reactive ways for the system to gain shorter exposure times and higher success rates. The dynamic cluster resizing method reduces the overhead of the system that occurs from dynamic workload fluctuations. The performance of the proposed ITS with various synthetic and real workloads using CloudSim showed that it guarantees higher availability and reliability of the system, even under malicious intrusions such as DDoS attacks.","Title: Customer Data Scanner For Hands-Free Mobile Payment\nYear: 2015\n\nAbstract: New payment technologies are coming that will raise user convenience. To support automatic hands-free payment, merchant devices will collect customer's information from the cloud of payment service providers or customer's smart phones, which should be removed after the transaction. Using Jaccard containment, we propose a proactive security approach of cleaning personal data at merchant-side point-of-sale terminals. We also propose a sampling method to reduce communication overhead by several orders of magnitude.","Title: Fully Batch Processing Enabled Memory Integrity Verification Algorithm Based on Merkle Tree.\nYear: 2015\n\nAbstract: Memory attacks have been increasing in number recently. Adversary can manipulate memory data or break system by doing active attacks. Especially, main memory is used as a target of attack, because main memory is more vulnerable than other components, such as CPU. To prevent adversary's active attack, memory integrity verification algorithm has been proposed. Protection of computer's memory integrity is important in situations where attacks on the computer systems are a threat. As technology has advanced, computer systems migrate from wire-based to wireless system. A lot of memory integrity verification algorithms are already developed, but these algorithms do not consider new wireless platform. Wireless platform is constrained by a lack of storage and power supply in comparison with wire-based system, therefore computational overhead and storage overhead must be considered when applying to algorithm, which is used in wireless system. In this study, integrity verification performance can be improved by doing batch-processing. Previous verification algorithms based on Merkle tree do not support fully batch processing verification. We propose fully batch processing enabled memory integrity verification algorithm based on Merkle tree. This algorithms can verify memory integrity in completely batches. For implement our algorithm, we use Incremental multiset hash function, and as a result, consume only 480-bit on-chip storage. Reducing consumption of on-chip storage leads to improving on the performance of computation. We implement our algorithm and previous memory integrity verification algorithms based on standard Merkle tree and lazy-processing Merkle tree in simulator to compare their performance. Our algorithm offers better system performance overall, especially when attack rarely occur.","Title: Sorting Method For Fully Homomorphic Encrypted Data Using The Cryptographic Single-Instruction Multiple-Data Operation\nYear: 2016\n\nAbstract: In this paper, we present a faster (wall-clock time) sorting method for numerical data subjected to fully homomorphic encryption (FHE). Owing to circuit-based construction and the FHE security property, most existing sorting methods cannot be applied to encrypted data without significantly compromising efficiency. The proposed algorithm utilizes the cryptographic single-instruction multiple-data (SIMD) operation, which is supported by most existing FHE algorithms, to reduce the computational overhead. We conducted a careful analysis of the number of required recryption operations, which are the computationally dominant operations in FHE. Accordingly, we verified that the proposed SIMD-based sorting algorithm completes the given task more quickly than existing sorting methods if the number of data items and (or) the maximum bit length of each data item exceed specific thresholds.","Title: Preemptive Real-Time Scheduling Incorporating Security Constraint For Cyber Physical Systems\nYear: 2016\n\nAbstract: Since many cyber-physical systems (CPSs) manipulate security-sensitive data, enhancing the quality of security in a CPS is a critical and challenging issue in CPS design. Although there has been a large body of research on securing general purpose PCs, directly applying such techniques to a CPS can compromise the real-time property of CPSs since the timely execution of tasks in a CPS typically relies on real-time scheduling. Recognizing this property, previous works have proposed approaches to add a security constraint to the real-time properties to cope with the information leakage problem that can arise between real-time tasks with different security levels. However, conventional works have mainly focused on non-preemptive scheduling and have suggested a very naive approach for preemptive scheduling, which shows limited analytical capability. In this paper, we present a new preemptive fixed-priority scheduling algorithm incorporating a security constraint, called lowest security-level first (LSF) and its strong schedulability analysis to reduce the potential of information leakage. Our simulation results show that LSF schedulability analysis outperforms state-of-the-art FP analysis when the security constraint has reasonable timing penalties.","Title: Efficient Privacy-Preserving Matrix Factorization via Fully Homomorphic Encryption: Extended Abstract.\nYear: 2016\n\nAbstract: Recommendation systems become popular in our daily life. It is well known that the more the release of users' personal data, the better the quality of recommendation. However, such services raise serious privacy concerns for users. In this paper, focusing on matrix factorization-based recommendation systems, we propose the first privacy-preserving matrix factorization using fully homomorphic encryption. On inputs of encrypted users' ratings, our protocol performs matrix factorization over the encrypted data and returns encrypted outputs so that the recommendation system knows nothing on rating values and resulting user/item profiles. It provides a way to obfuscate the number and list of items a user rated without harming the accuracy of recommendation, and additionally protects recommender's tuning parameters for business benefit and allows the recommender to optimize the parameters for quality of service. To overcome performance degradation caused by the use of fully homomorphic encryption, we introduce a novel data structure to perform computations over encrypted vectors, which are essential operations for matrix factorization, through secure 2-party computation in part. With the data structure, the proposed protocol requires dozens of times less computation cost over those of previous works. Our experiments on a personal computer with 3.4 GHz 6-cores 64 GB RAM show that the proposed protocol runs in 1.5 minutes per iteration. It is more efficient than Nikolaenko et al.'s work proposed in CCS 2013, in which it took about 170 minutes on two servers with 1.9 GHz 16-cores 128 GB RAM.","Title: Protecting Secret Keys In Networked Devices With Table Encoding Against Power Analysis Attacks\nYear: 2016\n\nAbstract: Nowadays, secret keys of networked devices are profoundly attacked by power analysis attacks, caused by the dramatic evolution of statistical analysis with a simple experimental setup. Recently, OpenSSL and CoreBitcoin running on Android and iOS have been broken by power analysis. Moreover, sensors and actuators can also be attacked thereby threatening user's privacy and security. To resolve these challenges, power-analysis-resistant implementations of cryptographic algorithms in networked devices have received a lot of attentions. Masking schemes have been developed to implement secure cryptographic algorithms against side-channel analysis (SCA) attacks. Technically, the first-order masking method is vulnerable to the second order differential power analysis (2ODPA) attacks, but the current solutions against 2ODPA are expensive to be implemented. Moreover, worse performance will be shown if the cryptographic algorithms include boolean and arithmetic operations. In this paper, we propose a new countermeasure scheme to resist SCA attacks. Our scheme randomizes all the intermediate values of block cipher by encoding functions in the algorithm to lookup table and makes it resistant to power analysis attack. We apply our scheme to the block cipher algorithm, HIGHT. Our protected implementation of HIGHT takes only 1.79 times compared to the straightforward algorithm, and it needs 25 kbytes to store lookup tables in memory.","Title: On-demand bootstrapping mechanism for isolated cryptographic operations on commodity accelerators.\nYear: 2016\n\nAbstract: General-Purpose computing on a Graphics Processing Unit (GPGPU) involves leveraging commodity GPUs as massively parallel processing units. GPGPU is an emerging computing paradigm for high-performance and data-intensive computations such as cryptographic operations. Although GPGPU is an attractive solution for accelerating modern cryptographic operations, the security challenges that stem from utilizing commodity GPUs remain an unresolved problem. In this paper, we present an On-demand Bootstrapping Mechanism for Isolated cryptographic operations (OBMI). OBMI transforms commodity GPUs into a securely isolated processing core for various cryptographic operations while maintaining cost-effective computations. By leveraging System Management Mode (SMM), a privileged execution mode provided by x86 architectures, OBMI implements a program and a secret key into the GPU such that they are securely isolated during the acceleration of cryptographic operations, even in the presence of compromised kernels. Our approach does not require an additional hardware-abstraction layer such as a hypervisor or micro-kernel, and it does not entail modifying the GPU driver. An evaluation of the proposed OBMI demonstrated that even adversaries with kernel privileges cannot gain access to the secret key, and it also showed that the proposed mechanism incurs negligible performance degradation for both the CPU and GPU.","Title: Motion-MiX DHT for Wireless Mobile Networks.\nYear: 2016\n\nAbstract: Last encounter routing (LER) is an excellent routing paradigm that exploits distributed mobility diffusion to achieve both moving object tracking and packet networking services in dynamic mobile networks. From our observations, we discover that LER can be easily extended to a mobile DHT protocol that introduces excellent performance to high-speed mobility environments. This is of particular interest when higher level of mobility and membership dynamics go hand in hand. In our simple but powerful DHT paradigm, a data publish/look-up process consists of a sequence of spatial motion tracking of the rendezvous node that is responsible for the data resource. Thus, we name the protocol MX-DHT (Motion-MiX-DHT). As opposed to existing topology based DHT schemes, MX-DHT does not require additional management of logical or overlying look-up topologies, except for the one-hop encounter records of logical metadata carried by mobile nodes. Therefore, in high-speed mobility and dynamic membership environments, MX-DHT achieves a significant reduction in the communication costs of the publish/look-up and join/leave operations as compared to existing mobile DHT schemes. An extensive set of experiments showed that MX-DHT is a cost-effective solution to providing a content centric networking service in different types of networks with dynamic mobility and membership changes.","Title: Designing And Implementing A Diversity Policy For Intrusion-Tolerant Systems\nYear: 2017\n\nAbstract: Research on intrusion-tolerant systems (ITSs) is being conducted to protect critical systems which provide useful information services. To provide services reliably, these critical systems must not have even a single point of failure (SPOF). Therefore, most ITSs employ redundant components to eliminate the SPOF problem and improve system reliability. However, systems that include identical components have common vulnerabilities that can be exploited to attack the servers. Attackers prefer to exploit these common vulnerabilities rather than general vulnerabilities because the former might provide an opportunity to compromise several servers. In this study, we analyze software vulnerability data from the National Vulnerability Database (NVD). Based on the analysis results, we present a scheme that finds software combinations that minimize the risk of common vulnerabilities. We implement this scheme with CSIM20, and simulation results prove that the proposed scheme is appropriate for a recovery-based intrusion tolerant architecture.","Title: Duo: Software Defined Intrusion Tolerant System Using Dual Cluster.\nYear: 2018\n\nAbstract: An intrusion tolerant system (ITS) is a network security system that is composed of redundant virtual servers that are online only in a short time window, called exposure time. The servers are periodically recovered to their clean state, and any infected servers are refreshed again, so attackers have insufficient time to succeed in breaking into the servers. However, there is a conflicting interest in determining exposure time, short for security and long for performance. In other words, the short exposure time can increase security but requires more servers to run in order to process requests in a timely manner. In this paper, we propose Duo, an ITS incorporated in SDN, which can reduce exposure time without consuming computing resources. In Duo, there are two types of servers: some servers with long exposure time (White server) and others with short exposure time (Gray server). Then, Duo classifies traffic into benign and suspicious with the help of SDN/NFV technology that also allows dynamically forwarding the classified traffic to White and Gray servers, respectively, based on the classification result. By reducing exposure time of a set of servers, Duo can decrease exposure time on average. We have implemented the prototype of Duo and evaluated its performance in a realistic environment.","Title: Captcha Image Generation Systems Using Generative Adversarial Networks\nYear: 2018\n\nAbstract: We propose new CAPTCHA image generation systems by using generative adversarial network (GAN) techniques to strengthen against CAPTCHA solvers. To verify whether a user is human, CAPTCHA images are widely used on the web industry today. We introduce two different systems for generating CAPTCHA images, namely, the distance GAN (D-GAN) and composite GAN (C-GAN). The D-GAN adds distance values to the original CAPTCHA images to generate new ones, and the C-GAN generates a CAPTCHA image by composing multiple source images. To evaluate the performance of the proposed schemes, we used the CAPTCHA breaker software as CAPTCHA solver. Then, we compared the resistance of the original source images and the generated CAPTCHA images against the CAPTCHA solver. The results show that the proposed schemes improve the resistance to the CAPTCHA solver by over 67.1% and 89.8% depending on the system.","Title: POSTER: Zero-Day Evasion Attack Analysis on Race between Attack and Defense.\nYear: 2018\n\nAbstract: Deep neural networks (DNNs) exhibit excellent performance in machine learning tasks such as image recognition, pattern recognition, speech recognition, and intrusion detection. However, the usage of adversarial examples, which are intentionally corrupted by noise, can lead to misclassification. As adversarial examples are serious threats to DNNs, both adversarial attacks and methods of defending against adversarial examples have been continuously studied. Zero-day adversarial examples are created with new test data and are unknown to the classifier; hence, they represent a more significant threat to DNNs. To the best of our knowledge, there are no analytical studies in the literature of zero-day adversarial examples with a focus on attack and defense methods through experiments using several scenarios. Therefore, in this study, zero-day adversarial examples are practically analyzed with an emphasis on attack and defense methods through experiments using various scenarios composed of a fixed target model and an adaptive target model. The Carlini method was used for a state-of-the-art attack, while an adversarial training method was used as a typical defense method. We used the MNIST dataset and analyzed success rates of zero-day adversarial examples, average distortions, and recognition of original samples through several scenarios of fixed and adaptive target models. Experimental results demonstrate that changing the parameters of the target model in real time leads to resistance to adversarial examples in both the fixed and adaptive target models.","Title: Detecting Online Game Chargeback Fraud Based on Transaction Sequence Modeling Using Recurrent Neural Network.\nYear: 2017\n\nAbstract: We propose an online game money chargeback fraud detection method using operation sequence, gradient of charge/purchase amount, time and country as features of a transaction. We model the sequence of transactions with a recurrent neural network which also combines charge and purchase transaction features in single feature vector. In experiments using real data (a 483,410 transaction log) from a famous online game company in Korea, the proposed method shows a 78% recall rate with a 0.057% false positive rate. This recall rate is 7% better than current methodology utilizing transaction statistics as features.","Title: A Simple and Enriched Closed-Form Formula for Cell Residence Time in 5G Heterogeneous Networks.\nYear: 2018\n\nAbstract: Cell residence time is a key system parameter for the performance analysis of cellular mobile systems and their related applications. In order to save cost and time before wide implementation and deployment of a mobile system, various network performance metrics, such as network resource occupancy probabilities, should be calculated through analytical results of the mean cell residence time. However, this is difficult to estimate in 5G heterogeneous networks, due to irregular-shaped multi-tier heterogeneous network topologies introduced by a diverse set of small cells. In order to find an efficient way of calculating the cell residence time for 5G heterogeneous networks, we first examine the well-known conventional formula. Then we propose an enhanced closed-form formula for cell residence time by considering more generalized cellular mobile systems, as well as more generic mobility models. The proposed closed-form formula can be directly applied to various complex and reliable network scenarios. It can be adapted to multi-tier heterogeneous networks with randomly positioned irregular-shaped small cells, even though they overlap with each other. Our simulation results show that the proposed formula provides excellent estimation accuracy for general random mobility models, including the Levy walk, which is known to realistically reflect human mobility.","Title: Advanced Ensemble Adversarial Example On Unknown Deep Neural Network Classifiers\nYear: 2018\n\nAbstract: Deep neural networks (DNNs) are widely used in many applications such as image, voice, and pattern recognition. However, it has recently been shown that a DNN can be vulnerable to a small distortion in images that humans cannot distinguish. This type of attack is known as an adversarial example and is a significant threat to deep learning systems. The unknown-target-oriented generalized adversarial example that can deceive most DNN classifiers is even more threatening. We propose a generalized adversarial example attack method that can effectively attack unknown classifiers by using a hierarchical ensemble method. Our proposed scheme creates advanced ensemble adversarial examples to achieve reasonable attack success rates for unknown classifiers. Our experiment results show that the proposed method can achieve attack success rates for an unknown classifier of up to 9.25% and 18.94% higher on MNIST data and 4.1% and 13% higher on CIFAR10 data compared with the previous ensemble method and the conventional baseline method, respectively.","Title: Efficient Privacy-Preserving Matrix Factorization for Recommendation via Fully Homomorphic Encryption.\nYear: 2018\n\nAbstract: There are recommendation systems everywhere in our daily life. The collection of personal data of users by a recommender in the system may cause serious privacy issues. In this article, we propose the first privacy-preserving matrix factorization for recommendation using fully homomorphic encryption. Our protocol performs matrix factorization over encrypted users’ rating data and returns encrypted outputs so that the recommendation system learns nothing on rating values and resulting user/item profiles. Furthermore, the protocol provides a privacy-preserving method to optimize the tuning parameters that can be a business benefit for the recommendation service providers. To overcome the performance degradation caused by the use of fully homomorphic encryption, we introduce a novel data structure to perform computations over encrypted vectors, which are essential for matrix factorization, through secure two-party computation in part. Our experiments demonstrate the efficiency of our protocol.","Title: Secure mobile device structure for trust IoT.\nYear: 2018\n\nAbstract: In the IoT environment, all devices are connected to each other, and mobile device is considered as key device. But hacking into mobile devices is increasing rapidly with the increase in mobile device users. As the market share of Android OS increases, hacking of mobile devices has focused on Android devices. Although there are many security solutions for mobile devices, they are fragmentary for mobile threats; that is, they are solutions for only several threats rather than comprehensive solutions. There is hence a limit to protecting user’s and company’s data stored or used on mobile devices from various types of hacking. To address this, we propose a mobile device protection technology based on domain isolation. Virtualization technology has emerged to increase CPU utilization in server-class PCs and to run various OSs in one system. As these virtualization technologies become lightweight, they are beginning to be applied to embedded devices. In this paper, we applied this lightweight embedded virtualization technology to mobile devices to divide mobile devices into two areas. Therefore, users can have hidden area from hacker attack in addition to Android OS area which can use same as existing mobile device. There is a hardware-based mobile security solution using an secure element, but this has to be reflected in the manufacturing process of the mobile device. However, since the domain separation technology using the virtualization, proposed in this paper, is a software solution, it has an advantage that it can be applied to a device that is already in use. In addition, to protect the hidden area, application authentication/authorization and user authentication technology were applied. And we use white-box cryptography to get root of trust of the key which is used for secure storage and data encryption/decryption. We believe this is a fundamental solution for protecting the mobile device users from hacking. We implemented and tested various mobile applications operating on a mobile device that incorporates our proposed structure based on domain isolation. There is some performance degradation caused by the domain separation, but it is negligible. According to https://www.wired.com/insights/2012/11/mobile-supercomputers/, the chips for mobile phones have evolved and mobile phones will soon become supercomputers. In this case, the addition of virtualization to the mobile device will have less impact on the computing power of the mobile device, and data protection stored in mobile devices and secure execution environment of security programs will become more important issues. Therefore, our TeeMo structure is a necessary technology to protect mobile device users.","Title: A Novel Positive Transfer Learning Approach for Telemonitoring of Parkinson's Disease.\nYear: 2019\n\nAbstract: Telemonitoring is the use of electronic devices to remotely monitor patients. Taking the Parkinson's disease (PD) as an example, the use of at-home testing device (AHTD) enables remote, internet-based measurement of PD vocal symptoms. Translating AHTD measurement into a unified PD rating scale (UPDRS) through predictive analytics enables cost-effective, convenient, and close tracking of PD progres...","Title: Fooling A Neural Network In Military Environments: Random Untargeted Adversarial Example\nYear: 2018\n\nAbstract: Deep neural networks (DNNs) show superior performance in machine learning tasks such as image recognition, speech recognition, intrusion detection, and pattern analysis. However, an adversarial example, created by adding a little noise to the original sample, can cause misclassification by the DNN. As adversarial examples are a serious threat to DNNs, there has been much research into the generation of adversarial examples designed for attacking DNNs. The adversarial example attack is divided into two categories: targeted adversarial example and untargeted adversarial example. The targeted adversarial example attack causes machines to misinterpret an object as the attacker's desired class. In contrast, the untargeted adversarial example causes machines to misinterpret an object as an incorrect class. In this paper, we focus on an untargeted adversarial example scenario because it has less distortion from the original sample and a faster learning time than a targeted adversarial example scenario. However, there is a pattern problem in generating untargeted adversarial examples: Because of the similarity between the original class and specific classes, it may be possible for the defending system to determine the original class by analyzing the output classes of the untargeted adversarial examples. To overcome this problem, we propose a new method for generating untargeted adversarial examples, one that uses an arbitrary class in the generation process. For experimental datasets, we used MNIST and CIFAR10, and the Tensorflow library was employed as the machine learning library. Through our experiment, we show that the proposed method can generate random untargeted adversarial examples that do not focus on a specific class for a given original class, while keeping distortion to a minimum (1.99 and 42.32 on MNIST and CIFAR10, respectively) and maintaining a 100% attack success rate.","Title: Random Untargeted Adversarial Example on Deep Neural Network.\nYear: 2018\n\nAbstract: Deep neural networks (DNNs) have demonstrated remarkable performance in machine learning areas such as image recognition, speech recognition, intrusion detection, and pattern analysis. However, it has been revealed that DNNs have weaknesses in the face of adversarial examples, which are created by adding a little noise to an original sample to cause misclassification by the DNN. Such adversarial examples can lead to fatal accidents in applications such as autonomous vehicles and disease diagnostics. Thus, the generation of adversarial examples has attracted extensive research attention recently. An adversarial example is categorized as targeted or untargeted. In this paper, we focus on the untargeted adversarial example scenario because it has a faster learning time and less distortion compared with the targeted adversarial example. However, there is a pattern vulnerability with untargeted adversarial examples: Because of the similarity between the original class and certain specific classes, it may be possible for the defending system to determine the original class by analyzing the output classes of the untargeted adversarial examples. To overcome this problem, we propose a new method for generating untargeted adversarial examples, one that uses an arbitrary class in the generation process. Moreover, we show that our proposed scheme can be applied to steganography. Through experiments, we show that our proposed scheme can achieve a 100% attack success rate with minimum distortion (1.99 and 42.32 using the MNIST and CIFAR10 datasets, respectively) and without the pattern vulnerability. Using a steganography test, we show that our proposed scheme can be used to fool humans, as demonstrated by the probability of their detecting hidden classes being equal to that of random selection.","Title: A new proxy signature scheme providing self-delegation\nYear: 2006\n\nAbstract: We improve Malkin et al's construction (Eurocrypt'04) of the proxy signature scheme in the random oracle model. Unlike Malkin et al's scheme, the proposed scheme does not assume the existence of the trusted secure device, which has a global secret key that all users' private keys can be recovered with. This makes the proposed scheme more scalable and efficient because users need not access and cooperate with the secure device to generate their public/private key pairs. We show that the proposed scheme is provably secure based on the modified strong RSA assumption that was made by the Itkis et al (CRYPTO'01).","Title: A Necessary and Sufficient Condition for Proper Routing in Omega-Omega Network\nYear: 1996\n\nAbstract: Recently, a routing algorithm, called inside-out algorithm, was proposed for routing an arbitrary permutation in the omega-based 2log2\nN stage networks including the omega-omega network. This paper discusses the problems of the inside-out routing algorithm and shows that the suggested condition for proper routing in the omega-omega network is insufficient. This paper suggests an extended necessary and sufficient condition for proper routing in the omega-omega network.","Title: An Efficient Audit Scheme for Cloud Storage.\nYear: 2014\n\nAbstract: As Internet technology has grown rapidly, a cloud storage service becomes more very popular and widespread recently. Since there is no guarantee for integrity of data in the cloud storage, audit mechanisms for outsourced data were proposed. However, the conventional audit process requires a lot of computation overheads due to the complex operations used in tag generation and verification. In this paper, we propose a light-weight audit scheme for the cloud storage with efficient cryptographic operations. While keeping high efficiency, our proposed scheme guarantees the consistency of data in the cloud storage. To evaluate the performance of our scheme, we measure the computation costs and the time required for audit process. As a result, our proposed scheme can provide correctness of auditing for outsourced data by an efficient manner.","Title: An efficient multi-path pipeline transmission for a bulk data transfer in IEEE 802.15.4 multi-hop networks.\nYear: 2019\n\nAbstract: A pipeline transmission is the state-of-the-art approach to transmit large amounts of data over IEEE 802.15.4 multi-hop networks, but the performance of the pipeline transmission can be degraded in unreliable networks. In this paper, we propose an efficient multi-path pipeline transmission (EMP) to support the large data transfer with low latency and high energy efficiency under various network conditions. The proposed EMP adjusts cycle time of the pipeline transmission, which allows nodes to retransmit dropped data packet efficiently. It helps to improve the transmission probability, so EMP can mitigate additional delay and energy consumption caused by frequent end-to-end retransmissions. In addition, EMP employs a multi-path transmission which distributes the large data transfer over multiple routes. It contributes not only to reduce transmission time but also to balance energy consumption of nodes. In this work, we evaluate the performance of EMP through theoretical and simulation-based analysis and compare the performance with other existing pipelines. The results show that EMP outperforms the existing protocols in terms of transmission time and energy efficiency, and then the improved performance of EMP can be maintained regardless of network environments such as link quality, hop counts, and network density.","Title: Efficient Approach For Mitigating Mobile Phishing Attacks\nYear: 2018\n\nAbstract: We propose a method for efficiently detecting phishing attacks in mobile environments. When a user visits a website of a certain URL, the proposed method first compares the URL to a generated whitelist. If the URL is not in the whitelist, it detects if the site is a phishing site based on the results of Google search with a carefully refined URL. In addition, the phishing detection is performed only when the user provides input to the website, thereby reducing the frequency of invoking phishing detection to decrease the amount of power used. We implemented the proposed method and used 8315 phishing sites and the same number of legitimate websites for evaluating the performance of the proposed method. We achieved a phishing detection rate of 99.22% with 81.22% reduction in energy consumption as compared to existing approaches that also use search engine for phishing detection. Moreover, because the proposed method does not employ any other algorithm, software, or comparison group, the proposed method can be easily deployed.","Title: Friend-safe evasion attack: An adversarial example that is correctly recognized by a friendly classifier.\nYear: 2018\n\nAbstract: Deep neural networks (DNNs) have been applied in several useful services, such as image recognition, intrusion detection, and pattern analysis of machine learning tasks. Recently proposed adversarial examples-slightly modified data that lead to incorrect classification-are a severe threat to the security of DNNs. In some situations, however, an adversarial example might be useful, such as when deceiving an enemy classifier on the battlefield. In such a scenario, it is necessary that a friendly classifier not be deceived. In this paper, we propose a friend-safe adversarial example, meaning that the friendly machine can classify the adversarial example correctly. To produce such examples, a transformation is carried out to minimize the probability of incorrect classification by the friend and that of correct classification by the adversary. We suggest two configurations for the scheme: targeted and untargeted class attacks. We performed experiments with this scheme using the MNIST and CIFAR10 datasets. Our proposed method shows a 100% attack success rate and 100% friend accuracy with only a small distortion: 2.18 and 1.54 for the two respective MNIST configurations, and 49.02 and 27.61 for the two respective CIFAR10 configurations. Additionally, we propose a new covert channel scheme and a mixed battlefield application for consideration in further applications.","Title: Multi-Targeted Adversarial Example in Evasion Attack on Deep Neural Network.\nYear: 2018\n\nAbstract: Deep neural networks (DNNs) are widely used for image recognition, speech recognition, pattern analysis, and intrusion detection. Recently, the adversarial example attack, in which the input data are only slightly modified, although not an issue for human interpretation, is a serious threat to a DNN as an attack as it causes the machine to misinterpret the data. The adversarial example attack has been receiving considerable attention owing to its potential threat to machine learning. It is divided into two categories: targeted adversarial example and untargeted adversarial example. The untargeted adversarial example happens when machines misclassify an object into an incorrect class. In contrast, the targeted adversarial example attack causes machines to misinterpret the image as the attacker's desired class. Thus, the latter is a more elaborate and powerful attack than the former. The existing targeted adversarial example is a single targeted attack that allows only one class to be recognized. However, in some cases, a multi-targeted adversarial example can be useful for an attacker to make multiple models recognize a single original image as different classes. For example, an attacker can use a single road sign generated by a multi-targeted adversarial example scheme to make model A recognize it as a stop sign and model B recognize it as a left turn, whereas a human might recognize it as a right turn. Therefore, in this paper, we propose a multi-targeted adversarial example that attacks multiple models within each target class with a single modified image. To produce such examples, we carried out a transformation to maximize the probability of different target classes by multiple models. We used the MNIST datasets and TensorFlow library for our experiment. The experimental results showed that the proposed scheme for generating a multi-targeted adversarial example achieved a 100% attack success rate.","Title: Priority Adversarial Example in Evasion Attack on Multiple Deep Neural Networks\nYear: 2019\n\nAbstract: Deep neural networks (DNNs) provide superior per-formance on machine learning tasks such as image recognition, speech recognition, pattern recognition, and intrusion detection. However, an adversarial example created by adding a little noise to the original data can lead to misclassification by the DNN, and the human eye cannot detect the difference from the original data. For example, if an attacker generates a modified left-turn road sign to be incorrectly categorized by a DNN, an autonomous vehicle with the DNN will incorrect classify the modified left-turn road sign as a right-turn sign, whereas a human will correctly classify the modified sign as a left-turn sign. Such an adversarial example is a serious threat to a DNN. Recently, a multi-target adversarial example was introduced that causes misclassification by several models within each target class using a single modified image. However, it has the vulnerability that as the number of target models increases, the overall attack success rate is reduced. Therefore, if there are several models that the attacker wishes to target, the attacker needs to control the attack success rate for each model by considering the attack priority for each model. In this paper, we propose a priority adversarial example that considers the attack priority for each model in cases targeting several models. The proposed method controls the attack success rate for each model by adjusting the weight of the attack function in the generation process, while maintaining minimum distortion. We used Tensorflow, a widely used machine learning library, and MNIST as the dataset. Experimental results show that the proposed method can control the attack success rate for each model by considering the attack priority of each model while maintaining minimum distortion (on average 3.95 and 2.45 in targeted and untargeted attacks, respectively).","Title: Deep Residual Inception Encoder-Decoder Network for Medical Imaging Synthesis.\nYear: 2020\n\nAbstract: Image synthesis is a novel solution in precision medicine for scenarios where important medical imaging is not otherwise available. The convolutional neural network (CNN) is an ideal model for this task because of its powerful learning capabilities through the large number of layers and trainable parameters. In this research, we propose a new architecture of residual inception encoder–decoder neural network (RIED-Net) to learn the nonlinear mapping between the input images and targeting output images. To evaluate the validity of the proposed approach, it is compared with two models from the literature: synthetic CT deep convolutional neural network (sCT-DCNN) and shallow CNN, using both an institutional mammogram dataset from Mayo Clinic Arizona and a public neuroimaging dataset from the Alzheimer's Disease Neuroimaging Initiative. Experimental results show that the proposed RIED-Net outperforms the two models on both datasets significantly in terms of structural similarity index, mean absolute percent error, and peak signal-to-noise ratio.","Title: Hybrid approach of parallel implementation on CPU–GPU for high-speed ECDSA verification\nYear: 2019\n\nAbstract: Since the advent of deep belief network deep learning technology in 2006, artificial intelligence technology has been utilized in various convergence areas, such as autonomous driving and medical care. Some services requiring fast decision making and action typically work seamlessly with edge computing service model. In autonomous driving of a connected vehicle with vehicle-to-everything (V2X) communication, roadside unit (RSU) acts as an edge computing device and it will make safer service by processing V2X messages fast, sent by vehicles or other devices. IEEE 1609.2 standard provides application message security technology to ensure the security and reliability of vehicle-to-vehicle communication messages. It uses elliptic curve digital signature algorithm (ECDSA) signatures based on the NIST p256 curve for message authenticity. In this paper, we investigate that RSU should be able to verify 3500 ECDSA signatures per second considering the expected maximum number of vehicles on nearby roads (e.g., during rush hour), message transmission rate, and IEEE 802.11p wireless channel capacity. RSU should satisfy this requirement without assistance of hardware-based cryptographic accelerator. For the requirement, we propose a hybrid approach of parallel ECDSA signature verification at high speed by using CPU and GPU, simultaneously. Moreover, we implemented the proposed method in various modern computing environments for RSU and edge computing devices. Through the experiments, we reach the conclusion that GPU can contribute to the required performance of ECDSA signature verification in RSU platform, which could not satisfy the above throughput only with CPU unit. The target platform with Intel Pentium E6500 CPU and GeForce GTX650 GPU can verify 5668 signatures per second with 30% utilization, while CPU in the platform can process only 2640 signatures. Even in a higher-performance edge computing device, we examine experimentally that the performance can be further improved by using the proposed hybrid approach.","Title: Restricted Evasion Attack: Generation Of Restricted-Area Adversarial Example\nYear: 2019\n\nAbstract: Deep neural networks (DNNs) show superior performance in image and speech recognition. However, adversarial examples created by adding a little noise to an original sample can lead to misclassification by a DNN. Conventional studies on adversarial examples have focused on ways of causing misclassification by a DNN by modulating the entire image. However, in some cases, a restricted adversarial example may be required in which only certain parts of the image are modified rather than the entire image and that results in misclassification by the DNN. For example, when the placement of a road sign has already been completed, an attack may be required that will change only a specific part of the sign, such as by placing a sticker on it, to cause misidentification of the entire image. As another example, an attack may be required that causes a DNN to misinterpret images according to a minimal modulation of the outside border of the image. In this paper, we propose a new restricted adversarial example that modifies only a restricted area to cause misclassification by a DNN while minimizing distortion from the original sample. It can also select the size of the restricted area. We used the CIFAR10 and ImageNet datasets to evaluate the performance. We measured the attack success rate and distortion of the restricted adversarial example while adjusting the size, shape, and position of the restricted area. The results show that the proposed scheme generates restricted adversarial examples with a 100% attack success rate in a restricted area of the whole image (approximately 14% for CIFAR10 and 1.07% for ImageNet) while minimizing the distortion distance.","Title: A Feature Transfer Enabled Multi-Task Deep Learning Model on Medical Imaging.\nYear: 2019\n\nAbstract: •Multi-task deep model for medical imaging-based applications.•First fully automatic system handling detection, segmentation and classification.•Cross-view feature transfer free of negative transferring issue.","Title: Selective Untargeted Evasion Attack: An Adversarial Example That Will Not Be Classified as Certain Avoided Classes.\nYear: 2019\n\nAbstract: Deep neural networks (DNNs) have useful applications in machine learning tasks involving recognition and pattern analysis. Despite the favorable applications of DNNs, these systems can be exploited by adversarial examples. An adversarial example, which is created by adding a small amount of noise to an original sample, can cause misclassification by the DNN. Under specific circumstances, it may be necessary to create a selective untargeted adversarial example that will not be classified as certain avoided classes. Such is the case, for example, if a modified tank cover can cause misclassification by a DNN, but the bandit equipped with the DNN must misclassify the modified tank as a class other than certain avoided classes, such as a tank, armored vehicle, or self-propelled gun. That is, selective untargeted adversarial examples are needed that will not be perceived as certain classes, such as tanks, armored vehicles, or self-propelled guns. In this study, we propose a selective untargeted adversarial example that exhibits 100% attack success with minimum distortions. The proposed scheme creates a selective untargeted adversarial example that will not be classified as certain avoided classes while minimizing distortions in the original sample. To generate untargeted adversarial examples, a transformation is performed to minimize the probability of certain avoided classes and distortions in the original sample. As experimental datasets, we used MNIST and CIFAR-10, including the Tensorflow library. The experimental results demonstrate that the proposed scheme creates a selective untargeted adversarial example that exhibits 100% attack success with minimum distortions (1.325 and 34.762 for MNIST and CIFAR-10, respectively).","Title: New Complementary Points of Cyber Security Schemes for Critical Digital Assets at Nuclear Power Plants.\nYear: 2019\n\nAbstract: As the importance of cyber security for nuclear power plants has recently increased, related organizations such as manufactures, operators, regulators, and research institutes have taken into consideration the many concerns for effective application of cyber security to critical digital assets at nuclear power plants. It is essential to analyze and research how to effectively apply nuclear cyber security requirements and what complementary factors are present in the overall process for a nuclear security technology system. The complementary points of nuclear cyber security technology could be analyzed through diverse cyber security methods such as cyber security assurance, assessments, and software development life cycle. It is necessary to apply nuclear cyber security technology effectively in accordance with the operation, assurance, and system development process of a nuclear facility. These analysis results could be applied to technical points for the cyber security of critical digital assets at nuclear power plants in the future and to technology that needs to be at the appropriate level of cyber security to ensure technical assurance. This paper is a new approach to compare and analyze various methods used in the cyber security field to discover complementary points for the application of cyber security to critical systems in nuclear power plants. Additionally, this paper also presents effective methods for developing and applying, evaluating, and regulating cyber security in nuclear digital critical systems using the complementary points of various cyber security methods identified in the analysis.","Title: ZeroKernel: Secure Context-Isolated Execution on Commodity GPUs\nYear: 2021\n\nAbstract: In the last decade, the dedicated graphics processing unit (GPU) has emerged as an architecture for high-performance computing workloads. Recently, researchers have also focused on the isolation property of a dedicated GPU and suggested GPU-based secure computing environments with several promising applications. However, despite the security analysis conducted by the prior studies, it has been unclear whether a dedicated GPU can be leveraged as a secure processor in the presence of a kernel-privileged attacker. In this paper, we first demonstrate the security of dedicated GPUs through comprehensive studies on context information for GPU execution. The paper shows that a kernel-privileged attacker can manipulate the GPU contexts to redirect memory accesses or execute arbitrary GPU codes on the running GPU kernel. Based on the security analysis, this paper proposes a new on-chip execution model for the dedicated GPU and a novel defense mechanism supporting the security of the on-chip execution. With comprehensive evaluation, the paper assures that the proposed solutions effectively isolate sensitive data in on-chip storages and defend against known attack vectors from a privileged attacker, supporting that the commodity GPUs can be leveraged as a secure processor."],"string":"[\n \"Title: Indirect Star-Type Networks for Large Multiprocessor Systems\\nYear: 1991\\n\\nAbstract: The authors propose three indirect star-type networks, the indirect star networks I and II and the star-delta network, and investigate their properties. An indirect star-type network is obtained by unfolding the star graph. The star-delta network is obtained through an unfolding scheme based on the recursive property of the star graph, and has n-1 switching stages. The star-delta network has the advantage of being controlled by the destination tag routing scheme. The indirect star-type networks are to the star graph as the indirect cube-type networks are to the n-cube. The authors analyze the performance of the indirect star-type networks under uniform traffic to investigate their potential as an alternative to the indirect cube-type networks for the future high-performance large multiprocessor systems.\",\n \"Title: Improved batch exponentiation\\nYear: 2009\\n\\nAbstract: An improved batch exponentiation algorithm is proposed that enhances the combination step of M'Raihi-Naccache's batch exponentiation algorithm with a decremental combination strategy. In comparison with M'Raihi-Naccache's algorithm for 160-bit and 1024-bit exponents, the proposed algorithm reduces the workload per exponentiation by about 15% in both cases when the optimal batch size is applied.\",\n \"Title: The deflection self-routing Banyan network: A large-scale ATM switch using the fully adaptive self-routing and its performance analyses\\nYear: 1999\\n\\nAbstract: Because the Internet traffic, that will be the major traffic of broadband integrated services digital networks, is bursty when cells are being switched within the multistage switching network, it has a higher possibility that multiple cells arriving simultaneously at a switching element through different incoming links may have to be forwarded along the same outgoing link. We propose a high-performance large-scale ATM switch dealing with such link contention problem. It is a new unbuffered augmented Banyan network using fully adaptive self-routing control: the deflection self-routing Banyan network. To utilize all the links of the network as alternate paths, we employ the deflection-routing algorithm in each switching element, such that cells failing to get selected for the intended link are sent along different links, in the hope that they later return, or detour the contended link and continue their journey to the destination. Cells are never dropped within the switching network, whereas the switch has no multiple cell buffers. The proposed routing is as simple as that of the generic Banyan network, and all the switch elements (SEs) have a uniform structure. To design the proposed network and its self-routing, we use the topological properties that all the SEs of the Banyan network are arranged in a regular pattern topologically. We formulate and prove these properties through an algebraic formalism. We also ran a performance analysis to provide quantitative comparison against the Banyan network and the replicated Banyan networks. As a result, we show that the new network has a far better performance and scalability than the other networks\",\n \"Title: Throttle and Preempt: A New Flow Control for Real-Time Communications in Wormhole Networks\\nYear: 1997\\n\\nAbstract: In this paper, we study wormhole routed networks and their suitability for real-time traffic in a priority-driven paradigm. A traditional blocking flow control in wormhole routing may lead to a priority inversion in the sense that high priority packets are blocked by low priority packets for unlimited time. This uncon-trolled priority inversion causes the frequent deadline missing. This paper therefore proposes a new flow control called throttle and preempt flow control, where high priority packets can preempt network resources held by low priority packets, if necessary. As a result, this flow control does not cause priority inversion. Our simulations show that the throttle and preempt flow control dramat-ically reduces deadline miss ratio without extra virtual channels. It is also observed that the throttle and preempt flow control offers shorter delay for non-real-time traffic than existing real-time flow control does.\",\n \"Title: Autonomous Clustering Scheme For Wireless Sensor Networks Using Coverage Estimation-Based Self-Pruning\\nYear: 2005\\n\\nAbstract: Energy-efficient operations are essential to prolonging the lifetime of wireless sensor networks. Clustering sensor nodes is one approach that can reduce energy consumption by aggregating data, controlling transmission power levels, and putting redundant sensor nodes to sleep. To distribute the role of a cluster head, clustering approaches should be based on efficient cluster configuration schemes. Therefore, low overhead in the cluster configuration process is one of the key constraints for energy-efficient clustering. In this paper, we present an autonomous clustering approach using a coverage estimation-based self-pruning algorithm. Our strategy for clustering is to allow the best candidate node within its own cluster range to declare itself as a cluster head and to dominate the other nodes in the range. This same self-declaration strategy is also used in the active sensor election process. As a result, the proposed scheme can minimize clustering overheads by obviating both the requirements of collecting neighbor information beforehand and the iterative negotiating steps of electing cluster heads. The proposed scheme allows any type of sensor network application, including spatial query execution or periodic environment monitoring, to operate in an energy-efficient manner.\",\n \"Title: An efficient pre-authentication scheme for IEEE 802.11-based vehicular networks\\nYear: 2007\\n\\nAbstract: In vehicular networks, supporting user mobility is one of the most challenging issues. Recently, as the desires for the high mobility and high-quality real-time services increase, fast handoff among base stations comes to a center of quality of connections. Therefore, minimizing re-authentication latency during handoff is crucial for revolutionizing the driving environment on public vehicular networks in terms of safety and convenience. In this study, we propose an efficient pre-authentication scheme for fast and secure handoff in IEEE 802.11-based vehicular networks. The proposed scheme reduces the handoff delay by reducing 4-way handshake to 2-way handshake between an access point and a mobile vehicle station during the re-authentication phase. Furthermore, the proposed scheme gives little burden over the proactive key predistribution scheme while satisfying 802.11i security requirements.\",\n \"Title: Active Queue Management for Flow Fairness and Stable Queue Length\\nYear: 2011\\n\\nAbstract: Two major goals of queue management are flow fairness and queue-length stability However, most prior works dealt with these goals independently. In this paper, we show that both goals can be effectively achieved at the same time. We propose a novel scheme that realizes flow fairness and queue-length stability. In the proposed scheme, high-bandwidth flows are identified via a multilevel caching technique. Then, we calculate the base drop probability for resolving congestion with a stable queue, and apply it to individual flows differently depending on their sending rates. Via extensive simulations, we show that the proposed scheme effectively realizes flow fairness between unresponsive and TCP flows, and among heterogeneous TCP flows, while maintaining a stable queue.\",\n \"Title: Bandwidth efficient key distribution for secure multicast in dynamic wireless mesh networks\\nYear: 2009\\n\\nAbstract: In the near future, various multicast based services will be provided over wireless mesh networks. For secure multicast services, various tree based group key management schemes have been introduced until now. Traditional tree based approaches mainly focus on reducing the number of rekeying messages transmitted by the key distribution center. However, they do not consider the network bandwidth used for transmitting each rekeying message.We propose a bandwidth efficient key tree management scheme for dynamic wireless mesh networks where membership changes occur frequently. Simulation results show that our scheme effectively reduces the bandwidth consumption used for rekeying compared to existing key tree schemes.\",\n \"Title: Improving Handover Quality in 4G Mobile Systems\\nYear: 2008\\n\\nAbstract: In this paper, we propose a new handover algorithm to guarantee handover quality in 4G mobile systems. The proposed algorithm limits the handover interruption time by improving the HARQ retransmission latency of the first packet transmitted from new serving cell. Through the simulations, we proved that our algorithm meets the requirement of handover interruption time for TCP services with high rate.\",\n \"Title: Efficient And Secure Self-Organized Public Key Management For Mobile Ad Hoc Networks\\nYear: 2008\\n\\nAbstract: This paper presents a fully self-organized key management scheme for mobile ad hoc networks. Unlike most previous schemes. there is no priori shared secret or no priori trust relationship in the proposed schemed every node plays the same role and carries out the same function of key management. The proposed scheme consists of (1) Handshaking (HS) and (2) Certificate request/reply (CRR) procedures. In HS, a node acquires the public key of the approaching node via a secure side channel. In CRR, a node requests certificates of a remote node via a radio channel to the nodes that it has HSed. If the number of received valid certificates that contain the same public key exceeds a given threshold, the node accepts the remote node's public key as valid. Security is rigorously analyzed against various known attacks and network costs are intensively analyzed mathematically. Using this analysis, we provide parameter selection guideline to optimize performance and to maintain security for diverse cases. Simulation results show that every node acquires the public keys of all other nodes at least 5 times faster than in a previous scheme.\",\n \"Title: Long-term flow fairness with fixed memory space.\\nYear: 2009\\n\\nAbstract: We provide a minimal overhead FIFO queue management scheme which controls link congestion and realizes long-term flow fairness. The proposed scheme is very simple but effective to regulate long term high bandwidth flows. The proposed scheme uses light resources including additional small fixed size memory. The evaluation result verifies the max-min convergence of the proposed mechanism and shows t...\",\n \"Title: An efficient identity-based proxy signcryption for secure broadcast\\nYear: 2010\\n\\nAbstract: To our best knowledge, no identity-based proxy broadcast signcryption scheme is known despite of the broadcast communication and the ability of delegation is widely used in many hierarchical ubiquitous computing groups. In this paper, we describe the first identity-based proxy broadcast signcryption scheme(IBPBSC). The proposed scheme supports the dynamic broadcast signcryption while achieving the security properties as well as providing the ability of delegating the signcrypting capacity from an original signcrypter to a proxy signcrypter. It is also very efficient in the aspect of communication because the size of ciphertexts and private keys in the scheme is constant, regardless of the size of receiver set. In addition, we analyze the proposed scheme from efficiency and security points of view.\",\n \"Title: Performance Analysis of Multipath Multistage Interconnection Networks with Nonuniform Output Traffic Distribution\\nYear: 1996\\n\\nAbstract: Abstract: Develops the analysis of output buffers in multipath multistage interconnection networks. The delay-throughput performance of output buffers in multipath interconnection networks is closely related to the output traffic distribution, which is the distribution of packet arrival rates at each output link connected to a given output module. Many multipath networks produce nonuniform output traffic distributions, even if the input traffic pattern is uniform. We propose in this paper a performance model for output buffer analysis with respect to such a nonuniform output traffic distribution and analyze the output buffer performance of several multipath networks by our proposed analysis model. It is shown that the output traffic distributions are different for the various multipath networks and the output buffer performance (such as packet loss probability and delay) gets better as the nonuniformity of the output traffic distribution becomes higher.\",\n \"Title: An efficient delegation protocol with delegation traceability in the X.509 proxy certificate environment for computational grids\\nYear: 2008\\n\\nAbstract: The X.509 proxy certificate is widely used to delegate an entity's right to another entity in the computational grid environment. However, this proxy certificate has two drawbacks: the potential security threat caused by non-traceability of a delegation chain and the inefficiency caused by an interactive communication between the right grantor and the right grantee on the delegation protocol. To address these problems, a new delegation protocol is presented. The proposed protocol employs an ID-based key generation technique to support delegation traceability and non-interactive delegation. Since access-right delegation occurs frequently in the computational grid environment, the proposed protocol can enhance security by providing delegation traceability and can enhance efficiency by reducing the inter-domain communication cost.\",\n \"Title: Congestion control for sudden bandwidth changes in TCP\\nYear: 2012\\n\\nAbstract: In this paper, we propose a novel technique to deal with sudden bandwidth changes in transmission control protocol (TCP). In the current Internet, sudden bandwidth changes may occur because of vertical handovers between heterogeneous access networks, routing path changes, cognitive ratio, and multi-rate wireless local area network. The current implementation of TCP is designed and optimized for stable networks and does not adapt well upon sudden bandwidth changes. Consequently, it might suffer from packet losses in burst upon sudden bandwidth decrement and under-utilization upon sudden bandwidth increment. To resolve this problem, we propose to modify the current TCP algorithm to include a new phase, called fast adaptation (FA). The FA phase is triggered upon detecting sudden bandwidth changes, and a TCP sender in the FA phase attempts to recover lost packets quickly to avoid spurious timeouts upon sudden bandwidth decrement. Upon sudden bandwidth increment, it increases its window size drastically to realize full utilization. Through extensive simulations, experiments, and analysis, it is shown that the proposed scheme can effectively deal with sudden bandwidth changes. Copyright © 2011 John Wiley & Sons, Ltd.\",\n \"Title: Optimal modulation and coding scheme selection in cellular networks with hybrid-ARQ error control\\nYear: 2008\\n\\nAbstract: We propose an optimal modulation and coding scheme (MCS) selection criterion for maximizing user throughput in cellular networks. The proposed criterion adopts both the Chase combining and incremental redundancy based hybrid automatic repeat request (HARQ) mechanisms and it selects an MCS level that maximizes the expected throughput which is estimated by considering both the number of transmissions and successful decoding probability in HARQ operation. We also prove that the conventional MCS selection rule is not optimized with mathematical analysis. Through link-level and system-level simulations, we show that the proposed MCS selection criterion yields higher average cell throughput than the conventional MCS selection schemes for slowly varying channels.\",\n \"Title: Rogue Public Key Registration Attack And The Importance Of 'Proof Of Possession' In The Pki Environment\\nYear: 2006\\n\\nAbstract: The security vulnerabilities of a number of provable secure proxy signature schemes are examined,with the assumption that users can register their public keys without having corresponding private keys. This assumption is different from that of a standard proxy signature in which the public keys of users are authentic. Under this assumption, both the Triple Schnorr scheme and Kang et al's scheme are shown to be vulnerable to a rogue public key registration attack. This attack gives an adversary the ability to generate a proxy signature without the valid agreement of the original signer. Moreover, it is shown that an adversary can manipulate the description of a delegated signing right at will. This work can be considered as an awakening to the importance of Proof of Possession (PoP) in the PKI environment, as in many cases certificate authorities do not require the PoP protocol, as has been stated [1].\",\n \"Title: Adaptive Hysteresis Using Mobility Correlation for Fast Handover\\nYear: 2008\\n\\nAbstract: The value of hysteresis has a significant effect on the handover performance measured in terms of the handover initiation delay and the number of handovers. To determine the hysteresis value effectively, a mobility correlation among users is utilized. In this letter, a grouping algorithm and an adaptive hysteresis algorithm based on the mobility correlation are proposed. It is shown that the propo...\",\n \"Title: Novel recursive multicast algorithms in multistage interconnection networks for ATM switches\\nYear: 1999\\n\\nAbstract: In this paper, we study the multicast communication in the self-routing multistage interconnection network (MIN) for constructing asynchronous transfer mode (ATM) switch architectures. Many of these applications require multicast communications in addition to conventional point-to-point communications. Multicast communication which the same message is delivered from a source to an arbitrary number of destinations, is fundamental in supporting collective communication primitives including cable TV, teleconferencing, and video-on-demand services. This paper presents a novel approach to supporting multicast communication, on the basis of the restricted address encoding scheme which constructs a short fixed-size multicast header and the recursive scheme that recycles a multicast packet one or more times through the network to reach at desired destinations. We also propose novel recursive multicast algorithms providing deadlock-freedom in MIN-based ATM switches. The emphasis is on evaluating the performance of these algorithms in terms of the number of recycling passes, the complexity of switching elements, and the number of links used. The proposed algorithms can be easily applicable to buffered MIN-based ATM switches.\",\n \"Title: Novel algorithms for multicast communication in self-routing MIN-based ATM switches\\nYear: 1996\\n\\nAbstract: We discuss the multicast communication in the self-routing multistage interconnection network (MIN) for constructing the internal architecture of asynchronous transfer mode (ATM) switches. Many applications of ATM switches require multicast communications in addition to conventional point-to-point communications. This paper presents a novel approach to supporting multicast communication, on the basis of the recursive scheme that recycles a multicast packet one or more times through the network to reach at desired destinations. We also propose cost-effective multicast algorithms providing deadlock-freedom in MIN-based ATM switches. The proposed algorithms require a small and fixed number of recycling passes and a reasonable number of links used. The proposed algorithms can be easily applicable to buffered MIN-based ATM switches.\",\n \"Title: An effective traffic management scheme using adaptive handover time in next-generation cellular networks\\nYear: 2007\\n\\nAbstract: Next-generation cellular networks are expected to support various multimedia services over IP networks with high spectral efficiency. In these networks, hotspot cells can occur when available wireless resources at some location are not sufficient to sustain the needs of users. The hotspot cell can potentially lead to blocked or dropped calls, which can deteriorate the service quality for users. A group of users enjoying multimedia services could move around in the networks generating heavy flows of traffic. This situation can generate a hotspot cell which has a short lifespan of only a few minutes. If there is a steady increase in the number of these users, the hotspot cell which has a short lifespan will occur more frequently in the overall service area. In this paper, we propose a handover-based traffic management scheme which can effectively deal with hotspot cells in next-generation cellular networks. With our scheme, the current serving cell can recognize the traffic load status of the target cell in advance, before handover execution. Then, according to the load status, it adaptively controls the handover time. The handover-based traffic management scheme can effectively and flexibly handle hotspot cells in the networks. Acceptable service quality can also be supported as users continuously maintain communication links. In the simulation results, we find that our scheme generates a smaller number of hotspot cells and supports higher service quality than the schemes compared.\",\n \"Title: Real-time analysis of intrusion detection alerts via correlation\\nYear: 2006\\n\\nAbstract: With the growing deployment of networks and the Internet, the importance of network security has increased. Recently, however, systems that detect intrusions, which are important in security countermeasures, have been unable to provide proper analysis or an effective defense mechanism. Instead, they have overwhelmed human operators with a large volume of intrusion detection alerts. This paper presents a fast and efficient system for analyzing alerts. Our system basically depends on the probabilistic correlation. However, we enhance the probabilistic correlation by applying more systematically defined similarity functions and also present a new correlation component that is absent in other correlation models. The system can produce meaningful information by aggregating and correlating the large volume of alerts and can detect large-scale attacks such as distributed denial of service (DDoS) in early stage. We measured the processing rate of each elementary component and carried out a scenario-based test in order to analyze the efficiency of our system. Although the system is still imperfect, we were able to reduce the numerous redundant alerts 5.5% of the original volume without distorting the meaning through two-phase reduction. This ability reduces the management overhead drastically and makes the analysis and correlation easy. Moreover, we were able to construct attack scenarios for multistep attacks and detect large-scale attacks in real time.\",\n \"Title: Distributed Fair Scheduling for Wireless Mesh Networks Using IEEE 802.11\\nYear: 2010\\n\\nAbstract: In IEEE-802.11-based wireless mesh networks (WMNs), unfair bandwidth sharing may arise, because the carrier sense multiple access with collision avoidance protocol is designed to provide per-station fairness only in one hop. As the hop count from a mobile client to the gateway node increases, the throughput of the node drastically decreases. In this paper, we propose a fair bandwidth allocation scheme for multiradio multichannel WMNs. This scheme provides fair bandwidth sharing among the nodes in a WMN, regardless of their hop distance from the gateway node. To achieve fairness, we first estimate the number of active nodes attached to each router and calculate the effective weights of routers based on the estimation. Then, we differentiate their contention window using their weights. For this method, we derive a multihop packet collision model. The proposed scheme is fully distributed and does not require any global information. Through an extensive simulation study, we show that our scheme ensures per-node fairness without loss of the total aggregate throughput.\",\n \"Title: On the modulo M translators for the prime memory system\\nYear: 1990\\n\\nAbstract: A fast binary to modulo M translation, where M is a prime number, is necessary in many applications. The prime memory system of an array processor which allows parallel, conflict-free access to various slices of a data array is one such application. We review existing modulo M translators, including the one used in the Burroughs Scientific Processor (BSP). We then propose a very simple and fast translator for a restricted but very useful class of M which includes, as a subclass, the form suggested by P. Budnik and D. J. Kuck ( IEEE Trans. Comput . C-20, 12 (Dec. 1971), 1566–1569). Our basic idea stems from the traditional “casting out nine” rule, which has been used as a quick method for checking the accuracy of remainders in the decimal number system. By generalizing this rule and extending it for the binary number system, we have obtained a new modulo M translator. The new modulo M translator is built upon a set of modulo adders, and is simpler and faster than previous modulo M translators such as the one employed in the BSP.\",\n \"Title: A Retransmission-Enhanced Duty-Cycle Mac Protocol Based On The Channel Quality For Wireless Sensor Networks\\nYear: 2010\\n\\nAbstract: Duty cycle MAC protocols have been proposed for wire less sensor networks (WSNs) to reduce the energy consumed by idle listening but they introduce significant end to end delivery latency Several works have attempted to mitigate this latency but they still have a problem on handling the packet loss The quality of the wireless channel in WSNs is quite bad so packets are frequently lost In this letter we present a novel duty cycle MAC protocol called REMAC (Retransmission Enhanced duty cycle MAC) which exploits both the network layer and the physical layer information REMAC estimates the quality of the wireless channel and properly reserves the wireless channel to handle the packet loss It can reduce the end to end packet delivery latency caused by the packet loss without sacrificing the energy efficiency Simulation results show that REMAC outperforms RMAC in terms of the end to end packet delivery latency\",\n \"Title: Tamper resistant software by integrity-based encryption\\nYear: 2004\\n\\nAbstract: There are many situations in which it is desirable to protect a piece of software from illegitimate tampering once it gets distributed to the users. Protecting the software code means some level of assurance that the program will execute as expected even if it encounters the illegitimated modifications. We provide the method of protecting software from unauthorized modification. One important technique is an integrity-based encryption, by which a program, while running, checks itself to verify that it has not been modified and conceals some privacy sensitive parts of program.\",\n \"Title: Handoff Procedure For Seamless Service In Ip And Ofdm Based 4g Mobile Systems\\nYear: 2005\\n\\nAbstract: We propose an efficient and practical seamless handoff scheme for 4G mobile systems based on IP and OFDM. The seamless handoff scheme obtains the physical channel for handoff in a contention-free manner with pre-synchronization and pre-forwarding IP contexts. As a result, it thoroughly decreases the physical channel blocking time as well as IP layer context-switching time to minimize total handoff delay.\",\n \"Title: A robotic service framework supporting automated integration of ubiquitous sensors and devices\\nYear: 2007\\n\\nAbstract: In recent years, due to the emergence of ubiquitous computing technology, a new class of networked robots called ubiquitous robots has been introduced. The Ubiquitous Robotic Companion (URC) is our conceptual vision of ubiquitous service robots that provides its user with the services the user needs, anytime and anywhere, in the ubiquitous computing environments. There are requirements to be met for the vision of URC. One of the essential requirements is that the robotic systems must support ubiquity of services. This means that a robot service must always be available even though there are changes in the service environment. More specifically, a robotic system needs to be interoperable with sensors and devices in its current service environments automatically, rather than statically pre-programmed for its environment. In this paper, the design and implementation of an infrastructure for URC called Ubiquitous Robotic Service Framework (URSF) is presented. URSF enables automated integration of networked robots in a ubiquitous computing environment by the use of Semantic Web Services Technologies.\",\n \"Title: Throttle and preempt: A flow control policy for real-time traffic in wormhole networks\\nYear: 1999\\n\\nAbstract: In this paper, we study wormhole routed networks and envision their suitability for real-time traffic in a priority-driven paradigm. A traditional blocking flow control in wormhole routing may lead to a priority inversion in the sense that high priority packets are blocked by low priority packets for unlimited time. This uncontrolled priority inversion causes the frequent deadline missing even at a low network load. This paper therefore proposes a new flow control called throttle and preempt flow control, where high priority packets can preempt network resources held by low priority packets, if necessary. As a result, this flow control does not cause priority inversion. Our simulations show that the throttle and preempt flow control dramatically reduces deadline miss ratio for various real-time traffic configurations without extra virtual channels. It is also observed that the throttle and preempt flow control offers shorter delay for non-real-time traffic than the existing real-time flow control does.\",\n \"Title: A Multi-service Group Key Management Scheme for Stateless Receivers in Wireless Mesh Networks\\nYear: 2010\\n\\nAbstract: Wireless mesh networks facilitate the development of the many group oriented applications by extending the coverage area of the group communication. Group communication in a wireless mesh network is complicated due to dynamic intermediate mesh points, access control for communications between different administrative domains, and the absence of a centralized network controller. In this study, we propose a topology-matching decentralized multi-service group key management scheme for wireless mesh networks. It allows service providers to update and deliver their group keys to valid members in a distributed manner using the identity-based encryption scheme. The analysis result indicates that the proposed scheme has advantages with regard to the rekeying cost and storage overhead for a member and a mesh point in multi-sender group communication environments. The stateless property is also achieved such that a stateless member, who could not be constantly online, can easily decrypt the rekeying messages without recording the past history of transmission.\",\n \"Title: An Efficient Tdma Scheduling Scheme For Wireless Sensor And Actor Networks\\nYear: 2008\\n\\nAbstract: This paper presents an efficient time slot assignment algorithm for a wireless sensor and actor network (WSAN), which consists of stationary sensors for detecting events and mobile actors for performing tasks. TDMA protocols are suitable for WSAN due to time-critical tasks, which are assigned to actors. In order to improve the performance of TDMA protocol, a time slot assignment algorithm should generate not only efficient TDMA scheduling but also reduce periodic run-time overhead. The proposed algorithm offers 0(delta(2)) run-time in the worst case, where delta is the maximum number of one-hop and two-hop neighbors in the network. The average run-time in simulation results is far less than 0(delta(2)), however, while the maximum number of assigned slots is bounded by 0(delta). In order to reduce the run-time further, we introduce two fundamental processes in the distributed slot assignment and design the algorithm to optimize these processes. We also present an analysis and verify it using ns-2 simulations. Although the algorithm requires time synchronization and prior knowledge of two-hop neighbors, simulation results show that it reduces the run-time significantly and has good scalability in dense networks.\",\n \"Title: Algorithm learning based neural network integrating feature selection and classification\\nYear: 2013\\n\\nAbstract: Feature selection and classification techniques have been studied independently without considering the interaction between both procedures, which leads to a degraded performance. In this paper, we present a new neural network approach, which is called an algorithm learning based neural network (ALBNN), to improve classification accuracy by integrating feature selection and classification procedures. In general, a knowledge-based artificial neural network operates on prior knowledge from domain experience, which provides it with better starting points for the target function and leads to better classification accuracy. However, prior knowledge is usually difficult to identify. Instead of using unknown background resources, the proposed method utilizes prior knowledge that is mathematically calculated from the properties of other learning algorithms such as PCA, LARS, C4.5, and SVM. We employ the extreme learning machine in this study to help obtain better initial points faster and avoid irrelevant time-consuming work, such as determining architecture and manual tuning. ALBNN correctly approximates a target hypothesis by both considering the interaction between two procedures and minimizing individual procedure errors. The approach produces new relevant features and improves the classification accuracy. Experimental results exhibit improved performance in various classification problems. ALBNN can be applied to various fields requiring high classification accuracy.\",\n \"Title: A design of macro-micro CDMA cellular overlays in the existing big urban areas\\nYear: 2001\\n\\nAbstract: We focus on delivering radio engineers a practical design of macro-micro code division multiple access (CDMA) cellular overlays. First, we review our algorithmic approach to jointly deploy macrocells and microcells over today's big urban areas having spatially nonuniform traffic distributions. Next, we identify several further issues related to the optimal design of macro-micro cellular overlays and enhance the cell-deploying algorithm to reflect these issues. The numerical results by extensive event-driven simulations show that the resulting macro-micro cellular overlays successfully cope with the existing conditions of today's big urban areas, such as spatial and temporal traffic distributions and user mobility characteristics. Finally, we discuss the practical guidelines for designing macro-micro cellular overlays in the existing big urban areas\",\n \"Title: Reactive Key Management Scheme For Access Control In Group Communications\\nYear: 2007\\n\\nAbstract: To control various access privileges in group-oriented applications having multiple data streams, we present a novel reactive key management scheme where each member can obtain the key of a data stream from public parameters only when necessary. Compared with the previous schemes, this scheme significantly reduces the amount of rekey messages for dynamic membership change due to its reactive nature.\",\n \"Title: B-banyan and B-delta networks for multiprocessor systems\\nYear: 1989\\n\\nAbstract: A new class of graphs, B-banyan, and a new class of multistage interconnection networks based on B-banyans, B-delta networks, in multiprocessor environments are proposed in this paper. A B-banyan is a regular rectangular SW-banyan augmented with a backward link at each vertex, in a way that preserves the destination control. A backward link can be taken when there is a conflict at a node, or when a faulty link or node is encountered. As such, backward links in B-delta networks significantly increase the network performance while at the same time providing a degree of fault tolerance. The performance of B-delta networks is analyzed and shown to be comparable to that of single-buffered delta networks. Different switches and control complexities of these two types of delta networks are discussed.\",\n \"Title: MRMAC: medium reservation MAC protocol for reducing end-to-end delay and energy consumption in wireless sensor networks\\nYear: 2010\\n\\nAbstract: This letter proposes a novel medium reservation MAC protocol that reduces both end-to-end delay and energy consumption for wireless sensor networks. MRMAC reduces endto-end delay by informing the intended receiver of NPAT (next packet arrival time) and MRI (medium reservation information) to reserve the medium in advance. Through this medium reservation phase, following packets can be delivered with reduced delay in multihop streaming phase. A simulation shows that the result of the proposed protocol outperforms previous works.\",\n \"Title: On The Best Fit Submesh Allocation Strategy In Mesh-Connected Multicomputers\\nYear: 1996\\n\\nAbstract: The submesh allocation problem is to recognize and locate a free submesh that can accommodate a request for a submesh of a specified size. An efficient submesh allocation strategy is required for achieving high performance on mesh multicomputers. In this paper, we propose a new best fit submesh allocation strategy. The proposed strategy maintains and uses a free submesh list to get global information for free submeshes. For an allocation request, the strategy tries to allocate a best fit submesh which causes the least amount of potential fragmentation so as to preserve the large free submeshes to be as many as possible and to prevent processor fragmentation for later requests. For this purpose, we introduce a novel function for quantifying the degree of potential fragmentation of submeshes. The proposed strategy has the complete submesh recognition capability. Extensive simulation is carried out to compare it with the previous strategies, and experimental results indicate that it exhibits the best performance along with an about 30% average improvement over the previous best strategy.\",\n \"Title: User-Perceived Qos In A Wireless Packet Network With Multiple Channel Conditions\\nYear: 2006\\n\\nAbstract: This paper analyzes the user-perceived QoS (Quality of Service) of interactive data application, the most notable being web browsing, in a shared wireless packet access network. We present elaborate analytical results regarding the mean page delay, the Equivalent Circuit Rate (ECR), and the system saturation point for interactive data users with different channel conditions. The ECR for a user in a shared access network is the channel rate that a dedicated connection would need in order to offer an equivalent user experience. The system saturation point is the maximum number of users that can be served with a given tolerable service delay in a shared packet access system. These analytical results are useful in the performance prediction of a real wireless environment with multiple user groups that are served with different service rates due to multiple channel conditions. Therefore, our analytical results can be used in the network planning, admission control, and service differentiation for different user groups. Our analysis is demonstrated to be in good accord with computer simulations performed on model systems based on EDGE TDMA and FEEE 802.16 OFDMA.\",\n \"Title: Miguard: Detecting And Guarding Against Malicious Iframe Through Api Hooking\\nYear: 2011\\n\\nAbstract: Recently, client-side attacks through the Microsoft Internet Explorer have increased. In this paper, we present a method to detect and block malware programs resulting from successful malicious iframe attacks. This method can detect malware program execution through distinguishing API sequences of normal execution and abnormal API sequences resulting from an exploit using Win32 API hooks. We implemented MiGuard (Guard against malicious iframes) and performed experiments. The evaluation results indicate that our approach can effectively detect and block malicious iframes. We also believe that our research can help prevent threats of malicious iframes.\",\n \"Title: Eventor: an authoring system for interactive multimedia applications\\nYear: 1994\\n\\nAbstract: Programming, testing, and maintaining interactive multimedia applications (IMAs) are still difficult and expensive, while substantial progress has been made to reduce the burden on authors. As IMAs get larger and more complex the difficulties will increase. To overcome the complexity of such IMAs, we argue that authoring systems should provide such facilities as (1) a traditional and intuitivedivide-and-conquer paradigm for solving large and complex problems in various fields, (2)formal specification of the behaviors of IMAs for checking the syntactic correctness of visual expressions or semantic anomalies, and (3)automatic aids like validation of temporal constraints and verification of visual expressions. In this paper, we investigate the properties of IMAs for recognizing the inherent interactivity and concurrency. We propose a specification method based on Milner'sCalculus of Communicating Systems (CCS), which is a well-known formal mechanism for specifying the concurrency in various distributed applications. We also design and implement an authoring system calledEventor (Event Editor), which is based on CCS and composed of three tools: a Temporal Synchronizer, a Spatial Synchronizer, and a User Interaction Builder. They focus on describing the temporal and spatial synchronizations and user interactions while they rely on existing tools in Intel's Digital Video Interactive (DVI) for supporting other functionalities. By editing a simple computer aided instruction (CAI) application, we illustrate that our specification mechanism is well-suited for handling the interactivity of multimedia applications, and Eventor is a simple, efficient, and powerful enough tool to handle practical applications. Especially the incremental refinement and the formal specification based on the CCS allow Eventor to be extended with formal verifications to cope with large and complex applications.\",\n \"Title: A practical approach of ID-based cryptosystem in ad hoc networks\\nYear: 2007\\n\\nAbstract: As research in ad hoc networking has advanced, the importance of security increases more and more. But it suffers from the restriction that ad hoc networks do not have the established infrastructure. In this paper we propose a security-enhanced model with ID-based cryptosystem which removes the necessity for any infrastructure and provides sound authentication. Also we provide a more secure and concrete routing protocol with aggregate signature. For confidentiality, we also provide a key exchange protocol. And our model not only distributes the role of key generation center in ID-based cryptosystem to the nodes with threshold cryptography, but also increases availability by providing a secret re-sharing protocol. Therefore, our model is very appropriate for ad hoc networks. Copyright (c) 2007 John Wiley & Sons, Ltd.\",\n \"Title: An Efficient Proactive Key Distribution Scheme for Fast Handoff in IEEE 802.11 Wireless Networks\\nYear: 2007\\n\\nAbstract: Supporting user mobility is one of the most challenging issues in wireless networks. Recently, as the desires for the user mobility and high-quality multimedia services increase, fast handoff among base stations comes to a center of quality of connections. Therefore, minimizing re-authentication latency during handoff is crucial for supporting various promising real-time applications such as Voice over IP (VoIP) on public wireless networks. In this study, we propose an enhanced proactive key distribution scheme for fast and secure handoff based on IEEE 802.11i authentication mechanism. The proposed scheme reduces the handoff delay by reducing 4-way handshake to 2-way handshake between an access point and a mobile station during the re-authentication phase. Furthermore, the proposed scheme gives little burden over the proactive key pre-distribution scheme while satisfying 802.11i security requirements.\",\n \"Title: Performance evaluation of a class of multipath packet switching interconnection networks\\nYear: 1993\\n\\nAbstract: This paper serves two purposes. First, it demonstrates that the analytic models and techniques used for unique-path networks can also be adopted for a class of multipath networks, called plusminus-2 (PM2I) networks. This is done by comparison of the analytic results and simulation results for unbuffered PM2I networks. Second, it extends earlier performance analysis of single-buffered delta networks, based on (2 x 2) crossbar switches, to handle different network types and switch sizes. From the extended performance model, numeric results for single-buffered delta networks based on switches larger than (2 )< 2) and single-buffered PM2I networks are obtained and compared.\",\n \"Title: An Analysis of Punctuated Equilibria in Simple Genetic Algorithms\\nYear: 1997\\n\\nAbstract: In the running of a genetic algorithm, the population is liable to be confined in the local optimum, that is the metastable state, making an equilibrium. It is known that, after a long time, the equilibrium is punctuated suddenly and the population transits into the better neighbor optimum. We adopt the formalization of Computational Ecosystems to show that the dynamics of the Simple Genetic Algorithm is represented by a differential equation focusing on the population mean of a phenotype. Referring to the studies of differential equations of this form, we show that the duration time of metastability is exponential in the population size and other parameters, on the one dimensional bistable fitness landscape which has one metastable and one stable state.\",\n \"Title: Drop-and-reroute: a new flow control policy for adaptive wormhole routing\\nYear: 1995\\n\\nAbstract: Adaptive routing is one of the promising approaches to overcome the packet contention problem in the network using wormhole routing. We propose a new flow control policy, called drop-and-reroute , for adaptive wormhole routing and describe architectural supports and its operation. Drop-and-reroute flow control further exploits the adaptivity achieved by adaptive routing by allowing a blocked packet to be routed again in the preceding node. We also measure the effect of drop-and-reroute flow control on the network performance by extensive simulation. Our simulation results show that drop-and-reroute flow control considerably improves the network performance of the adaptive routing algorithm under nonuniform traffic patterns.\",\n \"Title: A Resilient And Efficient Replication Attack Detection Scheme For Wireless Sensor Networks\\nYear: 2009\\n\\nAbstract: In a large-scale sensor network. replicated hostile nodes may be used for harsh inner attacks. To detect replicas, this paper presents a distributed, deterministic. and efficient approach robust to node compromise attacks Without incurring significant resource overheads.\",\n \"Title: Autonomous Broadcast Pruning In Wireless Ad Hoc Networks Using Coverage Estimation\\nYear: 2009\\n\\nAbstract: Purpose - The purpose of this paper is to find a practical method to minimize unnecessary rebroadcasts for ad hoc networks in a remote area.Design/methodology/approach - The authors design a theoretical coverage estimation model for autonomous broadcast pruning. To verify the effect of the model, simulations are performed to evaluate the effect of confidence level and the actual performance by using ns2 network simulator.Findings - The algorithm used to predict the coverage area of broadcasts can minimize unnecessary rebroadcasts.Originality/value - Autonomous broadcast pruning scheme based on the local prediction of a remained coverage area in the on-going broadcast process brings us beneficial results in terms of energy savings and limited bandwidth preservation.\",\n \"Title: An efficient mapping of Boltzmann Machine computations onto distributed-memory multiprocessors\\nYear: 1992\\n\\nAbstract: In this paper, an efficient mapping scheme of Boltzmann Machine computations onto a distributed-memory multiprocessor, which exploits the synchronous spatial parallelism, is presented. In this scheme, the neurons in the Boltzmann Machine are partitioned into p disjoint sets, and each set is mapped on a processor of a p-processor system. A parallel convergence and learning algorithms of Boltzmann Machine, necessary communication pattern among the processors, and their time complexities when neurons are partitioned and mapped onto a distributed-memory multiprocessor are investigated. An expected p-processor speed-up of the parallelizing scheme over a single processor is also analyzed theoretically. It can be used as a basis in determining the most cost-effective or optimal number of processors with respect to the communication capabilities and interconnection topologies of given distributed-memory multiprocessor.\",\n \"Title: On the Design of Delay based Admission Control in Hierarchical Networks.\\nYear: 2014\\n\\nAbstract: Today, as the hierarchical cellular system is getting more attention than before, some recent studies introduce delay based admission control (AC) scheme which delays the admission to the macro-embedded small cell for a relatively short time to prevent unnecessary handover caused by the short-term visitors of the small cell area. In such delay based ACs, when we use improper delay parameter, the system frequently makes incorrect handover decisions such as where unnecessary handover is allowed due to too short delaying, or where necessary handover is denied due to too long delaying. In order to avoid these undesirable situations as much as possible, we develop a new delay parameter decision method based on probabilistic cell residence time approximations. By the extensive numerical and analytical evaluations, we determine the proper delay parameter which prevents the incorrect handover decision as much as possible. We expect our delay parameter decision method can be useful system administration tips in hierarchical cellular system where delay based AC is adopted.\",\n \"Title: Performance analysis of multi-buffered packet-switching networks in multiprocessor systems\\nYear: 1987\\n\\nAbstract: In this paper, we extend Jenq [1]'s performance analysis method of single-buffered banyan networks, to be applicable for the multi-buffered packet-switching interconnection networks in multiprocessor systems. Earlier analyses on the buffered interconnection network performances assumed either single or infinite buffers at each input (output) port of a switch. As far as the multi-buffered interconnection network is concerned, only some simulation results for the delta networks have been known [2].\",\n \"Title: Reliability Measure For Sound Source Localization\\nYear: 2008\\n\\nAbstract: In this paper, we propose a method of measuring the reliability of direction-of-arrival (DOA) estimation results. One idea for the reliability measure arises from a waterbed effect in DOA estimation. The reliability measure can be obtained from the difference between the maximum and minimum point of cross-correlation. If the calculated reliability measure has a lower value than a predefined threshold, the estimated DOA is regarded as a wrong result and subsequently discarded. Some experiments show that the proposed method can reject perturbed results of the estimated DOA.\",\n \"Title: IEEE 802.11b WLAN performance with variable transmission rates: in view of high level throughput\\nYear: 2005\\n\\nAbstract: Wireless networks have been rapidly integrated with the wired Internet and have been widely deployed. In particular, IEEE 802.11b WLAN is the most widespread wireless network today. The IEEE 802.11b WLAN supports multiple transmission rates and the rate is chosen in an adaptive manner by an auto rate control algorithm. This auto rate control algorithm deeply affects the total system performance of the IEEE 802.11b WLAN. In this paper, we examine the WLAN performance with regard to the auto rate control algorithm, especially the ARF scheme, which is the most popular auto rate control algorithm in 802.11b based WLAN products. The experimental results indicate that the ARF scheme works well in the face of signal noise due to node location. However, the ARF scheme severely degrades system performance when multiple nodes contend to obtain the wireless channel and the packet is lost due to signal collision.\",\n \"Title: A Novel Intrusion Tolerant System Using Live Migration\\nYear: 2014\\n\\nAbstract: This letter proposes a novel intrusion tolerant system consisting of several virtual machines (VMs) that refresh the target system periodically and by live migration, which monitors the many features of the VMs to identify and replace exhausted VMS. The proposed scheme provides adequate performance and dependability against denial of service (DoS) attacks. To show its efficiency and security, we conduct experiments on the CSIM20 simulator, which showed 22% improvement in a normal situation and approximately 77.83% improvement in heavy traffic in terms of the response time compared to that reported in the literature. We measure and compare the response time. The result show that the proposed scheme has shorter response time and maintains than other systems and supports services during the heavy traffic.\",\n \"Title: Automated teleoperation of web-based devices using semantic web services\\nYear: 2005\\n\\nAbstract: In this paper, we present SWATS which supports task-oriented automated teleoperation of Web-based devices. The proposed system employs Semantic Web Services technology and AI planning technique to achieve operational automaticity.\",\n \"Title: Digital rights management with right delegation for home networks\\nYear: 2006\\n\\nAbstract: The purpose of digital rights management (DRM) is to protect the copyrights of content providers and to enable only designated user to access digital contents. For a user to share the contents among all his devices in the home network, several domain-based approaches that group multiple devices into a domain have been proposed. In these approaches, however, each device in a domain has equivalent rights on all contents although certain contents require an access control between the devices. In this paper, a new DRM system for home networks is presented. This system enables access control on the contents by a right delegation strategy with proxy certificates. Moreover, it also provides additional functionalities, including restricted sharing and temporal sharing of contents, which are necessary for ordinary scenarios in home networks.\",\n \"Title: An improved algorithm for protocol validation by extended circular exploration\\nYear: 1993\\n\\nAbstract: In this paper, to improve the efficiency of state exploration and to relieve the problem of state explosion, we propose another method of state exploration called the extended circular exploration (ECE) , which does not need to explore all the reachable global states and can be applied to N ( N ≥2)-party protocol with alternative routes, i.e. it is applicable to the protocol in which adaptive routing mechanism can be performed. Using this ECE, we can eliminate the topology restriction of the circular exploration (CE) through extending the notion of circular exploration. We explore only those global states which are reachable, provided that the participant processes of any group of transitions proceed at the same speed, and that they can be formed as a cycle. The state space thus explored is not exhaustive. The algorithm presented can detect deadlock error and unspecified reception error. It requires storage space and/or execution time much less than those of the conventional perturbation method. It can be viewed as a solution of the N -process collision and the interference mechanism.\",\n \"Title: Bit-permute multistage interconnection networks\\nYear: 1995\\n\\nAbstract: In this paper, we consider a class of log N stage interconnection networks called Bit-Permute Multistage Interconnection Networks (BPMIN's) where the ports of each switch of a stage are different at only one bit position of their labels. We describe the decomposition structure of the BPMIN's and prove that all of the BPMIN's are topologically equivalent and some of them are functionally equivalent. We also identify a class of 2 log N stage rearrangeable networks called symmetric BPMIN's where two log N stage BPMIN's are connected in sequence. The symmetric BPMIN's are either symmetric or asymmetric and regular or irregular in their inter-stage connections and can be reduced into 2 log N -1 stages by combining the two center stages. We show that the symmetric BPMIN's constitute larger class of rearrangeable networks than ever known. We also propose a general routing algorithm for the symmetric BPMIN's by modifying slightly the looping algorithm of the Benes network.\",\n \"Title: An Energy-Efficient Dispersion Method For Deployment Of Mobile Sensor Networks\\nYear: 2010\\n\\nAbstract: A novel method for deploying mobile sensors is proposed. The proposed method minimizes the energy consumption by deciding the target position before movement. It also deploys the sensor nodes in an optimal layout which provides maximum coverage and robust connectivity. To that end, we propose a novel topology, rake tree, in which each vertex can be matched to a position in the optimal layout. Simulation results show that the proposed method results in more coverage and consumes less energy compared to previous works.\",\n \"Title: Secure and efficient data retrieval over encrypted data using attribute-based encryption in cloud storage\\nYear: 2013\\n\\nAbstract: The cloud storage based information retrieval service is a promising technology that will form a vigorous market in the near future. Although there have been numerous studies proposed about secure data retrieval over encrypted data in cloud services, most of them focus on providing the strict security for the data stored in a third party domain. However, those approaches require stupendous costs centralized on the cloud service provider, which could be a principal impediment to achieve efficient data retrieval in cloud storage. In this paper, we propose an efficient data retrieval scheme using attribute-based encryption. The proposed scheme is best suited for cloud storage systems with massive amount of data. It provides rich expressiveness as regards access control and fast searches with simple comparisons of searching entities. The proposed scheme also guarantees data security and user privacy during the data retrieval process.\",\n \"Title: Active Trial-and-error Attack on SASC Protocols\\nYear: 2006\\n\\nAbstract: SASC (Server-Aided Secret Computation) protocols en- able a client (a smart card) to borrow computing power from a server (e.g., an untrustworthy auxiliary device like an ATM) without revealing its secret information. In this paper, we propose a new active attack on server-aided secret computation protocols. We describe our attack by using Beguin and Quisquater's protocol. (We modify the protocol in order to immunize it against Nguyen and Stern's lattice reduction attack.) The proposed attack re- duces the search space P to 1 p + pP, where 0 < p < 1. It is 2 p P for optimal p. Practically, it effectively threatens SASC protocols because an attacker can choose an appro- priate value p according to the situation. Therefore, the security parameters in the existing SASC protocols must be reconsidered.\",\n \"Title: Fault-tolerant multicasting in MINs for ATM switches\\nYear: 1998\\n\\nAbstract: This article proposes a fault-tolerant multicast routing algorithm in multistage interconnection networks (MINs) for ATM switch architectures. It employs both region and cube encoding schemes as the header encoding scheme. A multicast packet can be routed to its destinations in only two phases through the MIN having a single faulty element.\",\n \"Title: A new deadlock prevention scheme for nonminimal adaptive wormhole routing\\nYear: 1994\\n\\nAbstract: In multicomputer networks, the adaptive routing has been expected as a promising way to improve network performance by utilizing available network bandwidth. Previous adaptive routing algorithms in wormhole-routed multicomputer networks restrict the routing of messages to prevent deadlocks, and the routing restriction results in low degree of adaptiveness and low utilization of communication channels. In this paper, we examine the possibility of performing restriction-free, nonminimal adaptive routing in wormhole-routed networks as an approach to further improving the performance of these networks. A new flow control policy, called message cutting , is proposed, and two adaptive routing strategies are presented. Freedom of communication deadlock is achieved by the proposed flow control policy. The proposed adaptive routing strategies do not restrict routing and maximally utilize the physical and virtual channels. Simulation results show that the restriction-free adaptive routing approach is promising from the fact that it has the lowest latency and highest throughput depending on the number of virtual channels per physical channel and patterns of message traffic.\",\n \"Title: Cost-Effective Algorithms For Multicast Connection In Atm Switches Based On Self-Routing Multistage Networks\\nYear: 1998\\n\\nAbstract: In this paper, we discuss the multicast connection in the self-routing multistage interconnection network (MIN) for constructing the internal architecture of asynchronous transfer mode (ATM) switches. Many applications of ATM switches require multicast connections in addition to conventional point-to-point connections. Multicast connection, in which the same message is delivered from a source to an arbitrary number of destinations, is fundamental in supporting collective connection primitives including cable TV, teleconferencing, and video-on-demand services. This paper presents a novel approach to supporting multicast connection, on the basis of the recursive scheme that recycles a multicast packet one or mon times through the network to reach desired destinations. We also propose cost-effective multicast algorithms providing deadlock-freedom in MIN-based ATM switches. The proposed algorithms require a small and fixed number of recycling passes and a reasonable number of links used. The proposed algorithms can be easily applied to buffered MIN-based ATM switches. (C) 1998 Elsevier Science B.V.\",\n \"Title: A Smart Handover Decision Algorithm Using Location Prediction for Hierarchical Macro/Femto-Cell Networks\\nYear: 2011\\n\\nAbstract: To reduce the number of unnecessary handover in hierarchical macro/femto-cell networks, it is necessary to avoid macro → femto cell handovers of temporary femtocell visitors who stay in the femtocell for a relatively short time. In this paper, we propose a smart handover decision algorithm exploiting future mobility pattern prediction scheme to prevent macro → femto cell handovers of such temporary femtocell visitors. Our simulation result shows that the proposed algorithm effectively reduces the number of unnecessary handovers.\",\n \"Title: A new efficient submesh allocation strategy for mesh multicomputers\\nYear: 1996\\n\\nAbstract: The submesh allocation problem is to recognize and locate a free submesh that can accommodate a request for a submesh of a specified size. This paper proposes a new best-fit submesh allocation strategy. The proposed strategy maintains and uses a free submesh list to get global information for free submeshes. For an allocation request, the proposed strategy tries to allocate a best-fit submesh which causes the least amount of potential fragmentation so as to preserve as many large free submeshes as possible and to prevent processor fragmentation for later requests. For this purpose, we introduce a novel function for quantifying the degree of potential fragmentation of submeshes. The proposed strategy has the complete submesh recognition capability. Extensive simulation is carried out to compare the proposed strategy with previous strategies, and experimental results indicate that the proposed strategy exhibits the best performance, along with about a 30% average improvement over the best previous strategy\",\n \"Title: Enhanced cooperative communication MAC for mobile wireless networks\\nYear: 2013\\n\\nAbstract: In this paper, we propose a new MAC (medium access control) protocol called enhanced cooperative communication MAC (ECCMAC) based on IEEE 802.11. The major objective of ECCMAC is to maximize the benefits of cooperative communication. We first propose a scheme for selecting and maintaining the best relay node. Second, since both cooperative communication and network coding rely on the selection of a relay node, we consider exploiting a network coding technique for additional throughput improvement. Third, to accommodate asymmetric link rates between a sender and a relay node, we employ ECCMAC to measure forward and reverse link rates, whereas prior works have simply assumed symmetric rates. ECCMAC is evaluated in this paper through theoretical analysis, extensive simulation, and simulation with measured data, and the results show that ECCMAC effectively improves wireless network performance.\",\n \"Title: Adaptive Resource Reservation for Efficient Resource Utilization in the Wireless Multimedia Network\\nYear: 2007\\n\\nAbstract: With the increase of wireless network and demand for the variety of services, the handoff resource reservation for the Quality of Service(QoS) will be significant issue in future wireless networks. But on the other hand, the efficiency of resource usage also needs to be considered for preventing waste of network resources due to the reservation. Therefore, this paper suggests an efficient resource reservation scheme to support QoS provision as well as the efficient resource utilization simultaneously. In the proposed scheme, handoff probability estimation with the filtering algorithm is used to calculate the optimal amount of resources to be reserved in the neighbor cells. So, the mobile units with the proposed scheme can predict the amount of resources required in the future and reserve resource effectively. The performance simulation shows that the proposed scheme is advantageous with the balance of QoS provision and resource utilization in the wireless network.\",\n \"Title: Chosen ciphertext secure authenticated group communication using identity-based signcryption\\nYear: 2010\\n\\nAbstract: Efficient access control with scalable rekeying is one of the most important requirements for secure group communications, especially in a dynamic and large group. In a many-to-many group communication environment, sender authentication is also another important security requirement as each participant can potentially be both a sender and a receiver. In this study, we propose an authenticated group communication scheme which is secure against an adaptive chosen ciphertext attack using identity-based signcryption. The proposed scheme allows multiple senders to dynamically multicast messages into an arbitrary group of receivers determined by the senders. In the proposed scheme, the group member can be a stateless receiver. Data confidentiality of the group communication is guaranteed as well as the sender authentication.\",\n \"Title: Two Real-Time Flow Controls In Wormhole Networks\\nYear: 1997\\n\\nAbstract: In this paper, we study wormhole routed networks and envision their suitability for real-time traffic in a priority-driven paradigm. A traditional blocking flow control in wormhole routing may lead to a priority inversion in the sense that high priority packets are blocked by low priority packets for unlimited time. The priority inversion causes the frequent deadline missing even at a low network load. This paper therefore proposes two preemptive flow control policies where high priority packets can preempt network resources held by low priority packets. As a result, the proposed flow controls can resolve the priority inversion. Our simulations show that preemptive flow controls significantly reduce deadline miss ratios for Various real-time traffic configurations.\",\n \"Title: A heuristic processor allocation strategy in hypercube systems\\nYear: 1991\\n\\nAbstract: A new processor allocation scheme for hypercube systems, called the HPA (heuristic processor allocation) strategy, is presented. In this scheme, an undirected graph, called the SC-graph (Subcube-graph), is used to maintain the free subcubes available in system, which are represented by vertices. An allocation request for a k-cube is satisfied by finding a free subcube of dimension k in the SC-graph or by decomposing a nearest higher dimension subcube. If there are more than one subcube of dimension k, a subcube which has minimum degree in the SC-graph is selected to reduce the external fragmentation. For deallocating the released subcube a heuristic algorithm is used to maintain the dimension of free subcube as high as possible. It is theoretically shown that the HPA strategy is not only statically optimal but also it has a complete subcube recognition capability in a dynamic environment. Extensive simulation results show that the HPA strategy improves the performance and significantly reduces the allocation/deallocation time compared to the previously proposed schemes.\",\n \"Title: Customized Forwarding In Application Layer Multicast For Real-Time Services\\nYear: 2010\\n\\nAbstract: We propose a customized forwarding (CF) scheme for tree-based application layer multicasts to support real-time interactive multimedia services. We show that CF reduces residual loss rate and play pause frequency significantly.\",\n \"Title: Base station association schemes to reduce unnecessary handovers using location awareness in femtocell networks\\nYear: 2013\\n\\nAbstract: In this paper, we propose a predictive scheme to generate reduced neighbor lists. In addition, we formulate a predictive association game. The game is designed to determine an optimal base station for maximum throughput and reduced handover frequency. Several simulation scenarios are set forth herein for analysis to show how our schemes improve performance in terms of throughput and handover frequency.\",\n \"Title: Restriction-Free Adaptive Wormhole Routing In Multicomputer Networks\\nYear: 1995\\n\\nAbstract: The adaptive routing approach has been expected as a promising way to improve network performance by utilizing available network bandwidth. Previous adaptive routing strategies in wormhole-routed multicomputer networks restrict the routing of messages by the routing algorithm to prevent deadlock. This results in low degree of adaptivity and low utilization of physical or virtual channels. In this paper, we examine the possibility of performing restriction-free adaptive routing in wormhole-routed networks as an approach to further improving the performance of these networks. A new how control policy, called message cutting-in, is proposed, and two adaptive routing strategies are presented. Freedom of communication deadlock is achieved by the proposed flow control policy. The proposed adaptive routing strategies do not restrict routing and maximally utilize the physical and virtual channels. Simulation results show that the restriction-free adaptive routing approach is promising from the fact that it has the lowest latency and highest throughput depending on the number of virtual channels per physical channel and patterns of message traffic.\",\n \"Title: Accelerating Key Establishment Protocols for Mobile Communication\\nYear: 1999\\n\\nAbstract: Mobile communication is more vulnerable to security attacks such as interception and unauthorized access than fixed network communication. To overcome these problems, many protocols have been proposed to provide a secure channel between a mobile station and a base station. However, the public-key based protocols are not fully utilized due to the poor computing power and the small battery capacity of a mobile station. In this paper, we propose some techniques accelerating public-key based key establishment protocols between a mobile station and a base station. The proposed techniques enable a mobile station to borrow computing power from a base station without revealing its secret information. The proposed schemes accelerate the previous protocols up to five times and reduce the amount of power consumption of a mobile station. The proposed schemes use SASC (Server-Aided Secret Computation) protocols that are used for smart cards. Our insight is that the unbalanced property in computing power of the mobile communication is similar to that of the smart card system. The acceleration degrees of the proposed schemes are quite different from one another according to the used SASC protocols. In this paper, we analyze the acceleration factors of the proposed schemes and compare them with one another. The analysis shows that one of the approach presents outstanding performance among them.\",\n \"Title: Performance Analysis of a Multicast Switch Based on Multistage Interconnection Networks\\nYear: 1997\\n\\nAbstract: In this paper, we study multicasting in the self-routing multistage interconnection networks (MINs) for asynchronous transfer mode (ATM) switch architectures. Many B-ISDN applications require multicast connections in addition to conventional point-to-point connections. This paper presents a novel approach to support multicast connection, on the basis of a restricted address encoding scheme which constructs a short fixed-size multicast header and a recursive scheme that recycles a multicast packet one or more times through the network to send it to the desired destinations. The proposed two-phase multicast algorithm provides deadlock-free multiple multicast connections in MIN-based ATM switches. The emphasis is on analyzing the performance of an unbuffered MIN-based switch using the multicast algorithm in terms of network throughput. The proposed algorithm can be easily applied to buffered MIN-based ATM switches.\",\n \"Title: Security Considerations for Handover Schemes in Mobile WiMAX Networks\\nYear: 2008\\n\\nAbstract: IEEE 802.16e uses EAP-based authentication and key management for link layer security. Due to the lack of ability to support mobility, however, EAP-based key management becomes a principal impediment to the achievement of an efficient and secure handover in IEEE 802.16e mobile WiMAX networks. In this paper, an overview of the EAP-based handover procedures of the latest IEEE 802.16e standard is given and their security flaws are analyzed. Possible solutions for secure handover in IEEE 802.16e networks are also proposed in this paper. The proposed handover protocol guarantees a backward/forward secrecy while gives little burden over the previous handover protocols.\",\n \"Title: The cyclic banyan network: a fault tolerant multistage interconnection network with the fully-adaptive self-routing\\nYear: 1995\\n\\nAbstract: In this paper, we introduce a fault tolerant fully-adaptive self-routing scheme for a class of augmented banyan network. It can provide more multiple paths than the related previous networks between an input/output pair of a network by adding extra links between switching elements in the same stage and extending the self-routing scheme of the banyan network using the topological properties of the banyan network, which are discovered in this paper. Our routing scheme is as simple as that of the banyan network, which is based on the topological relationships among the switching elements (SEs) in the same stage. We show that the SEs in the same stage are arranged regularly, in other words, each stage of the banyan network is constructed with sequences of a cyclic group of SEs, so we can use not only the inter-stage relationships as the banyan network, but also intra-stage relationships among all switching elements for our scheme. We present an algebraic proof to show the correctness of this routing scheme, and an analytic reliability analysis to provide quantitative comparisons with other networks, which shows that the new network is more reliable than the banyan network and other augmented MINs\",\n \"Title: A Decentralized Multi-Group Key Management Scheme\\nYear: 2009\\n\\nAbstract: Scalability is one of the most important requirements for secure multicast in a multi-group environment. In this study, we propose a decentralized multi-group key management scheme that allows each multicast group sender to control the access to its group communication independently. Scalability is enhanced by local rekeying and inter-working among different subgroups. The group key secrecy and backward/forward secrecy are also guaranteed.\",\n \"Title: The B-Network: A Multistage Interconnection Network with Backward Links\\nYear: 1990\\n\\nAbstract: A multistage interconnection network (MIN) for multiprocessor systems is proposed. The proposed MIN, called the B-network, uses backward links to provide backward paths for the requests blocked at switches or memory due to contentions. The gamma network is known to contain a cube network (specifically, the inverse omega network) as a substructure. The B-network is obtained from the gamma network by preserving the cube structure but reversing the direction of all other links. These backward links are used as alternate paths for requests blocked due to path or memory contentions. The B-network can be controlled by the simple destination tag control algorithm; packets navigating through the B-network, using both regular forward links and backward links, can reach their destinations under the destination tag control. The performance of the B-network is analyzed under the uniform traffic model and compared to various networks of interest. It is shown that the B-network surpasses the performance of the gamma network, the crossbar switch, and single-buffered MINs based on (2*2) switches, while having the same hardware complexity as the gamma network.\",\n \"Title: Providing forwarding assurance in multi-hop wireless networks\\nYear: 2010\\n\\nAbstract: The Assured Forwarding (AF) service in a differentiated services network offers different levels of forwarding assurance for IP packets. In the wired Internet, the AF service is implemented as a queue management scheme associated with a drop policy in each router. In multi-hop wireless networks, however, queue management in each node is not enough to provide service differentiation globally since resource is shared among neighboring nodes. Hence, several studies for the AF service in wireless networks have proposed to manipulate the contention window size to provide service differentiation. The contention window size based differentiation scheme provides the AF service by transmitting packets in a higher class with a smaller contention window. However, since network congestion in multi-hop wireless networks causes packet losses at the link layer, the contention window size based differentiation scheme is not sufficient to provide the AF service. In this paper, we propose a rate control scheme to control load on the shared channel by adjusting sending rate according to the estimated channel state and the amount of AF traffic. We further propose a RTS retry limit adaptation scheme for loss rate differentiation at the link layer. Through extensive simulations, we show that proposed schemes are effective to differentiate forwarding assurance in multi-hop wireless networks.\",\n \"Title: A New Binary Image Authentication Scheme With Small Distortion And Low False Negative Rates\\nYear: 2007\\n\\nAbstract: In this study, a novel binary image authentication scheme is proposed, which can be used to detect any alteration of the host image. In the proposed scheme, the watermark is embedded into a host image using a Hamming-code-based embedding algorithm. A performance analysis shows that the proposed scheme achieves both smaller distortion and lower false negative rates than the previous schemes.\",\n \"Title: Authenticated public key broadcast encryption scheme secure against insiders’ attack\\nYear: 2012\\n\\nAbstract: Broadcast encryption schemes have been studied in the past decades. Recently, insiders’ attack on the broadcast encryption scheme has been attracted attention among researchers. So, several broadcast encryption schemes with sender authentication have been proposed. However, since broadcast message size in previous schemes increases linearly at the number of target members, the previous schemes are not suitable for the group with large members. In this paper, we propose a new authenticated public key broadcast encryption scheme called ω-APKBE scheme. The proposed ω-APKBE scheme provides sender authentication property with a constant size broadcast message which is nonlinear on the number of target members. Hence, the proposed scheme is more compatible to the dynamic group with large members than the previous schemes.\",\n \"Title: Parallelized Scalar Multiplication on Elliptic Curves Defined over Optimal Extension Field\\nYear: 2007\\n\\nAbstract: In this paper, we propose three algorithms to perform scalar multiplication on elliptic curves defined over higher characteristic finite fields such as the OEF (Optimal Ex- tension Field). First, we propose an efficient scalar multi- plication method in which the Frobenius expansion is used on an elliptic curve defined over OEF. Second, we pro- pose a new finite field multiplication algorithm. Third, we propose a particular polynomial squaring algorithm. We show that the proposed algorithms, when used together, accelerate the scalar multiplication on elliptic curves by two-fold.\",\n \"Title: A fast handoff scheme for IP over Bluetooth\\nYear: 2002\\n\\nAbstract: A lot of people use a cell phone, PDA, and laptop computer in their daily lives. Bluetooth which enables incompatible devices to communicate with each other over a wireless link can speed up the growth in popularity. BLUEPAC(BLUEtooth Public ACcess) is designed to support IP over Bluetooth and provide users with access to a public network such as the Internet. Our work is motivated by the need for a fast handoff in BLUEPAC A mobile device experiences frequent handoffs due to the short radius of the piconet which is about 10 m. A handoff requires a connection establishment between a mobile device and a new base station which is is a time-consuming procedure in the frequency hopping system. This situation may cause many connection losses which hurt the network performance as a result. We present a new scheme for improving the handoff performance. The basic idea is that a base station controller helps a mobile device to establish a connection to new base station by giving their addresses and clocks. Simulation results show that our scheme significantly reduces handoff delay and imposes a very light overhead on the base station.\",\n \"Title: Trust management for resilient wireless sensor networks\\nYear: 2005\\n\\nAbstract: Utilities of wireless sensor networks are standing out in bold relief in various fields such as home environmental, industrial, and military applications. Compared with the vivid applications of the sensor networks, however, the security and privacy issues of the sensor networks are still in their infancy because unique features of the sensor networks make it difficult to adopt conventional security policies. Especially, false reports are critical threats because they can drain out the finite amount of energy resources in a battery-powered sensor networks; thus, a novel trust management scheme is necessary to make resilient wireless sensor networks. Cryptographic authentication mechanisms and key management schemes cannot suggest solutions for the real root of the problem. In this paper, we propose a trust management scheme which can identify trustworthiness of sensor nodes and suggest a defensible approach against insider attacks beyond conventional cryptographic approaches.\",\n \"Title: A new appraoch to server-aided secret computation\\nYear: 1998\\n\\nAbstract: . Server-Aided Secret Computation(SASC) protocol enablesthe client(smart card) to borrow the computing power from the server(untrusted auxiliary device like ATM) without revealing the secret informationof the client. The previous approach is to decompose the client'ssecret information into many pieces and to reveal some of them to theserver. It significantly accelerates the secret computation, however thereare some problems. Firstly, many pieces that have linear relation enablean...\",\n \"Title: Sensor Node Localization By Three Mobile Anchors In The Wireless Sensor Networks\\nYear: 2011\\n\\nAbstract: For the deployment of sensor networks, the sensor localization, which finds the position of sensor nodes, is very important. Most previous localization schemes generally use the GPS signal for the sensor localization. However, the GPS signal is unavailable when there is an obstacle between the sensor nodes and satellites. Therefore, in this paper, we propose a new localization scheme which does not use the GPS signal. The proposed scheme localizes the sensors by using three mobile anchors. Because the three mobile anchors collaboratively move by themselves, it is self-localizable and can be adopted even when the sensors are randomly and sparsely deployed in the target field.\",\n \"Title: A Distributed Deterministic and Resilient Replication Attack Detection Protocol in Wireless Sensor Networks.\\nYear: 2009\\n\\nAbstract: To detect replica nodes in a wireless sensor network, we propose a distributed, deterministic and resilient (DDR) replica detection protocol developed from a witness node based strategy. In DDR, while a location claim message of each node is sent towards the designated verification location in the network, the consistency of the message is verified at intermediate nodes en route to its final destination. Compared with previous replication attack detection protocols, DDR achieves better computation and communication performance due to the use of symmetric key cryptography only and early replica detection.\",\n \"Title: Multi-rate aware partition and cooperation in WLANs\\nYear: 2011\\n\\nAbstract: Recently the prominent growing of smartphone users accelerates the demand of Wireless Local Area Network (WLAN) services more. In a large scale WLAN, the IEEE 802.11 Distributed Coordination Function (DCF) protocol has two problems; one is performance anomaly in multi-rate 802.11 environments [1]. The other is performance degradation due to frequent collision by a large number of stations [2].\",\n \"Title: Design And Performance Analysis Of Multistage Interconnection Networks Using A Recursive Multicast Algorithm\\nYear: 1996\\n\\nAbstract: In this paper, we study issues of the multicast communication in the multistage interconnection networks (MINs) for large-scale multicomputers. In addition to point-to-point communication among processing nodes, efficient collective communication is critical to the performance of multicomputers. Multicast communication in which the same packet is delivered from a source node to an arbitrary number of destination nodes is fundamental in supporting collective communication primitives including broadcast, reduction, and barrier synchronization operations. This paper presents a new approach to support multicast communication, on the basis of a restricted address encoding scheme which constructs a short fixed-size multicast header, and a recursive scheme that recycles a multicast packet one or more times through the MIN to send it to the desired destination nodes. We propose a recursive multicast algorithm which provides deadlock-freedom for multiple multicast packets in MIN-based multicomputers. We also present performance model for the unbuffered MIN using the multicast algorithm and analyze its performance in terms of the network throughput, where several multicast communications are considered. The proposed algorithm can be easily applied to wormhole or virtual cut-through MIN-based multicomputers.\",\n \"Title: New modular multiplication algorithms for fast modular exponentiation\\nYear: 1996\\n\\nAbstract: A modular exponentiation is one of the most important operations in public-key cryptography. However, it takes much time because the modular exponentiation deals with very large operands as 512-bit integers. The modular exponentiation is composed of repetition of modular multiplications. Therefore, we can rcducc the execution time of it by reducing thc execution time of each modular multiplication. In this paper, we propose two fast modular multiplication algorithms. One is for modular multiplications between different integers, and the other is for modular squarings. These proposed algorithms require single-precision multiplications fewer than those of Montgomery modular multiplication algorithms by 1/2 and 1/3 times, respectively. Implementing on PC, proposed algorithms reduce execution times by 50% and 30% compared with Montgomery algorithms, respectively.\",\n \"Title: A novel Adaptive Cluster Transformation (ACT)-based intrusion tolerant architecture for hybrid information technology\\nYear: 2013\\n\\nAbstract: Recently, the building of strong intrusion tolerant systems is in great demand since the openness and the distributed nature of information systems are easily used to compromise the systems by intentional attacks. To achieve intrusion tolerance by enabling the systems to survive various types of intrusions, we suggest a novel approach, Adaptive Cluster Transformation (ACT), in this paper. Instead of using a fixed cluster size as in conventional approaches, ACT adapts a variable cluster size depending on the system status. This is proved to maintain good quality of service (QoS). In addition, the early prediction of incoming massive packets makes ACT possible to replace any damaged clusters with new ones consisting of pristine virtual machines (VMs). This also contributes to defend the system against a Denial of Service (DoS). The performance of ACT is compared with other fixed sizes of VM cluster architectures by CSIM 20. And it is verified that the proposed method is more effective in maintaining the specific level of QoS as well as providing strong security to the targeted system.\",\n \"Title: An Efficient Causal Order Algorithm for Message Delivery in Distributed System\\nYear: 1997\\n\\nAbstract: Though causal order of message delivery simplifies the design and development of distributed applications, the overhead of enforcing it is not negligible. Causal order algorithm which does not send any redundant information is efficient in the sense of communication overhead. We characterize and classify the redundant information into four categories: information regarding just delivered, already delivered, just replaced, and already replaced messages. We propose an efficient causal order algorithm which prevents propagation of these redundant information. Our algorithm sends less amount of control information needed to ensure causal order than other existing algorithms. Since our algorithm's communication overhead increases relatively slowly as the number of processes increases, it shows good scalability feature. The potential of our algorithm is shown by simulation study.\",\n \"Title: Preamble allocation scheme using trajectory information for mobile relay stations\\nYear: 2009\\n\\nAbstract: We propose a novel preamble allocation scheme for mobile relay stations (MRS) in order to reduce the occurrence of preamble collision. We calculate the probability of preamble collision based on trajectory information of MRSs. Then, we select a preamble considering not only the probability of preamble collision but also interference from other base stations (BS) and relay stations (RS). Simulation results show that the proposed scheme significantly reduces the occurrence of preamble collision.\",\n \"Title: Self-Healing Key Distribution Scheme With Long Service Time\\nYear: 2010\\n\\nAbstract: Self-healing key distribution schemes for broadcast encryption are capable of recovering missed group keys and resisting a collusion attack of up to t revoked users. However, since previous schemes have been considered only m sessions, it does not suitable to support infinite sessions. In the paper, we propose an efficient delta-self healing key distribution scheme with revocation capability. The proposed scheme can support infinite sessions by updating the private key of users and supporting a partial revocation. The communication overhead of the proposed scheme is compared with several previous works.\",\n \"Title: A pseudo-random asynchronous duty cycle MAC protocol in wireless sensor networks\\nYear: 2010\\n\\nAbstract: We propose a pseudo-random asynchronous duty cycle MAC protocol in wireless sensor networks. The proposed protocol adopts a hash function to determine the next wake-up times. As the next wake-up times are known in advance, the sensor nodes do not need to remain active until their intended receivers wake up. Given an end-to-end delay requirement, the proposed protocol significantly decreases energy consumption by reducing the idle listening time.\",\n \"Title: Weakness of the Synchro-Difference LKH Scheme for Secure Multicast\\nYear: 2007\\n\\nAbstract: Zhu proposed a synchro-difference LKH scheme for secure multicast that appeared in IEEE Communications Letters, vol. 9, no. 5, pp. 477-479, May 2005, which is an enhancement of the well-known LKH scheme. In this letter, we show that Zhu's scheme is insecure against collusion of two or more malicious adversaries by presenting two kinds of attack scenarios.\",\n \"Title: Pre-authentication for fast handoff in wireless mesh networks with mobile APs\\nYear: 2006\\n\\nAbstract: Wireless mesh networks can extend the network service region by just adding APs. However wireless mesh networks also have the same security problems as the traditional wireless LAN. Until now, many methods have been proposed to solve the authentication problem, particularly for the fast handoff, in the traditional wireless LAN. However, previous methods are not efficient to the wireless mesh network with mobile APs because they just considered static APs. In this paper, we propose a new pre-authentication method for the wireless mesh network with mobile APs. We adapted the neighbor graph method of previous schemes for the compatibility. However, our method is suitable to the wireless mesh network by applying a Du et al's key distribution. Furthermore, we present a formal analysis about our method by using a logic based formal analysis method.\",\n \"Title: Performance of MCS selection for collaborative hybrid-ARQ protocol\\nYear: 2007\\n\\nAbstract: We propose a Modulation and Coding Scheme (MCS) selection algorithm for the collaborative hybrid Automatic-Repeat-reQuest (ARQ) protocol in order to provide high data rates. The collaborative hybrid ARQ protocol is designed to benefit from diversity gain. It exploits not only the broadcast nature of wireless channels, but also spatial diversity by formation of virtual antenna arrays and collaboration between a base station and relay nodes through Space-Time Block Coding (STBC). The proposed algorithm estimates both the effective Signal-to-Noise-Ratio (SNR) and the average throughput of each MCS level, and then selects the MCS level maximizing the average throughput. Simulation results show that the proposed algorithm outperforms conventional MCS selection algorithms in terms of the total throughput and satisfies packet delay constraints.\",\n \"Title: On Submesh Allocation for Mesh Multicomputers: A Best-Fit Allocation and a Virtual Submesh Allocation for Faulty Meshes\\nYear: 1998\\n\\nAbstract: The submesh allocation problem is to recognize and locate a free submesh that can accommodate a request for a submesh of a specified size. In this paper, we propose a new best-fit submesh allocation strategy for mesh-connected multiprocessor systems. The proposed strategy maintains and uses a free submesh list for an efficient allocation. For an allocation request, the strategy selects the best-fit submesh which causes the least amount of potential processor fragmentation. As many large free submeshes as possible are preserved for later allocations. For this purpose, we introduce a novel function quantifying the degree of potential fragmentation of submeshes. The proposed strategy has the capability of recognizing a complete submesh. We also propose an allocation strategy for faulty meshes which can maintain and allocate virtual submeshes derived from faulty submeshes. Extensive simulation is carried out to compare the proposed strategy with previous strategies. The proposed strategy has the best performance: a 6-50 percent improvement over the previous best strategy.\",\n \"Title: A systolic array exploiting the inherent parallelisms of artificial neural networks\\nYear: 1991\\n\\nAbstract: The systolic array implementation of artificial neural networks is one of the best solutions to the communication problems generated by the highly interconnected neurons. In this paper, a two-dimensional systolic array for backpropagation neural network is presented. The design is based on the classical systolic algorithm of matrix-by-vector multiplication, and exploits the inherent parallelisms of backpropagation neural networks. This design executes the forward and backward passes in parallel, and exploits the pipelined parallelism of multiple patterns in each pass. The estimated performance of this design shows that the pipelining of multiple patterns is an important factor in VLSI neural network implementations.\",\n \"Title: Wireless packet fair queueing algorithms with link level retransmission\\nYear: 2005\\n\\nAbstract: In order to provide quality of service to wireless networks, a number of wireless fair queueing algorithms have recently been proposed. They, however, require perfect channel prediction before transmission and rarely consider algorithms under the link layer. Instead of perfect channel prediction, most wireless systems adopt the Link Level Retransmission (LLR) algorithm within the link layer for recovering channel errors. However, the LLR algorithm does not work well with the previous prediction-based wireless fair queueing algorithms. Therefore, we propose a new wireless fair queueing algorithm, Wireless Fair Queueing with Retransmission (WFQ-R), which is well matched with the LLR algorithm and does not require channel prediction. In the WFQ-R algorithm, the share consumed by retransmission is regarded as a debt of the retransmitted flow to the other flows. So, the WFQ-R algorithm achieves wireless fairness with the LLR algorithm by penalizing flows that use wireless resources without permission in the link layer. Through analyses, we proved that the WFQ-R algorithm guarantees throughput and delay fairness. Through simulations, we showed that our WFQ-R algorithm maintains fairness adaptively. Furthermore, our WFQ-R algorithm is able to achieve flow separation and compensation.\",\n \"Title: An output queueing analysis of multipath ATM switches\\nYear: 2000\\n\\nAbstract: The performance of output buffers in multipath ATM switches is closely related to the output traffic distribution, which characterizes the packet arrival rate at each output link connected to the output buffer of a given output port. Many multipath switches produce nonuniform output traffic distributions even if the input traffic patterns are uniform. Focusing on the nonuniform output traffic distribution, we analyze the output buffer performances of several multipath switches under both uniform and nonuniform input traffic patterns. It is shown that the output traffic distributions are different for the various multipath switches and the output buffer performance measured in terms of packet loss probability and mean waiting time improves as the nonuniformity of the output traffic distribution becomes higher.\",\n \"Title: Fault-tolerant multicasting in multistage interconnection networks\\nYear: 1998\\n\\nAbstract: We study fault-tolerant multicasting in multistage interconnection networks (MINs) for constructing large-scale multicomputers. In addition to point-to-point routing among processor nodes, efficient multicasting is critical to the performance of multicomputers. This paper presents a new approach to provide fault-tolerant multicasting, which employs the restricted header encoding schemes. The proposed approach is based on a recursive scheme in order to send a multicast packet to the desired destinations detouring faulty element(s). In the proposed fault-tolerant multicasting, a multicast packet is routed to its own destinations in only two passes through the MIN having a number of faulty elements by exploiting its nonblocking property\",\n \"Title: A new call admission control scheme using the user conformity for non-uniform wireless systems\\nYear: 2001\\n\\nAbstract: Next generation wireless communication systems are expected to satisfy quality of service (QoS) requirements by preventing call droppings which are caused by user mobility. Call admission control (CAC) is becoming more important in guaranteeing the QoS of calls. Since existing CAC schemes estimate future requirements and reserve bandwidth to guarantee a certain level of QoS, the bandwidth estimation method is very important in CAC schemes. Most conventional schemes use the movement information of individual mobile terminals (MTs) to estimate the bandwidth requirements. While these schemes estimate the future bandwidth usage relatively well, it requires a high computational overhead and lots of memory space to keep and update the movement information of each mobile terminal, and therefore is hard to implement. We propose a call admission control scheme, which estimates the future handoffs using the handoff histories between BSs and the MT conformity. The proposed scheme achieves a low dropping ratio and high bandwidth utilization keeping a low computational overhead. The performance of the proposed scheme is evaluated through simulations\",\n \"Title: Rogue Public Key Registration Attack and the Importance of 'Proof of Possession' in the PKI Environment*This work was supported by the Ministry of Science and Technology (MOST)/Korea Science and Engineering Foundation (KOSEF) through the Advanced Information Technology Research Center (AITrc) and the MIC (Ministry of Information and Communication), Korea, under the ITRC (Information Technology Research Center) support program supervised by the IITA (Institute of Information Technology Assessment).\\nYear: 2006\\n\\nAbstract: The security vulnerabilities of a number of provable secure proxy signature schemes are examined with the assumption that users can register their public keys without having corresponding private keys. This assumption is different from that of a standard proxy signature in which the public keys of users are authentic. Under this assumption, both the Triple Schnorr scheme and Kang et al's scheme are shown to be vulnerable to a rogue public key registration attack. This attack gives an adversary the ability to generate a proxy signature without the valid agreement of the original signer. Moreover, it is shown that an adversary can manipulate the description of a delegated signing right at will. This work can be considered as an awakening to the importance of Proof of Possession (PoP) in the PKI environment, as in many cases certificate authorities do not require the PoP protocol, as has been stated [1].\",\n \"Title: EMBA: An Efficient Multihop Broadcast Protocol for Asynchronous Duty-Cycled Wireless Sensor Networks.\\nYear: 2013\\n\\nAbstract: In this paper, we propose an efficient multihop broadcast protocol for asynchronous duty-cycled wireless sensor networks (EMBA) where each node independently wakes up according to its own schedule. EMBA adopts two techniques of the forwarder's guidance and the overhearing of broadcast messages and ACKs. A node transmits broadcast messages with guidance to neighbor nodes. The guidance presents how ...\",\n \"Title: Decentralized group key management for dynamic networks using proxy cryptography\\nYear: 2007\\n\\nAbstract: Decentralized group key management mechanisms offer beneficial solutions to enhance the scalability and reliability of a secure multicast framework by confining the impact of a membership change in a local area. However, many of the previous decentralized solutions reveal the plain text to the intermediate relaying proxies,or require the key distribution center to coordinate secure group communications between subgroups. In this study, we propose a decentralized group key management scheme that features a mechanism allowing a service provider to deliver the group key to valid members in a distributed manner using the proxy cryptography. In the proposed scheme, the key distribution center is eliminated while data confidentiality of the transmitted message is provided during the message delivery process. The proposed scheme can support a secure group communication in dynamic network environments where there is no trusted central controller for the whole network and the network topology changes frequently.\",\n \"Title: Vamsd: Voronoi Diagram Based Autonomous Mobile Sensor Deployment For Maximizing Coverage\\nYear: 2010\\n\\nAbstract: This letter proposes a novel mobile sensor deployment scheme for maximizing coverage The basic idea is to force mobile sensors to move to predetermined target points that are the optimal layout in a distributed manner using Voronoi diagram data structure A simulation shows that the result of the proposed scheme is quite close to the optimal result and outperforms previous works.\",\n \"Title: Securing Mobile Agents by Integrity-Based Encryption\\nYear: 2005\\n\\nAbstract: The mobile agent paradigm is a promising technology to structure distributed applications. Since mobile agents physically move to a remote host that is under the control of a different principal, they need to be protected from this environment which is responsible for execution. In this paper, we provide a new cryptographic methodology of protecting mobile agents from unauthorized modification for the program code by malicious hosts.\",\n \"Title: A processor allocation strategy using cube coalescing in hypercube multicomputers\\nYear: 1993\\n\\nAbstract: We propose a new processor allocation strategy for hypercube multicomputers. The allocation strategy is called cube coalescing (CC) strategy since it uses a cube coalescing method to find an available subcube which is formed by the non-neighbored nodes. The main advantage of this strategy is that the nodes which do not contribute to form a requested subcube can be filtered out by coalescing, and thus the exhaustive search for possible subcubes can be avoided even in the worst case. It is theoretically shown that that proposed CC strategy is not only statistically optimal but also it has a complete subcube recognition capability.Extensive simulation is carried out to compare this strategy with the other strategies in terms of performance measures of interest such as processor utilization, completion time, and allocation/deallocation time. It is shown that the CC strategy gives better performance and the least allocation/deallocation time compared to the other schemes in a dynamic environment.\",\n \"Title: A Proximity-Based Self-Organizing Hierarchical Overlay Framework For Distributed Hash Tables\\nYear: 2007\\n\\nAbstract: Several structured peer-to-peer networks have been created to solve the scalability problem of previous peer-to-peer systems such as Gnutella and Napster. These peer-to-peer networks which support distributed hash table functionality construct a sort of structured overlay network, which can cause a topology mismatch between the overlay and the underlying physical network. To solve this mismatch problem, we propose a topology-aware hierarchical overlay framework for DHTs. The hierarchical approach for the overlay is based on the concept that the underlying global Internet is also a hierarchical architecture, that is, a network of networks. This hierarchical approach for the overlay puts forth two benefits: finding data in a physically near place with a high probability, and smaller lookup time. Our hierarchical overlay framework is different from other hierarchical architecture systems in a sense that it provides a specific self-organizing grouping algorithm. Our additional optimization schemes complete the basic algorithm which constructs a hierarchical structure without any central control.\",\n \"Title: Efficient service discovery mechanism for wireless sensor networks\\nYear: 2008\\n\\nAbstract: A ubiquitous computing environment is a large-scale mobile environment where numerous sensor devices are widely distributed and connected through wireless networks. However, existing service discovery protocols in sensor networks are inappropriate for such environments since they are not scalable. This paper proposes a DHT (Distributed Hash Table) based service discovery protocol including the mechanism that constructs topology-aware overlay networks in ubiquitous environments. Our protocol is scalable since it does not require a central lookup server and does not rely on multicast or flooding. Simulation results show that our protocol is scalable and outperforms traditional flooding-based protocols.\",\n \"Title: Peer-assisted multimedia delivery using periodic multicast.\\nYear: 2015\\n\\nAbstract: Multimedia content distribution is one of the most popular services available on the Internet, and several approaches have been developed to satisfy user requirements and reduce server load. IP multicast is one efficient mechanism for distributing content to multiple users with minimal server and network loads. Because its single stream of data is transmitted simultaneously to multiple users, IP multicast is well-suited for real-time broadcasting. However, it is difficult to offer on-demand multimedia services with IP multicast in which users request content without synchronization. In this paper, we propose an on-demand based multimedia delivery system using IP multicast. In our system, the multimedia server starts multicasting content upon receiving the first content request. Subsequent users then join the multicast session at any time during the transmission. The content that subsequent users have missed is provided by the users (peers) that joined earlier, or by server unicasts. To minimize server load, we quantitatively analyzed the optimal multicast starting point. Through simulation, we show that our proposed system outperforms prior related systems based on peer-to-peer delivery and multicasting.\",\n \"Title: A Novel Routing Protocol for Cognitive Radio Networks with Cooperation Process.\\nYear: 2014\\n\\nAbstract: Cognitive radio networks (CRNs) are composed of mobile users who can use multiple spectrum bands for communication. CRNs allow unlicensed users (called cognitive users) to efficiently utilize unused licensed spectrums without interfering with communications of licensed users (called primary users). The main goals of CRNs are to mitigate spectrum saturation and to improve spectrum utilization. This paper introduces state-of-the-art routing protocols for CRNs and addresses some limitations of these protocols. To resolve the limitations, we suggest a new research direction for routing protocols in CRNs. We implement our protocol to compare with the existing routing protocols for multi-hop CRNs. Our protocol shows good performance compared to the existing routing protocols in terms of network performance and PU protection.\",\n \"Title: Design of a Secure System Considering Quality of Service.\\nYear: 2014\\n\\nAbstract: Improvements in networking technologies have provided users with useful information services. Such information services may bring convenience and efficiency, but might be accompanied by vulnerabilities to a variety of attacks. Therefore, a variety of research to enhance the security of the systems and get the services at the same time has been carried out. Especially, research on intrusion-tolerant systems (ITSs) has been conducted in order to survive against every intrusion, rather than to detect and prevent them. In this paper, an ITS based on effective resource conversion (ERC) is presented to achieve the goal of intrusion-tolerance. Instead of using the fixed number of virtual machines (VMs) to process requests and recover as in conventional approaches, the ITS based on ERC can transform the assigned resources depending on the system status. This scheme is proved to maintain a certain level of quality of service (QoS) and quality of security service (QoSS) in threatening environments. The performance of ERC is compared with previous studies on ITS by CSIM 20, and it is verified that the proposed scheme is more effective in retaining a specific level of QoS and QoSS.\",\n \"Title: End-to-end mobility support in content centric networks.\\nYear: 2015\\n\\nAbstract: Content-centric networking CCN has been recently proposed as an alternative to traditional IP-based networking. In CCN, content is accessed by content name instead of a host identifier locational identifier. This new type of access methodology rapidly and efficiently disseminates content in combination with the in-network caching mechanism. For practical use of CCN, many network properties studied in IP-based networking are being revisited, and new types of CCN architecture components are being designed. Although mobility is an essential aspect of the future networking system, it has not been sufficiently studied. We therefore address fundamental mobility issues, such as seamless handover, optimal access point selection, network mobility, and handling of persistent interests. In addition, for each issue, we propose practical solutions that efficiently align to a CCN environment. To ensure seamless handoff, we propose various handoff schemes and compare their performance in terms of handoff latency using packet-level simulation. Because our proposed schemes are consistent with the characteristics and rules of CCN, we believe that they can easily be integrated as a part of CCN. Copyright © 2014 John Wiley & Sons, Ltd.\",\n \"Title: The PM221 interconnection network\\nYear: 1989\\n\\nAbstract: A scheme based on the standard multiplier recoding technique that enables the use of (5*5) switches for PM2I networks is presented. The connections between switching stages are based on the modified PM2I functions, and are called plus-minus-2/sup 2i/ (PM22I); hence, these networks are called the PM22I networks. Since the number of switching stages in the PM22I networks is one-half of those in the PM2I networks, with a moderate increase in the switch size from (3*3) to (5*5), these networks provide similar design tradeoffs availability by the (2*2) and (4*4) cube networks.<>\",\n \"Title: Application of fully recurrent neural networks for speech recognition\\nYear: 1991\\n\\nAbstract: The authors describe an extended backpropagation algorithm for fully connected recurrent neural networks applied to speech recognition. The extended delta rule is approximated by excluding some of the past activities of the dynamic neurons to reduce computational complexity without performance degradation. In speaker-dependent recognition of a confusable syllable set, the fully recurrent neural network with the approximated backpropagation algorithm showed better performance than the multilayer perceptron and the self-recurrent network with comparable time complexity. In addition, it is found that most self-recurrent connections become excitatory and most mutual recurrent connections become inhibitory\",\n \"Title: Design and analysis of a fair scheduling algorithm for QoS guarantees in high-speed packet-switched networks\\nYear: 1998\\n\\nAbstract: B-ISDNs are required to support a variety of services such as audio, data, and video, so that the guarantee of quality-of-service (QoS) has become an increasingly important problem. An effective fair scheduling algorithm permits high-speed switches to divide link bandwidth fairly among competing connections. Together with the connection admission control, it can guarantee the QoS of connections. We propose a novel fair scheduling algorithm, called “virtual-time-based round robin (VTRR)”. Our scheme maps the priorities of packets into classes and provides service to the first non-empty class in each round. Also, it uses an estimation method of the virtual time necessary to this service discipline. To find the first non-empty class, the VTRR adopts a priority queueing system of O(loglog c) which decreases the number of instructions which need to be carried out in one packet transmission time segment. These policies help the VTRR implementation in software, which presents flexibility for upgrades. Our analysis has demonstrated that the VTRR provides bounded unfairness and its performance is close to that of weighted fair queuing. Therefore, the VTRR has a good performance as well as simplicity, so that it is suitable for high-speed B-ISDN\",\n \"Title: Two-phase multicast in wormhole-switched bidirectional multistage Banyan networks\\nYear: 1998\\n\\nAbstract: A multistage interconnection network is a suitable class of interconnection architecture for constructing large-scale multicomputers. Broadcast and multicast communication are fundamental in supporting collective communication operations such as reduction and barrier synchronization. In this paper, we propose a new multicast technique in wormhole-switched bidirectional multistage banyan networks for constructing large-scale multicomputers. To efficiently support broadcast and multicast with simple additional hardware without deadlock, we propose a two-phase multicast algorithm which takes only two transmissions to perform a broadcast and a multicast to an arbitrary number of desired destinations. We encode a header as a cube and adopt the most upper input link first scheme with periodic priority rotation as arbitration mechanism on contented output links. We coalesce the desired destination addresses into multiple number of cubes. And then, we evaluate the performance of the proposed algorithm by simulation. The proposed two-phase multicast algorithm makes a significant improvement in terms of latency. It is noticeable that the two-phase algorithm keeps broadcast latency as efficient as the multicast latency of fanout $2^m$ where $m$ is the minimum integer satisfying $2^m \\\\geq \\\\sqrt{N}$($N$ is a network size).\",\n \"Title: Improved base-φ expansion method for Koblitz curves over optimal extension fields\\nYear: 2007\\n\\nAbstract: An improved base-phi expansion method is proposed, in which the bit-length of coefficients is shorter and the number of coefficients is smaller than in Kobayashi's expansion method. The proposed method meshes well with efficient multi-exponentiation algorithms. In addition, two efficient algorithms based on the proposed expansion method, named phi-wNAF and phi-SJSF, are presented which significantly reduce the computational effort involved in online precomputation by using the property of Frobenius endomorphism. The proposed algorithms noticeably accelerate computation of a scalar multiplication on Koblitz curves over optimal extension fields (OEFs). In particular, for OEFs where the characteristic is close to 32 bits or 64 bits, the required number of additions is reduced up to 50% in comparison with Kobayashi's base-phi scalar multiplication algorithm. Finally, a method that significantly reduces the memory usage of the precomputation table at the expense of slightly more computation is presented\",\n \"Title: The Design of a Robust Intrusion Tolerance System through Advanced Adaptive Cluster Transformation and Vulnerability-Based VM Selection\\nYear: 2013\\n\\nAbstract: In this paper, we suggested novel schemes which use advanced adaptive cluster transformation and VM selection policy for intrusion tolerant systems (ITSs). The cluster size is transformed adaptively in order to maintain a certain level of services by using the adaptive cluster expansion scheme. Also, all the servers in clusters can be substituted in case serious threat such as massive packets incoming is expected by using the adaptive cluster substitution method. If there exists reliable historical data, more fast transformation is possible. In addition, the less-vulnerable virtual machines (VMs) are chosen using evaluation policies to reduce data leakage occured from system's vulnerabilities. Simulation results done with CSIM 20 prove that the suggested schemes improve intrusion tolerance efficiently compared to other conventional methods.\",\n \"Title: Analysis of Cell Sojourn Time in Heterogeneous Networks with Small Cells\\nYear: 2016\\n\\nAbstract: Recently, heterogeneous networks with small cells have been widely used to increase the capacity of mobile systems. In such environments, accurate estimation of the mean cell sojourn time is critical for evaluating the performance of the network and its applications. It is especially important to analyze the cell sojourn times of mobile users as they reside in different network tiers: either macro-cell-only or small-cell-covered areas. But because small cells are deployed in an irregular manner, it is difficult to derive the analytical mean cell sojourn time in a macro-cell-only area. In this letter, we propose a novel approach to resolve this difficulty. We developed a simple but effective trick that approximates the heterogeneous network to a discrete grid so that it becomes tractable, making it possible to derive the analytical mean sojourn time in the macro-cell-only area. Simulation results confirm that the proposed method has excellent accuracy for general random walk mobility models including random waypoint, Brownian motion, tailored Brownian motion, and truncated Levy walk.\",\n \"Title: Efficient Packet Routing in Highly Mobile Wireless Networks\\nYear: 2015\\n\\nAbstract: Traditional mobile ad-hoc network (MANET) and delay tolerant network (DTN) routing schemes work properly only under the limited assumptions. MANET routing schemes are designed for densely populated networks while DTN routing schemes were developed for dealing with intermittently connected topologies. However, in a real environment, the assumption cannot be hold: the density of nodes constantly changes and a network can be separated due to node mobility even when it is composed of large number of nodes. We propose a routing scheme for maximizing the delivery rate in a network with varying density. In the proposed scheme, data packets are used to search an available path while they are delivered by DTN-based routing. If a path exists, subsequent packets are relayed along the path. Otherwise, packets are delivered continuously using DTN-based routing. Using simulation we show that the proposed scheme achieves high throughput and a low overhead in a dense part of the network, while it outperforms existing DTN routing schemes in an intermittently connected situation.\",\n \"Title: Hybrid Recovery-Based Intrusion Tolerant System For Practical Cyber-Defense\\nYear: 2016\\n\\nAbstract: Due to the periodic recovery of virtual machines regardless of whether malicious intrusions exist, proactive recovery-based Intrusion Tolerant Systems (ITSs) are being considered for mission-critical applications. However, the virtual replicas can easily be exposed to attacks during their working period, and additionally, proactive recovery based ITSs are ineffective in eliminating the vulnerability of exposure time, which is closely related to service availability. To address these problems, we propose a novel hybrid recovery-based ITS in this paper. The proposed method utilizes availability-driven recovery and dynamic cluster resizing. The availability-driven recovery method operates the recovery process by both proactive and reactive ways for the system to gain shorter exposure times and higher success rates. The dynamic cluster resizing method reduces the overhead of the system that occurs from dynamic workload fluctuations. The performance of the proposed ITS with various synthetic and real workloads using CloudSim showed that it guarantees higher availability and reliability of the system, even under malicious intrusions such as DDoS attacks.\",\n \"Title: Customer Data Scanner For Hands-Free Mobile Payment\\nYear: 2015\\n\\nAbstract: New payment technologies are coming that will raise user convenience. To support automatic hands-free payment, merchant devices will collect customer's information from the cloud of payment service providers or customer's smart phones, which should be removed after the transaction. Using Jaccard containment, we propose a proactive security approach of cleaning personal data at merchant-side point-of-sale terminals. We also propose a sampling method to reduce communication overhead by several orders of magnitude.\",\n \"Title: Fully Batch Processing Enabled Memory Integrity Verification Algorithm Based on Merkle Tree.\\nYear: 2015\\n\\nAbstract: Memory attacks have been increasing in number recently. Adversary can manipulate memory data or break system by doing active attacks. Especially, main memory is used as a target of attack, because main memory is more vulnerable than other components, such as CPU. To prevent adversary's active attack, memory integrity verification algorithm has been proposed. Protection of computer's memory integrity is important in situations where attacks on the computer systems are a threat. As technology has advanced, computer systems migrate from wire-based to wireless system. A lot of memory integrity verification algorithms are already developed, but these algorithms do not consider new wireless platform. Wireless platform is constrained by a lack of storage and power supply in comparison with wire-based system, therefore computational overhead and storage overhead must be considered when applying to algorithm, which is used in wireless system. In this study, integrity verification performance can be improved by doing batch-processing. Previous verification algorithms based on Merkle tree do not support fully batch processing verification. We propose fully batch processing enabled memory integrity verification algorithm based on Merkle tree. This algorithms can verify memory integrity in completely batches. For implement our algorithm, we use Incremental multiset hash function, and as a result, consume only 480-bit on-chip storage. Reducing consumption of on-chip storage leads to improving on the performance of computation. We implement our algorithm and previous memory integrity verification algorithms based on standard Merkle tree and lazy-processing Merkle tree in simulator to compare their performance. Our algorithm offers better system performance overall, especially when attack rarely occur.\",\n \"Title: Sorting Method For Fully Homomorphic Encrypted Data Using The Cryptographic Single-Instruction Multiple-Data Operation\\nYear: 2016\\n\\nAbstract: In this paper, we present a faster (wall-clock time) sorting method for numerical data subjected to fully homomorphic encryption (FHE). Owing to circuit-based construction and the FHE security property, most existing sorting methods cannot be applied to encrypted data without significantly compromising efficiency. The proposed algorithm utilizes the cryptographic single-instruction multiple-data (SIMD) operation, which is supported by most existing FHE algorithms, to reduce the computational overhead. We conducted a careful analysis of the number of required recryption operations, which are the computationally dominant operations in FHE. Accordingly, we verified that the proposed SIMD-based sorting algorithm completes the given task more quickly than existing sorting methods if the number of data items and (or) the maximum bit length of each data item exceed specific thresholds.\",\n \"Title: Preemptive Real-Time Scheduling Incorporating Security Constraint For Cyber Physical Systems\\nYear: 2016\\n\\nAbstract: Since many cyber-physical systems (CPSs) manipulate security-sensitive data, enhancing the quality of security in a CPS is a critical and challenging issue in CPS design. Although there has been a large body of research on securing general purpose PCs, directly applying such techniques to a CPS can compromise the real-time property of CPSs since the timely execution of tasks in a CPS typically relies on real-time scheduling. Recognizing this property, previous works have proposed approaches to add a security constraint to the real-time properties to cope with the information leakage problem that can arise between real-time tasks with different security levels. However, conventional works have mainly focused on non-preemptive scheduling and have suggested a very naive approach for preemptive scheduling, which shows limited analytical capability. In this paper, we present a new preemptive fixed-priority scheduling algorithm incorporating a security constraint, called lowest security-level first (LSF) and its strong schedulability analysis to reduce the potential of information leakage. Our simulation results show that LSF schedulability analysis outperforms state-of-the-art FP analysis when the security constraint has reasonable timing penalties.\",\n \"Title: Efficient Privacy-Preserving Matrix Factorization via Fully Homomorphic Encryption: Extended Abstract.\\nYear: 2016\\n\\nAbstract: Recommendation systems become popular in our daily life. It is well known that the more the release of users' personal data, the better the quality of recommendation. However, such services raise serious privacy concerns for users. In this paper, focusing on matrix factorization-based recommendation systems, we propose the first privacy-preserving matrix factorization using fully homomorphic encryption. On inputs of encrypted users' ratings, our protocol performs matrix factorization over the encrypted data and returns encrypted outputs so that the recommendation system knows nothing on rating values and resulting user/item profiles. It provides a way to obfuscate the number and list of items a user rated without harming the accuracy of recommendation, and additionally protects recommender's tuning parameters for business benefit and allows the recommender to optimize the parameters for quality of service. To overcome performance degradation caused by the use of fully homomorphic encryption, we introduce a novel data structure to perform computations over encrypted vectors, which are essential operations for matrix factorization, through secure 2-party computation in part. With the data structure, the proposed protocol requires dozens of times less computation cost over those of previous works. Our experiments on a personal computer with 3.4 GHz 6-cores 64 GB RAM show that the proposed protocol runs in 1.5 minutes per iteration. It is more efficient than Nikolaenko et al.'s work proposed in CCS 2013, in which it took about 170 minutes on two servers with 1.9 GHz 16-cores 128 GB RAM.\",\n \"Title: Protecting Secret Keys In Networked Devices With Table Encoding Against Power Analysis Attacks\\nYear: 2016\\n\\nAbstract: Nowadays, secret keys of networked devices are profoundly attacked by power analysis attacks, caused by the dramatic evolution of statistical analysis with a simple experimental setup. Recently, OpenSSL and CoreBitcoin running on Android and iOS have been broken by power analysis. Moreover, sensors and actuators can also be attacked thereby threatening user's privacy and security. To resolve these challenges, power-analysis-resistant implementations of cryptographic algorithms in networked devices have received a lot of attentions. Masking schemes have been developed to implement secure cryptographic algorithms against side-channel analysis (SCA) attacks. Technically, the first-order masking method is vulnerable to the second order differential power analysis (2ODPA) attacks, but the current solutions against 2ODPA are expensive to be implemented. Moreover, worse performance will be shown if the cryptographic algorithms include boolean and arithmetic operations. In this paper, we propose a new countermeasure scheme to resist SCA attacks. Our scheme randomizes all the intermediate values of block cipher by encoding functions in the algorithm to lookup table and makes it resistant to power analysis attack. We apply our scheme to the block cipher algorithm, HIGHT. Our protected implementation of HIGHT takes only 1.79 times compared to the straightforward algorithm, and it needs 25 kbytes to store lookup tables in memory.\",\n \"Title: On-demand bootstrapping mechanism for isolated cryptographic operations on commodity accelerators.\\nYear: 2016\\n\\nAbstract: General-Purpose computing on a Graphics Processing Unit (GPGPU) involves leveraging commodity GPUs as massively parallel processing units. GPGPU is an emerging computing paradigm for high-performance and data-intensive computations such as cryptographic operations. Although GPGPU is an attractive solution for accelerating modern cryptographic operations, the security challenges that stem from utilizing commodity GPUs remain an unresolved problem. In this paper, we present an On-demand Bootstrapping Mechanism for Isolated cryptographic operations (OBMI). OBMI transforms commodity GPUs into a securely isolated processing core for various cryptographic operations while maintaining cost-effective computations. By leveraging System Management Mode (SMM), a privileged execution mode provided by x86 architectures, OBMI implements a program and a secret key into the GPU such that they are securely isolated during the acceleration of cryptographic operations, even in the presence of compromised kernels. Our approach does not require an additional hardware-abstraction layer such as a hypervisor or micro-kernel, and it does not entail modifying the GPU driver. An evaluation of the proposed OBMI demonstrated that even adversaries with kernel privileges cannot gain access to the secret key, and it also showed that the proposed mechanism incurs negligible performance degradation for both the CPU and GPU.\",\n \"Title: Motion-MiX DHT for Wireless Mobile Networks.\\nYear: 2016\\n\\nAbstract: Last encounter routing (LER) is an excellent routing paradigm that exploits distributed mobility diffusion to achieve both moving object tracking and packet networking services in dynamic mobile networks. From our observations, we discover that LER can be easily extended to a mobile DHT protocol that introduces excellent performance to high-speed mobility environments. This is of particular interest when higher level of mobility and membership dynamics go hand in hand. In our simple but powerful DHT paradigm, a data publish/look-up process consists of a sequence of spatial motion tracking of the rendezvous node that is responsible for the data resource. Thus, we name the protocol MX-DHT (Motion-MiX-DHT). As opposed to existing topology based DHT schemes, MX-DHT does not require additional management of logical or overlying look-up topologies, except for the one-hop encounter records of logical metadata carried by mobile nodes. Therefore, in high-speed mobility and dynamic membership environments, MX-DHT achieves a significant reduction in the communication costs of the publish/look-up and join/leave operations as compared to existing mobile DHT schemes. An extensive set of experiments showed that MX-DHT is a cost-effective solution to providing a content centric networking service in different types of networks with dynamic mobility and membership changes.\",\n \"Title: Designing And Implementing A Diversity Policy For Intrusion-Tolerant Systems\\nYear: 2017\\n\\nAbstract: Research on intrusion-tolerant systems (ITSs) is being conducted to protect critical systems which provide useful information services. To provide services reliably, these critical systems must not have even a single point of failure (SPOF). Therefore, most ITSs employ redundant components to eliminate the SPOF problem and improve system reliability. However, systems that include identical components have common vulnerabilities that can be exploited to attack the servers. Attackers prefer to exploit these common vulnerabilities rather than general vulnerabilities because the former might provide an opportunity to compromise several servers. In this study, we analyze software vulnerability data from the National Vulnerability Database (NVD). Based on the analysis results, we present a scheme that finds software combinations that minimize the risk of common vulnerabilities. We implement this scheme with CSIM20, and simulation results prove that the proposed scheme is appropriate for a recovery-based intrusion tolerant architecture.\",\n \"Title: Duo: Software Defined Intrusion Tolerant System Using Dual Cluster.\\nYear: 2018\\n\\nAbstract: An intrusion tolerant system (ITS) is a network security system that is composed of redundant virtual servers that are online only in a short time window, called exposure time. The servers are periodically recovered to their clean state, and any infected servers are refreshed again, so attackers have insufficient time to succeed in breaking into the servers. However, there is a conflicting interest in determining exposure time, short for security and long for performance. In other words, the short exposure time can increase security but requires more servers to run in order to process requests in a timely manner. In this paper, we propose Duo, an ITS incorporated in SDN, which can reduce exposure time without consuming computing resources. In Duo, there are two types of servers: some servers with long exposure time (White server) and others with short exposure time (Gray server). Then, Duo classifies traffic into benign and suspicious with the help of SDN/NFV technology that also allows dynamically forwarding the classified traffic to White and Gray servers, respectively, based on the classification result. By reducing exposure time of a set of servers, Duo can decrease exposure time on average. We have implemented the prototype of Duo and evaluated its performance in a realistic environment.\",\n \"Title: Captcha Image Generation Systems Using Generative Adversarial Networks\\nYear: 2018\\n\\nAbstract: We propose new CAPTCHA image generation systems by using generative adversarial network (GAN) techniques to strengthen against CAPTCHA solvers. To verify whether a user is human, CAPTCHA images are widely used on the web industry today. We introduce two different systems for generating CAPTCHA images, namely, the distance GAN (D-GAN) and composite GAN (C-GAN). The D-GAN adds distance values to the original CAPTCHA images to generate new ones, and the C-GAN generates a CAPTCHA image by composing multiple source images. To evaluate the performance of the proposed schemes, we used the CAPTCHA breaker software as CAPTCHA solver. Then, we compared the resistance of the original source images and the generated CAPTCHA images against the CAPTCHA solver. The results show that the proposed schemes improve the resistance to the CAPTCHA solver by over 67.1% and 89.8% depending on the system.\",\n \"Title: POSTER: Zero-Day Evasion Attack Analysis on Race between Attack and Defense.\\nYear: 2018\\n\\nAbstract: Deep neural networks (DNNs) exhibit excellent performance in machine learning tasks such as image recognition, pattern recognition, speech recognition, and intrusion detection. However, the usage of adversarial examples, which are intentionally corrupted by noise, can lead to misclassification. As adversarial examples are serious threats to DNNs, both adversarial attacks and methods of defending against adversarial examples have been continuously studied. Zero-day adversarial examples are created with new test data and are unknown to the classifier; hence, they represent a more significant threat to DNNs. To the best of our knowledge, there are no analytical studies in the literature of zero-day adversarial examples with a focus on attack and defense methods through experiments using several scenarios. Therefore, in this study, zero-day adversarial examples are practically analyzed with an emphasis on attack and defense methods through experiments using various scenarios composed of a fixed target model and an adaptive target model. The Carlini method was used for a state-of-the-art attack, while an adversarial training method was used as a typical defense method. We used the MNIST dataset and analyzed success rates of zero-day adversarial examples, average distortions, and recognition of original samples through several scenarios of fixed and adaptive target models. Experimental results demonstrate that changing the parameters of the target model in real time leads to resistance to adversarial examples in both the fixed and adaptive target models.\",\n \"Title: Detecting Online Game Chargeback Fraud Based on Transaction Sequence Modeling Using Recurrent Neural Network.\\nYear: 2017\\n\\nAbstract: We propose an online game money chargeback fraud detection method using operation sequence, gradient of charge/purchase amount, time and country as features of a transaction. We model the sequence of transactions with a recurrent neural network which also combines charge and purchase transaction features in single feature vector. In experiments using real data (a 483,410 transaction log) from a famous online game company in Korea, the proposed method shows a 78% recall rate with a 0.057% false positive rate. This recall rate is 7% better than current methodology utilizing transaction statistics as features.\",\n \"Title: A Simple and Enriched Closed-Form Formula for Cell Residence Time in 5G Heterogeneous Networks.\\nYear: 2018\\n\\nAbstract: Cell residence time is a key system parameter for the performance analysis of cellular mobile systems and their related applications. In order to save cost and time before wide implementation and deployment of a mobile system, various network performance metrics, such as network resource occupancy probabilities, should be calculated through analytical results of the mean cell residence time. However, this is difficult to estimate in 5G heterogeneous networks, due to irregular-shaped multi-tier heterogeneous network topologies introduced by a diverse set of small cells. In order to find an efficient way of calculating the cell residence time for 5G heterogeneous networks, we first examine the well-known conventional formula. Then we propose an enhanced closed-form formula for cell residence time by considering more generalized cellular mobile systems, as well as more generic mobility models. The proposed closed-form formula can be directly applied to various complex and reliable network scenarios. It can be adapted to multi-tier heterogeneous networks with randomly positioned irregular-shaped small cells, even though they overlap with each other. Our simulation results show that the proposed formula provides excellent estimation accuracy for general random mobility models, including the Levy walk, which is known to realistically reflect human mobility.\",\n \"Title: Advanced Ensemble Adversarial Example On Unknown Deep Neural Network Classifiers\\nYear: 2018\\n\\nAbstract: Deep neural networks (DNNs) are widely used in many applications such as image, voice, and pattern recognition. However, it has recently been shown that a DNN can be vulnerable to a small distortion in images that humans cannot distinguish. This type of attack is known as an adversarial example and is a significant threat to deep learning systems. The unknown-target-oriented generalized adversarial example that can deceive most DNN classifiers is even more threatening. We propose a generalized adversarial example attack method that can effectively attack unknown classifiers by using a hierarchical ensemble method. Our proposed scheme creates advanced ensemble adversarial examples to achieve reasonable attack success rates for unknown classifiers. Our experiment results show that the proposed method can achieve attack success rates for an unknown classifier of up to 9.25% and 18.94% higher on MNIST data and 4.1% and 13% higher on CIFAR10 data compared with the previous ensemble method and the conventional baseline method, respectively.\",\n \"Title: Efficient Privacy-Preserving Matrix Factorization for Recommendation via Fully Homomorphic Encryption.\\nYear: 2018\\n\\nAbstract: There are recommendation systems everywhere in our daily life. The collection of personal data of users by a recommender in the system may cause serious privacy issues. In this article, we propose the first privacy-preserving matrix factorization for recommendation using fully homomorphic encryption. Our protocol performs matrix factorization over encrypted users’ rating data and returns encrypted outputs so that the recommendation system learns nothing on rating values and resulting user/item profiles. Furthermore, the protocol provides a privacy-preserving method to optimize the tuning parameters that can be a business benefit for the recommendation service providers. To overcome the performance degradation caused by the use of fully homomorphic encryption, we introduce a novel data structure to perform computations over encrypted vectors, which are essential for matrix factorization, through secure two-party computation in part. Our experiments demonstrate the efficiency of our protocol.\",\n \"Title: Secure mobile device structure for trust IoT.\\nYear: 2018\\n\\nAbstract: In the IoT environment, all devices are connected to each other, and mobile device is considered as key device. But hacking into mobile devices is increasing rapidly with the increase in mobile device users. As the market share of Android OS increases, hacking of mobile devices has focused on Android devices. Although there are many security solutions for mobile devices, they are fragmentary for mobile threats; that is, they are solutions for only several threats rather than comprehensive solutions. There is hence a limit to protecting user’s and company’s data stored or used on mobile devices from various types of hacking. To address this, we propose a mobile device protection technology based on domain isolation. Virtualization technology has emerged to increase CPU utilization in server-class PCs and to run various OSs in one system. As these virtualization technologies become lightweight, they are beginning to be applied to embedded devices. In this paper, we applied this lightweight embedded virtualization technology to mobile devices to divide mobile devices into two areas. Therefore, users can have hidden area from hacker attack in addition to Android OS area which can use same as existing mobile device. There is a hardware-based mobile security solution using an secure element, but this has to be reflected in the manufacturing process of the mobile device. However, since the domain separation technology using the virtualization, proposed in this paper, is a software solution, it has an advantage that it can be applied to a device that is already in use. In addition, to protect the hidden area, application authentication/authorization and user authentication technology were applied. And we use white-box cryptography to get root of trust of the key which is used for secure storage and data encryption/decryption. We believe this is a fundamental solution for protecting the mobile device users from hacking. We implemented and tested various mobile applications operating on a mobile device that incorporates our proposed structure based on domain isolation. There is some performance degradation caused by the domain separation, but it is negligible. According to https://www.wired.com/insights/2012/11/mobile-supercomputers/, the chips for mobile phones have evolved and mobile phones will soon become supercomputers. In this case, the addition of virtualization to the mobile device will have less impact on the computing power of the mobile device, and data protection stored in mobile devices and secure execution environment of security programs will become more important issues. Therefore, our TeeMo structure is a necessary technology to protect mobile device users.\",\n \"Title: A Novel Positive Transfer Learning Approach for Telemonitoring of Parkinson's Disease.\\nYear: 2019\\n\\nAbstract: Telemonitoring is the use of electronic devices to remotely monitor patients. Taking the Parkinson's disease (PD) as an example, the use of at-home testing device (AHTD) enables remote, internet-based measurement of PD vocal symptoms. Translating AHTD measurement into a unified PD rating scale (UPDRS) through predictive analytics enables cost-effective, convenient, and close tracking of PD progres...\",\n \"Title: Fooling A Neural Network In Military Environments: Random Untargeted Adversarial Example\\nYear: 2018\\n\\nAbstract: Deep neural networks (DNNs) show superior performance in machine learning tasks such as image recognition, speech recognition, intrusion detection, and pattern analysis. However, an adversarial example, created by adding a little noise to the original sample, can cause misclassification by the DNN. As adversarial examples are a serious threat to DNNs, there has been much research into the generation of adversarial examples designed for attacking DNNs. The adversarial example attack is divided into two categories: targeted adversarial example and untargeted adversarial example. The targeted adversarial example attack causes machines to misinterpret an object as the attacker's desired class. In contrast, the untargeted adversarial example causes machines to misinterpret an object as an incorrect class. In this paper, we focus on an untargeted adversarial example scenario because it has less distortion from the original sample and a faster learning time than a targeted adversarial example scenario. However, there is a pattern problem in generating untargeted adversarial examples: Because of the similarity between the original class and specific classes, it may be possible for the defending system to determine the original class by analyzing the output classes of the untargeted adversarial examples. To overcome this problem, we propose a new method for generating untargeted adversarial examples, one that uses an arbitrary class in the generation process. For experimental datasets, we used MNIST and CIFAR10, and the Tensorflow library was employed as the machine learning library. Through our experiment, we show that the proposed method can generate random untargeted adversarial examples that do not focus on a specific class for a given original class, while keeping distortion to a minimum (1.99 and 42.32 on MNIST and CIFAR10, respectively) and maintaining a 100% attack success rate.\",\n \"Title: Random Untargeted Adversarial Example on Deep Neural Network.\\nYear: 2018\\n\\nAbstract: Deep neural networks (DNNs) have demonstrated remarkable performance in machine learning areas such as image recognition, speech recognition, intrusion detection, and pattern analysis. However, it has been revealed that DNNs have weaknesses in the face of adversarial examples, which are created by adding a little noise to an original sample to cause misclassification by the DNN. Such adversarial examples can lead to fatal accidents in applications such as autonomous vehicles and disease diagnostics. Thus, the generation of adversarial examples has attracted extensive research attention recently. An adversarial example is categorized as targeted or untargeted. In this paper, we focus on the untargeted adversarial example scenario because it has a faster learning time and less distortion compared with the targeted adversarial example. However, there is a pattern vulnerability with untargeted adversarial examples: Because of the similarity between the original class and certain specific classes, it may be possible for the defending system to determine the original class by analyzing the output classes of the untargeted adversarial examples. To overcome this problem, we propose a new method for generating untargeted adversarial examples, one that uses an arbitrary class in the generation process. Moreover, we show that our proposed scheme can be applied to steganography. Through experiments, we show that our proposed scheme can achieve a 100% attack success rate with minimum distortion (1.99 and 42.32 using the MNIST and CIFAR10 datasets, respectively) and without the pattern vulnerability. Using a steganography test, we show that our proposed scheme can be used to fool humans, as demonstrated by the probability of their detecting hidden classes being equal to that of random selection.\",\n \"Title: A new proxy signature scheme providing self-delegation\\nYear: 2006\\n\\nAbstract: We improve Malkin et al's construction (Eurocrypt'04) of the proxy signature scheme in the random oracle model. Unlike Malkin et al's scheme, the proposed scheme does not assume the existence of the trusted secure device, which has a global secret key that all users' private keys can be recovered with. This makes the proposed scheme more scalable and efficient because users need not access and cooperate with the secure device to generate their public/private key pairs. We show that the proposed scheme is provably secure based on the modified strong RSA assumption that was made by the Itkis et al (CRYPTO'01).\",\n \"Title: A Necessary and Sufficient Condition for Proper Routing in Omega-Omega Network\\nYear: 1996\\n\\nAbstract: Recently, a routing algorithm, called inside-out algorithm, was proposed for routing an arbitrary permutation in the omega-based 2log2\\nN stage networks including the omega-omega network. This paper discusses the problems of the inside-out routing algorithm and shows that the suggested condition for proper routing in the omega-omega network is insufficient. This paper suggests an extended necessary and sufficient condition for proper routing in the omega-omega network.\",\n \"Title: An Efficient Audit Scheme for Cloud Storage.\\nYear: 2014\\n\\nAbstract: As Internet technology has grown rapidly, a cloud storage service becomes more very popular and widespread recently. Since there is no guarantee for integrity of data in the cloud storage, audit mechanisms for outsourced data were proposed. However, the conventional audit process requires a lot of computation overheads due to the complex operations used in tag generation and verification. In this paper, we propose a light-weight audit scheme for the cloud storage with efficient cryptographic operations. While keeping high efficiency, our proposed scheme guarantees the consistency of data in the cloud storage. To evaluate the performance of our scheme, we measure the computation costs and the time required for audit process. As a result, our proposed scheme can provide correctness of auditing for outsourced data by an efficient manner.\",\n \"Title: An efficient multi-path pipeline transmission for a bulk data transfer in IEEE 802.15.4 multi-hop networks.\\nYear: 2019\\n\\nAbstract: A pipeline transmission is the state-of-the-art approach to transmit large amounts of data over IEEE 802.15.4 multi-hop networks, but the performance of the pipeline transmission can be degraded in unreliable networks. In this paper, we propose an efficient multi-path pipeline transmission (EMP) to support the large data transfer with low latency and high energy efficiency under various network conditions. The proposed EMP adjusts cycle time of the pipeline transmission, which allows nodes to retransmit dropped data packet efficiently. It helps to improve the transmission probability, so EMP can mitigate additional delay and energy consumption caused by frequent end-to-end retransmissions. In addition, EMP employs a multi-path transmission which distributes the large data transfer over multiple routes. It contributes not only to reduce transmission time but also to balance energy consumption of nodes. In this work, we evaluate the performance of EMP through theoretical and simulation-based analysis and compare the performance with other existing pipelines. The results show that EMP outperforms the existing protocols in terms of transmission time and energy efficiency, and then the improved performance of EMP can be maintained regardless of network environments such as link quality, hop counts, and network density.\",\n \"Title: Efficient Approach For Mitigating Mobile Phishing Attacks\\nYear: 2018\\n\\nAbstract: We propose a method for efficiently detecting phishing attacks in mobile environments. When a user visits a website of a certain URL, the proposed method first compares the URL to a generated whitelist. If the URL is not in the whitelist, it detects if the site is a phishing site based on the results of Google search with a carefully refined URL. In addition, the phishing detection is performed only when the user provides input to the website, thereby reducing the frequency of invoking phishing detection to decrease the amount of power used. We implemented the proposed method and used 8315 phishing sites and the same number of legitimate websites for evaluating the performance of the proposed method. We achieved a phishing detection rate of 99.22% with 81.22% reduction in energy consumption as compared to existing approaches that also use search engine for phishing detection. Moreover, because the proposed method does not employ any other algorithm, software, or comparison group, the proposed method can be easily deployed.\",\n \"Title: Friend-safe evasion attack: An adversarial example that is correctly recognized by a friendly classifier.\\nYear: 2018\\n\\nAbstract: Deep neural networks (DNNs) have been applied in several useful services, such as image recognition, intrusion detection, and pattern analysis of machine learning tasks. Recently proposed adversarial examples-slightly modified data that lead to incorrect classification-are a severe threat to the security of DNNs. In some situations, however, an adversarial example might be useful, such as when deceiving an enemy classifier on the battlefield. In such a scenario, it is necessary that a friendly classifier not be deceived. In this paper, we propose a friend-safe adversarial example, meaning that the friendly machine can classify the adversarial example correctly. To produce such examples, a transformation is carried out to minimize the probability of incorrect classification by the friend and that of correct classification by the adversary. We suggest two configurations for the scheme: targeted and untargeted class attacks. We performed experiments with this scheme using the MNIST and CIFAR10 datasets. Our proposed method shows a 100% attack success rate and 100% friend accuracy with only a small distortion: 2.18 and 1.54 for the two respective MNIST configurations, and 49.02 and 27.61 for the two respective CIFAR10 configurations. Additionally, we propose a new covert channel scheme and a mixed battlefield application for consideration in further applications.\",\n \"Title: Multi-Targeted Adversarial Example in Evasion Attack on Deep Neural Network.\\nYear: 2018\\n\\nAbstract: Deep neural networks (DNNs) are widely used for image recognition, speech recognition, pattern analysis, and intrusion detection. Recently, the adversarial example attack, in which the input data are only slightly modified, although not an issue for human interpretation, is a serious threat to a DNN as an attack as it causes the machine to misinterpret the data. The adversarial example attack has been receiving considerable attention owing to its potential threat to machine learning. It is divided into two categories: targeted adversarial example and untargeted adversarial example. The untargeted adversarial example happens when machines misclassify an object into an incorrect class. In contrast, the targeted adversarial example attack causes machines to misinterpret the image as the attacker's desired class. Thus, the latter is a more elaborate and powerful attack than the former. The existing targeted adversarial example is a single targeted attack that allows only one class to be recognized. However, in some cases, a multi-targeted adversarial example can be useful for an attacker to make multiple models recognize a single original image as different classes. For example, an attacker can use a single road sign generated by a multi-targeted adversarial example scheme to make model A recognize it as a stop sign and model B recognize it as a left turn, whereas a human might recognize it as a right turn. Therefore, in this paper, we propose a multi-targeted adversarial example that attacks multiple models within each target class with a single modified image. To produce such examples, we carried out a transformation to maximize the probability of different target classes by multiple models. We used the MNIST datasets and TensorFlow library for our experiment. The experimental results showed that the proposed scheme for generating a multi-targeted adversarial example achieved a 100% attack success rate.\",\n \"Title: Priority Adversarial Example in Evasion Attack on Multiple Deep Neural Networks\\nYear: 2019\\n\\nAbstract: Deep neural networks (DNNs) provide superior per-formance on machine learning tasks such as image recognition, speech recognition, pattern recognition, and intrusion detection. However, an adversarial example created by adding a little noise to the original data can lead to misclassification by the DNN, and the human eye cannot detect the difference from the original data. For example, if an attacker generates a modified left-turn road sign to be incorrectly categorized by a DNN, an autonomous vehicle with the DNN will incorrect classify the modified left-turn road sign as a right-turn sign, whereas a human will correctly classify the modified sign as a left-turn sign. Such an adversarial example is a serious threat to a DNN. Recently, a multi-target adversarial example was introduced that causes misclassification by several models within each target class using a single modified image. However, it has the vulnerability that as the number of target models increases, the overall attack success rate is reduced. Therefore, if there are several models that the attacker wishes to target, the attacker needs to control the attack success rate for each model by considering the attack priority for each model. In this paper, we propose a priority adversarial example that considers the attack priority for each model in cases targeting several models. The proposed method controls the attack success rate for each model by adjusting the weight of the attack function in the generation process, while maintaining minimum distortion. We used Tensorflow, a widely used machine learning library, and MNIST as the dataset. Experimental results show that the proposed method can control the attack success rate for each model by considering the attack priority of each model while maintaining minimum distortion (on average 3.95 and 2.45 in targeted and untargeted attacks, respectively).\",\n \"Title: Deep Residual Inception Encoder-Decoder Network for Medical Imaging Synthesis.\\nYear: 2020\\n\\nAbstract: Image synthesis is a novel solution in precision medicine for scenarios where important medical imaging is not otherwise available. The convolutional neural network (CNN) is an ideal model for this task because of its powerful learning capabilities through the large number of layers and trainable parameters. In this research, we propose a new architecture of residual inception encoder–decoder neural network (RIED-Net) to learn the nonlinear mapping between the input images and targeting output images. To evaluate the validity of the proposed approach, it is compared with two models from the literature: synthetic CT deep convolutional neural network (sCT-DCNN) and shallow CNN, using both an institutional mammogram dataset from Mayo Clinic Arizona and a public neuroimaging dataset from the Alzheimer's Disease Neuroimaging Initiative. Experimental results show that the proposed RIED-Net outperforms the two models on both datasets significantly in terms of structural similarity index, mean absolute percent error, and peak signal-to-noise ratio.\",\n \"Title: Hybrid approach of parallel implementation on CPU–GPU for high-speed ECDSA verification\\nYear: 2019\\n\\nAbstract: Since the advent of deep belief network deep learning technology in 2006, artificial intelligence technology has been utilized in various convergence areas, such as autonomous driving and medical care. Some services requiring fast decision making and action typically work seamlessly with edge computing service model. In autonomous driving of a connected vehicle with vehicle-to-everything (V2X) communication, roadside unit (RSU) acts as an edge computing device and it will make safer service by processing V2X messages fast, sent by vehicles or other devices. IEEE 1609.2 standard provides application message security technology to ensure the security and reliability of vehicle-to-vehicle communication messages. It uses elliptic curve digital signature algorithm (ECDSA) signatures based on the NIST p256 curve for message authenticity. In this paper, we investigate that RSU should be able to verify 3500 ECDSA signatures per second considering the expected maximum number of vehicles on nearby roads (e.g., during rush hour), message transmission rate, and IEEE 802.11p wireless channel capacity. RSU should satisfy this requirement without assistance of hardware-based cryptographic accelerator. For the requirement, we propose a hybrid approach of parallel ECDSA signature verification at high speed by using CPU and GPU, simultaneously. Moreover, we implemented the proposed method in various modern computing environments for RSU and edge computing devices. Through the experiments, we reach the conclusion that GPU can contribute to the required performance of ECDSA signature verification in RSU platform, which could not satisfy the above throughput only with CPU unit. The target platform with Intel Pentium E6500 CPU and GeForce GTX650 GPU can verify 5668 signatures per second with 30% utilization, while CPU in the platform can process only 2640 signatures. Even in a higher-performance edge computing device, we examine experimentally that the performance can be further improved by using the proposed hybrid approach.\",\n \"Title: Restricted Evasion Attack: Generation Of Restricted-Area Adversarial Example\\nYear: 2019\\n\\nAbstract: Deep neural networks (DNNs) show superior performance in image and speech recognition. However, adversarial examples created by adding a little noise to an original sample can lead to misclassification by a DNN. Conventional studies on adversarial examples have focused on ways of causing misclassification by a DNN by modulating the entire image. However, in some cases, a restricted adversarial example may be required in which only certain parts of the image are modified rather than the entire image and that results in misclassification by the DNN. For example, when the placement of a road sign has already been completed, an attack may be required that will change only a specific part of the sign, such as by placing a sticker on it, to cause misidentification of the entire image. As another example, an attack may be required that causes a DNN to misinterpret images according to a minimal modulation of the outside border of the image. In this paper, we propose a new restricted adversarial example that modifies only a restricted area to cause misclassification by a DNN while minimizing distortion from the original sample. It can also select the size of the restricted area. We used the CIFAR10 and ImageNet datasets to evaluate the performance. We measured the attack success rate and distortion of the restricted adversarial example while adjusting the size, shape, and position of the restricted area. The results show that the proposed scheme generates restricted adversarial examples with a 100% attack success rate in a restricted area of the whole image (approximately 14% for CIFAR10 and 1.07% for ImageNet) while minimizing the distortion distance.\",\n \"Title: A Feature Transfer Enabled Multi-Task Deep Learning Model on Medical Imaging.\\nYear: 2019\\n\\nAbstract: •Multi-task deep model for medical imaging-based applications.•First fully automatic system handling detection, segmentation and classification.•Cross-view feature transfer free of negative transferring issue.\",\n \"Title: Selective Untargeted Evasion Attack: An Adversarial Example That Will Not Be Classified as Certain Avoided Classes.\\nYear: 2019\\n\\nAbstract: Deep neural networks (DNNs) have useful applications in machine learning tasks involving recognition and pattern analysis. Despite the favorable applications of DNNs, these systems can be exploited by adversarial examples. An adversarial example, which is created by adding a small amount of noise to an original sample, can cause misclassification by the DNN. Under specific circumstances, it may be necessary to create a selective untargeted adversarial example that will not be classified as certain avoided classes. Such is the case, for example, if a modified tank cover can cause misclassification by a DNN, but the bandit equipped with the DNN must misclassify the modified tank as a class other than certain avoided classes, such as a tank, armored vehicle, or self-propelled gun. That is, selective untargeted adversarial examples are needed that will not be perceived as certain classes, such as tanks, armored vehicles, or self-propelled guns. In this study, we propose a selective untargeted adversarial example that exhibits 100% attack success with minimum distortions. The proposed scheme creates a selective untargeted adversarial example that will not be classified as certain avoided classes while minimizing distortions in the original sample. To generate untargeted adversarial examples, a transformation is performed to minimize the probability of certain avoided classes and distortions in the original sample. As experimental datasets, we used MNIST and CIFAR-10, including the Tensorflow library. The experimental results demonstrate that the proposed scheme creates a selective untargeted adversarial example that exhibits 100% attack success with minimum distortions (1.325 and 34.762 for MNIST and CIFAR-10, respectively).\",\n \"Title: New Complementary Points of Cyber Security Schemes for Critical Digital Assets at Nuclear Power Plants.\\nYear: 2019\\n\\nAbstract: As the importance of cyber security for nuclear power plants has recently increased, related organizations such as manufactures, operators, regulators, and research institutes have taken into consideration the many concerns for effective application of cyber security to critical digital assets at nuclear power plants. It is essential to analyze and research how to effectively apply nuclear cyber security requirements and what complementary factors are present in the overall process for a nuclear security technology system. The complementary points of nuclear cyber security technology could be analyzed through diverse cyber security methods such as cyber security assurance, assessments, and software development life cycle. It is necessary to apply nuclear cyber security technology effectively in accordance with the operation, assurance, and system development process of a nuclear facility. These analysis results could be applied to technical points for the cyber security of critical digital assets at nuclear power plants in the future and to technology that needs to be at the appropriate level of cyber security to ensure technical assurance. This paper is a new approach to compare and analyze various methods used in the cyber security field to discover complementary points for the application of cyber security to critical systems in nuclear power plants. Additionally, this paper also presents effective methods for developing and applying, evaluating, and regulating cyber security in nuclear digital critical systems using the complementary points of various cyber security methods identified in the analysis.\",\n \"Title: ZeroKernel: Secure Context-Isolated Execution on Commodity GPUs\\nYear: 2021\\n\\nAbstract: In the last decade, the dedicated graphics processing unit (GPU) has emerged as an architecture for high-performance computing workloads. Recently, researchers have also focused on the isolation property of a dedicated GPU and suggested GPU-based secure computing environments with several promising applications. However, despite the security analysis conducted by the prior studies, it has been unclear whether a dedicated GPU can be leveraged as a secure processor in the presence of a kernel-privileged attacker. In this paper, we first demonstrate the security of dedicated GPUs through comprehensive studies on context information for GPU execution. The paper shows that a kernel-privileged attacker can manipulate the GPU contexts to redirect memory accesses or execute arbitrary GPU codes on the running GPU kernel. Based on the security analysis, this paper proposes a new on-chip execution model for the dedicated GPU and a novel defense mechanism supporting the security of the on-chip execution. With comprehensive evaluation, the paper assures that the proposed solutions effectively isolate sensitive data in on-chip storages and defend against known attack vectors from a privileged attacker, supporting that the commodity GPUs can be leveraged as a secure processor.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":161,"string":"161"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13552,"string":"13,552"}}},{"rowIdx":13553,"cells":{"question_id":{"kind":"string","value":"Rachid El Azouzi"},"question":{"kind":"string","value":"Generate an abstract for the title \"Improving connectivity in vehicular ad hoc networks: An analytical study\" using the following items: 1. Vehicular ad hoc networks \n2. Mobile base-stations \n3. Connectivity \n4. Queuing model \n5. Optimisation"},"answer":{"kind":"string","value":"Connectivity in vehicular ad hoc networks may degrade dramatically in sparse traffic and also high speed highways. In this paper we study a way to improve the connectivity by adding some extra nodes with higher transmission range which we call mobile base-stations. These nodes can also offer commercial services (e.g. advertisement, video, audio, etc.) to the vehicles on roads. Besides, the financial profit of those services also depends on a satisfactory connectivity. We use an equivalent M/G/∞ queuing model in order to investigate the connectivity. We further take into account the case when some vehicles do not participate in the network either because they are not equipped with wireless transceivers or some other reasons like security concerns. Moreover, connectivity in presence of fixed Road Side Units (RSUs) is also studied. Our proposed analytical model can be used to find the optimum values of the number of base-stations as well as their transmission range in order to achieve a desired degree of connectivity."},"correct_answer":{"kind":"string","value":"Connectivity in vehicular ad hoc networks may degrade dramatically in sparse traffic and also high speed highways. In this paper we study a way to improve the connectivity by adding some extra nodes with higher transmission range which we call mobile base-stations. These nodes can also offer commercial services (e.g. advertisement, video, audio, etc.) to the vehicles on roads. Besides, the financial profit of those services also depends on a satisfactory connectivity. We use an equivalent M/G/∞ queuing model in order to investigate the connectivity. We further take into account the case when some vehicles do not participate in the network either because they are not equipped with wireless transceivers or some other reasons like security concerns. Moreover, connectivity in presence of fixed Road Side Units (RSUs) is also studied. Our proposed analytical model can be used to find the optimum values of the number of base-stations as well as their transmission range in order to achieve a desired degree of connectivity."},"documents":{"kind":"list like","value":["Title: Information Concealing Games\nYear: 2010\n\nAbstract: A system with an n-dimensional state vector and a controller and an actor is considered. The controller has complete information about the system state, and reveals a certain “minimum” amount of information to the actor. The actor takes certain actions based on the information the controller reveals, and the actions fetch certain utilities for each entity. Both the controller and actor seek to maximize their individual utilities by respectively selecting the information to reveal and the actions to adopt. This decision problem forms the basis of several technical and social systems, and can be formulated as a signaling game. It is shown that the Perfect Bayesian Equilibrium of this game has several counterintuitive properties and can be obtained as a saddle point of a different two person zero sum game. The computation time for saddle points using standard linear programs however turns out to be superexponential in n, which leads to computational intractability even for moderate n. Algorithms for computing saddle point policies using a computation time that is exponential in n are presented. Finally, simple linear time computable policies that approximate the saddle-point policies within guaranteeable approximation ratios are obtained.","Title: Constrained cost-coupled stochastic games with independent state processes\nYear: 2007\n\nAbstract: We study non-cooperative constrained stochastic games in which each player controls its own Markov chain based on its own state and actions. Interactions between players occur through their costs and constraints which depend on the state and actions of all players. We provide an example from wireless communications.","Title: Emergence of equilibria from individual strategies in online content diffusion.\nYear: 2013\n\nAbstract: Social scientists have observed that human behavior in society can often be modeled as corresponding to a threshold type policy. A new behavior would propagate by a procedure in which an individual adopts the new behavior if the fraction of his neighbors or friends having adopted the new behavior exceeds some threshold. In this paper we study the question of whether the emergence of threshold policies may be modeled as a result of some rational process which would describe the behavior of non-cooperative rational members of some social network. We focus on situations in which individuals take the decision whether to access or not some content, based on the number of views that the content has. Our analysis aims at understanding not only the behavior of individuals, but also the way in which information about the quality of a given content can be deduced from view counts when only part of the viewers that access the content are informed about its quality. In this paper we present a game formulation for the behavior of individuals using a meanfield model: the number of individuals is approximated by a continuum of atomless players and for which the Wardrop equilibrium is the solution concept. We derive conditions on the problem's parameters that result indeed in the emergence of threshold equilibria policies. But we also identify some parameters in which other structures are obtained for the equilibrium behavior of individuals.","Title: Mean field asymptotics of Markov Decision Evolutionary Games and teams.\nYear: 2011\n\nAbstract: We introduce Markov Decision Evolutionary Games with N players, in which each individual in a large population interacts with other randomly selected players. The states and actions of each player in an interaction together determine the instantaneous payoff for all involved players. They also determine the transition probabilities to move to the next state. Each individual wishes to maximize the total expected discounted payoff over an infinite horizon. We provide a rigorous derivation of the asymptotic behavior of this system as the size of the population grows to infinity. We show that under any Markov strategy, the random process consisting of one specific player and the remaining population converges weakly to a jump process driven by the solution of a system of differential equations. We characterize the solutions to the team and to the game problems at the limit of infinite population and use these to construct almost optimal strategies for the case of a finite, but large, number of players. We show that the large population asymptotic of the microscopic model is equivalent to a (macroscopic) Markov decision evolutionary game in which a local interaction is described by a single player against a population profile. We illustrate our model to derive the equations for a dynamic evolutionary Hawk and Dove game with energy level.","Title: Online Algorithms for Adaptive Optimization in Heterogeneous Delay Tolerant Networks.\nYear: 2013\n\nAbstract: We study in this paper heterogeneous delay tolerant networks formed by different classes of nodes. We assume that two hop is the forwarding strategy to deliver messages from a source node to destination node and the goal is to optimize the probability of delivering a message in the presence of different classes of mobiles, while satisfying a given energy budget. Using our model, in particular, we transform the joint energy constraint into separate constraints, one per class. This allows us to characterize the optimal strategies and provides us with a suitable framework for the design of multi-dimensional stochastic control algorithms that achieve optimal performance at runtime in spite of the lack of full information on the network state and in presence of different classes of mobiles. A thorough analysis of the convergence properties and stability of our algorithms is presented.","Title: Dynamic spectrum allocation based on cognitive radio for QoS support\nYear: 2010\n\nAbstract: Cognitive radio is an emerging technology in wireless access, aimed at vastly improving the way radio spectrum is utilized. In this paper, we investigate the performance improvement gained by applying cognitive radio to a multiple Wireless Service Providers (WSPs). We consider several WSPs and two types of flows: classical (CL) and cognitive radio (CR) flows. We propose two admission control schemes for CR flows and we show the potential improvement that could be gained by utilizing cognitive radios in the presence of classical flows. We develop several performance metrics for CL and CR flows.","Title: Loss strategies for competing AIMD flows\nYear: 2006\n\nAbstract: We study in this paper two competing AIMD flows that share a common bottleneck link. When congestion occurs, one (or both) flows will suffer a loss that will cause its throughput to decrease by a multiplicative factor. The identity of the flow that will suffer a loss is determined by a randomized \"loss strategy\" that may depend on the throughputs of the flows at the congestion instant. We analyze several loss strategies: the one in which the identity of the flow experiencing the loss is independent of the current throughput and the one in which the flow with the largest throughput is to suffer the loss; this is compared with the strategy that assigns loss probabilities proportionally to the throughputs (thus a flow with a larger throughput has a larger loss probability). After deriving some results for the general asymmetric case, we focus in particular on the symmetric case and study the influence of the strategy on the average throughput and average utilization of the link. As the intuition says, a strategy that assigns a loss to a flow with a higher throughput is expected to give worse performance since the total instantaneous throughput after a loss is expected to be lower with such a strategy. Surprisingly, we show that this is not the case. We show that the average throughput and average link utilizations are invariant: they are the same under any possible strategy; the link utilization is 6/7 of the link capacity. We show, in contrast, that the second moment of the throughput does depend on the strategy.","Title: Modeling slotted aloha as a stochastic game with random discrete power selection algorithms\nYear: 2009\n\nAbstract: We consider the uplink case of a cellular system where m bufferless mobiles transmit over a common channel to a base station, using the slotted aloha medium access protocol. We study the performance of this system under several power differentiation schemes. Indeed, we consider a random set of selectable transmission powers and further study the impact of priorities given either to new arrival packets or to the backlogged ones. Later, we address a general capture model where a mobile transmits successfully a packet if its instantaneous SINR (signal to interferences plus noise ratio) is lager than some fixed threshold. Under this capture model, we analyze both the cooperative team in which a common goal is jointly optimized as well as the noncooperative game problem where mobiles reach to optimize their own objectives. Furthermore, we derive the throughput and the expected delay and use them as the objectives to optimize and provide a stability analysis as alternative study. Exhaustive performance evaluations were carried out, we show that schemes with power differentiation improve significantly the individual as well as global performances, and could eliminate in some cases the bi-stable nature of slotted aloha.","Title: Opportunistic Scheduling in Cellular Systems in the Presence of Noncooperative Mobiles\nYear: 2012\n\nAbstract: A central scheduling problem in wireless communications is that of allocating resources to one of many mobile stations that have a common radio channel. Much attention has been given to the design of efficient and fair scheduling schemes that are centrally controlled by a base station (BS) whose decisions depend on the channel conditions reported by each mobile. The BS is the only entity taking decisions in this framework. The decisions are based on the reports of mobiles on their radio channel conditions. In this paper, we study the scheduling problem from a game-theoretic perspective in which some of the mobiles may be noncooperative or strategic, and may not necessarily report their true channel conditions. We model this situation as a signaling game and study its equilibria. We demonstrate that the only Perfect Bayesian Equilibria (PBE) of the signaling game are of the babbling type: the noncooperative mobiles send signals independent of their channel states, the BS simply ignores them, and allocates channels based only on the prior information on the channel statistics. We then propose various approaches to enforce truthful signaling of the radio channel conditions: a pricing approach, an approach based on some knowledge of the mobiles' policies, and an approach that replaces this knowledge by a stochastic approximations approach that combines estimation and control. We further identify other equilibria that involve non-truthful signaling.","Title: Non-cooperative routing in loss networks\nYear: 2002\n\nAbstract: The paper studies routing in loss networks in the framework of a non-cooperative game with selfish users. Two solution concepts are considered: the Nash equilibrium, corresponding to the case of a finite number of agents (such as service providers) that take routing decisions, and the Wardrop equilibrium, in which routing decisions are taken by a very large number of individual users. We show that these equilibria do not fall into the standard frameworks of non-cooperative routing games. As a result, we show that uniqueness of equilibria or even of utilizations at equilibria may fail even in the case of simple topology of parallel links. However, we show that some of the problems disappear in the case in which the bandwidth required by all connections is the same. For the special case of a parallel link topology, we obtain some surprisingly simple way of solving the equilibrium for both cases of Wardrop as well as Nash equilibrium.","Title: Stackelberg approach for pricing differentiated services\nYear: 2007\n\nAbstract: We consider in this paper a set of connections sharing a common bottleneck link. We assume first that we have data transfer using TCP congestion control protocol and real-time traffic using some TCP-friendly transport protocol that satisfies the same square-root formula for throughput. The performance measures are determined according to the operational parameters of a RED buffer management. Those parameters are assumed to be able to give differentiated services to the applications according to their choice of service class. In terms of loss probabilities and of throughputs, we consider a best effort type of service differentiation where the QoS of connections is not guaranteed, but by choosing a better (more expensive) service class, the QoS parameters of a session can be improved (as long as that session is the only one to change its service class). We assume however, that the system is dimensioned so as to satisfy some average delay requirement. The choice of a service class of an application will depend both on the utility as well as on the cost it has to pay. We first study the performance of each traffic source as a function of the connections' parameters and the pricing policy of the network. We then study the Stackelberg equilibrium, i.e. the service provider's problem of how to choose the pricing so as to maximize its utility, taking into account the reaction of the users.","Title: Hierarchy sustains partial cooperation and induces a Braess-like paradox in slotted aloha-based networks\nYear: 2012\n\nAbstract: This paper studies a hierarchical distributed choice of retransmission probabilities in slotted aloha. In particular, we consider a wireless system composed of one central receiver and several selfish mobile users communicating via the slotted aloha protocol. The set of mobile users is split into two classes: leaders and followers. We then study the induced non-cooperative hierarchical game based on the Stackelberg equilibrium concept. Using a 4D Markovian model, we compute the steady state of the system and derive the average throughput and the expected delay as well. We start by discussing the protocol design and propose a controlled slotted aloha using a virtual controller. The virtual controller can sustain partial cooperation among concurrent mobile users when accessing the channel by making the channel lossy. This leads us to identify a Braess-like paradox in which reducing capacity to the system may improve the performance of all mobile users. We then investigate the impact of hierarchy among mobile users in such a random access protocol and discuss how to distribute leader/follower roles. We show that the global performance of the system is improved compared to standard slotted aloha system. However, slight performances slow-down may be observed for the followers group when the total number of mobile users is relatively small.","Title: Evolutionary power control games in wireless networks\nYear: 2008\n\nAbstract: In this paper, we apply evolutionary games to non-cooperative power control in wireless networks. Specifically, we focus our study in a power control in W-CDMA and WIMAX wireless systems. We study competitive power control within a large population of mobiles that interfere with each other through many local interactions. Each local interaction involves a random number of mobiles. An utility function is introduced as the difference between a utility function based on SIR of the mobile and pricing. The games are not necessarily reciprocal as the set of mobiles causing interference to a given mobile may differ from the set of those suffering from its interference. We show how the evolution dynamics and the equilibrium behavior (called Evolutionary Stable Strategy - ESS) are influenced by the characteristics of the wireless channel and pricing characteristics.","Title: From Altruism to Non-Cooperation in Routing Games\nYear: 2008\n\nAbstract: Abstract—The paper studies the routing in the network shared by several users. Each user seeks to optimize either its own,performance,or some,combination,between,its own performance and that of other users, by controlling the routing,of its given flow demand.,We parameterize the degree of cooperation,which,allows to cover the fully non-cooperative behavior, the fully cooperative behavior, and even more, the fully altruistic behavior, all these as special cases of the parameter’s,choice. A large part of the work,consists in exploring,the impact,of the degree of cooperation,on the equilibrium. Our first finding is to identify multiple Nash equilibria with cooperative behavior that do not occur in the non-cooperative,case under,the same conditions (cost, demand and topology). We then identify Braess like paradox,(in which,adding,capacity or adding a link to a network,results in worse performance,to all users) and study the impact of the degree of cooperation on it. We identify situations where,low cooperation,results in poorer performance,at equilibrium. We then pursue,the exploration,and carry it on to the setting of Stackelberg equilibrium. We finally obtain some theoretical results that show,that for low degree,of cooperation,the equilibrium is unique, confirming the results of our numerical study.","Title: Study Of Connectivity In Vehicular Ad Hoc Networks\nYear: 2007\n\nAbstract: We investigate connectivity in ad hoc network formed between vehicles that move in the same direction on a typical highway. We use the common model in vehicular traffic theory in which a fixed point on the highway sees cars passing it separated. by times with exponentially distributed duration. We obtain the distribution of the distances between cars, which allows us to use techniques from queuing theory for studying connectivity. We obtain explicit expressions for the expected connectivity distance as well as the probability distribution and expectation of the number of cars in a platoon. The analytical model we present is able to describe the effects of various system's parameters, including road traffic parameters (i.e. speed distribution and traffic flow) and transmission range of vehicles, on the connectivity. We further provide bounds obtained using stochastic ordering techniques. Our approach is based on the work of Morandi and Altman [13] that transformed the problem of connectivity distance distribution into that of the distribution of the busy period of an equivalent infinite server queue. We use our analytical results along with publicly available road traffic statistical data to understand connectivity in VANETs.","Title: Qos Differentiation For Initial And Bandwidth Request Ranging In Ieee802.16\nYear: 2008\n\nAbstract: In this paper, we propose and analyse a new scheme to handle ranging requests in IEEE802.16e networks based on differentiated backoff parameters and a novel hierarchical code partitioning principle. Our analysis is of the form of a fixed-point equation which characterizes the system operating points and which allows us to compute the attempt rate of each QoS class as well as the distribution of the number of ranging bandwidth requests received successfully by the base station as well as their average delay. Both analysis and simulation confirm that our scheme yields high throughput and low collision probability for real-time connections without altering the collision probability for non-real-time connections.","Title: Avoiding paradoxes in multi-agent competitive routing\nYear: 2003\n\nAbstract: Strange behavior may occur in networks due to the non-cooperative nature of decision making, when the latter are taken by individual agents. In particular, the well known Braess paradox illustrates that when upgrading a network by adding a link, the resulting equilibrium may exhibit larger delays for all users. We present here some guidelines to avoid the Braess paradox when upgrading a network. We furthermore present conditions for the delays to be monotone increasing in the total demand.","Title: Analysis of Buffer Starvation With Application to Objective QoE Optimization of Streaming Services\nYear: 2014\n\nAbstract: Our purpose in this paper is to characterize buffer starvations for streaming services. The buffer is modeled as a FIFO queue with exponential service time and Poisson arrivals. When the buffer is empty, the service restarts after a certain amount of packets are prefetched. With this goal, we propose two approaches to obtain exact distribution of the number of buffer starvations, one of which is based on Ballot theorem, and the other uses recursive equations. The Ballot theorem approach gives an explicit result. We extend this approach to the scenario with a constant playback rate using Tàkacs Ballot theorem. The recursive approach, though not offering an explicit result, allows us to obtain the distribution of starvations with non-independent and identically distributed (i.i.d.) arrival process in which an ON/OFF bursty arrival process is considered. We further compute the starvation probability as a function of the amount of prefetched packets for a large number of files via a fluid analysis. Among many potential applications of starvation analysis, we show how to apply it to optimize objective quality of experience (QoE) of media streaming, by exploiting the tradeoff between startup/rebuffering delay and starvations.","Title: Dynamic spectrum allocation with admission control based on cognitive radio for QoS support in multiple wireless network.\nYear: 2012\n\nAbstract: Cognitive radio is an emerging technology in wireless access, aimed at vastly improving the way radio spectrum is utilized. In this article, we investigate the performance improvement gained by applying cognitive radio to a multiple wireless service providers (WSPs). We consider several WSPs and two types of users: primary (licensed) and secondary (unlicensed) users. Two different schemes are proposed for unlicensed users to manage their handoff. These two schemes are different in the sense that one allows to each WSP to give more priority to his own unlicensed users by dropping low priority users, while the other does not allow that. The system is modeled by a Markov process, with continuous time and finite state space. Due to the very large number of states, we propose a robust method to approximate the stationary probability distribution vector of Markov process. Based on this approximation, we develop several performance metrics, blocking and dropping probabilities for both kinds of unlicensed users. Numerical results show that this approximation is very close to the exact solution. Finally, we show that spectrum utilization of cognitive users increases with the proposed schemes especially in medium and high traffic.","Title: Asymptotic delay analysis and timeout-based admission control for ad hoc wireless networks with asymmetric users\nYear: 2010\n\nAbstract: In this paper, we present an analytical model for an approximate calculation of the end-to-end delay performance in multi-hop wireless ad hoc networks. In contrast to literature that largely focuses on average delay, our paper focuses on the distribution of end-to-end delay. In this paper, we assume that each source injects packets in the network, which traverse intermediate nodes until they reach the destination. Firstly, we employ discrete-time queueing theory to derive the expressions for the queue length and the delay in terms of probability generating functions. Secondly, in order to improve the control routing and transmission scheduling, we adopt a cross-layer design that allows information sharing across different layers for efficient utilization of network resources, and meeting the end-to-end performance requirements of demanding applications. Thirdly, we propose a packet admission control scheme based on delay timeout mechanism. This guarantees quality of service for multimedia applications such as voice and video streaming. Finally, we conduct extensive simulations in order to verify our analytical results.","Title: Battery State-Dependent Access Control in Solar-Powered Broadband Wireless Networks\nYear: 2008\n\nAbstract: This paper studies both power control and multiple access control in solar-powered broadband wireless networks. We assume that the mobiles use power storage element, such as rechargeable Solar- powered batteries, in order to have energy available for later use. By modeling the energy-level of Solar-powered batteries as a stochastic process, we study noncooperative interactions within large population of mobiles that interfere with each other through many local interactions. Each local interaction involves a random number of mobiles. The actions taken by a mobile determine not only the immediate payoff but also the state transition probabilities of its battery. We define and characterize the evolutionary Stable Strategies (ESS) of the stochastic evolutionary game.","Title: Pricing differentiated services: a game-theoretic approach\nYear: 2006\n\nAbstract: The goal of this paper is to study pricing of differentiated services and its impact on the choice of service priority at equilibrium. We consider both TCP connections as well as noncontrolled (real-time) connections. The performance measures (such as throughput and loss rates) are determined according to the operational parameters of a RED (Random Early Discard) buffer management. The latter is assumed to be able to give differentiated services to the applications according to their choice of service class. We consider a service differentiation for both TCP as well as real-time traffic where the quality of service (QoS) of connections is not guaranteed, but by choosing a better (more expensive) service class, the QoS parameters of a session can improve (as long as the service class of other sessions are fixed). The choice of a service class of an application will depend both on the utility as well as on the cost it has to pay. We first study the performance of the system as a function of the connections' parameters and their choice of service classes. We then study the decision problem of how to choose the service classes. We model the problem as a noncooperative game. We establish conditions for an equilibrium to exist and to be uniquely defined. We further provide conditions for convergence to equilibrium from nonequilibria initial states. We finally study the pricing problem of how to choose prices so that the resulting equilibrium would maximize the network benefit.","Title: Modelling View-count Dynamics in YouTube.\nYear: 2014\n\nAbstract: The goal of this paper is to study the behaviour of view-count in YouTube. We first propose several bio-inspired models for the evolution of the view-count of YouTube videos. We show, using a large set of empirical data, that the view-count for 90% of videos in YouTube can indeed be associated to at least one of these models, with a Mean Error which does not exceed 5%. We derive automatic ways of classifying the view-count curve into one of these models and of extracting the most suitable parameters of the model. We study empirically the impact of videos' popularity and category on the evolution of its view-count. We finally use the above classification along with the automatic parameters extraction in order to predict the evolution of videos' view-count.","Title: La théorie des jeux non-coopératifs appliquée aux réseaux de télécommunication\nYear: 2007\n\nAbstract: In this paper, we present various tools originating from non-cooperative Game Theory, which allow us to study competition situations in telecommunication networks. We describe the mathematical tools while providing examples from a variety of areas in telecommunication networks.","Title: Stability-Throughput tradeoff and routing in multi-hop wireless ad-hoc networks\nYear: 2006\n\nAbstract: We study the throughput of multi-hop routes and stability of forwarding queues in a wireless Ad-Hoc network with random access channel. We focuse on wireless with stationary nodes, such as community wireless networks. Our main result is charactrerization of stability condition and the end-to-end throughput using the balance. We also investigate the impact of routing on end-to-end throughput and stability of intermediate nodes. We find that i) as long as the intermediate queues in the network are stable, the end-to-end throughput of a connection does not depend on the load on the intermediate nodes, ii) we showed that if the weight of a link originating from a node is set to the number of neighbors of this node, then shortest path routing maximizes the minimum probability of end-to-end packet delivery in a network of weighted fair queues with coupled servers. Numerical results are given and support the results of the analysis.","Title: Distributed energy-delay framework for opportunistic spectrum access\nYear: 2011\n\nAbstract: In this paper, we consider the problem of spectrum sharing among one primary and multiple secondary users. Particularly, we study the problem of non-cooperative spectrum sensing and access. Secondary users have to sense the channel before accessing it. There is a tradeoff between large packet delay, partially due to collisions between secondary users, and high energy consumption for sensing. To overcome this problem, we consider an opportunistic spectrum access in cognitive radio taking into account packet delay and energy consumption. Hence, there is an associated spectrum occupancy and packet delay (the time elapsed since the reception) state for each secondary user. Due to partial sensing, a Partial Observable Markov Decision Process (POMDP) is used to determine an equilibrium policy. The equilibrium is obtained by using a linear program. We identify a paradox in this cognitive radio context: when the channel is more occupied by the primary user, this may improve the spectrum utilization by secondary users. We also investigate the impact of transmission cost on the equilibrium and show numerically how this cost can be used to enhance the utilization of the spectrum.","Title: A Queuing Analysis of Packet Dropping over a Wireless Link with Retransmissions\nYear: 2003\n\nAbstract: One of the most important performance measure for real time applications is the packet loss probability. In a wireless environment losses are first due to noisy channels (fading, shadowing etc). In order to improve the radio link, one often retransmits packets that have not been well received (using the Automatic Retransmission reQuest-ARQ). This however may lead to queuing phenomena due to retransmissions and to losses of packets due to buffer overflow. We present a queuing analysis in order to compute the combined effect of the noisy channel, the retransmissions and the buffer overflow. We use advanced spectral methods of quasi-birth and death process in order to analyze the system for general multistate Markov channels and provide an approximation approach based on the theory of singular perturbation.","Title: The evolution of transport protocols: An evolutionary game perspective\nYear: 2009\n\nAbstract: Today's Internet is well adapted to the evolution of protocols at various network layers. Much of the intelligence of congestion control is delegated to the end users and they have a large amount of freedom in the choice of the protocols they use. In the absence of a centralized policy for a global deployment of a unique protocol to perform a given task, the Internet experiences a competitive evolution between various versions of protocols. The evolution manifests itself through the upgrading of existing protocols, abandonment of some protocols and appearance of new ones. We highlight in this paper the modeling capabilities of the evolutionary game paradigm for explaining past evolution and predicting the future one. In particular, using this paradigm we derive conditions under which (i) a successful protocol would dominate and wipe away other protocols, or (ii) various competing protocols could coexist. In the latter case we also predict the share of users that would use each of the protocols. We further use evolutionary games to propose guidelines for upgrading protocols in order to achieve desirable stability behavior of the system.","Title: Dynamic retransmission limit scheme in MAC layer for routing in multihop ad hoc networks\nYear: 2008\n\nAbstract: We consider a wireless ad hoc network with random access channel. We present a model that takes into account topology, routing, random access in MAC layer, and forwarding probability. In this paper, we focus on drawing benefit from the interaction of the MAC (governed by IEEE 802.11 or slotted Aloha) and routing by defining a new cross-layer scheme for routing based on the limit number of retransmission. By adjusting dynamically and judiciously this parameter in a saturated network, we have realized that both stability of forwarding queues and average throughput are significantly improved in linear networks with symmetric traffic: a gain of 100% can be reached. While in asymmetric topology network with asymmetric traffic, we achieve a better average delay (resp., throughput) for each connection without changing the average throughput (resp., delay). In addition, we show the efficiency of our new scheme in case of multimedia applications with delay constraint. A detailed performance study is presented using analytical and simulation evaluation.","Title: Evolutionary forwarding games in delay tolerant networks: Equilibria, mechanism design and stochastic approximation\nYear: 2013\n\nAbstract: In this paper, we apply evolutionary games to non-cooperative forwarding control in Delay Tolerant Networks (DTNs). The main focus is on mechanisms to rule the participation of the relays to the delivery of messages in DTNs. Thus, we express the success probability as a function of the competition that takes place within a large population of mobiles, and we characterize the effect of reward-based mechanisms on the performance of such systems. Devices acting as active relays, in fact, sacrifice part of their batteries in order to support message replication and thus increase the probability to reach the destination. In our scheme, a relay can choose the strategy by which they participate to the message relaying. A mobile that participates receives a unit of reward based on the reward mechanism selected by the network. A utility function is introduced as the difference between the expected reward and the energy cost, i.e., the cost spent by the relay to sustain forwarding operations. We show how the evolution dynamics and the equilibrium behavior (called Evolutionary Stable Strategy - ESS) are influenced by the characteristics of inter contact time, energy expenditure and pricing characteristics. We extend our analysis to mechanisms that the system can introduce in order to have the message delivered to the destination with high probability within a given deadline and under energy constraints which bound the number of released copies per message. Finally, we apply our findings in order to devise decentralized forwarding algorithms that are rooted in the theory of stochastic approximations. Thus, we demonstrate that the ESS can be attained without complete knowledge of the system state and letting the source monitor number of released copies per message only. We provide extensive numerical results to validate the proposed scheme.","Title: Introducing hierarchy in energy-efficient power control games.\nYear: 2008\n\nAbstract: We consider a multiple access channel where the users choose their best power control strategy in order to selfishly maximize their energy-efficiency. To increase the utilities with respect to the classical non-cooperative game, we introduce hierarchy in two ways. On the one hand, assuming single-user decoding at the receiver, we investigate a Stackelberg formulation of the game where one user is the leader. On the other hand, assuming neither leader nor followers among the users, we introduce hierarchy by using successive interference cancellation at the receiver. For both cases, we study the existence and uniqueness of an equilibrium and compare the individual performance obtained in the hierarchical game with that obtained in the non-cooperative game. An exhaustive comparative analysis of the two games is also conducted. In order to optimize the choice of the leader in the Stackelberg formulation (with single user decoding) and that of the decoding order (in the non-cooperative game with successive interference cancelation), we study two measures of global energy-efficiency for the network and discussions are provided for each case.","Title: A hierarchical slotted aloha game.\nYear: 2009\n\nAbstract: We consider in this paper a wireless system composed of one central receiver and several selfish transmitters communicating via the slotted aloha protocol. We study a noncooperative hierarchical system based on the Stackelberg game concept. Each user of our game tends to maximize his own throughput or minimize his expected delay of backlogged packets depending on his transmission probability and transmission probabilities of other users in the network. Using a 4D Markovian model, we compute the steady state of the system and derive the average throughput and expected delay. We then investigate the impact of introducing hierarchy in the random access protocol. Later, exhaustive performance evaluations were carried out, we show that the global performance of the system is improved compared to slotted aloha system without hierarchy. However, a not sharp performances slow-down may be observed for the followers when the number of users is not large, but they turn to outperform Nash equilibrium under average and heavy load.","Title: Asymmetric delay in evolutionary games\nYear: 2007\n\nAbstract: We study the effect of the time delays on the convergence of replicator dynamics to the evolutionary stable strategies in a symmetric evolutionary game in which each pure strategy is associated with its own delay. We compare the stability region to that obtained when all strategies have the same delay. Multiple Access Game and Hawk and Dove Game are studied.","Title: Delay Analysis For Real-Time Streaming Media In Multi-Hop Ad Hoc Networks\nYear: 2008\n\nAbstract: In this paper, we investigate an important issue for real-time multimedia over multi-hop ad-hoc networks on different layers. Such application requires receiver playback buffers to smooth network delay variation and reconstruct the periodic nature of the transmitted packets. Packets arriving after their schedule deadline are considered late and are not played out. This requires that the network is able to offer quality of service appropriate for the delay bounds of the real-time application constrains. Our primary contribution concerns the study of the end to end delay. Based on the latter results, we approximate the loss rate of a real-time traffic which requires a delay bounds constraints. Then, we propose a cross layer scheme from the network layer to the MAC layer to support real-time traffic. In this scheme, we reduce the loss rate by decreasing the packets that arrive after their schedule deadline. Through numerical and simulation results, we demonstrate the utility and efficiency of our approach.","Title: Evolutionary games in wireless networks.\nYear: 2010\n\nAbstract: We consider a noncooperative interaction among a large population of mobiles that interfere with each other through many local interactions. The first objective of this paper is to extend the evolutionary game framework to allow an arbitrary number of mobiles that are involved in a local interaction. We allow for interactions between mobiles that are not necessarily reciprocal. We study 1) multiple-access control in a slotted Aloha-based wireless network and 2) power control in wideband code-division multiple-access wireless networks. We define and characterize the equilibrium (called evolutionarily stable strategy) for these games and study the influence of wireless channels and pricing on the evolution of dynamics and the equilibrium.","Title: QoE analysis of media streaming in wireless data networks\nYear: 2012\n\nAbstract: The purpose of this paper is to model quality of experience (QoE) of media streaming service in a shared fast-fading channel. In this context, the arrival and the service processes of the playout buffer do not have the same job size. We present an analytical framework based on Takács Ballot theorem to compute the probability of buffer starvation and the distribution of playback intervals. We model the arrival processes of Proportional Fair and Round Robin schedulers, and feed them into this framework to study the impact of prefetching on the starvation behavior. Our simulations match the developed model very well if the base station knows the playback rate and the channel gain. Furthermore, we make an important observation that QoE metrics predicted by users are very sensitive to the measurement error of arrival process.","Title: TCP Optimization through FEC, ARQ, and Transmission Power Tradeoffs\nYear: 2004\n\nAbstract: TCP performance degrades when end-to-end connections extend over wireless connections - links which are characterized by high bit error rate and intermittent connectivity. Such link characteristics can significantly degrade TCP performance as the TCP sender assumes wireless losses to be congestion losses resulting in unnecessary congestion control actions. Link errors can be reduced by increasing transmission power, code redundancy (FEC) or number of retransmissions (ARQ). But increasing power costs resources, increasing code redundancy reduces available channel bandwidth and increasing persistency increases end-to-end delay. The paper proposes a TCP optimization through proper tuning of power management, FEC and ARQ in wireless environments (WLAN and WWAN). In particular, we conduct analytical and numerical analysis taking into account the three aforementioned factors, and evaluate TCP (and \"wireless-aware\" TCP) performance under different settings. Our results show that increasing power, redundancy and/or retransmission levels always improves TCP performance by reducing link-layer losses. However, such improvements are often associated with cost and arbitrary improvement cannot be realized without paying a lot in return. It is therefore important to consider some kind of net utility function that should be optimized, thus maximizing throughput at the least possible cost.","Title: Multiple access game in ad-hoc network\nYear: 2007\n\nAbstract: We study a Modified Multiple Access Game (MMAG) in two approach: repeated game approach and evolutionary game approach. We compute Nash equilibria and Evolutionary Stable Strategy (ESS) in the two cases. We study the delay impact on the performance of the evolutionary MMAG describing competition between mobile terminals over the access to a channel. We discuss about the convergence to the ESS in replicator dynamics and imitate better dynamics with delays.","Title: Impact of flow-level dynamics on QoE of video streaming in wireless networks\nYear: 2013\n\nAbstract: The Quality of Experience (QoE) of streaming service is often degraded by frequent playback interruptions. To mitigate the interruptions, the media player prefetches streaming contents before starting playback, at a cost of delay. We study the QoE of streaming from the perspective of flow dynamics. First, a framework is developed for QoE when streaming users join the network randomly and leave after downloading completion. We compute the distribution of prefetching delay using partial differential equations (PDEs), and the probability generating function of playout buffer starvations using ordinary differential equations (ODEs). Second, we extend our framework to characterize the throughput variation caused by opportunistic scheduling at the base station in the presence of fast fading. Our study reveals that the flow dynamics is the fundamental reason of playback starvation. The QoE of streaming service is dominated by the average throughput of opportunistic scheduling, while the variance of throughput has very limited impact on starvation behavior.","Title: A two-stage game theoretic approach for self-organizing networks.\nYear: 2013\n\nAbstract: Growth of network access technologies in the mobile environment has raised several new issues due to the interference between the available access. Thanks to the currently used access methods such as the orthogonal frequency division multiple access in mobile networks and the long term evolution-advanced systems, the intra-cell interferences are avoided and the quality of service has increased. Nevertheless, the diversity and multiplicity of base stations in the network have left behind a major problem of inter-cell interferences. In this paper, we focus on the optimization of the total throughput of cellular networks using fractional frequency reuse and allowing each mobile user to individually choose its serving base station. We derive analytically the utilities related to the network manager and mobile users and develop a Stackelberg game to obtain the equilibrium. We propose a distributed algorithm that allows the base stations, using a light collaboration, to achieve an efficient utilization of the frequencies, with the optic of maximizing the total system utility. This algorithm is based on stochastic gradient descent which requires some information to be exchanged between neighboring base stations. At user association level, we propose an iterative distributed algorithm based on automata learning algorithm. Both algorithms allow the system to converge to the Stackelberg equilibrium. Furthermore, simulation results carried out based on a realistic network setting show promising results in terms of global utility and convergence issues. In this setting, we include scenarios with a varying number of users and address the problem of robustness and scalability of the proposed approach.","Title: An end-to-end QoS framework for IEEE 802.16 and ad-hoc integrated networks\nYear: 2009\n\nAbstract: Current mobile users are often equipped with several network interfaces, which may be of different access technologies. Each access technology has specific characteristics in terms of coverage area and technical characteristics (bandwidth, QoS, etc.) and provides diverse commercial opportunities for the operators. It seems likely that these various technologies have to coexist and, from then, solutions of integration and interoperability will be necessary to deal with the technological diversity. We consider in this paper the WiMAX and ad-hoc integrated networks. A user who needs to establish a high reliable service prefers to use WiMAX network and another who needs to transmit Best Effort traffic with high rate but delay tolerant prefers to connect ad-hoc network that is assumed to have access to Internet. We aim through this combination to locally extend the coverage of WiMAX cell and to study the stability of involved nodes, in particular, gateway nodes. Under stability condition, our main result is characterization of the end-to-end throughput and delay using the rate balance equation and a G/G/1 queueing model. Through numerical results, we demonstrate the utility and efficiency of our approach.","Title: Fractional frequency reuse stackelberg model for self-organizing networks\nYear: 2011\n\nAbstract: Growth of network access technologies in the mobile environment has raised several new issues due to the interference between the available accesses. Thanks to the currently used access methods such as OFDMA in mobile networks, and LTE systems, the intra-cell interferences are avoided and the quality of service has increased. Nevertheless, the diversity and multiplicity of base stations in the network has left behind, a major problem of inter-cell interferences. In this paper, we focus on the optimization of the total throughput of cellular networks using fractional frequency reuse and allowing each mobile user to individually choose its serving base station. We derive analytically the utilities related to the network manager and mobile users and develop a Stackelberg game to obtain the equilibria. We propose a distributed algorithm that allows the base stations, using a light collaboration, to achieve an efficient utilization of the frequencies, with the optic of maximizing the total system utility. This algorithm is based on stochastic gradient descent which requires some information to be exchanged between neighboring base stations. At user association level, we propose an iterative distributed algorithm based on automata learning algorithm. Both algorithms allow the system to converge to the Stackelberg equilibrium. Furthermore, simulation results carried out based on a realistic network setting show promising results in terms of global utility and convergence issues.","Title: Loss Strategies for Competing TCP/IP Connections\nYear: 2004\n\nAbstract: We study in this paper two competing TCP connections that share a common bottleneck link. When congestion occurs, one (or both) connections will suffer a loss that will cause its throughput to decrease by a multiplicative factor. The identity of the connection that will suffer a loss is determined by a randomized \"loss strategy\" that may depend on the throughputs of the connections at the congestion instant. We analyze several such loss strategies. After deriving some results for the general asymmetric case, we focus in particular on the symmetric case and study the influence of the strategy on the average throughput and average utilization of the link. As the intuition says, a strategy that assigns a loss to a connection with a higher throughput is expected to give worse performance since the total instantaneous throughput after a loss is expected to be lower with such a strategy. We show that, surprisingly, the average throughput and average link utilizations are invariant: they are the same under any possible strategy; the link utilization is 6/7 of the link capacity. We show, in contrast, that the second moment of the throughput does depend on the strategy.","Title: Slotted Aloha as a game with partial information\nYear: 2004\n\nAbstract: This paper studies distributed choice of retransmission probabilities in slotted ALOHA. Both the cooperative team problem as well as the noncooperative game problem are considered. Unlike some previous work, we assume that mobiles do not know the number of backlogged packets at other nodes. A Markov chain analysis is used to obtain optimal and equilibrium retransmission probabilities and throughput. We then investigate the impact of adding retransmission costs (which may represent the disutility for power consumption) on the equilibrium and show how this pricing can be used to make the equilibrium throughput coincide with the optimal team throughput.","Title: Delay Tolerant Networks in Partially Overlapped Networks: A Non-cooperative Game Approach.\nYear: 2009\n\nAbstract: Epidemic forwarding protocol in Delay Tolerant Networks maximizes successful data delivery probability but at the same time incurs high costs in terms of redundancy of packet copies in the system and energy consumption. Two-hop routing on the other hand minimizes the packet flooding and the energy costs but degrades the delivery probability. This paper presents a framework to achieve a tradeoff between the successful data delivery probability and the energy costs. Each mobile has to decide which routing protocol it wants to use for packet delivering. In such a problem, we consider a non-cooperative game theory approach. We explore the scenario where the source and the destination mobiles are enclosed in two different regions, which are partially overlapped. We study the impact of the proportion of the surface covered by both regions on the Nash equilibrium and price of anarchy. We also design a fully distributed algorithm that can be employed for convergence to the Nash equilibrium. This algorithm does not require any knowledge of some parameter of the system as the number of mobiles or the rate of contacts between mobiles.","Title: A Stackelberg Model for Opportunistic Sensing in Cognitive Radio Networks.\nYear: 2013\n\nAbstract: We consider a non-cooperative Dynamic Spectrum Access (DSA) game where Secondary Users (SUs) access opportunistically the spectrum licensed for Primary Users (PUs). As SUs spend energy for sensing licensed channels, they may choose to be inactive during a given time slot in order to save energy. Then, there exists a tradeoff between large packet delay, partially due to collisions between SUs, and high-energy consumption spent for sensing the occupation of licensed channels. To overcome this problem, we take into account packet delay and energy consumption into our framework. Due to the partial spectrum sensing, we use a Partial Observable Stochastic Game (POSG) formalism, and we analyze the existence and some properties of the Nash equilibrium using a Linear Program (LP). We identify a paradox: when licensed channels are more occupied by PUs, this may improve the spectrum utilization by SUs. Based on this observation, we propose a Stackelberg formulation of our problem where the network manager may increase the occupation of licensed channels in order to improve the SUs' average throughput. We prove the existence of a Stackelberg equilibrium and we provide some simulations that validate our theoretical findings.","Title: Modeling the effect of forwarding in a multi-hop ad hoc networks with weighted fair queueing\nYear: 2007\n\nAbstract: Consider a wireless ad hoc network with random access channel. We present a model that takes into account topology, routing, random access in MAC layer (governed by IEEE 802.11orslotted aloha) and forwarding probability. In this paper, we are focusing to study the effect of cooperation on the stability and throughput of ad-hoc network. Forwarding packets of other nodes is an example of activity that requires such a collaboration. Hence, it may not be in interest of a node to always forward the requesting packet. We propose a new approach (based on cycle of transmissions) to derive throughput of multi-hop routes and stability of forwarding queues. With this cycle approach, we correct the analytical expressions derived in [2] and discover that their results are valid only in particular cases such as symmetric networks. However, in this paper, we get extended results for general network case. Moreover, we confirm that (i) the forwarding queues in a system of weighted fair queues has a special property and (ii) the end to end throughput of a connection does not depend on the load of the intermediate forwarding queues between a source and a destination. We perform extensive simulations and verify that the analytical results exactly match the results obtained from simulations.","Title: Forever Young: Aging Control In DTNs\nYear: 2010\n\nAbstract: The demand for Internet services that require frequent updates through small\nmessages, also known as microblogging, has tremendously grown in the past few\nyears. Although the use of such applications by domestic users is usually free,\ntheir access from mobile devices is subject to fees and consumes energy from\nlimited batteries. If a user activates his mobile device and is in range of a\nservice provider, a content update is received at the expense of monetary and\nenergy costs. Thus, users face a tradeoff between such costs and their messages\naging. The goal of this paper is to show how to cope with such a tradeoff, by\ndevising \\emph{aging control policies}. An aging control policy consists of\ndeciding, based on the current utility of the last message received, whether to\nactivate the mobile device, and if so, which technology to use (WiFi or 3G). We\npresent a model that yields the optimal aging control policy. Our model is\nbased on a Markov Decision Process in which states correspond to message ages.\nUsing our model, we show the existence of an optimal strategy in the class of\nthreshold strategies, wherein users activate their mobile devices if the age of\ntheir messages surpasses a given threshold and remain inactive otherwise. We\nthen consider strategic content providers (publishers) that offer \\emph{bonus\npackages} to users, so as to incent them to download updates of advertisement\ncampaigns. We provide simple algorithms for publishers to determine optimal\nbonus levels, leveraging the fact that users adopt their optimal aging control\nstrategies. The accuracy of our model is validated against traces from the\nUMass DieselNet bus network.","Title: Mobility management and call admission control for IEEE 802.16e wireless networks\nYear: 2010\n\nAbstract: The demand for wireless broadband access systems supporting mobility of the individual users has dramatically increased in recent years. Therefore, techniques for dynamically adapting the resources to the link condition become necessary for achieving high spectral efficiency. Our aim in this paper is to develop an admission control scheme that handles the intra-cell mobility issue in IEEE802.16e Wireless Networks with link adaptation. In particular we consider two types of intra-cell mobility: Low-mobility and High-mobility. We show how the admission threshold may adaptively controlled to achieve a better balance between guaranteing dropping probability and maximizing resource utilization. Simulation results show that the call admission control scheme under different types of mobility outperforms well by reserving amount of resource in order to provide an high priority of migration calls.","Title: Optimal sensing policies for smartphones in hybrid networks: A POMDP approach\nYear: 2012\n\nAbstract: The ubiquity of mobile devices is fostering a paradigm shift in the realm of Internet services. Consider, for instance, mobile users of social networks, that require frequent updates through small messages from their friends. If a user activates his mobile device and has a contact opportunity with an access point, an update can be received at the expense of monetary and energy costs. Thus, users face a tradeoff between such costs and the utilities of the messages received. The goal of this paper is to show how a user can cope with such a tradeoff, by deriving optimal sensing policies. A sensing policy consists of deciding, based on the age of the last message received and the belief about the future availability of a WiFi access point, whether to activate the mobile device or not. Alternatively, users may also decide to use 3G technology to receive updates, which provides broader coverage at the expense of higher monetary costs and lower bandwidth. To address the tradeoff faced by the users, we propose an analytical model based on a Partially Observed Markov Decision Process (POMDP) with an average reward criterion. Using the proposed model, we show properties of the optimal sensing policy. The applicability of the model and of the derived policy is illustrated through numerical case studies.","Title: Modeling the Effect of Forwarding in IEEE-802.11e-Based Ad Hoc Networks with Weighted Fair Queueing\nYear: 2009\n\nAbstract: We present in this paper a model that takes into account topology, routing, random access in MAC layer governed by IEEE 802.11 and forwarding probability. We are focusing to study the effect of cooperation and PHY/MAC layers (such as transmit power and physical carrier sense) on the stability and throughput of ad-hoc network. In this paper, we first extended the results of Yang's model [1] to a multi-hop ad-hoc network with asymmetric topology and traffic load. Based on this analysis, we reach a fixed-point formulation which characterizes the system operating points and allows us to compute the attempt rate of MAC layer which depends on the traffic flows on the network layer. With the use of the analytical model derived, we use an approach to derive throughput of multi-hop routes and stability of forwarding queues. However, in this paper, we get extended results for general network case.","Title: Perturbation of linear quadratic systems with jump parameters and hybrid controls\nYear: 2000\n\nAbstract: . We consider the problem of the perturbation of a class of linear-quadratic differential games with piecewise deterministic\n dynamics, where the changes from one structure (for the dynamics) to another are governed by a finite-state Markov process.\n Player 1 controls the continuous dynamics, whereas Player 2 controls the rate of transition for the finite-state Markov process;\n both have access to the states of both processes. Player 1 wishes to minimize a given quadratic performance index, while player\n 2 wishes to maximize or minimize the same quantity. The problem above leads to the analysis of some linearly coupled set of\n quadratic equations (Riccati equations). We obtain a Taylor expansion in the perturbation for the solution of the equation\n for a fixed stationary policy of the player 2. This allows us to solve the game or team problem as a function of the perturbation.","Title: Blind online optimal forwarding in heterogeneous delay tolerant networks\nYear: 2011\n\nAbstract: In this paper we consider a heterogeneous DTN formed by different classes of nodes. A core problem in DTNs literature is to optimize forwarding control such in a way to deliver messages from source to destination under a given energy budget. In this context, we model the presence of different classes and describe the structure of the optimal solution. The closed form of an optimal policy is found in the family of the multi-dimensional dynamic strategies. These policies assign a specific threshold to each class of mobiles. We also show that in the multi-class case there exists an order of the classes that depends on the system parameters in a very simple form. In particular, we exploit the ordering induced by the optimal control with the goal of designing a new class of multidimensional stochastic approximation algorithms that apply to the heterogeneous DTNs. These algorithms are blind because they drive online the source node to the optimal operating point without requiring explicit estimation of network parameters (intermeeting intensities and the number of nodes for each class).","Title: Analytical Framework for Contact Time Evaluation in Delay-Tolerant Networks.\nYear: 2009\n\nAbstract: In the last few years, there has been an increasing concern about stochastic properties of contact-based metrics under general mobility models in delay-tolerant networks. Such a concern will provide a first step toward detailed performance analysis of various routing/forwarding algorithms and shed light on better design of network protocols under realistic mobility patterns. However, throughout the variety of research works in this topic, most interests rather focused on the inter-contact time while other contact-based metrics such as the contact time received too little interest. In this paper, we provide an analytical framework to estimate the contact time in delay-tolerant networks based on some recent key results derived from biology and statistical physics while studying spontaneous displacement of insects such as ants. In particular, we analytically derive a closed-form expression for the average value of the contact time under the random waypoint mobility model and then give an approximation for its distribution function.","Title: Optimal number of users in wireless networks: A flat rate pricing\nYear: 2012\n\nAbstract: We study revenue-maximizing in a wireless system where interested users share a common spectrum and interfere with each other. Capacity is increased in proportion of the number of users. Our objective is to design a scheme that achieves an optimal solution for the provider with respect to users strategy, bandwidth management and some fairness criteria. We consider a flat rate pricing for users. Two cases are studied: firstly, an approximation to the original problem when the provider has a perfect knowledge about the channel state of each user is proposed. In the second case, a decision theoretic approach based on a POMDP framework is elaborated when the provider has only a partial information on the system state. Additionally, we provide numerical results that illustrate the performance of the proposed solutions.","Title: Dynamic retransmission limit scheme for routing in multi-hop ad hoc networks\nYear: 2007\n\nAbstract: In paper [2] and [3], we have studied the throughput and stability of forwarding queues in a wireless ad hoc network with random access channel. In this paper, we are focusing to draw benefit from the interaction of the MAC (governed by IEEE 802.11 or slotted aloha) and routing by defining a new cross-layer scheme for routing based on the limit number of retransmission. By adjusting dynamically and judiciously this parameter in a saturated network, we have realized that both stability and average throughput are significantly improved in linear symmetric networks: a gain of 100 % can be reached, while in asymmetric networks, we achieve a better average delay (resp. throughput) for each connection without changing the average throughput (resp. delay). A detailed performance study of our new scheme is presented using analytical and simulation evaluation.","Title: A survey on YouTube streaming service.\nYear: 2011\n\nAbstract: Established in 2005, YouTube is one of the fastest-growing websites, and has become one of the most accessed sites in the Internet. It has a significant impact on the Internet traffic distribution, but itself is suffering from severe scalability constraints and quality of service. Understanding the features of YouTube is thus crucial to network traffic engineering and to sustainable development of this new generation of services. In this survey, we first present an overview of previous works and analysis on YouTube with a particular attention on Quality of Experience. We then describe how the increased availability of meta-data in Web 2.0 (e.g., popularity distribution of video clips) could be effectively exploited to improve the performance and scalability of YouTube. In particular, we study the benefit gained by local caching along with prefetching in terms of reducing the client access time and start up delay in watching video.","Title: Stochastic learning solution for constrained Nash Equilibrium throughput in non saturated wireless collision channels\nYear: 2009\n\nAbstract: We consider finite number of users, with infinite buffer storage, sharing a single channel using the aloha medium access protocol. This is an interesting example of a non saturated collision channel. We investigate the uplink case of a cellular system where each user will select a desired throughput. The users then participate in a non cooperative game wherein they adjust their transmit rate to attain their desired throughput. We show that this game, in contrast to the saturated case, either has no Nash Equilibrium or has infinitely many Nash Equilibria. Further, we show that the region of NE coincides with an appropriate 'stability region'. We also discuss the efficiency of the equilibria in term of energy consumption and congestion rate. Next, we propose two learning algorithms using a stochastic iterative procedure that converges to the best Nash equilibrium. For instance, the first one needs partial information (transmit rates of other users during the last slot) whereas the second is an information less and fully distributed scheme. We approximate the control iterations by an equivalent ordinary differential equation in order to prove that the proposed stochastic learning algorithm converges to a Nash equilibrium even in the absence of any coordination or extra information. Extensive numerical examples and simulations are provided to validate our results.","Title: Altruism in groups: An evolutionary games approach\nYear: 2014\n\nAbstract: We revisit in this paper the relation between evolution of species and the mathematical tool of evolutionary games, which has been used to model and predict it. We indicate known shortcoming of this model that restricts the capacity of evolutionary games to model groups of individuals that share a common gene or a common fitness function. In this paper we provide a new concept to remedy this shortcoming in the standard evolutionary games in order to cover this kind of behavior. Further, we explore the relationship between this new concept and Nash equilibrium or ESS. We indicate through the study of some example in the biology as Hawk-Dove game, Stag-Hunt Game and Prisoner's Dilemma, that when taking into account a utility that is common to a group of individuals, the equilibrium structure may change dramatically. We also study the multiple access control in slotted Aloha based wireless networks. We analyze the impact of the altruism behavior on the performance at the equilibrium.","Title: Flow Level QoE of Video Streaming in Wireless Networks.\nYear: 2014\n\nAbstract: The Quality of Experience (QoE) of streaming service is often degraded by frequent playback interruptions. To mitigate the interruptions, the media player prefetches streaming contents before starting playback, at a cost of delay. We study the QoE of streaming from the perspective of flow dynamics. First, a framework is developed for QoE when streaming users join the network randomly and leave after downloading completion. We compute the distribution of prefetching delay using partial differential equations (PDEs), and the probability generating function of playout buffer starvations using ordinary differential equations (ODEs) for CBR streaming. Second, we extend our framework to characterize the throughput variation caused by opportunistic scheduling at the base station, and the playback variation of VBR streaming. Our study reveals that the flow dynamics is the fundamental reason of playback starvation. The QoE of streaming service is dominated by the first moments such as the average throughput of opportunistic scheduling and the mean playback rate. While the variances of throughput and playback rate have very limited impact on starvation behavior.","Title: A Game Theoretic Model For Network Virus Protection\nYear: 2014\n\nAbstract: Security is crucial for information systems. In a company, security management is traditionally controlled via a centralized single-point. However, when we deal with multiple computer systems interconnected in a wide area networks (WAN), the use of a central authority for security management is completely meaningless. In this paper, we propose a distributed decision-making designed to thwart viruses in a WAN. A key aspect is whether owners of devices are willing to update their anti-virus in order to protect their computers or not to pay for an anti-virus update and take the risk to be contaminated. Given the fact that computers are interconnected via networks and the Internet, the risk of being infected does not only depend on each computer's strategy, but also on the strategies chosen by other computers in the network. This makes the virus protection problem much more challenging. To do so, we model the interaction between nodes as a non-cooperative game in which each node decides individually whether to update the anti-virus or not. The virus spread is assumed to follow a biologically inspired epidemic model in which the dynamic of sources that disseminate the virus evolves as function of the popularity of virus using the influence linear threshold model. We first provide a full characterization of the equilibria of the game and then we investigate the impact of the update cost. In particular, we study the performance of the strategies at the equilibrium in terms of the update cost and the network size on both the security management system and the anti-virus producers. These results give some helpful insights on how secure is decentralizing antivirus update decisions.","Title: Evolutionary stable strategies in interacting communities.\nYear: 2013\n\nAbstract: In many social networks or biological systems, several populations may be influenced by interactions with not only individuals of their own communities but also with other communities in the network. Moreover, the inter-community interactions can affect the outcome of a community in addition to the strategies followed inside the community itself. However, throughout the variety of research works in this topic, classical model of evolutionary games rather assumes individuals are equally likely to interact with any other member of the population and the success of any individual depends only on the frequency with which all strategies are represented in the population. Instead, we devote in this paper a novel mathematical tool to model the evolution of a population composed of several communities where the interaction between them is non-uniform.\n\nThe main objective of our work is to model these interactions using evolutionary games and to find the evolutionary stable strategies (ESS). Hence, we show through the study of two communities and pairwise interactions that when taking into account the non-uniform interaction, any mixed Nash equilibrium is not an ESS wherein all communities are robust against a deviation of a fraction of the population from both communities which may wish to deviate. Moreover, the same mixed equilibrium is an ESS under some conditions when we consider the fitness of the whole population instead of the fitness of each community. We also demonstrate that this equilibrium is globally asymptotically stable for the replicator dynamic which describes the dynamic of the evolution of communities as well as the impact of the interaction between communities.","Title: Delayed Evolutionary game dynamics with non-uniform interactions in two communities.\nYear: 2014\n\nAbstract: In this paper we study the evolution of a population composed of several communities where the interaction between them is non-uniform. The main objective is to study the effect of slow time scale delays on the convergence of the replicator dynamics to the Evolutionary Stable Strategy (ESS). In particular, we investigate the effect of time delays on the stability of the replicator dynamics. We can distinguish two types of the time delay: strategic delay which is the delay associated with the strategies of players, and spatial delay, which is the delay associated with the communities of the players. We prove that for large strategic delays, the mixed intermediate ESS may become an unstable state for the replicator dynamics, while for any value of the spatial delay, the replicator dynamics converge to the intermediate ESS. We also show that when both types of the time delay coexist, the instability property is emphasized.","Title: Bio-inspired models for characterizing YouTube viewcout.\nYear: 2014\n\nAbstract: The goal of this paper is to study the behaviour of viewcount in YouTube. We first propose several bio-inspired models for the evolution of the viewcount of YouTube videos. We show, using a large set of empirical data, that the viewcount for 90% of videos in YouTube can indeed be associated to at least one of these models, with a Mean Error which does not exceed 5%. We derive automatic ways of classifying the viewcount curve into one of these models and of extracting the most suitable parameters of the model. We study empirically the impact of videos' popularity and category on the evolution of its viewcount. We finally use the above classification along with the automatic parameters extraction in order to predict the evolution of videos' viewcount.","Title: Adaptive Modulation and Coding scheme with intra- and inter-cell mobility for HSDPA system\nYear: 2009\n\nAbstract: The Adaptive Modulation and Coding (AMC) scheme which handles user's mobility issue plays a significant role to improve the desired quality of service in high Speed Downlink Packet Access (HSDPA) networks. We develop a resource allocation which maintains constant bit rate for real-time (RT) and non-real-time (NRT) calls independently of the user position in the cell. We present analytical results relating to user performance in terms of blocking and dropping probabilities, and the results relating to the system performance in term of total throughput. We propose a call admission control scheme to derive a good trad-off between the blocking and dropping probabilities, and the choice of scheduling that efficiently utilizes the available resources.","Title: Rejuvenation and the Spread of Epidemics in General Topologies\nYear: 2014\n\nAbstract: Epidemic models have received significant attention in the past few decades to study the propagation of viruses, worms and ideas in computer and social networks. In the case of viruses, the goal is to understand how the topology of the network and the properties of the nodes that comprise the network, together, impact the spread of the epidemics. In this paper, we propose rejuvenation as a way to cope with epidemics. Then, we present a model to study the effect of rejuvenation and of the topology on the steady-state number of infected and failed nodes. We distinguish between a state in which the virus is incubating and in which symptoms might not be visible and yet they may be contagious and infecting other nodes, and a state of failure where symptoms are clear. Sampling costs might be incurred to examine nodes in search for viruses at an early stage. Using the proposed model, we show that the sampling rate admits at most one local minimum greater than zero. Then, we numerically illustrate the impact of different system parameters on the optimal sampling rate, indicating when rejuvenation is more beneficial.","Title: Modeling rewards and incentive mechanisms for Delay Tolerant Networks\nYear: 2014\n\nAbstract: A central problem in Delay Tolerant Networks (DTNs) is to persuade mobile nodes to participate in relaying messages. Indeed, the delivery of a message incurs a certain number of costs for a relay. We consider a two-hop DTN in which a source node, wanting to get its message across to the destination as fast as possible, promises each relay it meets a reward. This reward is the minimum amount that offsets the expected delivery cost, as estimated by the relay from the information given by the source (number of existing copies of the message, age of these copies). A reward is given only to the relay that is the first one to deliver the message to the destination. For two relays and exponentially distributed inter-contact times, we show that the expected reward the source pays remains the same irrespective of the information it conveys, provided that the type of information does not vary dynamically over time. On the other hand, the source can gain by adapting the information that it conveys to a meeting relay.","Title: Fair Scheduling in Cellular Systems in the Presence of Noncooperative Mobiles\nYear: 2014\n\nAbstract: We consider the problem of centrally controlled 'fair' scheduling of resources to one of the many mobile stations connected to a base station (BS). The BS is the only entity making decisions in this framework based on truthful information from the mobiles on their radio channel. We study the well-known family of parametric ¿-fair scheduling problems from a game-theoretic perspective in which some of the mobiles may be noncooperative. We first show that if the BS is unaware of the noncooperative behavior from the mobiles, the noncooperative mobiles become successful in snatching the resources from the other cooperative mobiles, resulting in unfair allocations. If the BS is aware of the noncooperative mobiles, a new game arises with BS as an additional player. It can then do better by neglecting the signals from the noncooperative mobiles. The BS, however, becomes successful in eliciting the truthful signals from the mobiles only when it uses additional information (signal statistics). This new policy along with the truthful signals from mobiles forms a Nash Equilibrium (NE) called a Truth Revealing Equilibrium. Finally, we propose new iterative algorithms to implement fair scheduling policies that robustify the otherwise non-robust (in presence of noncooperation) ¿-fair scheduling algorithms.","Title: Delayed Evolutionary Game Dynamics applied to Medium Access Control\nYear: 2007\n\nAbstract: A major contribution of biology to competitive decision making is the development of the discipline of evolutionary games. Its ESS (evolutionary stable strategy) equilibrium concept, well adapted to large populations of players, describes robustness against deviations of a whole fraction of the population (in contrast to Nash equilibrium that requires robustness against a single user's deviation). The second appealing element of evolutionary games, the replicator dynamics, describes the evolution of strategies in time. Under suitable conditions, this (and some other) dynamics, converge to an ESS. In this paper we study the effect of slow time scale delays on the convergence of various dynamics. We apply this to an evolutionary game describing competition between mobile terminals over the access to a common channel.","Title: Analytical Model for Connectivity in Vehicular Ad Hoc Networks\nYear: 2008\n\nAbstract: We investigate connectivity in the ad hoc network formed between vehicles that move on a typical highway. We use a common model in vehicular traffic theory in which a fixed point on the highway sees cars passing it that are separated by times with an exponentially distributed duration. We obtain the distribution of the distances between the cars, which allows us to use techniques from queuing theory to study connectivity. We obtain the Laplace transform of the probability distribution of the connectivity distance, explicit expressions for the expected connectivity distance, and the probability distribution and expectation of the number of cars in a platoon. Then, we conduct extensive simulation studies to evaluate the obtained results. The analytical model that we present is able to describe the effects of various system parameters, including road traffic parameters (i.e., speed distribution and traffic flow) and the transmission range of vehicles, on the connectivity. To more precisely study the effect of speed on connectivity, we provide bounds obtained using stochastic ordering techniques. Our approach is based on the work of Miorandi and Altman, which transformed the problem of connectivity distance distribution into that of the distribution of the busy period of an equivalent infinite server queue. We use our analytical results, along with common road traffic statistical data, to understand connectivity in vehicular ad hoc networks.","Title: Evolutionary games with random number of interacting players applied to access control\nYear: 2008\n\nAbstract: We study access games within a large population of mobiles that interfere with each other through many local interactions. Each local interaction involves a random number of mobiles. The games are not necessarily reciprocal as the set of mobiles causing interference to a given mobile may differ from the set of those suffering from its interference. We model and study this using the theory of evolutionary games which we extend to cover a random number of players.","Title: An evolutionary game approach for the design of congestion control protocols in wireless networks\nYear: 2008\n\nAbstract: Evolutionary games have been developed in biological sciences in the aim of studying the evolution and equilibrium behavior (called evolutionary stable strategies - ESS) of large populations. While rich theoretical foundations of evolutionary games allow biologist to explain past and present evolution and predict future evolution, it can be further used in engineering to architect evolution. In this paper, we apply evolutionary games to non-cooperative networks containing large number of individual non-cooperative terminals or sensors. We study the evolution of congestion control protocols, and show how the evolution and the ESS are influenced by the characteristics of the wireless channel. We then consider the challenge of architecting the evolution: we propose some guidelines for designing a framework that supports evolution of protocols.","Title: Evolutionary game dynamics with migration for hybrid power control in wireless communications\nYear: 2008\n\nAbstract: We propose an evolutionary game dynamics with migration for hybrid population games with many local interactions at the same time. Each local interaction concerns a random number of interacting players. The strategies of a player have two components. Specifically, each player chooses both (i) the region or subpopulation and (ii) an action among a finite set of secondary pure strategies in each region. We assume that when updating a strategy, a player can change only the secondary strategies associate to the region at a time. We investigate what impact this restriction has on the population dynamics. We apply this model to the integrated power control and base station assignment problem in a multi-cell in code division multiple access (CDMA) wireless data networks with large number of mobiles. We show that global neutrally evolutionary stable strategies are stationary points of hybrid mean dynamics called dynamics with multicomponent strategies under the positive correlation conditions. We give some convergence results of our hybrid model in stable population games and potential population games.","Title: Markov decision evolutionary game for energy management in Delay Tolerant Networks\nYear: 2011\n\nAbstract: In this paper, we apply the concepts of Markov decision evolutionary games to non-cooperative forwarding control of Delay Tolerant Networks (DTN). Specifically, we rely on the design of mechanisms at the source node to study forwarding probability of the message in a DTN using the two-hop routing. We study the forwarding probability as a function of the competition within a large population of mobiles which need occasionally to make some action. In particular, for each message generated by a source, each mobile may take a decision that concerns the strategy by which the mobile participates to the relaying of the message from source to destination. A mobile that participates receives a unit of reward if it is the first to deliver a copy of the packet to the destination. The action taken by a mobile determine not only the immediate reward but also the transition probability to its next battery energy state. We characterize the Evolutionary Stable Strategies (ESS) for these games and propose a method to compute them. We also propose a mechanism design at the source in order to maximize the message delivery probability to the destination, given the equilibrium behavior (called Evolutionary Stable Strategy - ESS).","Title: Probabilistic analysis of buffer starvation in Markovian queues\nYear: 2011\n\nAbstract: Our purpose in this paper is to obtain the exact distribution of the number of buffer starvations within a sequence of N consecutive packet arrivals. The buffer is modeled as an M/M/1 queue. When the buffer is empty, the service restarts after a certain amount of packets are prefetched. With this goal, we propose two approaches, one of which is based on Ballot theorem, and the other uses recursive equations. The Ballot theorem approach gives an explicit solution, but at the cost of the high complexity order in certain circumstances. The recursive approach, though not offering an explicit result, needs fewer computations. We further propose a fluid analysis of starvation probability on the file level, given the distribution of file size and the traffic intensity. The starvation probabilities of this paper have many potential applications. We apply them to optimize the quality of experience (QoE) of media streaming service, by exploiting the tradeoff between the start-up delay and the starvation.","Title: Accumulative interference removal in multi-hop linear and circular ad hoc WLANs\nYear: 2012\n\nAbstract: This paper presents a cross-layered framework to study linear and circular ad hoc networks built on the IEEE 802.11e Enhanced Distributed Coordination Function. We investigate the intricate interactions among PHY, MAC and Network layers. Now, the carrier sense threshold, the transmit power, the contention window size, the retransmissions retry limit and the multi rates are jointly incorporated. We propose an equivalent network where each node is able to listen to other nodes, i.e., the effect of hidden nodes could be definitely eliminated. Then, we develop an analytical model that predicts the throughput of each connection as well as the stability of forwarding queues at intermediate nodes while each node is able to prevent the accumulative interference phenomenon. Further, performance of such a system is evaluated via simulation. We show that the performance measures of MAC layer are affected by the traffic intensity of flows to be forwarded and the average hop length. Moreover, attempt rate and collision probability are dependent on the traffic flows, PHY parameters and routing scheme.","Title: Optimal energy-delay tradeoff policies in cognitive radio networks\nYear: 2012\n\nAbstract: Cognitive radio (CR) has been considered as a promising technology to enhance spectrum efficiency via opportunistic transmission at link level. We consider Opportunistic Spectrum Access (OSA) mechanism that takes into account packet delay and energy consumption. We formulate the OSA problem as a Partially Observable Markov Decision Process (POMDP) by explicitly considering the energy constraint as well as, the delay constraint, which are often ignored in existing OSA solutions. Specifically, we consider a POMDP with an average reward criterion. We further consider that the secondary user (SU) may decide, at any moment, to use another dedicated way (3G) of communication in order to transmit its packets. We derive structural properties of the value function and we show the existence of optimal strategies in the class of the threshold strategies. In particular, numerical illustrations validate our theoretical findings. It is shown that optimal policy has a threshold structure.","Title: Interplay of contact times, fragmentation and coding in DTNs\nYear: 2013\n\nAbstract: Models describing DTNs proposed in the last few years have been focusing on message replication policies able to achieve high delivery probability at the cost of network resources, e.g., message copies. To this respect, the duration of contact events is the physical constraint that dictates how fine a large message should be fragmented into packets in order to match finite contacts duration. The price to pay, indeed, is that the source has to deliver a larger number of packets per message. In this paper we model the combined effect of message fragmentation and buffering and describe the structure of the forwarding process in closed form when the message is split into K packets and delivered to the destination. We consider the specific case of sequential forwarding, where the source delivers fragmented message packets in order to relays. In this case the interplay of forwarding and message fragmentation can be expressed in closed form by coupling the combinatorial structure for packet forwarding and fluid models for the replication of packets in the network. By deriving the closed form expression for the delivery probability, we are able to derive the optimal fragmentation K* as a function of the contact time distribution. Finally, results on sequential forwarding are applied to derive performance figures for the case when fountain coding is applied: redundancy is added to the original information with the aim to increase the message delivery probability. The paper is completed by numerical results.","Title: New insights from a bounded rationality analysis for strategic price-QoS war\nYear: 2012\n\nAbstract: We present a game theoretic framework for the dynamical behaviors of a duopoly game in telecommunications service providers' context. Competition between two Service Providers (SPs) is assumed to take place in terms of their pricing decisions and the Quality of Service (QoS) they offer. According to the SPs' rationality level, we consider two schemes: 1) Both SPs are rational, and 2) One SP is rational and the second SP is boundedly rational. We describe the competitive interaction and analyze the resulting equilibria. Later, we compute explicitly the steady states of the dynamical system induced by bounded rationality, and establish a necessary and sufficient condition for stability of its Nash equilibria (NEs). We prove that there exists exactly one NE which is fair whereas remaining equilibria are unfair. A special feature is that the stability condition of the Nash equilibrium coincides with the instability condition of the boundary equilibria. Thus the system would never be absorbed by any of the unfair equilibria which solves the equilibrium selection issue ! Moreover, we show that considering the delay case (i.e., assuming a market with memory) increases the stability of the system. Here, the size of the memory could be considered for multi-level rationality, which means that bounded rationality tends to rationality as the memory size increases. We finally show that boundedly rational SPs with delay have a higher chance of reaching the fair Nash equilibrium.","Title: A Theoretical Framework for Hierarchical Routing Games\nYear: 2010\n\nAbstract: Most theoretical research on routing games in telecommunication networks has so far dealt with reciprocal congestion effects between routed entities. Yet in networks that support differentiation between flows, the congestion experienced by a packet depends on its priority level. Another differentiation is made by compressing the packets in the low priority flow while leaving the high priority flow intact. In this paper we study such kind of routing scenarios for the case of non-atomic users and we establish conditions for the existence and uniqueness of equilibrium.","Title: Routing games : From egoism to altruism\nYear: 2010\n\nAbstract: The paper studies the routing in the network shared by several users. Each user seeks to optimize either its own performance or some combination between its own performance and that of other users, by controlling the routing of its given flow demand. We parameterize the degree of cooperation which allows to cover the fully non-cooperative behavior, the fully cooperative behavior, and even more, the fully al- truistic behavior, all these as special cases of the paramet er's choice. A large part of the work consists in exploring the impact of the degree of cooperation on the equilibrium. Our first finding is to identify multiple Nash equilibria with cooperative behavior that do not occur in the non-cooperative case under the same conditions (cost, demand and topology). We then identify Braess like paradox (in which adding capacity or adding a link to a network results in worse performance to all users) in presence of user's co- operation. We identify another type of paradox in cooperation scenario: when a given user increases its degree of cooperation while other users keep unchanged their degree of cooperation, this may lead to an improvement in performance of that given user. We then pursue the exploration and carry it on to the setting of Mixed equilibrium (i.e. some users are non atomic-they have infinitesimally small de mand, and other have fi- nite fixed demand). We finally obtain some theoretical result s that show that for low degree of cooperation the equilibrium is unique, confirming the results of our numerical study.","Title: Evolutionary forwarding games in Delay Tolerant Networks\nYear: 2010\n\nAbstract: In this paper, we apply evolutionary games to non-cooperative forwarding control of Delay Tolerant Networks (DTN). We focus our study on the probability to deliver a message from source to destination in a DTN using two-hop routing. We derive the success probability as a function of the competition within a large population of mobiles. In particular, for each message generated by a source, each mobile may take a decision that concerns the strategy by which the mobile participates to the relaying. A mobile that participates receives a unit of reward if it is the first to deliver a copy of the packet to the destination. A utility function is introduced as the difference between a reward unit and the energy cost. We show how the evolution dynamics and the equilibrium behavior (called Evolutionary Stable Strategy — ESS) are influenced by the characteristics of inter contact time, energy expenditure and pricing characteristics. We specialize our analysis also to mechanisms that the source may introduce in order to have the message delivered to the destination with high probability within a given deadline.","Title: Atomic hierarchical routing games in communication networks\nYear: 2010\n\nAbstract: Theoretical studies on routing games in networks have so far dealt with reciprocal congestion effects between routing entities. But, with the advent of technologies like Cognitive Radio, we have networks which support differentiation of flows. In a two priority level model a user can be high priority or low priority and there is a cost for such a classification. The point of departure of this model from the traditional routing scenarios is the absence of reciprocity in the congestion effects: The low priority flow faces congestion from both high priority as well as low priority flow while the high priority flow is immune to the congestion effects from the low priority flow. This hierarchy is naturally present in contexts where there are primary (licensed) users and secondary (unlicensed) users who can sense their environment because there are equipped with a cognitive radio. We study such kind of routing scenarios for the cases of atomic users. We establish the existence and the uniqueness of Nash equilibrium and further we show the existence of a potential function for linear congestion costs and a certain priority classification pricing scheme. Natural applications of this model to Cognitive Radio are also pointed out.","Title: Coordination minority games in delay tolerant networks\nYear: 2013\n\nAbstract: In this paper we introduce a novel framework for the distributed control of DTNs. The mechanism that we propose tackles a crucial aspect of such systems: in order to support message replication the devices acting as relays need to sacrifice part of their batteries. The aim is thus to provide a reward mechanism able to induce activation of relays in a coordinated fashion. The proposed scheme functions in non-cooperative fashion, and requires minimal message exchange to operate. In particular, relays choose among two strategies: either to participate to message relaying, or not to participate in order to save energy. The base for our mechanism design is to define the relays' utility function according to a minority game; in fact, relays compete to be in the population minority with respect to activation. By tuning the activation level, the system can hence control and optimize the DTN operating point in a distributed manner. To this respect, we characterize extensively the possible equilibria of this game. Finally, a stochastic learning algorithm is proposed which can provably drive the system to the equilibrium solution without requiring perfect state information at relay nodes. We provide extensive numerical results to validate the proposed scheme.","Title: Stochastic Geometric Models for Green Networking.\nYear: 2015\n\nAbstract: In this paper, we use a stochastic geometric approach in order to study the impact on energy consumption when base stations are switched OFF independently of each other. We present here both the uplink and downlink analysis based on the assumption that the base stations are distributed according to an independent stationary Poisson point process. This type of modeling allows us to make use of the property that the spatial distribution of the base stations after thinning (switching oFF) is still a Poisson process. This implies that the probability distribution of the signal to interference and noise ratio (SINR) can be kept unchanged when switching-oFF base stations provided that we scale up the transmission power of the remaining base stations. We then solve the problem of optimally selecting the switch-oFF probabilities so as to minimize the energy consumptions, while keeping unchanged the SINR probability distribution. We then study the tradeoff in the uplink performance involved in switching-oFF base stations. These include the energy consumption, the coverage and capacity, and the impact on amount of radiation absorbed by the transmitting user.","Title: Core Stable Algorithms for Coalition Games with Complementarities and Peer Effects.\nYear: 2015\n\nAbstract: In this short paper, we show two new algorithms for finding stable structures in ordinal coalition potential games. The first one is anytime and enumerative. It performs on a graph. The second one is a modified Deferred Acceptance Algorithm (DAA) using counter-proposals. It finds a many-to-one matching. We illustrate with the example of video caching from a content creator's servers to a service provider's servers.","Title: About Joint Stable User Association and Resource Allocation in Multi-Rate IEEE 802.11 WLANs\nYear: 2015\n\nAbstract: This paper aims at proposing an alternative approach for both the modeling of the IEEE 802.11 resource allocation scheme and the design of mechanisms to reduce the impact of the anomaly of the protocol. We use game theory to model the IEEE 802.11 resource allocation and mobiles users to APs association as a coalition matching game. We propose a new mechanism that gives mobile users and APs the incentive to associate with each others in a way that both absorbs the load and reduce the negative impact of the anomaly in IEEE 802.11.","Title: Controlled Matching Game for User Association and Resource Allocation in Multi-Rate WLANs\nYear: 2015\n\nAbstract: The deployment of IEEE 802.11 based WLANs in populated areas is such that many mobile terminals are covered by several Access Points (APs). These mobiles have the possibility to associate to the AP with the strongest signal (best-RSSI association scheme). This can lead to poor performances and overloaded APs. Moreover, the well known anomaly in the protocol at the MAC layer may also lead to very unpredictable performances and affect the system throughput due to the presence of heterogeneous data rate nodes and the shared nature of the 802.11 medium. The goal of this paper is to propose an alternative approach for the association. We model the joint resource allocation and mobile user association as a matching game with complementarities, peer effects and selfish players(1). We focus on the throughput fairness allocation induced by the saturated regime with equal packet sizes. We propose a novel three-stages mechanism for the modeling and control of load balancing, resource allocation and user association. We show that the proposed mechanism can greatly improve the efficiency of 802.11 with heterogeneous nodes and reduce the negative impact of peer effects such as the anomaly in IEEE 802.11.","Title: A fully distributed learning algorithm for power allocation in heterogeneous networks\nYear: 2015\n\nAbstract: In this work, we present a fully distributed Learning algorithm for power allocation in HetNets, referred to as the FLAPH, that reaches the global optimum given by the total social welfare. Using a mix of macro and femto base stations, we discuss opportunities to maximize users global throughput. We prove the convergence of the algorithm and compare its performance with the well-established Gibbs and Max-logit algorithms which ensure convergence to the global optimum. Algorithms are compared in terms of computational complexity, memory space, and time convergence.","Title: A Study of YouTube Recommendation Graph Based on Measurements and Stochastic Tools\nYear: 2015\n\nAbstract: The Youtube recommendation is one the most important view source of a video. In this paper, we focus on the recommendation system in boosting the popularity of videos. We first construct a graph that captures the recommendation system in Youtube and study empirically the relationship between the number of views of a video and the average number of views of the videos in its recommendation list. We then consider a random walker on the recommendation graph, i.e. a random user that browses through videos such that the video it chooses to watch is selected randomly among the videos in the recommendation list of the previous video it watched. We study the stability properties of this random process and we show that the trajectory obtained does not contain cycles if the number of videos in the recommendation list is small (which is the case if the computer's screen is small).","Title: Backward-shifted strategies based on SVC for HTTP adaptive video streaming\nYear: 2016\n\nAbstract: Although HTTP-based video streaming can easily penetrate firewalls and profit from Web caches, the underlying TCP may introduce large delays in case of a sudden capacity loss. To avoid an interruption of the video stream in such cases we propose the Backward-Shifted Coding (BSC). Based on Scalable Video Coding (SVC), BSC adds a time-shifted layer of redundancy to the video stream such that future frames are downloaded at any instant. This pre-fetched content maintains a fluent video stream even under highly variant network conditions and leads to high Quality of Experience (QoE). We characterize this QoE gain by analyzing initial buffering time, re-buffering time and content resolution using the Ballot theorem. The probability generating functions of the playback interruption and of the initial buffering latency are provided in closed form. We further compute the quasi-stationary distribution of the video quality, in order to compute the average quality, as well as temporal variability in video quality. Employing these analytic results to optimize QoE shows interesting trade-offs and video streaming at outstanding fluency.","Title: Analysis and modelling Quality of Experience of video streaming under time-varying bandwidth.\nYear: 2016\n\nAbstract: We take an analytical approach to study the Quality of user Experience (QoE) for video streaming applications. Our propose is to characterise buffer starvations for streaming video in the presence of bandwidth that varies over time. This variation of the bandwidth is caused due to the frequent changes of channel states and mobility. Specifically we develop an analytical framework to investigate Quality of user Experience (QoE) for streaming by considering a Markov Modulated Fluid Model (MMFM) that accurately approximates the nature of network traffic. We drive the close-form expressions for calculating the distribution of starvation as well as start-up delay using partial differential equations (PDEs) and solve them using the Laplace Transform. We illustrate the results with the cases of the two-state Markov Modulated Fluid Model that is commonly used in multimedia applications. We compare our analytical model with simulation results using ns-3 under various operating parameters. Finally, we apply our analysis results to optimize the objective quality of experience (QoE) of media streaming realizing the tradeoff among different metrics incorporating user preferences on buffering ratio, startup delay and perceived quality.","Title: NEWCAST: Anticipating resource management and QoE provisioning for mobile video streaming\nYear: 2016\n\nAbstract: The knowledge of the future capacity variations in wireless networks using smartphones becomes more and more possible by exploiting the rich contextual information from smartphone sensors through mobile applications and services. It is entirely likely that such contextual information, which may include the traffic, mobility and radio conditions, could lead to a novel agile resource management not yet thought of. Inspired by the attractive features and potential advantages of this agile resource management, several approaches have been proposed during the last period. However, agile resource management also comes with its own challenges, and there are significant technical issues that still need to be addressed for successful rollout and operation of this technique. In this paper, we propose an approach (called NEWCAST) for anticipating throughput variation for mobile video streaming services. The solution of the optimization problem realizes a fundamental trade-off among critical metrics that impact the user's perceptual quality of the experience (QoE) and system utilization. Both simulated and real-world traces collected from [1] are carried out to evaluate the performance of NEWCAST. In particular, it is shown that NEWCAST provides the efficiency, computational complexity and robustness that the new 5G architectures require.","Title: Coordinated scheduling via frequency and power allocation optimization in LTE cellular networks.\nYear: 2016\n\nAbstract: Due to Orthogonal Frequency Division Multiple Access (OFDMA) mechanism adopted in LTE cellular networks, intra-cell interference is nearly absent. Yet, as these networks are designed for a frequency reuse factor of 1 to maximize the utilization of the licensed bandwidth, inter-cell interference coordination remains an important challenge. In both homogeneous and heterogeneous cellular networks, there is a need for scheduling coordination techniques to efficiently distribute the resources and mitigate inter-cell interference. In this paper, we propose a dynamic solution of inter-cell interference coordination performing an optimization of frequency sub-band reuse and transmission power in order to maximize the overall network utility. The proposed framework, based on game theory, permits to dynamically define frequency and transmission power patterns for each cell in the coordinated cluster. Simulation results show significant benefits in average throughput and also cell edge user throughput of 40% and 55% gains when performing the frequency sub-band muting and power control. Furthermore, we also obtain a meaningful improvement in energy efficiency.","Title: Identifying A Volunteer-Like Dilemma In Cooperative Sensing-Empowered Cognitive Radio Networks\nYear: 2016\n\nAbstract: Cooperative sensing is a promising technique that enables secondary users (SUs) to combine their channel observations in order to improve the spectrum sensing accuracy. However, in an adversarial environment where the secondary nodes are particularly selfish and the spectrum sensing is energy costly, selfish SUs can easily exploit the spectrum sensing result without participating in the sensing process. In this paper, we model the cooperative sensing as a volunteer dilemma where SUs have the choice to volunteer in order to sense and share the spectrum sensing results, or to free ride the spectrum sensing and achieve possibly a higher profitability. We mainly give a full characterization of the Nash equilibria and prove a counter-intuitive property that claims: the probability of volunteering decreases when the number of SUs increases. Additionally, we propose a practical negotiation algorithm in order to select efficiently one SU to access to the channel. Finally, we propose a distributed algorithm that converges to Nash equilibria and show its performance through simulation results.","Title: Backward-Shifted Coding (BSC) based on Scalable Video Coding for HAS.\nYear: 2016\n\nAbstract: The main task of HTTP Adaptive Streaming is to adapt video quality dynamically under variable network conditions. This is a key feature for multimedia delivery especially when quality of service cannot be granted network-wide and, e.g., throughput may suffer short term fluctuations. Hence, robust bitrate adaptation schemes become crucial in order to improve video quality. The objective, in this context, is to control the filling level of the playback buffer and maximize the quality of the video, while avoiding unnecessary video quality variations. In this paper we study bitrate adaptation algorithms based on Backward-Shifted Coding (BSC), a scalable video coding scheme able to greatly improve video quality. We design bitrate adaptation algorithms that balance video rate smoothness and high network capacity utilization, leveraging both on throughput-based and buffer-based adaptation mechanisms. Extensive simulations using synthetic and real-world video traffic traces show that the proposed scheme performs remarkably well even under challenging network conditions.","Title: Optimal Energy-Delay Tradeoff for Opportunistic Spectrum Access in Cognitive Radio Networks.\nYear: 2018\n\nAbstract: Cognitive radio (CR) has been considered as a promising technology to enhance spectrum efficiency via opportunistic transmission at link level. Basic CR features allow secondary users (SUs) to transmit only when the licensed channel is not occupied by primary users (PUs). However, waiting for an idle time slot may lead to large packet delays and high energy consumption. We further consider that the SU may decide, at any moment, to use another dedicated way of communication (4G) in order to transmit his packets. Thus, we consider an Opportunistic Spectrum Access (OSA) mechanism that takes into account packet delay and energy consumption. We formulate the OSA problem as a Partially Observable Markov Decision Process (POMDP) by explicitly considering the energy consumption as well as packets’ delay, which are often ignored in existing OSA solutions. Specifically, we consider a POMDP with an average reward criterion. We derive structural properties of the value function and we show the existence of optimal strategies in the class of the threshold strategies. For implementation purposes, we propose online learning mechanisms that estimate the PU activity and determine the appropriate threshold strategy on the fly. In particular, numerical illustrations validate our theoretical findings.","Title: Context-Aware Mobility Resource Allocation For Qoe-Driven Streaming Services\nYear: 2016\n\nAbstract: Streaming services come with their own challenges and technical issues that still need to be addressed for satisfying the target quality of experience (QoE) of the end-users in mobile environments. In this paper, we explore the idea of combining users' context information with the packed prefetching process features to enhance users' QoE in heterogeneous networks. More specifically, we propose a scheduling mechanism for video streaming traffic, in which the access to the network resources is restricted to users with a signal-to-noise plus interference ratio (SINR) above a given threshold. This scheme benefits from the fact that, as users are in permanent motion, they may experience different SINR values during the same video streaming session offering the opportunity to only schedule users with good channel conditions. The proposed scheduling approach (subsequently referred to as context-aware mode switching (CAMS)) not only allows to achieve overall network spectral efficiency improvement, but also guarantees fairness and QoE among users. Our simulation results show that CAMS achieves almost 1 bit per second per Hertz gain compared to the conventional scheduler (without CAMS), and up to 87% improvement in the probability of no starvations when users move at 40 kmph.","Title: On the Design of a Reward-Based Incentive Mechanism for Delay Tolerant Networks.\nYear: 2017\n\nAbstract: A central problem in Delay Tolerant Networks (DTNs) is to persuade mobile nodes to participate in relaying messages. Indeed, the delivery of a message incurs a certain number of costs for a relay. We consider a two-hop DTN in which a source node, wanting to get its message across to the destination as fast as possible, promises each relay it meets a reward. This reward is the minimum amount that offsets the expected delivery cost, as estimated by the relay from the information given by the source (number of existing copies of the message, age of these copies). A reward is given only to the relay that is the first one to deliver the message to the destination. We show that under fairly weak assumptions, the expected reward the source pays remains the same irrespective of the information it conveys, provided that the type of information does not vary dynamically over time. On the other hand, the source can gain by adapting the information it conveys to a meeting relay. For the particular cases of two relays or exponentially distributed inter-contact times, we give some structural results on the optimal adaptive policy.","Title: Predicting popularity dynamics of online contents using data filtering methods.\nYear: 2016\n\nAbstract: This paper proposes a new prediction process to explain and predicts popularity evolution of YouTube videos. We exploit prior study on the classification of YouTube videos in order to predict the evolution of videos' view-count. This classification allows to identify important factors of the observed popularity dynamics. In particular, we use this classification as filtering method allowing to identify the factors responsible for this popularity evolution. Results given by extensive experiments show that the proposed prediction process is able to reduce the average prediction errors compared to a state-of-the-art baseline model. We also evaluate the impact of adding popularity criteria in the classification.","Title: Anticipating Resource Management and QoE for Mobile Video Streaming under Imperfect Prediction\nYear: 2016\n\nAbstract: By leveraging geolocation and contextual information for mobile users, the prediction of the future throughput becomes more and more feasible. Many approaches on contextaware content delivery have been explored to balance the operators' limited resources with users' requirements. However, the perfect knowledge of the future context cannot be easily performed in real world, which represents a hurdle for most context-aware approaches. In this paper, we address a contextaware delivery algorithm for adaptive video streaming (NEWCAST) that have already been explored in [1] under perfect knowledge of future capacity, to balance the user's perception of the video and the cost of network usage. In order to make NEWCAST more resistant to eventual throughput prediction errors and adapt it to short-term horizons, we propose 4 algorithms that efficiently reduce the number of stalls by at least 75%.","Title: User Association and Resource Allocation Optimization in LTE Cellular Networks.\nYear: 2017\n\nAbstract: As the demand for higher data rates is growing exponentially, homogeneous cellular networks have been facing limitations when handling data traffic. These limitations are related to the available spectrum and the capacity of the network. Heterogeneous networks (HetNets), composed of macro cells (MCs) and small cells (SCs), are seen as the key solution to improve spectral efficiency per unit area a...","Title: A Controlled Matching Game for WLANs.\nYear: 2017\n\nAbstract: In multi-rate IEEE 802.11 WLANs, the traditional user association based on the strongest received signal and the well-known anomaly of the MAC protocol can lead to overloaded access points (APs), and poor or heterogeneous performance. Our goal is to propose an alternative game-theoretic approach for association. We model the joint resource allocation and user association as a matching game with complementarities and peer effects consisting of selfish players solely interested in their individual throughputs. Using recent game-theoretic results, we first show that various resource sharing protocols actually fall in the scope of the set of stability-inducing resource allocation schemes. The game makes an extensive use of the Nash bargaining and some of its related properties that allow controlling the incentives of the players. We show that the proposed mechanism can greatly improve the efficiency of 802.11 with heterogeneous nodes and reduce the negative impact of peer effects such as its MAC anomaly. The mechanism can be implemented as a virtual connectivity management layer to achieve efficient APs-user associations without modification of the MAC layer.","Title: Hopf Bifurcations in Replicator Dynamics with Distributed Delays.\nYear: 2017\n\nAbstract: In this paper, we study the existence and the property of the Hopf bifurcation in the two-strategy replicator dynamics with distributed delays. In evolutionary games, we assume that a strategy would take an uncertain time delay to have a consequence on the fitness (or utility) of the players. As the mean delay increases, a change in the stability of the equilibrium (Hopf bifurcation) may occur at which a periodic oscillation appears. We consider Dirac, uniform, Gamma, and discrete delay distributions, and we use the Poincare- Lindstedtu0027s perturbation method to analyze the Hopf bifurcation. Our theoretical results are corroborated with numerical simulations.","Title: Learning from Experience: A Dynamic Closed-Loop QoE Optimization for Video Adaptation and Delivery.\nYear: 2017\n\nAbstract: The quality of experience (QoE) is known to be subjective and context-dependent. Identifying and calculating the factors that affect QoE is indeed a difficult task. Recently, a lot of effort has been devoted to estimate the usersu0027 QoE in order to improve video delivery. In the literature, most of the QoE-driven optimization schemes that realize trade-offs among different quality metrics have been addressed under the assumption of homogenous populations. Nevertheless, people perceptions on a given video quality may not be the same, which makes the QoE optimization a hard task. This paper aims at taking a step further in order to address this limitation and meet usersu0027 profiles. Specifically, we propose a closed-loop control framework based on the usersu0027 (subjective) feedbacks to learn the QoE function and optimize it at the same time. Extensive simulation results show that the proposed scheme converges to a steady state, where the resulting QoE function noticeably improves the usersu0027 feedbacks.","Title: Mobility Management Schemes For Real-Time Traffic And Resource Allocation In Ieee 802.16e Mobile Network\nYear: 2017\n\nAbstract: In this paper, our aim is to develop in IEEE802.16e Wireless Networks with link adaptation, in the presence of real-time traffic call admission control (CAC) schemes. These CAC propose various scenarios of resource splitting and handling the intracell mobility. In particular, we consider two types of intracell mobility classes : low mobility class for mobiles moving usually with low speed between the neighboring regions of the cell and high mobility class for those moving with high speed and that can skip more than one region before changing their modulation. For this reason, we assume a time threshold Tth that determines the minimum time a call must remain in a region before the base station changes its modulation. And we compare it with the time that a call may spend in a region to decide whether the base station will change its modulation or not.In the beginning, we introduce two CAC schemes. In the first one, we reserve a portion of resources to mobiles in migration with both high and low mobility. And in the second one, we give the priority just to mobiles in migration with high mobility. Then, we calculate the impact in the blocking and dropping probabilities. We show by numerical results that by the proposed CAC schemes, we can find a resource management that outperforms well under different types of mobility. But, to find a good tradeoff between dropping the calls in migration and blocking the new calls, we introduce the optimization problem in the second part. So, we are faced to the necessity of optimizing the results found in the first part. Therefore, we define an objective function to optimize, in order to ensure the highest quality of service for users and to give a better stability state between the dropping and blocking probability. We show that the proposed objective function gives the optimal resources allocation between the migrating and new arriving calls in the cell. Copyright (C) 2015 John Wiley & Sons, Ltd.","Title: A routing game in networks with lossy links\nYear: 2014\n\nAbstract: Standard assumptions in the theory of routing games are that costs are additive over links and that there is flow conservation. The assumptions typically hold when the costs represent delays. We introduce here a routing game where losses occur on links in a way that may depend on the congestion. In that case both assumptions fail. We study a load balancing network and identify a Kameda type paradox in which by adding capacity, all players suffer larger loss rates.","Title: Optimal open-loop control of online advertisement\nYear: 2014\n\nAbstract: Online advertisement through social media is studied in this paper. We describe the combined effect of two mechanisms: the acceleration service offered by online media platforms and the effect of content sharing which occurs in parallel over multiple online channels. In particular, the latter operates by reinforcing the effect of acceleration and may ultimately determine the viral diffusion of the advertisement. We describe the concurrence of these two phenomena by means of a generalized Bass model, which is inspired by the dynamics which is observed in real data. Also, the source of the advertisement pays the platform owner in order to accelerate its diffusion. We hence focus on the optimal control problem that solves the trade-off between the acceleration cost and the diffusion of the advertisement. We derive the structure of the optimal control in open loop form and we discuss its structure in connection with the virality footprint of the advertisement.","Title: Evolutionary games for multiple accessl control: From Egoism to Altruism\nYear: 2014\n\nAbstract: This paper studies multiple access games with a large population of mobiles decomposed into several groups. Mobiles interfere with each others through many local interactions. We assume that each mobile (or player) cooperates with his group by taking into account the performance of his group. The degree of cooperation not only covers the fully non-cooperative behavior and the fully cooperative behavior, but also the fully altruistic behavior. To do so, we make use of the evolutionary game theory which we extend to cover this kind of behavior. We define and characterize the equilibrium (called Evolutionary Stable Strategy) for these games and establish the optimal level of cooperation that maximizes the probability of successful transmission. We also study the game dynamics both in its classical form and in the presence of delay. Interestingly, we show that, in order to maximize the system performance, the mobiles should be less cooperative.","Title: Evolutionary Games in Interacting Communities.\nYear: 2017\n\nAbstract: In this paper, we extend the evolutionary games framework by considering a population composed of communities with each having its set of strategies and payoff functions. Assuming that the interactions among the communities occur with different probabilities, we define new evolutionarily stable strategies (ESS) with different levels of stability against mutations. In particular, through the analysis of two-community two-strategy model, we derive the conditions of existence of ESSs under different levels of stability. We also study the evolutionary game dynamics both in its classic form and with delays. The delays may be strategic, i.e., associated with the strategies, spatial, i.e., associated with the communities, or spatial strategic. We apply our model to the Hawk–Dove game played in two communities with an asymmetric level of aggressiveness, and we characterize the regions of ESSs as function of the interaction probabilities and the parameters of the model. We also show through numerical examples how the delays and the game parameters affect the stability of the mixed ESS.","Title: Bitrate Adaptation In Backward-Shifted Coding For Http Adaptive Video Streaming\nYear: 2017\n\nAbstract: HTTP Adaptive Streaming is able to dynamically match video quality to variable network conditions. This is a key feature for multimedia delivery when quality of service cannot be granted network-wide. For instance, the end-to-end throughput towards mobile terminals may suffer short term fluctuations due to fading. Hence, robust bitrate adaptation schemes become crucial in order to avoid degraded video reproduction. The objective, in this context, is to control the filling level of the playback buffer, maximize the video quality, and avoid unnecessary quality variations which may also impair the perceived quality of experience. In this work we study bitrate adaptation algorithms leveraging on Backward-Shifted Coding (BSC), a scalable video coding scheme able to cope with the effects of end-to-end throughput fluctuations. We have proposed a new adaptation scheme able to balance video rate smoothness and high network capacity utilization. Both the throughput-based and buffer-based variants of such scheme have been designed. Extensive simulations using synthetic and real-world video traffic traces show that the proposed solution performs remarkably well even under challenging network conditions.","Title: Quality Of Experience In Http Adaptive Video Streaming Systems\nYear: 2017\n\nAbstract: The main task of HTTP Adaptive Streaming (HAS) is to adapt video quality dynamically under variable network conditions. This is a key feature for multimedia delivery especially when quality of service cannot be granted network-wide and, e.g., throughput may suffer short term fluctuations. Hence, robust bitrate adaptation schemes become crucial in order to improve video quality of experience. The objective, in this context, is to control the filling level of the playback buffer and maximize the quality of the video, while avoiding unnecessary video quality variations. In this paper we study bitrate adaptation algorithms based on Backward-Shifted Coding (BSC), a scalable video coding scheme able to greatly improve video quality of experience. We design bitrate adaptation algorithms that balance video rate smoothness and high network capacity utilization, leveraging both on throughput-based and buffer-based adaptation mechanisms. Extensive simulations using an MPEG/DASH client server application on ns-3 show that the proposed scheme performs remarkably well even under challenging network conditions.","Title: A Multitype Hawk And Dove Game\nYear: 2017\n\nAbstract: We consider in this paper the Hawk-Dove game in which each of infinitely many individuals, involved with pairwise encounters with other individuals, can decide whether to act aggressively (Hawk) or peacefully (Dove). Each individual is characterized by its strength. The strength distribution among the population is assumed to be fixed and not to vary in time. If both individuals involved in an interaction are Hawks, there will be a fight, the result of which will be determined by the strength of each of the individuals involved. The larger the difference between the strength of the individuals is, the larger is the cost for the weaker player involved in the fight. Our goal is to study the influence of the parameters (such as the strength level distribution) on the equilibrium of the game. We show that for some parameters there exists a threshold equilibrium policy while for other parameters there is no equilibrium policy at all.","Title: Optimal Control of Storage Regeneration with Repair Codes\nYear: 2017\n\nAbstract: High availability of containerized applications requires to perform robust storage of applications' state. Since basic replication techniques are extremely costly at scale, storage space requirements can be reduced by means of erasure and/or repairing codes. In this paper we address storage regeneration using repair codes, a robust distributed storage technique with no need to fully restore the whole state in case of failure. In fact, only the lost servers' content is replaced. To do so, new clean-slate storage units are made operational at a cost for activating new storage servers and a cost for the transfer of repair data. Our goal is to guarantee maximal availability of containers' state files by a given deadline. Upon a fault occurring at a subset of the storage servers, we aim at ensuring that they are repaired by a given deadline. We introduce a controlled fluid model and derive the optimal activation policy to replace servers under such correlated faults. The solution concept is the optimal control of regeneration via the Pontryagin minimum principle. We characterize feasibility conditions and we prove that the optimal policy is of threshold type. Numerical results describe how to apply the model for system dimensioning and show the tradeoff between activation of servers and communication cost.","Title: Bio-inspired delayed evolutionary game dynamics with networking applications\nYear: 2011\n\nAbstract: We study in this paper some evolutionary games where competition between individuals from a large population occurs through many local interactions between randomly selected individuals. We focus on games that have the property of possessing a single interior evolutionarily stable strategy (ESS). We study in particular the effect of the time delays on the convergence of evolutionary dynamics to the ESS in an evolutionary game in which each pure strategy is associated with its own delay. In particular, we study a multiple access game as well as a Hawk and Dove game. We study the properties of the ESS in these games and also the effect of time delays on the convergence of various bio-inspired evolutionary game dynamics to the ESS.","Title: Discrete and Continuous Distributed Delays in Replicator Dynamics.\nYear: 2018\n\nAbstract: In this paper, we study evolutionary games and we examine the stability of the evolutionarily stable strategy (ESS) in the continuous-time replicator dynamics with distributed time delays. In many examples, the interactions between individuals take place instantaneously, but their impacts are not immediate and take a certain amount of time which is usually random. In this paper, we study the consequences of distributed delays on the stability of the replicator dynamics. The main results are that (i) under the exponential delay distribution, the ESS is asymptotically stable for any value of the rate parameter; (ii) the necessary and sufficient conditions for the asymptotic stability of the ESS under uniform and Erlang distributions; and (iii) for the discrete distributed delays, we derive a necessary and sufficient delay-independent stability conditions. We illustrate our results with numerical examples from the Hawk–Dove game.","Title: Analysis of QoE for adaptive video streaming over wireless networks\nYear: 2018\n\nAbstract: Adaptive video streaming improves users' quality of experience (QoE), while using the network efficiently. In the last few years, adaptive video streaming has seen widespread adoption and has attracted significant research effort. We study a dynamic system of random arrivals and departures for different classes of users using the adaptive streaming industry standard DASH (Dynamic Adaptive Streaming over HTTP). Using a Markov chain based analysis, we compute the user QoE metrics: probability of starvation, prefetching delay, average video quality and switching rate. We validate our model by simulations, which show a very close match. Our study of the playout buffer is based on client adaptation scheme, which makes efficient use of the network while improving users' QoE. We prove that for buffer-based variants, the average video bit-rate matches the average channel rate. Hence, we would see quality switches whenever the average channel rate does not match the available video bit rates. We give a sufficient condition for setting the playout buffer threshold to ensure that quality switches only between adjacent quality levels.","Title: Forever Young: Aging Control For Hybrid Networks\nYear: 2019\n\nAbstract: The demand for Internet services that require frequent updates through small messages, also known as microblogging, has tremendously grown in the past few years. Although the use of such applications by domestic users is usually free, their access from mobile devices is subject to fees and consumes energy from limited batteries. If a user activates his mobile device and is in the range of a publisher, an update is received at the expense of monetary and energy costs. Thus, users face a tradeoff between such costs and their messages aging. The goal of this paper is to show how to cope with such a tradeoff, by devising aging control policies. An aging control policy consists of deciding, based on the utility of the owned content, whether to activate the mobile device, and if so, which technology to use (WiFi or cellular). We present a model that yields the optimal aging control policy. Our model is based on a Markov Decision Process (MDP) in which states correspond to content ages. Using our model, we show the existence of an optimal strategy in the class of threshold strategies, wherein users activate their mobile devices if the age of their poadcasts surpasses a given threshold and remain inactive otherwise. The accuracy of our model is validated against traces from the UMass DieselNet bus network.","Title: DCoflow: Deadline-Aware Scheduling Algorithm for Coflows in Datacenter Networks\nYear: 2022\n\nAbstract: Datacenter networks routinely support the data transfers of distributed computing frameworks in the form of coflows, i.e., sets of concurrent flows related to a common task. The vast majority of the literature has focused on the problem of scheduling coflows for completion time minimization, i.e., to maximize the average rate at which coflows are dispatched in the network fabric. Modern applications, though, may generate coflows dedicated to online services and mission-critical computing tasks which have to comply with specific completion deadlines. In this paper, we introduce DCoflow, a lightweight deadline-aware scheduler for time-critical coflows in datacenter networks. The algorithm combines an online joint admission control and scheduling logic and returns a σ-order schedule which maximizes the number of coflows that attain their deadlines. Extensive numerical results demonstrate that the proposed solution outperforms existing ones."],"string":"[\n \"Title: Information Concealing Games\\nYear: 2010\\n\\nAbstract: A system with an n-dimensional state vector and a controller and an actor is considered. The controller has complete information about the system state, and reveals a certain “minimum” amount of information to the actor. The actor takes certain actions based on the information the controller reveals, and the actions fetch certain utilities for each entity. Both the controller and actor seek to maximize their individual utilities by respectively selecting the information to reveal and the actions to adopt. This decision problem forms the basis of several technical and social systems, and can be formulated as a signaling game. It is shown that the Perfect Bayesian Equilibrium of this game has several counterintuitive properties and can be obtained as a saddle point of a different two person zero sum game. The computation time for saddle points using standard linear programs however turns out to be superexponential in n, which leads to computational intractability even for moderate n. Algorithms for computing saddle point policies using a computation time that is exponential in n are presented. Finally, simple linear time computable policies that approximate the saddle-point policies within guaranteeable approximation ratios are obtained.\",\n \"Title: Constrained cost-coupled stochastic games with independent state processes\\nYear: 2007\\n\\nAbstract: We study non-cooperative constrained stochastic games in which each player controls its own Markov chain based on its own state and actions. Interactions between players occur through their costs and constraints which depend on the state and actions of all players. We provide an example from wireless communications.\",\n \"Title: Emergence of equilibria from individual strategies in online content diffusion.\\nYear: 2013\\n\\nAbstract: Social scientists have observed that human behavior in society can often be modeled as corresponding to a threshold type policy. A new behavior would propagate by a procedure in which an individual adopts the new behavior if the fraction of his neighbors or friends having adopted the new behavior exceeds some threshold. In this paper we study the question of whether the emergence of threshold policies may be modeled as a result of some rational process which would describe the behavior of non-cooperative rational members of some social network. We focus on situations in which individuals take the decision whether to access or not some content, based on the number of views that the content has. Our analysis aims at understanding not only the behavior of individuals, but also the way in which information about the quality of a given content can be deduced from view counts when only part of the viewers that access the content are informed about its quality. In this paper we present a game formulation for the behavior of individuals using a meanfield model: the number of individuals is approximated by a continuum of atomless players and for which the Wardrop equilibrium is the solution concept. We derive conditions on the problem's parameters that result indeed in the emergence of threshold equilibria policies. But we also identify some parameters in which other structures are obtained for the equilibrium behavior of individuals.\",\n \"Title: Mean field asymptotics of Markov Decision Evolutionary Games and teams.\\nYear: 2011\\n\\nAbstract: We introduce Markov Decision Evolutionary Games with N players, in which each individual in a large population interacts with other randomly selected players. The states and actions of each player in an interaction together determine the instantaneous payoff for all involved players. They also determine the transition probabilities to move to the next state. Each individual wishes to maximize the total expected discounted payoff over an infinite horizon. We provide a rigorous derivation of the asymptotic behavior of this system as the size of the population grows to infinity. We show that under any Markov strategy, the random process consisting of one specific player and the remaining population converges weakly to a jump process driven by the solution of a system of differential equations. We characterize the solutions to the team and to the game problems at the limit of infinite population and use these to construct almost optimal strategies for the case of a finite, but large, number of players. We show that the large population asymptotic of the microscopic model is equivalent to a (macroscopic) Markov decision evolutionary game in which a local interaction is described by a single player against a population profile. We illustrate our model to derive the equations for a dynamic evolutionary Hawk and Dove game with energy level.\",\n \"Title: Online Algorithms for Adaptive Optimization in Heterogeneous Delay Tolerant Networks.\\nYear: 2013\\n\\nAbstract: We study in this paper heterogeneous delay tolerant networks formed by different classes of nodes. We assume that two hop is the forwarding strategy to deliver messages from a source node to destination node and the goal is to optimize the probability of delivering a message in the presence of different classes of mobiles, while satisfying a given energy budget. Using our model, in particular, we transform the joint energy constraint into separate constraints, one per class. This allows us to characterize the optimal strategies and provides us with a suitable framework for the design of multi-dimensional stochastic control algorithms that achieve optimal performance at runtime in spite of the lack of full information on the network state and in presence of different classes of mobiles. A thorough analysis of the convergence properties and stability of our algorithms is presented.\",\n \"Title: Dynamic spectrum allocation based on cognitive radio for QoS support\\nYear: 2010\\n\\nAbstract: Cognitive radio is an emerging technology in wireless access, aimed at vastly improving the way radio spectrum is utilized. In this paper, we investigate the performance improvement gained by applying cognitive radio to a multiple Wireless Service Providers (WSPs). We consider several WSPs and two types of flows: classical (CL) and cognitive radio (CR) flows. We propose two admission control schemes for CR flows and we show the potential improvement that could be gained by utilizing cognitive radios in the presence of classical flows. We develop several performance metrics for CL and CR flows.\",\n \"Title: Loss strategies for competing AIMD flows\\nYear: 2006\\n\\nAbstract: We study in this paper two competing AIMD flows that share a common bottleneck link. When congestion occurs, one (or both) flows will suffer a loss that will cause its throughput to decrease by a multiplicative factor. The identity of the flow that will suffer a loss is determined by a randomized \\\"loss strategy\\\" that may depend on the throughputs of the flows at the congestion instant. We analyze several loss strategies: the one in which the identity of the flow experiencing the loss is independent of the current throughput and the one in which the flow with the largest throughput is to suffer the loss; this is compared with the strategy that assigns loss probabilities proportionally to the throughputs (thus a flow with a larger throughput has a larger loss probability). After deriving some results for the general asymmetric case, we focus in particular on the symmetric case and study the influence of the strategy on the average throughput and average utilization of the link. As the intuition says, a strategy that assigns a loss to a flow with a higher throughput is expected to give worse performance since the total instantaneous throughput after a loss is expected to be lower with such a strategy. Surprisingly, we show that this is not the case. We show that the average throughput and average link utilizations are invariant: they are the same under any possible strategy; the link utilization is 6/7 of the link capacity. We show, in contrast, that the second moment of the throughput does depend on the strategy.\",\n \"Title: Modeling slotted aloha as a stochastic game with random discrete power selection algorithms\\nYear: 2009\\n\\nAbstract: We consider the uplink case of a cellular system where m bufferless mobiles transmit over a common channel to a base station, using the slotted aloha medium access protocol. We study the performance of this system under several power differentiation schemes. Indeed, we consider a random set of selectable transmission powers and further study the impact of priorities given either to new arrival packets or to the backlogged ones. Later, we address a general capture model where a mobile transmits successfully a packet if its instantaneous SINR (signal to interferences plus noise ratio) is lager than some fixed threshold. Under this capture model, we analyze both the cooperative team in which a common goal is jointly optimized as well as the noncooperative game problem where mobiles reach to optimize their own objectives. Furthermore, we derive the throughput and the expected delay and use them as the objectives to optimize and provide a stability analysis as alternative study. Exhaustive performance evaluations were carried out, we show that schemes with power differentiation improve significantly the individual as well as global performances, and could eliminate in some cases the bi-stable nature of slotted aloha.\",\n \"Title: Opportunistic Scheduling in Cellular Systems in the Presence of Noncooperative Mobiles\\nYear: 2012\\n\\nAbstract: A central scheduling problem in wireless communications is that of allocating resources to one of many mobile stations that have a common radio channel. Much attention has been given to the design of efficient and fair scheduling schemes that are centrally controlled by a base station (BS) whose decisions depend on the channel conditions reported by each mobile. The BS is the only entity taking decisions in this framework. The decisions are based on the reports of mobiles on their radio channel conditions. In this paper, we study the scheduling problem from a game-theoretic perspective in which some of the mobiles may be noncooperative or strategic, and may not necessarily report their true channel conditions. We model this situation as a signaling game and study its equilibria. We demonstrate that the only Perfect Bayesian Equilibria (PBE) of the signaling game are of the babbling type: the noncooperative mobiles send signals independent of their channel states, the BS simply ignores them, and allocates channels based only on the prior information on the channel statistics. We then propose various approaches to enforce truthful signaling of the radio channel conditions: a pricing approach, an approach based on some knowledge of the mobiles' policies, and an approach that replaces this knowledge by a stochastic approximations approach that combines estimation and control. We further identify other equilibria that involve non-truthful signaling.\",\n \"Title: Non-cooperative routing in loss networks\\nYear: 2002\\n\\nAbstract: The paper studies routing in loss networks in the framework of a non-cooperative game with selfish users. Two solution concepts are considered: the Nash equilibrium, corresponding to the case of a finite number of agents (such as service providers) that take routing decisions, and the Wardrop equilibrium, in which routing decisions are taken by a very large number of individual users. We show that these equilibria do not fall into the standard frameworks of non-cooperative routing games. As a result, we show that uniqueness of equilibria or even of utilizations at equilibria may fail even in the case of simple topology of parallel links. However, we show that some of the problems disappear in the case in which the bandwidth required by all connections is the same. For the special case of a parallel link topology, we obtain some surprisingly simple way of solving the equilibrium for both cases of Wardrop as well as Nash equilibrium.\",\n \"Title: Stackelberg approach for pricing differentiated services\\nYear: 2007\\n\\nAbstract: We consider in this paper a set of connections sharing a common bottleneck link. We assume first that we have data transfer using TCP congestion control protocol and real-time traffic using some TCP-friendly transport protocol that satisfies the same square-root formula for throughput. The performance measures are determined according to the operational parameters of a RED buffer management. Those parameters are assumed to be able to give differentiated services to the applications according to their choice of service class. In terms of loss probabilities and of throughputs, we consider a best effort type of service differentiation where the QoS of connections is not guaranteed, but by choosing a better (more expensive) service class, the QoS parameters of a session can be improved (as long as that session is the only one to change its service class). We assume however, that the system is dimensioned so as to satisfy some average delay requirement. The choice of a service class of an application will depend both on the utility as well as on the cost it has to pay. We first study the performance of each traffic source as a function of the connections' parameters and the pricing policy of the network. We then study the Stackelberg equilibrium, i.e. the service provider's problem of how to choose the pricing so as to maximize its utility, taking into account the reaction of the users.\",\n \"Title: Hierarchy sustains partial cooperation and induces a Braess-like paradox in slotted aloha-based networks\\nYear: 2012\\n\\nAbstract: This paper studies a hierarchical distributed choice of retransmission probabilities in slotted aloha. In particular, we consider a wireless system composed of one central receiver and several selfish mobile users communicating via the slotted aloha protocol. The set of mobile users is split into two classes: leaders and followers. We then study the induced non-cooperative hierarchical game based on the Stackelberg equilibrium concept. Using a 4D Markovian model, we compute the steady state of the system and derive the average throughput and the expected delay as well. We start by discussing the protocol design and propose a controlled slotted aloha using a virtual controller. The virtual controller can sustain partial cooperation among concurrent mobile users when accessing the channel by making the channel lossy. This leads us to identify a Braess-like paradox in which reducing capacity to the system may improve the performance of all mobile users. We then investigate the impact of hierarchy among mobile users in such a random access protocol and discuss how to distribute leader/follower roles. We show that the global performance of the system is improved compared to standard slotted aloha system. However, slight performances slow-down may be observed for the followers group when the total number of mobile users is relatively small.\",\n \"Title: Evolutionary power control games in wireless networks\\nYear: 2008\\n\\nAbstract: In this paper, we apply evolutionary games to non-cooperative power control in wireless networks. Specifically, we focus our study in a power control in W-CDMA and WIMAX wireless systems. We study competitive power control within a large population of mobiles that interfere with each other through many local interactions. Each local interaction involves a random number of mobiles. An utility function is introduced as the difference between a utility function based on SIR of the mobile and pricing. The games are not necessarily reciprocal as the set of mobiles causing interference to a given mobile may differ from the set of those suffering from its interference. We show how the evolution dynamics and the equilibrium behavior (called Evolutionary Stable Strategy - ESS) are influenced by the characteristics of the wireless channel and pricing characteristics.\",\n \"Title: From Altruism to Non-Cooperation in Routing Games\\nYear: 2008\\n\\nAbstract: Abstract—The paper studies the routing in the network shared by several users. Each user seeks to optimize either its own,performance,or some,combination,between,its own performance and that of other users, by controlling the routing,of its given flow demand.,We parameterize the degree of cooperation,which,allows to cover the fully non-cooperative behavior, the fully cooperative behavior, and even more, the fully altruistic behavior, all these as special cases of the parameter’s,choice. A large part of the work,consists in exploring,the impact,of the degree of cooperation,on the equilibrium. Our first finding is to identify multiple Nash equilibria with cooperative behavior that do not occur in the non-cooperative,case under,the same conditions (cost, demand and topology). We then identify Braess like paradox,(in which,adding,capacity or adding a link to a network,results in worse performance,to all users) and study the impact of the degree of cooperation on it. We identify situations where,low cooperation,results in poorer performance,at equilibrium. We then pursue,the exploration,and carry it on to the setting of Stackelberg equilibrium. We finally obtain some theoretical results that show,that for low degree,of cooperation,the equilibrium is unique, confirming the results of our numerical study.\",\n \"Title: Study Of Connectivity In Vehicular Ad Hoc Networks\\nYear: 2007\\n\\nAbstract: We investigate connectivity in ad hoc network formed between vehicles that move in the same direction on a typical highway. We use the common model in vehicular traffic theory in which a fixed point on the highway sees cars passing it separated. by times with exponentially distributed duration. We obtain the distribution of the distances between cars, which allows us to use techniques from queuing theory for studying connectivity. We obtain explicit expressions for the expected connectivity distance as well as the probability distribution and expectation of the number of cars in a platoon. The analytical model we present is able to describe the effects of various system's parameters, including road traffic parameters (i.e. speed distribution and traffic flow) and transmission range of vehicles, on the connectivity. We further provide bounds obtained using stochastic ordering techniques. Our approach is based on the work of Morandi and Altman [13] that transformed the problem of connectivity distance distribution into that of the distribution of the busy period of an equivalent infinite server queue. We use our analytical results along with publicly available road traffic statistical data to understand connectivity in VANETs.\",\n \"Title: Qos Differentiation For Initial And Bandwidth Request Ranging In Ieee802.16\\nYear: 2008\\n\\nAbstract: In this paper, we propose and analyse a new scheme to handle ranging requests in IEEE802.16e networks based on differentiated backoff parameters and a novel hierarchical code partitioning principle. Our analysis is of the form of a fixed-point equation which characterizes the system operating points and which allows us to compute the attempt rate of each QoS class as well as the distribution of the number of ranging bandwidth requests received successfully by the base station as well as their average delay. Both analysis and simulation confirm that our scheme yields high throughput and low collision probability for real-time connections without altering the collision probability for non-real-time connections.\",\n \"Title: Avoiding paradoxes in multi-agent competitive routing\\nYear: 2003\\n\\nAbstract: Strange behavior may occur in networks due to the non-cooperative nature of decision making, when the latter are taken by individual agents. In particular, the well known Braess paradox illustrates that when upgrading a network by adding a link, the resulting equilibrium may exhibit larger delays for all users. We present here some guidelines to avoid the Braess paradox when upgrading a network. We furthermore present conditions for the delays to be monotone increasing in the total demand.\",\n \"Title: Analysis of Buffer Starvation With Application to Objective QoE Optimization of Streaming Services\\nYear: 2014\\n\\nAbstract: Our purpose in this paper is to characterize buffer starvations for streaming services. The buffer is modeled as a FIFO queue with exponential service time and Poisson arrivals. When the buffer is empty, the service restarts after a certain amount of packets are prefetched. With this goal, we propose two approaches to obtain exact distribution of the number of buffer starvations, one of which is based on Ballot theorem, and the other uses recursive equations. The Ballot theorem approach gives an explicit result. We extend this approach to the scenario with a constant playback rate using Tàkacs Ballot theorem. The recursive approach, though not offering an explicit result, allows us to obtain the distribution of starvations with non-independent and identically distributed (i.i.d.) arrival process in which an ON/OFF bursty arrival process is considered. We further compute the starvation probability as a function of the amount of prefetched packets for a large number of files via a fluid analysis. Among many potential applications of starvation analysis, we show how to apply it to optimize objective quality of experience (QoE) of media streaming, by exploiting the tradeoff between startup/rebuffering delay and starvations.\",\n \"Title: Dynamic spectrum allocation with admission control based on cognitive radio for QoS support in multiple wireless network.\\nYear: 2012\\n\\nAbstract: Cognitive radio is an emerging technology in wireless access, aimed at vastly improving the way radio spectrum is utilized. In this article, we investigate the performance improvement gained by applying cognitive radio to a multiple wireless service providers (WSPs). We consider several WSPs and two types of users: primary (licensed) and secondary (unlicensed) users. Two different schemes are proposed for unlicensed users to manage their handoff. These two schemes are different in the sense that one allows to each WSP to give more priority to his own unlicensed users by dropping low priority users, while the other does not allow that. The system is modeled by a Markov process, with continuous time and finite state space. Due to the very large number of states, we propose a robust method to approximate the stationary probability distribution vector of Markov process. Based on this approximation, we develop several performance metrics, blocking and dropping probabilities for both kinds of unlicensed users. Numerical results show that this approximation is very close to the exact solution. Finally, we show that spectrum utilization of cognitive users increases with the proposed schemes especially in medium and high traffic.\",\n \"Title: Asymptotic delay analysis and timeout-based admission control for ad hoc wireless networks with asymmetric users\\nYear: 2010\\n\\nAbstract: In this paper, we present an analytical model for an approximate calculation of the end-to-end delay performance in multi-hop wireless ad hoc networks. In contrast to literature that largely focuses on average delay, our paper focuses on the distribution of end-to-end delay. In this paper, we assume that each source injects packets in the network, which traverse intermediate nodes until they reach the destination. Firstly, we employ discrete-time queueing theory to derive the expressions for the queue length and the delay in terms of probability generating functions. Secondly, in order to improve the control routing and transmission scheduling, we adopt a cross-layer design that allows information sharing across different layers for efficient utilization of network resources, and meeting the end-to-end performance requirements of demanding applications. Thirdly, we propose a packet admission control scheme based on delay timeout mechanism. This guarantees quality of service for multimedia applications such as voice and video streaming. Finally, we conduct extensive simulations in order to verify our analytical results.\",\n \"Title: Battery State-Dependent Access Control in Solar-Powered Broadband Wireless Networks\\nYear: 2008\\n\\nAbstract: This paper studies both power control and multiple access control in solar-powered broadband wireless networks. We assume that the mobiles use power storage element, such as rechargeable Solar- powered batteries, in order to have energy available for later use. By modeling the energy-level of Solar-powered batteries as a stochastic process, we study noncooperative interactions within large population of mobiles that interfere with each other through many local interactions. Each local interaction involves a random number of mobiles. The actions taken by a mobile determine not only the immediate payoff but also the state transition probabilities of its battery. We define and characterize the evolutionary Stable Strategies (ESS) of the stochastic evolutionary game.\",\n \"Title: Pricing differentiated services: a game-theoretic approach\\nYear: 2006\\n\\nAbstract: The goal of this paper is to study pricing of differentiated services and its impact on the choice of service priority at equilibrium. We consider both TCP connections as well as noncontrolled (real-time) connections. The performance measures (such as throughput and loss rates) are determined according to the operational parameters of a RED (Random Early Discard) buffer management. The latter is assumed to be able to give differentiated services to the applications according to their choice of service class. We consider a service differentiation for both TCP as well as real-time traffic where the quality of service (QoS) of connections is not guaranteed, but by choosing a better (more expensive) service class, the QoS parameters of a session can improve (as long as the service class of other sessions are fixed). The choice of a service class of an application will depend both on the utility as well as on the cost it has to pay. We first study the performance of the system as a function of the connections' parameters and their choice of service classes. We then study the decision problem of how to choose the service classes. We model the problem as a noncooperative game. We establish conditions for an equilibrium to exist and to be uniquely defined. We further provide conditions for convergence to equilibrium from nonequilibria initial states. We finally study the pricing problem of how to choose prices so that the resulting equilibrium would maximize the network benefit.\",\n \"Title: Modelling View-count Dynamics in YouTube.\\nYear: 2014\\n\\nAbstract: The goal of this paper is to study the behaviour of view-count in YouTube. We first propose several bio-inspired models for the evolution of the view-count of YouTube videos. We show, using a large set of empirical data, that the view-count for 90% of videos in YouTube can indeed be associated to at least one of these models, with a Mean Error which does not exceed 5%. We derive automatic ways of classifying the view-count curve into one of these models and of extracting the most suitable parameters of the model. We study empirically the impact of videos' popularity and category on the evolution of its view-count. We finally use the above classification along with the automatic parameters extraction in order to predict the evolution of videos' view-count.\",\n \"Title: La théorie des jeux non-coopératifs appliquée aux réseaux de télécommunication\\nYear: 2007\\n\\nAbstract: In this paper, we present various tools originating from non-cooperative Game Theory, which allow us to study competition situations in telecommunication networks. We describe the mathematical tools while providing examples from a variety of areas in telecommunication networks.\",\n \"Title: Stability-Throughput tradeoff and routing in multi-hop wireless ad-hoc networks\\nYear: 2006\\n\\nAbstract: We study the throughput of multi-hop routes and stability of forwarding queues in a wireless Ad-Hoc network with random access channel. We focuse on wireless with stationary nodes, such as community wireless networks. Our main result is charactrerization of stability condition and the end-to-end throughput using the balance. We also investigate the impact of routing on end-to-end throughput and stability of intermediate nodes. We find that i) as long as the intermediate queues in the network are stable, the end-to-end throughput of a connection does not depend on the load on the intermediate nodes, ii) we showed that if the weight of a link originating from a node is set to the number of neighbors of this node, then shortest path routing maximizes the minimum probability of end-to-end packet delivery in a network of weighted fair queues with coupled servers. Numerical results are given and support the results of the analysis.\",\n \"Title: Distributed energy-delay framework for opportunistic spectrum access\\nYear: 2011\\n\\nAbstract: In this paper, we consider the problem of spectrum sharing among one primary and multiple secondary users. Particularly, we study the problem of non-cooperative spectrum sensing and access. Secondary users have to sense the channel before accessing it. There is a tradeoff between large packet delay, partially due to collisions between secondary users, and high energy consumption for sensing. To overcome this problem, we consider an opportunistic spectrum access in cognitive radio taking into account packet delay and energy consumption. Hence, there is an associated spectrum occupancy and packet delay (the time elapsed since the reception) state for each secondary user. Due to partial sensing, a Partial Observable Markov Decision Process (POMDP) is used to determine an equilibrium policy. The equilibrium is obtained by using a linear program. We identify a paradox in this cognitive radio context: when the channel is more occupied by the primary user, this may improve the spectrum utilization by secondary users. We also investigate the impact of transmission cost on the equilibrium and show numerically how this cost can be used to enhance the utilization of the spectrum.\",\n \"Title: A Queuing Analysis of Packet Dropping over a Wireless Link with Retransmissions\\nYear: 2003\\n\\nAbstract: One of the most important performance measure for real time applications is the packet loss probability. In a wireless environment losses are first due to noisy channels (fading, shadowing etc). In order to improve the radio link, one often retransmits packets that have not been well received (using the Automatic Retransmission reQuest-ARQ). This however may lead to queuing phenomena due to retransmissions and to losses of packets due to buffer overflow. We present a queuing analysis in order to compute the combined effect of the noisy channel, the retransmissions and the buffer overflow. We use advanced spectral methods of quasi-birth and death process in order to analyze the system for general multistate Markov channels and provide an approximation approach based on the theory of singular perturbation.\",\n \"Title: The evolution of transport protocols: An evolutionary game perspective\\nYear: 2009\\n\\nAbstract: Today's Internet is well adapted to the evolution of protocols at various network layers. Much of the intelligence of congestion control is delegated to the end users and they have a large amount of freedom in the choice of the protocols they use. In the absence of a centralized policy for a global deployment of a unique protocol to perform a given task, the Internet experiences a competitive evolution between various versions of protocols. The evolution manifests itself through the upgrading of existing protocols, abandonment of some protocols and appearance of new ones. We highlight in this paper the modeling capabilities of the evolutionary game paradigm for explaining past evolution and predicting the future one. In particular, using this paradigm we derive conditions under which (i) a successful protocol would dominate and wipe away other protocols, or (ii) various competing protocols could coexist. In the latter case we also predict the share of users that would use each of the protocols. We further use evolutionary games to propose guidelines for upgrading protocols in order to achieve desirable stability behavior of the system.\",\n \"Title: Dynamic retransmission limit scheme in MAC layer for routing in multihop ad hoc networks\\nYear: 2008\\n\\nAbstract: We consider a wireless ad hoc network with random access channel. We present a model that takes into account topology, routing, random access in MAC layer, and forwarding probability. In this paper, we focus on drawing benefit from the interaction of the MAC (governed by IEEE 802.11 or slotted Aloha) and routing by defining a new cross-layer scheme for routing based on the limit number of retransmission. By adjusting dynamically and judiciously this parameter in a saturated network, we have realized that both stability of forwarding queues and average throughput are significantly improved in linear networks with symmetric traffic: a gain of 100% can be reached. While in asymmetric topology network with asymmetric traffic, we achieve a better average delay (resp., throughput) for each connection without changing the average throughput (resp., delay). In addition, we show the efficiency of our new scheme in case of multimedia applications with delay constraint. A detailed performance study is presented using analytical and simulation evaluation.\",\n \"Title: Evolutionary forwarding games in delay tolerant networks: Equilibria, mechanism design and stochastic approximation\\nYear: 2013\\n\\nAbstract: In this paper, we apply evolutionary games to non-cooperative forwarding control in Delay Tolerant Networks (DTNs). The main focus is on mechanisms to rule the participation of the relays to the delivery of messages in DTNs. Thus, we express the success probability as a function of the competition that takes place within a large population of mobiles, and we characterize the effect of reward-based mechanisms on the performance of such systems. Devices acting as active relays, in fact, sacrifice part of their batteries in order to support message replication and thus increase the probability to reach the destination. In our scheme, a relay can choose the strategy by which they participate to the message relaying. A mobile that participates receives a unit of reward based on the reward mechanism selected by the network. A utility function is introduced as the difference between the expected reward and the energy cost, i.e., the cost spent by the relay to sustain forwarding operations. We show how the evolution dynamics and the equilibrium behavior (called Evolutionary Stable Strategy - ESS) are influenced by the characteristics of inter contact time, energy expenditure and pricing characteristics. We extend our analysis to mechanisms that the system can introduce in order to have the message delivered to the destination with high probability within a given deadline and under energy constraints which bound the number of released copies per message. Finally, we apply our findings in order to devise decentralized forwarding algorithms that are rooted in the theory of stochastic approximations. Thus, we demonstrate that the ESS can be attained without complete knowledge of the system state and letting the source monitor number of released copies per message only. We provide extensive numerical results to validate the proposed scheme.\",\n \"Title: Introducing hierarchy in energy-efficient power control games.\\nYear: 2008\\n\\nAbstract: We consider a multiple access channel where the users choose their best power control strategy in order to selfishly maximize their energy-efficiency. To increase the utilities with respect to the classical non-cooperative game, we introduce hierarchy in two ways. On the one hand, assuming single-user decoding at the receiver, we investigate a Stackelberg formulation of the game where one user is the leader. On the other hand, assuming neither leader nor followers among the users, we introduce hierarchy by using successive interference cancellation at the receiver. For both cases, we study the existence and uniqueness of an equilibrium and compare the individual performance obtained in the hierarchical game with that obtained in the non-cooperative game. An exhaustive comparative analysis of the two games is also conducted. In order to optimize the choice of the leader in the Stackelberg formulation (with single user decoding) and that of the decoding order (in the non-cooperative game with successive interference cancelation), we study two measures of global energy-efficiency for the network and discussions are provided for each case.\",\n \"Title: A hierarchical slotted aloha game.\\nYear: 2009\\n\\nAbstract: We consider in this paper a wireless system composed of one central receiver and several selfish transmitters communicating via the slotted aloha protocol. We study a noncooperative hierarchical system based on the Stackelberg game concept. Each user of our game tends to maximize his own throughput or minimize his expected delay of backlogged packets depending on his transmission probability and transmission probabilities of other users in the network. Using a 4D Markovian model, we compute the steady state of the system and derive the average throughput and expected delay. We then investigate the impact of introducing hierarchy in the random access protocol. Later, exhaustive performance evaluations were carried out, we show that the global performance of the system is improved compared to slotted aloha system without hierarchy. However, a not sharp performances slow-down may be observed for the followers when the number of users is not large, but they turn to outperform Nash equilibrium under average and heavy load.\",\n \"Title: Asymmetric delay in evolutionary games\\nYear: 2007\\n\\nAbstract: We study the effect of the time delays on the convergence of replicator dynamics to the evolutionary stable strategies in a symmetric evolutionary game in which each pure strategy is associated with its own delay. We compare the stability region to that obtained when all strategies have the same delay. Multiple Access Game and Hawk and Dove Game are studied.\",\n \"Title: Delay Analysis For Real-Time Streaming Media In Multi-Hop Ad Hoc Networks\\nYear: 2008\\n\\nAbstract: In this paper, we investigate an important issue for real-time multimedia over multi-hop ad-hoc networks on different layers. Such application requires receiver playback buffers to smooth network delay variation and reconstruct the periodic nature of the transmitted packets. Packets arriving after their schedule deadline are considered late and are not played out. This requires that the network is able to offer quality of service appropriate for the delay bounds of the real-time application constrains. Our primary contribution concerns the study of the end to end delay. Based on the latter results, we approximate the loss rate of a real-time traffic which requires a delay bounds constraints. Then, we propose a cross layer scheme from the network layer to the MAC layer to support real-time traffic. In this scheme, we reduce the loss rate by decreasing the packets that arrive after their schedule deadline. Through numerical and simulation results, we demonstrate the utility and efficiency of our approach.\",\n \"Title: Evolutionary games in wireless networks.\\nYear: 2010\\n\\nAbstract: We consider a noncooperative interaction among a large population of mobiles that interfere with each other through many local interactions. The first objective of this paper is to extend the evolutionary game framework to allow an arbitrary number of mobiles that are involved in a local interaction. We allow for interactions between mobiles that are not necessarily reciprocal. We study 1) multiple-access control in a slotted Aloha-based wireless network and 2) power control in wideband code-division multiple-access wireless networks. We define and characterize the equilibrium (called evolutionarily stable strategy) for these games and study the influence of wireless channels and pricing on the evolution of dynamics and the equilibrium.\",\n \"Title: QoE analysis of media streaming in wireless data networks\\nYear: 2012\\n\\nAbstract: The purpose of this paper is to model quality of experience (QoE) of media streaming service in a shared fast-fading channel. In this context, the arrival and the service processes of the playout buffer do not have the same job size. We present an analytical framework based on Takács Ballot theorem to compute the probability of buffer starvation and the distribution of playback intervals. We model the arrival processes of Proportional Fair and Round Robin schedulers, and feed them into this framework to study the impact of prefetching on the starvation behavior. Our simulations match the developed model very well if the base station knows the playback rate and the channel gain. Furthermore, we make an important observation that QoE metrics predicted by users are very sensitive to the measurement error of arrival process.\",\n \"Title: TCP Optimization through FEC, ARQ, and Transmission Power Tradeoffs\\nYear: 2004\\n\\nAbstract: TCP performance degrades when end-to-end connections extend over wireless connections - links which are characterized by high bit error rate and intermittent connectivity. Such link characteristics can significantly degrade TCP performance as the TCP sender assumes wireless losses to be congestion losses resulting in unnecessary congestion control actions. Link errors can be reduced by increasing transmission power, code redundancy (FEC) or number of retransmissions (ARQ). But increasing power costs resources, increasing code redundancy reduces available channel bandwidth and increasing persistency increases end-to-end delay. The paper proposes a TCP optimization through proper tuning of power management, FEC and ARQ in wireless environments (WLAN and WWAN). In particular, we conduct analytical and numerical analysis taking into account the three aforementioned factors, and evaluate TCP (and \\\"wireless-aware\\\" TCP) performance under different settings. Our results show that increasing power, redundancy and/or retransmission levels always improves TCP performance by reducing link-layer losses. However, such improvements are often associated with cost and arbitrary improvement cannot be realized without paying a lot in return. It is therefore important to consider some kind of net utility function that should be optimized, thus maximizing throughput at the least possible cost.\",\n \"Title: Multiple access game in ad-hoc network\\nYear: 2007\\n\\nAbstract: We study a Modified Multiple Access Game (MMAG) in two approach: repeated game approach and evolutionary game approach. We compute Nash equilibria and Evolutionary Stable Strategy (ESS) in the two cases. We study the delay impact on the performance of the evolutionary MMAG describing competition between mobile terminals over the access to a channel. We discuss about the convergence to the ESS in replicator dynamics and imitate better dynamics with delays.\",\n \"Title: Impact of flow-level dynamics on QoE of video streaming in wireless networks\\nYear: 2013\\n\\nAbstract: The Quality of Experience (QoE) of streaming service is often degraded by frequent playback interruptions. To mitigate the interruptions, the media player prefetches streaming contents before starting playback, at a cost of delay. We study the QoE of streaming from the perspective of flow dynamics. First, a framework is developed for QoE when streaming users join the network randomly and leave after downloading completion. We compute the distribution of prefetching delay using partial differential equations (PDEs), and the probability generating function of playout buffer starvations using ordinary differential equations (ODEs). Second, we extend our framework to characterize the throughput variation caused by opportunistic scheduling at the base station in the presence of fast fading. Our study reveals that the flow dynamics is the fundamental reason of playback starvation. The QoE of streaming service is dominated by the average throughput of opportunistic scheduling, while the variance of throughput has very limited impact on starvation behavior.\",\n \"Title: A two-stage game theoretic approach for self-organizing networks.\\nYear: 2013\\n\\nAbstract: Growth of network access technologies in the mobile environment has raised several new issues due to the interference between the available access. Thanks to the currently used access methods such as the orthogonal frequency division multiple access in mobile networks and the long term evolution-advanced systems, the intra-cell interferences are avoided and the quality of service has increased. Nevertheless, the diversity and multiplicity of base stations in the network have left behind a major problem of inter-cell interferences. In this paper, we focus on the optimization of the total throughput of cellular networks using fractional frequency reuse and allowing each mobile user to individually choose its serving base station. We derive analytically the utilities related to the network manager and mobile users and develop a Stackelberg game to obtain the equilibrium. We propose a distributed algorithm that allows the base stations, using a light collaboration, to achieve an efficient utilization of the frequencies, with the optic of maximizing the total system utility. This algorithm is based on stochastic gradient descent which requires some information to be exchanged between neighboring base stations. At user association level, we propose an iterative distributed algorithm based on automata learning algorithm. Both algorithms allow the system to converge to the Stackelberg equilibrium. Furthermore, simulation results carried out based on a realistic network setting show promising results in terms of global utility and convergence issues. In this setting, we include scenarios with a varying number of users and address the problem of robustness and scalability of the proposed approach.\",\n \"Title: An end-to-end QoS framework for IEEE 802.16 and ad-hoc integrated networks\\nYear: 2009\\n\\nAbstract: Current mobile users are often equipped with several network interfaces, which may be of different access technologies. Each access technology has specific characteristics in terms of coverage area and technical characteristics (bandwidth, QoS, etc.) and provides diverse commercial opportunities for the operators. It seems likely that these various technologies have to coexist and, from then, solutions of integration and interoperability will be necessary to deal with the technological diversity. We consider in this paper the WiMAX and ad-hoc integrated networks. A user who needs to establish a high reliable service prefers to use WiMAX network and another who needs to transmit Best Effort traffic with high rate but delay tolerant prefers to connect ad-hoc network that is assumed to have access to Internet. We aim through this combination to locally extend the coverage of WiMAX cell and to study the stability of involved nodes, in particular, gateway nodes. Under stability condition, our main result is characterization of the end-to-end throughput and delay using the rate balance equation and a G/G/1 queueing model. Through numerical results, we demonstrate the utility and efficiency of our approach.\",\n \"Title: Fractional frequency reuse stackelberg model for self-organizing networks\\nYear: 2011\\n\\nAbstract: Growth of network access technologies in the mobile environment has raised several new issues due to the interference between the available accesses. Thanks to the currently used access methods such as OFDMA in mobile networks, and LTE systems, the intra-cell interferences are avoided and the quality of service has increased. Nevertheless, the diversity and multiplicity of base stations in the network has left behind, a major problem of inter-cell interferences. In this paper, we focus on the optimization of the total throughput of cellular networks using fractional frequency reuse and allowing each mobile user to individually choose its serving base station. We derive analytically the utilities related to the network manager and mobile users and develop a Stackelberg game to obtain the equilibria. We propose a distributed algorithm that allows the base stations, using a light collaboration, to achieve an efficient utilization of the frequencies, with the optic of maximizing the total system utility. This algorithm is based on stochastic gradient descent which requires some information to be exchanged between neighboring base stations. At user association level, we propose an iterative distributed algorithm based on automata learning algorithm. Both algorithms allow the system to converge to the Stackelberg equilibrium. Furthermore, simulation results carried out based on a realistic network setting show promising results in terms of global utility and convergence issues.\",\n \"Title: Loss Strategies for Competing TCP/IP Connections\\nYear: 2004\\n\\nAbstract: We study in this paper two competing TCP connections that share a common bottleneck link. When congestion occurs, one (or both) connections will suffer a loss that will cause its throughput to decrease by a multiplicative factor. The identity of the connection that will suffer a loss is determined by a randomized \\\"loss strategy\\\" that may depend on the throughputs of the connections at the congestion instant. We analyze several such loss strategies. After deriving some results for the general asymmetric case, we focus in particular on the symmetric case and study the influence of the strategy on the average throughput and average utilization of the link. As the intuition says, a strategy that assigns a loss to a connection with a higher throughput is expected to give worse performance since the total instantaneous throughput after a loss is expected to be lower with such a strategy. We show that, surprisingly, the average throughput and average link utilizations are invariant: they are the same under any possible strategy; the link utilization is 6/7 of the link capacity. We show, in contrast, that the second moment of the throughput does depend on the strategy.\",\n \"Title: Slotted Aloha as a game with partial information\\nYear: 2004\\n\\nAbstract: This paper studies distributed choice of retransmission probabilities in slotted ALOHA. Both the cooperative team problem as well as the noncooperative game problem are considered. Unlike some previous work, we assume that mobiles do not know the number of backlogged packets at other nodes. A Markov chain analysis is used to obtain optimal and equilibrium retransmission probabilities and throughput. We then investigate the impact of adding retransmission costs (which may represent the disutility for power consumption) on the equilibrium and show how this pricing can be used to make the equilibrium throughput coincide with the optimal team throughput.\",\n \"Title: Delay Tolerant Networks in Partially Overlapped Networks: A Non-cooperative Game Approach.\\nYear: 2009\\n\\nAbstract: Epidemic forwarding protocol in Delay Tolerant Networks maximizes successful data delivery probability but at the same time incurs high costs in terms of redundancy of packet copies in the system and energy consumption. Two-hop routing on the other hand minimizes the packet flooding and the energy costs but degrades the delivery probability. This paper presents a framework to achieve a tradeoff between the successful data delivery probability and the energy costs. Each mobile has to decide which routing protocol it wants to use for packet delivering. In such a problem, we consider a non-cooperative game theory approach. We explore the scenario where the source and the destination mobiles are enclosed in two different regions, which are partially overlapped. We study the impact of the proportion of the surface covered by both regions on the Nash equilibrium and price of anarchy. We also design a fully distributed algorithm that can be employed for convergence to the Nash equilibrium. This algorithm does not require any knowledge of some parameter of the system as the number of mobiles or the rate of contacts between mobiles.\",\n \"Title: A Stackelberg Model for Opportunistic Sensing in Cognitive Radio Networks.\\nYear: 2013\\n\\nAbstract: We consider a non-cooperative Dynamic Spectrum Access (DSA) game where Secondary Users (SUs) access opportunistically the spectrum licensed for Primary Users (PUs). As SUs spend energy for sensing licensed channels, they may choose to be inactive during a given time slot in order to save energy. Then, there exists a tradeoff between large packet delay, partially due to collisions between SUs, and high-energy consumption spent for sensing the occupation of licensed channels. To overcome this problem, we take into account packet delay and energy consumption into our framework. Due to the partial spectrum sensing, we use a Partial Observable Stochastic Game (POSG) formalism, and we analyze the existence and some properties of the Nash equilibrium using a Linear Program (LP). We identify a paradox: when licensed channels are more occupied by PUs, this may improve the spectrum utilization by SUs. Based on this observation, we propose a Stackelberg formulation of our problem where the network manager may increase the occupation of licensed channels in order to improve the SUs' average throughput. We prove the existence of a Stackelberg equilibrium and we provide some simulations that validate our theoretical findings.\",\n \"Title: Modeling the effect of forwarding in a multi-hop ad hoc networks with weighted fair queueing\\nYear: 2007\\n\\nAbstract: Consider a wireless ad hoc network with random access channel. We present a model that takes into account topology, routing, random access in MAC layer (governed by IEEE 802.11orslotted aloha) and forwarding probability. In this paper, we are focusing to study the effect of cooperation on the stability and throughput of ad-hoc network. Forwarding packets of other nodes is an example of activity that requires such a collaboration. Hence, it may not be in interest of a node to always forward the requesting packet. We propose a new approach (based on cycle of transmissions) to derive throughput of multi-hop routes and stability of forwarding queues. With this cycle approach, we correct the analytical expressions derived in [2] and discover that their results are valid only in particular cases such as symmetric networks. However, in this paper, we get extended results for general network case. Moreover, we confirm that (i) the forwarding queues in a system of weighted fair queues has a special property and (ii) the end to end throughput of a connection does not depend on the load of the intermediate forwarding queues between a source and a destination. We perform extensive simulations and verify that the analytical results exactly match the results obtained from simulations.\",\n \"Title: Forever Young: Aging Control In DTNs\\nYear: 2010\\n\\nAbstract: The demand for Internet services that require frequent updates through small\\nmessages, also known as microblogging, has tremendously grown in the past few\\nyears. Although the use of such applications by domestic users is usually free,\\ntheir access from mobile devices is subject to fees and consumes energy from\\nlimited batteries. If a user activates his mobile device and is in range of a\\nservice provider, a content update is received at the expense of monetary and\\nenergy costs. Thus, users face a tradeoff between such costs and their messages\\naging. The goal of this paper is to show how to cope with such a tradeoff, by\\ndevising \\\\emph{aging control policies}. An aging control policy consists of\\ndeciding, based on the current utility of the last message received, whether to\\nactivate the mobile device, and if so, which technology to use (WiFi or 3G). We\\npresent a model that yields the optimal aging control policy. Our model is\\nbased on a Markov Decision Process in which states correspond to message ages.\\nUsing our model, we show the existence of an optimal strategy in the class of\\nthreshold strategies, wherein users activate their mobile devices if the age of\\ntheir messages surpasses a given threshold and remain inactive otherwise. We\\nthen consider strategic content providers (publishers) that offer \\\\emph{bonus\\npackages} to users, so as to incent them to download updates of advertisement\\ncampaigns. We provide simple algorithms for publishers to determine optimal\\nbonus levels, leveraging the fact that users adopt their optimal aging control\\nstrategies. The accuracy of our model is validated against traces from the\\nUMass DieselNet bus network.\",\n \"Title: Mobility management and call admission control for IEEE 802.16e wireless networks\\nYear: 2010\\n\\nAbstract: The demand for wireless broadband access systems supporting mobility of the individual users has dramatically increased in recent years. Therefore, techniques for dynamically adapting the resources to the link condition become necessary for achieving high spectral efficiency. Our aim in this paper is to develop an admission control scheme that handles the intra-cell mobility issue in IEEE802.16e Wireless Networks with link adaptation. In particular we consider two types of intra-cell mobility: Low-mobility and High-mobility. We show how the admission threshold may adaptively controlled to achieve a better balance between guaranteing dropping probability and maximizing resource utilization. Simulation results show that the call admission control scheme under different types of mobility outperforms well by reserving amount of resource in order to provide an high priority of migration calls.\",\n \"Title: Optimal sensing policies for smartphones in hybrid networks: A POMDP approach\\nYear: 2012\\n\\nAbstract: The ubiquity of mobile devices is fostering a paradigm shift in the realm of Internet services. Consider, for instance, mobile users of social networks, that require frequent updates through small messages from their friends. If a user activates his mobile device and has a contact opportunity with an access point, an update can be received at the expense of monetary and energy costs. Thus, users face a tradeoff between such costs and the utilities of the messages received. The goal of this paper is to show how a user can cope with such a tradeoff, by deriving optimal sensing policies. A sensing policy consists of deciding, based on the age of the last message received and the belief about the future availability of a WiFi access point, whether to activate the mobile device or not. Alternatively, users may also decide to use 3G technology to receive updates, which provides broader coverage at the expense of higher monetary costs and lower bandwidth. To address the tradeoff faced by the users, we propose an analytical model based on a Partially Observed Markov Decision Process (POMDP) with an average reward criterion. Using the proposed model, we show properties of the optimal sensing policy. The applicability of the model and of the derived policy is illustrated through numerical case studies.\",\n \"Title: Modeling the Effect of Forwarding in IEEE-802.11e-Based Ad Hoc Networks with Weighted Fair Queueing\\nYear: 2009\\n\\nAbstract: We present in this paper a model that takes into account topology, routing, random access in MAC layer governed by IEEE 802.11 and forwarding probability. We are focusing to study the effect of cooperation and PHY/MAC layers (such as transmit power and physical carrier sense) on the stability and throughput of ad-hoc network. In this paper, we first extended the results of Yang's model [1] to a multi-hop ad-hoc network with asymmetric topology and traffic load. Based on this analysis, we reach a fixed-point formulation which characterizes the system operating points and allows us to compute the attempt rate of MAC layer which depends on the traffic flows on the network layer. With the use of the analytical model derived, we use an approach to derive throughput of multi-hop routes and stability of forwarding queues. However, in this paper, we get extended results for general network case.\",\n \"Title: Perturbation of linear quadratic systems with jump parameters and hybrid controls\\nYear: 2000\\n\\nAbstract: . We consider the problem of the perturbation of a class of linear-quadratic differential games with piecewise deterministic\\n dynamics, where the changes from one structure (for the dynamics) to another are governed by a finite-state Markov process.\\n Player 1 controls the continuous dynamics, whereas Player 2 controls the rate of transition for the finite-state Markov process;\\n both have access to the states of both processes. Player 1 wishes to minimize a given quadratic performance index, while player\\n 2 wishes to maximize or minimize the same quantity. The problem above leads to the analysis of some linearly coupled set of\\n quadratic equations (Riccati equations). We obtain a Taylor expansion in the perturbation for the solution of the equation\\n for a fixed stationary policy of the player 2. This allows us to solve the game or team problem as a function of the perturbation.\",\n \"Title: Blind online optimal forwarding in heterogeneous delay tolerant networks\\nYear: 2011\\n\\nAbstract: In this paper we consider a heterogeneous DTN formed by different classes of nodes. A core problem in DTNs literature is to optimize forwarding control such in a way to deliver messages from source to destination under a given energy budget. In this context, we model the presence of different classes and describe the structure of the optimal solution. The closed form of an optimal policy is found in the family of the multi-dimensional dynamic strategies. These policies assign a specific threshold to each class of mobiles. We also show that in the multi-class case there exists an order of the classes that depends on the system parameters in a very simple form. In particular, we exploit the ordering induced by the optimal control with the goal of designing a new class of multidimensional stochastic approximation algorithms that apply to the heterogeneous DTNs. These algorithms are blind because they drive online the source node to the optimal operating point without requiring explicit estimation of network parameters (intermeeting intensities and the number of nodes for each class).\",\n \"Title: Analytical Framework for Contact Time Evaluation in Delay-Tolerant Networks.\\nYear: 2009\\n\\nAbstract: In the last few years, there has been an increasing concern about stochastic properties of contact-based metrics under general mobility models in delay-tolerant networks. Such a concern will provide a first step toward detailed performance analysis of various routing/forwarding algorithms and shed light on better design of network protocols under realistic mobility patterns. However, throughout the variety of research works in this topic, most interests rather focused on the inter-contact time while other contact-based metrics such as the contact time received too little interest. In this paper, we provide an analytical framework to estimate the contact time in delay-tolerant networks based on some recent key results derived from biology and statistical physics while studying spontaneous displacement of insects such as ants. In particular, we analytically derive a closed-form expression for the average value of the contact time under the random waypoint mobility model and then give an approximation for its distribution function.\",\n \"Title: Optimal number of users in wireless networks: A flat rate pricing\\nYear: 2012\\n\\nAbstract: We study revenue-maximizing in a wireless system where interested users share a common spectrum and interfere with each other. Capacity is increased in proportion of the number of users. Our objective is to design a scheme that achieves an optimal solution for the provider with respect to users strategy, bandwidth management and some fairness criteria. We consider a flat rate pricing for users. Two cases are studied: firstly, an approximation to the original problem when the provider has a perfect knowledge about the channel state of each user is proposed. In the second case, a decision theoretic approach based on a POMDP framework is elaborated when the provider has only a partial information on the system state. Additionally, we provide numerical results that illustrate the performance of the proposed solutions.\",\n \"Title: Dynamic retransmission limit scheme for routing in multi-hop ad hoc networks\\nYear: 2007\\n\\nAbstract: In paper [2] and [3], we have studied the throughput and stability of forwarding queues in a wireless ad hoc network with random access channel. In this paper, we are focusing to draw benefit from the interaction of the MAC (governed by IEEE 802.11 or slotted aloha) and routing by defining a new cross-layer scheme for routing based on the limit number of retransmission. By adjusting dynamically and judiciously this parameter in a saturated network, we have realized that both stability and average throughput are significantly improved in linear symmetric networks: a gain of 100 % can be reached, while in asymmetric networks, we achieve a better average delay (resp. throughput) for each connection without changing the average throughput (resp. delay). A detailed performance study of our new scheme is presented using analytical and simulation evaluation.\",\n \"Title: A survey on YouTube streaming service.\\nYear: 2011\\n\\nAbstract: Established in 2005, YouTube is one of the fastest-growing websites, and has become one of the most accessed sites in the Internet. It has a significant impact on the Internet traffic distribution, but itself is suffering from severe scalability constraints and quality of service. Understanding the features of YouTube is thus crucial to network traffic engineering and to sustainable development of this new generation of services. In this survey, we first present an overview of previous works and analysis on YouTube with a particular attention on Quality of Experience. We then describe how the increased availability of meta-data in Web 2.0 (e.g., popularity distribution of video clips) could be effectively exploited to improve the performance and scalability of YouTube. In particular, we study the benefit gained by local caching along with prefetching in terms of reducing the client access time and start up delay in watching video.\",\n \"Title: Stochastic learning solution for constrained Nash Equilibrium throughput in non saturated wireless collision channels\\nYear: 2009\\n\\nAbstract: We consider finite number of users, with infinite buffer storage, sharing a single channel using the aloha medium access protocol. This is an interesting example of a non saturated collision channel. We investigate the uplink case of a cellular system where each user will select a desired throughput. The users then participate in a non cooperative game wherein they adjust their transmit rate to attain their desired throughput. We show that this game, in contrast to the saturated case, either has no Nash Equilibrium or has infinitely many Nash Equilibria. Further, we show that the region of NE coincides with an appropriate 'stability region'. We also discuss the efficiency of the equilibria in term of energy consumption and congestion rate. Next, we propose two learning algorithms using a stochastic iterative procedure that converges to the best Nash equilibrium. For instance, the first one needs partial information (transmit rates of other users during the last slot) whereas the second is an information less and fully distributed scheme. We approximate the control iterations by an equivalent ordinary differential equation in order to prove that the proposed stochastic learning algorithm converges to a Nash equilibrium even in the absence of any coordination or extra information. Extensive numerical examples and simulations are provided to validate our results.\",\n \"Title: Altruism in groups: An evolutionary games approach\\nYear: 2014\\n\\nAbstract: We revisit in this paper the relation between evolution of species and the mathematical tool of evolutionary games, which has been used to model and predict it. We indicate known shortcoming of this model that restricts the capacity of evolutionary games to model groups of individuals that share a common gene or a common fitness function. In this paper we provide a new concept to remedy this shortcoming in the standard evolutionary games in order to cover this kind of behavior. Further, we explore the relationship between this new concept and Nash equilibrium or ESS. We indicate through the study of some example in the biology as Hawk-Dove game, Stag-Hunt Game and Prisoner's Dilemma, that when taking into account a utility that is common to a group of individuals, the equilibrium structure may change dramatically. We also study the multiple access control in slotted Aloha based wireless networks. We analyze the impact of the altruism behavior on the performance at the equilibrium.\",\n \"Title: Flow Level QoE of Video Streaming in Wireless Networks.\\nYear: 2014\\n\\nAbstract: The Quality of Experience (QoE) of streaming service is often degraded by frequent playback interruptions. To mitigate the interruptions, the media player prefetches streaming contents before starting playback, at a cost of delay. We study the QoE of streaming from the perspective of flow dynamics. First, a framework is developed for QoE when streaming users join the network randomly and leave after downloading completion. We compute the distribution of prefetching delay using partial differential equations (PDEs), and the probability generating function of playout buffer starvations using ordinary differential equations (ODEs) for CBR streaming. Second, we extend our framework to characterize the throughput variation caused by opportunistic scheduling at the base station, and the playback variation of VBR streaming. Our study reveals that the flow dynamics is the fundamental reason of playback starvation. The QoE of streaming service is dominated by the first moments such as the average throughput of opportunistic scheduling and the mean playback rate. While the variances of throughput and playback rate have very limited impact on starvation behavior.\",\n \"Title: A Game Theoretic Model For Network Virus Protection\\nYear: 2014\\n\\nAbstract: Security is crucial for information systems. In a company, security management is traditionally controlled via a centralized single-point. However, when we deal with multiple computer systems interconnected in a wide area networks (WAN), the use of a central authority for security management is completely meaningless. In this paper, we propose a distributed decision-making designed to thwart viruses in a WAN. A key aspect is whether owners of devices are willing to update their anti-virus in order to protect their computers or not to pay for an anti-virus update and take the risk to be contaminated. Given the fact that computers are interconnected via networks and the Internet, the risk of being infected does not only depend on each computer's strategy, but also on the strategies chosen by other computers in the network. This makes the virus protection problem much more challenging. To do so, we model the interaction between nodes as a non-cooperative game in which each node decides individually whether to update the anti-virus or not. The virus spread is assumed to follow a biologically inspired epidemic model in which the dynamic of sources that disseminate the virus evolves as function of the popularity of virus using the influence linear threshold model. We first provide a full characterization of the equilibria of the game and then we investigate the impact of the update cost. In particular, we study the performance of the strategies at the equilibrium in terms of the update cost and the network size on both the security management system and the anti-virus producers. These results give some helpful insights on how secure is decentralizing antivirus update decisions.\",\n \"Title: Evolutionary stable strategies in interacting communities.\\nYear: 2013\\n\\nAbstract: In many social networks or biological systems, several populations may be influenced by interactions with not only individuals of their own communities but also with other communities in the network. Moreover, the inter-community interactions can affect the outcome of a community in addition to the strategies followed inside the community itself. However, throughout the variety of research works in this topic, classical model of evolutionary games rather assumes individuals are equally likely to interact with any other member of the population and the success of any individual depends only on the frequency with which all strategies are represented in the population. Instead, we devote in this paper a novel mathematical tool to model the evolution of a population composed of several communities where the interaction between them is non-uniform.\\n\\nThe main objective of our work is to model these interactions using evolutionary games and to find the evolutionary stable strategies (ESS). Hence, we show through the study of two communities and pairwise interactions that when taking into account the non-uniform interaction, any mixed Nash equilibrium is not an ESS wherein all communities are robust against a deviation of a fraction of the population from both communities which may wish to deviate. Moreover, the same mixed equilibrium is an ESS under some conditions when we consider the fitness of the whole population instead of the fitness of each community. We also demonstrate that this equilibrium is globally asymptotically stable for the replicator dynamic which describes the dynamic of the evolution of communities as well as the impact of the interaction between communities.\",\n \"Title: Delayed Evolutionary game dynamics with non-uniform interactions in two communities.\\nYear: 2014\\n\\nAbstract: In this paper we study the evolution of a population composed of several communities where the interaction between them is non-uniform. The main objective is to study the effect of slow time scale delays on the convergence of the replicator dynamics to the Evolutionary Stable Strategy (ESS). In particular, we investigate the effect of time delays on the stability of the replicator dynamics. We can distinguish two types of the time delay: strategic delay which is the delay associated with the strategies of players, and spatial delay, which is the delay associated with the communities of the players. We prove that for large strategic delays, the mixed intermediate ESS may become an unstable state for the replicator dynamics, while for any value of the spatial delay, the replicator dynamics converge to the intermediate ESS. We also show that when both types of the time delay coexist, the instability property is emphasized.\",\n \"Title: Bio-inspired models for characterizing YouTube viewcout.\\nYear: 2014\\n\\nAbstract: The goal of this paper is to study the behaviour of viewcount in YouTube. We first propose several bio-inspired models for the evolution of the viewcount of YouTube videos. We show, using a large set of empirical data, that the viewcount for 90% of videos in YouTube can indeed be associated to at least one of these models, with a Mean Error which does not exceed 5%. We derive automatic ways of classifying the viewcount curve into one of these models and of extracting the most suitable parameters of the model. We study empirically the impact of videos' popularity and category on the evolution of its viewcount. We finally use the above classification along with the automatic parameters extraction in order to predict the evolution of videos' viewcount.\",\n \"Title: Adaptive Modulation and Coding scheme with intra- and inter-cell mobility for HSDPA system\\nYear: 2009\\n\\nAbstract: The Adaptive Modulation and Coding (AMC) scheme which handles user's mobility issue plays a significant role to improve the desired quality of service in high Speed Downlink Packet Access (HSDPA) networks. We develop a resource allocation which maintains constant bit rate for real-time (RT) and non-real-time (NRT) calls independently of the user position in the cell. We present analytical results relating to user performance in terms of blocking and dropping probabilities, and the results relating to the system performance in term of total throughput. We propose a call admission control scheme to derive a good trad-off between the blocking and dropping probabilities, and the choice of scheduling that efficiently utilizes the available resources.\",\n \"Title: Rejuvenation and the Spread of Epidemics in General Topologies\\nYear: 2014\\n\\nAbstract: Epidemic models have received significant attention in the past few decades to study the propagation of viruses, worms and ideas in computer and social networks. In the case of viruses, the goal is to understand how the topology of the network and the properties of the nodes that comprise the network, together, impact the spread of the epidemics. In this paper, we propose rejuvenation as a way to cope with epidemics. Then, we present a model to study the effect of rejuvenation and of the topology on the steady-state number of infected and failed nodes. We distinguish between a state in which the virus is incubating and in which symptoms might not be visible and yet they may be contagious and infecting other nodes, and a state of failure where symptoms are clear. Sampling costs might be incurred to examine nodes in search for viruses at an early stage. Using the proposed model, we show that the sampling rate admits at most one local minimum greater than zero. Then, we numerically illustrate the impact of different system parameters on the optimal sampling rate, indicating when rejuvenation is more beneficial.\",\n \"Title: Modeling rewards and incentive mechanisms for Delay Tolerant Networks\\nYear: 2014\\n\\nAbstract: A central problem in Delay Tolerant Networks (DTNs) is to persuade mobile nodes to participate in relaying messages. Indeed, the delivery of a message incurs a certain number of costs for a relay. We consider a two-hop DTN in which a source node, wanting to get its message across to the destination as fast as possible, promises each relay it meets a reward. This reward is the minimum amount that offsets the expected delivery cost, as estimated by the relay from the information given by the source (number of existing copies of the message, age of these copies). A reward is given only to the relay that is the first one to deliver the message to the destination. For two relays and exponentially distributed inter-contact times, we show that the expected reward the source pays remains the same irrespective of the information it conveys, provided that the type of information does not vary dynamically over time. On the other hand, the source can gain by adapting the information that it conveys to a meeting relay.\",\n \"Title: Fair Scheduling in Cellular Systems in the Presence of Noncooperative Mobiles\\nYear: 2014\\n\\nAbstract: We consider the problem of centrally controlled 'fair' scheduling of resources to one of the many mobile stations connected to a base station (BS). The BS is the only entity making decisions in this framework based on truthful information from the mobiles on their radio channel. We study the well-known family of parametric ¿-fair scheduling problems from a game-theoretic perspective in which some of the mobiles may be noncooperative. We first show that if the BS is unaware of the noncooperative behavior from the mobiles, the noncooperative mobiles become successful in snatching the resources from the other cooperative mobiles, resulting in unfair allocations. If the BS is aware of the noncooperative mobiles, a new game arises with BS as an additional player. It can then do better by neglecting the signals from the noncooperative mobiles. The BS, however, becomes successful in eliciting the truthful signals from the mobiles only when it uses additional information (signal statistics). This new policy along with the truthful signals from mobiles forms a Nash Equilibrium (NE) called a Truth Revealing Equilibrium. Finally, we propose new iterative algorithms to implement fair scheduling policies that robustify the otherwise non-robust (in presence of noncooperation) ¿-fair scheduling algorithms.\",\n \"Title: Delayed Evolutionary Game Dynamics applied to Medium Access Control\\nYear: 2007\\n\\nAbstract: A major contribution of biology to competitive decision making is the development of the discipline of evolutionary games. Its ESS (evolutionary stable strategy) equilibrium concept, well adapted to large populations of players, describes robustness against deviations of a whole fraction of the population (in contrast to Nash equilibrium that requires robustness against a single user's deviation). The second appealing element of evolutionary games, the replicator dynamics, describes the evolution of strategies in time. Under suitable conditions, this (and some other) dynamics, converge to an ESS. In this paper we study the effect of slow time scale delays on the convergence of various dynamics. We apply this to an evolutionary game describing competition between mobile terminals over the access to a common channel.\",\n \"Title: Analytical Model for Connectivity in Vehicular Ad Hoc Networks\\nYear: 2008\\n\\nAbstract: We investigate connectivity in the ad hoc network formed between vehicles that move on a typical highway. We use a common model in vehicular traffic theory in which a fixed point on the highway sees cars passing it that are separated by times with an exponentially distributed duration. We obtain the distribution of the distances between the cars, which allows us to use techniques from queuing theory to study connectivity. We obtain the Laplace transform of the probability distribution of the connectivity distance, explicit expressions for the expected connectivity distance, and the probability distribution and expectation of the number of cars in a platoon. Then, we conduct extensive simulation studies to evaluate the obtained results. The analytical model that we present is able to describe the effects of various system parameters, including road traffic parameters (i.e., speed distribution and traffic flow) and the transmission range of vehicles, on the connectivity. To more precisely study the effect of speed on connectivity, we provide bounds obtained using stochastic ordering techniques. Our approach is based on the work of Miorandi and Altman, which transformed the problem of connectivity distance distribution into that of the distribution of the busy period of an equivalent infinite server queue. We use our analytical results, along with common road traffic statistical data, to understand connectivity in vehicular ad hoc networks.\",\n \"Title: Evolutionary games with random number of interacting players applied to access control\\nYear: 2008\\n\\nAbstract: We study access games within a large population of mobiles that interfere with each other through many local interactions. Each local interaction involves a random number of mobiles. The games are not necessarily reciprocal as the set of mobiles causing interference to a given mobile may differ from the set of those suffering from its interference. We model and study this using the theory of evolutionary games which we extend to cover a random number of players.\",\n \"Title: An evolutionary game approach for the design of congestion control protocols in wireless networks\\nYear: 2008\\n\\nAbstract: Evolutionary games have been developed in biological sciences in the aim of studying the evolution and equilibrium behavior (called evolutionary stable strategies - ESS) of large populations. While rich theoretical foundations of evolutionary games allow biologist to explain past and present evolution and predict future evolution, it can be further used in engineering to architect evolution. In this paper, we apply evolutionary games to non-cooperative networks containing large number of individual non-cooperative terminals or sensors. We study the evolution of congestion control protocols, and show how the evolution and the ESS are influenced by the characteristics of the wireless channel. We then consider the challenge of architecting the evolution: we propose some guidelines for designing a framework that supports evolution of protocols.\",\n \"Title: Evolutionary game dynamics with migration for hybrid power control in wireless communications\\nYear: 2008\\n\\nAbstract: We propose an evolutionary game dynamics with migration for hybrid population games with many local interactions at the same time. Each local interaction concerns a random number of interacting players. The strategies of a player have two components. Specifically, each player chooses both (i) the region or subpopulation and (ii) an action among a finite set of secondary pure strategies in each region. We assume that when updating a strategy, a player can change only the secondary strategies associate to the region at a time. We investigate what impact this restriction has on the population dynamics. We apply this model to the integrated power control and base station assignment problem in a multi-cell in code division multiple access (CDMA) wireless data networks with large number of mobiles. We show that global neutrally evolutionary stable strategies are stationary points of hybrid mean dynamics called dynamics with multicomponent strategies under the positive correlation conditions. We give some convergence results of our hybrid model in stable population games and potential population games.\",\n \"Title: Markov decision evolutionary game for energy management in Delay Tolerant Networks\\nYear: 2011\\n\\nAbstract: In this paper, we apply the concepts of Markov decision evolutionary games to non-cooperative forwarding control of Delay Tolerant Networks (DTN). Specifically, we rely on the design of mechanisms at the source node to study forwarding probability of the message in a DTN using the two-hop routing. We study the forwarding probability as a function of the competition within a large population of mobiles which need occasionally to make some action. In particular, for each message generated by a source, each mobile may take a decision that concerns the strategy by which the mobile participates to the relaying of the message from source to destination. A mobile that participates receives a unit of reward if it is the first to deliver a copy of the packet to the destination. The action taken by a mobile determine not only the immediate reward but also the transition probability to its next battery energy state. We characterize the Evolutionary Stable Strategies (ESS) for these games and propose a method to compute them. We also propose a mechanism design at the source in order to maximize the message delivery probability to the destination, given the equilibrium behavior (called Evolutionary Stable Strategy - ESS).\",\n \"Title: Probabilistic analysis of buffer starvation in Markovian queues\\nYear: 2011\\n\\nAbstract: Our purpose in this paper is to obtain the exact distribution of the number of buffer starvations within a sequence of N consecutive packet arrivals. The buffer is modeled as an M/M/1 queue. When the buffer is empty, the service restarts after a certain amount of packets are prefetched. With this goal, we propose two approaches, one of which is based on Ballot theorem, and the other uses recursive equations. The Ballot theorem approach gives an explicit solution, but at the cost of the high complexity order in certain circumstances. The recursive approach, though not offering an explicit result, needs fewer computations. We further propose a fluid analysis of starvation probability on the file level, given the distribution of file size and the traffic intensity. The starvation probabilities of this paper have many potential applications. We apply them to optimize the quality of experience (QoE) of media streaming service, by exploiting the tradeoff between the start-up delay and the starvation.\",\n \"Title: Accumulative interference removal in multi-hop linear and circular ad hoc WLANs\\nYear: 2012\\n\\nAbstract: This paper presents a cross-layered framework to study linear and circular ad hoc networks built on the IEEE 802.11e Enhanced Distributed Coordination Function. We investigate the intricate interactions among PHY, MAC and Network layers. Now, the carrier sense threshold, the transmit power, the contention window size, the retransmissions retry limit and the multi rates are jointly incorporated. We propose an equivalent network where each node is able to listen to other nodes, i.e., the effect of hidden nodes could be definitely eliminated. Then, we develop an analytical model that predicts the throughput of each connection as well as the stability of forwarding queues at intermediate nodes while each node is able to prevent the accumulative interference phenomenon. Further, performance of such a system is evaluated via simulation. We show that the performance measures of MAC layer are affected by the traffic intensity of flows to be forwarded and the average hop length. Moreover, attempt rate and collision probability are dependent on the traffic flows, PHY parameters and routing scheme.\",\n \"Title: Optimal energy-delay tradeoff policies in cognitive radio networks\\nYear: 2012\\n\\nAbstract: Cognitive radio (CR) has been considered as a promising technology to enhance spectrum efficiency via opportunistic transmission at link level. We consider Opportunistic Spectrum Access (OSA) mechanism that takes into account packet delay and energy consumption. We formulate the OSA problem as a Partially Observable Markov Decision Process (POMDP) by explicitly considering the energy constraint as well as, the delay constraint, which are often ignored in existing OSA solutions. Specifically, we consider a POMDP with an average reward criterion. We further consider that the secondary user (SU) may decide, at any moment, to use another dedicated way (3G) of communication in order to transmit its packets. We derive structural properties of the value function and we show the existence of optimal strategies in the class of the threshold strategies. In particular, numerical illustrations validate our theoretical findings. It is shown that optimal policy has a threshold structure.\",\n \"Title: Interplay of contact times, fragmentation and coding in DTNs\\nYear: 2013\\n\\nAbstract: Models describing DTNs proposed in the last few years have been focusing on message replication policies able to achieve high delivery probability at the cost of network resources, e.g., message copies. To this respect, the duration of contact events is the physical constraint that dictates how fine a large message should be fragmented into packets in order to match finite contacts duration. The price to pay, indeed, is that the source has to deliver a larger number of packets per message. In this paper we model the combined effect of message fragmentation and buffering and describe the structure of the forwarding process in closed form when the message is split into K packets and delivered to the destination. We consider the specific case of sequential forwarding, where the source delivers fragmented message packets in order to relays. In this case the interplay of forwarding and message fragmentation can be expressed in closed form by coupling the combinatorial structure for packet forwarding and fluid models for the replication of packets in the network. By deriving the closed form expression for the delivery probability, we are able to derive the optimal fragmentation K* as a function of the contact time distribution. Finally, results on sequential forwarding are applied to derive performance figures for the case when fountain coding is applied: redundancy is added to the original information with the aim to increase the message delivery probability. The paper is completed by numerical results.\",\n \"Title: New insights from a bounded rationality analysis for strategic price-QoS war\\nYear: 2012\\n\\nAbstract: We present a game theoretic framework for the dynamical behaviors of a duopoly game in telecommunications service providers' context. Competition between two Service Providers (SPs) is assumed to take place in terms of their pricing decisions and the Quality of Service (QoS) they offer. According to the SPs' rationality level, we consider two schemes: 1) Both SPs are rational, and 2) One SP is rational and the second SP is boundedly rational. We describe the competitive interaction and analyze the resulting equilibria. Later, we compute explicitly the steady states of the dynamical system induced by bounded rationality, and establish a necessary and sufficient condition for stability of its Nash equilibria (NEs). We prove that there exists exactly one NE which is fair whereas remaining equilibria are unfair. A special feature is that the stability condition of the Nash equilibrium coincides with the instability condition of the boundary equilibria. Thus the system would never be absorbed by any of the unfair equilibria which solves the equilibrium selection issue ! Moreover, we show that considering the delay case (i.e., assuming a market with memory) increases the stability of the system. Here, the size of the memory could be considered for multi-level rationality, which means that bounded rationality tends to rationality as the memory size increases. We finally show that boundedly rational SPs with delay have a higher chance of reaching the fair Nash equilibrium.\",\n \"Title: A Theoretical Framework for Hierarchical Routing Games\\nYear: 2010\\n\\nAbstract: Most theoretical research on routing games in telecommunication networks has so far dealt with reciprocal congestion effects between routed entities. Yet in networks that support differentiation between flows, the congestion experienced by a packet depends on its priority level. Another differentiation is made by compressing the packets in the low priority flow while leaving the high priority flow intact. In this paper we study such kind of routing scenarios for the case of non-atomic users and we establish conditions for the existence and uniqueness of equilibrium.\",\n \"Title: Routing games : From egoism to altruism\\nYear: 2010\\n\\nAbstract: The paper studies the routing in the network shared by several users. Each user seeks to optimize either its own performance or some combination between its own performance and that of other users, by controlling the routing of its given flow demand. We parameterize the degree of cooperation which allows to cover the fully non-cooperative behavior, the fully cooperative behavior, and even more, the fully al- truistic behavior, all these as special cases of the paramet er's choice. A large part of the work consists in exploring the impact of the degree of cooperation on the equilibrium. Our first finding is to identify multiple Nash equilibria with cooperative behavior that do not occur in the non-cooperative case under the same conditions (cost, demand and topology). We then identify Braess like paradox (in which adding capacity or adding a link to a network results in worse performance to all users) in presence of user's co- operation. We identify another type of paradox in cooperation scenario: when a given user increases its degree of cooperation while other users keep unchanged their degree of cooperation, this may lead to an improvement in performance of that given user. We then pursue the exploration and carry it on to the setting of Mixed equilibrium (i.e. some users are non atomic-they have infinitesimally small de mand, and other have fi- nite fixed demand). We finally obtain some theoretical result s that show that for low degree of cooperation the equilibrium is unique, confirming the results of our numerical study.\",\n \"Title: Evolutionary forwarding games in Delay Tolerant Networks\\nYear: 2010\\n\\nAbstract: In this paper, we apply evolutionary games to non-cooperative forwarding control of Delay Tolerant Networks (DTN). We focus our study on the probability to deliver a message from source to destination in a DTN using two-hop routing. We derive the success probability as a function of the competition within a large population of mobiles. In particular, for each message generated by a source, each mobile may take a decision that concerns the strategy by which the mobile participates to the relaying. A mobile that participates receives a unit of reward if it is the first to deliver a copy of the packet to the destination. A utility function is introduced as the difference between a reward unit and the energy cost. We show how the evolution dynamics and the equilibrium behavior (called Evolutionary Stable Strategy — ESS) are influenced by the characteristics of inter contact time, energy expenditure and pricing characteristics. We specialize our analysis also to mechanisms that the source may introduce in order to have the message delivered to the destination with high probability within a given deadline.\",\n \"Title: Atomic hierarchical routing games in communication networks\\nYear: 2010\\n\\nAbstract: Theoretical studies on routing games in networks have so far dealt with reciprocal congestion effects between routing entities. But, with the advent of technologies like Cognitive Radio, we have networks which support differentiation of flows. In a two priority level model a user can be high priority or low priority and there is a cost for such a classification. The point of departure of this model from the traditional routing scenarios is the absence of reciprocity in the congestion effects: The low priority flow faces congestion from both high priority as well as low priority flow while the high priority flow is immune to the congestion effects from the low priority flow. This hierarchy is naturally present in contexts where there are primary (licensed) users and secondary (unlicensed) users who can sense their environment because there are equipped with a cognitive radio. We study such kind of routing scenarios for the cases of atomic users. We establish the existence and the uniqueness of Nash equilibrium and further we show the existence of a potential function for linear congestion costs and a certain priority classification pricing scheme. Natural applications of this model to Cognitive Radio are also pointed out.\",\n \"Title: Coordination minority games in delay tolerant networks\\nYear: 2013\\n\\nAbstract: In this paper we introduce a novel framework for the distributed control of DTNs. The mechanism that we propose tackles a crucial aspect of such systems: in order to support message replication the devices acting as relays need to sacrifice part of their batteries. The aim is thus to provide a reward mechanism able to induce activation of relays in a coordinated fashion. The proposed scheme functions in non-cooperative fashion, and requires minimal message exchange to operate. In particular, relays choose among two strategies: either to participate to message relaying, or not to participate in order to save energy. The base for our mechanism design is to define the relays' utility function according to a minority game; in fact, relays compete to be in the population minority with respect to activation. By tuning the activation level, the system can hence control and optimize the DTN operating point in a distributed manner. To this respect, we characterize extensively the possible equilibria of this game. Finally, a stochastic learning algorithm is proposed which can provably drive the system to the equilibrium solution without requiring perfect state information at relay nodes. We provide extensive numerical results to validate the proposed scheme.\",\n \"Title: Stochastic Geometric Models for Green Networking.\\nYear: 2015\\n\\nAbstract: In this paper, we use a stochastic geometric approach in order to study the impact on energy consumption when base stations are switched OFF independently of each other. We present here both the uplink and downlink analysis based on the assumption that the base stations are distributed according to an independent stationary Poisson point process. This type of modeling allows us to make use of the property that the spatial distribution of the base stations after thinning (switching oFF) is still a Poisson process. This implies that the probability distribution of the signal to interference and noise ratio (SINR) can be kept unchanged when switching-oFF base stations provided that we scale up the transmission power of the remaining base stations. We then solve the problem of optimally selecting the switch-oFF probabilities so as to minimize the energy consumptions, while keeping unchanged the SINR probability distribution. We then study the tradeoff in the uplink performance involved in switching-oFF base stations. These include the energy consumption, the coverage and capacity, and the impact on amount of radiation absorbed by the transmitting user.\",\n \"Title: Core Stable Algorithms for Coalition Games with Complementarities and Peer Effects.\\nYear: 2015\\n\\nAbstract: In this short paper, we show two new algorithms for finding stable structures in ordinal coalition potential games. The first one is anytime and enumerative. It performs on a graph. The second one is a modified Deferred Acceptance Algorithm (DAA) using counter-proposals. It finds a many-to-one matching. We illustrate with the example of video caching from a content creator's servers to a service provider's servers.\",\n \"Title: About Joint Stable User Association and Resource Allocation in Multi-Rate IEEE 802.11 WLANs\\nYear: 2015\\n\\nAbstract: This paper aims at proposing an alternative approach for both the modeling of the IEEE 802.11 resource allocation scheme and the design of mechanisms to reduce the impact of the anomaly of the protocol. We use game theory to model the IEEE 802.11 resource allocation and mobiles users to APs association as a coalition matching game. We propose a new mechanism that gives mobile users and APs the incentive to associate with each others in a way that both absorbs the load and reduce the negative impact of the anomaly in IEEE 802.11.\",\n \"Title: Controlled Matching Game for User Association and Resource Allocation in Multi-Rate WLANs\\nYear: 2015\\n\\nAbstract: The deployment of IEEE 802.11 based WLANs in populated areas is such that many mobile terminals are covered by several Access Points (APs). These mobiles have the possibility to associate to the AP with the strongest signal (best-RSSI association scheme). This can lead to poor performances and overloaded APs. Moreover, the well known anomaly in the protocol at the MAC layer may also lead to very unpredictable performances and affect the system throughput due to the presence of heterogeneous data rate nodes and the shared nature of the 802.11 medium. The goal of this paper is to propose an alternative approach for the association. We model the joint resource allocation and mobile user association as a matching game with complementarities, peer effects and selfish players(1). We focus on the throughput fairness allocation induced by the saturated regime with equal packet sizes. We propose a novel three-stages mechanism for the modeling and control of load balancing, resource allocation and user association. We show that the proposed mechanism can greatly improve the efficiency of 802.11 with heterogeneous nodes and reduce the negative impact of peer effects such as the anomaly in IEEE 802.11.\",\n \"Title: A fully distributed learning algorithm for power allocation in heterogeneous networks\\nYear: 2015\\n\\nAbstract: In this work, we present a fully distributed Learning algorithm for power allocation in HetNets, referred to as the FLAPH, that reaches the global optimum given by the total social welfare. Using a mix of macro and femto base stations, we discuss opportunities to maximize users global throughput. We prove the convergence of the algorithm and compare its performance with the well-established Gibbs and Max-logit algorithms which ensure convergence to the global optimum. Algorithms are compared in terms of computational complexity, memory space, and time convergence.\",\n \"Title: A Study of YouTube Recommendation Graph Based on Measurements and Stochastic Tools\\nYear: 2015\\n\\nAbstract: The Youtube recommendation is one the most important view source of a video. In this paper, we focus on the recommendation system in boosting the popularity of videos. We first construct a graph that captures the recommendation system in Youtube and study empirically the relationship between the number of views of a video and the average number of views of the videos in its recommendation list. We then consider a random walker on the recommendation graph, i.e. a random user that browses through videos such that the video it chooses to watch is selected randomly among the videos in the recommendation list of the previous video it watched. We study the stability properties of this random process and we show that the trajectory obtained does not contain cycles if the number of videos in the recommendation list is small (which is the case if the computer's screen is small).\",\n \"Title: Backward-shifted strategies based on SVC for HTTP adaptive video streaming\\nYear: 2016\\n\\nAbstract: Although HTTP-based video streaming can easily penetrate firewalls and profit from Web caches, the underlying TCP may introduce large delays in case of a sudden capacity loss. To avoid an interruption of the video stream in such cases we propose the Backward-Shifted Coding (BSC). Based on Scalable Video Coding (SVC), BSC adds a time-shifted layer of redundancy to the video stream such that future frames are downloaded at any instant. This pre-fetched content maintains a fluent video stream even under highly variant network conditions and leads to high Quality of Experience (QoE). We characterize this QoE gain by analyzing initial buffering time, re-buffering time and content resolution using the Ballot theorem. The probability generating functions of the playback interruption and of the initial buffering latency are provided in closed form. We further compute the quasi-stationary distribution of the video quality, in order to compute the average quality, as well as temporal variability in video quality. Employing these analytic results to optimize QoE shows interesting trade-offs and video streaming at outstanding fluency.\",\n \"Title: Analysis and modelling Quality of Experience of video streaming under time-varying bandwidth.\\nYear: 2016\\n\\nAbstract: We take an analytical approach to study the Quality of user Experience (QoE) for video streaming applications. Our propose is to characterise buffer starvations for streaming video in the presence of bandwidth that varies over time. This variation of the bandwidth is caused due to the frequent changes of channel states and mobility. Specifically we develop an analytical framework to investigate Quality of user Experience (QoE) for streaming by considering a Markov Modulated Fluid Model (MMFM) that accurately approximates the nature of network traffic. We drive the close-form expressions for calculating the distribution of starvation as well as start-up delay using partial differential equations (PDEs) and solve them using the Laplace Transform. We illustrate the results with the cases of the two-state Markov Modulated Fluid Model that is commonly used in multimedia applications. We compare our analytical model with simulation results using ns-3 under various operating parameters. Finally, we apply our analysis results to optimize the objective quality of experience (QoE) of media streaming realizing the tradeoff among different metrics incorporating user preferences on buffering ratio, startup delay and perceived quality.\",\n \"Title: NEWCAST: Anticipating resource management and QoE provisioning for mobile video streaming\\nYear: 2016\\n\\nAbstract: The knowledge of the future capacity variations in wireless networks using smartphones becomes more and more possible by exploiting the rich contextual information from smartphone sensors through mobile applications and services. It is entirely likely that such contextual information, which may include the traffic, mobility and radio conditions, could lead to a novel agile resource management not yet thought of. Inspired by the attractive features and potential advantages of this agile resource management, several approaches have been proposed during the last period. However, agile resource management also comes with its own challenges, and there are significant technical issues that still need to be addressed for successful rollout and operation of this technique. In this paper, we propose an approach (called NEWCAST) for anticipating throughput variation for mobile video streaming services. The solution of the optimization problem realizes a fundamental trade-off among critical metrics that impact the user's perceptual quality of the experience (QoE) and system utilization. Both simulated and real-world traces collected from [1] are carried out to evaluate the performance of NEWCAST. In particular, it is shown that NEWCAST provides the efficiency, computational complexity and robustness that the new 5G architectures require.\",\n \"Title: Coordinated scheduling via frequency and power allocation optimization in LTE cellular networks.\\nYear: 2016\\n\\nAbstract: Due to Orthogonal Frequency Division Multiple Access (OFDMA) mechanism adopted in LTE cellular networks, intra-cell interference is nearly absent. Yet, as these networks are designed for a frequency reuse factor of 1 to maximize the utilization of the licensed bandwidth, inter-cell interference coordination remains an important challenge. In both homogeneous and heterogeneous cellular networks, there is a need for scheduling coordination techniques to efficiently distribute the resources and mitigate inter-cell interference. In this paper, we propose a dynamic solution of inter-cell interference coordination performing an optimization of frequency sub-band reuse and transmission power in order to maximize the overall network utility. The proposed framework, based on game theory, permits to dynamically define frequency and transmission power patterns for each cell in the coordinated cluster. Simulation results show significant benefits in average throughput and also cell edge user throughput of 40% and 55% gains when performing the frequency sub-band muting and power control. Furthermore, we also obtain a meaningful improvement in energy efficiency.\",\n \"Title: Identifying A Volunteer-Like Dilemma In Cooperative Sensing-Empowered Cognitive Radio Networks\\nYear: 2016\\n\\nAbstract: Cooperative sensing is a promising technique that enables secondary users (SUs) to combine their channel observations in order to improve the spectrum sensing accuracy. However, in an adversarial environment where the secondary nodes are particularly selfish and the spectrum sensing is energy costly, selfish SUs can easily exploit the spectrum sensing result without participating in the sensing process. In this paper, we model the cooperative sensing as a volunteer dilemma where SUs have the choice to volunteer in order to sense and share the spectrum sensing results, or to free ride the spectrum sensing and achieve possibly a higher profitability. We mainly give a full characterization of the Nash equilibria and prove a counter-intuitive property that claims: the probability of volunteering decreases when the number of SUs increases. Additionally, we propose a practical negotiation algorithm in order to select efficiently one SU to access to the channel. Finally, we propose a distributed algorithm that converges to Nash equilibria and show its performance through simulation results.\",\n \"Title: Backward-Shifted Coding (BSC) based on Scalable Video Coding for HAS.\\nYear: 2016\\n\\nAbstract: The main task of HTTP Adaptive Streaming is to adapt video quality dynamically under variable network conditions. This is a key feature for multimedia delivery especially when quality of service cannot be granted network-wide and, e.g., throughput may suffer short term fluctuations. Hence, robust bitrate adaptation schemes become crucial in order to improve video quality. The objective, in this context, is to control the filling level of the playback buffer and maximize the quality of the video, while avoiding unnecessary video quality variations. In this paper we study bitrate adaptation algorithms based on Backward-Shifted Coding (BSC), a scalable video coding scheme able to greatly improve video quality. We design bitrate adaptation algorithms that balance video rate smoothness and high network capacity utilization, leveraging both on throughput-based and buffer-based adaptation mechanisms. Extensive simulations using synthetic and real-world video traffic traces show that the proposed scheme performs remarkably well even under challenging network conditions.\",\n \"Title: Optimal Energy-Delay Tradeoff for Opportunistic Spectrum Access in Cognitive Radio Networks.\\nYear: 2018\\n\\nAbstract: Cognitive radio (CR) has been considered as a promising technology to enhance spectrum efficiency via opportunistic transmission at link level. Basic CR features allow secondary users (SUs) to transmit only when the licensed channel is not occupied by primary users (PUs). However, waiting for an idle time slot may lead to large packet delays and high energy consumption. We further consider that the SU may decide, at any moment, to use another dedicated way of communication (4G) in order to transmit his packets. Thus, we consider an Opportunistic Spectrum Access (OSA) mechanism that takes into account packet delay and energy consumption. We formulate the OSA problem as a Partially Observable Markov Decision Process (POMDP) by explicitly considering the energy consumption as well as packets’ delay, which are often ignored in existing OSA solutions. Specifically, we consider a POMDP with an average reward criterion. We derive structural properties of the value function and we show the existence of optimal strategies in the class of the threshold strategies. For implementation purposes, we propose online learning mechanisms that estimate the PU activity and determine the appropriate threshold strategy on the fly. In particular, numerical illustrations validate our theoretical findings.\",\n \"Title: Context-Aware Mobility Resource Allocation For Qoe-Driven Streaming Services\\nYear: 2016\\n\\nAbstract: Streaming services come with their own challenges and technical issues that still need to be addressed for satisfying the target quality of experience (QoE) of the end-users in mobile environments. In this paper, we explore the idea of combining users' context information with the packed prefetching process features to enhance users' QoE in heterogeneous networks. More specifically, we propose a scheduling mechanism for video streaming traffic, in which the access to the network resources is restricted to users with a signal-to-noise plus interference ratio (SINR) above a given threshold. This scheme benefits from the fact that, as users are in permanent motion, they may experience different SINR values during the same video streaming session offering the opportunity to only schedule users with good channel conditions. The proposed scheduling approach (subsequently referred to as context-aware mode switching (CAMS)) not only allows to achieve overall network spectral efficiency improvement, but also guarantees fairness and QoE among users. Our simulation results show that CAMS achieves almost 1 bit per second per Hertz gain compared to the conventional scheduler (without CAMS), and up to 87% improvement in the probability of no starvations when users move at 40 kmph.\",\n \"Title: On the Design of a Reward-Based Incentive Mechanism for Delay Tolerant Networks.\\nYear: 2017\\n\\nAbstract: A central problem in Delay Tolerant Networks (DTNs) is to persuade mobile nodes to participate in relaying messages. Indeed, the delivery of a message incurs a certain number of costs for a relay. We consider a two-hop DTN in which a source node, wanting to get its message across to the destination as fast as possible, promises each relay it meets a reward. This reward is the minimum amount that offsets the expected delivery cost, as estimated by the relay from the information given by the source (number of existing copies of the message, age of these copies). A reward is given only to the relay that is the first one to deliver the message to the destination. We show that under fairly weak assumptions, the expected reward the source pays remains the same irrespective of the information it conveys, provided that the type of information does not vary dynamically over time. On the other hand, the source can gain by adapting the information it conveys to a meeting relay. For the particular cases of two relays or exponentially distributed inter-contact times, we give some structural results on the optimal adaptive policy.\",\n \"Title: Predicting popularity dynamics of online contents using data filtering methods.\\nYear: 2016\\n\\nAbstract: This paper proposes a new prediction process to explain and predicts popularity evolution of YouTube videos. We exploit prior study on the classification of YouTube videos in order to predict the evolution of videos' view-count. This classification allows to identify important factors of the observed popularity dynamics. In particular, we use this classification as filtering method allowing to identify the factors responsible for this popularity evolution. Results given by extensive experiments show that the proposed prediction process is able to reduce the average prediction errors compared to a state-of-the-art baseline model. We also evaluate the impact of adding popularity criteria in the classification.\",\n \"Title: Anticipating Resource Management and QoE for Mobile Video Streaming under Imperfect Prediction\\nYear: 2016\\n\\nAbstract: By leveraging geolocation and contextual information for mobile users, the prediction of the future throughput becomes more and more feasible. Many approaches on contextaware content delivery have been explored to balance the operators' limited resources with users' requirements. However, the perfect knowledge of the future context cannot be easily performed in real world, which represents a hurdle for most context-aware approaches. In this paper, we address a contextaware delivery algorithm for adaptive video streaming (NEWCAST) that have already been explored in [1] under perfect knowledge of future capacity, to balance the user's perception of the video and the cost of network usage. In order to make NEWCAST more resistant to eventual throughput prediction errors and adapt it to short-term horizons, we propose 4 algorithms that efficiently reduce the number of stalls by at least 75%.\",\n \"Title: User Association and Resource Allocation Optimization in LTE Cellular Networks.\\nYear: 2017\\n\\nAbstract: As the demand for higher data rates is growing exponentially, homogeneous cellular networks have been facing limitations when handling data traffic. These limitations are related to the available spectrum and the capacity of the network. Heterogeneous networks (HetNets), composed of macro cells (MCs) and small cells (SCs), are seen as the key solution to improve spectral efficiency per unit area a...\",\n \"Title: A Controlled Matching Game for WLANs.\\nYear: 2017\\n\\nAbstract: In multi-rate IEEE 802.11 WLANs, the traditional user association based on the strongest received signal and the well-known anomaly of the MAC protocol can lead to overloaded access points (APs), and poor or heterogeneous performance. Our goal is to propose an alternative game-theoretic approach for association. We model the joint resource allocation and user association as a matching game with complementarities and peer effects consisting of selfish players solely interested in their individual throughputs. Using recent game-theoretic results, we first show that various resource sharing protocols actually fall in the scope of the set of stability-inducing resource allocation schemes. The game makes an extensive use of the Nash bargaining and some of its related properties that allow controlling the incentives of the players. We show that the proposed mechanism can greatly improve the efficiency of 802.11 with heterogeneous nodes and reduce the negative impact of peer effects such as its MAC anomaly. The mechanism can be implemented as a virtual connectivity management layer to achieve efficient APs-user associations without modification of the MAC layer.\",\n \"Title: Hopf Bifurcations in Replicator Dynamics with Distributed Delays.\\nYear: 2017\\n\\nAbstract: In this paper, we study the existence and the property of the Hopf bifurcation in the two-strategy replicator dynamics with distributed delays. In evolutionary games, we assume that a strategy would take an uncertain time delay to have a consequence on the fitness (or utility) of the players. As the mean delay increases, a change in the stability of the equilibrium (Hopf bifurcation) may occur at which a periodic oscillation appears. We consider Dirac, uniform, Gamma, and discrete delay distributions, and we use the Poincare- Lindstedtu0027s perturbation method to analyze the Hopf bifurcation. Our theoretical results are corroborated with numerical simulations.\",\n \"Title: Learning from Experience: A Dynamic Closed-Loop QoE Optimization for Video Adaptation and Delivery.\\nYear: 2017\\n\\nAbstract: The quality of experience (QoE) is known to be subjective and context-dependent. Identifying and calculating the factors that affect QoE is indeed a difficult task. Recently, a lot of effort has been devoted to estimate the usersu0027 QoE in order to improve video delivery. In the literature, most of the QoE-driven optimization schemes that realize trade-offs among different quality metrics have been addressed under the assumption of homogenous populations. Nevertheless, people perceptions on a given video quality may not be the same, which makes the QoE optimization a hard task. This paper aims at taking a step further in order to address this limitation and meet usersu0027 profiles. Specifically, we propose a closed-loop control framework based on the usersu0027 (subjective) feedbacks to learn the QoE function and optimize it at the same time. Extensive simulation results show that the proposed scheme converges to a steady state, where the resulting QoE function noticeably improves the usersu0027 feedbacks.\",\n \"Title: Mobility Management Schemes For Real-Time Traffic And Resource Allocation In Ieee 802.16e Mobile Network\\nYear: 2017\\n\\nAbstract: In this paper, our aim is to develop in IEEE802.16e Wireless Networks with link adaptation, in the presence of real-time traffic call admission control (CAC) schemes. These CAC propose various scenarios of resource splitting and handling the intracell mobility. In particular, we consider two types of intracell mobility classes : low mobility class for mobiles moving usually with low speed between the neighboring regions of the cell and high mobility class for those moving with high speed and that can skip more than one region before changing their modulation. For this reason, we assume a time threshold Tth that determines the minimum time a call must remain in a region before the base station changes its modulation. And we compare it with the time that a call may spend in a region to decide whether the base station will change its modulation or not.In the beginning, we introduce two CAC schemes. In the first one, we reserve a portion of resources to mobiles in migration with both high and low mobility. And in the second one, we give the priority just to mobiles in migration with high mobility. Then, we calculate the impact in the blocking and dropping probabilities. We show by numerical results that by the proposed CAC schemes, we can find a resource management that outperforms well under different types of mobility. But, to find a good tradeoff between dropping the calls in migration and blocking the new calls, we introduce the optimization problem in the second part. So, we are faced to the necessity of optimizing the results found in the first part. Therefore, we define an objective function to optimize, in order to ensure the highest quality of service for users and to give a better stability state between the dropping and blocking probability. We show that the proposed objective function gives the optimal resources allocation between the migrating and new arriving calls in the cell. Copyright (C) 2015 John Wiley & Sons, Ltd.\",\n \"Title: A routing game in networks with lossy links\\nYear: 2014\\n\\nAbstract: Standard assumptions in the theory of routing games are that costs are additive over links and that there is flow conservation. The assumptions typically hold when the costs represent delays. We introduce here a routing game where losses occur on links in a way that may depend on the congestion. In that case both assumptions fail. We study a load balancing network and identify a Kameda type paradox in which by adding capacity, all players suffer larger loss rates.\",\n \"Title: Optimal open-loop control of online advertisement\\nYear: 2014\\n\\nAbstract: Online advertisement through social media is studied in this paper. We describe the combined effect of two mechanisms: the acceleration service offered by online media platforms and the effect of content sharing which occurs in parallel over multiple online channels. In particular, the latter operates by reinforcing the effect of acceleration and may ultimately determine the viral diffusion of the advertisement. We describe the concurrence of these two phenomena by means of a generalized Bass model, which is inspired by the dynamics which is observed in real data. Also, the source of the advertisement pays the platform owner in order to accelerate its diffusion. We hence focus on the optimal control problem that solves the trade-off between the acceleration cost and the diffusion of the advertisement. We derive the structure of the optimal control in open loop form and we discuss its structure in connection with the virality footprint of the advertisement.\",\n \"Title: Evolutionary games for multiple accessl control: From Egoism to Altruism\\nYear: 2014\\n\\nAbstract: This paper studies multiple access games with a large population of mobiles decomposed into several groups. Mobiles interfere with each others through many local interactions. We assume that each mobile (or player) cooperates with his group by taking into account the performance of his group. The degree of cooperation not only covers the fully non-cooperative behavior and the fully cooperative behavior, but also the fully altruistic behavior. To do so, we make use of the evolutionary game theory which we extend to cover this kind of behavior. We define and characterize the equilibrium (called Evolutionary Stable Strategy) for these games and establish the optimal level of cooperation that maximizes the probability of successful transmission. We also study the game dynamics both in its classical form and in the presence of delay. Interestingly, we show that, in order to maximize the system performance, the mobiles should be less cooperative.\",\n \"Title: Evolutionary Games in Interacting Communities.\\nYear: 2017\\n\\nAbstract: In this paper, we extend the evolutionary games framework by considering a population composed of communities with each having its set of strategies and payoff functions. Assuming that the interactions among the communities occur with different probabilities, we define new evolutionarily stable strategies (ESS) with different levels of stability against mutations. In particular, through the analysis of two-community two-strategy model, we derive the conditions of existence of ESSs under different levels of stability. We also study the evolutionary game dynamics both in its classic form and with delays. The delays may be strategic, i.e., associated with the strategies, spatial, i.e., associated with the communities, or spatial strategic. We apply our model to the Hawk–Dove game played in two communities with an asymmetric level of aggressiveness, and we characterize the regions of ESSs as function of the interaction probabilities and the parameters of the model. We also show through numerical examples how the delays and the game parameters affect the stability of the mixed ESS.\",\n \"Title: Bitrate Adaptation In Backward-Shifted Coding For Http Adaptive Video Streaming\\nYear: 2017\\n\\nAbstract: HTTP Adaptive Streaming is able to dynamically match video quality to variable network conditions. This is a key feature for multimedia delivery when quality of service cannot be granted network-wide. For instance, the end-to-end throughput towards mobile terminals may suffer short term fluctuations due to fading. Hence, robust bitrate adaptation schemes become crucial in order to avoid degraded video reproduction. The objective, in this context, is to control the filling level of the playback buffer, maximize the video quality, and avoid unnecessary quality variations which may also impair the perceived quality of experience. In this work we study bitrate adaptation algorithms leveraging on Backward-Shifted Coding (BSC), a scalable video coding scheme able to cope with the effects of end-to-end throughput fluctuations. We have proposed a new adaptation scheme able to balance video rate smoothness and high network capacity utilization. Both the throughput-based and buffer-based variants of such scheme have been designed. Extensive simulations using synthetic and real-world video traffic traces show that the proposed solution performs remarkably well even under challenging network conditions.\",\n \"Title: Quality Of Experience In Http Adaptive Video Streaming Systems\\nYear: 2017\\n\\nAbstract: The main task of HTTP Adaptive Streaming (HAS) is to adapt video quality dynamically under variable network conditions. This is a key feature for multimedia delivery especially when quality of service cannot be granted network-wide and, e.g., throughput may suffer short term fluctuations. Hence, robust bitrate adaptation schemes become crucial in order to improve video quality of experience. The objective, in this context, is to control the filling level of the playback buffer and maximize the quality of the video, while avoiding unnecessary video quality variations. In this paper we study bitrate adaptation algorithms based on Backward-Shifted Coding (BSC), a scalable video coding scheme able to greatly improve video quality of experience. We design bitrate adaptation algorithms that balance video rate smoothness and high network capacity utilization, leveraging both on throughput-based and buffer-based adaptation mechanisms. Extensive simulations using an MPEG/DASH client server application on ns-3 show that the proposed scheme performs remarkably well even under challenging network conditions.\",\n \"Title: A Multitype Hawk And Dove Game\\nYear: 2017\\n\\nAbstract: We consider in this paper the Hawk-Dove game in which each of infinitely many individuals, involved with pairwise encounters with other individuals, can decide whether to act aggressively (Hawk) or peacefully (Dove). Each individual is characterized by its strength. The strength distribution among the population is assumed to be fixed and not to vary in time. If both individuals involved in an interaction are Hawks, there will be a fight, the result of which will be determined by the strength of each of the individuals involved. The larger the difference between the strength of the individuals is, the larger is the cost for the weaker player involved in the fight. Our goal is to study the influence of the parameters (such as the strength level distribution) on the equilibrium of the game. We show that for some parameters there exists a threshold equilibrium policy while for other parameters there is no equilibrium policy at all.\",\n \"Title: Optimal Control of Storage Regeneration with Repair Codes\\nYear: 2017\\n\\nAbstract: High availability of containerized applications requires to perform robust storage of applications' state. Since basic replication techniques are extremely costly at scale, storage space requirements can be reduced by means of erasure and/or repairing codes. In this paper we address storage regeneration using repair codes, a robust distributed storage technique with no need to fully restore the whole state in case of failure. In fact, only the lost servers' content is replaced. To do so, new clean-slate storage units are made operational at a cost for activating new storage servers and a cost for the transfer of repair data. Our goal is to guarantee maximal availability of containers' state files by a given deadline. Upon a fault occurring at a subset of the storage servers, we aim at ensuring that they are repaired by a given deadline. We introduce a controlled fluid model and derive the optimal activation policy to replace servers under such correlated faults. The solution concept is the optimal control of regeneration via the Pontryagin minimum principle. We characterize feasibility conditions and we prove that the optimal policy is of threshold type. Numerical results describe how to apply the model for system dimensioning and show the tradeoff between activation of servers and communication cost.\",\n \"Title: Bio-inspired delayed evolutionary game dynamics with networking applications\\nYear: 2011\\n\\nAbstract: We study in this paper some evolutionary games where competition between individuals from a large population occurs through many local interactions between randomly selected individuals. We focus on games that have the property of possessing a single interior evolutionarily stable strategy (ESS). We study in particular the effect of the time delays on the convergence of evolutionary dynamics to the ESS in an evolutionary game in which each pure strategy is associated with its own delay. In particular, we study a multiple access game as well as a Hawk and Dove game. We study the properties of the ESS in these games and also the effect of time delays on the convergence of various bio-inspired evolutionary game dynamics to the ESS.\",\n \"Title: Discrete and Continuous Distributed Delays in Replicator Dynamics.\\nYear: 2018\\n\\nAbstract: In this paper, we study evolutionary games and we examine the stability of the evolutionarily stable strategy (ESS) in the continuous-time replicator dynamics with distributed time delays. In many examples, the interactions between individuals take place instantaneously, but their impacts are not immediate and take a certain amount of time which is usually random. In this paper, we study the consequences of distributed delays on the stability of the replicator dynamics. The main results are that (i) under the exponential delay distribution, the ESS is asymptotically stable for any value of the rate parameter; (ii) the necessary and sufficient conditions for the asymptotic stability of the ESS under uniform and Erlang distributions; and (iii) for the discrete distributed delays, we derive a necessary and sufficient delay-independent stability conditions. We illustrate our results with numerical examples from the Hawk–Dove game.\",\n \"Title: Analysis of QoE for adaptive video streaming over wireless networks\\nYear: 2018\\n\\nAbstract: Adaptive video streaming improves users' quality of experience (QoE), while using the network efficiently. In the last few years, adaptive video streaming has seen widespread adoption and has attracted significant research effort. We study a dynamic system of random arrivals and departures for different classes of users using the adaptive streaming industry standard DASH (Dynamic Adaptive Streaming over HTTP). Using a Markov chain based analysis, we compute the user QoE metrics: probability of starvation, prefetching delay, average video quality and switching rate. We validate our model by simulations, which show a very close match. Our study of the playout buffer is based on client adaptation scheme, which makes efficient use of the network while improving users' QoE. We prove that for buffer-based variants, the average video bit-rate matches the average channel rate. Hence, we would see quality switches whenever the average channel rate does not match the available video bit rates. We give a sufficient condition for setting the playout buffer threshold to ensure that quality switches only between adjacent quality levels.\",\n \"Title: Forever Young: Aging Control For Hybrid Networks\\nYear: 2019\\n\\nAbstract: The demand for Internet services that require frequent updates through small messages, also known as microblogging, has tremendously grown in the past few years. Although the use of such applications by domestic users is usually free, their access from mobile devices is subject to fees and consumes energy from limited batteries. If a user activates his mobile device and is in the range of a publisher, an update is received at the expense of monetary and energy costs. Thus, users face a tradeoff between such costs and their messages aging. The goal of this paper is to show how to cope with such a tradeoff, by devising aging control policies. An aging control policy consists of deciding, based on the utility of the owned content, whether to activate the mobile device, and if so, which technology to use (WiFi or cellular). We present a model that yields the optimal aging control policy. Our model is based on a Markov Decision Process (MDP) in which states correspond to content ages. Using our model, we show the existence of an optimal strategy in the class of threshold strategies, wherein users activate their mobile devices if the age of their poadcasts surpasses a given threshold and remain inactive otherwise. The accuracy of our model is validated against traces from the UMass DieselNet bus network.\",\n \"Title: DCoflow: Deadline-Aware Scheduling Algorithm for Coflows in Datacenter Networks\\nYear: 2022\\n\\nAbstract: Datacenter networks routinely support the data transfers of distributed computing frameworks in the form of coflows, i.e., sets of concurrent flows related to a common task. The vast majority of the literature has focused on the problem of scheduling coflows for completion time minimization, i.e., to maximize the average rate at which coflows are dispatched in the network fabric. Modern applications, though, may generate coflows dedicated to online services and mission-critical computing tasks which have to comply with specific completion deadlines. In this paper, we introduce DCoflow, a lightweight deadline-aware scheduler for time-critical coflows in datacenter networks. The algorithm combines an online joint admission control and scheduling logic and returns a σ-order schedule which maximizes the number of coflows that attain their deadlines. Extensive numerical results demonstrate that the proposed solution outperforms existing ones.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":119,"string":"119"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13553,"string":"13,553"}}},{"rowIdx":13554,"cells":{"question_id":{"kind":"string","value":"Alfredo Vellido"},"question":{"kind":"string","value":"Generate an abstract for the title \"Cartogram-Based Data Visualization Using the Growing Hierarchical SOM.\" using the following items: 1. Model interpretability \n2. Data visualization \n3. Nonlinear dimensionality reduction \n4. Cartogram method \n5. Growing Hierarchical Self Organising Map (GHSOM)"},"answer":{"kind":"string","value":"Model interpretability is a problem of knowledge extraction from the patterns found in data to which data visualization can contribute. Nonlinear dimensionality reduction techniques provide flexible visual insight, but their locally varying representation distortion makes interpretation far from intuitive. In this paper, we apply a cartogram method, based on techniques of geographic representation, to data visualization. It allows reintroducing this distortion, measured as a U-matrix, in the visual maps of the Growing Hierarchical Self Organising Map (GHSOM)."},"correct_answer":{"kind":"string","value":"Model interpretability is a problem of knowledge extraction from the patterns found in data to which data visualization can contribute. Nonlinear dimensionality reduction techniques provide flexible visual insight, but their locally varying representation distortion makes interpretation far from intuitive. In this paper, we apply a cartogram method, based on techniques of geographic representation, to data visualization. It allows reintroducing this distortion, measured as a U-matrix, in the visual maps of the Growing Hierarchical Self Organising Map (GHSOM)."},"documents":{"kind":"list like","value":["Title: The extracellular N-terminal domain suffices to discriminate class C G Protein-Coupled Receptor subtypes from n-grams of their sequences\nYear: 2015\n\nAbstract: The investigation of protein functionality often relies on the knowledge of crystal 3-D structure. This structure is not always known or easily unravelled, which is the case of eukaryotic cell membrane proteins such as G Protein-Coupled Receptors (GPCRs) and specially of those of class C, which are the target of the current study. In the absence of information about tertiary or quaternary structures, functionality can be investigated from the primary structure, that is, from the amino acid sequence. In previous research, we found that the different subtypes of class C GPCRs could be discriminated with a high level of accuracy from the n-gram transformation of their complete primary sequences, using a method that combined two-stage feature selection with kernel classifiers. This study aims at discovering whether subunits of the complete sequence retain such discrimination capabilities. We report experiments that show that the extracellular N-terminal domain of the receptor suffices to retain the classification accuracy of the complete sequence and that it does so using a reduced selection of n-grams whose length of up to five amino acids opens up an avenue for class C GPCR signature motif discovery.","Title: Advances in clustering and visualization of time series using GTM through time.\nYear: 2008\n\nAbstract: Most of the existing research on multivariate time series concerns supervised forecasting problems. In comparison, little research has been devoted to their exploration through unsupervised clustering and visualization. In this paper, the capabilities of Generative Topographic Mapping Through Time, a model with foundations in probability theory, that performs simultaneous time series clustering and visualization, are assessed in detail. Focus is placed on the visualization of the evolution of signal regimes and the exploration of sudden transitions, for which a novel identification index is defined. The interpretability of time series clustering results may become extremely difficult, even in exploratory visualization, for high dimensional datasets. Here, we define and test an unsupervised time series relevance determination method, fully integrated in the Generative Topographic Mapping Through Time model, that can be used as a basis for time series selection. This method should ease the interpretation of time series clustering results.","Title: Machine learning in cancer research: implications for personalised medicine\nYear: 2008\n\nAbstract: Driven by the growing demand of personalization of medical procedures, data-based, computer-aided cancer research in human patients is advancing at an accelerating pace, providing a broadening landscape of opportunity for Machine Learning methods. This landscape can be observed from the wide-reaching view of population studies down to the genotype detail. In this brief paper, we provide a sweeping glimpse, by no means exhaustive, of the state-of-the-art in this field at the different scales of data measurement and analysis.","Title: Feature Selection with Single-Layer Perceptrons for a Multicentre 1H-MRS Brain Tumour Database\nYear: 2009\n\nAbstract: A Feature Selection process with Single-Layer Perceptrons is shown to provide optimum discrimination of an international, multi-centre 1H-MRS database of brain tumors at reasonable computational cost. Results are both intuitively interpretable and very accurate. The method remains simple enough as to allow its easy integration in existing medical decision support systems.","Title: <Emphasis Type=\"Italic\">An Electronic Commerce Application of the Bayesian Framework for MLPs: The Effect of Marginalisation and ARD</Emphasis>\nYear: 2001\n\nAbstract: Automatic Relevance Determination (ARD), with multiple regularisation terms. A real-world classification problem, concerning the discrimination of online purchasers and non-purchasers using Internet’s WWW users’ opinions, is the test-bed for this assessment.","Title: Describing customer loyalty to spanish petrol stations through rule extraction\nYear: 2006\n\nAbstract: Globalization and deregulation are modifying the competitive framework in the majority of economic sectors and, as a result, many companies are changing their commercial model to focus on the preservation of existing customers. Understanding customer loyalty therefore represents an element of competitive advantage. In this brief paper, we investigate loyalty in the Spanish petrol station market, according to the customer satisfaction and switching barriers constructs. Satisfaction and behavioural intentions are analysed within a classification framework using Bayesian neural networks. The necessary interpretability and actionability of the results is achieved through the use of a feature selection process embedded in the network training and a novel rule extraction method.","Title: Semi-Supervised Analysis Of Human Brain Tumours From Partially Labeled Mrs Information, Using Manifold Learning Models\nYear: 2011\n\nAbstract: Medical diagnosis can often be understood as a classification problem. In oncology, this typically involves differentiating between tumour types and grades, or some type of discrete outcome prediction. From the viewpoint of computer-based medical decision support, this classification requires the availability of accurate diagnoses of past cases as training target examples. The availability of such labeled databases is scarce in most areas of oncology, and especially so in neuro-oncology. In such context, semi-supervised learning oriented towards classification can be a sensible data modeling choice. In this study, semi-supervised variants of Generative Topographic Mapping, a model of the manifold learning family, are applied to two neuro-oncology problems: the diagnostic discrimination between different brain tumour pathologies, and the prediction of outcomes for a specific type of aggressive brain tumours. Their performance compared favorably with those of the alternative Laplacian Eigenmaps and Semi-Supervised SVM for Manifold Learning models in most of the experiments.","Title: Handling outliers in brain tumour MRS data analysis through robust topographic mapping.\nYear: 2006\n\nAbstract: Uncertainty is inherent in medical decision making and poses a challenge for intelligent technologies. This paper focuses on magnetic resonance spectra (MRS) for discrimination of brain tumour types and grades. Modelling of this type of high-dimensional data is commonly affected by uncertainty caused by the presence of outliers. Multivariate data clustering and visualization of MRS data is proposed using the GTM framework with basis functions comprising Student t-distributions in order to minimize the negative impact on the model from outliers. The effectiveness of this model on the MRS data is demonstrated empirically.","Title: Sepsis mortality prediction with the Quotient Basis Kernel.\nYear: 2014\n\nAbstract: Objective: This paper presents an algorithm to assess the risk of death in patients with sepsis. Sepsis is a common clinical syndrome in the intensive care unit (ICU) that can lead to severe sepsis, a severe state of septic shock or multi-organ failure. The proposed algorithm may be implemented as part of a clinical decision support system that can be used in combination with the scores deployed in the ICU to improve the accuracy, sensitivity and specificity of mortality prediction for patients with sepsis. Methodology: In this paper, we used the Simplified Acute Physiology Score (SAPS) for ICU patients and the Sequential Organ Failure Assessment (SOFA) to build our kernels and algorithms. In the proposed method, we embed the available data in a suitable feature space and use algorithms based on linear algebra, geometry and statistics for inference. We present a simplified version of the Fisher kernel (practical Fisher kernel for multinomial distributions), as well as a novel kernel that we named the Quotient Basis Kernel (QBK). These kernels are used as the basis for mortality prediction using soft-margin support vector machines. The two new kernels presented are compared against other generative kernels based on the Jensen Shannon metric (centred, exponential and inverse) and other widely used kernels (linear, polynomial and Gaussian). Clinical relevance is also evaluated by comparing these results with logistic regression and the standard clinical prediction method based on the initial SAPS score. Results: As described in this paper, we tested the new methods via cross-validation with a cohort of 400 test patients. The results obtained using our methods compare favourably with those obtained using alternative kernels (80.18% accuracy for the QBK) and the standard clinical prediction method, which are based on the basal SAPS score or logistic regression (71.32% and 71.55%, respectively). The QBK presented a sensitivity and specificity of 79.34% and 83.24%, which outperformed the other kernels analysed, logistic regression and the standard clinical prediction method based on the basal SAPS score. Conclusion: Several scoring systems for patients with sepsis have been introduced and developed over the last 30 years. They allow for the assessment of the severity of disease and provide an estimate of in-hospital mortality. Physiology-based scoring systems are applied to critically ill patients and have a number of advantages over diagnosis-based systems. Severity score systems are often used to stratify critically ill patients for possible inclusion in clinical trials. In this paper, we present an effective algorithm that combines both scoring methodologies for the assessment of death in patients with sepsis that can be used to improve the sensitivity and specificity of the currently available methods. (C) 2014 Elsevier B.V. All rights reserved.","Title: DSS-oriented exploration of a multi-centre magnetic resonance spectroscopy brain tumour dataset through visualization\nYear: 2008\n\nAbstract: The exploration of brain tumours usually requires non- invasive techniques such as Magnetic Resonance Imaging (MRI) or Mag- netic Resonance Spectroscopy (MRS). While radiologists are used to inter- preting MRI, many of them are not used to the biochemical information provided by MRS. In this situation, radiologists may benefit from the use of computer-based support for their decisions. As part of the development of a medical Decision Support System (DSS), MRS data corresponding to various tumour pathologies are used to assist expert diagnosis. The high dimensionality of the data might obscure peculiarities and anomalies that would jeopardize automated DSS diagnostic assistance. We illustrate how visualization, combined with expert opinion, can be used to explore data in a process that should improve computer-based tumour classification.","Title: Comparative assessment of the robustness of missing data imputation through generative topographic mapping\nYear: 2005\n\nAbstract: The incompleteness of data is a most common source of uncertainty in real-world Data Mining applications. The management of this uncertainty is, therefore, a task of paramount importance for the data analyst. Many methods have been developed for missing data imputation, but few of them approach the problem of imputation as part of a general data density estimation scheme. Amongst the latter, a method for imputing and visualizing multivariate missing data using Generative Topographic Mapping was recently presented. This model and some of its extensions are tested under different experimental conditions. Its performance is compared to that of other missing data imputation techniques, thus assessing its relative capabilities and limitations.","Title: Rule-Based Assistance to Brain Tumour Diagnosis Using LR-FIR\nYear: 2008\n\nAbstract: This paper describes a process of rule-extraction from a multi-centre brain tumour database consisting of nuclear magnetic resonance spectroscopic signals. The expert diagnosis of human brain tumours can benefit from computer-aided assistance, which has to be readily interpretable by clinicians. Interpretation can be achieved through rule extraction, which is here performed using the LR-FIR algorithm, a method based on fuzzy logic. The experimental results of the classification of three groups of tumours indicate in this study that just three spectral frequencies, out of the 195 from a range pre-selected by experts, are enough to represent, in a simple and intuitive manner, most of the knowledge required to discriminate these groups.","Title: Time series relevance determination through a topology-constrained hidden markov model\nYear: 2006\n\nAbstract: Most of the existing research on multivariate time series concerns supervised forecasting problems. In comparison, little research has been devoted to unsupervised methods for the visual exploration of this type of data. The interpretability of time series clustering results may be difficult, even in exploratory visualization, for high dimensional datasets. In this paper, we define and test an unsupervised time series relevance determination method for Generative Topographic Mapping Through Time, a topology-constrained Hidden Markov Model that performs simultaneous time series data clustering and visualization. This relevance determination method can be used as a basis for time series selection, and should ease the interpretation of the time series clustering results.","Title: On the improvement of brain tumour data clustering using class information\nYear: 2006\n\nAbstract: Cluster analysis can benefit from the use of class labels, if these are available, in a semi-supervised approach. In this study, we propose the integration of class information in the clustering of Magnetic Resonance Spectra (MRS) corresponding to human brain tumours using an extension of Generative Topographic Mapping (GTM) that behaves robustly in the presence of outliers.","Title: Determination of feature relevance for the grouping of motor unit action potentials through a generative mixture model.\nYear: 2007\n\nAbstract: The study of motor unit action potential (MUAP) activity from electromyographic signals is important for neurological investigations aiming to understand the state of the neuromuscular system. In this context, the identification and clustering of MUAPs that exhibit common characteristics, and the assessment of which data features are most relevant for the definition of such cluster structure, are central issues. In this paper, we propose the application of an unsupervised feature relevance determination (FRD) method to the analysis of experimental MUAPs. This method is embedded in a constrained mixture of distributions model that simultaneously performs data clustering and visualization. The experimental results of the analysis of a data set consisting of MUAPs measured from the First Dorsal Interosseous, a hand muscle, indicate that the features corresponding to the hyperpolarization period in the physiological process of generating muscle fibre action potentials are consistently estimated to be the most relevant. Moreover, the MUAP cluster structure of the data is shown to be only partially attributable to inter-subject differences, with the hyperpolarization period providing the best discrimination of the data by subject.","Title: The Effect Of Noise And Sample Size On An Unsupervised Feature Selection Method For Manifold Learning\nYear: 2008\n\nAbstract: The research on unsupervised feature selection is scarce in comparison to that for supervised models, despite the fact that this is an important issue for many clustering problems. An unsupervised feature selection method for general Finite Mixture Models was recently proposed and subsequently extended to Generative Topographic Mapping (GTM), a manifold learning constrained mixture model that provides data visualization. Some of the results of a previous partial assessment of this unsupervised feature selection method for GTM suggested that its performance may be affected by insufficient sample size and by noisy data. In this brief study, we test in some detail such limitations of the method.","Title: Missing data imputation through GTM as a mixture of t-distributions.\nYear: 2006\n\nAbstract: The Generative Topographic Mapping (GTM) was originally conceived as a probabilistic alternative to the well-known, neural network-inspired, Self-Organizing Maps. The GTM can also be interpreted as a constrained mixture of distribution models. In recent years, much attention has been directed towards Student t-distributions as an alternative to Gaussians in mixture models due to their robustness towards outliers. In this paper, the GTM is redefined as a constrained mixture of t-distributions: the t-GTM, and the Expectation-Maximization algorithm that is used to fit the model to the data is modified to carry out missing data imputation. Several experiments show that the t-GTM successfully detects outliers, while minimizing their impact on the estimation of the model parameters. It is also shown that the t-GTM provides an overall more accurate imputation of missing values than the standard Gaussian GTM.","Title: Finding Relevant Features to Characterize Student Behavior on an e-Learning System\nYear: 2005\n\nAbstract: E-learning systems such as virtual campus environments have established themselves as a strong alternative to traditional distance university education. In this context, Internet allows the gathering of plenty of information on students' online behavior. The knowledge extracted from this information can be used to define personalization strategies tailored to the students' needs and requirements. In this brief study we introduce a model to assess the relative relevance of individual data features on the grouping structure of the users of a real virtual campus (Open University of Catalonia). Tests carried out on these data indicate that certain web navigation behaviors are best at explaining and discriminating the different typologies of online students.","Title: Advances In Semi-Supervised Alignment-Free Classification Of G Protein-Coupled Receptors\nYear: 2013\n\nAbstract: G Protein-coupled receptors (GPCRs) are integral cell membrane proteins of great relevance for pharmacology due to their role in transducing extracellular signals. The 3-D structure is unknown for most of them, and the investigation of their structure-function relationships usually relies on the construction of 3-D receptor models from amino acid sequence alignment onto those receptors of known structure. Sequence alignment risks the loss of relevant information. Different approaches have attempted the analysis of alignment-free sequences on the basis of amino acid physicochemical properties. In this paper, we use the Auto-Cross Covariance method and compare it to an amino acid composition representation. Novel semi-supervised manifold learning methods are then used to classify the several members of class C GPCRs on the basis of the transformed data. This approach is relevant because protein sequences are not always labeled and methods that provide robust classification for a limited amount of labels are required.","Title: Cartogram visualization for nonlinear manifold learning models\nYear: 2013\n\nAbstract: Real-world applications of multivariate data analysis often stumble upon the barrier of interpretability. Simple data analysis methods are usually easy to interpret, but they risk providing poor data models. More involved methods may instead yield faithful data models, but limited interpretability. This is the case of linear and nonlinear methods for multivariate data visualization through dimensionality reduction. Even though the latter have provided some of the most exciting visualization developments, their practicality is hindered by the difficulty of explaining them in an intuitive manner. The interpretability, and therefore the practical applicability, of data visualization through nonlinear dimensionality reduction (NLDR) methods would improve if, first, we could accurately calculate the distortion introduced by these methods in the visual representation and, second, if we could faithfully reintroduce this distortion into such representation. In this paper, we describe a technique for the reintroduction of the distortion into the visualization space of NLDR models. It is based on the concept of density-equalizing maps, or cartograms, recently developed for the representation of geographic information. We illustrate it using Generative Topographic Mapping (GTM), a nonlinear manifold learning method that can provide both multivariate data visualization and a measure of the local distortion that the model generates. Although illustrated here with GTM, it could easily be extended to other NLDR visualization methods, provided a local distortion measure could be calculated. It could also serve as a guiding tool for interactive data visualization.","Title: Assessment of an unsupervised feature selection method for generative topographic mapping\nYear: 2006\n\nAbstract: Feature selection (FS) has long been studied in classification and regression problems. In comparison, FS for unsupervised learning has received far less attention. For many real problems concerning unsupervised data clustering, FS becomes an issue of paramount importance. An unsupervised FS method for Gaussian Mixture Models, based on Feature Relevance Determination (FRD), was recently defined. Unfortunately, the data visualization capabilities of general mixture models are limited. Generative Topographic Mapping (GTM), a constrained mixture model, was originally defined to overcome such limitation. In this brief study, we test in some detail the capabilities of a recently described FRD method for GTM that allows the clustering results to be intuitively visualized and interpreted in terms of a reduced subset of selected relevant features.","Title: Classification, Dimensionality Reduction, and Maximally Discriminatory Visualization of a Multicentre 1H-MRS Database of Brain Tumors\nYear: 2008\n\nAbstract: Time-series segmentation in the fully unsupervised scenario in which the number of segment-types is a priori unknown is a fundamental problem in many applications. We propose a Bayesian approach to a segmentation model based on the switching linear Gaussian ...","Title: Severe sepsis mortality prediction with logistic regression over latent factors\nYear: 2012\n\nAbstract: Sepsis is one of the main causes of death for non-coronary ICU (Intensive Care Unit) patients and has become the 10th most common cause of death in western societies. This is a transversal condition affecting immunocompromised patients, critically ill patients, post-surgery patients, patients with AIDS, and the elderly. In western countries, septic patients account for as much as 25% of ICU bed utilization and the pathology affects 1-2% of all hospitalizations. Its mortality rates range from 12.8% for sepsis to 45.7% for septic shock. The prediction of mortality caused by sepsis is, therefore, a relevant research challenge from a medical viewpoint. The clinical indicators currently in use for this type of prediction have been criticized for their poor prognostic significance. In this study, we redescribe sepsis indicators through latent model-based feature extraction, using factor analysis. These extracted indicators are then applied to the prediction of mortality caused by sepsis. The reported results show that the proposed method improves on the results obtained with the current standard mortality predictor, which is based on the APACHE II score.","Title: Feature and model selection with discriminatory visualization for diagnostic classification of brain tumors\nYear: 2010\n\nAbstract: Machine Learning (ML) and related methods have of late made significant contributions to solving multidisciplinary problems in the field of oncology diagnosis. Human brain tumor diagnosis, in particular, often relies on the use of non-invasive techniques such as Magnetic Resonance Imaging (MRI) and Spectroscopy (MRS). In this paper, MRS data of human brain tumors are analyzed in detail. The high dimensionality of the MR spectra makes difficult both their classification and the interpretation of the obtained results, thus limiting their usability in practical medical settings. The use of dimensionality reduction techniques is therefore advisable. In this work, we apply feature selection methods and several off-the-shelf classifiers on various ^1H-MRS modalities: long and short echo times and an ad hoc combination of both. The introduction of bootstrap resampling techniques permits the obtention of mean performance estimates and their variability. Our experimental findings indicate that the feature selection process enhances the classification performance compared to using the full set of features. We also show that the use of combined information from the different echo times is a better strategy for small numbers of spectral frequencies; however, the use of ever greater numbers of short echo time frequencies permits the obtention of many models with similar performance. The final induced models offer very attractive solutions both in terms of prediction accuracy and number of involved spectral frequencies, which are also amenable to metabolic interpretation. A linear dimensionality-reduction technique that preserves class discrimination capabilities is used for visualizing the data corresponding to the selected frequencies.","Title: SVM-Based Classification of Class C GPCRs from Alignment-Free Physicochemical Transformations of Their Sequences\nYear: 2013\n\nAbstract: G protein-coupled receptors (GPCRs) have a key function in regulating the function of cells due to their ability to transmit extracelullar signals. Given that the 3D structure and the functionality of most GPCRs is unknown, there is a need to construct robust classification models based on the analysis of their amino acid sequences for protein homology detection. In this paper, we describe the supervised classification of the different subtypes of class C GPCRs using support vector machines (SVMs). These models are built on different transformations of the amino acid sequences based on their physicochemical properties. Previous research using semi-supervised methods on the same data has shown the usefulness of such transformations. The obtained classification models show a robust performance, as their Matthews correlation coefficient is close to 0.91 and their prediction accuracy is close to 0.93.","Title: Comparative Evaluation of Semi-supervised Geodesic GTM\nYear: 2009\n\nAbstract: In many real problems that ultimately require data classification, not all the class labels are readily available. This concerns the field of semi-supervised learning, in which missing class labels must be inferred from the available ones as well as from the natural cluster structure of the data. This structure can sometimes be quite convoluted. Previous research has shown the advantage, for these cases, of using the geodesic metric in clustering models of the manifold learning family to reveal the underlying true data structure. In this brief paper, we present a novel semi-supervised approach, namely Semi-Supervised Geo-GTM (SS-Geo-GTM). This is an extension of Geo-GTM, a variation on the Generative Topographic Mapping (GTM) manifold learning model for data clustering and visualization that resorts to the geodesic metric. SS-Geo-GTM uses a proximity graph built from Geo-GTM manifold as the basis for a label propagation algorithm that infers missing class labels. Its performance is compared to those of a semi-supervised version of the standard GTM and of the alternative Laplacian Eigenmaps method.","Title: Discriminant Convex Non-negative Matrix Factorization for the classification of human brain tumours\nYear: 2013\n\nAbstract: The medical analysis of human brain tumours commonly relies on indirect measurements. Among these, magnetic resonance imaging (MRI) and spectroscopy (MRS) predominate in clinical settings as tools for diagnostic assistance. Pattern recognition (PR) methods have successfully been used in this task, usually interpreting diagnosis as a supervised classification problem. In MRS, the acquired spectral signal can be analyzed in an unsupervised manner to extract its constituent sources. Recently, this has been successfully accomplished using Non-negative Matrix Factorization (NMF) methods. In this paper, we present a method to introduce the available class information into the unsupervised source extraction process of a convex variant of NMF. Novel techniques to generate diagnostic predictions for new, unseen spectra using the proposed Discriminant Convex-NMF are also described and experimentally assessed.","Title: On the Initialization of Two-Stage Clustering with Class-GTM\nYear: 2007\n\nAbstract: Generative Topographic Mapping is a probabilistic model for data clustering and visualization. It maps points, considered as prototype representatives of data clusters, from a low dimensional latent space onto the observed data space. In semi-supervised settings, class information can be added resulting in a model variation called class-GTM. The number of class-GTM latent points used is usually large for visualization purposes and does not necessarily reflect the class structure of the data. It is therefore convenient to group the clusters further in a two-stage procedure. In this paper, class-GTM is first used to obtain the basic cluster prototypes. Two novel methods are proposed to use this information as prior knowledge for the K-means-based second stage. We evaluate, using an entropy measure, whether these methods retain the class separability capabilities of class-GTM in the two-stage process, and whether the two-stage procedure improves on the direct clustering of the data using K-means.","Title: On the influence of class information in the two-stage clustering of a human brain tumour dataset\nYear: 2007\n\nAbstract: This paper analyzes, through clustering and visualization, Magnetic Resonance Spectra corresponding to a complex human brain tumour dataset. Clustering is performed as a two-stage process, in which the first stage model is Generative Topographic Mapping (GTM). In semi-supervised settings, class information can be added to refine the clustering process. A class information-enriched variant of GTM, class-GTM, is used here for a first cluster description of the data. The number of clusters used by GTM is usually large for visualization purposes and does not necessarily correspond to the overall class structure. Consequently, in a second stage, clusters are agglomerated using the K-means algorithm with different initialization strategies, some of them defined ad hoc for the GTM models. We aim to evaluate how and under what circumstances the use of class information influences tumour cluster-wise class separability in the final result of the two-stage clustering process.","Title: Brain Tumor Pathological Area Delimitation through Non-negative Matrix Factorization\nYear: 2011\n\nAbstract: Pattern Recognition and Data Mining can provide invaluable insights in the field of neuro oncology. This is because the clinical analysis of brain tumors requires the use of non-invasive methods that generate complex data in electronic format. Magnetic resonance, in the modalities of imaging and spectroscopy, is one of these methods that has been widely applied to this purpose. The heterogeneity of the tissue in the brain volumes analyzed by magnetic resonance remains a challenge in terms of pathological area delimitation. In this brief paper, we show that the Convex-Nonnegative Matrix Factorization technique can be used to extract MRS signal sources that are extremely tissue type-specific and that can be used to delimit these pathological areas with great accuracy.","Title: Bias reduction in skewed binary classification with Bayesian neural networks\nYear: 2000\n\nAbstract: The Bayesian evidence framework has become a standard of good practice for neural network estimation of class conditional probabilities. In this approach the conditional probability is marginalised over the distribution of network weights, which is usually approximated by an analytical expression that moderates the network output towards the midrange. In this paper, it is shown that the network calibration is considerably improved by marginalising to the prior distribution. Moreover, marginalisation to the midrange can seriously bias the estimates of the conditional probabilities calculated from the evidence framework. This is especially the case in the modelling of censored data. (C) 2000 Elsevier Science Ltd. All rights reserved.","Title: Two-Stage Clustering of a Human Brain Tumour Dataset using Manifold Learning Models\nYear: 2008\n\nAbstract: This paper analyzes, through clustering and visualization, Magnetic Resonance spectra of a complex multi-center human brain tumour dataset. Clustering is performed as a two-stage process, in which the models used in the first stage are variants of Generative Topographic Mapping (GTM). Class information-enriched variants of GTM are used to obtain a primary cluster description of the data. The number of clusters used by GTM is usually large and does not necessarily correspond to the overall class structure. Consequently, in a second stage, clusters are agglomerated using K-means with different initialization strategies, some of them defined ad hoc for the GTM models. We evaluate if the use of class information influence the brain tumour cluster-wise class separability resulting from the process. We also resort to a robust variant of GTM that detects outliers while effectively minimizing their negative impact in the clustering process.","Title: Robust analysis of MRS brain tumour data using t-GTM\nYear: 2006\n\nAbstract: This paper proposes a principled, self-organized, framework to manage two sources of uncertainty that are inherent in intelligent systems for medical decision support, namely outliers and missing data. The framework is applied to magnetic resonance spectra (MRS), which are indicators of the grade of malignancy in brain tumours. A model for multivariate data clustering and visualization, the generative topographic mapping (GTM), is re-formulated as a mixture of Student's t-distributions making it more robust to outliers while supporting the imputation of missing values. An important new development is the extension of the model to provide automatic feature relevance determination. Its effectiveness on the MRS data is demonstrated empirically.","Title: Non-negative matrix factorisation methods for the spectral decomposition of MRS data from human brain tumours.\nYear: 2012\n\nAbstract: In-vivo single voxel proton magnetic resonance spectroscopy (SV 1H-MRS), coupled with supervised pattern recognition (PR) methods, has been widely used in clinical studies of discrimination of brain tumour types and follow-up of patients bearing abnormal brain masses. SV 1H-MRS provides useful biochemical information about the metabolic state of tumours and can be performed at short (< 45 ms) or long (> 45 ms) echo time (TE), each with particular advantages. Short-TE spectra are more adequate for detecting lipids, while the long-TE provides a much flatter signal baseline in between peaks but also negative signals for metabolites such as lactate. Both, lipids and lactate, are respectively indicative of specific metabolic processes taking place. Ideally, the information provided by both TE should be of use for clinical purposes. In this study, we characterise the performance of a range of Non-negative Matrix Factorisation (NMF) methods in two respects: first, to derive sources correlated with the mean spectra of known tissue types (tumours and normal tissue); second, taking the best performing NMF method for source separation, we compare its accuracy for class assignment when using the mixing matrix directly as a basis for classification, as against using the method for dimensionality reduction (DR). For this, we used SV 1H-MRS data with positive and negative peaks, from a widely tested SV 1H-MRS human brain tumour database.The results reported in this paper reveal the advantage of using a recently described variant of NMF, namely Convex-NMF, as an unsupervised method of source extraction from SV1H-MRS. Most of the sources extracted in our experiments closely correspond to the mean spectra of some of the analysed tumour types. This similarity allows accurate diagnostic predictions to be made both in fully unsupervised mode and using Convex-NMF as a DR step previous to standard supervised classification. The obtained results are comparable to, or more accurate than those obtained with supervised techniques.The unsupervised properties of Convex-NMF place this approach one step ahead of classical label-requiring supervised methods for the discrimination of brain tumour types, as it accounts for their increasingly recognised molecular subtype heterogeneity. The application of Convex-NMF in computer assisted decision support systems is expected to facilitate further improvements in the uptake of MRS-derived information by clinicians.","Title: Classification of human brain tumours from MRS data using Discrete Wavelet Transform and Bayesian Neural Networks\nYear: 2012\n\nAbstract: The diagnosis of brain tumours is an extremely sensitive and complex clinical task that must rely upon information gathered through non-invasive techniques. One such technique is Magnetic Resonance Spectroscopy. In this task, radiology experts are likely to benefit from the support of computer-based systems built around robust classification processes. In this paper, a Discrete Wavelet Transform procedure was applied to the pre-processing of spectra corresponding to several brain tumour pathologies. This procedure does not alleviate the high dimensionality of the data by itself. For this reason, dimensionality reduction was subsequently implemented using Moving Window with Variance Analysis for feature selection or Principal Component Analysis for feature extraction. The combined method yielded very encouraging results in terms of diagnostic discriminatory binary classification using Bayesian Neural Networks. In most cases, the classification accuracy improved on previously reported results.","Title: Neural network models for language acquisition: a brief survey\nYear: 2006\n\nAbstract: Since the outbreak of connectionist modelling in the mid eighties, several problems in natural language processing have been tackled by employing neural network-based techniques. Neural network’s “biological plausibility” offers a promising framework in which the computational treatment of language may be linked to other disciplines such as cognitive science and psychology. With this brief survey, we set out to explore the landscape of artificial neural models for the acquisition of language that have been proposed in the research literature.","Title: Frequency Selection for the Diagnostic Characterization of Human Brain Tumours\nYear: 2009\n\nAbstract: The diagnosis of brain tumours is an extremely sensitive and complex clinical task that must rely upon information gathered through non-invasive techniques. One such technique is magnetic resonance, in the modalities of imaging or spectroscopy. The latter provides plenty of metabolic information about the tumour tissue, but its high dimensionality makes resorting to pattern recognition techniques advisable. In this brief paper, an international database of brain tumours is analyzed resorting to an ad hoc spectral frequency selection procedure combined with nonlinear classification.","Title: A variational Bayesian approach for the robust analysis of the cortical silent period from EMG recordings of brain stroke patients\nYear: 2011\n\nAbstract: Transcranial magnetic stimulation (TMS) is a powerful tool for the calculation of parameters related to the intracortical excitability and inhibition of the motor cortex. The cortical silent period (CSP) is one such parameter that corresponds to the suppression of muscle activity for a short period after a muscle response to TMS. The duration of the CSP is known to be correlated with the prognosis of brain stroke patients' motor ability. Current methods for the estimation of the CSP duration are very sensitive to the presence of noise. A variational Bayesian formulation of a manifold-constrained hidden Markov model is applied in this paper to the segmentation of a set of multivariate time series (MTS) of electromyographic recordings corresponding to stroke patients and control subjects. A novel index of variability associated to this model is defined and applied to the detection of the silent period interval of the signal and to the estimation of its duration. This model and its associated index are shown to behave robustly in the presence of noise and provide more reliable estimations than the current standard in clinical practice.","Title: Outlier exploration and diagnostic classification of a multi-centre 1H-MRS brain tumour database\nYear: 2009\n\nAbstract: Non-invasive techniques such as magnetic resonance spectroscopy (MRS) are often required for assisting the diagnosis of tumours. Radiologists are not always accustomed to make sense of the biochemical information provided by MRS and they may benefit from computer-based support in their decision making. The high dimensionality of the MR spectra obscures atypical aspects of the data that may jeopardize their classification. In this study, we describe a method to overcome this problem that combines nonlinear dimensionality reduction, outlier detection, and expert opinion. MR spectra subsequently undergo a feature selection process followed by classification. The impact of outlier removal on classification performance is assessed.","Title: Finding Class C GPCR Subtype-Discriminating N-grams through Feature Selection.\nYear: 2014\n\nAbstract: G protein-coupled receptors (GPCRs) are a large and heterogeneous superfamily of receptors that are key cell players for their role as extracellular signal transmitters. Class C GPCRs, in particular, are of great interest in pharmacology. The lack of knowledge about their full 3-D structure prompts the use of their primary amino acid sequences for the construction of robust classifiers, capable of discriminating their different subtypes. In this paper, we describe the use of feature selection techniques to build Support Vector Machine (SVM)-based classification models from selected receptor subsequences described as n-grams. We show that this approach to classification is useful for finding class C GPCR subtype-specific motifs.","Title: Semi-supervised outcome prediction for a type of human brain tumour using partially labeled MRS information\nYear: 2009\n\nAbstract: The diagnosis and prognosis of human brain tumours, especially when they are aggresive, are sensitive clinical tasks that usually require non-invasive measurement techniques. Outcome information for aggressive tumours, in particular, is usually scarce. In this paper, we aim to gauge the capability of a novel semi-supervised model, SS-Geo-GTM, to infer outcome stages from a very limited amount of available stage labels and Magnetic Resonance Spectroscopy (MRS) data corresponding to Glioblastoma, which is an aggressive tumor type. This model stems from a geodesic distance-based extension of Generative Topographic Mapping (Geo-GTM) that prioritizes neighbourhood relationships along a generated manifold embedded in the observed data space.","Title: On The Benefits For Model Regularization Of A Variational Formulation Of Gtm\nYear: 2008\n\nAbstract: Generative Topographic Mapping (GTM) is a manifold learning model for the simultaneous visualization and clustering of multivariate data. It was originally formulated as a constrained mixture of distributions, for which the adaptive parameters were determined by Maximum Likelihood (ML), using the Expectation-Maximization (EM) algorithm. In this formulation, GTM is prone to data overfitting unless a regularization mechanism is included. The theoretical principles of Variational GTM, an approximate method that provides a full Bayesian treatment to a Gaussian Process (GP)-based variation of the GTM, were recently introduced as alternative way to control data overfitting. In this paper we assess in some detail the generalization capabilities of Variational GTM and compare them with those of alternative regularization approaches in terms of test log-likelihood, using several artificial and real datasets.","Title: Neural networks and other machine learning methods in cancer research\nYear: 2007\n\nAbstract: Evidence-based medicine has grown in stature over the last three decades and is now regarded a key development of modern medicine. The evidence base can be heterogeneous, involving both qualitative knowledge and measured quantitative data. Machine Learning (ML) methods have also begun to establish themselves as an alternative and promising approach to computer-based data analysis in oncology, as this field moves gradually away from being the preserve of traditional statistical analysis. In this paper, we describe the main areas of cancer research in which ML methods are currently being applied, and briefly discuss some of the advantages and disadvantages of their application.","Title: Learning what is important: feature selection and rule extraction in a virtual course\nYear: 2006\n\nAbstract: Virtualcampus� environmentsare �becomingamain str eamalternative� to� traditionaldist ancehighereduc ation.�TheInter net �mediumtheyuse �allowsthe� gatheringofinf ormationonstud ents'�usagebeha vio ur.�Theknowledg eextracted� from� this� information� can� be� fed� back� to� the� e/lear ning� environment� to� ease� teachers'�worklo ad.�Inthiscont ext,�twoproblem sa readdressedin� thecurrent� study:�findingw hichusagefeatu resarebestat� pre dictingonlines tudents'�marks,� andexplainingm arkpredictioni ntheformofsi mpl eandinterpreta blerules.�To� thateffect,�two �methodsareuse d:�FuzzyInducti ve� Reasoning�(FIR)� forfeature� selection� and� Orthogonal� Search/Based� Rule� Extraction � (OSRE).� Experiments� carriedoutont heavailabledat aindicatethat� stu dents'�markscan �beaccurately� predicted� and� that� a� small� subset� of� variables� expl ains� the� accuracy� of� such� prediction,�whic hcanbedescrib edthroughaset �of �actionablerule s.��","Title: Automated classification of brain tumours from short echo time in vivo MRS data using Gaussian Decomposition and Bayesian Neural Networks.\nYear: 2014\n\nAbstract: •The paper describes a method for brain tumour type discrimination from MRS signal.•The method combines Gaussian Decomposition, dimensionality reduction and classification using Artificial Neural Networks.•The experiments yield encouraging brain tumour classification results, a difficult task even for expert radiologists.","Title: Semi-supervised geodesic Generative Topographic Mapping\nYear: 2010\n\nAbstract: We present a novel semi-supervised model, SS-Geo-GTM, which stems from a geodesic distance-based extension of Generative Topographic Mapping that prioritizes neighbourhood relationships along a generated manifold embedded in the observed data space. With this, it improves the trustworthiness and the continuity of the low-dimensional representations it provides, while behaving robustly in the presence of noise. In SS-Geo-GTM, the model prototypes are linked by the nearest neighbour to the data manifold constructed by Geo-GTM. The resulting proximity graph is used as the basis for a class label propagation algorithm. The performance of SS-Geo-GTM is experimentally assessed, comparing positively with that of an Euclidean distance-based counterpart and with those of alternative manifold learning methods.","Title: The influence of alignment-free sequence representations on the semi-supervised classification of class C G protein-coupled receptors.\nYear: 2015\n\nAbstract: G protein-coupled receptors (GPCRs) are integral cell membrane proteins of relevance for pharmacology. The tertiary structure of the transmembrane domain, a gate to the study of protein functionality, is unknown for almost all members of class C GPCRs, which are the target of the current study. As a result, their investigation must often rely on alignments of their amino acid sequences. Sequence alignment entails the risk of missing relevant information. Various approaches have attempted to circumvent this risk through alignment-free transformations of the sequences on the basis of different amino acid physicochemical properties. In this paper, we use several of these alignment-free methods, as well as a basic amino acid composition representation, to transform the available sequences. Novel semi-supervised statistical machine learning methods are then used to discriminate the different class C GPCRs types from the transformed data. This approach is relevant due to the existence of orphan proteins to which type labels should be assigned in a process of deorphanization or reverse pharmacology. The reported experiments show that the proposed techniques provide accurate classification even in settings of extreme class-label scarcity and that fair accuracy can be achieved even with very simple transformation strategies that ignore the sequence ordering.","Title: Exploratory Visualization Of Misclassified Gpcrs From Their Transformed Unaligned Sequences Using Manifold Learning Techniques\nYear: 2014\n\nAbstract: Class C G-protein-coupled receptors (GPCRs) are cell membrane proteins of great relevance to biology and pharmacology. Previous research has revealed an upper boundary on the accuracy that can be achieved in their classification into subtypes from the unaligned transformation of their sequences. To investigate this, we focus on sequences that have been misclassified using supervised methods. These are visualized, using a nonlinear dimensionality reduction technique and phylogenetic trees, and then characterized against the rest of the data and, particularly, against the rest of cases of their own subtype. This should help to discriminate between different types of misclassification and to build hypotheses about database quality problems and the extent to which GPCR sequence transformations limit subtype discriminability. The reported experiments provide a proof of concept for the proposed method.","Title: Visual interpretation of class C GPCR subtype overlapping from the nonlinear mapping of transformed primary sequences\nYear: 2014\n\nAbstract: This brief paper addresses the problem of visually assessing the natural discriminability of the different subtypes that characterize class C G-Protein-Coupled Receptors, which are membrane proteins of interest in pharmacology, from diverse transformations of their primary amino acid sequences. Visualization, achieved using nonlinear dimensionality reduction techniques, is used as an exploratory tool to increase the interpretability of the data subtype structure. Combined with a quantitative estimation of subtype overlapping, this approach should be of help to experts in proteomics.","Title: A MAP approach for convex non-negative matrix factorization in the diagnosis of brain tumors\nYear: 2014\n\nAbstract: Convex non-negative matrix factorization is a blind signal separation technique that has previously demonstrated to be well-suited for the task of human brain tumor diagnosis from magnetic resonance spectroscopy data. This is due to its ability to retrieve interpretable sources of mixed sign that highly correlate with tissue type prototypes. The current study provides a Bayesian formulation for such problem and derives a maximum a posteriori estimate based on a gradient descent algorithm specifically designed to deal with matrices with different sign restrictions. Its applicability to neuro-oncology diagnosis was experimentally assessed and the results were found to be comparable to those achieved by state of the art methods in tumor type discrimination and consistently better in source extraction.","Title: Manifold learning visualization of Metabotropic Glutamate Receptors.\nYear: 2014\n\nAbstract: G-Protein-Coupled Receptors (GPCRs) are cell membrane proteins with a key role in biological processes. GPCRs of class C, in particular, are of great interest in pharmacology. The lack of knowledge about their 3-D structures means they must be investigated through their primary amino acid sequences. Sequence visualization can help to explore the existing receptor sub-groupings at different partition levels. In this paper, we focus on Metabotropic Glutamate Receptors (mGluR), a subtype of class C GPCRs. Different versions of a probabilistic manifold learning model are employed to comparatively sub-group and visualize them through different transformations of their sequences.","Title: Probability Ridges and Distortion Flows: Visualizing Multivariate Time Series Using a Variational Bayesian Manifold Learning Method.\nYear: 2014\n\nAbstract: Time-dependent natural phenomena and artificial processes can often be quantitatively expressed as multivariate time series (MTS). As in any other process of knowledge extraction from data, the analyst can benefit from the exploration of the characteristics of MTS through data visualization. This visualization often becomes difficult to interpret when MTS are modelled using nonlinear techniques. Despite their flexibility, nonlinear models can be rendered useless if such interpretability is lacking. In this brief paper, we model MTS using Variational Bayesian Generative Topographic Mapping Through Time (VB-GTM-TT), a variational Bayesian variant of a constrained hidden Markov model of the manifold learning family defined for MTS visualization. We aim to increase its interpretability by taking advantage of two results of the probabilistic definition of the model: the explicit estimation of probabilities of transition between states described in the visualization space and the quantification of the nonlinear mapping distortion.","Title: A variational formulation for GTM through time\nYear: 2008\n\nAbstract: Generative topographic mapping (GTM) is a latent variable model that, in its original version, was conceived to provide clustering and visualization of multivariate, real-valued, i.i.d. data. It was also extended to deal with noni. i.d. data such as multivariate time series in a variant called GTM through time (GTM-TT), defined as a constrained hidden Markov model (HMM). In this paper, we provide the theoretical foundations of the reformulation of GTM-TT within the variational Bayesian framework and provide an illustrative example of its application. This approach handles the presence of noise in the time series, helping to avert the problem of data overfitting.","Title: Label noise in subtype discrimination of class C G protein-coupled receptors: A systematic approach to the analysis of classification errors\nYear: 2015\n\nAbstract: The characterization of proteins in families and subfamilies, at different levels, entails the definition and use of class labels. When the adscription of a protein to a family is uncertain, or even wrong, this becomes an instance of what has come to be known as a label noise problem. Label noise has a potentially negative effect on any quantitative analysis of proteins that depends on label information. This study investigates class C of G protein-coupled receptors, which are cell membrane proteins of relevance both to biology in general and pharmacology in particular. Their supervised classification into different known subtypes, based on primary sequence data, is hampered by label noise. The latter may stem from a combination of expert knowledge limitations and the lack of a clear correspondence between labels that mostly reflect GPCR functionality and the different representations of the protein primary sequences.","Title: Making nonlinear manifold learning models interpretable: The manifold grand tour\nYear: 2015\n\nAbstract: •Smooth nonlinear topographic maps of the data distribution to guide a Grand Tour visualisation.•Prioritisation of data linear views that are most consistent with data structure in the maps.•Useful visualisations that cannot be obtained by other more classical approaches.","Title: Quantitative characterization and prediction of on-line purchasing behavior: a latent variable approach\nYear: 2000\n\nAbstract: Realizing the full potential of the on-line consumer market requires careful identification of customer needs and expectations. As research on Internet consumer behavior is still in its infancy, a quantitative framework to characterize user profiles for e-commerce has not yet been established. This study proposes a quantitative framework that uses factor analysis to identify latent factor descriptors of Internet users' opinions on Web vendors and on-line shopping. Predictive models based on logistic discrimination and neural networks then select the factors most predictive of the propensity to buy on-line and classify Internet users accordingly. The application of this framework shows that the obtained latent factors agree in general with the major indicators identified in previous qualitative research. A small subset of the obtained factors is shown to retain the predictive power of the whole set. Neural networks are found to perform only marginally better than logistic discrimination in the task of classification.","Title: Random Forests for Quality Control in G-Protein Coupled Receptor Databases.\nYear: 2016\n\nAbstract: G protein-coupled receptors are a large and heterogeneous super-family of cell membrane proteins of interest to biology in general. One of its families, class C, is of particular interest to pharmacology and drug design. This family is quite heterogeneous on its own, and the discrimination of its several sub-families is a challenging problem. In the absence of known crystal structure, such discrimination must rely on their primary amino acid sequences. In this study, we are interested not as much in achieving maximum sub-family discrimination accuracy, but in exploring sequence misclassification behaviour. Specifically, we are interested in isolating those sequences showing consistent misclassification, that is, sequences that are very often misclassified and almost always to the same wrong sub-family. This analysis should assist database curators in receptor quality control tasks. Random Forests are used for this analysis due their ensemble nature, which makes them naturally suited to gauge the consistency of misclassification.","Title: Automated Quality Control for Proton Magnetic Resonance Spectroscopy Data Using Convex Non-negative Matrix Factorization.\nYear: 2016\n\nAbstract: Proton Magnetic Resonance Spectroscopy (H-1 MRS) has proven its diagnostic potential in a variety of conditions. However, MRS is not yet widely used in clinical routine because of the lack of experts on its diagnostic interpretation. Although data-based decision support systems exist to aid diagnosis, they often take for granted that the data is of good quality, which is not always the case in a real application context. Systems based on models built with bad quality data are likely to underperform in their decision support tasks. In this study, we propose a system to filter out such bad quality data. It is based on convex Non-Negative Matrix Factorization models, used as a dimensionality reduction procedure, and on the use of several classifiers to discriminate between good and bad quality data.","Title: Metrics For Probabilistic Geometries\nYear: 2014\n\nAbstract: We investigate the geometrical structure of probabilistic generative dimensionality reduction models using the tools of Riemannian geometry. We explicitly define a distribution over the natural metric given by the models. We provide the necessary algorithms to compute expected metric tensors where the distribution over mappings is given by a Gaussian process. We treat the corresponding latent variable model as a Riemannian manifold and we use the expectation of the metric under the Gaussian process prior to define interpolating paths and measure distance between latent points. We show how distances that respect the expected metric lead to more appropriate generation of new data.","Title: Visual Exploratory Assessment of Class C GPCR Extracellular Domains Discrimination Capabilities.\nYear: 2016\n\nAbstract: G protein-coupled receptors (GPCRs) are integral membrane-bound proteins. They are divided in five main classes with Class C members receiving recently special attention because of their involvement in many neurologic diseases. These receptors are composed of the seven-helix transmembrane domain (typical of all GPCRs) and a large extracellular domain where the endogenous ligand binds. In the present absence of crystal structures for complete Class C receptors, their primary sequences can provide limited but useful information that can be used for subtype discrimination in first instance. In this paper, we show that the extracellular part of these sequences provides as effective a discrimination as the complete sequence. With that purpose, we describe a process of exploratory sequence visualization using different data transformations and manifold learning techniques for dimensionality reduction. Class discriminability is assessed using an entropy-based measure.","Title: Computational Intelligence in Interior Design: State-of-the-Art and Outlook.\nYear: 2016\n\nAbstract: Tasks in architectural and interior design range from defining the building floor plans and ensuring desired functionality, to deciding furnishing styles and arrangement choices. The process of design, as a whole, has remained hard to master for computer-based optimization in general and for computational intelligence approaches in particular. Numerous attempts to tackle different subfields of this problem in a machine learning fashion have emerged over the last few years. In this paper, we present an overview of current advances of computational intelligence in architectural science with a focus on interior design. This is accompanied by the description of ongoing research towards the development of a commercial robust and scalable solution for automatic furniture arrangement.","Title: Intelligent data analysis approaches to churn as a business problem: a survey.\nYear: 2017\n\nAbstract: Globalization processes and market deregulation policies are rapidly changing the competitive environments of many economic sectors. The appearance of new competitors and technologies leads to an increase in competition and, with it, a growing preoccupation among service-providing companies with creating stronger customer bonds. In this context, anticipating the customer's intention to abandon the provider, a phenomenon known as churn, becomes a competitive advantage. Such anticipation can be the result of the correct application of information-based knowledge extraction in the form of business analytics. In particular, the use of intelligent data analysis, or data mining, for the analysis of market surveyed information can be of great assistance to churn management. In this paper, we provide a detailed survey of recent applications of business analytics to churn, with a focus on computational intelligence methods. This is preceded by an in-depth discussion of churn within the context of customer continuity management. The survey is structured according to the stages identified as basic for the building of the predictive models of churn, as well as according to the different types of predictive methods employed and the business areas of their application.","Title: Machine Learning for Critical Care: An Overview and a Sepsis Case Study.\nYear: 2017\n\nAbstract: Biology in general and medicine and healthcare in particular are facing the critical challenge of exponentially increasing data availability. The core of this challenge is putting these data to work through computer-based knowledge extraction methods. In the medical context this could take the form of medical decision support systems for diagnosis, prognosis or general management. Arguably, one of the most data dependent clinical environments is the critical care unit and by extension the whole area of critical care. Fresh approaches to data analysis in critical care are required, and Computational Intelligence and Machine Learning methods have already shown their usefulness in tackling problems in the area. This brief paper aims to be an introduction to the use of such methods in critical care.","Title: Topological Sequence Segments Discriminate Between Class C GPCR Subtypes.\nYear: 2017\n\nAbstract: G protein-coupled receptors are eukaryotic cell membrane proteins with a key role as extracellular signal transmitters. While GPCRs embrace a wide and heterogeneous super-family of proteins, our interest in this study is in its Class C, of great relevance to pharmacology. The scarcity of knowledge about their full 3-D crystal structure makes the use of their primary amino acid sequences important for analysis. In this paper, we systematically analyze whether segments of the receptor sequences are able to discriminate between the different class C GPCR subtypes according to their topological location on the extracellular, transmembrane or intracellular domain. For this, we build on previous research that showed that the use of the extracellular N-terminus domain on its own for this classification task did only entail a minor decrease in subtype discrimination when compared to the complete sequence. We use Support Vector Machine-based classification models to assess the subtype discriminating power of the topological segments.","Title: Discovering Subtype Specific n-Gram Motifs in Class C GPCR N-Termini.\nYear: 2017\n\nAbstract: Protein-Coupled Receptors are a family of cell membrane proteins, whose class C is a current main target for drug development. Their primary sequences are studied as a source of information for the characterization of their behaviour. In previous research, different alignment-free sequence transformations were explored as the basis for the supervised discrimination of the various class C subtypes using Support Vector Machines. We also investigated an alignment-free sequence transformation based on n-gram protein motifs, under the hypothesis that the sequences' extra-cellular N-terminus domain could suffice to retain most of the subtype-discrimination capabilities of the complete receptor, and that a parsimonious selection of n-grams would be responsible for such classification success. In the current study, these previous results are extended by investigating a different classification procedure that now employs a subtype-vs-all the rest of subtypes approach, shifting towards the selection of those sequence motifs that distinguish each class C subtype from the rest. The reported results indicate the adequacy of this new approach, both in terms of discrimination ability and motif selection parsimony.","Title: Bioinformatics and Medicine in the Era of Deep Learning.\nYear: 2018\n\nAbstract: Many of the current scientific advances in the life sciences have their origin in the intensive use of data for knowledge discovery. In no area this is so clear as in bioinformatics, led by technological breakthroughs in data acquisition technologies. It has been argued that bioinformatics could quickly become the field of research generating the largest data repositories, beating other data-intensive areas such as high-energy physics or astroinformatics. Over the last decade, deep learning has become a disruptive advance in machine learning, giving new live to the long-standing connectionist paradigm in artificial intelligence. Deep learning methods are ideally suited to large-scale data and, therefore, they should be ideally suited to knowledge discovery in bioinformatics and biomedicine at large. In this brief paper, we review key aspects of the application of deep learning in bioinformatics and medicine, drawing from the themes covered by the contributions to an ESANN 2018 special session devoted to this topic.","Title: Bayesian semi non-negative matrix factorisation.\nYear: 2016\n\nAbstract: Non-negative Matrix Factorisation (NMF) has become a standard method for source identification when data, sources and mixing coefficients are constrained to be positive-valued. The method has recently been extended to allow for negative-valued data and sources in the form of Semi-and Convex-NMF. In this paper, we re-elaborate Semi-NMF within a full Bayesian framework. This provides solid foundations for parameter estimation and, importantly, a principled method to address the problem of choosing the most adequate number of sources to describe the observed data. The proposed Bayesian Semi-NMF is preliminarily evaluated here in a real neuro-oncology problem.","Title: Physics and Machine Learning: Emerging Paradigms.\nYear: 2016\n\nAbstract: Current research in Machine Learning (ML) combines the study of variations on well-established methods with cutting-edge breakthroughs based on completely new approaches. Among the latter, emerging paradigms from Physics have taken special relevance in recent years. Although still in its initial stages, Quantum Machine Learning (QML) shows promising ways to speed up some of the costly ML calculations with a similar or even better performance than existing approaches. Two additional advantages are related to the intrinsic probabilistic approach of QML, since quantum states are genuinely probabilistic, and to the capability of finding the global optimum of a given cost function by means of adiabatic quantum optimization, thus circumventing the usual problem of local minima. Another Physics approach for ML comes from Statistical Physics and is linked to Information theory in supervised and semi-supervised learning frameworks. On the other hand, and from the perspective of Physics, ML can provide solutions by extracting knowledge from huge amounts of data, as it is common in many experiments in the field, such as those related to High Energy Physics for elementary-particle research and Observational Astronomy.","Title: A machine learning pipeline for supporting differentiation of glioblastomas from single brain metastases.\nYear: 2016\n\nAbstract: Machine learning has provided, over the last decades, tools for knowledge extraction in complex medical domains. Most of these tools, though, are ad hoc solutions and lack the systematic approach that would be required to become mainstream in medical practice. In this brief paper, we define a machine learning-based analysis pipeline for helping in a difficult problem in the field of neuro-oncology, namely the discrimination of brain glioblastomas from single brain metastases. This pipeline involves source extraction using k-Meansinitialized Convex Non-negative Matrix Factorization and a collection of classifiers, including Logistic Regression, Linear Discriminant Analysis, AdaBoost, and Random Forests.","Title: The Importance Of Interpretability And Visualization In Machine Learning For Applications In Medicine And Health Care\nYear: 2020\n\nAbstract: In a short period of time, many areas of science have made a sharp transition towards data-dependent methods. In some cases, this process has been enabled by simultaneous advances in data acquisition and the development of networked system technologies. This new situation is particularly clear in the life sciences, where data overabundance has sparked a flurry of new methodologies for data management and analysis. This can be seen as a perfect scenario for the use of machine learning and computational intelligence techniques to address problems in which more traditional data analysis approaches might struggle. But, this scenario also poses some serious challenges. One of them is model interpretability and explainability, especially for complex nonlinear models. In some areas such as medicine and health care, not addressing such challenge might seriously limit the chances of adoption, in real practice, of computer-based systems that rely on machine learning and computational intelligence methods for data analysis. In this paper, we reflect on recent investigations about the interpretability and explainability of machine learning methods and discuss their impact on medicine and health care. We pay specific attention to one of the ways in which interpretability and explainability in this context can be addressed, which is through data and model visualization. We argue that, beyond improving model interpretability as a goal in itself, we need to integrate the medical experts in the design of data analysis interpretation strategies. Otherwise, machine learning is unlikely to become a part of routine clinical and health care practice.","Title: Visual Mining of Industrial Gas Turbines Sensor Data as an Industry 4.0 Application\nYear: 2021\n\nAbstract: Industrial gas turbines for power generation are advanced engines that require constant and detailed monitorization using internal and external sensors. These sensors generate a large flow of data in the form of multivariate time series that are amenable to analysis using pattern recognition methods with the objective of improving and optimizing turbine operation. One aspect this may take is visual analytics, where dimensionality reduction methods can be used to intuitively visualize the multivariate time series. This brief paper provides a proof of concept and some case scenarios of a visual turbine-monitorization tool, based on the UMAP method, combined with clustering using HDBSCAN and unsupervised Agnostic Feature Selection. It can be considered as a first step towards a data-centered approach to gas turbine management within the industry 4.0 framework, based on the mining of turbine sensor data.","Title: Long Short-Term Memory to Predict 3D Amino Acids Positions in GPCR Molecular Dynamics.\nYear: 2022\n\nAbstract: G-Protein Coupled Receptors (GPCRs) are a big family of eukaryotic cell transmembrane proteins, responsible for numerous biological processes. From a practical viewpoint around 34\\% of the drugs approved by the US Food and Drug Administration target these receptors. They can be analyzed from their simulated molecular dynamics, including the prediction of their behavior in the presence of drugs. In this paper, the capability of Long Short-Term Memory Networks (LSTMs) are evaluated to learn and predict the molecular dynamic trajectories of a receptor. Several models were trained with the 3D position of the amino acids of the receptor considering different transformations on the position of the amino acid, such as their centers of mass, the geometric centers and the position of the $\\alpha$--carbon for each amino acid. The error of the prediction of the position was evaluated by the mean average error (MAE) and root-mean-square deviation (RMSD). The LSTM models show a robust performance, with results comparable to the state-of-the-art in non-dynamic 3D predictions. The best MAE and RMSD values were found for the mass center of the amino acids with 0.078 {\\AA} and 0.156 {\\AA} respectively. This work shows the potential of LSTM to predict the molecular dynamics of GPRCs.","Title: Molecular Dynamics forecasting of transmembrane Regions in GPRCs by Recurrent Neural Networks\nYear: 2022\n\nAbstract: G protein-coupled receptors are a large super-family of cell membrane proteins that play an important physiological role as transmitters of extra-cellular signals. Signal transmission through the cell membrane depends on the conformational changes of the transmembrane region of the receptor and the investigation of the dynamics in these regions is therefore key. Molecular Dynamics (MD) simulations can provide information of the receptor conformational states at the atom level and machine learning (ML) methods can be useful for the analysis of these data. In this paper, Recurrent Neural Networks (RNNs) are used to evaluate whether the MD can be modeled focusing on the different regions of the receptor (intra-cellular, extra-cellular and each transmembrane regions (TM)). The best results, as measured by root-mean-square deviation (RMSD), are 0.1228 Å for TM4 of the 2rh1 (inactive state) and 0.1325 Å for TM4 of the 3p0g (active state), which are comparable to the state-of-the-art in non-dynamic 3-D predictions, showing the potential of the proposed approach."],"string":"[\n \"Title: The extracellular N-terminal domain suffices to discriminate class C G Protein-Coupled Receptor subtypes from n-grams of their sequences\\nYear: 2015\\n\\nAbstract: The investigation of protein functionality often relies on the knowledge of crystal 3-D structure. This structure is not always known or easily unravelled, which is the case of eukaryotic cell membrane proteins such as G Protein-Coupled Receptors (GPCRs) and specially of those of class C, which are the target of the current study. In the absence of information about tertiary or quaternary structures, functionality can be investigated from the primary structure, that is, from the amino acid sequence. In previous research, we found that the different subtypes of class C GPCRs could be discriminated with a high level of accuracy from the n-gram transformation of their complete primary sequences, using a method that combined two-stage feature selection with kernel classifiers. This study aims at discovering whether subunits of the complete sequence retain such discrimination capabilities. We report experiments that show that the extracellular N-terminal domain of the receptor suffices to retain the classification accuracy of the complete sequence and that it does so using a reduced selection of n-grams whose length of up to five amino acids opens up an avenue for class C GPCR signature motif discovery.\",\n \"Title: Advances in clustering and visualization of time series using GTM through time.\\nYear: 2008\\n\\nAbstract: Most of the existing research on multivariate time series concerns supervised forecasting problems. In comparison, little research has been devoted to their exploration through unsupervised clustering and visualization. In this paper, the capabilities of Generative Topographic Mapping Through Time, a model with foundations in probability theory, that performs simultaneous time series clustering and visualization, are assessed in detail. Focus is placed on the visualization of the evolution of signal regimes and the exploration of sudden transitions, for which a novel identification index is defined. The interpretability of time series clustering results may become extremely difficult, even in exploratory visualization, for high dimensional datasets. Here, we define and test an unsupervised time series relevance determination method, fully integrated in the Generative Topographic Mapping Through Time model, that can be used as a basis for time series selection. This method should ease the interpretation of time series clustering results.\",\n \"Title: Machine learning in cancer research: implications for personalised medicine\\nYear: 2008\\n\\nAbstract: Driven by the growing demand of personalization of medical procedures, data-based, computer-aided cancer research in human patients is advancing at an accelerating pace, providing a broadening landscape of opportunity for Machine Learning methods. This landscape can be observed from the wide-reaching view of population studies down to the genotype detail. In this brief paper, we provide a sweeping glimpse, by no means exhaustive, of the state-of-the-art in this field at the different scales of data measurement and analysis.\",\n \"Title: Feature Selection with Single-Layer Perceptrons for a Multicentre 1H-MRS Brain Tumour Database\\nYear: 2009\\n\\nAbstract: A Feature Selection process with Single-Layer Perceptrons is shown to provide optimum discrimination of an international, multi-centre 1H-MRS database of brain tumors at reasonable computational cost. Results are both intuitively interpretable and very accurate. The method remains simple enough as to allow its easy integration in existing medical decision support systems.\",\n \"Title: <Emphasis Type=\\\"Italic\\\">An Electronic Commerce Application of the Bayesian Framework for MLPs: The Effect of Marginalisation and ARD</Emphasis>\\nYear: 2001\\n\\nAbstract: Automatic Relevance Determination (ARD), with multiple regularisation terms. A real-world classification problem, concerning the discrimination of online purchasers and non-purchasers using Internet’s WWW users’ opinions, is the test-bed for this assessment.\",\n \"Title: Describing customer loyalty to spanish petrol stations through rule extraction\\nYear: 2006\\n\\nAbstract: Globalization and deregulation are modifying the competitive framework in the majority of economic sectors and, as a result, many companies are changing their commercial model to focus on the preservation of existing customers. Understanding customer loyalty therefore represents an element of competitive advantage. In this brief paper, we investigate loyalty in the Spanish petrol station market, according to the customer satisfaction and switching barriers constructs. Satisfaction and behavioural intentions are analysed within a classification framework using Bayesian neural networks. The necessary interpretability and actionability of the results is achieved through the use of a feature selection process embedded in the network training and a novel rule extraction method.\",\n \"Title: Semi-Supervised Analysis Of Human Brain Tumours From Partially Labeled Mrs Information, Using Manifold Learning Models\\nYear: 2011\\n\\nAbstract: Medical diagnosis can often be understood as a classification problem. In oncology, this typically involves differentiating between tumour types and grades, or some type of discrete outcome prediction. From the viewpoint of computer-based medical decision support, this classification requires the availability of accurate diagnoses of past cases as training target examples. The availability of such labeled databases is scarce in most areas of oncology, and especially so in neuro-oncology. In such context, semi-supervised learning oriented towards classification can be a sensible data modeling choice. In this study, semi-supervised variants of Generative Topographic Mapping, a model of the manifold learning family, are applied to two neuro-oncology problems: the diagnostic discrimination between different brain tumour pathologies, and the prediction of outcomes for a specific type of aggressive brain tumours. Their performance compared favorably with those of the alternative Laplacian Eigenmaps and Semi-Supervised SVM for Manifold Learning models in most of the experiments.\",\n \"Title: Handling outliers in brain tumour MRS data analysis through robust topographic mapping.\\nYear: 2006\\n\\nAbstract: Uncertainty is inherent in medical decision making and poses a challenge for intelligent technologies. This paper focuses on magnetic resonance spectra (MRS) for discrimination of brain tumour types and grades. Modelling of this type of high-dimensional data is commonly affected by uncertainty caused by the presence of outliers. Multivariate data clustering and visualization of MRS data is proposed using the GTM framework with basis functions comprising Student t-distributions in order to minimize the negative impact on the model from outliers. The effectiveness of this model on the MRS data is demonstrated empirically.\",\n \"Title: Sepsis mortality prediction with the Quotient Basis Kernel.\\nYear: 2014\\n\\nAbstract: Objective: This paper presents an algorithm to assess the risk of death in patients with sepsis. Sepsis is a common clinical syndrome in the intensive care unit (ICU) that can lead to severe sepsis, a severe state of septic shock or multi-organ failure. The proposed algorithm may be implemented as part of a clinical decision support system that can be used in combination with the scores deployed in the ICU to improve the accuracy, sensitivity and specificity of mortality prediction for patients with sepsis. Methodology: In this paper, we used the Simplified Acute Physiology Score (SAPS) for ICU patients and the Sequential Organ Failure Assessment (SOFA) to build our kernels and algorithms. In the proposed method, we embed the available data in a suitable feature space and use algorithms based on linear algebra, geometry and statistics for inference. We present a simplified version of the Fisher kernel (practical Fisher kernel for multinomial distributions), as well as a novel kernel that we named the Quotient Basis Kernel (QBK). These kernels are used as the basis for mortality prediction using soft-margin support vector machines. The two new kernels presented are compared against other generative kernels based on the Jensen Shannon metric (centred, exponential and inverse) and other widely used kernels (linear, polynomial and Gaussian). Clinical relevance is also evaluated by comparing these results with logistic regression and the standard clinical prediction method based on the initial SAPS score. Results: As described in this paper, we tested the new methods via cross-validation with a cohort of 400 test patients. The results obtained using our methods compare favourably with those obtained using alternative kernels (80.18% accuracy for the QBK) and the standard clinical prediction method, which are based on the basal SAPS score or logistic regression (71.32% and 71.55%, respectively). The QBK presented a sensitivity and specificity of 79.34% and 83.24%, which outperformed the other kernels analysed, logistic regression and the standard clinical prediction method based on the basal SAPS score. Conclusion: Several scoring systems for patients with sepsis have been introduced and developed over the last 30 years. They allow for the assessment of the severity of disease and provide an estimate of in-hospital mortality. Physiology-based scoring systems are applied to critically ill patients and have a number of advantages over diagnosis-based systems. Severity score systems are often used to stratify critically ill patients for possible inclusion in clinical trials. In this paper, we present an effective algorithm that combines both scoring methodologies for the assessment of death in patients with sepsis that can be used to improve the sensitivity and specificity of the currently available methods. (C) 2014 Elsevier B.V. All rights reserved.\",\n \"Title: DSS-oriented exploration of a multi-centre magnetic resonance spectroscopy brain tumour dataset through visualization\\nYear: 2008\\n\\nAbstract: The exploration of brain tumours usually requires non- invasive techniques such as Magnetic Resonance Imaging (MRI) or Mag- netic Resonance Spectroscopy (MRS). While radiologists are used to inter- preting MRI, many of them are not used to the biochemical information provided by MRS. In this situation, radiologists may benefit from the use of computer-based support for their decisions. As part of the development of a medical Decision Support System (DSS), MRS data corresponding to various tumour pathologies are used to assist expert diagnosis. The high dimensionality of the data might obscure peculiarities and anomalies that would jeopardize automated DSS diagnostic assistance. We illustrate how visualization, combined with expert opinion, can be used to explore data in a process that should improve computer-based tumour classification.\",\n \"Title: Comparative assessment of the robustness of missing data imputation through generative topographic mapping\\nYear: 2005\\n\\nAbstract: The incompleteness of data is a most common source of uncertainty in real-world Data Mining applications. The management of this uncertainty is, therefore, a task of paramount importance for the data analyst. Many methods have been developed for missing data imputation, but few of them approach the problem of imputation as part of a general data density estimation scheme. Amongst the latter, a method for imputing and visualizing multivariate missing data using Generative Topographic Mapping was recently presented. This model and some of its extensions are tested under different experimental conditions. Its performance is compared to that of other missing data imputation techniques, thus assessing its relative capabilities and limitations.\",\n \"Title: Rule-Based Assistance to Brain Tumour Diagnosis Using LR-FIR\\nYear: 2008\\n\\nAbstract: This paper describes a process of rule-extraction from a multi-centre brain tumour database consisting of nuclear magnetic resonance spectroscopic signals. The expert diagnosis of human brain tumours can benefit from computer-aided assistance, which has to be readily interpretable by clinicians. Interpretation can be achieved through rule extraction, which is here performed using the LR-FIR algorithm, a method based on fuzzy logic. The experimental results of the classification of three groups of tumours indicate in this study that just three spectral frequencies, out of the 195 from a range pre-selected by experts, are enough to represent, in a simple and intuitive manner, most of the knowledge required to discriminate these groups.\",\n \"Title: Time series relevance determination through a topology-constrained hidden markov model\\nYear: 2006\\n\\nAbstract: Most of the existing research on multivariate time series concerns supervised forecasting problems. In comparison, little research has been devoted to unsupervised methods for the visual exploration of this type of data. The interpretability of time series clustering results may be difficult, even in exploratory visualization, for high dimensional datasets. In this paper, we define and test an unsupervised time series relevance determination method for Generative Topographic Mapping Through Time, a topology-constrained Hidden Markov Model that performs simultaneous time series data clustering and visualization. This relevance determination method can be used as a basis for time series selection, and should ease the interpretation of the time series clustering results.\",\n \"Title: On the improvement of brain tumour data clustering using class information\\nYear: 2006\\n\\nAbstract: Cluster analysis can benefit from the use of class labels, if these are available, in a semi-supervised approach. In this study, we propose the integration of class information in the clustering of Magnetic Resonance Spectra (MRS) corresponding to human brain tumours using an extension of Generative Topographic Mapping (GTM) that behaves robustly in the presence of outliers.\",\n \"Title: Determination of feature relevance for the grouping of motor unit action potentials through a generative mixture model.\\nYear: 2007\\n\\nAbstract: The study of motor unit action potential (MUAP) activity from electromyographic signals is important for neurological investigations aiming to understand the state of the neuromuscular system. In this context, the identification and clustering of MUAPs that exhibit common characteristics, and the assessment of which data features are most relevant for the definition of such cluster structure, are central issues. In this paper, we propose the application of an unsupervised feature relevance determination (FRD) method to the analysis of experimental MUAPs. This method is embedded in a constrained mixture of distributions model that simultaneously performs data clustering and visualization. The experimental results of the analysis of a data set consisting of MUAPs measured from the First Dorsal Interosseous, a hand muscle, indicate that the features corresponding to the hyperpolarization period in the physiological process of generating muscle fibre action potentials are consistently estimated to be the most relevant. Moreover, the MUAP cluster structure of the data is shown to be only partially attributable to inter-subject differences, with the hyperpolarization period providing the best discrimination of the data by subject.\",\n \"Title: The Effect Of Noise And Sample Size On An Unsupervised Feature Selection Method For Manifold Learning\\nYear: 2008\\n\\nAbstract: The research on unsupervised feature selection is scarce in comparison to that for supervised models, despite the fact that this is an important issue for many clustering problems. An unsupervised feature selection method for general Finite Mixture Models was recently proposed and subsequently extended to Generative Topographic Mapping (GTM), a manifold learning constrained mixture model that provides data visualization. Some of the results of a previous partial assessment of this unsupervised feature selection method for GTM suggested that its performance may be affected by insufficient sample size and by noisy data. In this brief study, we test in some detail such limitations of the method.\",\n \"Title: Missing data imputation through GTM as a mixture of t-distributions.\\nYear: 2006\\n\\nAbstract: The Generative Topographic Mapping (GTM) was originally conceived as a probabilistic alternative to the well-known, neural network-inspired, Self-Organizing Maps. The GTM can also be interpreted as a constrained mixture of distribution models. In recent years, much attention has been directed towards Student t-distributions as an alternative to Gaussians in mixture models due to their robustness towards outliers. In this paper, the GTM is redefined as a constrained mixture of t-distributions: the t-GTM, and the Expectation-Maximization algorithm that is used to fit the model to the data is modified to carry out missing data imputation. Several experiments show that the t-GTM successfully detects outliers, while minimizing their impact on the estimation of the model parameters. It is also shown that the t-GTM provides an overall more accurate imputation of missing values than the standard Gaussian GTM.\",\n \"Title: Finding Relevant Features to Characterize Student Behavior on an e-Learning System\\nYear: 2005\\n\\nAbstract: E-learning systems such as virtual campus environments have established themselves as a strong alternative to traditional distance university education. In this context, Internet allows the gathering of plenty of information on students' online behavior. The knowledge extracted from this information can be used to define personalization strategies tailored to the students' needs and requirements. In this brief study we introduce a model to assess the relative relevance of individual data features on the grouping structure of the users of a real virtual campus (Open University of Catalonia). Tests carried out on these data indicate that certain web navigation behaviors are best at explaining and discriminating the different typologies of online students.\",\n \"Title: Advances In Semi-Supervised Alignment-Free Classification Of G Protein-Coupled Receptors\\nYear: 2013\\n\\nAbstract: G Protein-coupled receptors (GPCRs) are integral cell membrane proteins of great relevance for pharmacology due to their role in transducing extracellular signals. The 3-D structure is unknown for most of them, and the investigation of their structure-function relationships usually relies on the construction of 3-D receptor models from amino acid sequence alignment onto those receptors of known structure. Sequence alignment risks the loss of relevant information. Different approaches have attempted the analysis of alignment-free sequences on the basis of amino acid physicochemical properties. In this paper, we use the Auto-Cross Covariance method and compare it to an amino acid composition representation. Novel semi-supervised manifold learning methods are then used to classify the several members of class C GPCRs on the basis of the transformed data. This approach is relevant because protein sequences are not always labeled and methods that provide robust classification for a limited amount of labels are required.\",\n \"Title: Cartogram visualization for nonlinear manifold learning models\\nYear: 2013\\n\\nAbstract: Real-world applications of multivariate data analysis often stumble upon the barrier of interpretability. Simple data analysis methods are usually easy to interpret, but they risk providing poor data models. More involved methods may instead yield faithful data models, but limited interpretability. This is the case of linear and nonlinear methods for multivariate data visualization through dimensionality reduction. Even though the latter have provided some of the most exciting visualization developments, their practicality is hindered by the difficulty of explaining them in an intuitive manner. The interpretability, and therefore the practical applicability, of data visualization through nonlinear dimensionality reduction (NLDR) methods would improve if, first, we could accurately calculate the distortion introduced by these methods in the visual representation and, second, if we could faithfully reintroduce this distortion into such representation. In this paper, we describe a technique for the reintroduction of the distortion into the visualization space of NLDR models. It is based on the concept of density-equalizing maps, or cartograms, recently developed for the representation of geographic information. We illustrate it using Generative Topographic Mapping (GTM), a nonlinear manifold learning method that can provide both multivariate data visualization and a measure of the local distortion that the model generates. Although illustrated here with GTM, it could easily be extended to other NLDR visualization methods, provided a local distortion measure could be calculated. It could also serve as a guiding tool for interactive data visualization.\",\n \"Title: Assessment of an unsupervised feature selection method for generative topographic mapping\\nYear: 2006\\n\\nAbstract: Feature selection (FS) has long been studied in classification and regression problems. In comparison, FS for unsupervised learning has received far less attention. For many real problems concerning unsupervised data clustering, FS becomes an issue of paramount importance. An unsupervised FS method for Gaussian Mixture Models, based on Feature Relevance Determination (FRD), was recently defined. Unfortunately, the data visualization capabilities of general mixture models are limited. Generative Topographic Mapping (GTM), a constrained mixture model, was originally defined to overcome such limitation. In this brief study, we test in some detail the capabilities of a recently described FRD method for GTM that allows the clustering results to be intuitively visualized and interpreted in terms of a reduced subset of selected relevant features.\",\n \"Title: Classification, Dimensionality Reduction, and Maximally Discriminatory Visualization of a Multicentre 1H-MRS Database of Brain Tumors\\nYear: 2008\\n\\nAbstract: Time-series segmentation in the fully unsupervised scenario in which the number of segment-types is a priori unknown is a fundamental problem in many applications. We propose a Bayesian approach to a segmentation model based on the switching linear Gaussian ...\",\n \"Title: Severe sepsis mortality prediction with logistic regression over latent factors\\nYear: 2012\\n\\nAbstract: Sepsis is one of the main causes of death for non-coronary ICU (Intensive Care Unit) patients and has become the 10th most common cause of death in western societies. This is a transversal condition affecting immunocompromised patients, critically ill patients, post-surgery patients, patients with AIDS, and the elderly. In western countries, septic patients account for as much as 25% of ICU bed utilization and the pathology affects 1-2% of all hospitalizations. Its mortality rates range from 12.8% for sepsis to 45.7% for septic shock. The prediction of mortality caused by sepsis is, therefore, a relevant research challenge from a medical viewpoint. The clinical indicators currently in use for this type of prediction have been criticized for their poor prognostic significance. In this study, we redescribe sepsis indicators through latent model-based feature extraction, using factor analysis. These extracted indicators are then applied to the prediction of mortality caused by sepsis. The reported results show that the proposed method improves on the results obtained with the current standard mortality predictor, which is based on the APACHE II score.\",\n \"Title: Feature and model selection with discriminatory visualization for diagnostic classification of brain tumors\\nYear: 2010\\n\\nAbstract: Machine Learning (ML) and related methods have of late made significant contributions to solving multidisciplinary problems in the field of oncology diagnosis. Human brain tumor diagnosis, in particular, often relies on the use of non-invasive techniques such as Magnetic Resonance Imaging (MRI) and Spectroscopy (MRS). In this paper, MRS data of human brain tumors are analyzed in detail. The high dimensionality of the MR spectra makes difficult both their classification and the interpretation of the obtained results, thus limiting their usability in practical medical settings. The use of dimensionality reduction techniques is therefore advisable. In this work, we apply feature selection methods and several off-the-shelf classifiers on various ^1H-MRS modalities: long and short echo times and an ad hoc combination of both. The introduction of bootstrap resampling techniques permits the obtention of mean performance estimates and their variability. Our experimental findings indicate that the feature selection process enhances the classification performance compared to using the full set of features. We also show that the use of combined information from the different echo times is a better strategy for small numbers of spectral frequencies; however, the use of ever greater numbers of short echo time frequencies permits the obtention of many models with similar performance. The final induced models offer very attractive solutions both in terms of prediction accuracy and number of involved spectral frequencies, which are also amenable to metabolic interpretation. A linear dimensionality-reduction technique that preserves class discrimination capabilities is used for visualizing the data corresponding to the selected frequencies.\",\n \"Title: SVM-Based Classification of Class C GPCRs from Alignment-Free Physicochemical Transformations of Their Sequences\\nYear: 2013\\n\\nAbstract: G protein-coupled receptors (GPCRs) have a key function in regulating the function of cells due to their ability to transmit extracelullar signals. Given that the 3D structure and the functionality of most GPCRs is unknown, there is a need to construct robust classification models based on the analysis of their amino acid sequences for protein homology detection. In this paper, we describe the supervised classification of the different subtypes of class C GPCRs using support vector machines (SVMs). These models are built on different transformations of the amino acid sequences based on their physicochemical properties. Previous research using semi-supervised methods on the same data has shown the usefulness of such transformations. The obtained classification models show a robust performance, as their Matthews correlation coefficient is close to 0.91 and their prediction accuracy is close to 0.93.\",\n \"Title: Comparative Evaluation of Semi-supervised Geodesic GTM\\nYear: 2009\\n\\nAbstract: In many real problems that ultimately require data classification, not all the class labels are readily available. This concerns the field of semi-supervised learning, in which missing class labels must be inferred from the available ones as well as from the natural cluster structure of the data. This structure can sometimes be quite convoluted. Previous research has shown the advantage, for these cases, of using the geodesic metric in clustering models of the manifold learning family to reveal the underlying true data structure. In this brief paper, we present a novel semi-supervised approach, namely Semi-Supervised Geo-GTM (SS-Geo-GTM). This is an extension of Geo-GTM, a variation on the Generative Topographic Mapping (GTM) manifold learning model for data clustering and visualization that resorts to the geodesic metric. SS-Geo-GTM uses a proximity graph built from Geo-GTM manifold as the basis for a label propagation algorithm that infers missing class labels. Its performance is compared to those of a semi-supervised version of the standard GTM and of the alternative Laplacian Eigenmaps method.\",\n \"Title: Discriminant Convex Non-negative Matrix Factorization for the classification of human brain tumours\\nYear: 2013\\n\\nAbstract: The medical analysis of human brain tumours commonly relies on indirect measurements. Among these, magnetic resonance imaging (MRI) and spectroscopy (MRS) predominate in clinical settings as tools for diagnostic assistance. Pattern recognition (PR) methods have successfully been used in this task, usually interpreting diagnosis as a supervised classification problem. In MRS, the acquired spectral signal can be analyzed in an unsupervised manner to extract its constituent sources. Recently, this has been successfully accomplished using Non-negative Matrix Factorization (NMF) methods. In this paper, we present a method to introduce the available class information into the unsupervised source extraction process of a convex variant of NMF. Novel techniques to generate diagnostic predictions for new, unseen spectra using the proposed Discriminant Convex-NMF are also described and experimentally assessed.\",\n \"Title: On the Initialization of Two-Stage Clustering with Class-GTM\\nYear: 2007\\n\\nAbstract: Generative Topographic Mapping is a probabilistic model for data clustering and visualization. It maps points, considered as prototype representatives of data clusters, from a low dimensional latent space onto the observed data space. In semi-supervised settings, class information can be added resulting in a model variation called class-GTM. The number of class-GTM latent points used is usually large for visualization purposes and does not necessarily reflect the class structure of the data. It is therefore convenient to group the clusters further in a two-stage procedure. In this paper, class-GTM is first used to obtain the basic cluster prototypes. Two novel methods are proposed to use this information as prior knowledge for the K-means-based second stage. We evaluate, using an entropy measure, whether these methods retain the class separability capabilities of class-GTM in the two-stage process, and whether the two-stage procedure improves on the direct clustering of the data using K-means.\",\n \"Title: On the influence of class information in the two-stage clustering of a human brain tumour dataset\\nYear: 2007\\n\\nAbstract: This paper analyzes, through clustering and visualization, Magnetic Resonance Spectra corresponding to a complex human brain tumour dataset. Clustering is performed as a two-stage process, in which the first stage model is Generative Topographic Mapping (GTM). In semi-supervised settings, class information can be added to refine the clustering process. A class information-enriched variant of GTM, class-GTM, is used here for a first cluster description of the data. The number of clusters used by GTM is usually large for visualization purposes and does not necessarily correspond to the overall class structure. Consequently, in a second stage, clusters are agglomerated using the K-means algorithm with different initialization strategies, some of them defined ad hoc for the GTM models. We aim to evaluate how and under what circumstances the use of class information influences tumour cluster-wise class separability in the final result of the two-stage clustering process.\",\n \"Title: Brain Tumor Pathological Area Delimitation through Non-negative Matrix Factorization\\nYear: 2011\\n\\nAbstract: Pattern Recognition and Data Mining can provide invaluable insights in the field of neuro oncology. This is because the clinical analysis of brain tumors requires the use of non-invasive methods that generate complex data in electronic format. Magnetic resonance, in the modalities of imaging and spectroscopy, is one of these methods that has been widely applied to this purpose. The heterogeneity of the tissue in the brain volumes analyzed by magnetic resonance remains a challenge in terms of pathological area delimitation. In this brief paper, we show that the Convex-Nonnegative Matrix Factorization technique can be used to extract MRS signal sources that are extremely tissue type-specific and that can be used to delimit these pathological areas with great accuracy.\",\n \"Title: Bias reduction in skewed binary classification with Bayesian neural networks\\nYear: 2000\\n\\nAbstract: The Bayesian evidence framework has become a standard of good practice for neural network estimation of class conditional probabilities. In this approach the conditional probability is marginalised over the distribution of network weights, which is usually approximated by an analytical expression that moderates the network output towards the midrange. In this paper, it is shown that the network calibration is considerably improved by marginalising to the prior distribution. Moreover, marginalisation to the midrange can seriously bias the estimates of the conditional probabilities calculated from the evidence framework. This is especially the case in the modelling of censored data. (C) 2000 Elsevier Science Ltd. All rights reserved.\",\n \"Title: Two-Stage Clustering of a Human Brain Tumour Dataset using Manifold Learning Models\\nYear: 2008\\n\\nAbstract: This paper analyzes, through clustering and visualization, Magnetic Resonance spectra of a complex multi-center human brain tumour dataset. Clustering is performed as a two-stage process, in which the models used in the first stage are variants of Generative Topographic Mapping (GTM). Class information-enriched variants of GTM are used to obtain a primary cluster description of the data. The number of clusters used by GTM is usually large and does not necessarily correspond to the overall class structure. Consequently, in a second stage, clusters are agglomerated using K-means with different initialization strategies, some of them defined ad hoc for the GTM models. We evaluate if the use of class information influence the brain tumour cluster-wise class separability resulting from the process. We also resort to a robust variant of GTM that detects outliers while effectively minimizing their negative impact in the clustering process.\",\n \"Title: Robust analysis of MRS brain tumour data using t-GTM\\nYear: 2006\\n\\nAbstract: This paper proposes a principled, self-organized, framework to manage two sources of uncertainty that are inherent in intelligent systems for medical decision support, namely outliers and missing data. The framework is applied to magnetic resonance spectra (MRS), which are indicators of the grade of malignancy in brain tumours. A model for multivariate data clustering and visualization, the generative topographic mapping (GTM), is re-formulated as a mixture of Student's t-distributions making it more robust to outliers while supporting the imputation of missing values. An important new development is the extension of the model to provide automatic feature relevance determination. Its effectiveness on the MRS data is demonstrated empirically.\",\n \"Title: Non-negative matrix factorisation methods for the spectral decomposition of MRS data from human brain tumours.\\nYear: 2012\\n\\nAbstract: In-vivo single voxel proton magnetic resonance spectroscopy (SV 1H-MRS), coupled with supervised pattern recognition (PR) methods, has been widely used in clinical studies of discrimination of brain tumour types and follow-up of patients bearing abnormal brain masses. SV 1H-MRS provides useful biochemical information about the metabolic state of tumours and can be performed at short (< 45 ms) or long (> 45 ms) echo time (TE), each with particular advantages. Short-TE spectra are more adequate for detecting lipids, while the long-TE provides a much flatter signal baseline in between peaks but also negative signals for metabolites such as lactate. Both, lipids and lactate, are respectively indicative of specific metabolic processes taking place. Ideally, the information provided by both TE should be of use for clinical purposes. In this study, we characterise the performance of a range of Non-negative Matrix Factorisation (NMF) methods in two respects: first, to derive sources correlated with the mean spectra of known tissue types (tumours and normal tissue); second, taking the best performing NMF method for source separation, we compare its accuracy for class assignment when using the mixing matrix directly as a basis for classification, as against using the method for dimensionality reduction (DR). For this, we used SV 1H-MRS data with positive and negative peaks, from a widely tested SV 1H-MRS human brain tumour database.The results reported in this paper reveal the advantage of using a recently described variant of NMF, namely Convex-NMF, as an unsupervised method of source extraction from SV1H-MRS. Most of the sources extracted in our experiments closely correspond to the mean spectra of some of the analysed tumour types. This similarity allows accurate diagnostic predictions to be made both in fully unsupervised mode and using Convex-NMF as a DR step previous to standard supervised classification. The obtained results are comparable to, or more accurate than those obtained with supervised techniques.The unsupervised properties of Convex-NMF place this approach one step ahead of classical label-requiring supervised methods for the discrimination of brain tumour types, as it accounts for their increasingly recognised molecular subtype heterogeneity. The application of Convex-NMF in computer assisted decision support systems is expected to facilitate further improvements in the uptake of MRS-derived information by clinicians.\",\n \"Title: Classification of human brain tumours from MRS data using Discrete Wavelet Transform and Bayesian Neural Networks\\nYear: 2012\\n\\nAbstract: The diagnosis of brain tumours is an extremely sensitive and complex clinical task that must rely upon information gathered through non-invasive techniques. One such technique is Magnetic Resonance Spectroscopy. In this task, radiology experts are likely to benefit from the support of computer-based systems built around robust classification processes. In this paper, a Discrete Wavelet Transform procedure was applied to the pre-processing of spectra corresponding to several brain tumour pathologies. This procedure does not alleviate the high dimensionality of the data by itself. For this reason, dimensionality reduction was subsequently implemented using Moving Window with Variance Analysis for feature selection or Principal Component Analysis for feature extraction. The combined method yielded very encouraging results in terms of diagnostic discriminatory binary classification using Bayesian Neural Networks. In most cases, the classification accuracy improved on previously reported results.\",\n \"Title: Neural network models for language acquisition: a brief survey\\nYear: 2006\\n\\nAbstract: Since the outbreak of connectionist modelling in the mid eighties, several problems in natural language processing have been tackled by employing neural network-based techniques. Neural network’s “biological plausibility” offers a promising framework in which the computational treatment of language may be linked to other disciplines such as cognitive science and psychology. With this brief survey, we set out to explore the landscape of artificial neural models for the acquisition of language that have been proposed in the research literature.\",\n \"Title: Frequency Selection for the Diagnostic Characterization of Human Brain Tumours\\nYear: 2009\\n\\nAbstract: The diagnosis of brain tumours is an extremely sensitive and complex clinical task that must rely upon information gathered through non-invasive techniques. One such technique is magnetic resonance, in the modalities of imaging or spectroscopy. The latter provides plenty of metabolic information about the tumour tissue, but its high dimensionality makes resorting to pattern recognition techniques advisable. In this brief paper, an international database of brain tumours is analyzed resorting to an ad hoc spectral frequency selection procedure combined with nonlinear classification.\",\n \"Title: A variational Bayesian approach for the robust analysis of the cortical silent period from EMG recordings of brain stroke patients\\nYear: 2011\\n\\nAbstract: Transcranial magnetic stimulation (TMS) is a powerful tool for the calculation of parameters related to the intracortical excitability and inhibition of the motor cortex. The cortical silent period (CSP) is one such parameter that corresponds to the suppression of muscle activity for a short period after a muscle response to TMS. The duration of the CSP is known to be correlated with the prognosis of brain stroke patients' motor ability. Current methods for the estimation of the CSP duration are very sensitive to the presence of noise. A variational Bayesian formulation of a manifold-constrained hidden Markov model is applied in this paper to the segmentation of a set of multivariate time series (MTS) of electromyographic recordings corresponding to stroke patients and control subjects. A novel index of variability associated to this model is defined and applied to the detection of the silent period interval of the signal and to the estimation of its duration. This model and its associated index are shown to behave robustly in the presence of noise and provide more reliable estimations than the current standard in clinical practice.\",\n \"Title: Outlier exploration and diagnostic classification of a multi-centre 1H-MRS brain tumour database\\nYear: 2009\\n\\nAbstract: Non-invasive techniques such as magnetic resonance spectroscopy (MRS) are often required for assisting the diagnosis of tumours. Radiologists are not always accustomed to make sense of the biochemical information provided by MRS and they may benefit from computer-based support in their decision making. The high dimensionality of the MR spectra obscures atypical aspects of the data that may jeopardize their classification. In this study, we describe a method to overcome this problem that combines nonlinear dimensionality reduction, outlier detection, and expert opinion. MR spectra subsequently undergo a feature selection process followed by classification. The impact of outlier removal on classification performance is assessed.\",\n \"Title: Finding Class C GPCR Subtype-Discriminating N-grams through Feature Selection.\\nYear: 2014\\n\\nAbstract: G protein-coupled receptors (GPCRs) are a large and heterogeneous superfamily of receptors that are key cell players for their role as extracellular signal transmitters. Class C GPCRs, in particular, are of great interest in pharmacology. The lack of knowledge about their full 3-D structure prompts the use of their primary amino acid sequences for the construction of robust classifiers, capable of discriminating their different subtypes. In this paper, we describe the use of feature selection techniques to build Support Vector Machine (SVM)-based classification models from selected receptor subsequences described as n-grams. We show that this approach to classification is useful for finding class C GPCR subtype-specific motifs.\",\n \"Title: Semi-supervised outcome prediction for a type of human brain tumour using partially labeled MRS information\\nYear: 2009\\n\\nAbstract: The diagnosis and prognosis of human brain tumours, especially when they are aggresive, are sensitive clinical tasks that usually require non-invasive measurement techniques. Outcome information for aggressive tumours, in particular, is usually scarce. In this paper, we aim to gauge the capability of a novel semi-supervised model, SS-Geo-GTM, to infer outcome stages from a very limited amount of available stage labels and Magnetic Resonance Spectroscopy (MRS) data corresponding to Glioblastoma, which is an aggressive tumor type. This model stems from a geodesic distance-based extension of Generative Topographic Mapping (Geo-GTM) that prioritizes neighbourhood relationships along a generated manifold embedded in the observed data space.\",\n \"Title: On The Benefits For Model Regularization Of A Variational Formulation Of Gtm\\nYear: 2008\\n\\nAbstract: Generative Topographic Mapping (GTM) is a manifold learning model for the simultaneous visualization and clustering of multivariate data. It was originally formulated as a constrained mixture of distributions, for which the adaptive parameters were determined by Maximum Likelihood (ML), using the Expectation-Maximization (EM) algorithm. In this formulation, GTM is prone to data overfitting unless a regularization mechanism is included. The theoretical principles of Variational GTM, an approximate method that provides a full Bayesian treatment to a Gaussian Process (GP)-based variation of the GTM, were recently introduced as alternative way to control data overfitting. In this paper we assess in some detail the generalization capabilities of Variational GTM and compare them with those of alternative regularization approaches in terms of test log-likelihood, using several artificial and real datasets.\",\n \"Title: Neural networks and other machine learning methods in cancer research\\nYear: 2007\\n\\nAbstract: Evidence-based medicine has grown in stature over the last three decades and is now regarded a key development of modern medicine. The evidence base can be heterogeneous, involving both qualitative knowledge and measured quantitative data. Machine Learning (ML) methods have also begun to establish themselves as an alternative and promising approach to computer-based data analysis in oncology, as this field moves gradually away from being the preserve of traditional statistical analysis. In this paper, we describe the main areas of cancer research in which ML methods are currently being applied, and briefly discuss some of the advantages and disadvantages of their application.\",\n \"Title: Learning what is important: feature selection and rule extraction in a virtual course\\nYear: 2006\\n\\nAbstract: Virtualcampus� environmentsare �becomingamain str eamalternative� to� traditionaldist ancehighereduc ation.�TheInter net �mediumtheyuse �allowsthe� gatheringofinf ormationonstud ents'�usagebeha vio ur.�Theknowledg eextracted� from� this� information� can� be� fed� back� to� the� e/lear ning� environment� to� ease� teachers'�worklo ad.�Inthiscont ext,�twoproblem sa readdressedin� thecurrent� study:�findingw hichusagefeatu resarebestat� pre dictingonlines tudents'�marks,� andexplainingm arkpredictioni ntheformofsi mpl eandinterpreta blerules.�To� thateffect,�two �methodsareuse d:�FuzzyInducti ve� Reasoning�(FIR)� forfeature� selection� and� Orthogonal� Search/Based� Rule� Extraction � (OSRE).� Experiments� carriedoutont heavailabledat aindicatethat� stu dents'�markscan �beaccurately� predicted� and� that� a� small� subset� of� variables� expl ains� the� accuracy� of� such� prediction,�whic hcanbedescrib edthroughaset �of �actionablerule s.��\",\n \"Title: Automated classification of brain tumours from short echo time in vivo MRS data using Gaussian Decomposition and Bayesian Neural Networks.\\nYear: 2014\\n\\nAbstract: •The paper describes a method for brain tumour type discrimination from MRS signal.•The method combines Gaussian Decomposition, dimensionality reduction and classification using Artificial Neural Networks.•The experiments yield encouraging brain tumour classification results, a difficult task even for expert radiologists.\",\n \"Title: Semi-supervised geodesic Generative Topographic Mapping\\nYear: 2010\\n\\nAbstract: We present a novel semi-supervised model, SS-Geo-GTM, which stems from a geodesic distance-based extension of Generative Topographic Mapping that prioritizes neighbourhood relationships along a generated manifold embedded in the observed data space. With this, it improves the trustworthiness and the continuity of the low-dimensional representations it provides, while behaving robustly in the presence of noise. In SS-Geo-GTM, the model prototypes are linked by the nearest neighbour to the data manifold constructed by Geo-GTM. The resulting proximity graph is used as the basis for a class label propagation algorithm. The performance of SS-Geo-GTM is experimentally assessed, comparing positively with that of an Euclidean distance-based counterpart and with those of alternative manifold learning methods.\",\n \"Title: The influence of alignment-free sequence representations on the semi-supervised classification of class C G protein-coupled receptors.\\nYear: 2015\\n\\nAbstract: G protein-coupled receptors (GPCRs) are integral cell membrane proteins of relevance for pharmacology. The tertiary structure of the transmembrane domain, a gate to the study of protein functionality, is unknown for almost all members of class C GPCRs, which are the target of the current study. As a result, their investigation must often rely on alignments of their amino acid sequences. Sequence alignment entails the risk of missing relevant information. Various approaches have attempted to circumvent this risk through alignment-free transformations of the sequences on the basis of different amino acid physicochemical properties. In this paper, we use several of these alignment-free methods, as well as a basic amino acid composition representation, to transform the available sequences. Novel semi-supervised statistical machine learning methods are then used to discriminate the different class C GPCRs types from the transformed data. This approach is relevant due to the existence of orphan proteins to which type labels should be assigned in a process of deorphanization or reverse pharmacology. The reported experiments show that the proposed techniques provide accurate classification even in settings of extreme class-label scarcity and that fair accuracy can be achieved even with very simple transformation strategies that ignore the sequence ordering.\",\n \"Title: Exploratory Visualization Of Misclassified Gpcrs From Their Transformed Unaligned Sequences Using Manifold Learning Techniques\\nYear: 2014\\n\\nAbstract: Class C G-protein-coupled receptors (GPCRs) are cell membrane proteins of great relevance to biology and pharmacology. Previous research has revealed an upper boundary on the accuracy that can be achieved in their classification into subtypes from the unaligned transformation of their sequences. To investigate this, we focus on sequences that have been misclassified using supervised methods. These are visualized, using a nonlinear dimensionality reduction technique and phylogenetic trees, and then characterized against the rest of the data and, particularly, against the rest of cases of their own subtype. This should help to discriminate between different types of misclassification and to build hypotheses about database quality problems and the extent to which GPCR sequence transformations limit subtype discriminability. The reported experiments provide a proof of concept for the proposed method.\",\n \"Title: Visual interpretation of class C GPCR subtype overlapping from the nonlinear mapping of transformed primary sequences\\nYear: 2014\\n\\nAbstract: This brief paper addresses the problem of visually assessing the natural discriminability of the different subtypes that characterize class C G-Protein-Coupled Receptors, which are membrane proteins of interest in pharmacology, from diverse transformations of their primary amino acid sequences. Visualization, achieved using nonlinear dimensionality reduction techniques, is used as an exploratory tool to increase the interpretability of the data subtype structure. Combined with a quantitative estimation of subtype overlapping, this approach should be of help to experts in proteomics.\",\n \"Title: A MAP approach for convex non-negative matrix factorization in the diagnosis of brain tumors\\nYear: 2014\\n\\nAbstract: Convex non-negative matrix factorization is a blind signal separation technique that has previously demonstrated to be well-suited for the task of human brain tumor diagnosis from magnetic resonance spectroscopy data. This is due to its ability to retrieve interpretable sources of mixed sign that highly correlate with tissue type prototypes. The current study provides a Bayesian formulation for such problem and derives a maximum a posteriori estimate based on a gradient descent algorithm specifically designed to deal with matrices with different sign restrictions. Its applicability to neuro-oncology diagnosis was experimentally assessed and the results were found to be comparable to those achieved by state of the art methods in tumor type discrimination and consistently better in source extraction.\",\n \"Title: Manifold learning visualization of Metabotropic Glutamate Receptors.\\nYear: 2014\\n\\nAbstract: G-Protein-Coupled Receptors (GPCRs) are cell membrane proteins with a key role in biological processes. GPCRs of class C, in particular, are of great interest in pharmacology. The lack of knowledge about their 3-D structures means they must be investigated through their primary amino acid sequences. Sequence visualization can help to explore the existing receptor sub-groupings at different partition levels. In this paper, we focus on Metabotropic Glutamate Receptors (mGluR), a subtype of class C GPCRs. Different versions of a probabilistic manifold learning model are employed to comparatively sub-group and visualize them through different transformations of their sequences.\",\n \"Title: Probability Ridges and Distortion Flows: Visualizing Multivariate Time Series Using a Variational Bayesian Manifold Learning Method.\\nYear: 2014\\n\\nAbstract: Time-dependent natural phenomena and artificial processes can often be quantitatively expressed as multivariate time series (MTS). As in any other process of knowledge extraction from data, the analyst can benefit from the exploration of the characteristics of MTS through data visualization. This visualization often becomes difficult to interpret when MTS are modelled using nonlinear techniques. Despite their flexibility, nonlinear models can be rendered useless if such interpretability is lacking. In this brief paper, we model MTS using Variational Bayesian Generative Topographic Mapping Through Time (VB-GTM-TT), a variational Bayesian variant of a constrained hidden Markov model of the manifold learning family defined for MTS visualization. We aim to increase its interpretability by taking advantage of two results of the probabilistic definition of the model: the explicit estimation of probabilities of transition between states described in the visualization space and the quantification of the nonlinear mapping distortion.\",\n \"Title: A variational formulation for GTM through time\\nYear: 2008\\n\\nAbstract: Generative topographic mapping (GTM) is a latent variable model that, in its original version, was conceived to provide clustering and visualization of multivariate, real-valued, i.i.d. data. It was also extended to deal with noni. i.d. data such as multivariate time series in a variant called GTM through time (GTM-TT), defined as a constrained hidden Markov model (HMM). In this paper, we provide the theoretical foundations of the reformulation of GTM-TT within the variational Bayesian framework and provide an illustrative example of its application. This approach handles the presence of noise in the time series, helping to avert the problem of data overfitting.\",\n \"Title: Label noise in subtype discrimination of class C G protein-coupled receptors: A systematic approach to the analysis of classification errors\\nYear: 2015\\n\\nAbstract: The characterization of proteins in families and subfamilies, at different levels, entails the definition and use of class labels. When the adscription of a protein to a family is uncertain, or even wrong, this becomes an instance of what has come to be known as a label noise problem. Label noise has a potentially negative effect on any quantitative analysis of proteins that depends on label information. This study investigates class C of G protein-coupled receptors, which are cell membrane proteins of relevance both to biology in general and pharmacology in particular. Their supervised classification into different known subtypes, based on primary sequence data, is hampered by label noise. The latter may stem from a combination of expert knowledge limitations and the lack of a clear correspondence between labels that mostly reflect GPCR functionality and the different representations of the protein primary sequences.\",\n \"Title: Making nonlinear manifold learning models interpretable: The manifold grand tour\\nYear: 2015\\n\\nAbstract: •Smooth nonlinear topographic maps of the data distribution to guide a Grand Tour visualisation.•Prioritisation of data linear views that are most consistent with data structure in the maps.•Useful visualisations that cannot be obtained by other more classical approaches.\",\n \"Title: Quantitative characterization and prediction of on-line purchasing behavior: a latent variable approach\\nYear: 2000\\n\\nAbstract: Realizing the full potential of the on-line consumer market requires careful identification of customer needs and expectations. As research on Internet consumer behavior is still in its infancy, a quantitative framework to characterize user profiles for e-commerce has not yet been established. This study proposes a quantitative framework that uses factor analysis to identify latent factor descriptors of Internet users' opinions on Web vendors and on-line shopping. Predictive models based on logistic discrimination and neural networks then select the factors most predictive of the propensity to buy on-line and classify Internet users accordingly. The application of this framework shows that the obtained latent factors agree in general with the major indicators identified in previous qualitative research. A small subset of the obtained factors is shown to retain the predictive power of the whole set. Neural networks are found to perform only marginally better than logistic discrimination in the task of classification.\",\n \"Title: Random Forests for Quality Control in G-Protein Coupled Receptor Databases.\\nYear: 2016\\n\\nAbstract: G protein-coupled receptors are a large and heterogeneous super-family of cell membrane proteins of interest to biology in general. One of its families, class C, is of particular interest to pharmacology and drug design. This family is quite heterogeneous on its own, and the discrimination of its several sub-families is a challenging problem. In the absence of known crystal structure, such discrimination must rely on their primary amino acid sequences. In this study, we are interested not as much in achieving maximum sub-family discrimination accuracy, but in exploring sequence misclassification behaviour. Specifically, we are interested in isolating those sequences showing consistent misclassification, that is, sequences that are very often misclassified and almost always to the same wrong sub-family. This analysis should assist database curators in receptor quality control tasks. Random Forests are used for this analysis due their ensemble nature, which makes them naturally suited to gauge the consistency of misclassification.\",\n \"Title: Automated Quality Control for Proton Magnetic Resonance Spectroscopy Data Using Convex Non-negative Matrix Factorization.\\nYear: 2016\\n\\nAbstract: Proton Magnetic Resonance Spectroscopy (H-1 MRS) has proven its diagnostic potential in a variety of conditions. However, MRS is not yet widely used in clinical routine because of the lack of experts on its diagnostic interpretation. Although data-based decision support systems exist to aid diagnosis, they often take for granted that the data is of good quality, which is not always the case in a real application context. Systems based on models built with bad quality data are likely to underperform in their decision support tasks. In this study, we propose a system to filter out such bad quality data. It is based on convex Non-Negative Matrix Factorization models, used as a dimensionality reduction procedure, and on the use of several classifiers to discriminate between good and bad quality data.\",\n \"Title: Metrics For Probabilistic Geometries\\nYear: 2014\\n\\nAbstract: We investigate the geometrical structure of probabilistic generative dimensionality reduction models using the tools of Riemannian geometry. We explicitly define a distribution over the natural metric given by the models. We provide the necessary algorithms to compute expected metric tensors where the distribution over mappings is given by a Gaussian process. We treat the corresponding latent variable model as a Riemannian manifold and we use the expectation of the metric under the Gaussian process prior to define interpolating paths and measure distance between latent points. We show how distances that respect the expected metric lead to more appropriate generation of new data.\",\n \"Title: Visual Exploratory Assessment of Class C GPCR Extracellular Domains Discrimination Capabilities.\\nYear: 2016\\n\\nAbstract: G protein-coupled receptors (GPCRs) are integral membrane-bound proteins. They are divided in five main classes with Class C members receiving recently special attention because of their involvement in many neurologic diseases. These receptors are composed of the seven-helix transmembrane domain (typical of all GPCRs) and a large extracellular domain where the endogenous ligand binds. In the present absence of crystal structures for complete Class C receptors, their primary sequences can provide limited but useful information that can be used for subtype discrimination in first instance. In this paper, we show that the extracellular part of these sequences provides as effective a discrimination as the complete sequence. With that purpose, we describe a process of exploratory sequence visualization using different data transformations and manifold learning techniques for dimensionality reduction. Class discriminability is assessed using an entropy-based measure.\",\n \"Title: Computational Intelligence in Interior Design: State-of-the-Art and Outlook.\\nYear: 2016\\n\\nAbstract: Tasks in architectural and interior design range from defining the building floor plans and ensuring desired functionality, to deciding furnishing styles and arrangement choices. The process of design, as a whole, has remained hard to master for computer-based optimization in general and for computational intelligence approaches in particular. Numerous attempts to tackle different subfields of this problem in a machine learning fashion have emerged over the last few years. In this paper, we present an overview of current advances of computational intelligence in architectural science with a focus on interior design. This is accompanied by the description of ongoing research towards the development of a commercial robust and scalable solution for automatic furniture arrangement.\",\n \"Title: Intelligent data analysis approaches to churn as a business problem: a survey.\\nYear: 2017\\n\\nAbstract: Globalization processes and market deregulation policies are rapidly changing the competitive environments of many economic sectors. The appearance of new competitors and technologies leads to an increase in competition and, with it, a growing preoccupation among service-providing companies with creating stronger customer bonds. In this context, anticipating the customer's intention to abandon the provider, a phenomenon known as churn, becomes a competitive advantage. Such anticipation can be the result of the correct application of information-based knowledge extraction in the form of business analytics. In particular, the use of intelligent data analysis, or data mining, for the analysis of market surveyed information can be of great assistance to churn management. In this paper, we provide a detailed survey of recent applications of business analytics to churn, with a focus on computational intelligence methods. This is preceded by an in-depth discussion of churn within the context of customer continuity management. The survey is structured according to the stages identified as basic for the building of the predictive models of churn, as well as according to the different types of predictive methods employed and the business areas of their application.\",\n \"Title: Machine Learning for Critical Care: An Overview and a Sepsis Case Study.\\nYear: 2017\\n\\nAbstract: Biology in general and medicine and healthcare in particular are facing the critical challenge of exponentially increasing data availability. The core of this challenge is putting these data to work through computer-based knowledge extraction methods. In the medical context this could take the form of medical decision support systems for diagnosis, prognosis or general management. Arguably, one of the most data dependent clinical environments is the critical care unit and by extension the whole area of critical care. Fresh approaches to data analysis in critical care are required, and Computational Intelligence and Machine Learning methods have already shown their usefulness in tackling problems in the area. This brief paper aims to be an introduction to the use of such methods in critical care.\",\n \"Title: Topological Sequence Segments Discriminate Between Class C GPCR Subtypes.\\nYear: 2017\\n\\nAbstract: G protein-coupled receptors are eukaryotic cell membrane proteins with a key role as extracellular signal transmitters. While GPCRs embrace a wide and heterogeneous super-family of proteins, our interest in this study is in its Class C, of great relevance to pharmacology. The scarcity of knowledge about their full 3-D crystal structure makes the use of their primary amino acid sequences important for analysis. In this paper, we systematically analyze whether segments of the receptor sequences are able to discriminate between the different class C GPCR subtypes according to their topological location on the extracellular, transmembrane or intracellular domain. For this, we build on previous research that showed that the use of the extracellular N-terminus domain on its own for this classification task did only entail a minor decrease in subtype discrimination when compared to the complete sequence. We use Support Vector Machine-based classification models to assess the subtype discriminating power of the topological segments.\",\n \"Title: Discovering Subtype Specific n-Gram Motifs in Class C GPCR N-Termini.\\nYear: 2017\\n\\nAbstract: Protein-Coupled Receptors are a family of cell membrane proteins, whose class C is a current main target for drug development. Their primary sequences are studied as a source of information for the characterization of their behaviour. In previous research, different alignment-free sequence transformations were explored as the basis for the supervised discrimination of the various class C subtypes using Support Vector Machines. We also investigated an alignment-free sequence transformation based on n-gram protein motifs, under the hypothesis that the sequences' extra-cellular N-terminus domain could suffice to retain most of the subtype-discrimination capabilities of the complete receptor, and that a parsimonious selection of n-grams would be responsible for such classification success. In the current study, these previous results are extended by investigating a different classification procedure that now employs a subtype-vs-all the rest of subtypes approach, shifting towards the selection of those sequence motifs that distinguish each class C subtype from the rest. The reported results indicate the adequacy of this new approach, both in terms of discrimination ability and motif selection parsimony.\",\n \"Title: Bioinformatics and Medicine in the Era of Deep Learning.\\nYear: 2018\\n\\nAbstract: Many of the current scientific advances in the life sciences have their origin in the intensive use of data for knowledge discovery. In no area this is so clear as in bioinformatics, led by technological breakthroughs in data acquisition technologies. It has been argued that bioinformatics could quickly become the field of research generating the largest data repositories, beating other data-intensive areas such as high-energy physics or astroinformatics. Over the last decade, deep learning has become a disruptive advance in machine learning, giving new live to the long-standing connectionist paradigm in artificial intelligence. Deep learning methods are ideally suited to large-scale data and, therefore, they should be ideally suited to knowledge discovery in bioinformatics and biomedicine at large. In this brief paper, we review key aspects of the application of deep learning in bioinformatics and medicine, drawing from the themes covered by the contributions to an ESANN 2018 special session devoted to this topic.\",\n \"Title: Bayesian semi non-negative matrix factorisation.\\nYear: 2016\\n\\nAbstract: Non-negative Matrix Factorisation (NMF) has become a standard method for source identification when data, sources and mixing coefficients are constrained to be positive-valued. The method has recently been extended to allow for negative-valued data and sources in the form of Semi-and Convex-NMF. In this paper, we re-elaborate Semi-NMF within a full Bayesian framework. This provides solid foundations for parameter estimation and, importantly, a principled method to address the problem of choosing the most adequate number of sources to describe the observed data. The proposed Bayesian Semi-NMF is preliminarily evaluated here in a real neuro-oncology problem.\",\n \"Title: Physics and Machine Learning: Emerging Paradigms.\\nYear: 2016\\n\\nAbstract: Current research in Machine Learning (ML) combines the study of variations on well-established methods with cutting-edge breakthroughs based on completely new approaches. Among the latter, emerging paradigms from Physics have taken special relevance in recent years. Although still in its initial stages, Quantum Machine Learning (QML) shows promising ways to speed up some of the costly ML calculations with a similar or even better performance than existing approaches. Two additional advantages are related to the intrinsic probabilistic approach of QML, since quantum states are genuinely probabilistic, and to the capability of finding the global optimum of a given cost function by means of adiabatic quantum optimization, thus circumventing the usual problem of local minima. Another Physics approach for ML comes from Statistical Physics and is linked to Information theory in supervised and semi-supervised learning frameworks. On the other hand, and from the perspective of Physics, ML can provide solutions by extracting knowledge from huge amounts of data, as it is common in many experiments in the field, such as those related to High Energy Physics for elementary-particle research and Observational Astronomy.\",\n \"Title: A machine learning pipeline for supporting differentiation of glioblastomas from single brain metastases.\\nYear: 2016\\n\\nAbstract: Machine learning has provided, over the last decades, tools for knowledge extraction in complex medical domains. Most of these tools, though, are ad hoc solutions and lack the systematic approach that would be required to become mainstream in medical practice. In this brief paper, we define a machine learning-based analysis pipeline for helping in a difficult problem in the field of neuro-oncology, namely the discrimination of brain glioblastomas from single brain metastases. This pipeline involves source extraction using k-Meansinitialized Convex Non-negative Matrix Factorization and a collection of classifiers, including Logistic Regression, Linear Discriminant Analysis, AdaBoost, and Random Forests.\",\n \"Title: The Importance Of Interpretability And Visualization In Machine Learning For Applications In Medicine And Health Care\\nYear: 2020\\n\\nAbstract: In a short period of time, many areas of science have made a sharp transition towards data-dependent methods. In some cases, this process has been enabled by simultaneous advances in data acquisition and the development of networked system technologies. This new situation is particularly clear in the life sciences, where data overabundance has sparked a flurry of new methodologies for data management and analysis. This can be seen as a perfect scenario for the use of machine learning and computational intelligence techniques to address problems in which more traditional data analysis approaches might struggle. But, this scenario also poses some serious challenges. One of them is model interpretability and explainability, especially for complex nonlinear models. In some areas such as medicine and health care, not addressing such challenge might seriously limit the chances of adoption, in real practice, of computer-based systems that rely on machine learning and computational intelligence methods for data analysis. In this paper, we reflect on recent investigations about the interpretability and explainability of machine learning methods and discuss their impact on medicine and health care. We pay specific attention to one of the ways in which interpretability and explainability in this context can be addressed, which is through data and model visualization. We argue that, beyond improving model interpretability as a goal in itself, we need to integrate the medical experts in the design of data analysis interpretation strategies. Otherwise, machine learning is unlikely to become a part of routine clinical and health care practice.\",\n \"Title: Visual Mining of Industrial Gas Turbines Sensor Data as an Industry 4.0 Application\\nYear: 2021\\n\\nAbstract: Industrial gas turbines for power generation are advanced engines that require constant and detailed monitorization using internal and external sensors. These sensors generate a large flow of data in the form of multivariate time series that are amenable to analysis using pattern recognition methods with the objective of improving and optimizing turbine operation. One aspect this may take is visual analytics, where dimensionality reduction methods can be used to intuitively visualize the multivariate time series. This brief paper provides a proof of concept and some case scenarios of a visual turbine-monitorization tool, based on the UMAP method, combined with clustering using HDBSCAN and unsupervised Agnostic Feature Selection. It can be considered as a first step towards a data-centered approach to gas turbine management within the industry 4.0 framework, based on the mining of turbine sensor data.\",\n \"Title: Long Short-Term Memory to Predict 3D Amino Acids Positions in GPCR Molecular Dynamics.\\nYear: 2022\\n\\nAbstract: G-Protein Coupled Receptors (GPCRs) are a big family of eukaryotic cell transmembrane proteins, responsible for numerous biological processes. From a practical viewpoint around 34\\\\% of the drugs approved by the US Food and Drug Administration target these receptors. They can be analyzed from their simulated molecular dynamics, including the prediction of their behavior in the presence of drugs. In this paper, the capability of Long Short-Term Memory Networks (LSTMs) are evaluated to learn and predict the molecular dynamic trajectories of a receptor. Several models were trained with the 3D position of the amino acids of the receptor considering different transformations on the position of the amino acid, such as their centers of mass, the geometric centers and the position of the $\\\\alpha$--carbon for each amino acid. The error of the prediction of the position was evaluated by the mean average error (MAE) and root-mean-square deviation (RMSD). The LSTM models show a robust performance, with results comparable to the state-of-the-art in non-dynamic 3D predictions. The best MAE and RMSD values were found for the mass center of the amino acids with 0.078 {\\\\AA} and 0.156 {\\\\AA} respectively. This work shows the potential of LSTM to predict the molecular dynamics of GPRCs.\",\n \"Title: Molecular Dynamics forecasting of transmembrane Regions in GPRCs by Recurrent Neural Networks\\nYear: 2022\\n\\nAbstract: G protein-coupled receptors are a large super-family of cell membrane proteins that play an important physiological role as transmitters of extra-cellular signals. Signal transmission through the cell membrane depends on the conformational changes of the transmembrane region of the receptor and the investigation of the dynamics in these regions is therefore key. Molecular Dynamics (MD) simulations can provide information of the receptor conformational states at the atom level and machine learning (ML) methods can be useful for the analysis of these data. In this paper, Recurrent Neural Networks (RNNs) are used to evaluate whether the MD can be modeled focusing on the different regions of the receptor (intra-cellular, extra-cellular and each transmembrane regions (TM)). The best results, as measured by root-mean-square deviation (RMSD), are 0.1228 Å for TM4 of the 2rh1 (inactive state) and 0.1325 Å for TM4 of the 3p0g (active state), which are comparable to the state-of-the-art in non-dynamic 3-D predictions, showing the potential of the proposed approach.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":73,"string":"73"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13554,"string":"13,554"}}},{"rowIdx":13555,"cells":{"question_id":{"kind":"string","value":"Martin Pinzger"},"question":{"kind":"string","value":"Generate an abstract for the title \"Analyzing and understanding architectural characteristics of COM+ components\" using the following items: COM+, component inspection, source code analysis, abstract model, architectural characteristics."},"answer":{"kind":"string","value":"Understanding architectural characteristics of software components that constitute distributed systems is crucial for maintaining and evolving them. One component framework heavily used for developing component-based software systems is Microsoft's COM+. In this paper we particularly concentrate on the analysis of COM+ components and introduce an iterative and interactive approach that combines component inspection techniques with source code analysis to obtain a complete abstract model of each COM+ component. The model describes important architectural characteristics such as transactions, security, and persistency, as well as creating and use dependencies between components, and maps these higher-level concepts down to their implementation in source files. Based on the model, engineers can browse the software system's COM+ components and navigate from the list of architectural characteristics to the corresponding source code statements. We also discuss the Island Hopper application with which our approach has been validated."},"correct_answer":{"kind":"string","value":"Understanding architectural characteristics of software components that constitute distributed systems is crucial for maintaining and evolving them. One component framework heavily used for developing component-based software systems is Microsoft's COM+. In this paper we particularly concentrate on the analysis of COM+ components and introduce an iterative and interactive approach that combines component inspection techniques with source code analysis to obtain a complete abstract model of each COM+ component. The model describes important architectural characteristics such as transactions, security, and persistency, as well as creating and use dependencies between components, and maps these higher-level concepts down to their implementation in source files. Based on the model, engineers can browse the software system's COM+ components and navigate from the list of architectural characteristics to the corresponding source code statements. We also discuss the Island Hopper application with which our approach has been validated."},"documents":{"kind":"list like","value":["Title: Using source code metrics to predict change-prone Java interfaces\nYear: 2011\n\nAbstract: Recent empirical studies have investigated the use of source code metrics to predict the change- and defect-proneness of source code files and classes. While results showed strong correlations and good predictive power of these metrics, they do not distinguish between interface, abstract or concrete classes. In particular, interfaces declare contracts that are meant to remain stable during the evolution of a software system while the implementation in concrete classes is more likely to change. This paper aims at investigating to which extent the existing source code metrics can be used for predicting change-prone Java interfaces. We empirically investigate the correlation between metrics and the number of fine-grained source code changes in interfaces of ten Java open-source systems. Then, we evaluate the metrics to calculate models for predicting change-prone Java interfaces. Our results show that the external interface cohesion metric exhibits the strongest correlation with the number of source code changes. This metric also improves the performance of prediction models to classify Java interfaces into change-prone and not change-prone.","Title: Relation of code clones and change couplings\nYear: 2006\n\nAbstract: Code clones have long been recognized as bad smells in software systems and are considered to cause maintenance problems during evolution. It is broadly assumed that the more clones two files share, the more often they have to be changed together. This relation between clones and change couplings has been postulated but neither demonstrated nor quantified yet. However, given such a relation it would simplify the identification of restructuring candidates and reduce change couplings. In this paper, we examine this relation and discuss if a correlation between code clones and change couplings can be verified. For that, we propose a framework to examine code clones and relate them to change couplings taken from release history analysis. We validated our framework with the open source project Mozilla and the results of the validation show that although the relation is statistically unverifiable it derives a reasonable amount of cases where the relation exists. Therefore, to discover clone candidates for restructuring we additionally propose a set of metrics and a visualization technique. This allows one to spot where a correlation between cloning and change coupling exists and, as a result, which files should be restructured to ease further evolution.","Title: Measuring Spreadsheet Formula Understandability\nYear: 2012\n\nAbstract: Spreadsheets are widely used in industry, because they are flexible and easy to use. Sometimes they are even used for business-critical applications. It is however difficult for spreadsheet users to correctly assess the quality of spreadsheets, especially with respect to their understandability. Understandability of spreadsheets is important, since spreadsheets often have a long lifespan, during which they are used by several users. In this paper, we establish a set of spreadsheet understandability metrics. We start by studying related work and interviewing 40 spreadsheet professionals to obtain a set of characteristics that might contribute to understandability problems in spreadsheets. Based on those characteristics we subsequently determine a number of understandability metrics. To evaluate the usefulness of our metrics, we conducted a series of experiments in which professional spreadsheet users performed a number of small maintenance tasks on a set of spreadsheets from the EUSES spreadsheet corpus. We subsequently calculate the correlation between the metrics and the performance of subjects on these tasks. The results clearly indicate that the number of ranges, the nesting depth and the presence of conditional operations in formulas significantly increase the difficulty of understanding a spreadsheet.","Title: Introduction to MSR 2006\nYear: 2006\n\nAbstract: Software repositories such as source control systems,defect tracking systems,or archived communications between project personnel are used to help manage the progress of software projects.Software practitioners and researchers are beginning to recognize the potential bene .t of mining this information to support the maintenance of software systems,improve software design/reuse,and empirically validate novel ideas and techniques.Research is now proceeding to uncover the ways in which mining these repositories can help to understand software development,to support predictions about software development,and to plan various aspects of software projects.Following the success of the first two iterations of the MSR workshop in 2004 and 2005,MSR 2006 attracted even more submissions:We received 45 papers from 15 different countries.The international program committee accepted 16 full and 12 short papers for presentation at the workshop and inclusion in the proceedings.We are grateful for the excellent and professional review job done by the reviewers on such a tight schedule.","Title: Visualizing multiple evolution metrics\nYear: 2005\n\nAbstract: Observing the evolution of very large software systems needs the analysis of large complex data models and visualization of condensed views on the system. For visualization software metrics have been used to compute such condensed views. However, current techniques concentrate on visualizing data of one particular release providing only insufficient support for visualizing data of several releases.In this paper we present the RelVis visualization approach that concentrates on providing integrated condensed graphical views on source code and release history data of up to n releases. Measures of metrics of source code entities and relationships are composed in Kiviat diagrams as annual rings. Diagrams highlight the good and bad times of an entity and facilitate the identification of entities and relationships with critical trends. They represent potential refactoring candidates that should be addressed first before further evolving the system. The paper provides needed background information and evaluation of the approach with a large open source software project.","Title: Pattern-Supported Architecture Recovery\nYear: 2002\n\nAbstract: Architectural patterns and styles represent important design decisions and thus are valuable abstractions for architecture recovery. Recognizing them is a challenge because styles and patterns basically span several architectural elements and can be implemented in various ways depending on the problem domain and the implementation variants. Our approach uses source code structures as patterns and introduces an iterative and interactive architecture recovery approach built upon such lower-level patterns extracted from source code. Associations between extracted pattern instances and architectural elements such as modules arise which result in new and higher-level views of the software system. These pattern views provide information for a consecutive refinement of pattern definitions to aggregate and abstract higher-level patterns which finally enable the description of a software system's architecture.","Title: Abstracting Module Views from Source Code\nYear: 2004\n\nAbstract: We have investigated an approach for abstracting and visualizing software module views from source code: ArchView computes abstraction metrics that are used to filter and provide architectural elements and relationships of major interest. Source code views can, therefore, be reduced in detail and size resulting in more reasonable and comprehensible module views on software architectures. ArchView focuses on modules and their relationships: source level relationships such as inheritance, call structures, or includes are abstracted to a module level to show basic dependency relationships of modules.","Title: \"A Bug's Life\" Visualizing a Bug Database\nYear: 2007\n\nAbstract: Visualization has long been accepted as a viable means to comprehend large amounts of information. Especially in the context of software evolution a well-designed visualiza- tion is crucial to be able to cope with the sheer data that needs to be analyzed. Many approaches have been investi- gated to visualize evolving systems, but most of them focus on structural data and are useful to answer questions about the structural evolution of a system. In this paper we consider an often neglected type of in- formation, namely the one provided by bug tracking sys- tems, which store data about the problems that various peo- ple, from developers to end users, detected and reported. We first briefly introduce into the context by reporting on the particularities of the present data, and then propose two visualizations to render bugs as first-level entities.","Title: Predicting defect densities in source code files with decision tree learners\nYear: 2006\n\nAbstract: With the advent of open source software repositories the data available for defect prediction in source files increased tremendously. Although traditional statistics turned out to derive reasonable results the sheer amount of data and the problem context of defect prediction demand sophisticated analysis such as provided by current data mining and machine learning techniques.In this work we focus on defect density prediction and present an approach that applies a decision tree learner on evolution data extracted from the Mozilla open source web browser project. The evolution data includes different source code, modification, and defect measures computed from seven recent Mozilla releases. Among the modification measures we also take into account the change coupling, a measure for the number of change-dependencies between source files. The main reason for choosing decision tree learners, instead of for example neural nets, was the goal of finding underlying rules which can be easily interpreted by humans. To find these rules, we set up a number of experiments to test common hypotheses regarding defects in software entities. Our experiments showed, that a simple tree learner can produce good results with various sets of input data.","Title: TUAnalyzer - analyzing templates in C++ code\nYear: 2004\n\nAbstract: We present TUAnalyzer, a novel tool that extracts the template structure of C++ programs on the basis of the GNU C/C++ compiler's internal representation of a C/C++ translation unit. In comparison to other such tools, our tool is capable of supporting the extraction of function invocations that depend on the particular instantiation of C++ templates and to relate them to their particular template instantiation. TUAnalyzer produces RSF format output that can be easily fed into existing visualization and analysis tools such as Rigi or Graphviz. We motivate why this kind of template analysis information is essential to understand real-world legacy C++ applications. We present how our tool extracts this kind of information to allow others to build on our results and further use the template information. The applicability of our tool has been validated on real code as proof of concept. The results obtained with TUAnalyzer enable us and other approaches and tools to perform detailed studies of large (open source) C/C++ projects in the near future.","Title: Extracting dynamic dependencies between web services using vector clocks\nYear: 2011\n\nAbstract: Service Oriented Architecture (SOA) enables organizations to react to requirement changes in an agile manner and to foster the reuse of existing services. However, the dynamic nature of service oriented systems and their agility bear the challenge of properly understanding such systems. In particular, understanding the dependencies among services is a non trivial task, especially if service oriented systems are distributed over several hosts belonging to different departments of an organization. In this paper, we propose an approach to extract dynamic dependencies among web services. The approach is based on the vector clocks, originally conceived and used to order events in a distributed environment. We use the vector clocks to order service executions and to infer causal dependencies among services. We show the feasibility of the approach by implementing it into the Apache CXF framework and instrumenting the SOAP messages. We designed and executed two experiments to investigate the impact of the approach on the response time. The results show a slight increase that is deemed to be low in typical industrial service oriented systems.","Title: Automatically extracting class diagrams from spreadsheets\nYear: 2010\n\nAbstract: The use of spreadsheets to capture information is widespread in industry. Spreadsheets can thus be a wealthy source of domain information. We propose to automatically extract this information and transform it into class diagrams. The resulting class diagram can be used by software engineers to understand, refine, or re-implement the spreadsheet's functionality. To enable the transformation into class diagrams we create a library of common spreadsheet usage patterns. These patterns are localized in the spreadsheet using a two- dimensional parsing algorithm. The resulting parse tree is transformed and enriched with information from the library. We evaluate our approach on the spreadsheets from the Euses Spreadsheet Corpus by comparing a subset of the generated class diagrams with reference class diagrams created manually.","Title: Using association rules to study the co-evolution of production & test code\nYear: 2009\n\nAbstract: Unit tests are generally acknowledged as an important aid to produce high quality code, as they provide quick feedback to developers on the correctness of their code. In order to achieve high quality, well-maintained tests are needed. Ideally, tests co-evolve with the production code to test changes as soon as possible. In this paper, we explore an approach based on association rule mining to determine whether production and test code co-evolve synchronously. Through two case studies, one with an open source and another one with an industrial software system, we show that our association rule mining approach allows one to assess the co-evolution of product and test code in a software project and, moreover, to uncover the distribution of programmer effort over pure coding, pure testing, or a more test-driven-like practice.","Title: Using Run-Time Data for Program Comprehension\nYear: 2003\n\nAbstract: Traditional approaches for program comprehension use static program analysis or dynamic program analysis in the form of execution traces. Our approach, however, makes use of runtime-data such as parameter and object values. Compared to traditional program comprehension techniques, this approach enables fundamentally new ways of program analysis which we have not seen so far. Reflection analysis which allows engineers to understand programs making use of reflective (dynamic) method invocations is one such analysis. Another is object tracing which allows engineers to trace and track the use of a given instance of a class within the program to be understood. In this paper, we present these techniques along with a case study to which we have applied them.","Title: Extracting feature model changes from the Linux kernel using FMDiff\nYear: 2014\n\nAbstract: The Linux kernel feature model has been studied as an example of large scale evolving feature model and yet details of its evolution are not known. We present here a classification of feature changes occurring on the Linux kernel feature model, as well as a tool, FMDiff, designed to automatically extract those changes. With this tool, we obtained the history of more than twenty architecture specific feature models, over ten releases and compared the recovered information with Kconfig file changes. We establish that FMDiff provides a comprehensive view of feature changes and show that the collected data contains promising information regarding the Linux feature model evolution.","Title: Supporting professional spreadsheet users by generating leveled dataflow diagrams\nYear: 2011\n\nAbstract: Thanks to their flexibility and intuitive programming model, spreadsheets are widely used in industry, often for businesscritical applications. Similar to software developers, professional spreadsheet users demand support for maintaining and transferring their spreadsheets. In this paper, we first study the problems and information needs of professional spreadsheet users by means of a survey conducted at a large financial company. Based on these needs, we then present an approach that extracts this information from spreadsheets and presents it in a compact and easy to understand way, with leveled dataflow diagrams. Our approach comes with three different views on the dataflow that allow the user to analyze the dataflow diagrams in a top-down fashion. To evaluate the usefulness of the proposed approach, we conducted a series of interviews as well as nine case studies in an industrial setting. The results of the evaluation clearly indicate the demand for and usefulness of our approach in ease the understanding of spreadsheets.","Title: Quality Assessment Based on Attribute Series of Software Evolution\nYear: 2007\n\nAbstract: Defect density and defect prediction are essential for ef- ficient resource allocation in software evolution. In an em- pirical study we applied data mining techniques for value series based on evolution attributes such as number of au- thors, commit messages, lines of code, bug fix count, etc. Daily data points of these evolution attributes were cap- tured over a period of two months to predict the defects in the subsequent two months in a project. For that, we developed models utilizing genetic programming and lin- ear regression to accurately predict software defects. In our study, we investigated the data of three independent projects, two open source and one commercial software system. The results show that by utilizing series of these attributes we obtain models with high correlation coeffi- cients (between 0.716 and 0.946). Further, we argue that prediction models based on series of a single variable are sometimes superior to the model including all attributes: in contrast to other studies that resulted in size or complexity measures as predictors, we have identified the number of authors and the number of commit messages to versioning systems as excellent predictors of defect densities.","Title: Can developer-module networks predict failures?\nYear: 2008\n\nAbstract: Software teams should follow a well defined goal and keep their work focused. Work fragmentation is bad for efficiency and quality. In this paper we empirically investigate the relationship between the fragmentation of developer contributions and the number of post-release failures. Our approach is to represent developer contributions with a developer-module network that we call contribution network. We use network centrality measures to measure the degree of fragmentation of developer contributions. Fragmentation is determined by the centrality of software modules in the contribution network. Our claim is that central software modules are more likely to be failure-prone than modules located in surrounding areas of the network. We analyze this hypothesis by exploring the network centrality of Microsoft Windows Vista binaries using several network centrality measures as well as linear and logistic regression analysis. In particular, we investigate which centrality measures are significant to predict the probability and number of post-release failures. Results of our experiments show that central modules are more failure-prone than modules located in surrounding areas of the network. Results further confirm that number of authors and number of commits are significant predictors for the probability of post-release failures. For predicting the number of post-release failures the closeness centrality measure is most significant.","Title: Smart views for analyzing problem reports: tool demo\nYear: 2009\n\nAbstract: Issue tracking repositories contain a wealth of information for reasoning about various aspects of software development processes. In this paper, we focus on bug triaging and provide visual means to explore the effort estimation quality and the bug life-cycle of reported problems. Our approach uses a combination of graphical views to investigate details of individual problem reports while maintaining the context provided by the surrounding data population. This enables the detection and detailed analysis of hidden patterns and facilitates the analysis of problem report outliers.","Title: Towards an Integrated View on Architecture and its Evolution\nYear: 2005\n\nAbstract: Information about the evolution of a software architecture can be found in the source basis of a project and in the release history data such as modification and problem reports. Existing approaches deal with these two data sources separately and do not exploit the integration of their analyses. In this paper, we present an architecture analysis approach that provides an integration of both kinds of evolution data. The analysis applies fact extraction and generates specific directed attributed graphs; nodes represent source code entities and edges represent relationships such as accesses, includes, inherits, invokes, and coupling between certain architectural elements. The integration of data is then performed on a meta-model level to enable the generation of architectural views using binary relational algebra. These integrated architectural views show intended and unintended couplings between architectural elements, hence pointing software engineers to locations in the system that may be critical for on-going and future maintenance activities. We demonstrate our analysis approach using a large open source software system.","Title: TA-RE: an exchange language for mining software repositories\nYear: 2006\n\nAbstract: Software repositories have been getting a lot of attention from researchers in recent years. In order to analyze software repositories, it is necessary to first extract raw data from the version control and problem tracking systems. This poses two challenges: (1) extraction requires a non-trivial effort, and (2) the results depend on the heuristics used during extraction. These challenges burden researchers that are new to the community and make it difficult to benchmark software repository mining since it is almost impossible to reproduce experiments done by another team. In this paper we present the TA-RE corpus. TA-RE collects extracted data from software repositories in order to build a collection of projects that will simplify extraction process. Additionally the collection can be used for benchmarking. As the first step we propose an exchange language capable of making sharing and reusing data as simple as possible.","Title: Data clone detection and visualization in spreadsheets\nYear: 2013\n\nAbstract: Spreadsheets are widely used in industry: it is estimated that end-user programmers outnumber programmers by a factor 5. However, spreadsheets are error-prone, numerous companies have lost money because of spreadsheet errors. One of the causes for spreadsheet problems is the prevalence of copy-pasting. In this paper, we study this cloning in spreadsheets. Based on existing text-based clone detection algorithms, we have developed an algorithm to detect data clones in spreadsheets: formulas whose values are copied as plain text in a different location. To evaluate the usefulness of the proposed approach, we conducted two evaluations. A quantitative evaluation in which we analyzed the EUSES corpus and a qualitative evaluation consisting of two case studies. The results of the evaluation clearly indicate that 1) data clones are common, 2) data clones pose threats to spreadsheet quality and 3) our approach supports users in finding and resolving data clones.","Title: Fine-Grained Analysis of Change Couplings\nYear: 2005\n\nAbstract: In software evolution analysis, many approaches analyze release history data available through versioning systems. The recent investigations of CVS data have shown that commonly committed files highlight their change couplings. However, CVS stores modifications on the basis of text but does not track structural changes, such as the insertion, removing, or modification of methods or classes. A detailed analysis whether change couplings are caused by source code couplings or by other textual modifications, such as updates in license terms, is not performed by current approaches. The focus of this paper is on adding structural change information to existing release history data. We present an approach that uses the structure compare services shipped with the Eclipse IDE to obtain the corresponding finegrained changes between two subsequent versions of any Java class. This information supports filtering those change couplings which result from structural changes. So we can distill the causes for change couplings along releases and filter out those that are structurally relevant. The first validation of our approach with a medium-sized open source software system showed that a reasonable amount of change couplings are not caused by source code changes.","Title: Combining micro-blogging and IDE interactions to support developers in their quests\nYear: 2010\n\nAbstract: Software engineers spend a considerable amount of time on program comprehension. Although vendors of Integrated Development Environments (IDEs) and analysis tools address this challenge, current support for reusing and sharing program comprehension knowledge is limited. As a consequence, developers have to go through the time-consuming program understanding phase multiple times, instead of recalling knowledge from their past or other's program comprehension activities. In this paper, we present an approach to making the knowledge gained during the program comprehension process accessible, by combining micro-blog messages with interaction data automatically collected from the IDE. We implemented the approach in an Eclipse plugin called James and performed a first evaluation of the underlying approach effectiveness, assessing the nature and usefulness of the collected messages, as well as the added benefit of combining them with interaction data.","Title: AWPS - Simulation Based Automated Web Performance Analysis and Prediction\nYear: 2010\n\nAbstract: The growing demand for quality and performance has become a discriminating factor in the field of web applications. Advances in software technology and project management have increased the amount of web applications which are developed under these aspects. At the same time the research in web performance simulation and prediction is focusing on complex modeling methods and techniques rather than providing easy to use tools to test and ensure the performance requirements of today’s and tomorrow’s web applications. As a tool for these tasks, the automatic web performance simulation and prediction system (AWPS) will be described. The AWPS is capable of (a) automatically creating an online web performance simulation and (b) conducting trend analysis of the system under test.","Title: Improving defect prediction using temporal features and non linear models\nYear: 2007\n\nAbstract: Predicting the defects in the next release of a large software system is a very valuable asset for the project manger to plan her resources. In this paper we argue that temporal features (or aspects) of the data are central to prediction performance. We also argue that the use of non-linear models, as opposed to traditional regression, is necessary to uncover some of the hidden interrelationships between the features and the defects and maintain the accuracy of the prediction in some cases. Using data obtained from the CVS and Bugzilla repositories of the Eclipse project, we extract a number of temporal features, such as the number of revisions and number of reported issues within the last three months. We then use these data to predict both the location of defects (i.e., the classes in which defects will occur) as well as the number of reported bugs in the next month of the project. To that end we use standard tree-based induction algorithms in comparison with the traditional regression. Our non-linear models uncover the hidden relationships between features and defects, and present them in easy to understand form. Results also show that using the temporal features our prediction model can predict whether a source file will have a defect with an accuracy of 99% (area under ROC curve 0.9251) and the number of defects with a mean absolute error of 0.019 (Spearman's correlation of 0.96).","Title: Change Analysis with Evolizer and ChangeDistiller\nYear: 2009\n\nAbstract: Changes are the heartbeat of a software system. Software must change to reflect new business, pro cess, and technology requirements, or it will become progressively less useful. Software typically grows and becomes more complex, inducing more time and effort for performing changes. Software archives such as source code version-control systems and issue-tracking systems (for bugs and change requests) are rich sources to examine what changes have what impact on the software. A software evolution analysis platform called Evolizer analyzes change histories and potential support for evolution. Change types, a core part of the analysis, help discover significant changes and change patterns. A tool called ChangeDistiller enables fine-grained change type extraction and analysis to reason about coding conventions, control or exception flow, and even code and comment coevolution. An investigation of open source and commercial software systems contributed to a deeper understanding of how researchers can actively support software evolution in an integrated development environment.","Title: Architecture Recovery for Product Families\nYear: 2003\n\nAbstract: Software product families are rarely created right away but they emerge when a domain becomes mature enough to sustain their long-term investments. The typical pattern is to start with a small set of products to quickly enter a new market. As soon as the business proves to be successful new investments are directed to consolidating the software assets. The various products are migrated towards a flexible platform where the assets are shared and new products can be derived from. In order to create and maintain the platform, the organization needs to carry out several activities such as recovering the architectures of single products and product families, designing the reference architecture, isolating the variable parts, and generalizing software components. In this paper, we introduce a product family construction process that exploits related systems and product families, and we describe methods and tools used. We also present an approach for classifying platforms according to platform coverage and variation and describe three techniques to handle variability across single products and whole product families.","Title: AWPS --- an architecture for pro-active web performance management\nYear: 2010\n\nAbstract: The growing demand for quality and performance has become a discriminating factor in the field of software applications. Specifically in the area of web applications, performance has become a key factor for success creating the need for new types of performance evaluation models and methods capable of representing the dynamic characteristics of web environments. In this paper we will recall seminal work in this area and present AWPS, a tool for automatic web performance simulation and prediction. AWPS is capable of (a) automatically creating a web performance simulation and (b) conducting trend analysis of the system under test. The operation and usage of this tool is demonstrated in a case study of a two-tier architecture system.","Title: Using the gini coefficient for bug prediction in eclipse\nYear: 2011\n\nAbstract: The Gini coefficient is a prominent measure to quantify the inequality of a distribution. It is often used in the field of economy to describe how goods, e.g., wealth or farmland, are distributed among people. We use the Gini coefficient to measure code ownership by investigating how changes made to source code are distributed among the developer population. The results of our study with data from the Eclipse platform show that less bugs can be expected if a large share of all changes are accumulated, i.e., carried out, by relatively few developers.","Title: Can we predict types of code changes?: an empirical analysis\nYear: 2012\n\nAbstract: There exist many approaches that help in pointing developers to the change-prone parts of a software system. Although beneficial, they mostly fall short in providing details of these changes. Fine-grained source code changes (SCC) capture such detailed code changes and their semantics on the statement level. These SCC can be condition changes, interface modifications, inserts or deletions of methods and attributes, or other kinds of statement changes. In this paper, we explore prediction models for whether a source file will be affected by a certain type of SCC. These predictions are computed on the static source code dependency graph and use social network centrality measures and object-oriented metrics. For that, we use change data of the Eclipse platform and the Azureus 3 project. The results show that Neural Network models can predict categories of SCC types. Furthermore, our models can output a list of the potentially change-prone files ranked according to their change-proneness, overall and per change type category.","Title: Detecting code smells in spreadsheet formulas\nYear: 2012\n\nAbstract: Spreadsheets are used extensively in business processes around the world and just like software, spreadsheets are changed throughout their lifetime causing maintainability issues. This paper adapts known code smells to spreadsheet formulas. To that end we present a list of metrics by which we can detect smelly formulas and a visualization technique to highlight these formulas in spreadsheets. We implemented the metrics and visualization technique in a prototype tool to evaluate our approach in two ways. Firstly, we analyze the EUSES spreadsheet corpus, to study the occurrence of the formula smells. Secondly, we analyze ten real life spreadsheets, and interview the spreadsheet owners about the identified smells. The results of these evaluations indicate that formula smells are common and that they can reveal real errors and weaknesses in spreadsheet formulas.","Title: Comparing fine-grained source code changes and code churn for bug prediction\nYear: 2011\n\nAbstract: A significant amount of research effort has been dedicated to learning prediction models that allow project managers to efficiently allocate resources to those parts of a software system that most likely are bug-prone and therefore critical. Prominent measures for building bug prediction models are product measures, e.g., complexity or process measures, such as code churn. Code churn in terms of lines modified (LM) and past changes turned out to be significant indicators of bugs. However, these measures are rather imprecise and do not reflect all the detailed changes of particular source code entities during maintenance activities. In this paper, we explore the advantage of using fine-grained source code changes (SCC) for bug prediction. SCC captures the exact code changes and their semantics down to statement level. We present a series of experiments using different machine learning algorithms with a dataset from the Eclipse platform to empirically evaluate the performance of SCC and LM. The results show that SCC outperforms LM for learning bug prediction models.","Title: Breviz: Visualizing Spreadsheets using Dataflow Diagrams\nYear: 2011\n\nAbstract: Spreadsheets are used extensively in industry, often for business critical purposes. In previous work we have analyzed the information needs of spreadsheet professionals and addressed their need for support with the transition of a spreadsheet to a colleague with the generation of data flow diagrams. In this paper we describe the application of these data flow diagrams for the purpose of understanding a spreadsheet with three example cases. We furthermore suggest an additional application of the data flow diagrams: the assessment of the quality of the spreadsheet's design.","Title: An exploratory study of the pull-based software development model\nYear: 2014\n\nAbstract: The advent of distributed version control systems has led to the development of a new paradigm for distributed software development; instead of pushing changes to a central repository, developers pull them from other repositories and merge them locally. Various code hosting sites, notably Github, have tapped on the opportunity to facilitate pull-based development by offering workflow support tools, such as code reviewing systems and integrated issue trackers. In this work, we explore how pull-based software development works, first on the GHTorrent corpus and then on a carefully selected sample of 291 projects. We find that the pull request model offers fast turnaround, increased opportunities for community engagement and decreased time to incorporate contributions. We show that a relatively small number of factors affect both the decision to merge a pull request and the time to process it. We also examine the reasons for pull request rejection and find that technical ones are only a small minority.","Title: EQ-mine: predicting short-term defects for software evolution\nYear: 2007\n\nAbstract: We use 63 features extracted from sources such as versioning and issue tracking systems to predict defects in short time frames of two months. Our multivariate approach covers aspects of software projects such as size, team structure, process orientation, complexity of existing solution, difficulty of problem, coupling aspects, time constrains, and testing data. We investigate the predictability of several severities of defects in software projects. Are defects with high severity difficult to predict? Are prediction models for defects that are discovered by internal staff similar to models for defects reported from the field? We present both an exact numerical prediction of future defect numbers based on regression models as well as a classification of software components as defect-prone based on the C4.5 decision tree. We create models to accurately predict short-term defects in a study of 5 applications composed of more than 8.000 classes and 700.000 lines of code. The model quality is assessed based on 10-fold cross validation.","Title: Asset Recovery and Their Incorporation into Product Lines\nYear: 2005\n\nAbstract: Software product lines aim in having a common platform from which several similar products can be derived. The elements of the platform are called assets and they are managed in an asset base being part of the product line infrastructure. Assets can include own developments, open source or third-party software modules, as well as design and project documents. In the context of the European-wide project FAMILIES we concentrated on techniques used to populate the asset base by recovering assets from existing systems. We present an approach on how to incorporate existing assets into the product line infrastructure. Thereby we explicitly distinguish the asset origins and the different information sources available. The incorporation is a quality-driven process that is backed up by a set of reverse engineering techniques to evaluate the asset's internal quality. The quality assessment of an asset is the critical measurement for industrial development organizations in order to incorporate assets into their product line infrastructure.","Title: Visual patterns in issue tracking data\nYear: 2010\n\nAbstract: Software development teams gather valuable data about features and bugs in issue tracking systems. This information can be used to measure and improve the efficiency and effectiveness of the development process. In this paper we present an approach that harnesses the extraordinary capability of the human brain to detect visual patterns. We specify generic visual process patterns that can be found in issue tracking data. With these patterns we can analyze information about effort estimation, and the length, and sequence of problem resolution activities. In an industrial case study we apply our interactive tool to identify instances of these patterns and discuss our observations. Our approach was validated through extensive discussions with multiple project managers and developers, as well as feedback from the project review board.","Title: Analyzing the Impact of Antipatterns on Change-Proneness Using Fine-Grained Source Code Changes\nYear: 2012\n\nAbstract: Antipatterns are poor solutions to design and implementation problems which are claimed to make object oriented systems hard to maintain. Our recent studies showed that classes with antipatterns change more frequently than classes without antipatterns. In this paper, we detail these analyses by taking into account fine-grained source code changes (SCC) extracted from 16 Java open source systems. In particular we investigate: whether classes with antipatterns are more change-prone (in terms of SCC) than classes without, (2) whether the type of antipattern impacts the change-proneness of Java classes, and (3) whether certain types of changes are performed more frequently in classes affected by a certain antipattern. Our results show that: 1) the number of SCC performed in classes affected by antipatterns is statistically greater than the number of SCC performed in classes with no antipattern, 2) classes participating in the three antipatterns Complex Class, Spaghetti Code, and SwissArmyKnife are more change-prone than classes affected by other antipatterns, and 3) certain types of changes are more likely to be performed in classes affected by certain antipatterns, such as API changes are likely to be performed in classes affected by the Complex Class, Spaghetti Code, and SwissArmyKnife antipatterns.","Title: Using vector clocks to monitor dependencies among services at runtime\nYear: 2011\n\nAbstract: Service-Oriented Architecture (SOA) enable organizations to react to requirement changes in an agile manner and to foster the reuse of existing services. However, the dynamic nature of Service-Oriented Systems and their agility bear the challenge of properly understanding such systems. In particular, understanding the dependencies among services is a non trivial task, especially if service-oriented systems are distributed over several hosts and/or using different SOA technologies. In this paper, we propose an approach to monitor dynamic dependencies among services. The approach is based on the vector clocks, originally conceived and used to order events in a distributed environment. We use the vector clocks to order service executions and to infer causal dependencies among services. In our future work we plan to use this information to study change and failure impact analysis in service-oriented systems.","Title: Analyzing the Evolution of Web Services Using Fine-Grained Changes\nYear: 2012\n\nAbstract: In the service-oriented paradigm web service interfaces are considered contracts between web service subscribers and providers. However, these interfaces are continuously evolving over time to satisfy changes in the requirements and to fix bugs. Changes in a web service interface typically affect the systems of its subscribers. Therefore, it is essential for subscribers to recognize which types of changes occur in a web service interface in order to analyze the impact on his/her systems. In this paper we propose a tool called WSDLDiff to extract fine-grained changes from subsequent versions of a web service interface defined in WSDL. In contrast to existing approaches, WSDLDiff takes into account the syntax of WSDL and extracts the WSDL elements affected by changes and the types of changes. With WSDLDiff we performed a study aimed at analyzing the evolution of web services using the fine-grained changes extracted from the subsequent versions of four real world WSDL interfaces. The results of our study show that the analysis of the fine-grained changes helps web service subscribers to highlight the most frequent types of changes affecting a WSDL interface. This information can be relevant for web service subscribers who want to assess the risk associated to the usage of web services and to subscribe to the most stable ones.","Title: Populating a Release History Database from Version Control and Bug Tracking Systems\nYear: 2003\n\nAbstract: Version control and bug tracking systems contain largeamounts of historical information that can give deep insightinto the evolution of a software project. Unfortunately,these systems provide only insufficient support for a detailedanalysis of software evolution aspects. We addressthis problem and introduce an approach for populating a releasehistory database that combines version data with bugtracking data and adds missing data not covered by versioncontrol systems such as merge points. Then simple queriescan be applied to the structured data to obtain meaningfulviews showing the evolution of a software project. Suchviews enable more accurate reasoning of evolutionary aspectsand facilitate the anticipation of software evolution.We demonstrate our approach on the large Open Sourceproject Mozilla that offers great opportunities to compareresults and validate our approach.","Title: Automated web performance analysis, with a special focus on prediction\nYear: 2008\n\nAbstract: Today the performance of web systems is getting ever more important, as the number of users and competitors is still increasing. Therefore performance analysis tools gain importance too. There are currently several tools on the market that ensure and test for adequate performance. There are a number of simulation tools and monitoring tools, but only few that automatise and combine both approaches. This paper outlines a system that is capable of (a) automatically creating a web performance simulation and (b) conducting trend analysis of the system under test (SUT). The system requires input information like monitoring points and static-information about the SUT. Based on this information a simulation model of the system is generated. Then the simulation model is refined stepwise e.g. by adding or removing connections between the model components or adjusting the parameters until the aimed accuracy is achieved. By using this simulation model the prediction module creates an analysis of the SUT, and thereby provides as much information as possible about the current state of the system and potential trends. This predictive information can be used for pro-active server tuning or other performance optimisations. The special focus of this work is on the adjustment and prediction parts of the system described here. For all the other parts existing tools and techniques will be used wherever possible.","Title: Towards the Integration of Versioning Systems, Bug Reports and Source Code Meta-Models\nYear: 2005\n\nAbstract: Versioning system repositories and bug tracking systems are valuable sources of information to study the evolution of large open source software systems. However, being conceived for specific purposes, i.e., to support the development or trigger maintenance activities, they do neither allow an easy information browsing nor support the study of software evolution. For example, queries such as locating and browsing the faultiest methods are not provided.This paper addresses such issues and proposes an approach and a framework to consistently merge information extracted from source code, versioning repositories and bug reports. Our information representation exploits the property concepts of the FAMIX information exchange meta-model, allowing to represent, browse, and query, at different level of abstractions, the concept of interest. This allows the user to navigate back and forth from versioning system modification reports to bug reports and to source code. This paper presents the analysis framework and approaches to populate it, tools developed and under development for it, as well as lessons learned while analyzing several releases of Mozilla.","Title: Improving Fact Extraction of Framework-Based Software Systems\nYear: 2003\n\nAbstract: Modern software frameworks provide a set of commonand prefabricated software artifacts that supportengineers in developing large-scale software systems.Framework-related information can be implemented insource code, comments or configuration files, but in thelatter two cases, current reverse engineering approachesmiss important facts reducing the quality of subsequentanalysis tasks. We introduce a generic fact extractionapproach for framework-based systems by combiningtraditional parsing with lexical pattern matching toobtain framework-specific facts from all three sources.We evaluate our approach with an industrial softwareapplication that was built using the Avalon/Phoenixframework. In particular we give examples to point outthe benefits of considering framework-related informationand reflect experiences made during the case study.","Title: Analyzing and Relating Bug Report Data for Feature Tracking\nYear: 2003\n\nAbstract: Gaining higher level evolutionary information aboutlarge software systems is a key in validating past and adjustingfuture development processes. In this paper, we analyzethe proximity of software features based on modificationand problem report data that capture the system's evolutionhistory. Features are instrumented and tracked, therelationships of modification and problem reports to thesefeatures are established, and the tracked features are visualizedto illustrate their otherwise hidden dependencies.Our approach uncovers these hidden relationships betweenfeatures via problem report analysis and presents them ineasy-to-evaluate visual form. Particular feature dependenciesthen can be selected to assess the feature evolution byzooming in into an arbitrary level of detail. Such visualizationof interwoven features, therefore, can indicate locationsof design erosion in the architectural evolution of asoftware system. Our approach has been validated usingthe large open source software project of Mozilla and itsbug reporting system Bugzilla.","Title: Detecting similar Java classes using tree algorithms\nYear: 2006\n\nAbstract: Similarity analysis of source code is helpful during development to provide, for instance, better support for code reuse. Consider a development environment that analyzes code while typing and that suggests similar code examples or existing implementations from a source code repository. Mining software repositories by means of similarity measures enables and enforces reusing existing code and reduces the developing effort needed by creating a shared knowledge base of code fragments. In information retrieval similarity measures are often used to find documents similar to a given query document. This paper extends this idea to source code repositories. It introduces our approach to detect similar Java classes in software projects using tree similarity algorithms. We show how our approach allows to find similar Java classes based on an evaluation of three tree-based similarity measures in the context of five user-defined test cases as well as a preliminary software evolution analysis of a medium-sized Java project. Initial results of our technique indicate that it (1) is indeed useful to identify similar Java classes, (2)successfully identifies the ex ante and ex post versions of refactored classes, and (3) provides some interesting insights into within-version and between-version dependencies of classes within a Java project.","Title: CodeCrawler: an information visualization tool for program comprehension\nYear: 2005\n\nAbstract: CodeCrawler (in the remainder of the text CC) is a language independent, interactive, information visualization tool. It is mainly targeted at visualizing object-oriented software, and has been successfully validated in several industrial case studies over the past few years. CC adheres to lightweight principles: it implements and visualizes polymetric views, visualizations of software enriched with information such as software metrics and other source code semantics. CC is built on top of Moose, an extensible language independent reengineering environment that implements the FAMIX metamodel. In its last implementation, CC has become a general-purpose information visualization tool.","Title: Domain-Specific Languages in Practice: A User Study on the Success Factors\nYear: 2009\n\nAbstract: In this paper we present an empirical study on the use of a domain-specific language(DSL) in industry. This DSL encapsulates the details of services that communicate using Windows Communication Foundation (WCF). From definitions of the data contracts between clients and servers, WCF/C# code for service plumbing is generated. We conducted a survey amongst developers that use this DSL while developing applications for customers. The DSL has been used in about 30 projects all around the world. We describe the known success factors of the use of DSLs, such as improved maintainability and ease of re-use, and assert how well this DSL scores on all of them. The analysis of the results of this case study also shows which conditions should be fulfilled in order to increase the chances of success in using a DSL in a real life case.","Title: Detecting and refactoring code smells in spreadsheet formulas\nYear: 2015\n\nAbstract: Spreadsheets are used extensively in business processes around the world and just like software, spreadsheets are changed throughout their lifetime causing understandability and maintainability issues. This paper adapts known code smells to spreadsheet formulas. To that end we present a list of metrics by which we can detect smelly formulas; a visualization technique to highlight these formulas in spreadsheets and a method to automatically suggest refactorings to resolve smells. We implemented the metrics, visualization and refactoring suggestions techniques in a prototype tool and evaluated our approach in three studies. Firstly, we analyze the EUSES spreadsheet corpus, to study the occurrence of the formula smells. Secondly, we analyze ten real life spreadsheets, and interview the spreadsheet owners about the identified smells. Finally, we generate refactoring suggestions for those ten spreadsheets and study the implications. The results of these evaluations indicate that formula smells are common, that they can reveal real errors and weaknesses in spreadsheet formulas and that in simple cases they can be refactored.","Title: Evaluating Feature Change Impact on Multi-product Line Configurations Using Partial Information.\nYear: 2015\n\nAbstract: Evolving large-scale, complex and highly variable systems is known to be a difficult task, where a single change can ripple through various parts of the system with potentially undesirable effects. In the case of product lines, and moreover multi-product lines, a change may affect only certain variants or certain combinations of features, making the evaluation of change effects more difficult.","Title: A Genetic Algorithm to Find the Adequate Granularity for Service Interfaces\nYear: 2014\n\nAbstract: The relevance of the service interfaces' granularity and its architectural impact have been widely investigated in literature. Existing studies show that the granularity of a service interface, in terms of exposed operations, should reflect their clients' usage. This idea has been formalized in the Consumer-Driven Contracts pattern (CDC). However, to the best of our knowledge, no studies propose techniques to assist providers in finding the right granularity and in easing the adoption of the CDC pattern. In this paper, we propose a genetic algorithm that mines the clients' usage of service operations and suggests Façade services whose granularity reflect the usage of each different type of clients. These services can be deployed on top of the original service and they become contracts for the different types of clients satisfying the CDC pattern. A first study shows that the genetic algorithm is capable of finding Façade services and outperforms a random search approach.","Title: A Manual Categorization of Android App Development Issues on Stack Overflow\nYear: 2014\n\nAbstract: While many tutorials, code examples, and documentation about Android APIs exist, developers still face various problems with the implementation of Android Apps. Many of these issues are discussed on Q&A-sites, such as Stack Overflow. In this paper we present a manual categorization of 450 Android related posts of Stack Overflow concerning their question and problem types. The idea is to find dependencies between certain problems and question types to get better insights into issues of Android App development. The categorization is developed using card sorting with three experienced Android App developers. An initial approach to automate the classification of Stack Overflow posts using Lucene is also presented. The study highlights that the most common question types are 'How to
?' and 'What is the problem
?'. The problems that are discussed most often are related to 'User Interface' and 'Core Elements'. In particular, the problem category 'Layout' is often related to 'What is the problem
?' and 'Frameworks' issues often come with 'Is it possible
?' questions.","Title: A Tool for Visual Understanding of Source Code Dependencies\nYear: 2008\n\nAbstract: Many program comprehension tools use graphs to visualize and analyze source code. The main issue is that existing approaches create graphs overloaded with too much information. Graphs contain hundreds of nodes and even more edges that cross each other. Understanding these graphs and using them for a given program comprehension task is tedious, and in the worst case developers stop using the tools. In this paper we present DA4Java, a graph-based approach for visualizing and analyzing static dependencies between Java source code entities. The main contribution of DA4Java is a set of features to incrementally compose graphs and remove irrelevant nodes and edges from graphs. This leads to graphs that contain significantly fewer nodes and edges and need less effort to understand.","Title: Automated Web Performance Analysis\nYear: 2008\n\nAbstract: Performance is a key feature in many systems nowadays. There are several tools on the market that ensure and test for adequate performance. They, can be divided into simulation tools and monitoring tools. But only a few automatise and combine both approaches. This paper describes a system capable of automatically creating a web performance simulation and conducting trend analysis of the system under test (SUT). To achieve this the system requires input information, like Monitoring Points and Static-Information about the SUT. The system monitors and analyses the SUT and based on this information generates a simulation model of the system. The simulation model is refinded stepwise e.g. by adding or removing connections between the model components or adjusting the parameters until the aimed accuracy is achieved. With the help of the simulation model a prediction module creates an analysis of the SUT, and thereby can give as much information about the current state of the system and potential trends as possible. This predictive information can be used for pro-active server tuning or other performance optimisations. The focus of my PhD thesis is on the adjustment and prediction part of the system described here. For all other parts, already existing tools and techniques will be used where possible. This initial paper outlines the complete system.","Title: Detecting and visualizing inter-worksheet smells in spreadsheets\nYear: 2012\n\nAbstract: Spreadsheets are often used in business, for simple tasks, as well as for mission critical tasks such as finance or forecasting. Similar to software, some spreadsheets are of better quality than others, for instance with respect to usability, maintainability or reliability. In contrast with software however, spreadsheets are rarely checked, tested or certified. In this paper, we aim at developing an approach for detecting smells that indicate weak points in a spreadsheet's design. To that end we first study code smells and transform these code smells to their spreadsheet counterparts. We then present an approach to detect the smells, and to communicate located smells to spreadsheet users with data flow diagrams. To evaluate our apporach, we analyzed occurrences of these smells in the Euses corpus. Furthermore we conducted ten case studies in an industrial setting. The results of the evaluation indicate that smells can indeed reveal weaknesses in a spreadsheet's design, and that data flow diagrams are an appropriate way to show those weaknesses.","Title: Interactive views for analyzing problem reports\nYear: 2009\n\nAbstract: Issue tracking repositories contain a wealth of informa- tion for reasoning about various aspects of software develop- ment processes. In this paper, we focus on bug triaging and provide visual means to explore the effort estimation quality and the bug life-cycle of reported problems. Our approach follows the Micro/Macro reading technique and uses a combination of graphical views to investigate de- tails of individual problem reports while maintaining the context provided by the surrounding data population. This enables the detection and detailed analysis of hidden pat- terns and facilitates the analysis of problem report outliers. In an industrial study, we use our approach in various problem report analysis scenarios and answer questions related to effort estimation and resource planning.","Title: Towards a Weighted Voting System for Q&A Sites\nYear: 2013\n\nAbstract: Q&A sites have become popular to share and look for valuable knowledge. Users can easily and quickly access high quality answers to common questions. The main mechanism to label good answers is to count the votes per answer. This mechanism, however, does not consider whether other answers were present at the time when a vote is given. Consequently, good answers that were given later are likely to receive less votes than they would have received if given earlier. In this paper we present a Weighted Votes (WV) metric that gives different weights to the votes depending on how many answers were present when the vote is performed. The idea behind WV is to emphasize the answer that receives most of the votes when most of the answers were already posted. Mining the Stack Overflow data dump we show that the WV metric is able to highlight between 4.07% and 10.82% answers that differ from the most voted ones.","Title: Refactoring Fat Interfaces Using a Genetic Algorithm\nYear: 2014\n\nAbstract: Recent studies have shown that the violation of the Interface Segregation Principle (ISP) is critical for maintaining and evolving software systems. Fat interfaces (i.e., interfaces violating the ISP) change more frequently and degrade the quality of the components coupled to them. According to the ISP the interfaces' design should force no client to depend on methods it does not invoke. Fat interfaces should be split into smaller interfaces exposing only the methods invoked by groups of clients. However, applying the ISP is a challenging task when fat interfaces are invoked differently by many clients. In this paper, we formulate the problem of applying the ISP as a multi-objective clustering problem and we propose a genetic algorithm to solve it. We evaluate the capability of the proposed genetic algorithm with 42,318 public Java APIs whose clients' usage has been mined from the Maven repository. The results of this study show that the genetic algorithm outperforms other search based approaches (i.e., random and simulated annealing approaches) in splitting the APIs according to the ISP.","Title: Method-level bug prediction\nYear: 2012\n\nAbstract: Researchers proposed a wide range of approaches to build effective bug prediction models that take into account multiple aspects of the software development process. Such models achieved good prediction performance, guiding developers towards those parts of their system where a large share of bugs can be expected. However, most of those approaches predict bugs on file-level. This often leaves developers with a considerable amount of effort to examine all methods of a file until a bug is located. This particular problem is reinforced by the fact that large files are typically predicted as the most bug-prone. In this paper, we present bug prediction models at the level of individual methods rather than at file-level. This increases the granularity of the prediction and thus reduces manual inspection efforts for developers. The models are based on change metrics and source code metrics that are typically used in bug prediction. Our experiments---performed on 21 Java open-source (sub-)systems---show that our prediction models reach a precision and recall of 84% and 88%, respectively. Furthermore, the results indicate that change metrics significantly outperform source code metrics.","Title: Synonym suggestion for tags on stack overflow\nYear: 2015\n\nAbstract: The amount of diverse tags used to classify posts on Stack Overflow increased in the last years to more than 38,000 tags. Many of these tags have the same or similar meaning. Stack Overflow provides an approach to reduce the amount of tags by allowing privileged users to manually create synonyms. However, currently exist only 2,765 synonym-pairs on Stack Overflow that is quite low compared to the total number of tags.\n\nTo comprehend how synonym-pairs are built, we manually analyzed the tags and how the synonyms could be created automatically. Based on our findings, we then present TSST, a tag synonym suggestion tool, that outputs a ranked list of possible synonyms for each input tag.\n\nWe first evaluated TSST with the 2,765 approved synonym-pairs of Stack Overflow. For 88.4% of the tags TSST finds the correct synonyms, for 72.2% the correct synonym is within the top 10 suggestions. In addition, we applied TSST to 10 randomly selected Android related tags and evaluated the suggested synonyms with 20 Android app developers in an online survey. Overall, in 80% of their ratings, developers found an adequate synonym suggested by TSST.","Title: Predicting Build Co-changes with Source Code Change and Commit Categories\nYear: 2016\n\nAbstract: When software is maintained and evolved the build configuration also needs to be updated. Knowing when to update the build configuration is typically done manually with the risk of missing an update and breaking the build. To mitigate this risk, previous work has investigated prediction models to help developers to identify commits that will likely involve an update of the build configuration. In this paper, we investigate whether we can improve these existing prediction models by taking into account detailed information on source code changes and commit categories. Our main hypothesis is that such detailed information on changes will significantly improve the prediction of build co-changes. To that extent, we extract information on changes from 10 Java open source projects and use a random forest classifier to train models that predict build co-changes within and across projects. Our results show significant improvements over existing prediction models: the AUC for intra-and cross-project prediction improves by 11.54% and 9.46% respectively. In addition, we investigate advanced resampling techniques to explore the effect of unbalanced data on our models. The results show that SMOTE can particularly improve prediction models with low performance that were trained on unbalanced data. Our models improve the prediction and enable a better understanding of build co-changes.","Title: FEVER: extracting feature-oriented changes from commits.\nYear: 2016\n\nAbstract: The study of the evolution of highly configurable systems requires a thorough understanding of thee core ingredients of such systems: (1) the underlying variability model; (2) the assets that together implement the configurable features; and (3) the mapping from variable features to actual assets. Unfortunately, to date no systematic way to obtain such information at a sufficiently fine grained level exists.\n\nTo remedy this problem we propose FEVER and its instantiation for the Linux kernel. FEVER extracts detailed information on changes in variability models (KConfig files), assets (preprocessor based C code), and mappings (Makefiles). We describe how FEVER works, and apply it to several releases of the Linux kernel. Our evaluation on 300 randomly selected commits, from two different releases, shows our results are accurate in 82.6% of the commits. Furthermore, we illustrate how the populated FEVER graph database thus obtained can be used in typical Linux engineering tasks.","Title: Grouping android tag synonyms on stack overflow.\nYear: 2016\n\nAbstract: On Stack Overflow, more than 38,000 diverse tags are used to classify posts. The Stack Overflow community provides tag synonyms to reduce the number of tags that have the same or similar meaning. In our previous research, we used those synonym pairs to derive a number of strategies to create tag synonyms automatically.\n\nIn this work, we continue this line of research and present an approach to group tag synonyms to meaningful topics. We represent our synonyms as directed, weighted graphs, and investigate several graph community detection algorithms to build meaningful groups of tags, also called tag communities.\n\nWe apply our approach to the tags obtained from Android-related Stack Overflow posts and evaluate the resulting tag communities quantitatively with various community metrics. In addition, we evaluate our approach qualitatively through a manual inspection and comparison of a random sample of tag communities. Our results show that we can cluster the Android tags to 2,481 meaningful tag communities. We also show how these tag communities can be used to derive trends of topics of Android-related questions on Stack Overflow.","Title: Extracting Build Changes with BUILDDIFF.\nYear: 2017\n\nAbstract: Build systems are an essential part of modern software engineering projects. As software projects change continuously, it is crucial to understand how the build system changes because neglecting its maintenance can lead to expensive build breakage. Recent studies have investigated the (co-)evolution of build configurations and reasons for build breakage, but they did this only on a coarse grained level. In this paper, we present BuildDiff, an approach to extract detailed build changes from Maven build files and classify them into 95 change types. In a manual evaluation of 400 build changing commits, we show that BuildDiff can extract and classify build changes with an average precision and recall of 0.96 and 0.98, respectively. We then present two studies using the build changes extracted from 30 open source Java projects to study the frequency and time of build changes. The results show that the top 10 most frequent change types account for 73% of the build changes. Among them, changes to version numbers and changes to dependencies of the projects occur most frequently. Furthermore, our results show that build changes occur frequently around releases. With these results, we provide the basis for further research, such as for analyzing the (co-)evolution of build files with other artifacts or improving effort estimation approaches. Furthermore, our detailed change information enables improvements of refactoring approaches for build configurations and improvements of models to identify error-prone build files.","Title: Extracting Timed Automata from Java Methods\nYear: 2017\n\nAbstract: The verification of the time behavior in distributed, multi-threaded programs is challenging, mainly because modern programming languages only provide means to represent time without a proper semantics. Current approaches to extract time models from source code represent time only as a sequence of events or require developers to manually provide a formal model of the time behavior. This makes it difficult for developers to verify various aspects of their systems, such as timeouts, delays and periodicity of the execution. In this paper, we introduce a definition of the time semantics of the Java programming language. Based on the semantics, we present an approach to automatically extract timed automata and their time constraints from the Java methods source code. First, we detect Java statements which involve time, from which we then extract the timed automata that are directly amenable to the verification of time properties of the methods. We evaluated the accuracy of our approach on ten open source Java projects that heavily use time in their source code. The results show a precision of 98.62% and recall of 95.37% in extracting time constraints from Java code. Finally, we demonstrate the effectiveness of our approach with five reported bugs of four different Apache systems that we could confirm.","Title: Automatically repairing dependency-related build breakage\nYear: 2018\n\nAbstract: Build systems are widely used in today's software projects to automate integration and build processes. Similar to source code, build specifications need to be maintained to avoid outdated specifications, and build breakage as a consequence. Recent work indicates that neglected build maintenance is one of the most frequently occurring reasons why open source and proprietary builds break. In this paper, we propose BuildMedic, an approach to automatically repair Maven builds that break due to dependency-related issues. Based on a manual investigation of 37 broken Maven builds in 23 open source Java projects, we derive three repair strategies to automatically repair the build, namely Version Update, Delete Dependency, and Add Repository. We evaluate the three strategies on 84 additional broken builds from the 23 studied projects in order to demonstrate the applicability of our approach. The evaluation shows that BuildMedic can automatically repair 45 of these broken builds (54%). Furthermore, in 36% of the successfully repaired build breakages, BuildMedic outputs at least one repair candidate that is considered a correct repair. Moreover, 76% of them could be repaired with only a single dependency correction.","Title: FEVER: An approach to analyze feature-oriented changes and artefact co-evolution in highly configurable systems.\nYear: 2018\n\nAbstract: The evolution of highly configurable systems is known to be a challenging task. Thorough understanding of configuration options their relationships, and their implementation in various types of artefacts (variability model, mapping, and implementation) is required to avoid compilation errors, invalid products, or dead code. Recent studies focusing on co-evolution of artefacts detailed feature-oriented change scenarios, describing how related artefacts might change over time. However, relying on manual analysis of commits, such work do not provide the means to obtain quantitative information on the frequency of described scenarios nor information on the exhaustiveness of the presented scenarios for the evolution of a large scale system. In this work, we propose FEVER and its instantiation for the Linux kernel. FEVER extracts detailed information on changes in variability models (KConfig files), assets (preprocessor based C code), and mappings (Makefiles). We apply this methodology to the Linux kernel and build a dataset comprised of 15 releases of the kernel history. We performed an evaluation of the FEVER approach by manually inspecting the data and compared it with commits in the system’s history. The evaluation shows that FEVER accurately captures feature related changes for more than 85% of the 810 manually inspected commits. We use the collected data to reflect on occurrences of co-evolution in practice. Our analysis shows that complex co-evolution scenarios occur in every studied release but are not among the most frequent change scenarios, as they only occur for 8 to 13% of the evolving features. Moreover, only a minority of developers working on a given release will make changes to all artefacts related to a feature (between 10% and 13% of authors). While our conclusions are derived from observations on the evolution of the Linux kernel, we believe that they may have implications for tool developers as well as guide further research in the field of co-evolution of artefacts.","Title: Modeling time in Java programs for automatic error detection.\nYear: 2018\n\nAbstract: Modern programming languages, such as Java, represent time as integer variables, called timestamps. Timestamps allow developers to tacitly model incorrect time values resulting in a program failure because any negative value or every positive value is not necessarily a valid time representation. Current approaches to automatically detect errors in programs, such as Randoop and FindBugs, cannot detect such errors because they treat timestamps as normal integer variables and test them with random values verifying if the program throws an exception. In this paper, we present an approach that considers the time semantics of the Java language to systematically detect time related errors in Java programs. With the formal time semantics, our approach determines which integer variables handle time and which statements use or alter their values. Based on this information, it translates these statements into an SMT model that is passed to an SMT solver. The solver formally verifies the correctness of the model and reports the violations of time properties in that program. For the evaluation, we have implemented our approach in a prototype tool and applied it to the source code of 20 Java open source projects. The results show that our approach is scalable and it is capable of detecting time errors precisely enough allowing its usability in real-world applications.","Title: Noise and heterogeneity in historical build data: an empirical study of Travis CI.\nYear: 2018\n\nAbstract: Automated builds, which may pass or fail, provide feedback to a development team about changes to the codebase. A passing build indicates that the change compiles cleanly and tests (continue to) pass. A failing (a.k.a., broken) build indicates that there are issues that require attention. Without a closer analysis of the nature of build outcome data, practitioners and researchers are likely to make two critical assumptions: (1) build results are not noisy; however, passing builds may contain failing or skipped jobs that are actively or passively ignored; and (2) builds are equal; however, builds vary in terms of the number of jobs and configurations. \n\n To investigate the degree to which these assumptions about build breakage hold, we perform an empirical study of 3.7 million build jobs spanning 1,276 open source projects. We find that: (1) 12% of passing builds have an actively ignored failure; (2) 9% of builds have a misleading or incorrect outcome on average; and (3) at least 44% of the broken builds contain passing jobs, i.e., the breakage is local to a subset of build variants. Like other software archives, build data is noisy and complex. Analysis of build data requires nuance.","Title: DiffViz: A Diff Algorithm Independent Visualization Tool for Edit Scripts\nYear: 2018\n\nAbstract: A number of approaches and tools exist that extract and visualize the changes between two versions of a file and thereby help developers to understand them. DiffViz is an interactive visualization tool that visualizes the changes independent from the differencing algorithm. It supports, but is not limited to, a granularity on the level of abstract syntax trees. Furthermore, it provides several new features, such as node matching and the mini-map, to navigate and analyze the changes. A demo of the installation and example usage of the tool is available here: https://youtu.be/RF93ey9GYoc.","Title: Towards Model Checking Security of Real Time Java Software.\nYear: 2018\n\nAbstract: More and more software libraries and applications in high-performance computing and distributed systems are coded using the Java programming language. The correctness of such pieces of code w.r.t. a given set of security policies often depends on the correct handling of timing between concurrent or recurrent events. Model-checking has proven to be an effective tool for verifying the correctness of software. In spite of the growing importance of this application area of formal methods, though, no approach exists that targets the problem of verifying the correctness of real-time software w.r.t. timed specifications. The few existing works focus on very different problems, such as schedulability analysis of Java tasks. In this paper we present an approach combining rule-based static analysis together with symbolic execution of Java code to extract networks of timed automata from existing software and then use Uppaal to model-check them against timed specifications. We show through a real-world case study that this approach can be helpful in model-checking security policies of real-time Java software.","Title: Automatically classifying posts into question categories on stack overflow.\nYear: 2018\n\nAbstract: Software developers frequently solve development issues with the help of question and answer web forums, such as Stack Overflow (SO). While tags exist to support question searching and browsing, they are more related to technological aspects than to the question purposes. Tagging questions with their purpose can add a new dimension to the investigation of topics discussed in posts on SO. In this paper, we aim to automate such a classification of SO posts into seven question categories. As a first step, we have manually created a curated data set of 500 SO posts, classified into the seven categories. Using this data set, we apply machine learning algorithms (Random Forest and Support Vector Machines) to build a classification model for SO questions. We then experiment with 82 different configurations regarding the preprocessing of the text and representation of the input data. The results of the best performing models show that our models can classify posts into the correct question category with an average precision and recall of 0.88 and 0.87 when using Random Forest and the phrases indicating a question category as input data for the training. The obtained model can be used to aid developers in browsing SO discussions or researchers in building recommenders based on SO.","Title: Can I Depend on you? Mapping the Dependency and Quality Landscape of ROS Packages\nYear: 2019\n\nAbstract: Since its beginnings ten years ago, the Robot Operating System (ROS) has created a huge community of developers and researchers and is now the most widespread open-source framework for robotics development. It is used in research, prototyping but also in commercial products and supports a wide range of robotic platforms, sensors and high-level data processing functions. While for a research platform, quality of the software developed with it is typically of lower importance, ROS is gradually moving towards industrial applications making software quality a premier topic. In this paper, we want to gain insights on how ROS is used in practice, how high the quality of the ROS packages and applications is, and where potential pitfalls in the use of ROS lie. To achieve this, we have analyzed several thousands of open-source ROS packages found on GitHub and Bitbucket for their quality and their interdependencies. Our results include packages on the rosdistro index and, more importantly, packages that are not. From our results we show common issues in ROS-applications, quality implications, and also which packages of what quality are particularly popular.","Title: Semantics-driven extraction of timed automata from Java programs\nYear: 2019\n\nAbstract: The automatic verification of time properties of models extracted from programs is challenging, mainly because modern programming languages, such as Java, represent time without a proper semantics. Current approaches to extract time models from source code either represent time only as a tree-like sequence of events or require developers to manually provide a formal model of the time behavior. This makes it difficult for software developers to verify various aspects of their systems, such as timeouts, delays and periodicity of the execution. In this paper, we introduce a formal definition of the time semantics for the Java programming language. Based on the semantics, we present an approach to automatically extract timed automata and their time constraints from Java programs at method level. First, our approach detects the Java statements that involve time, from which it then extracts the timed automata. Our extracted automata are directly amenable to the verification of time properties of the corresponding Java methods. We evaluated the accuracy of our approach on twenty open source Java projects that implement time behavior in their source code. The results show that our approach achieves 100% precision and recall in identifying time related information. They also show that 95% of the timed automata extracted from source code correctly model the time behavior of the method. Finally, we show the applicability of our timed automata to identify eight real errors in four open source Apache systems.","Title: A Method for the Joint Analysis of Numerical and Textual IT-System Data to Predict Critical System States\nYear: 2020\n\nAbstract: We present a method for the joint analysis of textual and numerical IT-system data usable to predict possibly critical system states. Towards a comparative discussion culminating in a justified model and method choice, we apply logistic regression, random forest and neural networks to the prediction of critical system states. Our models consume a set of different monitoring performance metrics and log file events. To ease the analysis of IT-systems, our models judge the future system state using one binary outcome variable for the system state's criticality as \"alarm\" or \"no alarm\". Moreover, we use feature importance measures to give IT-operators guidance on which system parameters, i.e., features, to consider primarily when responding to an alarm. We evaluate our models using different configurations, including (among others) the demanded lead time window for incident response, and a set of common performance measures. This paper is an extension to previous work that adds details on how to jointly process textual and numerical data.","Title: Comparing fine-grained source code changes and code churn for bug prediction - A retrospective\nYear: 2021\n\nAbstract: AbstractMore than two decades ago, researchers started to mine the data stored in software repositories to help software developers in making informed decisions for developing and testing software systems. Bug prediction was one of the most promising and popular research directions that uses the data stored in software repositories to predict the bug-proneness or number of bugs in source files. On that topic and as part of Emanuel's PhD studies, we submitted a paper with the title Comparing fine-grained source code changes and code churn for bug prediction [8] to the 8th Working Conference on Mining Software Engineering, held 2011 in beautiful Honolulu, Hawaii. Ten years later, it got selected as one of the finalists to receive the MSR 2021 Most Influential Paper Award. In the following, we provide a retrospective on our work, describing the road to publishing this paper, its impact in the field of bug prediction, and the road ahead.","Title: Comprehending Spreadsheets: Which Strategies do Users Apply?\nYear: 2021\n\nAbstract: Spreadsheets are a ubiquitous means of communication and decision making, and adequate comprehension is required for reading, interpretation as well as maintenance. Despite being around for forty years, only selected aspects of comprehension were studied, mainly focusing on error detection and correction, and largely disregarding a more holistic view on spreadsheet comprehension processes.In this ..."],"string":"[\n \"Title: Using source code metrics to predict change-prone Java interfaces\\nYear: 2011\\n\\nAbstract: Recent empirical studies have investigated the use of source code metrics to predict the change- and defect-proneness of source code files and classes. While results showed strong correlations and good predictive power of these metrics, they do not distinguish between interface, abstract or concrete classes. In particular, interfaces declare contracts that are meant to remain stable during the evolution of a software system while the implementation in concrete classes is more likely to change. This paper aims at investigating to which extent the existing source code metrics can be used for predicting change-prone Java interfaces. We empirically investigate the correlation between metrics and the number of fine-grained source code changes in interfaces of ten Java open-source systems. Then, we evaluate the metrics to calculate models for predicting change-prone Java interfaces. Our results show that the external interface cohesion metric exhibits the strongest correlation with the number of source code changes. This metric also improves the performance of prediction models to classify Java interfaces into change-prone and not change-prone.\",\n \"Title: Relation of code clones and change couplings\\nYear: 2006\\n\\nAbstract: Code clones have long been recognized as bad smells in software systems and are considered to cause maintenance problems during evolution. It is broadly assumed that the more clones two files share, the more often they have to be changed together. This relation between clones and change couplings has been postulated but neither demonstrated nor quantified yet. However, given such a relation it would simplify the identification of restructuring candidates and reduce change couplings. In this paper, we examine this relation and discuss if a correlation between code clones and change couplings can be verified. For that, we propose a framework to examine code clones and relate them to change couplings taken from release history analysis. We validated our framework with the open source project Mozilla and the results of the validation show that although the relation is statistically unverifiable it derives a reasonable amount of cases where the relation exists. Therefore, to discover clone candidates for restructuring we additionally propose a set of metrics and a visualization technique. This allows one to spot where a correlation between cloning and change coupling exists and, as a result, which files should be restructured to ease further evolution.\",\n \"Title: Measuring Spreadsheet Formula Understandability\\nYear: 2012\\n\\nAbstract: Spreadsheets are widely used in industry, because they are flexible and easy to use. Sometimes they are even used for business-critical applications. It is however difficult for spreadsheet users to correctly assess the quality of spreadsheets, especially with respect to their understandability. Understandability of spreadsheets is important, since spreadsheets often have a long lifespan, during which they are used by several users. In this paper, we establish a set of spreadsheet understandability metrics. We start by studying related work and interviewing 40 spreadsheet professionals to obtain a set of characteristics that might contribute to understandability problems in spreadsheets. Based on those characteristics we subsequently determine a number of understandability metrics. To evaluate the usefulness of our metrics, we conducted a series of experiments in which professional spreadsheet users performed a number of small maintenance tasks on a set of spreadsheets from the EUSES spreadsheet corpus. We subsequently calculate the correlation between the metrics and the performance of subjects on these tasks. The results clearly indicate that the number of ranges, the nesting depth and the presence of conditional operations in formulas significantly increase the difficulty of understanding a spreadsheet.\",\n \"Title: Introduction to MSR 2006\\nYear: 2006\\n\\nAbstract: Software repositories such as source control systems,defect tracking systems,or archived communications between project personnel are used to help manage the progress of software projects.Software practitioners and researchers are beginning to recognize the potential bene .t of mining this information to support the maintenance of software systems,improve software design/reuse,and empirically validate novel ideas and techniques.Research is now proceeding to uncover the ways in which mining these repositories can help to understand software development,to support predictions about software development,and to plan various aspects of software projects.Following the success of the first two iterations of the MSR workshop in 2004 and 2005,MSR 2006 attracted even more submissions:We received 45 papers from 15 different countries.The international program committee accepted 16 full and 12 short papers for presentation at the workshop and inclusion in the proceedings.We are grateful for the excellent and professional review job done by the reviewers on such a tight schedule.\",\n \"Title: Visualizing multiple evolution metrics\\nYear: 2005\\n\\nAbstract: Observing the evolution of very large software systems needs the analysis of large complex data models and visualization of condensed views on the system. For visualization software metrics have been used to compute such condensed views. However, current techniques concentrate on visualizing data of one particular release providing only insufficient support for visualizing data of several releases.In this paper we present the RelVis visualization approach that concentrates on providing integrated condensed graphical views on source code and release history data of up to n releases. Measures of metrics of source code entities and relationships are composed in Kiviat diagrams as annual rings. Diagrams highlight the good and bad times of an entity and facilitate the identification of entities and relationships with critical trends. They represent potential refactoring candidates that should be addressed first before further evolving the system. The paper provides needed background information and evaluation of the approach with a large open source software project.\",\n \"Title: Pattern-Supported Architecture Recovery\\nYear: 2002\\n\\nAbstract: Architectural patterns and styles represent important design decisions and thus are valuable abstractions for architecture recovery. Recognizing them is a challenge because styles and patterns basically span several architectural elements and can be implemented in various ways depending on the problem domain and the implementation variants. Our approach uses source code structures as patterns and introduces an iterative and interactive architecture recovery approach built upon such lower-level patterns extracted from source code. Associations between extracted pattern instances and architectural elements such as modules arise which result in new and higher-level views of the software system. These pattern views provide information for a consecutive refinement of pattern definitions to aggregate and abstract higher-level patterns which finally enable the description of a software system's architecture.\",\n \"Title: Abstracting Module Views from Source Code\\nYear: 2004\\n\\nAbstract: We have investigated an approach for abstracting and visualizing software module views from source code: ArchView computes abstraction metrics that are used to filter and provide architectural elements and relationships of major interest. Source code views can, therefore, be reduced in detail and size resulting in more reasonable and comprehensible module views on software architectures. ArchView focuses on modules and their relationships: source level relationships such as inheritance, call structures, or includes are abstracted to a module level to show basic dependency relationships of modules.\",\n \"Title: \\\"A Bug's Life\\\" Visualizing a Bug Database\\nYear: 2007\\n\\nAbstract: Visualization has long been accepted as a viable means to comprehend large amounts of information. Especially in the context of software evolution a well-designed visualiza- tion is crucial to be able to cope with the sheer data that needs to be analyzed. Many approaches have been investi- gated to visualize evolving systems, but most of them focus on structural data and are useful to answer questions about the structural evolution of a system. In this paper we consider an often neglected type of in- formation, namely the one provided by bug tracking sys- tems, which store data about the problems that various peo- ple, from developers to end users, detected and reported. We first briefly introduce into the context by reporting on the particularities of the present data, and then propose two visualizations to render bugs as first-level entities.\",\n \"Title: Predicting defect densities in source code files with decision tree learners\\nYear: 2006\\n\\nAbstract: With the advent of open source software repositories the data available for defect prediction in source files increased tremendously. Although traditional statistics turned out to derive reasonable results the sheer amount of data and the problem context of defect prediction demand sophisticated analysis such as provided by current data mining and machine learning techniques.In this work we focus on defect density prediction and present an approach that applies a decision tree learner on evolution data extracted from the Mozilla open source web browser project. The evolution data includes different source code, modification, and defect measures computed from seven recent Mozilla releases. Among the modification measures we also take into account the change coupling, a measure for the number of change-dependencies between source files. The main reason for choosing decision tree learners, instead of for example neural nets, was the goal of finding underlying rules which can be easily interpreted by humans. To find these rules, we set up a number of experiments to test common hypotheses regarding defects in software entities. Our experiments showed, that a simple tree learner can produce good results with various sets of input data.\",\n \"Title: TUAnalyzer - analyzing templates in C++ code\\nYear: 2004\\n\\nAbstract: We present TUAnalyzer, a novel tool that extracts the template structure of C++ programs on the basis of the GNU C/C++ compiler's internal representation of a C/C++ translation unit. In comparison to other such tools, our tool is capable of supporting the extraction of function invocations that depend on the particular instantiation of C++ templates and to relate them to their particular template instantiation. TUAnalyzer produces RSF format output that can be easily fed into existing visualization and analysis tools such as Rigi or Graphviz. We motivate why this kind of template analysis information is essential to understand real-world legacy C++ applications. We present how our tool extracts this kind of information to allow others to build on our results and further use the template information. The applicability of our tool has been validated on real code as proof of concept. The results obtained with TUAnalyzer enable us and other approaches and tools to perform detailed studies of large (open source) C/C++ projects in the near future.\",\n \"Title: Extracting dynamic dependencies between web services using vector clocks\\nYear: 2011\\n\\nAbstract: Service Oriented Architecture (SOA) enables organizations to react to requirement changes in an agile manner and to foster the reuse of existing services. However, the dynamic nature of service oriented systems and their agility bear the challenge of properly understanding such systems. In particular, understanding the dependencies among services is a non trivial task, especially if service oriented systems are distributed over several hosts belonging to different departments of an organization. In this paper, we propose an approach to extract dynamic dependencies among web services. The approach is based on the vector clocks, originally conceived and used to order events in a distributed environment. We use the vector clocks to order service executions and to infer causal dependencies among services. We show the feasibility of the approach by implementing it into the Apache CXF framework and instrumenting the SOAP messages. We designed and executed two experiments to investigate the impact of the approach on the response time. The results show a slight increase that is deemed to be low in typical industrial service oriented systems.\",\n \"Title: Automatically extracting class diagrams from spreadsheets\\nYear: 2010\\n\\nAbstract: The use of spreadsheets to capture information is widespread in industry. Spreadsheets can thus be a wealthy source of domain information. We propose to automatically extract this information and transform it into class diagrams. The resulting class diagram can be used by software engineers to understand, refine, or re-implement the spreadsheet's functionality. To enable the transformation into class diagrams we create a library of common spreadsheet usage patterns. These patterns are localized in the spreadsheet using a two- dimensional parsing algorithm. The resulting parse tree is transformed and enriched with information from the library. We evaluate our approach on the spreadsheets from the Euses Spreadsheet Corpus by comparing a subset of the generated class diagrams with reference class diagrams created manually.\",\n \"Title: Using association rules to study the co-evolution of production & test code\\nYear: 2009\\n\\nAbstract: Unit tests are generally acknowledged as an important aid to produce high quality code, as they provide quick feedback to developers on the correctness of their code. In order to achieve high quality, well-maintained tests are needed. Ideally, tests co-evolve with the production code to test changes as soon as possible. In this paper, we explore an approach based on association rule mining to determine whether production and test code co-evolve synchronously. Through two case studies, one with an open source and another one with an industrial software system, we show that our association rule mining approach allows one to assess the co-evolution of product and test code in a software project and, moreover, to uncover the distribution of programmer effort over pure coding, pure testing, or a more test-driven-like practice.\",\n \"Title: Using Run-Time Data for Program Comprehension\\nYear: 2003\\n\\nAbstract: Traditional approaches for program comprehension use static program analysis or dynamic program analysis in the form of execution traces. Our approach, however, makes use of runtime-data such as parameter and object values. Compared to traditional program comprehension techniques, this approach enables fundamentally new ways of program analysis which we have not seen so far. Reflection analysis which allows engineers to understand programs making use of reflective (dynamic) method invocations is one such analysis. Another is object tracing which allows engineers to trace and track the use of a given instance of a class within the program to be understood. In this paper, we present these techniques along with a case study to which we have applied them.\",\n \"Title: Extracting feature model changes from the Linux kernel using FMDiff\\nYear: 2014\\n\\nAbstract: The Linux kernel feature model has been studied as an example of large scale evolving feature model and yet details of its evolution are not known. We present here a classification of feature changes occurring on the Linux kernel feature model, as well as a tool, FMDiff, designed to automatically extract those changes. With this tool, we obtained the history of more than twenty architecture specific feature models, over ten releases and compared the recovered information with Kconfig file changes. We establish that FMDiff provides a comprehensive view of feature changes and show that the collected data contains promising information regarding the Linux feature model evolution.\",\n \"Title: Supporting professional spreadsheet users by generating leveled dataflow diagrams\\nYear: 2011\\n\\nAbstract: Thanks to their flexibility and intuitive programming model, spreadsheets are widely used in industry, often for businesscritical applications. Similar to software developers, professional spreadsheet users demand support for maintaining and transferring their spreadsheets. In this paper, we first study the problems and information needs of professional spreadsheet users by means of a survey conducted at a large financial company. Based on these needs, we then present an approach that extracts this information from spreadsheets and presents it in a compact and easy to understand way, with leveled dataflow diagrams. Our approach comes with three different views on the dataflow that allow the user to analyze the dataflow diagrams in a top-down fashion. To evaluate the usefulness of the proposed approach, we conducted a series of interviews as well as nine case studies in an industrial setting. The results of the evaluation clearly indicate the demand for and usefulness of our approach in ease the understanding of spreadsheets.\",\n \"Title: Quality Assessment Based on Attribute Series of Software Evolution\\nYear: 2007\\n\\nAbstract: Defect density and defect prediction are essential for ef- ficient resource allocation in software evolution. In an em- pirical study we applied data mining techniques for value series based on evolution attributes such as number of au- thors, commit messages, lines of code, bug fix count, etc. Daily data points of these evolution attributes were cap- tured over a period of two months to predict the defects in the subsequent two months in a project. For that, we developed models utilizing genetic programming and lin- ear regression to accurately predict software defects. In our study, we investigated the data of three independent projects, two open source and one commercial software system. The results show that by utilizing series of these attributes we obtain models with high correlation coeffi- cients (between 0.716 and 0.946). Further, we argue that prediction models based on series of a single variable are sometimes superior to the model including all attributes: in contrast to other studies that resulted in size or complexity measures as predictors, we have identified the number of authors and the number of commit messages to versioning systems as excellent predictors of defect densities.\",\n \"Title: Can developer-module networks predict failures?\\nYear: 2008\\n\\nAbstract: Software teams should follow a well defined goal and keep their work focused. Work fragmentation is bad for efficiency and quality. In this paper we empirically investigate the relationship between the fragmentation of developer contributions and the number of post-release failures. Our approach is to represent developer contributions with a developer-module network that we call contribution network. We use network centrality measures to measure the degree of fragmentation of developer contributions. Fragmentation is determined by the centrality of software modules in the contribution network. Our claim is that central software modules are more likely to be failure-prone than modules located in surrounding areas of the network. We analyze this hypothesis by exploring the network centrality of Microsoft Windows Vista binaries using several network centrality measures as well as linear and logistic regression analysis. In particular, we investigate which centrality measures are significant to predict the probability and number of post-release failures. Results of our experiments show that central modules are more failure-prone than modules located in surrounding areas of the network. Results further confirm that number of authors and number of commits are significant predictors for the probability of post-release failures. For predicting the number of post-release failures the closeness centrality measure is most significant.\",\n \"Title: Smart views for analyzing problem reports: tool demo\\nYear: 2009\\n\\nAbstract: Issue tracking repositories contain a wealth of information for reasoning about various aspects of software development processes. In this paper, we focus on bug triaging and provide visual means to explore the effort estimation quality and the bug life-cycle of reported problems. Our approach uses a combination of graphical views to investigate details of individual problem reports while maintaining the context provided by the surrounding data population. This enables the detection and detailed analysis of hidden patterns and facilitates the analysis of problem report outliers.\",\n \"Title: Towards an Integrated View on Architecture and its Evolution\\nYear: 2005\\n\\nAbstract: Information about the evolution of a software architecture can be found in the source basis of a project and in the release history data such as modification and problem reports. Existing approaches deal with these two data sources separately and do not exploit the integration of their analyses. In this paper, we present an architecture analysis approach that provides an integration of both kinds of evolution data. The analysis applies fact extraction and generates specific directed attributed graphs; nodes represent source code entities and edges represent relationships such as accesses, includes, inherits, invokes, and coupling between certain architectural elements. The integration of data is then performed on a meta-model level to enable the generation of architectural views using binary relational algebra. These integrated architectural views show intended and unintended couplings between architectural elements, hence pointing software engineers to locations in the system that may be critical for on-going and future maintenance activities. We demonstrate our analysis approach using a large open source software system.\",\n \"Title: TA-RE: an exchange language for mining software repositories\\nYear: 2006\\n\\nAbstract: Software repositories have been getting a lot of attention from researchers in recent years. In order to analyze software repositories, it is necessary to first extract raw data from the version control and problem tracking systems. This poses two challenges: (1) extraction requires a non-trivial effort, and (2) the results depend on the heuristics used during extraction. These challenges burden researchers that are new to the community and make it difficult to benchmark software repository mining since it is almost impossible to reproduce experiments done by another team. In this paper we present the TA-RE corpus. TA-RE collects extracted data from software repositories in order to build a collection of projects that will simplify extraction process. Additionally the collection can be used for benchmarking. As the first step we propose an exchange language capable of making sharing and reusing data as simple as possible.\",\n \"Title: Data clone detection and visualization in spreadsheets\\nYear: 2013\\n\\nAbstract: Spreadsheets are widely used in industry: it is estimated that end-user programmers outnumber programmers by a factor 5. However, spreadsheets are error-prone, numerous companies have lost money because of spreadsheet errors. One of the causes for spreadsheet problems is the prevalence of copy-pasting. In this paper, we study this cloning in spreadsheets. Based on existing text-based clone detection algorithms, we have developed an algorithm to detect data clones in spreadsheets: formulas whose values are copied as plain text in a different location. To evaluate the usefulness of the proposed approach, we conducted two evaluations. A quantitative evaluation in which we analyzed the EUSES corpus and a qualitative evaluation consisting of two case studies. The results of the evaluation clearly indicate that 1) data clones are common, 2) data clones pose threats to spreadsheet quality and 3) our approach supports users in finding and resolving data clones.\",\n \"Title: Fine-Grained Analysis of Change Couplings\\nYear: 2005\\n\\nAbstract: In software evolution analysis, many approaches analyze release history data available through versioning systems. The recent investigations of CVS data have shown that commonly committed files highlight their change couplings. However, CVS stores modifications on the basis of text but does not track structural changes, such as the insertion, removing, or modification of methods or classes. A detailed analysis whether change couplings are caused by source code couplings or by other textual modifications, such as updates in license terms, is not performed by current approaches. The focus of this paper is on adding structural change information to existing release history data. We present an approach that uses the structure compare services shipped with the Eclipse IDE to obtain the corresponding finegrained changes between two subsequent versions of any Java class. This information supports filtering those change couplings which result from structural changes. So we can distill the causes for change couplings along releases and filter out those that are structurally relevant. The first validation of our approach with a medium-sized open source software system showed that a reasonable amount of change couplings are not caused by source code changes.\",\n \"Title: Combining micro-blogging and IDE interactions to support developers in their quests\\nYear: 2010\\n\\nAbstract: Software engineers spend a considerable amount of time on program comprehension. Although vendors of Integrated Development Environments (IDEs) and analysis tools address this challenge, current support for reusing and sharing program comprehension knowledge is limited. As a consequence, developers have to go through the time-consuming program understanding phase multiple times, instead of recalling knowledge from their past or other's program comprehension activities. In this paper, we present an approach to making the knowledge gained during the program comprehension process accessible, by combining micro-blog messages with interaction data automatically collected from the IDE. We implemented the approach in an Eclipse plugin called James and performed a first evaluation of the underlying approach effectiveness, assessing the nature and usefulness of the collected messages, as well as the added benefit of combining them with interaction data.\",\n \"Title: AWPS - Simulation Based Automated Web Performance Analysis and Prediction\\nYear: 2010\\n\\nAbstract: The growing demand for quality and performance has become a discriminating factor in the field of web applications. Advances in software technology and project management have increased the amount of web applications which are developed under these aspects. At the same time the research in web performance simulation and prediction is focusing on complex modeling methods and techniques rather than providing easy to use tools to test and ensure the performance requirements of today’s and tomorrow’s web applications. As a tool for these tasks, the automatic web performance simulation and prediction system (AWPS) will be described. The AWPS is capable of (a) automatically creating an online web performance simulation and (b) conducting trend analysis of the system under test.\",\n \"Title: Improving defect prediction using temporal features and non linear models\\nYear: 2007\\n\\nAbstract: Predicting the defects in the next release of a large software system is a very valuable asset for the project manger to plan her resources. In this paper we argue that temporal features (or aspects) of the data are central to prediction performance. We also argue that the use of non-linear models, as opposed to traditional regression, is necessary to uncover some of the hidden interrelationships between the features and the defects and maintain the accuracy of the prediction in some cases. Using data obtained from the CVS and Bugzilla repositories of the Eclipse project, we extract a number of temporal features, such as the number of revisions and number of reported issues within the last three months. We then use these data to predict both the location of defects (i.e., the classes in which defects will occur) as well as the number of reported bugs in the next month of the project. To that end we use standard tree-based induction algorithms in comparison with the traditional regression. Our non-linear models uncover the hidden relationships between features and defects, and present them in easy to understand form. Results also show that using the temporal features our prediction model can predict whether a source file will have a defect with an accuracy of 99% (area under ROC curve 0.9251) and the number of defects with a mean absolute error of 0.019 (Spearman's correlation of 0.96).\",\n \"Title: Change Analysis with Evolizer and ChangeDistiller\\nYear: 2009\\n\\nAbstract: Changes are the heartbeat of a software system. Software must change to reflect new business, pro cess, and technology requirements, or it will become progressively less useful. Software typically grows and becomes more complex, inducing more time and effort for performing changes. Software archives such as source code version-control systems and issue-tracking systems (for bugs and change requests) are rich sources to examine what changes have what impact on the software. A software evolution analysis platform called Evolizer analyzes change histories and potential support for evolution. Change types, a core part of the analysis, help discover significant changes and change patterns. A tool called ChangeDistiller enables fine-grained change type extraction and analysis to reason about coding conventions, control or exception flow, and even code and comment coevolution. An investigation of open source and commercial software systems contributed to a deeper understanding of how researchers can actively support software evolution in an integrated development environment.\",\n \"Title: Architecture Recovery for Product Families\\nYear: 2003\\n\\nAbstract: Software product families are rarely created right away but they emerge when a domain becomes mature enough to sustain their long-term investments. The typical pattern is to start with a small set of products to quickly enter a new market. As soon as the business proves to be successful new investments are directed to consolidating the software assets. The various products are migrated towards a flexible platform where the assets are shared and new products can be derived from. In order to create and maintain the platform, the organization needs to carry out several activities such as recovering the architectures of single products and product families, designing the reference architecture, isolating the variable parts, and generalizing software components. In this paper, we introduce a product family construction process that exploits related systems and product families, and we describe methods and tools used. We also present an approach for classifying platforms according to platform coverage and variation and describe three techniques to handle variability across single products and whole product families.\",\n \"Title: AWPS --- an architecture for pro-active web performance management\\nYear: 2010\\n\\nAbstract: The growing demand for quality and performance has become a discriminating factor in the field of software applications. Specifically in the area of web applications, performance has become a key factor for success creating the need for new types of performance evaluation models and methods capable of representing the dynamic characteristics of web environments. In this paper we will recall seminal work in this area and present AWPS, a tool for automatic web performance simulation and prediction. AWPS is capable of (a) automatically creating a web performance simulation and (b) conducting trend analysis of the system under test. The operation and usage of this tool is demonstrated in a case study of a two-tier architecture system.\",\n \"Title: Using the gini coefficient for bug prediction in eclipse\\nYear: 2011\\n\\nAbstract: The Gini coefficient is a prominent measure to quantify the inequality of a distribution. It is often used in the field of economy to describe how goods, e.g., wealth or farmland, are distributed among people. We use the Gini coefficient to measure code ownership by investigating how changes made to source code are distributed among the developer population. The results of our study with data from the Eclipse platform show that less bugs can be expected if a large share of all changes are accumulated, i.e., carried out, by relatively few developers.\",\n \"Title: Can we predict types of code changes?: an empirical analysis\\nYear: 2012\\n\\nAbstract: There exist many approaches that help in pointing developers to the change-prone parts of a software system. Although beneficial, they mostly fall short in providing details of these changes. Fine-grained source code changes (SCC) capture such detailed code changes and their semantics on the statement level. These SCC can be condition changes, interface modifications, inserts or deletions of methods and attributes, or other kinds of statement changes. In this paper, we explore prediction models for whether a source file will be affected by a certain type of SCC. These predictions are computed on the static source code dependency graph and use social network centrality measures and object-oriented metrics. For that, we use change data of the Eclipse platform and the Azureus 3 project. The results show that Neural Network models can predict categories of SCC types. Furthermore, our models can output a list of the potentially change-prone files ranked according to their change-proneness, overall and per change type category.\",\n \"Title: Detecting code smells in spreadsheet formulas\\nYear: 2012\\n\\nAbstract: Spreadsheets are used extensively in business processes around the world and just like software, spreadsheets are changed throughout their lifetime causing maintainability issues. This paper adapts known code smells to spreadsheet formulas. To that end we present a list of metrics by which we can detect smelly formulas and a visualization technique to highlight these formulas in spreadsheets. We implemented the metrics and visualization technique in a prototype tool to evaluate our approach in two ways. Firstly, we analyze the EUSES spreadsheet corpus, to study the occurrence of the formula smells. Secondly, we analyze ten real life spreadsheets, and interview the spreadsheet owners about the identified smells. The results of these evaluations indicate that formula smells are common and that they can reveal real errors and weaknesses in spreadsheet formulas.\",\n \"Title: Comparing fine-grained source code changes and code churn for bug prediction\\nYear: 2011\\n\\nAbstract: A significant amount of research effort has been dedicated to learning prediction models that allow project managers to efficiently allocate resources to those parts of a software system that most likely are bug-prone and therefore critical. Prominent measures for building bug prediction models are product measures, e.g., complexity or process measures, such as code churn. Code churn in terms of lines modified (LM) and past changes turned out to be significant indicators of bugs. However, these measures are rather imprecise and do not reflect all the detailed changes of particular source code entities during maintenance activities. In this paper, we explore the advantage of using fine-grained source code changes (SCC) for bug prediction. SCC captures the exact code changes and their semantics down to statement level. We present a series of experiments using different machine learning algorithms with a dataset from the Eclipse platform to empirically evaluate the performance of SCC and LM. The results show that SCC outperforms LM for learning bug prediction models.\",\n \"Title: Breviz: Visualizing Spreadsheets using Dataflow Diagrams\\nYear: 2011\\n\\nAbstract: Spreadsheets are used extensively in industry, often for business critical purposes. In previous work we have analyzed the information needs of spreadsheet professionals and addressed their need for support with the transition of a spreadsheet to a colleague with the generation of data flow diagrams. In this paper we describe the application of these data flow diagrams for the purpose of understanding a spreadsheet with three example cases. We furthermore suggest an additional application of the data flow diagrams: the assessment of the quality of the spreadsheet's design.\",\n \"Title: An exploratory study of the pull-based software development model\\nYear: 2014\\n\\nAbstract: The advent of distributed version control systems has led to the development of a new paradigm for distributed software development; instead of pushing changes to a central repository, developers pull them from other repositories and merge them locally. Various code hosting sites, notably Github, have tapped on the opportunity to facilitate pull-based development by offering workflow support tools, such as code reviewing systems and integrated issue trackers. In this work, we explore how pull-based software development works, first on the GHTorrent corpus and then on a carefully selected sample of 291 projects. We find that the pull request model offers fast turnaround, increased opportunities for community engagement and decreased time to incorporate contributions. We show that a relatively small number of factors affect both the decision to merge a pull request and the time to process it. We also examine the reasons for pull request rejection and find that technical ones are only a small minority.\",\n \"Title: EQ-mine: predicting short-term defects for software evolution\\nYear: 2007\\n\\nAbstract: We use 63 features extracted from sources such as versioning and issue tracking systems to predict defects in short time frames of two months. Our multivariate approach covers aspects of software projects such as size, team structure, process orientation, complexity of existing solution, difficulty of problem, coupling aspects, time constrains, and testing data. We investigate the predictability of several severities of defects in software projects. Are defects with high severity difficult to predict? Are prediction models for defects that are discovered by internal staff similar to models for defects reported from the field? We present both an exact numerical prediction of future defect numbers based on regression models as well as a classification of software components as defect-prone based on the C4.5 decision tree. We create models to accurately predict short-term defects in a study of 5 applications composed of more than 8.000 classes and 700.000 lines of code. The model quality is assessed based on 10-fold cross validation.\",\n \"Title: Asset Recovery and Their Incorporation into Product Lines\\nYear: 2005\\n\\nAbstract: Software product lines aim in having a common platform from which several similar products can be derived. The elements of the platform are called assets and they are managed in an asset base being part of the product line infrastructure. Assets can include own developments, open source or third-party software modules, as well as design and project documents. In the context of the European-wide project FAMILIES we concentrated on techniques used to populate the asset base by recovering assets from existing systems. We present an approach on how to incorporate existing assets into the product line infrastructure. Thereby we explicitly distinguish the asset origins and the different information sources available. The incorporation is a quality-driven process that is backed up by a set of reverse engineering techniques to evaluate the asset's internal quality. The quality assessment of an asset is the critical measurement for industrial development organizations in order to incorporate assets into their product line infrastructure.\",\n \"Title: Visual patterns in issue tracking data\\nYear: 2010\\n\\nAbstract: Software development teams gather valuable data about features and bugs in issue tracking systems. This information can be used to measure and improve the efficiency and effectiveness of the development process. In this paper we present an approach that harnesses the extraordinary capability of the human brain to detect visual patterns. We specify generic visual process patterns that can be found in issue tracking data. With these patterns we can analyze information about effort estimation, and the length, and sequence of problem resolution activities. In an industrial case study we apply our interactive tool to identify instances of these patterns and discuss our observations. Our approach was validated through extensive discussions with multiple project managers and developers, as well as feedback from the project review board.\",\n \"Title: Analyzing the Impact of Antipatterns on Change-Proneness Using Fine-Grained Source Code Changes\\nYear: 2012\\n\\nAbstract: Antipatterns are poor solutions to design and implementation problems which are claimed to make object oriented systems hard to maintain. Our recent studies showed that classes with antipatterns change more frequently than classes without antipatterns. In this paper, we detail these analyses by taking into account fine-grained source code changes (SCC) extracted from 16 Java open source systems. In particular we investigate: whether classes with antipatterns are more change-prone (in terms of SCC) than classes without, (2) whether the type of antipattern impacts the change-proneness of Java classes, and (3) whether certain types of changes are performed more frequently in classes affected by a certain antipattern. Our results show that: 1) the number of SCC performed in classes affected by antipatterns is statistically greater than the number of SCC performed in classes with no antipattern, 2) classes participating in the three antipatterns Complex Class, Spaghetti Code, and SwissArmyKnife are more change-prone than classes affected by other antipatterns, and 3) certain types of changes are more likely to be performed in classes affected by certain antipatterns, such as API changes are likely to be performed in classes affected by the Complex Class, Spaghetti Code, and SwissArmyKnife antipatterns.\",\n \"Title: Using vector clocks to monitor dependencies among services at runtime\\nYear: 2011\\n\\nAbstract: Service-Oriented Architecture (SOA) enable organizations to react to requirement changes in an agile manner and to foster the reuse of existing services. However, the dynamic nature of Service-Oriented Systems and their agility bear the challenge of properly understanding such systems. In particular, understanding the dependencies among services is a non trivial task, especially if service-oriented systems are distributed over several hosts and/or using different SOA technologies. In this paper, we propose an approach to monitor dynamic dependencies among services. The approach is based on the vector clocks, originally conceived and used to order events in a distributed environment. We use the vector clocks to order service executions and to infer causal dependencies among services. In our future work we plan to use this information to study change and failure impact analysis in service-oriented systems.\",\n \"Title: Analyzing the Evolution of Web Services Using Fine-Grained Changes\\nYear: 2012\\n\\nAbstract: In the service-oriented paradigm web service interfaces are considered contracts between web service subscribers and providers. However, these interfaces are continuously evolving over time to satisfy changes in the requirements and to fix bugs. Changes in a web service interface typically affect the systems of its subscribers. Therefore, it is essential for subscribers to recognize which types of changes occur in a web service interface in order to analyze the impact on his/her systems. In this paper we propose a tool called WSDLDiff to extract fine-grained changes from subsequent versions of a web service interface defined in WSDL. In contrast to existing approaches, WSDLDiff takes into account the syntax of WSDL and extracts the WSDL elements affected by changes and the types of changes. With WSDLDiff we performed a study aimed at analyzing the evolution of web services using the fine-grained changes extracted from the subsequent versions of four real world WSDL interfaces. The results of our study show that the analysis of the fine-grained changes helps web service subscribers to highlight the most frequent types of changes affecting a WSDL interface. This information can be relevant for web service subscribers who want to assess the risk associated to the usage of web services and to subscribe to the most stable ones.\",\n \"Title: Populating a Release History Database from Version Control and Bug Tracking Systems\\nYear: 2003\\n\\nAbstract: Version control and bug tracking systems contain largeamounts of historical information that can give deep insightinto the evolution of a software project. Unfortunately,these systems provide only insufficient support for a detailedanalysis of software evolution aspects. We addressthis problem and introduce an approach for populating a releasehistory database that combines version data with bugtracking data and adds missing data not covered by versioncontrol systems such as merge points. Then simple queriescan be applied to the structured data to obtain meaningfulviews showing the evolution of a software project. Suchviews enable more accurate reasoning of evolutionary aspectsand facilitate the anticipation of software evolution.We demonstrate our approach on the large Open Sourceproject Mozilla that offers great opportunities to compareresults and validate our approach.\",\n \"Title: Automated web performance analysis, with a special focus on prediction\\nYear: 2008\\n\\nAbstract: Today the performance of web systems is getting ever more important, as the number of users and competitors is still increasing. Therefore performance analysis tools gain importance too. There are currently several tools on the market that ensure and test for adequate performance. There are a number of simulation tools and monitoring tools, but only few that automatise and combine both approaches. This paper outlines a system that is capable of (a) automatically creating a web performance simulation and (b) conducting trend analysis of the system under test (SUT). The system requires input information like monitoring points and static-information about the SUT. Based on this information a simulation model of the system is generated. Then the simulation model is refined stepwise e.g. by adding or removing connections between the model components or adjusting the parameters until the aimed accuracy is achieved. By using this simulation model the prediction module creates an analysis of the SUT, and thereby provides as much information as possible about the current state of the system and potential trends. This predictive information can be used for pro-active server tuning or other performance optimisations. The special focus of this work is on the adjustment and prediction parts of the system described here. For all the other parts existing tools and techniques will be used wherever possible.\",\n \"Title: Towards the Integration of Versioning Systems, Bug Reports and Source Code Meta-Models\\nYear: 2005\\n\\nAbstract: Versioning system repositories and bug tracking systems are valuable sources of information to study the evolution of large open source software systems. However, being conceived for specific purposes, i.e., to support the development or trigger maintenance activities, they do neither allow an easy information browsing nor support the study of software evolution. For example, queries such as locating and browsing the faultiest methods are not provided.This paper addresses such issues and proposes an approach and a framework to consistently merge information extracted from source code, versioning repositories and bug reports. Our information representation exploits the property concepts of the FAMIX information exchange meta-model, allowing to represent, browse, and query, at different level of abstractions, the concept of interest. This allows the user to navigate back and forth from versioning system modification reports to bug reports and to source code. This paper presents the analysis framework and approaches to populate it, tools developed and under development for it, as well as lessons learned while analyzing several releases of Mozilla.\",\n \"Title: Improving Fact Extraction of Framework-Based Software Systems\\nYear: 2003\\n\\nAbstract: Modern software frameworks provide a set of commonand prefabricated software artifacts that supportengineers in developing large-scale software systems.Framework-related information can be implemented insource code, comments or configuration files, but in thelatter two cases, current reverse engineering approachesmiss important facts reducing the quality of subsequentanalysis tasks. We introduce a generic fact extractionapproach for framework-based systems by combiningtraditional parsing with lexical pattern matching toobtain framework-specific facts from all three sources.We evaluate our approach with an industrial softwareapplication that was built using the Avalon/Phoenixframework. In particular we give examples to point outthe benefits of considering framework-related informationand reflect experiences made during the case study.\",\n \"Title: Analyzing and Relating Bug Report Data for Feature Tracking\\nYear: 2003\\n\\nAbstract: Gaining higher level evolutionary information aboutlarge software systems is a key in validating past and adjustingfuture development processes. In this paper, we analyzethe proximity of software features based on modificationand problem report data that capture the system's evolutionhistory. Features are instrumented and tracked, therelationships of modification and problem reports to thesefeatures are established, and the tracked features are visualizedto illustrate their otherwise hidden dependencies.Our approach uncovers these hidden relationships betweenfeatures via problem report analysis and presents them ineasy-to-evaluate visual form. Particular feature dependenciesthen can be selected to assess the feature evolution byzooming in into an arbitrary level of detail. Such visualizationof interwoven features, therefore, can indicate locationsof design erosion in the architectural evolution of asoftware system. Our approach has been validated usingthe large open source software project of Mozilla and itsbug reporting system Bugzilla.\",\n \"Title: Detecting similar Java classes using tree algorithms\\nYear: 2006\\n\\nAbstract: Similarity analysis of source code is helpful during development to provide, for instance, better support for code reuse. Consider a development environment that analyzes code while typing and that suggests similar code examples or existing implementations from a source code repository. Mining software repositories by means of similarity measures enables and enforces reusing existing code and reduces the developing effort needed by creating a shared knowledge base of code fragments. In information retrieval similarity measures are often used to find documents similar to a given query document. This paper extends this idea to source code repositories. It introduces our approach to detect similar Java classes in software projects using tree similarity algorithms. We show how our approach allows to find similar Java classes based on an evaluation of three tree-based similarity measures in the context of five user-defined test cases as well as a preliminary software evolution analysis of a medium-sized Java project. Initial results of our technique indicate that it (1) is indeed useful to identify similar Java classes, (2)successfully identifies the ex ante and ex post versions of refactored classes, and (3) provides some interesting insights into within-version and between-version dependencies of classes within a Java project.\",\n \"Title: CodeCrawler: an information visualization tool for program comprehension\\nYear: 2005\\n\\nAbstract: CodeCrawler (in the remainder of the text CC) is a language independent, interactive, information visualization tool. It is mainly targeted at visualizing object-oriented software, and has been successfully validated in several industrial case studies over the past few years. CC adheres to lightweight principles: it implements and visualizes polymetric views, visualizations of software enriched with information such as software metrics and other source code semantics. CC is built on top of Moose, an extensible language independent reengineering environment that implements the FAMIX metamodel. In its last implementation, CC has become a general-purpose information visualization tool.\",\n \"Title: Domain-Specific Languages in Practice: A User Study on the Success Factors\\nYear: 2009\\n\\nAbstract: In this paper we present an empirical study on the use of a domain-specific language(DSL) in industry. This DSL encapsulates the details of services that communicate using Windows Communication Foundation (WCF). From definitions of the data contracts between clients and servers, WCF/C# code for service plumbing is generated. We conducted a survey amongst developers that use this DSL while developing applications for customers. The DSL has been used in about 30 projects all around the world. We describe the known success factors of the use of DSLs, such as improved maintainability and ease of re-use, and assert how well this DSL scores on all of them. The analysis of the results of this case study also shows which conditions should be fulfilled in order to increase the chances of success in using a DSL in a real life case.\",\n \"Title: Detecting and refactoring code smells in spreadsheet formulas\\nYear: 2015\\n\\nAbstract: Spreadsheets are used extensively in business processes around the world and just like software, spreadsheets are changed throughout their lifetime causing understandability and maintainability issues. This paper adapts known code smells to spreadsheet formulas. To that end we present a list of metrics by which we can detect smelly formulas; a visualization technique to highlight these formulas in spreadsheets and a method to automatically suggest refactorings to resolve smells. We implemented the metrics, visualization and refactoring suggestions techniques in a prototype tool and evaluated our approach in three studies. Firstly, we analyze the EUSES spreadsheet corpus, to study the occurrence of the formula smells. Secondly, we analyze ten real life spreadsheets, and interview the spreadsheet owners about the identified smells. Finally, we generate refactoring suggestions for those ten spreadsheets and study the implications. The results of these evaluations indicate that formula smells are common, that they can reveal real errors and weaknesses in spreadsheet formulas and that in simple cases they can be refactored.\",\n \"Title: Evaluating Feature Change Impact on Multi-product Line Configurations Using Partial Information.\\nYear: 2015\\n\\nAbstract: Evolving large-scale, complex and highly variable systems is known to be a difficult task, where a single change can ripple through various parts of the system with potentially undesirable effects. In the case of product lines, and moreover multi-product lines, a change may affect only certain variants or certain combinations of features, making the evaluation of change effects more difficult.\",\n \"Title: A Genetic Algorithm to Find the Adequate Granularity for Service Interfaces\\nYear: 2014\\n\\nAbstract: The relevance of the service interfaces' granularity and its architectural impact have been widely investigated in literature. Existing studies show that the granularity of a service interface, in terms of exposed operations, should reflect their clients' usage. This idea has been formalized in the Consumer-Driven Contracts pattern (CDC). However, to the best of our knowledge, no studies propose techniques to assist providers in finding the right granularity and in easing the adoption of the CDC pattern. In this paper, we propose a genetic algorithm that mines the clients' usage of service operations and suggests Façade services whose granularity reflect the usage of each different type of clients. These services can be deployed on top of the original service and they become contracts for the different types of clients satisfying the CDC pattern. A first study shows that the genetic algorithm is capable of finding Façade services and outperforms a random search approach.\",\n \"Title: A Manual Categorization of Android App Development Issues on Stack Overflow\\nYear: 2014\\n\\nAbstract: While many tutorials, code examples, and documentation about Android APIs exist, developers still face various problems with the implementation of Android Apps. Many of these issues are discussed on Q&A-sites, such as Stack Overflow. In this paper we present a manual categorization of 450 Android related posts of Stack Overflow concerning their question and problem types. The idea is to find dependencies between certain problems and question types to get better insights into issues of Android App development. The categorization is developed using card sorting with three experienced Android App developers. An initial approach to automate the classification of Stack Overflow posts using Lucene is also presented. The study highlights that the most common question types are 'How to
?' and 'What is the problem
?'. The problems that are discussed most often are related to 'User Interface' and 'Core Elements'. In particular, the problem category 'Layout' is often related to 'What is the problem
?' and 'Frameworks' issues often come with 'Is it possible
?' questions.\",\n \"Title: A Tool for Visual Understanding of Source Code Dependencies\\nYear: 2008\\n\\nAbstract: Many program comprehension tools use graphs to visualize and analyze source code. The main issue is that existing approaches create graphs overloaded with too much information. Graphs contain hundreds of nodes and even more edges that cross each other. Understanding these graphs and using them for a given program comprehension task is tedious, and in the worst case developers stop using the tools. In this paper we present DA4Java, a graph-based approach for visualizing and analyzing static dependencies between Java source code entities. The main contribution of DA4Java is a set of features to incrementally compose graphs and remove irrelevant nodes and edges from graphs. This leads to graphs that contain significantly fewer nodes and edges and need less effort to understand.\",\n \"Title: Automated Web Performance Analysis\\nYear: 2008\\n\\nAbstract: Performance is a key feature in many systems nowadays. There are several tools on the market that ensure and test for adequate performance. They, can be divided into simulation tools and monitoring tools. But only a few automatise and combine both approaches. This paper describes a system capable of automatically creating a web performance simulation and conducting trend analysis of the system under test (SUT). To achieve this the system requires input information, like Monitoring Points and Static-Information about the SUT. The system monitors and analyses the SUT and based on this information generates a simulation model of the system. The simulation model is refinded stepwise e.g. by adding or removing connections between the model components or adjusting the parameters until the aimed accuracy is achieved. With the help of the simulation model a prediction module creates an analysis of the SUT, and thereby can give as much information about the current state of the system and potential trends as possible. This predictive information can be used for pro-active server tuning or other performance optimisations. The focus of my PhD thesis is on the adjustment and prediction part of the system described here. For all other parts, already existing tools and techniques will be used where possible. This initial paper outlines the complete system.\",\n \"Title: Detecting and visualizing inter-worksheet smells in spreadsheets\\nYear: 2012\\n\\nAbstract: Spreadsheets are often used in business, for simple tasks, as well as for mission critical tasks such as finance or forecasting. Similar to software, some spreadsheets are of better quality than others, for instance with respect to usability, maintainability or reliability. In contrast with software however, spreadsheets are rarely checked, tested or certified. In this paper, we aim at developing an approach for detecting smells that indicate weak points in a spreadsheet's design. To that end we first study code smells and transform these code smells to their spreadsheet counterparts. We then present an approach to detect the smells, and to communicate located smells to spreadsheet users with data flow diagrams. To evaluate our apporach, we analyzed occurrences of these smells in the Euses corpus. Furthermore we conducted ten case studies in an industrial setting. The results of the evaluation indicate that smells can indeed reveal weaknesses in a spreadsheet's design, and that data flow diagrams are an appropriate way to show those weaknesses.\",\n \"Title: Interactive views for analyzing problem reports\\nYear: 2009\\n\\nAbstract: Issue tracking repositories contain a wealth of informa- tion for reasoning about various aspects of software develop- ment processes. In this paper, we focus on bug triaging and provide visual means to explore the effort estimation quality and the bug life-cycle of reported problems. Our approach follows the Micro/Macro reading technique and uses a combination of graphical views to investigate de- tails of individual problem reports while maintaining the context provided by the surrounding data population. This enables the detection and detailed analysis of hidden pat- terns and facilitates the analysis of problem report outliers. In an industrial study, we use our approach in various problem report analysis scenarios and answer questions related to effort estimation and resource planning.\",\n \"Title: Towards a Weighted Voting System for Q&A Sites\\nYear: 2013\\n\\nAbstract: Q&A sites have become popular to share and look for valuable knowledge. Users can easily and quickly access high quality answers to common questions. The main mechanism to label good answers is to count the votes per answer. This mechanism, however, does not consider whether other answers were present at the time when a vote is given. Consequently, good answers that were given later are likely to receive less votes than they would have received if given earlier. In this paper we present a Weighted Votes (WV) metric that gives different weights to the votes depending on how many answers were present when the vote is performed. The idea behind WV is to emphasize the answer that receives most of the votes when most of the answers were already posted. Mining the Stack Overflow data dump we show that the WV metric is able to highlight between 4.07% and 10.82% answers that differ from the most voted ones.\",\n \"Title: Refactoring Fat Interfaces Using a Genetic Algorithm\\nYear: 2014\\n\\nAbstract: Recent studies have shown that the violation of the Interface Segregation Principle (ISP) is critical for maintaining and evolving software systems. Fat interfaces (i.e., interfaces violating the ISP) change more frequently and degrade the quality of the components coupled to them. According to the ISP the interfaces' design should force no client to depend on methods it does not invoke. Fat interfaces should be split into smaller interfaces exposing only the methods invoked by groups of clients. However, applying the ISP is a challenging task when fat interfaces are invoked differently by many clients. In this paper, we formulate the problem of applying the ISP as a multi-objective clustering problem and we propose a genetic algorithm to solve it. We evaluate the capability of the proposed genetic algorithm with 42,318 public Java APIs whose clients' usage has been mined from the Maven repository. The results of this study show that the genetic algorithm outperforms other search based approaches (i.e., random and simulated annealing approaches) in splitting the APIs according to the ISP.\",\n \"Title: Method-level bug prediction\\nYear: 2012\\n\\nAbstract: Researchers proposed a wide range of approaches to build effective bug prediction models that take into account multiple aspects of the software development process. Such models achieved good prediction performance, guiding developers towards those parts of their system where a large share of bugs can be expected. However, most of those approaches predict bugs on file-level. This often leaves developers with a considerable amount of effort to examine all methods of a file until a bug is located. This particular problem is reinforced by the fact that large files are typically predicted as the most bug-prone. In this paper, we present bug prediction models at the level of individual methods rather than at file-level. This increases the granularity of the prediction and thus reduces manual inspection efforts for developers. The models are based on change metrics and source code metrics that are typically used in bug prediction. Our experiments---performed on 21 Java open-source (sub-)systems---show that our prediction models reach a precision and recall of 84% and 88%, respectively. Furthermore, the results indicate that change metrics significantly outperform source code metrics.\",\n \"Title: Synonym suggestion for tags on stack overflow\\nYear: 2015\\n\\nAbstract: The amount of diverse tags used to classify posts on Stack Overflow increased in the last years to more than 38,000 tags. Many of these tags have the same or similar meaning. Stack Overflow provides an approach to reduce the amount of tags by allowing privileged users to manually create synonyms. However, currently exist only 2,765 synonym-pairs on Stack Overflow that is quite low compared to the total number of tags.\\n\\nTo comprehend how synonym-pairs are built, we manually analyzed the tags and how the synonyms could be created automatically. Based on our findings, we then present TSST, a tag synonym suggestion tool, that outputs a ranked list of possible synonyms for each input tag.\\n\\nWe first evaluated TSST with the 2,765 approved synonym-pairs of Stack Overflow. For 88.4% of the tags TSST finds the correct synonyms, for 72.2% the correct synonym is within the top 10 suggestions. In addition, we applied TSST to 10 randomly selected Android related tags and evaluated the suggested synonyms with 20 Android app developers in an online survey. Overall, in 80% of their ratings, developers found an adequate synonym suggested by TSST.\",\n \"Title: Predicting Build Co-changes with Source Code Change and Commit Categories\\nYear: 2016\\n\\nAbstract: When software is maintained and evolved the build configuration also needs to be updated. Knowing when to update the build configuration is typically done manually with the risk of missing an update and breaking the build. To mitigate this risk, previous work has investigated prediction models to help developers to identify commits that will likely involve an update of the build configuration. In this paper, we investigate whether we can improve these existing prediction models by taking into account detailed information on source code changes and commit categories. Our main hypothesis is that such detailed information on changes will significantly improve the prediction of build co-changes. To that extent, we extract information on changes from 10 Java open source projects and use a random forest classifier to train models that predict build co-changes within and across projects. Our results show significant improvements over existing prediction models: the AUC for intra-and cross-project prediction improves by 11.54% and 9.46% respectively. In addition, we investigate advanced resampling techniques to explore the effect of unbalanced data on our models. The results show that SMOTE can particularly improve prediction models with low performance that were trained on unbalanced data. Our models improve the prediction and enable a better understanding of build co-changes.\",\n \"Title: FEVER: extracting feature-oriented changes from commits.\\nYear: 2016\\n\\nAbstract: The study of the evolution of highly configurable systems requires a thorough understanding of thee core ingredients of such systems: (1) the underlying variability model; (2) the assets that together implement the configurable features; and (3) the mapping from variable features to actual assets. Unfortunately, to date no systematic way to obtain such information at a sufficiently fine grained level exists.\\n\\nTo remedy this problem we propose FEVER and its instantiation for the Linux kernel. FEVER extracts detailed information on changes in variability models (KConfig files), assets (preprocessor based C code), and mappings (Makefiles). We describe how FEVER works, and apply it to several releases of the Linux kernel. Our evaluation on 300 randomly selected commits, from two different releases, shows our results are accurate in 82.6% of the commits. Furthermore, we illustrate how the populated FEVER graph database thus obtained can be used in typical Linux engineering tasks.\",\n \"Title: Grouping android tag synonyms on stack overflow.\\nYear: 2016\\n\\nAbstract: On Stack Overflow, more than 38,000 diverse tags are used to classify posts. The Stack Overflow community provides tag synonyms to reduce the number of tags that have the same or similar meaning. In our previous research, we used those synonym pairs to derive a number of strategies to create tag synonyms automatically.\\n\\nIn this work, we continue this line of research and present an approach to group tag synonyms to meaningful topics. We represent our synonyms as directed, weighted graphs, and investigate several graph community detection algorithms to build meaningful groups of tags, also called tag communities.\\n\\nWe apply our approach to the tags obtained from Android-related Stack Overflow posts and evaluate the resulting tag communities quantitatively with various community metrics. In addition, we evaluate our approach qualitatively through a manual inspection and comparison of a random sample of tag communities. Our results show that we can cluster the Android tags to 2,481 meaningful tag communities. We also show how these tag communities can be used to derive trends of topics of Android-related questions on Stack Overflow.\",\n \"Title: Extracting Build Changes with BUILDDIFF.\\nYear: 2017\\n\\nAbstract: Build systems are an essential part of modern software engineering projects. As software projects change continuously, it is crucial to understand how the build system changes because neglecting its maintenance can lead to expensive build breakage. Recent studies have investigated the (co-)evolution of build configurations and reasons for build breakage, but they did this only on a coarse grained level. In this paper, we present BuildDiff, an approach to extract detailed build changes from Maven build files and classify them into 95 change types. In a manual evaluation of 400 build changing commits, we show that BuildDiff can extract and classify build changes with an average precision and recall of 0.96 and 0.98, respectively. We then present two studies using the build changes extracted from 30 open source Java projects to study the frequency and time of build changes. The results show that the top 10 most frequent change types account for 73% of the build changes. Among them, changes to version numbers and changes to dependencies of the projects occur most frequently. Furthermore, our results show that build changes occur frequently around releases. With these results, we provide the basis for further research, such as for analyzing the (co-)evolution of build files with other artifacts or improving effort estimation approaches. Furthermore, our detailed change information enables improvements of refactoring approaches for build configurations and improvements of models to identify error-prone build files.\",\n \"Title: Extracting Timed Automata from Java Methods\\nYear: 2017\\n\\nAbstract: The verification of the time behavior in distributed, multi-threaded programs is challenging, mainly because modern programming languages only provide means to represent time without a proper semantics. Current approaches to extract time models from source code represent time only as a sequence of events or require developers to manually provide a formal model of the time behavior. This makes it difficult for developers to verify various aspects of their systems, such as timeouts, delays and periodicity of the execution. In this paper, we introduce a definition of the time semantics of the Java programming language. Based on the semantics, we present an approach to automatically extract timed automata and their time constraints from the Java methods source code. First, we detect Java statements which involve time, from which we then extract the timed automata that are directly amenable to the verification of time properties of the methods. We evaluated the accuracy of our approach on ten open source Java projects that heavily use time in their source code. The results show a precision of 98.62% and recall of 95.37% in extracting time constraints from Java code. Finally, we demonstrate the effectiveness of our approach with five reported bugs of four different Apache systems that we could confirm.\",\n \"Title: Automatically repairing dependency-related build breakage\\nYear: 2018\\n\\nAbstract: Build systems are widely used in today's software projects to automate integration and build processes. Similar to source code, build specifications need to be maintained to avoid outdated specifications, and build breakage as a consequence. Recent work indicates that neglected build maintenance is one of the most frequently occurring reasons why open source and proprietary builds break. In this paper, we propose BuildMedic, an approach to automatically repair Maven builds that break due to dependency-related issues. Based on a manual investigation of 37 broken Maven builds in 23 open source Java projects, we derive three repair strategies to automatically repair the build, namely Version Update, Delete Dependency, and Add Repository. We evaluate the three strategies on 84 additional broken builds from the 23 studied projects in order to demonstrate the applicability of our approach. The evaluation shows that BuildMedic can automatically repair 45 of these broken builds (54%). Furthermore, in 36% of the successfully repaired build breakages, BuildMedic outputs at least one repair candidate that is considered a correct repair. Moreover, 76% of them could be repaired with only a single dependency correction.\",\n \"Title: FEVER: An approach to analyze feature-oriented changes and artefact co-evolution in highly configurable systems.\\nYear: 2018\\n\\nAbstract: The evolution of highly configurable systems is known to be a challenging task. Thorough understanding of configuration options their relationships, and their implementation in various types of artefacts (variability model, mapping, and implementation) is required to avoid compilation errors, invalid products, or dead code. Recent studies focusing on co-evolution of artefacts detailed feature-oriented change scenarios, describing how related artefacts might change over time. However, relying on manual analysis of commits, such work do not provide the means to obtain quantitative information on the frequency of described scenarios nor information on the exhaustiveness of the presented scenarios for the evolution of a large scale system. In this work, we propose FEVER and its instantiation for the Linux kernel. FEVER extracts detailed information on changes in variability models (KConfig files), assets (preprocessor based C code), and mappings (Makefiles). We apply this methodology to the Linux kernel and build a dataset comprised of 15 releases of the kernel history. We performed an evaluation of the FEVER approach by manually inspecting the data and compared it with commits in the system’s history. The evaluation shows that FEVER accurately captures feature related changes for more than 85% of the 810 manually inspected commits. We use the collected data to reflect on occurrences of co-evolution in practice. Our analysis shows that complex co-evolution scenarios occur in every studied release but are not among the most frequent change scenarios, as they only occur for 8 to 13% of the evolving features. Moreover, only a minority of developers working on a given release will make changes to all artefacts related to a feature (between 10% and 13% of authors). While our conclusions are derived from observations on the evolution of the Linux kernel, we believe that they may have implications for tool developers as well as guide further research in the field of co-evolution of artefacts.\",\n \"Title: Modeling time in Java programs for automatic error detection.\\nYear: 2018\\n\\nAbstract: Modern programming languages, such as Java, represent time as integer variables, called timestamps. Timestamps allow developers to tacitly model incorrect time values resulting in a program failure because any negative value or every positive value is not necessarily a valid time representation. Current approaches to automatically detect errors in programs, such as Randoop and FindBugs, cannot detect such errors because they treat timestamps as normal integer variables and test them with random values verifying if the program throws an exception. In this paper, we present an approach that considers the time semantics of the Java language to systematically detect time related errors in Java programs. With the formal time semantics, our approach determines which integer variables handle time and which statements use or alter their values. Based on this information, it translates these statements into an SMT model that is passed to an SMT solver. The solver formally verifies the correctness of the model and reports the violations of time properties in that program. For the evaluation, we have implemented our approach in a prototype tool and applied it to the source code of 20 Java open source projects. The results show that our approach is scalable and it is capable of detecting time errors precisely enough allowing its usability in real-world applications.\",\n \"Title: Noise and heterogeneity in historical build data: an empirical study of Travis CI.\\nYear: 2018\\n\\nAbstract: Automated builds, which may pass or fail, provide feedback to a development team about changes to the codebase. A passing build indicates that the change compiles cleanly and tests (continue to) pass. A failing (a.k.a., broken) build indicates that there are issues that require attention. Without a closer analysis of the nature of build outcome data, practitioners and researchers are likely to make two critical assumptions: (1) build results are not noisy; however, passing builds may contain failing or skipped jobs that are actively or passively ignored; and (2) builds are equal; however, builds vary in terms of the number of jobs and configurations. \\n\\n To investigate the degree to which these assumptions about build breakage hold, we perform an empirical study of 3.7 million build jobs spanning 1,276 open source projects. We find that: (1) 12% of passing builds have an actively ignored failure; (2) 9% of builds have a misleading or incorrect outcome on average; and (3) at least 44% of the broken builds contain passing jobs, i.e., the breakage is local to a subset of build variants. Like other software archives, build data is noisy and complex. Analysis of build data requires nuance.\",\n \"Title: DiffViz: A Diff Algorithm Independent Visualization Tool for Edit Scripts\\nYear: 2018\\n\\nAbstract: A number of approaches and tools exist that extract and visualize the changes between two versions of a file and thereby help developers to understand them. DiffViz is an interactive visualization tool that visualizes the changes independent from the differencing algorithm. It supports, but is not limited to, a granularity on the level of abstract syntax trees. Furthermore, it provides several new features, such as node matching and the mini-map, to navigate and analyze the changes. A demo of the installation and example usage of the tool is available here: https://youtu.be/RF93ey9GYoc.\",\n \"Title: Towards Model Checking Security of Real Time Java Software.\\nYear: 2018\\n\\nAbstract: More and more software libraries and applications in high-performance computing and distributed systems are coded using the Java programming language. The correctness of such pieces of code w.r.t. a given set of security policies often depends on the correct handling of timing between concurrent or recurrent events. Model-checking has proven to be an effective tool for verifying the correctness of software. In spite of the growing importance of this application area of formal methods, though, no approach exists that targets the problem of verifying the correctness of real-time software w.r.t. timed specifications. The few existing works focus on very different problems, such as schedulability analysis of Java tasks. In this paper we present an approach combining rule-based static analysis together with symbolic execution of Java code to extract networks of timed automata from existing software and then use Uppaal to model-check them against timed specifications. We show through a real-world case study that this approach can be helpful in model-checking security policies of real-time Java software.\",\n \"Title: Automatically classifying posts into question categories on stack overflow.\\nYear: 2018\\n\\nAbstract: Software developers frequently solve development issues with the help of question and answer web forums, such as Stack Overflow (SO). While tags exist to support question searching and browsing, they are more related to technological aspects than to the question purposes. Tagging questions with their purpose can add a new dimension to the investigation of topics discussed in posts on SO. In this paper, we aim to automate such a classification of SO posts into seven question categories. As a first step, we have manually created a curated data set of 500 SO posts, classified into the seven categories. Using this data set, we apply machine learning algorithms (Random Forest and Support Vector Machines) to build a classification model for SO questions. We then experiment with 82 different configurations regarding the preprocessing of the text and representation of the input data. The results of the best performing models show that our models can classify posts into the correct question category with an average precision and recall of 0.88 and 0.87 when using Random Forest and the phrases indicating a question category as input data for the training. The obtained model can be used to aid developers in browsing SO discussions or researchers in building recommenders based on SO.\",\n \"Title: Can I Depend on you? Mapping the Dependency and Quality Landscape of ROS Packages\\nYear: 2019\\n\\nAbstract: Since its beginnings ten years ago, the Robot Operating System (ROS) has created a huge community of developers and researchers and is now the most widespread open-source framework for robotics development. It is used in research, prototyping but also in commercial products and supports a wide range of robotic platforms, sensors and high-level data processing functions. While for a research platform, quality of the software developed with it is typically of lower importance, ROS is gradually moving towards industrial applications making software quality a premier topic. In this paper, we want to gain insights on how ROS is used in practice, how high the quality of the ROS packages and applications is, and where potential pitfalls in the use of ROS lie. To achieve this, we have analyzed several thousands of open-source ROS packages found on GitHub and Bitbucket for their quality and their interdependencies. Our results include packages on the rosdistro index and, more importantly, packages that are not. From our results we show common issues in ROS-applications, quality implications, and also which packages of what quality are particularly popular.\",\n \"Title: Semantics-driven extraction of timed automata from Java programs\\nYear: 2019\\n\\nAbstract: The automatic verification of time properties of models extracted from programs is challenging, mainly because modern programming languages, such as Java, represent time without a proper semantics. Current approaches to extract time models from source code either represent time only as a tree-like sequence of events or require developers to manually provide a formal model of the time behavior. This makes it difficult for software developers to verify various aspects of their systems, such as timeouts, delays and periodicity of the execution. In this paper, we introduce a formal definition of the time semantics for the Java programming language. Based on the semantics, we present an approach to automatically extract timed automata and their time constraints from Java programs at method level. First, our approach detects the Java statements that involve time, from which it then extracts the timed automata. Our extracted automata are directly amenable to the verification of time properties of the corresponding Java methods. We evaluated the accuracy of our approach on twenty open source Java projects that implement time behavior in their source code. The results show that our approach achieves 100% precision and recall in identifying time related information. They also show that 95% of the timed automata extracted from source code correctly model the time behavior of the method. Finally, we show the applicability of our timed automata to identify eight real errors in four open source Apache systems.\",\n \"Title: A Method for the Joint Analysis of Numerical and Textual IT-System Data to Predict Critical System States\\nYear: 2020\\n\\nAbstract: We present a method for the joint analysis of textual and numerical IT-system data usable to predict possibly critical system states. Towards a comparative discussion culminating in a justified model and method choice, we apply logistic regression, random forest and neural networks to the prediction of critical system states. Our models consume a set of different monitoring performance metrics and log file events. To ease the analysis of IT-systems, our models judge the future system state using one binary outcome variable for the system state's criticality as \\\"alarm\\\" or \\\"no alarm\\\". Moreover, we use feature importance measures to give IT-operators guidance on which system parameters, i.e., features, to consider primarily when responding to an alarm. We evaluate our models using different configurations, including (among others) the demanded lead time window for incident response, and a set of common performance measures. This paper is an extension to previous work that adds details on how to jointly process textual and numerical data.\",\n \"Title: Comparing fine-grained source code changes and code churn for bug prediction - A retrospective\\nYear: 2021\\n\\nAbstract: AbstractMore than two decades ago, researchers started to mine the data stored in software repositories to help software developers in making informed decisions for developing and testing software systems. Bug prediction was one of the most promising and popular research directions that uses the data stored in software repositories to predict the bug-proneness or number of bugs in source files. On that topic and as part of Emanuel's PhD studies, we submitted a paper with the title Comparing fine-grained source code changes and code churn for bug prediction [8] to the 8th Working Conference on Mining Software Engineering, held 2011 in beautiful Honolulu, Hawaii. Ten years later, it got selected as one of the finalists to receive the MSR 2021 Most Influential Paper Award. In the following, we provide a retrospective on our work, describing the road to publishing this paper, its impact in the field of bug prediction, and the road ahead.\",\n \"Title: Comprehending Spreadsheets: Which Strategies do Users Apply?\\nYear: 2021\\n\\nAbstract: Spreadsheets are a ubiquitous means of communication and decision making, and adequate comprehension is required for reading, interpretation as well as maintenance. Despite being around for forty years, only selected aspects of comprehension were studied, mainly focusing on error detection and correction, and largely disregarding a more holistic view on spreadsheet comprehension processes.In this ...\"\n]","isConversational":false},"profile_size":{"kind":"number","value":78,"string":"78"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13555,"string":"13,555"}}},{"rowIdx":13556,"cells":{"question_id":{"kind":"string","value":"Di Jin"},"question":{"kind":"string","value":"Generate an abstract for the title \"On Generalizing Static Node Embedding to Dynamic Settings\" using the following items: 1. Temporal graph embedding\n2. Dynamic tasks\n3. Static embedding methods\n4. Graph time-series representations\n5. Reachability graph"},"answer":{"kind":"string","value":"ABSTRACTTemporal graph embedding has been widely studied thanks to its superiority in tasks such as prediction and recommendation. Despite the advances in algorithms and novel frameworks such as deep learning, there has been relatively little work on systematically studying the properties of temporal network models and their cornerstones, the graph time-series representations that are used in these approaches. This paper aims to fill this gap by introducing a general framework that extends an arbitrary existing static embedding approach to handle dynamic tasks, and conducting a systematic study of seven base static embedding methods and six temporal network models. Our framework generalizes static node embeddings derived from the time-series representation of stream data to the dynamic setting by modeling the temporal dependencies with classic models such as the reachability graph. While previous works on dynamic modeling and embedding have focused on representing a stream of timestamped edges using a time-series of graphs based on a specific time-scale (\\eg, 1 month), we introduce the notion of an ε-graph time-series that uses a fixed number of edges for each graph, and show its superiority in practical settings over the standard solution. From the 42 methods that our framework subsumes, we find that leveraging the new ε-graph time-series representation and capturing temporal dependencies with the proposed reachability or summary graph tend to perform well. Furthermore, the new dynamic embedding methods based on our framework perform comparably and on average better than the state-of-the-art embedding methods designed specifically for temporal graphs in link prediction tasks."},"correct_answer":{"kind":"string","value":"ABSTRACTTemporal graph embedding has been widely studied thanks to its superiority in tasks such as prediction and recommendation. Despite the advances in algorithms and novel frameworks such as deep learning, there has been relatively little work on systematically studying the properties of temporal network models and their cornerstones, the graph time-series representations that are used in these approaches. This paper aims to fill this gap by introducing a general framework that extends an arbitrary existing static embedding approach to handle dynamic tasks, and conducting a systematic study of seven base static embedding methods and six temporal network models. Our framework generalizes static node embeddings derived from the time-series representation of stream data to the dynamic setting by modeling the temporal dependencies with classic models such as the reachability graph. While previous works on dynamic modeling and embedding have focused on representing a stream of timestamped edges using a time-series of graphs based on a specific time-scale (\\eg, 1 month), we introduce the notion of an ε-graph time-series that uses a fixed number of edges for each graph, and show its superiority in practical settings over the standard solution. From the 42 methods that our framework subsumes, we find that leveraging the new ε-graph time-series representation and capturing temporal dependencies with the proposed reachability or summary graph tend to perform well. Furthermore, the new dynamic embedding methods based on our framework perform comparably and on average better than the state-of-the-art embedding methods designed specifically for temporal graphs in link prediction tasks."},"documents":{"kind":"list like","value":["Title: Decentralized mining social network communities with agents.\nYear: 2013\n\nAbstract: Network community mining algorithms aim at efficiently and effectively discovering all such communities from a given network. Many related methods have been proposed and applied to different areas including social network analysis, gene network analysis and web clustering engines. Most of the existing methods for mining communities are centralized. In this paper, we present a multi-agent based decentralized algorithm, in which a group of autonomous agents work together to mine a network through a proposed self-aggregation and self-organization mechanism. Thanks to its decentralized feature, our method is potentially suitable for dealing with distributed networks, whose global structures are hard to obtain due to their geographical distributions, decentralized controls or huge sizes. The effectiveness of our method has been tested against different benchmark networks.","Title: Ant colony optimization for community detection in large-scale complex networks\nYear: 2011\n\nAbstract: In this paper we present a new ant colony optimization for community detection in large networks, which takes modularity Q as objective function. An important difference that distinguishes our algorithm from the former ant algorithms is the manner in which the ants are used in the algorithm. Unlike those existing methods in which each ant searches for a candidate solution, each ant in our algorithm only decides whether its current vertex joins the community of its previous vertex with the aid of a simulated annealing idea, whose purpose is to locally optimize function Q. In each iteration, the ants work collectively so as to uncover the community structure of the network. Moreover, we introduce a thought of “layer and rule” into this method for further improving its performance. Our algorithm doesn't employ the pheromone, which reduces its running time and makes it well suitable for large-scale networks. Meanwhile, it still performs very well on both computer-generated benchmark and some widely used real-world networks compared with a set of competing algorithm in terms of clustering quality.","Title: Genetic Algorithm With A Local Search Strategy For Discovering Communities In Complex Networks\nYear: 2013\n\nAbstract: In order to further improve the performance of current genetic algorithms aiming at discovering communities, a local search based genetic algorithm (GALS) is here proposed. The core of GALS is a local search based mutation technique. In order to overcome the drawbacks of traditional mutation methods, the paper develops the concept of marginal gene and then the local monotonicity of modularity function Q is deduced from each node's local view. Based on these two elements, a new mutation method combined with a local search strategy is presented. GALS has been evaluated on both synthetic benchmarks and several real networks, and compared with some presently competing algorithms. Experimental results show that GALS is highly effective and efficient for discovering community structure.","Title: Identification Of Hybrid Node And Link Communities In Complex Networks\nYear: 2013\n\nAbstract: Identifying communities in complex networks is an effective means for analyzing complex systems, with applications in diverse areas such as social science, engineering, biology and medicine. Finding communities of nodes and finding communities of links are two popular schemes for network analysis. These schemes, however, have inherent drawbacks and are inadequate to capture complex organizational structures in real networks. We introduce a new scheme and an effective approach for identifying complex mixture structures of node and link communities, called hybrid node-link communities. A central piece of our approach is a probabilistic model that accommodates node, link and hybrid node-link communities. Our extensive experiments on various real-world networks, including a large protein-protein interaction network and a large network of semantically associated words, illustrated that the scheme for hybrid communities is superior in revealing network characteristics. Moreover, the new approach outperformed the existing methods for finding node or link communities separately.","Title: The (un)supervised detection of overlapping communities as well as hubs and outliers via (bayesian) NMF\nYear: 2014\n\nAbstract: The detection of communities in various networks has been considered by many researchers. Moreover, it is preferable for a community detection method to detect hubs and outliers as well. This becomes even more interesting and challenging when taking the unsupervised assumption, that is, we do not assume the prior knowledge of the number K of communities. In this poster, we define a novel model to identify overlapping communities as well as hubs and outliers. When K is given, we propose a normalized symmetric nonnegative matrix factorization algorithm to learn the parameters of the model. Otherwise, we introduce a Bayesian symmetric nonnegative matrix factorization to learn the parameters of the model, while determining K. Our experiment indicate its superior performance on various networks.","Title: Modeling and simulation of performance analysis for a cluster-based Web server\nYear: 2006\n\nAbstract: Higher scalability and availability of Web servers are required as the traffic on the Internet has been increasing dramatically over the last few years. This paper focuses on modeling and simulation of the performance analysis for the test system of a cluster-based Web server consisting of five real servers. Three ways of load balancing are introduced, namely, network address translation, IP tunneling and direct routing. Calculation of packet delay is discussed for each part of the running system according to the transferred data in the system. The input model and the system model to measure and simulate the system performance are derived, and the probability distributions of the delay data are specified for using random inputs in the model according to the Q–Q plot and the cumulative distribution function. After the solution of performance tuning problems is evaluated, the maximum process capability of the system is found and a possible performance bottleneck is analyzed.","Title: Discovering link communities in complex networks by exploiting link dynamics\nYear: 2013\n\nAbstract: Discovery of communities in complex networks is a fundamental data analysis problem with applications in various domains. Most of the existing approaches have focused on discovering communities of nodes, while recent studies have shown great advantages and utilities of the knowledge of communities of links in networks. From this new perspective, we propose a link dynamics based algorithm, called UELC, for identifying link communities of networks. In UELC, the stochastic process of a link-node-link random walk is employed to unfold an embedded bipartition structure of links in a network. The local mixing properties of the Markov chain underlying the random walk are then utilized to extract two emerging link communities. Further, the random walk and the bipartitioning processes are wrapped in an iterative subdivision strategy to recursively identify link partitions that segregate the network links into multiple subdivisions. We evaluate the performance of the new method on synthetic benchmarks and demonstrate its utility on real-world networks. Our experimental results show that our method is highly effective for discovering link communities in complex networks. As a comparison, we also extend UELC to extracting communities of nodes, and show that it is effective for node community identification.","Title: Genetic Algorithm with Local Search for Community Mining in Complex Networks\nYear: 2010\n\nAbstract: Detecting communities from complex networks has triggered considerable attention in several application domains. Targeting this problem, a local search based genetic algorithm (GALS) which employs a graph-based representation (LAR) has been proposed in this work. The core of the GALS is a local search based mutation technique. Aiming to overcome the drawbacks of the existing mutation methods, a concept called marginal gene has been proposed, and then an effective and efficient mutation method, combined with a local search strategy which is based on the concept of marginal gene, has also been proposed by analyzing the modularity function. Moreover, in this paper the percolation theory on ER random graphs is employed to further clarify the effectiveness of LAR presentation; A Markov random walk based method is adopted to produce an accurate and diverse initial population; the solution space of GALS will be significantly reduced by using a graph based mechanism. The proposed GALS has been tested on both computer-generated and real-world networks, and compared with some competitive community mining algorithms. Experimental result has shown that GALS is highly effective and efficient for discovering community structure.","Title: Ant colony optimization with Markov random walk for community detection in graphs\nYear: 2011\n\nAbstract: Network clustering problem (NCP) is the problem associated to the detection of network community structures. Building on Markov random walks we address this problem with a new ant colony optimization strategy, named as ACOMRW, which improves prior results on the NCP problem and does not require knowledge of the number of communities present on a given network. The framework of ant colony optimization is taken as the basic framework in the ACOMRWalgorithm. At each iteration, a Markov random walk model is taken as heuristic rule; all of the ants' local solutions are aggregated to a global one through clustering ensemble, which then will be used to update a pheromone matrix. The strategy relies on the progressive strengthening of within-community links and the weakening of between-community links. Gradually this converges to a solution where the underlying community structure of the complex network will become clearly visible. The performance of algorithm ACOMRW was tested on a set of benchmark computer-generated networks, and as well on real-world network data sets. Experimental results confirm the validity and improvements met by this approach.","Title: An Ant-Based Algorithm with Local Optimization for Community Detection in Large-Scale Networks\nYear: 2012\n\nAbstract: In this paper, we propose a multi-layer ant-based algorithm (MABA), which detects communities from networks by means of locally optimizing modularity using individual ants. The basic version of MABA, namely SABA, combines a self-avoiding label propagation technique with a simulated annealing strategy for ant diffusion in networks. Once the communities are found by SABA, this method can be reapplied to a higher level network where each obtained community is regarded as a new vertex. The aforementioned process is repeated iteratively, and this corresponds to MABA. Thanks to the intrinsic multi-level nature of our algorithm, it possesses the potential ability to unfold multi-scale hierarchical structures. Furthermore, MABA has the ability that mitigates the resolution limit of modularity. The proposed MABA has been evaluated on both computer-generated benchmarks and widely used real-world networks, and has been compared with a set of competitive algorithms. Experimental results demonstrate that MABA is both effective and efficient (in near linear time with respect to the size of network) for discovering communities.","Title: Key Distribution Using Dual Quantum Channels\nYear: 2008\n\nAbstract: Quantum key distribution has been a relatively recent addition to data security and integrity. It is based on advanced ideas that emanate from Quantum theory. Additionally, the optical network and the quantum photon provide a formidable basis for the application of quantum key distribution. We propose a new way of distributing encryption keys over quantum channels. The proposed method takes advantage of dual quantum channels between the communicating entities and complementary measuring bases in order to improve the efficiency with which the keys can be exchanged.","Title: Fast convergent key distribution algorithms using a dual quantum channel\nYear: 2009\n\nAbstract: This paper presents three new algorithms over a dual quantum channel to enhance the convergence speed of the key distribution process. The proposed algorithms effectively use a dual quantum channel between the communicating entities and, when combined with the proposed selection of the basis, improve the efficiency of the key distribution process as well as deter Eve. With the use of an initialization vector (IV), the proposed algorithm does not necessitate information exchange about the bases between Alice and Bob, thus removing the requirement of the availability of a two-way conventional channel. Simultaneously, it allows the detection of the presence of Eve in real time. Copyright (C) 2008 John Wiley & Sons, Ltd.","Title: Ant Colony Optimization with a New Random Walk Model for Community Detection in Complex Networks\nYear: 2013\n\nAbstract: Detecting communities from complex networks has recently triggered great interest. Aiming at this problem, a new ant colony optimization strategy building on the Markov random walks theory, which is named as MACO, is proposed in this paper. The framework of ant colony optimization is taken as the basic framework in this algorithm. In each iteration, a Markov random walk model is employed as heuristic rule; all of the ants' local solutions are aggregated to a global one through an idea of clustering ensemble, which then will be used to update a pheromone matrix. The strategy relies on the progressive strengthening of within-community links and the weakening of between-community links. Gradually this converges to a solution where the underlying community structure of the complex network will become clearly visible. The proposed MACO has been evaluated both on synthetic benchmarks and on some real-world networks, and compared with some present competing algorithms. Experimental result has shown that MACO is highly effective for discovering communities.","Title: Fast Complex Network Clustering Algorithm Using Agents\nYear: 2013\n\nAbstract: Recently, the sizes of networks are always very huge, and they take on distributed nature. Aiming at this kind of network clustering problem, in the sight of local view, this paper proposes a fast network clustering algorithm in which each node is regarded as an agent, and each agent tries to maximize its local function in order to optimize network modularity defined by function Q, rather than optimize function Q from the global view as traditional methods. Both the efficiency and effectiveness of this algorithm are tested against computer-generated and real-world networks. Experimental result shows that this algorithm not only has the ability of clustering large-scale networks, but also can attain very good clustering quality compared with the existing algorithms. Furthermore, the parameters of this algorithm are analyzed.","Title: A Markov random walk under constraint for discovering overlapping communities in complex networks\nYear: 2013\n\nAbstract: The detection of overlapping communities in complex networks has motivated recent research in relevant fields. Aiming to address this problem, we propose a Markov-dynamics-based algorithm, called UEOC, which means 'unfold and extract overlapping communities'. In UEOC, when identifying each natural community that overlaps, a Markov random walk method combined with a constraint strategy, which is based on the corresponding annealed network (degree conserving random network), is performed to unfold the community. Then, a cutoff criterion with the aid of a local community function, called conductance, which can be thought of as the ratio between the number of edges inside the community and those leaving it, is presented to extract this emerged community from the entire network. The UEOC algorithm depends on only one parameter whose value can be easily set, and it requires no prior knowledge of the hidden community structures. The proposed UEOC has been evaluated both on synthetic benchmarks and on some real-world networks, and has been compared with a set of competing algorithms. The experimental result has shown that UEOC is highly effective and efficient for discovering overlapping communities.","Title: Cooperative Localization in WSNs Using Gaussian Mixture Modeling: Distributed ECM Algorithms\nYear: 2015\n\nAbstract: We study cooperative sensor network localization in a realistic scenario where the underlying measurement errors more probably follow a non-Gaussian distribution; the measurement error distribution is unknown without conducting massive offline calibrations; and non-line-of-sight identification is not performed due to the complexity constraint and/or storage limitation. The underlying measurement error distribution is approximated parametrically by a Gaussian mixture with finite number of components, and the expectation-conditional maximization (ECM) criterion is adopted to approximate the maximum-likelihood estimator of the unknown sensor positions and an extra set of Gaussian mixture model parameters. The resulting centralized ECM algorithms lead to easier inference tasks and meanwhile retain several convergence properties with a proof of the “space filling” condition. To meet the scalability requirement, we further develop two distributed ECM algorithms where an average consensus algorithm plays an important role for updating the Gaussian mixture model parameters locally. The proposed algorithms are analyzed systematically in terms of computational complexity and communication overhead. Various computer based tests are also conducted with both simulation and experimental data. The results pin down that the proposed distributed algorithms can provide overall good performance for the assumed scenario even under model mismatch, while the existing competing algorithms either cannot work without the prior knowledge of the measurement error statistics or merely provide degraded localization performance when the measurement error is clearly non-Gaussian.","Title: A Unified Semi-Supervised Community Detection Framework Using Latent Space Graph Regularization.\nYear: 2015\n\nAbstract: Community structure is one of the most important properties of complex networks and is a foundational concept in exploring and understanding networks. In real world, topology information alone is often inadequate to accurately find community structure due to its sparsity and noises. However, potential useful prior information can be obtained from domain knowledge in many applications. Thus, how to improve the community detection performance by combining network topology with prior information becomes an interesting and challenging problem. Previous efforts on utilizing such priors are either dedicated or insufficient. In this paper, we firstly present a unified interpretation to a group of existing community detection methods. And then based on this interpretation, we propose a unified semi-supervised framework to integrate network topology with prior information for community detection. If the prior information indicates that some nodes belong to the same community, we encode it by adding a graph regularization term to penalize the latent space dissimilarity of these nodes. This framework can be applied to many widely-used matrix-based community detection methods satisfying our interpretation, such as nonnegative matrix factorization, spectral clustering, and their variants. Extensive experiments on both synthetic and real networks show that the proposed framework significantly improves the accuracy of community detection, especially on networks with unclear structures.","Title: Security in Computing and Communications: Second International Symposium, SSCC 2014, Delhi, India, September 24-27, 2014. Proceedings\nYear: 2014\n\nAbstract: This book constitutes the refereed proceedings of the International Symposium on Security in Computing and Communications, SSCC 2014, held in Delhi, India, in September 2013. The 36 revised full papers presented together with 12 work-in-progress papers were carefully reviewed and selected from 132 submissions. The papers are organized in topical sections on security and privacy in networked systems; authentication and access control systems; encryption and cryptography; system and network security; work-in-progress.","Title: A Stochastic Model For Detecting Heterogeneous Link Communities In Complex Networks\nYear: 2015\n\nAbstract: Discovery of communities in networks is a fundamental data analysis problem. Most of the existing approaches have focused on discovering communities of nodes, while recent studies have shown great advantages and utilities of the knowledge of communities of links. Stochastic models provides a promising class of techniques for the identification of modular structures, but most stochastic models mainly focus on the detection of node communities rather than link communities. We propose a stochastic model, which not only describes the structure of link communities, but also considers the heterogeneous distribution of community sizes, a property which is often ignored by other models. We then learn the model parameters using a method of maximum likelihood based on an expectation-maximization algorithm. To deal with large complex real networks, we extend the method by a strategy of iterative bipartition. The extended method is not only efficient, but is also able to determine the number of communities for a given network. We test our approach on both synthetic benchmarks and real-world networks including an application to a large biological network, and also compare it with two existing methods. The results demonstrate the superior performance of our approach over the competing methods for detecting link communities.","Title: Modeling With Node Degree Preservation Can Accurately Find Communities\nYear: 2015\n\nAbstract: An important problem in analyzing complex networks is discovery of modular or community structures embedded in the networks. Although being promising for identifying network communities, the popular stochastic models often do not preserve node degrees, thus reducing their representation power and applicability to real-world networks. Here we address this critical problem. Instead of using a block-model, we adopted a random-graph null model to faithfully capture community structures by preserving in the model the expected node degrees. The new model, learned using nonnegative matrix factorization, is more accurate and robust in representing community structures than the existing methods. Our results from extensive experiments on synthetic benchmarks and real-world networks show the superior performance of the new method over the existing methods in detecting both disjoint and overlapping communities.","Title: Dialog Act classification in Chinese spoken language\nYear: 2013\n\nAbstract: Dialog Act (DA) is an important pragmatics feature for us to understand speakers' intention. Many methods have been proposed to recognize DA tags. However, little work has been conducted to address the problem of DA tagging in Chinese spoken dialog language. In this work, we employ both the lexical features and the inter-utterance dependency features for DA tagging. And we propose three different methods: n-gram, extended HMM and n-gram+KNN. The experimental results show that these methods are effective for the task.","Title: Perseus: an interactive large-scale graph mining and visualization tool\nYear: 2015\n\nAbstract: Given a large graph with several millions or billions of nodes and edges, such as a social network, how can we explore it efficiently and find out what is in the data? In this demo we present Perseus, a large-scale system that enables the comprehensive analysis of large graphs by supporting the coupled summarization of graph properties and structures, guiding attention to outliers, and allowing the user to interactively explore normal and anomalous node behaviors. Specifically, Perseus provides for the following operations: 1) It automatically extracts graph invariants (e.g., degree, PageRank, real eigenvectors) by performing scalable, offline batch processing on Hadoop; 2) It interactively visualizes univariate and bivariate distributions for those invariants; 3) It summarizes the properties of the nodes that the user selects; 4) It efficiently visualizes the induced subgraph of a selected node and its neighbors, by incrementally revealing its neighbors. In our demonstration, we invite the audience to interact with Perseus to explore a variety of multi-million-edge social networks including a Wikipedia vote network, a friendship/foeship network in Slashdot, and a trust network based on the consumer review website Epinions.com.","Title: Community Detection With Manifold Learning On Speaker I-Vector Space For Chinese\nYear: 2015\n\nAbstract: Speaker recognition with clustering speech signals of the same speaker is an important speech analysis task in various applications. Recent works have shown that there was an underlying manifold on which speaker utterances live in the model-parameter space. However, most speaker clustering methods work on the Euclidean space, and hence often fail to discover the intrinsic geometrical structure of the data space. For this problem, we consider to convert the speaker i-vector representation of utterances in the Euclidean space into a network structure constructed based on the local (k) nearest neighbor relationship of these signals. We then propose a community detection model on the network for clustering signals. The new model is based on the probabilistic community memberships, and is further refined with the idea that: if two connected nodes have a high similarity, their community membership distributions in the model should be made close. This refinement enhances the local invariance assumption, and thus better respects the structure of the underlying manifold than the existing community detection methods. Some experiments are conducted on speaker content network built from a Chinese speaker recognition database. The results confirmed the effectiveness of this new method.","Title: Efficient cooperative localization algorithm in LOS/NLOS environments\nYear: 2015\n\nAbstract: The well-known cooperative localization algorithm, 'sum-product algorithm over a wireless network' (SPAWN) has two major shortcomings, a relatively high computational complexity and a large communication load. Using the Gaussian mixture model with a model selection criterion and the sigma-point (SP) methods, we propose the SPAWN-SP to overcome these problems. The SPAWN-SP easily accommodates different localization scenarios due to its high flexibility in message representation. Furthermore, harsh LOS/NLOS environments are considered for the evaluation of cooperative localization algorithms. Our simulation results indicate that the proposed SPAWN-SP demonstrates high localization accuracy in different localization scenarios, thanks to its high flexibility in message representation.","Title: Dithering in quantized RSS based localization\nYear: 2015\n\nAbstract: We study maximum likelihood (ML) position estimation using quantized received signal strength measurements. In order to mitigate the undesired quantization effect in the observations, the dithering technique is adopted. Various dither noise distributions are considered and the corresponding likelihood functions are derived. Simulation results show that the proposed ML estimator with dithering is able to generate a significantly reduced bias but a modestly increased mean-square-error as compared to the conventional ML estimator without dithering.","Title: Semantic Community Identification in Large Attribute Networks.\nYear: 2016\n\nAbstract: Identification of modular or community structures of a network is a key to understanding the semantics and functions of the network. While many network community detection methods have been developed, which primarily explore network topologies, they provide little semantic information of the communities discovered. Although structures and semantics are closely related, little effort has been made to discover and analyze these two essential network properties together. By integrating network topology and semantic information on nodes, e.g., node attributes, we study the problems of detection of communities and inference of their semantics simultaneously. We propose a novel nonnegative matrix factorization (NMF) model with two sets of parameters, the community membership matrix and community attribute matrix, and present efficient updating rules to evaluate the parameters with a convergence guarantee. The use of node attributes improves upon community detection and provides a semantic interpretation to the resultant network communities. Extensive experimental results on synthetic and real-world networks not only show the superior performance of the new method over the state-of-the-art approaches, but also demonstrate its ability to semantically annotate the communities.","Title: Detect Overlapping Communities via Ranking Node Popularities.\nYear: 2016\n\nAbstract: Detection of overlapping communities has drawn much attention lately as they are essential properties of real complex networks. Despite its influence and popularity, the well studied and widely adopted stochastic model has not been made effective for finding overlapping communities. Here we extend the stochastic model method to detection of overlapping communities with the virtue of autonomous determination of the number of communities. Our approach hinges upon the idea of ranking node popularities within communities and using a Bayesian method to shrink communities to optimize an objective function based on the stochastic generative model. We evaluated the novel approach, showing its superior performance over five state-of-the-art methods, on large real networks and synthetic networks with ground-truths of overlapping communities.","Title: TeleLink: Link Prediction in Social Network Based on Multiplex Cohesive Structures.\nYear: 2016\n\nAbstract: Given a network where the same set of nodes have multiple types of relationships, how do we efficiently predict potential links in the future (e.g., interactions between social actors), and how do we predict links using information from other relationships? These problems have been widely studied recently, most of the existing methods either aggregate multiple types of relationships into a single network or consider them separately and ignore the correlations across relationships, leading to information loss. In this work, we present TeleLink, a general link prediction model that works for networks with single and multiple relationships. TeleLink predicts potential links based on community detection and improves link prediction by bringing in a cohesive structure across multiple networks constructed by different relationships or node attributes. To further improve the prediction performance, we extend TeleLink to a semi-supervised scheme, incorporating partially labeled information. Our extensive experiments show that TeleLink outperforms existing methods in predicting new links. Specifically, among the various datasets that we study, TeleLink achieves a precision improvement by up to 110 % compared to the baselines.","Title: Semi-supervised community detection based on non-negative matrix factorization with node popularity.\nYear: 2017\n\nAbstract: Based on the ideas of graph regularization, we present a semi-supervised and NMF-based model to utilize the prior information of networks.By introducing the parameters of node popularities, we propose a refined PSSNMF model, which is particularly suitable for networks with large degree heterogeneity and unbalanced community structure.The prior information with node popularity being introduced into the model is more effective than being directly encoded into the adjacent matrix. A plethora of exhaustive studies have proved that the community detection merely based on topological information often leads to relatively low accuracy. Several approaches aim to achieve performance improvement by utilizing the background information. But they ignore the effect of node degrees on the availability of prior information. In this paper, by combining the idea of graph regularization with the pairwise constraints, we present a semi-supervised non-negative matrix factorization (SSNMF) model for community detection. And then, to alleviate the influence of the heterogeneity of node degrees and community sizes, we propose an improved SSNMF model by introducing the node popularity, namely PSSNMF, which helps to utilize the prior information more effectively. At last, the extensive experiments on both artificial and real-world networks show that the proposed method improves, as expected, the accuracy of community detection, especially on networks with large degree heterogeneity and unbalanced community structure.","Title: Fast and accurate cooperative localization in wireless sensor networks.\nYear: 2016\n\nAbstract: Cooperative localization capability is a highly desirable characteristic of wireless sensor networks. It has attracted considerable research attention in academia and industry. The sumproduct algorithm over a wireless sensor network (SPAWN) is a powerful method to cooperatively estimate the positions of many sensors (agents) using knowledge of the absolute positions of a few sensors (anchors). Drawbacks of the SPAWN, however, are its high computational complexity and communication load. In this paper we address the complexity issue, reformulate it as convolution problem and utilize the fast Fourier transform (FFT), culminating in a fast and accurate localization algorithm, which we named SPAWN-FFT. Our simulation results show SPAWN-FFT's superiority over SPAWN regarding the computational effort, while maintaining its full flexibility and localization performance.","Title: Joint Identification of Network Communities and Semantics via Integrative Modeling of Network Topologies and Node Contents.\nYear: 2017\n\nAbstract: The objective of discovering network communities, an essential step in complex systems analysis, is two-fold: identification of functional modules and their semantics at the same time. However, most existing community-finding methods have focused on finding communities using network topologies, and the problem of extracting module semantics has not been well studied and node contents, which often contain semantic information of nodes and networks, have not been fully utilized. We considered the problem of identifying network communities and module semantics at the same time. We introduced a novel generative model with two closely correlated parts, one for communities and the other for semantics. We developed a co-learning strategy to jointly train the two parts of the model by combining a nested EM algorithm and belief propagation. By extracting the latent correlation between the two parts, our new method is not only robust for finding communities and semantics, but also able to provide more than one semantic explanation to a community. We evaluated the new method on artificial benchmarks and analyzed the semantic interpretability by a case study. We compared the new method with eight state-of-the-art methods on ten real-world networks, showing its superior performance over the existing methods.","Title: Leveraging Behavioral And Social Information For Weakly Supervised Collective Classification Of Political Discourse On Twitter\nYear: 2017\n\nAbstract: Framing is a political strategy in which politicians carefully word their statements in order to control public perception of issues. Previous works exploring political framing typically analyze frame usage in longer texts, such as congressional speeches. We present a collection of weakly supervised models which harness collective classification to predict the frames used in political discourse on the microblogging platform, Twitter. Our global probabilistic models show that by combining both lexical features of tweets and network-based behavioral features of Twitter, we are able to increase the average, unsupervised F-1 score by 21.52 points over a lexical baseline alone.","Title: Adaptive Community Detection Incorporating Topology and Content in Social Networks.\nYear: 2018\n\nAbstract: In social network analysis, community detection is a basic step to understand the structure, function and semantics of networks. Some conventional community detection methods may have limited performance because they merely focus on topological structure of networks. In addition to topology, content information is another significant aspect of social networks. Some state-of-the-art methods started to combine these two aspects of information, but they often assume that topology and content share the same characteristics. However, for some examples of social networks, content may mismatch with topological structure. In order to better cope with such situations, we introduce a novel community detection method under the framework of non-negative matrix factorization (NMF). Our proposed method integrates topology and content of networks, and introduces a novel adaptive parameter for controlling the contribution of content with respect to the identified mismatch degree between the topological and content information. The case study using real social networks show that our new method can simultaneously obtain community partition and the corresponding semantic descriptions. Experiments on both artificial networks and real social networks further indicate that our method outperforms some state-of-the-art methods while exhibiting more robust behaviour when the mismatch topological and content information is observed.","Title: Perseus-Hub: Interactive And Collective Exploration Of Large-Scale Graphs\nYear: 2017\n\nAbstract: Graphs emerge naturally in many domains, such as social science, neuroscience, transportation engineering, and more. In many cases, such graphs have millions or billions of nodes and edges, and their sizes increase daily at a fast pace. How can researchers from various domains explore large graphs interactively and efficiently to find out what is 'important'? How can multiple researchers explore a new graph dataset collectively and \"help\" each other with their findings? In this article, we present PERSEUS-HUB, a large-scale graph mining tool that computes a set of graph properties in a distributed manner, performs ensemble, multi-view anomaly detection to highlight regions that are worth investigating, and provides users with uncluttered visualization and easy interaction with complex graph statistics. PERSEUS-HUB uses a Spark cluster to calculate various statistics of large-scale graphs efficiently, and aggregates the results in a summary on the master node to support interactive user exploration. In PERSEUS-HUB, the visualized distributions of graph statistics provide preliminary analysis to understand a graph. To perform a deeper analysis, users with little prior knowledge can leverage patterns (e.g., spikes in the power-law degree distribution) marked by other users or experts. Moreover, PERSEUS-HUB guides users to regions of interest by highlighting anomalous nodes and helps users establish a more comprehensive understanding about the graph at hand. We demonstrate our system through the case study on real, large-scale networks.","Title: Exploratory Analysis of Graph Data by Leveraging Domain Knowledge\nYear: 2017\n\nAbstract: Given the soaring amount of data being generated daily, graph mining tasks are becoming increasingly challenging, leading to tremendous demand for summarization techniques. Feature selection is a representative approach that simplifies a dataset by choosing features that are relevant to a specific task, such as classification, prediction, and anomaly detection. Although it can be viewed as a way to summarize a graph in terms of a few features, it is not well-defined for exploratory analysis, and it operates on a set of observations jointly rather than conditionally (i.e., feature selection from many graphs vs. selection for an input graph conditioned on other graphs). In this work, we introduce EAGLE (Exploratory Analysis of Graphs with domain knowLEdge), a novel method that creates interpretable, feature-based, and domain-specific graph summaries in a fully automatic way. That is, the same graph in different domains-e.g., social science and neuroscience-will be described via different EAGLE summaries, which automatically leverage the domain knowledge and expectations. We propose an optimization formulation that seeks to find an interpretable summary with the most representative features for the input graph so that it is: diverse, concise, domain-specific, and efficient. Extensive experiments on synthetic and real-world datasets with up to ~1M edges and ~400 features demonstrate the effectiveness and efficiency of EAGLE and its benefits over existing methods. We also show how our method can be applied to various graph mining tasks, such as classification and exploratory analysis.","Title: High-throughput, high-resolution Generated Adversarial Network Microscopy.\nYear: 2018\n\nAbstract: We for the first time combine generated adversarial network (GAN) with wide-field light microscopy to achieve deep learning super-resolution under a large field of view (FOV). By appropriately adopting prior microscopy data in an adversarial training, the network can recover a high-resolution, accurate image of new specimen from its single low-resolution measurement. This capacity has been adequately demonstrated by imaging various types of samples, such as USAF resolution target, human pathological slides and fluorescence-labelled fibroblast cells. Their gigapixel, multi-color reconstructions verify a successful GAN-based single image super-resolution procedure. Furthermore, this deep learning-based imaging approach doesn;t necessarily introduce any change to the setup of a conventional wide-filed microscope, reconstructing large FOV (about 95 mm^2), high-resolution (about 1.7 {mu}m) image at a high speed (in 1 second). As a result, GAN-microscopy opens a new way to computationally overcome the general challenge of high-throughput, high-resolution microscopy that is originally coupled to the physical limitation of systemu0027s optics.","Title: A Network-Specific Markov Random Field Approach to Community Detection.\nYear: 2018\n\nAbstract: Markov Random Field (MRF) is a powerful framework for developing probabilistic models of complex problems. MRF models possess rich structures to represent properties and constraints of a problem. It has been successful on many application problems, particularly those of computer vision and image processing, where data are structured, e.g., pixels are organized on grids. The problem of identifying communities in networks, which is essential for network analysis, is in principle analogous to finding objects in images. It is surprising that MRF has not yet been explored for network community detection. It is challenging to apply MRF to network analysis problems where data are organized on graphs with irregular structures. Here we present a network specific MRF approach to community detection. The new method effectively encodes the structural properties of an irregular network in an energy function (the core of an MRF model) so that the minimization of the function gives rise to the best community structures. We analyzed the new MRF-based method on several synthetic benchmarks and real world networks, showing its superior performance over the state-of-the-art methods for community identification.","Title: Robust Detection of Link Communities in Large Social Networks by Exploiting Link Semantics.\nYear: 2018\n\nAbstract: Community detection has been extensively studied for various applications, focusing primarily on network topologies. Recent research has started to explore node contents to identify semantically meaningful communities and interpret their structures using selected words. However, links in real networks typically have semantic descriptions, e.g., comments and emails in social media, supporting the notion of communities of links Indeed, communities of links can better describe multiple roles that nodes may play and provide a richer characterization of community behaviors than communities of nodes. The second issue in community finding is that most existing methods assume network topologies and descriptive contents to be consistent and to carry the compatible information of node group membership, which is generally violated in real networks. These methods are also restricted to interpret one community with one topic. The third problem is that the existing methods have used top ranked words or phrases to label topics when interpreting communities. However, it is often difficult to comprehend the derived topics using words or phrases, which may be irrelevant. To address these issues altogether, we propose a new unified probabilistic model that can be learned by a dual nested expectation-maximization algorithm. Our new method explores the intrinsic correlation between communities and topics to discover link communities robustly and extract adequate community summaries in sentences instead of words for topic labeling at the same time. It is able to derive more than one topical summary per community to provide rich explanations. We present experimental results to show the effectiveness of our new approach, and evaluate the quality of the results by a case study.","Title: Incorporating network structure with node contents for community detection on large networks using deep learning.\nYear: 2018\n\nAbstract: Community detection is an important task in social network analysis. In community detection, in general, there exist two types of the models that utilize either network topology or node contents. Some studies endeavor to incorporate these two types of models under the framework of spectral clustering for a better community detection. However, it was not successful to obtain a big achievement since they used a simple way for the combination. To reach a better community detection, it requires to realize a seamless combination of these two methods. For this purpose, we re-examine the properties of the modularity maximization and normalized-cut models and fund out a certain approach to realize a seamless combination of these two models. These two models seek for a low-rank embedding to represent of the community structure and reconstruct the network topology and node contents, respectively. Meanwhile, we found that autoencoder and spectral clustering have a similar framework in their low- rank matrix reconstruction. Based on this property, we proposed a new approach to seamlessly combine the models of modularity and normalized-cut via the autoencoder. The proposed method also utilized the advantages of the deep structure by means of deep learning. The experiment demonstrated that the proposed method can provide a nonlinearly deep representation for a large-scale network and reached an efficient community detection. The evaluation results showed that our proposed method outperformed the existing leading methods on nine real-world networks.","Title: Tracking and Detecting Dynamic Communities with Node Popularity Preservation\nYear: 2017\n\nAbstract: Temporal networks are increasingly prevalent in modern data, and the analysis of such networks can uncover the important phenomenon and characterize the properties of networks of dynamical units. This paper presents a new community detection model using nonnegative matrix factorization (NMF) which is capable of detecting and tracking dynamic communities in temporal networks. A novel temporal smoothness constraint by preserving the expected node popularity of two consecutive timeslices is proposed, and the community membership transition matrices are introduced to capture the sudden changes in network topologies in dynamic cases. We then propose a gradient descent algorithm to optimize objective function and prove its correctness and convergence. Experimental results on some synthetic and real benchmarked networks show the effectiveness of the proposed new method in detecting communities as well as finding their changes in dynamic networks.","Title: Using Deep Learning for Community Discovery in Social Networks\nYear: 2017\n\nAbstract: Community detection is an important task in social network analysis. Existing methods typically use the topological information alone, and ignore the rich information available in the content data. Recently, some researchers have noticed that user profiles can also benefit to community detection, and hence the combination of topology and node contents has become a new hot topic. Some methods using both topology and content have been proposed. However, they often suffer from two drawbacks: 1) they cannot extract a potential deep representation of the network; 2) they cannot automatically weight different information sources with adequate balance parameters. To overcome these issues, we propose a deep integration representation (DIR) algorithm via deep joint reconstruction, which is motivated by the similarity between deep feedforward auto-encoders and spectral clustering in terms of matrix reconstruction. Thanks to spectral clustering which is one of the best community detection methods, the proposed new method is also good at community discovery task. In addition, DIR has further benefit because it not only provides a nonlinear and deep representation of the network, but also learns the most suitable balance between different components automatically. We compare the proposed new approach with nine state-of-the-art community detection methods on eight real relatively large networks. The experimental results show the definite superiority of this new approach.","Title: Finding Communities with Hierarchical Semantics by Distinguishing General and Specialized topics.\nYear: 2018\n\nAbstract: Using network topology and semantic contents to find topic-related communities is a new trend in the field of community detection. By analyzing texts in social networks, we find that topics in networked contents are often hierarchical. In most cases, they have a two-level semantic structure with general and specialized topics, to respectively denote common and specific interests of communities. However, the existing community detection methods ignore such a hierarchy and take all words used to describe node semantics from an identical perspective. This indiscriminate use of words leads to natural defects in depicting networked content in which the deep semantics is not fully utilized. To address this problem, we propose a novel probabilistic generative model. By distinguishing the general and specialized topics of words, our model not only can find community structures more accurately, but also provide two-level semantic interpretation for each community. We train the model by deriving an efficient inference method under the framework of variational expectation-maximization. We provide a case study to show the ability of our algorithm in deep semantic interpretability of communities. The superiority of our algorithm for community detection is further demonstrated in comparison with eight state-of-the-art algorithms on eight real-world networks.","Title: PICO Element Detection in Medical Text via Long Short-Term Memory Neural Networks.\nYear: 2018\n\nAbstract: Successful evidence-based medicine (EBM) applications rely on answering clinical questions by analyzing large medical literature databases. In order to formulate a well-defined, focused clinical question, a framework called PICO is widely used, which identifies the sentences in a given medical text that belong to the four components: Participants/Problem (P), Intervention (I), Comparison (C) and Outcome (O). In this work, we present a Long Short-Term Memory (LSTM) neural network based model to automatically detect PICO elements. By jointly classifying subsequent sentences in the given text, we achieve state-of-the-art results on PICO element classification compared to several strong baseline models. We also make our curated data public as a benchmarking dataset so that the community can benefit from it.","Title: Robust Detection of Communities with Multi-semantics in Large Attributed Networks.\nYear: 2018\n\nAbstract: In this paper, we are interested in how to explore and utilize the relationship between network communities and semantic topics in order to find the strong explanatory communities robustly. First, the relationship between communities and topics displays different situations. For example, from the viewpoint of semantic mapping, their relationship can be one-to-one, one-to-many or many-to-one. But from the standpoint of underlying community structures, the relationship can be consistent, partially consistent or completely inconsistent. Second, it will be helpful to not only find communities more precise but also reveal the communities’ semantics that shows the relationship between communities and topics. To better describe this relationship, we introduce the transition probability which is an important concept in Markov chain into a well-designed nonnegative matrix factorization framework. This new transition probability matrix with a suitable prior which plays the role of depicting the relationship between communities and topics can perform well in this task. To illustrate the effectiveness of the proposed new approach, we conduct some experiments on both synthetic and real networks. The results show that our new method is superior to baselines in accuracy. We finally conduct a case study analysis to validate the new method’s strong interpretability to detected communities.","Title: Autoencoder Based Community Detection with Adaptive Integration of Network Topology and Node Contents.\nYear: 2018\n\nAbstract: Community detection plays an important role in understanding the structure and laws of social networks. Many community detection approaches have been proposed and focus on topological structure alone. In addition to topology, node contents exist in real-world networks, and may help for community detection. Recently, some studies try to combine topological structure and node contents. However, it is difficult to address an inherent situation in real-world networks, that is the mismatch between topological structure and node contents in term of community patterns. When considering both topology and content of networks, the performance of those community detection methods is often limited by this mismatch. Besides, networks are often full of nonlinear features, making those methods less effective in practice. In this paper, we present an adaptive method for community detection, which is reached by a graph regularized autoencoder approach. This new method introduces a novel adaptive parameter to achieve robust integration of the topological and content information when there exists the mismatch between those two types of information in term of communities. Experiments on both synthetic networks and real-world networks further indicate that the proposed new method exhibits more robust behavior and outperforms the leading methods when there exists the mismatch between topology and content.","Title: Hierarchical Neural Networks for Sequential Sentence Classification in Medical Scientific Abstracts.\nYear: 2018\n\nAbstract: Prevalent models based on artificial neural network (ANN) for sentence classification often classify sentences in isolation without considering the context in which sentences appear. This hampers the traditional sentence classification approaches to the problem of sequential sentence classification, where structured prediction is needed for better overall classification performance. In this work, we present a hierarchical sequential labeling network to make use of the contextual information within surrounding sentences to help classify the current sentence. Our model outperforms the state-of-the-art results by 2%-3% on two benchmarking datasets for sequential sentence classification in medical scientific abstracts.","Title: Bridging Network Embedding and Graph Summarization.\nYear: 2018\n\nAbstract: An important reason behind the prevalence of node representation learning is their superiority in downstream machine learning tasks on graphs. However, storing the vector-based node representation of massive real-world graphs often requires space that is orders of magnitude larger. To alleviate this issue, we introduce the problem of latent network summarization that is complementary to the problem of network embedding, and propose a general framework called Multi-LENS. Instead of deriving dense node-wise representations, the goal of latent network summarization is to summarize the structural properties of the graph in a latent space with dimensionality that is independent of the nodes or edges in the input graph. The size-independent graph summary given by Multi-LENS consists of (i) a set of relational aggregators with their compositions (relational functions) that captures structural features of multi-order node-centric subgraphs, and (ii) the low-rank approximations to matrices that incorporate captured structural features. In addition, Multi-LENS is able to derive node embeddings on the fly from this latent summary due to its inductive properties. Multi-LENS bridges advantages brought by both embeddings and graph summarization, and applies to graphs with or without directionality, weights, attributes or labels. Extensive experiments on both synthetic and real-world graphs show that Multi-LENS achieves 2 - 89% improvement in AUC for link prediction, while requiring less than 79x space compared to existing representation learning approaches. We also show the effectiveness of Multi-LENS summaries in anomaly and event detection on two real-world graphs.","Title: Advancing PICO Element Detection in Medical Text via Deep Neural Networks.\nYear: 2018\n\nAbstract: In evidence-based medicine (EBM), structured medical questions are always favored for efficient search of the best available evidence for treatments. PICO element detection is widely used to help structurize the clinical studies and question by identifying the sentences in a given medical text that belong to one of the four components: Participants (P), Intervention (I), Comparison (C), and Outcome (O). In this work, we propose a hierarchical deep neural network (DNN) architecture that contains dual bi-directional long short-term memory (bi-LSTM) layers to automatically detect the PICO element in medical texts. Within the model, the lower layer of bi-LSTM is for sentence encoding while the upper one is to contextualize the encoded sentence representation vector. In addition, we adopt adversarial and virtual adversarial training to regularize the model. Overall, we advance the PICO element detection to new state-of-the-art performance, outperforming the previous works by at least 4% in F1 score for all P/I/O categories.","Title: Context-Aware Data Dissemination For Icn-Based Vehicular Ad Hoc Networks\nYear: 2018\n\nAbstract: Information-centric networking (ICN) technology matches many major requirements of vehicular ad hoc networks (VANETs) in terms of its connectionless networking paradigm accordant with the dynamic environments of VANETs and is increasingly being applied to VANETs. However, wireless transmissions of packets in VANETs using ICN mechanisms can lead to broadcast storms and channel contention, severely affecting the performance of data dissemination. At the same time, frequent changes of topology due to driving at high speeds and environmental obstacles can also lead to link interruptions when too few vehicles are involved in data forwarding. Hence, balancing the number of forwarding vehicular nodes and the number of copies of packets that are forwarded is essential for improving the performance of data dissemination in information-centric networking for vehicular ad-hoc networks. In this paper, we propose a context-aware packet-forwarding mechanism for ICN-based VANETs. The relative geographical position of vehicles, the density and relative distribution of vehicles, and the priority of content are considered during the packet forwarding. Simulation results show that the proposed mechanism can improve the performance of data dissemination in ICN-based VANET in terms of a successful data delivery ratio, packet loss rate, bandwidth usage, data response time, and traversed hops.","Title: Edge Content Enhanced Network Embedding\nYear: 2018\n\nAbstract: Network embedding, aiming at learning the low dimensional representations of nodes in a network, is a key to many network analysis tasks. All the current network embedding methods primarily explore the network topology or node attributes, while no effort has been made to analyze the edge content for network embedding. The edge content, such as the email content between two users in an email network, is often naturally associated with edges. They carry rich information to describe the interaction between nodes, and provide valuable supervision to learn the representations of nodes. In this paper, we propose a novel edge content enhanced network embedding model, which incorporates the edge content to guide the network representation learning process. We provide the efficient updating rules to infer the parameters in the model, along with theoretical analysis on correctness and convergence guarantees. Extensive experiments, in comparison with the state-of-the-arts, show the superior performance of our proposed new approach on different network analysis tasks.","Title: Analysis of Activity States of Local Neuronal Microcircuits in Mouse Brain.\nYear: 2018\n\nAbstract: Time series of neuronal activity corresponding to different activity states in mouse brain are analyzed in the time domain and the time-frequency domain. The signals are associated with either a slow wave brain state or a persistent brain state. For both states, characteristic spectral features are identified and a simple detector is proposed that is able to identify the brain state with low latency and high accuracy. In practice, being able to monitor the brain state online and in real time is crucial for improved in vivo experiments and, ultimately, for a causal understanding of brain dynamics.","Title: Constrained common cluster based model for community detection in temporal and multiplex networks.\nYear: 2018\n\nAbstract: On one hand, the detection of tightly connected groups, also known as community detection in complex networks, is a prominent problem for network analysis and mining. On the other hand, almost all of social, biological, bibliographic, communication and computer systems are modeled as temporal networks, the topological structures of which evolve with time, or multiplex networks, each pair of nodes of which has multiple linked relations. Current methods of community detection for temporal networks are based on incremental, independent or evolutionary clustering, and for multiplex networks are based on fusion of the multiple links. However, all these methods ignore the common structure hidden in the networks, which is denoted as the common cluster here. So in this paper, we propose a constrained common cluster based model (C3 model) to analyze the temporal and multiplex networks, which can not only detect the community structure, but also identify the importance of each node based on the common cluster structure of both two classes of networks. The intrinsic assumption of the proposed model is that there are common or coincident clusters hidden in these networks. In detail, we first construct the Markov steady-state matrices of each snapshot of temporal network or each slice of multiplex network. Next, we propose the object function of C3 model by combining the Markov steady-state matrices, similarity matrices with community membership matrices of each snapshot or slice of the network. Last, a gradient descent algorithm based on non-negative matrix factorization is proposed for the object function. Experiments on both synthetic datasets and real-world networks demonstrate that the proposed C3 model has competitive performance based on the evaluation indexes NMI and error of community detection, otherwise, the proposed model could identify the importance of nodes of the temporal or multiplex networks.","Title: A Unified Weakly Supervised Framework for Community Detection and Semantic Matching.\nYear: 2018\n\nAbstract: Due to the sparsity of network, some community detection methods only based on topology often lead to relatively low accuracy. Although some methods have been proposed to improve the detection accuracy by using few known semi-supervised information or node content, the research of community detection not only pursues the enhancement of community accuracy, but also pays more attention to the semantic description for communities. In this paper, we proposed a unified nonnegative matrix factorization framework simultaneously for community detection and semantic matching by integrating both semi-supervised information and node content. The framework reveals two-fold community structures as well as their coupling relationship matrix, which helps to identify accurate community structure and at the same time assign specific semantic information to each community. Experiments on some real networks show that the framework is efficient to match each community with specific semantic information, and the performance are superior over the compared methods.","Title: Integrative Network Embedding via Deep Joint Reconstruction.\nYear: 2018\n\nAbstract: Network embedding is to learn a low-dimensional representation for a network in order to capture intrinsic features of the network. It has been applied to many applications, e.g., network community detection and user recommendation. One of the recent research topics for network embedding has been focusing on exploitation of diverse information, including network topology and semantic information on nodes of networks. However, such diverse information has not been fully utilized nor adequately integrated in the existing methods, so that the resulting network embedding is far from satisfactory. In this paper, we develop a weight-free multi-component network embedding approach by network reconstruction via a deep Autoencoder. Three key components make our new approach effective, i.e., a uniformed graph representation of network topology and semantic information, enhancement to the graph representation using local network structure (i.e., pairwise relationship on nodes) by sampling with latent space regularization, and integration of the diverse information in graph forms in a deep Autoencoder. Extensive experimental results on seven real-world networks demonstrate a superior performance of our method over nine state-of-the-art methods for embedding.","Title: Combination of links and node contents for community discovery using a graph regularization approach.\nYear: 2019\n\nAbstract: With the rapid growth of the networked data, the study of community detection is drawing increasing attention of researchers. A number of algorithms have been proposed and some of them have been well applied in many research fields, such as recommendation systems, information retrieval, etc. Traditionally, the community detection methods mainly use the knowledge of the topological structure which contains the most important clue for finding potential groups or communities. However, as we know, a wealth of content information exists on the nodes in real-world networks, and may help for community detection. Considering the above problem, we introduce a novel community detection method under the framework of nonnegative matrix factorization (NMF), and adopt the idea that two nodes with similar content will be most likely to belong to the same community to achieve the incorporation of links and node contents, i.e., we employ a graph regularization to penalize the dissimilarity of nodes denoted by community memberships. Besides, we introduce an intuitive manifold learning strategy to recover the intrinsic geometrical structure of the content information, i.e., K-near neighbor consistency. In addition, we found that, there are still drawbacks in this framework due to it does not consider the heterogeneous distribution of node degrees. This heterogeneous distribution can affect the function of graph regularization and isolates the original community memberships. We first proposed the node popularities satisfying the above interpretation and develop a new NMF-based model, named as Combination of Links and Node Contents for Community Discovery (CLNCCD). The experiments on both artificial and real-world networks compared with the state-of-the-art methods show that, the new model obtains significant improvement for community detection by incorporating node contents effectively.","Title: RoSeq: Robust Sequence Labeling.\nYear: 2020\n\nAbstract: In this paper, we mainly investigate two issues for sequence labeling, namely, label imbalance and noisy data that are commonly seen in the scenario of named entity recognition (NER) and are largely ignored in the existing works. To address these two issues, a new method termed robust sequence labeling (RoSeq) is proposed. Specifically, to handle the label imbalance issue, we first incorporate label statistics in a novel conditional random field (CRF) loss. In addition, we design an additional loss to reduce the weights of overwhelming easy tokens for augmenting the CRF loss. To address the noisy training data, we adopt an adversarial training strategy to improve model generalization. In experiments, the proposed RoSeq achieves the state-of-the-art performances on CoNLL and English Twitter NER—88.07% on CoNLL-2002 Dutch, 87.33% on CoNLL-2002 Spanish, 52.94% on WNUT-2016 Twitter, and 43.03% on WNUT-2017 Twitter without using the additional data.","Title: A Novel Generative Topic Embedding Model by Introducing Network Communities\nYear: 2019\n\nAbstract: Topic models have many important applications in fields such as Natural Language Processing. Topic embedding modelling aims at introducing word and topic embeddings into topic models to describe correlations between topics. Existing topic embedding methods use documents alone, which suffer from the topical fuzziness problem brought by the introduction of embeddings of semantic fuzzy words, e.g. polysemous words or some misleading academic terms. Links often exist between documents which form document networks. The use of links may alleviate this semantic fuzziness, but they are sparse and noisy which may meanwhile mislead topics. In this paper, we utilize community structure to solve these problems. It can not only alleviate the topical fuzziness of topic embeddings since communities are often believed to be topic related, but also can overcome the drawbacks brought by the sparsity and noise of networks (because community is a high-order network information). We give a new generative topic embedding model which incorporates documents (with topics) and network (with communities) together, and uses probability transition to describe the relationship between topics and communities to make it robust when topics and communities do not match. An efficient variational inference algorithm is then proposed to learn the model. We validate the superiority of our new approach on two tasks, document classifications and visualization of topic embeddings, respectively.","Title: Robust distributed cooperative RSS-based localization for directed graphs in mixed LoS/NLoS environments.\nYear: 2019\n\nAbstract: The accurate and low-cost localization of sensors using a wireless sensor network is critically required in a wide range of today’s applications. We propose a novel, robust maximum likelihood-type method for distributed cooperative received signal strength-based localization in wireless sensor networks. To cope with mixed LoS/NLoS conditions, we model the measurements using a two-component Gaussian mixture model. The relevant channel parameters, including the reference path loss, the path loss exponent, and the variance of the measurement error, for both LoS and NLoS conditions, are assumed to be unknown deterministic parameters and are adaptively estimated. Unlike existing algorithms, the proposed method naturally takes into account the (possible) asymmetry of links between nodes. The proposed approach has a communication overhead upper-bounded by a quadratic function of the number of nodes and computational complexity scaling linearly with it. The convergence of the proposed method is guaranteed for compatible network graphs, and compatibility can be tested a priori by restating the problem as a graph coloring problem. Simulation results, carried out in comparison to a centralized benchmark algorithm, demonstrate the good overall performance and high robustness in mixed LoS/NLoS environments.","Title: Unsupervised Text Style Transfer via Iterative Matching and Translation.\nYear: 2019\n\nAbstract: Text style transfer seeks to learn how to automatically rewrite sentences from a source domain to the target domain in different styles, while simultaneously preserving their semantic contents. A major challenge in this task stems from the lack of parallel data that connects the source and target styles. Existing approaches try to disentangle content and style, but this is quite difficult and often results in poor content-preservation and grammaticality. In contrast, we propose a novel approach by first constructing a pseudo-parallel resource that aligns a subset of sentences with similar content between source and target corpus. And then a standard sequence-to-sequence model can be applied to learn the style transfer. Subsequently, we iteratively refine the learned style transfer function while improving upon the imperfections in our original alignment. Our method is applied to the tasks of sentiment modification and formality transfer, where it outperforms state-of-the-art systems by a large margin. As an auxiliary contribution, we produced a publicly-available test set with human-generated style transfers for future community use.","Title: Publicly Available Clinical BERT Embeddings.\nYear: 2019\n\nAbstract: Contextual word embedding models such as ELMo (Peters et al., 2018) and BERT (Devlin et al., 2018) have dramatically improved performance for many natural language processing (NLP) tasks in recent months. However, these models have been minimally explored on specialty corpora, such as clinical text; moreover, in the clinical domain, no publicly-available pre-trained BERT models yet exist. In this work, we address this need by exploring and releasing BERT models for clinical text: one for generic clinical text and another for discharge summaries specifically. We demonstrate that using a domain-specific model yields performance improvements on three common clinical NLP tasks as compared to nonspecific embeddings. These domain-specific models are not as performant on two clinical de-identification tasks, and argue that this is a natural consequence of the differences between de-identified source text and synthetically non de-identified task text.","Title: Incorporating Network Embedding into Markov Random Field for Better Community Detection\nYear: 2019\n\nAbstract: Recent research on community detection focuses on learning representations of nodes using different network embedding methods, and then feeding them as normal features to clustering algorithms. However, we find that though one may have good results by direct clustering based on such network embedding features, there is ample room for improvement. More seriously, in many real networks, some statistically significant nodes which play pivotal roles are often divided into incorrect communities using network embedding methods. This is because while some distance measures are used to capture the spatial relationship between nodes by embedding, the nodes after mapping to feature vectors are essentially not coupled any more, losing important structural information. To address this problem, we propose a general Markov Random Field (MRF) framework to incorporate coupling in network embedding which allows better detecting network communities. By smartly utilizing properties of MRF, the new framework not only preserves the advantages of network embedding (e.g. low complexity, high parallelizability and applicability for traditional machine learning), but also alleviates its core drawback of inadequate representations of dependencies via making up the missing coupling relationships. Experiments on real networks show that our new approach improves the accuracy of existing embedding methods (e.g. Node2Vec, DeepWalk and MNMF), and corrects most wrongly-divided statistically-significant nodes, which makes network embedding essentially suitable for real community detection applications. The new approach also outperforms other state-of-the-art conventional community detection methods.","Title: The Image-Based Analysis And Classification Of Urine Sediments Using A Lenet-5 Neural Network\nYear: 2020\n\nAbstract: In this work, we presented a deep learning approach based on the LeNet-5 network for analysing and classifying recognisable shapes in urine sample images. The approach is based on shape analysis to recognise and classify red blood cells, white blood cells, epithelial cells and crystals observed under microscopes in urine samples. We modified the LeNet-5 neural network by changing the numbers of output nodes and convolutional layers. We compared the results of our method with those obtained by traditional feature extraction followed by back-propagation neural networks. Our testing showed that our method achieved a higher accuracy, sensitivity and specificity. The performance of our method demonstrated its broad applicability in urine sample analysis.","Title: Morphological Verb-Aware Tibetan Language Model.\nYear: 2019\n\nAbstract: The Tibetan language model (TLM) is the key to Tibetan natural language processing. In this paper, we first observe that, different from widely used languages, Tibetan contains many morphological verbs that rarely appear in natural sentences but play a key role in accurate text prediction. This property is usually ignored by existing methods and makes traditional training strategies less effective in constructing accurate and robust TLMs. Hence, we propose a morphological verb-aware TLM by offline learning via a character frequency reweighting strategy and online tuning of discriminative weights conditioned on morphological verbs. However, because of the influence of morphological verbs on the tense and semantics of sentences, it is necessary to consider the morphological verbs in Tibetan. As a result, compared with stateof-the-art methods, our method not only reduces the perplexity but also improves the character error on tasks of the text prediction and automatic speech recognition (ASR).","Title: Smart Roles: Inferring Professional Roles in Email Networks\nYear: 2019\n\nAbstract: Email is ubiquitous in the workplace. Naturally, machine learning models that make third-party email clients \"smarter\" can dramatically impact employees' productivity and efficiency. Motivated by this potential, we study the task of professional role inference from email data, which is crucial for email prioritization and contact recommendation systems. The central question we address is: Given limited data about employees, as is common in third-party email applications, can we infer where in the organizational hierarchy these employees belong based on their email behavior? Toward our goal, in this paper we study professional role inference on a unique new email dataset comprising billions of email exchanges across thousands of organizations. Taking a network approach in which nodes are employees and edges represent email communication, we propose EMBER, or EMBedding Email-based Roles, which finds email-centric embeddings of network nodes to be used in professional role inference tasks. EMBER automatically captures behavioral similarity between employees in the email network, leading to embeddings that naturally distinguish employees of different hierarchical roles. EMBER often outperforms the state-of-the-art by 2-20% in role inference accuracy and 2.5-344x in speed. We also use EMBER with our unique dataset to study how inferred professional roles compare between organizations of different sizes and sectors, gaining new insights into organizational hierarchy.","Title: Latent Network Summarization: Bridging Network Embedding and Summarization\nYear: 2019\n\nAbstract: Motivated by the computational and storage challenges that dense embeddings pose, we introduce the problem of latent network summarization that aims to learn a compact, latent representation of the graph structure with dimensionality that is independent of the input graph size (\\i.e., #nodes and #edges), while retaining the ability to derive node representations on the fly. We propose Multi-LENS, an inductive multi-level latent network summarization approach that leverages a set of relational operators and relational functions (compositions of operators) to capture the structure of egonets and higher-order subgraphs, respectively. The structure is stored in low-rank, size-independent structural feature matrices, which along with the relational functions comprise our latent network summary. Multi-LENS is general and naturally supports both homogeneous and heterogeneous graphs with or without directionality, weights, attributes or labels. Extensive experiments on real graphs show 3.5-34.3% improvement in AUC for link prediction, while requiring 80-2152x less output storage space than baseline embedding methods on large datasets. As application areas, we show the effectiveness of Multi-LENS in detecting anomalies and events in the Enron email communication graph and Twitter co-mention graph.","Title: Heterogeneous Graph Neural Network via Attribute Completion\nYear: 2021\n\nAbstract: ABSTRACT Heterogeneous information networks (HINs), also called heterogeneous graphs, are composed of multiple types of nodes and edges, and contain comprehensive information and rich semantics. Graph neural networks (GNNs), as powerful tools for graph data, have shown superior performance on network analysis. Recently, many excellent models have been proposed to process hetero-graph data using GNNs and have achieved great success. These GNN-based heterogeneous models can be interpreted as smooth node attributes guided by graph structure, which requires all nodes to have attributes. However, this is not easy to satisfy, as some types of nodes often have no attributes in heterogeneous graphs. Previous studies take some handcrafted methods to solve this problem, which separate the attribute completion from the graph learning process and, in turn, result in poor performance. In this paper, we hold that missing attributes can be acquired by a learnable manner, and propose a general framework for Heterogeneous Graph Neural Network via Attribute Completion (HGNN-AC), including pre-learning of topological embedding and attribute completion with attention mechanism. HGNN-AC first uses existing HIN-Embedding methods to obtain node topological embedding. Then it uses the topological relationship between nodes as guidance to complete attributes for no-attribute nodes by weighted aggregation of the attributes from these attributed nodes. Our complement mechanism can be easily combined with an arbitrary GNN-based heterogeneous model making the whole system end-to-end. We conduct extensive experiments on three real-world heterogeneous graphs. The results demonstrate the superiority of the proposed framework over state-of-the-art baselines.","Title: Can I Be Of Further Assistance? Using Unstructured Knowledge Access To Improve Task-Oriented Conversational Modeling\nYear: 2021\n\nAbstract: Most prior work on task-oriented dialogue systems are restricted to limited coverage of domain APIs. However, users oftentimes have requests that are out of the scope of these APIs. This work focuses on responding to these beyond-API-coverage user turns by incorporating external, unstructured knowledge sources. Our approach works in a pipelined manner with knowledge-seeking turn detection, knowledge selection, and response generation in sequence. We introduce novel data augmentation methods for the first two steps and demonstrate that the use of information extracted from dialogue context improves the knowledge selection and end-to-end performances. Through experiments, we achieve state-of-theart performance for both automatic and human evaluation metrics on the DSTC9 Track 1 benchmark dataset, validating the effectiveness of our contributions.","Title: \"HOW ROBUST R U?\": EVALUATING TASK-ORIENTED DIALOGUE SYSTEMS ON SPOKEN CONVERSATIONS\nYear: 2021\n\nAbstract: Most prior work in dialogue modeling has been on written conversations mostly because of existing data sets. However, written dialogues are not sufficient to fully capture the nature of spoken conversations as well as the potential speech recognition errors in practical spoken dialogue systems. This work presents a new benchmark on spoken task-oriented conversations, which is intended to study multi-domain dialogue state tracking and knowledge-grounded dialogue modeling. We report that the existing state-of-the-art models trained on written conversations are not performing well on our spoken data, as expected. Furthermore, we observe improvements in task performances when leveraging n-best speech recognition hypotheses such as by combining predictions based on individual hypotheses. Our data set enables speech-based bench-marking of task-oriented dialogue systems.","Title: TransFusion: Multi-Modal Fusion for Video Tag Inference via Translation-based Knowledge Embedding\nYear: 2021\n\nAbstract: ABSTRACTTag inference is an important task in the business of video platforms with wide applications such as recommendation, interpretation, and more. Existing works are mainly based on extracting video information from multiple modalities such as frames or music, and then infer tags through classification or object detection. This, however, does not apply to inferring generic tags or taxonomy that are less relevant to video contents, such as video originality or its broader category, which are important in practice. In this paper, we claim that these generic tags can be modeled through the semantic relations between videos and tags, and can be utilized simultaneously with the multi-modal features to achieve better video tagging. We propose TransFusion, an end-to-end supervised learning framework that fuses multi-modal embeddings (e.g., vision, audio, texts, etc.) with the knowledge embedding to derive the video representation. To infer the diverse tags following heterogeneous relations, TransFusion adopts a dual attentive approach to learn both the modality importance in fusion and relation importance in inference. Besides, it is general enough and can be used with the existing translation-based knowledge embedding approaches. Extensive experiments show that TransFusion outperforms the baseline methods with lowered mean rank and at least 9.59% improvement in [email protected] on the real-world video knowledge graph.","Title: Deep Transfer Learning for Multi-source Entity Linkage via Domain Adaptation\nYear: 2021\n\nAbstract: Multi-source entity linkage focuses on integrating knowledge from multiple sources by linking the records that represent the same real world entity. This is critical in high-impact applications such as data cleaning and user stitching. The state-of-the-art entity linkage pipelines mainly depend on supervised learning that requires abundant amounts of training data. However, collecting well-labeled training data becomes expensive when the data from many sources arrives incrementally over time. Moreover, the trained models can easily overfit to specific data sources, and thus fail to generalize to new sources due to significant differences in data and label distributions. To address these challenges, we present ADAMEL, a deep transfer learning framework that learns generic high-level knowledge to perform multi-source entity linkage. ADAMEL models the attribute importance that is used to match entities through an attribute-level self-attention mechanism, and leverages the massive unlabeled data from new data sources through domain adaptation to make it generic and data-source agnostic. In addition, ADAMEL is capable of incorporating an additional set of labeled data to more accurately integrate data sources with different attribute importance. Extensive experiments show that our framework achieves state-of-the-art results with 8.21% improvement on average over methods based on supervised learning. Besides, it is more stable in handling different sets of data sources in less runtime.","Title: Budget-constrained Truss Maximization over Large Graphs: A Component-based Approach\nYear: 2021\n\nAbstract: ABSTRACTCohesive substructure identification is one fundamental task of graph analytics. Recently, a useful problem of dense subgraph maximization has attracted significant attentions, which aims at enlarging a dense subgraph pattern using a few new edge insertions, e.g., k-core maximization. As a more cohesive subgraph of k-core, k-truss requires that each edge has at least k-2 triangles within this subgraph. However, the problem of k-truss maximization has not been studied yet. In this paper, we motivate and formulate a new problem of budget-constrained k-truss maximization. Given a budget of b edges and an integer k≥2, the problem is to find and insert b new edges into a graph G such that the resulted k-truss of G is maximized. We theoretically prove the NP-hardness of k-truss maximization problem. To efficiently tackle it, we analyze non-submodular property of k-truss newcomers function and develop non-conventional heuristic strategies for edge insertions. We first identify high-quality candidate edges with regard to (k-1)-light subgraphs and propose a greedy algorithm using per-edge insertion. Besides further improving the efficiency by pruning disqualified candidate edges, we finally develop a component-based dynamic programming algorithm for enlarging k-truss mostly, which makes a balance of budget assignment and inserts multiple edges simultaneously into all (k-1)-light components. Extensive experiments on nine real-world graphs demonstrate the efficiency and effectiveness of our proposed methods.","Title: Artificial-Intelligence-Enabled Reagent-Free Imaging Hematology Analyzer\nYear: 2021\n\nAbstract: Leukocyte differential test is a widely carried out clinical procedure for screening infectious diseases. Existing hematology analyzers require labor-intensive work and a panel of expensive reagents. Herein, an artificial-intelligence-enabled reagent-free imaging hematology analyzer (AIRFIHA) modality is reported that can accurately classify subpopulations of leukocytes with minimal sample preparation. AIRFIHA is realized through training a two-step residual neural network using label-free images of isolated leukocytes acquired from a custom-built quantitative phase microscope. By leveraging the rich information contained in quantitative phase images, not only high accuracy is achieved in differentiating B and T lymphocytes, but also CD4 and CD8 T cells are classified, therefore outperforming the classification accuracy of most current hematology analyzers. The performance of AIRFIHA in a randomly selected test set is validated and is cross-validated across all blood donors. Due to its easy operation, low cost, and accurate discerning capability of complex leukocyte subpopulations, AIRFIHA is clinically translatable and can also be deployed in resource-limited settings, e.g., during pandemic situations for the rapid screening of infectious diseases.","Title: Continuous-Variable Quantum Key Distribution Based on Heralded Hybrid Linear Amplifier with a Local Local Oscillator\nYear: 2021\n\nAbstract: An improved continuous variable quantum key distribution (CVQKD) approach based on a heralded hybrid linear amplifier (HLA) is proposed in this study, which includes an ideal deterministic linear amplifier and a probabilistic noiseless linear amplifier. The CVQKD, which is based on an amplifier, enhances the signal-to-noise ratio and provides for fine control between high gain and strong noise reduction. We focus on the impact of two types of optical amplifiers on system performance: phase sensitive amplifiers (PSA) and phase insensitive amplifiers (PIA). The results indicate that employing amplifiers, local local oscillation-based CVQKD systems can enhance key rates and communication distances. In addition, the PIA-based CVQKD system has a broader application than the PSA-based system.","Title: Universal Graph Convolutional Networks.\nYear: 2021\n\nAbstract: Use the \"Report an Issue\" link to request a name change.","Title: Sketching as a Tool for Understanding and Accelerating Self-attention for Long Sequences\nYear: 2022\n\nAbstract: Transformer-based models are not efficient in processing long sequences due to the quadratic space and time complexity of the self-attention modules. To address this limitation, Linformer and Informer are proposed to reduce the quadratic complexity to linear (modulo logarithmic factors) via low-dimensional projection and row selection respectively. These two models are intrinsically connected, and to understand their connection, we introduce a theoretical framework of matrix sketching. Based on the theoretical analysis, we propose Skeinformer to accelerate self-attention and further improve the accuracy of matrix approximation to self-attention with three carefully designed components: column sampling, adaptive row normalization and pilot sampling reutilization. Experiments on the Long Range Arena (LRA) benchmark demonstrate that our methods outperform alternatives with a consistently smaller time/space footprint.","Title: A Deep Dive Into Understanding The Random Walk-Based Temporal Graph Learning\nYear: 2021\n\nAbstract: Machine learning on graph data has gained significant interest because of its applicability to various domains ranging from product recommendations to drug discovery. While there is a rapid growth in the algorithmic community, the computer architecture community has so far focused on a subset of graph learning algorithms including Graph Convolution Network (GCN), and a few others. In this paper, w...","Title: AS-GCN: Adaptive Semantic Architecture of Graph Convolutional Networks for Text-Rich Networks\nYear: 2021\n\nAbstract: Graph Neural Networks (GNNs) have demonstrated great power in many network analytical tasks. However, graphs (i.e., networks) in the real world are usually text-rich, implying that valuable semantic information needs to be carefully considered. Existing GNNs for text-rich networks typically treat text as attribute words alone, which inevitably leads to the loss of important semantic structures, li...","Title: An Effective and Robust Framework by Modeling Correlations of Multiplex Network Embedding\nYear: 2021\n\nAbstract: The dependencies across different layers are an important property in multiplex networks and a few methods have been proposed to learn the dependencies in various ways. When capturing the dependencies across different layers, some of them assumed the structure among layers following consistent connectivity to force two nodes with a link in one layer tend to have links in other layers, some introduced a common vector to model the shared information across all layers. However, the correlations among layers in multiplex networks are diverse, which go beyond the connectivity consistency. In this paper, we propose a novel Modeling Correlations for Multiplex network Embedding (MCME) framework to learn the robust node representations for each layer. It can deal with complex correlations with a common structure, layer similarity and node heterogeneity through a unified framework in multiplex networks. To evaluate our proposed model, we conduct extensive experiments on several real-world datasets and the results demonstrate that our proposed model consistently outperforms state-of-the-art methods.","Title: Towards Textual Out-of-Domain Detection Without In-Domain Labels\nYear: 2022\n\nAbstract: In many real-world settings, machine learning models need to identify user inputs that are out-of-domain (OOD) so as to avoid performing wrong actions. This work focuses on a challenging case of OOD detection, where no labels for in-domain data are accessible (e.g., no intent labels for the intent classification task). To this end, we first evaluate different language model based approaches that predict likelihood for a sequence of tokens. Furthermore, we propose a novel representation learning based method by combining unsupervised clustering and contrastive learning so that better data representations for OOD detection can be learned. Through extensive experiments, we demonstrate that this method can significantly outperform likelihood-based methods and can be even competitive to the state-of-the-art supervised approaches with label information.","Title: Inflation Improves Graph Neural Networks\nYear: 2022\n\nAbstract: ABSTRACT Graph neural networks (GNNs) have gained significant success in graph representation learning and become the go-to approach for many graph-based tasks. Despite their effectiveness, the performance of GNNs is known to decline gradually as the number of layers increases. This attenuation is mainly caused by noise propagation, which refers to the useless or negative information propagated (directly or indirectly) from other nodes during the multi-layer graph convolution for node representation learning. This noise increases more severely as the layers of GNNs deepen, which is also a main reason of over-smoothing. In this paper, we propose a new convolution strategy for GNNs to address this problem via suppressing the noise propagation. Specifically, we first find that the feature propagation process of GNNs can be taken as a Markov chain. And then, based on the idea of Markov clustering, we introduce a new graph inflation layer (i.e., using a power function over the distribution) into GNNs to prevent noise propagating from local neighbourhoods to the whole graph with the increase of network layers. Our method is simple in design, which does not require any changes on the original basis and therefore can be easily extended. We conduct extensive experiments on real-world networks and have a stable improved performance as the network depth increases over existing GNNs.","Title: Graph Neural Network for Higher-Order Dependency Networks\nYear: 2022\n\nAbstract: ABSTRACT Graph neural network (GNN) has become a popular tool to analyze the graph data. Existing GNNs only focus on networks with first-order dependency, that is, conventional networks following the Markov property. However, many networks in real life own the higher-order dependency, such as click-stream data where the choice of the next page depends not only on the current page but also on previous pages. This kind of sequential data from complex systems (including natural dependencies) are often ignored by existing GNNs which makes them ineffective. To address this problem, we propose for the first time new GNN approaches for higher-order networks in this paper. First, we form sequence fragments by the current node and its predecessor nodes of different orders as candidate higher-order dependencies. When the fragment significantly affects the probability distribution of different successor nodes of the current node, we include it in the higher-order dependency set. We formulize the network with higher-order dependency as an augmented conventional first-order network, and then feed it into GNNs to derive network embeddings. Moreover, we further propose a new end-to-end GNN framework for dealing with higher-order networks directly in the model. Specifically, the higher-order dependency is used as the neighbor aggregation controller when the node is embedded and updated. In the graph convolutional layer, in addition to the first-order neighbor information, we also aggregate the middle node information from the higher-order dependency segment. We finally test the new approaches on three real networks with higher-order dependency, and compare with some state-of-the-art methods. The results show significant improvements of the new approaches which consider higher-order dependency.","Title: Empowering parameter-efficient transfer learning by recognizing the kernel structure in self-attention.\nYear: 2022\n\nAbstract: The massive amount of trainable parameters in the pre-trained language models (PLMs) makes them hard to be deployed to multiple downstream tasks. To address this issue, parameter-efficient transfer learning methods have been proposed to tune only a few parameters during fine-tuning while freezing the rest. This paper looks at existing methods along this line through the \\textit{kernel lens}. Motivated by the connection between self-attention in transformer-based PLMs and kernel learning, we propose \\textit{kernel-wise adapters}, namely \\textit{Kernel-mix}, that utilize the kernel structure in self-attention to guide the assignment of the tunable parameters. These adapters use guidelines found in classical kernel learning and enable separate parameter tuning for each attention head. Our empirical results, over a diverse set of natural language generation and understanding tasks, show that our proposed adapters can attain or improve the strong performance of existing baselines.","Title: On the Limits of Evaluating Embodied Agent Model Generalization Using Validation Sets\nYear: 2022\n\nAbstract: Natural language guided embodied task completion is a challenging problem since it requires understanding natural language instructions, aligning them with egocentric visual observations, and choosing appropriate actions to execute in the environment to produce desired changes. We experiment with augmenting a transformer model for this task with modules that effectively utilize a wider field of view and learn to choose whether the next step requires a navigation or manipulation action. We observed that the proposed modules resulted in improved, and in fact state-of-the-art performance on an unseen validation set of a popular benchmark dataset, ALFRED. However, our best model selected using the unseen validation set underperforms on the unseen test split of ALFRED, indicating that performance on the unseen validation set may not in itself be a sufficient indicator of whether model improvements generalize to unseen test sets. We highlight this result as we believe it may be a wider phenomenon in machine learning tasks but primarily noticeable only in benchmarks that limit evaluations on test splits, and highlights the need to modify benchmark design to better account for variance in model performance.","Title: DMIX: Adaptive Distance-aware Interpolative Mixup\nYear: 2022\n\nAbstract: Interpolation-based regularisation methods such as Mixup, which generate virtual training samples, have proven to be effective for various tasks and modalities. We extend Mixup and propose DMIX, an adaptive distanceaware interpolative Mixup that selects samples based on their diversity in the embedding space. DMIX leverages the hyperbolic space as a similarity measure among input samples for a richer encoded representation. DMIX achieves state-of-the-art results on sentence classification over existing data augmentation methods on 8 benchmark datasets across English, Arabic, Turkish, and Hindi languages while achieving benchmark F1 scores in 3 times less number of iterations. We probe the effectiveness of DMIX in conjunction with various similarity measures and qualitatively analyze the different components. DMIX being generalizable, can be applied to various tasks, models and modalities.","Title: CGMN: A Contrastive Graph Matching Network for Self-Supervised Graph Similarity Learning.\nYear: 2022\n\nAbstract: Graph similarity learning refers to calculating the similarity score between two graphs, which is required in many realistic applications, such as visual tracking, graph classification, and collaborative filtering. As most of the existing graph neural networks yield effective graph representations of a single graph, little effort has been made for jointly learning two graph representations and calculating their similarity score. In addition, existing unsupervised graph similarity learning methods are mainly clustering-based, which ignores the valuable information embodied in graph pairs. To this end, we propose a contrastive graph matching network (CGMN) for self-supervised graph similarity learning in order to calculate the similarity between any two input graph objects. Specifically, we generate two augmented views for each graph in a pair respectively. Then, we employ two strategies, namely cross-view interaction and cross-graph interaction, for effective node representation learning. The former is resorted to strengthen the consistency of node representations in two views. The latter is utilized to identify node differences between different graphs. Finally, we transform node representations into graph-level representations via pooling operations for graph similarity computation. We have evaluated CGMN on eight real-world datasets, and the experiment results show that the proposed new approach is superior to the state-of-the-art methods in graph similarity learning downstream tasks.","Title: Enhancing Knowledge Selection for Grounded Dialogues via Document Semantic Graphs\nYear: 2022\n\nAbstract: Providing conversation models with background knowledge has been shown to make open-domain dialogues more informative and engaging. Existing models treat knowledge selection as a sentence ranking or classification problem where each sentence is handled individually, ignoring the internal semantic connection among sentences in the background document. In this work, we propose to automatically convert the background knowledge documents into document semantic graphs and then perform knowledge selection over such graphs. Our document semantic graphs preserve sentence-level information through the use of sentence nodes and provide concept connections between sentences. We jointly apply multi-task learning for sentence-level and concept-level knowledge selection and show that it improves sentence-level selection. Our experiments show that our semantic graph-based knowledge selection improves over sentence selection baselines for both the knowledge selection task and the end-to-end response generation task on HollE and improves generalization on unseen topics in WoW.","Title: RAW-GNN: RAndom Walk Aggregation based Graph Neural Network.\nYear: 2022\n\nAbstract: Graph-Convolution-based methods have been successfully applied to representation learning on homophily graphs where nodes with the same label or similar attributes tend to connect with one another. Due to the homophily assumption of Graph Convolutional Networks (GCNs) that these methods use, they are not suitable for heterophily graphs where nodes with different labels or dissimilar attributes tend to be adjacent. Several methods have attempted to address this heterophily problem, but they do not change the fundamental aggregation mechanism of GCNs because they rely on summation operators to aggregate information from neighboring nodes, which is implicitly subject to the homophily assumption. Here, we introduce a novel aggregation mechanism and develop a RAndom Walk Aggregation-based Graph Neural Network (called RAW-GNN) method. The proposed approach integrates the random walk strategy with graph neural networks. The new method utilizes breadth-first random walk search to capture homophily information and depth-first search to collect heterophily information. It replaces the conventional neighborhoods with path-based neighborhoods and introduces a new path-based aggregator based on Recurrent Neural Networks. These designs make RAW-GNN suitable for both homophily and heterophily graphs. Extensive experimental results showed that the new method achieved state-of-the-art performance on a variety of homophily and heterophily graphs.","Title: Orbital angular momentum-encoded quantum digital signature over atmospheric channel\nYear: 2022\n\nAbstract: Quantum digital signature (QDS) ensures the integrity of a classical message and the authenticity of its sender based on information-theoretical limits and quantum mechanical mechanisms. The existing polarization-encoded or phase-encoded QDS protocols are susceptible to the interference from non-ideal factors in the quantum channel, resulting in the mismatch of the measurement reference frame. The channel capacities of these protocols are also limited, reducing the practicability of these protocols. To overcome these defects, we present a time-bin orbital angular momentum (OAM)-encoded continuous variable QDS (CVQDS) protocol, which exploits the rotation invariance of the vortex beam in the transmission direction to circumvent the real-time calibration of the reference frame and utilizes the infinite-dimensional eigenstate characteristics of the OAM states to beat the channel capacity limit. Security analysis and numerical simulations show that it is feasible to use this protocol to sign 1-bit message in atmospheric channel, and the channel capacity can be improved by increasing the dimensionality of the OAM code.","Title: Improving Bot Response Contradiction Detection via Utterance Rewriting.\nYear: 2022\n\nAbstract: Though chatbots based on large neural models can often produce fluent responses in open domain conversations, one salient error type is contradiction or inconsistency with the preceding conversation turns. Previous work has treated contradiction detection in bot responses as a task similar to natural language inference, e.g., detect the contradiction between a pair of bot utterances. However, utterances in conversations may contain co-references or ellipsis, and using these utterances as is may not always be sufficient for identifying contradictions. This work aims to improve the contradiction detection via rewriting all bot utterances to restore antecedents and ellipsis. We curated a new dataset for utterance rewriting and built a rewriting model on it. We empirically demonstrate that this model can produce satisfactory rewrites to make bot utterances more complete. Furthermore, using rewritten utterances improves contradiction detection performance significantly, e.g., the AUPR and joint accuracy scores (detecting contradiction along with evidence) increase by 6.5% and 4.5% (absolute increase), respectively.","Title: A Systematic Evaluation of Response Selection for Open Domain Dialogue.\nYear: 2022\n\nAbstract: Recent progress on neural approaches for language processing has triggered a resurgence of interest on building intelligent open-domain chatbots. However, even the state-of-the-art neural chatbots cannot produce satisfying responses for every turn in a dialog. A practical solution is to generate multiple response candidates for the same context, and then perform response ranking/selection to determine which candidate is the best. Previous work in response selection typically trains response rankers using synthetic data that is formed from existing dialogs by using a ground truth response as the single appropriate response and constructing inappropriate responses via random selection or using adversarial methods. In this work, we curated a dataset where responses from multiple response generators produced for the same dialog context are manually annotated as appropriate (positive) and inappropriate (negative). We argue that such training data better matches the actual use case examples, enabling the models to learn to rank responses effectively. With this new dataset, we conduct a systematic evaluation of state-of-the-art methods for response selection, and demonstrate that both strategies of using multiple positive candidates and using manually verified hard negative candidates can bring in significant performance improvement in comparison to using the adversarial training data, e.g., increase of 3% and 13% in Recall@1 score, respectively.","Title: Community Detection in Social Networks Considering Social Behaviors\nYear: 2022\n\nAbstract: The study of community detection in networks has drawn great attention in recent years. To find communities and to understand community semantics, both network topology and network content are utilized. Unfortunately, none of them can explain the driving factors of generating community structure with semantics, which is significant for understanding the mechanisms of community generation. Our observations on a large number of networks show that specific user social behaviors are underlying factors for the generation of community structure. We exploit four types of social behaviors that widely exist in networks, i.e., reciprocity of interactions, posting preference, multitopic preference, and temporal variation of topics. We investigate their impacts on the formation process of links and content in networks, during which communities with topics form. Our analysis shows that they are highly related to community structure. Consequently, a generative community detection model SBCD (social behavior-based community detection) is proposed by combining network topology and content, in which the above social behaviors play a core role. The model is evaluated on two real datasets. The experimental results show that SBCD outperforms state-of-the-art baselines. Finally, a case study illustrates several significant observations with respect to the proposed social behaviors."],"string":"[\n \"Title: Decentralized mining social network communities with agents.\\nYear: 2013\\n\\nAbstract: Network community mining algorithms aim at efficiently and effectively discovering all such communities from a given network. Many related methods have been proposed and applied to different areas including social network analysis, gene network analysis and web clustering engines. Most of the existing methods for mining communities are centralized. In this paper, we present a multi-agent based decentralized algorithm, in which a group of autonomous agents work together to mine a network through a proposed self-aggregation and self-organization mechanism. Thanks to its decentralized feature, our method is potentially suitable for dealing with distributed networks, whose global structures are hard to obtain due to their geographical distributions, decentralized controls or huge sizes. The effectiveness of our method has been tested against different benchmark networks.\",\n \"Title: Ant colony optimization for community detection in large-scale complex networks\\nYear: 2011\\n\\nAbstract: In this paper we present a new ant colony optimization for community detection in large networks, which takes modularity Q as objective function. An important difference that distinguishes our algorithm from the former ant algorithms is the manner in which the ants are used in the algorithm. Unlike those existing methods in which each ant searches for a candidate solution, each ant in our algorithm only decides whether its current vertex joins the community of its previous vertex with the aid of a simulated annealing idea, whose purpose is to locally optimize function Q. In each iteration, the ants work collectively so as to uncover the community structure of the network. Moreover, we introduce a thought of “layer and rule” into this method for further improving its performance. Our algorithm doesn't employ the pheromone, which reduces its running time and makes it well suitable for large-scale networks. Meanwhile, it still performs very well on both computer-generated benchmark and some widely used real-world networks compared with a set of competing algorithm in terms of clustering quality.\",\n \"Title: Genetic Algorithm With A Local Search Strategy For Discovering Communities In Complex Networks\\nYear: 2013\\n\\nAbstract: In order to further improve the performance of current genetic algorithms aiming at discovering communities, a local search based genetic algorithm (GALS) is here proposed. The core of GALS is a local search based mutation technique. In order to overcome the drawbacks of traditional mutation methods, the paper develops the concept of marginal gene and then the local monotonicity of modularity function Q is deduced from each node's local view. Based on these two elements, a new mutation method combined with a local search strategy is presented. GALS has been evaluated on both synthetic benchmarks and several real networks, and compared with some presently competing algorithms. Experimental results show that GALS is highly effective and efficient for discovering community structure.\",\n \"Title: Identification Of Hybrid Node And Link Communities In Complex Networks\\nYear: 2013\\n\\nAbstract: Identifying communities in complex networks is an effective means for analyzing complex systems, with applications in diverse areas such as social science, engineering, biology and medicine. Finding communities of nodes and finding communities of links are two popular schemes for network analysis. These schemes, however, have inherent drawbacks and are inadequate to capture complex organizational structures in real networks. We introduce a new scheme and an effective approach for identifying complex mixture structures of node and link communities, called hybrid node-link communities. A central piece of our approach is a probabilistic model that accommodates node, link and hybrid node-link communities. Our extensive experiments on various real-world networks, including a large protein-protein interaction network and a large network of semantically associated words, illustrated that the scheme for hybrid communities is superior in revealing network characteristics. Moreover, the new approach outperformed the existing methods for finding node or link communities separately.\",\n \"Title: The (un)supervised detection of overlapping communities as well as hubs and outliers via (bayesian) NMF\\nYear: 2014\\n\\nAbstract: The detection of communities in various networks has been considered by many researchers. Moreover, it is preferable for a community detection method to detect hubs and outliers as well. This becomes even more interesting and challenging when taking the unsupervised assumption, that is, we do not assume the prior knowledge of the number K of communities. In this poster, we define a novel model to identify overlapping communities as well as hubs and outliers. When K is given, we propose a normalized symmetric nonnegative matrix factorization algorithm to learn the parameters of the model. Otherwise, we introduce a Bayesian symmetric nonnegative matrix factorization to learn the parameters of the model, while determining K. Our experiment indicate its superior performance on various networks.\",\n \"Title: Modeling and simulation of performance analysis for a cluster-based Web server\\nYear: 2006\\n\\nAbstract: Higher scalability and availability of Web servers are required as the traffic on the Internet has been increasing dramatically over the last few years. This paper focuses on modeling and simulation of the performance analysis for the test system of a cluster-based Web server consisting of five real servers. Three ways of load balancing are introduced, namely, network address translation, IP tunneling and direct routing. Calculation of packet delay is discussed for each part of the running system according to the transferred data in the system. The input model and the system model to measure and simulate the system performance are derived, and the probability distributions of the delay data are specified for using random inputs in the model according to the Q–Q plot and the cumulative distribution function. After the solution of performance tuning problems is evaluated, the maximum process capability of the system is found and a possible performance bottleneck is analyzed.\",\n \"Title: Discovering link communities in complex networks by exploiting link dynamics\\nYear: 2013\\n\\nAbstract: Discovery of communities in complex networks is a fundamental data analysis problem with applications in various domains. Most of the existing approaches have focused on discovering communities of nodes, while recent studies have shown great advantages and utilities of the knowledge of communities of links in networks. From this new perspective, we propose a link dynamics based algorithm, called UELC, for identifying link communities of networks. In UELC, the stochastic process of a link-node-link random walk is employed to unfold an embedded bipartition structure of links in a network. The local mixing properties of the Markov chain underlying the random walk are then utilized to extract two emerging link communities. Further, the random walk and the bipartitioning processes are wrapped in an iterative subdivision strategy to recursively identify link partitions that segregate the network links into multiple subdivisions. We evaluate the performance of the new method on synthetic benchmarks and demonstrate its utility on real-world networks. Our experimental results show that our method is highly effective for discovering link communities in complex networks. As a comparison, we also extend UELC to extracting communities of nodes, and show that it is effective for node community identification.\",\n \"Title: Genetic Algorithm with Local Search for Community Mining in Complex Networks\\nYear: 2010\\n\\nAbstract: Detecting communities from complex networks has triggered considerable attention in several application domains. Targeting this problem, a local search based genetic algorithm (GALS) which employs a graph-based representation (LAR) has been proposed in this work. The core of the GALS is a local search based mutation technique. Aiming to overcome the drawbacks of the existing mutation methods, a concept called marginal gene has been proposed, and then an effective and efficient mutation method, combined with a local search strategy which is based on the concept of marginal gene, has also been proposed by analyzing the modularity function. Moreover, in this paper the percolation theory on ER random graphs is employed to further clarify the effectiveness of LAR presentation; A Markov random walk based method is adopted to produce an accurate and diverse initial population; the solution space of GALS will be significantly reduced by using a graph based mechanism. The proposed GALS has been tested on both computer-generated and real-world networks, and compared with some competitive community mining algorithms. Experimental result has shown that GALS is highly effective and efficient for discovering community structure.\",\n \"Title: Ant colony optimization with Markov random walk for community detection in graphs\\nYear: 2011\\n\\nAbstract: Network clustering problem (NCP) is the problem associated to the detection of network community structures. Building on Markov random walks we address this problem with a new ant colony optimization strategy, named as ACOMRW, which improves prior results on the NCP problem and does not require knowledge of the number of communities present on a given network. The framework of ant colony optimization is taken as the basic framework in the ACOMRWalgorithm. At each iteration, a Markov random walk model is taken as heuristic rule; all of the ants' local solutions are aggregated to a global one through clustering ensemble, which then will be used to update a pheromone matrix. The strategy relies on the progressive strengthening of within-community links and the weakening of between-community links. Gradually this converges to a solution where the underlying community structure of the complex network will become clearly visible. The performance of algorithm ACOMRW was tested on a set of benchmark computer-generated networks, and as well on real-world network data sets. Experimental results confirm the validity and improvements met by this approach.\",\n \"Title: An Ant-Based Algorithm with Local Optimization for Community Detection in Large-Scale Networks\\nYear: 2012\\n\\nAbstract: In this paper, we propose a multi-layer ant-based algorithm (MABA), which detects communities from networks by means of locally optimizing modularity using individual ants. The basic version of MABA, namely SABA, combines a self-avoiding label propagation technique with a simulated annealing strategy for ant diffusion in networks. Once the communities are found by SABA, this method can be reapplied to a higher level network where each obtained community is regarded as a new vertex. The aforementioned process is repeated iteratively, and this corresponds to MABA. Thanks to the intrinsic multi-level nature of our algorithm, it possesses the potential ability to unfold multi-scale hierarchical structures. Furthermore, MABA has the ability that mitigates the resolution limit of modularity. The proposed MABA has been evaluated on both computer-generated benchmarks and widely used real-world networks, and has been compared with a set of competitive algorithms. Experimental results demonstrate that MABA is both effective and efficient (in near linear time with respect to the size of network) for discovering communities.\",\n \"Title: Key Distribution Using Dual Quantum Channels\\nYear: 2008\\n\\nAbstract: Quantum key distribution has been a relatively recent addition to data security and integrity. It is based on advanced ideas that emanate from Quantum theory. Additionally, the optical network and the quantum photon provide a formidable basis for the application of quantum key distribution. We propose a new way of distributing encryption keys over quantum channels. The proposed method takes advantage of dual quantum channels between the communicating entities and complementary measuring bases in order to improve the efficiency with which the keys can be exchanged.\",\n \"Title: Fast convergent key distribution algorithms using a dual quantum channel\\nYear: 2009\\n\\nAbstract: This paper presents three new algorithms over a dual quantum channel to enhance the convergence speed of the key distribution process. The proposed algorithms effectively use a dual quantum channel between the communicating entities and, when combined with the proposed selection of the basis, improve the efficiency of the key distribution process as well as deter Eve. With the use of an initialization vector (IV), the proposed algorithm does not necessitate information exchange about the bases between Alice and Bob, thus removing the requirement of the availability of a two-way conventional channel. Simultaneously, it allows the detection of the presence of Eve in real time. Copyright (C) 2008 John Wiley & Sons, Ltd.\",\n \"Title: Ant Colony Optimization with a New Random Walk Model for Community Detection in Complex Networks\\nYear: 2013\\n\\nAbstract: Detecting communities from complex networks has recently triggered great interest. Aiming at this problem, a new ant colony optimization strategy building on the Markov random walks theory, which is named as MACO, is proposed in this paper. The framework of ant colony optimization is taken as the basic framework in this algorithm. In each iteration, a Markov random walk model is employed as heuristic rule; all of the ants' local solutions are aggregated to a global one through an idea of clustering ensemble, which then will be used to update a pheromone matrix. The strategy relies on the progressive strengthening of within-community links and the weakening of between-community links. Gradually this converges to a solution where the underlying community structure of the complex network will become clearly visible. The proposed MACO has been evaluated both on synthetic benchmarks and on some real-world networks, and compared with some present competing algorithms. Experimental result has shown that MACO is highly effective for discovering communities.\",\n \"Title: Fast Complex Network Clustering Algorithm Using Agents\\nYear: 2013\\n\\nAbstract: Recently, the sizes of networks are always very huge, and they take on distributed nature. Aiming at this kind of network clustering problem, in the sight of local view, this paper proposes a fast network clustering algorithm in which each node is regarded as an agent, and each agent tries to maximize its local function in order to optimize network modularity defined by function Q, rather than optimize function Q from the global view as traditional methods. Both the efficiency and effectiveness of this algorithm are tested against computer-generated and real-world networks. Experimental result shows that this algorithm not only has the ability of clustering large-scale networks, but also can attain very good clustering quality compared with the existing algorithms. Furthermore, the parameters of this algorithm are analyzed.\",\n \"Title: A Markov random walk under constraint for discovering overlapping communities in complex networks\\nYear: 2013\\n\\nAbstract: The detection of overlapping communities in complex networks has motivated recent research in relevant fields. Aiming to address this problem, we propose a Markov-dynamics-based algorithm, called UEOC, which means 'unfold and extract overlapping communities'. In UEOC, when identifying each natural community that overlaps, a Markov random walk method combined with a constraint strategy, which is based on the corresponding annealed network (degree conserving random network), is performed to unfold the community. Then, a cutoff criterion with the aid of a local community function, called conductance, which can be thought of as the ratio between the number of edges inside the community and those leaving it, is presented to extract this emerged community from the entire network. The UEOC algorithm depends on only one parameter whose value can be easily set, and it requires no prior knowledge of the hidden community structures. The proposed UEOC has been evaluated both on synthetic benchmarks and on some real-world networks, and has been compared with a set of competing algorithms. The experimental result has shown that UEOC is highly effective and efficient for discovering overlapping communities.\",\n \"Title: Cooperative Localization in WSNs Using Gaussian Mixture Modeling: Distributed ECM Algorithms\\nYear: 2015\\n\\nAbstract: We study cooperative sensor network localization in a realistic scenario where the underlying measurement errors more probably follow a non-Gaussian distribution; the measurement error distribution is unknown without conducting massive offline calibrations; and non-line-of-sight identification is not performed due to the complexity constraint and/or storage limitation. The underlying measurement error distribution is approximated parametrically by a Gaussian mixture with finite number of components, and the expectation-conditional maximization (ECM) criterion is adopted to approximate the maximum-likelihood estimator of the unknown sensor positions and an extra set of Gaussian mixture model parameters. The resulting centralized ECM algorithms lead to easier inference tasks and meanwhile retain several convergence properties with a proof of the “space filling” condition. To meet the scalability requirement, we further develop two distributed ECM algorithms where an average consensus algorithm plays an important role for updating the Gaussian mixture model parameters locally. The proposed algorithms are analyzed systematically in terms of computational complexity and communication overhead. Various computer based tests are also conducted with both simulation and experimental data. The results pin down that the proposed distributed algorithms can provide overall good performance for the assumed scenario even under model mismatch, while the existing competing algorithms either cannot work without the prior knowledge of the measurement error statistics or merely provide degraded localization performance when the measurement error is clearly non-Gaussian.\",\n \"Title: A Unified Semi-Supervised Community Detection Framework Using Latent Space Graph Regularization.\\nYear: 2015\\n\\nAbstract: Community structure is one of the most important properties of complex networks and is a foundational concept in exploring and understanding networks. In real world, topology information alone is often inadequate to accurately find community structure due to its sparsity and noises. However, potential useful prior information can be obtained from domain knowledge in many applications. Thus, how to improve the community detection performance by combining network topology with prior information becomes an interesting and challenging problem. Previous efforts on utilizing such priors are either dedicated or insufficient. In this paper, we firstly present a unified interpretation to a group of existing community detection methods. And then based on this interpretation, we propose a unified semi-supervised framework to integrate network topology with prior information for community detection. If the prior information indicates that some nodes belong to the same community, we encode it by adding a graph regularization term to penalize the latent space dissimilarity of these nodes. This framework can be applied to many widely-used matrix-based community detection methods satisfying our interpretation, such as nonnegative matrix factorization, spectral clustering, and their variants. Extensive experiments on both synthetic and real networks show that the proposed framework significantly improves the accuracy of community detection, especially on networks with unclear structures.\",\n \"Title: Security in Computing and Communications: Second International Symposium, SSCC 2014, Delhi, India, September 24-27, 2014. Proceedings\\nYear: 2014\\n\\nAbstract: This book constitutes the refereed proceedings of the International Symposium on Security in Computing and Communications, SSCC 2014, held in Delhi, India, in September 2013. The 36 revised full papers presented together with 12 work-in-progress papers were carefully reviewed and selected from 132 submissions. The papers are organized in topical sections on security and privacy in networked systems; authentication and access control systems; encryption and cryptography; system and network security; work-in-progress.\",\n \"Title: A Stochastic Model For Detecting Heterogeneous Link Communities In Complex Networks\\nYear: 2015\\n\\nAbstract: Discovery of communities in networks is a fundamental data analysis problem. Most of the existing approaches have focused on discovering communities of nodes, while recent studies have shown great advantages and utilities of the knowledge of communities of links. Stochastic models provides a promising class of techniques for the identification of modular structures, but most stochastic models mainly focus on the detection of node communities rather than link communities. We propose a stochastic model, which not only describes the structure of link communities, but also considers the heterogeneous distribution of community sizes, a property which is often ignored by other models. We then learn the model parameters using a method of maximum likelihood based on an expectation-maximization algorithm. To deal with large complex real networks, we extend the method by a strategy of iterative bipartition. The extended method is not only efficient, but is also able to determine the number of communities for a given network. We test our approach on both synthetic benchmarks and real-world networks including an application to a large biological network, and also compare it with two existing methods. The results demonstrate the superior performance of our approach over the competing methods for detecting link communities.\",\n \"Title: Modeling With Node Degree Preservation Can Accurately Find Communities\\nYear: 2015\\n\\nAbstract: An important problem in analyzing complex networks is discovery of modular or community structures embedded in the networks. Although being promising for identifying network communities, the popular stochastic models often do not preserve node degrees, thus reducing their representation power and applicability to real-world networks. Here we address this critical problem. Instead of using a block-model, we adopted a random-graph null model to faithfully capture community structures by preserving in the model the expected node degrees. The new model, learned using nonnegative matrix factorization, is more accurate and robust in representing community structures than the existing methods. Our results from extensive experiments on synthetic benchmarks and real-world networks show the superior performance of the new method over the existing methods in detecting both disjoint and overlapping communities.\",\n \"Title: Dialog Act classification in Chinese spoken language\\nYear: 2013\\n\\nAbstract: Dialog Act (DA) is an important pragmatics feature for us to understand speakers' intention. Many methods have been proposed to recognize DA tags. However, little work has been conducted to address the problem of DA tagging in Chinese spoken dialog language. In this work, we employ both the lexical features and the inter-utterance dependency features for DA tagging. And we propose three different methods: n-gram, extended HMM and n-gram+KNN. The experimental results show that these methods are effective for the task.\",\n \"Title: Perseus: an interactive large-scale graph mining and visualization tool\\nYear: 2015\\n\\nAbstract: Given a large graph with several millions or billions of nodes and edges, such as a social network, how can we explore it efficiently and find out what is in the data? In this demo we present Perseus, a large-scale system that enables the comprehensive analysis of large graphs by supporting the coupled summarization of graph properties and structures, guiding attention to outliers, and allowing the user to interactively explore normal and anomalous node behaviors. Specifically, Perseus provides for the following operations: 1) It automatically extracts graph invariants (e.g., degree, PageRank, real eigenvectors) by performing scalable, offline batch processing on Hadoop; 2) It interactively visualizes univariate and bivariate distributions for those invariants; 3) It summarizes the properties of the nodes that the user selects; 4) It efficiently visualizes the induced subgraph of a selected node and its neighbors, by incrementally revealing its neighbors. In our demonstration, we invite the audience to interact with Perseus to explore a variety of multi-million-edge social networks including a Wikipedia vote network, a friendship/foeship network in Slashdot, and a trust network based on the consumer review website Epinions.com.\",\n \"Title: Community Detection With Manifold Learning On Speaker I-Vector Space For Chinese\\nYear: 2015\\n\\nAbstract: Speaker recognition with clustering speech signals of the same speaker is an important speech analysis task in various applications. Recent works have shown that there was an underlying manifold on which speaker utterances live in the model-parameter space. However, most speaker clustering methods work on the Euclidean space, and hence often fail to discover the intrinsic geometrical structure of the data space. For this problem, we consider to convert the speaker i-vector representation of utterances in the Euclidean space into a network structure constructed based on the local (k) nearest neighbor relationship of these signals. We then propose a community detection model on the network for clustering signals. The new model is based on the probabilistic community memberships, and is further refined with the idea that: if two connected nodes have a high similarity, their community membership distributions in the model should be made close. This refinement enhances the local invariance assumption, and thus better respects the structure of the underlying manifold than the existing community detection methods. Some experiments are conducted on speaker content network built from a Chinese speaker recognition database. The results confirmed the effectiveness of this new method.\",\n \"Title: Efficient cooperative localization algorithm in LOS/NLOS environments\\nYear: 2015\\n\\nAbstract: The well-known cooperative localization algorithm, 'sum-product algorithm over a wireless network' (SPAWN) has two major shortcomings, a relatively high computational complexity and a large communication load. Using the Gaussian mixture model with a model selection criterion and the sigma-point (SP) methods, we propose the SPAWN-SP to overcome these problems. The SPAWN-SP easily accommodates different localization scenarios due to its high flexibility in message representation. Furthermore, harsh LOS/NLOS environments are considered for the evaluation of cooperative localization algorithms. Our simulation results indicate that the proposed SPAWN-SP demonstrates high localization accuracy in different localization scenarios, thanks to its high flexibility in message representation.\",\n \"Title: Dithering in quantized RSS based localization\\nYear: 2015\\n\\nAbstract: We study maximum likelihood (ML) position estimation using quantized received signal strength measurements. In order to mitigate the undesired quantization effect in the observations, the dithering technique is adopted. Various dither noise distributions are considered and the corresponding likelihood functions are derived. Simulation results show that the proposed ML estimator with dithering is able to generate a significantly reduced bias but a modestly increased mean-square-error as compared to the conventional ML estimator without dithering.\",\n \"Title: Semantic Community Identification in Large Attribute Networks.\\nYear: 2016\\n\\nAbstract: Identification of modular or community structures of a network is a key to understanding the semantics and functions of the network. While many network community detection methods have been developed, which primarily explore network topologies, they provide little semantic information of the communities discovered. Although structures and semantics are closely related, little effort has been made to discover and analyze these two essential network properties together. By integrating network topology and semantic information on nodes, e.g., node attributes, we study the problems of detection of communities and inference of their semantics simultaneously. We propose a novel nonnegative matrix factorization (NMF) model with two sets of parameters, the community membership matrix and community attribute matrix, and present efficient updating rules to evaluate the parameters with a convergence guarantee. The use of node attributes improves upon community detection and provides a semantic interpretation to the resultant network communities. Extensive experimental results on synthetic and real-world networks not only show the superior performance of the new method over the state-of-the-art approaches, but also demonstrate its ability to semantically annotate the communities.\",\n \"Title: Detect Overlapping Communities via Ranking Node Popularities.\\nYear: 2016\\n\\nAbstract: Detection of overlapping communities has drawn much attention lately as they are essential properties of real complex networks. Despite its influence and popularity, the well studied and widely adopted stochastic model has not been made effective for finding overlapping communities. Here we extend the stochastic model method to detection of overlapping communities with the virtue of autonomous determination of the number of communities. Our approach hinges upon the idea of ranking node popularities within communities and using a Bayesian method to shrink communities to optimize an objective function based on the stochastic generative model. We evaluated the novel approach, showing its superior performance over five state-of-the-art methods, on large real networks and synthetic networks with ground-truths of overlapping communities.\",\n \"Title: TeleLink: Link Prediction in Social Network Based on Multiplex Cohesive Structures.\\nYear: 2016\\n\\nAbstract: Given a network where the same set of nodes have multiple types of relationships, how do we efficiently predict potential links in the future (e.g., interactions between social actors), and how do we predict links using information from other relationships? These problems have been widely studied recently, most of the existing methods either aggregate multiple types of relationships into a single network or consider them separately and ignore the correlations across relationships, leading to information loss. In this work, we present TeleLink, a general link prediction model that works for networks with single and multiple relationships. TeleLink predicts potential links based on community detection and improves link prediction by bringing in a cohesive structure across multiple networks constructed by different relationships or node attributes. To further improve the prediction performance, we extend TeleLink to a semi-supervised scheme, incorporating partially labeled information. Our extensive experiments show that TeleLink outperforms existing methods in predicting new links. Specifically, among the various datasets that we study, TeleLink achieves a precision improvement by up to 110 % compared to the baselines.\",\n \"Title: Semi-supervised community detection based on non-negative matrix factorization with node popularity.\\nYear: 2017\\n\\nAbstract: Based on the ideas of graph regularization, we present a semi-supervised and NMF-based model to utilize the prior information of networks.By introducing the parameters of node popularities, we propose a refined PSSNMF model, which is particularly suitable for networks with large degree heterogeneity and unbalanced community structure.The prior information with node popularity being introduced into the model is more effective than being directly encoded into the adjacent matrix. A plethora of exhaustive studies have proved that the community detection merely based on topological information often leads to relatively low accuracy. Several approaches aim to achieve performance improvement by utilizing the background information. But they ignore the effect of node degrees on the availability of prior information. In this paper, by combining the idea of graph regularization with the pairwise constraints, we present a semi-supervised non-negative matrix factorization (SSNMF) model for community detection. And then, to alleviate the influence of the heterogeneity of node degrees and community sizes, we propose an improved SSNMF model by introducing the node popularity, namely PSSNMF, which helps to utilize the prior information more effectively. At last, the extensive experiments on both artificial and real-world networks show that the proposed method improves, as expected, the accuracy of community detection, especially on networks with large degree heterogeneity and unbalanced community structure.\",\n \"Title: Fast and accurate cooperative localization in wireless sensor networks.\\nYear: 2016\\n\\nAbstract: Cooperative localization capability is a highly desirable characteristic of wireless sensor networks. It has attracted considerable research attention in academia and industry. The sumproduct algorithm over a wireless sensor network (SPAWN) is a powerful method to cooperatively estimate the positions of many sensors (agents) using knowledge of the absolute positions of a few sensors (anchors). Drawbacks of the SPAWN, however, are its high computational complexity and communication load. In this paper we address the complexity issue, reformulate it as convolution problem and utilize the fast Fourier transform (FFT), culminating in a fast and accurate localization algorithm, which we named SPAWN-FFT. Our simulation results show SPAWN-FFT's superiority over SPAWN regarding the computational effort, while maintaining its full flexibility and localization performance.\",\n \"Title: Joint Identification of Network Communities and Semantics via Integrative Modeling of Network Topologies and Node Contents.\\nYear: 2017\\n\\nAbstract: The objective of discovering network communities, an essential step in complex systems analysis, is two-fold: identification of functional modules and their semantics at the same time. However, most existing community-finding methods have focused on finding communities using network topologies, and the problem of extracting module semantics has not been well studied and node contents, which often contain semantic information of nodes and networks, have not been fully utilized. We considered the problem of identifying network communities and module semantics at the same time. We introduced a novel generative model with two closely correlated parts, one for communities and the other for semantics. We developed a co-learning strategy to jointly train the two parts of the model by combining a nested EM algorithm and belief propagation. By extracting the latent correlation between the two parts, our new method is not only robust for finding communities and semantics, but also able to provide more than one semantic explanation to a community. We evaluated the new method on artificial benchmarks and analyzed the semantic interpretability by a case study. We compared the new method with eight state-of-the-art methods on ten real-world networks, showing its superior performance over the existing methods.\",\n \"Title: Leveraging Behavioral And Social Information For Weakly Supervised Collective Classification Of Political Discourse On Twitter\\nYear: 2017\\n\\nAbstract: Framing is a political strategy in which politicians carefully word their statements in order to control public perception of issues. Previous works exploring political framing typically analyze frame usage in longer texts, such as congressional speeches. We present a collection of weakly supervised models which harness collective classification to predict the frames used in political discourse on the microblogging platform, Twitter. Our global probabilistic models show that by combining both lexical features of tweets and network-based behavioral features of Twitter, we are able to increase the average, unsupervised F-1 score by 21.52 points over a lexical baseline alone.\",\n \"Title: Adaptive Community Detection Incorporating Topology and Content in Social Networks.\\nYear: 2018\\n\\nAbstract: In social network analysis, community detection is a basic step to understand the structure, function and semantics of networks. Some conventional community detection methods may have limited performance because they merely focus on topological structure of networks. In addition to topology, content information is another significant aspect of social networks. Some state-of-the-art methods started to combine these two aspects of information, but they often assume that topology and content share the same characteristics. However, for some examples of social networks, content may mismatch with topological structure. In order to better cope with such situations, we introduce a novel community detection method under the framework of non-negative matrix factorization (NMF). Our proposed method integrates topology and content of networks, and introduces a novel adaptive parameter for controlling the contribution of content with respect to the identified mismatch degree between the topological and content information. The case study using real social networks show that our new method can simultaneously obtain community partition and the corresponding semantic descriptions. Experiments on both artificial networks and real social networks further indicate that our method outperforms some state-of-the-art methods while exhibiting more robust behaviour when the mismatch topological and content information is observed.\",\n \"Title: Perseus-Hub: Interactive And Collective Exploration Of Large-Scale Graphs\\nYear: 2017\\n\\nAbstract: Graphs emerge naturally in many domains, such as social science, neuroscience, transportation engineering, and more. In many cases, such graphs have millions or billions of nodes and edges, and their sizes increase daily at a fast pace. How can researchers from various domains explore large graphs interactively and efficiently to find out what is 'important'? How can multiple researchers explore a new graph dataset collectively and \\\"help\\\" each other with their findings? In this article, we present PERSEUS-HUB, a large-scale graph mining tool that computes a set of graph properties in a distributed manner, performs ensemble, multi-view anomaly detection to highlight regions that are worth investigating, and provides users with uncluttered visualization and easy interaction with complex graph statistics. PERSEUS-HUB uses a Spark cluster to calculate various statistics of large-scale graphs efficiently, and aggregates the results in a summary on the master node to support interactive user exploration. In PERSEUS-HUB, the visualized distributions of graph statistics provide preliminary analysis to understand a graph. To perform a deeper analysis, users with little prior knowledge can leverage patterns (e.g., spikes in the power-law degree distribution) marked by other users or experts. Moreover, PERSEUS-HUB guides users to regions of interest by highlighting anomalous nodes and helps users establish a more comprehensive understanding about the graph at hand. We demonstrate our system through the case study on real, large-scale networks.\",\n \"Title: Exploratory Analysis of Graph Data by Leveraging Domain Knowledge\\nYear: 2017\\n\\nAbstract: Given the soaring amount of data being generated daily, graph mining tasks are becoming increasingly challenging, leading to tremendous demand for summarization techniques. Feature selection is a representative approach that simplifies a dataset by choosing features that are relevant to a specific task, such as classification, prediction, and anomaly detection. Although it can be viewed as a way to summarize a graph in terms of a few features, it is not well-defined for exploratory analysis, and it operates on a set of observations jointly rather than conditionally (i.e., feature selection from many graphs vs. selection for an input graph conditioned on other graphs). In this work, we introduce EAGLE (Exploratory Analysis of Graphs with domain knowLEdge), a novel method that creates interpretable, feature-based, and domain-specific graph summaries in a fully automatic way. That is, the same graph in different domains-e.g., social science and neuroscience-will be described via different EAGLE summaries, which automatically leverage the domain knowledge and expectations. We propose an optimization formulation that seeks to find an interpretable summary with the most representative features for the input graph so that it is: diverse, concise, domain-specific, and efficient. Extensive experiments on synthetic and real-world datasets with up to ~1M edges and ~400 features demonstrate the effectiveness and efficiency of EAGLE and its benefits over existing methods. We also show how our method can be applied to various graph mining tasks, such as classification and exploratory analysis.\",\n \"Title: High-throughput, high-resolution Generated Adversarial Network Microscopy.\\nYear: 2018\\n\\nAbstract: We for the first time combine generated adversarial network (GAN) with wide-field light microscopy to achieve deep learning super-resolution under a large field of view (FOV). By appropriately adopting prior microscopy data in an adversarial training, the network can recover a high-resolution, accurate image of new specimen from its single low-resolution measurement. This capacity has been adequately demonstrated by imaging various types of samples, such as USAF resolution target, human pathological slides and fluorescence-labelled fibroblast cells. Their gigapixel, multi-color reconstructions verify a successful GAN-based single image super-resolution procedure. Furthermore, this deep learning-based imaging approach doesn;t necessarily introduce any change to the setup of a conventional wide-filed microscope, reconstructing large FOV (about 95 mm^2), high-resolution (about 1.7 {mu}m) image at a high speed (in 1 second). As a result, GAN-microscopy opens a new way to computationally overcome the general challenge of high-throughput, high-resolution microscopy that is originally coupled to the physical limitation of systemu0027s optics.\",\n \"Title: A Network-Specific Markov Random Field Approach to Community Detection.\\nYear: 2018\\n\\nAbstract: Markov Random Field (MRF) is a powerful framework for developing probabilistic models of complex problems. MRF models possess rich structures to represent properties and constraints of a problem. It has been successful on many application problems, particularly those of computer vision and image processing, where data are structured, e.g., pixels are organized on grids. The problem of identifying communities in networks, which is essential for network analysis, is in principle analogous to finding objects in images. It is surprising that MRF has not yet been explored for network community detection. It is challenging to apply MRF to network analysis problems where data are organized on graphs with irregular structures. Here we present a network specific MRF approach to community detection. The new method effectively encodes the structural properties of an irregular network in an energy function (the core of an MRF model) so that the minimization of the function gives rise to the best community structures. We analyzed the new MRF-based method on several synthetic benchmarks and real world networks, showing its superior performance over the state-of-the-art methods for community identification.\",\n \"Title: Robust Detection of Link Communities in Large Social Networks by Exploiting Link Semantics.\\nYear: 2018\\n\\nAbstract: Community detection has been extensively studied for various applications, focusing primarily on network topologies. Recent research has started to explore node contents to identify semantically meaningful communities and interpret their structures using selected words. However, links in real networks typically have semantic descriptions, e.g., comments and emails in social media, supporting the notion of communities of links Indeed, communities of links can better describe multiple roles that nodes may play and provide a richer characterization of community behaviors than communities of nodes. The second issue in community finding is that most existing methods assume network topologies and descriptive contents to be consistent and to carry the compatible information of node group membership, which is generally violated in real networks. These methods are also restricted to interpret one community with one topic. The third problem is that the existing methods have used top ranked words or phrases to label topics when interpreting communities. However, it is often difficult to comprehend the derived topics using words or phrases, which may be irrelevant. To address these issues altogether, we propose a new unified probabilistic model that can be learned by a dual nested expectation-maximization algorithm. Our new method explores the intrinsic correlation between communities and topics to discover link communities robustly and extract adequate community summaries in sentences instead of words for topic labeling at the same time. It is able to derive more than one topical summary per community to provide rich explanations. We present experimental results to show the effectiveness of our new approach, and evaluate the quality of the results by a case study.\",\n \"Title: Incorporating network structure with node contents for community detection on large networks using deep learning.\\nYear: 2018\\n\\nAbstract: Community detection is an important task in social network analysis. In community detection, in general, there exist two types of the models that utilize either network topology or node contents. Some studies endeavor to incorporate these two types of models under the framework of spectral clustering for a better community detection. However, it was not successful to obtain a big achievement since they used a simple way for the combination. To reach a better community detection, it requires to realize a seamless combination of these two methods. For this purpose, we re-examine the properties of the modularity maximization and normalized-cut models and fund out a certain approach to realize a seamless combination of these two models. These two models seek for a low-rank embedding to represent of the community structure and reconstruct the network topology and node contents, respectively. Meanwhile, we found that autoencoder and spectral clustering have a similar framework in their low- rank matrix reconstruction. Based on this property, we proposed a new approach to seamlessly combine the models of modularity and normalized-cut via the autoencoder. The proposed method also utilized the advantages of the deep structure by means of deep learning. The experiment demonstrated that the proposed method can provide a nonlinearly deep representation for a large-scale network and reached an efficient community detection. The evaluation results showed that our proposed method outperformed the existing leading methods on nine real-world networks.\",\n \"Title: Tracking and Detecting Dynamic Communities with Node Popularity Preservation\\nYear: 2017\\n\\nAbstract: Temporal networks are increasingly prevalent in modern data, and the analysis of such networks can uncover the important phenomenon and characterize the properties of networks of dynamical units. This paper presents a new community detection model using nonnegative matrix factorization (NMF) which is capable of detecting and tracking dynamic communities in temporal networks. A novel temporal smoothness constraint by preserving the expected node popularity of two consecutive timeslices is proposed, and the community membership transition matrices are introduced to capture the sudden changes in network topologies in dynamic cases. We then propose a gradient descent algorithm to optimize objective function and prove its correctness and convergence. Experimental results on some synthetic and real benchmarked networks show the effectiveness of the proposed new method in detecting communities as well as finding their changes in dynamic networks.\",\n \"Title: Using Deep Learning for Community Discovery in Social Networks\\nYear: 2017\\n\\nAbstract: Community detection is an important task in social network analysis. Existing methods typically use the topological information alone, and ignore the rich information available in the content data. Recently, some researchers have noticed that user profiles can also benefit to community detection, and hence the combination of topology and node contents has become a new hot topic. Some methods using both topology and content have been proposed. However, they often suffer from two drawbacks: 1) they cannot extract a potential deep representation of the network; 2) they cannot automatically weight different information sources with adequate balance parameters. To overcome these issues, we propose a deep integration representation (DIR) algorithm via deep joint reconstruction, which is motivated by the similarity between deep feedforward auto-encoders and spectral clustering in terms of matrix reconstruction. Thanks to spectral clustering which is one of the best community detection methods, the proposed new method is also good at community discovery task. In addition, DIR has further benefit because it not only provides a nonlinear and deep representation of the network, but also learns the most suitable balance between different components automatically. We compare the proposed new approach with nine state-of-the-art community detection methods on eight real relatively large networks. The experimental results show the definite superiority of this new approach.\",\n \"Title: Finding Communities with Hierarchical Semantics by Distinguishing General and Specialized topics.\\nYear: 2018\\n\\nAbstract: Using network topology and semantic contents to find topic-related communities is a new trend in the field of community detection. By analyzing texts in social networks, we find that topics in networked contents are often hierarchical. In most cases, they have a two-level semantic structure with general and specialized topics, to respectively denote common and specific interests of communities. However, the existing community detection methods ignore such a hierarchy and take all words used to describe node semantics from an identical perspective. This indiscriminate use of words leads to natural defects in depicting networked content in which the deep semantics is not fully utilized. To address this problem, we propose a novel probabilistic generative model. By distinguishing the general and specialized topics of words, our model not only can find community structures more accurately, but also provide two-level semantic interpretation for each community. We train the model by deriving an efficient inference method under the framework of variational expectation-maximization. We provide a case study to show the ability of our algorithm in deep semantic interpretability of communities. The superiority of our algorithm for community detection is further demonstrated in comparison with eight state-of-the-art algorithms on eight real-world networks.\",\n \"Title: PICO Element Detection in Medical Text via Long Short-Term Memory Neural Networks.\\nYear: 2018\\n\\nAbstract: Successful evidence-based medicine (EBM) applications rely on answering clinical questions by analyzing large medical literature databases. In order to formulate a well-defined, focused clinical question, a framework called PICO is widely used, which identifies the sentences in a given medical text that belong to the four components: Participants/Problem (P), Intervention (I), Comparison (C) and Outcome (O). In this work, we present a Long Short-Term Memory (LSTM) neural network based model to automatically detect PICO elements. By jointly classifying subsequent sentences in the given text, we achieve state-of-the-art results on PICO element classification compared to several strong baseline models. We also make our curated data public as a benchmarking dataset so that the community can benefit from it.\",\n \"Title: Robust Detection of Communities with Multi-semantics in Large Attributed Networks.\\nYear: 2018\\n\\nAbstract: In this paper, we are interested in how to explore and utilize the relationship between network communities and semantic topics in order to find the strong explanatory communities robustly. First, the relationship between communities and topics displays different situations. For example, from the viewpoint of semantic mapping, their relationship can be one-to-one, one-to-many or many-to-one. But from the standpoint of underlying community structures, the relationship can be consistent, partially consistent or completely inconsistent. Second, it will be helpful to not only find communities more precise but also reveal the communities’ semantics that shows the relationship between communities and topics. To better describe this relationship, we introduce the transition probability which is an important concept in Markov chain into a well-designed nonnegative matrix factorization framework. This new transition probability matrix with a suitable prior which plays the role of depicting the relationship between communities and topics can perform well in this task. To illustrate the effectiveness of the proposed new approach, we conduct some experiments on both synthetic and real networks. The results show that our new method is superior to baselines in accuracy. We finally conduct a case study analysis to validate the new method’s strong interpretability to detected communities.\",\n \"Title: Autoencoder Based Community Detection with Adaptive Integration of Network Topology and Node Contents.\\nYear: 2018\\n\\nAbstract: Community detection plays an important role in understanding the structure and laws of social networks. Many community detection approaches have been proposed and focus on topological structure alone. In addition to topology, node contents exist in real-world networks, and may help for community detection. Recently, some studies try to combine topological structure and node contents. However, it is difficult to address an inherent situation in real-world networks, that is the mismatch between topological structure and node contents in term of community patterns. When considering both topology and content of networks, the performance of those community detection methods is often limited by this mismatch. Besides, networks are often full of nonlinear features, making those methods less effective in practice. In this paper, we present an adaptive method for community detection, which is reached by a graph regularized autoencoder approach. This new method introduces a novel adaptive parameter to achieve robust integration of the topological and content information when there exists the mismatch between those two types of information in term of communities. Experiments on both synthetic networks and real-world networks further indicate that the proposed new method exhibits more robust behavior and outperforms the leading methods when there exists the mismatch between topology and content.\",\n \"Title: Hierarchical Neural Networks for Sequential Sentence Classification in Medical Scientific Abstracts.\\nYear: 2018\\n\\nAbstract: Prevalent models based on artificial neural network (ANN) for sentence classification often classify sentences in isolation without considering the context in which sentences appear. This hampers the traditional sentence classification approaches to the problem of sequential sentence classification, where structured prediction is needed for better overall classification performance. In this work, we present a hierarchical sequential labeling network to make use of the contextual information within surrounding sentences to help classify the current sentence. Our model outperforms the state-of-the-art results by 2%-3% on two benchmarking datasets for sequential sentence classification in medical scientific abstracts.\",\n \"Title: Bridging Network Embedding and Graph Summarization.\\nYear: 2018\\n\\nAbstract: An important reason behind the prevalence of node representation learning is their superiority in downstream machine learning tasks on graphs. However, storing the vector-based node representation of massive real-world graphs often requires space that is orders of magnitude larger. To alleviate this issue, we introduce the problem of latent network summarization that is complementary to the problem of network embedding, and propose a general framework called Multi-LENS. Instead of deriving dense node-wise representations, the goal of latent network summarization is to summarize the structural properties of the graph in a latent space with dimensionality that is independent of the nodes or edges in the input graph. The size-independent graph summary given by Multi-LENS consists of (i) a set of relational aggregators with their compositions (relational functions) that captures structural features of multi-order node-centric subgraphs, and (ii) the low-rank approximations to matrices that incorporate captured structural features. In addition, Multi-LENS is able to derive node embeddings on the fly from this latent summary due to its inductive properties. Multi-LENS bridges advantages brought by both embeddings and graph summarization, and applies to graphs with or without directionality, weights, attributes or labels. Extensive experiments on both synthetic and real-world graphs show that Multi-LENS achieves 2 - 89% improvement in AUC for link prediction, while requiring less than 79x space compared to existing representation learning approaches. We also show the effectiveness of Multi-LENS summaries in anomaly and event detection on two real-world graphs.\",\n \"Title: Advancing PICO Element Detection in Medical Text via Deep Neural Networks.\\nYear: 2018\\n\\nAbstract: In evidence-based medicine (EBM), structured medical questions are always favored for efficient search of the best available evidence for treatments. PICO element detection is widely used to help structurize the clinical studies and question by identifying the sentences in a given medical text that belong to one of the four components: Participants (P), Intervention (I), Comparison (C), and Outcome (O). In this work, we propose a hierarchical deep neural network (DNN) architecture that contains dual bi-directional long short-term memory (bi-LSTM) layers to automatically detect the PICO element in medical texts. Within the model, the lower layer of bi-LSTM is for sentence encoding while the upper one is to contextualize the encoded sentence representation vector. In addition, we adopt adversarial and virtual adversarial training to regularize the model. Overall, we advance the PICO element detection to new state-of-the-art performance, outperforming the previous works by at least 4% in F1 score for all P/I/O categories.\",\n \"Title: Context-Aware Data Dissemination For Icn-Based Vehicular Ad Hoc Networks\\nYear: 2018\\n\\nAbstract: Information-centric networking (ICN) technology matches many major requirements of vehicular ad hoc networks (VANETs) in terms of its connectionless networking paradigm accordant with the dynamic environments of VANETs and is increasingly being applied to VANETs. However, wireless transmissions of packets in VANETs using ICN mechanisms can lead to broadcast storms and channel contention, severely affecting the performance of data dissemination. At the same time, frequent changes of topology due to driving at high speeds and environmental obstacles can also lead to link interruptions when too few vehicles are involved in data forwarding. Hence, balancing the number of forwarding vehicular nodes and the number of copies of packets that are forwarded is essential for improving the performance of data dissemination in information-centric networking for vehicular ad-hoc networks. In this paper, we propose a context-aware packet-forwarding mechanism for ICN-based VANETs. The relative geographical position of vehicles, the density and relative distribution of vehicles, and the priority of content are considered during the packet forwarding. Simulation results show that the proposed mechanism can improve the performance of data dissemination in ICN-based VANET in terms of a successful data delivery ratio, packet loss rate, bandwidth usage, data response time, and traversed hops.\",\n \"Title: Edge Content Enhanced Network Embedding\\nYear: 2018\\n\\nAbstract: Network embedding, aiming at learning the low dimensional representations of nodes in a network, is a key to many network analysis tasks. All the current network embedding methods primarily explore the network topology or node attributes, while no effort has been made to analyze the edge content for network embedding. The edge content, such as the email content between two users in an email network, is often naturally associated with edges. They carry rich information to describe the interaction between nodes, and provide valuable supervision to learn the representations of nodes. In this paper, we propose a novel edge content enhanced network embedding model, which incorporates the edge content to guide the network representation learning process. We provide the efficient updating rules to infer the parameters in the model, along with theoretical analysis on correctness and convergence guarantees. Extensive experiments, in comparison with the state-of-the-arts, show the superior performance of our proposed new approach on different network analysis tasks.\",\n \"Title: Analysis of Activity States of Local Neuronal Microcircuits in Mouse Brain.\\nYear: 2018\\n\\nAbstract: Time series of neuronal activity corresponding to different activity states in mouse brain are analyzed in the time domain and the time-frequency domain. The signals are associated with either a slow wave brain state or a persistent brain state. For both states, characteristic spectral features are identified and a simple detector is proposed that is able to identify the brain state with low latency and high accuracy. In practice, being able to monitor the brain state online and in real time is crucial for improved in vivo experiments and, ultimately, for a causal understanding of brain dynamics.\",\n \"Title: Constrained common cluster based model for community detection in temporal and multiplex networks.\\nYear: 2018\\n\\nAbstract: On one hand, the detection of tightly connected groups, also known as community detection in complex networks, is a prominent problem for network analysis and mining. On the other hand, almost all of social, biological, bibliographic, communication and computer systems are modeled as temporal networks, the topological structures of which evolve with time, or multiplex networks, each pair of nodes of which has multiple linked relations. Current methods of community detection for temporal networks are based on incremental, independent or evolutionary clustering, and for multiplex networks are based on fusion of the multiple links. However, all these methods ignore the common structure hidden in the networks, which is denoted as the common cluster here. So in this paper, we propose a constrained common cluster based model (C3 model) to analyze the temporal and multiplex networks, which can not only detect the community structure, but also identify the importance of each node based on the common cluster structure of both two classes of networks. The intrinsic assumption of the proposed model is that there are common or coincident clusters hidden in these networks. In detail, we first construct the Markov steady-state matrices of each snapshot of temporal network or each slice of multiplex network. Next, we propose the object function of C3 model by combining the Markov steady-state matrices, similarity matrices with community membership matrices of each snapshot or slice of the network. Last, a gradient descent algorithm based on non-negative matrix factorization is proposed for the object function. Experiments on both synthetic datasets and real-world networks demonstrate that the proposed C3 model has competitive performance based on the evaluation indexes NMI and error of community detection, otherwise, the proposed model could identify the importance of nodes of the temporal or multiplex networks.\",\n \"Title: A Unified Weakly Supervised Framework for Community Detection and Semantic Matching.\\nYear: 2018\\n\\nAbstract: Due to the sparsity of network, some community detection methods only based on topology often lead to relatively low accuracy. Although some methods have been proposed to improve the detection accuracy by using few known semi-supervised information or node content, the research of community detection not only pursues the enhancement of community accuracy, but also pays more attention to the semantic description for communities. In this paper, we proposed a unified nonnegative matrix factorization framework simultaneously for community detection and semantic matching by integrating both semi-supervised information and node content. The framework reveals two-fold community structures as well as their coupling relationship matrix, which helps to identify accurate community structure and at the same time assign specific semantic information to each community. Experiments on some real networks show that the framework is efficient to match each community with specific semantic information, and the performance are superior over the compared methods.\",\n \"Title: Integrative Network Embedding via Deep Joint Reconstruction.\\nYear: 2018\\n\\nAbstract: Network embedding is to learn a low-dimensional representation for a network in order to capture intrinsic features of the network. It has been applied to many applications, e.g., network community detection and user recommendation. One of the recent research topics for network embedding has been focusing on exploitation of diverse information, including network topology and semantic information on nodes of networks. However, such diverse information has not been fully utilized nor adequately integrated in the existing methods, so that the resulting network embedding is far from satisfactory. In this paper, we develop a weight-free multi-component network embedding approach by network reconstruction via a deep Autoencoder. Three key components make our new approach effective, i.e., a uniformed graph representation of network topology and semantic information, enhancement to the graph representation using local network structure (i.e., pairwise relationship on nodes) by sampling with latent space regularization, and integration of the diverse information in graph forms in a deep Autoencoder. Extensive experimental results on seven real-world networks demonstrate a superior performance of our method over nine state-of-the-art methods for embedding.\",\n \"Title: Combination of links and node contents for community discovery using a graph regularization approach.\\nYear: 2019\\n\\nAbstract: With the rapid growth of the networked data, the study of community detection is drawing increasing attention of researchers. A number of algorithms have been proposed and some of them have been well applied in many research fields, such as recommendation systems, information retrieval, etc. Traditionally, the community detection methods mainly use the knowledge of the topological structure which contains the most important clue for finding potential groups or communities. However, as we know, a wealth of content information exists on the nodes in real-world networks, and may help for community detection. Considering the above problem, we introduce a novel community detection method under the framework of nonnegative matrix factorization (NMF), and adopt the idea that two nodes with similar content will be most likely to belong to the same community to achieve the incorporation of links and node contents, i.e., we employ a graph regularization to penalize the dissimilarity of nodes denoted by community memberships. Besides, we introduce an intuitive manifold learning strategy to recover the intrinsic geometrical structure of the content information, i.e., K-near neighbor consistency. In addition, we found that, there are still drawbacks in this framework due to it does not consider the heterogeneous distribution of node degrees. This heterogeneous distribution can affect the function of graph regularization and isolates the original community memberships. We first proposed the node popularities satisfying the above interpretation and develop a new NMF-based model, named as Combination of Links and Node Contents for Community Discovery (CLNCCD). The experiments on both artificial and real-world networks compared with the state-of-the-art methods show that, the new model obtains significant improvement for community detection by incorporating node contents effectively.\",\n \"Title: RoSeq: Robust Sequence Labeling.\\nYear: 2020\\n\\nAbstract: In this paper, we mainly investigate two issues for sequence labeling, namely, label imbalance and noisy data that are commonly seen in the scenario of named entity recognition (NER) and are largely ignored in the existing works. To address these two issues, a new method termed robust sequence labeling (RoSeq) is proposed. Specifically, to handle the label imbalance issue, we first incorporate label statistics in a novel conditional random field (CRF) loss. In addition, we design an additional loss to reduce the weights of overwhelming easy tokens for augmenting the CRF loss. To address the noisy training data, we adopt an adversarial training strategy to improve model generalization. In experiments, the proposed RoSeq achieves the state-of-the-art performances on CoNLL and English Twitter NER—88.07% on CoNLL-2002 Dutch, 87.33% on CoNLL-2002 Spanish, 52.94% on WNUT-2016 Twitter, and 43.03% on WNUT-2017 Twitter without using the additional data.\",\n \"Title: A Novel Generative Topic Embedding Model by Introducing Network Communities\\nYear: 2019\\n\\nAbstract: Topic models have many important applications in fields such as Natural Language Processing. Topic embedding modelling aims at introducing word and topic embeddings into topic models to describe correlations between topics. Existing topic embedding methods use documents alone, which suffer from the topical fuzziness problem brought by the introduction of embeddings of semantic fuzzy words, e.g. polysemous words or some misleading academic terms. Links often exist between documents which form document networks. The use of links may alleviate this semantic fuzziness, but they are sparse and noisy which may meanwhile mislead topics. In this paper, we utilize community structure to solve these problems. It can not only alleviate the topical fuzziness of topic embeddings since communities are often believed to be topic related, but also can overcome the drawbacks brought by the sparsity and noise of networks (because community is a high-order network information). We give a new generative topic embedding model which incorporates documents (with topics) and network (with communities) together, and uses probability transition to describe the relationship between topics and communities to make it robust when topics and communities do not match. An efficient variational inference algorithm is then proposed to learn the model. We validate the superiority of our new approach on two tasks, document classifications and visualization of topic embeddings, respectively.\",\n \"Title: Robust distributed cooperative RSS-based localization for directed graphs in mixed LoS/NLoS environments.\\nYear: 2019\\n\\nAbstract: The accurate and low-cost localization of sensors using a wireless sensor network is critically required in a wide range of today’s applications. We propose a novel, robust maximum likelihood-type method for distributed cooperative received signal strength-based localization in wireless sensor networks. To cope with mixed LoS/NLoS conditions, we model the measurements using a two-component Gaussian mixture model. The relevant channel parameters, including the reference path loss, the path loss exponent, and the variance of the measurement error, for both LoS and NLoS conditions, are assumed to be unknown deterministic parameters and are adaptively estimated. Unlike existing algorithms, the proposed method naturally takes into account the (possible) asymmetry of links between nodes. The proposed approach has a communication overhead upper-bounded by a quadratic function of the number of nodes and computational complexity scaling linearly with it. The convergence of the proposed method is guaranteed for compatible network graphs, and compatibility can be tested a priori by restating the problem as a graph coloring problem. Simulation results, carried out in comparison to a centralized benchmark algorithm, demonstrate the good overall performance and high robustness in mixed LoS/NLoS environments.\",\n \"Title: Unsupervised Text Style Transfer via Iterative Matching and Translation.\\nYear: 2019\\n\\nAbstract: Text style transfer seeks to learn how to automatically rewrite sentences from a source domain to the target domain in different styles, while simultaneously preserving their semantic contents. A major challenge in this task stems from the lack of parallel data that connects the source and target styles. Existing approaches try to disentangle content and style, but this is quite difficult and often results in poor content-preservation and grammaticality. In contrast, we propose a novel approach by first constructing a pseudo-parallel resource that aligns a subset of sentences with similar content between source and target corpus. And then a standard sequence-to-sequence model can be applied to learn the style transfer. Subsequently, we iteratively refine the learned style transfer function while improving upon the imperfections in our original alignment. Our method is applied to the tasks of sentiment modification and formality transfer, where it outperforms state-of-the-art systems by a large margin. As an auxiliary contribution, we produced a publicly-available test set with human-generated style transfers for future community use.\",\n \"Title: Publicly Available Clinical BERT Embeddings.\\nYear: 2019\\n\\nAbstract: Contextual word embedding models such as ELMo (Peters et al., 2018) and BERT (Devlin et al., 2018) have dramatically improved performance for many natural language processing (NLP) tasks in recent months. However, these models have been minimally explored on specialty corpora, such as clinical text; moreover, in the clinical domain, no publicly-available pre-trained BERT models yet exist. In this work, we address this need by exploring and releasing BERT models for clinical text: one for generic clinical text and another for discharge summaries specifically. We demonstrate that using a domain-specific model yields performance improvements on three common clinical NLP tasks as compared to nonspecific embeddings. These domain-specific models are not as performant on two clinical de-identification tasks, and argue that this is a natural consequence of the differences between de-identified source text and synthetically non de-identified task text.\",\n \"Title: Incorporating Network Embedding into Markov Random Field for Better Community Detection\\nYear: 2019\\n\\nAbstract: Recent research on community detection focuses on learning representations of nodes using different network embedding methods, and then feeding them as normal features to clustering algorithms. However, we find that though one may have good results by direct clustering based on such network embedding features, there is ample room for improvement. More seriously, in many real networks, some statistically significant nodes which play pivotal roles are often divided into incorrect communities using network embedding methods. This is because while some distance measures are used to capture the spatial relationship between nodes by embedding, the nodes after mapping to feature vectors are essentially not coupled any more, losing important structural information. To address this problem, we propose a general Markov Random Field (MRF) framework to incorporate coupling in network embedding which allows better detecting network communities. By smartly utilizing properties of MRF, the new framework not only preserves the advantages of network embedding (e.g. low complexity, high parallelizability and applicability for traditional machine learning), but also alleviates its core drawback of inadequate representations of dependencies via making up the missing coupling relationships. Experiments on real networks show that our new approach improves the accuracy of existing embedding methods (e.g. Node2Vec, DeepWalk and MNMF), and corrects most wrongly-divided statistically-significant nodes, which makes network embedding essentially suitable for real community detection applications. The new approach also outperforms other state-of-the-art conventional community detection methods.\",\n \"Title: The Image-Based Analysis And Classification Of Urine Sediments Using A Lenet-5 Neural Network\\nYear: 2020\\n\\nAbstract: In this work, we presented a deep learning approach based on the LeNet-5 network for analysing and classifying recognisable shapes in urine sample images. The approach is based on shape analysis to recognise and classify red blood cells, white blood cells, epithelial cells and crystals observed under microscopes in urine samples. We modified the LeNet-5 neural network by changing the numbers of output nodes and convolutional layers. We compared the results of our method with those obtained by traditional feature extraction followed by back-propagation neural networks. Our testing showed that our method achieved a higher accuracy, sensitivity and specificity. The performance of our method demonstrated its broad applicability in urine sample analysis.\",\n \"Title: Morphological Verb-Aware Tibetan Language Model.\\nYear: 2019\\n\\nAbstract: The Tibetan language model (TLM) is the key to Tibetan natural language processing. In this paper, we first observe that, different from widely used languages, Tibetan contains many morphological verbs that rarely appear in natural sentences but play a key role in accurate text prediction. This property is usually ignored by existing methods and makes traditional training strategies less effective in constructing accurate and robust TLMs. Hence, we propose a morphological verb-aware TLM by offline learning via a character frequency reweighting strategy and online tuning of discriminative weights conditioned on morphological verbs. However, because of the influence of morphological verbs on the tense and semantics of sentences, it is necessary to consider the morphological verbs in Tibetan. As a result, compared with stateof-the-art methods, our method not only reduces the perplexity but also improves the character error on tasks of the text prediction and automatic speech recognition (ASR).\",\n \"Title: Smart Roles: Inferring Professional Roles in Email Networks\\nYear: 2019\\n\\nAbstract: Email is ubiquitous in the workplace. Naturally, machine learning models that make third-party email clients \\\"smarter\\\" can dramatically impact employees' productivity and efficiency. Motivated by this potential, we study the task of professional role inference from email data, which is crucial for email prioritization and contact recommendation systems. The central question we address is: Given limited data about employees, as is common in third-party email applications, can we infer where in the organizational hierarchy these employees belong based on their email behavior? Toward our goal, in this paper we study professional role inference on a unique new email dataset comprising billions of email exchanges across thousands of organizations. Taking a network approach in which nodes are employees and edges represent email communication, we propose EMBER, or EMBedding Email-based Roles, which finds email-centric embeddings of network nodes to be used in professional role inference tasks. EMBER automatically captures behavioral similarity between employees in the email network, leading to embeddings that naturally distinguish employees of different hierarchical roles. EMBER often outperforms the state-of-the-art by 2-20% in role inference accuracy and 2.5-344x in speed. We also use EMBER with our unique dataset to study how inferred professional roles compare between organizations of different sizes and sectors, gaining new insights into organizational hierarchy.\",\n \"Title: Latent Network Summarization: Bridging Network Embedding and Summarization\\nYear: 2019\\n\\nAbstract: Motivated by the computational and storage challenges that dense embeddings pose, we introduce the problem of latent network summarization that aims to learn a compact, latent representation of the graph structure with dimensionality that is independent of the input graph size (\\\\i.e., #nodes and #edges), while retaining the ability to derive node representations on the fly. We propose Multi-LENS, an inductive multi-level latent network summarization approach that leverages a set of relational operators and relational functions (compositions of operators) to capture the structure of egonets and higher-order subgraphs, respectively. The structure is stored in low-rank, size-independent structural feature matrices, which along with the relational functions comprise our latent network summary. Multi-LENS is general and naturally supports both homogeneous and heterogeneous graphs with or without directionality, weights, attributes or labels. Extensive experiments on real graphs show 3.5-34.3% improvement in AUC for link prediction, while requiring 80-2152x less output storage space than baseline embedding methods on large datasets. As application areas, we show the effectiveness of Multi-LENS in detecting anomalies and events in the Enron email communication graph and Twitter co-mention graph.\",\n \"Title: Heterogeneous Graph Neural Network via Attribute Completion\\nYear: 2021\\n\\nAbstract: ABSTRACT Heterogeneous information networks (HINs), also called heterogeneous graphs, are composed of multiple types of nodes and edges, and contain comprehensive information and rich semantics. Graph neural networks (GNNs), as powerful tools for graph data, have shown superior performance on network analysis. Recently, many excellent models have been proposed to process hetero-graph data using GNNs and have achieved great success. These GNN-based heterogeneous models can be interpreted as smooth node attributes guided by graph structure, which requires all nodes to have attributes. However, this is not easy to satisfy, as some types of nodes often have no attributes in heterogeneous graphs. Previous studies take some handcrafted methods to solve this problem, which separate the attribute completion from the graph learning process and, in turn, result in poor performance. In this paper, we hold that missing attributes can be acquired by a learnable manner, and propose a general framework for Heterogeneous Graph Neural Network via Attribute Completion (HGNN-AC), including pre-learning of topological embedding and attribute completion with attention mechanism. HGNN-AC first uses existing HIN-Embedding methods to obtain node topological embedding. Then it uses the topological relationship between nodes as guidance to complete attributes for no-attribute nodes by weighted aggregation of the attributes from these attributed nodes. Our complement mechanism can be easily combined with an arbitrary GNN-based heterogeneous model making the whole system end-to-end. We conduct extensive experiments on three real-world heterogeneous graphs. The results demonstrate the superiority of the proposed framework over state-of-the-art baselines.\",\n \"Title: Can I Be Of Further Assistance? Using Unstructured Knowledge Access To Improve Task-Oriented Conversational Modeling\\nYear: 2021\\n\\nAbstract: Most prior work on task-oriented dialogue systems are restricted to limited coverage of domain APIs. However, users oftentimes have requests that are out of the scope of these APIs. This work focuses on responding to these beyond-API-coverage user turns by incorporating external, unstructured knowledge sources. Our approach works in a pipelined manner with knowledge-seeking turn detection, knowledge selection, and response generation in sequence. We introduce novel data augmentation methods for the first two steps and demonstrate that the use of information extracted from dialogue context improves the knowledge selection and end-to-end performances. Through experiments, we achieve state-of-theart performance for both automatic and human evaluation metrics on the DSTC9 Track 1 benchmark dataset, validating the effectiveness of our contributions.\",\n \"Title: \\\"HOW ROBUST R U?\\\": EVALUATING TASK-ORIENTED DIALOGUE SYSTEMS ON SPOKEN CONVERSATIONS\\nYear: 2021\\n\\nAbstract: Most prior work in dialogue modeling has been on written conversations mostly because of existing data sets. However, written dialogues are not sufficient to fully capture the nature of spoken conversations as well as the potential speech recognition errors in practical spoken dialogue systems. This work presents a new benchmark on spoken task-oriented conversations, which is intended to study multi-domain dialogue state tracking and knowledge-grounded dialogue modeling. We report that the existing state-of-the-art models trained on written conversations are not performing well on our spoken data, as expected. Furthermore, we observe improvements in task performances when leveraging n-best speech recognition hypotheses such as by combining predictions based on individual hypotheses. Our data set enables speech-based bench-marking of task-oriented dialogue systems.\",\n \"Title: TransFusion: Multi-Modal Fusion for Video Tag Inference via Translation-based Knowledge Embedding\\nYear: 2021\\n\\nAbstract: ABSTRACTTag inference is an important task in the business of video platforms with wide applications such as recommendation, interpretation, and more. Existing works are mainly based on extracting video information from multiple modalities such as frames or music, and then infer tags through classification or object detection. This, however, does not apply to inferring generic tags or taxonomy that are less relevant to video contents, such as video originality or its broader category, which are important in practice. In this paper, we claim that these generic tags can be modeled through the semantic relations between videos and tags, and can be utilized simultaneously with the multi-modal features to achieve better video tagging. We propose TransFusion, an end-to-end supervised learning framework that fuses multi-modal embeddings (e.g., vision, audio, texts, etc.) with the knowledge embedding to derive the video representation. To infer the diverse tags following heterogeneous relations, TransFusion adopts a dual attentive approach to learn both the modality importance in fusion and relation importance in inference. Besides, it is general enough and can be used with the existing translation-based knowledge embedding approaches. Extensive experiments show that TransFusion outperforms the baseline methods with lowered mean rank and at least 9.59% improvement in [email protected] on the real-world video knowledge graph.\",\n \"Title: Deep Transfer Learning for Multi-source Entity Linkage via Domain Adaptation\\nYear: 2021\\n\\nAbstract: Multi-source entity linkage focuses on integrating knowledge from multiple sources by linking the records that represent the same real world entity. This is critical in high-impact applications such as data cleaning and user stitching. The state-of-the-art entity linkage pipelines mainly depend on supervised learning that requires abundant amounts of training data. However, collecting well-labeled training data becomes expensive when the data from many sources arrives incrementally over time. Moreover, the trained models can easily overfit to specific data sources, and thus fail to generalize to new sources due to significant differences in data and label distributions. To address these challenges, we present ADAMEL, a deep transfer learning framework that learns generic high-level knowledge to perform multi-source entity linkage. ADAMEL models the attribute importance that is used to match entities through an attribute-level self-attention mechanism, and leverages the massive unlabeled data from new data sources through domain adaptation to make it generic and data-source agnostic. In addition, ADAMEL is capable of incorporating an additional set of labeled data to more accurately integrate data sources with different attribute importance. Extensive experiments show that our framework achieves state-of-the-art results with 8.21% improvement on average over methods based on supervised learning. Besides, it is more stable in handling different sets of data sources in less runtime.\",\n \"Title: Budget-constrained Truss Maximization over Large Graphs: A Component-based Approach\\nYear: 2021\\n\\nAbstract: ABSTRACTCohesive substructure identification is one fundamental task of graph analytics. Recently, a useful problem of dense subgraph maximization has attracted significant attentions, which aims at enlarging a dense subgraph pattern using a few new edge insertions, e.g., k-core maximization. As a more cohesive subgraph of k-core, k-truss requires that each edge has at least k-2 triangles within this subgraph. However, the problem of k-truss maximization has not been studied yet. In this paper, we motivate and formulate a new problem of budget-constrained k-truss maximization. Given a budget of b edges and an integer k≥2, the problem is to find and insert b new edges into a graph G such that the resulted k-truss of G is maximized. We theoretically prove the NP-hardness of k-truss maximization problem. To efficiently tackle it, we analyze non-submodular property of k-truss newcomers function and develop non-conventional heuristic strategies for edge insertions. We first identify high-quality candidate edges with regard to (k-1)-light subgraphs and propose a greedy algorithm using per-edge insertion. Besides further improving the efficiency by pruning disqualified candidate edges, we finally develop a component-based dynamic programming algorithm for enlarging k-truss mostly, which makes a balance of budget assignment and inserts multiple edges simultaneously into all (k-1)-light components. Extensive experiments on nine real-world graphs demonstrate the efficiency and effectiveness of our proposed methods.\",\n \"Title: Artificial-Intelligence-Enabled Reagent-Free Imaging Hematology Analyzer\\nYear: 2021\\n\\nAbstract: Leukocyte differential test is a widely carried out clinical procedure for screening infectious diseases. Existing hematology analyzers require labor-intensive work and a panel of expensive reagents. Herein, an artificial-intelligence-enabled reagent-free imaging hematology analyzer (AIRFIHA) modality is reported that can accurately classify subpopulations of leukocytes with minimal sample preparation. AIRFIHA is realized through training a two-step residual neural network using label-free images of isolated leukocytes acquired from a custom-built quantitative phase microscope. By leveraging the rich information contained in quantitative phase images, not only high accuracy is achieved in differentiating B and T lymphocytes, but also CD4 and CD8 T cells are classified, therefore outperforming the classification accuracy of most current hematology analyzers. The performance of AIRFIHA in a randomly selected test set is validated and is cross-validated across all blood donors. Due to its easy operation, low cost, and accurate discerning capability of complex leukocyte subpopulations, AIRFIHA is clinically translatable and can also be deployed in resource-limited settings, e.g., during pandemic situations for the rapid screening of infectious diseases.\",\n \"Title: Continuous-Variable Quantum Key Distribution Based on Heralded Hybrid Linear Amplifier with a Local Local Oscillator\\nYear: 2021\\n\\nAbstract: An improved continuous variable quantum key distribution (CVQKD) approach based on a heralded hybrid linear amplifier (HLA) is proposed in this study, which includes an ideal deterministic linear amplifier and a probabilistic noiseless linear amplifier. The CVQKD, which is based on an amplifier, enhances the signal-to-noise ratio and provides for fine control between high gain and strong noise reduction. We focus on the impact of two types of optical amplifiers on system performance: phase sensitive amplifiers (PSA) and phase insensitive amplifiers (PIA). The results indicate that employing amplifiers, local local oscillation-based CVQKD systems can enhance key rates and communication distances. In addition, the PIA-based CVQKD system has a broader application than the PSA-based system.\",\n \"Title: Universal Graph Convolutional Networks.\\nYear: 2021\\n\\nAbstract: Use the \\\"Report an Issue\\\" link to request a name change.\",\n \"Title: Sketching as a Tool for Understanding and Accelerating Self-attention for Long Sequences\\nYear: 2022\\n\\nAbstract: Transformer-based models are not efficient in processing long sequences due to the quadratic space and time complexity of the self-attention modules. To address this limitation, Linformer and Informer are proposed to reduce the quadratic complexity to linear (modulo logarithmic factors) via low-dimensional projection and row selection respectively. These two models are intrinsically connected, and to understand their connection, we introduce a theoretical framework of matrix sketching. Based on the theoretical analysis, we propose Skeinformer to accelerate self-attention and further improve the accuracy of matrix approximation to self-attention with three carefully designed components: column sampling, adaptive row normalization and pilot sampling reutilization. Experiments on the Long Range Arena (LRA) benchmark demonstrate that our methods outperform alternatives with a consistently smaller time/space footprint.\",\n \"Title: A Deep Dive Into Understanding The Random Walk-Based Temporal Graph Learning\\nYear: 2021\\n\\nAbstract: Machine learning on graph data has gained significant interest because of its applicability to various domains ranging from product recommendations to drug discovery. While there is a rapid growth in the algorithmic community, the computer architecture community has so far focused on a subset of graph learning algorithms including Graph Convolution Network (GCN), and a few others. In this paper, w...\",\n \"Title: AS-GCN: Adaptive Semantic Architecture of Graph Convolutional Networks for Text-Rich Networks\\nYear: 2021\\n\\nAbstract: Graph Neural Networks (GNNs) have demonstrated great power in many network analytical tasks. However, graphs (i.e., networks) in the real world are usually text-rich, implying that valuable semantic information needs to be carefully considered. Existing GNNs for text-rich networks typically treat text as attribute words alone, which inevitably leads to the loss of important semantic structures, li...\",\n \"Title: An Effective and Robust Framework by Modeling Correlations of Multiplex Network Embedding\\nYear: 2021\\n\\nAbstract: The dependencies across different layers are an important property in multiplex networks and a few methods have been proposed to learn the dependencies in various ways. When capturing the dependencies across different layers, some of them assumed the structure among layers following consistent connectivity to force two nodes with a link in one layer tend to have links in other layers, some introduced a common vector to model the shared information across all layers. However, the correlations among layers in multiplex networks are diverse, which go beyond the connectivity consistency. In this paper, we propose a novel Modeling Correlations for Multiplex network Embedding (MCME) framework to learn the robust node representations for each layer. It can deal with complex correlations with a common structure, layer similarity and node heterogeneity through a unified framework in multiplex networks. To evaluate our proposed model, we conduct extensive experiments on several real-world datasets and the results demonstrate that our proposed model consistently outperforms state-of-the-art methods.\",\n \"Title: Towards Textual Out-of-Domain Detection Without In-Domain Labels\\nYear: 2022\\n\\nAbstract: In many real-world settings, machine learning models need to identify user inputs that are out-of-domain (OOD) so as to avoid performing wrong actions. This work focuses on a challenging case of OOD detection, where no labels for in-domain data are accessible (e.g., no intent labels for the intent classification task). To this end, we first evaluate different language model based approaches that predict likelihood for a sequence of tokens. Furthermore, we propose a novel representation learning based method by combining unsupervised clustering and contrastive learning so that better data representations for OOD detection can be learned. Through extensive experiments, we demonstrate that this method can significantly outperform likelihood-based methods and can be even competitive to the state-of-the-art supervised approaches with label information.\",\n \"Title: Inflation Improves Graph Neural Networks\\nYear: 2022\\n\\nAbstract: ABSTRACT Graph neural networks (GNNs) have gained significant success in graph representation learning and become the go-to approach for many graph-based tasks. Despite their effectiveness, the performance of GNNs is known to decline gradually as the number of layers increases. This attenuation is mainly caused by noise propagation, which refers to the useless or negative information propagated (directly or indirectly) from other nodes during the multi-layer graph convolution for node representation learning. This noise increases more severely as the layers of GNNs deepen, which is also a main reason of over-smoothing. In this paper, we propose a new convolution strategy for GNNs to address this problem via suppressing the noise propagation. Specifically, we first find that the feature propagation process of GNNs can be taken as a Markov chain. And then, based on the idea of Markov clustering, we introduce a new graph inflation layer (i.e., using a power function over the distribution) into GNNs to prevent noise propagating from local neighbourhoods to the whole graph with the increase of network layers. Our method is simple in design, which does not require any changes on the original basis and therefore can be easily extended. We conduct extensive experiments on real-world networks and have a stable improved performance as the network depth increases over existing GNNs.\",\n \"Title: Graph Neural Network for Higher-Order Dependency Networks\\nYear: 2022\\n\\nAbstract: ABSTRACT Graph neural network (GNN) has become a popular tool to analyze the graph data. Existing GNNs only focus on networks with first-order dependency, that is, conventional networks following the Markov property. However, many networks in real life own the higher-order dependency, such as click-stream data where the choice of the next page depends not only on the current page but also on previous pages. This kind of sequential data from complex systems (including natural dependencies) are often ignored by existing GNNs which makes them ineffective. To address this problem, we propose for the first time new GNN approaches for higher-order networks in this paper. First, we form sequence fragments by the current node and its predecessor nodes of different orders as candidate higher-order dependencies. When the fragment significantly affects the probability distribution of different successor nodes of the current node, we include it in the higher-order dependency set. We formulize the network with higher-order dependency as an augmented conventional first-order network, and then feed it into GNNs to derive network embeddings. Moreover, we further propose a new end-to-end GNN framework for dealing with higher-order networks directly in the model. Specifically, the higher-order dependency is used as the neighbor aggregation controller when the node is embedded and updated. In the graph convolutional layer, in addition to the first-order neighbor information, we also aggregate the middle node information from the higher-order dependency segment. We finally test the new approaches on three real networks with higher-order dependency, and compare with some state-of-the-art methods. The results show significant improvements of the new approaches which consider higher-order dependency.\",\n \"Title: Empowering parameter-efficient transfer learning by recognizing the kernel structure in self-attention.\\nYear: 2022\\n\\nAbstract: The massive amount of trainable parameters in the pre-trained language models (PLMs) makes them hard to be deployed to multiple downstream tasks. To address this issue, parameter-efficient transfer learning methods have been proposed to tune only a few parameters during fine-tuning while freezing the rest. This paper looks at existing methods along this line through the \\\\textit{kernel lens}. Motivated by the connection between self-attention in transformer-based PLMs and kernel learning, we propose \\\\textit{kernel-wise adapters}, namely \\\\textit{Kernel-mix}, that utilize the kernel structure in self-attention to guide the assignment of the tunable parameters. These adapters use guidelines found in classical kernel learning and enable separate parameter tuning for each attention head. Our empirical results, over a diverse set of natural language generation and understanding tasks, show that our proposed adapters can attain or improve the strong performance of existing baselines.\",\n \"Title: On the Limits of Evaluating Embodied Agent Model Generalization Using Validation Sets\\nYear: 2022\\n\\nAbstract: Natural language guided embodied task completion is a challenging problem since it requires understanding natural language instructions, aligning them with egocentric visual observations, and choosing appropriate actions to execute in the environment to produce desired changes. We experiment with augmenting a transformer model for this task with modules that effectively utilize a wider field of view and learn to choose whether the next step requires a navigation or manipulation action. We observed that the proposed modules resulted in improved, and in fact state-of-the-art performance on an unseen validation set of a popular benchmark dataset, ALFRED. However, our best model selected using the unseen validation set underperforms on the unseen test split of ALFRED, indicating that performance on the unseen validation set may not in itself be a sufficient indicator of whether model improvements generalize to unseen test sets. We highlight this result as we believe it may be a wider phenomenon in machine learning tasks but primarily noticeable only in benchmarks that limit evaluations on test splits, and highlights the need to modify benchmark design to better account for variance in model performance.\",\n \"Title: DMIX: Adaptive Distance-aware Interpolative Mixup\\nYear: 2022\\n\\nAbstract: Interpolation-based regularisation methods such as Mixup, which generate virtual training samples, have proven to be effective for various tasks and modalities. We extend Mixup and propose DMIX, an adaptive distanceaware interpolative Mixup that selects samples based on their diversity in the embedding space. DMIX leverages the hyperbolic space as a similarity measure among input samples for a richer encoded representation. DMIX achieves state-of-the-art results on sentence classification over existing data augmentation methods on 8 benchmark datasets across English, Arabic, Turkish, and Hindi languages while achieving benchmark F1 scores in 3 times less number of iterations. We probe the effectiveness of DMIX in conjunction with various similarity measures and qualitatively analyze the different components. DMIX being generalizable, can be applied to various tasks, models and modalities.\",\n \"Title: CGMN: A Contrastive Graph Matching Network for Self-Supervised Graph Similarity Learning.\\nYear: 2022\\n\\nAbstract: Graph similarity learning refers to calculating the similarity score between two graphs, which is required in many realistic applications, such as visual tracking, graph classification, and collaborative filtering. As most of the existing graph neural networks yield effective graph representations of a single graph, little effort has been made for jointly learning two graph representations and calculating their similarity score. In addition, existing unsupervised graph similarity learning methods are mainly clustering-based, which ignores the valuable information embodied in graph pairs. To this end, we propose a contrastive graph matching network (CGMN) for self-supervised graph similarity learning in order to calculate the similarity between any two input graph objects. Specifically, we generate two augmented views for each graph in a pair respectively. Then, we employ two strategies, namely cross-view interaction and cross-graph interaction, for effective node representation learning. The former is resorted to strengthen the consistency of node representations in two views. The latter is utilized to identify node differences between different graphs. Finally, we transform node representations into graph-level representations via pooling operations for graph similarity computation. We have evaluated CGMN on eight real-world datasets, and the experiment results show that the proposed new approach is superior to the state-of-the-art methods in graph similarity learning downstream tasks.\",\n \"Title: Enhancing Knowledge Selection for Grounded Dialogues via Document Semantic Graphs\\nYear: 2022\\n\\nAbstract: Providing conversation models with background knowledge has been shown to make open-domain dialogues more informative and engaging. Existing models treat knowledge selection as a sentence ranking or classification problem where each sentence is handled individually, ignoring the internal semantic connection among sentences in the background document. In this work, we propose to automatically convert the background knowledge documents into document semantic graphs and then perform knowledge selection over such graphs. Our document semantic graphs preserve sentence-level information through the use of sentence nodes and provide concept connections between sentences. We jointly apply multi-task learning for sentence-level and concept-level knowledge selection and show that it improves sentence-level selection. Our experiments show that our semantic graph-based knowledge selection improves over sentence selection baselines for both the knowledge selection task and the end-to-end response generation task on HollE and improves generalization on unseen topics in WoW.\",\n \"Title: RAW-GNN: RAndom Walk Aggregation based Graph Neural Network.\\nYear: 2022\\n\\nAbstract: Graph-Convolution-based methods have been successfully applied to representation learning on homophily graphs where nodes with the same label or similar attributes tend to connect with one another. Due to the homophily assumption of Graph Convolutional Networks (GCNs) that these methods use, they are not suitable for heterophily graphs where nodes with different labels or dissimilar attributes tend to be adjacent. Several methods have attempted to address this heterophily problem, but they do not change the fundamental aggregation mechanism of GCNs because they rely on summation operators to aggregate information from neighboring nodes, which is implicitly subject to the homophily assumption. Here, we introduce a novel aggregation mechanism and develop a RAndom Walk Aggregation-based Graph Neural Network (called RAW-GNN) method. The proposed approach integrates the random walk strategy with graph neural networks. The new method utilizes breadth-first random walk search to capture homophily information and depth-first search to collect heterophily information. It replaces the conventional neighborhoods with path-based neighborhoods and introduces a new path-based aggregator based on Recurrent Neural Networks. These designs make RAW-GNN suitable for both homophily and heterophily graphs. Extensive experimental results showed that the new method achieved state-of-the-art performance on a variety of homophily and heterophily graphs.\",\n \"Title: Orbital angular momentum-encoded quantum digital signature over atmospheric channel\\nYear: 2022\\n\\nAbstract: Quantum digital signature (QDS) ensures the integrity of a classical message and the authenticity of its sender based on information-theoretical limits and quantum mechanical mechanisms. The existing polarization-encoded or phase-encoded QDS protocols are susceptible to the interference from non-ideal factors in the quantum channel, resulting in the mismatch of the measurement reference frame. The channel capacities of these protocols are also limited, reducing the practicability of these protocols. To overcome these defects, we present a time-bin orbital angular momentum (OAM)-encoded continuous variable QDS (CVQDS) protocol, which exploits the rotation invariance of the vortex beam in the transmission direction to circumvent the real-time calibration of the reference frame and utilizes the infinite-dimensional eigenstate characteristics of the OAM states to beat the channel capacity limit. Security analysis and numerical simulations show that it is feasible to use this protocol to sign 1-bit message in atmospheric channel, and the channel capacity can be improved by increasing the dimensionality of the OAM code.\",\n \"Title: Improving Bot Response Contradiction Detection via Utterance Rewriting.\\nYear: 2022\\n\\nAbstract: Though chatbots based on large neural models can often produce fluent responses in open domain conversations, one salient error type is contradiction or inconsistency with the preceding conversation turns. Previous work has treated contradiction detection in bot responses as a task similar to natural language inference, e.g., detect the contradiction between a pair of bot utterances. However, utterances in conversations may contain co-references or ellipsis, and using these utterances as is may not always be sufficient for identifying contradictions. This work aims to improve the contradiction detection via rewriting all bot utterances to restore antecedents and ellipsis. We curated a new dataset for utterance rewriting and built a rewriting model on it. We empirically demonstrate that this model can produce satisfactory rewrites to make bot utterances more complete. Furthermore, using rewritten utterances improves contradiction detection performance significantly, e.g., the AUPR and joint accuracy scores (detecting contradiction along with evidence) increase by 6.5% and 4.5% (absolute increase), respectively.\",\n \"Title: A Systematic Evaluation of Response Selection for Open Domain Dialogue.\\nYear: 2022\\n\\nAbstract: Recent progress on neural approaches for language processing has triggered a resurgence of interest on building intelligent open-domain chatbots. However, even the state-of-the-art neural chatbots cannot produce satisfying responses for every turn in a dialog. A practical solution is to generate multiple response candidates for the same context, and then perform response ranking/selection to determine which candidate is the best. Previous work in response selection typically trains response rankers using synthetic data that is formed from existing dialogs by using a ground truth response as the single appropriate response and constructing inappropriate responses via random selection or using adversarial methods. In this work, we curated a dataset where responses from multiple response generators produced for the same dialog context are manually annotated as appropriate (positive) and inappropriate (negative). We argue that such training data better matches the actual use case examples, enabling the models to learn to rank responses effectively. With this new dataset, we conduct a systematic evaluation of state-of-the-art methods for response selection, and demonstrate that both strategies of using multiple positive candidates and using manually verified hard negative candidates can bring in significant performance improvement in comparison to using the adversarial training data, e.g., increase of 3% and 13% in Recall@1 score, respectively.\",\n \"Title: Community Detection in Social Networks Considering Social Behaviors\\nYear: 2022\\n\\nAbstract: The study of community detection in networks has drawn great attention in recent years. To find communities and to understand community semantics, both network topology and network content are utilized. Unfortunately, none of them can explain the driving factors of generating community structure with semantics, which is significant for understanding the mechanisms of community generation. Our observations on a large number of networks show that specific user social behaviors are underlying factors for the generation of community structure. We exploit four types of social behaviors that widely exist in networks, i.e., reciprocity of interactions, posting preference, multitopic preference, and temporal variation of topics. We investigate their impacts on the formation process of links and content in networks, during which communities with topics form. Our analysis shows that they are highly related to community structure. Consequently, a generative community detection model SBCD (social behavior-based community detection) is proposed by combining network topology and content, in which the above social behaviors play a core role. The model is evaluated on two real datasets. The experimental results show that SBCD outperforms state-of-the-art baselines. Finally, a case study illustrates several significant observations with respect to the proposed social behaviors.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":91,"string":"91"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13556,"string":"13,556"}}},{"rowIdx":13557,"cells":{"question_id":{"kind":"string","value":"Aaron Harwood"},"question":{"kind":"string","value":"Generate an abstract for the title \"ScoreTree: a decentralised framework for credibility management of user-generated content\" using the following items: 1. Peer-to-peer \n2. User-generated content \n3. Credibility \n4. Decentralized \n5. ScoreTree"},"answer":{"kind":"string","value":"Peer-to-peer applications are used in sharing User-Generated Content (UGC) on the Internet and there is a significant need for UGC to be analysed for credibility/quality. A number of schemes have been proposed for deriving credibility of content items by analysing users' feedback, mostly using centralised computations and/or semi-decentralised approaches. In this paper, we propose our P2P schema, ScoreTree, that decentralises a relatively complex credibility management algorithm by aggregating distributed evaluations and delivering an estimate of credibility for each interested content item. Our experiments show that our schema compares favourably with existing decentralised approaches including a gossip message based implementation of ScoreFinder, and a widely adopted P2P application called Vuze."},"correct_answer":{"kind":"string","value":"Peer-to-peer applications are used in sharing User-Generated Content (UGC) on the Internet and there is a significant need for UGC to be analysed for credibility/quality. A number of schemes have been proposed for deriving credibility of content items by analysing users' feedback, mostly using centralised computations and/or semi-decentralised approaches. In this paper, we propose our P2P schema, ScoreTree, that decentralises a relatively complex credibility management algorithm by aggregating distributed evaluations and delivering an estimate of credibility for each interested content item. Our experiments show that our schema compares favourably with existing decentralised approaches including a gossip message based implementation of ScoreFinder, and a widely adopted P2P application called Vuze."},"documents":{"kind":"list like","value":["Title: ScoreFinder: A method for collaborative quality inference on user-generated content\nYear: 2010\n\nAbstract: User-generated content is quickly becoming the greatest source of information on the World Wide Web. Shared content items are initially considered unconfirmed in the sense that their credibility has not yet been established. Conventional, centralized confirmation of credibility is infeasible at the Internet scale and so making use of the annotators themselves to evaluate each item is essential. However, users usually differ in opinions to the same item, and the existence of bias, variance and maliciousness makes the problem of aggregating opinions more difficult. Addressing this problem, we propose the use of an Author-Annotator model with an iterative algorithm, called ScoreFinder, for inferring credibility by ranking shared items. In order to reduce the influence from a variety of error sources, we identify reliable users on each topic, and adaptively aggregate scores from them. Moreover, we transform the users' input to remove errors/anomalies, by identifying patterns of misbehaviour learned from a real data set. We show how our algorithm performs on both real data sets and synthetic data sets, and a significant improvement was achieved in the experiment.","Title: Impacts of Asynchrony on Epidemic-Style Aggregation Protocols\nYear: 2010\n\nAbstract: The large scale and dynamic nature of a distributed system makes it difficult to collect the attributes of the individual nodes in the network. In these systems, often an aggregate (e.g. AVG, COUNT, MIN, MAX, SUM etc) of these attributes is adequate. Epidemic-style protocols are one of the popular approaches to estimate aggregates in such systems. In existing epidemic-style aggregation protocols the accuracy of the estimated aggregate at local nodes heavily depends upon synchronization of aggregation rounds. To enforce synchronization in these protocols, length of aggregation round should be long enough so that all the nodes in the system complete their aggregation information exchange. In this paper, we study the impacts of asynchrony in epidemic-style aggregation protocols. We present a simple asynchronous technique to estimate system aggregates in a distributed system. Based upon this technique, we analyze two popular existing epidemic-style aggregation protocols, Push-Pull and Push-Sum. Through detailed simulations, we evaluate accuracy and cost of asynchronous version of these protocols. We found that to obtain an estimate of the true system aggregate, aggregation protocols do not need to be synchronized and hence an efficient estimate can be obtained in lesser time.","Title: Flow Generation for IP/ATM Label-Switched Routing over Random Networks\nYear: 2001\n\nAbstract: We propose a dynamic periodicity detector (DPD) for the estimation of periodicities in data series obtained from the execution of applications. We analyze the algorithm used by the periodicity detector and its performance on a number of data streams. ...","Title: A serverless 3D world\nYear: 2004\n\nAbstract: Online multi-participant virtual-world systems have attracted significant interest from the Internet community but are hindered by their inability to efficiently support interactivity for a large number of participants. Current solutions divide a large virtual-world into a few mutually exclusive zones, with each zone controlled by a different server, and/or limit the number of participants per server or per virtual-world. Peer-to-Peer (P2P) systems are known to provide excellent scalability in a networked environment (one peer is introduced to the system by each participant), however current P2P applications can only provide file sharing and other forms of relatively simple data communications. In this paper, we present a generic 3D virtual-world application that runs on a P2P network with no central administration or server. Two issues are addressed by this paper to enable such a spatial application on a P2P network. First, we demonstrate how to index and query a 3D space on a dynamic distributed network. Second, we show how to build such a complex application from the ground level of a P2P routing algorithm. Our work leads to new directions for the development of online virtual-worlds that we believe can be used for many government, industry, and public domain applications.","Title: Directions for Peer-to-Peer based mobile pervasive augmented reality gaming\nYear: 2007\n\nAbstract: Pervasive augmented reality gaming and mobile P2P are both attracting significant focus and much progress has occurred recently. Augmented reality games extend reality with virtual elements to enhance game experience and mobile P2P networks provide autonomous, self-organizing, scalable and robust communication platforms. We review the technological challenges in both mobile P2P computing and AR domains by investigating some existing related works. Then we propose possible directions of the mobile AR collaborative gaming over P2P networks and show its advantages over other AR gaming and discuss the feasibility and future requirements regarding the support for this type of game.","Title: An Energy Efficient Distributed Scheduling Scheme for Wireless Mesh Networks\nYear: 2013\n\nAbstract: The usefulness of wireless mesh networks (WMNs) highly depends on the effective usage of frequency spectrum, scalability and energy efficiency of the underlying technologies [1]. While maximizing network throughput for given frequency spectrum, and scalability have been extensively studied in multi-hop environments, energy efficiency has not received due attention. In this paper, we propose an enhanced pseudo random access (EPRA) scheme based on the IEEE 802.16-2004 distributed mesh framework to improve energy efficiency. Our proposal addresses issues that we identify with the existing election based transmission timing (EBTT) mechanism of IEEE 802.16-2004. Our extensive simulation results show a very significant improvement in the reception energy efficiency by up to 85%. Moreover, notable improvements are also obtained in the transmission energy efficiency. While we focus mainly on the aspect of energy efficiency, the proposed enhancement also helps to improve the spatial reuse of the frequency spectrum.","Title: Indexing distributed complex data for complex queries\nYear: 2004\n\nAbstract: Peer-to-peer networks are becoming a common form of online data exchange. Querying data, mostly files, using keywords on peer-to-peer networks is well-known. But users cannot perform many types of queries on complex data and on many of the attributes of the data on such networks other than mostly exact-match queries. We introduce a distributed hashing-based index for enabling more powerful accesses on complex data over peer-to-peer networks that we expect to be commonly deployed for digital government applications. Preliminary experiments show that our index scales well and we believe that it can be extended to obtain similar indices for many other data types for performing various complex queries, such as range queries.","Title: A Peer-to-Peer Browsable File Index using a Popularity Based Global Namespace\nYear: 2007\n\nAbstract: The distribution of files using decentralized, peer-to- peer (P2P) systems, has significant advantages over cen- tralized approaches. It is however more difficult to settle on the best approach for file sharing. Most file sharing systems are based on query string searches, leading to a relatively simple but inefficient broadcast or to an ef- ficient but relatively complicated index in a structured environment. In this paper we use a browsable peer- to-peer file index consisting of files which serve as di- rectory nodes, interconnecting to form a directory net- work. We implemented the system based on BitTorrent and Kademlia. The directory network inherits all of the advantages of decentralization and provides browsable, efficient searching. To avoid conflict between users in the P2P system while also imposing no additional re- strictions, we allow multiple versions of each directory node to simultaneously exist - using popularity as the basis for default browsing behavior. Users can freely add files and directory nodes to the network. We show, using a simulation of user behavior and file quality, that the popularity based system consistently leads users to a high quality directory network; above the average qual- ity of user updates.","Title: Analytical formulations of Peer-to-Peer Connection Efficiency\nYear: 2003\n\nAbstract: Use of Peer-to-Peer (P2P) service networks in- troduces a new communication paradigm be- cause peers are both clients and servers and so each peer may provide/request services to/from other peers. Empirical studies of P2P networks have been undertaken and reveal useful charac- teristics. However there is to date little analyti- cal work to describe P2P networks with respect to their communication paradigm and their in- terconnections. This paper provides an analyti- cal formulation and optimisation of peer connec- tion efficiency, in terms of minimising the frac- tion of wasted connection time. Peer connection efficiency is analysed for both a uni- and multi- connected peer. Given this fundamental optimi- sation, the paper optimises the number of con- nections that peers should make use of as a func- tion of network load, in terms of minimising the total queue size that requests in the P2P network experience. The results of this paper provide a basis for engineering high performance P2P in- terconnection networks. The optimisations are useful for reducing bandwidth and power con- sumption, e.g. in the case of peers being mo- bile devices with a limited power supply. Also these results could be used to determine when a (virtual) circuit should be switched to support a","Title: Realizing the e-science desktop peer using a peer-to-peer distributed virtual machine middleware\nYear: 2006\n\nAbstract: Emerging e-Science applications face the challenge of providing both high performance and low maintenance cost infrastructure. We propose the e-Science Desktop Peer system and a prototype middleware implementation called P2P-DVM to address this challenge, that is designed to allow scientists' resources to be readily shared over the Internet in a Peer-to-Peer (P2P) manner. Our P2P-DVM currently supports the Message Passing Interface and Bulk Synchronous Processing model over an adaptive P2P network and we introduce a Programming Environment Abstraction Layer to allow users to quickly adapt other programming environments to our P2P approach. P2P-DVM provides decentralized coordinated checkpoint and restart functionality, message passing over P2P networks, and distributed storage for scalability. Our experimental results on the PlanetLab testbed shows support for our work.","Title: On Problem for Aggregate Node Selection for Unstructured Overlay Networks\nYear: 2010\n\nAbstract: With the success of the peer-to-peer systems, there is an emerging trend to develop an aggregation service that can provide global information about these distributed and decentralized overlays. Existing aggregation protocols have focused on providing an efficient and cost effective solution to compute aggregates. In this paper, we are interested to search the node based on an aggregate query over rapidly changing data e.g., to seek a node with minimum/maximum workload in the network. Solution to such queries can benefit number of applications such as, node-ranking, reputation building, resource scheduling, etc. Here we propose a novel protocol to solve such queries. Our proposed protocol is reactive and pull-based in nature. It intrinsically builds and manages a routing table which routes aggregate query to the desired node in the system. We evaluated our protocol for accuracy, efficiency and communication cost through detailed simulations. We also discuss effects of an explicit routing table management scheme and show that this explicit scheme causes our protocol to work better in presence of churn and node failures.","Title: An analytical model for performance evaluation in sparse mobile ad hoc networks\nYear: 2009\n\nAbstract: Routing protocols for sparse mobile ad hoc networks are becoming important due to applications such as vehicular networks and emergency services. Simulation based performance evaluation of a given protocol is computationally demanding, requiring a significant amount of time and resources. For this reason, analytical approaches to performance evaluation are desirable; however, such approaches are challenging to develop. In this paper, we present an analytical model to calculate two important performance metrics, namely throughput and average end-to-end delay, for a sparse mobile ad hoc network routing protocol. In our protocol, independent mobile nodes move in a bounded region and communicate indirectly through passive RFID tags that act like mailboxes. We analyze our protocol for two different buffer policies by providing analytical equations to calculate throughput and delay for each buffer policy. Finally, we support our equations via simulation.","Title: A comparative study on Peer-to-Peer failure rate estimation\nYear: 2007\n\nAbstract: The robustness of Peer-to-Peer systems is challenged by its highly dynamic nature. Frequent peer failure and departure events introduce uncertainty for which is considered exceptional in traditional distributed systems. The difficulty of monitoring such large scale networks is further exacerbated because it has to be done in a completely decentralized way for both scalability and reliability concerns. Some methods for estimating peer failure rate have been applied in Peer-to-Peer systems, however their comparative performance has not yet been reported in the literature. We simulate three different failure rate estimation methods and compare their accuracy and response time with respect to sample size, stabilization interval and neighbour set size. We conclude that the Maximum Likelihood Method introduced is better than the Failure Frequency based Methods commonly used in current Peer-to-Peer systems.","Title: Diffusive Load Balancing of Loosely-Synchronous Parallel Programs over Peer-to-Peer Networks\nYear: 2004\n\nAbstract: The use of under-utilized Internet resources is widely recognized as a viable\nform of high performance computing. Sustained processing power of roughly 40T\nFLOPS using 4 million volunteered Internet hosts has been reported for\nembarrassingly parallel problems. At the same time, peer-to-peer (P2P) file\nsharing networks, with more than 50 million participants, have demonstrated the\ncapacity for scale in distributed systems. This paper contributes a study of\nload balancing techniques for a general class of loosely-synchronous parallel\nalgorithms when executed over a P2P network. We show that decentralized,\ndiffusive load balancing can be effective at balancing load and is facilitated\nby the dynamic properties of P2P. While a moderate degree of dynamicity can\nbenefit load balancing, significant dynamicity hinders the parallel program\nperformance due to the need for increased load migration. To the best of our\nknowledge this study provides new insight into the performance of\nloosely-synchronous parallel programs over the Internet.","Title: Improving performance in delay/disruption tolerant networks through passive relay points\nYear: 2012\n\nAbstract: In this paper, we study the case of a limited number of mobile nodes trying to communicate in a large geographic area, forming a delay/disruption tolerant network (DTN). In such networks the mobile nodes are disconnected for significantly long periods of time. Traditional routing protocols proposed for mobile ad hoc networks or mesh networks, which assume at least one path between each source and destination, are ineffective in DTNs. One approach to improve communication is through gossip based protocols because these protocols do not rely on a fixed path. Another approach is to control the movement of the mobile nodes and/or use special mobile nodes called ferry nodes. Others try to employ a fixed infrastructure including stationary relay points. One scheme in stationary relay point approach is to use base stations as relay points which need their own power supply. In this paper, we study a passive approach where mobile nodes deposit/retrieve messages to/ from known stationary locations in the geographic region. Messages are delivered from a source by being deposited at one or more locations that are later visited by the destination. A proposed implementation of our approach using read/writable passive Radio Frequency Identification (RFID) tags, one per point location, is considered in this work. Passive RFID technology is desirable because it operates wirelessly and without the need for attached power. Our simulation results indicate that our approach can achieve competitive message delay and delivery rates. We also demonstrate several techniques for optimizing the stationary relay node placement, namely relay pruning, probability based relay distribution and a genetic algorithm; the genetic algorithm is shown to provide the best solutions to this problem.","Title: Batched Circuit Switched Routing for Efficient Service of Requests\nYear: 2000\n\nAbstract: Circuit switching protocols were largely developed in light of ATM virtual circuits and also for the case of all-optical networks. The merging of ATM and IP packet switching technology may banish true circuit switching in the digital domain due to fast switching networks using TDM. Even so, the characteristics of all-optical light paths (an optical circuit switched path) are so desirable that digital WAN's using ATM may eventually be completely augmented by optical WAN's using a hybrid opto-electronic or complete optical switching fabric. Afundamental problem is servicing a large number of requests while maintaining a low blocking probability where a request is blocked if the path it requires is in use by a circuit. We present a simple routing method, namely Batched Circuit Switching (BCS), that batches requests according to the rate of arrival and service time. Compared to orthodox Circuit Switching (CS), BCS provides superior request service times due to the observation that many requests may traverse the same circuit (or channel) during the batch time interval, rather than being blocked by preceding requests (as is the case for CS). Thus, BCS exhibits better utilization of thenetwork resources than CS. We demonstrate the requirements for this system to be stable, namely that the batch size must not increase to infinity before being serviced and the relationship between \\lambda (the arrival rate of requests) and µ (the service rate ofrequests) to ensure stability.","Title: Coordinated load management in Peer-to-Peer coupled federated grid systems\nYear: 2012\n\nAbstract: This paper proposes a coordinated load management protocol for Peer-to-Peer (P2P) coupled federated Grid systems. The participants in the system, such as the resource providers and the consumers who belong to multiple control domains, work together to enable a coordinated federation. The coordinated load management protocol embeds a logical spatial index over a Distributed Hash Table (DHT) space for efficient management of the coordination objects; the DHT-based space serves as a kind of decentralized blackboard system. We show that our coordination protocol has a message complexity that is logarithmic to the number of nodes in the system, which is significantly better than existing broadcast based coordination protocols.The proposed load management protocol can be applied for efficiently coordinating resource brokering services of distributed computing systems such as grids and PlanetLab. Resource brokering services are the main components that control the way applications are scheduled, managed and allocated in a distributed, heterogeneous, and dynamic Grid computing environments. Existing Grid resource brokers, e-Science application work-flow schedulers, operate in tandem but still lack a coordination mechanism that can lead to efficient application schedules across distributed resources. Further, lack of coordination exacerbates the utilization of various resources (such as computing cycles and network bandwidth). The feasibility of the proposed coordinated load management protocol is studied through extensive simulations.","Title: Incorporating Fault Tolerance with Replication on Very Large Scale Grids\nYear: 2007\n\nAbstract: Providing fault tolerance for message passing parallel application on a distributed environment is a rule rather than an exception. A node failure can cause the whole computation to stop and has to be restarted from the begin- ning if no fault tolerance is available. However, introducing fault tolerance has some overhead on speedup that can be achieved. In this paper, we introduce a new technique called replication with cross-over packets for reliability and to in- crease fault tolerance over Very Large Scale Grids (VLSG). This technique has two pronged effect of avoiding single point of failure and single link of failure. We incorporate this new technique into the L-BSP model and show the pos- sible speedup of parallel process. We also derive the achiev- able speedup for some fundamental parallel algorithms us- ing this technique.","Title: P2P-Tuple: Towards a Robust Volunteer Computing Platform\nYear: 2009\n\nAbstract: The evolution of emerging volunteer computing systems is placing an ever greater emphasis on the use of a loosely coupled decentralized architecture, mainly to address concerns of scalability and robustness. We propose to build the next generation volunteer computing systems on top of well studied peer-to-peer techniques to fully take advantage of its decentralized characteristic and its very large, shared data storage capacity. This paper sketches the design of such a completely decentralized and fault tolerant volunteer computing system by using a peer-to-peer based tuple space formed by volunteer peers. Both simulation and software prototype evaluation results show support for the proposed design even when all peers are having a high churn rate of Mean Time Before Failure of 2 hours.","Title: Using a distributed quadtree index in peer-to-peer networks\nYear: 2007\n\nAbstract: Peer-to-peer (P2P) networks have become a powerful means for online data exchange. Currently, users are primarily utilizing these networks to perform exact-match queries and retrieve complete files. However, future more data intensive applications, such as P2P auction networks, P2P job-search networks, P2P multiplayer games, will require the capability to respond to more complex queries such as range queries involving numerous data types including those that have a spatial component. In this paper, a distributed quadtree index that adapts the MX-CIF quadtree is described that enables more powerful accesses to data in P2P networks. This index has been implemented for various prototype P2P applications and results of experiments are presented. Our index is easy to use, scalable, and exhibits good load-balancing properties. Similar indices can be constructed for various multidimensional data types with both spatial and non-spatial components.","Title: Decentralisation of scorefinder: a framework for credibility management on user-generated contents\nYear: 2010\n\nAbstract: User-generated content (UGC) from Internet users has significant value only when its credibility can be established A basic approach to establishing credibility is to take an average of scores from annotators, while more sophisticated approaches have been used to eliminate anomalous scoring behaviour by giving different weights to scores from different annotator profiles A number of applications such as file sharing and article reviewing use a decentralised architecture While computing a weighted average of static values in a decentralised application is well studied, sophisticated UGC algorithms are more complicated since source values to be aggregated and their weights may change in time In our work we consider a centralised credibility management algorithm, ScoreFinder, as an example, and show both structured and unstructured approaches for computing time-dependent weighted average values in peer-to-peer (P2P) networks Experimental results on two real data sets demonstrate that our approaches converge and deliver results comparable to those from the centralised version of ScoreFinder.","Title: A Case for Cooperative and Incentive-Based Coupling of Distributed Clusters\nYear: 2006\n\nAbstract: Interest in grid computing has grown significantly over the past five years. Management of distributed cluster resources is a key issue in grid computing. Central to management of resources is the effectiveness of resource allocation, as it determines the overall utility of the system. In this paper, we propose a new grid system that consists of grid federation agents which couple together distributed cluster resources to enable a cooperative environment. The agents use a computational economy methodology, that facilitates QoS scheduling, with a cost-time scheduling heuristic based on a scalable, shared federation directory. We show by simulation, while some users that are local to popular resources can experience higher cost and/or longer delays, the overall users' QoS demands across the federation are better met. Also, the federation's average case message passing complexity is seen to be scalable, though some jobs in the system may lead to large numbers of messages before being scheduled","Title: A case for cooperative and incentive-based federation of distributed clusters\nYear: 2008\n\nAbstract: Research interest in Grid computing has grown significantly over the past five years. Management of distributed resources is one of the key issues in Grid computing. Central to management of resources is the effectiveness of resource allocation as it determines the overall utility of the system. The current approaches to brokering in a Grid environment are non-coordinated since application-level schedulers or brokers make scheduling decisions independently of the others in the system. Clearly, this can exacerbate the load sharing and utilization problems of distributed resources due to sub-optimal schedules that are likely to occur. To overcome these limitations, we propose a mechanism for coordinated sharing of distributed clusters based on computational economy. The resulting environment, called Grid-Federation, allows the transparent use of resources from the federation when local resources are insufficient to meet its users' requirements. The use of computational economy methodology in coordinating resource allocation not only facilitates the Quality of Service (QoS)-based scheduling, but also enhances utility delivered by resources. We show by simulation, while some users that are local to popular resources can experience higher cost and/or longer delays, the overall users' QoS demands across the federation are better met. Also, the federation's average case message-passing complexity is seen to be scalable, though some jobs in the system may lead to large numbers of messages before being scheduled.","Title: On Placement of Passive Stationary Relay Points in Delay Tolerant Networking\nYear: 2011\n\nAbstract: Recently, there has been focus on augmenting Delay/Disruption Tolerant Networks (DTNs) with easily deployable stationary relay nodes making an unconnected infrastructure to facilitate the data delivery by increasing forwarding opportunities. Relay nodes are capable of downloading, storing, and forwarding the data messages from/to the mobile nodes. Placing the relay nodes is an important issue in DTNs as the performance of the network is dependent to their positions. Relay placement is an NP-hard problem hence it makes it a more complicated issue in DTNs. Existing works in the literature are based on simulation which are suffering from computational complexities dictated by simulation. Moreover, they are optimizing the relay placement only based on specific scenarios. In this paper, we propose a generic analytical model in order to evaluate the performance of DTNs in presence of relay nodes. Our model is dependent on the mobile nodes' mobility pattern, and we consider the case when the mobile nodes move according to the random waypoint model. In order to use the proposed model for placing the relays efficiently, we utilize two heuristic approaches. The first approach is based on optimization of the network performance using simulated annealing and the second one relies on a greedy approach to find the best location for each relay one at a time. Our simulation results show that our approaches outperform the simulation based approaches in terms of data delivery performance.","Title: Distributed scheduling schemes for wireless mesh networks: A survey\nYear: 2013\n\nAbstract: An efficient scheduling scheme is a crucial part of Wireless Mesh Networks (WMNs)—an emerging communication infrastructure solution for autonomy, scalability, higher throughput, lower delay metrics, energy efficiency, and other service-level guarantees. Distributed schedulers are preferred due to better scalability, smaller setup delays, smaller management overheads, no single point of failure, and for avoiding bottlenecks. Based on the sequence in which nodes access the shared medium, repetitiveness, and determinism, distributed schedulers that are supported by wireless mesh standards can be classified as either random, pseudo-random, or cyclic schemes. We performed qualitative and quantitative studies that show the strengths and weaknesses of each category, and how the schemes complement each other. We discuss how wireless standards with mesh definitions have evolved by incorporating and enhancing one or more of these schemes. Emerging trends and research problems remaining for future research also have been identified.","Title: Cooperative repair of wireless broadcasts\nYear: 2011\n\nAbstract: Wireless broadcasting systems, such as Digital Video Broadcasting (DVB), are subject to signal degradation, having an effect on end users' reception quality. Clearly, reception quality can be improved by increasing signal strength. This, however, comes at a significantly increased energy use, with adverse environmental and financial consequences, notably in either sparsely populated rural regions, or overly built and difficult to penetrate dense urban areas. This paper discusses our ongoing work on an alternative approach to improving reception quality, based on the collaborative repair of lossy packet streams among the community of DVB viewers. We present our main idea, the crucial design decisions, the algorithm, as well as preliminary results demonstrating the feasibility and efficiency of this approach.","Title: Event Based Simulator for Parallel Computing over the Wide Area Network for Real Time Visualization\nYear: 2009\n\nAbstract: As the computational requirement of applications in computational science continues to grow tremendously, the use of computational resources distributed across the Wide Area Network (WAN) becomes advantageous. However, not all applications can be executed over the WAN due to communication overhead that can drastically slowdown the computation. In this paper, we introduce an event based simulator to investigate the performance of parallel algorithms executed over the WAN. The event based simulator known as SIMPAR (SIMulator for PARallel computation), simulates the actual computations and communications involved in parallel computation over the WAN using time stamps. Visualization of real time applications require steady stream of processed data flow for visualization purposes. Hence, SIMPAR may prove to be a valuable tool to investigate types of applications and computing resource requirements to provide uninterrupted flow of processed data for real time visualization purposes. The results obtained from the simulation show concurrence with the expected performance using the L-BSP model.","Title: Stable High-Capacity One-Hop Distributed Hash Tables\nYear: 2006\n\nAbstract: Most research on Distributed Hash Tables (DHTs) assumes ephemeral, lightly loaded deployments. Each node has a lifetime of a few hours and initiates a lookup once every few seconds or minutes. However, in giant internet data centers, each node has a lifetime of weeks or months and initiates hundreds or thousands of lookups every second. In such an environment, one-hop DHTs are superior to multi-hop DHTs. They use lookup bandwidth more efficiently. We qualify conflicting research to show that a single onehop DHT can indeed scale to at least a few hundred thousand nodes in stable, high-capacity enterprise networks. Two new designs are presented: One Hop Sites (1HS), a high-capacity DHT tailored for site redundancy; and the One Hop Federation (1HF), a global, hierarchic DHT that resolves an open latency problem. For both, the analysis a) confirms linear scalability to at least a few hundred thousand nodes and b) identifies the most sensitive design parameters.","Title: Global Detection of Flooding-Based DDoS Attacks Using a Cooperative Overlay Network\nYear: 2010\n\nAbstract: Flooding-based Distributed Denial of Service (DDoS) attacks present a serious threat to the stability of the Internet. Identifying the attacks rapidly and accurately is significant for the efficient operation of Internet applications and services. Recent observations in the U.S. indicate a significant increase of cyber attacks on U.S. military information systems in 2009. Current technologies are still unable to withstand large-scale DDoS attacks. Single point detection and response is a first step to defeat such distributed attacks. Distributed global defense systems, using a coordinated effort, go much further towards thwarting such attacks. In this paper, we propose a distributed defense infrastructure to detect DDoS attacks globally using a cooperative overlay network and a gossip-based information exchange protocol. Our NS2 based simulation results show that the proposed solution can detect attacks with a detection rate as high as 0.99 with false alarms below 0.01. This compares favorably against other widely known methods including change-point detection, TTL analysis and wavelet analysis.","Title: Using fundamental electrical theory for varying time quantum uni-processor scheduling\nYear: 2001\n\nAbstract: Given the total number of instructions to be completed on a uni-processor system and the average cycle time per instruction we introduce a method of calculating time quantum allocation to individual fine grain tasks. The main theory behind our method is based on fundamental equations describing electrical phenomenon. We show how electric circuit analysis can be used to describe the fundamental scheduling problem, and provide a framework for defining more elaborate scheduling problems such as multiprocessor and multicomputer task scheduling. As a matter of physical soundness we demonstrate through unit derivation that our electrical analogy provides proper physical quantities that are supported by current literature. Our analysis shows that variable time round-robin scheduling (VTRR) provides a more appropriate means of scheduling fine-grain tasks than constant time round-robin scheduling (CTRR). We prove that, our VTRR scheduler always completes at least one task per cycle . We show through numerical comparisons some differences between VTRR and CTRR performance.","Title: Mean Field Models of Message Throughput in Dynamic Peer-to-Peer Systems\nYear: 2007\n\nAbstract: The churn rate of a peer-to-peer system places direct limitations on the rate at which messages can be effectively communicated to a group of peers. These limitations are independent of the topology and message transmission latency. In this paper we consider a peer-to-peer network, based on the Engset model, where peers arrive and depart independently at random. We show how the arrival and departure rates directly limit the capacity for message streams to be broadcast to all other peers, by deriving mean field models that accurately describe the system behavior. Our models cover the unit and more general k buffer cases, i.e. where a peer can buffer at most k messages at any one time, and we give results for both single and multi-source message streams. We define coverage rate as peer-messages per unit time, i.e. the rate at which a number of peers receive messages, and show that the coverage rate is limited by the churn rate and buffer size. Our theory introduces an Instantaneous Message Exchange (IME) model and provides a template for further analysis of more complicated systems. Using the IME model, and assuming random processes, we have obtained very accurate equations of the system dynam- ics in a variety of interesting cases, that allow us to tune a peer-to-peer system. It remains to be seen if we can maintain this accuracy for general processes and when applying a non-instantaneous model.","Title: Fast Channel Establishment for IEEE 802.16 Wireless Mesh Networks\nYear: 2010\n\nAbstract: IEEE 802.16 based wireless mesh networks (WMNs) are a promising broadband access solution to support flexibility, cost effectiveness and fast deployment of the fourth generation infrastructure based wireless networks. Reducing the time for channel establishment is critical for low latency/interactive applications. In this paper, we propose two modifications to the election based transmission timing (EBTT) mechanism used in the coordinated distributed scheduling (CDS) within the IEEE 802.16 standard, which makes use of the standardized signaling messages and formats. Simulation results show that our proposed modifications results in significantly reduced channel establishment delay - up to 75% with low neighborhood densities - over the standard, and at least 10% with high neighborhood densities. Moreover, the proposed mechanisms are also applicable for the three-hop extended neighborhood CDS when required.","Title: An Adaptive Checkpointing Scheme for Peer-to-Peer Based Volunteer Computing Work Flows\nYear: 2008\n\nAbstract: Volunteer computing, sometimes called public resource computing, is an emerging computational model that is very suitable for work-pooled parallel processing. As more complex grid applications make use of work flows in their design and deployment it is reasonable to consider the impact of work flow deployment over a volunteer computing infrastructure. In this case, the inter work flow I/O can lead to a significant increase in I/O demands at the work pool server. A possible solution is the use of a peer-to-peer based parallel computing architecture to off-load this I/O demand to the workers; where the workers can fulfill some aspects of work flow coordination and I/O checking, etc. However, achieving robustness in such a large scale system is a challenging hurdle towards the decentralized execution of work flows and general parallel processes. To increase robustness, we propose and show the merits of using an adaptive checkpoint scheme that efficiently checkpoints the status of the parallel processes according to the estimation of relevant network and peer parameters. Based on our proposed mathematical checkpoint model, our scheme uses statistical data observed during runtime to dynamically make checkpoint decisions in a completely decentralized manner. The results of simulation show support for our proposed approach in terms of reduced required runtime.","Title: Map-based spatio-temporal interpolation in vehicle trajectory data using routing web-services\nYear: 2012\n\nAbstract: Interpolation of motion data is a challenging problem that is often overlooked by researchers when using GPS data with low sampling rate. In this paper we define a spatio-temporal query called Historical Spatio-Temporal Query (HSTQ). This query receives a time and returns a waypoint containing spatio-temporal data of a moving object at that particular time. To respond to the query for any given time (e.g. every second), we need to interpolate missing waypoints of the trajectory. Linear Interpolation (LI) is the most commonly used method although it can be grossly inaccurate. To deal with this problem, we propose a method called Map-Based Interpolation (MBI). This method uses routing web-services to find significant points (path) between two waypoints. Instead of using road networks, which is typically used by methods such as map-matching, we send a query to a routing web service and analyse the returned data to find missing waypoints. To be able to respond to queries sent to HSTQ for any input time within a trajectory period, we use a combination of MBI and LI (MBI+LI) to interpolate the missing data. We also propose two measures for comparing performance of our interpolation method with LI. Experimental results using realistic trajectories show significant improvement in quality and accuracy of interpolated and down-sampled trajectory data using HSTQ with MBI+LI in comparison to the earlier, widely used LI method.","Title: Lossy Bulk Synchronous Parallel Processing Model for Very Large Scale Grids\nYear: 2006\n\nAbstract: The performance of a parallel algorithm in a very large scale grid is significantly influenced by the underlyin g Internet protocols and inter-connectivity. Many grid program- ming platforms use TCP due to its reliability, usually with some optimizations to reduce its costs. However, TCP does not perform well in a high bandwidth and high delay network environment. On the other hand, UDP is the fastest protocol avail- able because it omits connection setup process, acknowledgments and retransmissions sacrificing reliable transfer. Many ne w bulk data transfer schemes using UDP for data transmission such as RBUDP, Tsunami, and SABUL have been introduced and shown to have better performance compared to TCP. In this paper, we consider the use of UDP and examine the relationship between packet loss and speedup with respect to the number of grid nodes. Our measurement suggests that packet loss rates between 5%- 15% on average are not uncommon between PlanetLab nodes that are widely distributed over the Internet. We show that transmitting multiple copies of same packet produces higher speedup. We show the minimum number of packet duplication required to maximize the possible speedup for a given number of nodes using a BSP based model. Our work demonstrates that by using an appropriate number of packet copies, we can increase performance of parallel program.","Title: Towards Parallel Computing on the Internet: Applications, Architectures, Models and Programming Tools\nYear: 2006\n\nAbstract: The development of Internet wide resources for general purpose parallel\ncomputing poses the challenging task of matching computation and communication\ncomplexity. A number of parallel computing models exist that address this for\ntraditional parallel architectures, and there are a number of emerging models\nthat attempt to do this for large scale Internet-based systems like\ncomputational grids. In this survey we cover the three fundamental aspects --\napplication, architecture and model, and we show how they have been developed\nover the last decade. We also cover programming tools that are currently being\nused for parallel programming in computational grids. The trend in conventional\ncomputational models are to put emphasis on efficient communication between\nparticipating nodes by adapting different types of communication to network\nconditions. Effects of dynamism and uncertainties that arise in large scale\nsystems are evidently important to understand and yet there is currently little\nwork that addresses this from a parallel computing perspective.","Title: A method of trading diameter for reduced degree to construct low cost interconnection networks\nYear: 1999\n\nAbstract: Classical network topology has identified many “bottom-up” approaches to designing low cost interconnection networks. The foremost figure of merit is considered to be cost, the product of degree and diameter. We argue that average cost, being the product of average degree and diameter, is more applicable than the conventional cost and propose a general method of Yap-down” network construction that provides networks of average cost 0(log, N) where B = o(N) is a bound on maximum degree and N is the number of nodes. From this we show an example topology that has averagecostofQ(,J~~~N). By doing so we examine a class of networks with constant average cost, that is, networks whose average cost is fixed to a constant value as the size of the network increases to infinity. We then identify those aspects that are undesirable, namely some nodes have infinite dcgree, and show methods of trading an increase in diameter for a reduction in degree.","Title: A support platform for event detection using social intelligence\nYear: 2012\n\nAbstract: This paper describes a system designed to support event detection over Twitter. The system operates by querying the data stream with a user-specified set of keywords, filtering out non-English messages, and probabilistically geolocating each message. The user can dynamically set a probability threshold over the geolocation predictions, and also the time interval to present data for.","Title: A distributed network architecture for robust internet voting systems\nYear: 2005\n\nAbstract: Web-based Internet voting services cannot be provided in a reliable manner where a publicly visible website forms the basis of interacting with voters and collecting votes, as the service cannot resist a distributed denial of service attack. The problem is a profound one for all interactive web services, however, an Internet voting session is a well defined and quite simple interaction and an interactive service similar to web-based voting is proposed. This architecture uses a P2P web cache as the source of a Java voting applet. The applet interacts with the voter via their web browser to collect a vote. The applet then uses a common P2P file sharing network to deposit encrypted votes for collection by the electoral returning officer. The issues of how to authenticate voters and how to provide receipting in such an asynchronous system are described along with premininary .ndings using a large P2P network established for this purpose. A successful demonstration of this approach may have implications for other mission critical web services which must collect information.","Title: Building and Querying a P2P Virtual World\nYear: 2006\n\nAbstract: Peer-to-Peer (P2P) systems are known to provide excellent scalability in a networked environment. One peer is introduced to the system by each participant. However current P2P applications can only provide file sharing and other forms of relatively simple data communications, and, in this paper, we demonstrate how this limitation can be bridged by indexing and querying a 3D virtual-world on a dynamic distributed network. We present an algorithm for 3D range queries as well as an algorithm for nearest neighbor queries. We also show how to build such a complex application from the ground level of a P2P routing algorithm.","Title: Topology Dissemination for Reliable One-Hop Distributed Hash Tables\nYear: 2009\n\nAbstract: Many distributed hash tables (DHTs) resolve lookups in O(\\log n) hops, where n is the number of nodes. One-hop DHTs give lower lookup latencies and lower lookup failure rates. However, it is hard to maintain large, wide-area one-hop topologies. We contribute aecast, a new topology dissemination algorithm for one-hop DHTs. It avoids expensive repair mechanisms and critical points of failure in existing one-hop DHTs. When a node discovers by anti-entropy that it has missed a topology update, it initiates \"controlled flooding,” sending the update to nodes in the multicast tree that also missed the update. We compare aecast with a widely cited epidemic multicasting algorithm, pbcast, by analysis and simulation. Aecast gives at least fivefold fewer out-of-date nodes on average within one round of a topology update. We support it with a fault-tolerant topology agreement protocol, so that only legitimate topology changes propagate throughout the overlay. Consequently, we argue that one-hop DHTs deserve greater attention for Internet applications in which reasonably reliable nodes carry high lookup loads.","Title: CAESAR: middleware for complex service-oriented peer-to-peer applications\nYear: 2007\n\nAbstract: Recent research advances in Peer-to-Peer (P2P) computing have enabled the P2P paradigm to be used for developing complex applications beyond file sharing and data storage. These applications have drawn significant benefits, specifically scalability and low cost, from the P2P paradigm. However, the current approach for designing P2P applications introduce issues that prevent the development of high quality complex P2P applications. These issues, namely tight coupling to P2P protocols, limited logic sharing between peers and complicated recovery processes, motivate us to introduce a service-oriented architecture for P2P applications. We have developed a middleware called CAESAR to support the development of service-oriented P2P applications applying the principles of abstraction, dynamic binding, loose coupling and information hiding. In this paper, we discuss the design principles and the components of CAESAR middleware, as well as our experiences in using CAESAR to develop several service-oriented P2P applications.","Title: SLA-Based Coordinated Superscheduling Scheme and Performance for Computational Grids\nYear: 2006\n\nAbstract: The Service Level Agreement (SLA) based grid su- perscheduling approach promotes coordinated resource sharing. Superscheduling is facilitated between admin- istratively and topologically distributed grid sites by gr id schedulers such as Resource brokers. In this work, we present a market-based SLA coordination mechanism. We based our SLA model on a well knowncontract net protocol. The key advantages of our approach are that it al- lows: (i) resource owners to have finer degree of con- trol over the resource allocation that was previously not possible through traditional mechanism; and (ii) super- schedulers to bid for SLA contracts in the contract net with focus on completing the job within the user speci- fied deadline. In this work, we use simulation to show the effectiveness of our proposed approach.","Title: Adaptive expression based routing protocol for P2P systems\nYear: 2007\n\nAbstract: There is an emerging trend of using P2P systems for computational and data intensive tasks, such as online collaborations, distributed database applications and message passing interface (MPI) algorithms. For better resource utilization in these and similar applications, it is necessary to discover the resource capabilities of the collaborating peers. Another requirement of these applications is to locate the optimal resource based on a search criteria, e.g., to seek a resource with earliest execution time. We define this kind of routing as expression based routing. However, in the absence of the centralized controlling node, tracking the capabilities of the participating peers is very difficult. Moreover peers join and leave the system dynamically and, thus, makes the discovery of a desired resource even more complex. In this paper we investigate a novel algorithm for expression based routing in a P2P system. It evaluates peer u such that lu (the value of u) is minimum where e.g. lu is load of the peer u. Our contribution includes a detailed algorithm to search for the least loaded peer in the system. We analyze the accuracy and cost of our proposed algorithm through detailed simulations.","Title: A Scalable Anonymous Server Overlay Network\nYear: 2006\n\nAbstract: Large scale malicious activities continue to take place on the Internet. Safeguarding vulnerable computers and servers is therefore important. Recently, systems have been proposed to protect network addresses of computers from malicious adversaries on the Internet. Some of these systems focus specifically on providing client-side anonymity. However, as the Internet bandwidth for average users continues to increase, individuals are increasingly capable of carrying out complicated tasks as servers. It is in the interest of the individuals to gain server anonymity in which the network address of the machine remains protected from malicious activities. In this paper, we present a Scalable Anonymous Server Overlay Network (SASON) that aims at protecting network addresses of all participating servers. SASON also provides reliable server discovery and routing through the use of server pseudonyms and distributed hashtables.","Title: A Taxonomy of Peer-to-Peer Based Complex Queries: a Grid perspective\nYear: 2006\n\nAbstract: Grid superscheduling requires support for efficient and sca lable discovery of resources. Resource discovery ac- tivities involve searching for the appropriate resource ty pes that match the user's job requirements. To accomplish this goal, a resource discovery system that supports the desired look-up operation is mandatory. Various kinds of so- lutions to this problem have been suggested, including the centralised and hierarchical information server approach. However, both of these approaches have serious limitations in regards to scalability, fault-tolerance and network congestion. To overcome these limitations, organising resource information using Peer-to-Peer (P2P) network model has been proposed. Existing approaches advocate an extension to structured P2P protocols, to support the Grid re- source information system (GRIS). In this paper, we identify issues related to the design of such an efficient, scalable, fault-tolerant, consistent and practical GRIS system usin g a P2P network model. We compile these issues into various taxonomies in sections 3 and 4. Further, we look into existing works that apply P2P based network protocols to GRIS. We think that this taxonomy and its mapping to relevant systems would be useful for academic and industry based researchers who are engaged in the design of scalable Grid systems.","Title: An unstructured peer-to-peer approach to aggregate node selection\nYear: 2008\n\nAbstract: In distributed and high performance applications it it is necessary to discover the aggregate resource capabilities of the collaborating peers. Another requirement of these applications is to locate the resources meeting the given aggregate search criteria. In this paper we present an unstructured peer-to-peer approach to aggregate node selection for resource management. Our objective is to efficiently calculate the aggregate value and search the aggregate node whose attribute matches or falls in the proximity of this value. Our contribution also includes the inclusion of an explicit network for aggregate query routing. We evaluated our protocol for accuracy, efficiency and communication cost through detailed simulations.","Title: Decentralised Resource Discovery Service for Large Scale Federated Grids\nYear: 2007\n\nAbstract: Efficient Resource discovery mechanism is one of the fundamental requirement for Grid computing systems, as it aids in resource management and scheduling of applications. Resource discovery involves searching for resources that match the user's application requirements. Various kinds of solutions to Grid resource discovery have been developed, including the centralised and hierarchical information server approach. However, these approaches have serious limitations in regards to scalability, fault-tolerance and network congestion. To overcome such limitations, we propose a decentralised Grid resource discovery system based on a spatial publish/subscribe index. It utilises a Distributed Hash Table (DHT) routing substrate for delegation of d-dimensional service messages. Our approach has been validated using a simulated publish/subscribe index that assigns regions of a d-dimensional resource attribute space to the Grid peers in the system. We generated the resource attribute distribution using the configurations obtained from the Top 500 Supercomputer list. The simulation study takes into account various parameters such as resource query rate, index load distribution, number of index messages generated, overlay routing hops and system size. Our results show that grid resource query rate directly affects the performance of the decentralised resource discovery system, and that at higher rates the queries can experience considerable latencies. Further, contrary to what one can expect, system size does not have a significant impact on the performance of the system, in particular the query latency.","Title: Achieving Ubiquitous Network Connectivity Using an RFID Tag-Based Routing Protocol\nYear: 2008\n\nAbstract: Ubiquitous network connectivity is challenged in areas outside of the range of wireless base stations and cellular networks. In such networks the mobile nodes are disconnected for significantly long periods of time. In this paper, we propose an approach where mobile nodes deposit/retrieve messages to/from known point locations in the geographic region. Messages are delivered from a source by being deposited at one or more locations that are later visited by the destination. We propose the use of passive, read/writable radio frequency identification tags, one per point location. Our simulation results indicate that our approach can achieve competitive message latency and delivery rates. We also provide techniques for optimizing the tag placement, including tag pruning, probability based tag distribution and a genetic algorithm; the genetic algorithm is shown to provide the best solutions to this problem.","Title: Algorithmic-Parameter optimization of a parallelized split-step fourier transform using a modified BSP cost model\nYear: 2004\n\nAbstract: Adaptive algorithms are increasingly acknowledged in leading parallel and distributed research. In the past, algorithms were manually tuned to be executed efficiently on a particular architecture. However, interest has shifted towards algorithms that can adapt themselves to the computational resources. A cost model representing the behavior of the system (i.e. system parameters) and the algorithm (i.e algorithm parameters) plays an important role in adaptive parallel algorithms. In this paper, we contribute a computational model based on Bulk Synchronous Parallel processing that predicts performance of a parallelized split-step Fourier transform. We extracted the system parameters of a cluster (upon which our algorithm was executed) and showed the use of an algorithmic parameter in the model that exhibits optimal behavior. Our model can thus be used for the purpose of self-adaption.","Title: GWISFI: A universal GPU interface for exhaustive search of pairwise interactions in case-control GWAS in minutes\nYear: 2014\n\nAbstract: Epistatic interactions between genes are believed to be a critical component in the genetic architecture of complex diseases. Genome Wide Association Studies (GWAS) may be able to detect such genetic interactions indirectly, via the identification of associated SNP markers. Major obstacles to progress in this area are: the unknown nature of epistatic interactions, little understanding of the capabilities of different filtering methods, and the computational difficulties for exhaustive analysis. A common platform enabling various detection methods is needed to avoid practical issues such as software compatibility and portability, incompatible input and output formats and varying demands on computational resources. We developed a highly optimised GPU system capable of exhaustively analysing all SNP-pairs in typical GWAS data (0.5M SNPs, 5K samples) in a few minutes on a standard desktop computer. A number of programming elements provided by a functional interface can be used to construct user-defined statistical tests to efficiently score every SNP pair. As a proof of principle, we have implemented 8 methods from the literature via our interface. We have applied all of them using a single GPU to exhaustively scan the 7 popular WTCCC case-control GWAS datasets. We present timing results for these methods, both in their original software implementations and using our platform. Significant improvements in timing are observed, up to 10000 times for CPU implementations of the popular FastEpistasis in PLINK and up to 2 orders of magnitude for some GPU implementations in the literature. As an initial discovery we show plots for overlaps of list of selected pairs by 8 algorithms for Type 2 Diabetes, WTCCC data.","Title: Multi-space distributed hash tables for multiple transport domains\nYear: 2003\n\nAbstract: The use of consistent hashing to construct a dis- tributed hash table (DHT) has recently gained signiflcant interest from the distributed computing community, partic- ularly because algorithms have been devised which imple- ment a DHT using peer-to-peer (P2P) technology. However a common assumption is that every node that participates in the DHT is within the same transport domain. Some so- lutions make to this problem make use of an intermediate overlay network which itself can become messy. Other solu- tions rely on the widespread use of IPv6 or \\work-arounds\" for the limited IPv4. We propose a solution that uses a multi-space DHT algorithm to manifest a DHT space in each transport domain, with bridge nodes linking DHT spaces be- tween transport domains. The ostensive problem with this solution is to ensure that key requests are evenly distributed over all spaces. We provide a framework to solve this prob- lem and give an example solution for a simple 2-space DHT network.","Title: Delay sensitive identity protection in peer-to-peer online gaming environments\nYear: 2007\n\nAbstract: Peer-to-peer computing overcomes communication bottleneck problems associated with centralized game servers and provides an alternative mechanism to support massively multi-player online gaming (MMOG) applications. A potential security problem associated with this approach relates to player identity. In an MMOG, users from different parts of the network interact with each other in a virtual world. In a peer-to-peer model, these interactions happen directly between the peers and this leads to the IP addresses of peers being available in the network packets. Malicious users can extract the IP addresses of their opponents and the information can be used to gain unfair advantage by compromising their opponents’ computers. The well known approach to this problem is the use of anonymizing networks (onion routing and mixing), that anonymize the sender and the destination IP addresses from the peers along a path. However this approach introduces significant delays which are not desirable in MMOG applications. This paper proposes the use of a secret shared key to reduce computational delay and also provides a theoretical framework for trading off the strength of anonymity for reduced delay with respect to classes of interactions in the MMOG. The results suggest that appropriately low delays can be achieved with small reductions in anonymity strength.","Title: Demonstrating BooSTER: The broadcast stream transmission epidemic repair\nYear: 2012\n\nAbstract: Wireless broadcasting systems, such as Digital Video Broadcasting (DVB), are subject to signal degradation, having an effect on end users' reception quality. Reception quality can be improved by increasing signal strength, but this comes at a significantly increased energy use and still without guaranteeing error-free reception. BOOSTER is a system for cooperative repair of DVB streams. It operates based on a fully decentralized epidemic algorithm that boosts reception quality by cooperatively repairing lossy packet streams among the community of DVB viewers. In this demo, we demonstrate BOOSTER in a special configurable implementation that allows on-the-fly tweaking of the error rate, network topology, and some protocol parameters, allowing the audience to get a feeling on how they affect end-user experience and load distribution.","Title: A gossip-based asynchronous aggregation protocol for P2P systems\nYear: 2010\n\nAbstract: Gossip-based protocols are a popular approach to estimate aggregates in P2P systems. For gossip-based aggregation protocols, the accuracy of the estimated aggregate heavily depends upon conservation of the system mass. In this paper, we present a gossip-based asynchronous aggregation protocol. Our protocol conserves the system mass under asynchronous settings and churn. Simulation results show that our proposed protocol converges to the correct system aggregate even in the presence of very dynamic and adverse network delays and under churn.","Title: Feature selection for user motion pattern recognition in mobile networks\nYear: 2012\n\nAbstract: Mobility is a challenging issues in mobile networks which significantly impacts the performance of a variety of network protocols. Understanding user motion behavior can improve the performance of mobile network protocols in different aspects. Therefore, we consider how to analyze the motion behavior of mobile nodes in various environments. In this paper we propose a few mobility metrics useful for analysis of individual, collective and geographical behavior of mobile nodes and a simple supervised mobility pattern recognition method which is able to classify mobility traces into different mobility model classes using our proposed mobility metrics. The remaining challenge is to find appropriate features, extracted from mobility traces, which are able to distinguish between different motion behaviors with maximum accuracy and minimum overhead. We propose a method to find the most suitable feature sets for classification of different mobility traces with different motion behaviors. Simulation results show significant performance of our proposed mobility pattern recognition method using appropriate mobility metrics and feature sets.","Title: Geometry of Interest (GOI): Spatio-Temporal Destination Extraction and Partitioning in GPS Trajectory Data.\nYear: 2017\n\nAbstract: Nowadays large amounts of GPS trajectory data is continuously collected by GPS-enabled devices such as vehicles navigation systems and mobile phones. GPS trajectory data is useful for applications such as traffic management, location forecasting, and itinerary planning. Such applications often need to extract the time-stamped sequence of visited locations (SVLs) of the mobile objects. The nearest neighbor query (NNQ) is the most applied method for labeling the visited locations based on the IDs of the points of interests (POIs) in the process of SVL generation. NNQ in some scenarios is not accurate enough. To improve the quality of the extracted SVLs, instead of using NNQ, we label the visited locations as the IDs of the POIs which geometrically intersect with the GPS observations. Intersection operator requires the accurate geometries of the POIs which we refer to them as the geometries of interest (GOIs). In some application domains (e.g. movement trajectories of animals), adequate information about the POIs and their geometries may not be available a priori, or they may not be publicly accessible and, therefore, they need to be derived from GPS trajectory data. In this paper we propose a novel method for estimating the GOIs, which consists of three phases: (1) extracting the geometries of the stay regions; (2) constructing the geometry of destination regions based on the extracted stay regions; and (3) constructing the GOIs based on the geometries of the destination regions. Using the geometric similarity to known GOIs as the major evaluation criterion, the experiments we performed using long-term GPS trajectory data show that our method outperforms the state-of-the-art.","Title: Using Para-Virtualization as the Basis for a Federated PlanetLab Architecture\nYear: 2006\n\nAbstract: The federation of islands of large scale distributed systems is a difficult problem, in part because of the variety of systems involved. We categorise solutions into either the systems integration or the virtualization approach. Using the virtualization approach, we enable a federated architecture for the PlanetLab Internet testbed. The current implementation of PlanetLab has a central management authority responsible for the maintenance of the system's nodes and creation of virtual machines for users. We propose two additional entities, which separate aspects of a node and site's management, allowing viable federation, called the Site Manager and Federation Authority. Building on the existing virtualization used by PlanetLab, the realisation of our proposal makes use of Xen, a para-virtualization system for x86 hardware. We present the uses of Xen domains, or virtual machines, as hosts for both PlanetLab nodes and slices. This additional layer of virtualization is the basis of our federation architecture.","Title: BrowserCloud: A Personal Cloud for Browser Session Migration and Management\nYear: 2015\n\nAbstract: Web browsers are de facto clients for an ever-increasing range of web applications. At the same time, web users are accessing these applications from a wide range of devices. This paper presents a solution for runtime browser session migration and management, called BrowserCloud, which allows a user to securely manage multiple browsers from a personal or third-party Cloud service, migrate snapshots of active browser sessions between browsers over different devices, using a robust security module. The design of BrowserCloud is based on browser extensions/plugins that can preserve and restore browser session state, and a PHP server that stores browser sessions securely. We have tested our implementation over a range of increasingly complex web applications, including WebRTC and HTML5 video. To the best of our knowledge, our implementation is the most robust and secure approach to runtime browser session management to date.","Title: Finding High-Level Topics and Tweet Labeling Using Topic Models\nYear: 2015\n\nAbstract: Making sense of Twitter data streams is challenging due to the extremely high volume of data. One way to address this challenge is to consider these data streams as containing a set of high-level topics. In this research we address the problem of: given a collection of tweets about K high-level topics, how to find topic words that describe these topics as well as how to label each tweet with one of the K topics using a topic modeling approach. Current research has shown that applying topic modeling algorithms directly on tweets does not lead to good results. Hence one approach is to group related tweets together, so as to form a single “pseudo-document”, which is more informative than a single tweet. In this paper we evaluate different grouping schemes found in the literature and propose a new grouping scheme utilizing named entities and word collocations. Results show that our proposed scheme performs better than the existing approaches, to a some extent for all the test cases, and for both finding high-level topics and tweet labeling.","Title: Optimizing indoor location recognition through wireless fingerprinting at the Ian Potter Museum of Art\nYear: 2014\n\nAbstract: Indoor tracking of smartphones adds context to smartphone applications, enabling a range of smarter behaviours. The predicted use cases are many and varied, and include navigation, planning, advertising and communication. Potentially, indoor tracking could become as ubiquitous as GPS - however, all of these possibilities depend on being able to produce a reasonably accurate, reliable system which does not require specialised infrastructure. While professional systems using custom devices are able to achieve very high levels of accuracy (<;1 cm), consumer no-infrastructure systems struggle to achieve reliable room-level tracking. This paper focuses on the use of WiFi received signal strength indicator (RSSI) fingerprinting, a machine learning approach which currently seems to be the most promising option for consumer smartphones. We have undertaken experimentation and optimisation in a real-world, noisy environment - the Ian Potter Museum of Art - where we developed and deployed a no-infrastructure, indoor visitor tracking application. Data was collected in a trial involving several dozen users over a few weeks, who used the system extensively. This data was analysed with a range of current WiFi RSSI fingerprinting techniques and algorithms (WASP, Redpin (kNN), SSD, SVM, Gaussian Naive Bayes and Random Forests), and their efficacy was compared and improved where possible. Known challenges such as device heterogeneity are explored, and the consistency of signal levels, including magnetic fields, are examined. Large Random Forests (200 trees) were found to have the best performance, which was further improved by calibrating for average differences in RSSI between phone models, to achieve an average of 90% correct classification of exhibits within the top five hits.","Title: Discriminative Concept Learning Network: Reveal High-Level Differential Concepts From Shallow Architecture\nYear: 2015\n\nAbstract: A desired capability of deep learning is to understand the high-level, class-specific features via hierarchical features learning. However the training of deep architectures is costly comparing to simple shallow models. Bringing the high-level feature understanding into a simple shallow architecture remains an open question.We proposed a supervised learning algorithm, enabling binary classification along with an intrinsic ability of learning high-level discriminative concepts via a shallow neural network architecture. The physical architecture of the network has one hidden layer (also serving as the output layer) responsible for the classification and an input layer directly identifies the informative features that constitute the high-level differential concepts between the two classes.Compared to other shallow classifiers, we demonstrate its practicability in real world classification problems. We also illustrate the human-understandable, discriminative concepts learned from the two image recognition exercises. Lastly, we show how it is useful in validating the disease-associated genetic variants in human genome as a real diagnostic genomics application.","Title: Benchmarking the performance of hadoop triple replication and erasure coding on a nation-wide distributed cloud\nYear: 2015\n\nAbstract: Large Scale distributed storage systems play a vital role in maintaining data across storage locations globally. These systems use replication as the default mechanism for providing fault-tolerance. Recently, erasure codes are being used as a viable alternative to replication, since they provide the same fault-tolerance for reduced storage overhead. However, their performance is unclear in a geographically diverse distributed storage system. This paper compares the performance of triple replication with the erasure coding (Reed-Solomon codes) used in Apache Hadoop’s implementation of a distributed file system, on a cluster distributed across Australia that runs on the NeCTAR research cloud. Our results show that using erasure coding does not degrade the read performance in such a setting. We also compare the Hadoop’s code with a local reconstruction code, implemented in the XORBAS version of Hadoop. These codes perform well in our clusters but the performance gain observed in our results does not conform to the results reported. Hence, we need new codes that perform better, addressing the geographical diversity issue. We believe that our framework is readily usable to test a range of novel erasure codes that are being introduced in the literature.","Title: On improving recovery performance in erasure code based geo-diverse storage clusters\nYear: 2016\n\nAbstract: Erasure code based distributed storage systems are increasingly being used by storage providers for big data storage since they offer the same reliability as replication with significant decrease in the amount of storage required. But, when it comes to a storage system with data nodes spread across a very large geographical area, the code's recovery performance is affected by various factors, both network and computation related. In this paper, we expose an XOR based code supplemented with the ideas of parity duplication and rack awareness that could be adopted in such storage clusters to improve the recovery performance during node failures. We have implemented them on the erasure code module of the XORBAS version of the Hadoop Distributed File System (HDFS). For evaluating the performance of the proposed ideas, we employ a geo-diverse cluster on the NeCTAR research cloud. The experimental results show that the techniques aid in bringing down the data read for repair by a factor of 85% and repair duration by a factor of 57% during node failures, though resulting in an increased storage requirement of 21% as compared to the traditional Reed-Solomon codes used in HDFS. The sum of all these ideas could offer a better solution for a code based storage system spanning a wide geographical area that has storage constraints such that a triple replicated system is not affordable and at the same time has strict requirements on ensuring the minimal recovery time.","Title: Distributed stream clustering using micro-clusters on Apache Storm.\nYear: 2017\n\nAbstract: The recent need to extract real-time insights from data has driven the need for machine learning algorithms that can operate on data streams. Given the current extreme rates of data generation (around 5000 messages per second), these algorithms need to be able to handle data streams of very high velocity. Many current algorithms do not reach this requirement, in some cases processing only tens of messages per second.","Title: A Multi-Stage Hierarchical Window Model With Application To Real-Time Graph Analysis\nYear: 2017\n\nAbstract: The dynamic nature of rea-world networks, such as social networks and communication networks, has increased the focus towards real-time dynamic graph analysis. Observations made based on real-time analysis of dynamic graphs reflect the latest properties of the graph and have the most value in real-time analysis. Computing graph properties of large-scale, fast-evolving graphs in real-time is challenging due, not only to the high computational and memory cost, but also to the understanding of the result with respect to the data from which it was derived. This paper proposes a multi-stage hierarchical window model that can aid in rigorous understanding of complicated real-time results and we apply it to generate graphs based on real-time updates along with periodic computations on graph snapshots for processing dynamic graphs. Moreover, the paper discusses the utilization of parallel window computation. The paper evaluates the hierarchical model through analyzing graphs formed by coccurring hashtags in a Twitter data-stream.","Title: Predicting The Stability Of Large-Scale Distributed Stream Processing Systems On The Cloud\nYear: 2017\n\nAbstract: Large-scale topology-based stream processing systems are non-trivial to build and deploy. They require understanding of the performance, cost of deployment and considerations of potential downtime. Our work considers stability as a primary characteristic of these systems. By stability, we mean that unstable systems exhibit large-spikes in latency and can drop throughput frequently or unpredictably. Such instabilities can be due to variations of workloads or underlying hardware platforms that are often difficult to predict. To understand and tackle this for large-scale stream processing systems, we apply queueing theory and simulate the results through a series of experiments on the Cloud.","Title: Optimum Benefit Protocol: A fast converging, bandwidth-efficient decentralized similarity overlay.\nYear: 2017\n\nAbstract: Due to large volumes of data available online, techniques such as document classification and clustering are required for organization, analysis and management of data. Similarity-based Clustering (SBC) is used by many existing systems for filtering information. Decentralized gossip-based overlays offer a simple, robust and scalable solution to SBC clustering. Convergence and communication complexity are the two key areas of concern when SBC is implemented using these overlays. Convergence guarantees accurate clustering but costs bandwidth because these systems rely on message passing to achieve convergence. In this work, we address the long tail problem, experienced by low in-degree nodes in a long tailed similarity distribution, such as power-law distribution and propose a new SBC approach, Optimum Benefit Protocol (OBP), that converges more rapidly than existing approaches and reduces the long tail. The proposed protocol only sends messages that could benefit the receiver, reducing bandwidth to one fifth of the default. Our protocol obtains at least 90% convergence for a 900 node network, starting in a random configuration, in less than 10 cycles, for all observed experiments. We used real world distributions from Yahoo, Movielens, and Epinion datasets, for experimentation.","Title: Search Result Personalization in Twitter Using Neural Word Embeddings.\nYear: 2017\n\nAbstract: In recent years, Twitter has become one of the most popular microblogging avenues. Today it has more than 300 million monthly active users generating more than 500 million tweets everyday. Twitter users both post messages as well as search for messages. Current search results given by Twitter are chronologically ordered and often users have to manually scan through an overwhelming number of the tweets to find content of interest. This process can quickly become infeasible. Personalization techniques address this problem by learning the user interests and tailoring search results by matching them with the user's interests. Recent research on neural word embedding models, which represents each word in the vocabulary as a vector of real values, has gained much attention. These models learn word embeddings in such a way that contextually similar words have similar vectors. In this paper we propose a novel approach, PWEBA, for personalizing Twitter search, which uses neural word embeddings to model user interests. Our experimental results show that PWEBA outperforms existing approaches for all the evaluation metrics we have considered in this paper.","Title: E-Storm: Replication-Based State Management in Distributed Stream Processing Systems\nYear: 2017\n\nAbstract: Apache Storm is a fault-tolerant, distributed inmemory computation system for processing large volumes of high-velocity data in real-time. As an integral part of the fault-tolerance mechanism, Storm's state management is achieved by a checkpointing framework, which commits states regularly and recovers lost states from the latest checkpoint. However, this method involves a remote data store for state preservation and access, resulting in significant overheads to the performance of error-free execution.In this paper, we propose E-Storm, a replication-based state management system that actively maintains multiple state backups on different worker nodes. We build a prototype on top of Storm by extending it with monitoring and recovery modules to support inter-task state transfer whenever needed. The experiments carried out on synthetic and real-world streaming applications confirm that E-Storm outperforms the existing checkpointing method in terms of the resulting application performance, obtaining as much as 9.44 times throughput improvement while reducing the application latency down to 9.8%.","Title: Enhancing the quality of geometries of interest (GOIs) extracted from GPS trajectory data using spatio-temporal data aggregation and outlier detection.\nYear: 2018\n\nAbstract: One of the initial phases in the applications dealing with data processing on GPS trajectory data is to generate the time-stamped Sequence of Visited Locations (SVLs) of the mobile objects. The sequence is constructed by labeling each of the GPS observations of the trajectory using the ID of their intersecting Geometries of Interest (GOIs). In this paper, we enhance the performance of the state-of-the-art scheme for constructing the GOIs of a mobile object by proposing a data aggregation and outlier detection method. Our experimental results using geometric similarity metrics show that our improved GOI construction method outperforms the baseline methods by constructing the GOIs remarkably more geometrically similar to the real world GOIs. The geometric similarity metrics are only applicable when we have access to the geometries of the real world GOIs (ground truth). To be able to analyse the performance of the GOI extraction methods in environments which we do not have access to the ground truth, we propose two useful spatio-temporal metrics to measure the quality of GOIs based on the quality of the generated SVLs based on them. Our experimental results show that these two metrics are able to discriminate between the results of our different outlier detection methods and select the best scheme without using any external knowledge about the geometries of the real world GOIs.","Title: Geo-aware erasure coding for high-performance erasure-coded storage clusters.\nYear: 2018\n\nAbstract: Erasure code-based distributed storage systems are increasingly being used by storage providers for big data storage since they offer the same reliability as replication with a significant decrease in the amount of storage required. But, when it comes to a storage system with data nodes spread across a very large geographical area, the node’s recovery performance is affected by various factors that are both network and computation related. In this paper, we present a XOR-based code supplemented with the ideas of parity duplication and rack awareness that could be adopted in such storage clusters to improve the recovery performance during node failures and compare it with popular implementations of erasure codes, namely Facebook’s Reed-Solomon codes and XORBAS local recovery codes. The code performance along with the proposed ideas are evaluated on a geo-diverse cluster deployed on the NeCTAR research cloud. We also present a scheme for intelligently placing blocks of coded storage depending on the design of the code, inspired by local reconstruction codes. The sum of all these propositions could offer a better solution for applications that are deployed on coded storage systems that are geographically distributed, in which storage constraints make triple replication not affordable, at the same time ensuring minimal recovery time is a strict requirement.","Title: Handling Churn in Similarity Based Clustering Overlays Using Weighted Benefit\nYear: 2017\n\nAbstract: Similarity based clustering (SBC) overlays are decentralized networks of nodes on the Internet edge, where each node maintains some number of direct connections to other nodes that are most \"similar\" to it. The challenge is: how do the nodes in the overlay converge to and maintain the most similar neighbors, given that the network is decentralized, is subject to churn and that similarity varies over time. Protocols that simultaneously provide fast convergence and low bandwidth consumption are the objective of this research. We present a protocol, that we call Weighted Benefit Scheme (WBS), that improves upon existing state-of-the-art in this area: it has equivalent convergence rate to the Optimum Benefit Protocol while simultaneously handling churn competitively to the Vicinity protocol. We use real world datasets from Yahoo WebScope that comprises of 15,400 users with 354,000 ratings about 1000 songs and our experiments are performed on the simulation test-bed PeerNet.","Title: A Data Stream Processing Optimisation Framework for Edge Computing Applications\nYear: 2018\n\nAbstract: Data Stream Processing (DSP) is a widely used programming paradigm to process an unbounded event stream. Often, DSP frameworks are deployed on the cloud with a scalable resource model. One of the key requirements of DSP is to produce results with low latency. With the emergence of IoT, many event sources have been located outside the cloud which can result in higher end-to-end latency due to communication overhead. However, due to the abundance of resources at the IoT layer, Edge computing has emerged as a viable computational paradigm. In this paper, we devise an optimisation framework, consisting of a constraint satisfaction formulation and a system model, that aims to minimise end-to-end latency through appropriate placement of DSP operators either on cloud nodes or edge devices, i.e. deployed in an edge-cloud integrated environment. We test our optimisation framework using OMNeT++, with realistic topologies and power consumption data, and show that it is capable of achieving approx 1.65 times reduction of latency compared to edge-only and cloud-only placements, which in turn also reduces the energy consumption per event by up to approx 4% at the edge layer. To the best of our knowledge our optimisation framework is the first of its kind to integrate power, bandwidth and CPU constraints with latency minimisation.","Title: Performance Analysis Of Large-Scale Distributed Stream Processing Systems On The Cloud\nYear: 2018\n\nAbstract: Real-time data processing is often a necessity as it can provide insights that have less value if discovered off-line or after the fact. However, large-scale stream processing systems are non-trivial to build and deploy. While there are many frameworks that allow users to create large-scale distributed systems, there remains many challenges in understanding the performance, cost of deployment and considerations and impact of potential (partial) outages on real-time systems performance. Our work considers the performance of Cloud-based stream processing systems in terms of back-pressure and expected utilization. The performance of an exemplar stream application is explored using different Cloud-based virtual machine resources and where the scale of deployment and cost benefits are taken into consideration in relation to the overall performance.To achieve this, we develop an algorithm based on queueing theory to predict the throughput and latency of stream data processing while supporting system stability. Our methodology for making fundamental measurements is applicable to mainstream stream processing frameworks such as Apache Storm and Heron. The method is especially suitable for large-scale distributed stream processing where jobs can run for extended time periods. We benchmark the performance of the system on the national research cloud of Australia (Nectar), and present a performance analysis based on estimating the overall effective utilization.","Title: Stepping stone graph for public movement analysis.\nYear: 2018\n\nAbstract: There are many real world applications that require to identify movement of users such as identifying movement corridors, most popular paths, and nearest neighbours. If one is not given trajectories mapping to movement of people but rather sporadic location data, such as location based social network data, finding movement related information becomes difficult. Rather than processing all points in a data set given a query, a clever approach is to construct a graph, based on user locations, and query this graph for all queries. One example is the shortest path graph. However the shortest path graph can be inefficient and ineffective analysing movement, as it calculates the graph considering all points in a data set. We propose the stepping stone graph, which calculates graph considering point pairs rather than all points, that focuses on local possible movement, making it both efficient and effective for location based social network related queries. We demonstrate its uses by applying it in the aforementioned domain and comparing with the shortest path graph. We also compare its properties to a range of other graphs.","Title: Communication Model for Parallel Iterative Stream Processing\nYear: 2018\n\nAbstract: Stream processing systems are used for a plethora of low-latency applications that deal with high volumes and varieties of data. In general, streaming applications are formulated as a fixed number of inter-connected operators, where the operator graph indicates the sequence of computations that apply to the streaming data in motion. Typically an iterative operation in a streaming application is done by embedding the iteration in a single operator (which excludes multiple operator instances and hence not scalable) or by using a sequence of operators, one for each iteration (which requires knowledge of the number of iterations a priori). However, the number of iterations of some applications such as k-means clustering cannot be determined at the creation of the application as it depends on the properties of the data received at runtime and the convergence criterion. Hence only a limited number of iterative computations can be executed using this approach. In this paper, we propose a communication model to support intra-operator communication so that iterative computations, including those with an arbitrary number of iterations, can be efficiently executed in streaming applications. We show that the proposed model can support different iterative algorithms that have complex communication patterns. Finally, through evaluating a number of parallel iterative algorithms using large-scale datasets, we demonstrate the scalability and performance of our proposed communication model and compare it to the existing approaches used for constructing iterative streaming applications.","Title: Geotagging tweets to landmarks using convolutional neural networks with text and posting time\nYear: 2019\n\nAbstract: Geotagged tweets serve many important applications, e.g., crisis management, but only a small proportion of tweets are explicitly geotagged. We propose a Convolutional Neural Network (CNN) architecture for geotagging tweets to landmarks, based on the text in tweets and other meta information, such as posting time and source. Using a dataset of Melbourne tweets, experimental results show that our algorithm out-performed various state-of-the-art baselines.","Title: Power-law Verification for Event Detection at Multi-spatial Scales from Geo-tagged Tweet Streams.\nYear: 2019\n\nAbstract: Compared with traditional news media, social media nowadays provides a richer and more timely source of news. We are interested in multi-spatial level event detection from geo-tagged tweet streams. Specifically, in this paper we (1) examine the statistical characteristic for the time series of the number of geo-tagged tweets posted from specific regions during a short time interval, e.g., ten seconds or one minute; (2) verify from over thirty datasets that while almost all such time series exhibit self-similarity, those that correspond to events, especially short-term and unplanned outbursts, follow a power-law distribution; (3) demonstrate that these findings can be applied to facilitate event detection from tweet streams---we propose a simple algorithm that only checks the existence of power-law distributions in the time series from tweet streams at multi-spatial scales, without looking into the content of each tweet. Our experiments on multiple datasets show that by considering spatio-temporal statistical distributions of tweets alone, this seemingly naive algorithm achieves comparable results with event detection methods that perform semantic analysis. We further discuss how to integrate the proposed technique into existing algorithms for better performance.","Title: Spatio-Temporal Event Detection Using Poisson Model And Quad-Tree On Geotagged Social Media\nYear: 2019\n\nAbstract: Identifying events happening in a specific locality is important as an early warning for accidents, protests, elections or breaking news. However, this location-specific event detection is challenging as the locations and types of events are not known beforehand. To address this problem, we propose an online spatiotemporal event detection system using social media that is able to detect events at different time and space resolutions. First, we exploit a quad-tree method to split the geographical space into multiscale regions based on the density of social media data. Then, we implement a statistical unsupervised approach using Poisson distribution and a smoothing method for highlighting regions with unexpected density of social posts. Further, event duration is estimated by merging events happening in the same region at consecutive time intervals. A post processing stage is introduced to filter out events that are spam, fake or wrong. Finally, we incorporate simple semantics by using social media entities to assess the integrity, and accuracy of detected events. The proposed method is evaluated using Twitter and Flickr for the city of Melbourne based on recall and precision measures. We also propose a new quality measure named strength index, which automatically measures how accurate the reported event is.","Title: Multi-Spatial Scale Event Detection From Geo-Tagged Tweet Streams Via Power-Law Verification\nYear: 2019\n\nAbstract: Compared with traditional news media, social media nowadays provides a richer and more timely source of news. We are interested in multi-spatial level event detection from geo-tagged tweet streams. Specifically, in this paper we (1) examine the statistical characteristic for the time series of the number of geo-tagged tweets posted from specific regions during a short time interval, e.g., one minute; (2) verify from over thirty datasets that while almost all such time series exhibit self-similarity, those that correspond to events, especially short-term and unplanned outbursts, follow a power-law distribution; (3) demonstrate that these findings can be applied to facilitate event detection from tweet streams. We propose two algorithms-Power-law basic and Power-law advanced, where Power-law basic only checks the existence of power-law distributions in the time series from tweet streams at multi-spatial scales, without looking into the content of each tweet, and Power-law advanced integrates power-law verification with semantic analysis via word embedding. Our experiments on multiple datasets show that when combined with a Quad-tree, the seemingly naive algorithm of Power-law basic achieves comparable results with more advanced event detection methods, while the semantic analysis enhanced version, Power-law advanced, can significantly increase both the precision and the recall.","Title: Sentiment-Aware And Personalized Tour Recommendation\nYear: 2019\n\nAbstract: Itinerary planning is one of the most important tasks in tourism. A well-planned itinerary enhances the tourist experience and their visit satisfaction in new cities. However, the task of planning personalized tour itineraries is complicated by tourists with different interest preferences. Furthermore, there is an added complexity of recommending an itinerary with discrete budget, time and cost. Due to an increase in web-technologies and online geo-location services, there is emerging research targeting itinerary recommendation based on each tourist's interest, preferences and trip constraints. While several research works consider tourist interest, they adopt a simple measure based on the number of times a tourist has visited a place or the number of photos taken by the tourist at a place. Our research proposes an improved sentiment-aware personalized tour planner that considers each tourist's interests based on his/her sentiments on specific categories relative to his/her overall preferences. Unlike the previous approaches that do not consider the actual opinion based preferences, our proposed approach determines user interests based on their sentiments associated with their written text about a place of their visit. This interest measure is based on the intuition that users are more likely to post favorable comments about places they like. Using a dataset from Twitter, we compare our proposed algorithm against the baseline and experimental results show that our algorithm obtained superior performance in terms of tour precision, recall, Fl-score and overall popularity.","Title: User Identification across Social Networking Sites using User Profiles and Posting Patterns\nYear: 2021\n\nAbstract: With the prevalence of online social networking sites (OSNs) and mobile devices, people are increasingly reliant on a variety of OSNs for keeping in touch with family and friends, and using it as a source of information. For example, a user might utilise multiple OSNs for different purposes, such as using Flickr to share holiday pictures with family and friends, and Twitter to post short messages about their thoughts. Identifying the same user across multiple OSNs is an important task as this allows us to understand the usage patterns of users among different OSNs, make recommendations when a user registers for a new OSN, and various other useful applications. To address this problem, we proposed an algorithm based on the multilayer perceptron using various types of features, namely: (i) user profile, such as name, location, description; (ii) temporal distribution of user generated content; and (iii) embedding based on user name, real name and description. Using a Twitter and Flickr dataset of users and their posting activities, we perform an empirical study on how these features affect the performance of user identification across the two OSNs and discuss our main findings based on the different features.","Title: Spatial-Based Topic Modelling Using Wikidata Knowledge Base\nYear: 2017\n\nAbstract: Topic modelling is a well-studied field that aims to identify topics from traditional documents such as news articles and reports. More recently, Latent Dirichlet Allocation (LDA) and its variants, have been applied on social media platforms to model and study topics relating to sports, politics and companies. While these applications were able to successfully identify the general topics, we posit that standard LDA can be augmented with spatial and temporal considerations based on the geo-coordinates and timestamps of social media posts. Towards this effort, we propose a spatial and temporal variant of LDA to better detect more specific topics, such as a particular art exhibit held at a museum or a security incident happening on a particular day. We validate our approach on a Twitter dataset and find that the detected topics are well-aligned to real-life events happening on the specific days and locations.","Title: Multi-Step Prediction With Missing Smart Sensor Data Using Multi-Task Gaussian Processes\nYear: 2017\n\nAbstract: With the proliferation of sensors and the increased connectivity of citizens, many global cities are increasingly adopting Smart City initiatives. Such initiatives provide real-time monitoring capabilities, and effective modelling techniques allow the prediction of future states in a city. For example, urban electricity smart meter data can be utilised to predict future demand in order to facilitate capacity planning. However, the accuracy of this foresight is often marred by low quality and missing sensor data in real-world systems. In this work, we focus on the problem of reliable forecasting by mitigating the effect of missing data on forecast accuracy. In order to mitigate the effects of missing data, we develop a multi-task learning scheme to jointly learn Gaussian Process Regression models between highly correlated sensors. We demonstrate that our methods are robust in a variety of error generation scenarios. We validate our methods based on publicly available and real-world datasets related to electricity smart meters in a university campus and pedestrian counts in a global city, where we achieve significant improvement over competitive baselines and other effective forecasting methods.","Title: Clustop: A Clustering-Based Topic Modelling Algorithm For Twitter Using Word Networks\nYear: 2017\n\nAbstract: Twitter is a popular microblogging service, where users frequently engage in discussions about various topics of interest, ranging from popular topics (e.g., music) to niche topics (e.g., politics). With the large amount of tweets, a key challenge is to automatically model and determine the discussion topics without having prior knowledge of the types and number of topics, or requiring the technical expertise to define various algorithmic parameters. For this purpose, we propose the Clustering-based Topic Modelling (ClusTop) algorithm that constructs various types of word network and automatically determines the discussion topics using community detection approaches. Unlike traditional topic models, ClusTop is able to automatically determine the appropriate number of topics and does not require numerous parameters to be set. The ClusTop algorithm is also able to capture the syntactic meaning in tweets via the use of bigrams, trigrams and other word combinations in constructing the word network graph. Using three Twitter datasets with labelled crises and events as topics, ClusTop has been shown to outperform various baselines in terms of topic coherence, pointwise mutual information, precision, recall and F-score.","Title: Optimizing checkpoint-based fault-tolerance in distributed stream processing systems: Theory to practice\nYear: 2022\n\nAbstract: Fault-tolerance is an essential part of a stream processing system that guarantees data analysis could continue even after failures. State-of-the-art distributed stream processing systems use checkpointing to support fault-tolerance for stateful computations where the state of the computations is periodically persisted. However, the frequency of performing checkpoints impacts the performance (utilization, latency, and throughput) of the system as the checkpointing process consumes resources and time that can be used for actual computations. In practice, systems are often configured to perform checkpoints based on crude values ignoring factors such as checkpoint and restart costs, leading to suboptimal performance. In our previous work, we proposed a theoretical optimal checkpoint interval that maximizes the system utilization for stream processing systems to minimize the impact of checkpointing on system performance. In this article, we investigate the practical benefits of our proposed theoretical optimal by conducting experiments in a real-world cloud setting using different streaming applications; we use Apache Flink, a well-known stream processing system for our experiments. The experiment results demonstrate that an optimal interval can achieve better utilization, confirming the practicality of the theoretical model when applied to real-world applications. We observed utilization improvements from 10% to 200% for a range of failure rates from 0.3 failures per hour to 0.075 failures per minute. Moreover, we explore how performance measures: latency and throughput are affected by the optimal interval. Our observations demonstrate that significant improvements can be achieved using the optimal interval for both latency and throughput.","Title: A Systematic Review of Density Grid-Based Clustering for Data Streams\nYear: 2022\n\nAbstract: Various applications, such as electronic business, satellite remote sensing, intrusion discovery, and network traffic monitoring, generate large unbounded data stream sequences at a rapid pace. The clustering of data streams has attracted considerable interest due to the increasing usage of evolving data streams. In particular, evolving data streams affect clustering because they introduce numerous challenges, such as time and memory limits and one-pass clustering. Furthermore, researchers need to be able to determine arbitrarily shaped clusters present in evolving data streams from applications. Due to these characteristics, conventional density grid-based clustering techniques cannot be used. Moreover, the existing density grid-based clustering algorithms have low cluster quality for clustering evolving data streams. This study conducted a systematic literature review (SLR) and noted numerous research-related issues encountered in solving the aforementioned problems. We summarized numerous grid-based clustering algorithms that have been used and determined their distinctive and limited features. We also observed how these algorithms address the challenges affecting the clustering of evolving data streams and studied their advantages and disadvantages. SLR was based on 104 articles published between 2010 and 2021. Numerous challenges remain for grid-based clustering algorithms, particularly in terms of time-limited and high-dimensional data handling. Last, our findings indicated a variety of active studies on density grid-based clustering.","Title: A clustering-based topic model using word networks and word embeddings\nYear: 2022\n\nAbstract: Online social networking services like Twitter are frequently used for discussions on numerous topics of interest, which range from mainstream and popular topics (e.g., music and movies) to niche and specialized topics (e.g., politics). Due to the popularity of such services, it is a challenging task to automatically model and determine the numerous discussion topics given the large amount of tweets. Adding on this complexity is the need to identify these topics with the absence of prior knowledge about both the types and number of topics, while having the requirement of the relevant technical expertise to tune the numerous parameters for the various models. To address this challenge, we develop the Clustering-based Topic Modelling (ClusTop) algorithm that first constructs different types of word networks based on different types of n-grams co-occurrence and word embedding distances. Using these word networks, ClusTop is then able to automatically determine the discussion topics using community detection approaches. In contrast to traditional topic models, ClusTop does not require the tuning or setting of numerous parameters and instead uses community detection approaches to automatically determine the appropriate number of topics. The ClusTop algorithm is also able to capture the syntactic meaning in tweets via the use of bigrams, trigrams, other word combinations and word embedding techniques in constructing the word network graph, and utilizes edge weights based on word embedding. Using three Twitter datasets with labelled crises and events as topics, we show that ClusTop outperforms various traditional baselines in terms of topic coherence, pointwise mutual information, precision, recall and F-score."],"string":"[\n \"Title: ScoreFinder: A method for collaborative quality inference on user-generated content\\nYear: 2010\\n\\nAbstract: User-generated content is quickly becoming the greatest source of information on the World Wide Web. Shared content items are initially considered unconfirmed in the sense that their credibility has not yet been established. Conventional, centralized confirmation of credibility is infeasible at the Internet scale and so making use of the annotators themselves to evaluate each item is essential. However, users usually differ in opinions to the same item, and the existence of bias, variance and maliciousness makes the problem of aggregating opinions more difficult. Addressing this problem, we propose the use of an Author-Annotator model with an iterative algorithm, called ScoreFinder, for inferring credibility by ranking shared items. In order to reduce the influence from a variety of error sources, we identify reliable users on each topic, and adaptively aggregate scores from them. Moreover, we transform the users' input to remove errors/anomalies, by identifying patterns of misbehaviour learned from a real data set. We show how our algorithm performs on both real data sets and synthetic data sets, and a significant improvement was achieved in the experiment.\",\n \"Title: Impacts of Asynchrony on Epidemic-Style Aggregation Protocols\\nYear: 2010\\n\\nAbstract: The large scale and dynamic nature of a distributed system makes it difficult to collect the attributes of the individual nodes in the network. In these systems, often an aggregate (e.g. AVG, COUNT, MIN, MAX, SUM etc) of these attributes is adequate. Epidemic-style protocols are one of the popular approaches to estimate aggregates in such systems. In existing epidemic-style aggregation protocols the accuracy of the estimated aggregate at local nodes heavily depends upon synchronization of aggregation rounds. To enforce synchronization in these protocols, length of aggregation round should be long enough so that all the nodes in the system complete their aggregation information exchange. In this paper, we study the impacts of asynchrony in epidemic-style aggregation protocols. We present a simple asynchronous technique to estimate system aggregates in a distributed system. Based upon this technique, we analyze two popular existing epidemic-style aggregation protocols, Push-Pull and Push-Sum. Through detailed simulations, we evaluate accuracy and cost of asynchronous version of these protocols. We found that to obtain an estimate of the true system aggregate, aggregation protocols do not need to be synchronized and hence an efficient estimate can be obtained in lesser time.\",\n \"Title: Flow Generation for IP/ATM Label-Switched Routing over Random Networks\\nYear: 2001\\n\\nAbstract: We propose a dynamic periodicity detector (DPD) for the estimation of periodicities in data series obtained from the execution of applications. We analyze the algorithm used by the periodicity detector and its performance on a number of data streams. ...\",\n \"Title: A serverless 3D world\\nYear: 2004\\n\\nAbstract: Online multi-participant virtual-world systems have attracted significant interest from the Internet community but are hindered by their inability to efficiently support interactivity for a large number of participants. Current solutions divide a large virtual-world into a few mutually exclusive zones, with each zone controlled by a different server, and/or limit the number of participants per server or per virtual-world. Peer-to-Peer (P2P) systems are known to provide excellent scalability in a networked environment (one peer is introduced to the system by each participant), however current P2P applications can only provide file sharing and other forms of relatively simple data communications. In this paper, we present a generic 3D virtual-world application that runs on a P2P network with no central administration or server. Two issues are addressed by this paper to enable such a spatial application on a P2P network. First, we demonstrate how to index and query a 3D space on a dynamic distributed network. Second, we show how to build such a complex application from the ground level of a P2P routing algorithm. Our work leads to new directions for the development of online virtual-worlds that we believe can be used for many government, industry, and public domain applications.\",\n \"Title: Directions for Peer-to-Peer based mobile pervasive augmented reality gaming\\nYear: 2007\\n\\nAbstract: Pervasive augmented reality gaming and mobile P2P are both attracting significant focus and much progress has occurred recently. Augmented reality games extend reality with virtual elements to enhance game experience and mobile P2P networks provide autonomous, self-organizing, scalable and robust communication platforms. We review the technological challenges in both mobile P2P computing and AR domains by investigating some existing related works. Then we propose possible directions of the mobile AR collaborative gaming over P2P networks and show its advantages over other AR gaming and discuss the feasibility and future requirements regarding the support for this type of game.\",\n \"Title: An Energy Efficient Distributed Scheduling Scheme for Wireless Mesh Networks\\nYear: 2013\\n\\nAbstract: The usefulness of wireless mesh networks (WMNs) highly depends on the effective usage of frequency spectrum, scalability and energy efficiency of the underlying technologies [1]. While maximizing network throughput for given frequency spectrum, and scalability have been extensively studied in multi-hop environments, energy efficiency has not received due attention. In this paper, we propose an enhanced pseudo random access (EPRA) scheme based on the IEEE 802.16-2004 distributed mesh framework to improve energy efficiency. Our proposal addresses issues that we identify with the existing election based transmission timing (EBTT) mechanism of IEEE 802.16-2004. Our extensive simulation results show a very significant improvement in the reception energy efficiency by up to 85%. Moreover, notable improvements are also obtained in the transmission energy efficiency. While we focus mainly on the aspect of energy efficiency, the proposed enhancement also helps to improve the spatial reuse of the frequency spectrum.\",\n \"Title: Indexing distributed complex data for complex queries\\nYear: 2004\\n\\nAbstract: Peer-to-peer networks are becoming a common form of online data exchange. Querying data, mostly files, using keywords on peer-to-peer networks is well-known. But users cannot perform many types of queries on complex data and on many of the attributes of the data on such networks other than mostly exact-match queries. We introduce a distributed hashing-based index for enabling more powerful accesses on complex data over peer-to-peer networks that we expect to be commonly deployed for digital government applications. Preliminary experiments show that our index scales well and we believe that it can be extended to obtain similar indices for many other data types for performing various complex queries, such as range queries.\",\n \"Title: A Peer-to-Peer Browsable File Index using a Popularity Based Global Namespace\\nYear: 2007\\n\\nAbstract: The distribution of files using decentralized, peer-to- peer (P2P) systems, has significant advantages over cen- tralized approaches. It is however more difficult to settle on the best approach for file sharing. Most file sharing systems are based on query string searches, leading to a relatively simple but inefficient broadcast or to an ef- ficient but relatively complicated index in a structured environment. In this paper we use a browsable peer- to-peer file index consisting of files which serve as di- rectory nodes, interconnecting to form a directory net- work. We implemented the system based on BitTorrent and Kademlia. The directory network inherits all of the advantages of decentralization and provides browsable, efficient searching. To avoid conflict between users in the P2P system while also imposing no additional re- strictions, we allow multiple versions of each directory node to simultaneously exist - using popularity as the basis for default browsing behavior. Users can freely add files and directory nodes to the network. We show, using a simulation of user behavior and file quality, that the popularity based system consistently leads users to a high quality directory network; above the average qual- ity of user updates.\",\n \"Title: Analytical formulations of Peer-to-Peer Connection Efficiency\\nYear: 2003\\n\\nAbstract: Use of Peer-to-Peer (P2P) service networks in- troduces a new communication paradigm be- cause peers are both clients and servers and so each peer may provide/request services to/from other peers. Empirical studies of P2P networks have been undertaken and reveal useful charac- teristics. However there is to date little analyti- cal work to describe P2P networks with respect to their communication paradigm and their in- terconnections. This paper provides an analyti- cal formulation and optimisation of peer connec- tion efficiency, in terms of minimising the frac- tion of wasted connection time. Peer connection efficiency is analysed for both a uni- and multi- connected peer. Given this fundamental optimi- sation, the paper optimises the number of con- nections that peers should make use of as a func- tion of network load, in terms of minimising the total queue size that requests in the P2P network experience. The results of this paper provide a basis for engineering high performance P2P in- terconnection networks. The optimisations are useful for reducing bandwidth and power con- sumption, e.g. in the case of peers being mo- bile devices with a limited power supply. Also these results could be used to determine when a (virtual) circuit should be switched to support a\",\n \"Title: Realizing the e-science desktop peer using a peer-to-peer distributed virtual machine middleware\\nYear: 2006\\n\\nAbstract: Emerging e-Science applications face the challenge of providing both high performance and low maintenance cost infrastructure. We propose the e-Science Desktop Peer system and a prototype middleware implementation called P2P-DVM to address this challenge, that is designed to allow scientists' resources to be readily shared over the Internet in a Peer-to-Peer (P2P) manner. Our P2P-DVM currently supports the Message Passing Interface and Bulk Synchronous Processing model over an adaptive P2P network and we introduce a Programming Environment Abstraction Layer to allow users to quickly adapt other programming environments to our P2P approach. P2P-DVM provides decentralized coordinated checkpoint and restart functionality, message passing over P2P networks, and distributed storage for scalability. Our experimental results on the PlanetLab testbed shows support for our work.\",\n \"Title: On Problem for Aggregate Node Selection for Unstructured Overlay Networks\\nYear: 2010\\n\\nAbstract: With the success of the peer-to-peer systems, there is an emerging trend to develop an aggregation service that can provide global information about these distributed and decentralized overlays. Existing aggregation protocols have focused on providing an efficient and cost effective solution to compute aggregates. In this paper, we are interested to search the node based on an aggregate query over rapidly changing data e.g., to seek a node with minimum/maximum workload in the network. Solution to such queries can benefit number of applications such as, node-ranking, reputation building, resource scheduling, etc. Here we propose a novel protocol to solve such queries. Our proposed protocol is reactive and pull-based in nature. It intrinsically builds and manages a routing table which routes aggregate query to the desired node in the system. We evaluated our protocol for accuracy, efficiency and communication cost through detailed simulations. We also discuss effects of an explicit routing table management scheme and show that this explicit scheme causes our protocol to work better in presence of churn and node failures.\",\n \"Title: An analytical model for performance evaluation in sparse mobile ad hoc networks\\nYear: 2009\\n\\nAbstract: Routing protocols for sparse mobile ad hoc networks are becoming important due to applications such as vehicular networks and emergency services. Simulation based performance evaluation of a given protocol is computationally demanding, requiring a significant amount of time and resources. For this reason, analytical approaches to performance evaluation are desirable; however, such approaches are challenging to develop. In this paper, we present an analytical model to calculate two important performance metrics, namely throughput and average end-to-end delay, for a sparse mobile ad hoc network routing protocol. In our protocol, independent mobile nodes move in a bounded region and communicate indirectly through passive RFID tags that act like mailboxes. We analyze our protocol for two different buffer policies by providing analytical equations to calculate throughput and delay for each buffer policy. Finally, we support our equations via simulation.\",\n \"Title: A comparative study on Peer-to-Peer failure rate estimation\\nYear: 2007\\n\\nAbstract: The robustness of Peer-to-Peer systems is challenged by its highly dynamic nature. Frequent peer failure and departure events introduce uncertainty for which is considered exceptional in traditional distributed systems. The difficulty of monitoring such large scale networks is further exacerbated because it has to be done in a completely decentralized way for both scalability and reliability concerns. Some methods for estimating peer failure rate have been applied in Peer-to-Peer systems, however their comparative performance has not yet been reported in the literature. We simulate three different failure rate estimation methods and compare their accuracy and response time with respect to sample size, stabilization interval and neighbour set size. We conclude that the Maximum Likelihood Method introduced is better than the Failure Frequency based Methods commonly used in current Peer-to-Peer systems.\",\n \"Title: Diffusive Load Balancing of Loosely-Synchronous Parallel Programs over Peer-to-Peer Networks\\nYear: 2004\\n\\nAbstract: The use of under-utilized Internet resources is widely recognized as a viable\\nform of high performance computing. Sustained processing power of roughly 40T\\nFLOPS using 4 million volunteered Internet hosts has been reported for\\nembarrassingly parallel problems. At the same time, peer-to-peer (P2P) file\\nsharing networks, with more than 50 million participants, have demonstrated the\\ncapacity for scale in distributed systems. This paper contributes a study of\\nload balancing techniques for a general class of loosely-synchronous parallel\\nalgorithms when executed over a P2P network. We show that decentralized,\\ndiffusive load balancing can be effective at balancing load and is facilitated\\nby the dynamic properties of P2P. While a moderate degree of dynamicity can\\nbenefit load balancing, significant dynamicity hinders the parallel program\\nperformance due to the need for increased load migration. To the best of our\\nknowledge this study provides new insight into the performance of\\nloosely-synchronous parallel programs over the Internet.\",\n \"Title: Improving performance in delay/disruption tolerant networks through passive relay points\\nYear: 2012\\n\\nAbstract: In this paper, we study the case of a limited number of mobile nodes trying to communicate in a large geographic area, forming a delay/disruption tolerant network (DTN). In such networks the mobile nodes are disconnected for significantly long periods of time. Traditional routing protocols proposed for mobile ad hoc networks or mesh networks, which assume at least one path between each source and destination, are ineffective in DTNs. One approach to improve communication is through gossip based protocols because these protocols do not rely on a fixed path. Another approach is to control the movement of the mobile nodes and/or use special mobile nodes called ferry nodes. Others try to employ a fixed infrastructure including stationary relay points. One scheme in stationary relay point approach is to use base stations as relay points which need their own power supply. In this paper, we study a passive approach where mobile nodes deposit/retrieve messages to/ from known stationary locations in the geographic region. Messages are delivered from a source by being deposited at one or more locations that are later visited by the destination. A proposed implementation of our approach using read/writable passive Radio Frequency Identification (RFID) tags, one per point location, is considered in this work. Passive RFID technology is desirable because it operates wirelessly and without the need for attached power. Our simulation results indicate that our approach can achieve competitive message delay and delivery rates. We also demonstrate several techniques for optimizing the stationary relay node placement, namely relay pruning, probability based relay distribution and a genetic algorithm; the genetic algorithm is shown to provide the best solutions to this problem.\",\n \"Title: Batched Circuit Switched Routing for Efficient Service of Requests\\nYear: 2000\\n\\nAbstract: Circuit switching protocols were largely developed in light of ATM virtual circuits and also for the case of all-optical networks. The merging of ATM and IP packet switching technology may banish true circuit switching in the digital domain due to fast switching networks using TDM. Even so, the characteristics of all-optical light paths (an optical circuit switched path) are so desirable that digital WAN's using ATM may eventually be completely augmented by optical WAN's using a hybrid opto-electronic or complete optical switching fabric. Afundamental problem is servicing a large number of requests while maintaining a low blocking probability where a request is blocked if the path it requires is in use by a circuit. We present a simple routing method, namely Batched Circuit Switching (BCS), that batches requests according to the rate of arrival and service time. Compared to orthodox Circuit Switching (CS), BCS provides superior request service times due to the observation that many requests may traverse the same circuit (or channel) during the batch time interval, rather than being blocked by preceding requests (as is the case for CS). Thus, BCS exhibits better utilization of thenetwork resources than CS. We demonstrate the requirements for this system to be stable, namely that the batch size must not increase to infinity before being serviced and the relationship between \\\\lambda (the arrival rate of requests) and µ (the service rate ofrequests) to ensure stability.\",\n \"Title: Coordinated load management in Peer-to-Peer coupled federated grid systems\\nYear: 2012\\n\\nAbstract: This paper proposes a coordinated load management protocol for Peer-to-Peer (P2P) coupled federated Grid systems. The participants in the system, such as the resource providers and the consumers who belong to multiple control domains, work together to enable a coordinated federation. The coordinated load management protocol embeds a logical spatial index over a Distributed Hash Table (DHT) space for efficient management of the coordination objects; the DHT-based space serves as a kind of decentralized blackboard system. We show that our coordination protocol has a message complexity that is logarithmic to the number of nodes in the system, which is significantly better than existing broadcast based coordination protocols.The proposed load management protocol can be applied for efficiently coordinating resource brokering services of distributed computing systems such as grids and PlanetLab. Resource brokering services are the main components that control the way applications are scheduled, managed and allocated in a distributed, heterogeneous, and dynamic Grid computing environments. Existing Grid resource brokers, e-Science application work-flow schedulers, operate in tandem but still lack a coordination mechanism that can lead to efficient application schedules across distributed resources. Further, lack of coordination exacerbates the utilization of various resources (such as computing cycles and network bandwidth). The feasibility of the proposed coordinated load management protocol is studied through extensive simulations.\",\n \"Title: Incorporating Fault Tolerance with Replication on Very Large Scale Grids\\nYear: 2007\\n\\nAbstract: Providing fault tolerance for message passing parallel application on a distributed environment is a rule rather than an exception. A node failure can cause the whole computation to stop and has to be restarted from the begin- ning if no fault tolerance is available. However, introducing fault tolerance has some overhead on speedup that can be achieved. In this paper, we introduce a new technique called replication with cross-over packets for reliability and to in- crease fault tolerance over Very Large Scale Grids (VLSG). This technique has two pronged effect of avoiding single point of failure and single link of failure. We incorporate this new technique into the L-BSP model and show the pos- sible speedup of parallel process. We also derive the achiev- able speedup for some fundamental parallel algorithms us- ing this technique.\",\n \"Title: P2P-Tuple: Towards a Robust Volunteer Computing Platform\\nYear: 2009\\n\\nAbstract: The evolution of emerging volunteer computing systems is placing an ever greater emphasis on the use of a loosely coupled decentralized architecture, mainly to address concerns of scalability and robustness. We propose to build the next generation volunteer computing systems on top of well studied peer-to-peer techniques to fully take advantage of its decentralized characteristic and its very large, shared data storage capacity. This paper sketches the design of such a completely decentralized and fault tolerant volunteer computing system by using a peer-to-peer based tuple space formed by volunteer peers. Both simulation and software prototype evaluation results show support for the proposed design even when all peers are having a high churn rate of Mean Time Before Failure of 2 hours.\",\n \"Title: Using a distributed quadtree index in peer-to-peer networks\\nYear: 2007\\n\\nAbstract: Peer-to-peer (P2P) networks have become a powerful means for online data exchange. Currently, users are primarily utilizing these networks to perform exact-match queries and retrieve complete files. However, future more data intensive applications, such as P2P auction networks, P2P job-search networks, P2P multiplayer games, will require the capability to respond to more complex queries such as range queries involving numerous data types including those that have a spatial component. In this paper, a distributed quadtree index that adapts the MX-CIF quadtree is described that enables more powerful accesses to data in P2P networks. This index has been implemented for various prototype P2P applications and results of experiments are presented. Our index is easy to use, scalable, and exhibits good load-balancing properties. Similar indices can be constructed for various multidimensional data types with both spatial and non-spatial components.\",\n \"Title: Decentralisation of scorefinder: a framework for credibility management on user-generated contents\\nYear: 2010\\n\\nAbstract: User-generated content (UGC) from Internet users has significant value only when its credibility can be established A basic approach to establishing credibility is to take an average of scores from annotators, while more sophisticated approaches have been used to eliminate anomalous scoring behaviour by giving different weights to scores from different annotator profiles A number of applications such as file sharing and article reviewing use a decentralised architecture While computing a weighted average of static values in a decentralised application is well studied, sophisticated UGC algorithms are more complicated since source values to be aggregated and their weights may change in time In our work we consider a centralised credibility management algorithm, ScoreFinder, as an example, and show both structured and unstructured approaches for computing time-dependent weighted average values in peer-to-peer (P2P) networks Experimental results on two real data sets demonstrate that our approaches converge and deliver results comparable to those from the centralised version of ScoreFinder.\",\n \"Title: A Case for Cooperative and Incentive-Based Coupling of Distributed Clusters\\nYear: 2006\\n\\nAbstract: Interest in grid computing has grown significantly over the past five years. Management of distributed cluster resources is a key issue in grid computing. Central to management of resources is the effectiveness of resource allocation, as it determines the overall utility of the system. In this paper, we propose a new grid system that consists of grid federation agents which couple together distributed cluster resources to enable a cooperative environment. The agents use a computational economy methodology, that facilitates QoS scheduling, with a cost-time scheduling heuristic based on a scalable, shared federation directory. We show by simulation, while some users that are local to popular resources can experience higher cost and/or longer delays, the overall users' QoS demands across the federation are better met. Also, the federation's average case message passing complexity is seen to be scalable, though some jobs in the system may lead to large numbers of messages before being scheduled\",\n \"Title: A case for cooperative and incentive-based federation of distributed clusters\\nYear: 2008\\n\\nAbstract: Research interest in Grid computing has grown significantly over the past five years. Management of distributed resources is one of the key issues in Grid computing. Central to management of resources is the effectiveness of resource allocation as it determines the overall utility of the system. The current approaches to brokering in a Grid environment are non-coordinated since application-level schedulers or brokers make scheduling decisions independently of the others in the system. Clearly, this can exacerbate the load sharing and utilization problems of distributed resources due to sub-optimal schedules that are likely to occur. To overcome these limitations, we propose a mechanism for coordinated sharing of distributed clusters based on computational economy. The resulting environment, called Grid-Federation, allows the transparent use of resources from the federation when local resources are insufficient to meet its users' requirements. The use of computational economy methodology in coordinating resource allocation not only facilitates the Quality of Service (QoS)-based scheduling, but also enhances utility delivered by resources. We show by simulation, while some users that are local to popular resources can experience higher cost and/or longer delays, the overall users' QoS demands across the federation are better met. Also, the federation's average case message-passing complexity is seen to be scalable, though some jobs in the system may lead to large numbers of messages before being scheduled.\",\n \"Title: On Placement of Passive Stationary Relay Points in Delay Tolerant Networking\\nYear: 2011\\n\\nAbstract: Recently, there has been focus on augmenting Delay/Disruption Tolerant Networks (DTNs) with easily deployable stationary relay nodes making an unconnected infrastructure to facilitate the data delivery by increasing forwarding opportunities. Relay nodes are capable of downloading, storing, and forwarding the data messages from/to the mobile nodes. Placing the relay nodes is an important issue in DTNs as the performance of the network is dependent to their positions. Relay placement is an NP-hard problem hence it makes it a more complicated issue in DTNs. Existing works in the literature are based on simulation which are suffering from computational complexities dictated by simulation. Moreover, they are optimizing the relay placement only based on specific scenarios. In this paper, we propose a generic analytical model in order to evaluate the performance of DTNs in presence of relay nodes. Our model is dependent on the mobile nodes' mobility pattern, and we consider the case when the mobile nodes move according to the random waypoint model. In order to use the proposed model for placing the relays efficiently, we utilize two heuristic approaches. The first approach is based on optimization of the network performance using simulated annealing and the second one relies on a greedy approach to find the best location for each relay one at a time. Our simulation results show that our approaches outperform the simulation based approaches in terms of data delivery performance.\",\n \"Title: Distributed scheduling schemes for wireless mesh networks: A survey\\nYear: 2013\\n\\nAbstract: An efficient scheduling scheme is a crucial part of Wireless Mesh Networks (WMNs)—an emerging communication infrastructure solution for autonomy, scalability, higher throughput, lower delay metrics, energy efficiency, and other service-level guarantees. Distributed schedulers are preferred due to better scalability, smaller setup delays, smaller management overheads, no single point of failure, and for avoiding bottlenecks. Based on the sequence in which nodes access the shared medium, repetitiveness, and determinism, distributed schedulers that are supported by wireless mesh standards can be classified as either random, pseudo-random, or cyclic schemes. We performed qualitative and quantitative studies that show the strengths and weaknesses of each category, and how the schemes complement each other. We discuss how wireless standards with mesh definitions have evolved by incorporating and enhancing one or more of these schemes. Emerging trends and research problems remaining for future research also have been identified.\",\n \"Title: Cooperative repair of wireless broadcasts\\nYear: 2011\\n\\nAbstract: Wireless broadcasting systems, such as Digital Video Broadcasting (DVB), are subject to signal degradation, having an effect on end users' reception quality. Clearly, reception quality can be improved by increasing signal strength. This, however, comes at a significantly increased energy use, with adverse environmental and financial consequences, notably in either sparsely populated rural regions, or overly built and difficult to penetrate dense urban areas. This paper discusses our ongoing work on an alternative approach to improving reception quality, based on the collaborative repair of lossy packet streams among the community of DVB viewers. We present our main idea, the crucial design decisions, the algorithm, as well as preliminary results demonstrating the feasibility and efficiency of this approach.\",\n \"Title: Event Based Simulator for Parallel Computing over the Wide Area Network for Real Time Visualization\\nYear: 2009\\n\\nAbstract: As the computational requirement of applications in computational science continues to grow tremendously, the use of computational resources distributed across the Wide Area Network (WAN) becomes advantageous. However, not all applications can be executed over the WAN due to communication overhead that can drastically slowdown the computation. In this paper, we introduce an event based simulator to investigate the performance of parallel algorithms executed over the WAN. The event based simulator known as SIMPAR (SIMulator for PARallel computation), simulates the actual computations and communications involved in parallel computation over the WAN using time stamps. Visualization of real time applications require steady stream of processed data flow for visualization purposes. Hence, SIMPAR may prove to be a valuable tool to investigate types of applications and computing resource requirements to provide uninterrupted flow of processed data for real time visualization purposes. The results obtained from the simulation show concurrence with the expected performance using the L-BSP model.\",\n \"Title: Stable High-Capacity One-Hop Distributed Hash Tables\\nYear: 2006\\n\\nAbstract: Most research on Distributed Hash Tables (DHTs) assumes ephemeral, lightly loaded deployments. Each node has a lifetime of a few hours and initiates a lookup once every few seconds or minutes. However, in giant internet data centers, each node has a lifetime of weeks or months and initiates hundreds or thousands of lookups every second. In such an environment, one-hop DHTs are superior to multi-hop DHTs. They use lookup bandwidth more efficiently. We qualify conflicting research to show that a single onehop DHT can indeed scale to at least a few hundred thousand nodes in stable, high-capacity enterprise networks. Two new designs are presented: One Hop Sites (1HS), a high-capacity DHT tailored for site redundancy; and the One Hop Federation (1HF), a global, hierarchic DHT that resolves an open latency problem. For both, the analysis a) confirms linear scalability to at least a few hundred thousand nodes and b) identifies the most sensitive design parameters.\",\n \"Title: Global Detection of Flooding-Based DDoS Attacks Using a Cooperative Overlay Network\\nYear: 2010\\n\\nAbstract: Flooding-based Distributed Denial of Service (DDoS) attacks present a serious threat to the stability of the Internet. Identifying the attacks rapidly and accurately is significant for the efficient operation of Internet applications and services. Recent observations in the U.S. indicate a significant increase of cyber attacks on U.S. military information systems in 2009. Current technologies are still unable to withstand large-scale DDoS attacks. Single point detection and response is a first step to defeat such distributed attacks. Distributed global defense systems, using a coordinated effort, go much further towards thwarting such attacks. In this paper, we propose a distributed defense infrastructure to detect DDoS attacks globally using a cooperative overlay network and a gossip-based information exchange protocol. Our NS2 based simulation results show that the proposed solution can detect attacks with a detection rate as high as 0.99 with false alarms below 0.01. This compares favorably against other widely known methods including change-point detection, TTL analysis and wavelet analysis.\",\n \"Title: Using fundamental electrical theory for varying time quantum uni-processor scheduling\\nYear: 2001\\n\\nAbstract: Given the total number of instructions to be completed on a uni-processor system and the average cycle time per instruction we introduce a method of calculating time quantum allocation to individual fine grain tasks. The main theory behind our method is based on fundamental equations describing electrical phenomenon. We show how electric circuit analysis can be used to describe the fundamental scheduling problem, and provide a framework for defining more elaborate scheduling problems such as multiprocessor and multicomputer task scheduling. As a matter of physical soundness we demonstrate through unit derivation that our electrical analogy provides proper physical quantities that are supported by current literature. Our analysis shows that variable time round-robin scheduling (VTRR) provides a more appropriate means of scheduling fine-grain tasks than constant time round-robin scheduling (CTRR). We prove that, our VTRR scheduler always completes at least one task per cycle . We show through numerical comparisons some differences between VTRR and CTRR performance.\",\n \"Title: Mean Field Models of Message Throughput in Dynamic Peer-to-Peer Systems\\nYear: 2007\\n\\nAbstract: The churn rate of a peer-to-peer system places direct limitations on the rate at which messages can be effectively communicated to a group of peers. These limitations are independent of the topology and message transmission latency. In this paper we consider a peer-to-peer network, based on the Engset model, where peers arrive and depart independently at random. We show how the arrival and departure rates directly limit the capacity for message streams to be broadcast to all other peers, by deriving mean field models that accurately describe the system behavior. Our models cover the unit and more general k buffer cases, i.e. where a peer can buffer at most k messages at any one time, and we give results for both single and multi-source message streams. We define coverage rate as peer-messages per unit time, i.e. the rate at which a number of peers receive messages, and show that the coverage rate is limited by the churn rate and buffer size. Our theory introduces an Instantaneous Message Exchange (IME) model and provides a template for further analysis of more complicated systems. Using the IME model, and assuming random processes, we have obtained very accurate equations of the system dynam- ics in a variety of interesting cases, that allow us to tune a peer-to-peer system. It remains to be seen if we can maintain this accuracy for general processes and when applying a non-instantaneous model.\",\n \"Title: Fast Channel Establishment for IEEE 802.16 Wireless Mesh Networks\\nYear: 2010\\n\\nAbstract: IEEE 802.16 based wireless mesh networks (WMNs) are a promising broadband access solution to support flexibility, cost effectiveness and fast deployment of the fourth generation infrastructure based wireless networks. Reducing the time for channel establishment is critical for low latency/interactive applications. In this paper, we propose two modifications to the election based transmission timing (EBTT) mechanism used in the coordinated distributed scheduling (CDS) within the IEEE 802.16 standard, which makes use of the standardized signaling messages and formats. Simulation results show that our proposed modifications results in significantly reduced channel establishment delay - up to 75% with low neighborhood densities - over the standard, and at least 10% with high neighborhood densities. Moreover, the proposed mechanisms are also applicable for the three-hop extended neighborhood CDS when required.\",\n \"Title: An Adaptive Checkpointing Scheme for Peer-to-Peer Based Volunteer Computing Work Flows\\nYear: 2008\\n\\nAbstract: Volunteer computing, sometimes called public resource computing, is an emerging computational model that is very suitable for work-pooled parallel processing. As more complex grid applications make use of work flows in their design and deployment it is reasonable to consider the impact of work flow deployment over a volunteer computing infrastructure. In this case, the inter work flow I/O can lead to a significant increase in I/O demands at the work pool server. A possible solution is the use of a peer-to-peer based parallel computing architecture to off-load this I/O demand to the workers; where the workers can fulfill some aspects of work flow coordination and I/O checking, etc. However, achieving robustness in such a large scale system is a challenging hurdle towards the decentralized execution of work flows and general parallel processes. To increase robustness, we propose and show the merits of using an adaptive checkpoint scheme that efficiently checkpoints the status of the parallel processes according to the estimation of relevant network and peer parameters. Based on our proposed mathematical checkpoint model, our scheme uses statistical data observed during runtime to dynamically make checkpoint decisions in a completely decentralized manner. The results of simulation show support for our proposed approach in terms of reduced required runtime.\",\n \"Title: Map-based spatio-temporal interpolation in vehicle trajectory data using routing web-services\\nYear: 2012\\n\\nAbstract: Interpolation of motion data is a challenging problem that is often overlooked by researchers when using GPS data with low sampling rate. In this paper we define a spatio-temporal query called Historical Spatio-Temporal Query (HSTQ). This query receives a time and returns a waypoint containing spatio-temporal data of a moving object at that particular time. To respond to the query for any given time (e.g. every second), we need to interpolate missing waypoints of the trajectory. Linear Interpolation (LI) is the most commonly used method although it can be grossly inaccurate. To deal with this problem, we propose a method called Map-Based Interpolation (MBI). This method uses routing web-services to find significant points (path) between two waypoints. Instead of using road networks, which is typically used by methods such as map-matching, we send a query to a routing web service and analyse the returned data to find missing waypoints. To be able to respond to queries sent to HSTQ for any input time within a trajectory period, we use a combination of MBI and LI (MBI+LI) to interpolate the missing data. We also propose two measures for comparing performance of our interpolation method with LI. Experimental results using realistic trajectories show significant improvement in quality and accuracy of interpolated and down-sampled trajectory data using HSTQ with MBI+LI in comparison to the earlier, widely used LI method.\",\n \"Title: Lossy Bulk Synchronous Parallel Processing Model for Very Large Scale Grids\\nYear: 2006\\n\\nAbstract: The performance of a parallel algorithm in a very large scale grid is significantly influenced by the underlyin g Internet protocols and inter-connectivity. Many grid program- ming platforms use TCP due to its reliability, usually with some optimizations to reduce its costs. However, TCP does not perform well in a high bandwidth and high delay network environment. On the other hand, UDP is the fastest protocol avail- able because it omits connection setup process, acknowledgments and retransmissions sacrificing reliable transfer. Many ne w bulk data transfer schemes using UDP for data transmission such as RBUDP, Tsunami, and SABUL have been introduced and shown to have better performance compared to TCP. In this paper, we consider the use of UDP and examine the relationship between packet loss and speedup with respect to the number of grid nodes. Our measurement suggests that packet loss rates between 5%- 15% on average are not uncommon between PlanetLab nodes that are widely distributed over the Internet. We show that transmitting multiple copies of same packet produces higher speedup. We show the minimum number of packet duplication required to maximize the possible speedup for a given number of nodes using a BSP based model. Our work demonstrates that by using an appropriate number of packet copies, we can increase performance of parallel program.\",\n \"Title: Towards Parallel Computing on the Internet: Applications, Architectures, Models and Programming Tools\\nYear: 2006\\n\\nAbstract: The development of Internet wide resources for general purpose parallel\\ncomputing poses the challenging task of matching computation and communication\\ncomplexity. A number of parallel computing models exist that address this for\\ntraditional parallel architectures, and there are a number of emerging models\\nthat attempt to do this for large scale Internet-based systems like\\ncomputational grids. In this survey we cover the three fundamental aspects --\\napplication, architecture and model, and we show how they have been developed\\nover the last decade. We also cover programming tools that are currently being\\nused for parallel programming in computational grids. The trend in conventional\\ncomputational models are to put emphasis on efficient communication between\\nparticipating nodes by adapting different types of communication to network\\nconditions. Effects of dynamism and uncertainties that arise in large scale\\nsystems are evidently important to understand and yet there is currently little\\nwork that addresses this from a parallel computing perspective.\",\n \"Title: A method of trading diameter for reduced degree to construct low cost interconnection networks\\nYear: 1999\\n\\nAbstract: Classical network topology has identified many “bottom-up” approaches to designing low cost interconnection networks. The foremost figure of merit is considered to be cost, the product of degree and diameter. We argue that average cost, being the product of average degree and diameter, is more applicable than the conventional cost and propose a general method of Yap-down” network construction that provides networks of average cost 0(log, N) where B = o(N) is a bound on maximum degree and N is the number of nodes. From this we show an example topology that has averagecostofQ(,J~~~N). By doing so we examine a class of networks with constant average cost, that is, networks whose average cost is fixed to a constant value as the size of the network increases to infinity. We then identify those aspects that are undesirable, namely some nodes have infinite dcgree, and show methods of trading an increase in diameter for a reduction in degree.\",\n \"Title: A support platform for event detection using social intelligence\\nYear: 2012\\n\\nAbstract: This paper describes a system designed to support event detection over Twitter. The system operates by querying the data stream with a user-specified set of keywords, filtering out non-English messages, and probabilistically geolocating each message. The user can dynamically set a probability threshold over the geolocation predictions, and also the time interval to present data for.\",\n \"Title: A distributed network architecture for robust internet voting systems\\nYear: 2005\\n\\nAbstract: Web-based Internet voting services cannot be provided in a reliable manner where a publicly visible website forms the basis of interacting with voters and collecting votes, as the service cannot resist a distributed denial of service attack. The problem is a profound one for all interactive web services, however, an Internet voting session is a well defined and quite simple interaction and an interactive service similar to web-based voting is proposed. This architecture uses a P2P web cache as the source of a Java voting applet. The applet interacts with the voter via their web browser to collect a vote. The applet then uses a common P2P file sharing network to deposit encrypted votes for collection by the electoral returning officer. The issues of how to authenticate voters and how to provide receipting in such an asynchronous system are described along with premininary .ndings using a large P2P network established for this purpose. A successful demonstration of this approach may have implications for other mission critical web services which must collect information.\",\n \"Title: Building and Querying a P2P Virtual World\\nYear: 2006\\n\\nAbstract: Peer-to-Peer (P2P) systems are known to provide excellent scalability in a networked environment. One peer is introduced to the system by each participant. However current P2P applications can only provide file sharing and other forms of relatively simple data communications, and, in this paper, we demonstrate how this limitation can be bridged by indexing and querying a 3D virtual-world on a dynamic distributed network. We present an algorithm for 3D range queries as well as an algorithm for nearest neighbor queries. We also show how to build such a complex application from the ground level of a P2P routing algorithm.\",\n \"Title: Topology Dissemination for Reliable One-Hop Distributed Hash Tables\\nYear: 2009\\n\\nAbstract: Many distributed hash tables (DHTs) resolve lookups in O(\\\\log n) hops, where n is the number of nodes. One-hop DHTs give lower lookup latencies and lower lookup failure rates. However, it is hard to maintain large, wide-area one-hop topologies. We contribute aecast, a new topology dissemination algorithm for one-hop DHTs. It avoids expensive repair mechanisms and critical points of failure in existing one-hop DHTs. When a node discovers by anti-entropy that it has missed a topology update, it initiates \\\"controlled flooding,” sending the update to nodes in the multicast tree that also missed the update. We compare aecast with a widely cited epidemic multicasting algorithm, pbcast, by analysis and simulation. Aecast gives at least fivefold fewer out-of-date nodes on average within one round of a topology update. We support it with a fault-tolerant topology agreement protocol, so that only legitimate topology changes propagate throughout the overlay. Consequently, we argue that one-hop DHTs deserve greater attention for Internet applications in which reasonably reliable nodes carry high lookup loads.\",\n \"Title: CAESAR: middleware for complex service-oriented peer-to-peer applications\\nYear: 2007\\n\\nAbstract: Recent research advances in Peer-to-Peer (P2P) computing have enabled the P2P paradigm to be used for developing complex applications beyond file sharing and data storage. These applications have drawn significant benefits, specifically scalability and low cost, from the P2P paradigm. However, the current approach for designing P2P applications introduce issues that prevent the development of high quality complex P2P applications. These issues, namely tight coupling to P2P protocols, limited logic sharing between peers and complicated recovery processes, motivate us to introduce a service-oriented architecture for P2P applications. We have developed a middleware called CAESAR to support the development of service-oriented P2P applications applying the principles of abstraction, dynamic binding, loose coupling and information hiding. In this paper, we discuss the design principles and the components of CAESAR middleware, as well as our experiences in using CAESAR to develop several service-oriented P2P applications.\",\n \"Title: SLA-Based Coordinated Superscheduling Scheme and Performance for Computational Grids\\nYear: 2006\\n\\nAbstract: The Service Level Agreement (SLA) based grid su- perscheduling approach promotes coordinated resource sharing. Superscheduling is facilitated between admin- istratively and topologically distributed grid sites by gr id schedulers such as Resource brokers. In this work, we present a market-based SLA coordination mechanism. We based our SLA model on a well knowncontract net protocol. The key advantages of our approach are that it al- lows: (i) resource owners to have finer degree of con- trol over the resource allocation that was previously not possible through traditional mechanism; and (ii) super- schedulers to bid for SLA contracts in the contract net with focus on completing the job within the user speci- fied deadline. In this work, we use simulation to show the effectiveness of our proposed approach.\",\n \"Title: Adaptive expression based routing protocol for P2P systems\\nYear: 2007\\n\\nAbstract: There is an emerging trend of using P2P systems for computational and data intensive tasks, such as online collaborations, distributed database applications and message passing interface (MPI) algorithms. For better resource utilization in these and similar applications, it is necessary to discover the resource capabilities of the collaborating peers. Another requirement of these applications is to locate the optimal resource based on a search criteria, e.g., to seek a resource with earliest execution time. We define this kind of routing as expression based routing. However, in the absence of the centralized controlling node, tracking the capabilities of the participating peers is very difficult. Moreover peers join and leave the system dynamically and, thus, makes the discovery of a desired resource even more complex. In this paper we investigate a novel algorithm for expression based routing in a P2P system. It evaluates peer u such that lu (the value of u) is minimum where e.g. lu is load of the peer u. Our contribution includes a detailed algorithm to search for the least loaded peer in the system. We analyze the accuracy and cost of our proposed algorithm through detailed simulations.\",\n \"Title: A Scalable Anonymous Server Overlay Network\\nYear: 2006\\n\\nAbstract: Large scale malicious activities continue to take place on the Internet. Safeguarding vulnerable computers and servers is therefore important. Recently, systems have been proposed to protect network addresses of computers from malicious adversaries on the Internet. Some of these systems focus specifically on providing client-side anonymity. However, as the Internet bandwidth for average users continues to increase, individuals are increasingly capable of carrying out complicated tasks as servers. It is in the interest of the individuals to gain server anonymity in which the network address of the machine remains protected from malicious activities. In this paper, we present a Scalable Anonymous Server Overlay Network (SASON) that aims at protecting network addresses of all participating servers. SASON also provides reliable server discovery and routing through the use of server pseudonyms and distributed hashtables.\",\n \"Title: A Taxonomy of Peer-to-Peer Based Complex Queries: a Grid perspective\\nYear: 2006\\n\\nAbstract: Grid superscheduling requires support for efficient and sca lable discovery of resources. Resource discovery ac- tivities involve searching for the appropriate resource ty pes that match the user's job requirements. To accomplish this goal, a resource discovery system that supports the desired look-up operation is mandatory. Various kinds of so- lutions to this problem have been suggested, including the centralised and hierarchical information server approach. However, both of these approaches have serious limitations in regards to scalability, fault-tolerance and network congestion. To overcome these limitations, organising resource information using Peer-to-Peer (P2P) network model has been proposed. Existing approaches advocate an extension to structured P2P protocols, to support the Grid re- source information system (GRIS). In this paper, we identify issues related to the design of such an efficient, scalable, fault-tolerant, consistent and practical GRIS system usin g a P2P network model. We compile these issues into various taxonomies in sections 3 and 4. Further, we look into existing works that apply P2P based network protocols to GRIS. We think that this taxonomy and its mapping to relevant systems would be useful for academic and industry based researchers who are engaged in the design of scalable Grid systems.\",\n \"Title: An unstructured peer-to-peer approach to aggregate node selection\\nYear: 2008\\n\\nAbstract: In distributed and high performance applications it it is necessary to discover the aggregate resource capabilities of the collaborating peers. Another requirement of these applications is to locate the resources meeting the given aggregate search criteria. In this paper we present an unstructured peer-to-peer approach to aggregate node selection for resource management. Our objective is to efficiently calculate the aggregate value and search the aggregate node whose attribute matches or falls in the proximity of this value. Our contribution also includes the inclusion of an explicit network for aggregate query routing. We evaluated our protocol for accuracy, efficiency and communication cost through detailed simulations.\",\n \"Title: Decentralised Resource Discovery Service for Large Scale Federated Grids\\nYear: 2007\\n\\nAbstract: Efficient Resource discovery mechanism is one of the fundamental requirement for Grid computing systems, as it aids in resource management and scheduling of applications. Resource discovery involves searching for resources that match the user's application requirements. Various kinds of solutions to Grid resource discovery have been developed, including the centralised and hierarchical information server approach. However, these approaches have serious limitations in regards to scalability, fault-tolerance and network congestion. To overcome such limitations, we propose a decentralised Grid resource discovery system based on a spatial publish/subscribe index. It utilises a Distributed Hash Table (DHT) routing substrate for delegation of d-dimensional service messages. Our approach has been validated using a simulated publish/subscribe index that assigns regions of a d-dimensional resource attribute space to the Grid peers in the system. We generated the resource attribute distribution using the configurations obtained from the Top 500 Supercomputer list. The simulation study takes into account various parameters such as resource query rate, index load distribution, number of index messages generated, overlay routing hops and system size. Our results show that grid resource query rate directly affects the performance of the decentralised resource discovery system, and that at higher rates the queries can experience considerable latencies. Further, contrary to what one can expect, system size does not have a significant impact on the performance of the system, in particular the query latency.\",\n \"Title: Achieving Ubiquitous Network Connectivity Using an RFID Tag-Based Routing Protocol\\nYear: 2008\\n\\nAbstract: Ubiquitous network connectivity is challenged in areas outside of the range of wireless base stations and cellular networks. In such networks the mobile nodes are disconnected for significantly long periods of time. In this paper, we propose an approach where mobile nodes deposit/retrieve messages to/from known point locations in the geographic region. Messages are delivered from a source by being deposited at one or more locations that are later visited by the destination. We propose the use of passive, read/writable radio frequency identification tags, one per point location. Our simulation results indicate that our approach can achieve competitive message latency and delivery rates. We also provide techniques for optimizing the tag placement, including tag pruning, probability based tag distribution and a genetic algorithm; the genetic algorithm is shown to provide the best solutions to this problem.\",\n \"Title: Algorithmic-Parameter optimization of a parallelized split-step fourier transform using a modified BSP cost model\\nYear: 2004\\n\\nAbstract: Adaptive algorithms are increasingly acknowledged in leading parallel and distributed research. In the past, algorithms were manually tuned to be executed efficiently on a particular architecture. However, interest has shifted towards algorithms that can adapt themselves to the computational resources. A cost model representing the behavior of the system (i.e. system parameters) and the algorithm (i.e algorithm parameters) plays an important role in adaptive parallel algorithms. In this paper, we contribute a computational model based on Bulk Synchronous Parallel processing that predicts performance of a parallelized split-step Fourier transform. We extracted the system parameters of a cluster (upon which our algorithm was executed) and showed the use of an algorithmic parameter in the model that exhibits optimal behavior. Our model can thus be used for the purpose of self-adaption.\",\n \"Title: GWISFI: A universal GPU interface for exhaustive search of pairwise interactions in case-control GWAS in minutes\\nYear: 2014\\n\\nAbstract: Epistatic interactions between genes are believed to be a critical component in the genetic architecture of complex diseases. Genome Wide Association Studies (GWAS) may be able to detect such genetic interactions indirectly, via the identification of associated SNP markers. Major obstacles to progress in this area are: the unknown nature of epistatic interactions, little understanding of the capabilities of different filtering methods, and the computational difficulties for exhaustive analysis. A common platform enabling various detection methods is needed to avoid practical issues such as software compatibility and portability, incompatible input and output formats and varying demands on computational resources. We developed a highly optimised GPU system capable of exhaustively analysing all SNP-pairs in typical GWAS data (0.5M SNPs, 5K samples) in a few minutes on a standard desktop computer. A number of programming elements provided by a functional interface can be used to construct user-defined statistical tests to efficiently score every SNP pair. As a proof of principle, we have implemented 8 methods from the literature via our interface. We have applied all of them using a single GPU to exhaustively scan the 7 popular WTCCC case-control GWAS datasets. We present timing results for these methods, both in their original software implementations and using our platform. Significant improvements in timing are observed, up to 10000 times for CPU implementations of the popular FastEpistasis in PLINK and up to 2 orders of magnitude for some GPU implementations in the literature. As an initial discovery we show plots for overlaps of list of selected pairs by 8 algorithms for Type 2 Diabetes, WTCCC data.\",\n \"Title: Multi-space distributed hash tables for multiple transport domains\\nYear: 2003\\n\\nAbstract: The use of consistent hashing to construct a dis- tributed hash table (DHT) has recently gained signiflcant interest from the distributed computing community, partic- ularly because algorithms have been devised which imple- ment a DHT using peer-to-peer (P2P) technology. However a common assumption is that every node that participates in the DHT is within the same transport domain. Some so- lutions make to this problem make use of an intermediate overlay network which itself can become messy. Other solu- tions rely on the widespread use of IPv6 or \\\\work-arounds\\\" for the limited IPv4. We propose a solution that uses a multi-space DHT algorithm to manifest a DHT space in each transport domain, with bridge nodes linking DHT spaces be- tween transport domains. The ostensive problem with this solution is to ensure that key requests are evenly distributed over all spaces. We provide a framework to solve this prob- lem and give an example solution for a simple 2-space DHT network.\",\n \"Title: Delay sensitive identity protection in peer-to-peer online gaming environments\\nYear: 2007\\n\\nAbstract: Peer-to-peer computing overcomes communication bottleneck problems associated with centralized game servers and provides an alternative mechanism to support massively multi-player online gaming (MMOG) applications. A potential security problem associated with this approach relates to player identity. In an MMOG, users from different parts of the network interact with each other in a virtual world. In a peer-to-peer model, these interactions happen directly between the peers and this leads to the IP addresses of peers being available in the network packets. Malicious users can extract the IP addresses of their opponents and the information can be used to gain unfair advantage by compromising their opponents’ computers. The well known approach to this problem is the use of anonymizing networks (onion routing and mixing), that anonymize the sender and the destination IP addresses from the peers along a path. However this approach introduces significant delays which are not desirable in MMOG applications. This paper proposes the use of a secret shared key to reduce computational delay and also provides a theoretical framework for trading off the strength of anonymity for reduced delay with respect to classes of interactions in the MMOG. The results suggest that appropriately low delays can be achieved with small reductions in anonymity strength.\",\n \"Title: Demonstrating BooSTER: The broadcast stream transmission epidemic repair\\nYear: 2012\\n\\nAbstract: Wireless broadcasting systems, such as Digital Video Broadcasting (DVB), are subject to signal degradation, having an effect on end users' reception quality. Reception quality can be improved by increasing signal strength, but this comes at a significantly increased energy use and still without guaranteeing error-free reception. BOOSTER is a system for cooperative repair of DVB streams. It operates based on a fully decentralized epidemic algorithm that boosts reception quality by cooperatively repairing lossy packet streams among the community of DVB viewers. In this demo, we demonstrate BOOSTER in a special configurable implementation that allows on-the-fly tweaking of the error rate, network topology, and some protocol parameters, allowing the audience to get a feeling on how they affect end-user experience and load distribution.\",\n \"Title: A gossip-based asynchronous aggregation protocol for P2P systems\\nYear: 2010\\n\\nAbstract: Gossip-based protocols are a popular approach to estimate aggregates in P2P systems. For gossip-based aggregation protocols, the accuracy of the estimated aggregate heavily depends upon conservation of the system mass. In this paper, we present a gossip-based asynchronous aggregation protocol. Our protocol conserves the system mass under asynchronous settings and churn. Simulation results show that our proposed protocol converges to the correct system aggregate even in the presence of very dynamic and adverse network delays and under churn.\",\n \"Title: Feature selection for user motion pattern recognition in mobile networks\\nYear: 2012\\n\\nAbstract: Mobility is a challenging issues in mobile networks which significantly impacts the performance of a variety of network protocols. Understanding user motion behavior can improve the performance of mobile network protocols in different aspects. Therefore, we consider how to analyze the motion behavior of mobile nodes in various environments. In this paper we propose a few mobility metrics useful for analysis of individual, collective and geographical behavior of mobile nodes and a simple supervised mobility pattern recognition method which is able to classify mobility traces into different mobility model classes using our proposed mobility metrics. The remaining challenge is to find appropriate features, extracted from mobility traces, which are able to distinguish between different motion behaviors with maximum accuracy and minimum overhead. We propose a method to find the most suitable feature sets for classification of different mobility traces with different motion behaviors. Simulation results show significant performance of our proposed mobility pattern recognition method using appropriate mobility metrics and feature sets.\",\n \"Title: Geometry of Interest (GOI): Spatio-Temporal Destination Extraction and Partitioning in GPS Trajectory Data.\\nYear: 2017\\n\\nAbstract: Nowadays large amounts of GPS trajectory data is continuously collected by GPS-enabled devices such as vehicles navigation systems and mobile phones. GPS trajectory data is useful for applications such as traffic management, location forecasting, and itinerary planning. Such applications often need to extract the time-stamped sequence of visited locations (SVLs) of the mobile objects. The nearest neighbor query (NNQ) is the most applied method for labeling the visited locations based on the IDs of the points of interests (POIs) in the process of SVL generation. NNQ in some scenarios is not accurate enough. To improve the quality of the extracted SVLs, instead of using NNQ, we label the visited locations as the IDs of the POIs which geometrically intersect with the GPS observations. Intersection operator requires the accurate geometries of the POIs which we refer to them as the geometries of interest (GOIs). In some application domains (e.g. movement trajectories of animals), adequate information about the POIs and their geometries may not be available a priori, or they may not be publicly accessible and, therefore, they need to be derived from GPS trajectory data. In this paper we propose a novel method for estimating the GOIs, which consists of three phases: (1) extracting the geometries of the stay regions; (2) constructing the geometry of destination regions based on the extracted stay regions; and (3) constructing the GOIs based on the geometries of the destination regions. Using the geometric similarity to known GOIs as the major evaluation criterion, the experiments we performed using long-term GPS trajectory data show that our method outperforms the state-of-the-art.\",\n \"Title: Using Para-Virtualization as the Basis for a Federated PlanetLab Architecture\\nYear: 2006\\n\\nAbstract: The federation of islands of large scale distributed systems is a difficult problem, in part because of the variety of systems involved. We categorise solutions into either the systems integration or the virtualization approach. Using the virtualization approach, we enable a federated architecture for the PlanetLab Internet testbed. The current implementation of PlanetLab has a central management authority responsible for the maintenance of the system's nodes and creation of virtual machines for users. We propose two additional entities, which separate aspects of a node and site's management, allowing viable federation, called the Site Manager and Federation Authority. Building on the existing virtualization used by PlanetLab, the realisation of our proposal makes use of Xen, a para-virtualization system for x86 hardware. We present the uses of Xen domains, or virtual machines, as hosts for both PlanetLab nodes and slices. This additional layer of virtualization is the basis of our federation architecture.\",\n \"Title: BrowserCloud: A Personal Cloud for Browser Session Migration and Management\\nYear: 2015\\n\\nAbstract: Web browsers are de facto clients for an ever-increasing range of web applications. At the same time, web users are accessing these applications from a wide range of devices. This paper presents a solution for runtime browser session migration and management, called BrowserCloud, which allows a user to securely manage multiple browsers from a personal or third-party Cloud service, migrate snapshots of active browser sessions between browsers over different devices, using a robust security module. The design of BrowserCloud is based on browser extensions/plugins that can preserve and restore browser session state, and a PHP server that stores browser sessions securely. We have tested our implementation over a range of increasingly complex web applications, including WebRTC and HTML5 video. To the best of our knowledge, our implementation is the most robust and secure approach to runtime browser session management to date.\",\n \"Title: Finding High-Level Topics and Tweet Labeling Using Topic Models\\nYear: 2015\\n\\nAbstract: Making sense of Twitter data streams is challenging due to the extremely high volume of data. One way to address this challenge is to consider these data streams as containing a set of high-level topics. In this research we address the problem of: given a collection of tweets about K high-level topics, how to find topic words that describe these topics as well as how to label each tweet with one of the K topics using a topic modeling approach. Current research has shown that applying topic modeling algorithms directly on tweets does not lead to good results. Hence one approach is to group related tweets together, so as to form a single “pseudo-document”, which is more informative than a single tweet. In this paper we evaluate different grouping schemes found in the literature and propose a new grouping scheme utilizing named entities and word collocations. Results show that our proposed scheme performs better than the existing approaches, to a some extent for all the test cases, and for both finding high-level topics and tweet labeling.\",\n \"Title: Optimizing indoor location recognition through wireless fingerprinting at the Ian Potter Museum of Art\\nYear: 2014\\n\\nAbstract: Indoor tracking of smartphones adds context to smartphone applications, enabling a range of smarter behaviours. The predicted use cases are many and varied, and include navigation, planning, advertising and communication. Potentially, indoor tracking could become as ubiquitous as GPS - however, all of these possibilities depend on being able to produce a reasonably accurate, reliable system which does not require specialised infrastructure. While professional systems using custom devices are able to achieve very high levels of accuracy (<;1 cm), consumer no-infrastructure systems struggle to achieve reliable room-level tracking. This paper focuses on the use of WiFi received signal strength indicator (RSSI) fingerprinting, a machine learning approach which currently seems to be the most promising option for consumer smartphones. We have undertaken experimentation and optimisation in a real-world, noisy environment - the Ian Potter Museum of Art - where we developed and deployed a no-infrastructure, indoor visitor tracking application. Data was collected in a trial involving several dozen users over a few weeks, who used the system extensively. This data was analysed with a range of current WiFi RSSI fingerprinting techniques and algorithms (WASP, Redpin (kNN), SSD, SVM, Gaussian Naive Bayes and Random Forests), and their efficacy was compared and improved where possible. Known challenges such as device heterogeneity are explored, and the consistency of signal levels, including magnetic fields, are examined. Large Random Forests (200 trees) were found to have the best performance, which was further improved by calibrating for average differences in RSSI between phone models, to achieve an average of 90% correct classification of exhibits within the top five hits.\",\n \"Title: Discriminative Concept Learning Network: Reveal High-Level Differential Concepts From Shallow Architecture\\nYear: 2015\\n\\nAbstract: A desired capability of deep learning is to understand the high-level, class-specific features via hierarchical features learning. However the training of deep architectures is costly comparing to simple shallow models. Bringing the high-level feature understanding into a simple shallow architecture remains an open question.We proposed a supervised learning algorithm, enabling binary classification along with an intrinsic ability of learning high-level discriminative concepts via a shallow neural network architecture. The physical architecture of the network has one hidden layer (also serving as the output layer) responsible for the classification and an input layer directly identifies the informative features that constitute the high-level differential concepts between the two classes.Compared to other shallow classifiers, we demonstrate its practicability in real world classification problems. We also illustrate the human-understandable, discriminative concepts learned from the two image recognition exercises. Lastly, we show how it is useful in validating the disease-associated genetic variants in human genome as a real diagnostic genomics application.\",\n \"Title: Benchmarking the performance of hadoop triple replication and erasure coding on a nation-wide distributed cloud\\nYear: 2015\\n\\nAbstract: Large Scale distributed storage systems play a vital role in maintaining data across storage locations globally. These systems use replication as the default mechanism for providing fault-tolerance. Recently, erasure codes are being used as a viable alternative to replication, since they provide the same fault-tolerance for reduced storage overhead. However, their performance is unclear in a geographically diverse distributed storage system. This paper compares the performance of triple replication with the erasure coding (Reed-Solomon codes) used in Apache Hadoop’s implementation of a distributed file system, on a cluster distributed across Australia that runs on the NeCTAR research cloud. Our results show that using erasure coding does not degrade the read performance in such a setting. We also compare the Hadoop’s code with a local reconstruction code, implemented in the XORBAS version of Hadoop. These codes perform well in our clusters but the performance gain observed in our results does not conform to the results reported. Hence, we need new codes that perform better, addressing the geographical diversity issue. We believe that our framework is readily usable to test a range of novel erasure codes that are being introduced in the literature.\",\n \"Title: On improving recovery performance in erasure code based geo-diverse storage clusters\\nYear: 2016\\n\\nAbstract: Erasure code based distributed storage systems are increasingly being used by storage providers for big data storage since they offer the same reliability as replication with significant decrease in the amount of storage required. But, when it comes to a storage system with data nodes spread across a very large geographical area, the code's recovery performance is affected by various factors, both network and computation related. In this paper, we expose an XOR based code supplemented with the ideas of parity duplication and rack awareness that could be adopted in such storage clusters to improve the recovery performance during node failures. We have implemented them on the erasure code module of the XORBAS version of the Hadoop Distributed File System (HDFS). For evaluating the performance of the proposed ideas, we employ a geo-diverse cluster on the NeCTAR research cloud. The experimental results show that the techniques aid in bringing down the data read for repair by a factor of 85% and repair duration by a factor of 57% during node failures, though resulting in an increased storage requirement of 21% as compared to the traditional Reed-Solomon codes used in HDFS. The sum of all these ideas could offer a better solution for a code based storage system spanning a wide geographical area that has storage constraints such that a triple replicated system is not affordable and at the same time has strict requirements on ensuring the minimal recovery time.\",\n \"Title: Distributed stream clustering using micro-clusters on Apache Storm.\\nYear: 2017\\n\\nAbstract: The recent need to extract real-time insights from data has driven the need for machine learning algorithms that can operate on data streams. Given the current extreme rates of data generation (around 5000 messages per second), these algorithms need to be able to handle data streams of very high velocity. Many current algorithms do not reach this requirement, in some cases processing only tens of messages per second.\",\n \"Title: A Multi-Stage Hierarchical Window Model With Application To Real-Time Graph Analysis\\nYear: 2017\\n\\nAbstract: The dynamic nature of rea-world networks, such as social networks and communication networks, has increased the focus towards real-time dynamic graph analysis. Observations made based on real-time analysis of dynamic graphs reflect the latest properties of the graph and have the most value in real-time analysis. Computing graph properties of large-scale, fast-evolving graphs in real-time is challenging due, not only to the high computational and memory cost, but also to the understanding of the result with respect to the data from which it was derived. This paper proposes a multi-stage hierarchical window model that can aid in rigorous understanding of complicated real-time results and we apply it to generate graphs based on real-time updates along with periodic computations on graph snapshots for processing dynamic graphs. Moreover, the paper discusses the utilization of parallel window computation. The paper evaluates the hierarchical model through analyzing graphs formed by coccurring hashtags in a Twitter data-stream.\",\n \"Title: Predicting The Stability Of Large-Scale Distributed Stream Processing Systems On The Cloud\\nYear: 2017\\n\\nAbstract: Large-scale topology-based stream processing systems are non-trivial to build and deploy. They require understanding of the performance, cost of deployment and considerations of potential downtime. Our work considers stability as a primary characteristic of these systems. By stability, we mean that unstable systems exhibit large-spikes in latency and can drop throughput frequently or unpredictably. Such instabilities can be due to variations of workloads or underlying hardware platforms that are often difficult to predict. To understand and tackle this for large-scale stream processing systems, we apply queueing theory and simulate the results through a series of experiments on the Cloud.\",\n \"Title: Optimum Benefit Protocol: A fast converging, bandwidth-efficient decentralized similarity overlay.\\nYear: 2017\\n\\nAbstract: Due to large volumes of data available online, techniques such as document classification and clustering are required for organization, analysis and management of data. Similarity-based Clustering (SBC) is used by many existing systems for filtering information. Decentralized gossip-based overlays offer a simple, robust and scalable solution to SBC clustering. Convergence and communication complexity are the two key areas of concern when SBC is implemented using these overlays. Convergence guarantees accurate clustering but costs bandwidth because these systems rely on message passing to achieve convergence. In this work, we address the long tail problem, experienced by low in-degree nodes in a long tailed similarity distribution, such as power-law distribution and propose a new SBC approach, Optimum Benefit Protocol (OBP), that converges more rapidly than existing approaches and reduces the long tail. The proposed protocol only sends messages that could benefit the receiver, reducing bandwidth to one fifth of the default. Our protocol obtains at least 90% convergence for a 900 node network, starting in a random configuration, in less than 10 cycles, for all observed experiments. We used real world distributions from Yahoo, Movielens, and Epinion datasets, for experimentation.\",\n \"Title: Search Result Personalization in Twitter Using Neural Word Embeddings.\\nYear: 2017\\n\\nAbstract: In recent years, Twitter has become one of the most popular microblogging avenues. Today it has more than 300 million monthly active users generating more than 500 million tweets everyday. Twitter users both post messages as well as search for messages. Current search results given by Twitter are chronologically ordered and often users have to manually scan through an overwhelming number of the tweets to find content of interest. This process can quickly become infeasible. Personalization techniques address this problem by learning the user interests and tailoring search results by matching them with the user's interests. Recent research on neural word embedding models, which represents each word in the vocabulary as a vector of real values, has gained much attention. These models learn word embeddings in such a way that contextually similar words have similar vectors. In this paper we propose a novel approach, PWEBA, for personalizing Twitter search, which uses neural word embeddings to model user interests. Our experimental results show that PWEBA outperforms existing approaches for all the evaluation metrics we have considered in this paper.\",\n \"Title: E-Storm: Replication-Based State Management in Distributed Stream Processing Systems\\nYear: 2017\\n\\nAbstract: Apache Storm is a fault-tolerant, distributed inmemory computation system for processing large volumes of high-velocity data in real-time. As an integral part of the fault-tolerance mechanism, Storm's state management is achieved by a checkpointing framework, which commits states regularly and recovers lost states from the latest checkpoint. However, this method involves a remote data store for state preservation and access, resulting in significant overheads to the performance of error-free execution.In this paper, we propose E-Storm, a replication-based state management system that actively maintains multiple state backups on different worker nodes. We build a prototype on top of Storm by extending it with monitoring and recovery modules to support inter-task state transfer whenever needed. The experiments carried out on synthetic and real-world streaming applications confirm that E-Storm outperforms the existing checkpointing method in terms of the resulting application performance, obtaining as much as 9.44 times throughput improvement while reducing the application latency down to 9.8%.\",\n \"Title: Enhancing the quality of geometries of interest (GOIs) extracted from GPS trajectory data using spatio-temporal data aggregation and outlier detection.\\nYear: 2018\\n\\nAbstract: One of the initial phases in the applications dealing with data processing on GPS trajectory data is to generate the time-stamped Sequence of Visited Locations (SVLs) of the mobile objects. The sequence is constructed by labeling each of the GPS observations of the trajectory using the ID of their intersecting Geometries of Interest (GOIs). In this paper, we enhance the performance of the state-of-the-art scheme for constructing the GOIs of a mobile object by proposing a data aggregation and outlier detection method. Our experimental results using geometric similarity metrics show that our improved GOI construction method outperforms the baseline methods by constructing the GOIs remarkably more geometrically similar to the real world GOIs. The geometric similarity metrics are only applicable when we have access to the geometries of the real world GOIs (ground truth). To be able to analyse the performance of the GOI extraction methods in environments which we do not have access to the ground truth, we propose two useful spatio-temporal metrics to measure the quality of GOIs based on the quality of the generated SVLs based on them. Our experimental results show that these two metrics are able to discriminate between the results of our different outlier detection methods and select the best scheme without using any external knowledge about the geometries of the real world GOIs.\",\n \"Title: Geo-aware erasure coding for high-performance erasure-coded storage clusters.\\nYear: 2018\\n\\nAbstract: Erasure code-based distributed storage systems are increasingly being used by storage providers for big data storage since they offer the same reliability as replication with a significant decrease in the amount of storage required. But, when it comes to a storage system with data nodes spread across a very large geographical area, the node’s recovery performance is affected by various factors that are both network and computation related. In this paper, we present a XOR-based code supplemented with the ideas of parity duplication and rack awareness that could be adopted in such storage clusters to improve the recovery performance during node failures and compare it with popular implementations of erasure codes, namely Facebook’s Reed-Solomon codes and XORBAS local recovery codes. The code performance along with the proposed ideas are evaluated on a geo-diverse cluster deployed on the NeCTAR research cloud. We also present a scheme for intelligently placing blocks of coded storage depending on the design of the code, inspired by local reconstruction codes. The sum of all these propositions could offer a better solution for applications that are deployed on coded storage systems that are geographically distributed, in which storage constraints make triple replication not affordable, at the same time ensuring minimal recovery time is a strict requirement.\",\n \"Title: Handling Churn in Similarity Based Clustering Overlays Using Weighted Benefit\\nYear: 2017\\n\\nAbstract: Similarity based clustering (SBC) overlays are decentralized networks of nodes on the Internet edge, where each node maintains some number of direct connections to other nodes that are most \\\"similar\\\" to it. The challenge is: how do the nodes in the overlay converge to and maintain the most similar neighbors, given that the network is decentralized, is subject to churn and that similarity varies over time. Protocols that simultaneously provide fast convergence and low bandwidth consumption are the objective of this research. We present a protocol, that we call Weighted Benefit Scheme (WBS), that improves upon existing state-of-the-art in this area: it has equivalent convergence rate to the Optimum Benefit Protocol while simultaneously handling churn competitively to the Vicinity protocol. We use real world datasets from Yahoo WebScope that comprises of 15,400 users with 354,000 ratings about 1000 songs and our experiments are performed on the simulation test-bed PeerNet.\",\n \"Title: A Data Stream Processing Optimisation Framework for Edge Computing Applications\\nYear: 2018\\n\\nAbstract: Data Stream Processing (DSP) is a widely used programming paradigm to process an unbounded event stream. Often, DSP frameworks are deployed on the cloud with a scalable resource model. One of the key requirements of DSP is to produce results with low latency. With the emergence of IoT, many event sources have been located outside the cloud which can result in higher end-to-end latency due to communication overhead. However, due to the abundance of resources at the IoT layer, Edge computing has emerged as a viable computational paradigm. In this paper, we devise an optimisation framework, consisting of a constraint satisfaction formulation and a system model, that aims to minimise end-to-end latency through appropriate placement of DSP operators either on cloud nodes or edge devices, i.e. deployed in an edge-cloud integrated environment. We test our optimisation framework using OMNeT++, with realistic topologies and power consumption data, and show that it is capable of achieving approx 1.65 times reduction of latency compared to edge-only and cloud-only placements, which in turn also reduces the energy consumption per event by up to approx 4% at the edge layer. To the best of our knowledge our optimisation framework is the first of its kind to integrate power, bandwidth and CPU constraints with latency minimisation.\",\n \"Title: Performance Analysis Of Large-Scale Distributed Stream Processing Systems On The Cloud\\nYear: 2018\\n\\nAbstract: Real-time data processing is often a necessity as it can provide insights that have less value if discovered off-line or after the fact. However, large-scale stream processing systems are non-trivial to build and deploy. While there are many frameworks that allow users to create large-scale distributed systems, there remains many challenges in understanding the performance, cost of deployment and considerations and impact of potential (partial) outages on real-time systems performance. Our work considers the performance of Cloud-based stream processing systems in terms of back-pressure and expected utilization. The performance of an exemplar stream application is explored using different Cloud-based virtual machine resources and where the scale of deployment and cost benefits are taken into consideration in relation to the overall performance.To achieve this, we develop an algorithm based on queueing theory to predict the throughput and latency of stream data processing while supporting system stability. Our methodology for making fundamental measurements is applicable to mainstream stream processing frameworks such as Apache Storm and Heron. The method is especially suitable for large-scale distributed stream processing where jobs can run for extended time periods. We benchmark the performance of the system on the national research cloud of Australia (Nectar), and present a performance analysis based on estimating the overall effective utilization.\",\n \"Title: Stepping stone graph for public movement analysis.\\nYear: 2018\\n\\nAbstract: There are many real world applications that require to identify movement of users such as identifying movement corridors, most popular paths, and nearest neighbours. If one is not given trajectories mapping to movement of people but rather sporadic location data, such as location based social network data, finding movement related information becomes difficult. Rather than processing all points in a data set given a query, a clever approach is to construct a graph, based on user locations, and query this graph for all queries. One example is the shortest path graph. However the shortest path graph can be inefficient and ineffective analysing movement, as it calculates the graph considering all points in a data set. We propose the stepping stone graph, which calculates graph considering point pairs rather than all points, that focuses on local possible movement, making it both efficient and effective for location based social network related queries. We demonstrate its uses by applying it in the aforementioned domain and comparing with the shortest path graph. We also compare its properties to a range of other graphs.\",\n \"Title: Communication Model for Parallel Iterative Stream Processing\\nYear: 2018\\n\\nAbstract: Stream processing systems are used for a plethora of low-latency applications that deal with high volumes and varieties of data. In general, streaming applications are formulated as a fixed number of inter-connected operators, where the operator graph indicates the sequence of computations that apply to the streaming data in motion. Typically an iterative operation in a streaming application is done by embedding the iteration in a single operator (which excludes multiple operator instances and hence not scalable) or by using a sequence of operators, one for each iteration (which requires knowledge of the number of iterations a priori). However, the number of iterations of some applications such as k-means clustering cannot be determined at the creation of the application as it depends on the properties of the data received at runtime and the convergence criterion. Hence only a limited number of iterative computations can be executed using this approach. In this paper, we propose a communication model to support intra-operator communication so that iterative computations, including those with an arbitrary number of iterations, can be efficiently executed in streaming applications. We show that the proposed model can support different iterative algorithms that have complex communication patterns. Finally, through evaluating a number of parallel iterative algorithms using large-scale datasets, we demonstrate the scalability and performance of our proposed communication model and compare it to the existing approaches used for constructing iterative streaming applications.\",\n \"Title: Geotagging tweets to landmarks using convolutional neural networks with text and posting time\\nYear: 2019\\n\\nAbstract: Geotagged tweets serve many important applications, e.g., crisis management, but only a small proportion of tweets are explicitly geotagged. We propose a Convolutional Neural Network (CNN) architecture for geotagging tweets to landmarks, based on the text in tweets and other meta information, such as posting time and source. Using a dataset of Melbourne tweets, experimental results show that our algorithm out-performed various state-of-the-art baselines.\",\n \"Title: Power-law Verification for Event Detection at Multi-spatial Scales from Geo-tagged Tweet Streams.\\nYear: 2019\\n\\nAbstract: Compared with traditional news media, social media nowadays provides a richer and more timely source of news. We are interested in multi-spatial level event detection from geo-tagged tweet streams. Specifically, in this paper we (1) examine the statistical characteristic for the time series of the number of geo-tagged tweets posted from specific regions during a short time interval, e.g., ten seconds or one minute; (2) verify from over thirty datasets that while almost all such time series exhibit self-similarity, those that correspond to events, especially short-term and unplanned outbursts, follow a power-law distribution; (3) demonstrate that these findings can be applied to facilitate event detection from tweet streams---we propose a simple algorithm that only checks the existence of power-law distributions in the time series from tweet streams at multi-spatial scales, without looking into the content of each tweet. Our experiments on multiple datasets show that by considering spatio-temporal statistical distributions of tweets alone, this seemingly naive algorithm achieves comparable results with event detection methods that perform semantic analysis. We further discuss how to integrate the proposed technique into existing algorithms for better performance.\",\n \"Title: Spatio-Temporal Event Detection Using Poisson Model And Quad-Tree On Geotagged Social Media\\nYear: 2019\\n\\nAbstract: Identifying events happening in a specific locality is important as an early warning for accidents, protests, elections or breaking news. However, this location-specific event detection is challenging as the locations and types of events are not known beforehand. To address this problem, we propose an online spatiotemporal event detection system using social media that is able to detect events at different time and space resolutions. First, we exploit a quad-tree method to split the geographical space into multiscale regions based on the density of social media data. Then, we implement a statistical unsupervised approach using Poisson distribution and a smoothing method for highlighting regions with unexpected density of social posts. Further, event duration is estimated by merging events happening in the same region at consecutive time intervals. A post processing stage is introduced to filter out events that are spam, fake or wrong. Finally, we incorporate simple semantics by using social media entities to assess the integrity, and accuracy of detected events. The proposed method is evaluated using Twitter and Flickr for the city of Melbourne based on recall and precision measures. We also propose a new quality measure named strength index, which automatically measures how accurate the reported event is.\",\n \"Title: Multi-Spatial Scale Event Detection From Geo-Tagged Tweet Streams Via Power-Law Verification\\nYear: 2019\\n\\nAbstract: Compared with traditional news media, social media nowadays provides a richer and more timely source of news. We are interested in multi-spatial level event detection from geo-tagged tweet streams. Specifically, in this paper we (1) examine the statistical characteristic for the time series of the number of geo-tagged tweets posted from specific regions during a short time interval, e.g., one minute; (2) verify from over thirty datasets that while almost all such time series exhibit self-similarity, those that correspond to events, especially short-term and unplanned outbursts, follow a power-law distribution; (3) demonstrate that these findings can be applied to facilitate event detection from tweet streams. We propose two algorithms-Power-law basic and Power-law advanced, where Power-law basic only checks the existence of power-law distributions in the time series from tweet streams at multi-spatial scales, without looking into the content of each tweet, and Power-law advanced integrates power-law verification with semantic analysis via word embedding. Our experiments on multiple datasets show that when combined with a Quad-tree, the seemingly naive algorithm of Power-law basic achieves comparable results with more advanced event detection methods, while the semantic analysis enhanced version, Power-law advanced, can significantly increase both the precision and the recall.\",\n \"Title: Sentiment-Aware And Personalized Tour Recommendation\\nYear: 2019\\n\\nAbstract: Itinerary planning is one of the most important tasks in tourism. A well-planned itinerary enhances the tourist experience and their visit satisfaction in new cities. However, the task of planning personalized tour itineraries is complicated by tourists with different interest preferences. Furthermore, there is an added complexity of recommending an itinerary with discrete budget, time and cost. Due to an increase in web-technologies and online geo-location services, there is emerging research targeting itinerary recommendation based on each tourist's interest, preferences and trip constraints. While several research works consider tourist interest, they adopt a simple measure based on the number of times a tourist has visited a place or the number of photos taken by the tourist at a place. Our research proposes an improved sentiment-aware personalized tour planner that considers each tourist's interests based on his/her sentiments on specific categories relative to his/her overall preferences. Unlike the previous approaches that do not consider the actual opinion based preferences, our proposed approach determines user interests based on their sentiments associated with their written text about a place of their visit. This interest measure is based on the intuition that users are more likely to post favorable comments about places they like. Using a dataset from Twitter, we compare our proposed algorithm against the baseline and experimental results show that our algorithm obtained superior performance in terms of tour precision, recall, Fl-score and overall popularity.\",\n \"Title: User Identification across Social Networking Sites using User Profiles and Posting Patterns\\nYear: 2021\\n\\nAbstract: With the prevalence of online social networking sites (OSNs) and mobile devices, people are increasingly reliant on a variety of OSNs for keeping in touch with family and friends, and using it as a source of information. For example, a user might utilise multiple OSNs for different purposes, such as using Flickr to share holiday pictures with family and friends, and Twitter to post short messages about their thoughts. Identifying the same user across multiple OSNs is an important task as this allows us to understand the usage patterns of users among different OSNs, make recommendations when a user registers for a new OSN, and various other useful applications. To address this problem, we proposed an algorithm based on the multilayer perceptron using various types of features, namely: (i) user profile, such as name, location, description; (ii) temporal distribution of user generated content; and (iii) embedding based on user name, real name and description. Using a Twitter and Flickr dataset of users and their posting activities, we perform an empirical study on how these features affect the performance of user identification across the two OSNs and discuss our main findings based on the different features.\",\n \"Title: Spatial-Based Topic Modelling Using Wikidata Knowledge Base\\nYear: 2017\\n\\nAbstract: Topic modelling is a well-studied field that aims to identify topics from traditional documents such as news articles and reports. More recently, Latent Dirichlet Allocation (LDA) and its variants, have been applied on social media platforms to model and study topics relating to sports, politics and companies. While these applications were able to successfully identify the general topics, we posit that standard LDA can be augmented with spatial and temporal considerations based on the geo-coordinates and timestamps of social media posts. Towards this effort, we propose a spatial and temporal variant of LDA to better detect more specific topics, such as a particular art exhibit held at a museum or a security incident happening on a particular day. We validate our approach on a Twitter dataset and find that the detected topics are well-aligned to real-life events happening on the specific days and locations.\",\n \"Title: Multi-Step Prediction With Missing Smart Sensor Data Using Multi-Task Gaussian Processes\\nYear: 2017\\n\\nAbstract: With the proliferation of sensors and the increased connectivity of citizens, many global cities are increasingly adopting Smart City initiatives. Such initiatives provide real-time monitoring capabilities, and effective modelling techniques allow the prediction of future states in a city. For example, urban electricity smart meter data can be utilised to predict future demand in order to facilitate capacity planning. However, the accuracy of this foresight is often marred by low quality and missing sensor data in real-world systems. In this work, we focus on the problem of reliable forecasting by mitigating the effect of missing data on forecast accuracy. In order to mitigate the effects of missing data, we develop a multi-task learning scheme to jointly learn Gaussian Process Regression models between highly correlated sensors. We demonstrate that our methods are robust in a variety of error generation scenarios. We validate our methods based on publicly available and real-world datasets related to electricity smart meters in a university campus and pedestrian counts in a global city, where we achieve significant improvement over competitive baselines and other effective forecasting methods.\",\n \"Title: Clustop: A Clustering-Based Topic Modelling Algorithm For Twitter Using Word Networks\\nYear: 2017\\n\\nAbstract: Twitter is a popular microblogging service, where users frequently engage in discussions about various topics of interest, ranging from popular topics (e.g., music) to niche topics (e.g., politics). With the large amount of tweets, a key challenge is to automatically model and determine the discussion topics without having prior knowledge of the types and number of topics, or requiring the technical expertise to define various algorithmic parameters. For this purpose, we propose the Clustering-based Topic Modelling (ClusTop) algorithm that constructs various types of word network and automatically determines the discussion topics using community detection approaches. Unlike traditional topic models, ClusTop is able to automatically determine the appropriate number of topics and does not require numerous parameters to be set. The ClusTop algorithm is also able to capture the syntactic meaning in tweets via the use of bigrams, trigrams and other word combinations in constructing the word network graph. Using three Twitter datasets with labelled crises and events as topics, ClusTop has been shown to outperform various baselines in terms of topic coherence, pointwise mutual information, precision, recall and F-score.\",\n \"Title: Optimizing checkpoint-based fault-tolerance in distributed stream processing systems: Theory to practice\\nYear: 2022\\n\\nAbstract: Fault-tolerance is an essential part of a stream processing system that guarantees data analysis could continue even after failures. State-of-the-art distributed stream processing systems use checkpointing to support fault-tolerance for stateful computations where the state of the computations is periodically persisted. However, the frequency of performing checkpoints impacts the performance (utilization, latency, and throughput) of the system as the checkpointing process consumes resources and time that can be used for actual computations. In practice, systems are often configured to perform checkpoints based on crude values ignoring factors such as checkpoint and restart costs, leading to suboptimal performance. In our previous work, we proposed a theoretical optimal checkpoint interval that maximizes the system utilization for stream processing systems to minimize the impact of checkpointing on system performance. In this article, we investigate the practical benefits of our proposed theoretical optimal by conducting experiments in a real-world cloud setting using different streaming applications; we use Apache Flink, a well-known stream processing system for our experiments. The experiment results demonstrate that an optimal interval can achieve better utilization, confirming the practicality of the theoretical model when applied to real-world applications. We observed utilization improvements from 10% to 200% for a range of failure rates from 0.3 failures per hour to 0.075 failures per minute. Moreover, we explore how performance measures: latency and throughput are affected by the optimal interval. Our observations demonstrate that significant improvements can be achieved using the optimal interval for both latency and throughput.\",\n \"Title: A Systematic Review of Density Grid-Based Clustering for Data Streams\\nYear: 2022\\n\\nAbstract: Various applications, such as electronic business, satellite remote sensing, intrusion discovery, and network traffic monitoring, generate large unbounded data stream sequences at a rapid pace. The clustering of data streams has attracted considerable interest due to the increasing usage of evolving data streams. In particular, evolving data streams affect clustering because they introduce numerous challenges, such as time and memory limits and one-pass clustering. Furthermore, researchers need to be able to determine arbitrarily shaped clusters present in evolving data streams from applications. Due to these characteristics, conventional density grid-based clustering techniques cannot be used. Moreover, the existing density grid-based clustering algorithms have low cluster quality for clustering evolving data streams. This study conducted a systematic literature review (SLR) and noted numerous research-related issues encountered in solving the aforementioned problems. We summarized numerous grid-based clustering algorithms that have been used and determined their distinctive and limited features. We also observed how these algorithms address the challenges affecting the clustering of evolving data streams and studied their advantages and disadvantages. SLR was based on 104 articles published between 2010 and 2021. Numerous challenges remain for grid-based clustering algorithms, particularly in terms of time-limited and high-dimensional data handling. Last, our findings indicated a variety of active studies on density grid-based clustering.\",\n \"Title: A clustering-based topic model using word networks and word embeddings\\nYear: 2022\\n\\nAbstract: Online social networking services like Twitter are frequently used for discussions on numerous topics of interest, which range from mainstream and popular topics (e.g., music and movies) to niche and specialized topics (e.g., politics). Due to the popularity of such services, it is a challenging task to automatically model and determine the numerous discussion topics given the large amount of tweets. Adding on this complexity is the need to identify these topics with the absence of prior knowledge about both the types and number of topics, while having the requirement of the relevant technical expertise to tune the numerous parameters for the various models. To address this challenge, we develop the Clustering-based Topic Modelling (ClusTop) algorithm that first constructs different types of word networks based on different types of n-grams co-occurrence and word embedding distances. Using these word networks, ClusTop is then able to automatically determine the discussion topics using community detection approaches. In contrast to traditional topic models, ClusTop does not require the tuning or setting of numerous parameters and instead uses community detection approaches to automatically determine the appropriate number of topics. The ClusTop algorithm is also able to capture the syntactic meaning in tweets via the use of bigrams, trigrams, other word combinations and word embedding techniques in constructing the word network graph, and utilizes edge weights based on word embedding. Using three Twitter datasets with labelled crises and events as topics, we show that ClusTop outperforms various traditional baselines in terms of topic coherence, pointwise mutual information, precision, recall and F-score.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":89,"string":"89"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13557,"string":"13,557"}}},{"rowIdx":13558,"cells":{"question_id":{"kind":"string","value":"Er Wei Bai"},"question":{"kind":"string","value":"Generate an abstract for the title \"Variable selection via RIVAL (removing irrelevant variables amidst Lasso iterations) and its application to nuclear material detection\" using the following items: 1. RIVAL \n2. Model Selection \n3. Non-negative parameters \n4. Positive Lasso \n5. Convergence"},"answer":{"kind":"string","value":"In many situations, the number of data points is fixed, and the asymptotic convergence results of popular model selection tools may not be useful. A new algorithm for model selection, RIVAL (removing irrelevant variables amidst Lasso iterations), is presented and shown to be particularly effective for a large but fixed number of data points. The algorithm is motivated by an application of nuclear material detection where all unknown parameters are to be non-negative. Thus, positive Lasso and its variants are analyzed. Then, RIVAL is proposed and is shown to have some desirable properties, namely the number of data points needed to have convergence is smaller than existing methods."},"correct_answer":{"kind":"string","value":"In many situations, the number of data points is fixed, and the asymptotic convergence results of popular model selection tools may not be useful. A new algorithm for model selection, RIVAL (removing irrelevant variables amidst Lasso iterations), is presented and shown to be particularly effective for a large but fixed number of data points. The algorithm is motivated by an application of nuclear material detection where all unknown parameters are to be non-negative. Thus, positive Lasso and its variants are analyzed. Then, RIVAL is proposed and is shown to have some desirable properties, namely the number of data points needed to have convergence is smaller than existing methods."},"documents":{"kind":"list like","value":["Title: Brief A random least-trimmed-squares identification algorithm\nYear: 2003\n\nAbstract: The Least-trimmed-squares (LTS) estimator is a well known robust estimator in terms of protecting the estimate from the outliers. Its high computational complexity is however a problem in practice. In this paper, we propose a random LTS algorithm which has a low computational complexity that can be calculated a priori as a function of the required error bound and the confidence interval. Moreover, if the number of data points goes to infinite, the algorithm becomes a deterministic one that converges to the true LTS in some probability sense.","Title: Brief Variable gain parameter estimation algorithms for fast tracking and smooth steady state\nYear: 2000\n\nAbstract: Based on the ideas of the set-membership identification, we present in this paper variable gain least mean-square (LMS) and weighted recursive least-squares (WRLS) algorithms. It is shown that the proposed algorithms possess a fast tracking ability to the variations of the parameter and at the same time are less sensitive to noise leading to a smooth steady state.","Title: Constrained logarithmic least squares in parameter estimation.\nYear: 1999\n\nAbstract: The contribution of this paper is twofolds. First, it is shown that while robust in termsof the average output error, the least squares estimate is sensitive to outliers with respectto the maximum output error. In fact the worst-case output error of the least squares cango unbounded. Then, a constrained logarithmic least squares for system identification isproposed. Analytic center algorithms are presented to solve this constrained logarithmicleast squares problem.1 IntroductionThe...","Title: Decoupling the linear and nonlinear parts in Hammerstein model identification\nYear: 2004\n\nAbstract: The main result of this paper is to show that the linear part can be made decoupled from the nonlinear part in Hammerstein model identification. Therefore, identification of the linear part for a Hammerstein model becomes a linear problem and accordingly enjoys the same convergence and consistency results as if the unknown nonlinearity is absent.","Title: Least squares solutions of bilinear equations\nYear: 2006\n\nAbstract: The problem of finding a least squares solution for a system of bilinear equations is investigated. Sufficient conditions to have a unique minimum are given in the case of random inputs. Three methods, the normalized iterative method, the over-parameterization method and the numerical method are presented for solving the problem along with their convergence properties. Simulation examples are provided.","Title: PageRank computation via a distributed randomized approach with lossy communication.\nYear: 2012\n\nAbstract: For ranking web pages in search results, Google employs the so-called PageRank algorithm, which provides a measure of importance to each page based on the web structure. Recently, we have developed a distributed randomized approach for this algorithm, where pages compute their own PageRank by communicating over selected links. In this paper, the focus is on the effects of unreliability in communication channels. Specifically, we consider random data losses modeled as a Markov chain and introduce a generalized version of the distributed scheme. Its convergence properties and the error in the approximated PageRank are analyzed.","Title: Optimization with few violated constraints for linear bounded error parameter estimation\nYear: 2002\n\nAbstract: In the context of linear constrained optimization, we study in this paper the problem of finding an optimal solution sat- isfying all but of the given constraints. A solution is obtained by means of an algorithm of the complexity , where is the dimension of the problem. We then use these results to solve the problem of robust identification in the presence of outliers in the setting of bounded error parameter iden- tification. Finally, we show that the estimate obtained converges to the true but unknown parameter in the presence of outliers. Index Terms—Constrainted optimization, parameter estima- tion, system identification, unknown but bounded error.","Title: Distributed randomized PageRank computation based on web aggregation\nYear: 2009\n\nAbstract: The so-called PageRank algorithm has been used at Google for properly ranking search results. It quantifies the importance of each page by the structure of links in the web. In our recent work, we have proposed a distributed randomized approach for the PageRank computation, where the pages find their own values by communicating with linked pages. This paper builds upon this approach to improve the computation and communication load for the algorithm by systematically aggregating the webpages while maintaining performance.","Title: A Web Aggregation Approach for Distributed Randomized PageRank Algorithms\nYear: 2012\n\nAbstract: The PageRank algorithm employed at Google assigns a measure of importance to each web page for rankings in search results. In our recent papers, we have proposed a distributed randomized approach for this algorithm, where web pages are treated as agents computing their own PageRank by communicating with linked pages. This paper builds upon this approach to reduce the computation and communication loads for the algorithms. In particular, we develop a method to systematically aggregate the web pages into groups by exploiting the sparsity inherent in the web. For each group, an aggregated PageRank value is computed, which can then be distributed among the group members. We provide a distributed update scheme for the aggregated PageRank along with an analysis on its convergence properties. The method is especially motivated by results on singular perturbation techniques for large-scale Markov chains and multi-agent consensus. A numerical example is provided to illustrate the level of reduction in computation while keeping the error in rankings small.","Title: A Recursive Local Linear Estimator for Identification of Nonlinear ARX Systems: Asymptotical Convergence and Applications\nYear: 2013\n\nAbstract: In this paper, we propose a recursive local linear estimator (RLLE) for nonparametric identification of nonlinear autoregressive systems with exogenous inputs (NARX). First, the RLLE is introduced. Next, the strong consistency as well as the asymptotical mean square error properties of the RLLE are established, and then an application of the RLLE to an additive nonlinear system is discussed. The RLLE provides recursive estimates not only for the function values but also their gradients at fixed points. A simulation example is provided to confirm the theoretical analysis.","Title: An optimal two stage identification algorithm for Hammerstein-Wiener nonlinear systems\nYear: 1998\n\nAbstract: An optimal two stage identification algorithm is presented for Hammerstein-Wiener systems where two static nonlinear elements surround a linear block. The proposed algorithm consists of two steps: The first one is the recursive least squares and the second one is the singular value decomposition of two matrices whose dimensions are fixed and do not increase as the number of the data point increases. Moreover, the algorithm is shown to be convergent in the absence of noise and convergent with probability one in the presence of white noise","Title: Blind decision feedback equalization of time-varying channels with DPSK inputs\nYear: 2001\n\nAbstract: This paper presents a new method to the blind equalization of a class of linear time-varying mobile channels for differentially encoded channel input signals. Through re-parameterization, the discrete system can be modeled equivalently as a single input multiple output system. Unlike some existing methods, this new algorithm does not require prior knowledge on the cyclic frequency of the channel response. The algorithm is simple and relies only on a decision feedback structure and the differential encoding nature of the channel input signals. The convergence analysis of the decision feedback system is presented along with simulation examples","Title: A blind approach to the Hammerstein-Wiener model identification\nYear: 2002\n\nAbstract: The work of this chapter is a continuation of [3]. Unlike in [3] where a very special structure is assumed, however, the blind\n approach in this chapter allows a very general structure on the nonlinearities. In particular, the input nonlinearity structure\n can be arbitrary and is not assumed to be known. By using the blind identification approach, all the unknown internal variables\n can be recovered solely based on the output measurements.","Title: Worst-case properties of the uniform distribution and randomized algorithms for robustness analysis\nYear: 1998\n\nAbstract: In this paper we study a probabilistic approach which is an alternative to the classical worst-case algorithms for robustness analysis and design of uncertain control systems. That is, we aim to estimate the probability that a control system with uncertain parameters q restricted to a box I! attains a given level of performance gamma. Since this probability depends on the underlying distribution, we address the following question: What is a \"reasonable\" distribution so that the estimated probability makes sense? To answer this question, we define two worst-case criteria and prove that the uniform distribution is optimal in both cases. In the second part of the paper we turn our attention to a subsequent problem. That is, we estimate the sizes of both the so-called \"good\" and \"bad\" sets via sampling. Roughly speaking, the good set contains the parameters q is an element of Q with a performance level better than or equal to gamma while the bad set is the set of parameters q is an element of Q with a performance level worse than gamma. We give bounds on the minimum sample size to attain a good estimate of these sets in a certain probabilistic sense.","Title: Adaptive bolus-chasing computed tomography angiography in the cases of symmetric and asymmetric arterial flows in peripheral arteries.\nYear: 2009\n\nAbstract: Synchronization of the contrast bolus peak and CT imaging aperture is a crucial issue for computed tomography angiography (CTA). It affects the CTA image quality and the amount of contrast dose. A whole-body CTA procedure means to scan from the abdominal aorta to pedal arteries. In this context, the synchronization is much more difficult with the asymmetric arterial flow in lower extremities than in the case of symmetric arterial flow. In this paper, we propose an adaptive optimal controller to chase the contrast bolus peak while it propagates in the aorta and lower extremities with symmetric flow. In the case of asymmetric flow after the contrast bolus splitting into two lower limbs, we propose a dynamic programming approach to cover the lower limbs optimally. Simulation and experimental results show that the proposed methods outperform the current constant-speed method substantially.","Title: Two-stage mixed discrete-continuous identification of radial basis function (RBF) neural models for nonlinear systems\nYear: 2009\n\nAbstract: The identification of nonlinear dynamic systems using radial basis function (RBF) neural models is studied in this paper. Given a model selection criterion, the main objective is to effectively and efficiently build a parsimonious compact neural model that generalizes well over unseen data. This is achieved by simultaneous model structure selection and optimization of the parameters over the continuous parameter space. It is a mixed-integer hard problem, and a unified analytic framework is proposed to enable an effective and efficient two-stage mixed discrete-continuous identification procedure. This novel framework combines the advantages of an iterative discrete two-stage subset selection technique for model structure determination and the calculus-based continuous optimization of the model parameters. Computational complexity analysis and simulation studies confirm the efficacy of the proposed algorithm.","Title: Brief Limiting zero distribution of sampled systems\nYear: 2002\n\nAbstract: In this paper, we consider the sampling of a continuous time system with possibly a non-rational transfer function. Limiting zero distribution of the sampled system represented by an FIR model is derived. It is shown that the zeros of the FIR sampled system converge evenly to the unit circle or to the unit disk plus unstable roots of the reciprocal polynomials depending on assumptions.","Title: An Adaptive Optimal Control Design for a Bolus Chasing Computed Tomography Angiography\nYear: 2008\n\nAbstract: To improve imaging quality and to reduce contrast dose and radiation exposure, an optimal adaptive bolus chasing controller is proposed and tested based on actual patient data. The controller estimates and predicts the unknown 2D bolus density online and then determines the optimal control actions. Tracking errors are mathematically quantified in terms of estimation errors. The test results not only support the analytical analysis and exhibit its superior performance over the current constant-speed controller, but also demonstrate the clinical feasibility.","Title: Representation and identification of non-parametric nonlinear systems of short term memory and low degree of interaction\nYear: 2010\n\nAbstract: In this paper, we propose a new representation which is particularly useful for a class of non-parametric nonlinear systems that have short term memory and low degree of interaction. Advantages and disadvantages of this representation are discussed and compared to existing methods both theoretically and numerically. Furthermore, results regarding structural estimation based on the analysis of variance and on full scale identification are also provided.","Title: Frequency domain identification of Wiener models\nYear: 2003\n\nAbstract: This paper proposes a frequency domain algorithm for Wiener model identifications based on exploring the fundamental frequency and harmonics generated by the unknown nonlinearity. The convergence of the algorithm is established in the presence of white noise. No a priori knowledge of the structure of the nonlinearity is required and the linear part can be nonparametric.","Title: Iterative identification of Hammerstein systems\nYear: 2007\n\nAbstract: Convergences of iterative algorithms have been established for identification of Hammerstein systems in the case that the unknown nonlinearities are odd. Then, the results are further extended to nonsmooth nonlinearities.","Title: Blind system identification and channel equalization of IIR systemswithout statistical information\nYear: 1999\n\nAbstract: A novel approach is proposed to blindly identify an unknown IIR system. The approach is based on faster sampling at the system output and requires neither a priori statistical information on the unknown input nor training signals. The methods presented are linear in the parameters of the unknown system so that many standard recursive algorithms can be readily applied. It is also shown in the paper that under a generic condition, any finite-order IIR system is identifiable, provided the oversampling ratio is appropriately chosen","Title: Adaptive bolus chasing computed tomography angiography: Control scheme and experimental results.\nYear: 2008\n\nAbstract: In this paper, a new adaptive bolus-chasing control scheme is proposed to synchronize the bolus peak in a patient's vascular system and the imaging aperture of a computed tomography (CT) scanner. The proposed control scheme is theoretically evaluated and experimentally tested on a modified Siemens SOMATOM Volume Zoom CT scanner. The first set of experimental results is reported on bolus-chasing CT angiography using realistic bolus dynamics, real-time CT imaging and adaptive table control with physical vasculature phantoms. The data demonstrate that the proposed control approach tracks the bolus propagation well, and clearly outperforms the constant-speed scheme that is the current clinical standard.","Title: Blind identifiability of IIR systems\nYear: 2002\n\nAbstract: It is shown that, with no prior information, a necessary and sufficient condition for an nth order IIR system to be blindly identifiable is that the over-sampling ratio p⩾n+1.","Title: Capacitors and the method of moments\nYear: 2004\n\nAbstract: The inhomogeneous charge distribution in several plate capacitors is calculated using the Method of Moments. Effects of global and local “inhomogeneities” in the capacitor are simulated.","Title: Toeplitz matrix based sparse error correction in system identification: Outliers and random noises\nYear: 2013\n\nAbstract: In this paper, we consider robust system identification under sparse outliers and random noises. In our problem, system parameters are observed through a Toeplitz matrix. All observations are subject to random noises and a few are corrupted with outliers. We reduce this problem of system identification to a sparse error correcting problem using a Toeplitz structured real-numbered codingmatrix. We prove the performance guarantee of Toeplitz structured matrix in sparse error correction. Thresholds on the percentage of correctable errors for Toeplitz structured matrices are also established. When both outliers and observation noise are present, we have shown that the estimation error goes to 0 asymptotically as long as the probability density function for observation noise is not “vanishing” around 0.","Title: Zeros of sampled data systems represented by FIR models\nYear: 2000\n\nAbstract: In this note, we investigate zero locations of FIR linear systems that are finite length approximations of sampled continuous-time systems. For linear systems with rational transfer functions, it is shown that with a relative degree higher than two and a short sampling interval, the resultant FIR sampled data system is always a nonminimum phase.","Title: Towards identification of Wiener systems with the least amount of a priori information on the nonlinearity\nYear: 2008\n\nAbstract: In this paper, we investigate what constitutes the least amount of a priori information on the nonlinearity so that the FIR linear part is identifiable in the non-Gaussian input case. Three types of a priori information are considered including quadrant information, point information and locally monotonous information. In all three cases, identifiability has been established and corresponding identification algorithms are developed with their convergence proofs.","Title: Convergence of the iterative algorithm for a general Hammerstein system identification\nYear: 2010\n\nAbstract: The convergence of the iterative identification algorithm for a general Hammerstein system has been an open problem for a long time. In this paper, it is shown that the convergence can be achieved by incorporating a regularization procedure on the nonlinearity in addition to a normalization step on the parameters.","Title: Fatigue and non-fatigue mathematical muscle models during functional electrical stimulation of paralyzed muscle.\nYear: 2010\n\nAbstract: Electrical muscle stimulation demonstrates potential for preventing muscle atrophy and restoring functional movement after spinal cord injury (SCI). Control systems used to optimize delivery of electrical stimulation protocols depend upon the algorithms generated using computational models of paralyzed muscle force output. The Hill–Huxley-type model, while being highly accurate, is also very complex, making it difficult for real-time implementation. In this paper, we propose a Wiener–Hammerstein system to model the paralyzed skeletal muscle under electrical stimulus conditions. The proposed model has substantial advantages in identification algorithm analysis and implementation including computational complexity and convergence, which enable it to be used in real-time model implementation. Experimental data sets from the soleus muscles of 14 subjects with SCI were collected and tested. The simulation results show that the proposed model outperforms the Hill–Huxley-type model not only in peak force prediction, but also in fitting performance for force output of each individual stimulation train.","Title: Convergence results of the analytic center estimator\nYear: 2000\n\nAbstract: The analytic center approach for bounded error parameter estimation was recentlyproposed as an alternative to the well-known Chybyshev and least squares estimates.In this paper, we show the asymptotic performance of this approach and prove that theanalytic center converges to the true parameter under mild conditions.This work was supported in part by grants of NSF(USA)1;4, ARC(Australia)2and CENS-CNR(Italy)1;3.The first author especially thanks the support of CNR short-term...","Title: A comparative study on interpolation methods for controlled cardiac CT\nYear: 2007\n\nAbstract: Bai et al. recently proposed to acquire random fan-beam-cone-beam projections with a linear-planar detector from a circular scanning locus for controlled cardiac computed tomography (CT). After specifying a uniform acquisition geometry required by FBP (filtration-backprojection), we rebin the random fan-beam-cone-beam data via nearest-neighbor, quadrilateral and triangle-based linear interpolation methods. The fan-beam and parallel-beam FBP algorithms are employed for rebinned fan-beam projections. The FDK (Feldkamp–Davis–Kress) and t-FDK (tent-FDK) methods are employed for rebinned cone-beam data. Also, nonuniform weighting fan-beam-FDK methods are used to reconstruct without rebinning. As a benchmark, the images are reconstructed using uniform weighting fan-beam-FDK method from data collected at the specified uniform grid. To evaluate the different methods, effects of increasing the number of projections and adding Poisson noise are studied. The root mean square error (RMSE) is used to quantify the image quality by numerical tests with the cardiac phantom. Our results show that it is helpful to perform data interpolation for improvements of the image quality in controlled cardiac CT from random projections. Our simulations indicate that triangular interpolation gives the most satisfactory result for improved image quality whereas quadrilateral interpolationgives the best noise performance. © 2007 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 17, 91–98, 2007","Title: Towards Identification of Wiener Systems with the Least Amount of a priori Information: IIR Cases.\nYear: 2009\n\nAbstract: In this paper, we investigate what constitutes the least amount of a priori information on the nonlinearity so that the linear part is identifiable in the non-Gaussian input case. Under the white noise input, three types of a priori information are considered including quadrant information, point information and monotonic information. In all three cases, identifiability has been established and the corresponding nonparametric identification algorithms are developed along with their convergence proofs.","Title: On the capacitance of a cube\nYear: 2002\n\nAbstract: The capacitance of a metallic cube is analyzed using the method of subareas and modern computing tools. This calculation extends the work of Reitan and Higgins in order to obtain a more accurate value for the capacitance.","Title: Analytic center approach to parameter estimation: convergence analysis\nYear: 1998\n\nAbstract: The so-called analytic center approach to parameter estimation has been proposed as an alternative to the well-known least squares approach. This new approach offers a parameter estimate that is consistent with the past data observations, has a simple geometric interpretation, and is computable sequentially. We study the asymptotic performance of the analytic center approach and show that the resulting estimate converges to the true parameter asymptotically, provided some mild conditions are satisfied. These conditions involve some weak persistent excitation and independence between noise and regressor, similar to the least squares case. This result is used to derive a new parameter estimation approach which offers both good transient and asymptotic performance","Title: A New Approach For Aggregated Pagerank Computation Via Distributed Randomized Algorithms\nYear: 2011\n\nAbstract: At Google, the PageRank algorithm helps rankings in search results by providing measures of web page importance. This paper builds upon the distributed randomized approach for this algorithm proposed in our recent works. To reduce computation and communication, we develop a method to systematically aggregate web pages into groups by exploiting the sparsity inherent in the web. Each group computes an aggregated PageRank, which can be distributed among group members. We provide a decentralized scheme for its computation and analyze convergence properties.","Title: Identification of non-parametric FIR non-linear systems with low-degree interactive terms\nYear: 2009\n\nAbstract: In this paper, an interactive term identification approach is proposed for identification of non-parametric finite impulse response non-linear systems under i.i.d. random sequences. The idea is to make a high-dimensional non-linear identification problem into a number of low-dimensional problems and thus to effectively combat the problem of the curse of dimensionality. Convergence results are esta...","Title: A two-stage algorithm for identification of nonlinear dynamic systems\nYear: 2006\n\nAbstract: This paper investigates the two-stage stepwise identification for a class of nonlinear dynamic systems that can be described by linear-in-the-parameters models, and the model has to be built from a very large pool of basis functions or model terms. The main objective is to improve the compactness of the model that is obtained by the forward stepwise methods, while retaining the computational efficiency. The proposed algorithm first generates an initial model using a forward stepwise procedure. The significance of each selected term is then reviewed at the second stage and all insignificant ones are replaced, resulting in an optimised compact model with significantly improved performance. The main contribution of this paper is that these two stages are performed within a well-defined regression context, leading to significantly reduced computational complexity. The efficiency of the algorithm is confirmed by the computational complexity analysis, and its effectiveness is demonstrated by the simulation results.","Title: Brief Identification of linear systems with hard input nonlinearities of known structure\nYear: 2002\n\nAbstract: This paper studies identification of systems with input nonlinearities of known structure. For input nonlinearities parameterized by one parameter, a deterministic approach is proposed based on the idea of separable least squares. The identification problem is shown to be equivalent to an one-dimensional minimization problem. The method is very effective for several common static and nonstatic input nonlinearities. For a general input nonlinearity, a correlation analysis based identification algorithm is presented which is shown to be convergent.","Title: Non-Parametric Nonlinear System Identification: A Data-Driven Orthogonal Basis Function Approach\nYear: 2008\n\nAbstract: In this paper, a data driven orthogonal basis function approach is proposed for non-parametric FIR nonlinear system identification. The basis functions are not fixed a priori and match the structure of the unknown system automatically. This eliminates the problem of blindly choosing the basis functions without a priori structural information. Further, based on the proposed basis functions, approaches are proposed for model order determination and regressor selection along with their theoretical justifications.","Title: Making parametric Hammerstein system identification a linear problem\nYear: 2011\n\nAbstract: In this paper, we study the identification of parametric Hammerstein systems with FIR linear parts. By a proper normalization and a clever characterization, it is shown that the average squared error cost function for identification can be expressed in terms of the inner product between the true but unknown parameter vector and its estimate. Further, the cost function is concave in the inner product and linear in the inner product square. Therefore, the identification of parametric Hammerstein systems with FIR linear parts is a globally convergent problem and has one and only one (local and global) minimum. This implies that the identification of such systems is a linear problem in terms of the inner product square and any local search based identification algorithm converges globally.","Title: A New Extension of Newton Algorithm for Nonlinear System Modelling Using RBF Neural Networks\nYear: 2013\n\nAbstract: Model performance and convergence rate are two key measures for assessing the methods used in nonlinear system identification using Radial Basis Function neural networks. A new extension of the Newton algorithm is proposed to further improve these two aspects by extending the results of recently proposed continuous forward algorithm (CFA) and hybrid forward algorithm (HFA). Computational complexity analysis confirms its efficiency, and numerical examples show that it converges faster and potentially outperforms CFA and HFA.","Title: Non-Parametric Nonlinear System Identification: An Asymptotic Minimum Mean Squared Error Estimator\nYear: 2009\n\nAbstract: This paper studies the problem of the minimum mean squared error estimator for non-parametric nonlinear system identification. It is shown that for a wide class of nonlinear systems, the local linear estimator is a linear (in outputs) asymptotic minimum mean squared error estimator. The class of the systems allowed is characterized by a stability condition that is related to many well studied stability notions in the literature. Numerical simulations support the analytical analysis.","Title: Kernel based approaches to local nonlinear non-parametric variable selection\nYear: 2014\n\nAbstract: In this paper, we consider the variable selection problem for a nonlinear non-parametric system. Two approaches are proposed, one top-down approach and one bottom-up approach. The top-down algorithm selects a variable by detecting if the corresponding partial derivative is zero or not at the point of interest. The algorithm is shown to have not only the parameter but also the set convergence. This is critical because the variable selection problem is binary, a variable is either selected or not selected. The bottom-up approach is based on the forward/backward stepwise selection which is designed to work if the data length is limited. Both approaches determine the most important variables locally and allow the unknown non-parametric nonlinear system to have different local dimensions at different points of interest. Further, two potential applications along with numerical simulations are provided to illustrate the usefulness of the proposed algorithms.","Title: Identification of a Modified Wiener-Hammerstein System and Its Application in Electrically Stimulated Paralyzed Skeletal Muscle Modeling.\nYear: 2009\n\nAbstract: Electrical muscle stimulation demonstrates potential for restoring functional movement and preventing muscle atrophy after spinal cord injury (SCI). Control systems used to optimize delivery of electrical stimulation protocols depend upon mathematical models of paralyzed muscle force outputs. While accurate, the Hill–Huxley-type model is very complex, making it difficult to implement for real-time control. As an alternative, we propose a modified Wiener–Hammerstein system to model the paralyzed skeletal muscle dynamics under electrical stimulus conditions. Experimental data from the soleus muscles of individuals with SCI was used to quantify the model performance. It is shown that the proposed Wiener–Hammerstein system is at least comparable to the Hill–Huxley-type model. On the other hand, the proposed system involves a much smaller number of unknown coefficients. This has substantial advantages in identification algorithm analysis and implementation including computational complexity, convergence and also in real-time model implementation for control purposes.","Title: How Nonlinear Parametric Wiener System Identification is Under Gaussian Inputs?\nYear: 2012\n\nAbstract: A number of methods exist for identifying nonlinear Wiener systems. However, there is no attempt to address the fundamental question of how nonlinear these identification problems really are? In this technical note, we try to address this question by investigating the average squared error cost function used in identification. By a proper normalization and a clever characterization of the cost function in terms of the angle between the true but unknown parameter vector and its estimate, it is shown in the technical note that under iid Gaussian inputs for parametric Wiener systems with polynomial nonlinear parts and FIR linear parts, the cost function is globally monotonic and has one and only one (local and global) minimum. The implication is that identification of such systems is nonlinear but very close to linear. Further, any local search based identification algorithms would converge globally for such systems.","Title: Robust system identification with noisy experimental data: projection operator and linear algorithms\nYear: 1994\n\nAbstract: In this paper we consider the problem of robust system identification with noisy time or frequency response measurement data. It is shown here that any linear identification algorithm which is convergent in the noise free case can be made robustly convergent in the presence of noise by incorporating a simple projection operator into the algorithm. The computational simplicity and faster rate of convergence distinguish this approach from other existing robustly convergent nonlinear identification techniques.","Title: Identification of an additive nonlinear system and its applications in generalized Hammerstein models\nYear: 2008\n\nAbstract: An identification algorithm is developed for a class of nonlinear systems that are multi-input and multi-output in an additive form. The convergence results are achieved and its applications to identification of a generalized Hammerstein system is also discussed.","Title: Recursive Direct Weight Optimization in Nonlinear System Identification: A Minimal Probability Approach\nYear: 2007\n\nAbstract: In this paper, a direct weight optimization method is proposed for nonlinear system identification based on a minimal probability idea. The approach has several quite attractive features and is very different from existing ones. It is optimal for any given number of finite data points and at the same time possesses asymptotic convergence. The estimator admits a closed form and no numerical optimization is needed. Theoretical analysis and numerical simulations show that the approach is a very competitive alternative to existing nonlinear identification methods.","Title: Identification of nonlinear additive FIR systems\nYear: 2005\n\nAbstract: In this paper, we consider identification of a nonlinear additive system. An input signal is designed in such a way that the problem of identification of nonlinear additive systems is reduced to a problem of identification of static nonlinear functions. Then, three approaches are established to estimate the order of the system. The methods exploit the structure of the nonlinear additive model so that their implementations are easy.","Title: Blind source separation/channel equalization of nonlinear channels with binary inputs\nYear: 2005\n\nAbstract: In this paper, we consider blind source separation/channel equalization problems for an unknown nonlinear channel. By using the property of the binary alphabets, we show that blind nonlinear source separation/channel equalization problems can be easily solved by existing linear methods.","Title: On the Long-term Health Care Crisis. A Possible Eradication Scenario\nYear: 2011\n\nAbstract: The purpose of the present essay is to suggest a possible model to describe the worldwide healthcare crisis, where diseases that have been considered to be eradicated or under our control are re-emerging today.","Title: A Markov Chain Monte Carlo Approach to Nonlinear Parametric System Identification\nYear: 2015\n\nAbstract: Nonlinear system identification is discussed in a mixed set-membership and statistical setting. A Markov chain Monte Carlo (MCMC) approach is proposed that es- timates the feasible parameter set, the minimum volume outer-bounding ellipsoid and the minimum variance esti- mate. The proposed algorithm is proved to be convergent and enjoys some desirable properties. Further, its compu- tational complexity and numerical accuracy are studied.","Title: Kernel-based local order estimation of nonlinear nonparametric systems.\nYear: 2015\n\nAbstract: We consider the local order estimation of nonlinear autoregressive systems with exogenous inputs (NARX), which may have different local dimensions at different points. By minimizing the kernel-based local information criterion introduced in this paper, the strongly consistent estimates for the local orders of the NARX system at points of interest are obtained. The modification of the criterion and a simple procedure of searching the minimum of the criterion, are also discussed. The theoretical results derived here are tested by simulation examples.","Title: Identifiability and convergence analysis of the MINLIP estimator.\nYear: 2015\n\nAbstract: This paper studies the problem of identifying a monotone Wiener system from observed, noiseless input–output signals. Specifically, we study identifiability of such systems, as well as the convergence properties of the recently introduced MINimal LIPschitz (MINLIP) estimator. This estimator finds a system with minimal complexity by exploiting ordering of the output samples. This makes the approach conceptual quite different from traditional identification schemes based on least squares, prediction errors, maximum likelihood or numerical projections. Sufficient conditions from which the result follows are given in terms of ‘Rotational Complete Inputs (RCI)’, a generalization of the notion of Persistency of Excitation (PE) as used for the study of linear identification methods. Finally, it is derived that a Gaussian i.i.d. sequence satisfies this RCI property, while any binary sequence does not.","Title: System identification in the presence of outliers and random noises: A compressed sensing approach.\nYear: 2014\n\nAbstract: In this paper, we consider robust system identification of FIR systems when both sparse outliers and random noises are present. We reduce this problem of system identification to a sparse error correcting problem using a Toeplitz structured real-numbered coding matrix and prove the performance guarantee. Thresholds on the percentage of correctable errors for Toeplitz structured matrices are established. When both outliers and observation noise are present, we have shown that the estimation error goes to 0 asymptotically as long as the probability density function for observation noise is not “vanishing” around origin. No probabilistic assumptions are imposed on the outliers.","Title: Source location estimation for possibly unknown propagation models\nYear: 2014\n\nAbstract: In this paper the source localization problem is considered without any knowledge of the signal propagation model, beyond the fact that the measured signal, e.g. received signal strength, strictly decreases with distance. Three algorithms dedicated to various scenarios are developed. Their convergence and other properties are established.","Title: The maximum likelihood estimate for radiation source localization: Initializing an iterative search\nYear: 2014\n\nAbstract: The maximum likelihood estimate approach is adopted in this paper for finding the unknown radiation source location and strength. The problem is nonlinear and has to rely on iterative numerical algorithms. Since the problem has possibly multiple local maxima, the initial estimate in those iterative algorithms plays a critical role in guaranteeing the global optimum. This paper proposes a way to generate such an initial estimate which is easy to calculate. Besides some insights that justifies the proposed approach, it is shown that the proposed initial estimate actually converges to the true but unknown maximum likelihood estimate asymptotically thus ensuring that the initial estimate is indeed in a neighborhood of the maximum likelihood estimate and consequently the convergence to the global optimum by local iterative numerical algorithms.","Title: A distributed control law for optimum sensor placement for source localization\nYear: 2014\n\nAbstract: We formulate a nonlinear distributed control law that guides the motion a group of sensors to achieve a configuration that permits them to optimally localize a hazardous source they must keep a prescribed distance from. Earlier work shows that such a configuration involves the sensors being placed in an equispaced manner on a prescribed circle. The nonlinear control law we propose assumes that each sensor resides and moves on the prescribed circle, by accessing only the states of its two immediate clockwise and counterclockwise neighbors. We theoretically prove and verify through simulations, that the law allows the sensors to achieve the desired configuration while avoiding collisions.","Title: Two-Stage Orthogonal Least Squares Methods for Neural Network Construction.\nYear: 2015\n\nAbstract: A number of neural networks can be formulated as the linear-in-the-parameters models. Training such networks can be transformed to a model selection problem where a compact model is selected from all the candidates using subset selection algorithms. Forward selection methods are popular fast subset selection approaches. However, they may only produce suboptimal models and can be trapped into a local minimum. More recently, a two-stage fast recursive algorithm (TSFRA) combining forward selection and backward model refinement has been proposed to improve the compactness and generalization performance of the model. This paper proposes unified two-stage orthogonal least squares methods instead of the fast recursive-based methods. In contrast to the TSFRA, this paper derives a new simplified relationship between the forward and the backward stages to avoid repetitive computations using the inherent orthogonal properties of the least squares methods. Furthermore, a new term exchanging scheme for backward model refinement is introduced to reduce computational demand. Finally, given the error reduction ratio criterion, effective and efficient forward and backward subset selection procedures are proposed. Extensive examples are presented to demonstrate the improved model compactness constructed by the proposed technique in comparison with some popular methods.","Title: On the Gradient Descent Localization of Radioactive Sources\nYear: 2013\n\nAbstract: We consider the robust localization of radioactive sources by using their gamma-ray count at the smallest number of sensors needed to theoretically localize. We formulate a class of non-convex cost functions and consider their gradient descent optimization. We show that in N-dimensions, if there are exactly N + 1 sensors and the source lies in their open convex hull, then this convex hull is devoid of false stationary points. Thus we augment gradient descent with random projections into the convex hull, when an estimate leaves it. We argue that convergence in probability to the correct source location, will occur. Simulations demonstrate the efficacy of this algorithm.","Title: Time series wind power forecasting based on variant Gaussian Process and TLBO\nYear: 2016\n\nAbstract: Due to the variability and stochastic nature of wind power, accurate wind power forecasting plays an important role in developing reliable and economic power system operation and control strategies. As wind variability is stochastic, Gaussian Process regression has recently been introduced to capture the randomness of wind energy. However, the disadvantages of Gaussian Process regression include its computation complexity and incapability to adapt to time varying time-series systems. A variant Gaussian Process for time series forecasting is introduced in this study to address these issues. This new method is shown to be capable of reducing computational complexity and increasing prediction accuracy. The convergence of the forecasting results is also proved. Further, a teaching learning based optimization (TLBO) method is used to train the model and to accelerate the learning rate. The proposed modelling and optimization method is applied to forecast both the wind power generation in Ireland and that from a single wind farm to demonstrate the effectiveness of the proposed method.","Title: Robust tracking of piecewise linear trajectories with binary sensor networks\nYear: 2015\n\nAbstract: We present a novel approach to the problem of tracking objects moving along smooth trajectories using a network of simple inexpensive binary sensors. Specifically, we consider object trajectories that can be well-approximated by piecewise linear curves and sensors that can only detect whether an object is in their sensing range. We start by considering objects moving along straight-line trajectories with an unknown speed and show that such objects can be tracked using the measurements of just three generically placed binary sensors whose sensing ranges intersect the trajectory. Next, we present an asymptotic analysis that shows that a trajectory consisting of a finite number of straight line segments can be recovered with high probability using an arbitrarily low spatial density of sensors in the limit when the area to be covered gets larger and larger. We also present efficient algorithms that effectively recover piecewise linear trajectories. Finally we present analysis and simulations to demonstrate the high tracking accuracy of our approach and its robustness to sensing errors.","Title: Modelling And Parameter Estimation Of A Cell System\nYear: 2009\n\nAbstract: A non-linear identification approach for modelling endothelial cells is mathematically derived and experimentally tested under the realistic alternating current voltage and a square-like shape assumption. Results demonstrate the applicability and effectiveness of the approach. The derived model is an important step in understanding a cell system and in developing controlling and treating certain disorders by new drugs.","Title: Bolus Chasing Computed Tomography Angiography Using Local Maximum Tracking Method\nYear: 2007\n\nAbstract: Tracking bolus peak position is a crucial issue for Computed Tomography Angiography(CTA). In this paper, a local maximum tracking method is proposed and tested using real patient data. The method uses a second order polynomial to approximate bolus density function locally. By estimating the local maximum density position at next sampling time, bolus peak trajectory is closely followed. Experiments with Magnetic Resonance Angiography (MRA) patient data showa significant performance improvement using the proposed tracking method over the conventional constant speed counterpart.","Title: An Optimization Based Robust Identification Algorithm in the Presence of Outliers\nYear: 2002\n\nAbstract: In this paper, the robust identification problem is formulated in the framework of optimization with few violated constraints and an efficient numerical algorithm is presented with a low complexity. Moreover, it is shown that the proposed method requires minimum a priori information and is convergent in the presence of outliers.","Title: Some theoretical limits on nuclear source localization and tracking\nYear: 2015\n\nAbstract: We consider the simple and general estimation problem of finding the location of a nuclear source from radiation measurements. Our objective is to study the effect of the inherent quantum randomness of radioactive emissions on the accuracy to which nuclear sources can be localized. To this end, we consider an ideal mobile detector making perfect, noiseless measurements and formulate a general problem of maximum likelihood estimation of source location using such measurements. For the case of a stationary source and a detector moving with uniform speed in a straight line, we derive solutions to the maximum likelihood location estimate as well as the corresponding Cramer-Rao lower bounds. We present a simple iterative procedure for calculating the ML estimate, and argue that in the asymptotic case of source strength becoming large, the procedure converges to the ML estimate with high probability and this estimate is unbiased. We also present simulations showing that the maximum likelihood estimates approach the Cramer-Rao bounds, and comment on the implications of these theoretical results with ideal detectors and perfect estimators to the problem of nuclear source localization.","Title: A globally consistent nonlinear least squares estimator for identification of nonlinear rational systems.\nYear: 2017\n\nAbstract: This paper considers identification of nonlinear rational systems defined as the ratio of two nonlinear functions of past inputs and outputs. Despite its long history, a globally consistent identification algorithm remains illusive. This paper proposes a globally convergent identification algorithm for such nonlinear rational systems. To the best of our knowledge, this is the first globally convergent algorithm for the nonlinear rational systems. The technique employed is a two-step estimator. Though two-step estimators are known to produce consistent nonlinear least squares estimates if a N consistent estimate can be determined in the first step, how to find such a N consistent estimate in the first step for nonlinear rational systems is nontrivial and is not answered by any two-step estimators. The technical contribution of the paper is to develop a globally consistent estimator for nonlinear rational systems in the first step. This is achieved by involving model transformation, bias analysis, noise variance estimation, and bias compensation in the paper. Two simulation examples and a practical example are provided to verify the good performance of the proposed two-step estimator.","Title: Consistent Variable Selection For High-Dimensional Nonparametric Additive Nonlinear Systems\nYear: 2016\n\nAbstract: In this paper, the problem of variable selection is addressed for high-dimensional nonparametric additive nonlinear systems. The purpose of variable selection is to determine contributing additive functions and to remove non-contributing ones from the underlying nonlinear system. A two-step method is developed to conduct variable selection. The first step is concerned with estimating each additive function by virtue of kernel-based nonparametric approaches. The second step is to apply a nonnegative garrote estimator to identify which additive functions are nonzero in terms of the obtained non parametric estimates of each function. The proposed variable selection method is workable without suffering from the curse of dimensionality, and it is able to find the correct variables with probability one under weak conditions as the sample size approaches infinity. The good performance of the proposed variable selection method is demonstrated by a numerical example.","Title: Variable Selection and Identification of High-Dimensional Nonparametric Additive Nonlinear Systems.\nYear: 2017\n\nAbstract: This paper considers variable selection and identification of dynamic additive nonlinear systems via kernel-based nonparametric approaches, where the number of variables and additive functions may be large. Variable selection aims to find which additive functions contribute and which do not. The proposed variable selection consists of two successive steps. At the first step, one estimates each additive function by kernel-based nonparametric identification approaches without suffering from the curse of dimensionality. At the second step, a nonnegative garrote estimator is applied to identify which additive functions are nonzero by utilizing the obtained nonparametric estimates of each function. Under weak conditions, the nonparametric estimates of each additive function can achieve the same asymptotic properties as for 1D nonparametric identification based on kernel functions. It is also established that the nonnegative garrote estimator turns a consistent estimate for each additive function into a consistent variable selection with probability one as the number of samples tends to infinity. Two simulation examples are presented to verify the effectiveness of the variable selection and identification approaches proposed in the paper.","Title: Variable Selection Of High-Dimensional Non-Parametric Nonlinear Systems: A Way To Avoid The Curse Of Dimensionality\nYear: 2017\n\nAbstract: This paper presents a numerically efficient algorithm for variable selection for high-dimensional nonlinear non-parametric systems. It is based on the average derivatives and relies on one dimensional estimates of the density function and its derivative. Thus it avoids the curse of dimensionality usually encountered for high-dimensional systems. Theoretical analysis is provided and shows that the conditions derived are sufficient for a variable to contribute, and necessary and almost sufficient for a variable not to contribute for a class of nonlinear systems. Further, convergence results are established.","Title: Source Localization by a Binary Sensor Network in the Presence of Imperfection, Noise, and Outliers.\nYear: 2018\n\nAbstract: In this paper, source localization by a network of primitive binary sensors under various imperfections are studied. Detailed analysis and mathematical modeling of imperfect binary sensors are presented. Imperfections include sensor failures of two types, drifting, uncertainty, and heterogeneity in binary sensor trigger thresholds, presence of noise, and nonradial symmetry of sensing ranges. Theoretical results, including asymptotical convergence, are established, in particular in the presence of substantial outliers due to sensor failure and large noise. Efficient numerical algorithms are proposed and simulated supporting the theoretical analysis.","Title: Local variable selection of nonlinear nonparametric systems by first order expansion.\nYear: 2018\n\nAbstract: Local variable selection by first order expansion for nonlinear nonparametric systems is investigated in the paper. By substantially modifying the algorithms developed in our earlier work (Bai et al., 2014), the previous results have been considerably strengthened under much less restrictive conditions. Firstly, the estimates generated by the modified algorithms are shown to have both the set and parameter convergence with probability one, rather than only the set convergence in probability given in our earlier work. Secondly, several technical assumptions, e.g., the lower and upper bounds on the growth of some random sequences, which practically are uncheckable, have been removed. Thirdly, not only the consistency but also the convergence rate of estimates have been established. Besides, a generalization of the proposed algorithms is also introduced.","Title: Variable selection of high-dimensional non-parametric nonlinear systems by derivative averaging to avoid the curse of dimensionality.\nYear: 2019\n\nAbstract: In identification, variable selection for a nonlinear non-parametric and high-dimensional system is the first and often one of the most difficult problems. The issue is the curse of dimensionality. This paper presents a numerically efficient algorithm of variable selection for high-dimensional nonlinear non-parametric systems. It is based on averaging derivatives and relies on one dimensional estimates of the density function and its derivative. Thus, it avoids the curse of dimensionality usually encountered for high-dimensional systems. Theoretical analysis is provided and the conditions are derived for a variable to contribute or not to contribute for a large class of nonlinear systems. Further, convergence results are established.","Title: Frequency domain identification of hammerstein models\nYear: 2003\n\nAbstract: In this chapter, we discuss a frequency approach for Hammerstein model identification. The method is based on the fundamental\n frequency and therefore, no a priori information on the structure of the nonlinearity is required. Moreover, the method is not limited to Hammerstein models whose\n linear part is a finite-order rational transfer functions, but applies to Hammerstein models with a non-parametric linear\n part. The method can be easily extended to Wiener models with minor modifications. The chapter is based on an article [1],\n IEEE Trans. on Automatic Control, Vol. 48, pp.530-542, 2003 with permission from IEEE Intellectual Property Rights Office.\n All the proofs can be found in [1].","Title: Ranking Variables Based on Goodness of Fits in Nonlinear Nonparametric System Identification.\nYear: 2018\n\nAbstract: Identification of a high dimensional nonlinear nonparametric system is costly. On the other hand for many real-world problems, they are sparse in the sense that not all variables contribute or contribute significantly. If these variables that do not contribute or contribute little can be identified and removed prior to system identification, the identification problem is of lower dimension. In this paper, methods to rank variables based on the Goodness of Fits are proposed without full scale identification.","Title: Ranking the importance of variables in nonlinear system identification\nYear: 2019\n\nAbstract: In this paper, the importance measure based on Goodness of Fit (GOF) is defined. Based on the measure, ways to rank the importance of variables are proposed. In particular, this paper addresses three questions: (1) What is the importance measure based on Goodness of Fit (GOF)? (2) How to rank the variables based on GOF, prior to actual identification? (3) Given an integer d, can we find theoretically d-variables that collectively have the largest contribution among all d-variable subsets based on GOF? An affirmative answer is provided along with numerical algorithms. The problem of ranking variables prior to actual identification is important in identification of high dimensional nonlinear systems. If variables that contribute little can be identified and removed prior to system identification, then the identification problem is of lower dimension and relatively easy to deal with.","Title: Identification Of An Additive Nfir System And Its Applications In Generalized Hammerstein Models\nYear: 2005\n\nAbstract: Identification of a nonlinear additive system is considered. An input signal is designed in such a way that the problem of identification of nonlinear additive systems is reduced to a problem of identification of static nonlinear functions. Then, three approaches are established to estimate the order of the system. The methods exploit the structure of the nonlinear additive model so that their implementations are easy.","Title: Persistence of excitation for identifying switched linear systems\nYear: 2022\n\nAbstract: This paper investigates the uniqueness of parameters via persistence of excitation for switched linear systems. The main contribution is a much weaker sufficient condition on the regressors to be persistently exciting that guarantees the uniqueness of the parameter sets and also provides new insights in understanding the relation among different subsystems. It is found that for uniquely determining the parameters of switched linear systems, the needed minimum number of samples derived from our sufficient condition is much smaller than that reported in the literature."],"string":"[\n \"Title: Brief A random least-trimmed-squares identification algorithm\\nYear: 2003\\n\\nAbstract: The Least-trimmed-squares (LTS) estimator is a well known robust estimator in terms of protecting the estimate from the outliers. Its high computational complexity is however a problem in practice. In this paper, we propose a random LTS algorithm which has a low computational complexity that can be calculated a priori as a function of the required error bound and the confidence interval. Moreover, if the number of data points goes to infinite, the algorithm becomes a deterministic one that converges to the true LTS in some probability sense.\",\n \"Title: Brief Variable gain parameter estimation algorithms for fast tracking and smooth steady state\\nYear: 2000\\n\\nAbstract: Based on the ideas of the set-membership identification, we present in this paper variable gain least mean-square (LMS) and weighted recursive least-squares (WRLS) algorithms. It is shown that the proposed algorithms possess a fast tracking ability to the variations of the parameter and at the same time are less sensitive to noise leading to a smooth steady state.\",\n \"Title: Constrained logarithmic least squares in parameter estimation.\\nYear: 1999\\n\\nAbstract: The contribution of this paper is twofolds. First, it is shown that while robust in termsof the average output error, the least squares estimate is sensitive to outliers with respectto the maximum output error. In fact the worst-case output error of the least squares cango unbounded. Then, a constrained logarithmic least squares for system identification isproposed. Analytic center algorithms are presented to solve this constrained logarithmicleast squares problem.1 IntroductionThe...\",\n \"Title: Decoupling the linear and nonlinear parts in Hammerstein model identification\\nYear: 2004\\n\\nAbstract: The main result of this paper is to show that the linear part can be made decoupled from the nonlinear part in Hammerstein model identification. Therefore, identification of the linear part for a Hammerstein model becomes a linear problem and accordingly enjoys the same convergence and consistency results as if the unknown nonlinearity is absent.\",\n \"Title: Least squares solutions of bilinear equations\\nYear: 2006\\n\\nAbstract: The problem of finding a least squares solution for a system of bilinear equations is investigated. Sufficient conditions to have a unique minimum are given in the case of random inputs. Three methods, the normalized iterative method, the over-parameterization method and the numerical method are presented for solving the problem along with their convergence properties. Simulation examples are provided.\",\n \"Title: PageRank computation via a distributed randomized approach with lossy communication.\\nYear: 2012\\n\\nAbstract: For ranking web pages in search results, Google employs the so-called PageRank algorithm, which provides a measure of importance to each page based on the web structure. Recently, we have developed a distributed randomized approach for this algorithm, where pages compute their own PageRank by communicating over selected links. In this paper, the focus is on the effects of unreliability in communication channels. Specifically, we consider random data losses modeled as a Markov chain and introduce a generalized version of the distributed scheme. Its convergence properties and the error in the approximated PageRank are analyzed.\",\n \"Title: Optimization with few violated constraints for linear bounded error parameter estimation\\nYear: 2002\\n\\nAbstract: In the context of linear constrained optimization, we study in this paper the problem of finding an optimal solution sat- isfying all but of the given constraints. A solution is obtained by means of an algorithm of the complexity , where is the dimension of the problem. We then use these results to solve the problem of robust identification in the presence of outliers in the setting of bounded error parameter iden- tification. Finally, we show that the estimate obtained converges to the true but unknown parameter in the presence of outliers. Index Terms—Constrainted optimization, parameter estima- tion, system identification, unknown but bounded error.\",\n \"Title: Distributed randomized PageRank computation based on web aggregation\\nYear: 2009\\n\\nAbstract: The so-called PageRank algorithm has been used at Google for properly ranking search results. It quantifies the importance of each page by the structure of links in the web. In our recent work, we have proposed a distributed randomized approach for the PageRank computation, where the pages find their own values by communicating with linked pages. This paper builds upon this approach to improve the computation and communication load for the algorithm by systematically aggregating the webpages while maintaining performance.\",\n \"Title: A Web Aggregation Approach for Distributed Randomized PageRank Algorithms\\nYear: 2012\\n\\nAbstract: The PageRank algorithm employed at Google assigns a measure of importance to each web page for rankings in search results. In our recent papers, we have proposed a distributed randomized approach for this algorithm, where web pages are treated as agents computing their own PageRank by communicating with linked pages. This paper builds upon this approach to reduce the computation and communication loads for the algorithms. In particular, we develop a method to systematically aggregate the web pages into groups by exploiting the sparsity inherent in the web. For each group, an aggregated PageRank value is computed, which can then be distributed among the group members. We provide a distributed update scheme for the aggregated PageRank along with an analysis on its convergence properties. The method is especially motivated by results on singular perturbation techniques for large-scale Markov chains and multi-agent consensus. A numerical example is provided to illustrate the level of reduction in computation while keeping the error in rankings small.\",\n \"Title: A Recursive Local Linear Estimator for Identification of Nonlinear ARX Systems: Asymptotical Convergence and Applications\\nYear: 2013\\n\\nAbstract: In this paper, we propose a recursive local linear estimator (RLLE) for nonparametric identification of nonlinear autoregressive systems with exogenous inputs (NARX). First, the RLLE is introduced. Next, the strong consistency as well as the asymptotical mean square error properties of the RLLE are established, and then an application of the RLLE to an additive nonlinear system is discussed. The RLLE provides recursive estimates not only for the function values but also their gradients at fixed points. A simulation example is provided to confirm the theoretical analysis.\",\n \"Title: An optimal two stage identification algorithm for Hammerstein-Wiener nonlinear systems\\nYear: 1998\\n\\nAbstract: An optimal two stage identification algorithm is presented for Hammerstein-Wiener systems where two static nonlinear elements surround a linear block. The proposed algorithm consists of two steps: The first one is the recursive least squares and the second one is the singular value decomposition of two matrices whose dimensions are fixed and do not increase as the number of the data point increases. Moreover, the algorithm is shown to be convergent in the absence of noise and convergent with probability one in the presence of white noise\",\n \"Title: Blind decision feedback equalization of time-varying channels with DPSK inputs\\nYear: 2001\\n\\nAbstract: This paper presents a new method to the blind equalization of a class of linear time-varying mobile channels for differentially encoded channel input signals. Through re-parameterization, the discrete system can be modeled equivalently as a single input multiple output system. Unlike some existing methods, this new algorithm does not require prior knowledge on the cyclic frequency of the channel response. The algorithm is simple and relies only on a decision feedback structure and the differential encoding nature of the channel input signals. The convergence analysis of the decision feedback system is presented along with simulation examples\",\n \"Title: A blind approach to the Hammerstein-Wiener model identification\\nYear: 2002\\n\\nAbstract: The work of this chapter is a continuation of [3]. Unlike in [3] where a very special structure is assumed, however, the blind\\n approach in this chapter allows a very general structure on the nonlinearities. In particular, the input nonlinearity structure\\n can be arbitrary and is not assumed to be known. By using the blind identification approach, all the unknown internal variables\\n can be recovered solely based on the output measurements.\",\n \"Title: Worst-case properties of the uniform distribution and randomized algorithms for robustness analysis\\nYear: 1998\\n\\nAbstract: In this paper we study a probabilistic approach which is an alternative to the classical worst-case algorithms for robustness analysis and design of uncertain control systems. That is, we aim to estimate the probability that a control system with uncertain parameters q restricted to a box I! attains a given level of performance gamma. Since this probability depends on the underlying distribution, we address the following question: What is a \\\"reasonable\\\" distribution so that the estimated probability makes sense? To answer this question, we define two worst-case criteria and prove that the uniform distribution is optimal in both cases. In the second part of the paper we turn our attention to a subsequent problem. That is, we estimate the sizes of both the so-called \\\"good\\\" and \\\"bad\\\" sets via sampling. Roughly speaking, the good set contains the parameters q is an element of Q with a performance level better than or equal to gamma while the bad set is the set of parameters q is an element of Q with a performance level worse than gamma. We give bounds on the minimum sample size to attain a good estimate of these sets in a certain probabilistic sense.\",\n \"Title: Adaptive bolus-chasing computed tomography angiography in the cases of symmetric and asymmetric arterial flows in peripheral arteries.\\nYear: 2009\\n\\nAbstract: Synchronization of the contrast bolus peak and CT imaging aperture is a crucial issue for computed tomography angiography (CTA). It affects the CTA image quality and the amount of contrast dose. A whole-body CTA procedure means to scan from the abdominal aorta to pedal arteries. In this context, the synchronization is much more difficult with the asymmetric arterial flow in lower extremities than in the case of symmetric arterial flow. In this paper, we propose an adaptive optimal controller to chase the contrast bolus peak while it propagates in the aorta and lower extremities with symmetric flow. In the case of asymmetric flow after the contrast bolus splitting into two lower limbs, we propose a dynamic programming approach to cover the lower limbs optimally. Simulation and experimental results show that the proposed methods outperform the current constant-speed method substantially.\",\n \"Title: Two-stage mixed discrete-continuous identification of radial basis function (RBF) neural models for nonlinear systems\\nYear: 2009\\n\\nAbstract: The identification of nonlinear dynamic systems using radial basis function (RBF) neural models is studied in this paper. Given a model selection criterion, the main objective is to effectively and efficiently build a parsimonious compact neural model that generalizes well over unseen data. This is achieved by simultaneous model structure selection and optimization of the parameters over the continuous parameter space. It is a mixed-integer hard problem, and a unified analytic framework is proposed to enable an effective and efficient two-stage mixed discrete-continuous identification procedure. This novel framework combines the advantages of an iterative discrete two-stage subset selection technique for model structure determination and the calculus-based continuous optimization of the model parameters. Computational complexity analysis and simulation studies confirm the efficacy of the proposed algorithm.\",\n \"Title: Brief Limiting zero distribution of sampled systems\\nYear: 2002\\n\\nAbstract: In this paper, we consider the sampling of a continuous time system with possibly a non-rational transfer function. Limiting zero distribution of the sampled system represented by an FIR model is derived. It is shown that the zeros of the FIR sampled system converge evenly to the unit circle or to the unit disk plus unstable roots of the reciprocal polynomials depending on assumptions.\",\n \"Title: An Adaptive Optimal Control Design for a Bolus Chasing Computed Tomography Angiography\\nYear: 2008\\n\\nAbstract: To improve imaging quality and to reduce contrast dose and radiation exposure, an optimal adaptive bolus chasing controller is proposed and tested based on actual patient data. The controller estimates and predicts the unknown 2D bolus density online and then determines the optimal control actions. Tracking errors are mathematically quantified in terms of estimation errors. The test results not only support the analytical analysis and exhibit its superior performance over the current constant-speed controller, but also demonstrate the clinical feasibility.\",\n \"Title: Representation and identification of non-parametric nonlinear systems of short term memory and low degree of interaction\\nYear: 2010\\n\\nAbstract: In this paper, we propose a new representation which is particularly useful for a class of non-parametric nonlinear systems that have short term memory and low degree of interaction. Advantages and disadvantages of this representation are discussed and compared to existing methods both theoretically and numerically. Furthermore, results regarding structural estimation based on the analysis of variance and on full scale identification are also provided.\",\n \"Title: Frequency domain identification of Wiener models\\nYear: 2003\\n\\nAbstract: This paper proposes a frequency domain algorithm for Wiener model identifications based on exploring the fundamental frequency and harmonics generated by the unknown nonlinearity. The convergence of the algorithm is established in the presence of white noise. No a priori knowledge of the structure of the nonlinearity is required and the linear part can be nonparametric.\",\n \"Title: Iterative identification of Hammerstein systems\\nYear: 2007\\n\\nAbstract: Convergences of iterative algorithms have been established for identification of Hammerstein systems in the case that the unknown nonlinearities are odd. Then, the results are further extended to nonsmooth nonlinearities.\",\n \"Title: Blind system identification and channel equalization of IIR systemswithout statistical information\\nYear: 1999\\n\\nAbstract: A novel approach is proposed to blindly identify an unknown IIR system. The approach is based on faster sampling at the system output and requires neither a priori statistical information on the unknown input nor training signals. The methods presented are linear in the parameters of the unknown system so that many standard recursive algorithms can be readily applied. It is also shown in the paper that under a generic condition, any finite-order IIR system is identifiable, provided the oversampling ratio is appropriately chosen\",\n \"Title: Adaptive bolus chasing computed tomography angiography: Control scheme and experimental results.\\nYear: 2008\\n\\nAbstract: In this paper, a new adaptive bolus-chasing control scheme is proposed to synchronize the bolus peak in a patient's vascular system and the imaging aperture of a computed tomography (CT) scanner. The proposed control scheme is theoretically evaluated and experimentally tested on a modified Siemens SOMATOM Volume Zoom CT scanner. The first set of experimental results is reported on bolus-chasing CT angiography using realistic bolus dynamics, real-time CT imaging and adaptive table control with physical vasculature phantoms. The data demonstrate that the proposed control approach tracks the bolus propagation well, and clearly outperforms the constant-speed scheme that is the current clinical standard.\",\n \"Title: Blind identifiability of IIR systems\\nYear: 2002\\n\\nAbstract: It is shown that, with no prior information, a necessary and sufficient condition for an nth order IIR system to be blindly identifiable is that the over-sampling ratio p⩾n+1.\",\n \"Title: Capacitors and the method of moments\\nYear: 2004\\n\\nAbstract: The inhomogeneous charge distribution in several plate capacitors is calculated using the Method of Moments. Effects of global and local “inhomogeneities” in the capacitor are simulated.\",\n \"Title: Toeplitz matrix based sparse error correction in system identification: Outliers and random noises\\nYear: 2013\\n\\nAbstract: In this paper, we consider robust system identification under sparse outliers and random noises. In our problem, system parameters are observed through a Toeplitz matrix. All observations are subject to random noises and a few are corrupted with outliers. We reduce this problem of system identification to a sparse error correcting problem using a Toeplitz structured real-numbered codingmatrix. We prove the performance guarantee of Toeplitz structured matrix in sparse error correction. Thresholds on the percentage of correctable errors for Toeplitz structured matrices are also established. When both outliers and observation noise are present, we have shown that the estimation error goes to 0 asymptotically as long as the probability density function for observation noise is not “vanishing” around 0.\",\n \"Title: Zeros of sampled data systems represented by FIR models\\nYear: 2000\\n\\nAbstract: In this note, we investigate zero locations of FIR linear systems that are finite length approximations of sampled continuous-time systems. For linear systems with rational transfer functions, it is shown that with a relative degree higher than two and a short sampling interval, the resultant FIR sampled data system is always a nonminimum phase.\",\n \"Title: Towards identification of Wiener systems with the least amount of a priori information on the nonlinearity\\nYear: 2008\\n\\nAbstract: In this paper, we investigate what constitutes the least amount of a priori information on the nonlinearity so that the FIR linear part is identifiable in the non-Gaussian input case. Three types of a priori information are considered including quadrant information, point information and locally monotonous information. In all three cases, identifiability has been established and corresponding identification algorithms are developed with their convergence proofs.\",\n \"Title: Convergence of the iterative algorithm for a general Hammerstein system identification\\nYear: 2010\\n\\nAbstract: The convergence of the iterative identification algorithm for a general Hammerstein system has been an open problem for a long time. In this paper, it is shown that the convergence can be achieved by incorporating a regularization procedure on the nonlinearity in addition to a normalization step on the parameters.\",\n \"Title: Fatigue and non-fatigue mathematical muscle models during functional electrical stimulation of paralyzed muscle.\\nYear: 2010\\n\\nAbstract: Electrical muscle stimulation demonstrates potential for preventing muscle atrophy and restoring functional movement after spinal cord injury (SCI). Control systems used to optimize delivery of electrical stimulation protocols depend upon the algorithms generated using computational models of paralyzed muscle force output. The Hill–Huxley-type model, while being highly accurate, is also very complex, making it difficult for real-time implementation. In this paper, we propose a Wiener–Hammerstein system to model the paralyzed skeletal muscle under electrical stimulus conditions. The proposed model has substantial advantages in identification algorithm analysis and implementation including computational complexity and convergence, which enable it to be used in real-time model implementation. Experimental data sets from the soleus muscles of 14 subjects with SCI were collected and tested. The simulation results show that the proposed model outperforms the Hill–Huxley-type model not only in peak force prediction, but also in fitting performance for force output of each individual stimulation train.\",\n \"Title: Convergence results of the analytic center estimator\\nYear: 2000\\n\\nAbstract: The analytic center approach for bounded error parameter estimation was recentlyproposed as an alternative to the well-known Chybyshev and least squares estimates.In this paper, we show the asymptotic performance of this approach and prove that theanalytic center converges to the true parameter under mild conditions.This work was supported in part by grants of NSF(USA)1;4, ARC(Australia)2and CENS-CNR(Italy)1;3.The first author especially thanks the support of CNR short-term...\",\n \"Title: A comparative study on interpolation methods for controlled cardiac CT\\nYear: 2007\\n\\nAbstract: Bai et al. recently proposed to acquire random fan-beam-cone-beam projections with a linear-planar detector from a circular scanning locus for controlled cardiac computed tomography (CT). After specifying a uniform acquisition geometry required by FBP (filtration-backprojection), we rebin the random fan-beam-cone-beam data via nearest-neighbor, quadrilateral and triangle-based linear interpolation methods. The fan-beam and parallel-beam FBP algorithms are employed for rebinned fan-beam projections. The FDK (Feldkamp–Davis–Kress) and t-FDK (tent-FDK) methods are employed for rebinned cone-beam data. Also, nonuniform weighting fan-beam-FDK methods are used to reconstruct without rebinning. As a benchmark, the images are reconstructed using uniform weighting fan-beam-FDK method from data collected at the specified uniform grid. To evaluate the different methods, effects of increasing the number of projections and adding Poisson noise are studied. The root mean square error (RMSE) is used to quantify the image quality by numerical tests with the cardiac phantom. Our results show that it is helpful to perform data interpolation for improvements of the image quality in controlled cardiac CT from random projections. Our simulations indicate that triangular interpolation gives the most satisfactory result for improved image quality whereas quadrilateral interpolationgives the best noise performance. © 2007 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 17, 91–98, 2007\",\n \"Title: Towards Identification of Wiener Systems with the Least Amount of a priori Information: IIR Cases.\\nYear: 2009\\n\\nAbstract: In this paper, we investigate what constitutes the least amount of a priori information on the nonlinearity so that the linear part is identifiable in the non-Gaussian input case. Under the white noise input, three types of a priori information are considered including quadrant information, point information and monotonic information. In all three cases, identifiability has been established and the corresponding nonparametric identification algorithms are developed along with their convergence proofs.\",\n \"Title: On the capacitance of a cube\\nYear: 2002\\n\\nAbstract: The capacitance of a metallic cube is analyzed using the method of subareas and modern computing tools. This calculation extends the work of Reitan and Higgins in order to obtain a more accurate value for the capacitance.\",\n \"Title: Analytic center approach to parameter estimation: convergence analysis\\nYear: 1998\\n\\nAbstract: The so-called analytic center approach to parameter estimation has been proposed as an alternative to the well-known least squares approach. This new approach offers a parameter estimate that is consistent with the past data observations, has a simple geometric interpretation, and is computable sequentially. We study the asymptotic performance of the analytic center approach and show that the resulting estimate converges to the true parameter asymptotically, provided some mild conditions are satisfied. These conditions involve some weak persistent excitation and independence between noise and regressor, similar to the least squares case. This result is used to derive a new parameter estimation approach which offers both good transient and asymptotic performance\",\n \"Title: A New Approach For Aggregated Pagerank Computation Via Distributed Randomized Algorithms\\nYear: 2011\\n\\nAbstract: At Google, the PageRank algorithm helps rankings in search results by providing measures of web page importance. This paper builds upon the distributed randomized approach for this algorithm proposed in our recent works. To reduce computation and communication, we develop a method to systematically aggregate web pages into groups by exploiting the sparsity inherent in the web. Each group computes an aggregated PageRank, which can be distributed among group members. We provide a decentralized scheme for its computation and analyze convergence properties.\",\n \"Title: Identification of non-parametric FIR non-linear systems with low-degree interactive terms\\nYear: 2009\\n\\nAbstract: In this paper, an interactive term identification approach is proposed for identification of non-parametric finite impulse response non-linear systems under i.i.d. random sequences. The idea is to make a high-dimensional non-linear identification problem into a number of low-dimensional problems and thus to effectively combat the problem of the curse of dimensionality. Convergence results are esta...\",\n \"Title: A two-stage algorithm for identification of nonlinear dynamic systems\\nYear: 2006\\n\\nAbstract: This paper investigates the two-stage stepwise identification for a class of nonlinear dynamic systems that can be described by linear-in-the-parameters models, and the model has to be built from a very large pool of basis functions or model terms. The main objective is to improve the compactness of the model that is obtained by the forward stepwise methods, while retaining the computational efficiency. The proposed algorithm first generates an initial model using a forward stepwise procedure. The significance of each selected term is then reviewed at the second stage and all insignificant ones are replaced, resulting in an optimised compact model with significantly improved performance. The main contribution of this paper is that these two stages are performed within a well-defined regression context, leading to significantly reduced computational complexity. The efficiency of the algorithm is confirmed by the computational complexity analysis, and its effectiveness is demonstrated by the simulation results.\",\n \"Title: Brief Identification of linear systems with hard input nonlinearities of known structure\\nYear: 2002\\n\\nAbstract: This paper studies identification of systems with input nonlinearities of known structure. For input nonlinearities parameterized by one parameter, a deterministic approach is proposed based on the idea of separable least squares. The identification problem is shown to be equivalent to an one-dimensional minimization problem. The method is very effective for several common static and nonstatic input nonlinearities. For a general input nonlinearity, a correlation analysis based identification algorithm is presented which is shown to be convergent.\",\n \"Title: Non-Parametric Nonlinear System Identification: A Data-Driven Orthogonal Basis Function Approach\\nYear: 2008\\n\\nAbstract: In this paper, a data driven orthogonal basis function approach is proposed for non-parametric FIR nonlinear system identification. The basis functions are not fixed a priori and match the structure of the unknown system automatically. This eliminates the problem of blindly choosing the basis functions without a priori structural information. Further, based on the proposed basis functions, approaches are proposed for model order determination and regressor selection along with their theoretical justifications.\",\n \"Title: Making parametric Hammerstein system identification a linear problem\\nYear: 2011\\n\\nAbstract: In this paper, we study the identification of parametric Hammerstein systems with FIR linear parts. By a proper normalization and a clever characterization, it is shown that the average squared error cost function for identification can be expressed in terms of the inner product between the true but unknown parameter vector and its estimate. Further, the cost function is concave in the inner product and linear in the inner product square. Therefore, the identification of parametric Hammerstein systems with FIR linear parts is a globally convergent problem and has one and only one (local and global) minimum. This implies that the identification of such systems is a linear problem in terms of the inner product square and any local search based identification algorithm converges globally.\",\n \"Title: A New Extension of Newton Algorithm for Nonlinear System Modelling Using RBF Neural Networks\\nYear: 2013\\n\\nAbstract: Model performance and convergence rate are two key measures for assessing the methods used in nonlinear system identification using Radial Basis Function neural networks. A new extension of the Newton algorithm is proposed to further improve these two aspects by extending the results of recently proposed continuous forward algorithm (CFA) and hybrid forward algorithm (HFA). Computational complexity analysis confirms its efficiency, and numerical examples show that it converges faster and potentially outperforms CFA and HFA.\",\n \"Title: Non-Parametric Nonlinear System Identification: An Asymptotic Minimum Mean Squared Error Estimator\\nYear: 2009\\n\\nAbstract: This paper studies the problem of the minimum mean squared error estimator for non-parametric nonlinear system identification. It is shown that for a wide class of nonlinear systems, the local linear estimator is a linear (in outputs) asymptotic minimum mean squared error estimator. The class of the systems allowed is characterized by a stability condition that is related to many well studied stability notions in the literature. Numerical simulations support the analytical analysis.\",\n \"Title: Kernel based approaches to local nonlinear non-parametric variable selection\\nYear: 2014\\n\\nAbstract: In this paper, we consider the variable selection problem for a nonlinear non-parametric system. Two approaches are proposed, one top-down approach and one bottom-up approach. The top-down algorithm selects a variable by detecting if the corresponding partial derivative is zero or not at the point of interest. The algorithm is shown to have not only the parameter but also the set convergence. This is critical because the variable selection problem is binary, a variable is either selected or not selected. The bottom-up approach is based on the forward/backward stepwise selection which is designed to work if the data length is limited. Both approaches determine the most important variables locally and allow the unknown non-parametric nonlinear system to have different local dimensions at different points of interest. Further, two potential applications along with numerical simulations are provided to illustrate the usefulness of the proposed algorithms.\",\n \"Title: Identification of a Modified Wiener-Hammerstein System and Its Application in Electrically Stimulated Paralyzed Skeletal Muscle Modeling.\\nYear: 2009\\n\\nAbstract: Electrical muscle stimulation demonstrates potential for restoring functional movement and preventing muscle atrophy after spinal cord injury (SCI). Control systems used to optimize delivery of electrical stimulation protocols depend upon mathematical models of paralyzed muscle force outputs. While accurate, the Hill–Huxley-type model is very complex, making it difficult to implement for real-time control. As an alternative, we propose a modified Wiener–Hammerstein system to model the paralyzed skeletal muscle dynamics under electrical stimulus conditions. Experimental data from the soleus muscles of individuals with SCI was used to quantify the model performance. It is shown that the proposed Wiener–Hammerstein system is at least comparable to the Hill–Huxley-type model. On the other hand, the proposed system involves a much smaller number of unknown coefficients. This has substantial advantages in identification algorithm analysis and implementation including computational complexity, convergence and also in real-time model implementation for control purposes.\",\n \"Title: How Nonlinear Parametric Wiener System Identification is Under Gaussian Inputs?\\nYear: 2012\\n\\nAbstract: A number of methods exist for identifying nonlinear Wiener systems. However, there is no attempt to address the fundamental question of how nonlinear these identification problems really are? In this technical note, we try to address this question by investigating the average squared error cost function used in identification. By a proper normalization and a clever characterization of the cost function in terms of the angle between the true but unknown parameter vector and its estimate, it is shown in the technical note that under iid Gaussian inputs for parametric Wiener systems with polynomial nonlinear parts and FIR linear parts, the cost function is globally monotonic and has one and only one (local and global) minimum. The implication is that identification of such systems is nonlinear but very close to linear. Further, any local search based identification algorithms would converge globally for such systems.\",\n \"Title: Robust system identification with noisy experimental data: projection operator and linear algorithms\\nYear: 1994\\n\\nAbstract: In this paper we consider the problem of robust system identification with noisy time or frequency response measurement data. It is shown here that any linear identification algorithm which is convergent in the noise free case can be made robustly convergent in the presence of noise by incorporating a simple projection operator into the algorithm. The computational simplicity and faster rate of convergence distinguish this approach from other existing robustly convergent nonlinear identification techniques.\",\n \"Title: Identification of an additive nonlinear system and its applications in generalized Hammerstein models\\nYear: 2008\\n\\nAbstract: An identification algorithm is developed for a class of nonlinear systems that are multi-input and multi-output in an additive form. The convergence results are achieved and its applications to identification of a generalized Hammerstein system is also discussed.\",\n \"Title: Recursive Direct Weight Optimization in Nonlinear System Identification: A Minimal Probability Approach\\nYear: 2007\\n\\nAbstract: In this paper, a direct weight optimization method is proposed for nonlinear system identification based on a minimal probability idea. The approach has several quite attractive features and is very different from existing ones. It is optimal for any given number of finite data points and at the same time possesses asymptotic convergence. The estimator admits a closed form and no numerical optimization is needed. Theoretical analysis and numerical simulations show that the approach is a very competitive alternative to existing nonlinear identification methods.\",\n \"Title: Identification of nonlinear additive FIR systems\\nYear: 2005\\n\\nAbstract: In this paper, we consider identification of a nonlinear additive system. An input signal is designed in such a way that the problem of identification of nonlinear additive systems is reduced to a problem of identification of static nonlinear functions. Then, three approaches are established to estimate the order of the system. The methods exploit the structure of the nonlinear additive model so that their implementations are easy.\",\n \"Title: Blind source separation/channel equalization of nonlinear channels with binary inputs\\nYear: 2005\\n\\nAbstract: In this paper, we consider blind source separation/channel equalization problems for an unknown nonlinear channel. By using the property of the binary alphabets, we show that blind nonlinear source separation/channel equalization problems can be easily solved by existing linear methods.\",\n \"Title: On the Long-term Health Care Crisis. A Possible Eradication Scenario\\nYear: 2011\\n\\nAbstract: The purpose of the present essay is to suggest a possible model to describe the worldwide healthcare crisis, where diseases that have been considered to be eradicated or under our control are re-emerging today.\",\n \"Title: A Markov Chain Monte Carlo Approach to Nonlinear Parametric System Identification\\nYear: 2015\\n\\nAbstract: Nonlinear system identification is discussed in a mixed set-membership and statistical setting. A Markov chain Monte Carlo (MCMC) approach is proposed that es- timates the feasible parameter set, the minimum volume outer-bounding ellipsoid and the minimum variance esti- mate. The proposed algorithm is proved to be convergent and enjoys some desirable properties. Further, its compu- tational complexity and numerical accuracy are studied.\",\n \"Title: Kernel-based local order estimation of nonlinear nonparametric systems.\\nYear: 2015\\n\\nAbstract: We consider the local order estimation of nonlinear autoregressive systems with exogenous inputs (NARX), which may have different local dimensions at different points. By minimizing the kernel-based local information criterion introduced in this paper, the strongly consistent estimates for the local orders of the NARX system at points of interest are obtained. The modification of the criterion and a simple procedure of searching the minimum of the criterion, are also discussed. The theoretical results derived here are tested by simulation examples.\",\n \"Title: Identifiability and convergence analysis of the MINLIP estimator.\\nYear: 2015\\n\\nAbstract: This paper studies the problem of identifying a monotone Wiener system from observed, noiseless input–output signals. Specifically, we study identifiability of such systems, as well as the convergence properties of the recently introduced MINimal LIPschitz (MINLIP) estimator. This estimator finds a system with minimal complexity by exploiting ordering of the output samples. This makes the approach conceptual quite different from traditional identification schemes based on least squares, prediction errors, maximum likelihood or numerical projections. Sufficient conditions from which the result follows are given in terms of ‘Rotational Complete Inputs (RCI)’, a generalization of the notion of Persistency of Excitation (PE) as used for the study of linear identification methods. Finally, it is derived that a Gaussian i.i.d. sequence satisfies this RCI property, while any binary sequence does not.\",\n \"Title: System identification in the presence of outliers and random noises: A compressed sensing approach.\\nYear: 2014\\n\\nAbstract: In this paper, we consider robust system identification of FIR systems when both sparse outliers and random noises are present. We reduce this problem of system identification to a sparse error correcting problem using a Toeplitz structured real-numbered coding matrix and prove the performance guarantee. Thresholds on the percentage of correctable errors for Toeplitz structured matrices are established. When both outliers and observation noise are present, we have shown that the estimation error goes to 0 asymptotically as long as the probability density function for observation noise is not “vanishing” around origin. No probabilistic assumptions are imposed on the outliers.\",\n \"Title: Source location estimation for possibly unknown propagation models\\nYear: 2014\\n\\nAbstract: In this paper the source localization problem is considered without any knowledge of the signal propagation model, beyond the fact that the measured signal, e.g. received signal strength, strictly decreases with distance. Three algorithms dedicated to various scenarios are developed. Their convergence and other properties are established.\",\n \"Title: The maximum likelihood estimate for radiation source localization: Initializing an iterative search\\nYear: 2014\\n\\nAbstract: The maximum likelihood estimate approach is adopted in this paper for finding the unknown radiation source location and strength. The problem is nonlinear and has to rely on iterative numerical algorithms. Since the problem has possibly multiple local maxima, the initial estimate in those iterative algorithms plays a critical role in guaranteeing the global optimum. This paper proposes a way to generate such an initial estimate which is easy to calculate. Besides some insights that justifies the proposed approach, it is shown that the proposed initial estimate actually converges to the true but unknown maximum likelihood estimate asymptotically thus ensuring that the initial estimate is indeed in a neighborhood of the maximum likelihood estimate and consequently the convergence to the global optimum by local iterative numerical algorithms.\",\n \"Title: A distributed control law for optimum sensor placement for source localization\\nYear: 2014\\n\\nAbstract: We formulate a nonlinear distributed control law that guides the motion a group of sensors to achieve a configuration that permits them to optimally localize a hazardous source they must keep a prescribed distance from. Earlier work shows that such a configuration involves the sensors being placed in an equispaced manner on a prescribed circle. The nonlinear control law we propose assumes that each sensor resides and moves on the prescribed circle, by accessing only the states of its two immediate clockwise and counterclockwise neighbors. We theoretically prove and verify through simulations, that the law allows the sensors to achieve the desired configuration while avoiding collisions.\",\n \"Title: Two-Stage Orthogonal Least Squares Methods for Neural Network Construction.\\nYear: 2015\\n\\nAbstract: A number of neural networks can be formulated as the linear-in-the-parameters models. Training such networks can be transformed to a model selection problem where a compact model is selected from all the candidates using subset selection algorithms. Forward selection methods are popular fast subset selection approaches. However, they may only produce suboptimal models and can be trapped into a local minimum. More recently, a two-stage fast recursive algorithm (TSFRA) combining forward selection and backward model refinement has been proposed to improve the compactness and generalization performance of the model. This paper proposes unified two-stage orthogonal least squares methods instead of the fast recursive-based methods. In contrast to the TSFRA, this paper derives a new simplified relationship between the forward and the backward stages to avoid repetitive computations using the inherent orthogonal properties of the least squares methods. Furthermore, a new term exchanging scheme for backward model refinement is introduced to reduce computational demand. Finally, given the error reduction ratio criterion, effective and efficient forward and backward subset selection procedures are proposed. Extensive examples are presented to demonstrate the improved model compactness constructed by the proposed technique in comparison with some popular methods.\",\n \"Title: On the Gradient Descent Localization of Radioactive Sources\\nYear: 2013\\n\\nAbstract: We consider the robust localization of radioactive sources by using their gamma-ray count at the smallest number of sensors needed to theoretically localize. We formulate a class of non-convex cost functions and consider their gradient descent optimization. We show that in N-dimensions, if there are exactly N + 1 sensors and the source lies in their open convex hull, then this convex hull is devoid of false stationary points. Thus we augment gradient descent with random projections into the convex hull, when an estimate leaves it. We argue that convergence in probability to the correct source location, will occur. Simulations demonstrate the efficacy of this algorithm.\",\n \"Title: Time series wind power forecasting based on variant Gaussian Process and TLBO\\nYear: 2016\\n\\nAbstract: Due to the variability and stochastic nature of wind power, accurate wind power forecasting plays an important role in developing reliable and economic power system operation and control strategies. As wind variability is stochastic, Gaussian Process regression has recently been introduced to capture the randomness of wind energy. However, the disadvantages of Gaussian Process regression include its computation complexity and incapability to adapt to time varying time-series systems. A variant Gaussian Process for time series forecasting is introduced in this study to address these issues. This new method is shown to be capable of reducing computational complexity and increasing prediction accuracy. The convergence of the forecasting results is also proved. Further, a teaching learning based optimization (TLBO) method is used to train the model and to accelerate the learning rate. The proposed modelling and optimization method is applied to forecast both the wind power generation in Ireland and that from a single wind farm to demonstrate the effectiveness of the proposed method.\",\n \"Title: Robust tracking of piecewise linear trajectories with binary sensor networks\\nYear: 2015\\n\\nAbstract: We present a novel approach to the problem of tracking objects moving along smooth trajectories using a network of simple inexpensive binary sensors. Specifically, we consider object trajectories that can be well-approximated by piecewise linear curves and sensors that can only detect whether an object is in their sensing range. We start by considering objects moving along straight-line trajectories with an unknown speed and show that such objects can be tracked using the measurements of just three generically placed binary sensors whose sensing ranges intersect the trajectory. Next, we present an asymptotic analysis that shows that a trajectory consisting of a finite number of straight line segments can be recovered with high probability using an arbitrarily low spatial density of sensors in the limit when the area to be covered gets larger and larger. We also present efficient algorithms that effectively recover piecewise linear trajectories. Finally we present analysis and simulations to demonstrate the high tracking accuracy of our approach and its robustness to sensing errors.\",\n \"Title: Modelling And Parameter Estimation Of A Cell System\\nYear: 2009\\n\\nAbstract: A non-linear identification approach for modelling endothelial cells is mathematically derived and experimentally tested under the realistic alternating current voltage and a square-like shape assumption. Results demonstrate the applicability and effectiveness of the approach. The derived model is an important step in understanding a cell system and in developing controlling and treating certain disorders by new drugs.\",\n \"Title: Bolus Chasing Computed Tomography Angiography Using Local Maximum Tracking Method\\nYear: 2007\\n\\nAbstract: Tracking bolus peak position is a crucial issue for Computed Tomography Angiography(CTA). In this paper, a local maximum tracking method is proposed and tested using real patient data. The method uses a second order polynomial to approximate bolus density function locally. By estimating the local maximum density position at next sampling time, bolus peak trajectory is closely followed. Experiments with Magnetic Resonance Angiography (MRA) patient data showa significant performance improvement using the proposed tracking method over the conventional constant speed counterpart.\",\n \"Title: An Optimization Based Robust Identification Algorithm in the Presence of Outliers\\nYear: 2002\\n\\nAbstract: In this paper, the robust identification problem is formulated in the framework of optimization with few violated constraints and an efficient numerical algorithm is presented with a low complexity. Moreover, it is shown that the proposed method requires minimum a priori information and is convergent in the presence of outliers.\",\n \"Title: Some theoretical limits on nuclear source localization and tracking\\nYear: 2015\\n\\nAbstract: We consider the simple and general estimation problem of finding the location of a nuclear source from radiation measurements. Our objective is to study the effect of the inherent quantum randomness of radioactive emissions on the accuracy to which nuclear sources can be localized. To this end, we consider an ideal mobile detector making perfect, noiseless measurements and formulate a general problem of maximum likelihood estimation of source location using such measurements. For the case of a stationary source and a detector moving with uniform speed in a straight line, we derive solutions to the maximum likelihood location estimate as well as the corresponding Cramer-Rao lower bounds. We present a simple iterative procedure for calculating the ML estimate, and argue that in the asymptotic case of source strength becoming large, the procedure converges to the ML estimate with high probability and this estimate is unbiased. We also present simulations showing that the maximum likelihood estimates approach the Cramer-Rao bounds, and comment on the implications of these theoretical results with ideal detectors and perfect estimators to the problem of nuclear source localization.\",\n \"Title: A globally consistent nonlinear least squares estimator for identification of nonlinear rational systems.\\nYear: 2017\\n\\nAbstract: This paper considers identification of nonlinear rational systems defined as the ratio of two nonlinear functions of past inputs and outputs. Despite its long history, a globally consistent identification algorithm remains illusive. This paper proposes a globally convergent identification algorithm for such nonlinear rational systems. To the best of our knowledge, this is the first globally convergent algorithm for the nonlinear rational systems. The technique employed is a two-step estimator. Though two-step estimators are known to produce consistent nonlinear least squares estimates if a N consistent estimate can be determined in the first step, how to find such a N consistent estimate in the first step for nonlinear rational systems is nontrivial and is not answered by any two-step estimators. The technical contribution of the paper is to develop a globally consistent estimator for nonlinear rational systems in the first step. This is achieved by involving model transformation, bias analysis, noise variance estimation, and bias compensation in the paper. Two simulation examples and a practical example are provided to verify the good performance of the proposed two-step estimator.\",\n \"Title: Consistent Variable Selection For High-Dimensional Nonparametric Additive Nonlinear Systems\\nYear: 2016\\n\\nAbstract: In this paper, the problem of variable selection is addressed for high-dimensional nonparametric additive nonlinear systems. The purpose of variable selection is to determine contributing additive functions and to remove non-contributing ones from the underlying nonlinear system. A two-step method is developed to conduct variable selection. The first step is concerned with estimating each additive function by virtue of kernel-based nonparametric approaches. The second step is to apply a nonnegative garrote estimator to identify which additive functions are nonzero in terms of the obtained non parametric estimates of each function. The proposed variable selection method is workable without suffering from the curse of dimensionality, and it is able to find the correct variables with probability one under weak conditions as the sample size approaches infinity. The good performance of the proposed variable selection method is demonstrated by a numerical example.\",\n \"Title: Variable Selection and Identification of High-Dimensional Nonparametric Additive Nonlinear Systems.\\nYear: 2017\\n\\nAbstract: This paper considers variable selection and identification of dynamic additive nonlinear systems via kernel-based nonparametric approaches, where the number of variables and additive functions may be large. Variable selection aims to find which additive functions contribute and which do not. The proposed variable selection consists of two successive steps. At the first step, one estimates each additive function by kernel-based nonparametric identification approaches without suffering from the curse of dimensionality. At the second step, a nonnegative garrote estimator is applied to identify which additive functions are nonzero by utilizing the obtained nonparametric estimates of each function. Under weak conditions, the nonparametric estimates of each additive function can achieve the same asymptotic properties as for 1D nonparametric identification based on kernel functions. It is also established that the nonnegative garrote estimator turns a consistent estimate for each additive function into a consistent variable selection with probability one as the number of samples tends to infinity. Two simulation examples are presented to verify the effectiveness of the variable selection and identification approaches proposed in the paper.\",\n \"Title: Variable Selection Of High-Dimensional Non-Parametric Nonlinear Systems: A Way To Avoid The Curse Of Dimensionality\\nYear: 2017\\n\\nAbstract: This paper presents a numerically efficient algorithm for variable selection for high-dimensional nonlinear non-parametric systems. It is based on the average derivatives and relies on one dimensional estimates of the density function and its derivative. Thus it avoids the curse of dimensionality usually encountered for high-dimensional systems. Theoretical analysis is provided and shows that the conditions derived are sufficient for a variable to contribute, and necessary and almost sufficient for a variable not to contribute for a class of nonlinear systems. Further, convergence results are established.\",\n \"Title: Source Localization by a Binary Sensor Network in the Presence of Imperfection, Noise, and Outliers.\\nYear: 2018\\n\\nAbstract: In this paper, source localization by a network of primitive binary sensors under various imperfections are studied. Detailed analysis and mathematical modeling of imperfect binary sensors are presented. Imperfections include sensor failures of two types, drifting, uncertainty, and heterogeneity in binary sensor trigger thresholds, presence of noise, and nonradial symmetry of sensing ranges. Theoretical results, including asymptotical convergence, are established, in particular in the presence of substantial outliers due to sensor failure and large noise. Efficient numerical algorithms are proposed and simulated supporting the theoretical analysis.\",\n \"Title: Local variable selection of nonlinear nonparametric systems by first order expansion.\\nYear: 2018\\n\\nAbstract: Local variable selection by first order expansion for nonlinear nonparametric systems is investigated in the paper. By substantially modifying the algorithms developed in our earlier work (Bai et al., 2014), the previous results have been considerably strengthened under much less restrictive conditions. Firstly, the estimates generated by the modified algorithms are shown to have both the set and parameter convergence with probability one, rather than only the set convergence in probability given in our earlier work. Secondly, several technical assumptions, e.g., the lower and upper bounds on the growth of some random sequences, which practically are uncheckable, have been removed. Thirdly, not only the consistency but also the convergence rate of estimates have been established. Besides, a generalization of the proposed algorithms is also introduced.\",\n \"Title: Variable selection of high-dimensional non-parametric nonlinear systems by derivative averaging to avoid the curse of dimensionality.\\nYear: 2019\\n\\nAbstract: In identification, variable selection for a nonlinear non-parametric and high-dimensional system is the first and often one of the most difficult problems. The issue is the curse of dimensionality. This paper presents a numerically efficient algorithm of variable selection for high-dimensional nonlinear non-parametric systems. It is based on averaging derivatives and relies on one dimensional estimates of the density function and its derivative. Thus, it avoids the curse of dimensionality usually encountered for high-dimensional systems. Theoretical analysis is provided and the conditions are derived for a variable to contribute or not to contribute for a large class of nonlinear systems. Further, convergence results are established.\",\n \"Title: Frequency domain identification of hammerstein models\\nYear: 2003\\n\\nAbstract: In this chapter, we discuss a frequency approach for Hammerstein model identification. The method is based on the fundamental\\n frequency and therefore, no a priori information on the structure of the nonlinearity is required. Moreover, the method is not limited to Hammerstein models whose\\n linear part is a finite-order rational transfer functions, but applies to Hammerstein models with a non-parametric linear\\n part. The method can be easily extended to Wiener models with minor modifications. The chapter is based on an article [1],\\n IEEE Trans. on Automatic Control, Vol. 48, pp.530-542, 2003 with permission from IEEE Intellectual Property Rights Office.\\n All the proofs can be found in [1].\",\n \"Title: Ranking Variables Based on Goodness of Fits in Nonlinear Nonparametric System Identification.\\nYear: 2018\\n\\nAbstract: Identification of a high dimensional nonlinear nonparametric system is costly. On the other hand for many real-world problems, they are sparse in the sense that not all variables contribute or contribute significantly. If these variables that do not contribute or contribute little can be identified and removed prior to system identification, the identification problem is of lower dimension. In this paper, methods to rank variables based on the Goodness of Fits are proposed without full scale identification.\",\n \"Title: Ranking the importance of variables in nonlinear system identification\\nYear: 2019\\n\\nAbstract: In this paper, the importance measure based on Goodness of Fit (GOF) is defined. Based on the measure, ways to rank the importance of variables are proposed. In particular, this paper addresses three questions: (1) What is the importance measure based on Goodness of Fit (GOF)? (2) How to rank the variables based on GOF, prior to actual identification? (3) Given an integer d, can we find theoretically d-variables that collectively have the largest contribution among all d-variable subsets based on GOF? An affirmative answer is provided along with numerical algorithms. The problem of ranking variables prior to actual identification is important in identification of high dimensional nonlinear systems. If variables that contribute little can be identified and removed prior to system identification, then the identification problem is of lower dimension and relatively easy to deal with.\",\n \"Title: Identification Of An Additive Nfir System And Its Applications In Generalized Hammerstein Models\\nYear: 2005\\n\\nAbstract: Identification of a nonlinear additive system is considered. An input signal is designed in such a way that the problem of identification of nonlinear additive systems is reduced to a problem of identification of static nonlinear functions. Then, three approaches are established to estimate the order of the system. The methods exploit the structure of the nonlinear additive model so that their implementations are easy.\",\n \"Title: Persistence of excitation for identifying switched linear systems\\nYear: 2022\\n\\nAbstract: This paper investigates the uniqueness of parameters via persistence of excitation for switched linear systems. The main contribution is a much weaker sufficient condition on the regressors to be persistently exciting that guarantees the uniqueness of the parameter sets and also provides new insights in understanding the relation among different subsystems. It is found that for uniquely determining the parameters of switched linear systems, the needed minimum number of samples derived from our sufficient condition is much smaller than that reported in the literature.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":79,"string":"79"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13558,"string":"13,558"}}},{"rowIdx":13559,"cells":{"question_id":{"kind":"string","value":"Volker Diekert"},"question":{"kind":"string","value":"Generate an abstract for the title \"The Krohn-Rhodes Theorem and Local Divisors\" using the following items: 1. Proof\n2. Krohn-Rhodes theorem\n3. Local divisors\n4. Monoids\n5. Holonomy decomposition"},"answer":{"kind":"string","value":"We give a new proof of the Krohn-Rhodes theorem using local divisors. The proof provides nearly as good a decomposition in terms of size as the holonomy decomposition of Eilenberg, avoids induction on the size of the state set, and works exclusively with monoids with the base case of the induction being that of a group."},"correct_answer":{"kind":"string","value":"We give a new proof of the Krohn-Rhodes theorem using local divisors. The proof provides nearly as good a decomposition in terms of size as the holonomy decomposition of Eilenberg, avoids induction on the size of the state set, and works exclusively with monoids with the base case of the induction being that of a group."},"documents":{"kind":"list like","value":["Title: Approximating Traces\nYear: 1998\n\nAbstract: . In order to give a semantics to concurrent processes we need a model having some good mathematical properties. To this end\n we generalize (infinite) Mazurkiewicz traces by adding some alphabetical information concerning the possible continuations\n of a process. This allows to define an approximation order compatible with the composition. We obtain a prime algebraic and\n coherently complete domain where the compact elements are exactly the finite approximations of processes. The composition\n is shown to be monotone and -continuous. We define a suitable metric which induces the Lawson topology and which yields a compact metric space being therefore\n complete. The finite approximations of processes form a dense and open subset and the composition is (uniformly) continuous.\n A preliminary version of this work appeared in [7].","Title: Regular frequency computations\nYear: 2005\n\nAbstract: An (m, n)-computation of a function f is given by a deterministic Turing machine which on n pairwise different inputs produces n output values where at least m of the n values are in accordance with f. In such a case, we say that the Turing machine computes f with frequency ≥ m/n. The most prominent result for frequency computations is due to Trakhtenbrot: The class of (m, n)-computable functions equals the class of computable functions if and only if 2m n.Via characteristic functions the definition of (m, n)-computability carries over to sets. Here Trakhtenbrot's result reads as: The class of (m, n)-computable sets equals the class of recursive sets if and only if 2m n.The notion of frequency computation can be extended to other models of computation. For resource bounded computations, the behavior is completely different: for e.g., whenever n' - m' n - m, it is known that under any reasonable resource bound there are sets (m', n')-computable, but not (m, n)- computable.However, scaling down to finite automata, the analogue of Trakhtenbrot's result holds again: We show here that the class of languages (m, n)-recognizable by deterministic finite automata equals the class of regular languages if and only if 2m n. This was originally stated by Kinber, but his proof has a flaw, as pointed out by Tantau.Conversely, for 2m ≤ n, the class of languages (m, n)-recognizable by deterministic finite automata is uncountable for a two-letter alphabet. When restricted to a one-letter alphabet, then every (m, n)- recognizable language is regular. This was also shown by Kinber. We give a new and more direct proof for this result.","Title: Quadratic Word Equations\nYear: 1999\n\nAbstract: IntroductionThe existential theory of equations over free monoids is decidable. This majorresult was obtained by Makanin [10], who showed that the satisability ofword equations with constants is decidable. For the background we refer to[14], to the corresponding chapter in the Handbook of Formal Languages(Eds. G. Rozenberg and A. Salomaa), [2], or to the forthcoming [3]. In 1990Schulz [19] showed an important generalization: Makanin's result remainstrue when adding regular constraints. ...","Title: Star-free languages are Church-Rosser congruential\nYear: 2012\n\nAbstract: The class of Church-Rosser congruential languages has been introduced by McNaughton, Narendran, and Otto in 1988. A language L is Church-Rosser congruential (belongs to CRCL), if there is a finite, confluent, and length-reducing semi-Thue system S such that L is a finite union of congruence classes modulo S. To date, it is still open whether every regular language is in CRCL. In this paper, we show that every star-free language is in CRCL. In fact, we prove a stronger statement: for every star-free language L there exists a finite, confluent, and subword-reducing semi-Thue system S such that the total number of congruence classes modulo S is finite and such that L is a union of congruence classes modulo S. The construction turns out to be effective.","Title: Solvability Of Equations In Graph Groups Is Decidable\nYear: 2006\n\nAbstract: We show that the existential theory of free partially commutative monoids with involution is decidable. As a consequence the existential theory of graph groups is also decidable. If the underlying alphabet of generators is fixed, we obtain a PSPACE-completeness result, otherwise (in the uniform setting) our decision procedure is in EXPSPACE. Our proof is a reduction to the main result of [6].","Title: A remark about quadratic trace equations\nYear: 2002\n\nAbstract: We present a simple PSPACE-algorithm for computing the most general solution of a system of quadratic trace equations with involution. We extend the known linear time algorithm for quadratic word equations with length constraints to cope with involutions. Finally, we show that the same linear-time result cannot be expected for trace equations. We obtain an NP-hardness result.","Title: Word problems over traces which are solvable in linear time\nYear: 1989\n\nAbstract: Finite complete replacement systems over free partially commutative monoids have a decidable word problem, and if these systems are weight-reducing, then the word problem is solvable in square time. It is unknown whether this time bound is optimal, but the restriction to semi-Thus systems or vector replacement systems yields linear time. In direction to close this gap, we first give a global condition for trace replacement systems which also ensures a linear time bound: and we prove that this condition is decidable. Then we give a stronger but local condition which leads to the notion of cones and blocks which are certain traces. Trivial examples of cones and blocks are words and vectors, respectively. We show that under a mild additional assumption we may decide the confluence of a noetherian trace replacement system if all left-hand sides are cones or blocks. (In general, confluence is undecidable even for length-reducing systems). Altogether we obtain a sufficient and computable condition such that the word problem of a trace replacement system is solvable in linear time.","Title: Code Problems on Traces\nYear: 1996\n\nAbstract: . The topic of codes in the framework of trace monoids leadsto interesting and challenging decision problems of combinatorial flavour.We give an overview of the current state of some basic questions in thisfield. Among these, we consider the existence problem for strong codings,clique-preserving morphisms and the unique decipherability problem(code problem).1 IntroductionFree partially commutative monoids [7] offer a mathematically sound frameworkfor modelling and analyzing concurrent...","Title: Combinatorial rewriting on traces\nYear: 1990\n\nAbstract: There are two main problems in working with replacement systems over free partially commutative monoids: For finite noetherian systems confluence is undecidable, in general, and the known algorithm to compute irreducible normal forms need time square in the derivation length instead of linear. We first give a decidable and sufficient condition for finite noetherian systems such that confluence becomes decidable. This condition is weaker than the known ones before. Then we give a decidable and sufficient condition such that irreducible normal forms are computable in time linear to the derivation length. Furthermore, we prove that the first condition is implied by the second. We also present a new uniform algorithm for computing normal forms using Zielonka's theory of asynchronous automata.","Title: On Codings of Traces\nYear: 1995\n\nAbstract: Abstract. The paper solves the main open problem of [BFG94]. We show that given two dependence alphabets ( ; D) and (,; D,), it is de-cidable whether there exists a strong coding h : M ( ; D) ! M (,; D,) between the associated trace monoids. In fact, we show that the problem is NP-complete. (A coding is an injective homomorphism, it is strong if independent letters are mapped to independent traces. ) We exhibit an example of trace monoids where a coding between them exists, but no strong coding. The decidability of codings remains open, in general. We have a lower and an upper bound, which show both to be strict. We fur-ther discuss encodings of free products of trace monoids and give almost optimal constructions. In the nal section, we state that the coding property is undecidable in a naturally arising class of homomorphisms.","Title: Recognizable Complex Trace Languages\nYear: 1991\n\nAbstract: A. Mazurkiewicz defined traces in order to modelize non-sequential processes. Complex traces have been recently introduced as a generalization of both traces and infinite words. This paper studies the family of recognizable complex trace languages. It is proved that this family is closed under boolean operations, concatenation, left and right quotients. Then sufficient conditions ensuring the recognizability of the finite and infinite iterations of a recognizable complex trace language are given. The notion of co-iteration is defined and the Kleene-Ochmanski theorem is generalized to complex traces.","Title: Some remarks about stabilizers\nYear: 2009\n\nAbstract: We continue our study of stabilizers of infinite words over finite alphabets, begun in [D. Krieger, On stabilizers of infinite words, Theoret. Comput. Sci. 400 (2008), 169-181]. Let w be an aperiodic infinite word over a finite alphabet, and let Stab(w) be its stabilizer. We show that Stab(w) can be partitioned into the monoid of morphisms that stabilize w by finite fixed points and the ideal of morphisms that stabilize w by iteration. We also settle a conjecture given in the paper mentioned above, by showing that in some cases Stab(w) is infinitely generated. If the aforementioned ideal is nonempty, then it contains either polynomially growing morphisms or exponentially growing morphisms, but not both. Moreover, in the polynomial case, the degree of the polynomial is fixed. We also show how to compute the polynomial degree from the dependency graph of a polynomially growing morphism.","Title: On quadratic word equations\nYear: 1999\n\nAbstract: We investigate the satisfiability problem of word equations where each variable occurs at most twice (quadratic systems). We obtain various new results: The satisfiability problem is NP-hard (even for a single equation). The main result says that once we have fixed the lengths of a possible solution, then we can decide in linear time whether there is a corresponding solution. If the lengths of a minimal solution were at most exponential, then the satisfiability problem of quadratic systems would be NP-complete. (The inclusion in NP follows also from [21]) In the second part we address the problem with regular constraints: The uniform version is PSPACE-complete. Fixing the lengths of a possible solution doesn't make the problem much easier. The non-uniform version remains NP-hard (in contrast to the linear time result above). The uniform version remains PSPACE-complete.","Title: Transitive Orientations, Möbius Functions, and Complete Semi-Thue Systems for Free Partially Commutative Monoids\nYear: 1988\n\nAbstract: Let I\n\nÍ</font\n>\\subseteq\n\nX×X be an independence relation over a finite alphabet X and M=X*/{(ab, ba)|(a, b)teI} the associated free partially commutative monoid. The Möbius function of M is a polynomial in the ring of formal power series Z Í</font\n>\\subseteq\n\nX*×X* which define M. We obtain a one-to-one correspondence between unambiguous Möbius functions, transitive orientations and finite (normalized) complete semi-Thue systems.","Title: Hotz-Isomorphism Theorems in Formal Language Theory\nYear: 1989\n\nAbstract: A language \nL Í</font\n> X*L \\subseteq X*\nis called a language with a Hotz-isomorphism if there is a generating grammar such that its Hotz group is canonically isomorphic to F(X)/L. The main result of the paper states that L is a language with a Hotz-isomorphism if and only if F(X)/L is a finitely presentable group. We also prove an analogous result for Hotz-monoids. Further, we show when the construction of a grammar which allows the Hotz-isomorphism is effective; and we prove various undecidability results concerning this construction.","Title: On Efficient Reduction Algorithms for Some Trace Rewriting Systems\nYear: 1993\n\nAbstract: . We consider some basic problems on the decidability andcomplexity of trace rewriting systems. The new contribution of this paperis an O(n log(n)) algorithm for some computing irreducible normal formsin the case of certain one-rule systems.1 IntroductionThe notes of this paper are based on the Font Romeu Lecture and on an invitedlecture at FCT-93 conference [11] of the second author. In the first part of thepaper we give some basic background about trace rewriting systems. There is...","Title: Language theoretical properties of hairpin formations\nYear: 2012\n\nAbstract: Hairpin formations arise in biochemical processes and play an important role in DNA-computing. We study language theoretical properties of hairpin formations and our new results concern the hairpin completion H\"@k(L\"1,L\"2) of two regular languages L\"1 and L\"2 and the iterated hairpin lengthening HL\"@k^*(L) of any language L. Assume that L\"1 and L\"2 belong to a certain variety of regular languages which satisfies a mild closure property (being closed by a restricted concatenation), then either H\"@k(L\"1,L\"2) is not regular or it belongs to the same variety as L\"1 and L\"2. This result applies, in particular, to the class of first-order definable languages (which is the class of aperiodic or star-free languages) and it applies to the class of first-order definable languages in two variables with predicates","Title: From local to global temporal logics over Mazurkiewicz traces\nYear: 2006\n\nAbstract: We review some results on global and local temporal logic on Mazurkiewicz traces. Our main contribution is to show how to derive the expressive completeness of global temporal logic with respect to first-order logic [V. Diekert, P. Gastin, LTL is expressively complete for Mazurkiewicz traces, J. Comput. System Sci. 64 (2002) 396-418] from the similar result on local temporal logic [V. Diekert, P. Gastin, Pure future local temporal logics are expressively complete for Mazurkiewicz traces, in: M. Farach-Colton (Ed.), Proc. LATIN'04, Lecture Notes in Computer Science, Vol. 2976, Springer, Berlin, 2004, pp. 232-241, Full version available as Research Report LSV-05-22, Laboratoire Spécification et Vérification, ENS Cachan, France].","Title: An Expressively Complete Temporal Logic without Past Tense Operators for Mazurkiewicz Traces\nYear: 1999\n\nAbstract: Mazurkiewicz traces are a widely accepted model of concurrent systems. We introduce a linear time temporal logic LTLf which has the same expressive power as the first order theory FO(","Title: Slp Compression For Solutions Of Equations With Constraints In Free And Hyperbolic Groups\nYear: 2013\n\nAbstract: The paper is a part of an ongoing program which aims to show that the problem of satisfiability of a system of equations in a free group (hyperbolic or even toral relatively hyperbolic group) is NP-complete. For that, we study compression of solutions with straight-line programs (SLPs) as suggested originally by Plandowski and Rytter in the context of a single word equation. We review some basic results on SLPs and give full proofs in order to keep this fundamental part of the program self-contained. Next we study systems of equations with constraints in free groups and more generally in free products of abelian groups. We show how to compress minimal solutions with extended Parikh-constraints. This type of constraints allows to express semi-linear conditions as e.g. alphabetic information. The result relies on some combinatorial analysis and has not been shown elsewhere. We show similar compression results for Boolean formula of equations over a torsion-free delta-hyperbolic group. The situation is much more delicate than in free groups. As byproduct we improve the estimation of the \"capacity\" constant used by Rips and Sela in their paper \"Canonical representatives and equations in hyperbolic groups\" from a double-exponential bound in delta to some single-exponential bound. The final section shows compression results for toral relatively hyperbolic groups using the work of Dahmani: We show that given a system of equations over a fixed toral relatively hyperbolic group, for every solution of length N there is an SLP for another solution such that the size of the SLP is bounded by some polynomial p( s + log N) where s is the size of the system.","Title: Confluence of One-Rule Thue Systems\nYear: 1990\n\nAbstract: Without Abstract","Title: Logspace computations in graph groups and coxeter groups\nYear: 2012\n\nAbstract: Computing normal forms in groups (or monoids) is in general harder than solving the word problem (equality testing). However, normal form computation has a much wider range of applications. It is therefore interesting to investigate the complexity of computing normal forms for important classes of groups. We show that shortlex normal forms in graph groups and in right-angled Coxeter groups can be computed in logspace. Graph groups are also known as free partially commutative groups or as right-angled Artin groups in the literature. (Artin groups can be realized as subgroups of Coxeter groups.) Graph groups arise in many areas and have a close connection to concurrency theory. The connection is used here. Indeed, for our result we use a representation of group elements by Mazurkiewicz traces. These are directed acyclic node-labelled graphs (i.e. pomsets). They form an algebraic model to describe runs of concurrent systems. Concurrent systems which are deterministic and co-deterministic can be studied via inverse monoids. As an application of our results we show that the word problem for free partially commutative inverse monoids is in logspace. This result generalizes a result of Ondrusch and the third author on free inverse monoids. All Coxeter groups are linear, so the word problem can be solved in logspace, but it is open (in the non-right-angled case) whether shortlex normal forms can be computed in logspace, or, less demanding, whether they can be computed efficiently in parallel. We show that for all Coxeter groups the set of letters occurring in the shortlex normal form of an element can be computed in logspace.","Title: Rational and recognizable complex trace languages\nYear: 1995\n\nAbstract: Mazurkiewicz defined traces as an algebraic model of finite concurrent processes. In order to model non-terminating processes a good notion of infinite traces was needed, which finally led to the notion of complex traces. For complex traces an associative concatenation and an ω-iteration are defined. This paper defines and investigates rational and recognizable complex trace languages. We prove various closure results such as the closure under boolean operations (for recognizable languages), concatenation, and left and right quotients by recognizable sets. Then we study sufficient conditions ensuring the recognizability of the finite and infinite iterations of complex trace languages. We introduce a generalization of the notion of concurrent iteration which leads to the main result of the paper: the generalization of Kleene′s and Ochmański′s theorems to complex trace languages.","Title: Existential and Positive Theories of Equations in Graph Products\nYear: 2002\n\nAbstract: We prove that the existential theory of equations with normalized rational constraints in a fixed graph product of finite\n monoids, free monoids, and free groups is PSPACE-complete. Under certain restrictions this result also holds if the graph\n product is part of the input. As the second main result we prove that the positive theory of equations with recognizable constraints\n in graph products of finite and free groups is decidable.","Title: A structural property of regular frequency computations\nYear: 2003\n\nAbstract: The notion of an (m,n)-computation was already introduced in 1960 by Rose and further investigated by Trakhtenbrot in 1963. It has been extended to finite automata by Kinber in 1976 and he has shown an analogue of Trakhtenbrot's result: The class of languages (m,n)-recognizable by deterministic finite automata is equal to the class of regular languages if and only if 2m n. Furthermore, for a unary alphabet, the class of (m,n)-recognizable languages coincides with the class of regular languages for all m and n. In this paper, we will present the first structural property of (m,n)-recognizable languages which is valid for all 1 ≤ m ≤ n and for all alphabets. Kinber's result for unary alphabets becomes a corollary. This property is also used to show that certain non-unary languages are not (m,n)-regular and that the class of all (m,n)-recognizable languages is not closed under the reversal operation. However, this class forms a Boolean algebra.","Title: Mo¨bius functions and confluent semi-commutations\nYear: 1993\n\nAbstract: Dickert (1988) settled a well-known conjecture on unambiguous liftings to words of Möbius functions for free partially commutative monoids. He established thereby a new bridge from the theory of formal power series to the theory of complete semi-Thue systems. As an open question, it was asked whether this approach has a generalization to a relative situation where we do not lift to the free (word-) case but only to some level of less commutation. This question has been reconsidered by König (1992) and a conjecture has been formulated there. Our results solve the open problem and settle this conjecture (Remark 4.8). We show that there is a canonical one-to-one correspondence between unambiguous Möbius functions and confluent semi-commutation systems (Theorem 5.1). This identification is due to some graph-theoretic characterizationobtained by Diekert (1991) and which allows one to apply some interesting complexity results (Section 6). Our results can be viewed as a contribution to the combinatorial theory of Möbius functions and to the theory of rewriting on traces.","Title: Removing epsilon-Transitions in Timed Automata\nYear: 1997\n\nAbstract: . Timed automata are among the most widely studied modelsfor real-time systems. Silent transitions, i.e., "-transitions, have alreadybeen proposed in the original paper on timed automata by Alur and Dill[2]. In [7] it is shown that "-transitions can be removed, if they do notreset clocks; moreover "-transitions strictly increase the power of timedautomata, if there is a self-loop containing "-transitions which reset someclocks. The authors of [7] left open the problem about the power of...","Title: On the Hairpin Completion of Regular Languages\nYear: 2009\n\nAbstract: The hairpin completion is a natural operation of formal languages which has been inspired by molecular phenomena in biology and by DNA-computing. The hairpin completion of a regular language is linear context-free and we consider the problem to decide whether the hairpin completion remains regular. This problem has been open since the first formal definition of the operation. In this paper we present a positive solution to this problem. Our solution yields more than decidability because we present a polynomial time procedure. The degree of the polynomial is however unexpectedly high, since in our approach it is more than n 14. Nevertheless, the polynomial time result is surprising, because even if the hairpin completion $\\mathcal{H}$ of a regular language L is regular, there can be an exponential gap between the size of a minimal DFA for L and the size of a smallest NFA for $\\mathcal{H}$.","Title: On confluent semi-commutations: decidability and complexity results\nYear: 1991\n\nAbstract: The subject of this paper is the confluence of finite semi-commutation systems. Confluence of such systems is proved to be decidable property. Existence of a finite complete presentation of a trace monoid using rules only from another given trace monoid is proved to be reducible to the existence of a confluent semi-commutation system. Complexity results related to the preceding problems are proved: deciding the existence of finite complete presentations is Σ P 2 -complete, whereas deciding confluence of semi-commutation systems is Co-NP-complete. Additionally, an open problem about trace synchronizations is solved: The local checking property is Co-NP-complete.","Title: Group Extensions Over Infinite Words\nYear: 2012\n\nAbstract: Non-Archimedean words have been introduced as a new type of infinite words which can be investigated through classical methods in combinatorics on words due to a length function. The length function, however, takes values in the additive group of polynomials Z[t] (and not, as traditionally, in N), which yields various new properties. Non-Archimedean words allow to solve a number of interesting algorithmic problems in geometric and algorithmic group theory. There is also a connection to logic and the first-order theory in free groups (Tarski Problems).In the present paper we provide a general method to use infinite words over a discretely ordered abelian group as a tool to investigate certain group extensions for an arbitrary group G. The central object is a group E(A, G) which is defined in terms of a non-terminating, but confluent rewriting system. The group G as well as some natural HNN-extensions of G embed into E(A, G) (and still \"behave like\" G), which makes it interesting to study its algorithmic properties. In order to show that every group G embeds into E(A, G) we combine methods from combinatorics on words, string rewriting and group theory.","Title: On the Knuth–Bendix completion for concurrent processes\nYear: 1987\n\nAbstract: We consider critical pairs for replacement systems over free partially commutative monoids. This is done in order to apply the Knuth-Bendix completion procedure to concurrent processes. We will see that there are one-rule systems which have no finite set of critical pairs. Therefore we develop a sufficient (and computable) condition such that finite trace replacement systems have a finite set of critical pairs. This condition is always satisfied in the purely free case of semi-Thue systems or purely commutative case of vector replacement systems. In fact, it generalizes (and unifies) both cases. We will give examples of how one can use this generalization.","Title: Algorithmic Problems On Inverse Monoids Over Virtually Free Groups\nYear: 2008\n\nAbstract: Let G be a finitely generated virtually-free group. We consider the Birget-Rhodes expansion of G, which yields an inverse monoid and which is denoted by IM(G) in the following. We show that for a finite idempotent presentation P, the word problem of a quotient monoid IM(G)/P can be solved in linear time on a RAM. The uniform word problem, where G and the presentation P are also part of the input, is EXPTIME-complete. With IM(G)/P we associate a relational structure, which contains for every rational subset L of IM(G)/P a binary relation, consisting of all pairs (x, y) such that y can be obtained from x by right multiplication with an element from L. We prove that the first-order theory of this structure is decidable. This result implies that the emptiness problem for Boolean combinations of rational subsets of IM(G)/P is decidable, which, in turn implies the decidability of the submonoid membership problem of IM(G)/P. These results were known previously for free groups, only. Moreover, we provide a new algorithmic approach for these problems, which seems to be of independent interest even for free groups. We also show that one cannot expect decidability results in much larger frameworks than virtually-free groups because the subgroup membership problem of a subgroup H in an arbitrary group G can be reduced to a word problem of some IM(G)/P, where P depends only on H. A consequence is that there is a hyperbolic group G and a finite idempotent presentation P such that the word problem is undecidable for some finitely generated submonoid of IM(G)/P. In particular, the word problem of IM(G)/P is undecidable.","Title: Conjugacy in Baumslag's Group, Generic Case Complexity, and Division in Power Circuiats.\nYear: 2013\n\nAbstract: The conjugacy problem asks whether two words over generators of a fixed group G are conjugated, i.e., it is the problem to decide on input words x, y whether there exists z such that $$zx z^{-1} =y$$zxz-1=y in G. The conjugacy problem is more difficult than the word problem, in general. We investigate the conjugacy problem for two prominent groups: the Baumslag---Solitar group $$\\\\mathbf{{BS}}_{1,2}$$BS1,2 and the Baumslag group $${\\\\mathbf{{G}}}_{1,2}$$G1,2 (also known as Baumslag---Gersten group). The conjugacy problem in $${\\\\mathbf{{BS}}}_{1,2}$$BS1,2 is complete for the circuit class $$\\\\mathsf {TC}^0$$TC0. To the best of our knowledge $${\\\\mathbf{{BS}}}_{1,2}$$BS1,2 is the first natural infinite non-commutative group where such a precise and low complexity is shown. The Baumslag group $${\\\\mathbf{{G}}}_{1,2}$$G1,2 is an HNN-extension of $${\\\\mathbf{{BS}}}_{1,2}$$BS1,2. Hence, decidability of the conjugacy problem in $$\\\\mathbf{{G}}_{1,2}$$G1,2 outside the so-called \\\"black hole\\\" follows from Borovik et al. (Int J Algebra Comput 17(5/6):963---997, 2007). Decidability everywhere is due to Beese. Moreover, he showed exponential time for the set of elements which cannot be conjugated into $$\\\\mathbf{{BS}}_{1,2}$$BS1,2 (Beese 2012). Here we improve Beese's result in two directions by showing that the conjugacy problem in $${\\\\mathbf{{G}}}_{1,2}$$G1,2 can be solved in polynomial time in a strongly generic setting. This means that essentially for all inputs, conjugacy in $${\\\\mathbf{{G}}}_{1,2}$$G1,2 can be decided efficiently. In contrast, we show that under a plausible assumption the average case complexity of the same problem is non-elementary. Moreover, we provide a lower bound for the conjugacy problem in $${\\\\mathbf{{G}}}_{1,2}$$G1,2 by reducing the divisibility problem in power circuits to the conjugacy problem in $${\\\\mathbf{{G}}}_{1,2}$$G1,2. The complexity of the divisibility problem in power circuits is an open and interesting problem in integer arithmetic. We conjecture that it cannot be solved in elementary time because we can show that it cannot be solved in elementary time by calculating modulo values in power circuits.","Title: Rewriting, Möbius Functions and Semi-Commutations\nYear: 1993\n\nAbstract: Without Abstract","Title: Regular languages are church-rosser congruential\nYear: 2012\n\nAbstract: This paper proves a long standing conjecture in formal language theory. It shows that all regular languages are Church-Rosser congruential. The class of Church-Rosser congruential languages was introduced by McNaughton, Narendran, and Otto in 1988. A language L is Church-Rosser congruential if there exists a finite, confluent, and length-reducing semi-Thue system S such that L is a finite union of congruence classes modulo S. It was known that there are deterministic linear context-free languages which are not Church-Rosser congruential, but on the other hand it was strongly believed that all regular languages are of this form. This paper solves the conjecture affirmatively by actually proving a more general result.","Title: Cyclic Rewriting And Conjugacy Problems\nYear: 2012\n\nAbstract: Cyclic words are equivalence classes of cyclic permutations of ordinary words. When a group is given by a rewriting relation, a rewriting system on cyclic words is induced, which is used to construct algorithms to find minimal length elements of conjugacy classes in the group. These techniques are applied to the universal groups of Stallings pre-groups and in particular to free products with amalgamation, HNN- extensions and virtually free groups, to yield simple and intuitive algorithms and proofs of conjugacy criteria.","Title: QuickHeapsort: Modifications and Improved Analysis.\nYear: 2012\n\nAbstract: QuickHeapsort is a combination of Quicksort and Heapsort. We show that the expected number of comparisons for QuickHeapsort is always better than for Quicksort if a usual median-of-constant strategy is used for choosing pivot elements. In order to obtain the result we present a new analysis for QuickHeapsort splitting it into the analysis of the partition-phases and the analysis of the heap-phases. This enables us to consider samples of non-constant size for the pivot selection and leads to better theoretical bounds for the algorithm. Furthermore, we introduce some modifications of QuickHeapsort. We show that for every input the expected number of comparisons is at most nlog2nź0.03n+o(n)$n\\\\log _{2}n - 0.03n + o(n)$ for the in-place variant. If we allow n extra bits, then we can lower the bound to nlog2nź0.997n+o(n)$ n\\\\log _{2} n -0.997 n+ o (n)$. Thus, spending n extra bits we can save more that 0.96n comparisons if n is large enough. Both estimates improve the previously known results. Moreover, our non-in-place variant does essentially use the same number of comparisons as index based Heapsort variants and Relaxed-Weak-Heapsort which use nlog2nź0.9n+o(n)$ n\\\\log _{2}n -0.9 n+ o (n)$ comparisons in the worst case. However, index based Heapsort variants and Relaxed-Weak-Heapsort require ź(nlogn)${\\\\Theta }(n\\\\log n)$ extra bits whereas we need n bits only. Our theoretical results are upper bounds and valid for every input. Our computer experiments show that the gap between our bounds and the actual values on random inputs is small. Moreover, the computer experiments establish QuickHeapsort as competitive with Quicksort in terms of running time.","Title: One-Rule Trace-Rewriting Systems and Confluence\nYear: 1992\n\nAbstract: Confluence is an undecidable property even for finite noetherian trace-rewriting systems. Here we investigate this property\n for the special case of trace-rewriting systems with a single rule. For certain classes of one-rule trace-rewriting systems\n R we present syntactic characterizations that are necessary and sufficient for R to be confluent. Based on these characterizations\n confluence is easily decidable for these systems. In fact, each system satisfying one of these characterizations is strongly\n confluent. This raises the question of whether every confluent onerule trace-rewriting system is strongly confluent. We close\n this paper with an example that answers this question in the negative.","Title: Context-Free Groups And Their Structure Trees\nYear: 2013\n\nAbstract: Let G be a connected, locally finite graph of finite tree width and G be a group acting on it with finitely many orbits and finite node stabilizers. We provide an elementary and direct construction of a tree T on which G acts with finitely many orbits and finite vertex stabilizers. Moreover, the tree is defined directly in terms of the structure tree of optimally nested cuts of G. Once the tree is constructed, Bass-Serre theory yields that G is virtually free. This approach simplifies the existing proofs for the fundamental result of Muller and Schupp that characterizes context-free groups as finitely generated (f.g.) virtually free groups. Our construction avoids the explicit use of Stallings' structure theorem and it is self-contained. We also give a simplified proof for an important consequence of the structure tree theory by Dicks and Dunwoody which has been stated by Thomassen and Woess. It says that a f.g. group is accessible if and only if its Cayley graph is accessible.","Title: On the Synchronization of Traces.\nYear: 1989\n\nAbstract: Trace theory has been developed to describe the behavior of concurrent systems. Synchronization of traces is of special interest for a modular approach. We characterize those trace monoids for which synchronization can be described locally.","Title: Deterministic asynchronous automata for infinite traces\nYear: 1993\n\nAbstract: This paper shows the equivalence between the family of recognizable languages over infinite traces and the family of languages which are recognized by deterministic asynchronous cellular Muller automata. We thus give a proper generalization of McNaughton's Theorem from infinite words to infinite traces. Thereby we solve one of the main open problems in this field. As a special case we obtain that every closed (w.r.t. the independence relation) word language is accepted by someI-diamond deterministic Muller automaton.","Title: It Is Nl-Complete To Decide Whether A Hairpin Completion Of Regular Languages Is Regular\nYear: 2011\n\nAbstract: The hairpin completion is an operation on formal languages which is inspired by the hairpin formation in biochemistry. Hairpin formations occur naturally within DNA-computing. It has been known that the hairpin completion of a regular language is linear context-free, but not regular, in general. However, for some time it is was open whether the regularity of the hairpin completion of a regular language is decidable. In 2009 this decidability problem has been solved positively in [5] by providing a polynomial time algorithm. In this paper we improve the complexity bound by showing that the decision problem is actually NL-complete. This complexity bound holds for both, the one-sided and the two-sided hairpin completions.","Title: Logspace computations in graph products\nYear: 2013\n\nAbstract: We consider three important and well-studied algorithmic problems in group theory: the word, geodesic, and conjugacy problem. We show transfer results from individual groups to graph products. We concentrate on logspace complexity because the challenge is actually in small complexity classes, only. The most difficult transfer result is for the conjugacy problem. We have a general result for graph products, but even in the special case of a graph group the result is new. Graph groups are closely linked to the theory of Mazurkiewicz traces which form an algebraic model for concurrent processes. Our proofs are combinatorial and based on well-known concepts in trace theory. We also use rewriting techniques over traces. For the group-theoretical part we apply Bass-Serre theory. But as we need explicit formulae and as we design concrete algorithms all our group-theoretical calculations are completely explicit and accessible to non-specialists.","Title: Local Checking of Trace Synchroniziability\nYear: 1988\n\nAbstract: Trace theory has been developed to describe the behaviour of concurrent systems. For a modular approach synchronization of traces is of special interest. We characterize those trace monoids for which synchronization can be described locally.","Title: On the concatenation of infinite traces\nYear: 1993\n\nAbstract: There is a straightforward generalization of traces to infinite traces as dependence graphs where every vertex has finitely many predecessors, or what is the same, as a backward closed and directed set of traces with respect to prefix ordering. However, this direct approach has a drawback since it does not allow one to describe some basic phenomena which are related to concatenation. We solve this problem by adding to an infinite trace a second component. This second component is a finite alphabetic information which is called the alphabet at infinity . We obtain a compact and complete ultra-metric space where the concatenation is uniformly continuous and where the set of finite traces is an open, discrete, and dense subset. Our objects arise in a natural way from the consideration of dependence graphs where the induced partial order is well-founded. Such a graph splits into a so-called real part and a transfinite part . From the transfinite part only its alphabet is of importance. Our approach is a nontrivial generalization of the well-known construction for words and yields a convenient semantics for infinite concurrent processes.","Title: Investigations on Hotz groups for arbitrary grammars\nYear: 1986\n\nAbstract: The Hotz group H(G) and the Hotz monoid M(G) of an arbitrary grammar G=(V, X, P, S) are defined by H(G)=F(V?X)/P and M(G) =(V?X)*/P respectively. A language L?X* is called a language with Hotz isomorphism if there exists a grammar G with L=L(G) such that the natural homomorphism F(X)/L?H(G) is an isomorphism. The main result of this paper states that homomorphic images of sentential form languages are languages with Hotz isomorphism. This is a generalization of a result of Frougny, Sakarovitch, and Valkema on context-free languages.","Title: On Computing Geodesics In Baumslag-Solitar Groups\nYear: 2011\n\nAbstract: We introduce the peak normal form for elements of the Baumslag-Solitar groups BS(p, q). This normal form is very close to the length-lexicographical normal form, but more symmetric. Both normal forms are geodesic. This means the normal form of an element u(-1)v yields the shortest path between u and v in the Cayley graph. For horocyclic elements the peak normal form and the length-lexicographical normal form coincide. The main result of this paper is that we can compute the peak normal form in polynomial time if p divides q. As consequence we can compute geodesic lengths in this case. In particular, this gives a partial answer to Question 1 in [4].For arbitrary p and q it is possible to compute the peak normal form (length-lexicographical normal form resp.) also the for elements in the horocyclic subgroup and, more generally, for elements which we call hills. This approach leads to a linear time reduction of the problem of computing geodesics to the problem of computing geodesics for Britton-reduced words where the t-sequence starts with t(-1) and ends with t.","Title: A Survey On Small Fragments Of First-Order Logic Over Finite Words\nYear: 2008\n\nAbstract: We consider fragments of first-order logic over finite words. In particular, we deal with first-order logic with a restricted number of variables and with the lower levels of the alternation hierarchy. We use the algebraic approach to show decidability of expressibility within these fragments. As a byproduct, we survey several characterizations of the respective fragments. We give complete proofs for all characterizations and we provide all necessary background. Some of the proofs seem to be new and simpler than those which can be found elsewhere. We also give a proof of Simon's theorem on factorization forests restricted to aperiodic monoids because this is simpler and sufficient for our purpose.","Title: Safety and Liveness Properties for Real Traces and a Direct Translation from LTL to Monoids\nYear: 2002\n\nAbstract: For infinite words there are well-known characterizations of safety and liveness properties. We extend these results to real Mazurkiewicz traces. This is possible due to a result, which has been established recently: Every first-order definable real trace language is definable in linear temporal logic using future tense operators, only. We show that the canonical choice for a topological characterization of safety and liveness properties is given by the Scott topology. In this paper we use an algebraic approach where we work with aperiodic monoids. Therefore we also give a direct translation from temporal logic to aperiodic monoids which is of independent interest.","Title: On distributed monitoring of asynchronous systems\nYear: 2012\n\nAbstract: Distributed systems are notoriously difficult to understand and analyze in order to assert their correction w.r.t. given properties. They often exhibit a huge number of different behaviors, as soon as the active entities (peers, agents, processes, etc) behave in an asynchronous manner. Already the modelization of such systems is a non-trivial task, let alone their formal verification. The purpose of this paper is to discuss the problem of distributed monitoring on a simple model of finite-state distributed automata based on shared actions, called asynchronous automata. Monitoring is a question related to runtime verification: assume that we have to check a property $L$ against an unknown or very complex system $A$, so that classical static analysis is not possible. Therefore instead of model-checking a monitor is used, that checks the property on the underlying system at runtime. We are interested here in monitoring distributed systems modeled as asynchronous automata. It is natural to require that monitors should be of the same kind as the underlying system, so we consider here distributed monitoring. A distributed monitor does not have a global view of the system, therefore we propose the notion of locally monitorable trace language. Our main result shows that if the distributed alphabet of actions is connected and if $L$ is a set of infinite traces such that both $L$ and its complement $L^c$ are countable unions of locally safety languages, then $L$ is locally monitorable. We also show that over infinite traces, recognizable countable unions of locally safety languages are precisely the complements of deterministic languages.","Title: A Note On The Existential Theory Of Equations In Plain Groups\nYear: 2002\n\nAbstract: Based on a PSPACE-completeness result for free monoids with involution [4] it is shown that the existential theory of equations with rational constraints in plain groups is PSPACE-complete. As a corollary this settles a question from [16].","Title: A Domain for Concurrent Termination: A Generalization of Mazurkiewicz Traces (Extended Abstract)\nYear: 1995\n\nAbstract: This paper generalizes the concept of Mazurkiewicz traces to a description of a concurrent process, where a known prefix is given as a trace in a first component and a second alphabetic component yields some information about future actions. This allows to define a good semantic domain where the concatenation is continuous with respect to the Scott- and to the Lawson topology. For this we define the notion of - and of -trace. We show various mathematical results proving thereby the soundness of our approach. Our theory is a proper generalization of the theory of finite and infinite words (with explicit termination) and of the theory of finite and infinite (real and complex) traces. We make use of trace theory, domain theory, and topology.","Title: LTL is expressively complete for Mazurkiewicz traces\nYear: 2000\n\nAbstract: A long standing open problem in the theory of (Mazurkiewicz) traces has been the question whether LTL (linear temporal logic) is expressively complete with respect to the first order theory. We solve this problem positively for finite and infinite traces and for the simplest temporal logic, which is based only on next and until modalities. Similar results were established previously, but they were all weaker, since they used additional past or future modalities. Another feature of our work is that our proof is direct and does not use any reduction to the word case.","Title: Solving Trace Equations Using Lexicographical Normal Forms\nYear: 1997\n\nAbstract: . Very recently, the second author showed that the questionwhether an equation over a trace monoid has a solution or not is decidable[11,12]. In the original proof this question is reduced to the solvabilityof word equations with constraints, by induction on the size ofthe commutation relation. In the present paper we give another proof ofthis result using lexicographical normal forms. Our method is a directreduction of a trace equation system to a word equation system withregular...","Title: Word Equations Over Graph Products\nYear: 2008\n\nAbstract: For monoids that satisfy a weak cancellation condition, it is shown that the decidability of the existential theory of word equations is preserved under graph products. Furthermore, it is shown that the positive theory of a graph product of groups can be reduced to the positive theories of those factors, which commute with all other factors, and the existential theories of the remaining factors. Both results also include suitable constraints for the variables. Larger classes of constraints lead in many cases to undecidability results.","Title: Local temporal logic is expressively complete for cograph dependence alphabets\nYear: 2001\n\nAbstract: Recently, local logics for Mazurkiewicz traces are of increas- ing interest. This is mainly due to the fact that the satisabilit y problem has the same complexity as in the word case. If we focus on a purely lo- cal interpretation of formulae at vertices (or events) of a trace, then the satisabilit y problem of linear temporal logics over traces turns out to be PSPACE{complete. But now the dicult problem is to obtain expressive completeness results with respect to rst order logic. The main result of the paper shows such an expressive completeness result, if the underlying dependence alphabet is a cograph, i.e., if all traces are series parallel posets. Moreover, we show that this is the best we can expect in our setting: If the dependence alphabet is not a cograph, then we cannot express all rst order properties. Keywords Temporal logics, Mazurkiewicz traces, concurrency","Title: Efficient algorithms for highly compressed data: The Word Problem in Higman's group is in P.\nYear: 2012\n\nAbstract: Power circuits are data structures which support efficient algorithms for highly compressed integers. Using this new data structure it has been shown recently by Myasnikov, Ushakov and Won that the Word Problem of the one-relator Baumslag group is is decidable in polynomial time. Before that the best known upper bound was non-elementary. In the present paper we provide new results for power circuits and we give new applications in algorithmic group theory: 1. We define a modified reduction procedure on power circuits which runs in quadratic time thereby improving the known cubic time complexity. 2. We improve the complexity of the Word Problem for the Baumslag group to cubic time thereby providing the first practical algorithm for that problem. 3. The Word Problem of Higman's group is decidable in polynomial time. It is due to the last result that we were forced to advance the theory of power circuits.","Title: Solvability of Equations in Free Partially Commutative Groups Is Decidable\nYear: 2001\n\nAbstract: Trace monoids are well-studied objects in computer science where they serve as a basic algebraic tool for analyzing concurrent systems. The question whether the existential theory of trace equations is decidable has been solved positively in 1996. Free partially commutative groups (graph groups) generalize trace monoids in the sense that every element has an inverse. In this paper we show that the existential theory of equations over graph groups is decidable, too. Our decision procedure is non-elementary, but if a certain graph theoretical parameter is viewed as a constant, then we obtain a PSPACE-completeness result. Restricting ourselves to trace monoids we still obtain a better complexity result, as it was known previously.","Title: On First-Order Fragments for Mazurkiewicz Traces\nYear: 2007\n\nAbstract: Mazurkiewicz traces form a model for concurrency. Temporal logic and first-order logic are important tools in order to deal with the abstract behavior of such systems. Since typical properties can be described by rather simple logical formulas one is interested in logical fragments. One focus of this paper is unary temporal logic and first-order logic in two variables. Over words, this corresponds to the variety of finite monoids calledDA. However, overMazurkiewicz traces it is crucial whether traces are given as dependence graphs or as partial orders (over words these notions coincide). The main technical contribution is a generalization of important characterizations of DA from words to dependence graphs, whereas the use of partial orders leads to strictly larger classes. As a consequence we can decide whether a first-order formula over dependence graphs is equivalent to a first-order formula in two variables. The corresponding result for partial orders is not known. This difference between dependence graphs and partial orders also affects the complexity of the satisfiability problems for the fragments under consideration: for first-order formulas in two variables we prove an NEXPTIME upper bound, whereas the corresponding problem for partial orders leads to EXPSPACE. Furthermore, we give several separation results for the alternation hierarchy for first-order logic. It turns out that even for those levels at which one can express the partial order relation in terms of dependence graphs, the fragments over partial orders have more expressive power.","Title: Trace Rewriting: Computing Normal Forms in Time O(n log n)\nYear: 1996\n\nAbstract: . We develop an O(n log n) algorithm for computing normalforms in the case of finite weight-reducing trace rewriting systems withconnected left-hand sides. The time complexity of previously known algorithmssolving this problem has been square time in the worst-case.Keywords: Mazurkiewicz traces, rewriting modulo a partial commutation.IntroductionFree partially commutative monoids (trace monoids) have been introduced byCartier and Foata [6] in order to solve some combinatorial problems...","Title: On “inherently context-sensitive” languages —an application of complexity cores\nYear: 1991\n\nAbstract: The concept of an ''inherently context-sensitive language'' is introduced. It is shown that every context-sensitive language that is not in the Boolean closure of the context-free languages has a subset that is inherently context-sensitive.","Title: Solving word problems in group extensions over infinite words\nYear: 2011\n\nAbstract: Non-Archimedean words have been introduced as a new type of infinite words which can be investigated through classical methods in combinatorics on words due to a length function. The length function, however, takes values in the additive group of polynomials Z[t] (and not, as traditionally, in N), which yields various new properties. Non-Archimedean words allow to solve a number of algorithmic problems in geometric and algorithmic group theory. There is a connection to the first-order theory in free groups (Tarski Problems), too. In the present paper we provide a general method to use infinite words over a discretely ordered abelian group as a tool to investigate certain group extensions for an arbitrary group G. The central object is a group E(A, G) which is defined in terms of a non-terminating, but confluent rewriting system. The group G as well as some natural HNN-extensions of G embed into E(A, G) (and still \"behave like\" G), which makes it interesting to study its algorithmic properties. The main result characterizes when the Word Problem (WP) is decidable in all finitely generated subgroups of E(A, G). We show that this property holds if and only if the Cyclic Membership Problem \"u ∈ 〈v〉?\" is decidable for all v ∈ G. Our methods combine combinatorics on words, string rewriting, and group theory.","Title: Fragments of First-Order Logic over Infinite Words\nYear: 2009\n\nAbstract: We give topological and algebraic characterizations as well as language theoretic descriptions of the following subclasses of first-order logic FO[ω-languages: Σ2, FO2, FO2∩Σ2, and Δ2 (and by duality Π2 and FO2∩Π2). These descriptions extend the respective results for finite words. In particular, we relate the above fragments to language classes of certain (unambiguous) polynomials. An immediate consequence is the decidability of the membership problem of these classes, but this was shown before by Wilke (Classifying Discrete Temporal Properties. Habilitationsschrift, Universität Kiel, April 1998) and Bojańczyk (Lecture Notes in Computer Science, vol. 4962, pp. 172–185, 2008) and is therefore not our main focus. The paper is about the interplay of algebraic, topological, and language theoretic properties.","Title: Complexity Results And The Growths Of Hairpin Completions Of Regular Languages (Extended Abstract)\nYear: 2010\n\nAbstract: The hairpin completion is a natural operation on formal languages which has been inspired by molecular phenomena in biology and by DNA-computing. In 2009 we presented in [6] a (polynomial time) decision algorithm to decide regularity of the hairpin completion. In this paper we provide four new results: 1.) We show that the decision problem is NL-complete. 2.) There is a polynomial time decision algorithm which runs in time O(n(8)), this improves [6], which provided O(n(20)). 3.) For the one-sided case (which is closer to DNA computing) the time is O(n(2)), only. 4.) The hairpin completion is unambiguous linear context-free. This result allows to compute the growth (generating function) of the hairpin completion and to compare it with the growth of the underlying regular language.","Title: Characterization of the expressive power of silent transitions in timed automata\nYear: 1998\n\nAbstract: Timed automata are among the most widely studied models for real-time systems. Silent transitions, i.e., ε-transitions, have already been proposed in the original paper on timed automata by Alur and Dill [3]. We show that the class TL ε of timed languages recognized by automata with ε-transitions, is more robust and more expressive than the corresponding class TL without ε-transitions. We then focus on ε-transitions without reset, i.e. ε-transitions which do not reset clocks. We propose an algorithm to construct, given a timed automaton, an equivalent one without such transitions. This algorithm is in two steps, it first suppresses the cycles of ε-transitions without reset and then the remaining ones. Then, we prove that a timed automaton such that no ε-transition which resets clocks lies on any directed cycle, can be effectively transformed into a timed automaton without εtransitions. Interestingly, this main result holds under the assumption of non-Zenoness and it is false otherwise. To complete the picture, we exhibit a simple timed automaton with an ε-transition, which resets some clock, on a cycle and which is not equivalent to any ε-free timed automaton. To show this, we develop a promising new technique based on the notion of precise action. This paper presents a synthesis of the two conference communications [9] and [13].","Title: Deciding regularity of hairpin completions of regular languages in polynomial time\nYear: 2011\n\nAbstract: Hairpin completion is an operation on formal languages that has been inspired by hairpin formation in DNA biochemistry and by DNA computing. In this paper we investigate the one- and two-sided hairpin completion of regular languages. We solve an open problem from the literature by showing that the regularity problem for hairpin completions is decidable. Actually, we show that the problem is decidable in polynomial time if the input is specified by DFAs. Furthermore, we prove that the hairpin completion of regular languages is an unambiguous linear context-free language. Beforehand, it was known only that it is linear context-free. Unambiguity is a strong additional property because it allows to compute the growth function or the topological entropy. In particular, we can compare the growth of the hairpin completion with the growth of the defining regular languages. We show that the growth of the hairpin completion is exponential if and only if the growth of the underlying languages is exponential. Even if both growth functions are exponential, they can be as far apart as 2^@Q^(^n^) for the hairpin completion and 2^@Q^(^n^) for the defining regular languages. However, if the hairpin completion is still regular, then the hairpin completion and its underlying language have essentially the same growth and same topological entropy.","Title: Solving word equations modulo partial commutations\nYear: 1999\n\nAbstract: It is shown that it is decidable whether an equation over a free partially commutative monoid has a solution. We give a proof of this result using normal forms. Our method is a direct reduction of a trace equation system to a word equation system with regular constraints. Hereby we use the extension of Makanin's theorem on the decidability of word equations to word equations with regular constraints, which is due to Schulz.","Title: Topology, monitorable properties and runtime verification.\nYear: 2014\n\nAbstract: We review concepts like safety, liveness, and monitorability from a rigorous topological viewpoint. Thus, monitorability of an ω-language means that the boundary in the Cantor topology has an empty interior. We show that all ω-regular languages which are deterministic and co-deterministic are monitorable, but certain deterministic liveness properties like “infinitely many a's” cannot be written as a countable union of monitorable languages. We briefly discuss model checking with LTL, its three-valued variant LTL3 and monitor constructions based upon LTL3.","Title: Omega-Rational Expressions with Bounded Synchronization Delay\nYear: 2015\n\nAbstract: In 1965 Schützenberger published his famous result that star-free languages ( $\\operatorname{SF}$ ) and aperiodic languages ( $\\operatorname{AP}$ ) coincide over finite words, often written as $\\operatorname{SF}= \\operatorname {AP}$ . Perrin generalized $\\operatorname{SF} = \\operatorname{AP}$ to infinite words in the mid 1980s. In 1973 Schützenberger presented another (and less known) characterization of aperiodic languages in terms of rational expressions where the use of the star operation is restricted to prefix codes with bounded synchronization delay and no complementation is used. We denote this class of languages by $\\operatorname{SD}$ . In this paper, we present a generalization of $\\operatorname{SD}= \\operatorname{AP}$ to infinite words. This became possible via a substantial simplification of the proof for the corresponding result for finite words. Moreover, we show that $\\operatorname{SD}= \\operatorname{AP}$ can be viewed as more fundamental than $\\operatorname{SF}= \\operatorname{AP}$ in the sense that the classical 1965 result of Schützenberger and its 1980s extension to infinite words by Perrin are immediate consequences of $\\operatorname{SD}= \\operatorname{AP}$ .","Title: A survey on the local divisor technique.\nYear: 2016\n\nAbstract: Local divisors allow a powerful induction scheme on the size of a monoid. We survey this technique by giving several examples of this proof method. These applications include linear temporal logic, rational expressions with Kleene stars restricted to prefix codes with bounded synchronization delay, Church–Rosser congruential languages, and Simon's Factorization Forest Theorem. We also introduce the notion of a localizable language class as a new abstract concept which unifies some of the proofs for the results above.","Title: Equations over free inverse monoids with idempotent variables.\nYear: 2014\n\nAbstract: We introduce the notion of idempotent variables for studying equations in inverse monoids. It is proved that it is decidable in singly exponential time (DEXPTIME) whether a system of equations in idempotent variables over a free inverse monoid has a solution. Moreover the problem becomes hard for DEXPTIME, as soon as the quotient group of the free inverse monoid has rank at least two. The upper bound is proved by a direct reduction to solve language equations with one-sided concatenation and a known complexity result by Baader and Narendran (J. Symb. Comput. 31, 277---305 2001). For the lower bound we show hardness for a restricted class of language equations. Decidability for systems of typed equations over a free inverse monoid with one irreducible variable and at least one unbalanced equation is proved with the same complexity upper-bound. Our results improve known complexity bounds by Deis et al. (IJAC 17, 761---795 2007). Our results also apply to larger families of equations where no decidability has been previously known. The lower bound confirms a conjecture made in the conference version of this paper.","Title: Characterizing Classes Of Regular Languages Using Prefix Codes Of Bounded Synchronization Delay\nYear: 2016\n\nAbstract: This paper is motivated by the work of Schutzenberger on codes with bounded synchronization delay. He was interested in characterizing those regular languages where groups in the syntactic monoid belong to a variety H. On the language side he allowed the operations union, intersection, concatenation and modified Kleene-star involving a mapping of a prefix code of bounded synchronization delay to a group G is an element of H, but no complementation. In our notation, this leads to the language classes SDH(A8). Our main result shows that SDH(A8) coincides with the class of languages having syntactic monoids where all subgroups are in H. We show that this statement holds for all varieties H of finite groups, whereas Schutzenberger proved this result for varieties H containing Abelian groups, only. Our method shows the result for all H simultaneously on finite and infinite words. Furthermore, we introduce the notion of local Rees extension which refers to a restricted type of the classical Rees extension. We give a decomposition of a monoid in terms of its groups and local Rees extensions. This gives a somewhat similar, but simpler decomposition than in the synthesis theorem of Rhodes and Allen. Moreover, we need a singly exponential number of operations, only. Finally, our decomposition yields an answer to a question in a recent paper of Almeida and Klima about varieties that are closed under Rees extensions.","Title: Solution sets for equations over free groups are EDT0L languages.\nYear: 2016\n\nAbstract: We show that, given a word equation over a finitely generated free group, the set of all solutions in reduced words forms an EDT0L language. In particular, it is an indexed language in the sense of Aho. The question of whether a description of solution sets in reduced words as an indexed language is possible has been open for some years [9, 10], apparently without much hope that a positive answer could hold. Nevertheless, our answer goes far beyond: they are EDT0L, which is a proper subclass of indexed languages. We can additionally handle the existential theory of equations with rational constraints in free products $$\\\\star _{1 \\\\le i \\\\le s}F_i$$, where each $$F_i$$ is either a free or finite group, or a free monoid with involution. In all cases the result is the same: the set of all solutions in reduced words is EDT0L. This was known only for quadratic word equations by [8], which is a very restricted case. Our general result became possible due to the recent recompression technique of Je﾿. In this paper we use a new method to integrate solutions of linear Diophantine equations into the process and obtain more general results than in the related paper [5]. For example, we improve the complexity from quadratic nondeterministic space in [5] to quasi-linear nondeterministic space here. This implies an improved complexity for deciding the existential theory of non-abelian free groups: $$\\\\mathsf {NSPACE}n\\\\log n$$. The conjectured complexity is $$\\\\mathsf {NP}$$, however, we believe that our results are optimal with respect to space complexity, independent of the conjectured $$\\\\mathsf {NP}$$.","Title: Solutions of Word Equations over Partially Commutative Structures.\nYear: 2016\n\nAbstract: We give NSPACE(n*log(n)) algorithms solving the following decision problems. Satisfiability: Is the given equation over a free partially commutative monoid with involution (resp. a free partially commutative group) solvable? Finiteness: Are there only finitely many solutions of such an equation? PSPACE algorithms with worse complexities for the first problem are known, but so far, a PSPACE algorithm for the second problem was out of reach. Our results are much stronger: Given such an equation, its solutions form an EDT0L language effectively representable in NSPACE(n*log(n)). In particular, we give an effective description of the set of all solutions for equations with constraints in free partially commutative monoids and groups.","Title: Amenability of Schreier graphs and strongly generic algorithms for the conjugacy problem.\nYear: 2015\n\nAbstract: In various occasions the conjugacy problem in finitely generated amalgamated products and HNN extensions can be decided efficiently for elements which cannot be conjugated into the base groups. Thus, the question arises how many such elements there are. This question can be formalized using the notion of strongly generic sets and lower bounds can be proven by applying the theory of amenable graphs.In this work we examine Schreier graphs of amalgamated products and HNN extensions. For an amalgamated product G=HAK with [H:A][K:A]2, the Schreier graph with respect to H or K turns out to be non-amenable if and only if [H:A]3. Moreover, for an HNN extension of the form G=H,b|bab1=(a),aA, we show that the Schreier graph of G with respect to the subgroup H is non-amenable if and only if AH(A).As application of these characterizations we show that the conjugacy problem in fundamental groups of finite graphs of groups with finitely generated free abelian vertex groups can be solved in polynomial time on a strongly generic set. Furthermore, the conjugacy problem in groups with more than one end can be solved with a strongly generic algorithm which has essentially the same time complexity as the word problem. These are rather striking results as the word problem might be easy, but the conjugacy problem might be even undecidable. Finally, our results yield a new proof that the set where the conjugacy problem of the Baumslag group G1,2 is decidable in polynomial time is also strongly generic.","Title: More Than 1700 Years of Word Equations\nYear: 2015\n\nAbstract: Geometry and Diophantine equations have been ever-present in mathematics. According to the existing literature the work of Diophantus of Alexandria was mentioned before 364 AD, but a systematic mathematical study of word equations began only in the 20th century. So, the title of the present article does not seem to be justified at all. However, a Diophantine equation can be viewed as a special case of a system of word equations over a unary alphabet, and, more importantly, a word equation can be viewed as a special case of a Diophantine equation. Hence, the problem WordEquations: \"Is a given word equation solvable?\", is intimately related to Hilbert's 10th problem on the solvability of Diophantine equations. This became clear to the Russian school of mathematics at the latest in the mid 1960s, after which a systematic study of that relation began. Here, we review some recent developments which led to an amazingly simple decision procedure for WordEquations, and to the description of the set of all solutions as an EDT0L language.","Title: A Note on Monitors and Büchi automata.\nYear: 2015\n\nAbstract: When a property needs to be checked against an unknown or very complex system, classical exploration techniques like model-checking are not applicable anymore. Sometimes a﾿monitor can be used, that checks a given property on the underlying system at runtime. A monitor for a property L is a deterministic finite automaton $$\\\\mathcal {M}_L$$ that after each finite execution tells whether 1 every possible extension of the execution is in L, or 2 every possible extension is in the complement of L, or neither 1 nor 2 holds. Moreover, L being monitorable means that it is always possible that in some future the monitor reaches 1 or 2. Classical examples for monitorable properties are safety and cosafety properties. On the other hand, deterministic liveness properties like \\\"infinitely many a's\\\" are not monitorable. We discuss various monitor constructions with a focus on deterministic $$\\\\omega $$-regular languages. We locate a proper subclass of deterministic $$\\\\omega $$-regular languages but also strictly larger than the subclass of languages which are deterministic and codeterministic; and for this subclass there exist canonical monitors which also accept the language itself. We also address the problem to decide monitorability in comparison with deciding liveness. The state of the art is as follows. Given a Büchi automaton, it is PSPACE-complete to decide liveness or monitorability. Given an LTL formula, deciding liveness becomes EXPSPACE-complete, but the complexity to decide monitorability remains open.","Title: Solutions of twisted word equations, EDT0L languages, and context-free groups.\nYear: 2017\n\nAbstract: We prove that the full solution set of a twisted word equation with regular constraints is an EDT0L language. It follows that the set of solutions to equations with rational constraints in a context-free group (= finitely generated virtually free group) in reduced normal forms is EDT0L. We can also decide whether or not the solution set is finite, which was an open problem. Moreover, this can all be done in PSPACE. Our results generalize the work by Lohrey and Senizergues (ICALP 2006) and Dahmani and Guirardel (J. of Topology 2010) with respect to complexity and with respect to expressive power. Both papers show that satisfiability is decidable, but neither gave any concrete complexity bound. Our results concern all solutions, and give, in some sense, the optimal formal language characterization.","Title: Asymptotic Approximation for the Quotient Complexities of Atoms.\nYear: 2015\n\nAbstract: In a series of papers, Brzozowski together with Tamm, Davies, andSzykuła studied the quotient complexities of atoms of regularlanguages. The authors obtained precise bounds in terms of binomialsums for the most complex situations in the following five cases:G: general, R: right ideals, L: left ideals, T:two-sided ideals and S: suffix-free languages. In each caselet κCn be the maximal complexity of an atom of aregular language L, where L has complexity n≧2 and belongsto the class C∈{G,R,L,T,S}. It is known thatκTn ≦ κLn = κRn ≦ κGn< 3n and κSn = κLn-1. We show that the ratioκCn / κCn-1 tends exponentiallyfast to 3 in all five cases but it remains different from 3.This behaviour was suggested by experimental results of Brzozowskiand Tamm; and the result for G was shown independently by LukeSchaeffer and the first author soon after the paper of Brzozowskiand Tamm appeared in 2012. However, proofs for the asymptotic behaviorof κGn / κGn-1 were never published;and the results here are valid for all five classes above. Moreover,there is an interesting oscillation for all C: for almost alln we have κCn / κCn-1 >3$ if and only ifκCn+1 / κCn<3.","Title: Church-Rosser Systems, Codes with Bounded Synchronization Delay and Local Rees Extensions.\nYear: 2017\n\nAbstract: What is the common link, if there is any, between Church-Rosser systems, prefix codes with bounded synchronization delay, and local Rees extensions? The first obvious answer is that each of these notions relates to topics of interest for WORDS: Church-Rosser systems are certain rewriting systems over words, codes are given by sets of words which form a basis of a free submonoid in the free monoid of all words (over a given alphabet) and local Rees extensions provide structural insight into regular languages over words. So, it seems to be a legitimate title for an extended abstract presented at the conference WORDS 2017. However, this work is more ambitious, it outlines some less obvious but much more interesting link between these topics. This link is based on a structure theory of finite monoids with varieties of groups and the concept of local divisors playing a prominent role. Parts of this work appeared in a similar form in conference proceedings [6, 10] where proofs and further material can be found.","Title: A Remark on Trace Equations\nYear: 1997\n\nAbstract: Without Abstract","Title: Finding All Solutions of Equations in Free Groups and Monoids with Involution.\nYear: 2014\n\nAbstract: This paper presents a PSPACE algorithm which yields a finite graph of exponential size that describes the set of all solutions of equations in free groups as well as the set of all solutions of equations with rational constraints in free monoids. This became possible due to the recent recompression technique.While this technique was successfully applied for pure word equations without involution or rational constraints it could not be used as a black box for free groups. Actually, the presence of an involution and rational constraints complicates the situation and some additional analysis is necessary. Still, the technique is general enough to accommodate both extensions. In the end, it simplifies proofs that satisfiability of word equations is in PSPACE and the corresponding result for equations in free groups with rational constraints. As a byproduct we can decide in PSPACE whether the solution set is finite.","Title: Pure future local temporal logics are expressively complete for Mazurkiewicz traces\nYear: 2006\n\nAbstract: The paper settles a long standing problem for Mazurkiewicz traces: the pure future local temporal logic defined with the basic modal- ities exists-next and until is expressively complete. The analogous result with a global interpretation was solved some years ago by Thiagarajan and Walukiewicz (1997) and in its final form without any reference to past tense constants by Diekert and Gastin (2000). Each, the (previously known) global or the (new) local result generalizes Kamp's Theorem for words, because for sequences local and global viewpoints coincide. But traces are labelled partial orders and then the difference between an inter- pretation globally over cuts (configurations) or locally at points (events) is significant. For global temporal logics the satisfiability problem is non- elementary (Walukiewicz 1998), whereas for local temporal logics both the satisfiability problem and the model checking problem are sovable in Pspace (Gastin and Kuske 2003) as in the case of words. This makes local temporal logics much more attractive.","Title: Reachability Games and Parity Games\nYear: 2022\n\nAbstract: Parity games are positionally determined. This is a fundamental and classical result. In 2010, Calude et al. showed a breakthrough result for finite parity games: the winning regions and their positional winning strategies can be computed in quasi-polynomial time. In the present paper we give a self-contained and detailed proofs for both results. The results in this paper are not meant to be original. The positional determinacy result is shown for possibly infinite parity games using the ideas of Zielonka which he published in 1998. In order to show quasi-polynomial time, we follow Lehtinen's register games, which she introduced in 2018. Although the time complexity of Lehtinen's algorithm is not optimal, register games are conceptually simple and interesting in their own right. Various of our proofs are either new or simplifications of the original proofs. The topics in this paper include the definition and the computation of optimal attractors for reachability games, too.","Title: Properties of graphs specified by a regular language\nYear: 2022\n\nAbstract: Traditionally, graph algorithms get a single graph as input, and then they should decide if this graph satisfies a certain property \n \n \n \n $$\\varPhi $$\n \n . What happens if this question is modified in a way that we get a possibly infinite family of graphs as an input, and the question is if there is a graph satisfying \n \n \n \n $$\\varPhi $$\n \n in the family? We approach this question by using formal languages for specifying families of graphs, in particular by regular sets of words. We show that certain graph properties can be decided by studying the syntactic monoid of the specification language L if a certain torsion condition is satisfied. This condition holds trivially if L is regular. More specifically, we use a natural binary encoding of finite graphs over a binary alphabet \n \n \n \n $$\\varSigma $$\n \n , and we define a regular set \n \n \n \n $$\\mathbb {G}\\subseteq \\varSigma ^*$$\n \n such that every nonempty word \n \n \n \n $$w\\in \\mathbb {G}$$\n \n defines a finite and nonempty graph. Also, graph properties can then be syntactically defined as languages over \n \n \n \n $$\\varSigma $$\n \n . Then, we ask whether the automaton \n \n \n \n $$\\mathcal {A}$$\n \n specifies some graph satisfying a certain property \n \n \n \n $$\\varPhi $$\n \n . Our structural results show that we can answer this question for all “typical” graph properties. In order to show our results, we split L into a finite union of subsets and every subset of this union defines in a natural way a single finite graph F where some edges and vertices are marked. The marked graph in turn defines an infinite graph \n \n \n \n $$F^\\infty $$\n \n and therefore the family of finite subgraphs of \n \n \n \n $$F^\\infty $$\n \n where F appears as an induced subgraph. This yields a geometric description of all graphs specified by L based on splitting L into finitely many pieces; then using the notion of graph retraction, we obtain an easily understandable description of the graphs in each piece."],"string":"[\n \"Title: Approximating Traces\\nYear: 1998\\n\\nAbstract: . In order to give a semantics to concurrent processes we need a model having some good mathematical properties. To this end\\n we generalize (infinite) Mazurkiewicz traces by adding some alphabetical information concerning the possible continuations\\n of a process. This allows to define an approximation order compatible with the composition. We obtain a prime algebraic and\\n coherently complete domain where the compact elements are exactly the finite approximations of processes. The composition\\n is shown to be monotone and -continuous. We define a suitable metric which induces the Lawson topology and which yields a compact metric space being therefore\\n complete. The finite approximations of processes form a dense and open subset and the composition is (uniformly) continuous.\\n A preliminary version of this work appeared in [7].\",\n \"Title: Regular frequency computations\\nYear: 2005\\n\\nAbstract: An (m, n)-computation of a function f is given by a deterministic Turing machine which on n pairwise different inputs produces n output values where at least m of the n values are in accordance with f. In such a case, we say that the Turing machine computes f with frequency ≥ m/n. The most prominent result for frequency computations is due to Trakhtenbrot: The class of (m, n)-computable functions equals the class of computable functions if and only if 2m n.Via characteristic functions the definition of (m, n)-computability carries over to sets. Here Trakhtenbrot's result reads as: The class of (m, n)-computable sets equals the class of recursive sets if and only if 2m n.The notion of frequency computation can be extended to other models of computation. For resource bounded computations, the behavior is completely different: for e.g., whenever n' - m' n - m, it is known that under any reasonable resource bound there are sets (m', n')-computable, but not (m, n)- computable.However, scaling down to finite automata, the analogue of Trakhtenbrot's result holds again: We show here that the class of languages (m, n)-recognizable by deterministic finite automata equals the class of regular languages if and only if 2m n. This was originally stated by Kinber, but his proof has a flaw, as pointed out by Tantau.Conversely, for 2m ≤ n, the class of languages (m, n)-recognizable by deterministic finite automata is uncountable for a two-letter alphabet. When restricted to a one-letter alphabet, then every (m, n)- recognizable language is regular. This was also shown by Kinber. We give a new and more direct proof for this result.\",\n \"Title: Quadratic Word Equations\\nYear: 1999\\n\\nAbstract: IntroductionThe existential theory of equations over free monoids is decidable. This majorresult was obtained by Makanin [10], who showed that the satisability ofword equations with constants is decidable. For the background we refer to[14], to the corresponding chapter in the Handbook of Formal Languages(Eds. G. Rozenberg and A. Salomaa), [2], or to the forthcoming [3]. In 1990Schulz [19] showed an important generalization: Makanin's result remainstrue when adding regular constraints. ...\",\n \"Title: Star-free languages are Church-Rosser congruential\\nYear: 2012\\n\\nAbstract: The class of Church-Rosser congruential languages has been introduced by McNaughton, Narendran, and Otto in 1988. A language L is Church-Rosser congruential (belongs to CRCL), if there is a finite, confluent, and length-reducing semi-Thue system S such that L is a finite union of congruence classes modulo S. To date, it is still open whether every regular language is in CRCL. In this paper, we show that every star-free language is in CRCL. In fact, we prove a stronger statement: for every star-free language L there exists a finite, confluent, and subword-reducing semi-Thue system S such that the total number of congruence classes modulo S is finite and such that L is a union of congruence classes modulo S. The construction turns out to be effective.\",\n \"Title: Solvability Of Equations In Graph Groups Is Decidable\\nYear: 2006\\n\\nAbstract: We show that the existential theory of free partially commutative monoids with involution is decidable. As a consequence the existential theory of graph groups is also decidable. If the underlying alphabet of generators is fixed, we obtain a PSPACE-completeness result, otherwise (in the uniform setting) our decision procedure is in EXPSPACE. Our proof is a reduction to the main result of [6].\",\n \"Title: A remark about quadratic trace equations\\nYear: 2002\\n\\nAbstract: We present a simple PSPACE-algorithm for computing the most general solution of a system of quadratic trace equations with involution. We extend the known linear time algorithm for quadratic word equations with length constraints to cope with involutions. Finally, we show that the same linear-time result cannot be expected for trace equations. We obtain an NP-hardness result.\",\n \"Title: Word problems over traces which are solvable in linear time\\nYear: 1989\\n\\nAbstract: Finite complete replacement systems over free partially commutative monoids have a decidable word problem, and if these systems are weight-reducing, then the word problem is solvable in square time. It is unknown whether this time bound is optimal, but the restriction to semi-Thus systems or vector replacement systems yields linear time. In direction to close this gap, we first give a global condition for trace replacement systems which also ensures a linear time bound: and we prove that this condition is decidable. Then we give a stronger but local condition which leads to the notion of cones and blocks which are certain traces. Trivial examples of cones and blocks are words and vectors, respectively. We show that under a mild additional assumption we may decide the confluence of a noetherian trace replacement system if all left-hand sides are cones or blocks. (In general, confluence is undecidable even for length-reducing systems). Altogether we obtain a sufficient and computable condition such that the word problem of a trace replacement system is solvable in linear time.\",\n \"Title: Code Problems on Traces\\nYear: 1996\\n\\nAbstract: . The topic of codes in the framework of trace monoids leadsto interesting and challenging decision problems of combinatorial flavour.We give an overview of the current state of some basic questions in thisfield. Among these, we consider the existence problem for strong codings,clique-preserving morphisms and the unique decipherability problem(code problem).1 IntroductionFree partially commutative monoids [7] offer a mathematically sound frameworkfor modelling and analyzing concurrent...\",\n \"Title: Combinatorial rewriting on traces\\nYear: 1990\\n\\nAbstract: There are two main problems in working with replacement systems over free partially commutative monoids: For finite noetherian systems confluence is undecidable, in general, and the known algorithm to compute irreducible normal forms need time square in the derivation length instead of linear. We first give a decidable and sufficient condition for finite noetherian systems such that confluence becomes decidable. This condition is weaker than the known ones before. Then we give a decidable and sufficient condition such that irreducible normal forms are computable in time linear to the derivation length. Furthermore, we prove that the first condition is implied by the second. We also present a new uniform algorithm for computing normal forms using Zielonka's theory of asynchronous automata.\",\n \"Title: On Codings of Traces\\nYear: 1995\\n\\nAbstract: Abstract. The paper solves the main open problem of [BFG94]. We show that given two dependence alphabets ( ; D) and (,; D,), it is de-cidable whether there exists a strong coding h : M ( ; D) ! M (,; D,) between the associated trace monoids. In fact, we show that the problem is NP-complete. (A coding is an injective homomorphism, it is strong if independent letters are mapped to independent traces. ) We exhibit an example of trace monoids where a coding between them exists, but no strong coding. The decidability of codings remains open, in general. We have a lower and an upper bound, which show both to be strict. We fur-ther discuss encodings of free products of trace monoids and give almost optimal constructions. In the nal section, we state that the coding property is undecidable in a naturally arising class of homomorphisms.\",\n \"Title: Recognizable Complex Trace Languages\\nYear: 1991\\n\\nAbstract: A. Mazurkiewicz defined traces in order to modelize non-sequential processes. Complex traces have been recently introduced as a generalization of both traces and infinite words. This paper studies the family of recognizable complex trace languages. It is proved that this family is closed under boolean operations, concatenation, left and right quotients. Then sufficient conditions ensuring the recognizability of the finite and infinite iterations of a recognizable complex trace language are given. The notion of co-iteration is defined and the Kleene-Ochmanski theorem is generalized to complex traces.\",\n \"Title: Some remarks about stabilizers\\nYear: 2009\\n\\nAbstract: We continue our study of stabilizers of infinite words over finite alphabets, begun in [D. Krieger, On stabilizers of infinite words, Theoret. Comput. Sci. 400 (2008), 169-181]. Let w be an aperiodic infinite word over a finite alphabet, and let Stab(w) be its stabilizer. We show that Stab(w) can be partitioned into the monoid of morphisms that stabilize w by finite fixed points and the ideal of morphisms that stabilize w by iteration. We also settle a conjecture given in the paper mentioned above, by showing that in some cases Stab(w) is infinitely generated. If the aforementioned ideal is nonempty, then it contains either polynomially growing morphisms or exponentially growing morphisms, but not both. Moreover, in the polynomial case, the degree of the polynomial is fixed. We also show how to compute the polynomial degree from the dependency graph of a polynomially growing morphism.\",\n \"Title: On quadratic word equations\\nYear: 1999\\n\\nAbstract: We investigate the satisfiability problem of word equations where each variable occurs at most twice (quadratic systems). We obtain various new results: The satisfiability problem is NP-hard (even for a single equation). The main result says that once we have fixed the lengths of a possible solution, then we can decide in linear time whether there is a corresponding solution. If the lengths of a minimal solution were at most exponential, then the satisfiability problem of quadratic systems would be NP-complete. (The inclusion in NP follows also from [21]) In the second part we address the problem with regular constraints: The uniform version is PSPACE-complete. Fixing the lengths of a possible solution doesn't make the problem much easier. The non-uniform version remains NP-hard (in contrast to the linear time result above). The uniform version remains PSPACE-complete.\",\n \"Title: Transitive Orientations, Möbius Functions, and Complete Semi-Thue Systems for Free Partially Commutative Monoids\\nYear: 1988\\n\\nAbstract: Let I\\n\\nÍ</font\\n>\\\\subseteq\\n\\nX×X be an independence relation over a finite alphabet X and M=X*/{(ab, ba)|(a, b)teI} the associated free partially commutative monoid. The Möbius function of M is a polynomial in the ring of formal power series Z Í</font\\n>\\\\subseteq\\n\\nX*×X* which define M. We obtain a one-to-one correspondence between unambiguous Möbius functions, transitive orientations and finite (normalized) complete semi-Thue systems.\",\n \"Title: Hotz-Isomorphism Theorems in Formal Language Theory\\nYear: 1989\\n\\nAbstract: A language \\nL Í</font\\n> X*L \\\\subseteq X*\\nis called a language with a Hotz-isomorphism if there is a generating grammar such that its Hotz group is canonically isomorphic to F(X)/L. The main result of the paper states that L is a language with a Hotz-isomorphism if and only if F(X)/L is a finitely presentable group. We also prove an analogous result for Hotz-monoids. Further, we show when the construction of a grammar which allows the Hotz-isomorphism is effective; and we prove various undecidability results concerning this construction.\",\n \"Title: On Efficient Reduction Algorithms for Some Trace Rewriting Systems\\nYear: 1993\\n\\nAbstract: . We consider some basic problems on the decidability andcomplexity of trace rewriting systems. The new contribution of this paperis an O(n log(n)) algorithm for some computing irreducible normal formsin the case of certain one-rule systems.1 IntroductionThe notes of this paper are based on the Font Romeu Lecture and on an invitedlecture at FCT-93 conference [11] of the second author. In the first part of thepaper we give some basic background about trace rewriting systems. There is...\",\n \"Title: Language theoretical properties of hairpin formations\\nYear: 2012\\n\\nAbstract: Hairpin formations arise in biochemical processes and play an important role in DNA-computing. We study language theoretical properties of hairpin formations and our new results concern the hairpin completion H\\\"@k(L\\\"1,L\\\"2) of two regular languages L\\\"1 and L\\\"2 and the iterated hairpin lengthening HL\\\"@k^*(L) of any language L. Assume that L\\\"1 and L\\\"2 belong to a certain variety of regular languages which satisfies a mild closure property (being closed by a restricted concatenation), then either H\\\"@k(L\\\"1,L\\\"2) is not regular or it belongs to the same variety as L\\\"1 and L\\\"2. This result applies, in particular, to the class of first-order definable languages (which is the class of aperiodic or star-free languages) and it applies to the class of first-order definable languages in two variables with predicates\",\n \"Title: From local to global temporal logics over Mazurkiewicz traces\\nYear: 2006\\n\\nAbstract: We review some results on global and local temporal logic on Mazurkiewicz traces. Our main contribution is to show how to derive the expressive completeness of global temporal logic with respect to first-order logic [V. Diekert, P. Gastin, LTL is expressively complete for Mazurkiewicz traces, J. Comput. System Sci. 64 (2002) 396-418] from the similar result on local temporal logic [V. Diekert, P. Gastin, Pure future local temporal logics are expressively complete for Mazurkiewicz traces, in: M. Farach-Colton (Ed.), Proc. LATIN'04, Lecture Notes in Computer Science, Vol. 2976, Springer, Berlin, 2004, pp. 232-241, Full version available as Research Report LSV-05-22, Laboratoire Spécification et Vérification, ENS Cachan, France].\",\n \"Title: An Expressively Complete Temporal Logic without Past Tense Operators for Mazurkiewicz Traces\\nYear: 1999\\n\\nAbstract: Mazurkiewicz traces are a widely accepted model of concurrent systems. We introduce a linear time temporal logic LTLf which has the same expressive power as the first order theory FO(\",\n \"Title: Slp Compression For Solutions Of Equations With Constraints In Free And Hyperbolic Groups\\nYear: 2013\\n\\nAbstract: The paper is a part of an ongoing program which aims to show that the problem of satisfiability of a system of equations in a free group (hyperbolic or even toral relatively hyperbolic group) is NP-complete. For that, we study compression of solutions with straight-line programs (SLPs) as suggested originally by Plandowski and Rytter in the context of a single word equation. We review some basic results on SLPs and give full proofs in order to keep this fundamental part of the program self-contained. Next we study systems of equations with constraints in free groups and more generally in free products of abelian groups. We show how to compress minimal solutions with extended Parikh-constraints. This type of constraints allows to express semi-linear conditions as e.g. alphabetic information. The result relies on some combinatorial analysis and has not been shown elsewhere. We show similar compression results for Boolean formula of equations over a torsion-free delta-hyperbolic group. The situation is much more delicate than in free groups. As byproduct we improve the estimation of the \\\"capacity\\\" constant used by Rips and Sela in their paper \\\"Canonical representatives and equations in hyperbolic groups\\\" from a double-exponential bound in delta to some single-exponential bound. The final section shows compression results for toral relatively hyperbolic groups using the work of Dahmani: We show that given a system of equations over a fixed toral relatively hyperbolic group, for every solution of length N there is an SLP for another solution such that the size of the SLP is bounded by some polynomial p( s + log N) where s is the size of the system.\",\n \"Title: Confluence of One-Rule Thue Systems\\nYear: 1990\\n\\nAbstract: Without Abstract\",\n \"Title: Logspace computations in graph groups and coxeter groups\\nYear: 2012\\n\\nAbstract: Computing normal forms in groups (or monoids) is in general harder than solving the word problem (equality testing). However, normal form computation has a much wider range of applications. It is therefore interesting to investigate the complexity of computing normal forms for important classes of groups. We show that shortlex normal forms in graph groups and in right-angled Coxeter groups can be computed in logspace. Graph groups are also known as free partially commutative groups or as right-angled Artin groups in the literature. (Artin groups can be realized as subgroups of Coxeter groups.) Graph groups arise in many areas and have a close connection to concurrency theory. The connection is used here. Indeed, for our result we use a representation of group elements by Mazurkiewicz traces. These are directed acyclic node-labelled graphs (i.e. pomsets). They form an algebraic model to describe runs of concurrent systems. Concurrent systems which are deterministic and co-deterministic can be studied via inverse monoids. As an application of our results we show that the word problem for free partially commutative inverse monoids is in logspace. This result generalizes a result of Ondrusch and the third author on free inverse monoids. All Coxeter groups are linear, so the word problem can be solved in logspace, but it is open (in the non-right-angled case) whether shortlex normal forms can be computed in logspace, or, less demanding, whether they can be computed efficiently in parallel. We show that for all Coxeter groups the set of letters occurring in the shortlex normal form of an element can be computed in logspace.\",\n \"Title: Rational and recognizable complex trace languages\\nYear: 1995\\n\\nAbstract: Mazurkiewicz defined traces as an algebraic model of finite concurrent processes. In order to model non-terminating processes a good notion of infinite traces was needed, which finally led to the notion of complex traces. For complex traces an associative concatenation and an ω-iteration are defined. This paper defines and investigates rational and recognizable complex trace languages. We prove various closure results such as the closure under boolean operations (for recognizable languages), concatenation, and left and right quotients by recognizable sets. Then we study sufficient conditions ensuring the recognizability of the finite and infinite iterations of complex trace languages. We introduce a generalization of the notion of concurrent iteration which leads to the main result of the paper: the generalization of Kleene′s and Ochmański′s theorems to complex trace languages.\",\n \"Title: Existential and Positive Theories of Equations in Graph Products\\nYear: 2002\\n\\nAbstract: We prove that the existential theory of equations with normalized rational constraints in a fixed graph product of finite\\n monoids, free monoids, and free groups is PSPACE-complete. Under certain restrictions this result also holds if the graph\\n product is part of the input. As the second main result we prove that the positive theory of equations with recognizable constraints\\n in graph products of finite and free groups is decidable.\",\n \"Title: A structural property of regular frequency computations\\nYear: 2003\\n\\nAbstract: The notion of an (m,n)-computation was already introduced in 1960 by Rose and further investigated by Trakhtenbrot in 1963. It has been extended to finite automata by Kinber in 1976 and he has shown an analogue of Trakhtenbrot's result: The class of languages (m,n)-recognizable by deterministic finite automata is equal to the class of regular languages if and only if 2m n. Furthermore, for a unary alphabet, the class of (m,n)-recognizable languages coincides with the class of regular languages for all m and n. In this paper, we will present the first structural property of (m,n)-recognizable languages which is valid for all 1 ≤ m ≤ n and for all alphabets. Kinber's result for unary alphabets becomes a corollary. This property is also used to show that certain non-unary languages are not (m,n)-regular and that the class of all (m,n)-recognizable languages is not closed under the reversal operation. However, this class forms a Boolean algebra.\",\n \"Title: Mo¨bius functions and confluent semi-commutations\\nYear: 1993\\n\\nAbstract: Dickert (1988) settled a well-known conjecture on unambiguous liftings to words of Möbius functions for free partially commutative monoids. He established thereby a new bridge from the theory of formal power series to the theory of complete semi-Thue systems. As an open question, it was asked whether this approach has a generalization to a relative situation where we do not lift to the free (word-) case but only to some level of less commutation. This question has been reconsidered by König (1992) and a conjecture has been formulated there. Our results solve the open problem and settle this conjecture (Remark 4.8). We show that there is a canonical one-to-one correspondence between unambiguous Möbius functions and confluent semi-commutation systems (Theorem 5.1). This identification is due to some graph-theoretic characterizationobtained by Diekert (1991) and which allows one to apply some interesting complexity results (Section 6). Our results can be viewed as a contribution to the combinatorial theory of Möbius functions and to the theory of rewriting on traces.\",\n \"Title: Removing epsilon-Transitions in Timed Automata\\nYear: 1997\\n\\nAbstract: . Timed automata are among the most widely studied modelsfor real-time systems. Silent transitions, i.e., "-transitions, have alreadybeen proposed in the original paper on timed automata by Alur and Dill[2]. In [7] it is shown that "-transitions can be removed, if they do notreset clocks; moreover "-transitions strictly increase the power of timedautomata, if there is a self-loop containing "-transitions which reset someclocks. The authors of [7] left open the problem about the power of...\",\n \"Title: On the Hairpin Completion of Regular Languages\\nYear: 2009\\n\\nAbstract: The hairpin completion is a natural operation of formal languages which has been inspired by molecular phenomena in biology and by DNA-computing. The hairpin completion of a regular language is linear context-free and we consider the problem to decide whether the hairpin completion remains regular. This problem has been open since the first formal definition of the operation. In this paper we present a positive solution to this problem. Our solution yields more than decidability because we present a polynomial time procedure. The degree of the polynomial is however unexpectedly high, since in our approach it is more than n 14. Nevertheless, the polynomial time result is surprising, because even if the hairpin completion $\\\\mathcal{H}$ of a regular language L is regular, there can be an exponential gap between the size of a minimal DFA for L and the size of a smallest NFA for $\\\\mathcal{H}$.\",\n \"Title: On confluent semi-commutations: decidability and complexity results\\nYear: 1991\\n\\nAbstract: The subject of this paper is the confluence of finite semi-commutation systems. Confluence of such systems is proved to be decidable property. Existence of a finite complete presentation of a trace monoid using rules only from another given trace monoid is proved to be reducible to the existence of a confluent semi-commutation system. Complexity results related to the preceding problems are proved: deciding the existence of finite complete presentations is Σ P 2 -complete, whereas deciding confluence of semi-commutation systems is Co-NP-complete. Additionally, an open problem about trace synchronizations is solved: The local checking property is Co-NP-complete.\",\n \"Title: Group Extensions Over Infinite Words\\nYear: 2012\\n\\nAbstract: Non-Archimedean words have been introduced as a new type of infinite words which can be investigated through classical methods in combinatorics on words due to a length function. The length function, however, takes values in the additive group of polynomials Z[t] (and not, as traditionally, in N), which yields various new properties. Non-Archimedean words allow to solve a number of interesting algorithmic problems in geometric and algorithmic group theory. There is also a connection to logic and the first-order theory in free groups (Tarski Problems).In the present paper we provide a general method to use infinite words over a discretely ordered abelian group as a tool to investigate certain group extensions for an arbitrary group G. The central object is a group E(A, G) which is defined in terms of a non-terminating, but confluent rewriting system. The group G as well as some natural HNN-extensions of G embed into E(A, G) (and still \\\"behave like\\\" G), which makes it interesting to study its algorithmic properties. In order to show that every group G embeds into E(A, G) we combine methods from combinatorics on words, string rewriting and group theory.\",\n \"Title: On the Knuth–Bendix completion for concurrent processes\\nYear: 1987\\n\\nAbstract: We consider critical pairs for replacement systems over free partially commutative monoids. This is done in order to apply the Knuth-Bendix completion procedure to concurrent processes. We will see that there are one-rule systems which have no finite set of critical pairs. Therefore we develop a sufficient (and computable) condition such that finite trace replacement systems have a finite set of critical pairs. This condition is always satisfied in the purely free case of semi-Thue systems or purely commutative case of vector replacement systems. In fact, it generalizes (and unifies) both cases. We will give examples of how one can use this generalization.\",\n \"Title: Algorithmic Problems On Inverse Monoids Over Virtually Free Groups\\nYear: 2008\\n\\nAbstract: Let G be a finitely generated virtually-free group. We consider the Birget-Rhodes expansion of G, which yields an inverse monoid and which is denoted by IM(G) in the following. We show that for a finite idempotent presentation P, the word problem of a quotient monoid IM(G)/P can be solved in linear time on a RAM. The uniform word problem, where G and the presentation P are also part of the input, is EXPTIME-complete. With IM(G)/P we associate a relational structure, which contains for every rational subset L of IM(G)/P a binary relation, consisting of all pairs (x, y) such that y can be obtained from x by right multiplication with an element from L. We prove that the first-order theory of this structure is decidable. This result implies that the emptiness problem for Boolean combinations of rational subsets of IM(G)/P is decidable, which, in turn implies the decidability of the submonoid membership problem of IM(G)/P. These results were known previously for free groups, only. Moreover, we provide a new algorithmic approach for these problems, which seems to be of independent interest even for free groups. We also show that one cannot expect decidability results in much larger frameworks than virtually-free groups because the subgroup membership problem of a subgroup H in an arbitrary group G can be reduced to a word problem of some IM(G)/P, where P depends only on H. A consequence is that there is a hyperbolic group G and a finite idempotent presentation P such that the word problem is undecidable for some finitely generated submonoid of IM(G)/P. In particular, the word problem of IM(G)/P is undecidable.\",\n \"Title: Conjugacy in Baumslag's Group, Generic Case Complexity, and Division in Power Circuiats.\\nYear: 2013\\n\\nAbstract: The conjugacy problem asks whether two words over generators of a fixed group G are conjugated, i.e., it is the problem to decide on input words x, y whether there exists z such that $$zx z^{-1} =y$$zxz-1=y in G. The conjugacy problem is more difficult than the word problem, in general. We investigate the conjugacy problem for two prominent groups: the Baumslag---Solitar group $$\\\\\\\\mathbf{{BS}}_{1,2}$$BS1,2 and the Baumslag group $${\\\\\\\\mathbf{{G}}}_{1,2}$$G1,2 (also known as Baumslag---Gersten group). The conjugacy problem in $${\\\\\\\\mathbf{{BS}}}_{1,2}$$BS1,2 is complete for the circuit class $$\\\\\\\\mathsf {TC}^0$$TC0. To the best of our knowledge $${\\\\\\\\mathbf{{BS}}}_{1,2}$$BS1,2 is the first natural infinite non-commutative group where such a precise and low complexity is shown. The Baumslag group $${\\\\\\\\mathbf{{G}}}_{1,2}$$G1,2 is an HNN-extension of $${\\\\\\\\mathbf{{BS}}}_{1,2}$$BS1,2. Hence, decidability of the conjugacy problem in $$\\\\\\\\mathbf{{G}}_{1,2}$$G1,2 outside the so-called \\\\\\\"black hole\\\\\\\" follows from Borovik et al. (Int J Algebra Comput 17(5/6):963---997, 2007). Decidability everywhere is due to Beese. Moreover, he showed exponential time for the set of elements which cannot be conjugated into $$\\\\\\\\mathbf{{BS}}_{1,2}$$BS1,2 (Beese 2012). Here we improve Beese's result in two directions by showing that the conjugacy problem in $${\\\\\\\\mathbf{{G}}}_{1,2}$$G1,2 can be solved in polynomial time in a strongly generic setting. This means that essentially for all inputs, conjugacy in $${\\\\\\\\mathbf{{G}}}_{1,2}$$G1,2 can be decided efficiently. In contrast, we show that under a plausible assumption the average case complexity of the same problem is non-elementary. Moreover, we provide a lower bound for the conjugacy problem in $${\\\\\\\\mathbf{{G}}}_{1,2}$$G1,2 by reducing the divisibility problem in power circuits to the conjugacy problem in $${\\\\\\\\mathbf{{G}}}_{1,2}$$G1,2. The complexity of the divisibility problem in power circuits is an open and interesting problem in integer arithmetic. We conjecture that it cannot be solved in elementary time because we can show that it cannot be solved in elementary time by calculating modulo values in power circuits.\",\n \"Title: Rewriting, Möbius Functions and Semi-Commutations\\nYear: 1993\\n\\nAbstract: Without Abstract\",\n \"Title: Regular languages are church-rosser congruential\\nYear: 2012\\n\\nAbstract: This paper proves a long standing conjecture in formal language theory. It shows that all regular languages are Church-Rosser congruential. The class of Church-Rosser congruential languages was introduced by McNaughton, Narendran, and Otto in 1988. A language L is Church-Rosser congruential if there exists a finite, confluent, and length-reducing semi-Thue system S such that L is a finite union of congruence classes modulo S. It was known that there are deterministic linear context-free languages which are not Church-Rosser congruential, but on the other hand it was strongly believed that all regular languages are of this form. This paper solves the conjecture affirmatively by actually proving a more general result.\",\n \"Title: Cyclic Rewriting And Conjugacy Problems\\nYear: 2012\\n\\nAbstract: Cyclic words are equivalence classes of cyclic permutations of ordinary words. When a group is given by a rewriting relation, a rewriting system on cyclic words is induced, which is used to construct algorithms to find minimal length elements of conjugacy classes in the group. These techniques are applied to the universal groups of Stallings pre-groups and in particular to free products with amalgamation, HNN- extensions and virtually free groups, to yield simple and intuitive algorithms and proofs of conjugacy criteria.\",\n \"Title: QuickHeapsort: Modifications and Improved Analysis.\\nYear: 2012\\n\\nAbstract: QuickHeapsort is a combination of Quicksort and Heapsort. We show that the expected number of comparisons for QuickHeapsort is always better than for Quicksort if a usual median-of-constant strategy is used for choosing pivot elements. In order to obtain the result we present a new analysis for QuickHeapsort splitting it into the analysis of the partition-phases and the analysis of the heap-phases. This enables us to consider samples of non-constant size for the pivot selection and leads to better theoretical bounds for the algorithm. Furthermore, we introduce some modifications of QuickHeapsort. We show that for every input the expected number of comparisons is at most nlog2nź0.03n+o(n)$n\\\\\\\\log _{2}n - 0.03n + o(n)$ for the in-place variant. If we allow n extra bits, then we can lower the bound to nlog2nź0.997n+o(n)$ n\\\\\\\\log _{2} n -0.997 n+ o (n)$. Thus, spending n extra bits we can save more that 0.96n comparisons if n is large enough. Both estimates improve the previously known results. Moreover, our non-in-place variant does essentially use the same number of comparisons as index based Heapsort variants and Relaxed-Weak-Heapsort which use nlog2nź0.9n+o(n)$ n\\\\\\\\log _{2}n -0.9 n+ o (n)$ comparisons in the worst case. However, index based Heapsort variants and Relaxed-Weak-Heapsort require ź(nlogn)${\\\\\\\\Theta }(n\\\\\\\\log n)$ extra bits whereas we need n bits only. Our theoretical results are upper bounds and valid for every input. Our computer experiments show that the gap between our bounds and the actual values on random inputs is small. Moreover, the computer experiments establish QuickHeapsort as competitive with Quicksort in terms of running time.\",\n \"Title: One-Rule Trace-Rewriting Systems and Confluence\\nYear: 1992\\n\\nAbstract: Confluence is an undecidable property even for finite noetherian trace-rewriting systems. Here we investigate this property\\n for the special case of trace-rewriting systems with a single rule. For certain classes of one-rule trace-rewriting systems\\n R we present syntactic characterizations that are necessary and sufficient for R to be confluent. Based on these characterizations\\n confluence is easily decidable for these systems. In fact, each system satisfying one of these characterizations is strongly\\n confluent. This raises the question of whether every confluent onerule trace-rewriting system is strongly confluent. We close\\n this paper with an example that answers this question in the negative.\",\n \"Title: Context-Free Groups And Their Structure Trees\\nYear: 2013\\n\\nAbstract: Let G be a connected, locally finite graph of finite tree width and G be a group acting on it with finitely many orbits and finite node stabilizers. We provide an elementary and direct construction of a tree T on which G acts with finitely many orbits and finite vertex stabilizers. Moreover, the tree is defined directly in terms of the structure tree of optimally nested cuts of G. Once the tree is constructed, Bass-Serre theory yields that G is virtually free. This approach simplifies the existing proofs for the fundamental result of Muller and Schupp that characterizes context-free groups as finitely generated (f.g.) virtually free groups. Our construction avoids the explicit use of Stallings' structure theorem and it is self-contained. We also give a simplified proof for an important consequence of the structure tree theory by Dicks and Dunwoody which has been stated by Thomassen and Woess. It says that a f.g. group is accessible if and only if its Cayley graph is accessible.\",\n \"Title: On the Synchronization of Traces.\\nYear: 1989\\n\\nAbstract: Trace theory has been developed to describe the behavior of concurrent systems. Synchronization of traces is of special interest for a modular approach. We characterize those trace monoids for which synchronization can be described locally.\",\n \"Title: Deterministic asynchronous automata for infinite traces\\nYear: 1993\\n\\nAbstract: This paper shows the equivalence between the family of recognizable languages over infinite traces and the family of languages which are recognized by deterministic asynchronous cellular Muller automata. We thus give a proper generalization of McNaughton's Theorem from infinite words to infinite traces. Thereby we solve one of the main open problems in this field. As a special case we obtain that every closed (w.r.t. the independence relation) word language is accepted by someI-diamond deterministic Muller automaton.\",\n \"Title: It Is Nl-Complete To Decide Whether A Hairpin Completion Of Regular Languages Is Regular\\nYear: 2011\\n\\nAbstract: The hairpin completion is an operation on formal languages which is inspired by the hairpin formation in biochemistry. Hairpin formations occur naturally within DNA-computing. It has been known that the hairpin completion of a regular language is linear context-free, but not regular, in general. However, for some time it is was open whether the regularity of the hairpin completion of a regular language is decidable. In 2009 this decidability problem has been solved positively in [5] by providing a polynomial time algorithm. In this paper we improve the complexity bound by showing that the decision problem is actually NL-complete. This complexity bound holds for both, the one-sided and the two-sided hairpin completions.\",\n \"Title: Logspace computations in graph products\\nYear: 2013\\n\\nAbstract: We consider three important and well-studied algorithmic problems in group theory: the word, geodesic, and conjugacy problem. We show transfer results from individual groups to graph products. We concentrate on logspace complexity because the challenge is actually in small complexity classes, only. The most difficult transfer result is for the conjugacy problem. We have a general result for graph products, but even in the special case of a graph group the result is new. Graph groups are closely linked to the theory of Mazurkiewicz traces which form an algebraic model for concurrent processes. Our proofs are combinatorial and based on well-known concepts in trace theory. We also use rewriting techniques over traces. For the group-theoretical part we apply Bass-Serre theory. But as we need explicit formulae and as we design concrete algorithms all our group-theoretical calculations are completely explicit and accessible to non-specialists.\",\n \"Title: Local Checking of Trace Synchroniziability\\nYear: 1988\\n\\nAbstract: Trace theory has been developed to describe the behaviour of concurrent systems. For a modular approach synchronization of traces is of special interest. We characterize those trace monoids for which synchronization can be described locally.\",\n \"Title: On the concatenation of infinite traces\\nYear: 1993\\n\\nAbstract: There is a straightforward generalization of traces to infinite traces as dependence graphs where every vertex has finitely many predecessors, or what is the same, as a backward closed and directed set of traces with respect to prefix ordering. However, this direct approach has a drawback since it does not allow one to describe some basic phenomena which are related to concatenation. We solve this problem by adding to an infinite trace a second component. This second component is a finite alphabetic information which is called the alphabet at infinity . We obtain a compact and complete ultra-metric space where the concatenation is uniformly continuous and where the set of finite traces is an open, discrete, and dense subset. Our objects arise in a natural way from the consideration of dependence graphs where the induced partial order is well-founded. Such a graph splits into a so-called real part and a transfinite part . From the transfinite part only its alphabet is of importance. Our approach is a nontrivial generalization of the well-known construction for words and yields a convenient semantics for infinite concurrent processes.\",\n \"Title: Investigations on Hotz groups for arbitrary grammars\\nYear: 1986\\n\\nAbstract: The Hotz group H(G) and the Hotz monoid M(G) of an arbitrary grammar G=(V, X, P, S) are defined by H(G)=F(V?X)/P and M(G) =(V?X)*/P respectively. A language L?X* is called a language with Hotz isomorphism if there exists a grammar G with L=L(G) such that the natural homomorphism F(X)/L?H(G) is an isomorphism. The main result of this paper states that homomorphic images of sentential form languages are languages with Hotz isomorphism. This is a generalization of a result of Frougny, Sakarovitch, and Valkema on context-free languages.\",\n \"Title: On Computing Geodesics In Baumslag-Solitar Groups\\nYear: 2011\\n\\nAbstract: We introduce the peak normal form for elements of the Baumslag-Solitar groups BS(p, q). This normal form is very close to the length-lexicographical normal form, but more symmetric. Both normal forms are geodesic. This means the normal form of an element u(-1)v yields the shortest path between u and v in the Cayley graph. For horocyclic elements the peak normal form and the length-lexicographical normal form coincide. The main result of this paper is that we can compute the peak normal form in polynomial time if p divides q. As consequence we can compute geodesic lengths in this case. In particular, this gives a partial answer to Question 1 in [4].For arbitrary p and q it is possible to compute the peak normal form (length-lexicographical normal form resp.) also the for elements in the horocyclic subgroup and, more generally, for elements which we call hills. This approach leads to a linear time reduction of the problem of computing geodesics to the problem of computing geodesics for Britton-reduced words where the t-sequence starts with t(-1) and ends with t.\",\n \"Title: A Survey On Small Fragments Of First-Order Logic Over Finite Words\\nYear: 2008\\n\\nAbstract: We consider fragments of first-order logic over finite words. In particular, we deal with first-order logic with a restricted number of variables and with the lower levels of the alternation hierarchy. We use the algebraic approach to show decidability of expressibility within these fragments. As a byproduct, we survey several characterizations of the respective fragments. We give complete proofs for all characterizations and we provide all necessary background. Some of the proofs seem to be new and simpler than those which can be found elsewhere. We also give a proof of Simon's theorem on factorization forests restricted to aperiodic monoids because this is simpler and sufficient for our purpose.\",\n \"Title: Safety and Liveness Properties for Real Traces and a Direct Translation from LTL to Monoids\\nYear: 2002\\n\\nAbstract: For infinite words there are well-known characterizations of safety and liveness properties. We extend these results to real Mazurkiewicz traces. This is possible due to a result, which has been established recently: Every first-order definable real trace language is definable in linear temporal logic using future tense operators, only. We show that the canonical choice for a topological characterization of safety and liveness properties is given by the Scott topology. In this paper we use an algebraic approach where we work with aperiodic monoids. Therefore we also give a direct translation from temporal logic to aperiodic monoids which is of independent interest.\",\n \"Title: On distributed monitoring of asynchronous systems\\nYear: 2012\\n\\nAbstract: Distributed systems are notoriously difficult to understand and analyze in order to assert their correction w.r.t. given properties. They often exhibit a huge number of different behaviors, as soon as the active entities (peers, agents, processes, etc) behave in an asynchronous manner. Already the modelization of such systems is a non-trivial task, let alone their formal verification. The purpose of this paper is to discuss the problem of distributed monitoring on a simple model of finite-state distributed automata based on shared actions, called asynchronous automata. Monitoring is a question related to runtime verification: assume that we have to check a property $L$ against an unknown or very complex system $A$, so that classical static analysis is not possible. Therefore instead of model-checking a monitor is used, that checks the property on the underlying system at runtime. We are interested here in monitoring distributed systems modeled as asynchronous automata. It is natural to require that monitors should be of the same kind as the underlying system, so we consider here distributed monitoring. A distributed monitor does not have a global view of the system, therefore we propose the notion of locally monitorable trace language. Our main result shows that if the distributed alphabet of actions is connected and if $L$ is a set of infinite traces such that both $L$ and its complement $L^c$ are countable unions of locally safety languages, then $L$ is locally monitorable. We also show that over infinite traces, recognizable countable unions of locally safety languages are precisely the complements of deterministic languages.\",\n \"Title: A Note On The Existential Theory Of Equations In Plain Groups\\nYear: 2002\\n\\nAbstract: Based on a PSPACE-completeness result for free monoids with involution [4] it is shown that the existential theory of equations with rational constraints in plain groups is PSPACE-complete. As a corollary this settles a question from [16].\",\n \"Title: A Domain for Concurrent Termination: A Generalization of Mazurkiewicz Traces (Extended Abstract)\\nYear: 1995\\n\\nAbstract: This paper generalizes the concept of Mazurkiewicz traces to a description of a concurrent process, where a known prefix is given as a trace in a first component and a second alphabetic component yields some information about future actions. This allows to define a good semantic domain where the concatenation is continuous with respect to the Scott- and to the Lawson topology. For this we define the notion of - and of -trace. We show various mathematical results proving thereby the soundness of our approach. Our theory is a proper generalization of the theory of finite and infinite words (with explicit termination) and of the theory of finite and infinite (real and complex) traces. We make use of trace theory, domain theory, and topology.\",\n \"Title: LTL is expressively complete for Mazurkiewicz traces\\nYear: 2000\\n\\nAbstract: A long standing open problem in the theory of (Mazurkiewicz) traces has been the question whether LTL (linear temporal logic) is expressively complete with respect to the first order theory. We solve this problem positively for finite and infinite traces and for the simplest temporal logic, which is based only on next and until modalities. Similar results were established previously, but they were all weaker, since they used additional past or future modalities. Another feature of our work is that our proof is direct and does not use any reduction to the word case.\",\n \"Title: Solving Trace Equations Using Lexicographical Normal Forms\\nYear: 1997\\n\\nAbstract: . Very recently, the second author showed that the questionwhether an equation over a trace monoid has a solution or not is decidable[11,12]. In the original proof this question is reduced to the solvabilityof word equations with constraints, by induction on the size ofthe commutation relation. In the present paper we give another proof ofthis result using lexicographical normal forms. Our method is a directreduction of a trace equation system to a word equation system withregular...\",\n \"Title: Word Equations Over Graph Products\\nYear: 2008\\n\\nAbstract: For monoids that satisfy a weak cancellation condition, it is shown that the decidability of the existential theory of word equations is preserved under graph products. Furthermore, it is shown that the positive theory of a graph product of groups can be reduced to the positive theories of those factors, which commute with all other factors, and the existential theories of the remaining factors. Both results also include suitable constraints for the variables. Larger classes of constraints lead in many cases to undecidability results.\",\n \"Title: Local temporal logic is expressively complete for cograph dependence alphabets\\nYear: 2001\\n\\nAbstract: Recently, local logics for Mazurkiewicz traces are of increas- ing interest. This is mainly due to the fact that the satisabilit y problem has the same complexity as in the word case. If we focus on a purely lo- cal interpretation of formulae at vertices (or events) of a trace, then the satisabilit y problem of linear temporal logics over traces turns out to be PSPACE{complete. But now the dicult problem is to obtain expressive completeness results with respect to rst order logic. The main result of the paper shows such an expressive completeness result, if the underlying dependence alphabet is a cograph, i.e., if all traces are series parallel posets. Moreover, we show that this is the best we can expect in our setting: If the dependence alphabet is not a cograph, then we cannot express all rst order properties. Keywords Temporal logics, Mazurkiewicz traces, concurrency\",\n \"Title: Efficient algorithms for highly compressed data: The Word Problem in Higman's group is in P.\\nYear: 2012\\n\\nAbstract: Power circuits are data structures which support efficient algorithms for highly compressed integers. Using this new data structure it has been shown recently by Myasnikov, Ushakov and Won that the Word Problem of the one-relator Baumslag group is is decidable in polynomial time. Before that the best known upper bound was non-elementary. In the present paper we provide new results for power circuits and we give new applications in algorithmic group theory: 1. We define a modified reduction procedure on power circuits which runs in quadratic time thereby improving the known cubic time complexity. 2. We improve the complexity of the Word Problem for the Baumslag group to cubic time thereby providing the first practical algorithm for that problem. 3. The Word Problem of Higman's group is decidable in polynomial time. It is due to the last result that we were forced to advance the theory of power circuits.\",\n \"Title: Solvability of Equations in Free Partially Commutative Groups Is Decidable\\nYear: 2001\\n\\nAbstract: Trace monoids are well-studied objects in computer science where they serve as a basic algebraic tool for analyzing concurrent systems. The question whether the existential theory of trace equations is decidable has been solved positively in 1996. Free partially commutative groups (graph groups) generalize trace monoids in the sense that every element has an inverse. In this paper we show that the existential theory of equations over graph groups is decidable, too. Our decision procedure is non-elementary, but if a certain graph theoretical parameter is viewed as a constant, then we obtain a PSPACE-completeness result. Restricting ourselves to trace monoids we still obtain a better complexity result, as it was known previously.\",\n \"Title: On First-Order Fragments for Mazurkiewicz Traces\\nYear: 2007\\n\\nAbstract: Mazurkiewicz traces form a model for concurrency. Temporal logic and first-order logic are important tools in order to deal with the abstract behavior of such systems. Since typical properties can be described by rather simple logical formulas one is interested in logical fragments. One focus of this paper is unary temporal logic and first-order logic in two variables. Over words, this corresponds to the variety of finite monoids calledDA. However, overMazurkiewicz traces it is crucial whether traces are given as dependence graphs or as partial orders (over words these notions coincide). The main technical contribution is a generalization of important characterizations of DA from words to dependence graphs, whereas the use of partial orders leads to strictly larger classes. As a consequence we can decide whether a first-order formula over dependence graphs is equivalent to a first-order formula in two variables. The corresponding result for partial orders is not known. This difference between dependence graphs and partial orders also affects the complexity of the satisfiability problems for the fragments under consideration: for first-order formulas in two variables we prove an NEXPTIME upper bound, whereas the corresponding problem for partial orders leads to EXPSPACE. Furthermore, we give several separation results for the alternation hierarchy for first-order logic. It turns out that even for those levels at which one can express the partial order relation in terms of dependence graphs, the fragments over partial orders have more expressive power.\",\n \"Title: Trace Rewriting: Computing Normal Forms in Time O(n log n)\\nYear: 1996\\n\\nAbstract: . We develop an O(n log n) algorithm for computing normalforms in the case of finite weight-reducing trace rewriting systems withconnected left-hand sides. The time complexity of previously known algorithmssolving this problem has been square time in the worst-case.Keywords: Mazurkiewicz traces, rewriting modulo a partial commutation.IntroductionFree partially commutative monoids (trace monoids) have been introduced byCartier and Foata [6] in order to solve some combinatorial problems...\",\n \"Title: On “inherently context-sensitive” languages —an application of complexity cores\\nYear: 1991\\n\\nAbstract: The concept of an ''inherently context-sensitive language'' is introduced. It is shown that every context-sensitive language that is not in the Boolean closure of the context-free languages has a subset that is inherently context-sensitive.\",\n \"Title: Solving word problems in group extensions over infinite words\\nYear: 2011\\n\\nAbstract: Non-Archimedean words have been introduced as a new type of infinite words which can be investigated through classical methods in combinatorics on words due to a length function. The length function, however, takes values in the additive group of polynomials Z[t] (and not, as traditionally, in N), which yields various new properties. Non-Archimedean words allow to solve a number of algorithmic problems in geometric and algorithmic group theory. There is a connection to the first-order theory in free groups (Tarski Problems), too. In the present paper we provide a general method to use infinite words over a discretely ordered abelian group as a tool to investigate certain group extensions for an arbitrary group G. The central object is a group E(A, G) which is defined in terms of a non-terminating, but confluent rewriting system. The group G as well as some natural HNN-extensions of G embed into E(A, G) (and still \\\"behave like\\\" G), which makes it interesting to study its algorithmic properties. The main result characterizes when the Word Problem (WP) is decidable in all finitely generated subgroups of E(A, G). We show that this property holds if and only if the Cyclic Membership Problem \\\"u ∈ 〈v〉?\\\" is decidable for all v ∈ G. Our methods combine combinatorics on words, string rewriting, and group theory.\",\n \"Title: Fragments of First-Order Logic over Infinite Words\\nYear: 2009\\n\\nAbstract: We give topological and algebraic characterizations as well as language theoretic descriptions of the following subclasses of first-order logic FO[ω-languages: Σ2, FO2, FO2∩Σ2, and Δ2 (and by duality Π2 and FO2∩Π2). These descriptions extend the respective results for finite words. In particular, we relate the above fragments to language classes of certain (unambiguous) polynomials. An immediate consequence is the decidability of the membership problem of these classes, but this was shown before by Wilke (Classifying Discrete Temporal Properties. Habilitationsschrift, Universität Kiel, April 1998) and Bojańczyk (Lecture Notes in Computer Science, vol. 4962, pp. 172–185, 2008) and is therefore not our main focus. The paper is about the interplay of algebraic, topological, and language theoretic properties.\",\n \"Title: Complexity Results And The Growths Of Hairpin Completions Of Regular Languages (Extended Abstract)\\nYear: 2010\\n\\nAbstract: The hairpin completion is a natural operation on formal languages which has been inspired by molecular phenomena in biology and by DNA-computing. In 2009 we presented in [6] a (polynomial time) decision algorithm to decide regularity of the hairpin completion. In this paper we provide four new results: 1.) We show that the decision problem is NL-complete. 2.) There is a polynomial time decision algorithm which runs in time O(n(8)), this improves [6], which provided O(n(20)). 3.) For the one-sided case (which is closer to DNA computing) the time is O(n(2)), only. 4.) The hairpin completion is unambiguous linear context-free. This result allows to compute the growth (generating function) of the hairpin completion and to compare it with the growth of the underlying regular language.\",\n \"Title: Characterization of the expressive power of silent transitions in timed automata\\nYear: 1998\\n\\nAbstract: Timed automata are among the most widely studied models for real-time systems. Silent transitions, i.e., ε-transitions, have already been proposed in the original paper on timed automata by Alur and Dill [3]. We show that the class TL ε of timed languages recognized by automata with ε-transitions, is more robust and more expressive than the corresponding class TL without ε-transitions. We then focus on ε-transitions without reset, i.e. ε-transitions which do not reset clocks. We propose an algorithm to construct, given a timed automaton, an equivalent one without such transitions. This algorithm is in two steps, it first suppresses the cycles of ε-transitions without reset and then the remaining ones. Then, we prove that a timed automaton such that no ε-transition which resets clocks lies on any directed cycle, can be effectively transformed into a timed automaton without εtransitions. Interestingly, this main result holds under the assumption of non-Zenoness and it is false otherwise. To complete the picture, we exhibit a simple timed automaton with an ε-transition, which resets some clock, on a cycle and which is not equivalent to any ε-free timed automaton. To show this, we develop a promising new technique based on the notion of precise action. This paper presents a synthesis of the two conference communications [9] and [13].\",\n \"Title: Deciding regularity of hairpin completions of regular languages in polynomial time\\nYear: 2011\\n\\nAbstract: Hairpin completion is an operation on formal languages that has been inspired by hairpin formation in DNA biochemistry and by DNA computing. In this paper we investigate the one- and two-sided hairpin completion of regular languages. We solve an open problem from the literature by showing that the regularity problem for hairpin completions is decidable. Actually, we show that the problem is decidable in polynomial time if the input is specified by DFAs. Furthermore, we prove that the hairpin completion of regular languages is an unambiguous linear context-free language. Beforehand, it was known only that it is linear context-free. Unambiguity is a strong additional property because it allows to compute the growth function or the topological entropy. In particular, we can compare the growth of the hairpin completion with the growth of the defining regular languages. We show that the growth of the hairpin completion is exponential if and only if the growth of the underlying languages is exponential. Even if both growth functions are exponential, they can be as far apart as 2^@Q^(^n^) for the hairpin completion and 2^@Q^(^n^) for the defining regular languages. However, if the hairpin completion is still regular, then the hairpin completion and its underlying language have essentially the same growth and same topological entropy.\",\n \"Title: Solving word equations modulo partial commutations\\nYear: 1999\\n\\nAbstract: It is shown that it is decidable whether an equation over a free partially commutative monoid has a solution. We give a proof of this result using normal forms. Our method is a direct reduction of a trace equation system to a word equation system with regular constraints. Hereby we use the extension of Makanin's theorem on the decidability of word equations to word equations with regular constraints, which is due to Schulz.\",\n \"Title: Topology, monitorable properties and runtime verification.\\nYear: 2014\\n\\nAbstract: We review concepts like safety, liveness, and monitorability from a rigorous topological viewpoint. Thus, monitorability of an ω-language means that the boundary in the Cantor topology has an empty interior. We show that all ω-regular languages which are deterministic and co-deterministic are monitorable, but certain deterministic liveness properties like “infinitely many a's” cannot be written as a countable union of monitorable languages. We briefly discuss model checking with LTL, its three-valued variant LTL3 and monitor constructions based upon LTL3.\",\n \"Title: Omega-Rational Expressions with Bounded Synchronization Delay\\nYear: 2015\\n\\nAbstract: In 1965 Schützenberger published his famous result that star-free languages ( $\\\\operatorname{SF}$ ) and aperiodic languages ( $\\\\operatorname{AP}$ ) coincide over finite words, often written as $\\\\operatorname{SF}= \\\\operatorname {AP}$ . Perrin generalized $\\\\operatorname{SF} = \\\\operatorname{AP}$ to infinite words in the mid 1980s. In 1973 Schützenberger presented another (and less known) characterization of aperiodic languages in terms of rational expressions where the use of the star operation is restricted to prefix codes with bounded synchronization delay and no complementation is used. We denote this class of languages by $\\\\operatorname{SD}$ . In this paper, we present a generalization of $\\\\operatorname{SD}= \\\\operatorname{AP}$ to infinite words. This became possible via a substantial simplification of the proof for the corresponding result for finite words. Moreover, we show that $\\\\operatorname{SD}= \\\\operatorname{AP}$ can be viewed as more fundamental than $\\\\operatorname{SF}= \\\\operatorname{AP}$ in the sense that the classical 1965 result of Schützenberger and its 1980s extension to infinite words by Perrin are immediate consequences of $\\\\operatorname{SD}= \\\\operatorname{AP}$ .\",\n \"Title: A survey on the local divisor technique.\\nYear: 2016\\n\\nAbstract: Local divisors allow a powerful induction scheme on the size of a monoid. We survey this technique by giving several examples of this proof method. These applications include linear temporal logic, rational expressions with Kleene stars restricted to prefix codes with bounded synchronization delay, Church–Rosser congruential languages, and Simon's Factorization Forest Theorem. We also introduce the notion of a localizable language class as a new abstract concept which unifies some of the proofs for the results above.\",\n \"Title: Equations over free inverse monoids with idempotent variables.\\nYear: 2014\\n\\nAbstract: We introduce the notion of idempotent variables for studying equations in inverse monoids. It is proved that it is decidable in singly exponential time (DEXPTIME) whether a system of equations in idempotent variables over a free inverse monoid has a solution. Moreover the problem becomes hard for DEXPTIME, as soon as the quotient group of the free inverse monoid has rank at least two. The upper bound is proved by a direct reduction to solve language equations with one-sided concatenation and a known complexity result by Baader and Narendran (J. Symb. Comput. 31, 277---305 2001). For the lower bound we show hardness for a restricted class of language equations. Decidability for systems of typed equations over a free inverse monoid with one irreducible variable and at least one unbalanced equation is proved with the same complexity upper-bound. Our results improve known complexity bounds by Deis et al. (IJAC 17, 761---795 2007). Our results also apply to larger families of equations where no decidability has been previously known. The lower bound confirms a conjecture made in the conference version of this paper.\",\n \"Title: Characterizing Classes Of Regular Languages Using Prefix Codes Of Bounded Synchronization Delay\\nYear: 2016\\n\\nAbstract: This paper is motivated by the work of Schutzenberger on codes with bounded synchronization delay. He was interested in characterizing those regular languages where groups in the syntactic monoid belong to a variety H. On the language side he allowed the operations union, intersection, concatenation and modified Kleene-star involving a mapping of a prefix code of bounded synchronization delay to a group G is an element of H, but no complementation. In our notation, this leads to the language classes SDH(A8). Our main result shows that SDH(A8) coincides with the class of languages having syntactic monoids where all subgroups are in H. We show that this statement holds for all varieties H of finite groups, whereas Schutzenberger proved this result for varieties H containing Abelian groups, only. Our method shows the result for all H simultaneously on finite and infinite words. Furthermore, we introduce the notion of local Rees extension which refers to a restricted type of the classical Rees extension. We give a decomposition of a monoid in terms of its groups and local Rees extensions. This gives a somewhat similar, but simpler decomposition than in the synthesis theorem of Rhodes and Allen. Moreover, we need a singly exponential number of operations, only. Finally, our decomposition yields an answer to a question in a recent paper of Almeida and Klima about varieties that are closed under Rees extensions.\",\n \"Title: Solution sets for equations over free groups are EDT0L languages.\\nYear: 2016\\n\\nAbstract: We show that, given a word equation over a finitely generated free group, the set of all solutions in reduced words forms an EDT0L language. In particular, it is an indexed language in the sense of Aho. The question of whether a description of solution sets in reduced words as an indexed language is possible has been open for some years [9, 10], apparently without much hope that a positive answer could hold. Nevertheless, our answer goes far beyond: they are EDT0L, which is a proper subclass of indexed languages. We can additionally handle the existential theory of equations with rational constraints in free products $$\\\\\\\\star _{1 \\\\\\\\le i \\\\\\\\le s}F_i$$, where each $$F_i$$ is either a free or finite group, or a free monoid with involution. In all cases the result is the same: the set of all solutions in reduced words is EDT0L. This was known only for quadratic word equations by [8], which is a very restricted case. Our general result became possible due to the recent recompression technique of Je﾿. In this paper we use a new method to integrate solutions of linear Diophantine equations into the process and obtain more general results than in the related paper [5]. For example, we improve the complexity from quadratic nondeterministic space in [5] to quasi-linear nondeterministic space here. This implies an improved complexity for deciding the existential theory of non-abelian free groups: $$\\\\\\\\mathsf {NSPACE}n\\\\\\\\log n$$. The conjectured complexity is $$\\\\\\\\mathsf {NP}$$, however, we believe that our results are optimal with respect to space complexity, independent of the conjectured $$\\\\\\\\mathsf {NP}$$.\",\n \"Title: Solutions of Word Equations over Partially Commutative Structures.\\nYear: 2016\\n\\nAbstract: We give NSPACE(n*log(n)) algorithms solving the following decision problems. Satisfiability: Is the given equation over a free partially commutative monoid with involution (resp. a free partially commutative group) solvable? Finiteness: Are there only finitely many solutions of such an equation? PSPACE algorithms with worse complexities for the first problem are known, but so far, a PSPACE algorithm for the second problem was out of reach. Our results are much stronger: Given such an equation, its solutions form an EDT0L language effectively representable in NSPACE(n*log(n)). In particular, we give an effective description of the set of all solutions for equations with constraints in free partially commutative monoids and groups.\",\n \"Title: Amenability of Schreier graphs and strongly generic algorithms for the conjugacy problem.\\nYear: 2015\\n\\nAbstract: In various occasions the conjugacy problem in finitely generated amalgamated products and HNN extensions can be decided efficiently for elements which cannot be conjugated into the base groups. Thus, the question arises how many such elements there are. This question can be formalized using the notion of strongly generic sets and lower bounds can be proven by applying the theory of amenable graphs.In this work we examine Schreier graphs of amalgamated products and HNN extensions. For an amalgamated product G=HAK with [H:A][K:A]2, the Schreier graph with respect to H or K turns out to be non-amenable if and only if [H:A]3. Moreover, for an HNN extension of the form G=H,b|bab1=(a),aA, we show that the Schreier graph of G with respect to the subgroup H is non-amenable if and only if AH(A).As application of these characterizations we show that the conjugacy problem in fundamental groups of finite graphs of groups with finitely generated free abelian vertex groups can be solved in polynomial time on a strongly generic set. Furthermore, the conjugacy problem in groups with more than one end can be solved with a strongly generic algorithm which has essentially the same time complexity as the word problem. These are rather striking results as the word problem might be easy, but the conjugacy problem might be even undecidable. Finally, our results yield a new proof that the set where the conjugacy problem of the Baumslag group G1,2 is decidable in polynomial time is also strongly generic.\",\n \"Title: More Than 1700 Years of Word Equations\\nYear: 2015\\n\\nAbstract: Geometry and Diophantine equations have been ever-present in mathematics. According to the existing literature the work of Diophantus of Alexandria was mentioned before 364 AD, but a systematic mathematical study of word equations began only in the 20th century. So, the title of the present article does not seem to be justified at all. However, a Diophantine equation can be viewed as a special case of a system of word equations over a unary alphabet, and, more importantly, a word equation can be viewed as a special case of a Diophantine equation. Hence, the problem WordEquations: \\\"Is a given word equation solvable?\\\", is intimately related to Hilbert's 10th problem on the solvability of Diophantine equations. This became clear to the Russian school of mathematics at the latest in the mid 1960s, after which a systematic study of that relation began. Here, we review some recent developments which led to an amazingly simple decision procedure for WordEquations, and to the description of the set of all solutions as an EDT0L language.\",\n \"Title: A Note on Monitors and Büchi automata.\\nYear: 2015\\n\\nAbstract: When a property needs to be checked against an unknown or very complex system, classical exploration techniques like model-checking are not applicable anymore. Sometimes a﾿monitor can be used, that checks a given property on the underlying system at runtime. A monitor for a property L is a deterministic finite automaton $$\\\\\\\\mathcal {M}_L$$ that after each finite execution tells whether 1 every possible extension of the execution is in L, or 2 every possible extension is in the complement of L, or neither 1 nor 2 holds. Moreover, L being monitorable means that it is always possible that in some future the monitor reaches 1 or 2. Classical examples for monitorable properties are safety and cosafety properties. On the other hand, deterministic liveness properties like \\\\\\\"infinitely many a's\\\\\\\" are not monitorable. We discuss various monitor constructions with a focus on deterministic $$\\\\\\\\omega $$-regular languages. We locate a proper subclass of deterministic $$\\\\\\\\omega $$-regular languages but also strictly larger than the subclass of languages which are deterministic and codeterministic; and for this subclass there exist canonical monitors which also accept the language itself. We also address the problem to decide monitorability in comparison with deciding liveness. The state of the art is as follows. Given a Büchi automaton, it is PSPACE-complete to decide liveness or monitorability. Given an LTL formula, deciding liveness becomes EXPSPACE-complete, but the complexity to decide monitorability remains open.\",\n \"Title: Solutions of twisted word equations, EDT0L languages, and context-free groups.\\nYear: 2017\\n\\nAbstract: We prove that the full solution set of a twisted word equation with regular constraints is an EDT0L language. It follows that the set of solutions to equations with rational constraints in a context-free group (= finitely generated virtually free group) in reduced normal forms is EDT0L. We can also decide whether or not the solution set is finite, which was an open problem. Moreover, this can all be done in PSPACE. Our results generalize the work by Lohrey and Senizergues (ICALP 2006) and Dahmani and Guirardel (J. of Topology 2010) with respect to complexity and with respect to expressive power. Both papers show that satisfiability is decidable, but neither gave any concrete complexity bound. Our results concern all solutions, and give, in some sense, the optimal formal language characterization.\",\n \"Title: Asymptotic Approximation for the Quotient Complexities of Atoms.\\nYear: 2015\\n\\nAbstract: In a series of papers, Brzozowski together with Tamm, Davies, andSzykuła studied the quotient complexities of atoms of regularlanguages. The authors obtained precise bounds in terms of binomialsums for the most complex situations in the following five cases:G: general, R: right ideals, L: left ideals, T:two-sided ideals and S: suffix-free languages. In each caselet κCn be the maximal complexity of an atom of aregular language L, where L has complexity n≧2 and belongsto the class C∈{G,R,L,T,S}. It is known thatκTn ≦ κLn = κRn ≦ κGn< 3n and κSn = κLn-1. We show that the ratioκCn / κCn-1 tends exponentiallyfast to 3 in all five cases but it remains different from 3.This behaviour was suggested by experimental results of Brzozowskiand Tamm; and the result for G was shown independently by LukeSchaeffer and the first author soon after the paper of Brzozowskiand Tamm appeared in 2012. However, proofs for the asymptotic behaviorof κGn / κGn-1 were never published;and the results here are valid for all five classes above. Moreover,there is an interesting oscillation for all C: for almost alln we have κCn / κCn-1 >3$ if and only ifκCn+1 / κCn<3.\",\n \"Title: Church-Rosser Systems, Codes with Bounded Synchronization Delay and Local Rees Extensions.\\nYear: 2017\\n\\nAbstract: What is the common link, if there is any, between Church-Rosser systems, prefix codes with bounded synchronization delay, and local Rees extensions? The first obvious answer is that each of these notions relates to topics of interest for WORDS: Church-Rosser systems are certain rewriting systems over words, codes are given by sets of words which form a basis of a free submonoid in the free monoid of all words (over a given alphabet) and local Rees extensions provide structural insight into regular languages over words. So, it seems to be a legitimate title for an extended abstract presented at the conference WORDS 2017. However, this work is more ambitious, it outlines some less obvious but much more interesting link between these topics. This link is based on a structure theory of finite monoids with varieties of groups and the concept of local divisors playing a prominent role. Parts of this work appeared in a similar form in conference proceedings [6, 10] where proofs and further material can be found.\",\n \"Title: A Remark on Trace Equations\\nYear: 1997\\n\\nAbstract: Without Abstract\",\n \"Title: Finding All Solutions of Equations in Free Groups and Monoids with Involution.\\nYear: 2014\\n\\nAbstract: This paper presents a PSPACE algorithm which yields a finite graph of exponential size that describes the set of all solutions of equations in free groups as well as the set of all solutions of equations with rational constraints in free monoids. This became possible due to the recent recompression technique.While this technique was successfully applied for pure word equations without involution or rational constraints it could not be used as a black box for free groups. Actually, the presence of an involution and rational constraints complicates the situation and some additional analysis is necessary. Still, the technique is general enough to accommodate both extensions. In the end, it simplifies proofs that satisfiability of word equations is in PSPACE and the corresponding result for equations in free groups with rational constraints. As a byproduct we can decide in PSPACE whether the solution set is finite.\",\n \"Title: Pure future local temporal logics are expressively complete for Mazurkiewicz traces\\nYear: 2006\\n\\nAbstract: The paper settles a long standing problem for Mazurkiewicz traces: the pure future local temporal logic defined with the basic modal- ities exists-next and until is expressively complete. The analogous result with a global interpretation was solved some years ago by Thiagarajan and Walukiewicz (1997) and in its final form without any reference to past tense constants by Diekert and Gastin (2000). Each, the (previously known) global or the (new) local result generalizes Kamp's Theorem for words, because for sequences local and global viewpoints coincide. But traces are labelled partial orders and then the difference between an inter- pretation globally over cuts (configurations) or locally at points (events) is significant. For global temporal logics the satisfiability problem is non- elementary (Walukiewicz 1998), whereas for local temporal logics both the satisfiability problem and the model checking problem are sovable in Pspace (Gastin and Kuske 2003) as in the case of words. This makes local temporal logics much more attractive.\",\n \"Title: Reachability Games and Parity Games\\nYear: 2022\\n\\nAbstract: Parity games are positionally determined. This is a fundamental and classical result. In 2010, Calude et al. showed a breakthrough result for finite parity games: the winning regions and their positional winning strategies can be computed in quasi-polynomial time. In the present paper we give a self-contained and detailed proofs for both results. The results in this paper are not meant to be original. The positional determinacy result is shown for possibly infinite parity games using the ideas of Zielonka which he published in 1998. In order to show quasi-polynomial time, we follow Lehtinen's register games, which she introduced in 2018. Although the time complexity of Lehtinen's algorithm is not optimal, register games are conceptually simple and interesting in their own right. Various of our proofs are either new or simplifications of the original proofs. The topics in this paper include the definition and the computation of optimal attractors for reachability games, too.\",\n \"Title: Properties of graphs specified by a regular language\\nYear: 2022\\n\\nAbstract: Traditionally, graph algorithms get a single graph as input, and then they should decide if this graph satisfies a certain property \\n \\n \\n \\n $$\\\\varPhi $$\\n \\n . What happens if this question is modified in a way that we get a possibly infinite family of graphs as an input, and the question is if there is a graph satisfying \\n \\n \\n \\n $$\\\\varPhi $$\\n \\n in the family? We approach this question by using formal languages for specifying families of graphs, in particular by regular sets of words. We show that certain graph properties can be decided by studying the syntactic monoid of the specification language L if a certain torsion condition is satisfied. This condition holds trivially if L is regular. More specifically, we use a natural binary encoding of finite graphs over a binary alphabet \\n \\n \\n \\n $$\\\\varSigma $$\\n \\n , and we define a regular set \\n \\n \\n \\n $$\\\\mathbb {G}\\\\subseteq \\\\varSigma ^*$$\\n \\n such that every nonempty word \\n \\n \\n \\n $$w\\\\in \\\\mathbb {G}$$\\n \\n defines a finite and nonempty graph. Also, graph properties can then be syntactically defined as languages over \\n \\n \\n \\n $$\\\\varSigma $$\\n \\n . Then, we ask whether the automaton \\n \\n \\n \\n $$\\\\mathcal {A}$$\\n \\n specifies some graph satisfying a certain property \\n \\n \\n \\n $$\\\\varPhi $$\\n \\n . Our structural results show that we can answer this question for all “typical” graph properties. In order to show our results, we split L into a finite union of subsets and every subset of this union defines in a natural way a single finite graph F where some edges and vertices are marked. The marked graph in turn defines an infinite graph \\n \\n \\n \\n $$F^\\\\infty $$\\n \\n and therefore the family of finite subgraphs of \\n \\n \\n \\n $$F^\\\\infty $$\\n \\n where F appears as an induced subgraph. This yields a geometric description of all graphs specified by L based on splitting L into finitely many pieces; then using the notion of graph retraction, we obtain an easily understandable description of the graphs in each piece.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":85,"string":"85"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13559,"string":"13,559"}}},{"rowIdx":13560,"cells":{"question_id":{"kind":"string","value":"Radu Prodan"},"question":{"kind":"string","value":"Generate an abstract for the title \"DEE: a distributed fault tolerant workflow enactment engine for grid computing\" using the following items: 1. Workflow management system design\n2. Scalable executions\n3. Grid computing\n4. Distributed workflow enactment engine\n5. Experimental results"},"answer":{"kind":"string","value":"It is a complex task to design and implement a workflow management system that supports scalable executions of large-scale scientific workflows for dynamic and heterogeneous Grid environments. In this paper we describe the Distributed workflow Enactment Engine (DEE) of the ASKALON Grid application development environment for Grid computing. DEE proposes a de-centralized architecture that simplifies and reduces the overhead for managing large workflows through partitioning, improved data locality, and reduced workflow-level checkpointing overhead. We report experimental results for a real-world material science workflow application."},"correct_answer":{"kind":"string","value":"It is a complex task to design and implement a workflow management system that supports scalable executions of large-scale scientific workflows for dynamic and heterogeneous Grid environments. In this paper we describe the Distributed workflow Enactment Engine (DEE) of the ASKALON Grid application development environment for Grid computing. DEE proposes a de-centralized architecture that simplifies and reduces the overhead for managing large workflows through partitioning, improved data locality, and reduced workflow-level checkpointing overhead. We report experimental results for a real-world material science workflow application."},"documents":{"kind":"list like","value":["Title: A new business model for massively multiplayer online games\nYear: 2011\n\nAbstract: Today, highly successful Massively Multiplayer Online Games (MMOGs) have millions of registered users and hundreds of thousands of active concurrent players. To sustain their highly variable load, game operators over-provision a large static infrastructure capable of sustaining the game peak load, even though a large portion of the resources is unused most of the time. This inefficient resource utilisation has negative economic impacts by preventing any but the largest hosting centres from joining the market and dramatically increases prices. In this paper, we propose a new business model of hosting and operating MMOGs based on Cloud computing principles involving four actors: resource provider, game operator, game provider, and client. Our model efficiently provisions on-demand virtualised resources to game sessions based on their dynamic client load, which dramatically decreases prices and gives small and medium enterprises the opportunity of joining the market through zero initial investment. We validate our new model and its underlying business relationships through trace-based simulations utilising six months worth of monitoring data from a real-life MMOG using emulated resources from 16 of the largest Cloud resource providers currently on the market. We demonstrate that our model can operate state-of-the-art MMOGs with an average monthly gross profit of nearly $6 million excluding game purchase prices, overheads and taxation, while being able to maintain and control the QoS offered to all clients. Finally, we show how our approach is capable of operating next generation very highly interactive MMOGs with a small increase of 5.8% in the subscription price.","Title: ZENTURIO: A Grid Service-Based Tool for Optimising Parallel and Grid Applications\nYear: 2004\n\nAbstract: ZENTURIO (R. Prodan and T. Fahringer, \"ZENTURIO: A Grid Middleware-based Tool for Experiment Man- agement of Parallel and Distributed Applications\", Journal of Parallel and Distributed Computing, 2003. http://authors.elsevier.com/sd/article/S0743731503001977 (to appear)) is a semi-automatic experiment manage- ment tool for performance and parameter studies of parallel and distributed applications on cluster and Grid architectures. ZENTURIO has been designed as an Open Grid Services Architecture (OGSA) - compliant Grid application built on top of standard Web and Grid services technologies. In this paper we first comparatively present various issues from our transition to an Open Grid Services Infrastructure (OGSI) - compliant prototype. Then we introduce a generic framework for solving NP-complete optimisation problems for parallel and Grid applications. We present a case study on high throughput scheduling for large sets of computational tasks on the Grid using genetic algorithms. Our algorithm has a complexity of O(n2) and delivers a fivefold improvement in solution over 500 generations in a Grid with uniformly distributed computational resources.","Title: Edutain@Grid: A Business Grid Infrastructure for Real-Time On-Line Interactive Applications\nYear: 2008\n\nAbstract: Grid infrastructures are maturing to a point where they are attracting the interest of businesses in many application domains. While many large-scale on-line gaming platforms exist, they fail to take into consideration the potential business to business relationships when it comes to dynamic on-line game hosting. This work presents an initial implementation of the edutain@grid architecture to support business value chains identified for on-line gaming and e-learning application hosting. An analysis of business actors and value chains is presented briefly before a detailed description of the edutain@grid implementation. We also consider first results concerning how best to construct appropriate value chains using bipartite and bi-directional Service Level Agreements.","Title: On Using ZENTURIO for Performance and Parameter Studies on Cluster and Grid Architectures\nYear: 2003\n\nAbstract: Over the last decade, a dramatic increase has been ob- served in the need for generating and organising data in the course of large parameter studies, performance analy- sis, and software testings. We have developed the ZENTU- RIO experiment management tool for performance and pa- rameter studies on cluster and Grid architectures. In this paper we describe our experience with ZENTURIO for per- formance and parameter studies of a material science ker- nel, a three-dimensional particle-in-cell simulation, a fast fourier transform, and a financial modeling application. Ex- periments have been conducted on an SMP cluster with Fast Ethernet and Myrinet communication networks, using PBS (Portable Batch System) and GRAM (Globus Resource Allo- cation Manager) as job managers.","Title: Dynamic Resource Provisioning in Massively Multiplayer Online Games\nYear: 2011\n\nAbstract: Today's Massively Multiplayer Online Games (MMOGs) can include millions of concurrent players spread across the world and interacting with each other within a single session. Faced with high resource demand variability and with misfit resource renting policies, the current industry practice is to overprovision for each game tens of self-owned data centers, making the market entry affordable only for big companies. Focusing on the reduction of entry and operational costs, we investigate a new dynamic resource provisioning method for MMOG operation using external data centers as low-cost resource providers. First, we identify in the various types of player interaction a source of short-term load variability, which complements the long-term load variability due to the size of the player population. Then, we introduce a combined MMOG processor, network, and memory load model that takes into account both the player interaction type and the population size. Our model is best used for estimating the MMOG resource demand dynamically, and thus, for dynamic resource provisioning based on the game world entity distribution. We evaluate several classes of online predictors for MMOG entity distribution and propose and tune a neural network-based predictor to deliver good accuracy consistently under real-time performance constraints. We assess using trace-based simulation the impact of the data center policies on the quality of resource provisioning. We find that the dynamic resource provisioning can be much more efficient than its static alternative even when the external data centers are busy, and that data centers with policies unsuitable for MMOGs are penalized by our dynamic resource provisioning method. Finally, we present experimental results showing the real-time parallelization and load balancing of a real game prototype using data center resources provisioned using our method and show its advantage against a rudimentary client threshold approach.","Title: ZENTURIO: a grid middleware-based tool for experiment management of parallel and distributed applications\nYear: 2004\n\nAbstract: So far there is very limited tool support for performance-oriented code development, software testing, and parameter studies of high-performance scientific applications. Such an effort commonly requires to conduct a large number of experimental executions. We have developed the ZENTURIO experiment management tool for multi-experimental parameter and performance studies of parallel and distributed applications on cluster and Grid architectures. We describe the design and implementation of ZENTURIO as a Grid application based on state-of-the-art Grid middleware technologies, such as Web services standards and the Globus toolkit. We offer solutions for building Open Grid Services Architecture-compliant Grid services using a pure Web services implementation toolkit. Existing Web services standards such as Universal Description Discovery and Integration and WSDL compatibility are refined for a more coherent integration with the Grid characteristics. We compare our design and implementation with an early prototype built on top of the OGSI technology. We report results on using ZENTURIO for performance study of a material science kernel and parameter study of a financial application.","Title: The edutain@grid Project\nYear: 2007\n\nAbstract: edutain@grid is an exciting and ground breaking new project making use of Grid technology. The project will identify and define\n a new class of applications that are highly significant for Grid computing but have not been studied in the past, which we\n characterise as Real-Time Online Interactive Applications (ROIA). The distinctive features that make ROIA unique include large\n user concurrency to a single application instance, ad-hoc connections, competition-oriented Virtual Organisations, real-time\n interactive response, dynamically changing control and data application flows whilst maintaining high Quality of Service (QoS),\n user friendly security, and novel Business-to-Consumer market models. In order to meet these challenges, the project team\n will develop a new middleware layer that will allow ROIA to exploit Grid computing and validate the system using two pilot\n applications from online gaming and e-learning domains.","Title: Negotiation-Based Scheduling of Scientific Grid Workflows Through Advance Reservations\nYear: 2010\n\nAbstract: In its broadest sense, scheduling of Grid applications can be viewed as a negotiation process between a scheduling service optimising user-centric objectives such as execution time, and a resource manager optimising provider-centric metrics such as resource utilisation or fairness. In this paper we enhance an existing list scheduling algorithm designed for minimising the workflow makespan with advance reservation-based negotiation functionality. As an instantiation of the new negotiation phase, we investigate two advance reservation functionality from the resource provider perspective: attentive and progressive. We illustrate through real-world experiments a two-fold benefit of our approach: improved execution predictability from the user's perspective, and higher resource utilisation fairness through a new progressive allocation strategy from the provider's perspective.","Title: A Lightweight C++ Interface to MPI\nYear: 2012\n\nAbstract: The Message Passing Interface (MPI) provides bindings for the three programming languages commonly used in High Performance Computing (HPC): C, C++ and Fortran. Unfortunately, MPI supports only the lowest common denominator of the three languages, providing a level of abstraction far lower than typical C++ libraries. Lately, after the decision of the MPI committee to deprecate and remove the C++ bindings from the MPI standard, programmers are forced to use either the C API or rely on third-party libraries. In this paper we present a lightweight, header-only C++ interface to MPI which uses object oriented and generic programming concepts to improve its integration into the C++ programming language. We compare our wrapper with a related approach called Boost. MPI showing how MPP facilitates the interaction with C++ objects. Performance wise, MPP outperforms Boost. MPI by reducing the interface overhead by a factor of eight. Additionally, MPP's handling of user-defined data types allows transferring of STL containers (e.g. std::list) up to 20 times faster than Boost. MPI for small linked lists by relying on software serialization.","Title: Meeting Soft Deadlines in Scientific Workflows Using Resubmission Impact\nYear: 2012\n\nAbstract: We propose a new heuristic called Resubmission Impact to support fault tolerant execution of scientific workflows in heterogeneous parallel and distributed computing environments. In contrast to related approaches, our method can be effectively used on new or unfamiliar environments, even in the absence of historical executions or failure trace models. On top of this method, we propose a dynamic enactment and rescheduling heuristic able to execute workflows with a high degree of fault tolerance, while taking into account soft deadlines. Simulated experiments of three real-world workflows in the Austrian Grid demonstrate that our method significantly reduces the resource waste compared to conservative task replication and resubmission techniques, while having a comparable makespan and only a slight decrease in the success probability. On the other hand, the dynamic enactment method manages to successfully meet soft deadlines in faulty environments in the absence of historical failure trace information or models.","Title: Applying Advance Reservation to Increase Predictability of Workflow Execution on the Grid\nYear: 2006\n\nAbstract: In this paper we present an extension to devise and implement advance reservation as part of the scheduling and resource management services of the ASKALON Grid application development and runtime environment. The scheduling service has been enhanced to offer a list of resources that can execute a specific task and to negotiatewith the resource manager about resources capable of processing tasks in the shortest possible time. We introduce progressive reservation approach which tries to allocate resources based on a fair-share principle. Experiments are shown that demonstrate the effectiveness of our approach, and that reflect different QoS parameters including performance, predictability, resource usage and resource fairness.","Title: Autonomous massively multiplayer online game operation on unreliable resources\nYear: 2013\n\nAbstract: Massively Multiplayer Online Games (MMOGs) are a new type of large-scale distributed applications characterised by seamless virtual worlds in which millions of world-wide players act and interact in real-time. Although for the past decade the number of MMOG players has grown exponentially, to the current tens of millions, this very growth may now hamper the progress of this important branch of the entertainment business. To guarantee Quality of Service (QoS) to a highly variable number of concurrent users, operators statically over-provision a large infrastructure capable of sustaining the game peak load, even though a large portion of the resources is unused most of the time. To address this problem, we propose a Cloud middleware-based system for autonomous operation of MMOGs. Our system provisions resources on-demand from multiple Cloud providers, automatically distributes the MMOG load between these resources, and self-heals when confronted with unforeseen resource failures. This new operational model allows small and medium enterprises to join the competitive MMOG market through near-zero initial infrastructure investment and operate MMOGs at given levels of QoS with small human intervention. We evaluate through simulations based on real-life MMOG traces the impact of resource availability on the QoS offered to the MMOG clients. We find that: (1) our proposed MMOG operation system can mitigate the negative effects of resource failures in under four minutes, (2) MMOG server consolidation in a resource-scarce environment can accentuate the negative effects of resource failures, and (3) the competition for resources can indirectly affect the QoS of the MMOG sessions.","Title: A Sequential Cooperative Game Theoretic Approach to Storage-Aware Scheduling of Multiple Large-Scale Workflow Applications in Grids\nYear: 2012\n\nAbstract: Scheduling large-scale applications in heterogeneous Grid and Cloud systems is a fundamental NP-complete problem for obtaining good performance and execution costs. We address the problem of scheduling an important class of large-scale Grid applications inspired from real-world, characterised by a large number of homogeneous, concurrent, and computationally-intensive tasks that are the main sources of performance, cost, and storage bottlenecks. We propose a new formulation of this problem based on a cooperative distributed game theoretic method for makespan and cost optimisation of a multiple such applications while fulfilling important storage constraints. We present experimental results using simulation and real-world applications that demonstrate the effectiveness of our method in terms of the solution delivered, algorithm execution time, and fairness compared to other related approaches.","Title: MOHEFT: A multi-objective list-based method for workflow scheduling\nYear: 2012\n\nAbstract: Nowadays, scientists and companies are confronted with multiple competing goals such as makespan in high-performance computing and economic cost in Clouds that have to be simultaneously optimized. Multi-objective scheduling of scientific workflows in distributed systems is therefore receiving increasing research attention. Most existing approaches typically aggregate all objectives in a single function, defined a-priori without any knowledge about the problem being solved, which negatively impacts the quality of the solutions. In contrast, Pareto-based approaches having as outcome a set of several (nearly-) optimal solutions that represent a tradeoff among the different objectives, have been scarcely studied. In this paper, we propose a new Pareto-based list scheduling heuristic that provides the user with a set of tradeoff optimal solutions from where the one that better suits the user requirements can be manually selected. We demonstrate the potential of MOHEFT for a bi-objective scheduling problem that optimizes makespan and economic cost in a Cloud-based computing scenario. We compare MOHEFT with two state-of-the-art approaches using different synthetic and real-world workflows: the classical HEFT algorithm used in single-objective scheduling and the SPEA2∗ genetic algorithm used for multi-objective optimisation problems.","Title: A new business model for massively multiplayer online games (abstracts only)\nYear: 2011\n\nAbstract: Today, highly successfulMassivelyMultiplayer Online Games (MMOGs) have millions of registered users and hundreds of thousands of active concurrent players. To sustain their highly variable load, game operators over-provision a large static infrastructure capable of sustaining the game peak load, even though a large portion of the resources is unused most of the time. This inefficient resource utilisation has negative economic impacts by preventing any but the largest hosting centres from joining the market and dramatically increases prices. In this paper, we propose a new business model of hosting and operating MMOGs based on Cloud computing principles involving four actors: resource provider, game operator, game provider, and client. Our model efficiently provisions on-demand virtualised resources to game sessions based on their dynamic client load, which dramatically decreases prices and gives small and medium enterprises the opportunity of joining the market through zero initial investment. We validate our new model and its underlying business relationships through trace-based simulations utilising six months worth of monitoring data from a real-life MMOG using emulated resources from 16 of the largest Cloud resource providers currently on the market. We demonstrate that our model can operate state-of-the-art MMOGs with an average monthly gross profit of nearly $6 million excluding game purchase prices, overheads and taxation, while being able to maintain and control the QoS offered to all clients. Finally, we show how our approach is capable of operating next generation very highly interactive MMOGs with a small increase of 5.8% in the subscription price.","Title: Kalipy: A Tool for Online Performance Analysis of Grid Workflows through Event Correlation\nYear: 2006\n\nAbstract: Static scheduling and execution of Grid workflows is prone to severe performance losses due to inaccurate predictions or the dynamic nature of the Grid environment. In this paper we present an online tool for analysing the performance overheads that appear during the real-time execution of workflow applications in Grid environments. We employ event correlation techniques and a distributed superpeer architecture in which each peer correlates local lowlevel activity and middleware events to infer performance overheads related to larger workflow regions at a higher level of abstraction. The rule-based correlation technique provides full extensibility to our approach that requires no source code modification. We demonstrate the functionality of our tool through online performance analysis of a real-world workflow application executed in a Grid environment. We present automatically generated online graphs of correlated events that promptly signal to the end-users the real reasons of run-time performance overheads in their executions.","Title: Double Auction-based Scheduling of Scientific Applications in Distributed Grid and Cloud Environments\nYear: 2011\n\nAbstract: Economy models have long been considered as a promising complement to the classical distributed resource management not only due of their dynamic and decentralized nature, but also because the concept of financial valuation of resources and services is an inherent part of any such model. In its broadest sense, scheduling of scientific applications in distributed Grid and Cloud environments can be regarded as a market-based negotiation between a scheduling service optimizing user-centric objectives (execution time, budget), and a resource manager optimizing provider-centric metrics (resource utilization, income, job throughput). In this paper, we propose a new instantiation of the negotiation protocol between the scheduler and resource manager using a market-based Continuous Double Auction (CDA) model. We analyze different scheduling strategies that can be applied and identify general strategic patterns that can lead to a fast and cheap work ow execution. In the experimental study, we demonstrate that under certain circumstances one can benefit by applying an aggressive scheduling strategy.","Title: Dynamic Real-Time Resource Provisioning for Massively Multiplayer Online Games\nYear: 2009\n\nAbstract: Massively Multiplayer Online Games (MMOG) are a class of computationally-intensive client-server applications with severe real-time Quality of Service (QoS) requirements, such the number of updates per second each client needs to receive from the servers for a fluent and realistic experience. To guarantee the QoS requirements, game providers currently over-provision a large amount of their resources, which makes the overall efficiency of provisioning and utilisation of resources rather low and prohibits any but the largest providers from joining the market. To address this deficiency, we propose a new prediction-based method for dynamic resource provisioning and scaling of MMOGs in distributed Grid environments. Firstly, a load prediction service anticipates the future game world entity distribution from historical trace data using a fast and flexible neural network-based method. On top of it, we developed generic analytical game load models used to foresee future hot-spots that congest the game servers and make the overall environment fragmented and unplayable. Finally, a resource allocation service performs dynamic load distribution, balancing, and migration of entities that keep the game servers reasonably loaded such that the real-time QoS requirements are maintained.","Title: A novel graph based approach for automatic composition of high quality grid workflows\nYear: 2009\n\nAbstract: The workflow paradigm is one of the most important programming models for the Grid. The composition of Grid workflows has been widely studied in the Grid community. However, there is still a lack of a general and efficient approach for automatic composition of Grid workflows. In this paper, we present a STRIPS (Stanford Research Institute Problem Solver) based formal definition of the Grid workflow composition problem, followed by a novel graph based algorithm for automatic composition of high quality (portable, fault tolerant and optimized) Grid workflows. Our algorithm searches for semantic descriptions of workflow activities, i.e., Activity Functions (AFs), defined by ontologies and composes them into Grid workflows using AF Data Dependence (ADD) graphs. The composition process consists of three phases: ADD graph creation, workflow extraction, and workflow optimization. The worst case complexity of our algorithm is quadratic in the number of AFs. An extension of our algorithm to compose Grid workflows with branches and loops is also presented. Experimental results illustrate the effectiveness and efficiency of our approach: (i) the measured worst case execution time of our algorithm further proofs the analyzed time complexity; (ii) the composition of the real world meteorology Grid workflow application MeteoAG with our algorithm takes approximate half a second; and (iii) the execution time of the MeteoAG workflow when running on the Austrian Grid is reduced by up to 25% and the speedup is increased by up to 2.24 by applying our workflow optimization techniques.","Title: Tuning MPI Runtime Parameter Setting for High Performance Computing\nYear: 2012\n\nAbstract: The performance of MPI applications on parallel computers can be considerably improved by tuning the runtime parameters provided by modern MPI libraries. However, due to the large and increasing number of tunable parameters, finding a parameter setting which optimizes the execution of several user programs on a chosen target machine is challenging. Existing tools execute input programs multiple times with varying parameter settings until a satisfying performance is reached. Several hundred runs of the input programs are nevertheless needed making this approach appealing only when the cost of the tuning phase can be amortized over many runs of the optimized programs. In this paper, we introduce a novel technique for tuning MPI runtime parameter values to better suit the underlying system architecture. The MPI parameter values are determined by performing the analysis of variance (ANOVA) on experimental data collected by randomly exploring the optimization space of a set of computational kernels commonly employed in High Performance Computing (HPC). We use our new technique to derive optimized values for 27 runtime parameters of the Open MPI library for two different parallel architectures. Results show an average performance improvement up to 20% for codes from the SPEC MPI 2007 benchmark suite with respect to Open MPI's default parameter setting.","Title: Leveraging C++ meta-programming capabilities to simplify the message passing programming model\nYear: 2011\n\nAbstract: Message passing is the primary programming model utilized for distributed memory systems. Because it aims at performance, the level of abstraction is low, making distributed memory programming often difficult and error-prone. In this paper, we leverage the expressivity and meta-programming capabilities of the C++ language to raise the abstraction level and simplify message passing programming. We redefine the semantics of the assignment operator to work in a distributed memory fashion and leave to the compiler the burden of generating the required communication operations. By enforcing more severe checks at compile-time we are able to statically capture common programming errors without causing runtime overhead.","Title: On the Evaluation of JavaSymphony for Heterogeneous Multi-core Clusters.\nYear: 2010\n\nAbstract: Programming hybrid heterogeneous multi-core cluster architectures is today an important topic in scientific and mainstream communities. To address this challenge, we developed JavaSymphony providing high-level programming abstraction and a middle-ware that facilitates the development and high-performance execution of Java applications on modern shared and distributed memory architectures. In this paper we present results of programming and executing a three-dimensional ray tracing application on a heterogeneous many-core cluster architecture.","Title: Optimizing performance of automatic training phase for application performance prediction in the grid\nYear: 2007\n\nAbstract: Automatic execution time prediction of the Grid applications plays a critical role in making the pervasive Grid more reliable and predictable. However, automatic execution time prediction has not been addressed due to the diversity of the Grid applications, usability of an application in multiple contexts, dynamic nature of the Grid, and concerns about result accuracy and time expensive experimental training. We introduce an optimized, low-cost, and efficient yet automatic training phase for automatic execution time prediction of Grid applications. Our approach is supported by intra - and inter-platform performance sharing and translation mechanisms. We are able to reduce the total number of experiments from an polynomial complexity to a linear complexity.","Title: Multi-objective list scheduling of workflow applications in distributed computing infrastructures\nYear: 2014\n\nAbstract: Executing large-scale applications in distributed computing infrastructures (DCI), for example modern Cloud environments, involves optimization of several conflicting objectives such as makespan, reliability, energy, or economic cost. Despite this trend, scheduling in heterogeneous DCIs has been traditionally approached as a single or bi-criteria optimization problem. In this paper, we propose a generic multi-objective optimization framework supported by a list scheduling heuristic for scientific workflows in heterogeneous DCIs. The algorithm approximates the optimal solution by considering user-specified constraints on objectives in a dual strategy: maximizing the distance to the user's constraints for dominant solutions and minimizing it otherwise. We instantiate the framework and algorithm for a four-objective case study comprising makespan, economic cost, energy consumption, and reliability as optimization goals. We implemented our method as part of the ASKALON environment (Fahringer et al., 2007) for Grid and Cloud computing and demonstrate through extensive real and synthetic simulation experiments that our algorithm outperforms related bi-criteria heuristics while meeting the user constraints most of the time.","Title: SLA-based operation of massively multiplayer online games in competition-based environments\nYear: 2013\n\nAbstract: To sustain the variable load of Massively Multiplayer Online Games (MMOGs) with guaranteed Quality of Service (QoS), game operators over-provision a static infrastructure capable of sustaining the peak load, even though a large portion of the resources is unused most of the time. This inefficient way of provisioning resources has negative impacts, leading to inefficient resource utilisation, high service prices, and limited market participation accessible only to the large companies. We propose a new ecosystem and model for hosting and operating MMOGs based on cloud computing principles involving four smaller and better focused business actors whose interaction is regulated through Service Level Agreements (SLAs): resource provider, game operator, game provider, and client. In our model, game providers lease operation SLAs from the game operators to satisfy all client requests and manage multiple distributed MMOG sessions. In turn, game operators efficiently lease on-demand cloud resources based on the dynamic MMOG load and ensure proper game operation that maintains QoS to all clients. In this paper, we focus on the business interaction between the game provider and the game operator by defining the SLA terms and the underlying negotiation protocol, including a model for compensations for QoS violations. We propose a method for ranking operational offers based on price, compensation and resource fitness, and study its impact on game provider's profit in an environment with several providers competing for SLAs from multiple game operators.","Title: Askalon: A Tool Set For Cluster And Grid Computing\nYear: 2005\n\nAbstract: Performance engineering of parallel and distributed applications is a complex task that iterates through various phases, ranging from modeling and prediction, to performance measurement, experiment management, data collection, and bottleneck analysis. There is no evidence so far that all of these phases should/can be integrated into a single monolithic tool. Moreover, the emergence of computational Grids as a common single wide-area platform for high-performance computing raises the idea to provide tools as interacting Grid services that share resources, support interoperability among different users and tools, and, most importantly, provide omnipresent services over the Grid. We have developed the ASKALON tool set to support performance-oriented development of parallel and distributed (Grid) applications. ASKALON comprises four tools, coherently integrated into a service-oriented architecture. SCALEA is a performance instrumentation, measurement, and analysis tool of parallel and distributed applications. ZENTURIO is a general purpose experiment management tool with advanced support for multi-experiment performance analysis and parameter studies. AKSUM provides semi-automatic highlevel performance bottleneck detection through a special-purpose performance property specification language. The PerformanceProphet enables the user to model and predict the performance of parallel applications at the early stages of development. In this paper we describe the overall architecture of the ASKALON tool set and outline the basic functionality of the four constituent tools. The structure of each tool is based on the composition and sharing of remote Grid services, thus enabling tool interoperability. In addition, a data repository allows the tools to share the common application performance and output data that have been derived by the individual tools. A service repository is used to store common portable Grid service implementations. A general-purpose Factory service is employed to create service instances on arbitrary remote Grid sites. Discovering and dynamically binding to existing remote services is achieved through registry services. The ASKALON visualization diagrams support both online and postmortem visualization of performance and output data. We demonstrate the usefulness and effectiveness of ASKALON by applying the tools to real-world applications. Copyright (C) 2005 John Wiley Sons, Ltd.","Title: Execution time prediction for grid infrastructures based on runtime provenance data\nYear: 2013\n\nAbstract: An accurate performance prediction service can be very useful for resource management and the scheduler service and help them make better resource utilization decisions by providing better execution time estimates. In this paper we present a novel approach of predicting the execution time of computational tasks for Grid infrastructures using machine learning models based on multilayer perceptron combined with a principal feature selection algorithm for selecting the most important runtime features. Our technique uses runtime provenance information as input to multilayer perceptron (MLP) based neural network which trains a model that can predict the execution time of programs with reasonably good accuracy. For the development and training of our machine learning models, we used provenance data collected from three different computational Grids by executing real-world applications. By using our MLP based method we were able to minimise the prediction error to as low as 22% for real-world applications on various Grid infrastructures.","Title: Using a new event-based simulation framework for investigating resource provisioning in Clouds\nYear: 2011\n\nAbstract: Today, Cloud computing proposes an attractive alternative to building large-scale distributed computing environments by which resources are no longer hosted by the scientists' computational facilities, but leased from specialised data centres only when and for how long they are needed. This new class of Cloud resources raises new interesting research questions in the fields of resource management, scheduling, fault tolerance, or quality of service, requiring hundreds to thousands of experiments for finding valid solutions. To enable such research, a scalable simulation framework is typically required for early prototyping, extensive testing and validation of results before the real deployment is performed. The scope of this paper is twofold. In the first part we present GroudSim, a Grid and Cloud simulation toolkit for scientific computing based on a scalable simulation-independent discrete-event engine. GroudSim provides a comprehensive set of features for complex simulation scenarios from simple job executions on leased computing resources to file transfers, calculation of costs and background load on resources. Simulations can be parameterised and are easily extendable by probability distribution packages for failures which normally occur in complex distributed environments. Experimental results demonstrate the improved scalability of GroudSim compared to a related process-based simulation approach. In the second part, we show the use of the GroudSim simulator to analyse the problem of dynamic provisioning of Cloud resources to scientific workflows that do not benefit from sufficient Grid resources as required by their computational demands. We propose and study four strategies for provisioning and releasing Cloud resources that take into account the general leasing model encountered in today's commercial Cloud environments based on resource bulks, fuzzy descriptions and hourly payment intervals. We study the impact of our techniques to the overall execution time, overall cost and cost per unit of saved time with respect to various instance types offered by the Amazon EC2.","Title: Scientific computing with Google App Engine\nYear: 2013\n\nAbstract: We investigate the potential of Google App Engine (GAE) for scientific computing. We designed a generic master-slave framework that enables implementation and integration of new algorithms by instantiating one interface and two abstract classes. Applications are dynamically scheduled, executed, and monitored by the framework on the Google infrastructure, including a fault tolerant resubmission mechanism upon certain failures. We describe the implementation of a parallel rank sort algorithm using our framework and analyse the speedup, the execution overheads, and the cost of executing it on the Google Cloud infrastructure. Experimental results show that important speedup can be obtained from using GAE, especially for embarrassingly parallel Monte Carlo simulations. The two main obstacles that hinder the performance of applications on GAE are the resource quotas upon free use (especially the 30 s sequential execution time limit) and the middleware overheads. Cost-wise, GAE offers opportunities for cheaper computation due to the CPU cycle-based payment granularity, as opposed to the hourly billing intervals of other providers such as Amazon EC2.","Title: Characterizing, Modeling and Predicting Dynamic Resource Availability in a Large Scale Multi-purpose Grid\nYear: 2008\n\nAbstract: The functional heterogeneity of computational Grids has highly increased due to inclusion of resources other than dedicated to Grid, like from non-dedicated desktop Grids, on-demand systems and even from P2P systems and mobile Grids. At such a diversified scale, resources exhibit different availability properties mainly due to administrators' policies for resource availability in the Grid, and their failure/unavailability properties. These make resources' availability predictions for optimized resource selection, a challenging problem. Addressing this problem, we characterize resource availability properties against their availability policies to understand their availability behavior and quantify it through availability models. We further exploit the availability/failure properties to make predictions about their availability through pattern recognition and classification. We have achieved, on average, accuracy of more than 90% and 75% in our predictions for resource instance availability and lifetime respectively.","Title: A Multi-objective Approach for Workflow Scheduling in Heterogeneous Environments\nYear: 2012\n\nAbstract: Traditional scheduling research usually targets make span as the only optimization goal, while several isolated efforts addressed the problem by considering at most two objectives. In this paper we propose a general framework and heuristic algorithm for multi-objective static scheduling of scientific workflows in heterogeneous computing environments. The algorithm uses constraints specified by the user for each objective and approximates the optimal solution by applying a double strategy: maximizing the distance to the constraint vector for dominant solutions and minimizing it otherwise. We analyze and classify different objectives with respect to their impact on the optimization process and present a four-objective case study comprising make span, economic cost, energy consumption, and reliability. We implemented the algorithm as part of the ASKALON environment for Grid and Cloud computing. Results for two real-world applications demonstrate that the solutions generated by our algorithm are superior to user-defined constraints most of the time. Moreover, the algorithm outperforms a related bi-criteria heuristic and a bi-criteria genetic algorithm.","Title: A Distributed Object-Oriented Framework for Tool Development\nYear: 2000\n\nAbstract: In recent years, there has been a substantial increase in the availability and quality of software engineering tools; such tools are invaluable in ensuring program correctness and identifying performance problems. The majority of these, however, does not interoperate and are available on a limited platform set.We analyze such deficiencies and propose an extensible architecture for a distributed software engineering tool framework using CORBA object-oriented technology. The resulting framework provides a unified interface for parallel, distributed and single-processor systems, facilitates tool development, promotes tool interoperability, and can be extended by the integration of new tools and services. This flexibility is demonstrated by the specification of an extension to support the MPI programming paradigm and a wide selection of tools that have been built using the system.","Title: GroudSim: an event-based simulation framework for computational grids and clouds\nYear: 2010\n\nAbstract: We present GroudSim, a Grid and Cloud simulation toolkit for scientific applications based on a scalable simulation-independent discrete-event core. GroudSim provides a comprehensive set of features for complex simulation scenarios from simple job executions on leased computing resources to calculation of costs, and background load on resources. Simulations can be parameterised and are easily extendable by probability distribution packages for failures which normally occur in complex environments. Experimental results demonstrate the improved scalability of GroudSim compared to a related process-based approach.","Title: Multi-Objectiveworkflow Scheduling: An Analysis Of The Energy Efficiency And Makespan Tradeoff\nYear: 2013\n\nAbstract: While in the past scheduling algorithms were almost exclusively targeted at optimizing applications' makespan, today they must simultaneously optimise several goals. Among these goals, energy efficiency is receiving increasing attention for environmental and financial reasons. In contrast to related work that optimises energy consumption as a single objective function, we reformulate in this paper the problem as a bi-objective optimisation by considering both makespan and energy as goals. We study the potential benefits of using a Pareto-based workflow scheduling algorithm called MOHEFT using realistic energy consumption and performance models for task executions. We analyse the tradeoff solutions computed by MOHET for different workflows (different in shapes and sizes) in different execution scenarios (different resources in terms of energy consumption). The obtained results show that our bi-objective approach found in some cases schedules that reduce the energy consumption up to 85% with only 3.3% of makespan concessions.","Title: Budget-Constrained Resource Provisioning for Scientific Applications in Clouds\nYear: 2013\n\nAbstract: Public commercial clouds emerged as new and attractive resource provisioning option for scientific computing. This new alternative raises new challenges for users of such clouds, since optimizing the completion time of scientific applications might substantially increase the monetary cost of leasing cloud resources. In this paper, we first propose a set of basic rescheduling operations covering a broad set of scenarios for reducing the costs of running scientific workflows in clouds. Based on them, we design two heuristic scheduling algorithms. The first algorithm aims at reducing the cost of resource provisioning while still attaining the optimal make span. The second algorithm further reduces the costs to meet a budget constraint with a small increase in the make span. The experiments conducted using real-world and synthetic workflow applications demonstrate important benefits compared to related state-of-the-art approaches.","Title: Towards a general model of the multi-criteria workflow scheduling on the grid\nYear: 2009\n\nAbstract: Workflow scheduling on the Grid becomes more challenging when multiple scheduling criteria are considered. Existing studies provide different approaches to the multi-criteria Grid workflow scheduling problem, and address different variants of the problem. A profound understanding of the problem's nature can be an important step towards more generic scheduling approaches. Based on the related work and on our own experience, we propose several novel taxonomies of the problem, considering five facets: workflow model, scheduling criteria, scheduling process, resource model, and task model. We make a survey of the existing related work, and classify it according to the proposed taxonomies, identifying the most common use cases and the areas that have not been sufficiently explored yet.","Title: A sequential cooperative game theoretic approach to scheduling multiple large-scale applications in grids.\nYear: 2014\n\nAbstract: Scheduling large-scale applications in heterogeneous distributed computing systems is a fundamental NP-complete problem that is critical to obtaining good performance and execution cost. In this paper, we address the scheduling problem of an important class of large-scale Grid applications inspired by the real world, characterized by a huge number of homogeneous, concurrent, and computationally intensive tasks that are the main sources of performance, cost, and storage bottlenecks. We propose a new formulation of this problem based on a cooperative distributed game-theory-based method applied using three algorithms with low time complexity for optimizing three important metrics in scientific computing: execution time, economic cost, and storage requirements. We present comprehensive experiments using simulation and real-world applications that demonstrate the effectiveness of our approach in terms of time and fairness compared to other related algorithms. We schedule large-scale Grid applications with a huge identical parallel tasks.We formulate the problem using a cooperative distributed game-theory-based method.We design three algorithms for optimizing time, cost, and storage requirements.We evaluate our method using simulation and real-world applications.We compare our results with related algorithms.","Title: Prediction-based real-time resource provisioning for massively multiplayer online games\nYear: 2009\n\nAbstract: Massively Multiplayer Online Games (MMOGs) are a class of computationally intensive client-server applications with severe real-time Quality of Service (QoS) requirements, such as the number of updates per second each client needs to receive from the servers for a fluent and realistic experience. To guarantee the QoS requirements, game providers currently over-provision a large amount of their resources, which makes the overall efficiency of provisioning and utilization of resources rather low and prohibits any but the largest providers from joining the market. To address this deficiency, we propose a new prediction-based method for dynamic resource provisioning and scaling of MMOGs in distributed Grid environments. Firstly, a load prediction service anticipates the future game world entity distribution from historical trace data using a fast and flexible neural network-based method. On top of it, we developed generic analytical game load models used to foresee future hot-spots that congest the game servers and make the overall environment fragmented and unplayable. Finally, a resource allocation service performs dynamic load distribution, balancing, and migration of entities that keep the game servers reasonably loaded such that the real-time QoS requirements are maintained. Experimental results based on a realistic simulation environment demonstrate the advantages of our prediction service compared to other conventional methods, especially due to its ability to adapt to different user load patterns, and a reduction of the average over-allocation from 250% (in the case of static over-provisioning) to around 25% using our dynamic provisioning method.","Title: A Framework for an Interoperable Tool Environment (Research Note)\nYear: 2000\n\nAbstract: Software engineering tools are indispensable for parallel and distributed program development, yet the desire to combine them to provide enhanced functionality has still to be realised. Existing tool environments integrate a number of tools, but do not achieve interoperability and lack extensibility. Integration of new tools can necessitate the redesign of the whole system. We describe the FIRST tool framework, its initial tool-set, classify different types of tool interaction and describe a variety of tool scenarios which are being used to investigate tool interoperability.","Title: Monitoring and fault tolerance for real-time online interactive applications\nYear: 2009\n\nAbstract: The edutain@grid European project [1] is developing a support platform for deployment, management and execution of Real-Time Online Interactive Applications (ROIA) on Grid. In this paper we present a monitoring system we developed which collects data from all the resources in a distributed environment and from the ROIA managed by our platform. We also describe a fault tolerance service which addresses not only the faults commonly encountered in distributed systems, but also faults manifesting at service level, within the platform's management services. Finally, a use-case consisting of the platform running a massively multiplayer online game as a concrete ROIA, is presented in order to demonstrate the roles of the monitoring and fault tolerance services.","Title: Meteorological simulations in the cloud with the ASKALON environment\nYear: 2012\n\nAbstract: Precipitation in mountainous regions is an essential process in meteorological research for its strong impact on the hydrological cycle. To support scientists, we present the design of a meteorological application using the ASKALON environment comprising graphical workflow modeling and execution in a Cloud computing environment. We illustrate performance results that demonstrate that, although limited by Amdahl's law, our workflow can gain important speedup when executed in a virtualized Cloud environment with important operational cost reductions. Results from the meteorological research show the usefulness of our model for determining precipitation distribution in the case of two field campaigns over Norway.","Title: Semi-automatic composition of ontologies for ASKALON grid workflows\nYear: 2011\n\nAbstract: Automatic workflow composition with the help of ontologies has been addressed by numerous researchers in the past. While ontologies are very useful for automatic and semiautomatic workflow composition, ontology creation itself remains a very important and complex task. In this paper we present a novel tool to synthesize ontologies for the Abstract Grid Workflow Language (AGWL) which has been used for years to successfully create Grid workflow applications at a high level of abstraction. In order to semi-automatically generate ontologies we use an AGWL Ontology (AGWO - an ontological description of the AGWL language), structural information of one or several input workflows of a given application domain, and semantic enrichment of the structural information with the help of the user. Experiments based on two separate application domains (movie rendering and meteorology) will be shown that demonstrate the effectiveness of our approach by semi-automatically generating ontologies which are then used to automatically create workflow applications.","Title: The impact of virtualisation on the performance and operational costs of Massively Multiplayer Online Games\nYear: 2010\n\nAbstract: To serve millions of players, Massively Multiplayer Online Game (MMOG) operators pre-provision and then maintain thousands of computer resources. We investigate a hybrid resource provisioning model that uses smaller and cheaper data centers, complemented during peak hours by virtualised cloud computing resources. Through trace-based simulation and empirical experimentation, we assess the impact of provisioning virtualised cloud resources, analyse the virtualisation overhead, and compare provisioning of virtualised resources with resource ownership. Using a simple cost model, we also investigate the costs of hosting MMOGs on the resources leased independently from three commercial cloud providers, including Amazon.","Title: A Web Service-Based Experiment Management System for the Grid\nYear: 2003\n\nAbstract: We have developed the ZENTURIO experiment management system for performance studies, parameter studies, and software testing of parallel applications on cluster and Grid architectures. In this paper we describe the design and implementation of ZENTURIO as a Web service-based distributed architecture. We highlight limitations of Web services for deployment on the Grid and propose solutions. Experiment Generator and Experiment Executor services are used to generate, respectively submit and manage experiments on remote Grid sites. A UDDI-based Service Repository publishes information about existing Web service implementations. Factory and Registry services are employed to create and register Web services on the Grid. A GMA-compliant event infrastructure supports high-level notifications about important events. We compare our design with the Open Grid Service Architecture (OGSA) and highlight similarities and differences. We report results of using ZENTURIO to conduct performance analysis of a material science code executed on the Grid by using GRAM.","Title: An Information System for Real-Time Online Interactive Applications\nYear: 2008\n\nAbstract: The edutain@grid European project [1] is developing a support platform for deployment, management and execution of Real-Time Online Interactive Applications (ROIA) on Grid. In this paper we present an information system designed by the edutain@grid project which provides support for ROIA deployment and monitoring, and offers a generic frontend for ROIA-specific optimisations. We conduct a variety of experiments that justify various decisions of our design, and investigate the performance and scalability of our system with respect to various types of queries.","Title: A Bi-Criteria Truthful Mechanism for Scheduling of Workflows in Clouds\nYear: 2011\n\nAbstract: Commercial distributed systems such as Clouds are managed by selfish providers that strategically try to increase their revenues regardless of the utility of other providers and users. These selfish behaviors affect the efficiency of using such environments. In this paper, based on a general game theoretic truthful reverse auction mechanism, we investigate the scheduling problem of dependent tasks on distributed Cloud resources owned by selfish providers. The social cost of the game is to minimize the make span and monetary cost simultaneously. Extensive simulation experiments show that the schedules obtained are approximately Pareto optimal.","Title: A Performance Analysis of EC2 Cloud Computing Services for Scientific Computing.\nYear: 2009\n\nAbstract: Cloud Computing is emerging today as a commercial infrastructure that eliminates the need for maintaining expensive computing hardware. Through the use of virtualization, clouds promise to address with the same shared set of physical resources a large user base with different needs. Thus, clouds promise to be for scientists an alternative to clusters, grids, and supercomputers. However, virtualization may induce significant performance penalties for the demanding scientific computing workloads. In this work we present an evaluation of the usefulness of the current cloud computing services for scientific computing. We analyze the performance of the Amazon EC2 platform using micro-benchmarks and kernels. While clouds are still changing, our results indicate that the current cloud services need an order of magnitude in performance improvement to be useful to the scientific community.","Title: A Collection of Java Class Libraries for Stochastic Modeling and Simulation\nYear: 2002\n\nAbstract: The paper focuses on a set of Java class libraries for stochastic modeling and simulation, created and implemented by the Department of Mathematics and Informatics from Iuliu Ha驴ieganu University Cluj-Napoca. Previous research has shown that stochastic models are advantageous tools for representation of the real world activities. Due to actual spread of fast and inexpensive computational power everywhere in the world, the best approach is to model the real phenomenon as faithfully as possible, and then rely on a simulation study to analyze it. We approached the new bootstrapping methods, which are very useful in analyzing a simulation. Our aim is to implement bootstrapping strategies in software tools which will be used by both the teaching staff and the students in didactic and research activities, with the purpose to optimize the use of substances and reactants.","Title: Soft Benchmarks-Based Application Performance Prediction Using a Minimum Training Set\nYear: 2006\n\nAbstract: Application execution time prediction is of key importance in making decisions about efficient usage of Grid resources. Grid services lack support of a generic application execution time prediction service due to environment specific solutions provided by the existing prediction techniques. To remedy this, we present a generic and comprehensive system to provide execution time predictions of applications on different Grid-sites. Our system is based on a two layered training phase to minimize the training effort, which is our first main contribution. The training phase is driven by a novel experimental design. We also introduce a mechanism of sharing performance measurements across the Grid, on the basis of soft benchmarks, which is our second contribution. Both of these phases support our prediction engine to serve robust predictions. Experiments from the prototype implementation are shown to demonstrate the effectiveness of our proposed system.","Title: Overhead Analysis of Scientific Workflows in Grid Environments\nYear: 2008\n\nAbstract: Scientific workflows are a topic of great interest in the Grid community that sees in the workflow model an attractive paradigm for programming distributed wide-area Grid infrastructures. Traditionally, the Grid workflow execution is approached as a pure best-effort scheduling problem that maps the activities onto the Grid processors based on appropriate optimisation or local matchmaking heuristics such that the overall execution time is minimised. Even though such heuristics often deliver effective results, the execution in dynamic and unpredictable Grid environments is prone to severe performance losses that must be understood for minimising the completion time or for efficient use of high- erformance resources. In this paper, we propose a new systematic approach to help the scientists and middleware developers understand the most severe sources of performance losses that occur when executing scientific workflows in dynamic Grid environments. We introduce an ideal model for the lowest execution time that can be achieved by a workflow and explain the difference to the real measured Grid execution time based on a hierarchy of performance overheads for Grid computing. We describe how to systematically measure and compute the overheads from individual activities to larger workflow regions and adjust well-known parallel processing metrics to the scope of Grid computing, including speedup and efficiency. We present a distributed online tool for computing and analysing the performance overheads in realtime based on event correlation techniques and introduce several performance contracts, as quality of service parameters to be enforced during the workflow execution beyond traditional best-effort practices. We illustrate our method through post-mortem and online performance analysis of two real-world workflow applications executed in the Austrian Grid environment.","Title: JavaSymphony: a programming and execution environment for parallel and distributed many-core architectures\nYear: 2010\n\nAbstract: Today, software developers face the challenge of re-engineering their applications to exploit the full power of the new emerging many-core processors. However, a uniform high-level programming model and interface for parallelising Java applications is still missing. In this paper, we propose a new Java-based programming model for shared memory many-core parallel computers as an extension to the JavaSymphony distributed programming environment. The concept of dynamic virtual architecture allows modelling of hierarchical resource topologies ranging from individual cores and multi-core processors to more complex parallel computers and distributed Grid infrastructures. On top of this virtual architecture, objects can be explicitly distributed, migrated, and invoked, enabling high-level user control of parallelism, locality, and load balancing. We evaluate the JavaSymphony programming model and the new shared memory run-time environment for six real applications and benchmarks on a modern multi-core parallel computer. We report scalability analysis results that demonstrate that JavaSymphony outperforms pure Java implementations, as well as other alternative related solutions.","Title: Multi-objective workflow scheduling in Amazon EC2\nYear: 2014\n\nAbstract: Nowadays, scientists and companies are confronted with multiple competing goals such as makespan in high-performance computing and economic cost in Clouds that have to be simultaneously optimised. Multi-objective scheduling of scientific applications in these systems is therefore receiving increasing research attention. Most existing approaches typically aggregate all objectives in a single function, defined a-priori without any knowledge about the problem being solved, which negatively impacts the quality of the solutions. In contrast, Pareto-based approaches having as outcome a set of (nearly) optimal solutions that represent a tradeoff among the different objectives, have been scarcely studied. In this paper, we analyse MOHEFT, a Pareto-based list scheduling heuristic that provides the user with a set of tradeoff optimal solutions from which the one that better suits the user requirements can be manually selected. We demonstrate the potential of our method for multi-objective workflow scheduling on the commercial Amazon EC2 Cloud. We compare the quality of the MOHEFT tradeoff solutions with two state-of-the-art approaches using different synthetic and real-world workflows: the classical HEFT algorithm for single-objective scheduling and the SPEA2* genetic algorithm used in multi-objective optimisation problems. The results demonstrate that our approach is able to compute solutions of higher quality than SPEA2*. In addition, we show that MOHEFT is more suitable than SPEA2* for workflow scheduling in the context of commercial Clouds, since the genetic-based approach is unable of dealing with some of the constraints imposed by these systems.","Title: Performance Analysis and Benchmarking of the Intel SCC\nYear: 2011\n\nAbstract: Over the past years there has been a steady change in CPU design towards both many-core processors and power-aware hardware architectures. These two trends are combined in the Intel Single-chip Cloud Computer (SCC), an experimental prototype with 48 Pentium cores created by Intel Labs. The SCC is a highly configurable many-core chip which provides unique opportunities to optimize run time, communication and memory access as well as power/energy consumption of parallel programs. The aim of this paper is to characterize the performance behavior of the chip with various power settings, mappings of processes/cores to memory controllers, etc through benchmarking. Analytical models are used to verify and interpret the results. Conclusions drawn from our benchmark outcomes are that data exchange based on message passing is faster than shared memory data exchange. Contrary to popular belief, lowest energy consumption is not achieved for the fastest execution time. Furthermore in order to improve the memory access behavior one should increase the clock frequency of both, mesh network and memory controllers. In general, the results of our investigations can be used to analyze the effect of power settings and architecture properties on the performance and energy consumption of parallel programs as well as assist in choosing appropriate settings for specific workloads.","Title: Specification-correct and scalable coordination of Grid applications\nYear: 2007\n\nAbstract: The so called ''invisible Grid'', transparent to application developers, is still far from being a reality. One reason is that the workflow model, which emerged as a widely accepted paradigm for high-level composition of Grid applications, is based on a low-level imperative programming model prone to programming errors. The issue of developing correct (bug-free) Grid applications has not been addressed by the community. We propose a new model for building Grid applications based on two programming phases: (1) formal functional specification, written by the application developer not interested in any Grid-related issues; (2) imperative workflow-based coordination, written by the computer scientist, which ports and efficiently executes the specification onto the Grid. A correctness checker automatically connects both parts at compile-time and insures the correct execution of the workflow coordination with respect to the formal specification. We validate our approach for three scientific applications and show real-world experimental results that demonstrate the scalability of our coordination model and the fact that the overhead introduced by our correctness checker is insignificant when compared to the latencies exhibited by the Grid middleware software.","Title: Optimising parallel applications on the grid using irregular array distributions\nYear: 2005\n\nAbstract: In this paper we propose a new approach for scheduling data parallel applications on the Grid using irregular array distributions. We implement the scheduler as a new case study for using a general purpose experiment management tool that we developed in previous work for performance tuning and optimisation of scientific applications. We report results on scheduling a Jacobi relaxation application on a simulation testbed of the Austrian Grid [2] using a problem independent plug-and-play genetic algorithm.","Title: A Truthful Dynamic Workflow Scheduling Mechanism for Commercial Multicloud Environments\nYear: 2013\n\nAbstract: The ultimate goal of cloud providers by providing resources is increasing their revenues. This goal leads to a selfish behavior that negatively affects the users of a commercial multicloud environment. In this paper, we introduce a pricing model and a truthful mechanism for scheduling single tasks considering two objectives: monetary cost and completion time. With respect to the social cost of the mechanism, i.e., minimizing the completion time and monetary cost, we extend the mechanism for dynamic scheduling of scientific workflows. We theoretically analyze the truthfulness and the efficiency of the mechanism and present extensive experimental results showing significant impact of the selfish behavior of the cloud providers on the efficiency of the whole system. The experiments conducted using real-world and synthetic workflow applications demonstrate that our solutions dominate in most cases the Pareto-optimal solutions estimated by two classical multiobjective evolutionary algorithms.","Title: Bi-criteria Scheduling of Scientific Workflows for the Grid\nYear: 2008\n\nAbstract: The drift towards new challenges in grid computing, including the utility grid paradigm and service level agreements based on quality-of-service guarantees, implies the need for new, robust, multi-criteria scheduling algorithms that can be applied by the user in an intuitive way. Multiple scheduling criteria addressed by the related grid research include execution time, the cost of running a task on a machine, reliability, and different data quality metrics. The existing bi-criteria scheduling approaches are usually dedicated for certain criterion pairs only that require the user to define one's preferences either as weights assigned to the criteria or as fixed constraints defined for one of the criteria. These requirements are often not feasible for the user and not suited to the specificity of the multi- criteria scheduling problem. We propose a novel requirement specification method based on a sliding constraint, and we model the problem as an extension of the multiple-choice knapsack problem. We propose a general bi-criteria scheduling heuristic called dynamic constraint algorithm (DCA) based on dynamic programming, dedicated to the problem model defined by us. In the experimental study, we show that in most of the problem variants, DCA outperforms two existing algorithms designed for the same problem. It also shows relatively low scheduling times for workflows of medium size.","Title: Online Gaming In The Cloud\nYear: 2010\n\nAbstract: Computational Grids and Clouds remain highly specialized technologies that are only used by scientists and large commercial organizations. To overcome this gap, University of Innsbruck is conducting basic research that is unusual compared with previous academic research projects in that it addresses a new class of application that appeals to the public for leisure reasons: Massively Multiplayer Online Games. Online games have the potential to raise strong interest, providing societal benefits through increased technological awareness and engagement. By standardizing on a Cloud-based platform and removing the need for large investments in hosting facilities, this research may remove the technical barrier and the costs of hosting MMOGs, and thus significantly increase the number of players while keeping the high quality of responsiveness of action games.","Title: Specification-correct and Scalable Coordination of Scientific Applications in Grid Environments\nYear: 2007\n\nAbstract: The workflow model for composing Grid applications is based on an imperative model of computation prone to programming errors, which is an issue yet to consider in the Grid community. In this paper, we propose a new unconventional model for programming Grid applications based on two programming phases: (1) formal functional specification, written by the application scientist not interested in any Grid-related issues and representing \"what\" has to be computed; (2) imperative workflow-based coordination, written by the computer scientist which indicates \"how\" to efficiently execute the specification on the Grid. A correctness checker automatically connects both parts at compile- time and ensures the correct execution of the workflow coordination with respect to the formal specification. We validate our approach for three real-world applications and show experimental results that demonstrate the correctness and scalability of our coordination model.","Title: Scheduling of scientific workflows in the ASKALON grid environment\nYear: 2005\n\nAbstract: Scheduling is a key concern for the execution of performance-driven Grid applications. In this paper we comparatively examine different existing approaches for scheduling of scientific workflow applications in a Grid environment. We evaluate three algorithms namely genetic, HEFT, and simple \"myopic\" and compare incremental workflow partitioning against the full-graph scheduling strategy. We demonstrate experiments using real-world scientific applications covering both balanced (symmetric) and unbalanced (asymmetric) workflows. Our results demonstrate that full-graph scheduling with the HEFT algorithm performs best compared to the other strategies examined in this paper.","Title: IaaS cloud benchmarking: approaches, challenges, and experience\nYear: 2013\n\nAbstract: Over the past five years, Infrastructure-as-a-Service (IaaS) clouds have grown into the branch of ICT that offers services related to on-demand lease of storage, computation, and network. One of the major impediments in the selection and even use of (commercial) IaaS clouds is the lack of benchmarking results, that is, the lack of trustworthy quantitative information that allows (potential) cloud users to compare and reason about IaaS clouds. In this talk we discuss empirical approaches to quantitative evaluation of IaaS clouds, toward cloud benchmarking. Both industry and academia have used empirical approaches for years, but with limited success for IaaS clouds and similar systems (e.g., grids) due to the complexity and size of challenges. We present initial results of our research into cloud workload characterization, including Big Data applications. We present the lessons we have learned in developing the SkyMark framework for cloud performance evaluation and the results of our SkyMark-based investigation of three research questions: What is the performance of production IaaS cloud services? How variable is the performance of widely used production cloud services? and What is the impact on performance of the provisioning and allocation policies that interact with IaaS clouds? In contrast to previous attempts, our research combines empirical and other approaches, for example modeling and simulation, for gaining more insight; is based on a combination of short-term and multi-year measurements, for ensuring the longevity of our results; and uses large, comprehensive studies of several real clouds, for reducing the threats to validity related to the experimental environment. This presentation can also provide useful insights for fields related to benchmarking, for example experimental evaluation conducted in other large-scale distributed systems. Last but not least, we present a road-map toward cloud benchmarking and the way we plan to progress on it with other members of the RG Cloud WG of the Standard Performance Evaluation Corporation (SPEC) http://research.spec.org/working-groups/rg-cloud-working-group.html). Note: talk based on the homonym article [1].","Title: Integration of an event-based simulation framework into a scientific workflow execution environment for Grids and clouds\nYear: 2011\n\nAbstract: The utilisation of Grid and Cloud-based computing environments for solving scientific problems has become an increasingly used practice in the last decade. To ease the use of these global distributed resources, sophisticated middleware systems have been developed, enabling the transparent execution of applications by hiding low-level technology details from the user. The ASKALON environment is such a system, which supports the development and execution of distributed applications such as scientific workflows or parameter studies in Grid and Cloud computing environments. On the other hand, simulation is a widely accepted approach to analyse and further optimise the behaviour of software systems. Beside the advantage of enabling repeatable deterministic evaluations, simulations are able to circumvent the difficulties in setting up and operating multi-institutional Grid systems, thus providing a lightweight simulated distributed environment on a single machine. In this paper, we present the integration of the GroudSim Grid and Cloud event-based simulator into the ASKALON environment. This enables system, application developers, and users to perform simulations using their accustomed environment, thereby benefiting from the combination of an established real-world platform and the advantages of a simulation.","Title: Bringing Scientific Workflows to Amazon SWF\nYear: 2013\n\nAbstract: In response to the ever-increasing needs of scientific applications for resources, Cloud computing emerged as an alternative on-demand and cost-effective resource provisioning approach. In this context, Cloud providers have recognised the importance of workflow applications to science and provide their own native solutions, such as the Amazon Simple Workflow Service (SWF). Nevertheless, an important downside of SWF is its incompatibility with existing workflow systems, and lack of means for reusing scientific legacy code. Similarly, existing workflow middlewares and applications require non-trivial extensions to take advantage of Cloud resources. We present in this paper a software engineering solution that allows the scientific workflow community access the Amazon Cloud through one single front-end converter, and propose a legacy wrapper service for executing legacy code using SWF. Empirical results using a real-world scientific workflow demonstrate that our automatically generated SWF application performs almost as fast as a native manually-optimised version, and outperforms other workflow middleware systems using the Amazon Cloud.","Title: Comparison of workflow scheduling strategies on the grid\nYear: 2005\n\nAbstract: Scheduling is a key concern for the execution of performance-driven Grid applications. In this paper we comparatively examine different existing approaches for scheduling of scientific workflow applications in a Grid environment. We evaluate three algorithms namely genetic, HEFT, and simple “myopic” and compare incremental workflow partitioning against the full-graph scheduling strategy. We demonstrate experiments using real-world scientific applications covering both balanced (symmetric) and unbalanced (asymmetric) workflows. Our results demonstrate that full-graph scheduling with the HEFT algorithm performs best compared to the other strategies examined in this paper.","Title: The impact of virtualization on the performance of Massively Multiplayer Online Games\nYear: 2009\n\nAbstract: Today's highly successful Massively Multiplayer Online Games (MMOGs) have millions of registered users and hundreds of thousands of active concurrent users. As a result of the highly dynamic MMOG usage patterns, the MMOG operators pre-provision and then maintain throughout the lifetime of the game tens of thousands of compute resources in data centers located across the world. Until recently, the difficulty of porting the MMOG software services to different platforms made it impractical to dynamically provision resources external to the MMOG operators' data centers. However, virtualization is a new technology that promises to alleviate this problem by providing a uniform computing platform with minimal overhead. To investigate the potential of this new technology, in this paper we propose a new hybrid resource provisioning model that uses a smaller and less expensive set of self-owned data centers, complemented by virtualized cloud computing resources during peak hours. Using real traces from RuneScape, one of the most successful contemporary MMOGs, we evaluate with simulations the effectiveness of the on-demand cloud resource provisioning strategy for MMOGs. We assess the impact of provisioning of virtualized cloud resources, analyze the components of virtualization overhead, and compare provisioning of virtualized resources with direct provisioning of data center resources.","Title: Workflow Scheduling in Amazon EC2.\nYear: 2013\n\nAbstract: Workflow scheduling has been traditionally targeted to map the execution of set of tasks onto a set of resources for makespan minimization. With the increasing popularity of Cloud computing systems, the financial cost entailed for executing these tasks plays also an important role. Existing works have however combined both, makespan and cost, on a single function and no analysis of the tradeoff between both criteria has been produced. In addition, no work in the context a real commercial cloud system exists. This paper includes a comparison of two real multi-objective workflow scheduling, MOHEFT and SPEA2*, in the context of Amazon EC2. The carried experiments show that MOHEFT outperforms SPEA2*, and that the analysis of the tradeoff solutions can help in selecting good scheduling solutions.","Title: Real world workflow applications in the askalon grid environment\nYear: 2005\n\nAbstract: The workflow paradigm is widely considered as an important class of truly distributed Grid applications which poses many challenges for a Grid computing environment. Still to this time, rather few real-world applications have been successfully ported as Grid-enabled workflows. We present the Askalon programming and computing environment for the Grid which comprises a high-level abstract workflow language and a sophisticated service-oriented runtime environment including meta-scheduling, performance monitoring and analysis, and resource brokerage services.We demonstrate the development of a real-world river modelling distributed workflow system in the Askalon environment that harnesses the computational power of multiple Grid sites to optimise the overall execution time.","Title: Dynamic scheduling of scientific workflow applications on the grid: a case study\nYear: 2005\n\nAbstract: The existing Grid workflow scheduling projects do not handle recursive loops which are characteristic to many scientific problems. We propose a hybrid approach for scheduling Directed Graph (DG)-based workflows in a Grid environment with dynamically changing computational and network resources. Our dynamic scheduling algorithm is based on the iterative invocation of classical static Directed Acyclic Graphs (DAGs) scheduling heuristics generated using well-defined cycle elimination and task migration techniques. We approach the static scheduling problem as an application of a modular optimisation tool using genetic algorithms. We report successful implementation and experimental results on a pilot real-world material science workflow application.","Title: A Hybrid Intelligent Method for Performance Modeling and Prediction of Workflow Activities in Grids\nYear: 2009\n\nAbstract: Grid schedulers require individual activity performance predictions to map workflow activities on different Grid sites. The effectiveness of the scheduling systems is hampered by inaccurate predictions due to the inability of existing predictors to effectively model the dynamic and heterogeneous nature of Grid resources, or the wide range of problem sizes and runtime arguments. To address this deficiency, we propose a hybrid Bayesian-neural network approach to dynamically model and predict the execution time of activities in real workflow applications. Bayesian network is used for a high-level representation of activities performance probability distribution against different factors affecting the performance. The important attributes are dynamically selected by the Bayesian network and fed into a radial basis function neural network to make further predictions. Our approach is generic to any type of scientific applications, and flexible to import expert knowledge to further improve accuracies. Experimental results for activities from three realworld workflow applications are presented to show effectiveness of our approach.","Title: Impact of variable priced cloud resources on scientific workflow scheduling\nYear: 2012\n\nAbstract: We analyze the problem of provisioning Cloud instances to large scientific workflows that do not benefit from sufficient Grid resources as required by their computational requirements. We propose an extension to the dynamic critical path scheduling algorithm to deal with the general resource leasing model encountered in today's commercial Clouds. We analyze the availability of the cheaper and unreliable Spot instances and study their potential to complement the unavailability of Grid resources for large workflow executions. Experimental results demonstrate that Spot instances represent a 60% cheaper but equally reliable alternative to Standard instances provided that a correct user bet is made.","Title: Enhancing Grids for Massively Multiplayer Online Computer Games\nYear: 2008\n\nAbstract: Massively multiplayer online games (MMOG) are an innovative and challenging class of applications for Grid computing that require large amounts of computational resources for providing a responsive and scalable gameplay for concurrently participating players connected via Internet. We present our Real-Time Framework (RTF)--- a Grid-based middleware for scaling game sessions through a variety of parallelization and distribution techniques. RTF is described within a novel multi-layer service-oriented architecture that comprises three advanced services --- monitoring, capacity planning, and runtime steering --- that use the potential of Grid computing to provide pervasive access to a potentially unbounded number of resources. We report experimental results on the quality of our capacity planning and scalability of the RTF distribution mechanism.","Title: Neural Network-Based Load Prediction for Highly Dynamic Distributed Online Games\nYear: 2008\n\nAbstract: We propose a neural network-based prediction method for the future entity layout in massively multiplayer online games. Our service has the potential to timely foresee critical hot-spots in fast-paced First Person Shooter action games that saturate the game servers which no longer respond to user actions at the required rate. Using our service, proactive load balancing (and redistribution) actions can be triggered.We show results based on a realistic simulation environment that demonstrate the advantages of our method compared to other conventional ones, especially due to its ability to adapt to different load patterns.","Title: Scheduling javasymphony applications on many-core parallel computers\nYear: 2011\n\nAbstract: JavaSymphony is a Java-based programming and execution environment for programming and scheduling the performance oriented applications on multi-core parallel computers. In this paper, we present a multi-core aware scheduling extension to JavaSymphony capable of mapping parallel applications on large multi-core machines and heterogeneous clusters. JavaSymphony scheduler considers several multicore specific performance parameters and application types, and uses these parameters to optimise the mapping of JavaSymphony objects and tasks. We evaluate the performance of JavaSymphony scheduler using several real scientific applications and benchmarks on a multi-core shared memory machine and a heterogeneous cluster.","Title: A New Fault Tolerance Heuristic for Scientific Workflows in Highly Distributed Environments Based on Resubmission Impact\nYear: 2009\n\nAbstract: Even though highly distributed environments such as Clouds and Grids are increasingly used for e-Science high performance applications, they still cannot deliver the robustness and reliability needed for widespread acceptance as ubiquitous scientific tools. To overcome this problem, existing systems resort to fault tolerance mechanisms such as task replication and task resubmission. In this paper we propose a new heuristic called Resubmission Impact to enhance the fault tolerance support for scientific workflows in highly distributed systems. In contrast to related approaches, our method can be used effectively on systems even in the absence of historic failure trace data. Simulated experiments of three real scientific workflows in the Austrian Grid environment show that our algorithm drastically reduces the resource waste compared to conservative task replication and resubmission techniques, while having a comparable execution performance and only a slight decrease in the success probability.","Title: Online Analysis and Runtime Steering of Dynamic Workflows in the ASKALON Grid Environment\nYear: 2007\n\nAbstract: We present a new distributed performance analysis service of the ASKALON integrated Grid environment for computing runtime overheads of dynamic workflows in realtime based on event correlation techniques. We illustrate a formal method to express precise overhead correlation rules, including several performance contracts as quality of service parameters based on fuzzy logic to be enforced in dynamic environments though rescheduling, various runtime optimisations, and steering techniques. We demonstrate experimental results for two real applications from material chemistry and graphics rendering domains.","Title: Applying double auctions for scheduling of workflows on the Grid\nYear: 2008\n\nAbstract: Grid economy models have long been considered as a promising alternative for the classical Grid resource management, due to their dynamic and decentralized nature, and because the financial valuation of resources and services is inherent in any such model. In particular, auction models are widely used in the existing Grid research, as they are easy to implement and are shown to successfully manage resource allocation on the Grid market. The focus on the current work is on workflow scheduling in the Grid resource allocation model based on Continuous Double Auctions (CDA). We analyze different scheduling strategies that can be applied by the user to execute workflows in such an environment, and try to identify the general behavioral patterns that can lead to a fast and cheap workflow execution. In the experimental study, we show that under certain circumstances some benefit can be gained by applying an \"aggressive\" scheduling strategy.","Title: Overhead Analysis of Grid Workflow Applications\nYear: 2006\n\nAbstract: In this paper we propose a systematic approach to performance analysis of workflow applications on the Grid. We introduce an ideal model for the workflow execution time and explain the difference to the real measured times based on a hierarchy of performance overheads for Grid computing. We describe how to systematically measure and compute the overheads from individual activities to entire workflow applications. We adjusted well-known parallel processing metrics to the scope of Grid computing, comprising speedup and efficiency. We have implemented and largely automatised our analysis approach in the context of the ASKALON Grid application development and computing environment. We present experimental results that show detailed overhead analysis of two real-world workflow applications executed in a national Grid environment.","Title: IWIR: a language enabling portability across grid workflow systems.\nYear: 2011\n\nAbstract: Today there are many different scientific Grid workflow management systems using a wide array of custom workflow languages. Some of them are geared towards a data-based view, some are geared towards a control-flow based view and others try to be as generic, and therefore often complex, as possible. All of these languages and custom workflow management system front-ends fulfill special needs and workflow creation paradigms for their respective user communities. The problem is that once a workflow application has been created in one of these systems, it becomes very hard to share the workflow with users working with different systems. Portability and interoperability between current systems barely exists. In this work, we present a common workflow language for use as an intermediate exchange representation by multiple workflow systems. It comprises atomic tasks, compound tasks including conditionals, sequential and parallel loops as well as an expressive set of data types and data flow constructs.","Title: ZENTURIO: An Experiment Management System for Cluster and Grid Computing\nYear: 2002\n\nAbstract: The need to conduct and manage large sets of experiments for scientific applications dramatically increased over the last decade. However, there is still very little tool support for this complex and tedious process. In this paper we introduce the ZENTURIO experiment management system for parameter studies, performance analysis, and software testing for cluster and Grid architectures. ZENTURIO uses the ZEN directive-based language to specify arbitrary complex program executions. ZENTURIO is designed as a collection of Grid services that comprise: (1) a registry service which supports registering and locating Grid services; (2) an experiment generator that parses files with ZEN directives and instruments applications for performance analysis and parameter studies; (3) an experiment executor that compiles and controls the execution of experiments on the target machine. A graphical user portal allows the user to control and monitor the experiments and to automatically visualise performance and output data across multiple experiments. ZENTURIO has been implemented based on Java/Jini distributed technology. It supports experiment management on cluster architectures via PBS and on Grid infrastructures through GRAM. We report results of using ZENTURIO for performance analysis of an ocean simulation application and a parameter study of a computational finance code.","Title: Fine-Grain Interoperability of Scientific Workflows in Distributed Computing Infrastructures\nYear: 2013\n\nAbstract: Today there exist a wide variety of scientific workflow management systems, each designed to fulfill the needs of a certain scientific community. Unfortunately, once a workflow application has been designed in one particular system it becomes very hard to share it with users working with different systems. Portability of workflows and interoperability between current systems barely exists. In this work, we present the fine-grained interoperability solution proposed in the SHIWA European project that brings together four representative European workflow systems: ASKALON, MOTEUR, WS-PGRADE, and Triana. The proposed interoperability is realised at two levels of abstraction: abstract and concrete. At the abstract level, we propose a generic Interoperable Workflow Intermediate Representation (IWIR) that can be used as a common bridge for translating workflows between different languages independent of the underlying distributed computing infrastructure. At the concrete level, we propose a bundling technique that aggregates the abstract IWIR representation and concrete task representations to enable workflow instantiation, execution and scheduling. We illustrate case studies using two real-workflow applications designed in a native environment and then translated and executed by a foreign workflow system in a foreign distributed computing infrastructure.","Title: ZEN: a directive based experiment specification language for performance and parameter studies of parallel scientific applications\nYear: 2005\n\nAbstract: Currently there exists little support to specify and control the execution of a large number of experiments required for performance oriented development of scientific applications. We have developed a new directive based language called ZEN for the specification of wide sets of synthetic experiments in the context of large scale parameter and performance studies of parallel applications. We report experimental results for the performance analysis of an ocean simulation application and for the parameter study of a computational finance code.","Title: Performance and cost optimization for multiple large-scale grid workflow applications\nYear: 2007\n\nAbstract: Scheduling large-scale applications on the Grid is a fundamental challenge and is critical to application performance and cost. Large-scale applications typically contain a large number of homogeneous and concurrent activities which are main bottlenecks, but open great potentials for optimization. This paper presents a new formulation of the well-known NP-complete problems and two novel algorithms that addresses the problems. The optimization problems are formulated as sequential cooperative games among workflow managers. Experimental results indicate that we have successfully devised and implemented one group of effective, efficient, and feasible approaches. They can produce soultuins of significantly better performance and cost than traditional algorithms. Our algorithms have considerably low time complexity and can assign 1,000,000 activities to 10,000 processors within 0.4 second on one Opteron processor. Moreover, the solutions can be practically performed by workflow managers, and the violation of QoS can be easily detected, which are critical to fault tolerance.","Title: The first framework for interoperable run-time software tools\nYear: 2001\n\nAbstract: A variety of different program correctness tools are now available to help the software developer write, test and maintain faster, correctly conforming code. The majority of these software tools are independent in nature and can not be used in conjunction with other similar tools. Such tools are therefore not interoperable and, typically, the program must be recompiled with a different set of tool-dependent flags in order to move from one tool to another.The FIRST project uses CORBA to build an object-oriented framework for real-time software performance and correctness tools. These tools use framework-provided high-level tool services to generate instrumentation which is dynamically inserted into the already running target process.This paper describes in detail the FIRST tool development framework, classifies different types of tool interoperability and discusses various scenarios in which FIRST profiling and debugging tools can be combined to achieve synergies through interoperability.","Title: ZEN: A Directive-Based Language for Automatic Experiment Management of Distributed and Parallel Programs\nYear: 2002\n\nAbstract: So far there exists very little support to specify and to control execution of a large number of experiments on distributed and parallel architectures. This paper describes ZEN, a directive-based language for the specification of arbitrarily complex program executions by varying problem, system, or machine parameters for parallel and distributed applications. ZEN introduces directives to substitute strings and to insert assignment statements inside arbitrary files, such as program, input, script, or makefiles. The programmer thus can invoke experiments for arbitrary value ranges of any problem parameter, including program variables, file names, compiler options, target machines, machine sizes, scheduling strategies, data distributions, etc. The number of experiments can be controlled through ZEN constraintdirectives. Finally, the programmer may request a large set of performance metrics to be computed for any code region of interest. The scope of ZEN directives can be restrictedto arbitrary file or code regions. We have implemented a prototype tool for automatic experiment management that is based on ZEN. We will report results for performance analysisof an ocean simulation application and for parameter study of a computational finance code.","Title: From Web Services to OGSA: Experiences in Implementing an OGSA-based Grid Application\nYear: 2003\n\nAbstract: In previous work we have presented the ZENTURIO experimentmanagement system for performance and parameterstudies of parallel and distributed applications on clusterand Grid architectures. In this paper we describe experiencesof an on-going work, targeting the implementationof ZENTURIO on top of the Open Grid Services Architecture(OGSA). We analyse the opportunities offered by aWeb services toolkit to develop Grid services as required byOGSA and compare them with the solutions offered by theOpen Grid Services Infrastructure (OGSI) specification. Issuesregarding proxy management, service lifecycle, UDDIservice repository, firewall management, Factory and Registryservices, service throughput, and security are comparativelyanalysed in both implementations.","Title: Autonomic operation of massively multiplayer online games in clouds\nYear: 2013\n\nAbstract: To support the variable load of Massively Multiplayer On-line Games (MMOGs) with millions of registered users and thousands of active concurrent players, game operators over-provision a large static infrastructure capable of sustaining the peak load with guaranteed Quality of Service (QoS). This leads to inefficient resource utilisation, high service prices, and limited market participation accessible only to the large companies. To address this problem, we propose a new autonomic ecosystem for hosting and operating MMOGs based on cloud computing principles involving four smaller and better focused business actors whose interaction is regulated through Service Level Agreements (SLAs): resource provider, game operator, game provider, and client. In our model, game providers acquire operation SLAs from game operators to satisfy client requests and manage multiple distributed MMOG sessions. Game operators lease on-demand cloud resources based on the dynamic MMOG load and guarantee the required QoS to all clients. We evaluate through simulations based on real MMOG traces and commercial cloud SLAs different methods of ranking MMOG operation offers. We show that considering compensations for SLA faults in the offer selection can lead to over 11% gains in game providers' income, and that adequate ranking of offers can reduce operational costs by up to 60%.","Title: Run-time Optimisation of Grid Workflow Applications\nYear: 2006\n\nAbstract: The execution of workflow applications on the grid is a complex issue because of its dynamic and heterogeneous nature. While the grid provides good potential for achieving high performance, it also introduces a broad set of unpredictable overheads and possible failures. In this paper we present new methods for scalable and fault tolerant coordination of workflows in dynamic grid environments, including partitioning, static and dynamic optimisation, as well as virtual single execution environment, incorporated into the ASKALON distributed workflow enactment engine. We demonstrate the effectiveness of our methods on a material science workflow application executed in a real-world grid environment","Title: Dynamic load management for MMOGs in distributed environments\nYear: 2010\n\nAbstract: To support thousands of concurrent players in virtual worlds simulated by contemporary Massively Multiplayer Online Games, most implementations employ static game world partitioning for distributing the load among multiple game server instances. Further, the resources that manage the resulting subregions are statically allocated, independent of the actual game load. As a result, due to the high variability of the user demand, this approach leads to a low resource utilization causing much higher provisioning costs than necessary. In addition, the number of players supported by a region is limited by the maximum load that can be handled by a single server instance. We propose in this paper a novel game load management technique divided in two (global and a local) layers, capable of dynamically adjusting the amount of allocated resources to the present user demand. The global level assigns the responsibility of serving particular game regions to data centers using a peer-to-peer infrastructure, while the local level within individual facilities maintains the necessary server instances for the assigned obligations. We device two generic heuristics based on the well-known bin-packing problem to achieve the ultimate goal of maximizing the resource utilization on both levels while maintaining user-level Quality of Service (QoS). We evaluate the performance of our proposed solution using simulation-based experiments, which demonstrate a potential cost reduction in maintaining MMOG sessions by up to 60% while maintaining QoS in 99% of the cases.","Title: Multi-Objective Game Theoretic Schedulingof Bag-of-Tasks Workflows on Hybrid Clouds\nYear: 2014\n\nAbstract: Scheduling multiple large-scale parallel workflow applications on heterogeneous computing systems like hybrid clouds is a fundamental NP-complete problem that is critical to meeting various types of QoS (Quality of Service) requirements. This paper addresses the scheduling problem of large-scale applications inspired from real-world, characterized by a huge number of homogeneous and concurrent bags-of-tasks that are the main sources of bottlenecks but open great potential for optimization. The scheduling problem is formulated as a new sequential cooperative game and propose a communication and storage-aware multi-objective algorithm that optimizes two user objectives (execution time and economic cost) while fulfilling two constraints (network bandwidth and storage requirements). We present comprehensive experiments using both simulation and real-world applications that demonstrate the efficiency and effectiveness of our approach in terms of algorithm complexity, makespan, cost, system-level efficiency, fairness, and other aspects compared with other related algorithms.","Title: SLA-based operations of massively multiplayer online games in clouds\nYear: 2014\n\nAbstract: Successful massively multiplayer online games (MMOGs) have today millions of registered users and hundreds of thousands of active concurrent players. To be able to guarantee quality of service (QoS) to a highly variable number of concurrent users, game operators statically over-provision a large infrastructure capable of sustaining the game peak load, even though a large portion of the resources is unused most of the time. To address this problem, we introduce in this work a new MMOG ecosystem for hosting and provisioning of MMOGs which effectively splits the traditional monolithic MMOG companies into three main service providers: game providers, game operators, and resource providers. Their interaction is regulated through comprehensive service level agreements (SLA) that establish the price, terms of operation, and compensation for service violations. In our model, game operators efficiently provision resources for MMOGs from multiple cloud providers, based on dynamic load forecasts, and ensure proper game operation that maintains the required QoS to all clients under varying resource availability. Game providers manage multiple distributed MMOGs for which they lease services under strict operational SLAs from game operators to satisfy all client requests. These three self-standing, smaller, more agile service providers enable access to the MMOG market for the small and medium enterprises, and to the current commercial cloud providers. We evaluate, through simulations based on real-life MMOG traces and commercial cloud SLAs, the impact of resource availability on the QoS offered to the MMOG clients. We find that our model can mitigate the negative effects of resource failures within four minutes and that MMOG server consolidation can accentuate the negative effects of failures in a resource-scarce environment. We further investigate different methods of ranking MMOG operational offers with either single or multiple (competing) MMOG providers. Our results show that compensations for SLA faults in the offer selection process can lead up to 11---16 % gain in the game providers' income. We also demonstrate that adequate ranking of offers can lead to MMOG operational cost reductions from 20 up to 60 %.","Title: Multi-objective energy-efficient workflow scheduling using list-based heuristics.\nYear: 2014\n\nAbstract: Workflow applications are a popular paradigm used by scientists for modelling applications to be run on heterogeneous high-performance parallel and distributed computing systems. Today, the increase in the number and heterogeneity of multi-core parallel systems facilitates the access to high-performance computing to almost every scientist, yet entailing additional challenges to be addressed. One of the critical problems today is the power required for operating these systems for both environmental and financial reasons. To decrease the energy consumption in heterogeneous systems, different methods such as energy-efficient scheduling are receiving increasing attention. Current schedulers are, however, based on simplistic energy models not matching the reality, use techniques like DVFS not available on all types of systems, or do not approach the problem as a multi-objective optimisation considering both performance and energy as simultaneous objectives. In this paper, we present a new Pareto-based multi-objective workflow scheduling algorithm as an extension to an existing state-of-the-art heuristic capable of computing a set of tradeoff optimal solutions in terms of makespan and energy efficiency. Our approach is based on empirical models which capture the real behaviour of energy consumption in heterogeneous parallel systems. We compare our new approach with a classical mono-objective scheduling heuristic and state-of-the-art multi-objective optimisation algorithm and demonstrate that it computes better or similar results in different scenarios. We analyse the different tradeoff solutions computed by our algorithm under different experimental configurations and we observe that in some cases it finds solutions which reduce the energy consumption by up to 34.5% with a slight increase of 2% in the makespan.","Title: Workflow Scheduling on Federated Clouds.\nYear: 2014\n\nAbstract: Federated Clouds, or the orchestration of multiple Cloud services for fulfilling applications' requirements, is receiving increasing attention. Despite their many advantages, federated Clouds also present some downsides since different services may reside in different geographically located areas. This paper focuses on evaluating the advantages and disadvantages, from the point of view of performance and financial costs, of using a federation of Clouds for executing scientific workflows. It evaluates a wide range of different workflow types with different requirements in terms of computation and communication (produced and consumed data), and discusses which kind of workflow applications can benefit from a Cloud federation and how.","Title: Comparison of Static and Dynamic Resource Allocations for Massively Multiplayer Online Games on Unreliable Resources.\nYear: 2014\n\nAbstract: We investigate the use of a new Massively Multiplayer Online Gaming (MMOG) ecosystem consisting of end-users, game providers, game operators, and Cloud resource providers, for autonomous, self-adaptive hosting and operation of MMOGs on unreliable resources. For this purpose, we developed an MMOG simulator compliant with our MMOG ecosystem in which we inject traces collected from a real-world MMOG and resource characteristics from 16 Cloud providers. Using our simulator, we study the impact on the involved actors by considering different resource availability levels, and highlight the advantages of dynamic resource allocation over the static overprovisioning with respect to two types of metrics: QoS offered to the clients and financial profit of game providers and operators.","Title: Cooperative Scheduling of Bag-of-Tasks Workflows on Hybrid Clouds\nYear: 2014\n\nAbstract: We address the problem of scheduling a class of large-scale applications inspired from real-world on hybrid Clouds, characterized by a large number of homogeneous and concurrent tasks that are the main sources of bottlenecks but open great potential for optimization. We formulate the scheduling problem as a new sequential cooperative game and propose a communication- and storage-aware multi-objective algorithm that optimizes two user objectives (execution time and economic cost) while fulfilling two constraints (network bandwidth and storage requirements). We present comprehensive experiments using both simulation and real-world applications that demonstrate the efficiency and effectiveness of our approach in terms of algorithm complexity, make span, cost, system-level efficiency, fairness, and other aspects compared with other related algorithms.","Title: Massivizing online games using cloud computing: A vision\nYear: 2014\n\nAbstract: Online gaming, a large market with hundreds of millions of active players, is still struggling to scale without risky investments in infrastructure. In this work, we propose a cloudbased platform to massivize online gaming-the challenges and opportunities of scaling on-demand, while paying only for what is used. We discuss the major aspects of cloud-based gaming, virtual-world management, game-data processing, and game-content generation.","Title: Optimizations to enhance sustainability of MPI applications\nYear: 2014\n\nAbstract: Ultrascale computing systems are likely to reach speeds of two or three orders of magnitude greater than today's computing systems. However, to achieve this level of performance, we need to design and implement more sustainable solutions for ultra-scale computing systems, at both the hardware and software levels, while understanding sustainability in a holistic manner in order to address challenges in economy-of-scale, agile elastic scalability, heterogeneity, programmability, fault resilience, energy efficiency, and storage. Some solutions could be integrated into MPI, but others should be devised as higher level concepts, less general, but adapted to applicative domains, possibly as programming patterns or libraries. In this paper, we layout some proposals to extend MPI to cover major relevant domains in a move towards sustainability, including: MPI programming optimizations and programming models, resilience, data management, and their usage for applications.","Title: Meteorological applications utilizing Grid and Cloud computing\nYear: 2014\n\nAbstract: Three practical meteorological applications with different characteristics are used to highlight the usability of a computer science workflow middleware called ASKALON by allowing easy access to distributed computing for atmospheric scientists. Utilizing Cloud and Grid computing, this paper shows use case scenarios fitting a wide range of applications from operational to research studies with real world examples from meteorological research. The paper concludes that distributed computing is easily usable for meteorological problems using ASKALON. This powerful tool allows simple and cost effective access to computing capacity from Grid and Cloud environments.","Title: Fostering energy-awareness in scientific cloud users\nYear: 2014\n\nAbstract: Academic cloud infrastructures are constructed and maintained so they minimally constrain their users. Since they are free and do not limit usage patterns, academics developed such behavior that jeopardizes fair and flexible resource provisioning. For efficiency, related work either explicitly limits user access to resources, or introduces automatic rationing techniques. Surprisingly, the root cause (i.e., the user behavior) is disregarded by these approaches. This paper compares academic cloud user behavior to its commercial equivalent. We deduce, that academics should behave like commercial cloud users to relieve resource provisioning. To encourage this behavior, we propose an architectural extension to academic infrastructure clouds. We evaluate our extension via a simulation using real life academic resource request traces. We show a potential resource usage reduction while maintaining the unlimited nature of academic clouds.","Title: Fostering Energy-Awareness in Simulations behind Scientific Workflow Management Systems\nYear: 2014\n\nAbstract: Scientific workflow management systems face a new challenge in the era of cloud computing. The past availability of rich information regarding the state of the used infrastructures is gone. Thus, organising virtual infrastructures so that they not only support the workflow being executed, but also optimise for several service level objectives (e.g., Maximum energy consumption limit, cost, reliability, availability) become dependent on good infrastructure modelling and prediction techniques. While simulators have been successfully used in the past to aid research on such workflow management systems, the currently available cloud related simulation toolkits suffer form several issues (e.g., Scalability, narrow scope) that hinder their applicability. To address this need, this paper introduces techniques for unifying two existing simulation toolkits by first analysing the problems with the current simulators, and then by illustrating the problems faced by workflow systems through the example of the ASKALON environment. Finally, we show how the unification of the selected simulators improve on the the discussed problems.","Title: Evaluating Energy Efficiency of Gigabit Ethernet and Infiniband Software Stacks in Data Centres\nYear: 2014\n\nAbstract: Reducing energy consumption has become a key issue for data centres, not only because of economical benefits but also for environmental and marketing reasons. Many approaches tackle this problem from the point of view of different hardware components, such as CPUs, storage and network interface cards (NIC). To this date, few works focused on the energy consumption of network transfers at the software level comprising their complete stacks with different energy characteristics, and the way the NIC selection impacts the energy consumption of applications. Since data centres often install multiple NICs on each node, investigating and comparing them at the software level has high potential to enhance the energy efficiency of applications on Cloud infrastructures. We present a comparative analysis of the energy consumption of the software stack of two of today's most used NICs in data centres, Ethernet and Infiniband. We carefully design for this purpose a set of benchmark experiments to assess the impact of different traffic patterns and interface settings on energy consumption. Using our benchmark results, we derive an energy consumption model for network transfers and evaluate its accuracy for a virtual machine migration scenario. Finally, we propose guidelines for NIC selection from an energy efficiency perspective for different application classes.","Title: Bi-Criteria Scheduling of Scientific Grid Workflows\nYear: 2010\n\nAbstract: The drift towards new challenges in Grid computing including scientific workflow management implies the need for new, robust, multicriteria scheduling algorithms that can be applied by the user in an intuitive way. Currently existing bi-criteria scheduling approaches for scientific workflows are usually restricted to certain criterion pairs and require the user to define his preferences either as weights assigned each criterion or as fixed constraints defined for one criterion. The first approach has the drawback that combining multiple criteria into a single objective function is not always intuitive to the end-user, while the second requires a priori knowledge about the result of the first criteria scheduling result. We propose a new bi-criteria scheduling specification method defining for the secondary criterion a sliding constraint as a function of the primary criterion. We model the problem as an extension of the multiple-choice knapsack problem and propose a general bi-criteria scheduling heuristic called dynamic constraint algorithm (DCA) based on dynamic programming. We show through simulation that in most of the experimental cases DCA outperforms two existing algorithms designed for the same problem at the expense of an increased execution time, which is still relatively low for workflows of medium size. Finally, we confirm our simulation results for a real-world hydrological application executed in the Austrian Grid environment.","Title: Evaluating High-Performance Computing on Google App Engine\nYear: 2012\n\nAbstract: An experimental approach employs the Google App Engine (GAE) for high-performance parallel computing. A generic master-slave framework enables fast prototyping and integration of parallel algorithms that are transparently scheduled and executed on the Google cloud infrastructure. Compared to Amazon Elastic Compute Cloud (EC2), GAE offers lower resource-provisioning overhead and is cheaper for jobs shorter than one hour. Experiments demonstrated good scalability of a Monte Carlo simulation algorithm. Although this approach produced important speedup, two main obstacles limited its performance: middleware overhead and resource quotas.","Title: Extending Grids with cloud resource management for scientific computing\nYear: 2009\n\nAbstract: From its start using supercomputers, scientific computing constantly evolved to the next levels such as cluster computing, meta-computing, or computational Grids. Today, Cloud Computing is emerging as the paradigm for the next generation of large-scale scientific computing, eliminating the need of hosting expensive computing hardware. Scientists still have their Grid environments in place and can benefit from extending them by leased Cloud resources whenever needed. This paradigm shift opens new problems that need to be analyzed, such as integration of this new resource class into existing environments, applications on the resources and security. The virtualization overheads for deployment and starting of a virtual machine image are new factors which will need to be considered when choosing scheduling mechanisms. In this paper we investigate the usability of compute Clouds to extend a Grid workflow middleware and show on a real implementation that this can speed up executions of scientific workflows.","Title: Performance analysis of Grid applications in the ASKALON environment\nYear: 2009\n\nAbstract: In this paper we describe the approach taken by the ASKALON Grid application development and computing environment for scalability and overhead analysis of scientific applications in the Austrian Grid. We present a technique imported from parallel processing for overhead and scalability analysis of Grid applications based on speedup and efficiency metrics required primarily for tuning the middleware services and improving the executions. We present experimental results that validate our techniques for five real-world applications in the Austrian Grid environment.","Title: The JavaSymphony Extensions for Parallel GPU Computing\nYear: 2012\n\nAbstract: Today, the use of GPUs as coprocessors to accelerate high-performance scientific applications is becoming an important practice. Still, some of the high-level programming languages such as Java require extensions or new interfaces for utilising the huge parallelism of these new devices. In this paper, we propose extensions to an existing Java-based programming and parallel computing environment called Java Symphony to enable Java applications use accelerating devices such as GPUs with little API programmability change. With Java Symphony, a parallel Java application can be uniformly programmed and executed on heterogeneous platforms consisting of conventional parallel computers enhanced with data-parallel coprocessors such as GPUs. We report results on using Java Symphony for programming and improving the performance of six real applications and benchmarks in a heterogeneous environment consisting of a combination of different multi-core CPU and GPU devices.","Title: Cost-efficient hosting and load balancing of Massively Multiplayer Online Games\nYear: 2010\n\nAbstract: Massively Multiplayer Online Games (MMOG) are a class of computationally-intensive client-server applications with severe real-time Quality of Service (QoS) requirements, such as the number of updates per second each client needs to receive from the servers for a fluent and realistic experience. To guarantee the QoS requirements, game providers over-provision to game sessions a large amount of their resources, which is very inefficient and prohibits any but the largest providers from joining the market. In this paper, we present a new approach for cost-efficient hosting of MMOG sessions on Cloud resources, provisioned on-demand in the correct amount based on the current number of connected players. Simulation results on real MMOG traces demonstrate that compute Clouds can reduce the hosting costs by a factor between two and five. The resource allocation is driven by a load balancing algorithm that appropriately distributes the load such that the QoS requirements are fulfilled at all times. Experimental results on a fast-paced game demonstrator executed on resources owned by a specialised hosting company demonstrate that our algorithm is able to adjust the number of game servers and load distribution to the highly dynamic client load, while maintaining the QoS in 99.34% of the monitored events.","Title: Massively Multiplayer Online Games on unreliable resources\nYear: 2012\n\nAbstract: We have previously introduced a cloud-based model for Massively Multiplayer Online Games (MMOGs) operation, which allows small and medium enterprises to join the MMOG market through near-zero initial infrastructure investment. In this work we propose an architectural extension to this MMOG operational model designed to increase and guarantee the MMOG session availability through fault tolerance mechanisms.","Title: PIASA: A power and interference aware resource management strategy for heterogeneous workloads in cloud data centers\nYear: 2015\n\nAbstract: •A scheduling strategy for heterogeneous workloads with specific SLAs requirements.•Deviations detection of applications performance, with noisy data samples.•Multiple types of SLAs, according to mixed/heterogeneous applications.•Support for mitigation of performance degradation caused by resource contention.•Auto-scaling of resources in order to satisfy SLAs and cope with peak time demands.","Title: Pareto tradeoff scheduling of workflows on federated commercial Clouds\nYear: 2015\n\nAbstract: •We extend the HEFT algorithm for multi-objective workflow optimisation.•We apply the algorithm for performance and cost optimisation.•We adapt the new algorithm to fit the scope of public commercial Clouds.•We evaluate Pareto fronts algorithm using the hypervolume metric.•We demonstrate better results compared with two related algorithms.","Title: On Processing Extreme Data\nYear: 2015\n\nAbstract: Extreme Data is an incarnation of Big Data concept distinguished by the massive amounts of data that must be queried, communicated and analyzed in near real-time by using a very large number of memory or storage elements and exascale computing systems. Immediate examples are the scientific data produced at a rate of hundreds of gigabits-per-second that must be stored, filtered and analyzed, the millions of images per day that must be analyzed in parallel, the one billion of social data posts queried in real-time on an in-memory components database. Traditional disks or commercial storage nowadays cannot handle the extreme scale of such application data.Following the need of improvement of current concepts and technologies, we focus in this paper on the needs of data intensive applications running on systems composed of up to millions of computing elements (exascale systems). We propose in this paper a methodology to advance the state-of-the-art. The starting point is the definition of new programming paradigms, APIs, runtime tools and methodologies for expressing data-intensive tasks on exascale systems. This will pave the way for the exploitation of massive parallelism over a simplified model of the system architecture, thus promoting high performance and efficiency, offering powerful operations and mechanisms for processing extreme data sources at high speed and/or real time.","Title: Modelling energy consumption of network transfers and virtual machine migration.\nYear: 2016\n\nAbstract: Reducing energy consumption has become a key issue for data centres, not only because of economical benefits but also for environmental and marketing reasons. Therefore, assessing their energy consumption requires precise models. In the past years, many models targeting different hardware components, such as CPU, storage and network interface cards (NIC) have been proposed. However, most of them neglect energy consumption related to VM migration. Since VM migration is a network-intensive process, to accurately model its energy consumption we also need energy models for network transfers, comprising their complete software stacks with different energy characteristics.","Title: Low-time complexity budget–deadline constrained workflow scheduling on heterogeneous resources\nYear: 2016\n\nAbstract: The execution of scientific applications, under the utility computing model, is constrained to Quality of Service (QoS) parameters. Commonly, applications have time and cost constraints such that all tasks of an application need to be finished within a user-specified Deadline and Budget. Several algorithms have been proposed for multiple QoS workflow scheduling, but most of them use search-based strategies that generally have a high time complexity, making them less useful in realistic scenarios. In this paper, we present a heuristic scheduling algorithm with quadratic time complexity that considers two important constraints for QoS-based workflow scheduling, time and cost, named Deadline–Budget Constrained Scheduling (DBCS). From the deadline and budget defined by the user, the DBCS algorithm finds a feasible solution that accomplishes both constraints with a success rate similar to other state-of-the-art search-based algorithms in terms of the successful rate of feasible solutions, consuming in the worst case only approximately 4% of the time. The DBCS algorithm has a low-time complexity of O(n2.p) for n tasks and p processors.","Title: A Workload-Aware Energy Model for Virtual Machine Migration\nYear: 2015\n\nAbstract: Energy consumption has become a significant issue for data centres. Assessing their consumption requires precise and detailed models. In the latter years, many models have been proposed, but most of them either do not consider energy consumption related to virtual machine migration or do not consider the variation of the workload on (1) the virtual machines (VM) and (2) the physical machines hosting the VMs. In this paper, we show that omitting migration and workload variation from the models could lead to misleading consumption estimates. Then, we propose a new model for data centre energy consumption that takes into account the previously omitted model parameters and provides accurate energy consumption predictions for paravirtualised virtual machines running on homogeneous hosts. The new model's accuracy is evaluated with a comprehensive set of operational scenarios. With the use of these scenarios we present a comparative analysis of our model with similar state-of-the-art models for energy consumption of VM Migration, showing an improvement up to 24% in accuracy of prediction.","Title: A workflow runtime environment for manycore parallel architectures\nYear: 2015\n\nAbstract: We introduce a new Manycore Workflow Runtime Environment (MWRE) to efficiently enact traditional scientific workflows on modern manycore computing architectures. In contrast to existing engines that enact workflows acting as external services, MWRE is compiler-based and translates workflows specified in the XML-based Interoperable Workflow Intermediate Representation (IWIR) into an equivalent C++-based program. This program efficiently enacts the workflow as a stand-alone executable by means of a new callback mechanism that resolves dependencies, transfers data, and handles composite activities. Experimental results on a number of real-world workflows demonstrate that MWRE clearly outperforms existing Java-based workflow engines designed for distributed (Grid/Cloud) computing infrastructures in terms of enactment time, is generally better than an existing script-based engine for manycore architectures (Swift), and sometimes gets even close to an artificial baseline implementation of the workflows in the standard OpenMP language for shared memory systems.","Title: Energy Prediction of OpenMP Applications Using Random Forest Modeling Approach\nYear: 2015\n\nAbstract: OpenMP, with its extended parallelism features and support for radically changing HPC architectures, spurred to a surge in developing parallel applications among the HPC application developers community, leading to severe energy consumption issues. Consequently, a notion of addressing the energy consumption issue of HPC applications in an automated fashion increased among compiler developers although the underlying optimization search space could increase tremendously. This paper proposes a Random Forest Modeling (RFM) approach for predicting the energy consumption of OpenMP applications in compilers. The approach was tested using OpenMP applications, such as, NAS benchmarks, matrix multiplication, n-body simulations, and stencil applications while tuning the applications based on energy, problem size, and other performance concerns. The proposed RFM approach predicted the energy consumption of code variants with less than 0.699 Mean Square Error (MSE) and 0.998 R2 value when the testing dataset had energy variations between 0.024 joules and 150.23 joules. In addition, the influences of energy variations, number of independent variables used, and the proportion of testing dataset used during the RFM modeling process are discussed.","Title: Multi-layered simulations at the heart of workflow enactment on clouds\nYear: 2016\n\nAbstract: AbstractScientific workflow systems face new challenges when supporting Cloud computing, as the information on the state of the used infrastructures is much less detailed than before. Thus, organising virtual infrastructures in a way that not only supports the workflow execution but also optimises it for several service level objectives e.g. maximum energy consumption limit, cost, reliability, availability become reliant on good Cloud modelling and prediction information. While simulators were successfully aiding research on such workflow management systems, the currently available Cloud related simulation toolkits suffer from several issues e.g. scalability and narrow scope that hinder their applicability. To address these issues, this article introduces techniques for unifying two existing simulation toolkits by first analysing the problems with the current simulators, and then by illustrating the problems faced by workflow systems. We use for this purpose the example of the ASKALON environment, a scientific workflow composition and execution tool for cloud and grid environments. We illustrate the advantages of a workflow system with directly integrated simulation back-end and how the unification of the selected simulators does not affect the overall workflow execution simulation performance. Copyright © 2015 John Wiley & Sons, Ltd.","Title: Large scale distributed cooperative environments on clouds and P2P.\nYear: 2016\n\nAbstract: The recent advances in networking have determined an increasing use of information technology to support distributed cooperative applications. Several novel applications have emerged in this area, like computer supported collaborative work (CSCW), large scale distributed virtual worlds, collaborative recommender systems, collaborative learning systems. The definition of these applications requires to afford several challenges, like the definition of user interfaces, of coordination protocols, and of proper middle-ware and architectures supporting distributed cooperation. Collaborative applications may greatly benefits also from the support of P2P and cloud architectures. As a matter of fact, with the emergence of readily available P2P and cloud platforms, collaborative applications developers have the opportunity of deploying their collaborative application by exploiting a hybrid P2P/Cloud architectures with a support dynamically adapting . This brings possibilities to smaller developers that were reserved for the big companies until recently. The integration of mobile/Cloud platforms for Collaborative Applications is another challenge for the widespread of these applications.","Title: An Improved Model for Live Migration in Data Centre Simulators.\nYear: 2016\n\nAbstract: Due to the difficulty of employing real data centres' infrastructure for assessing the effectiveness of energy-aware algorithms, many researchers resort to use simulation tools. These tools require precise and detailed models for virtualized data centres in order to deliver accurate results. In recent years, many models have been proposed, but most of them either do not consider energy consumption related to virtual machine (VM) migration or do not investigate some of the energy impacting components (e.g. CPU, network, storage). We propose a new model for data centre energy consumption that takes into account the previously omitted components and provides more accurate energy consumption predictions compared to other state-of-the-art solutions. We evaluate our model's accuracy in a comprehensive set of scenarios implemented in the combined GroudSim/DISSECT-CF simulator. Our analysis revealed a significant improvement in accuracy (up to 42.5%) in the modelled energy consumption compared to a similar state-of-the-art simulator.","Title: On the Next Generations of Infrastructure-as-a-Services.\nYear: 2016\n\nAbstract: Following the wide adoption by industry of the cloud computing technologies, we can talk about a second generation of cloud services and products that are currently under design phase. However, it is not yet clear how the third generation of cloud products and services of the next decade will look like, especially at the delivery level of Infrastructure-as-a-Service. In order to answer at least partially to such a challenging question, we initiated a literature overview and two surveys involving the members of a cluster of European research and innovation actions. The results are interpreted in this paper and a set of topics of interest for the third generation are identified.","Title: Multi-Objective Middleware For Distributed Vmi Repositories In Federated Cloud Environment\nYear: 2016\n\nAbstract: Virtualization represents an essential technology in Cloud computing, which allows virtual machines (VM) to be executed within their own environment on top of physical hardware. The modern methods for software delivery are utilizing the concept of Vitual Machine as a efficient tool for software packaging. Typically, VMs are created using specific templates that are stored in proprietary repositories, thus leading to provider lock-in and reduced portability in the cases of simultaneous usage of multiple federated Clouds. Unfortunately, the current state-of-the-art does not provide any efficient means for streamlined management of VM images across multiple repositories, especially within federated Cloud environments. In this paper we present a novel multi-objective middleware for distributed VMI repositories in federated Cloud environment. The middleware has been designed to provide easy to use interface capable of receiving unmodified and functionally complete VM images from its users, and transparently distribute them to a specific Cloud infrastructure in a federation with respect to their size, configuration, and geographical distribution, such that they are loaded, delivered, and executed faster and with improved QoS compared to their current behaviour.","Title: Superlinear Speedup in HPC Systems: why and when?\nYear: 2016\n\nAbstract: The speedup is usually limited by two main laws in high-performance computing, that is, the Amdahlu0027s and Gustafsonu0027s laws. However, the speedup sometimes can reach far beyond the limited linear speedup, known as superlinear speedup, which means that the speedup is greater than the number of processors that are used. Although the superlinear speedup is not a new concept and many authors have already reported its existence, most of them reported it as a side effect, without explaining why and how it is happening. In this paper, we analyze several different superlinear speedup types and define a taxonomy for them. Additionally, we present several explanations and cases of superlinearity existence for different types of granular algorithms (tasks), which means that they can be divided into many sub-tasks and scattered to the processors for execution. Apart from frequent explanation that having more cache memory in parallel execution is the main reason, we summarize other different effects that cause the superlinearity, including the superlinear speedup in cloud virtual environment for both vertical and horizontal scaling.","Title: Using Constraint-Based Reasoning for Multi-objective Optimisation of the ENTICE Environment\nYear: 2016\n\nAbstract: ENTICE is a set of innovative software services currently being developed to facilitate efficient operations of distributed Virtual Machine and container images (VMI/CI) repositories. Its operation necessitates various decision making for which a solver for Multi-Objective Optimisation (MOO) problems is used. However, the solver is a bottleneck due to its computational complexity. In order to be able to reduce the search space for the solver, we have developed an ontology and corresponding Knowledge Base (KB) that underpins the operation of the ENTICE environment. The Knowledge Base is developed based on the Jena Fuseki technology. To address the problem of computational complexity, constraint based queries and different reasoning mechanisms are applied. The Knowledge Base services are then integrated with other ENTICE services including the MOO solver. It is shown that this approach significantly reduces the computational complexity for the MOO, thus it shortens the optimisation time, and makes it possible to use the MOO for both strategic (decisions that can be made up to one day in advance) and dynamic (decisions requiring response within one minute) decision making possible.","Title: A Two-Stage Multi-Objective Optimization of Erasure Coding in Overlay Networks.\nYear: 2017\n\nAbstract: In the recent years, overlay networks have emerged as a crucial platform for deployment of various distributed applications. Many of these applications rely on data redundancy techniques, such as erasure coding, to achieve higher fault tolerance. However, erasure coding applied in large scale overlay networks entails various overheads in terms of storage, latency and data rebuilding costs. These overheads are largely attributed to the selected erasure coding scheme and the encoded chunk placement in the overlay network. This paper explores a multi-objective optimization approach for identifying appropriate erasure coding schemes and encoded chunk placement in overlay networks. The uniqueness of our approach lies in the consideration of multiple erasure coding objectives such as encoding rate and redundancy factor, with overlay network performance characteristics like storage consumption, latency and system reliability. Our approach enables a variety of tradeoff solutions with respect to these objectives to be identified in the form of a Pareto front. To solve this problem, we propose a novel two stage multi-objective evolutionary algorithm, where the first stage determines the optimal set of encoding schemes, while the second stage optimizes placement of the corresponding encoded data chunks in overlay networks of varying sizes. We study the performance of our method by generating and analyzing the Pareto optimal sets of tradeoff solutions. Experimental results demonstrate that the Pareto optimal set produced by our multi-objective approach includes and even dominates the chunk placements delivered by a related state-of-the-art weighted sum method.","Title: VM Image Repository and Distribution Models for Federated Clouds: State of the Art, Possible Directions and Open Issues.\nYear: 2016\n\nAbstract: The emerging trend of Federated Cloud models enlist virtualization as a significant concept to offer a large scale distributed Infrastructure as a Service collaborative paradigm to end users. Virtualization leverage Virtual Machines (VM) instantiated from user specific templates labelled as VM Images (VMI). To this extent, the rapid provisioning of VMs with varying user requests ensuring Quality of Service (QoS) across multiple cloud providers largely depends upon the image repository architecture and distribution policies. We discuss the possible state-of-art in VMI storage repository and distribution mechanisms for efficient VM provisioning in federated clouds. In addition, we present and compare various representative systems in this realm. Furthermore, we define a design space, identify current limitations, challenges and open trends for VMI repositories and distribution techniques within federated infrastructure.","Title: Multi-objective Optimization Framework for VMI Distribution in Federated Cloud Repositories.\nYear: 2016\n\nAbstract: Cloud Federation facilitates the concept of aggregation of multiple services administered by different providers, thus opening the possibility for the customers to profit from lower cost and better performance, while allowing for the cloud providers to offer more sophisticated services. Unfortunately, current state-of-the-art does not provide any substantial means for streamlined adaptation of federated Cloud environments. One of the essential barriers that prevents Cloud federation is the inefficient management of distributed storage repositories for Virtual Machine Images (VMI). In such environments, the VMIs are currently stored by Cloud providers in proprietary centralised repositories without considering application characteristics and their runtime requirements, causing high deployment and instantiation overheads. In this paper, a novel multi-objective optimization framework for VMI placement across distributed repositories in federated Cloud environment has been proposed. Based on the communication performance requirements, VMI use patterns, and structure of images or input data, the framework provides efficient means for transparent optimization of the distribution and placement of VMIs across distributed repositories to significantly lower their provisioning time for complex resource requests and for executing the user applications.","Title: Analysing the Performance Instability Correlation with Various Workflow and Cloud Parameters\nYear: 2017\n\nAbstract: Cloud is an eco-system in which virtual machine instances are starting and terminating asynchronously on user demand or automatically when the load is rapidly increased or decreased. Although this dynamic environment allows to rent computing or storage resources cheaper rather than buying them, still it does not guarantee the stable execution during a period of time as the traditional physical environment. This is emphasised even more for workflows execution, since they consist of many data and control dependencies, which cause the makespan to be instable when a workflow is being executed in different periods of time in Cloud. In this paper we analyse several parameters of workflow and the cloud environment that are expected to impact the workflow execution instability and investigate the correlation between them. The cloud parameters include the number of instances and their type, as well as the correlation with the efficient or inefficient execution of workflow parallel sections. We conduct a series of experiments, repeating each experiment by 30 test cases in order to evaluate instability for different cloud and workflow parameters. The results show a neglectful correlation between each pair of parameters, as well as the tasks and file transfers within the workflow. Oppose to the expectations, the distribution of the makespan per experiment does not always comply with the normal distribution, which is also not correlated to a particular cloud or workflow parameter.","Title: A Serverless Real-Time Data Analytics Platform for Edge Computing.\nYear: 2017\n\nAbstract: Contemporary solutions for cloud-supported, edge-data analytics mostly apply analytics techniques in a rigid bottom-up approach, regardless of the data's origin. Typically, data are generated at the edge of the infrastructure and transmitted to the cloud, where traditional data analytics techniques are applied. Currently, developers are forced to resort to ad hoc solutions specifically tailored fo...","Title: A Simplified Model for Simulating the Execution of a Workflow in Cloud.\nYear: 2017\n\nAbstract: Although simulators provide approximate, faster and easier simulation of an application execution in Clouds, still many researchers argue that these results cannot be always generalized for complex application types, which consist of many dependencies among tasks and various scheduling possibilities, such as workflows. DynamicCloudSim, the extension of the well known CloudSim simulator, offers users the capability to simulate the Cloud heterogeneity by introducing noisiness in dozens parameters. Still, it is difficult, or sometimes even impossible to determine appropriate values for all these parameters because they are usually Cloud or application-dependent. In this paper, we propose a new model that simplifies the simulation setup for a workflow and reduces the bias between the behavior of simulated and real Cloud environments based on one parameter only, the Cloud noisiness. It represents the noise produced by the Cloud’s interference including the application’s (in our case a workflow) noisiness too. Another novelty in our model is that it does not use a normal distribution naively to create noised values, but shifts the mean value of the task execution time by the cloud noisiness and uses its deviation as a standard deviation. Besides our model reduces the complexity of DynamicCloudSim’s heterogeneity model, evaluation conducted in Amazon EC2 shows that it is also more accurate, with better trueness (closeness to the real mean values) of up to (9.2 %) and precision (closeness to the real deviation) of up to 8.39 times.","Title: HPS-HDS: High Performance Scheduling for Heterogeneous Distributed Systems.\nYear: 2018\n\nAbstract: Heterogeneous Distributed Systems (HDS) are often characterized by a variety of resources that may or may not be coupled with specific platforms or environments. Such type of systems are Cluster Computing, Grid Computing, Peer-to-Peer Computing, Cloud Computing and Ubiquitous Computing all involving elements of heterogeneity, having a large variety of tools and software to manage them. As computing and data storage needs grow exponentially in HDS, increasing the size of data centers brings important diseconomies of scale. In this context, major solutions for scalability, mobility, reliability, fault tolerance and security are required to achieve high performance. More, HDS are highly dynamic in its structure, because the user requests must be respected as an agreement rule (SLA) and ensure QoS, so new algorithm for events and tasks scheduling and new methods for resource management should be designed to increase the performance of such systems. In this special issues, the accepted papers address the advance on scheduling algorithms, energy-aware models, self-organizing resource management, data-aware service allocation, Big Data management and processing, performance analysis and optimization.","Title: Towards Digital Twins Cloud Platform: Microservices And Computational Workflows To Rule A Smart Factory\nYear: 2017\n\nAbstract: The concept of \"Industry 4.0\" considers smart factories as data-driven and knowledge enabled enterprise intelligence. In such kind of factory, manufacturing processes and final products are accompanied by virtual models - Digital Twins. To support Digital Twins concept, a simulation model for each process or system should be implemented as independent computational service. The only way to implement an orchestration of a set of independent services and provide scalability for simulation is to use a cloud computing platform as a provider of the computing infrastructure. In this paper, we describe a Digital Twin-as-a-Service (DTaaS) model for simulation and prediction of industrial processes using Digital Twins.","Title: Simulation of a workflow execution as a real Cloud by adding noise.\nYear: 2017\n\nAbstract: •A new model and methodology to add a noise to workflow execution simulations.•Noisiness of tasks within a workflow and environmental noisiness over a workflow.•Up to 59 times simulation inaccuracy reduction of makespan’s mean value.•Instance type impacts the cloud performance instability.•Sensitivity analysis determines minimum data to be collected in Learning Phase in order to achieve accuracy and precision.","Title: A compiler transformation-based approach to scientific workflow enactment\nYear: 2017\n\nAbstract: We investigate in this paper the application of compiler transformations to workflow applications using the Manycore Workflow Runtime Environment (MWRE), a compiler-based workflow environment for modern manycore computing architectures. MWRE translates scientific workflows into equivalent C++ programs and efficiently executes them using a novel callback mechanism for dependency resolution and data transfers, with explicit support for full-ahead scheduling. We evaluate four different classes of compiler transformations, analyse their advantages and possible solutions to overcome their limitations, and present experimental results for improving the performance of a combination of real-world and synthetic workflows through compiler transformations. Our experiments were able to improve the workflow enactment by a factor of two and to reduce the memory usage of the engine by up to 33%. We achieved a speedup of up to 1.7 by eliminating unnecessary activity invocations, an improved parallel throughput up to 2.8 times by transforming the workflow structure, and a better performance of the HEFT scheduling algorithm by up to 36%.","Title: Multi-objective Service Oriented Network Provisioning in Ultra-Scale Systems.\nYear: 2017\n\nAbstract: The paradigm of ultra-scale computing has been recently pushed forward by the current trends in distributed computing. This novel architecture concept is focused towards a federation of multiple geographically distributed heterogeneous systems under a single system image, thus allowing efficient deployment and management of very complex architectures applications. To enable sustainable ultra-scale computing, there are multiple major challenges, which have to be tackled, such as, improved data distribution, increased systems scalability, enhanced fault tolerance, elastic resource management, low latency communication and etc. Regrettably, the current research initiatives in the area of ultra-scale computing are in a very early stage of research and are predominantly concentrated on the management of the computational and storage resources, thus leaving the networking aspects unexplored. In this paper we introduce a promising new paradigm for cluster-based Multi-objective service-oriented network provisioning for ultra-scale computing environments by unifying the management of the local communication resources and the external inter-domain network services under a single point of view. We explore the potentials for representing the local network resources within a single distributed or parallel system and combine them together with the external communication services.","Title: Adaptive Nature-Inspired Fog Architecture\nYear: 2018\n\nAbstract: During the last decade, Cloud computing has efficiently exploited the economy of scale by providing low cost computational and storage resources over the Internet, eventually leading to consolidation of computing resources into large data centers. However, the nascent of the highly decentralized Internet of Things (IoT) technologies that cannot effectively utilize the centralized Cloud infrastructures pushes computing towards resource dispersion. Fog computing extends the Cloud paradigm by enabling dispersion of the computational and storage resources at the edge of the network in a close proximity to where the data is generated. In its essence, Fog computing facilitates the operation of the limited compute, storage and networking resources physically located close to the edge devices. However, the shared complexity of the Fog and the influence of the recent IoT trends moving towards deploying and interconnecting extremely large sets of pervasive devices and sensors, requires exploration of adaptive Fog architectural approaches capable of adapting and scaling in response to the unpredictable load patterns of the distributed IoT applications. In this paper we introduce a promising new nature- inspired Fog architecture, named SmartFog, capable of providing low decision making latency and adaptive resource management. By utilizing novel algorithms and techniques from the fields of multi- criteria decision making, graph theory and machine learning we model the Fog as a distributed intelligent processing system, therefore emulating the function of the human brain.","Title: h-Fair: Asymptotic Scheduling of Heavy Workloads in Heterogeneous Data Centers\nYear: 2018\n\nAbstract: Large scale computing solutions are increasingly used in the context of Big Data platforms, where efficient scheduling algorithms play an important role in providing optimized cluster resource utilization, throughput and fairness. This paper deals with the problem of scheduling a set of jobs across a cluster of machines handling the specific use case of fair scheduling for jobs and machines with heterogeneous characteristics. Although job and cluster diversity is unprecedented, most schedulers do not provide implementations that handle multiple resource type fairness in a heterogeneous system. We propose in this paper a new scheduler called h-Fair that selects jobs for scheduling based on a global dominant resource fairness heterogeneous policy, and dispatches them on machines with similar characteristics to the resource demands using the cosine similarity. We implemented h-Fair in Apache Hadoop YARN and we compare it with the existing Fair Scheduler that uses the dominant resource fairness policy based on the Google workload trace. We show that our implementation provides better cluster resource utilization and allocates more containers when jobs and machines have heterogeneous characteristics.","Title: RM-BDP: Resource management for Big Data platforms.\nYear: 2018\n\nAbstract: Nowadays, when we face with numerous data, when data cannot be classified into regular relational databases and new solutions are required, and when data are generated and processed rapidly, we need powerful platforms and infrastructure as support. Extracting valuable information from raw data is especially difficult considering the velocity of growing data from year to year and the fact that 80% of data is unstructured. In addition, data sources are heterogeneous (various sensors, users with different profiles, etc.) and are located in different situations or contexts. Cloud computing, which concerns large-scale interconnected systems with the main purpose of aggregation and efficient exploiting the power of widely distributed resources, represent one viable solution. Resource management and task scheduling play an essential role, in cases where one is concerned with optimized use of resources (Negru et al., 2017) [1].","Title: E-OSched: a load balancing scheduler for heterogeneous multicores.\nYear: 2018\n\nAbstract: The contemporary multicore era has adhered to the heterogeneous computing devices as one of the proficient platforms to execute compute-intensive applications. These heterogeneous devices are based on CPUs and GPUs. OpenCL is deemed as one of the industry standards to program heterogeneous machines. The conventional application scheduling mechanisms allocate most of the applications to GPUs while leaving CPU device underutilized. This underutilization of slower devices (such as CPU) often originates the sub-optimal performance of data-parallel applications in terms of load balance, execution time, and throughput. Moreover, multiple scheduled applications on a heterogeneous system further aggravate the problem of performance inefficiency. This paper is an attempt to evade the aforementioned deficiencies via initiating a novel scheduling strategy named OSched. An enhancement to the OSched named E-OSched is also part of this study. The OSched performs the resource-aware assignment of jobs to both CPUs and GPUs while ensuring a balanced load. The load balancing is achieved via contemplation on computational requirements of jobs and computing potential of a device. The load-balanced execution is beneficiary in terms of lower execution time, higher throughput, and improved utilization. The E-OSched reduces the magnitude of the main memory contention during concurrent job execution phase. The mathematical model of the proposed algorithms is evaluated by comparison of simulation results with different state-of-the-art scheduling heuristics. The results revealed that the proposed E-OSched has performed significantly well than the state-of-the-art scheduling heuristics by obtaining up to 8.09% improved execution time and up to 7.07% better throughput.","Title: Semantic approach for multi-objective optimisation of the ENTICE distributed Virtual Machine and container images repository.\nYear: 2019\n\nAbstract: New software engineering technologies facilitate development of applications from reusable software components, such as Virtual Machine and container images (VMI/CIs). Key requirements for the storage of VMI/CIs in public or private repositories are their fast delivery and cloud deployment times. ENTICE is a federated storage facility for VMI/CIs that provides optimisation mechanisms through the use of fragmentation and replication of images and a Pareto Multi-Objective Optimisation (MO) solver. The operation of the MO solver is, however, time-consuming due to the size and complexity of the metadata, specifying various non-functional requirements for the management of VMI/CIs, such as geolocation, operational cost, and delivery time. In this work, we address this problem with a new semantic approach, which uses an ontology of the federated ENTICE repository, knowledge base, and constraint-based reasoning mechanism. Open Source technologies such as Protege, Jena Fuseki, and Pellet were used to develop a solution. Two specific use cases, (1) repository optimisation with offline and (2) online redistribution of VMI/CIs, are presented in detail. In both use cases, data from the knowledge base are provided to the MO solver. It is shown that Pellet-based reasoning can be used to reduce the input metadata size used in the optimisation process by taking into consideration the geographic location of the VMI/CIs and the provenance of the VMI fragments. It is shown that this process leads to reduction of the input metadata size for the MO solver by up to 60% and reduction of the total optimisation time of the MO solver by up to 68%, while fully preserving the quality of the solution, which is significant.","Title: DRUMS - Demand Response Management in a Smart City Using Deep Learning and SVR.\nYear: 2018\n\nAbstract: Demand response management in smart cities is one of the most challenging tasks to be performed due to the continuous changes in the load profile of the home users. The existing proposals in the literature fail to observe the hidden patterns in the load profile of these users. So, to fill these gaps, the concept of deep learning has been used in this paper for smart energy management in a smart city. The consumption data from smart homes (SHs) is gathered and taken as an input to the deep learning model, convolution neural network (CNN). The CNN model learns the hidden patterns in the data and outputs different load curves. These load curves are then used to train a support vector regression (SVR) model, which predicts the overall load consumption of all SHs in the smart city. This prediction is then compared with the power generation from the grid and consequently the demand response (DR) of the connected SHs is managed so as to minimize the gap between predicted demand and supply. The proposed scheme has been evaluated on the dataset collected from PJM and open energy information with respect to load demand prediction and DR management. The results obtained prove the efficacy of the proposed scheme. The prediction errors, i.e., root mean squared error and mean absolute percentage error are observed less in comparison to the cases when CNN and SVR are used individually.","Title: Scheduling scientific workflows to meet soft deadlines in the absence of failure models\nYear: 2010\n\nAbstract: Highly distributed systems such as Clouds and Grids are used to execute complex scientific workflow applications by researchers from various areas of science. While scientists rightfully expect efficient and reliable execution of their applications, current systems often cannot deliver the required Quality of Service. We propose a dynamic execution and scheduling heuristic able to schedule workflow applications with a high degree of fault tolerance, while taking into account soft deadlines. Experimental results show that our method meets soft deadlines in volatile highly distributed systems in the absence of historic failure trace data or complex failure models of the target system.","Title: Towards an Environment for Efficient and Transparent Virtual Machine Operations: The ENTICE Approach\nYear: 2016\n\nAbstract: Cloud computing is based on Virtual Machines (VM) or containers, which provide their own software execution environment that can be deployed by facilitating technologies on top of various physical hardware. The use of VMs or containers represents an efficient way to automatize the overall software engineering and operation life-cycle. Some of the benefits include elasticity and high scalability, which increases the utilization efficiency and decreases the operational costs. VMs or containers as software artifacts are created using provider-specific templates and are stored in proprietary or public repositories for further use. However, technology specific choices may reduce their portability, lead to a vendor lock-in, particularly when applications need to run in federated Clouds. In this paper we present the current state of development of the novel concept of a VM repository and operational environment for federated Clouds named ENTICE. The ENTICE environment has been designed to receive unmodified and functionally complete VM images from its users, and transparently tailor and optimise them for specific Cloud infrastructures with respect to their size, configuration, and geographical distribution, such that they are loaded, delivered, and executed faster and with improved QoS compared to their current behaviour. Furthermore, in this work a specific use case scenario for the ENTICE environment has been provided and the underlying novel technologies have been presented.","Title: Handling the uncertainty in resource performance for executing workflow applications in clouds.\nYear: 2016\n\nAbstract: Execution of workflow applications in Cloud environments involves many uncertainties because of elastic resource provisioning and unstable performance of multitenant virtual machines (VM) instances over time. These uncertainties are usually either neglected by existing researches, or modeled with some probability distribution function. To address this gap, we extend a multi-objective workflow scheduling algorithm (MOHEFT) in two directions: (1) to deal with the dynamic nature of Cloud environments offering a potentially infinite amount of on-demand resources, and (2) to consider robustness as an objective that mitigates the variability in VM performance over time. Our new robust model, called R-MOHEFT, considers uncertainty in processing times of workflow activities without a precise estimation or known distribution function within an uncertainty interval. We approach this scheduling problem as a three-objective optimisation that considers makespan, monetary cost, and robustness as simultaneous objectives of a commercial Cloud environment. Our new algorithm is able to estimate the Pareto optimal set of scheduling solutions that resist against fluctuations in processing times three times better than its MOHEFT predecessor, with a tradeoff of only 15% worse Pareto frontier. R-MOHEFT's hypervolume suffers by only 5% to 16%, compared to the MOHEFT's drawback of 38% to surprisingly 87%, when the processing time fluctuates up to its double value.","Title: Use Cases towards a Decentralized Repository for Transparent and Efficient Virtual Machine Operations\nYear: 2017\n\nAbstract: Virtualization is a key enabling technology in Cloud computing that allows users to run multiple virtual machines (VMs) with their own application environment on top of physical hardware. It permits scaling up and down of applications by elastic on-demand provisioning of VMs in response to their variable load to achieve increased utilization efficiency at a lower operational cost, while guaranteeing the desired level of Quality of Service (QoS) to the end-users. Typically, VMs are created using provider-specific templates that are stored in proprietary repositories, leading to provider lock-in and hampering portability or simultaneous usage of multiple federated Clouds. In this context, optimization at the level of the virtual machine image is needed both by the applications and by the underlying Cloud providers for improved resource usage, operational costs, elasticity, storage use, and other desired QoS-related features. To overcome those issues, the ENTICE project researches and creates a novel VM repository and operational environment for federated Cloud infrastructures. There exists a large variety of industrial applications that can strongly benefit by the ENTICE environment. In this paper we present an interesting selection of complementary use cases that drive the definition of the essential requirements for the ENTICE environment, and more importantly, validate the introduced innovations.","Title: Distributed environment for efficient virtual machine image management in federated Cloud architectures.\nYear: 2018\n\nAbstract: The use of virtual machines (VMs) in Cloud computing provides various benefits in the overall software engineering lifecycle. These include efficient elasticity mechanisms resulting in higher resource utilization and lower operational costs. The VMs as software artifacts are created using provider-specific templates, called virtual machine images (VMI), and are stored in proprietary or public repositories for further use. However, some technology-specific choices can limit the interoperability among various Cloud providers and bundle the VMIs with nonessential or redundant software packages, leading to increased storage size, prolonged VMI delivery, stagnant VMI instantiation, and ultimately vendor lock-in. To address these challenges, we present a set of novel functionalities and design approaches for efficient operation of distributed VMI repositories, specifically tailored for enabling (1) simplified creation of lightweight and size optimized VMIs tuned for specific application requirements; (2) multi-objective VMI repository optimization; and (3) efficient reasoning mechanism to help optimizing complex VMI operations. The evaluation results confirm that the presented approaches can enable VMI size reduction by up to 55%, while trimming the image creation time by 66%. Furthermore, the repository optimization algorithms can reduce the VMI delivery time by up to 51% and cut down the storage expenses by 3%. Moreover, by implementing replication strategies, the optimization algorithms can increase the system reliability by 74%.","Title: A New Model For Cloud Elastic Services Efficiency\nYear: 2019\n\nAbstract: The speedup measures the improvement in performance when the computational resources are being scaled. The efficiency, on the other side, provides the ratio between the achieved speedup and the number of scaled computational resources (processors). Both parameters (speedup and efficiency), which are defined according to Amdahl's Law, provide very important information about performance of a computer system with scaled resources compared with a computer system with a single processor. However, as cloud elastic services' load is variable, apart of the scaled resources, it is vital to analyse the load in order to determine which system is more effective and efficient. Unfortunately, both the speedup and efficiency are not sufficient enough for proper modeling of cloud elastic services, as the assumptions for both the speedup and efficiency are that the system's resources are scaled, while the load is constant. In this paper, we extend the scaling of resources and define two additional scaled systems by (i) scaling the load and (ii) scaling both the load and resources. We introduce a model to determine the efficiency for each scaled system, which can be used to compare the efficiencies of all scaled systems, regardless if they are scaled in terms of load or resources. We have evaluated the model by using Windows Azure and the experimental results confirm the theoretical analysis. Although one can argue that web services are scalable and comply with Gustafson's Law only, we provide a taxonomy that classifies scaled systems based on the compliance with both the Amdahl's and Gustafson's laws.[GRAPHICS]For three different scaled systems (scaled resources R, scaled load L and combination RL), we introduce a model to determine the scaling efficiency. Our model extends the current definition of efficiency according to Amdahl's Law, which assumes scaling the resources, and not the load.E-R = S-R/p = T(1)/T-R center dot 1/p; E-L = S-L center dot N = T(1)/T-L center dot N; E-RL = S-RL center dot N/p = T(1)/T-RL center dot N/p","Title: A New Model For Cloud Elastic Services Efficiency\nYear: 2019\n\nAbstract: The speedup measures the improvement in performance when the computational resources are being scaled. The efficiency, on the other side, provides the ratio between the achieved speedup and the number of scaled computational resources (processors). Both parameters (speedup and efficiency), which are defined according to Amdahl's Law, provide very important information about performance of a computer system with scaled resources compared with a computer system with a single processor. However, as cloud elastic services' load is variable, apart of the scaled resources, it is vital to analyse the load in order to determine which system is more effective and efficient. Unfortunately, both the speedup and efficiency are not sufficient enough for proper modeling of cloud elastic services, as the assumptions for both the speedup and efficiency are that the system's resources are scaled, while the load is constant. In this paper, we extend the scaling of resources and define two additional scaled systems by (i) scaling the load and (ii) scaling both the load and resources. We introduce a model to determine the efficiency for each scaled system, which can be used to compare the efficiencies of all scaled systems, regardless if they are scaled in terms of load or resources. We have evaluated the model by using Windows Azure and the experimental results confirm the theoretical analysis. Although one can argue that web services are scalable and comply with Gustafson's Law only, we provide a taxonomy that classifies scaled systems based on the compliance with both the Amdahl's and Gustafson's laws.[GRAPHICS]For three different scaled systems (scaled resources R, scaled load L and combination RL), we introduce a model to determine the scaling efficiency. Our model extends the current definition of efficiency according to Amdahl's Law, which assumes scaling the resources, and not the load.E-R = S-R/p = T(1)/T-R center dot 1/p; E-L = S-L center dot N = T(1)/T-L center dot N; E-RL = S-RL center dot N/p = T(1)/T-RL center dot N/p","Title: The Workflow Trace Archive: Open-Access Data from Public and Private Computing Infrastructures - Technical Report.\nYear: 2019\n\nAbstract: Realistic, relevant, and reproducible experiments often need input traces collected from real-world environments. We focus in this work on traces of workflows---common in datacenters, clouds, and HPC infrastructures. We show that the state-of-the-art in using workflow-traces raises important issues: (1) the use of realistic traces is infrequent, and (2) the use of realistic, {\\it open-access} traces even more so. Alleviating these issues, we introduce the Workflow Trace Archive (WTA), an open-access archive of workflow traces from diverse computing infrastructures and tooling to parse, validate, and analyze traces. The WTA includes ${>}48$ million workflows captured from ${>}10$ computing infrastructures, representing a broad diversity of trace domains and characteristics. To emphasize the importance of trace diversity, we characterize the WTA contents and analyze in simulation the impact of trace diversity on experiment results. Our results indicate significant differences in characteristics, properties, and workflow structures between workload sources, domains, and fields.","Title: Semantics-Aware Virtual Machine Image Management in IaaS Clouds\nYear: 2019\n\nAbstract: Infrastructure-as-a-service (IaaS) Clouds concurrently accommodate diverse sets of user requests, requiring an efficient strategy for storing and retrieving virtual machine images (VMIs) at a large scale. The VMI storage management require dealing with multiple VMIs, typically in the magnitude of gigabytes, which entails VMI sprawl issues hindering the elastic resource management and provisioning. Nevertheless, existing techniques to facilitate VMI management overlook VMI semantics (i.e at the level of base image and software packages) with either restricted possibility to identify and extract reusable functionalities or with higher VMI publish and retrieval overheads. In this paper, we design, implement and evaluate Expelliarmus, a novel VMI management system that helps to minimize storage, publish and retrieval overheads. To achieve this goal, Expelliarmus incorporates three complementary features. First, it makes use of VMIs modelled as semantic graphs to expedite the similarity computation between multiple VMIs. Second, Expelliarmus provides a semantic aware VMI decomposition and base image selection to extract and store non-redundant base image and software packages. Third, Expelliarmus can also assemble VMIs based on the required software packages upon user request. We evaluate Expelliarmus through a representative set of synthetic Cloud VMIs on the real test-bed. Experimental results show that our semantic-centric approach is able to optimize repository size by 2.2 - 16 times compared to state-of-the-art systems (e.g. IBM's Mirage and Hemera) with significant VMI publish and slight retrieval performance improvement.","Title: Where to Encode: A Performance Analysis of x86 and Arm-based Amazon EC2 Instances\nYear: 2021\n\nAbstract: Video streaming became an undivided part of the Internet. To efficiently utilise the limited network bandwidth it is essential to encode the video content. However, encoding is a computationally intensive task, involving high-performance resources provided by private infrastructures or public clouds. Public clouds, such as Amazon EC2, provide a large portfolio of services and instances optimized f...","Title: Fusioncl: A Machine-Learning Based Approach For Opencl Kernel Fusion To Increase System Performance\nYear: 2021\n\nAbstract: Employing general-purpose graphics processing units (GPGPU) with the help of OpenCL has resulted in greatly reducing the execution time of data-parallel applications by taking advantage of the massive available parallelism. However, when a small data size application is executed on GPU there is a wastage of GPU resources as the application cannot fully utilize GPU compute-cores. There is no mechanism to share a GPU between two kernels due to the lack of operating system support on GPU. In this paper, we propose the provision of a GPU sharing mechanism between two kernels that will lead to increasing GPU occupancy, and as a result, reduce execution time of a job pool. However, if a pair of the kernel is competing for the same set of resources (i.e., both applications are compute-intensive or memory-intensive), kernel fusion may also result in a significant increase in execution time of fused kernels. Therefore, it is pertinent to select an optimal pair of kernels for fusion that will result in significant speedup over their serial execution. This research presents FusionCL, a machine learning-based GPU sharing mechanism between a pair of OpenCL kernels. FusionCL identifies each pair of kernels (from the job pool), which are suitable candidates for fusion using a machine learning-based fusion suitability classifier. Thereafter, from all the candidates, it selects a pair of candidate kernels that will produce maximum speedup after fusion over their serial execution using a fusion speedup predictor. The experimental evaluation shows that the proposed kernel fusion mechanism reduces execution time by 2.83x when compared to a baseline scheduling scheme. When compared to state-of-the-art, the reduction in execution time is up to 8%.","Title: Handover Authentication Latency Reduction Using Mobile Edge Computing And Mobility Patterns\nYear: 2021\n\nAbstract: With the advancement in technology and the exponential growth of mobile devices, network traffic has increased manifold in cellular networks. Due to this reason, latency reduction has become a challenging issue for mobile devices. In order to achieve seamless connectivity and minimal disruption during movement, latency reduction is crucial in the handover authentication process. Handover authentication is a process in which the legitimacy of a mobile node is checked when it crosses the boundary of an access network. This paper proposes an efficient technique that utilizes mobility patterns of the mobile node and mobile Edge computing framework to reduce handover authentication latency. The key idea of the proposed technique is to categorize mobile nodes on the basis of their mobility patterns. We perform simulations to measure the networking latency. Besides, we use queuing model to measure the processing time of an authentication query at an Edge servers. The results show that the proposed approach reduces the handover authentication latency up to 54% in comparison with the existing approach.","Title: FAUST: Fast Per-Scene Encoding Using Entropy-Based Scene Detection and Machine Learning\nYear: 2021\n\nAbstract: HTTP adaptive video streaming is a widespread and sought-after technology on the Internet that allows clients to dynamically switch between different stream qualities presented in the bitrate ladder to optimize overall received video quality. Currently, there exist several approaches of different complexity for building such a ladder. The simplest method is to use a static bitrate ladder, and the ...","Title: Autotuning of Exascale Applications With Anomalies Detection.\nYear: 2021\n\nAbstract: The execution of complex distributed applications in exascale systems faces many challenges, as it involves empirical evaluation of countless code variations and application runtime parameters over a heterogeneous set of resources. To mitigate these challenges, the research field of autotuning has gained momentum. The autotuning automates identifying the most desirable application implementation in terms of code variations and runtime parameters. However, the complexity and size of the exascale systems make the autotuning process very difficult, especially considering the number of parameter variations that have to be identified. Therefore, we introduce a novel approach for autotuning exascale applications based on a genetic multi-objective optimization algorithm integrated within the ASPIDE exascale computing framework. The approach considers multi-dimensional search space with support for pluggable objective functions, including execution time and energy requirements. Furthermore, the autotuner employs a machine learning-based event detection approach to detect events and anomalies during application execution, such as hardware failures or communication bottlenecks.","Title: Decentralized Machine Learning for Intelligent Health-Care Systems on the Computing Continuum\nYear: 2022\n\nAbstract: The introduction of electronic personal health records (EHRs) enables nationwide information exchange and curation among different health-care systems. However, current EHR systems are centrally orchestrated, which could potentially lead to a single point of failure.","Title: Hybrid On/Off Blockchain Approach for Vehicle Data Management, Processing and Visualization Exemplified by the ADAPT Platform.\nYear: 2022\n\nAbstract: Hybrid on/off-blockchain vehicle data management approaches have received a lot of attention in recent years. However, there are various technical challenges remained to deal with. In this paper we relied on real-world data from Austria to investigate the effects of connectivity on the transport of personal protective equipment. We proposed a three-step mechanism to process, simulate, and store/visualize aggregated vehicle datasets together with a formal pipeline process workflow model. To this end, we implemented a hybrid blockchain platform based on the hyperledger fabric and gluster file systems. The obtained results demonstrated efficiency and stability for both hyperledger fabric and gluster file systems and ability of the both on/off-blockchain mechanisms to meet the platform quality of service requirements","Title: Big Data Pipeline Scheduling and Adaptation on the Computing Continuum\nYear: 2022\n\nAbstract: The Computing Continuum, covering Cloud, Fog, and Edge systems, promises to provide on-demand resource-as-a-service for Internet applications with diverse requirements, ranging from extremely low latency to high-performance processing. However, eminent challenges in automating the resources management of Big Data pipelines across the Computing Continuum remain. The resource management and adaptation for Big Data pipelines across the Computing Continuum require significant research effort, as the current data processing pipelines are dynamic. In contrast, traditional resource management strategies are static, leading to inefficient pipeline scheduling and overly complex process deployment. To address these needs, we propose in this work a scheduling and adaptation approach implemented as a software tool to lower the technological barriers to the management of Big Data pipelines over the Computing Continuum. The approach separates the static scheduling from the run-time execution, empowering domain experts with little infrastructure and software knowledge to take an active part in the Big Data pipeline adaptation. We conduct a feasibility study using a digital healthcare use case to validate our approach. We illustrate concrete scenarios supported by demonstrating how the scheduling and adaptation tool and its implementation automate the management of the lifecycle of a remote patient monitoring, treatment, and care pipeline.","Title: Big Data Pipelines on the Computing Continuum: Tapping the Dark Data\nYear: 2022\n\nAbstract: The computing continuum enables new opportunities for managing big data pipelines concerning efficient management of heterogeneous and untrustworthy resources. We discuss the big data pipelines lifecycle on the computing continuum and its associated challenges, and we outline a future research agenda in this area.","Title: Matching-based Scheduling of Asynchronous Data Processing Workflows on the Computing Continuum\nYear: 2022\n\nAbstract: Today's distributed computing infrastructures en-compass complex workflows for real-time data gathering, transferring, storage, and processing, quickly overwhelming centralized cloud centers. Recently, the computing continuum that federates the Cloud services with emerging Fog and Edge devices represents a relevant alternative for supporting the next-generation data processing workflows. However, eminent challenges in automating data processing across the computing continuum still exist, such as scheduling heterogeneous devices across the Cloud, Fog, and Edge layers. We propose a new scheduling algorithm called C\n<sup xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">3</sup>\n -MATCH, based on matching theory principles, involving two sets of players negotiating different utility functions: 1) workflow microservices that prefer computing devices with lower data processing and queuing times; 2) computing continuum devices that prefer microservices with corresponding resource requirements and less data transmission time. We evaluate \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">$C^{3}$</tex>\n-MATCH using real-world road sign inspection and sentiment analysis workflows on a federated computing continuum across four Cloud, Fog, and Edge providers. Our combined simulation and real execution results reveal that \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">$C^{3}$</tex>\n-MATCH achieves up to 67% lower completion time than three state-of-the-art methods with 10 ms-1000 ms higher transmission time."],"string":"[\n \"Title: A new business model for massively multiplayer online games\\nYear: 2011\\n\\nAbstract: Today, highly successful Massively Multiplayer Online Games (MMOGs) have millions of registered users and hundreds of thousands of active concurrent players. To sustain their highly variable load, game operators over-provision a large static infrastructure capable of sustaining the game peak load, even though a large portion of the resources is unused most of the time. This inefficient resource utilisation has negative economic impacts by preventing any but the largest hosting centres from joining the market and dramatically increases prices. In this paper, we propose a new business model of hosting and operating MMOGs based on Cloud computing principles involving four actors: resource provider, game operator, game provider, and client. Our model efficiently provisions on-demand virtualised resources to game sessions based on their dynamic client load, which dramatically decreases prices and gives small and medium enterprises the opportunity of joining the market through zero initial investment. We validate our new model and its underlying business relationships through trace-based simulations utilising six months worth of monitoring data from a real-life MMOG using emulated resources from 16 of the largest Cloud resource providers currently on the market. We demonstrate that our model can operate state-of-the-art MMOGs with an average monthly gross profit of nearly $6 million excluding game purchase prices, overheads and taxation, while being able to maintain and control the QoS offered to all clients. Finally, we show how our approach is capable of operating next generation very highly interactive MMOGs with a small increase of 5.8% in the subscription price.\",\n \"Title: ZENTURIO: A Grid Service-Based Tool for Optimising Parallel and Grid Applications\\nYear: 2004\\n\\nAbstract: ZENTURIO (R. Prodan and T. Fahringer, \\\"ZENTURIO: A Grid Middleware-based Tool for Experiment Man- agement of Parallel and Distributed Applications\\\", Journal of Parallel and Distributed Computing, 2003. http://authors.elsevier.com/sd/article/S0743731503001977 (to appear)) is a semi-automatic experiment manage- ment tool for performance and parameter studies of parallel and distributed applications on cluster and Grid architectures. ZENTURIO has been designed as an Open Grid Services Architecture (OGSA) - compliant Grid application built on top of standard Web and Grid services technologies. In this paper we first comparatively present various issues from our transition to an Open Grid Services Infrastructure (OGSI) - compliant prototype. Then we introduce a generic framework for solving NP-complete optimisation problems for parallel and Grid applications. We present a case study on high throughput scheduling for large sets of computational tasks on the Grid using genetic algorithms. Our algorithm has a complexity of O(n2) and delivers a fivefold improvement in solution over 500 generations in a Grid with uniformly distributed computational resources.\",\n \"Title: Edutain@Grid: A Business Grid Infrastructure for Real-Time On-Line Interactive Applications\\nYear: 2008\\n\\nAbstract: Grid infrastructures are maturing to a point where they are attracting the interest of businesses in many application domains. While many large-scale on-line gaming platforms exist, they fail to take into consideration the potential business to business relationships when it comes to dynamic on-line game hosting. This work presents an initial implementation of the edutain@grid architecture to support business value chains identified for on-line gaming and e-learning application hosting. An analysis of business actors and value chains is presented briefly before a detailed description of the edutain@grid implementation. We also consider first results concerning how best to construct appropriate value chains using bipartite and bi-directional Service Level Agreements.\",\n \"Title: On Using ZENTURIO for Performance and Parameter Studies on Cluster and Grid Architectures\\nYear: 2003\\n\\nAbstract: Over the last decade, a dramatic increase has been ob- served in the need for generating and organising data in the course of large parameter studies, performance analy- sis, and software testings. We have developed the ZENTU- RIO experiment management tool for performance and pa- rameter studies on cluster and Grid architectures. In this paper we describe our experience with ZENTURIO for per- formance and parameter studies of a material science ker- nel, a three-dimensional particle-in-cell simulation, a fast fourier transform, and a financial modeling application. Ex- periments have been conducted on an SMP cluster with Fast Ethernet and Myrinet communication networks, using PBS (Portable Batch System) and GRAM (Globus Resource Allo- cation Manager) as job managers.\",\n \"Title: Dynamic Resource Provisioning in Massively Multiplayer Online Games\\nYear: 2011\\n\\nAbstract: Today's Massively Multiplayer Online Games (MMOGs) can include millions of concurrent players spread across the world and interacting with each other within a single session. Faced with high resource demand variability and with misfit resource renting policies, the current industry practice is to overprovision for each game tens of self-owned data centers, making the market entry affordable only for big companies. Focusing on the reduction of entry and operational costs, we investigate a new dynamic resource provisioning method for MMOG operation using external data centers as low-cost resource providers. First, we identify in the various types of player interaction a source of short-term load variability, which complements the long-term load variability due to the size of the player population. Then, we introduce a combined MMOG processor, network, and memory load model that takes into account both the player interaction type and the population size. Our model is best used for estimating the MMOG resource demand dynamically, and thus, for dynamic resource provisioning based on the game world entity distribution. We evaluate several classes of online predictors for MMOG entity distribution and propose and tune a neural network-based predictor to deliver good accuracy consistently under real-time performance constraints. We assess using trace-based simulation the impact of the data center policies on the quality of resource provisioning. We find that the dynamic resource provisioning can be much more efficient than its static alternative even when the external data centers are busy, and that data centers with policies unsuitable for MMOGs are penalized by our dynamic resource provisioning method. Finally, we present experimental results showing the real-time parallelization and load balancing of a real game prototype using data center resources provisioned using our method and show its advantage against a rudimentary client threshold approach.\",\n \"Title: ZENTURIO: a grid middleware-based tool for experiment management of parallel and distributed applications\\nYear: 2004\\n\\nAbstract: So far there is very limited tool support for performance-oriented code development, software testing, and parameter studies of high-performance scientific applications. Such an effort commonly requires to conduct a large number of experimental executions. We have developed the ZENTURIO experiment management tool for multi-experimental parameter and performance studies of parallel and distributed applications on cluster and Grid architectures. We describe the design and implementation of ZENTURIO as a Grid application based on state-of-the-art Grid middleware technologies, such as Web services standards and the Globus toolkit. We offer solutions for building Open Grid Services Architecture-compliant Grid services using a pure Web services implementation toolkit. Existing Web services standards such as Universal Description Discovery and Integration and WSDL compatibility are refined for a more coherent integration with the Grid characteristics. We compare our design and implementation with an early prototype built on top of the OGSI technology. We report results on using ZENTURIO for performance study of a material science kernel and parameter study of a financial application.\",\n \"Title: The edutain@grid Project\\nYear: 2007\\n\\nAbstract: edutain@grid is an exciting and ground breaking new project making use of Grid technology. The project will identify and define\\n a new class of applications that are highly significant for Grid computing but have not been studied in the past, which we\\n characterise as Real-Time Online Interactive Applications (ROIA). The distinctive features that make ROIA unique include large\\n user concurrency to a single application instance, ad-hoc connections, competition-oriented Virtual Organisations, real-time\\n interactive response, dynamically changing control and data application flows whilst maintaining high Quality of Service (QoS),\\n user friendly security, and novel Business-to-Consumer market models. In order to meet these challenges, the project team\\n will develop a new middleware layer that will allow ROIA to exploit Grid computing and validate the system using two pilot\\n applications from online gaming and e-learning domains.\",\n \"Title: Negotiation-Based Scheduling of Scientific Grid Workflows Through Advance Reservations\\nYear: 2010\\n\\nAbstract: In its broadest sense, scheduling of Grid applications can be viewed as a negotiation process between a scheduling service optimising user-centric objectives such as execution time, and a resource manager optimising provider-centric metrics such as resource utilisation or fairness. In this paper we enhance an existing list scheduling algorithm designed for minimising the workflow makespan with advance reservation-based negotiation functionality. As an instantiation of the new negotiation phase, we investigate two advance reservation functionality from the resource provider perspective: attentive and progressive. We illustrate through real-world experiments a two-fold benefit of our approach: improved execution predictability from the user's perspective, and higher resource utilisation fairness through a new progressive allocation strategy from the provider's perspective.\",\n \"Title: A Lightweight C++ Interface to MPI\\nYear: 2012\\n\\nAbstract: The Message Passing Interface (MPI) provides bindings for the three programming languages commonly used in High Performance Computing (HPC): C, C++ and Fortran. Unfortunately, MPI supports only the lowest common denominator of the three languages, providing a level of abstraction far lower than typical C++ libraries. Lately, after the decision of the MPI committee to deprecate and remove the C++ bindings from the MPI standard, programmers are forced to use either the C API or rely on third-party libraries. In this paper we present a lightweight, header-only C++ interface to MPI which uses object oriented and generic programming concepts to improve its integration into the C++ programming language. We compare our wrapper with a related approach called Boost. MPI showing how MPP facilitates the interaction with C++ objects. Performance wise, MPP outperforms Boost. MPI by reducing the interface overhead by a factor of eight. Additionally, MPP's handling of user-defined data types allows transferring of STL containers (e.g. std::list) up to 20 times faster than Boost. MPI for small linked lists by relying on software serialization.\",\n \"Title: Meeting Soft Deadlines in Scientific Workflows Using Resubmission Impact\\nYear: 2012\\n\\nAbstract: We propose a new heuristic called Resubmission Impact to support fault tolerant execution of scientific workflows in heterogeneous parallel and distributed computing environments. In contrast to related approaches, our method can be effectively used on new or unfamiliar environments, even in the absence of historical executions or failure trace models. On top of this method, we propose a dynamic enactment and rescheduling heuristic able to execute workflows with a high degree of fault tolerance, while taking into account soft deadlines. Simulated experiments of three real-world workflows in the Austrian Grid demonstrate that our method significantly reduces the resource waste compared to conservative task replication and resubmission techniques, while having a comparable makespan and only a slight decrease in the success probability. On the other hand, the dynamic enactment method manages to successfully meet soft deadlines in faulty environments in the absence of historical failure trace information or models.\",\n \"Title: Applying Advance Reservation to Increase Predictability of Workflow Execution on the Grid\\nYear: 2006\\n\\nAbstract: In this paper we present an extension to devise and implement advance reservation as part of the scheduling and resource management services of the ASKALON Grid application development and runtime environment. The scheduling service has been enhanced to offer a list of resources that can execute a specific task and to negotiatewith the resource manager about resources capable of processing tasks in the shortest possible time. We introduce progressive reservation approach which tries to allocate resources based on a fair-share principle. Experiments are shown that demonstrate the effectiveness of our approach, and that reflect different QoS parameters including performance, predictability, resource usage and resource fairness.\",\n \"Title: Autonomous massively multiplayer online game operation on unreliable resources\\nYear: 2013\\n\\nAbstract: Massively Multiplayer Online Games (MMOGs) are a new type of large-scale distributed applications characterised by seamless virtual worlds in which millions of world-wide players act and interact in real-time. Although for the past decade the number of MMOG players has grown exponentially, to the current tens of millions, this very growth may now hamper the progress of this important branch of the entertainment business. To guarantee Quality of Service (QoS) to a highly variable number of concurrent users, operators statically over-provision a large infrastructure capable of sustaining the game peak load, even though a large portion of the resources is unused most of the time. To address this problem, we propose a Cloud middleware-based system for autonomous operation of MMOGs. Our system provisions resources on-demand from multiple Cloud providers, automatically distributes the MMOG load between these resources, and self-heals when confronted with unforeseen resource failures. This new operational model allows small and medium enterprises to join the competitive MMOG market through near-zero initial infrastructure investment and operate MMOGs at given levels of QoS with small human intervention. We evaluate through simulations based on real-life MMOG traces the impact of resource availability on the QoS offered to the MMOG clients. We find that: (1) our proposed MMOG operation system can mitigate the negative effects of resource failures in under four minutes, (2) MMOG server consolidation in a resource-scarce environment can accentuate the negative effects of resource failures, and (3) the competition for resources can indirectly affect the QoS of the MMOG sessions.\",\n \"Title: A Sequential Cooperative Game Theoretic Approach to Storage-Aware Scheduling of Multiple Large-Scale Workflow Applications in Grids\\nYear: 2012\\n\\nAbstract: Scheduling large-scale applications in heterogeneous Grid and Cloud systems is a fundamental NP-complete problem for obtaining good performance and execution costs. We address the problem of scheduling an important class of large-scale Grid applications inspired from real-world, characterised by a large number of homogeneous, concurrent, and computationally-intensive tasks that are the main sources of performance, cost, and storage bottlenecks. We propose a new formulation of this problem based on a cooperative distributed game theoretic method for makespan and cost optimisation of a multiple such applications while fulfilling important storage constraints. We present experimental results using simulation and real-world applications that demonstrate the effectiveness of our method in terms of the solution delivered, algorithm execution time, and fairness compared to other related approaches.\",\n \"Title: MOHEFT: A multi-objective list-based method for workflow scheduling\\nYear: 2012\\n\\nAbstract: Nowadays, scientists and companies are confronted with multiple competing goals such as makespan in high-performance computing and economic cost in Clouds that have to be simultaneously optimized. Multi-objective scheduling of scientific workflows in distributed systems is therefore receiving increasing research attention. Most existing approaches typically aggregate all objectives in a single function, defined a-priori without any knowledge about the problem being solved, which negatively impacts the quality of the solutions. In contrast, Pareto-based approaches having as outcome a set of several (nearly-) optimal solutions that represent a tradeoff among the different objectives, have been scarcely studied. In this paper, we propose a new Pareto-based list scheduling heuristic that provides the user with a set of tradeoff optimal solutions from where the one that better suits the user requirements can be manually selected. We demonstrate the potential of MOHEFT for a bi-objective scheduling problem that optimizes makespan and economic cost in a Cloud-based computing scenario. We compare MOHEFT with two state-of-the-art approaches using different synthetic and real-world workflows: the classical HEFT algorithm used in single-objective scheduling and the SPEA2∗ genetic algorithm used for multi-objective optimisation problems.\",\n \"Title: A new business model for massively multiplayer online games (abstracts only)\\nYear: 2011\\n\\nAbstract: Today, highly successfulMassivelyMultiplayer Online Games (MMOGs) have millions of registered users and hundreds of thousands of active concurrent players. To sustain their highly variable load, game operators over-provision a large static infrastructure capable of sustaining the game peak load, even though a large portion of the resources is unused most of the time. This inefficient resource utilisation has negative economic impacts by preventing any but the largest hosting centres from joining the market and dramatically increases prices. In this paper, we propose a new business model of hosting and operating MMOGs based on Cloud computing principles involving four actors: resource provider, game operator, game provider, and client. Our model efficiently provisions on-demand virtualised resources to game sessions based on their dynamic client load, which dramatically decreases prices and gives small and medium enterprises the opportunity of joining the market through zero initial investment. We validate our new model and its underlying business relationships through trace-based simulations utilising six months worth of monitoring data from a real-life MMOG using emulated resources from 16 of the largest Cloud resource providers currently on the market. We demonstrate that our model can operate state-of-the-art MMOGs with an average monthly gross profit of nearly $6 million excluding game purchase prices, overheads and taxation, while being able to maintain and control the QoS offered to all clients. Finally, we show how our approach is capable of operating next generation very highly interactive MMOGs with a small increase of 5.8% in the subscription price.\",\n \"Title: Kalipy: A Tool for Online Performance Analysis of Grid Workflows through Event Correlation\\nYear: 2006\\n\\nAbstract: Static scheduling and execution of Grid workflows is prone to severe performance losses due to inaccurate predictions or the dynamic nature of the Grid environment. In this paper we present an online tool for analysing the performance overheads that appear during the real-time execution of workflow applications in Grid environments. We employ event correlation techniques and a distributed superpeer architecture in which each peer correlates local lowlevel activity and middleware events to infer performance overheads related to larger workflow regions at a higher level of abstraction. The rule-based correlation technique provides full extensibility to our approach that requires no source code modification. We demonstrate the functionality of our tool through online performance analysis of a real-world workflow application executed in a Grid environment. We present automatically generated online graphs of correlated events that promptly signal to the end-users the real reasons of run-time performance overheads in their executions.\",\n \"Title: Double Auction-based Scheduling of Scientific Applications in Distributed Grid and Cloud Environments\\nYear: 2011\\n\\nAbstract: Economy models have long been considered as a promising complement to the classical distributed resource management not only due of their dynamic and decentralized nature, but also because the concept of financial valuation of resources and services is an inherent part of any such model. In its broadest sense, scheduling of scientific applications in distributed Grid and Cloud environments can be regarded as a market-based negotiation between a scheduling service optimizing user-centric objectives (execution time, budget), and a resource manager optimizing provider-centric metrics (resource utilization, income, job throughput). In this paper, we propose a new instantiation of the negotiation protocol between the scheduler and resource manager using a market-based Continuous Double Auction (CDA) model. We analyze different scheduling strategies that can be applied and identify general strategic patterns that can lead to a fast and cheap work ow execution. In the experimental study, we demonstrate that under certain circumstances one can benefit by applying an aggressive scheduling strategy.\",\n \"Title: Dynamic Real-Time Resource Provisioning for Massively Multiplayer Online Games\\nYear: 2009\\n\\nAbstract: Massively Multiplayer Online Games (MMOG) are a class of computationally-intensive client-server applications with severe real-time Quality of Service (QoS) requirements, such the number of updates per second each client needs to receive from the servers for a fluent and realistic experience. To guarantee the QoS requirements, game providers currently over-provision a large amount of their resources, which makes the overall efficiency of provisioning and utilisation of resources rather low and prohibits any but the largest providers from joining the market. To address this deficiency, we propose a new prediction-based method for dynamic resource provisioning and scaling of MMOGs in distributed Grid environments. Firstly, a load prediction service anticipates the future game world entity distribution from historical trace data using a fast and flexible neural network-based method. On top of it, we developed generic analytical game load models used to foresee future hot-spots that congest the game servers and make the overall environment fragmented and unplayable. Finally, a resource allocation service performs dynamic load distribution, balancing, and migration of entities that keep the game servers reasonably loaded such that the real-time QoS requirements are maintained.\",\n \"Title: A novel graph based approach for automatic composition of high quality grid workflows\\nYear: 2009\\n\\nAbstract: The workflow paradigm is one of the most important programming models for the Grid. The composition of Grid workflows has been widely studied in the Grid community. However, there is still a lack of a general and efficient approach for automatic composition of Grid workflows. In this paper, we present a STRIPS (Stanford Research Institute Problem Solver) based formal definition of the Grid workflow composition problem, followed by a novel graph based algorithm for automatic composition of high quality (portable, fault tolerant and optimized) Grid workflows. Our algorithm searches for semantic descriptions of workflow activities, i.e., Activity Functions (AFs), defined by ontologies and composes them into Grid workflows using AF Data Dependence (ADD) graphs. The composition process consists of three phases: ADD graph creation, workflow extraction, and workflow optimization. The worst case complexity of our algorithm is quadratic in the number of AFs. An extension of our algorithm to compose Grid workflows with branches and loops is also presented. Experimental results illustrate the effectiveness and efficiency of our approach: (i) the measured worst case execution time of our algorithm further proofs the analyzed time complexity; (ii) the composition of the real world meteorology Grid workflow application MeteoAG with our algorithm takes approximate half a second; and (iii) the execution time of the MeteoAG workflow when running on the Austrian Grid is reduced by up to 25% and the speedup is increased by up to 2.24 by applying our workflow optimization techniques.\",\n \"Title: Tuning MPI Runtime Parameter Setting for High Performance Computing\\nYear: 2012\\n\\nAbstract: The performance of MPI applications on parallel computers can be considerably improved by tuning the runtime parameters provided by modern MPI libraries. However, due to the large and increasing number of tunable parameters, finding a parameter setting which optimizes the execution of several user programs on a chosen target machine is challenging. Existing tools execute input programs multiple times with varying parameter settings until a satisfying performance is reached. Several hundred runs of the input programs are nevertheless needed making this approach appealing only when the cost of the tuning phase can be amortized over many runs of the optimized programs. In this paper, we introduce a novel technique for tuning MPI runtime parameter values to better suit the underlying system architecture. The MPI parameter values are determined by performing the analysis of variance (ANOVA) on experimental data collected by randomly exploring the optimization space of a set of computational kernels commonly employed in High Performance Computing (HPC). We use our new technique to derive optimized values for 27 runtime parameters of the Open MPI library for two different parallel architectures. Results show an average performance improvement up to 20% for codes from the SPEC MPI 2007 benchmark suite with respect to Open MPI's default parameter setting.\",\n \"Title: Leveraging C++ meta-programming capabilities to simplify the message passing programming model\\nYear: 2011\\n\\nAbstract: Message passing is the primary programming model utilized for distributed memory systems. Because it aims at performance, the level of abstraction is low, making distributed memory programming often difficult and error-prone. In this paper, we leverage the expressivity and meta-programming capabilities of the C++ language to raise the abstraction level and simplify message passing programming. We redefine the semantics of the assignment operator to work in a distributed memory fashion and leave to the compiler the burden of generating the required communication operations. By enforcing more severe checks at compile-time we are able to statically capture common programming errors without causing runtime overhead.\",\n \"Title: On the Evaluation of JavaSymphony for Heterogeneous Multi-core Clusters.\\nYear: 2010\\n\\nAbstract: Programming hybrid heterogeneous multi-core cluster architectures is today an important topic in scientific and mainstream communities. To address this challenge, we developed JavaSymphony providing high-level programming abstraction and a middle-ware that facilitates the development and high-performance execution of Java applications on modern shared and distributed memory architectures. In this paper we present results of programming and executing a three-dimensional ray tracing application on a heterogeneous many-core cluster architecture.\",\n \"Title: Optimizing performance of automatic training phase for application performance prediction in the grid\\nYear: 2007\\n\\nAbstract: Automatic execution time prediction of the Grid applications plays a critical role in making the pervasive Grid more reliable and predictable. However, automatic execution time prediction has not been addressed due to the diversity of the Grid applications, usability of an application in multiple contexts, dynamic nature of the Grid, and concerns about result accuracy and time expensive experimental training. We introduce an optimized, low-cost, and efficient yet automatic training phase for automatic execution time prediction of Grid applications. Our approach is supported by intra - and inter-platform performance sharing and translation mechanisms. We are able to reduce the total number of experiments from an polynomial complexity to a linear complexity.\",\n \"Title: Multi-objective list scheduling of workflow applications in distributed computing infrastructures\\nYear: 2014\\n\\nAbstract: Executing large-scale applications in distributed computing infrastructures (DCI), for example modern Cloud environments, involves optimization of several conflicting objectives such as makespan, reliability, energy, or economic cost. Despite this trend, scheduling in heterogeneous DCIs has been traditionally approached as a single or bi-criteria optimization problem. In this paper, we propose a generic multi-objective optimization framework supported by a list scheduling heuristic for scientific workflows in heterogeneous DCIs. The algorithm approximates the optimal solution by considering user-specified constraints on objectives in a dual strategy: maximizing the distance to the user's constraints for dominant solutions and minimizing it otherwise. We instantiate the framework and algorithm for a four-objective case study comprising makespan, economic cost, energy consumption, and reliability as optimization goals. We implemented our method as part of the ASKALON environment (Fahringer et al., 2007) for Grid and Cloud computing and demonstrate through extensive real and synthetic simulation experiments that our algorithm outperforms related bi-criteria heuristics while meeting the user constraints most of the time.\",\n \"Title: SLA-based operation of massively multiplayer online games in competition-based environments\\nYear: 2013\\n\\nAbstract: To sustain the variable load of Massively Multiplayer Online Games (MMOGs) with guaranteed Quality of Service (QoS), game operators over-provision a static infrastructure capable of sustaining the peak load, even though a large portion of the resources is unused most of the time. This inefficient way of provisioning resources has negative impacts, leading to inefficient resource utilisation, high service prices, and limited market participation accessible only to the large companies. We propose a new ecosystem and model for hosting and operating MMOGs based on cloud computing principles involving four smaller and better focused business actors whose interaction is regulated through Service Level Agreements (SLAs): resource provider, game operator, game provider, and client. In our model, game providers lease operation SLAs from the game operators to satisfy all client requests and manage multiple distributed MMOG sessions. In turn, game operators efficiently lease on-demand cloud resources based on the dynamic MMOG load and ensure proper game operation that maintains QoS to all clients. In this paper, we focus on the business interaction between the game provider and the game operator by defining the SLA terms and the underlying negotiation protocol, including a model for compensations for QoS violations. We propose a method for ranking operational offers based on price, compensation and resource fitness, and study its impact on game provider's profit in an environment with several providers competing for SLAs from multiple game operators.\",\n \"Title: Askalon: A Tool Set For Cluster And Grid Computing\\nYear: 2005\\n\\nAbstract: Performance engineering of parallel and distributed applications is a complex task that iterates through various phases, ranging from modeling and prediction, to performance measurement, experiment management, data collection, and bottleneck analysis. There is no evidence so far that all of these phases should/can be integrated into a single monolithic tool. Moreover, the emergence of computational Grids as a common single wide-area platform for high-performance computing raises the idea to provide tools as interacting Grid services that share resources, support interoperability among different users and tools, and, most importantly, provide omnipresent services over the Grid. We have developed the ASKALON tool set to support performance-oriented development of parallel and distributed (Grid) applications. ASKALON comprises four tools, coherently integrated into a service-oriented architecture. SCALEA is a performance instrumentation, measurement, and analysis tool of parallel and distributed applications. ZENTURIO is a general purpose experiment management tool with advanced support for multi-experiment performance analysis and parameter studies. AKSUM provides semi-automatic highlevel performance bottleneck detection through a special-purpose performance property specification language. The PerformanceProphet enables the user to model and predict the performance of parallel applications at the early stages of development. In this paper we describe the overall architecture of the ASKALON tool set and outline the basic functionality of the four constituent tools. The structure of each tool is based on the composition and sharing of remote Grid services, thus enabling tool interoperability. In addition, a data repository allows the tools to share the common application performance and output data that have been derived by the individual tools. A service repository is used to store common portable Grid service implementations. A general-purpose Factory service is employed to create service instances on arbitrary remote Grid sites. Discovering and dynamically binding to existing remote services is achieved through registry services. The ASKALON visualization diagrams support both online and postmortem visualization of performance and output data. We demonstrate the usefulness and effectiveness of ASKALON by applying the tools to real-world applications. Copyright (C) 2005 John Wiley Sons, Ltd.\",\n \"Title: Execution time prediction for grid infrastructures based on runtime provenance data\\nYear: 2013\\n\\nAbstract: An accurate performance prediction service can be very useful for resource management and the scheduler service and help them make better resource utilization decisions by providing better execution time estimates. In this paper we present a novel approach of predicting the execution time of computational tasks for Grid infrastructures using machine learning models based on multilayer perceptron combined with a principal feature selection algorithm for selecting the most important runtime features. Our technique uses runtime provenance information as input to multilayer perceptron (MLP) based neural network which trains a model that can predict the execution time of programs with reasonably good accuracy. For the development and training of our machine learning models, we used provenance data collected from three different computational Grids by executing real-world applications. By using our MLP based method we were able to minimise the prediction error to as low as 22% for real-world applications on various Grid infrastructures.\",\n \"Title: Using a new event-based simulation framework for investigating resource provisioning in Clouds\\nYear: 2011\\n\\nAbstract: Today, Cloud computing proposes an attractive alternative to building large-scale distributed computing environments by which resources are no longer hosted by the scientists' computational facilities, but leased from specialised data centres only when and for how long they are needed. This new class of Cloud resources raises new interesting research questions in the fields of resource management, scheduling, fault tolerance, or quality of service, requiring hundreds to thousands of experiments for finding valid solutions. To enable such research, a scalable simulation framework is typically required for early prototyping, extensive testing and validation of results before the real deployment is performed. The scope of this paper is twofold. In the first part we present GroudSim, a Grid and Cloud simulation toolkit for scientific computing based on a scalable simulation-independent discrete-event engine. GroudSim provides a comprehensive set of features for complex simulation scenarios from simple job executions on leased computing resources to file transfers, calculation of costs and background load on resources. Simulations can be parameterised and are easily extendable by probability distribution packages for failures which normally occur in complex distributed environments. Experimental results demonstrate the improved scalability of GroudSim compared to a related process-based simulation approach. In the second part, we show the use of the GroudSim simulator to analyse the problem of dynamic provisioning of Cloud resources to scientific workflows that do not benefit from sufficient Grid resources as required by their computational demands. We propose and study four strategies for provisioning and releasing Cloud resources that take into account the general leasing model encountered in today's commercial Cloud environments based on resource bulks, fuzzy descriptions and hourly payment intervals. We study the impact of our techniques to the overall execution time, overall cost and cost per unit of saved time with respect to various instance types offered by the Amazon EC2.\",\n \"Title: Scientific computing with Google App Engine\\nYear: 2013\\n\\nAbstract: We investigate the potential of Google App Engine (GAE) for scientific computing. We designed a generic master-slave framework that enables implementation and integration of new algorithms by instantiating one interface and two abstract classes. Applications are dynamically scheduled, executed, and monitored by the framework on the Google infrastructure, including a fault tolerant resubmission mechanism upon certain failures. We describe the implementation of a parallel rank sort algorithm using our framework and analyse the speedup, the execution overheads, and the cost of executing it on the Google Cloud infrastructure. Experimental results show that important speedup can be obtained from using GAE, especially for embarrassingly parallel Monte Carlo simulations. The two main obstacles that hinder the performance of applications on GAE are the resource quotas upon free use (especially the 30 s sequential execution time limit) and the middleware overheads. Cost-wise, GAE offers opportunities for cheaper computation due to the CPU cycle-based payment granularity, as opposed to the hourly billing intervals of other providers such as Amazon EC2.\",\n \"Title: Characterizing, Modeling and Predicting Dynamic Resource Availability in a Large Scale Multi-purpose Grid\\nYear: 2008\\n\\nAbstract: The functional heterogeneity of computational Grids has highly increased due to inclusion of resources other than dedicated to Grid, like from non-dedicated desktop Grids, on-demand systems and even from P2P systems and mobile Grids. At such a diversified scale, resources exhibit different availability properties mainly due to administrators' policies for resource availability in the Grid, and their failure/unavailability properties. These make resources' availability predictions for optimized resource selection, a challenging problem. Addressing this problem, we characterize resource availability properties against their availability policies to understand their availability behavior and quantify it through availability models. We further exploit the availability/failure properties to make predictions about their availability through pattern recognition and classification. We have achieved, on average, accuracy of more than 90% and 75% in our predictions for resource instance availability and lifetime respectively.\",\n \"Title: A Multi-objective Approach for Workflow Scheduling in Heterogeneous Environments\\nYear: 2012\\n\\nAbstract: Traditional scheduling research usually targets make span as the only optimization goal, while several isolated efforts addressed the problem by considering at most two objectives. In this paper we propose a general framework and heuristic algorithm for multi-objective static scheduling of scientific workflows in heterogeneous computing environments. The algorithm uses constraints specified by the user for each objective and approximates the optimal solution by applying a double strategy: maximizing the distance to the constraint vector for dominant solutions and minimizing it otherwise. We analyze and classify different objectives with respect to their impact on the optimization process and present a four-objective case study comprising make span, economic cost, energy consumption, and reliability. We implemented the algorithm as part of the ASKALON environment for Grid and Cloud computing. Results for two real-world applications demonstrate that the solutions generated by our algorithm are superior to user-defined constraints most of the time. Moreover, the algorithm outperforms a related bi-criteria heuristic and a bi-criteria genetic algorithm.\",\n \"Title: A Distributed Object-Oriented Framework for Tool Development\\nYear: 2000\\n\\nAbstract: In recent years, there has been a substantial increase in the availability and quality of software engineering tools; such tools are invaluable in ensuring program correctness and identifying performance problems. The majority of these, however, does not interoperate and are available on a limited platform set.We analyze such deficiencies and propose an extensible architecture for a distributed software engineering tool framework using CORBA object-oriented technology. The resulting framework provides a unified interface for parallel, distributed and single-processor systems, facilitates tool development, promotes tool interoperability, and can be extended by the integration of new tools and services. This flexibility is demonstrated by the specification of an extension to support the MPI programming paradigm and a wide selection of tools that have been built using the system.\",\n \"Title: GroudSim: an event-based simulation framework for computational grids and clouds\\nYear: 2010\\n\\nAbstract: We present GroudSim, a Grid and Cloud simulation toolkit for scientific applications based on a scalable simulation-independent discrete-event core. GroudSim provides a comprehensive set of features for complex simulation scenarios from simple job executions on leased computing resources to calculation of costs, and background load on resources. Simulations can be parameterised and are easily extendable by probability distribution packages for failures which normally occur in complex environments. Experimental results demonstrate the improved scalability of GroudSim compared to a related process-based approach.\",\n \"Title: Multi-Objectiveworkflow Scheduling: An Analysis Of The Energy Efficiency And Makespan Tradeoff\\nYear: 2013\\n\\nAbstract: While in the past scheduling algorithms were almost exclusively targeted at optimizing applications' makespan, today they must simultaneously optimise several goals. Among these goals, energy efficiency is receiving increasing attention for environmental and financial reasons. In contrast to related work that optimises energy consumption as a single objective function, we reformulate in this paper the problem as a bi-objective optimisation by considering both makespan and energy as goals. We study the potential benefits of using a Pareto-based workflow scheduling algorithm called MOHEFT using realistic energy consumption and performance models for task executions. We analyse the tradeoff solutions computed by MOHET for different workflows (different in shapes and sizes) in different execution scenarios (different resources in terms of energy consumption). The obtained results show that our bi-objective approach found in some cases schedules that reduce the energy consumption up to 85% with only 3.3% of makespan concessions.\",\n \"Title: Budget-Constrained Resource Provisioning for Scientific Applications in Clouds\\nYear: 2013\\n\\nAbstract: Public commercial clouds emerged as new and attractive resource provisioning option for scientific computing. This new alternative raises new challenges for users of such clouds, since optimizing the completion time of scientific applications might substantially increase the monetary cost of leasing cloud resources. In this paper, we first propose a set of basic rescheduling operations covering a broad set of scenarios for reducing the costs of running scientific workflows in clouds. Based on them, we design two heuristic scheduling algorithms. The first algorithm aims at reducing the cost of resource provisioning while still attaining the optimal make span. The second algorithm further reduces the costs to meet a budget constraint with a small increase in the make span. The experiments conducted using real-world and synthetic workflow applications demonstrate important benefits compared to related state-of-the-art approaches.\",\n \"Title: Towards a general model of the multi-criteria workflow scheduling on the grid\\nYear: 2009\\n\\nAbstract: Workflow scheduling on the Grid becomes more challenging when multiple scheduling criteria are considered. Existing studies provide different approaches to the multi-criteria Grid workflow scheduling problem, and address different variants of the problem. A profound understanding of the problem's nature can be an important step towards more generic scheduling approaches. Based on the related work and on our own experience, we propose several novel taxonomies of the problem, considering five facets: workflow model, scheduling criteria, scheduling process, resource model, and task model. We make a survey of the existing related work, and classify it according to the proposed taxonomies, identifying the most common use cases and the areas that have not been sufficiently explored yet.\",\n \"Title: A sequential cooperative game theoretic approach to scheduling multiple large-scale applications in grids.\\nYear: 2014\\n\\nAbstract: Scheduling large-scale applications in heterogeneous distributed computing systems is a fundamental NP-complete problem that is critical to obtaining good performance and execution cost. In this paper, we address the scheduling problem of an important class of large-scale Grid applications inspired by the real world, characterized by a huge number of homogeneous, concurrent, and computationally intensive tasks that are the main sources of performance, cost, and storage bottlenecks. We propose a new formulation of this problem based on a cooperative distributed game-theory-based method applied using three algorithms with low time complexity for optimizing three important metrics in scientific computing: execution time, economic cost, and storage requirements. We present comprehensive experiments using simulation and real-world applications that demonstrate the effectiveness of our approach in terms of time and fairness compared to other related algorithms. We schedule large-scale Grid applications with a huge identical parallel tasks.We formulate the problem using a cooperative distributed game-theory-based method.We design three algorithms for optimizing time, cost, and storage requirements.We evaluate our method using simulation and real-world applications.We compare our results with related algorithms.\",\n \"Title: Prediction-based real-time resource provisioning for massively multiplayer online games\\nYear: 2009\\n\\nAbstract: Massively Multiplayer Online Games (MMOGs) are a class of computationally intensive client-server applications with severe real-time Quality of Service (QoS) requirements, such as the number of updates per second each client needs to receive from the servers for a fluent and realistic experience. To guarantee the QoS requirements, game providers currently over-provision a large amount of their resources, which makes the overall efficiency of provisioning and utilization of resources rather low and prohibits any but the largest providers from joining the market. To address this deficiency, we propose a new prediction-based method for dynamic resource provisioning and scaling of MMOGs in distributed Grid environments. Firstly, a load prediction service anticipates the future game world entity distribution from historical trace data using a fast and flexible neural network-based method. On top of it, we developed generic analytical game load models used to foresee future hot-spots that congest the game servers and make the overall environment fragmented and unplayable. Finally, a resource allocation service performs dynamic load distribution, balancing, and migration of entities that keep the game servers reasonably loaded such that the real-time QoS requirements are maintained. Experimental results based on a realistic simulation environment demonstrate the advantages of our prediction service compared to other conventional methods, especially due to its ability to adapt to different user load patterns, and a reduction of the average over-allocation from 250% (in the case of static over-provisioning) to around 25% using our dynamic provisioning method.\",\n \"Title: A Framework for an Interoperable Tool Environment (Research Note)\\nYear: 2000\\n\\nAbstract: Software engineering tools are indispensable for parallel and distributed program development, yet the desire to combine them to provide enhanced functionality has still to be realised. Existing tool environments integrate a number of tools, but do not achieve interoperability and lack extensibility. Integration of new tools can necessitate the redesign of the whole system. We describe the FIRST tool framework, its initial tool-set, classify different types of tool interaction and describe a variety of tool scenarios which are being used to investigate tool interoperability.\",\n \"Title: Monitoring and fault tolerance for real-time online interactive applications\\nYear: 2009\\n\\nAbstract: The edutain@grid European project [1] is developing a support platform for deployment, management and execution of Real-Time Online Interactive Applications (ROIA) on Grid. In this paper we present a monitoring system we developed which collects data from all the resources in a distributed environment and from the ROIA managed by our platform. We also describe a fault tolerance service which addresses not only the faults commonly encountered in distributed systems, but also faults manifesting at service level, within the platform's management services. Finally, a use-case consisting of the platform running a massively multiplayer online game as a concrete ROIA, is presented in order to demonstrate the roles of the monitoring and fault tolerance services.\",\n \"Title: Meteorological simulations in the cloud with the ASKALON environment\\nYear: 2012\\n\\nAbstract: Precipitation in mountainous regions is an essential process in meteorological research for its strong impact on the hydrological cycle. To support scientists, we present the design of a meteorological application using the ASKALON environment comprising graphical workflow modeling and execution in a Cloud computing environment. We illustrate performance results that demonstrate that, although limited by Amdahl's law, our workflow can gain important speedup when executed in a virtualized Cloud environment with important operational cost reductions. Results from the meteorological research show the usefulness of our model for determining precipitation distribution in the case of two field campaigns over Norway.\",\n \"Title: Semi-automatic composition of ontologies for ASKALON grid workflows\\nYear: 2011\\n\\nAbstract: Automatic workflow composition with the help of ontologies has been addressed by numerous researchers in the past. While ontologies are very useful for automatic and semiautomatic workflow composition, ontology creation itself remains a very important and complex task. In this paper we present a novel tool to synthesize ontologies for the Abstract Grid Workflow Language (AGWL) which has been used for years to successfully create Grid workflow applications at a high level of abstraction. In order to semi-automatically generate ontologies we use an AGWL Ontology (AGWO - an ontological description of the AGWL language), structural information of one or several input workflows of a given application domain, and semantic enrichment of the structural information with the help of the user. Experiments based on two separate application domains (movie rendering and meteorology) will be shown that demonstrate the effectiveness of our approach by semi-automatically generating ontologies which are then used to automatically create workflow applications.\",\n \"Title: The impact of virtualisation on the performance and operational costs of Massively Multiplayer Online Games\\nYear: 2010\\n\\nAbstract: To serve millions of players, Massively Multiplayer Online Game (MMOG) operators pre-provision and then maintain thousands of computer resources. We investigate a hybrid resource provisioning model that uses smaller and cheaper data centers, complemented during peak hours by virtualised cloud computing resources. Through trace-based simulation and empirical experimentation, we assess the impact of provisioning virtualised cloud resources, analyse the virtualisation overhead, and compare provisioning of virtualised resources with resource ownership. Using a simple cost model, we also investigate the costs of hosting MMOGs on the resources leased independently from three commercial cloud providers, including Amazon.\",\n \"Title: A Web Service-Based Experiment Management System for the Grid\\nYear: 2003\\n\\nAbstract: We have developed the ZENTURIO experiment management system for performance studies, parameter studies, and software testing of parallel applications on cluster and Grid architectures. In this paper we describe the design and implementation of ZENTURIO as a Web service-based distributed architecture. We highlight limitations of Web services for deployment on the Grid and propose solutions. Experiment Generator and Experiment Executor services are used to generate, respectively submit and manage experiments on remote Grid sites. A UDDI-based Service Repository publishes information about existing Web service implementations. Factory and Registry services are employed to create and register Web services on the Grid. A GMA-compliant event infrastructure supports high-level notifications about important events. We compare our design with the Open Grid Service Architecture (OGSA) and highlight similarities and differences. We report results of using ZENTURIO to conduct performance analysis of a material science code executed on the Grid by using GRAM.\",\n \"Title: An Information System for Real-Time Online Interactive Applications\\nYear: 2008\\n\\nAbstract: The edutain@grid European project [1] is developing a support platform for deployment, management and execution of Real-Time Online Interactive Applications (ROIA) on Grid. In this paper we present an information system designed by the edutain@grid project which provides support for ROIA deployment and monitoring, and offers a generic frontend for ROIA-specific optimisations. We conduct a variety of experiments that justify various decisions of our design, and investigate the performance and scalability of our system with respect to various types of queries.\",\n \"Title: A Bi-Criteria Truthful Mechanism for Scheduling of Workflows in Clouds\\nYear: 2011\\n\\nAbstract: Commercial distributed systems such as Clouds are managed by selfish providers that strategically try to increase their revenues regardless of the utility of other providers and users. These selfish behaviors affect the efficiency of using such environments. In this paper, based on a general game theoretic truthful reverse auction mechanism, we investigate the scheduling problem of dependent tasks on distributed Cloud resources owned by selfish providers. The social cost of the game is to minimize the make span and monetary cost simultaneously. Extensive simulation experiments show that the schedules obtained are approximately Pareto optimal.\",\n \"Title: A Performance Analysis of EC2 Cloud Computing Services for Scientific Computing.\\nYear: 2009\\n\\nAbstract: Cloud Computing is emerging today as a commercial infrastructure that eliminates the need for maintaining expensive computing hardware. Through the use of virtualization, clouds promise to address with the same shared set of physical resources a large user base with different needs. Thus, clouds promise to be for scientists an alternative to clusters, grids, and supercomputers. However, virtualization may induce significant performance penalties for the demanding scientific computing workloads. In this work we present an evaluation of the usefulness of the current cloud computing services for scientific computing. We analyze the performance of the Amazon EC2 platform using micro-benchmarks and kernels. While clouds are still changing, our results indicate that the current cloud services need an order of magnitude in performance improvement to be useful to the scientific community.\",\n \"Title: A Collection of Java Class Libraries for Stochastic Modeling and Simulation\\nYear: 2002\\n\\nAbstract: The paper focuses on a set of Java class libraries for stochastic modeling and simulation, created and implemented by the Department of Mathematics and Informatics from Iuliu Ha驴ieganu University Cluj-Napoca. Previous research has shown that stochastic models are advantageous tools for representation of the real world activities. Due to actual spread of fast and inexpensive computational power everywhere in the world, the best approach is to model the real phenomenon as faithfully as possible, and then rely on a simulation study to analyze it. We approached the new bootstrapping methods, which are very useful in analyzing a simulation. Our aim is to implement bootstrapping strategies in software tools which will be used by both the teaching staff and the students in didactic and research activities, with the purpose to optimize the use of substances and reactants.\",\n \"Title: Soft Benchmarks-Based Application Performance Prediction Using a Minimum Training Set\\nYear: 2006\\n\\nAbstract: Application execution time prediction is of key importance in making decisions about efficient usage of Grid resources. Grid services lack support of a generic application execution time prediction service due to environment specific solutions provided by the existing prediction techniques. To remedy this, we present a generic and comprehensive system to provide execution time predictions of applications on different Grid-sites. Our system is based on a two layered training phase to minimize the training effort, which is our first main contribution. The training phase is driven by a novel experimental design. We also introduce a mechanism of sharing performance measurements across the Grid, on the basis of soft benchmarks, which is our second contribution. Both of these phases support our prediction engine to serve robust predictions. Experiments from the prototype implementation are shown to demonstrate the effectiveness of our proposed system.\",\n \"Title: Overhead Analysis of Scientific Workflows in Grid Environments\\nYear: 2008\\n\\nAbstract: Scientific workflows are a topic of great interest in the Grid community that sees in the workflow model an attractive paradigm for programming distributed wide-area Grid infrastructures. Traditionally, the Grid workflow execution is approached as a pure best-effort scheduling problem that maps the activities onto the Grid processors based on appropriate optimisation or local matchmaking heuristics such that the overall execution time is minimised. Even though such heuristics often deliver effective results, the execution in dynamic and unpredictable Grid environments is prone to severe performance losses that must be understood for minimising the completion time or for efficient use of high- erformance resources. In this paper, we propose a new systematic approach to help the scientists and middleware developers understand the most severe sources of performance losses that occur when executing scientific workflows in dynamic Grid environments. We introduce an ideal model for the lowest execution time that can be achieved by a workflow and explain the difference to the real measured Grid execution time based on a hierarchy of performance overheads for Grid computing. We describe how to systematically measure and compute the overheads from individual activities to larger workflow regions and adjust well-known parallel processing metrics to the scope of Grid computing, including speedup and efficiency. We present a distributed online tool for computing and analysing the performance overheads in realtime based on event correlation techniques and introduce several performance contracts, as quality of service parameters to be enforced during the workflow execution beyond traditional best-effort practices. We illustrate our method through post-mortem and online performance analysis of two real-world workflow applications executed in the Austrian Grid environment.\",\n \"Title: JavaSymphony: a programming and execution environment for parallel and distributed many-core architectures\\nYear: 2010\\n\\nAbstract: Today, software developers face the challenge of re-engineering their applications to exploit the full power of the new emerging many-core processors. However, a uniform high-level programming model and interface for parallelising Java applications is still missing. In this paper, we propose a new Java-based programming model for shared memory many-core parallel computers as an extension to the JavaSymphony distributed programming environment. The concept of dynamic virtual architecture allows modelling of hierarchical resource topologies ranging from individual cores and multi-core processors to more complex parallel computers and distributed Grid infrastructures. On top of this virtual architecture, objects can be explicitly distributed, migrated, and invoked, enabling high-level user control of parallelism, locality, and load balancing. We evaluate the JavaSymphony programming model and the new shared memory run-time environment for six real applications and benchmarks on a modern multi-core parallel computer. We report scalability analysis results that demonstrate that JavaSymphony outperforms pure Java implementations, as well as other alternative related solutions.\",\n \"Title: Multi-objective workflow scheduling in Amazon EC2\\nYear: 2014\\n\\nAbstract: Nowadays, scientists and companies are confronted with multiple competing goals such as makespan in high-performance computing and economic cost in Clouds that have to be simultaneously optimised. Multi-objective scheduling of scientific applications in these systems is therefore receiving increasing research attention. Most existing approaches typically aggregate all objectives in a single function, defined a-priori without any knowledge about the problem being solved, which negatively impacts the quality of the solutions. In contrast, Pareto-based approaches having as outcome a set of (nearly) optimal solutions that represent a tradeoff among the different objectives, have been scarcely studied. In this paper, we analyse MOHEFT, a Pareto-based list scheduling heuristic that provides the user with a set of tradeoff optimal solutions from which the one that better suits the user requirements can be manually selected. We demonstrate the potential of our method for multi-objective workflow scheduling on the commercial Amazon EC2 Cloud. We compare the quality of the MOHEFT tradeoff solutions with two state-of-the-art approaches using different synthetic and real-world workflows: the classical HEFT algorithm for single-objective scheduling and the SPEA2* genetic algorithm used in multi-objective optimisation problems. The results demonstrate that our approach is able to compute solutions of higher quality than SPEA2*. In addition, we show that MOHEFT is more suitable than SPEA2* for workflow scheduling in the context of commercial Clouds, since the genetic-based approach is unable of dealing with some of the constraints imposed by these systems.\",\n \"Title: Performance Analysis and Benchmarking of the Intel SCC\\nYear: 2011\\n\\nAbstract: Over the past years there has been a steady change in CPU design towards both many-core processors and power-aware hardware architectures. These two trends are combined in the Intel Single-chip Cloud Computer (SCC), an experimental prototype with 48 Pentium cores created by Intel Labs. The SCC is a highly configurable many-core chip which provides unique opportunities to optimize run time, communication and memory access as well as power/energy consumption of parallel programs. The aim of this paper is to characterize the performance behavior of the chip with various power settings, mappings of processes/cores to memory controllers, etc through benchmarking. Analytical models are used to verify and interpret the results. Conclusions drawn from our benchmark outcomes are that data exchange based on message passing is faster than shared memory data exchange. Contrary to popular belief, lowest energy consumption is not achieved for the fastest execution time. Furthermore in order to improve the memory access behavior one should increase the clock frequency of both, mesh network and memory controllers. In general, the results of our investigations can be used to analyze the effect of power settings and architecture properties on the performance and energy consumption of parallel programs as well as assist in choosing appropriate settings for specific workloads.\",\n \"Title: Specification-correct and scalable coordination of Grid applications\\nYear: 2007\\n\\nAbstract: The so called ''invisible Grid'', transparent to application developers, is still far from being a reality. One reason is that the workflow model, which emerged as a widely accepted paradigm for high-level composition of Grid applications, is based on a low-level imperative programming model prone to programming errors. The issue of developing correct (bug-free) Grid applications has not been addressed by the community. We propose a new model for building Grid applications based on two programming phases: (1) formal functional specification, written by the application developer not interested in any Grid-related issues; (2) imperative workflow-based coordination, written by the computer scientist, which ports and efficiently executes the specification onto the Grid. A correctness checker automatically connects both parts at compile-time and insures the correct execution of the workflow coordination with respect to the formal specification. We validate our approach for three scientific applications and show real-world experimental results that demonstrate the scalability of our coordination model and the fact that the overhead introduced by our correctness checker is insignificant when compared to the latencies exhibited by the Grid middleware software.\",\n \"Title: Optimising parallel applications on the grid using irregular array distributions\\nYear: 2005\\n\\nAbstract: In this paper we propose a new approach for scheduling data parallel applications on the Grid using irregular array distributions. We implement the scheduler as a new case study for using a general purpose experiment management tool that we developed in previous work for performance tuning and optimisation of scientific applications. We report results on scheduling a Jacobi relaxation application on a simulation testbed of the Austrian Grid [2] using a problem independent plug-and-play genetic algorithm.\",\n \"Title: A Truthful Dynamic Workflow Scheduling Mechanism for Commercial Multicloud Environments\\nYear: 2013\\n\\nAbstract: The ultimate goal of cloud providers by providing resources is increasing their revenues. This goal leads to a selfish behavior that negatively affects the users of a commercial multicloud environment. In this paper, we introduce a pricing model and a truthful mechanism for scheduling single tasks considering two objectives: monetary cost and completion time. With respect to the social cost of the mechanism, i.e., minimizing the completion time and monetary cost, we extend the mechanism for dynamic scheduling of scientific workflows. We theoretically analyze the truthfulness and the efficiency of the mechanism and present extensive experimental results showing significant impact of the selfish behavior of the cloud providers on the efficiency of the whole system. The experiments conducted using real-world and synthetic workflow applications demonstrate that our solutions dominate in most cases the Pareto-optimal solutions estimated by two classical multiobjective evolutionary algorithms.\",\n \"Title: Bi-criteria Scheduling of Scientific Workflows for the Grid\\nYear: 2008\\n\\nAbstract: The drift towards new challenges in grid computing, including the utility grid paradigm and service level agreements based on quality-of-service guarantees, implies the need for new, robust, multi-criteria scheduling algorithms that can be applied by the user in an intuitive way. Multiple scheduling criteria addressed by the related grid research include execution time, the cost of running a task on a machine, reliability, and different data quality metrics. The existing bi-criteria scheduling approaches are usually dedicated for certain criterion pairs only that require the user to define one's preferences either as weights assigned to the criteria or as fixed constraints defined for one of the criteria. These requirements are often not feasible for the user and not suited to the specificity of the multi- criteria scheduling problem. We propose a novel requirement specification method based on a sliding constraint, and we model the problem as an extension of the multiple-choice knapsack problem. We propose a general bi-criteria scheduling heuristic called dynamic constraint algorithm (DCA) based on dynamic programming, dedicated to the problem model defined by us. In the experimental study, we show that in most of the problem variants, DCA outperforms two existing algorithms designed for the same problem. It also shows relatively low scheduling times for workflows of medium size.\",\n \"Title: Online Gaming In The Cloud\\nYear: 2010\\n\\nAbstract: Computational Grids and Clouds remain highly specialized technologies that are only used by scientists and large commercial organizations. To overcome this gap, University of Innsbruck is conducting basic research that is unusual compared with previous academic research projects in that it addresses a new class of application that appeals to the public for leisure reasons: Massively Multiplayer Online Games. Online games have the potential to raise strong interest, providing societal benefits through increased technological awareness and engagement. By standardizing on a Cloud-based platform and removing the need for large investments in hosting facilities, this research may remove the technical barrier and the costs of hosting MMOGs, and thus significantly increase the number of players while keeping the high quality of responsiveness of action games.\",\n \"Title: Specification-correct and Scalable Coordination of Scientific Applications in Grid Environments\\nYear: 2007\\n\\nAbstract: The workflow model for composing Grid applications is based on an imperative model of computation prone to programming errors, which is an issue yet to consider in the Grid community. In this paper, we propose a new unconventional model for programming Grid applications based on two programming phases: (1) formal functional specification, written by the application scientist not interested in any Grid-related issues and representing \\\"what\\\" has to be computed; (2) imperative workflow-based coordination, written by the computer scientist which indicates \\\"how\\\" to efficiently execute the specification on the Grid. A correctness checker automatically connects both parts at compile- time and ensures the correct execution of the workflow coordination with respect to the formal specification. We validate our approach for three real-world applications and show experimental results that demonstrate the correctness and scalability of our coordination model.\",\n \"Title: Scheduling of scientific workflows in the ASKALON grid environment\\nYear: 2005\\n\\nAbstract: Scheduling is a key concern for the execution of performance-driven Grid applications. In this paper we comparatively examine different existing approaches for scheduling of scientific workflow applications in a Grid environment. We evaluate three algorithms namely genetic, HEFT, and simple \\\"myopic\\\" and compare incremental workflow partitioning against the full-graph scheduling strategy. We demonstrate experiments using real-world scientific applications covering both balanced (symmetric) and unbalanced (asymmetric) workflows. Our results demonstrate that full-graph scheduling with the HEFT algorithm performs best compared to the other strategies examined in this paper.\",\n \"Title: IaaS cloud benchmarking: approaches, challenges, and experience\\nYear: 2013\\n\\nAbstract: Over the past five years, Infrastructure-as-a-Service (IaaS) clouds have grown into the branch of ICT that offers services related to on-demand lease of storage, computation, and network. One of the major impediments in the selection and even use of (commercial) IaaS clouds is the lack of benchmarking results, that is, the lack of trustworthy quantitative information that allows (potential) cloud users to compare and reason about IaaS clouds. In this talk we discuss empirical approaches to quantitative evaluation of IaaS clouds, toward cloud benchmarking. Both industry and academia have used empirical approaches for years, but with limited success for IaaS clouds and similar systems (e.g., grids) due to the complexity and size of challenges. We present initial results of our research into cloud workload characterization, including Big Data applications. We present the lessons we have learned in developing the SkyMark framework for cloud performance evaluation and the results of our SkyMark-based investigation of three research questions: What is the performance of production IaaS cloud services? How variable is the performance of widely used production cloud services? and What is the impact on performance of the provisioning and allocation policies that interact with IaaS clouds? In contrast to previous attempts, our research combines empirical and other approaches, for example modeling and simulation, for gaining more insight; is based on a combination of short-term and multi-year measurements, for ensuring the longevity of our results; and uses large, comprehensive studies of several real clouds, for reducing the threats to validity related to the experimental environment. This presentation can also provide useful insights for fields related to benchmarking, for example experimental evaluation conducted in other large-scale distributed systems. Last but not least, we present a road-map toward cloud benchmarking and the way we plan to progress on it with other members of the RG Cloud WG of the Standard Performance Evaluation Corporation (SPEC) http://research.spec.org/working-groups/rg-cloud-working-group.html). Note: talk based on the homonym article [1].\",\n \"Title: Integration of an event-based simulation framework into a scientific workflow execution environment for Grids and clouds\\nYear: 2011\\n\\nAbstract: The utilisation of Grid and Cloud-based computing environments for solving scientific problems has become an increasingly used practice in the last decade. To ease the use of these global distributed resources, sophisticated middleware systems have been developed, enabling the transparent execution of applications by hiding low-level technology details from the user. The ASKALON environment is such a system, which supports the development and execution of distributed applications such as scientific workflows or parameter studies in Grid and Cloud computing environments. On the other hand, simulation is a widely accepted approach to analyse and further optimise the behaviour of software systems. Beside the advantage of enabling repeatable deterministic evaluations, simulations are able to circumvent the difficulties in setting up and operating multi-institutional Grid systems, thus providing a lightweight simulated distributed environment on a single machine. In this paper, we present the integration of the GroudSim Grid and Cloud event-based simulator into the ASKALON environment. This enables system, application developers, and users to perform simulations using their accustomed environment, thereby benefiting from the combination of an established real-world platform and the advantages of a simulation.\",\n \"Title: Bringing Scientific Workflows to Amazon SWF\\nYear: 2013\\n\\nAbstract: In response to the ever-increasing needs of scientific applications for resources, Cloud computing emerged as an alternative on-demand and cost-effective resource provisioning approach. In this context, Cloud providers have recognised the importance of workflow applications to science and provide their own native solutions, such as the Amazon Simple Workflow Service (SWF). Nevertheless, an important downside of SWF is its incompatibility with existing workflow systems, and lack of means for reusing scientific legacy code. Similarly, existing workflow middlewares and applications require non-trivial extensions to take advantage of Cloud resources. We present in this paper a software engineering solution that allows the scientific workflow community access the Amazon Cloud through one single front-end converter, and propose a legacy wrapper service for executing legacy code using SWF. Empirical results using a real-world scientific workflow demonstrate that our automatically generated SWF application performs almost as fast as a native manually-optimised version, and outperforms other workflow middleware systems using the Amazon Cloud.\",\n \"Title: Comparison of workflow scheduling strategies on the grid\\nYear: 2005\\n\\nAbstract: Scheduling is a key concern for the execution of performance-driven Grid applications. In this paper we comparatively examine different existing approaches for scheduling of scientific workflow applications in a Grid environment. We evaluate three algorithms namely genetic, HEFT, and simple “myopic” and compare incremental workflow partitioning against the full-graph scheduling strategy. We demonstrate experiments using real-world scientific applications covering both balanced (symmetric) and unbalanced (asymmetric) workflows. Our results demonstrate that full-graph scheduling with the HEFT algorithm performs best compared to the other strategies examined in this paper.\",\n \"Title: The impact of virtualization on the performance of Massively Multiplayer Online Games\\nYear: 2009\\n\\nAbstract: Today's highly successful Massively Multiplayer Online Games (MMOGs) have millions of registered users and hundreds of thousands of active concurrent users. As a result of the highly dynamic MMOG usage patterns, the MMOG operators pre-provision and then maintain throughout the lifetime of the game tens of thousands of compute resources in data centers located across the world. Until recently, the difficulty of porting the MMOG software services to different platforms made it impractical to dynamically provision resources external to the MMOG operators' data centers. However, virtualization is a new technology that promises to alleviate this problem by providing a uniform computing platform with minimal overhead. To investigate the potential of this new technology, in this paper we propose a new hybrid resource provisioning model that uses a smaller and less expensive set of self-owned data centers, complemented by virtualized cloud computing resources during peak hours. Using real traces from RuneScape, one of the most successful contemporary MMOGs, we evaluate with simulations the effectiveness of the on-demand cloud resource provisioning strategy for MMOGs. We assess the impact of provisioning of virtualized cloud resources, analyze the components of virtualization overhead, and compare provisioning of virtualized resources with direct provisioning of data center resources.\",\n \"Title: Workflow Scheduling in Amazon EC2.\\nYear: 2013\\n\\nAbstract: Workflow scheduling has been traditionally targeted to map the execution of set of tasks onto a set of resources for makespan minimization. With the increasing popularity of Cloud computing systems, the financial cost entailed for executing these tasks plays also an important role. Existing works have however combined both, makespan and cost, on a single function and no analysis of the tradeoff between both criteria has been produced. In addition, no work in the context a real commercial cloud system exists. This paper includes a comparison of two real multi-objective workflow scheduling, MOHEFT and SPEA2*, in the context of Amazon EC2. The carried experiments show that MOHEFT outperforms SPEA2*, and that the analysis of the tradeoff solutions can help in selecting good scheduling solutions.\",\n \"Title: Real world workflow applications in the askalon grid environment\\nYear: 2005\\n\\nAbstract: The workflow paradigm is widely considered as an important class of truly distributed Grid applications which poses many challenges for a Grid computing environment. Still to this time, rather few real-world applications have been successfully ported as Grid-enabled workflows. We present the Askalon programming and computing environment for the Grid which comprises a high-level abstract workflow language and a sophisticated service-oriented runtime environment including meta-scheduling, performance monitoring and analysis, and resource brokerage services.We demonstrate the development of a real-world river modelling distributed workflow system in the Askalon environment that harnesses the computational power of multiple Grid sites to optimise the overall execution time.\",\n \"Title: Dynamic scheduling of scientific workflow applications on the grid: a case study\\nYear: 2005\\n\\nAbstract: The existing Grid workflow scheduling projects do not handle recursive loops which are characteristic to many scientific problems. We propose a hybrid approach for scheduling Directed Graph (DG)-based workflows in a Grid environment with dynamically changing computational and network resources. Our dynamic scheduling algorithm is based on the iterative invocation of classical static Directed Acyclic Graphs (DAGs) scheduling heuristics generated using well-defined cycle elimination and task migration techniques. We approach the static scheduling problem as an application of a modular optimisation tool using genetic algorithms. We report successful implementation and experimental results on a pilot real-world material science workflow application.\",\n \"Title: A Hybrid Intelligent Method for Performance Modeling and Prediction of Workflow Activities in Grids\\nYear: 2009\\n\\nAbstract: Grid schedulers require individual activity performance predictions to map workflow activities on different Grid sites. The effectiveness of the scheduling systems is hampered by inaccurate predictions due to the inability of existing predictors to effectively model the dynamic and heterogeneous nature of Grid resources, or the wide range of problem sizes and runtime arguments. To address this deficiency, we propose a hybrid Bayesian-neural network approach to dynamically model and predict the execution time of activities in real workflow applications. Bayesian network is used for a high-level representation of activities performance probability distribution against different factors affecting the performance. The important attributes are dynamically selected by the Bayesian network and fed into a radial basis function neural network to make further predictions. Our approach is generic to any type of scientific applications, and flexible to import expert knowledge to further improve accuracies. Experimental results for activities from three realworld workflow applications are presented to show effectiveness of our approach.\",\n \"Title: Impact of variable priced cloud resources on scientific workflow scheduling\\nYear: 2012\\n\\nAbstract: We analyze the problem of provisioning Cloud instances to large scientific workflows that do not benefit from sufficient Grid resources as required by their computational requirements. We propose an extension to the dynamic critical path scheduling algorithm to deal with the general resource leasing model encountered in today's commercial Clouds. We analyze the availability of the cheaper and unreliable Spot instances and study their potential to complement the unavailability of Grid resources for large workflow executions. Experimental results demonstrate that Spot instances represent a 60% cheaper but equally reliable alternative to Standard instances provided that a correct user bet is made.\",\n \"Title: Enhancing Grids for Massively Multiplayer Online Computer Games\\nYear: 2008\\n\\nAbstract: Massively multiplayer online games (MMOG) are an innovative and challenging class of applications for Grid computing that require large amounts of computational resources for providing a responsive and scalable gameplay for concurrently participating players connected via Internet. We present our Real-Time Framework (RTF)--- a Grid-based middleware for scaling game sessions through a variety of parallelization and distribution techniques. RTF is described within a novel multi-layer service-oriented architecture that comprises three advanced services --- monitoring, capacity planning, and runtime steering --- that use the potential of Grid computing to provide pervasive access to a potentially unbounded number of resources. We report experimental results on the quality of our capacity planning and scalability of the RTF distribution mechanism.\",\n \"Title: Neural Network-Based Load Prediction for Highly Dynamic Distributed Online Games\\nYear: 2008\\n\\nAbstract: We propose a neural network-based prediction method for the future entity layout in massively multiplayer online games. Our service has the potential to timely foresee critical hot-spots in fast-paced First Person Shooter action games that saturate the game servers which no longer respond to user actions at the required rate. Using our service, proactive load balancing (and redistribution) actions can be triggered.We show results based on a realistic simulation environment that demonstrate the advantages of our method compared to other conventional ones, especially due to its ability to adapt to different load patterns.\",\n \"Title: Scheduling javasymphony applications on many-core parallel computers\\nYear: 2011\\n\\nAbstract: JavaSymphony is a Java-based programming and execution environment for programming and scheduling the performance oriented applications on multi-core parallel computers. In this paper, we present a multi-core aware scheduling extension to JavaSymphony capable of mapping parallel applications on large multi-core machines and heterogeneous clusters. JavaSymphony scheduler considers several multicore specific performance parameters and application types, and uses these parameters to optimise the mapping of JavaSymphony objects and tasks. We evaluate the performance of JavaSymphony scheduler using several real scientific applications and benchmarks on a multi-core shared memory machine and a heterogeneous cluster.\",\n \"Title: A New Fault Tolerance Heuristic for Scientific Workflows in Highly Distributed Environments Based on Resubmission Impact\\nYear: 2009\\n\\nAbstract: Even though highly distributed environments such as Clouds and Grids are increasingly used for e-Science high performance applications, they still cannot deliver the robustness and reliability needed for widespread acceptance as ubiquitous scientific tools. To overcome this problem, existing systems resort to fault tolerance mechanisms such as task replication and task resubmission. In this paper we propose a new heuristic called Resubmission Impact to enhance the fault tolerance support for scientific workflows in highly distributed systems. In contrast to related approaches, our method can be used effectively on systems even in the absence of historic failure trace data. Simulated experiments of three real scientific workflows in the Austrian Grid environment show that our algorithm drastically reduces the resource waste compared to conservative task replication and resubmission techniques, while having a comparable execution performance and only a slight decrease in the success probability.\",\n \"Title: Online Analysis and Runtime Steering of Dynamic Workflows in the ASKALON Grid Environment\\nYear: 2007\\n\\nAbstract: We present a new distributed performance analysis service of the ASKALON integrated Grid environment for computing runtime overheads of dynamic workflows in realtime based on event correlation techniques. We illustrate a formal method to express precise overhead correlation rules, including several performance contracts as quality of service parameters based on fuzzy logic to be enforced in dynamic environments though rescheduling, various runtime optimisations, and steering techniques. We demonstrate experimental results for two real applications from material chemistry and graphics rendering domains.\",\n \"Title: Applying double auctions for scheduling of workflows on the Grid\\nYear: 2008\\n\\nAbstract: Grid economy models have long been considered as a promising alternative for the classical Grid resource management, due to their dynamic and decentralized nature, and because the financial valuation of resources and services is inherent in any such model. In particular, auction models are widely used in the existing Grid research, as they are easy to implement and are shown to successfully manage resource allocation on the Grid market. The focus on the current work is on workflow scheduling in the Grid resource allocation model based on Continuous Double Auctions (CDA). We analyze different scheduling strategies that can be applied by the user to execute workflows in such an environment, and try to identify the general behavioral patterns that can lead to a fast and cheap workflow execution. In the experimental study, we show that under certain circumstances some benefit can be gained by applying an \\\"aggressive\\\" scheduling strategy.\",\n \"Title: Overhead Analysis of Grid Workflow Applications\\nYear: 2006\\n\\nAbstract: In this paper we propose a systematic approach to performance analysis of workflow applications on the Grid. We introduce an ideal model for the workflow execution time and explain the difference to the real measured times based on a hierarchy of performance overheads for Grid computing. We describe how to systematically measure and compute the overheads from individual activities to entire workflow applications. We adjusted well-known parallel processing metrics to the scope of Grid computing, comprising speedup and efficiency. We have implemented and largely automatised our analysis approach in the context of the ASKALON Grid application development and computing environment. We present experimental results that show detailed overhead analysis of two real-world workflow applications executed in a national Grid environment.\",\n \"Title: IWIR: a language enabling portability across grid workflow systems.\\nYear: 2011\\n\\nAbstract: Today there are many different scientific Grid workflow management systems using a wide array of custom workflow languages. Some of them are geared towards a data-based view, some are geared towards a control-flow based view and others try to be as generic, and therefore often complex, as possible. All of these languages and custom workflow management system front-ends fulfill special needs and workflow creation paradigms for their respective user communities. The problem is that once a workflow application has been created in one of these systems, it becomes very hard to share the workflow with users working with different systems. Portability and interoperability between current systems barely exists. In this work, we present a common workflow language for use as an intermediate exchange representation by multiple workflow systems. It comprises atomic tasks, compound tasks including conditionals, sequential and parallel loops as well as an expressive set of data types and data flow constructs.\",\n \"Title: ZENTURIO: An Experiment Management System for Cluster and Grid Computing\\nYear: 2002\\n\\nAbstract: The need to conduct and manage large sets of experiments for scientific applications dramatically increased over the last decade. However, there is still very little tool support for this complex and tedious process. In this paper we introduce the ZENTURIO experiment management system for parameter studies, performance analysis, and software testing for cluster and Grid architectures. ZENTURIO uses the ZEN directive-based language to specify arbitrary complex program executions. ZENTURIO is designed as a collection of Grid services that comprise: (1) a registry service which supports registering and locating Grid services; (2) an experiment generator that parses files with ZEN directives and instruments applications for performance analysis and parameter studies; (3) an experiment executor that compiles and controls the execution of experiments on the target machine. A graphical user portal allows the user to control and monitor the experiments and to automatically visualise performance and output data across multiple experiments. ZENTURIO has been implemented based on Java/Jini distributed technology. It supports experiment management on cluster architectures via PBS and on Grid infrastructures through GRAM. We report results of using ZENTURIO for performance analysis of an ocean simulation application and a parameter study of a computational finance code.\",\n \"Title: Fine-Grain Interoperability of Scientific Workflows in Distributed Computing Infrastructures\\nYear: 2013\\n\\nAbstract: Today there exist a wide variety of scientific workflow management systems, each designed to fulfill the needs of a certain scientific community. Unfortunately, once a workflow application has been designed in one particular system it becomes very hard to share it with users working with different systems. Portability of workflows and interoperability between current systems barely exists. In this work, we present the fine-grained interoperability solution proposed in the SHIWA European project that brings together four representative European workflow systems: ASKALON, MOTEUR, WS-PGRADE, and Triana. The proposed interoperability is realised at two levels of abstraction: abstract and concrete. At the abstract level, we propose a generic Interoperable Workflow Intermediate Representation (IWIR) that can be used as a common bridge for translating workflows between different languages independent of the underlying distributed computing infrastructure. At the concrete level, we propose a bundling technique that aggregates the abstract IWIR representation and concrete task representations to enable workflow instantiation, execution and scheduling. We illustrate case studies using two real-workflow applications designed in a native environment and then translated and executed by a foreign workflow system in a foreign distributed computing infrastructure.\",\n \"Title: ZEN: a directive based experiment specification language for performance and parameter studies of parallel scientific applications\\nYear: 2005\\n\\nAbstract: Currently there exists little support to specify and control the execution of a large number of experiments required for performance oriented development of scientific applications. We have developed a new directive based language called ZEN for the specification of wide sets of synthetic experiments in the context of large scale parameter and performance studies of parallel applications. We report experimental results for the performance analysis of an ocean simulation application and for the parameter study of a computational finance code.\",\n \"Title: Performance and cost optimization for multiple large-scale grid workflow applications\\nYear: 2007\\n\\nAbstract: Scheduling large-scale applications on the Grid is a fundamental challenge and is critical to application performance and cost. Large-scale applications typically contain a large number of homogeneous and concurrent activities which are main bottlenecks, but open great potentials for optimization. This paper presents a new formulation of the well-known NP-complete problems and two novel algorithms that addresses the problems. The optimization problems are formulated as sequential cooperative games among workflow managers. Experimental results indicate that we have successfully devised and implemented one group of effective, efficient, and feasible approaches. They can produce soultuins of significantly better performance and cost than traditional algorithms. Our algorithms have considerably low time complexity and can assign 1,000,000 activities to 10,000 processors within 0.4 second on one Opteron processor. Moreover, the solutions can be practically performed by workflow managers, and the violation of QoS can be easily detected, which are critical to fault tolerance.\",\n \"Title: The first framework for interoperable run-time software tools\\nYear: 2001\\n\\nAbstract: A variety of different program correctness tools are now available to help the software developer write, test and maintain faster, correctly conforming code. The majority of these software tools are independent in nature and can not be used in conjunction with other similar tools. Such tools are therefore not interoperable and, typically, the program must be recompiled with a different set of tool-dependent flags in order to move from one tool to another.The FIRST project uses CORBA to build an object-oriented framework for real-time software performance and correctness tools. These tools use framework-provided high-level tool services to generate instrumentation which is dynamically inserted into the already running target process.This paper describes in detail the FIRST tool development framework, classifies different types of tool interoperability and discusses various scenarios in which FIRST profiling and debugging tools can be combined to achieve synergies through interoperability.\",\n \"Title: ZEN: A Directive-Based Language for Automatic Experiment Management of Distributed and Parallel Programs\\nYear: 2002\\n\\nAbstract: So far there exists very little support to specify and to control execution of a large number of experiments on distributed and parallel architectures. This paper describes ZEN, a directive-based language for the specification of arbitrarily complex program executions by varying problem, system, or machine parameters for parallel and distributed applications. ZEN introduces directives to substitute strings and to insert assignment statements inside arbitrary files, such as program, input, script, or makefiles. The programmer thus can invoke experiments for arbitrary value ranges of any problem parameter, including program variables, file names, compiler options, target machines, machine sizes, scheduling strategies, data distributions, etc. The number of experiments can be controlled through ZEN constraintdirectives. Finally, the programmer may request a large set of performance metrics to be computed for any code region of interest. The scope of ZEN directives can be restrictedto arbitrary file or code regions. We have implemented a prototype tool for automatic experiment management that is based on ZEN. We will report results for performance analysisof an ocean simulation application and for parameter study of a computational finance code.\",\n \"Title: From Web Services to OGSA: Experiences in Implementing an OGSA-based Grid Application\\nYear: 2003\\n\\nAbstract: In previous work we have presented the ZENTURIO experimentmanagement system for performance and parameterstudies of parallel and distributed applications on clusterand Grid architectures. In this paper we describe experiencesof an on-going work, targeting the implementationof ZENTURIO on top of the Open Grid Services Architecture(OGSA). We analyse the opportunities offered by aWeb services toolkit to develop Grid services as required byOGSA and compare them with the solutions offered by theOpen Grid Services Infrastructure (OGSI) specification. Issuesregarding proxy management, service lifecycle, UDDIservice repository, firewall management, Factory and Registryservices, service throughput, and security are comparativelyanalysed in both implementations.\",\n \"Title: Autonomic operation of massively multiplayer online games in clouds\\nYear: 2013\\n\\nAbstract: To support the variable load of Massively Multiplayer On-line Games (MMOGs) with millions of registered users and thousands of active concurrent players, game operators over-provision a large static infrastructure capable of sustaining the peak load with guaranteed Quality of Service (QoS). This leads to inefficient resource utilisation, high service prices, and limited market participation accessible only to the large companies. To address this problem, we propose a new autonomic ecosystem for hosting and operating MMOGs based on cloud computing principles involving four smaller and better focused business actors whose interaction is regulated through Service Level Agreements (SLAs): resource provider, game operator, game provider, and client. In our model, game providers acquire operation SLAs from game operators to satisfy client requests and manage multiple distributed MMOG sessions. Game operators lease on-demand cloud resources based on the dynamic MMOG load and guarantee the required QoS to all clients. We evaluate through simulations based on real MMOG traces and commercial cloud SLAs different methods of ranking MMOG operation offers. We show that considering compensations for SLA faults in the offer selection can lead to over 11% gains in game providers' income, and that adequate ranking of offers can reduce operational costs by up to 60%.\",\n \"Title: Run-time Optimisation of Grid Workflow Applications\\nYear: 2006\\n\\nAbstract: The execution of workflow applications on the grid is a complex issue because of its dynamic and heterogeneous nature. While the grid provides good potential for achieving high performance, it also introduces a broad set of unpredictable overheads and possible failures. In this paper we present new methods for scalable and fault tolerant coordination of workflows in dynamic grid environments, including partitioning, static and dynamic optimisation, as well as virtual single execution environment, incorporated into the ASKALON distributed workflow enactment engine. We demonstrate the effectiveness of our methods on a material science workflow application executed in a real-world grid environment\",\n \"Title: Dynamic load management for MMOGs in distributed environments\\nYear: 2010\\n\\nAbstract: To support thousands of concurrent players in virtual worlds simulated by contemporary Massively Multiplayer Online Games, most implementations employ static game world partitioning for distributing the load among multiple game server instances. Further, the resources that manage the resulting subregions are statically allocated, independent of the actual game load. As a result, due to the high variability of the user demand, this approach leads to a low resource utilization causing much higher provisioning costs than necessary. In addition, the number of players supported by a region is limited by the maximum load that can be handled by a single server instance. We propose in this paper a novel game load management technique divided in two (global and a local) layers, capable of dynamically adjusting the amount of allocated resources to the present user demand. The global level assigns the responsibility of serving particular game regions to data centers using a peer-to-peer infrastructure, while the local level within individual facilities maintains the necessary server instances for the assigned obligations. We device two generic heuristics based on the well-known bin-packing problem to achieve the ultimate goal of maximizing the resource utilization on both levels while maintaining user-level Quality of Service (QoS). We evaluate the performance of our proposed solution using simulation-based experiments, which demonstrate a potential cost reduction in maintaining MMOG sessions by up to 60% while maintaining QoS in 99% of the cases.\",\n \"Title: Multi-Objective Game Theoretic Schedulingof Bag-of-Tasks Workflows on Hybrid Clouds\\nYear: 2014\\n\\nAbstract: Scheduling multiple large-scale parallel workflow applications on heterogeneous computing systems like hybrid clouds is a fundamental NP-complete problem that is critical to meeting various types of QoS (Quality of Service) requirements. This paper addresses the scheduling problem of large-scale applications inspired from real-world, characterized by a huge number of homogeneous and concurrent bags-of-tasks that are the main sources of bottlenecks but open great potential for optimization. The scheduling problem is formulated as a new sequential cooperative game and propose a communication and storage-aware multi-objective algorithm that optimizes two user objectives (execution time and economic cost) while fulfilling two constraints (network bandwidth and storage requirements). We present comprehensive experiments using both simulation and real-world applications that demonstrate the efficiency and effectiveness of our approach in terms of algorithm complexity, makespan, cost, system-level efficiency, fairness, and other aspects compared with other related algorithms.\",\n \"Title: SLA-based operations of massively multiplayer online games in clouds\\nYear: 2014\\n\\nAbstract: Successful massively multiplayer online games (MMOGs) have today millions of registered users and hundreds of thousands of active concurrent players. To be able to guarantee quality of service (QoS) to a highly variable number of concurrent users, game operators statically over-provision a large infrastructure capable of sustaining the game peak load, even though a large portion of the resources is unused most of the time. To address this problem, we introduce in this work a new MMOG ecosystem for hosting and provisioning of MMOGs which effectively splits the traditional monolithic MMOG companies into three main service providers: game providers, game operators, and resource providers. Their interaction is regulated through comprehensive service level agreements (SLA) that establish the price, terms of operation, and compensation for service violations. In our model, game operators efficiently provision resources for MMOGs from multiple cloud providers, based on dynamic load forecasts, and ensure proper game operation that maintains the required QoS to all clients under varying resource availability. Game providers manage multiple distributed MMOGs for which they lease services under strict operational SLAs from game operators to satisfy all client requests. These three self-standing, smaller, more agile service providers enable access to the MMOG market for the small and medium enterprises, and to the current commercial cloud providers. We evaluate, through simulations based on real-life MMOG traces and commercial cloud SLAs, the impact of resource availability on the QoS offered to the MMOG clients. We find that our model can mitigate the negative effects of resource failures within four minutes and that MMOG server consolidation can accentuate the negative effects of failures in a resource-scarce environment. We further investigate different methods of ranking MMOG operational offers with either single or multiple (competing) MMOG providers. Our results show that compensations for SLA faults in the offer selection process can lead up to 11---16 % gain in the game providers' income. We also demonstrate that adequate ranking of offers can lead to MMOG operational cost reductions from 20 up to 60 %.\",\n \"Title: Multi-objective energy-efficient workflow scheduling using list-based heuristics.\\nYear: 2014\\n\\nAbstract: Workflow applications are a popular paradigm used by scientists for modelling applications to be run on heterogeneous high-performance parallel and distributed computing systems. Today, the increase in the number and heterogeneity of multi-core parallel systems facilitates the access to high-performance computing to almost every scientist, yet entailing additional challenges to be addressed. One of the critical problems today is the power required for operating these systems for both environmental and financial reasons. To decrease the energy consumption in heterogeneous systems, different methods such as energy-efficient scheduling are receiving increasing attention. Current schedulers are, however, based on simplistic energy models not matching the reality, use techniques like DVFS not available on all types of systems, or do not approach the problem as a multi-objective optimisation considering both performance and energy as simultaneous objectives. In this paper, we present a new Pareto-based multi-objective workflow scheduling algorithm as an extension to an existing state-of-the-art heuristic capable of computing a set of tradeoff optimal solutions in terms of makespan and energy efficiency. Our approach is based on empirical models which capture the real behaviour of energy consumption in heterogeneous parallel systems. We compare our new approach with a classical mono-objective scheduling heuristic and state-of-the-art multi-objective optimisation algorithm and demonstrate that it computes better or similar results in different scenarios. We analyse the different tradeoff solutions computed by our algorithm under different experimental configurations and we observe that in some cases it finds solutions which reduce the energy consumption by up to 34.5% with a slight increase of 2% in the makespan.\",\n \"Title: Workflow Scheduling on Federated Clouds.\\nYear: 2014\\n\\nAbstract: Federated Clouds, or the orchestration of multiple Cloud services for fulfilling applications' requirements, is receiving increasing attention. Despite their many advantages, federated Clouds also present some downsides since different services may reside in different geographically located areas. This paper focuses on evaluating the advantages and disadvantages, from the point of view of performance and financial costs, of using a federation of Clouds for executing scientific workflows. It evaluates a wide range of different workflow types with different requirements in terms of computation and communication (produced and consumed data), and discusses which kind of workflow applications can benefit from a Cloud federation and how.\",\n \"Title: Comparison of Static and Dynamic Resource Allocations for Massively Multiplayer Online Games on Unreliable Resources.\\nYear: 2014\\n\\nAbstract: We investigate the use of a new Massively Multiplayer Online Gaming (MMOG) ecosystem consisting of end-users, game providers, game operators, and Cloud resource providers, for autonomous, self-adaptive hosting and operation of MMOGs on unreliable resources. For this purpose, we developed an MMOG simulator compliant with our MMOG ecosystem in which we inject traces collected from a real-world MMOG and resource characteristics from 16 Cloud providers. Using our simulator, we study the impact on the involved actors by considering different resource availability levels, and highlight the advantages of dynamic resource allocation over the static overprovisioning with respect to two types of metrics: QoS offered to the clients and financial profit of game providers and operators.\",\n \"Title: Cooperative Scheduling of Bag-of-Tasks Workflows on Hybrid Clouds\\nYear: 2014\\n\\nAbstract: We address the problem of scheduling a class of large-scale applications inspired from real-world on hybrid Clouds, characterized by a large number of homogeneous and concurrent tasks that are the main sources of bottlenecks but open great potential for optimization. We formulate the scheduling problem as a new sequential cooperative game and propose a communication- and storage-aware multi-objective algorithm that optimizes two user objectives (execution time and economic cost) while fulfilling two constraints (network bandwidth and storage requirements). We present comprehensive experiments using both simulation and real-world applications that demonstrate the efficiency and effectiveness of our approach in terms of algorithm complexity, make span, cost, system-level efficiency, fairness, and other aspects compared with other related algorithms.\",\n \"Title: Massivizing online games using cloud computing: A vision\\nYear: 2014\\n\\nAbstract: Online gaming, a large market with hundreds of millions of active players, is still struggling to scale without risky investments in infrastructure. In this work, we propose a cloudbased platform to massivize online gaming-the challenges and opportunities of scaling on-demand, while paying only for what is used. We discuss the major aspects of cloud-based gaming, virtual-world management, game-data processing, and game-content generation.\",\n \"Title: Optimizations to enhance sustainability of MPI applications\\nYear: 2014\\n\\nAbstract: Ultrascale computing systems are likely to reach speeds of two or three orders of magnitude greater than today's computing systems. However, to achieve this level of performance, we need to design and implement more sustainable solutions for ultra-scale computing systems, at both the hardware and software levels, while understanding sustainability in a holistic manner in order to address challenges in economy-of-scale, agile elastic scalability, heterogeneity, programmability, fault resilience, energy efficiency, and storage. Some solutions could be integrated into MPI, but others should be devised as higher level concepts, less general, but adapted to applicative domains, possibly as programming patterns or libraries. In this paper, we layout some proposals to extend MPI to cover major relevant domains in a move towards sustainability, including: MPI programming optimizations and programming models, resilience, data management, and their usage for applications.\",\n \"Title: Meteorological applications utilizing Grid and Cloud computing\\nYear: 2014\\n\\nAbstract: Three practical meteorological applications with different characteristics are used to highlight the usability of a computer science workflow middleware called ASKALON by allowing easy access to distributed computing for atmospheric scientists. Utilizing Cloud and Grid computing, this paper shows use case scenarios fitting a wide range of applications from operational to research studies with real world examples from meteorological research. The paper concludes that distributed computing is easily usable for meteorological problems using ASKALON. This powerful tool allows simple and cost effective access to computing capacity from Grid and Cloud environments.\",\n \"Title: Fostering energy-awareness in scientific cloud users\\nYear: 2014\\n\\nAbstract: Academic cloud infrastructures are constructed and maintained so they minimally constrain their users. Since they are free and do not limit usage patterns, academics developed such behavior that jeopardizes fair and flexible resource provisioning. For efficiency, related work either explicitly limits user access to resources, or introduces automatic rationing techniques. Surprisingly, the root cause (i.e., the user behavior) is disregarded by these approaches. This paper compares academic cloud user behavior to its commercial equivalent. We deduce, that academics should behave like commercial cloud users to relieve resource provisioning. To encourage this behavior, we propose an architectural extension to academic infrastructure clouds. We evaluate our extension via a simulation using real life academic resource request traces. We show a potential resource usage reduction while maintaining the unlimited nature of academic clouds.\",\n \"Title: Fostering Energy-Awareness in Simulations behind Scientific Workflow Management Systems\\nYear: 2014\\n\\nAbstract: Scientific workflow management systems face a new challenge in the era of cloud computing. The past availability of rich information regarding the state of the used infrastructures is gone. Thus, organising virtual infrastructures so that they not only support the workflow being executed, but also optimise for several service level objectives (e.g., Maximum energy consumption limit, cost, reliability, availability) become dependent on good infrastructure modelling and prediction techniques. While simulators have been successfully used in the past to aid research on such workflow management systems, the currently available cloud related simulation toolkits suffer form several issues (e.g., Scalability, narrow scope) that hinder their applicability. To address this need, this paper introduces techniques for unifying two existing simulation toolkits by first analysing the problems with the current simulators, and then by illustrating the problems faced by workflow systems through the example of the ASKALON environment. Finally, we show how the unification of the selected simulators improve on the the discussed problems.\",\n \"Title: Evaluating Energy Efficiency of Gigabit Ethernet and Infiniband Software Stacks in Data Centres\\nYear: 2014\\n\\nAbstract: Reducing energy consumption has become a key issue for data centres, not only because of economical benefits but also for environmental and marketing reasons. Many approaches tackle this problem from the point of view of different hardware components, such as CPUs, storage and network interface cards (NIC). To this date, few works focused on the energy consumption of network transfers at the software level comprising their complete stacks with different energy characteristics, and the way the NIC selection impacts the energy consumption of applications. Since data centres often install multiple NICs on each node, investigating and comparing them at the software level has high potential to enhance the energy efficiency of applications on Cloud infrastructures. We present a comparative analysis of the energy consumption of the software stack of two of today's most used NICs in data centres, Ethernet and Infiniband. We carefully design for this purpose a set of benchmark experiments to assess the impact of different traffic patterns and interface settings on energy consumption. Using our benchmark results, we derive an energy consumption model for network transfers and evaluate its accuracy for a virtual machine migration scenario. Finally, we propose guidelines for NIC selection from an energy efficiency perspective for different application classes.\",\n \"Title: Bi-Criteria Scheduling of Scientific Grid Workflows\\nYear: 2010\\n\\nAbstract: The drift towards new challenges in Grid computing including scientific workflow management implies the need for new, robust, multicriteria scheduling algorithms that can be applied by the user in an intuitive way. Currently existing bi-criteria scheduling approaches for scientific workflows are usually restricted to certain criterion pairs and require the user to define his preferences either as weights assigned each criterion or as fixed constraints defined for one criterion. The first approach has the drawback that combining multiple criteria into a single objective function is not always intuitive to the end-user, while the second requires a priori knowledge about the result of the first criteria scheduling result. We propose a new bi-criteria scheduling specification method defining for the secondary criterion a sliding constraint as a function of the primary criterion. We model the problem as an extension of the multiple-choice knapsack problem and propose a general bi-criteria scheduling heuristic called dynamic constraint algorithm (DCA) based on dynamic programming. We show through simulation that in most of the experimental cases DCA outperforms two existing algorithms designed for the same problem at the expense of an increased execution time, which is still relatively low for workflows of medium size. Finally, we confirm our simulation results for a real-world hydrological application executed in the Austrian Grid environment.\",\n \"Title: Evaluating High-Performance Computing on Google App Engine\\nYear: 2012\\n\\nAbstract: An experimental approach employs the Google App Engine (GAE) for high-performance parallel computing. A generic master-slave framework enables fast prototyping and integration of parallel algorithms that are transparently scheduled and executed on the Google cloud infrastructure. Compared to Amazon Elastic Compute Cloud (EC2), GAE offers lower resource-provisioning overhead and is cheaper for jobs shorter than one hour. Experiments demonstrated good scalability of a Monte Carlo simulation algorithm. Although this approach produced important speedup, two main obstacles limited its performance: middleware overhead and resource quotas.\",\n \"Title: Extending Grids with cloud resource management for scientific computing\\nYear: 2009\\n\\nAbstract: From its start using supercomputers, scientific computing constantly evolved to the next levels such as cluster computing, meta-computing, or computational Grids. Today, Cloud Computing is emerging as the paradigm for the next generation of large-scale scientific computing, eliminating the need of hosting expensive computing hardware. Scientists still have their Grid environments in place and can benefit from extending them by leased Cloud resources whenever needed. This paradigm shift opens new problems that need to be analyzed, such as integration of this new resource class into existing environments, applications on the resources and security. The virtualization overheads for deployment and starting of a virtual machine image are new factors which will need to be considered when choosing scheduling mechanisms. In this paper we investigate the usability of compute Clouds to extend a Grid workflow middleware and show on a real implementation that this can speed up executions of scientific workflows.\",\n \"Title: Performance analysis of Grid applications in the ASKALON environment\\nYear: 2009\\n\\nAbstract: In this paper we describe the approach taken by the ASKALON Grid application development and computing environment for scalability and overhead analysis of scientific applications in the Austrian Grid. We present a technique imported from parallel processing for overhead and scalability analysis of Grid applications based on speedup and efficiency metrics required primarily for tuning the middleware services and improving the executions. We present experimental results that validate our techniques for five real-world applications in the Austrian Grid environment.\",\n \"Title: The JavaSymphony Extensions for Parallel GPU Computing\\nYear: 2012\\n\\nAbstract: Today, the use of GPUs as coprocessors to accelerate high-performance scientific applications is becoming an important practice. Still, some of the high-level programming languages such as Java require extensions or new interfaces for utilising the huge parallelism of these new devices. In this paper, we propose extensions to an existing Java-based programming and parallel computing environment called Java Symphony to enable Java applications use accelerating devices such as GPUs with little API programmability change. With Java Symphony, a parallel Java application can be uniformly programmed and executed on heterogeneous platforms consisting of conventional parallel computers enhanced with data-parallel coprocessors such as GPUs. We report results on using Java Symphony for programming and improving the performance of six real applications and benchmarks in a heterogeneous environment consisting of a combination of different multi-core CPU and GPU devices.\",\n \"Title: Cost-efficient hosting and load balancing of Massively Multiplayer Online Games\\nYear: 2010\\n\\nAbstract: Massively Multiplayer Online Games (MMOG) are a class of computationally-intensive client-server applications with severe real-time Quality of Service (QoS) requirements, such as the number of updates per second each client needs to receive from the servers for a fluent and realistic experience. To guarantee the QoS requirements, game providers over-provision to game sessions a large amount of their resources, which is very inefficient and prohibits any but the largest providers from joining the market. In this paper, we present a new approach for cost-efficient hosting of MMOG sessions on Cloud resources, provisioned on-demand in the correct amount based on the current number of connected players. Simulation results on real MMOG traces demonstrate that compute Clouds can reduce the hosting costs by a factor between two and five. The resource allocation is driven by a load balancing algorithm that appropriately distributes the load such that the QoS requirements are fulfilled at all times. Experimental results on a fast-paced game demonstrator executed on resources owned by a specialised hosting company demonstrate that our algorithm is able to adjust the number of game servers and load distribution to the highly dynamic client load, while maintaining the QoS in 99.34% of the monitored events.\",\n \"Title: Massively Multiplayer Online Games on unreliable resources\\nYear: 2012\\n\\nAbstract: We have previously introduced a cloud-based model for Massively Multiplayer Online Games (MMOGs) operation, which allows small and medium enterprises to join the MMOG market through near-zero initial infrastructure investment. In this work we propose an architectural extension to this MMOG operational model designed to increase and guarantee the MMOG session availability through fault tolerance mechanisms.\",\n \"Title: PIASA: A power and interference aware resource management strategy for heterogeneous workloads in cloud data centers\\nYear: 2015\\n\\nAbstract: •A scheduling strategy for heterogeneous workloads with specific SLAs requirements.•Deviations detection of applications performance, with noisy data samples.•Multiple types of SLAs, according to mixed/heterogeneous applications.•Support for mitigation of performance degradation caused by resource contention.•Auto-scaling of resources in order to satisfy SLAs and cope with peak time demands.\",\n \"Title: Pareto tradeoff scheduling of workflows on federated commercial Clouds\\nYear: 2015\\n\\nAbstract: •We extend the HEFT algorithm for multi-objective workflow optimisation.•We apply the algorithm for performance and cost optimisation.•We adapt the new algorithm to fit the scope of public commercial Clouds.•We evaluate Pareto fronts algorithm using the hypervolume metric.•We demonstrate better results compared with two related algorithms.\",\n \"Title: On Processing Extreme Data\\nYear: 2015\\n\\nAbstract: Extreme Data is an incarnation of Big Data concept distinguished by the massive amounts of data that must be queried, communicated and analyzed in near real-time by using a very large number of memory or storage elements and exascale computing systems. Immediate examples are the scientific data produced at a rate of hundreds of gigabits-per-second that must be stored, filtered and analyzed, the millions of images per day that must be analyzed in parallel, the one billion of social data posts queried in real-time on an in-memory components database. Traditional disks or commercial storage nowadays cannot handle the extreme scale of such application data.Following the need of improvement of current concepts and technologies, we focus in this paper on the needs of data intensive applications running on systems composed of up to millions of computing elements (exascale systems). We propose in this paper a methodology to advance the state-of-the-art. The starting point is the definition of new programming paradigms, APIs, runtime tools and methodologies for expressing data-intensive tasks on exascale systems. This will pave the way for the exploitation of massive parallelism over a simplified model of the system architecture, thus promoting high performance and efficiency, offering powerful operations and mechanisms for processing extreme data sources at high speed and/or real time.\",\n \"Title: Modelling energy consumption of network transfers and virtual machine migration.\\nYear: 2016\\n\\nAbstract: Reducing energy consumption has become a key issue for data centres, not only because of economical benefits but also for environmental and marketing reasons. Therefore, assessing their energy consumption requires precise models. In the past years, many models targeting different hardware components, such as CPU, storage and network interface cards (NIC) have been proposed. However, most of them neglect energy consumption related to VM migration. Since VM migration is a network-intensive process, to accurately model its energy consumption we also need energy models for network transfers, comprising their complete software stacks with different energy characteristics.\",\n \"Title: Low-time complexity budget–deadline constrained workflow scheduling on heterogeneous resources\\nYear: 2016\\n\\nAbstract: The execution of scientific applications, under the utility computing model, is constrained to Quality of Service (QoS) parameters. Commonly, applications have time and cost constraints such that all tasks of an application need to be finished within a user-specified Deadline and Budget. Several algorithms have been proposed for multiple QoS workflow scheduling, but most of them use search-based strategies that generally have a high time complexity, making them less useful in realistic scenarios. In this paper, we present a heuristic scheduling algorithm with quadratic time complexity that considers two important constraints for QoS-based workflow scheduling, time and cost, named Deadline–Budget Constrained Scheduling (DBCS). From the deadline and budget defined by the user, the DBCS algorithm finds a feasible solution that accomplishes both constraints with a success rate similar to other state-of-the-art search-based algorithms in terms of the successful rate of feasible solutions, consuming in the worst case only approximately 4% of the time. The DBCS algorithm has a low-time complexity of O(n2.p) for n tasks and p processors.\",\n \"Title: A Workload-Aware Energy Model for Virtual Machine Migration\\nYear: 2015\\n\\nAbstract: Energy consumption has become a significant issue for data centres. Assessing their consumption requires precise and detailed models. In the latter years, many models have been proposed, but most of them either do not consider energy consumption related to virtual machine migration or do not consider the variation of the workload on (1) the virtual machines (VM) and (2) the physical machines hosting the VMs. In this paper, we show that omitting migration and workload variation from the models could lead to misleading consumption estimates. Then, we propose a new model for data centre energy consumption that takes into account the previously omitted model parameters and provides accurate energy consumption predictions for paravirtualised virtual machines running on homogeneous hosts. The new model's accuracy is evaluated with a comprehensive set of operational scenarios. With the use of these scenarios we present a comparative analysis of our model with similar state-of-the-art models for energy consumption of VM Migration, showing an improvement up to 24% in accuracy of prediction.\",\n \"Title: A workflow runtime environment for manycore parallel architectures\\nYear: 2015\\n\\nAbstract: We introduce a new Manycore Workflow Runtime Environment (MWRE) to efficiently enact traditional scientific workflows on modern manycore computing architectures. In contrast to existing engines that enact workflows acting as external services, MWRE is compiler-based and translates workflows specified in the XML-based Interoperable Workflow Intermediate Representation (IWIR) into an equivalent C++-based program. This program efficiently enacts the workflow as a stand-alone executable by means of a new callback mechanism that resolves dependencies, transfers data, and handles composite activities. Experimental results on a number of real-world workflows demonstrate that MWRE clearly outperforms existing Java-based workflow engines designed for distributed (Grid/Cloud) computing infrastructures in terms of enactment time, is generally better than an existing script-based engine for manycore architectures (Swift), and sometimes gets even close to an artificial baseline implementation of the workflows in the standard OpenMP language for shared memory systems.\",\n \"Title: Energy Prediction of OpenMP Applications Using Random Forest Modeling Approach\\nYear: 2015\\n\\nAbstract: OpenMP, with its extended parallelism features and support for radically changing HPC architectures, spurred to a surge in developing parallel applications among the HPC application developers community, leading to severe energy consumption issues. Consequently, a notion of addressing the energy consumption issue of HPC applications in an automated fashion increased among compiler developers although the underlying optimization search space could increase tremendously. This paper proposes a Random Forest Modeling (RFM) approach for predicting the energy consumption of OpenMP applications in compilers. The approach was tested using OpenMP applications, such as, NAS benchmarks, matrix multiplication, n-body simulations, and stencil applications while tuning the applications based on energy, problem size, and other performance concerns. The proposed RFM approach predicted the energy consumption of code variants with less than 0.699 Mean Square Error (MSE) and 0.998 R2 value when the testing dataset had energy variations between 0.024 joules and 150.23 joules. In addition, the influences of energy variations, number of independent variables used, and the proportion of testing dataset used during the RFM modeling process are discussed.\",\n \"Title: Multi-layered simulations at the heart of workflow enactment on clouds\\nYear: 2016\\n\\nAbstract: AbstractScientific workflow systems face new challenges when supporting Cloud computing, as the information on the state of the used infrastructures is much less detailed than before. Thus, organising virtual infrastructures in a way that not only supports the workflow execution but also optimises it for several service level objectives e.g. maximum energy consumption limit, cost, reliability, availability become reliant on good Cloud modelling and prediction information. While simulators were successfully aiding research on such workflow management systems, the currently available Cloud related simulation toolkits suffer from several issues e.g. scalability and narrow scope that hinder their applicability. To address these issues, this article introduces techniques for unifying two existing simulation toolkits by first analysing the problems with the current simulators, and then by illustrating the problems faced by workflow systems. We use for this purpose the example of the ASKALON environment, a scientific workflow composition and execution tool for cloud and grid environments. We illustrate the advantages of a workflow system with directly integrated simulation back-end and how the unification of the selected simulators does not affect the overall workflow execution simulation performance. Copyright © 2015 John Wiley & Sons, Ltd.\",\n \"Title: Large scale distributed cooperative environments on clouds and P2P.\\nYear: 2016\\n\\nAbstract: The recent advances in networking have determined an increasing use of information technology to support distributed cooperative applications. Several novel applications have emerged in this area, like computer supported collaborative work (CSCW), large scale distributed virtual worlds, collaborative recommender systems, collaborative learning systems. The definition of these applications requires to afford several challenges, like the definition of user interfaces, of coordination protocols, and of proper middle-ware and architectures supporting distributed cooperation. Collaborative applications may greatly benefits also from the support of P2P and cloud architectures. As a matter of fact, with the emergence of readily available P2P and cloud platforms, collaborative applications developers have the opportunity of deploying their collaborative application by exploiting a hybrid P2P/Cloud architectures with a support dynamically adapting . This brings possibilities to smaller developers that were reserved for the big companies until recently. The integration of mobile/Cloud platforms for Collaborative Applications is another challenge for the widespread of these applications.\",\n \"Title: An Improved Model for Live Migration in Data Centre Simulators.\\nYear: 2016\\n\\nAbstract: Due to the difficulty of employing real data centres' infrastructure for assessing the effectiveness of energy-aware algorithms, many researchers resort to use simulation tools. These tools require precise and detailed models for virtualized data centres in order to deliver accurate results. In recent years, many models have been proposed, but most of them either do not consider energy consumption related to virtual machine (VM) migration or do not investigate some of the energy impacting components (e.g. CPU, network, storage). We propose a new model for data centre energy consumption that takes into account the previously omitted components and provides more accurate energy consumption predictions compared to other state-of-the-art solutions. We evaluate our model's accuracy in a comprehensive set of scenarios implemented in the combined GroudSim/DISSECT-CF simulator. Our analysis revealed a significant improvement in accuracy (up to 42.5%) in the modelled energy consumption compared to a similar state-of-the-art simulator.\",\n \"Title: On the Next Generations of Infrastructure-as-a-Services.\\nYear: 2016\\n\\nAbstract: Following the wide adoption by industry of the cloud computing technologies, we can talk about a second generation of cloud services and products that are currently under design phase. However, it is not yet clear how the third generation of cloud products and services of the next decade will look like, especially at the delivery level of Infrastructure-as-a-Service. In order to answer at least partially to such a challenging question, we initiated a literature overview and two surveys involving the members of a cluster of European research and innovation actions. The results are interpreted in this paper and a set of topics of interest for the third generation are identified.\",\n \"Title: Multi-Objective Middleware For Distributed Vmi Repositories In Federated Cloud Environment\\nYear: 2016\\n\\nAbstract: Virtualization represents an essential technology in Cloud computing, which allows virtual machines (VM) to be executed within their own environment on top of physical hardware. The modern methods for software delivery are utilizing the concept of Vitual Machine as a efficient tool for software packaging. Typically, VMs are created using specific templates that are stored in proprietary repositories, thus leading to provider lock-in and reduced portability in the cases of simultaneous usage of multiple federated Clouds. Unfortunately, the current state-of-the-art does not provide any efficient means for streamlined management of VM images across multiple repositories, especially within federated Cloud environments. In this paper we present a novel multi-objective middleware for distributed VMI repositories in federated Cloud environment. The middleware has been designed to provide easy to use interface capable of receiving unmodified and functionally complete VM images from its users, and transparently distribute them to a specific Cloud infrastructure in a federation with respect to their size, configuration, and geographical distribution, such that they are loaded, delivered, and executed faster and with improved QoS compared to their current behaviour.\",\n \"Title: Superlinear Speedup in HPC Systems: why and when?\\nYear: 2016\\n\\nAbstract: The speedup is usually limited by two main laws in high-performance computing, that is, the Amdahlu0027s and Gustafsonu0027s laws. However, the speedup sometimes can reach far beyond the limited linear speedup, known as superlinear speedup, which means that the speedup is greater than the number of processors that are used. Although the superlinear speedup is not a new concept and many authors have already reported its existence, most of them reported it as a side effect, without explaining why and how it is happening. In this paper, we analyze several different superlinear speedup types and define a taxonomy for them. Additionally, we present several explanations and cases of superlinearity existence for different types of granular algorithms (tasks), which means that they can be divided into many sub-tasks and scattered to the processors for execution. Apart from frequent explanation that having more cache memory in parallel execution is the main reason, we summarize other different effects that cause the superlinearity, including the superlinear speedup in cloud virtual environment for both vertical and horizontal scaling.\",\n \"Title: Using Constraint-Based Reasoning for Multi-objective Optimisation of the ENTICE Environment\\nYear: 2016\\n\\nAbstract: ENTICE is a set of innovative software services currently being developed to facilitate efficient operations of distributed Virtual Machine and container images (VMI/CI) repositories. Its operation necessitates various decision making for which a solver for Multi-Objective Optimisation (MOO) problems is used. However, the solver is a bottleneck due to its computational complexity. In order to be able to reduce the search space for the solver, we have developed an ontology and corresponding Knowledge Base (KB) that underpins the operation of the ENTICE environment. The Knowledge Base is developed based on the Jena Fuseki technology. To address the problem of computational complexity, constraint based queries and different reasoning mechanisms are applied. The Knowledge Base services are then integrated with other ENTICE services including the MOO solver. It is shown that this approach significantly reduces the computational complexity for the MOO, thus it shortens the optimisation time, and makes it possible to use the MOO for both strategic (decisions that can be made up to one day in advance) and dynamic (decisions requiring response within one minute) decision making possible.\",\n \"Title: A Two-Stage Multi-Objective Optimization of Erasure Coding in Overlay Networks.\\nYear: 2017\\n\\nAbstract: In the recent years, overlay networks have emerged as a crucial platform for deployment of various distributed applications. Many of these applications rely on data redundancy techniques, such as erasure coding, to achieve higher fault tolerance. However, erasure coding applied in large scale overlay networks entails various overheads in terms of storage, latency and data rebuilding costs. These overheads are largely attributed to the selected erasure coding scheme and the encoded chunk placement in the overlay network. This paper explores a multi-objective optimization approach for identifying appropriate erasure coding schemes and encoded chunk placement in overlay networks. The uniqueness of our approach lies in the consideration of multiple erasure coding objectives such as encoding rate and redundancy factor, with overlay network performance characteristics like storage consumption, latency and system reliability. Our approach enables a variety of tradeoff solutions with respect to these objectives to be identified in the form of a Pareto front. To solve this problem, we propose a novel two stage multi-objective evolutionary algorithm, where the first stage determines the optimal set of encoding schemes, while the second stage optimizes placement of the corresponding encoded data chunks in overlay networks of varying sizes. We study the performance of our method by generating and analyzing the Pareto optimal sets of tradeoff solutions. Experimental results demonstrate that the Pareto optimal set produced by our multi-objective approach includes and even dominates the chunk placements delivered by a related state-of-the-art weighted sum method.\",\n \"Title: VM Image Repository and Distribution Models for Federated Clouds: State of the Art, Possible Directions and Open Issues.\\nYear: 2016\\n\\nAbstract: The emerging trend of Federated Cloud models enlist virtualization as a significant concept to offer a large scale distributed Infrastructure as a Service collaborative paradigm to end users. Virtualization leverage Virtual Machines (VM) instantiated from user specific templates labelled as VM Images (VMI). To this extent, the rapid provisioning of VMs with varying user requests ensuring Quality of Service (QoS) across multiple cloud providers largely depends upon the image repository architecture and distribution policies. We discuss the possible state-of-art in VMI storage repository and distribution mechanisms for efficient VM provisioning in federated clouds. In addition, we present and compare various representative systems in this realm. Furthermore, we define a design space, identify current limitations, challenges and open trends for VMI repositories and distribution techniques within federated infrastructure.\",\n \"Title: Multi-objective Optimization Framework for VMI Distribution in Federated Cloud Repositories.\\nYear: 2016\\n\\nAbstract: Cloud Federation facilitates the concept of aggregation of multiple services administered by different providers, thus opening the possibility for the customers to profit from lower cost and better performance, while allowing for the cloud providers to offer more sophisticated services. Unfortunately, current state-of-the-art does not provide any substantial means for streamlined adaptation of federated Cloud environments. One of the essential barriers that prevents Cloud federation is the inefficient management of distributed storage repositories for Virtual Machine Images (VMI). In such environments, the VMIs are currently stored by Cloud providers in proprietary centralised repositories without considering application characteristics and their runtime requirements, causing high deployment and instantiation overheads. In this paper, a novel multi-objective optimization framework for VMI placement across distributed repositories in federated Cloud environment has been proposed. Based on the communication performance requirements, VMI use patterns, and structure of images or input data, the framework provides efficient means for transparent optimization of the distribution and placement of VMIs across distributed repositories to significantly lower their provisioning time for complex resource requests and for executing the user applications.\",\n \"Title: Analysing the Performance Instability Correlation with Various Workflow and Cloud Parameters\\nYear: 2017\\n\\nAbstract: Cloud is an eco-system in which virtual machine instances are starting and terminating asynchronously on user demand or automatically when the load is rapidly increased or decreased. Although this dynamic environment allows to rent computing or storage resources cheaper rather than buying them, still it does not guarantee the stable execution during a period of time as the traditional physical environment. This is emphasised even more for workflows execution, since they consist of many data and control dependencies, which cause the makespan to be instable when a workflow is being executed in different periods of time in Cloud. In this paper we analyse several parameters of workflow and the cloud environment that are expected to impact the workflow execution instability and investigate the correlation between them. The cloud parameters include the number of instances and their type, as well as the correlation with the efficient or inefficient execution of workflow parallel sections. We conduct a series of experiments, repeating each experiment by 30 test cases in order to evaluate instability for different cloud and workflow parameters. The results show a neglectful correlation between each pair of parameters, as well as the tasks and file transfers within the workflow. Oppose to the expectations, the distribution of the makespan per experiment does not always comply with the normal distribution, which is also not correlated to a particular cloud or workflow parameter.\",\n \"Title: A Serverless Real-Time Data Analytics Platform for Edge Computing.\\nYear: 2017\\n\\nAbstract: Contemporary solutions for cloud-supported, edge-data analytics mostly apply analytics techniques in a rigid bottom-up approach, regardless of the data's origin. Typically, data are generated at the edge of the infrastructure and transmitted to the cloud, where traditional data analytics techniques are applied. Currently, developers are forced to resort to ad hoc solutions specifically tailored fo...\",\n \"Title: A Simplified Model for Simulating the Execution of a Workflow in Cloud.\\nYear: 2017\\n\\nAbstract: Although simulators provide approximate, faster and easier simulation of an application execution in Clouds, still many researchers argue that these results cannot be always generalized for complex application types, which consist of many dependencies among tasks and various scheduling possibilities, such as workflows. DynamicCloudSim, the extension of the well known CloudSim simulator, offers users the capability to simulate the Cloud heterogeneity by introducing noisiness in dozens parameters. Still, it is difficult, or sometimes even impossible to determine appropriate values for all these parameters because they are usually Cloud or application-dependent. In this paper, we propose a new model that simplifies the simulation setup for a workflow and reduces the bias between the behavior of simulated and real Cloud environments based on one parameter only, the Cloud noisiness. It represents the noise produced by the Cloud’s interference including the application’s (in our case a workflow) noisiness too. Another novelty in our model is that it does not use a normal distribution naively to create noised values, but shifts the mean value of the task execution time by the cloud noisiness and uses its deviation as a standard deviation. Besides our model reduces the complexity of DynamicCloudSim’s heterogeneity model, evaluation conducted in Amazon EC2 shows that it is also more accurate, with better trueness (closeness to the real mean values) of up to (9.2 %) and precision (closeness to the real deviation) of up to 8.39 times.\",\n \"Title: HPS-HDS: High Performance Scheduling for Heterogeneous Distributed Systems.\\nYear: 2018\\n\\nAbstract: Heterogeneous Distributed Systems (HDS) are often characterized by a variety of resources that may or may not be coupled with specific platforms or environments. Such type of systems are Cluster Computing, Grid Computing, Peer-to-Peer Computing, Cloud Computing and Ubiquitous Computing all involving elements of heterogeneity, having a large variety of tools and software to manage them. As computing and data storage needs grow exponentially in HDS, increasing the size of data centers brings important diseconomies of scale. In this context, major solutions for scalability, mobility, reliability, fault tolerance and security are required to achieve high performance. More, HDS are highly dynamic in its structure, because the user requests must be respected as an agreement rule (SLA) and ensure QoS, so new algorithm for events and tasks scheduling and new methods for resource management should be designed to increase the performance of such systems. In this special issues, the accepted papers address the advance on scheduling algorithms, energy-aware models, self-organizing resource management, data-aware service allocation, Big Data management and processing, performance analysis and optimization.\",\n \"Title: Towards Digital Twins Cloud Platform: Microservices And Computational Workflows To Rule A Smart Factory\\nYear: 2017\\n\\nAbstract: The concept of \\\"Industry 4.0\\\" considers smart factories as data-driven and knowledge enabled enterprise intelligence. In such kind of factory, manufacturing processes and final products are accompanied by virtual models - Digital Twins. To support Digital Twins concept, a simulation model for each process or system should be implemented as independent computational service. The only way to implement an orchestration of a set of independent services and provide scalability for simulation is to use a cloud computing platform as a provider of the computing infrastructure. In this paper, we describe a Digital Twin-as-a-Service (DTaaS) model for simulation and prediction of industrial processes using Digital Twins.\",\n \"Title: Simulation of a workflow execution as a real Cloud by adding noise.\\nYear: 2017\\n\\nAbstract: •A new model and methodology to add a noise to workflow execution simulations.•Noisiness of tasks within a workflow and environmental noisiness over a workflow.•Up to 59 times simulation inaccuracy reduction of makespan’s mean value.•Instance type impacts the cloud performance instability.•Sensitivity analysis determines minimum data to be collected in Learning Phase in order to achieve accuracy and precision.\",\n \"Title: A compiler transformation-based approach to scientific workflow enactment\\nYear: 2017\\n\\nAbstract: We investigate in this paper the application of compiler transformations to workflow applications using the Manycore Workflow Runtime Environment (MWRE), a compiler-based workflow environment for modern manycore computing architectures. MWRE translates scientific workflows into equivalent C++ programs and efficiently executes them using a novel callback mechanism for dependency resolution and data transfers, with explicit support for full-ahead scheduling. We evaluate four different classes of compiler transformations, analyse their advantages and possible solutions to overcome their limitations, and present experimental results for improving the performance of a combination of real-world and synthetic workflows through compiler transformations. Our experiments were able to improve the workflow enactment by a factor of two and to reduce the memory usage of the engine by up to 33%. We achieved a speedup of up to 1.7 by eliminating unnecessary activity invocations, an improved parallel throughput up to 2.8 times by transforming the workflow structure, and a better performance of the HEFT scheduling algorithm by up to 36%.\",\n \"Title: Multi-objective Service Oriented Network Provisioning in Ultra-Scale Systems.\\nYear: 2017\\n\\nAbstract: The paradigm of ultra-scale computing has been recently pushed forward by the current trends in distributed computing. This novel architecture concept is focused towards a federation of multiple geographically distributed heterogeneous systems under a single system image, thus allowing efficient deployment and management of very complex architectures applications. To enable sustainable ultra-scale computing, there are multiple major challenges, which have to be tackled, such as, improved data distribution, increased systems scalability, enhanced fault tolerance, elastic resource management, low latency communication and etc. Regrettably, the current research initiatives in the area of ultra-scale computing are in a very early stage of research and are predominantly concentrated on the management of the computational and storage resources, thus leaving the networking aspects unexplored. In this paper we introduce a promising new paradigm for cluster-based Multi-objective service-oriented network provisioning for ultra-scale computing environments by unifying the management of the local communication resources and the external inter-domain network services under a single point of view. We explore the potentials for representing the local network resources within a single distributed or parallel system and combine them together with the external communication services.\",\n \"Title: Adaptive Nature-Inspired Fog Architecture\\nYear: 2018\\n\\nAbstract: During the last decade, Cloud computing has efficiently exploited the economy of scale by providing low cost computational and storage resources over the Internet, eventually leading to consolidation of computing resources into large data centers. However, the nascent of the highly decentralized Internet of Things (IoT) technologies that cannot effectively utilize the centralized Cloud infrastructures pushes computing towards resource dispersion. Fog computing extends the Cloud paradigm by enabling dispersion of the computational and storage resources at the edge of the network in a close proximity to where the data is generated. In its essence, Fog computing facilitates the operation of the limited compute, storage and networking resources physically located close to the edge devices. However, the shared complexity of the Fog and the influence of the recent IoT trends moving towards deploying and interconnecting extremely large sets of pervasive devices and sensors, requires exploration of adaptive Fog architectural approaches capable of adapting and scaling in response to the unpredictable load patterns of the distributed IoT applications. In this paper we introduce a promising new nature- inspired Fog architecture, named SmartFog, capable of providing low decision making latency and adaptive resource management. By utilizing novel algorithms and techniques from the fields of multi- criteria decision making, graph theory and machine learning we model the Fog as a distributed intelligent processing system, therefore emulating the function of the human brain.\",\n \"Title: h-Fair: Asymptotic Scheduling of Heavy Workloads in Heterogeneous Data Centers\\nYear: 2018\\n\\nAbstract: Large scale computing solutions are increasingly used in the context of Big Data platforms, where efficient scheduling algorithms play an important role in providing optimized cluster resource utilization, throughput and fairness. This paper deals with the problem of scheduling a set of jobs across a cluster of machines handling the specific use case of fair scheduling for jobs and machines with heterogeneous characteristics. Although job and cluster diversity is unprecedented, most schedulers do not provide implementations that handle multiple resource type fairness in a heterogeneous system. We propose in this paper a new scheduler called h-Fair that selects jobs for scheduling based on a global dominant resource fairness heterogeneous policy, and dispatches them on machines with similar characteristics to the resource demands using the cosine similarity. We implemented h-Fair in Apache Hadoop YARN and we compare it with the existing Fair Scheduler that uses the dominant resource fairness policy based on the Google workload trace. We show that our implementation provides better cluster resource utilization and allocates more containers when jobs and machines have heterogeneous characteristics.\",\n \"Title: RM-BDP: Resource management for Big Data platforms.\\nYear: 2018\\n\\nAbstract: Nowadays, when we face with numerous data, when data cannot be classified into regular relational databases and new solutions are required, and when data are generated and processed rapidly, we need powerful platforms and infrastructure as support. Extracting valuable information from raw data is especially difficult considering the velocity of growing data from year to year and the fact that 80% of data is unstructured. In addition, data sources are heterogeneous (various sensors, users with different profiles, etc.) and are located in different situations or contexts. Cloud computing, which concerns large-scale interconnected systems with the main purpose of aggregation and efficient exploiting the power of widely distributed resources, represent one viable solution. Resource management and task scheduling play an essential role, in cases where one is concerned with optimized use of resources (Negru et al., 2017) [1].\",\n \"Title: E-OSched: a load balancing scheduler for heterogeneous multicores.\\nYear: 2018\\n\\nAbstract: The contemporary multicore era has adhered to the heterogeneous computing devices as one of the proficient platforms to execute compute-intensive applications. These heterogeneous devices are based on CPUs and GPUs. OpenCL is deemed as one of the industry standards to program heterogeneous machines. The conventional application scheduling mechanisms allocate most of the applications to GPUs while leaving CPU device underutilized. This underutilization of slower devices (such as CPU) often originates the sub-optimal performance of data-parallel applications in terms of load balance, execution time, and throughput. Moreover, multiple scheduled applications on a heterogeneous system further aggravate the problem of performance inefficiency. This paper is an attempt to evade the aforementioned deficiencies via initiating a novel scheduling strategy named OSched. An enhancement to the OSched named E-OSched is also part of this study. The OSched performs the resource-aware assignment of jobs to both CPUs and GPUs while ensuring a balanced load. The load balancing is achieved via contemplation on computational requirements of jobs and computing potential of a device. The load-balanced execution is beneficiary in terms of lower execution time, higher throughput, and improved utilization. The E-OSched reduces the magnitude of the main memory contention during concurrent job execution phase. The mathematical model of the proposed algorithms is evaluated by comparison of simulation results with different state-of-the-art scheduling heuristics. The results revealed that the proposed E-OSched has performed significantly well than the state-of-the-art scheduling heuristics by obtaining up to 8.09% improved execution time and up to 7.07% better throughput.\",\n \"Title: Semantic approach for multi-objective optimisation of the ENTICE distributed Virtual Machine and container images repository.\\nYear: 2019\\n\\nAbstract: New software engineering technologies facilitate development of applications from reusable software components, such as Virtual Machine and container images (VMI/CIs). Key requirements for the storage of VMI/CIs in public or private repositories are their fast delivery and cloud deployment times. ENTICE is a federated storage facility for VMI/CIs that provides optimisation mechanisms through the use of fragmentation and replication of images and a Pareto Multi-Objective Optimisation (MO) solver. The operation of the MO solver is, however, time-consuming due to the size and complexity of the metadata, specifying various non-functional requirements for the management of VMI/CIs, such as geolocation, operational cost, and delivery time. In this work, we address this problem with a new semantic approach, which uses an ontology of the federated ENTICE repository, knowledge base, and constraint-based reasoning mechanism. Open Source technologies such as Protege, Jena Fuseki, and Pellet were used to develop a solution. Two specific use cases, (1) repository optimisation with offline and (2) online redistribution of VMI/CIs, are presented in detail. In both use cases, data from the knowledge base are provided to the MO solver. It is shown that Pellet-based reasoning can be used to reduce the input metadata size used in the optimisation process by taking into consideration the geographic location of the VMI/CIs and the provenance of the VMI fragments. It is shown that this process leads to reduction of the input metadata size for the MO solver by up to 60% and reduction of the total optimisation time of the MO solver by up to 68%, while fully preserving the quality of the solution, which is significant.\",\n \"Title: DRUMS - Demand Response Management in a Smart City Using Deep Learning and SVR.\\nYear: 2018\\n\\nAbstract: Demand response management in smart cities is one of the most challenging tasks to be performed due to the continuous changes in the load profile of the home users. The existing proposals in the literature fail to observe the hidden patterns in the load profile of these users. So, to fill these gaps, the concept of deep learning has been used in this paper for smart energy management in a smart city. The consumption data from smart homes (SHs) is gathered and taken as an input to the deep learning model, convolution neural network (CNN). The CNN model learns the hidden patterns in the data and outputs different load curves. These load curves are then used to train a support vector regression (SVR) model, which predicts the overall load consumption of all SHs in the smart city. This prediction is then compared with the power generation from the grid and consequently the demand response (DR) of the connected SHs is managed so as to minimize the gap between predicted demand and supply. The proposed scheme has been evaluated on the dataset collected from PJM and open energy information with respect to load demand prediction and DR management. The results obtained prove the efficacy of the proposed scheme. The prediction errors, i.e., root mean squared error and mean absolute percentage error are observed less in comparison to the cases when CNN and SVR are used individually.\",\n \"Title: Scheduling scientific workflows to meet soft deadlines in the absence of failure models\\nYear: 2010\\n\\nAbstract: Highly distributed systems such as Clouds and Grids are used to execute complex scientific workflow applications by researchers from various areas of science. While scientists rightfully expect efficient and reliable execution of their applications, current systems often cannot deliver the required Quality of Service. We propose a dynamic execution and scheduling heuristic able to schedule workflow applications with a high degree of fault tolerance, while taking into account soft deadlines. Experimental results show that our method meets soft deadlines in volatile highly distributed systems in the absence of historic failure trace data or complex failure models of the target system.\",\n \"Title: Towards an Environment for Efficient and Transparent Virtual Machine Operations: The ENTICE Approach\\nYear: 2016\\n\\nAbstract: Cloud computing is based on Virtual Machines (VM) or containers, which provide their own software execution environment that can be deployed by facilitating technologies on top of various physical hardware. The use of VMs or containers represents an efficient way to automatize the overall software engineering and operation life-cycle. Some of the benefits include elasticity and high scalability, which increases the utilization efficiency and decreases the operational costs. VMs or containers as software artifacts are created using provider-specific templates and are stored in proprietary or public repositories for further use. However, technology specific choices may reduce their portability, lead to a vendor lock-in, particularly when applications need to run in federated Clouds. In this paper we present the current state of development of the novel concept of a VM repository and operational environment for federated Clouds named ENTICE. The ENTICE environment has been designed to receive unmodified and functionally complete VM images from its users, and transparently tailor and optimise them for specific Cloud infrastructures with respect to their size, configuration, and geographical distribution, such that they are loaded, delivered, and executed faster and with improved QoS compared to their current behaviour. Furthermore, in this work a specific use case scenario for the ENTICE environment has been provided and the underlying novel technologies have been presented.\",\n \"Title: Handling the uncertainty in resource performance for executing workflow applications in clouds.\\nYear: 2016\\n\\nAbstract: Execution of workflow applications in Cloud environments involves many uncertainties because of elastic resource provisioning and unstable performance of multitenant virtual machines (VM) instances over time. These uncertainties are usually either neglected by existing researches, or modeled with some probability distribution function. To address this gap, we extend a multi-objective workflow scheduling algorithm (MOHEFT) in two directions: (1) to deal with the dynamic nature of Cloud environments offering a potentially infinite amount of on-demand resources, and (2) to consider robustness as an objective that mitigates the variability in VM performance over time. Our new robust model, called R-MOHEFT, considers uncertainty in processing times of workflow activities without a precise estimation or known distribution function within an uncertainty interval. We approach this scheduling problem as a three-objective optimisation that considers makespan, monetary cost, and robustness as simultaneous objectives of a commercial Cloud environment. Our new algorithm is able to estimate the Pareto optimal set of scheduling solutions that resist against fluctuations in processing times three times better than its MOHEFT predecessor, with a tradeoff of only 15% worse Pareto frontier. R-MOHEFT's hypervolume suffers by only 5% to 16%, compared to the MOHEFT's drawback of 38% to surprisingly 87%, when the processing time fluctuates up to its double value.\",\n \"Title: Use Cases towards a Decentralized Repository for Transparent and Efficient Virtual Machine Operations\\nYear: 2017\\n\\nAbstract: Virtualization is a key enabling technology in Cloud computing that allows users to run multiple virtual machines (VMs) with their own application environment on top of physical hardware. It permits scaling up and down of applications by elastic on-demand provisioning of VMs in response to their variable load to achieve increased utilization efficiency at a lower operational cost, while guaranteeing the desired level of Quality of Service (QoS) to the end-users. Typically, VMs are created using provider-specific templates that are stored in proprietary repositories, leading to provider lock-in and hampering portability or simultaneous usage of multiple federated Clouds. In this context, optimization at the level of the virtual machine image is needed both by the applications and by the underlying Cloud providers for improved resource usage, operational costs, elasticity, storage use, and other desired QoS-related features. To overcome those issues, the ENTICE project researches and creates a novel VM repository and operational environment for federated Cloud infrastructures. There exists a large variety of industrial applications that can strongly benefit by the ENTICE environment. In this paper we present an interesting selection of complementary use cases that drive the definition of the essential requirements for the ENTICE environment, and more importantly, validate the introduced innovations.\",\n \"Title: Distributed environment for efficient virtual machine image management in federated Cloud architectures.\\nYear: 2018\\n\\nAbstract: The use of virtual machines (VMs) in Cloud computing provides various benefits in the overall software engineering lifecycle. These include efficient elasticity mechanisms resulting in higher resource utilization and lower operational costs. The VMs as software artifacts are created using provider-specific templates, called virtual machine images (VMI), and are stored in proprietary or public repositories for further use. However, some technology-specific choices can limit the interoperability among various Cloud providers and bundle the VMIs with nonessential or redundant software packages, leading to increased storage size, prolonged VMI delivery, stagnant VMI instantiation, and ultimately vendor lock-in. To address these challenges, we present a set of novel functionalities and design approaches for efficient operation of distributed VMI repositories, specifically tailored for enabling (1) simplified creation of lightweight and size optimized VMIs tuned for specific application requirements; (2) multi-objective VMI repository optimization; and (3) efficient reasoning mechanism to help optimizing complex VMI operations. The evaluation results confirm that the presented approaches can enable VMI size reduction by up to 55%, while trimming the image creation time by 66%. Furthermore, the repository optimization algorithms can reduce the VMI delivery time by up to 51% and cut down the storage expenses by 3%. Moreover, by implementing replication strategies, the optimization algorithms can increase the system reliability by 74%.\",\n \"Title: A New Model For Cloud Elastic Services Efficiency\\nYear: 2019\\n\\nAbstract: The speedup measures the improvement in performance when the computational resources are being scaled. The efficiency, on the other side, provides the ratio between the achieved speedup and the number of scaled computational resources (processors). Both parameters (speedup and efficiency), which are defined according to Amdahl's Law, provide very important information about performance of a computer system with scaled resources compared with a computer system with a single processor. However, as cloud elastic services' load is variable, apart of the scaled resources, it is vital to analyse the load in order to determine which system is more effective and efficient. Unfortunately, both the speedup and efficiency are not sufficient enough for proper modeling of cloud elastic services, as the assumptions for both the speedup and efficiency are that the system's resources are scaled, while the load is constant. In this paper, we extend the scaling of resources and define two additional scaled systems by (i) scaling the load and (ii) scaling both the load and resources. We introduce a model to determine the efficiency for each scaled system, which can be used to compare the efficiencies of all scaled systems, regardless if they are scaled in terms of load or resources. We have evaluated the model by using Windows Azure and the experimental results confirm the theoretical analysis. Although one can argue that web services are scalable and comply with Gustafson's Law only, we provide a taxonomy that classifies scaled systems based on the compliance with both the Amdahl's and Gustafson's laws.[GRAPHICS]For three different scaled systems (scaled resources R, scaled load L and combination RL), we introduce a model to determine the scaling efficiency. Our model extends the current definition of efficiency according to Amdahl's Law, which assumes scaling the resources, and not the load.E-R = S-R/p = T(1)/T-R center dot 1/p; E-L = S-L center dot N = T(1)/T-L center dot N; E-RL = S-RL center dot N/p = T(1)/T-RL center dot N/p\",\n \"Title: A New Model For Cloud Elastic Services Efficiency\\nYear: 2019\\n\\nAbstract: The speedup measures the improvement in performance when the computational resources are being scaled. The efficiency, on the other side, provides the ratio between the achieved speedup and the number of scaled computational resources (processors). Both parameters (speedup and efficiency), which are defined according to Amdahl's Law, provide very important information about performance of a computer system with scaled resources compared with a computer system with a single processor. However, as cloud elastic services' load is variable, apart of the scaled resources, it is vital to analyse the load in order to determine which system is more effective and efficient. Unfortunately, both the speedup and efficiency are not sufficient enough for proper modeling of cloud elastic services, as the assumptions for both the speedup and efficiency are that the system's resources are scaled, while the load is constant. In this paper, we extend the scaling of resources and define two additional scaled systems by (i) scaling the load and (ii) scaling both the load and resources. We introduce a model to determine the efficiency for each scaled system, which can be used to compare the efficiencies of all scaled systems, regardless if they are scaled in terms of load or resources. We have evaluated the model by using Windows Azure and the experimental results confirm the theoretical analysis. Although one can argue that web services are scalable and comply with Gustafson's Law only, we provide a taxonomy that classifies scaled systems based on the compliance with both the Amdahl's and Gustafson's laws.[GRAPHICS]For three different scaled systems (scaled resources R, scaled load L and combination RL), we introduce a model to determine the scaling efficiency. Our model extends the current definition of efficiency according to Amdahl's Law, which assumes scaling the resources, and not the load.E-R = S-R/p = T(1)/T-R center dot 1/p; E-L = S-L center dot N = T(1)/T-L center dot N; E-RL = S-RL center dot N/p = T(1)/T-RL center dot N/p\",\n \"Title: The Workflow Trace Archive: Open-Access Data from Public and Private Computing Infrastructures - Technical Report.\\nYear: 2019\\n\\nAbstract: Realistic, relevant, and reproducible experiments often need input traces collected from real-world environments. We focus in this work on traces of workflows---common in datacenters, clouds, and HPC infrastructures. We show that the state-of-the-art in using workflow-traces raises important issues: (1) the use of realistic traces is infrequent, and (2) the use of realistic, {\\\\it open-access} traces even more so. Alleviating these issues, we introduce the Workflow Trace Archive (WTA), an open-access archive of workflow traces from diverse computing infrastructures and tooling to parse, validate, and analyze traces. The WTA includes ${>}48$ million workflows captured from ${>}10$ computing infrastructures, representing a broad diversity of trace domains and characteristics. To emphasize the importance of trace diversity, we characterize the WTA contents and analyze in simulation the impact of trace diversity on experiment results. Our results indicate significant differences in characteristics, properties, and workflow structures between workload sources, domains, and fields.\",\n \"Title: Semantics-Aware Virtual Machine Image Management in IaaS Clouds\\nYear: 2019\\n\\nAbstract: Infrastructure-as-a-service (IaaS) Clouds concurrently accommodate diverse sets of user requests, requiring an efficient strategy for storing and retrieving virtual machine images (VMIs) at a large scale. The VMI storage management require dealing with multiple VMIs, typically in the magnitude of gigabytes, which entails VMI sprawl issues hindering the elastic resource management and provisioning. Nevertheless, existing techniques to facilitate VMI management overlook VMI semantics (i.e at the level of base image and software packages) with either restricted possibility to identify and extract reusable functionalities or with higher VMI publish and retrieval overheads. In this paper, we design, implement and evaluate Expelliarmus, a novel VMI management system that helps to minimize storage, publish and retrieval overheads. To achieve this goal, Expelliarmus incorporates three complementary features. First, it makes use of VMIs modelled as semantic graphs to expedite the similarity computation between multiple VMIs. Second, Expelliarmus provides a semantic aware VMI decomposition and base image selection to extract and store non-redundant base image and software packages. Third, Expelliarmus can also assemble VMIs based on the required software packages upon user request. We evaluate Expelliarmus through a representative set of synthetic Cloud VMIs on the real test-bed. Experimental results show that our semantic-centric approach is able to optimize repository size by 2.2 - 16 times compared to state-of-the-art systems (e.g. IBM's Mirage and Hemera) with significant VMI publish and slight retrieval performance improvement.\",\n \"Title: Where to Encode: A Performance Analysis of x86 and Arm-based Amazon EC2 Instances\\nYear: 2021\\n\\nAbstract: Video streaming became an undivided part of the Internet. To efficiently utilise the limited network bandwidth it is essential to encode the video content. However, encoding is a computationally intensive task, involving high-performance resources provided by private infrastructures or public clouds. Public clouds, such as Amazon EC2, provide a large portfolio of services and instances optimized f...\",\n \"Title: Fusioncl: A Machine-Learning Based Approach For Opencl Kernel Fusion To Increase System Performance\\nYear: 2021\\n\\nAbstract: Employing general-purpose graphics processing units (GPGPU) with the help of OpenCL has resulted in greatly reducing the execution time of data-parallel applications by taking advantage of the massive available parallelism. However, when a small data size application is executed on GPU there is a wastage of GPU resources as the application cannot fully utilize GPU compute-cores. There is no mechanism to share a GPU between two kernels due to the lack of operating system support on GPU. In this paper, we propose the provision of a GPU sharing mechanism between two kernels that will lead to increasing GPU occupancy, and as a result, reduce execution time of a job pool. However, if a pair of the kernel is competing for the same set of resources (i.e., both applications are compute-intensive or memory-intensive), kernel fusion may also result in a significant increase in execution time of fused kernels. Therefore, it is pertinent to select an optimal pair of kernels for fusion that will result in significant speedup over their serial execution. This research presents FusionCL, a machine learning-based GPU sharing mechanism between a pair of OpenCL kernels. FusionCL identifies each pair of kernels (from the job pool), which are suitable candidates for fusion using a machine learning-based fusion suitability classifier. Thereafter, from all the candidates, it selects a pair of candidate kernels that will produce maximum speedup after fusion over their serial execution using a fusion speedup predictor. The experimental evaluation shows that the proposed kernel fusion mechanism reduces execution time by 2.83x when compared to a baseline scheduling scheme. When compared to state-of-the-art, the reduction in execution time is up to 8%.\",\n \"Title: Handover Authentication Latency Reduction Using Mobile Edge Computing And Mobility Patterns\\nYear: 2021\\n\\nAbstract: With the advancement in technology and the exponential growth of mobile devices, network traffic has increased manifold in cellular networks. Due to this reason, latency reduction has become a challenging issue for mobile devices. In order to achieve seamless connectivity and minimal disruption during movement, latency reduction is crucial in the handover authentication process. Handover authentication is a process in which the legitimacy of a mobile node is checked when it crosses the boundary of an access network. This paper proposes an efficient technique that utilizes mobility patterns of the mobile node and mobile Edge computing framework to reduce handover authentication latency. The key idea of the proposed technique is to categorize mobile nodes on the basis of their mobility patterns. We perform simulations to measure the networking latency. Besides, we use queuing model to measure the processing time of an authentication query at an Edge servers. The results show that the proposed approach reduces the handover authentication latency up to 54% in comparison with the existing approach.\",\n \"Title: FAUST: Fast Per-Scene Encoding Using Entropy-Based Scene Detection and Machine Learning\\nYear: 2021\\n\\nAbstract: HTTP adaptive video streaming is a widespread and sought-after technology on the Internet that allows clients to dynamically switch between different stream qualities presented in the bitrate ladder to optimize overall received video quality. Currently, there exist several approaches of different complexity for building such a ladder. The simplest method is to use a static bitrate ladder, and the ...\",\n \"Title: Autotuning of Exascale Applications With Anomalies Detection.\\nYear: 2021\\n\\nAbstract: The execution of complex distributed applications in exascale systems faces many challenges, as it involves empirical evaluation of countless code variations and application runtime parameters over a heterogeneous set of resources. To mitigate these challenges, the research field of autotuning has gained momentum. The autotuning automates identifying the most desirable application implementation in terms of code variations and runtime parameters. However, the complexity and size of the exascale systems make the autotuning process very difficult, especially considering the number of parameter variations that have to be identified. Therefore, we introduce a novel approach for autotuning exascale applications based on a genetic multi-objective optimization algorithm integrated within the ASPIDE exascale computing framework. The approach considers multi-dimensional search space with support for pluggable objective functions, including execution time and energy requirements. Furthermore, the autotuner employs a machine learning-based event detection approach to detect events and anomalies during application execution, such as hardware failures or communication bottlenecks.\",\n \"Title: Decentralized Machine Learning for Intelligent Health-Care Systems on the Computing Continuum\\nYear: 2022\\n\\nAbstract: The introduction of electronic personal health records (EHRs) enables nationwide information exchange and curation among different health-care systems. However, current EHR systems are centrally orchestrated, which could potentially lead to a single point of failure.\",\n \"Title: Hybrid On/Off Blockchain Approach for Vehicle Data Management, Processing and Visualization Exemplified by the ADAPT Platform.\\nYear: 2022\\n\\nAbstract: Hybrid on/off-blockchain vehicle data management approaches have received a lot of attention in recent years. However, there are various technical challenges remained to deal with. In this paper we relied on real-world data from Austria to investigate the effects of connectivity on the transport of personal protective equipment. We proposed a three-step mechanism to process, simulate, and store/visualize aggregated vehicle datasets together with a formal pipeline process workflow model. To this end, we implemented a hybrid blockchain platform based on the hyperledger fabric and gluster file systems. The obtained results demonstrated efficiency and stability for both hyperledger fabric and gluster file systems and ability of the both on/off-blockchain mechanisms to meet the platform quality of service requirements\",\n \"Title: Big Data Pipeline Scheduling and Adaptation on the Computing Continuum\\nYear: 2022\\n\\nAbstract: The Computing Continuum, covering Cloud, Fog, and Edge systems, promises to provide on-demand resource-as-a-service for Internet applications with diverse requirements, ranging from extremely low latency to high-performance processing. However, eminent challenges in automating the resources management of Big Data pipelines across the Computing Continuum remain. The resource management and adaptation for Big Data pipelines across the Computing Continuum require significant research effort, as the current data processing pipelines are dynamic. In contrast, traditional resource management strategies are static, leading to inefficient pipeline scheduling and overly complex process deployment. To address these needs, we propose in this work a scheduling and adaptation approach implemented as a software tool to lower the technological barriers to the management of Big Data pipelines over the Computing Continuum. The approach separates the static scheduling from the run-time execution, empowering domain experts with little infrastructure and software knowledge to take an active part in the Big Data pipeline adaptation. We conduct a feasibility study using a digital healthcare use case to validate our approach. We illustrate concrete scenarios supported by demonstrating how the scheduling and adaptation tool and its implementation automate the management of the lifecycle of a remote patient monitoring, treatment, and care pipeline.\",\n \"Title: Big Data Pipelines on the Computing Continuum: Tapping the Dark Data\\nYear: 2022\\n\\nAbstract: The computing continuum enables new opportunities for managing big data pipelines concerning efficient management of heterogeneous and untrustworthy resources. We discuss the big data pipelines lifecycle on the computing continuum and its associated challenges, and we outline a future research agenda in this area.\",\n \"Title: Matching-based Scheduling of Asynchronous Data Processing Workflows on the Computing Continuum\\nYear: 2022\\n\\nAbstract: Today's distributed computing infrastructures en-compass complex workflows for real-time data gathering, transferring, storage, and processing, quickly overwhelming centralized cloud centers. Recently, the computing continuum that federates the Cloud services with emerging Fog and Edge devices represents a relevant alternative for supporting the next-generation data processing workflows. However, eminent challenges in automating data processing across the computing continuum still exist, such as scheduling heterogeneous devices across the Cloud, Fog, and Edge layers. We propose a new scheduling algorithm called C\\n<sup xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">3</sup>\\n -MATCH, based on matching theory principles, involving two sets of players negotiating different utility functions: 1) workflow microservices that prefer computing devices with lower data processing and queuing times; 2) computing continuum devices that prefer microservices with corresponding resource requirements and less data transmission time. We evaluate \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">$C^{3}$</tex>\\n-MATCH using real-world road sign inspection and sentiment analysis workflows on a federated computing continuum across four Cloud, Fog, and Edge providers. Our combined simulation and real execution results reveal that \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">$C^{3}$</tex>\\n-MATCH achieves up to 67% lower completion time than three state-of-the-art methods with 10 ms-1000 ms higher transmission time.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":158,"string":"158"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13560,"string":"13,560"}}},{"rowIdx":13561,"cells":{"question_id":{"kind":"string","value":"Alexander G Loukianov"},"question":{"kind":"string","value":"Generate an abstract for the title \"Robust sliding mode regulation of nonlinear systems.\" using the following items: nonlinear sliding mode, output regulation, state feedback, affine control systems, Pendubot system"},"answer":{"kind":"string","value":"In this paper, the problem of nonlinear Sliding Mode (SM) output regulation is addressed. In particular, a state feedback SM regulator problem is formulated, taking the concepts related to the zero output tracking submanifold as a starting point, and a solution is proposed for general nonlinear affine control systems subject to unmodeled time-varying disturbance. Then, the problem is studied for the particular class of nonlinear systems presented in the so-called Regular form. The effectiveness of the proposed method is demonstrated by the application to the Pendubot system."},"correct_answer":{"kind":"string","value":"In this paper, the problem of nonlinear Sliding Mode (SM) output regulation is addressed. In particular, a state feedback SM regulator problem is formulated, taking the concepts related to the zero output tracking submanifold as a starting point, and a solution is proposed for general nonlinear affine control systems subject to unmodeled time-varying disturbance. Then, the problem is studied for the particular class of nonlinear systems presented in the so-called Regular form. The effectiveness of the proposed method is demonstrated by the application to the Pendubot system."},"documents":{"kind":"list like","value":["Title: Hybrid Control of Induction Motors via Sampled Closed Representations\nYear: 2008\n\nAbstract: In this paper, a controller for induction motors is proposed. A continuous feedback is first applied to obtain a discrete-time model in closed form. Then, on the basis of these exact sampled dynamics, a discrete-time controller ensuring speed and flux modulus reference tracking is determined, making use of the sliding mode technique. The resulting controller is hence hybrid, in the sense that it contains both continuous and discrete-time terms. It is shown how to implement such a hybrid controller using the so-called exponential holder, which is the only device to be implemented analogically, together with an analog integrator. Moreover, a discrete-time reduced-order observer is designed for rotor fluxes and load torque estimation. The performance of the proposed controller is finally studied by simulation and experimental tests.","Title: Discrete-Time Nonlinear Systems Inverse Optimal Control: A Control Lyapunov Function Approach\nYear: 2011\n\nAbstract: This paper presents an inverse optimal control approach for exponential stabilization of discrete-time non-linear systems, avoiding to solve the associated Hamilton-Jacobi-Bellman (HJB) equation, and minimizing a meaningful cost function. This stabilizing optimal controller is based on a discrete-time control Lyapunov function. The applicability of the proposed approach is illustrated via simulations by stabilization of an example.","Title: Observer based controller for internal combustion engine\nYear: 2013\n\nAbstract: In this work, an observer-based controller for an internal combustion engine is presented. At first, an algorithm for the estimation of an unknown function of the internal combustion engine is designed, since it is very difficult to obtain direct measurements of this variable. This estimator is based on sliding mode algorithms, providing a finite time and robust estimation, using only measurements from the velocity of the engine. On the other hand, with the measured velocity and the estimates of the other variables, a robust controller is synthesized for the engine. In order to considerate the actuator dynamics, the proposed control scheme is based on the master-slave structure, regarding the controller for the actuator as the slave one. For this scheme, the backstepping algorithm is used to design the master controller. Then, the calculated control input signal to the engine is used as a reference for the throttle actuator which is driven by a direct current (DC) motor. Thus, a high-order sliding mode controller is applied to the actuator in order to track the control input signal and reject perturbations, as the applied mechanical load, regulating the velocity of the combustion engine. Numerical simulations show the efficient performance of this proposal.","Title: Neural identification and control for linear induction motors\nYear: 2005\n\nAbstract: We propose a new adaptive control scheme, composed of a neural identifier and a nonlinear controller and applied it to a linear induction motor (LIM). In order to compare the performance of LIM, we use α - β and d - q models. A neural identifier of triangular form is proposed for both models as a nonlinear block controllable form (NBC). Then, a reduced order observer is designed in order to estimate no measured variables. Learning law for neural network weights ensure that the identification error converges to zero exponentially. Sliding mode control is developed to track velocity and flux magnitude. Simulations are presented to compare the behaviour of both models of LIM and the applicability of the proposed identification and control scheme.","Title: Mean-square filtering for uncertain linear stochastic systems\nYear: 2010\n\nAbstract: This paper presents the mean-square joint filtering and parameter identification problem for uncertain linear stochastic systems with unknown parameters in both, state and observation, equations, where the unknown parameters are considered Wiener processes. The original problem is reduced to the filtering problem for an extended state vector that incorporates parameters as additional states. The resulting filtering system is polynomial in state and linear in observations. The obtained mean-square filter for the extended state vector also serves as the mean-square identifier for the unknown parameters. A simulation example is included to show convergence of the designed mean-square state filter and parameter identifier for both, positive and negative, parameter values.","Title: Recurrent neural networks with fixed time convergence for linear and quadratic programming.\nYear: 2013\n\nAbstract: In this paper, a new class of recurrent neural networks which solve linear and quadratic programs are presented. Their design is considered as a sliding mode control problem, where the network structure is based on the KarushKuhn-Tucker (KKT) optimality conditions with the KKT multipliers considered as control inputs to be implemented with fixed time stabilizing terms, instead of common used activation functions. Thus, the main feature of the proposed network is its fixed convergence time to the solution. That means, there is time independent to the initial conditions in which the network converges to the optimization solution. Simulations show the feasibility of the current approach.","Title: Mean-square state and noise intensity estimation for uncertain linear systems\nYear: 2010\n\nAbstract: This paper presents the mean-square joint state and diffusion coefficient (noise intensity) estimator for linear stochastic systems with unknown noise intensity over linear observations, where unknown parameters are considered Wiener processes. The original problem is reduced to the filtering problem for an extended state vector that incorporates parameters as additional states. Since the noise intensities cannot be observable in the original linear system, the new quadratic vector variable formed by the diagonal of the matrix square of the system state is introduced. The obtained mean-square filter for the extended state vector also serves as the optimal identifier for the unknown parameters. Performance of the designed mean-square state filter and parameter identifier is verified in an illustrative example.","Title: Real-time torque control using discrete-time recurrent high-order neural networks.\nYear: 2013\n\nAbstract: This paper presents a discrete-time direct current (DC) motor torque tracking controller, based on a recurrent high-order neural network to identify the plant model. In order to train the neural identifier, the extended Kalman filter (EKF) based training algorithm is used. The neural identifier is in series-parallel configuration that constitutes a well approximation method of the real plant by the neural identifier. Using the neural identifier structure that is in the nonlinear controllable form, the block control (BC) combined with sliding modes (SM) control techniques in discrete-time are applied. The BC technique is used to design a nonlinear sliding manifold such that the resulting sliding mode dynamics are described by a desired linear system. For the SM control technique, the equivalent control law is used in order to the plant output tracks a reference signal. For reducing the effect of unknown terms, it is proposed a specific desired dynamics for the sliding variables. The control problem is solved by the indirect approach, where an appropriate neural network (NN) identification model is selected; the NN parameters (synaptic weights) are adjusted according to a specific adaptive law (EKF), such that the response of the NN identifier approximates the response of the real plant for the same input. Then, based on the designed NN identifier a stabilizing or reference tracking controller is proposed (BC combined with SM). The proposed neural identifier and control applicability are illustrated by torque trajectory tracking for a DC motor with separate winding excitation via real-time implementation.","Title: Neural block control for synchronous generators\nYear: 2009\n\nAbstract: In this paper, a novel approach to control a single generator, connected to an infinite bus, is presented. Modifying published results for nonlinear identification using recurrent neural networks, a block controllable neural identifier is proposed, based on this neural model a control law is derived, which combines sliding modes and block control. The neural identifier and the proposed control law allows to reject external disturbances caused by generator terminal short circuits and mechanical power variations. Applicability of the approach is tested via simulations.","Title: Dynamic triangular neural controller for stepper motor trajectory tracking.\nYear: 2002\n\nAbstract: In this paper, we present a novel neural controller for this kind of motor. This controller is developed as follows. Modifying published results for nonlinear identification using dynamic neural networks (NNs), an NN identifier of triangular form is implemented. Then, based on this mode, a control law using sliding modes is derived. This neural identifier and the proposed control law allow trajectory tracking for stepping motors. Applicability of the approach is tested via simulations.","Title: Discrete-Time Neural Inverse Optimal Control for Nonlinear Systems via Passivation.\nYear: 2012\n\nAbstract: This paper presents a discrete-time inverse optimal neural controller, which is constituted by combination of two techniques: 1) inverse optimal control to avoid solving the Hamilton-Jacobi-Bellman equation associated with nonlinear system optimal control and 2) on-line neural identification, using a recurrent neural network trained with an extended Kalman filter, in order to build a model of the ...","Title: Decentralized neural identification and control for uncertain nonlinear systems: Application to planar robot\nYear: 2010\n\nAbstract: This paper presents a discrete-time decentralized neural identification and control for large-scale uncertain nonlinear systems, which is developed using recurrent high order neural networks (RHONN); the neural network learning algorithm uses an extended Kalman filter (EKF). The discrete-time control law proposed is based on block control and sliding mode techniques. The control algorithm is first simulated, and then implemented in real time for a two degree of freedom (DOF) planar robot.","Title: Particle swarm optimization for discrete-time inverse optimal control of a doubly fed induction generator.\nYear: 2013\n\nAbstract: In this paper, the authors propose a particle swarm optimization (PSO) for a discrete-time inverse optimal control scheme of a doubly fed induction generator (DFIG). For the inverse optimal scheme, a control Lyapunov function (CLF) is proposed to obtain an inverse optimal control law in order to achieve trajectory tracking. A posteriori, it is established that this control law minimizes a meaningful cost function. The CLFs depend on matrix selection in order to achieve the control objectives; this matrix is determined by two mechanisms: initially, fixed parameters are proposed for this matrix by a trial-and-error method and then by using the PSO algorithm. The inverse optimal control scheme is illustrated via simulations for the DFIG, including the comparison between both mechanisms.","Title: Discrete-time inverse optimal control for nonlinear systems trajectory tracking.\nYear: 2010\n\nAbstract: This paper presents an inverse optimal control approach for output tracking of discrete-time nonlinear systems, avoiding to solve the associated Hamilton-Jacobi-Bellman (HJB) equation, and minimizing a meaningful cost function. This stabilizing optimal controller is based on discrete-time passivity theory. The applicability of the proposed approach is illustrated via simulations by trajectory tracking control of a planar robot.","Title: An optimal sliding mode-like regulator for nonlinear polynomial systems\nYear: 2011\n\nAbstract: This article presents the optimal control problem for a nonlinear polynomial system with respect to a Bolza-Meyer criterion with non-quadratic state energy terms. A distinctive feature of the obtained result is that a part of the optimally controlled state trajectory occurs to be in a sliding mode, i.e. represents a motion along a certain manifold. The optimal solution is obtained as a sliding mode-like control, whereas the conventional linear feedback control fails to provide a causal solution. Performance of the obtained optimal controller is verified in the illustrative example against the conventional polynomial-quadratic regulator that is optimal for the quadratic Bolza-Meyer criterion. The simulation results confirm an advantage in favour of the designed sliding mode control.","Title: Sliding Mode Block Control Regulation Of The Pendubot\nYear: 2011\n\nAbstract: In this work, the problem of nonlinear regulation of an underactuated system is treated by means of sliding mode continuous control actions combined with block control technique. The sliding mode state feedback output regulator based on the super twisting algorithm, is applied to the Pendubot system. The transformation of the original system to regular form and then block control technique are used to design a sliding manifold with asymptotically stable motion. To verify the effectiveness of the method, simulations are carried out.","Title: Copper And Core Loss Minimization For Induction Motors Using High-Order Sliding-Mode Control\nYear: 2012\n\nAbstract: A novel nonlinear affine model for an induction motor with core loss is developed in the well-known (alpha, beta) stationary reference frame, where the core is represented with a resistance in parallel with a magnetization inductance. Then, an optimal rotor flux modulus is calculated such that the power loss due to stator, rotor, and core resistances is minimized, and as a consequence, the motor efficiency is raised; therefore, this flux modulus is forced to be tracked by the induction motor along with a desired rotor velocity by means of a high-order sliding-mode controller, the supertwisting algorithm. Using a novel Lyapunov function, the closed-loop stability of the system is demonstrated. Moreover, a classical sliding-mode observer is designed for the estimation of unmeasurable variables like rotor fluxes and magnetization currents. For the load torque, a Luenberger observer is designed. The performance of the proposed controller is finally studied by simulation and experimental tests. It was observed that the steady-state optimal flux signal corresponds to the load torque profile. This fact suggests that the flux demand is the necessary one to produce the electric torque that can cancel out the load torque.","Title: Discrete Integral Sliding Mode Control In Visual Object Tracking\nYear: 2010\n\nAbstract: Object tracking is an important process for many applications in computer vision. This process must be implemented in a discrete manner because the images are available only at certain periods. A discrete integral sliding mode algorithm is proposed to control a stereo vision system and perform the aforementioned second task. The kinematic model of the structure is obtained using geometric algebra. The performance of the controller is compared with proportional-integral-derivative control via simulation. The implementation for a pan tilt unit is presented in real time. The algorithm presents a good and robust performance.","Title: Second order sliding mode sensorless torque regulator for induction motor.\nYear: 2010\n\nAbstract: In this paper the torque and the square of the rotor fluxes magnitude using a second order sliding mode (SM) controller for an induction motor, is proposed. The main contribution of the control scheme presented here is the output regulation as well as the asymptotically stabilization of the zero dynamics. The sliding manifold is designed such that the speed zero dynamics become asymptotically stable on this manifold. To enforce the SM motion on the designed manifold a super-twisting SM algorithm is used. This algorithm permits to reduce chattering that is inherent in standard sliding mode control and to improve accuracy. The rotor fluxes and speed are estimated by means of an adaptive observer, whereas the estimate of the load torque is obtained via design of a robust differentiator. The effectiveness of the designed observer-control scheme is confirmed via simulation.","Title: Discrete-Time Recurrent Neural Dc Motor Control Using Kalman Learning\nYear: 2008\n\nAbstract: An adaptive tracking controller for a discrete-time direct current (DC) motor model in presence of bounded disturbances is presented. A high order neural network is used to identify the plant model; this network is trained with an extended Kalman filter. Then, the discrete-time block control and sliding modes techniques are used to develop the reference tracking control. This paper includes also the respective stability analysis and a strategy to avoid specific adaptive weights zero-crossing. The scheme is illustrated via simulations for a discrete-time nonlinear model of an electric DC motor.","Title: Discrete time nonlinear identification via recurrent high order neural networks for a three phase induction motor\nYear: 2010\n\nAbstract: This paper deals with the problem of discrete-time nonlinear system identification via Recurrent High Order Neural Networks It includes the respective stability analysis on the basis of the Lyapunov approach for the extended Kalman filter (EKF)-based NN training algorithm, which is applied for learning Applicability of the scheme is illustrated via simulation for a discrete-time nonlinear model of an electric induction motor.","Title: Real-time Discrete Backstepping Neural Control for Induction Motors.\nYear: 2011\n\nAbstract: This brief focuses on real-time implementation, as applied to a three-phase induction motor, of results already published in 2007. The proposed controller is based on a high-order neural network, trained online using Kalman filter learning, to approximate a control law designed by the backstepping technique.","Title: Discrete Time Block Control Of A Double Fed Induction Generator Using Sliding Modes\nYear: 2009\n\nAbstract: In this paper, the authors propose a block control scheme using sliding modes, for a double fed induction generator connected to an infinity bus. To estimate the mechanical torque, a nonlinear observer is utilized; additionally, as a viable alternative, to eliminate the requirement on this estimation, an integrator is included. To reduce the effect of parameter variations, the one step delayed disturbance algorithm is used. The stability of the closed loop system is analyzed by the Lyapunov techniques. The applicability of the proposed scheme is tested via simulations.","Title: Robust Trajectory Tracking for an Electrohydraulic Actuator\nYear: 2009\n\nAbstract: Various robust control techniques, such as integral-block, sliding-mode, and H-infinity controls, are combined to design a controller, forcing an electrohydraulic actuator which is driven by a servovalve to track a chaotic reference trajectory. This approach enables one to compensate the inherent nonlinearities of the actuator and reject matched external disturbances and attenuate mismatched exter...","Title: Recurrent neural block form control\nYear: 2003\n\nAbstract: By modifying some previously developed results for nonlinear identification and control using recurrent neural networks, the present authors propose a new neural network identifier in block form, and, based on this model, a control law is developed by combining sliding mode and block controls. This neural identifier and control law allow satisfactory trajectory tracking for general nonlinear systems. Applicability of the new design is illustrated, via simulations, for robust tracking control of stepping motors.","Title: Real-Time Discrete Nonlinear Identification Via Recurrent High Order Neural Networks\nYear: 2010\n\nAbstract: This paper deals with the discrete-time nonlinear system identification via Recurrent High Order Neural Networks, trained with an extended Kalman filter (EKF) based algorithm. The paper also includes the respective stability analysis on the basis of the Lyapunov approach for the whole scheme. Applicability of the scheme is illustrated via real-time implementation for a three phase induction motor.","Title: Super-twisting sensorless control of linear induction motors\nYear: 2012\n\nAbstract: In this work, a sensorless control scheme is presented for linear induction motors. The secondary fluxes are algebraically calculated by first determining the primary fluxes, then, a super-twisitng observer for secondary fluxes is designed in order to retrieve the back-EMF components by means of the equivalent control method. Based on these components, the linear velocity is determined and used in a linear velocity and load force observer, where the estimated variables along with primary current and voltage measurements are used to control the linear induction motor for the tracking of a reference linear velocity signal and a square secondary flux modulus, all by means of a super-twisting controller. Simulations show that the proposed observer based controller scheme performs well when tracking a time varying linear velocity signal.","Title: Discrete-Time Backstepping Neural Control for Synchronous Generators.\nYear: 2007\n\nAbstract: This paper deals with adaptive tracking for discrete-time MIMO nonlinear systems in presence of hounded disturbances. In this paper, a high order neural network structure is used to approximate a control law designed by the backstepping technique, applied to a block strict feedback form (BSFF). The learning algorithm for the HONN is based on an Extended Kalman Filter (EKF). This paper also includes the respective stability analysis, using the Lyapunov approach. The viability of the proposed approach is tested via simulations, by its application to synchronous generators control.","Title: Decentralized adaptive recurrent neural control structure\nYear: 2007\n\nAbstract: This paper presents a novel decentralized variable structure neural control approach for large-scale uncertain systems, which is developed using recurrent high-order neural networks (RHONN). It is assumed that each subsystem belongs to a class of block-controllable nonlinear systems whose vector fields includes interconnection terms, which are bounded by nonlinear functions. A decentralized RHONN structure and the respective learning law are proposed in order to approximate online the dynamical behavior of each nonlinear subsystem. The control law, which is able to regulate and to track the desired reference signals, is designed using the well-known variable structure theory. The stability of the whole system is analyzed via the Lyapunov methodology. The applicability of the proposed decentralized identification and control algorithm is illustrated via simulations as applied to an interconnected double inverted pendulum.","Title: Speed-gradient inverse optimal control for discrete-time nonlinear systems.\nYear: 2011\n\nAbstract: This paper presents a speed-gradient-based inverse optimal control approach for the asymptotic stabilization of discrete-time nonlinear systems. With the solution presented, we avoid to solve the associated Hamilton-Jacobi-Bellman equation, and a meaningful cost function is minimized. The proposed stabilizing optimal controller uses the speed-gradient algorithm and is based on the proposal of what is called a discrete-time control Lyapunov function. This combined approach is referred to as the speed-gradient inverse optimal control. An example is used to illustrate the methodology. Several simulations are provided.","Title: Discrete-time block control for a doubly fed induction generator coupled to a wind turbine\nYear: 2011\n\nAbstract: This paper proposes a control scheme based on the block control technique using sliding modes, for a doubly fed induction generator connected to an infinity bus. This control scheme is compared with a linear control using PIs. Simulation results in PSCAD/EMTDC are included to illustrate the applicability of the scheme. A comparison between the PI controller and the proposed block controller is presented, from which, it is verified that our approach is faster to reach the speed reference.","Title: Discrete-time recurrent high order neural networks for nonlinear identification\nYear: 2010\n\nAbstract: This paper focuses on the problem of discrete-time nonlinear system identification via recurrent high order neural networks. It includes the respective stability analysis on the basis of the Lyapunov approach for the NN training algorithm. Applicability of the proposed scheme is illustrated via simulation for a discrete-time nonlinear model of an electric induction motor.","Title: Discrete-Time Recurrent High Order Neural Observer for Induction Motors\nYear: 2007\n\nAbstract: A nonlinear discrete-time neural observer for the state estimation of a discrete-time induction motor model, in presence of external and internal uncertainties is presented. The observer is based on a discrete-time recurrent high order neural network (RHONN) trained with an extended Kalman filter (EKF)-based algorithm. This observer estimates the state of the unknown discrete-time nonlinear system, using a parallel configuration. The paper also includes the stability proof on the basis of the Lyapunov approach. To illustrate the applicability simulation results are included.","Title: Digital Sliding-Mode Sensorless Control For Surface-Mounted Pmsm\nYear: 2014\n\nAbstract: In this work, a sampled-data model for surface-mounted permanent magnet synchronous motors (SMPMSMs) is proposed based on the symplectic Euler method. Then, based on the obtained model, a digital sliding-mode controller is designed for rotor velocity output tracking. Moreover, for rotor position and velocity and load estimation, a digital observer is designed. A separation principle is introduced in order to verify that the proposed controller can accomplish its task with estimated signals. Open-loop simulations predict that the sampled-data model obtained via the symplectic Euler method performs better than a model obtained with the classical explicit Euler method. Moreover, closed-loop simulations are carried on where the proposed controller is compared with a sampled continuous-time controller, where the controller here designed yields to better results even when the sampling period is increased. Finally, the performance of the proposed controller is verified by experimental tests.","Title: Real-Time Discrete Recurrent High Order Neural Observer For Induction Motors\nYear: 2008\n\nAbstract: A nonlinear discrete-time neural observer for the state estimation of a discrete-time induction motor model, in presence of external and internal uncertainties is presented. The observer is based on a discrete-time recurrent high order neural network (RHONN) trained with an extended Kalman filter (EKF)-based algorithm. This observer estimates the state of the unknown discrete-time nonlinear system, using a parallel configuration. The paper also includes the stability proof on the basis of the Lyapunov approach. To illustrate the applicability real-time results are included.","Title: Real-time sliding mode control for a doubly fed induction generator\nYear: 2011\n\nAbstract: This paper proposes a control scheme based on the block control technique using sliding modes, for a doubly fed induction generator (DFIG) prototype connected to an infinity bus. The DFIG is widely used as a wind generator; it allows the rotor speed to vary while synchronizing the stator directly to a fixed frequency power system. The generation scheme for the DFIG has one back-to-back PWM voltage-source converter connected between the rotor and the electrical grid. The rotor side converter (RSC) is connected via a dc link to the grid side converter (GSC), which is in turn connected to the stator terminals directly or through a step-up transformer. A block control scheme for the RSC is proposed to control the electric torque and the reactive power independently. The variables to be controlled by the GSC are the dc voltage in the dc link and the reactive power in the step-up terminals. The performance of the control scheme proposed is illustrated via real-time implementation in a 1/4HP DFIG prototype.","Title: Multimachine Power-System Control: Integral-SM Approach\nYear: 2009\n\nAbstract: In this paper, a robust nonlinear control scheme based on the combination of the block-control linearization and integral sliding modes is proposed. This control technique is applied to multimachine electric-power systems. The plant model takes into account the interactions in power systems between the electrical and mechanical dynamics and load constraints. A nonlinear observer is designed to est...","Title: Integral high order sliding mode control of single-phase induction motor\nYear: 2013\n\nAbstract: An observer-based controller for the single-phase induction motor is proposed in this paper. The scheme presented is formulated using block control feedback linearization technique and high order sliding mode algorithms with measurements of the rotor speed and stator currents. A second order sliding mode observer is included into the controller design in order to obtain estimates of the rotor flux. The stability of the complete closed-loop system is analyzed in the presence of model uncertainty, namely, rotor resistance variation and bounded time-varying load torque.","Title: Backstepping control with sliding mode estimation for a hexacopter\nYear: 2013\n\nAbstract: This paper presents the dynamic nonlinear model and the design of a controller for a hexarotor aerial robot. The control algorithm is based on the backstepping technique and it is applied to the trajectory tracking problem. The wind parameters resulting from the aerodynamic forces and moments acting on the hexacopter are estimated via a sliding mode observer with the super twisting algorithm in order to ensure robustness against aerial external disturbances and parameter variations. The performance and effectiveness of the proposed controller are tested in a simulation study taking into account external disturbances which are function of time.","Title: Discrete-time neural block control for a doubly fed induction generator\nYear: 2011\n\nAbstract: This paper proposes a control scheme based on the discrete-time block control technique using sliding modes, for a doubly fed induction generator connected to an infinity bus. In order to obtain the generator mathematical model, it is proposed to use a recurrent high order neural network (RHONN) identifier, which is trained with an extended Kalman filter (EFK) algorithm. Parameter changes are applied to test the scheme robustness. Its performance is illustrated via simulations.","Title: High-Order Sliding Mode Block Control of Single-Phase Induction Motor.\nYear: 2014\n\nAbstract: A new sliding mode (SM) observer-based controller for single-phase induction motor is designed. The proposed control scheme is formulated using block control feedback linearization technique and high-order SM algorithms with measurements of the rotor speed and stator currents. The stability of the complete closed-loop system, including the rotor flux second-order SM observer, is analyzed in the presence of model uncertainty, namely, rotor resistance variation and bounded time-varying load torque.","Title: A simple recurrent neural network for solution of linear programming: Application to a Microgrid\nYear: 2014\n\nAbstract: The aim of this paper is to present a simple new class of recurrent neural networks, which solves linear programming. It is considered as a sliding mode control problem, where the network structure is based on the Karush-Kuhn-Tucker (KKT) optimality conditions, and the KKT multipliers are the control inputs to be implemented with finite time stabilizing terms based on the unit control, instead of common used activation functions. Thus, the main feature of the proposed network is the fixed number of parameters despite of the optimization problem dimension, which means, the network can be easily scaled from a small to a higher dimension problem. The applicability of the proposed scheme is tested on real-time optimization of an electrical Microgrid prototype.","Title: A discontinuous recurrent neural network with predefined time convergence for solution of linear programming\nYear: 2014\n\nAbstract: The aim of this paper is to introduce a new recurrent neural network to solve linear programming. The main characteristic of the proposed scheme is its design based on the predefined-time stability. The predefined-time stability is a stronger form of finite-time stability which allows the a priori definition of a convergence time that does not depend on the network initial state. The network structure is based on the Karush-Kuhn-Tucker (KKT) conditions and the KKT multipliers are proposed as sliding mode control inputs. This selection yields to an one-layer recurrent neural network in which the only parameter to be tuned is the desired convergence time. With this features, the network can be easily scaled from a small to a higher dimension problem. The simulation of a simple example shows the feasibility of the current approach.","Title: Torque controller via second order sliding modes of WRIG impelled by DC-motor for application in wind systems\nYear: 2012\n\nAbstract: In this paper, the authors propose three robust non-linear controllers based on a second order sliding mode technique named super-twisting method which can be applied into work bench for research related to wind system. The main control scheme is proposed to control the electromagnetic torque and stator power factor of a wound rotor induction generator connected to a local electrical network and mechanically coupled with a DC-motor whose velocity is controlled applying feedback linearization technique and super-twisting algorithm; it is known that the DC-motor can be emulate the operation of the wind turbine. Finally, a grid side converter controller is proposed to regulate the DC link bus located between to power converters connected in “back to back” standard configuration, applying linearization by block control and super-twisting algorithm. The performance of the designed controllers for the DC motor-WR induction generator group is validated through real time implementation.","Title: Visual servoing and robust object manipulation using symmetries and Conformai Geometric Algebra\nYear: 2014\n\nAbstract: Object tracking and manipulation is an important process for many applications in robotics and computer vision. A novel 3D pose estimation of objects using reflectionally symmetry formulated in Conformai Geometric Algebra (CGA) is proposed in this work. The synthesis of the kinematics model for robots and a sliding mode controller using the CGA approach is described. Real time implementation results are presented for the pose estimation of object using a stereo vision system.","Title: Discrete-time Neural Network Control for a Linear Induction Motor\nYear: 2008\n\nAbstract: This paper presents a discrete-time control for a linear induction motor (LIM). First, an identifier is proposed with a nonlinear block controllable form (NBC) structure. This identifier is based on a discrete-time high order neural network trained on-line with an extended Kalman filter (EKF)-based algorithm. Then, a sliding mode control is used to achieve the purpose of tracking velocity and magnitude flux. The neural control performance is illustrated via simulations.","Title: Real-Time Inverse Optimal Control For A Planar Robot\nYear: 2010\n\nAbstract: This paper presents an inverse optimal neural controller, which is constituted by the combination of two techniques: a) inverse optimal control to avoid solving the Hamilton Jacobi Bellman (HJB) equation associated to nonlinear system optimal control, and b) an on-line neural identifier, which uses a recurrent neural network, trained with the extended Kalman filter (EKF), in order to build a model of an assumed unknown nonlinear system. The applicability of the proposed approach is illustrated by real-time control of a planar robot.","Title: Real-Time Output Trajectory Tracking using a Discrete Neural Backstepping Controller.\nYear: 2008\n\nAbstract: This paper deals with adaptive tracking for discrete-time MIMO nonlinear systems in presence of bounded disturbances. A high order neural network (HONN) structure is used to approximate a control law designed by the backstepping technique, applied to a block strict feedback form (BSFF). The learning algorithrm for the HONN is based on an Extended Kalman Filter (EKF). This paper also includes the respective stability analysis, using the Lyapunov approach. The proposed scheme is implemented in real-time to control a three phase induction motor, as to track a time-variying speed reference and a constant flux magnitude reference.","Title: A fixed time convergent dynamical system to solve linear programming\nYear: 2014\n\nAbstract: The aim of this paper is to present a new dynamical system which solves linear programming. Its design is considered as a sliding mode control problem, where its structure is based on the Karush-Kuhn-Tucker optimality conditions, and its multipliers are the control inputs to be implemented by using fixed time stabilizing terms with vectorial structure, based on the unit control, instead of common terms used in other approaches. Thus, the main features of the proposed system are the fixed convergence time to the programming solution and the fixed parameters number despite of the optimization problem dimension. That is, there is a time independent to the initial conditions in which the system converges to the solution and, the proposed structure can be easily scaled from a small to a higher dimension problem. The applicability of the proposed scheme is tested on real-time optimization of an electrical Microgrid prototype.","Title: A recurrent neural network for real time electrical microgrid prototype optimization\nYear: 2014\n\nAbstract: The aim of this paper is to present a new class of recurrent neural networks, which solve linear programming. It is considered as a sliding mode control problem, where the network structure is based on the Karush-Kuhn-Tucker (KKT) optimality conditions, and the KKT multipliers are the control inputs to be implemented with fixed time stabilizing terms, instead of common used activation functions. Thus, the main feature of the proposed network is its fixed convergence time to the solution, which means, there it is a time independent to the initial conditions in which the network converges to the optimization solution. The applicability of the proposed scheme is tested on real-time optimization of an electrical microgrid prototype.","Title: High order integral nested sliding mode control\nYear: 2014\n\nAbstract: This paper exposes a controller for the nonlinear systems in the block controllable form. This proposal guarantees exponential exact tracking in the presence of unknown matched and unmatched disturbances by means a combination of the block control method and integral terms designed with high-order sliding-modes algorithms. Both, matched and unmatched, disturbances are compensated by those integral continuous terms and the tracking is achieved with the design of a nominal control law.","Title: Sliding control variation for ball-plate system stabilization\nYear: 2014\n\nAbstract: The fundamental purpose of this paper is to present a stabilizing control for rolling ball (ball-plate system), which is developed using the framework of the theory of high order control variation in combination with integrator backstepping, and sliding mode control. Under this scheme, control laws are proposed to solve the problem of stabilization for this system.","Title: A fixed-time second order sliding mode observer for a class of nonlinear systems\nYear: 2014\n\nAbstract: This paper presents a second order fixed time sliding mode observer based on an extension of the super-twisting algorithm. This observer can be applied to a class of nonlinear system with a block-wise representation. The block structure provides a straightforward form to the application of the proposed second order sliding mode algorithm, yielding to finite-time convergence with a settling time independent to the system initial conditions. Finally, as numerical simulation example, the case of a linear induction motor is studied, exposing the efficiency and feasibility of the proposal.","Title: A discontinuous mean-square filter for stochastic differential systems\nYear: 2012\n\nAbstract: This paper presents a mean-square filter for Lipschitz stochastic differential systems, which contains a discontinuous innovations term in the form of unit control. It is demonstrated that the designed filter yields the ultimate boundedness of the estimation error variance. The developed algorithm is then applied to a third degree polynomial system and compared to the extended Kalman-Bucy filter. The simulation results show the faster convergence and better approximation properties of the proposed filter.","Title: An equivalent control based sliding mode observer using high order uniform robust sliding operators\nYear: 2012\n\nAbstract: In this paper a sliding-mode observer based on the equivalent control method for discontinuous functions for a class of non-linear systems is proposed. The observer structure and its existence conditions are presented. Besides, a class of high order sliding operators with the properties of uniform (w.r.t. initial conditions) finite time convergence and with reduction of chattering effect are exposed. The use of these operators in the observer design allows the calculation of the equivalent control and the observer convergence uniformly in finite time. A simulation example is presented to illustrate the proposed method.","Title: Optimal filtering for uncertain linear stochastic systems\nYear: 2009\n\nAbstract: This paper presents the optimal joint filtering and parameter identification problem for uncertain linear stochastic systems with unknown parameters in both state and observation equations, where the unknown parameters are considered Wiener processes. The original problem is reduced to the filtering problem for an extended state vector that incorporates parameters as additional states. The resulting filtering system is polynomial in state and linear in observations. The obtained optimal filter for the extended state vector also serves as the optimal identifier for the unknown parameters. Performance of the designed optimal state filter and parameter identifier is verified for both, positive and negative, parameter values.","Title: HOSM block control of SPIM\nYear: 2012\n\nAbstract: In the continuation of authors' studies on control and estimation methods for Single-Phase Induction Motor (SPIM), a new observer-based controller using High Order Sliding Mode (HOSM) algorithms is proposed here. This observer-controller scheme only uses measurements of the rotor speed and stator currents. The complete scheme is robust to uncertainties in the rotor resistance, and a bounded time-varying load torque.","Title: ℋ∞ Filtering for Polynomial Systems Over Switching Delayed Observations\nYear: 2013\n\nAbstract: The problem of designing an ℋ∞ filter for a class of nonlinear polynomial systems over linear delayed observations is presented in this paper. Output observations may or may not experience sensor delay due to a random variable which is modelled as a Bernoulli distributed one which takes the quantities of zero and one. A closed form of this filter is obtained by expressing the conditional expectations of polynomial terms as functions of the estimate and the gain matrix. As a particular case, a third degree polynomial is considered to obtain the finite-dimensional filtering equations. Simulations results applied to an induction motor show the effectiveness of proposed filter.","Title: Discontinuous output regulation for a DC-to-AC boost converter\nYear: 2012\n\nAbstract: This work presents the design of a discontinuous output regulator for a DC-to-AC power conversion problem on a boost converter, for asymptotic output trajectory tracking. The steady state reference signal for the inductor current is proposed as an approximated solution of the well known Francis-Isidori-Byrnes equations, then, by a proper selection of a sliding surface one can stabilize the non-minimum phase inductor current tracking variable. Finally, by means of numerical simulations one verifies the performance of the proposed control strategy.","Title: Super-twisting sliding mode control of synchronous motors\nYear: 2012\n\nAbstract: Based on a nonlinear complete order model of a synchronous motor, a robust second order sliding mode observer based control scheme is proposed. To design this scheme a super-twisting sliding mode control and adaptive backstepping techniques, are implemented. A nonlinear reduced order observer is designed for estimation of the rotor fluxes. The simulation result shows the effectiveness of the proposed controller. Copyright © 2012 CCE.","Title: Discrete-time sliding mode regulator for nonminimum phase systems\nYear: 2012\n\nAbstract: This work presents an approach to solve the output regulation problem for a class of nonlinear discrete-time nonminimum phase perturbed systems. Based on feedback linearization block control technique and discrete-time sliding mode (SM) control, we propose a sliding manifold on which the zero dynamics are stabilized. To enforce the robust SM motion on the designed manifold, a discrete-time super-twisting SM algorithm is implemented. The effectiveness of the proposed methodology is verified via the design of a position tracking controller for an under-actuated robotic system, the Pendubot.","Title: Optimal Controller for Stochastic Polynomial Systems with State-Dependent Polynomial Input\nYear: 2011\n\nAbstract: AbstractThis paper presents an optimal quadratic-Gaussian controller for stochastic polynomial systems with a state-dependent polynomial control input and a quadratic criterion over linear observations. The optimal closed-form controller equations are obtained using the separation principle, whose applicability to the considered problem is substantiated. As an intermediate result, the paper gives a closed-form solution of the optimal regulator (control) problem for polynomial systems with a state-dependent polynomial control input and a quadratic criterion. Performance of the obtained optimal controller is verified in an illustrative example against a conventional linear-quadratic-Gaussian (LQG) controller that is optimal for linearized systems. Simulation graphs demonstrating overall performance and computational accuracy of the designed optimal controller are included.","Title: Optimal Filtering for Polynomial Systems over Switching Delayed Observations\nYear: 2013\n\nAbstract: The problem of designing an optimal filter for a class of nonlinear systems confused with white Gaussian noise over linear time delayed observations is presented in this paper. Output observations may or may not experience sensor delays due to a random variable which is modeled as a Bernoulli distributed one that takes the quantities of zero and one. A closed form of this filter is obtained by expressing the conditional expectation of polynomial terms as a function of its expectation and covariance matrix. As a particular case, finite-dimensional filtering equations are obtained for a third degree polynomial function. Numerical simulations are performed on an induction motor in order to show the effectiveness of proposed filter.","Title: Modeling And Pose Control Of Robotic Manipulators And Legs Using Conformal Geometric Algebra\nYear: 2015\n\nAbstract: Controlling the pose of a manipulator involves finding the correct configuration of the robot's elements to move the end effector to a desired position and orientation. In order to find the geometric relationships between the elements of a robot manipulator, it is necessary to define the kinematics of the robot. We present a synthesis of the kinematical model of the pose for this type of robot using the conformal geometric algebra framework. In addition, two controllers are developed, one for the position tracking problem and another for the orientation tracking problem, both using an error feedback controller. The stability analysis is carried out for both controllers, and their application to a 6-DOF serial manipulator and the legs of a biped robot are presented. By proposing the error feedback and Lyapunov functions in terms of geometric algebra, we are opening a new venue of research in control of manipulators and robot legs that involves the use of geometric primitives, such as lines, circles, planes, spheres.","Title: A Robust Automotive Controller Design\nYear: 2008\n\nAbstract: This paper describes an application of the block control and sliding mode control techniques to form a stabilising controller for an internal combustion engine with a throttle driven by a DC motor. This combined approach enables the inherent non-linearities of the engine to be compensated and high-level external disturbances to be rejected. Also, the control system utilises a generalised observer for non-linear plants, which incorporates a high gain technique.","Title: Distributed super twisting controller for multiple quadrotors\nYear: 2015\n\nAbstract: The distributed cooperative tracking control problem for a group of quadrotors in a three-dimensional space is addressed in this paper. The controller design is divided in two stages. In the first stage, local distributed controllers for the translational dynamics are designed, forcing the quadrotor to asymptotically track the desired trajectory with the required separation and therefore a specific formation. In the second stage, the rotational dynamics is asymptotically stabilized. The controllers implemented in both stages are based in a combination of block control technique and the super twisting control algorithm which ensures robustness with respect to external disturbances and parameter uncertainties. Moreover, a first order exact differentiator is used to estimate the virtual control inputs, simplifying the control law design. The stability proof of the complete closed-loop system is shown to be asymptotically stable. Finally, numerical simulations are carried out to show that theoretical conclusions are effective.","Title: Quaternion-based trajectory tracking robust control for a quadrotor\nYear: 2015\n\nAbstract: This paper tackles the problem of trajectory tracking problem of a quadrotor. A sliding mode block control algorithm based on the super twisting algorithm using unit quaternion feedback is applied to the dynamic model of the quadrotor helicopter. The derivative of the virtual control inputs are obtained using exact first order differentiator, and the aerodynamic forces and moments acting on the quadrotor are estimated via a sliding mode observer to ensure robustness against external disturbances and model uncertainties. Moreover, a comparison of the computational complexity of the controller between the proposed control scheme and the one considering Euler angles is carried out. The stability proof showing the asymptotic stability of the proposed control scheme is presented. Simulations show the good performance of the proposed control scheme.","Title: Real-time neural inverse optimal control for position trajectory tracking of an induction motor\nYear: 2015\n\nAbstract: This paper describes a neural inverse optimal control approach for a three-phase induction motor position trajectory and flux magnitude tracking. A recurrent high order neural network (RHONN) is used to identify the plant model, trained with an Extended Kalman Filter (EKF) algorithm; the control law minimize a cost functional avoiding to solve the Hamilton Jacobi Bellman (HBJ) equation. The applicability of the approach is illustrated via experimental results. The proposed scheme allows the easy integration of this kind of motors into a system of systems configuration.","Title: Predefined-time stability of dynamical systems with sliding modes\nYear: 2015\n\nAbstract: This paper introduces a class of fixed-time stable dynamical systems with settling time as a explicit parameter, namely the inverse the gain. Those systems are defined as predefined-timed stable dynamical systems. Continuous and discontinuous are cases are presented. A detailed Lyapunov characterization of this class of systems is also shown. Finally, the application to the design of a class of first order sliding mode controllers is exposed.","Title: Control for partial state constraining of nonlinear systems in regular form\nYear: 2015\n\nAbstract: A method for enforcing non-symmetrical constraints on the first block of the state of a nonlinear system in regular form is presented. Two cases are considered: the first block of the system stays within some arbitrarily selected limits, a) displaying a conservative behavior, shown as limit-cycle plots, and b) some sort of damping is introduced, forcing the second block of the state to tend to zero as time increases. Both cases emerged from a convenient selection of the stabilizing function in the control law. Simulation results present the closed-loop behavior of an open-loop unstable MIMO system, showing the effectiveness of the approach.","Title: Integral nested ℋ∞ block control\nYear: 2015\n\nAbstract: This paper exposes a robust controller for the tracking of nonlinear systems in the block controllable form. This proposal is based on the block control method and integral terms designed with ℋ\n<sub xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">∞</sub>\n algorithms. The controller guarantees high performance the presence of uncertainty. The disturbances are compensated by those integral continuous terms and the tracking is achieved with the design of a nominal control law.","Title: Enhanced Discrete-Time Modeling via Variational Integrators and Digital Controller Design for Ground Vehicles.\nYear: 2016\n\nAbstract: In this paper, a discrete-time controller for ground vehicles is proposed, for the tracking of desired references for lateral and yaw velocities. A novel discrete-time model of the vehicle dynamics is considered, which is obtained by means of a variational integrator known as symplectic Euler. The control inputs are given by active front steering and rear torque vectoring. Simulations have been ca...","Title: Predefined-time tracking of a class of mechanical systems\nYear: 2016\n\nAbstract: In this paper the problem of predefined-time exact tracking of fully actuated and unperturbed mechanical systems is solved by means of a continuous controller. It is assumed the availability of the state and the desired trajectory as well as its two first derivatives. This is accomplished introducing the idea of second-order predefined-time stable systems, which is based on the nested application of the first-order predefined-time stabilizing function. As an example, the proposed solution is applied over a two-link planar manipulator and numerical simulations are conducted to show its performance.","Title: Arbitrary references output tracking in linear systems\nYear: 2016\n\nAbstract: The problem of arbitrary references output tracking for linear systems is solved through an approximate required steady state. This estimate is obtained using a dynamic estimator based on a state observer. The state-feedback control law is proposed using this estimate and the closed-loop system is then analyzed. The simulations show the effectiveness of the proposed technique.","Title: Dynamic Predictor For Linear Time-Delay Systems With Disturbances\nYear: 2016\n\nAbstract: A dynamic predictor is proposed for linear time invariant disturbed systems with delay in the control input. The proposed dynamic predictor has a structure similar to the Luenberger observer. It outperforms the current predictors because it does not need the full state, as long as the plant is observable, it only requires the output from the system. The controller is proposed making use of the state vector from the dynamic predictor. The full control scheme including predictor and control law is applied in simulation. The advantages of the proposed predictor include robustness against external disturbances, and also the fact that the controllers developed making use of predictor are capable of tracking time-variable signals.","Title: Robust Observer-Based Sliding Mode Controller For Vehicles With Roll Dynamics\nYear: 2016\n\nAbstract: In this work, an observer-based sliding mode (SM) control scheme for a vehicle considering roll dynamics, is presented. The proposal considers a nonlinear higher order sliding mode reduced observer for the lateral velocity, roll angle and roll velocity. Based on this observer, the controller is designed for the reference tracking of the lateral and yaw velocities of the vehicle, using adaptive super-twisting algorithm. In order to demonstrate the stability of the closed-loop system, Lyapunov functions for each of the algorithms are used. The proposed control scheme has finite-time convergence and the robustness in presence of plant parameters variations. This is demonstrated through the performed simulation.","Title: On The Discrete-Time Modeling And Control Of Synchronous Generators By Means Of Variational Integrators And Sliding Modes\nYear: 2016\n\nAbstract: This work deals with the discrete-time modeling and control design for synchronous generators by means of a variational integrator and sliding modes. First, a continuous non-conservative Lagrangian is formulated for the plant, then, the respective discrete Lagrangian is determined. Based on this, discrete-time rules are derived for synchronous generators. Second, a sliding-mode controller is proposed for the velocity stabilization of the system, and a nonlinear observer is designed for the non-measurable states. Simulations show the good performance of the synchronous generator when closed-loop with the novel discrete-time controller.","Title: Causal Output Tracking For Discrete-Time Linear Systems\nYear: 2017\n\nAbstract: A new approach to the solution of the causal output tracking problem for discrete-time linear systems is presented. A dynamic steady state estimator is proposed in the form of a state observer taking as output, the reference to be tracked. Once the steady state has been estimated, the output tracking problem is converted into a stabilization problem and a control law is designed such that the output tracking is achieved while the system dynamics keep stable. The main applications of the techniques presented in this work are the non-minimum phase systems, where classical input-output techniques cannot be applied.","Title: Robust sensorless observer-based adaptive sliding modes control of synchronous motors.\nYear: 2018\n\nAbstract: •The full order model of the synchronous motor is considered with matched and unmatched perturbations.•Adaptive backstepping and the super-twisting sliding mode control techniques are used to ensure robustness of the system.•A generalized super-twisting algorithm is proposed to design a robust observer for a 3rd order nonlinear system.•The observer proposed is used to estimate the rotor fluxes in presence of plant parameter variations.•The rotor speed estimator proposed is used in the design of the controller and flux observer, avoiding the use of sensor.","Title: Sliding mode predictive control of linear uncertain systems with delays.\nYear: 2018\n\nAbstract: In this paper, a continuous control strategy for robust stabilization of a class of uncertain multivariable linear systems with delays in both the state and control variables is proposed. A predictor is designed to compensate the delay effect in the control input, and then an integral sliding mode control technique along with super-twisting algorithm is applied to compensate partially the effect of the perturbation term. Finally, a nominal delay-free part of the control input is designed to stabilize the sliding mode dynamics. The proposed control scheme is extended to the class of systems modeled in Regular form. For this class of perturbed systems with delay in the state, a transformation to the systems with the delay-free state is proposed. The stability conditions of the closed-loop uncertain system are derived, and the results obtained in this work are compared against previous works. To show the effectiveness of the proposed method, simulation results are presented.","Title: Higher Order Sliding Mode And Adaptive Backstepping Controllers For A Full-Car Model With Active Suspension\nYear: 2018\n\nAbstract: The robust stabilization of the vertical position, pitch and roll dynamics of the center of gravity for a vehicle with active suspension is addressed in this work. Three different higher order sliding mode (HOSM) control algorithms and an adaptive backstepping controller (ABC) are designed to fulfill the control objective. First, two second order sliding mode controllers, the quasi-continuous and the nested, are designed to stabilize the 2-relative degree vertical position. Then, a super-twisting (ST) algorithm is designed, by means of a 1-relative degree sliding function, obtaining asymptotic stability. For the ABC, adaptive parameters dynamics are proposed, as well as the control law that adds desired linear dynamics. Finally, a performance comparison between the proposed controllers is presented based on simulations results.","Title: A second order predefined-time control algorithm\nYear: 2017\n\nAbstract: The predefined-time stabilization of second-order systems, i.e. the fixed-time stabilization with settling time as a function of the controller parameters, is revisited in this paper. The proposed controller is a time-based switched controller which first drive the system trajectories to a linear manifold in predefined time and then uses a nested second-order controller. The application of the results is demonstrated for the trajectory tracking control in fully actuated mechanical systems. An illustrative example of the control of a two-link planar manipulator with predefined-time convergence shows the effectiveness of the proposed algorithm.","Title: Semi-global predefined-time stable systems\nYear: 2017\n\nAbstract: A Lyapunov-based construction of a predefined-time stabilizing function (a function that stabilizes a system in fixed-time with settling time as function of the controller parameters) for scalar systems is considered in this paper. The constructed function involves the inverse incomplete gamma function, causing this function to be semi-global, i.e., the domain of definition of the function can be made as large as wanted with an appropriate parameter selection. Finally, the constructed function is used to design predefined-time stabilizing controllers which are robust against vanishing and non-vanishing perturbations.","Title: Semi-Global Predefined-Time Stable Vector Systems\nYear: 2018\n\nAbstract: In this paper, we expose a control function which allows the semi-global predefined-time stabilization of first-order vector systems. The predefined-time stability is a stronger class of finite-time stability that has as main advantage the settling time as a tunable parameter of the proposed function. To design that stabilizing function, we use the unit control principle jointly to the inverse incomplete gamma function. For the resulting expression, the domain of definition the inverse incomplete gamma function can be made as large as wanted with an appropriate parameter selection, and, as consequence, the attraction domain of the systems. Therefore, we say that the system exhibits semi-global predefined-time stability. As an essential feature, the parameter which defines the settling time bound and those that tune the attraction domain are independent of each other. Finally, the constructed function is used to design predefined-time stabilizing controllers which are robust against vanishing and non-vanishing perturbations.","Title: Robust Discrete-Time Sliding Mode Control With Arbitrarily Small Boundary Layer\nYear: 2018\n\nAbstract: The robust discrete-time sliding mode control problem is addressed. A novel approach to estimate the matched disturbance is presented. This approach is based on the approximation of a discrete-time function actual value by using various previous steps of itself. The matched disturbance previous step is calculated from the system model and then, using the mentioned approach, its actual value is estimated with an error of exponential order w.r.t. the sampling time and the used number of previous steps. Simulations show the effectiveness of the proposed approach to estimate a discrete time function and to robustly stabilize a dynamic system.","Title: Robust Discrete-Time Sliding Mode Controller For A Class Of Nonlinear Perturbed Systems\nYear: 2018\n\nAbstract: The output tracking problem for a class of nonlinear systems presented in Nonlinear Block Controllable (NBC) form is addressed. Both matched and unmatched perturbations are considered. First, the Block Control iterative feedback linearization technique combined with a perturbation estimation are employed to design a sliding manifold. With the perturbation estimation, the effect of unmatched perturbation is mitigated. Then, a discrete-time sliding mode controller is synthesized such that the system state is driven toward a vicinity of the designed sliding manifold and stays there for all sampled time instants, avoiding chattering and reducing the matched perturbation effect. The effectiveness of the proposed methodology is confirmed by simulation.","Title: Robust Control for Perturbed Linear Systems with Time-Varying Delay via Sliding Mode Control\nYear: 2018\n\nAbstract: In the present paper, a robust predictor-based sliding mode control for perturbed time-varying delay systems is proposed. A predictor is designed to compensate the time-varying delay effect in control input, and then an integral sliding mode control technique along with super-twisting algorithm is applied to compensate partially the effect of the perturbation term. Finally, a nominal delay-free part of the control input is designed to stabilize the sliding mode dynamics. Conditions for the stability of the closed-loop perturbed system are derived. Simulation result is presented.","Title: Adaptive back-stepping speed Tracking controller for a Brush-less DC motor with a SM observer for back EMFs\nYear: 2018\n\nAbstract: This paper proposes a robust speed tracking controller design for Brush-less DC Motor (BLDC) under parametric uncertainness based on the combination of a sliding modes (SM) observer and adaptive back-stepping control techniques. The adaptive controller developed in this work in combination with the SM observer allows also a reduction on the torque rippling since electrical torque is controlled in an indirect way based on the obtained estimates of the back EMFs. The effectiveness of the proposed controller is demonstrated by simulations with MATLAB/SIMULINK.","Title: Sliding mode identification and control for a class of nonlinear systems\nYear: 2018\n\nAbstract: In this work, the problem of designing a control scheme capable of controlling dynamical systems with unknown time-varying parameters and disturbances, is proposed. In contrast with other works, based on two techniques: adaptive backstepping and sliding modes, an improved method that guaranties the output asymptotic tracking of a smooth reference signal, the stability of the closed-loop system and the identification errors boundedness, is developed. By means of sliding mode observers, the adaptation errors required information is extracted and injected to adaptive laws. The behavior of the proposed control scheme is analyzed by Lyapunov method. The performance of the proposed system is verified with an academic example.","Title: Adaptive Discrete Time Sliding Mode Control for a Class of Nonlinear Systems\nYear: 2018\n\nAbstract: In this paper, a robust sliding mode adaptive tracking state feedback control scheme is designed for a class of nonlinear discrete time (DT) perturbed systems in parametric strict feedback form. The design procedure is based on the adaptive backstepping methodology proposed for continuous time systems, using step-by-step a Lyapunov function. First, an adaptive sliding manifold is formulated, and then, a DT sliding mode controller is proposed to achieve the designed manifold be attractive in presence of bounded matched uncertainty. To show the effectiveness of the proposed method, simulation results are performed.","Title: Output-Feedback Discrete-Time Sliding Mode Control via Disturbance Estimation\nYear: 2018\n\nAbstract: In this paper, a novel technique to achieve robustness in sliding mode control for discrete-time systems by measuring only its output is presented. The proposed technique is based on the approximation of an arbitrary discrete-time function by using various of its previous steps. Using this approach, the system disturbance is approximated by a dynamical system and as result, the plant model is extended. Based on the extended model, a state observer is designed to estimate both the system state and the disturbance. Using the observer state, a discrete-time sliding mode controller is then designed to robustly stabilize the perturbed system. The simulation results show the effectiveness of the proposed technique.","Title: Second Order Sliding Mode Control of a STATCOM with Saturated Inputs\nYear: 2018\n\nAbstract: This paper presents a robust controller for a STATCOM device with saturated inputs. As the primary assumption, the proposed design considers the presence of unknown but bounded external perturbations and parametric variations. This proposal has a cascade structure, where a saturated super twisting control algorithm closes the currents control loop, and a high-gain proportional-integral (PI) algorithm ensures the voltage regulation. Thus, the exposed scheme provides an adequate performance of the STATCOM, considering the saturation of the inputs with the anti-windup feature. Posteriorly, a proper stability analysis presents the conditions for the appropriate operation of the closed-loop system in saturation and non-saturation regimes. Numerical simulations are also included to show the performance of the proposed controller.","Title: Predefined-Time Stabilization Of High Order Systems\nYear: 2017\n\nAbstract: The aim of this paper is to introduce a controller that stabilizes a class of arbitrary order systems in predefined-time. The proposed controller is designed with basis on the block-control principle yielding in a nested structure similar to high order sliding mode algorithms and terminal sliding mode algorithms. For this case, it is assumed the availability of the state and the absence of perturbations. Numerical simulations expose the desired performance of this controller.","Title: Passivity Sliding Mode Control of Large-Scale Power Systems\nYear: 2019\n\nAbstract: In this brief, a sliding mode (SM) decentralized excitation controller for multimachine power systems is designed to stabilize the speed and regulate the terminal voltage of each generator in the system. To achieve the speed stabilization, a sliding manifold is designed using the passivity technique. An integral SM controller is proposed to reject internal and external perturbations, taking advant...","Title: Approximate Causal Output Tracking For Linear Perturbed Systems Via Sliding Mode Control\nYear: 2020\n\nAbstract: A new approach to the solution of what is termed causal output tracking problem for linear time-invariant systems in the presence of both matched and unmatched unmodelled disturbances is presented. The proposed solution is addressed through the design of a dynamic steady state estimator based on the desired system structure and an observer, considering the reference as the system output. The unmatched disturbance is estimated and used in the steady state estimator to achieve invariance, while the matched disturbance robustness is provided via sliding mode control using the super-twisting algorithm. A useful application of the presented techniques is output tracking for non-minimum phase systems; thus, the performed simulation results confirm the effectiveness of the proposed control scheme for this kind of systems.","Title: Observer based sliding mode controller for vehicles with roll dynamics.\nYear: 2019\n\nAbstract: In this work, considering the roll dynamics and actuator dynamics, an observer-based control scheme for a vehicle is proposed. The proposal considers a nonlinear higher order sliding mode observer to estimate unmeasurable lateral velocity, roll angle and roll velocity. Using the observer information, a controller based on block control with sliding mode technique is designed for the reference trajectory tracking of the lateral and yaw velocities of the vehicle. The stability of the complete closed-loop system including zero dynamics is analyzed. The effectiveness of the proposed scheme is demonstrated through CarSim simulations.","Title: Predefined-time robust contour tracking of robotic manipulators\nYear: 2019\n\nAbstract: Robotic tasks are primarily defined to comply to spatial and time constraints, and in several of such tasks, it is required to track a nominal contour depending just on spatial coordinates, without involving the time variable in the definition of such a contour, in contrast to conventional trajectory tracking schemes. With the aim to satisfy the salient requirements of these robotic tasks, this paper presents an integrated robust controller. The contour is encoded by using a robust velocity field, such that, any particle immerse in such field converges in predefined-time to a predefined vicinity of the given contour. Then, a robust controller is proposed to enforce the velocity field, in turns, in predefined-time. In this sense, a predefined-time-precision contour tracking is assured, by considering a robust controller that does not assume the exact knowledge of the robot parameters for its real-time implementation. Numerical simulations are conducted to highlight the reliability of the proposed scheme.","Title: Synchronization of complex networks in discrete-time for trajectory tracking using neural sliding modes with pinning control\nYear: 2021\n\nAbstract: This paper considers a novel solution for synchronization of complex networks in discrete-time, based on Lyapunov theory and passivity degree to define a reliable interval of coupling strength required to achieve synchronization in a directed complex network with non-identical nodes. By means of a discrete-time recurrent high-order neural network trained with an extended Kalman filter algorithm, the unknows pinned nodes are identified. Later, using an adaptive approach of sliding modes in discrete-time to complex networks, a selected pin node is locally controlled to guarantee trajectory tracking. Once the pin node is controlled, network synchronization is achieved, all its nodes follow selected trajectories. The results are illustrated via simulations considering that each node corresponds to a discretized nonlinear chaotic attractor.","Title: Adaptive Control Methodology for a Class of Nonlinear Systems With Speed Tracking Implementation for a BLDC Motor\nYear: 2022\n\nAbstract: This work proposesan adaptive control methodology for a class of time-varying uncertain systems. The goal of this work is to address problems in which the model is known, but the parameters involved are either unknown, time-varying, or the system is being affected by disturbances. The proposed methodology relies on Lyapunov control concepts and the bound of the time-varying elements (only the boundedness feature is known) to extend traditional adaptive control techniques restricted to linear constant parameterization. In order to show the efficiency of this methodology, we present a comparison simulation and an implementation for speed tracking in a Brush-Less Direct Current (BLDC) motor under parametric uncertainty."],"string":"[\n \"Title: Hybrid Control of Induction Motors via Sampled Closed Representations\\nYear: 2008\\n\\nAbstract: In this paper, a controller for induction motors is proposed. A continuous feedback is first applied to obtain a discrete-time model in closed form. Then, on the basis of these exact sampled dynamics, a discrete-time controller ensuring speed and flux modulus reference tracking is determined, making use of the sliding mode technique. The resulting controller is hence hybrid, in the sense that it contains both continuous and discrete-time terms. It is shown how to implement such a hybrid controller using the so-called exponential holder, which is the only device to be implemented analogically, together with an analog integrator. Moreover, a discrete-time reduced-order observer is designed for rotor fluxes and load torque estimation. The performance of the proposed controller is finally studied by simulation and experimental tests.\",\n \"Title: Discrete-Time Nonlinear Systems Inverse Optimal Control: A Control Lyapunov Function Approach\\nYear: 2011\\n\\nAbstract: This paper presents an inverse optimal control approach for exponential stabilization of discrete-time non-linear systems, avoiding to solve the associated Hamilton-Jacobi-Bellman (HJB) equation, and minimizing a meaningful cost function. This stabilizing optimal controller is based on a discrete-time control Lyapunov function. The applicability of the proposed approach is illustrated via simulations by stabilization of an example.\",\n \"Title: Observer based controller for internal combustion engine\\nYear: 2013\\n\\nAbstract: In this work, an observer-based controller for an internal combustion engine is presented. At first, an algorithm for the estimation of an unknown function of the internal combustion engine is designed, since it is very difficult to obtain direct measurements of this variable. This estimator is based on sliding mode algorithms, providing a finite time and robust estimation, using only measurements from the velocity of the engine. On the other hand, with the measured velocity and the estimates of the other variables, a robust controller is synthesized for the engine. In order to considerate the actuator dynamics, the proposed control scheme is based on the master-slave structure, regarding the controller for the actuator as the slave one. For this scheme, the backstepping algorithm is used to design the master controller. Then, the calculated control input signal to the engine is used as a reference for the throttle actuator which is driven by a direct current (DC) motor. Thus, a high-order sliding mode controller is applied to the actuator in order to track the control input signal and reject perturbations, as the applied mechanical load, regulating the velocity of the combustion engine. Numerical simulations show the efficient performance of this proposal.\",\n \"Title: Neural identification and control for linear induction motors\\nYear: 2005\\n\\nAbstract: We propose a new adaptive control scheme, composed of a neural identifier and a nonlinear controller and applied it to a linear induction motor (LIM). In order to compare the performance of LIM, we use α - β and d - q models. A neural identifier of triangular form is proposed for both models as a nonlinear block controllable form (NBC). Then, a reduced order observer is designed in order to estimate no measured variables. Learning law for neural network weights ensure that the identification error converges to zero exponentially. Sliding mode control is developed to track velocity and flux magnitude. Simulations are presented to compare the behaviour of both models of LIM and the applicability of the proposed identification and control scheme.\",\n \"Title: Mean-square filtering for uncertain linear stochastic systems\\nYear: 2010\\n\\nAbstract: This paper presents the mean-square joint filtering and parameter identification problem for uncertain linear stochastic systems with unknown parameters in both, state and observation, equations, where the unknown parameters are considered Wiener processes. The original problem is reduced to the filtering problem for an extended state vector that incorporates parameters as additional states. The resulting filtering system is polynomial in state and linear in observations. The obtained mean-square filter for the extended state vector also serves as the mean-square identifier for the unknown parameters. A simulation example is included to show convergence of the designed mean-square state filter and parameter identifier for both, positive and negative, parameter values.\",\n \"Title: Recurrent neural networks with fixed time convergence for linear and quadratic programming.\\nYear: 2013\\n\\nAbstract: In this paper, a new class of recurrent neural networks which solve linear and quadratic programs are presented. Their design is considered as a sliding mode control problem, where the network structure is based on the KarushKuhn-Tucker (KKT) optimality conditions with the KKT multipliers considered as control inputs to be implemented with fixed time stabilizing terms, instead of common used activation functions. Thus, the main feature of the proposed network is its fixed convergence time to the solution. That means, there is time independent to the initial conditions in which the network converges to the optimization solution. Simulations show the feasibility of the current approach.\",\n \"Title: Mean-square state and noise intensity estimation for uncertain linear systems\\nYear: 2010\\n\\nAbstract: This paper presents the mean-square joint state and diffusion coefficient (noise intensity) estimator for linear stochastic systems with unknown noise intensity over linear observations, where unknown parameters are considered Wiener processes. The original problem is reduced to the filtering problem for an extended state vector that incorporates parameters as additional states. Since the noise intensities cannot be observable in the original linear system, the new quadratic vector variable formed by the diagonal of the matrix square of the system state is introduced. The obtained mean-square filter for the extended state vector also serves as the optimal identifier for the unknown parameters. Performance of the designed mean-square state filter and parameter identifier is verified in an illustrative example.\",\n \"Title: Real-time torque control using discrete-time recurrent high-order neural networks.\\nYear: 2013\\n\\nAbstract: This paper presents a discrete-time direct current (DC) motor torque tracking controller, based on a recurrent high-order neural network to identify the plant model. In order to train the neural identifier, the extended Kalman filter (EKF) based training algorithm is used. The neural identifier is in series-parallel configuration that constitutes a well approximation method of the real plant by the neural identifier. Using the neural identifier structure that is in the nonlinear controllable form, the block control (BC) combined with sliding modes (SM) control techniques in discrete-time are applied. The BC technique is used to design a nonlinear sliding manifold such that the resulting sliding mode dynamics are described by a desired linear system. For the SM control technique, the equivalent control law is used in order to the plant output tracks a reference signal. For reducing the effect of unknown terms, it is proposed a specific desired dynamics for the sliding variables. The control problem is solved by the indirect approach, where an appropriate neural network (NN) identification model is selected; the NN parameters (synaptic weights) are adjusted according to a specific adaptive law (EKF), such that the response of the NN identifier approximates the response of the real plant for the same input. Then, based on the designed NN identifier a stabilizing or reference tracking controller is proposed (BC combined with SM). The proposed neural identifier and control applicability are illustrated by torque trajectory tracking for a DC motor with separate winding excitation via real-time implementation.\",\n \"Title: Neural block control for synchronous generators\\nYear: 2009\\n\\nAbstract: In this paper, a novel approach to control a single generator, connected to an infinite bus, is presented. Modifying published results for nonlinear identification using recurrent neural networks, a block controllable neural identifier is proposed, based on this neural model a control law is derived, which combines sliding modes and block control. The neural identifier and the proposed control law allows to reject external disturbances caused by generator terminal short circuits and mechanical power variations. Applicability of the approach is tested via simulations.\",\n \"Title: Dynamic triangular neural controller for stepper motor trajectory tracking.\\nYear: 2002\\n\\nAbstract: In this paper, we present a novel neural controller for this kind of motor. This controller is developed as follows. Modifying published results for nonlinear identification using dynamic neural networks (NNs), an NN identifier of triangular form is implemented. Then, based on this mode, a control law using sliding modes is derived. This neural identifier and the proposed control law allow trajectory tracking for stepping motors. Applicability of the approach is tested via simulations.\",\n \"Title: Discrete-Time Neural Inverse Optimal Control for Nonlinear Systems via Passivation.\\nYear: 2012\\n\\nAbstract: This paper presents a discrete-time inverse optimal neural controller, which is constituted by combination of two techniques: 1) inverse optimal control to avoid solving the Hamilton-Jacobi-Bellman equation associated with nonlinear system optimal control and 2) on-line neural identification, using a recurrent neural network trained with an extended Kalman filter, in order to build a model of the ...\",\n \"Title: Decentralized neural identification and control for uncertain nonlinear systems: Application to planar robot\\nYear: 2010\\n\\nAbstract: This paper presents a discrete-time decentralized neural identification and control for large-scale uncertain nonlinear systems, which is developed using recurrent high order neural networks (RHONN); the neural network learning algorithm uses an extended Kalman filter (EKF). The discrete-time control law proposed is based on block control and sliding mode techniques. The control algorithm is first simulated, and then implemented in real time for a two degree of freedom (DOF) planar robot.\",\n \"Title: Particle swarm optimization for discrete-time inverse optimal control of a doubly fed induction generator.\\nYear: 2013\\n\\nAbstract: In this paper, the authors propose a particle swarm optimization (PSO) for a discrete-time inverse optimal control scheme of a doubly fed induction generator (DFIG). For the inverse optimal scheme, a control Lyapunov function (CLF) is proposed to obtain an inverse optimal control law in order to achieve trajectory tracking. A posteriori, it is established that this control law minimizes a meaningful cost function. The CLFs depend on matrix selection in order to achieve the control objectives; this matrix is determined by two mechanisms: initially, fixed parameters are proposed for this matrix by a trial-and-error method and then by using the PSO algorithm. The inverse optimal control scheme is illustrated via simulations for the DFIG, including the comparison between both mechanisms.\",\n \"Title: Discrete-time inverse optimal control for nonlinear systems trajectory tracking.\\nYear: 2010\\n\\nAbstract: This paper presents an inverse optimal control approach for output tracking of discrete-time nonlinear systems, avoiding to solve the associated Hamilton-Jacobi-Bellman (HJB) equation, and minimizing a meaningful cost function. This stabilizing optimal controller is based on discrete-time passivity theory. The applicability of the proposed approach is illustrated via simulations by trajectory tracking control of a planar robot.\",\n \"Title: An optimal sliding mode-like regulator for nonlinear polynomial systems\\nYear: 2011\\n\\nAbstract: This article presents the optimal control problem for a nonlinear polynomial system with respect to a Bolza-Meyer criterion with non-quadratic state energy terms. A distinctive feature of the obtained result is that a part of the optimally controlled state trajectory occurs to be in a sliding mode, i.e. represents a motion along a certain manifold. The optimal solution is obtained as a sliding mode-like control, whereas the conventional linear feedback control fails to provide a causal solution. Performance of the obtained optimal controller is verified in the illustrative example against the conventional polynomial-quadratic regulator that is optimal for the quadratic Bolza-Meyer criterion. The simulation results confirm an advantage in favour of the designed sliding mode control.\",\n \"Title: Sliding Mode Block Control Regulation Of The Pendubot\\nYear: 2011\\n\\nAbstract: In this work, the problem of nonlinear regulation of an underactuated system is treated by means of sliding mode continuous control actions combined with block control technique. The sliding mode state feedback output regulator based on the super twisting algorithm, is applied to the Pendubot system. The transformation of the original system to regular form and then block control technique are used to design a sliding manifold with asymptotically stable motion. To verify the effectiveness of the method, simulations are carried out.\",\n \"Title: Copper And Core Loss Minimization For Induction Motors Using High-Order Sliding-Mode Control\\nYear: 2012\\n\\nAbstract: A novel nonlinear affine model for an induction motor with core loss is developed in the well-known (alpha, beta) stationary reference frame, where the core is represented with a resistance in parallel with a magnetization inductance. Then, an optimal rotor flux modulus is calculated such that the power loss due to stator, rotor, and core resistances is minimized, and as a consequence, the motor efficiency is raised; therefore, this flux modulus is forced to be tracked by the induction motor along with a desired rotor velocity by means of a high-order sliding-mode controller, the supertwisting algorithm. Using a novel Lyapunov function, the closed-loop stability of the system is demonstrated. Moreover, a classical sliding-mode observer is designed for the estimation of unmeasurable variables like rotor fluxes and magnetization currents. For the load torque, a Luenberger observer is designed. The performance of the proposed controller is finally studied by simulation and experimental tests. It was observed that the steady-state optimal flux signal corresponds to the load torque profile. This fact suggests that the flux demand is the necessary one to produce the electric torque that can cancel out the load torque.\",\n \"Title: Discrete Integral Sliding Mode Control In Visual Object Tracking\\nYear: 2010\\n\\nAbstract: Object tracking is an important process for many applications in computer vision. This process must be implemented in a discrete manner because the images are available only at certain periods. A discrete integral sliding mode algorithm is proposed to control a stereo vision system and perform the aforementioned second task. The kinematic model of the structure is obtained using geometric algebra. The performance of the controller is compared with proportional-integral-derivative control via simulation. The implementation for a pan tilt unit is presented in real time. The algorithm presents a good and robust performance.\",\n \"Title: Second order sliding mode sensorless torque regulator for induction motor.\\nYear: 2010\\n\\nAbstract: In this paper the torque and the square of the rotor fluxes magnitude using a second order sliding mode (SM) controller for an induction motor, is proposed. The main contribution of the control scheme presented here is the output regulation as well as the asymptotically stabilization of the zero dynamics. The sliding manifold is designed such that the speed zero dynamics become asymptotically stable on this manifold. To enforce the SM motion on the designed manifold a super-twisting SM algorithm is used. This algorithm permits to reduce chattering that is inherent in standard sliding mode control and to improve accuracy. The rotor fluxes and speed are estimated by means of an adaptive observer, whereas the estimate of the load torque is obtained via design of a robust differentiator. The effectiveness of the designed observer-control scheme is confirmed via simulation.\",\n \"Title: Discrete-Time Recurrent Neural Dc Motor Control Using Kalman Learning\\nYear: 2008\\n\\nAbstract: An adaptive tracking controller for a discrete-time direct current (DC) motor model in presence of bounded disturbances is presented. A high order neural network is used to identify the plant model; this network is trained with an extended Kalman filter. Then, the discrete-time block control and sliding modes techniques are used to develop the reference tracking control. This paper includes also the respective stability analysis and a strategy to avoid specific adaptive weights zero-crossing. The scheme is illustrated via simulations for a discrete-time nonlinear model of an electric DC motor.\",\n \"Title: Discrete time nonlinear identification via recurrent high order neural networks for a three phase induction motor\\nYear: 2010\\n\\nAbstract: This paper deals with the problem of discrete-time nonlinear system identification via Recurrent High Order Neural Networks It includes the respective stability analysis on the basis of the Lyapunov approach for the extended Kalman filter (EKF)-based NN training algorithm, which is applied for learning Applicability of the scheme is illustrated via simulation for a discrete-time nonlinear model of an electric induction motor.\",\n \"Title: Real-time Discrete Backstepping Neural Control for Induction Motors.\\nYear: 2011\\n\\nAbstract: This brief focuses on real-time implementation, as applied to a three-phase induction motor, of results already published in 2007. The proposed controller is based on a high-order neural network, trained online using Kalman filter learning, to approximate a control law designed by the backstepping technique.\",\n \"Title: Discrete Time Block Control Of A Double Fed Induction Generator Using Sliding Modes\\nYear: 2009\\n\\nAbstract: In this paper, the authors propose a block control scheme using sliding modes, for a double fed induction generator connected to an infinity bus. To estimate the mechanical torque, a nonlinear observer is utilized; additionally, as a viable alternative, to eliminate the requirement on this estimation, an integrator is included. To reduce the effect of parameter variations, the one step delayed disturbance algorithm is used. The stability of the closed loop system is analyzed by the Lyapunov techniques. The applicability of the proposed scheme is tested via simulations.\",\n \"Title: Robust Trajectory Tracking for an Electrohydraulic Actuator\\nYear: 2009\\n\\nAbstract: Various robust control techniques, such as integral-block, sliding-mode, and H-infinity controls, are combined to design a controller, forcing an electrohydraulic actuator which is driven by a servovalve to track a chaotic reference trajectory. This approach enables one to compensate the inherent nonlinearities of the actuator and reject matched external disturbances and attenuate mismatched exter...\",\n \"Title: Recurrent neural block form control\\nYear: 2003\\n\\nAbstract: By modifying some previously developed results for nonlinear identification and control using recurrent neural networks, the present authors propose a new neural network identifier in block form, and, based on this model, a control law is developed by combining sliding mode and block controls. This neural identifier and control law allow satisfactory trajectory tracking for general nonlinear systems. Applicability of the new design is illustrated, via simulations, for robust tracking control of stepping motors.\",\n \"Title: Real-Time Discrete Nonlinear Identification Via Recurrent High Order Neural Networks\\nYear: 2010\\n\\nAbstract: This paper deals with the discrete-time nonlinear system identification via Recurrent High Order Neural Networks, trained with an extended Kalman filter (EKF) based algorithm. The paper also includes the respective stability analysis on the basis of the Lyapunov approach for the whole scheme. Applicability of the scheme is illustrated via real-time implementation for a three phase induction motor.\",\n \"Title: Super-twisting sensorless control of linear induction motors\\nYear: 2012\\n\\nAbstract: In this work, a sensorless control scheme is presented for linear induction motors. The secondary fluxes are algebraically calculated by first determining the primary fluxes, then, a super-twisitng observer for secondary fluxes is designed in order to retrieve the back-EMF components by means of the equivalent control method. Based on these components, the linear velocity is determined and used in a linear velocity and load force observer, where the estimated variables along with primary current and voltage measurements are used to control the linear induction motor for the tracking of a reference linear velocity signal and a square secondary flux modulus, all by means of a super-twisting controller. Simulations show that the proposed observer based controller scheme performs well when tracking a time varying linear velocity signal.\",\n \"Title: Discrete-Time Backstepping Neural Control for Synchronous Generators.\\nYear: 2007\\n\\nAbstract: This paper deals with adaptive tracking for discrete-time MIMO nonlinear systems in presence of hounded disturbances. In this paper, a high order neural network structure is used to approximate a control law designed by the backstepping technique, applied to a block strict feedback form (BSFF). The learning algorithm for the HONN is based on an Extended Kalman Filter (EKF). This paper also includes the respective stability analysis, using the Lyapunov approach. The viability of the proposed approach is tested via simulations, by its application to synchronous generators control.\",\n \"Title: Decentralized adaptive recurrent neural control structure\\nYear: 2007\\n\\nAbstract: This paper presents a novel decentralized variable structure neural control approach for large-scale uncertain systems, which is developed using recurrent high-order neural networks (RHONN). It is assumed that each subsystem belongs to a class of block-controllable nonlinear systems whose vector fields includes interconnection terms, which are bounded by nonlinear functions. A decentralized RHONN structure and the respective learning law are proposed in order to approximate online the dynamical behavior of each nonlinear subsystem. The control law, which is able to regulate and to track the desired reference signals, is designed using the well-known variable structure theory. The stability of the whole system is analyzed via the Lyapunov methodology. The applicability of the proposed decentralized identification and control algorithm is illustrated via simulations as applied to an interconnected double inverted pendulum.\",\n \"Title: Speed-gradient inverse optimal control for discrete-time nonlinear systems.\\nYear: 2011\\n\\nAbstract: This paper presents a speed-gradient-based inverse optimal control approach for the asymptotic stabilization of discrete-time nonlinear systems. With the solution presented, we avoid to solve the associated Hamilton-Jacobi-Bellman equation, and a meaningful cost function is minimized. The proposed stabilizing optimal controller uses the speed-gradient algorithm and is based on the proposal of what is called a discrete-time control Lyapunov function. This combined approach is referred to as the speed-gradient inverse optimal control. An example is used to illustrate the methodology. Several simulations are provided.\",\n \"Title: Discrete-time block control for a doubly fed induction generator coupled to a wind turbine\\nYear: 2011\\n\\nAbstract: This paper proposes a control scheme based on the block control technique using sliding modes, for a doubly fed induction generator connected to an infinity bus. This control scheme is compared with a linear control using PIs. Simulation results in PSCAD/EMTDC are included to illustrate the applicability of the scheme. A comparison between the PI controller and the proposed block controller is presented, from which, it is verified that our approach is faster to reach the speed reference.\",\n \"Title: Discrete-time recurrent high order neural networks for nonlinear identification\\nYear: 2010\\n\\nAbstract: This paper focuses on the problem of discrete-time nonlinear system identification via recurrent high order neural networks. It includes the respective stability analysis on the basis of the Lyapunov approach for the NN training algorithm. Applicability of the proposed scheme is illustrated via simulation for a discrete-time nonlinear model of an electric induction motor.\",\n \"Title: Discrete-Time Recurrent High Order Neural Observer for Induction Motors\\nYear: 2007\\n\\nAbstract: A nonlinear discrete-time neural observer for the state estimation of a discrete-time induction motor model, in presence of external and internal uncertainties is presented. The observer is based on a discrete-time recurrent high order neural network (RHONN) trained with an extended Kalman filter (EKF)-based algorithm. This observer estimates the state of the unknown discrete-time nonlinear system, using a parallel configuration. The paper also includes the stability proof on the basis of the Lyapunov approach. To illustrate the applicability simulation results are included.\",\n \"Title: Digital Sliding-Mode Sensorless Control For Surface-Mounted Pmsm\\nYear: 2014\\n\\nAbstract: In this work, a sampled-data model for surface-mounted permanent magnet synchronous motors (SMPMSMs) is proposed based on the symplectic Euler method. Then, based on the obtained model, a digital sliding-mode controller is designed for rotor velocity output tracking. Moreover, for rotor position and velocity and load estimation, a digital observer is designed. A separation principle is introduced in order to verify that the proposed controller can accomplish its task with estimated signals. Open-loop simulations predict that the sampled-data model obtained via the symplectic Euler method performs better than a model obtained with the classical explicit Euler method. Moreover, closed-loop simulations are carried on where the proposed controller is compared with a sampled continuous-time controller, where the controller here designed yields to better results even when the sampling period is increased. Finally, the performance of the proposed controller is verified by experimental tests.\",\n \"Title: Real-Time Discrete Recurrent High Order Neural Observer For Induction Motors\\nYear: 2008\\n\\nAbstract: A nonlinear discrete-time neural observer for the state estimation of a discrete-time induction motor model, in presence of external and internal uncertainties is presented. The observer is based on a discrete-time recurrent high order neural network (RHONN) trained with an extended Kalman filter (EKF)-based algorithm. This observer estimates the state of the unknown discrete-time nonlinear system, using a parallel configuration. The paper also includes the stability proof on the basis of the Lyapunov approach. To illustrate the applicability real-time results are included.\",\n \"Title: Real-time sliding mode control for a doubly fed induction generator\\nYear: 2011\\n\\nAbstract: This paper proposes a control scheme based on the block control technique using sliding modes, for a doubly fed induction generator (DFIG) prototype connected to an infinity bus. The DFIG is widely used as a wind generator; it allows the rotor speed to vary while synchronizing the stator directly to a fixed frequency power system. The generation scheme for the DFIG has one back-to-back PWM voltage-source converter connected between the rotor and the electrical grid. The rotor side converter (RSC) is connected via a dc link to the grid side converter (GSC), which is in turn connected to the stator terminals directly or through a step-up transformer. A block control scheme for the RSC is proposed to control the electric torque and the reactive power independently. The variables to be controlled by the GSC are the dc voltage in the dc link and the reactive power in the step-up terminals. The performance of the control scheme proposed is illustrated via real-time implementation in a 1/4HP DFIG prototype.\",\n \"Title: Multimachine Power-System Control: Integral-SM Approach\\nYear: 2009\\n\\nAbstract: In this paper, a robust nonlinear control scheme based on the combination of the block-control linearization and integral sliding modes is proposed. This control technique is applied to multimachine electric-power systems. The plant model takes into account the interactions in power systems between the electrical and mechanical dynamics and load constraints. A nonlinear observer is designed to est...\",\n \"Title: Integral high order sliding mode control of single-phase induction motor\\nYear: 2013\\n\\nAbstract: An observer-based controller for the single-phase induction motor is proposed in this paper. The scheme presented is formulated using block control feedback linearization technique and high order sliding mode algorithms with measurements of the rotor speed and stator currents. A second order sliding mode observer is included into the controller design in order to obtain estimates of the rotor flux. The stability of the complete closed-loop system is analyzed in the presence of model uncertainty, namely, rotor resistance variation and bounded time-varying load torque.\",\n \"Title: Backstepping control with sliding mode estimation for a hexacopter\\nYear: 2013\\n\\nAbstract: This paper presents the dynamic nonlinear model and the design of a controller for a hexarotor aerial robot. The control algorithm is based on the backstepping technique and it is applied to the trajectory tracking problem. The wind parameters resulting from the aerodynamic forces and moments acting on the hexacopter are estimated via a sliding mode observer with the super twisting algorithm in order to ensure robustness against aerial external disturbances and parameter variations. The performance and effectiveness of the proposed controller are tested in a simulation study taking into account external disturbances which are function of time.\",\n \"Title: Discrete-time neural block control for a doubly fed induction generator\\nYear: 2011\\n\\nAbstract: This paper proposes a control scheme based on the discrete-time block control technique using sliding modes, for a doubly fed induction generator connected to an infinity bus. In order to obtain the generator mathematical model, it is proposed to use a recurrent high order neural network (RHONN) identifier, which is trained with an extended Kalman filter (EFK) algorithm. Parameter changes are applied to test the scheme robustness. Its performance is illustrated via simulations.\",\n \"Title: High-Order Sliding Mode Block Control of Single-Phase Induction Motor.\\nYear: 2014\\n\\nAbstract: A new sliding mode (SM) observer-based controller for single-phase induction motor is designed. The proposed control scheme is formulated using block control feedback linearization technique and high-order SM algorithms with measurements of the rotor speed and stator currents. The stability of the complete closed-loop system, including the rotor flux second-order SM observer, is analyzed in the presence of model uncertainty, namely, rotor resistance variation and bounded time-varying load torque.\",\n \"Title: A simple recurrent neural network for solution of linear programming: Application to a Microgrid\\nYear: 2014\\n\\nAbstract: The aim of this paper is to present a simple new class of recurrent neural networks, which solves linear programming. It is considered as a sliding mode control problem, where the network structure is based on the Karush-Kuhn-Tucker (KKT) optimality conditions, and the KKT multipliers are the control inputs to be implemented with finite time stabilizing terms based on the unit control, instead of common used activation functions. Thus, the main feature of the proposed network is the fixed number of parameters despite of the optimization problem dimension, which means, the network can be easily scaled from a small to a higher dimension problem. The applicability of the proposed scheme is tested on real-time optimization of an electrical Microgrid prototype.\",\n \"Title: A discontinuous recurrent neural network with predefined time convergence for solution of linear programming\\nYear: 2014\\n\\nAbstract: The aim of this paper is to introduce a new recurrent neural network to solve linear programming. The main characteristic of the proposed scheme is its design based on the predefined-time stability. The predefined-time stability is a stronger form of finite-time stability which allows the a priori definition of a convergence time that does not depend on the network initial state. The network structure is based on the Karush-Kuhn-Tucker (KKT) conditions and the KKT multipliers are proposed as sliding mode control inputs. This selection yields to an one-layer recurrent neural network in which the only parameter to be tuned is the desired convergence time. With this features, the network can be easily scaled from a small to a higher dimension problem. The simulation of a simple example shows the feasibility of the current approach.\",\n \"Title: Torque controller via second order sliding modes of WRIG impelled by DC-motor for application in wind systems\\nYear: 2012\\n\\nAbstract: In this paper, the authors propose three robust non-linear controllers based on a second order sliding mode technique named super-twisting method which can be applied into work bench for research related to wind system. The main control scheme is proposed to control the electromagnetic torque and stator power factor of a wound rotor induction generator connected to a local electrical network and mechanically coupled with a DC-motor whose velocity is controlled applying feedback linearization technique and super-twisting algorithm; it is known that the DC-motor can be emulate the operation of the wind turbine. Finally, a grid side converter controller is proposed to regulate the DC link bus located between to power converters connected in “back to back” standard configuration, applying linearization by block control and super-twisting algorithm. The performance of the designed controllers for the DC motor-WR induction generator group is validated through real time implementation.\",\n \"Title: Visual servoing and robust object manipulation using symmetries and Conformai Geometric Algebra\\nYear: 2014\\n\\nAbstract: Object tracking and manipulation is an important process for many applications in robotics and computer vision. A novel 3D pose estimation of objects using reflectionally symmetry formulated in Conformai Geometric Algebra (CGA) is proposed in this work. The synthesis of the kinematics model for robots and a sliding mode controller using the CGA approach is described. Real time implementation results are presented for the pose estimation of object using a stereo vision system.\",\n \"Title: Discrete-time Neural Network Control for a Linear Induction Motor\\nYear: 2008\\n\\nAbstract: This paper presents a discrete-time control for a linear induction motor (LIM). First, an identifier is proposed with a nonlinear block controllable form (NBC) structure. This identifier is based on a discrete-time high order neural network trained on-line with an extended Kalman filter (EKF)-based algorithm. Then, a sliding mode control is used to achieve the purpose of tracking velocity and magnitude flux. The neural control performance is illustrated via simulations.\",\n \"Title: Real-Time Inverse Optimal Control For A Planar Robot\\nYear: 2010\\n\\nAbstract: This paper presents an inverse optimal neural controller, which is constituted by the combination of two techniques: a) inverse optimal control to avoid solving the Hamilton Jacobi Bellman (HJB) equation associated to nonlinear system optimal control, and b) an on-line neural identifier, which uses a recurrent neural network, trained with the extended Kalman filter (EKF), in order to build a model of an assumed unknown nonlinear system. The applicability of the proposed approach is illustrated by real-time control of a planar robot.\",\n \"Title: Real-Time Output Trajectory Tracking using a Discrete Neural Backstepping Controller.\\nYear: 2008\\n\\nAbstract: This paper deals with adaptive tracking for discrete-time MIMO nonlinear systems in presence of bounded disturbances. A high order neural network (HONN) structure is used to approximate a control law designed by the backstepping technique, applied to a block strict feedback form (BSFF). The learning algorithrm for the HONN is based on an Extended Kalman Filter (EKF). This paper also includes the respective stability analysis, using the Lyapunov approach. The proposed scheme is implemented in real-time to control a three phase induction motor, as to track a time-variying speed reference and a constant flux magnitude reference.\",\n \"Title: A fixed time convergent dynamical system to solve linear programming\\nYear: 2014\\n\\nAbstract: The aim of this paper is to present a new dynamical system which solves linear programming. Its design is considered as a sliding mode control problem, where its structure is based on the Karush-Kuhn-Tucker optimality conditions, and its multipliers are the control inputs to be implemented by using fixed time stabilizing terms with vectorial structure, based on the unit control, instead of common terms used in other approaches. Thus, the main features of the proposed system are the fixed convergence time to the programming solution and the fixed parameters number despite of the optimization problem dimension. That is, there is a time independent to the initial conditions in which the system converges to the solution and, the proposed structure can be easily scaled from a small to a higher dimension problem. The applicability of the proposed scheme is tested on real-time optimization of an electrical Microgrid prototype.\",\n \"Title: A recurrent neural network for real time electrical microgrid prototype optimization\\nYear: 2014\\n\\nAbstract: The aim of this paper is to present a new class of recurrent neural networks, which solve linear programming. It is considered as a sliding mode control problem, where the network structure is based on the Karush-Kuhn-Tucker (KKT) optimality conditions, and the KKT multipliers are the control inputs to be implemented with fixed time stabilizing terms, instead of common used activation functions. Thus, the main feature of the proposed network is its fixed convergence time to the solution, which means, there it is a time independent to the initial conditions in which the network converges to the optimization solution. The applicability of the proposed scheme is tested on real-time optimization of an electrical microgrid prototype.\",\n \"Title: High order integral nested sliding mode control\\nYear: 2014\\n\\nAbstract: This paper exposes a controller for the nonlinear systems in the block controllable form. This proposal guarantees exponential exact tracking in the presence of unknown matched and unmatched disturbances by means a combination of the block control method and integral terms designed with high-order sliding-modes algorithms. Both, matched and unmatched, disturbances are compensated by those integral continuous terms and the tracking is achieved with the design of a nominal control law.\",\n \"Title: Sliding control variation for ball-plate system stabilization\\nYear: 2014\\n\\nAbstract: The fundamental purpose of this paper is to present a stabilizing control for rolling ball (ball-plate system), which is developed using the framework of the theory of high order control variation in combination with integrator backstepping, and sliding mode control. Under this scheme, control laws are proposed to solve the problem of stabilization for this system.\",\n \"Title: A fixed-time second order sliding mode observer for a class of nonlinear systems\\nYear: 2014\\n\\nAbstract: This paper presents a second order fixed time sliding mode observer based on an extension of the super-twisting algorithm. This observer can be applied to a class of nonlinear system with a block-wise representation. The block structure provides a straightforward form to the application of the proposed second order sliding mode algorithm, yielding to finite-time convergence with a settling time independent to the system initial conditions. Finally, as numerical simulation example, the case of a linear induction motor is studied, exposing the efficiency and feasibility of the proposal.\",\n \"Title: A discontinuous mean-square filter for stochastic differential systems\\nYear: 2012\\n\\nAbstract: This paper presents a mean-square filter for Lipschitz stochastic differential systems, which contains a discontinuous innovations term in the form of unit control. It is demonstrated that the designed filter yields the ultimate boundedness of the estimation error variance. The developed algorithm is then applied to a third degree polynomial system and compared to the extended Kalman-Bucy filter. The simulation results show the faster convergence and better approximation properties of the proposed filter.\",\n \"Title: An equivalent control based sliding mode observer using high order uniform robust sliding operators\\nYear: 2012\\n\\nAbstract: In this paper a sliding-mode observer based on the equivalent control method for discontinuous functions for a class of non-linear systems is proposed. The observer structure and its existence conditions are presented. Besides, a class of high order sliding operators with the properties of uniform (w.r.t. initial conditions) finite time convergence and with reduction of chattering effect are exposed. The use of these operators in the observer design allows the calculation of the equivalent control and the observer convergence uniformly in finite time. A simulation example is presented to illustrate the proposed method.\",\n \"Title: Optimal filtering for uncertain linear stochastic systems\\nYear: 2009\\n\\nAbstract: This paper presents the optimal joint filtering and parameter identification problem for uncertain linear stochastic systems with unknown parameters in both state and observation equations, where the unknown parameters are considered Wiener processes. The original problem is reduced to the filtering problem for an extended state vector that incorporates parameters as additional states. The resulting filtering system is polynomial in state and linear in observations. The obtained optimal filter for the extended state vector also serves as the optimal identifier for the unknown parameters. Performance of the designed optimal state filter and parameter identifier is verified for both, positive and negative, parameter values.\",\n \"Title: HOSM block control of SPIM\\nYear: 2012\\n\\nAbstract: In the continuation of authors' studies on control and estimation methods for Single-Phase Induction Motor (SPIM), a new observer-based controller using High Order Sliding Mode (HOSM) algorithms is proposed here. This observer-controller scheme only uses measurements of the rotor speed and stator currents. The complete scheme is robust to uncertainties in the rotor resistance, and a bounded time-varying load torque.\",\n \"Title: ℋ∞ Filtering for Polynomial Systems Over Switching Delayed Observations\\nYear: 2013\\n\\nAbstract: The problem of designing an ℋ∞ filter for a class of nonlinear polynomial systems over linear delayed observations is presented in this paper. Output observations may or may not experience sensor delay due to a random variable which is modelled as a Bernoulli distributed one which takes the quantities of zero and one. A closed form of this filter is obtained by expressing the conditional expectations of polynomial terms as functions of the estimate and the gain matrix. As a particular case, a third degree polynomial is considered to obtain the finite-dimensional filtering equations. Simulations results applied to an induction motor show the effectiveness of proposed filter.\",\n \"Title: Discontinuous output regulation for a DC-to-AC boost converter\\nYear: 2012\\n\\nAbstract: This work presents the design of a discontinuous output regulator for a DC-to-AC power conversion problem on a boost converter, for asymptotic output trajectory tracking. The steady state reference signal for the inductor current is proposed as an approximated solution of the well known Francis-Isidori-Byrnes equations, then, by a proper selection of a sliding surface one can stabilize the non-minimum phase inductor current tracking variable. Finally, by means of numerical simulations one verifies the performance of the proposed control strategy.\",\n \"Title: Super-twisting sliding mode control of synchronous motors\\nYear: 2012\\n\\nAbstract: Based on a nonlinear complete order model of a synchronous motor, a robust second order sliding mode observer based control scheme is proposed. To design this scheme a super-twisting sliding mode control and adaptive backstepping techniques, are implemented. A nonlinear reduced order observer is designed for estimation of the rotor fluxes. The simulation result shows the effectiveness of the proposed controller. Copyright © 2012 CCE.\",\n \"Title: Discrete-time sliding mode regulator for nonminimum phase systems\\nYear: 2012\\n\\nAbstract: This work presents an approach to solve the output regulation problem for a class of nonlinear discrete-time nonminimum phase perturbed systems. Based on feedback linearization block control technique and discrete-time sliding mode (SM) control, we propose a sliding manifold on which the zero dynamics are stabilized. To enforce the robust SM motion on the designed manifold, a discrete-time super-twisting SM algorithm is implemented. The effectiveness of the proposed methodology is verified via the design of a position tracking controller for an under-actuated robotic system, the Pendubot.\",\n \"Title: Optimal Controller for Stochastic Polynomial Systems with State-Dependent Polynomial Input\\nYear: 2011\\n\\nAbstract: AbstractThis paper presents an optimal quadratic-Gaussian controller for stochastic polynomial systems with a state-dependent polynomial control input and a quadratic criterion over linear observations. The optimal closed-form controller equations are obtained using the separation principle, whose applicability to the considered problem is substantiated. As an intermediate result, the paper gives a closed-form solution of the optimal regulator (control) problem for polynomial systems with a state-dependent polynomial control input and a quadratic criterion. Performance of the obtained optimal controller is verified in an illustrative example against a conventional linear-quadratic-Gaussian (LQG) controller that is optimal for linearized systems. Simulation graphs demonstrating overall performance and computational accuracy of the designed optimal controller are included.\",\n \"Title: Optimal Filtering for Polynomial Systems over Switching Delayed Observations\\nYear: 2013\\n\\nAbstract: The problem of designing an optimal filter for a class of nonlinear systems confused with white Gaussian noise over linear time delayed observations is presented in this paper. Output observations may or may not experience sensor delays due to a random variable which is modeled as a Bernoulli distributed one that takes the quantities of zero and one. A closed form of this filter is obtained by expressing the conditional expectation of polynomial terms as a function of its expectation and covariance matrix. As a particular case, finite-dimensional filtering equations are obtained for a third degree polynomial function. Numerical simulations are performed on an induction motor in order to show the effectiveness of proposed filter.\",\n \"Title: Modeling And Pose Control Of Robotic Manipulators And Legs Using Conformal Geometric Algebra\\nYear: 2015\\n\\nAbstract: Controlling the pose of a manipulator involves finding the correct configuration of the robot's elements to move the end effector to a desired position and orientation. In order to find the geometric relationships between the elements of a robot manipulator, it is necessary to define the kinematics of the robot. We present a synthesis of the kinematical model of the pose for this type of robot using the conformal geometric algebra framework. In addition, two controllers are developed, one for the position tracking problem and another for the orientation tracking problem, both using an error feedback controller. The stability analysis is carried out for both controllers, and their application to a 6-DOF serial manipulator and the legs of a biped robot are presented. By proposing the error feedback and Lyapunov functions in terms of geometric algebra, we are opening a new venue of research in control of manipulators and robot legs that involves the use of geometric primitives, such as lines, circles, planes, spheres.\",\n \"Title: A Robust Automotive Controller Design\\nYear: 2008\\n\\nAbstract: This paper describes an application of the block control and sliding mode control techniques to form a stabilising controller for an internal combustion engine with a throttle driven by a DC motor. This combined approach enables the inherent non-linearities of the engine to be compensated and high-level external disturbances to be rejected. Also, the control system utilises a generalised observer for non-linear plants, which incorporates a high gain technique.\",\n \"Title: Distributed super twisting controller for multiple quadrotors\\nYear: 2015\\n\\nAbstract: The distributed cooperative tracking control problem for a group of quadrotors in a three-dimensional space is addressed in this paper. The controller design is divided in two stages. In the first stage, local distributed controllers for the translational dynamics are designed, forcing the quadrotor to asymptotically track the desired trajectory with the required separation and therefore a specific formation. In the second stage, the rotational dynamics is asymptotically stabilized. The controllers implemented in both stages are based in a combination of block control technique and the super twisting control algorithm which ensures robustness with respect to external disturbances and parameter uncertainties. Moreover, a first order exact differentiator is used to estimate the virtual control inputs, simplifying the control law design. The stability proof of the complete closed-loop system is shown to be asymptotically stable. Finally, numerical simulations are carried out to show that theoretical conclusions are effective.\",\n \"Title: Quaternion-based trajectory tracking robust control for a quadrotor\\nYear: 2015\\n\\nAbstract: This paper tackles the problem of trajectory tracking problem of a quadrotor. A sliding mode block control algorithm based on the super twisting algorithm using unit quaternion feedback is applied to the dynamic model of the quadrotor helicopter. The derivative of the virtual control inputs are obtained using exact first order differentiator, and the aerodynamic forces and moments acting on the quadrotor are estimated via a sliding mode observer to ensure robustness against external disturbances and model uncertainties. Moreover, a comparison of the computational complexity of the controller between the proposed control scheme and the one considering Euler angles is carried out. The stability proof showing the asymptotic stability of the proposed control scheme is presented. Simulations show the good performance of the proposed control scheme.\",\n \"Title: Real-time neural inverse optimal control for position trajectory tracking of an induction motor\\nYear: 2015\\n\\nAbstract: This paper describes a neural inverse optimal control approach for a three-phase induction motor position trajectory and flux magnitude tracking. A recurrent high order neural network (RHONN) is used to identify the plant model, trained with an Extended Kalman Filter (EKF) algorithm; the control law minimize a cost functional avoiding to solve the Hamilton Jacobi Bellman (HBJ) equation. The applicability of the approach is illustrated via experimental results. The proposed scheme allows the easy integration of this kind of motors into a system of systems configuration.\",\n \"Title: Predefined-time stability of dynamical systems with sliding modes\\nYear: 2015\\n\\nAbstract: This paper introduces a class of fixed-time stable dynamical systems with settling time as a explicit parameter, namely the inverse the gain. Those systems are defined as predefined-timed stable dynamical systems. Continuous and discontinuous are cases are presented. A detailed Lyapunov characterization of this class of systems is also shown. Finally, the application to the design of a class of first order sliding mode controllers is exposed.\",\n \"Title: Control for partial state constraining of nonlinear systems in regular form\\nYear: 2015\\n\\nAbstract: A method for enforcing non-symmetrical constraints on the first block of the state of a nonlinear system in regular form is presented. Two cases are considered: the first block of the system stays within some arbitrarily selected limits, a) displaying a conservative behavior, shown as limit-cycle plots, and b) some sort of damping is introduced, forcing the second block of the state to tend to zero as time increases. Both cases emerged from a convenient selection of the stabilizing function in the control law. Simulation results present the closed-loop behavior of an open-loop unstable MIMO system, showing the effectiveness of the approach.\",\n \"Title: Integral nested ℋ∞ block control\\nYear: 2015\\n\\nAbstract: This paper exposes a robust controller for the tracking of nonlinear systems in the block controllable form. This proposal is based on the block control method and integral terms designed with ℋ\\n<sub xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">∞</sub>\\n algorithms. The controller guarantees high performance the presence of uncertainty. The disturbances are compensated by those integral continuous terms and the tracking is achieved with the design of a nominal control law.\",\n \"Title: Enhanced Discrete-Time Modeling via Variational Integrators and Digital Controller Design for Ground Vehicles.\\nYear: 2016\\n\\nAbstract: In this paper, a discrete-time controller for ground vehicles is proposed, for the tracking of desired references for lateral and yaw velocities. A novel discrete-time model of the vehicle dynamics is considered, which is obtained by means of a variational integrator known as symplectic Euler. The control inputs are given by active front steering and rear torque vectoring. Simulations have been ca...\",\n \"Title: Predefined-time tracking of a class of mechanical systems\\nYear: 2016\\n\\nAbstract: In this paper the problem of predefined-time exact tracking of fully actuated and unperturbed mechanical systems is solved by means of a continuous controller. It is assumed the availability of the state and the desired trajectory as well as its two first derivatives. This is accomplished introducing the idea of second-order predefined-time stable systems, which is based on the nested application of the first-order predefined-time stabilizing function. As an example, the proposed solution is applied over a two-link planar manipulator and numerical simulations are conducted to show its performance.\",\n \"Title: Arbitrary references output tracking in linear systems\\nYear: 2016\\n\\nAbstract: The problem of arbitrary references output tracking for linear systems is solved through an approximate required steady state. This estimate is obtained using a dynamic estimator based on a state observer. The state-feedback control law is proposed using this estimate and the closed-loop system is then analyzed. The simulations show the effectiveness of the proposed technique.\",\n \"Title: Dynamic Predictor For Linear Time-Delay Systems With Disturbances\\nYear: 2016\\n\\nAbstract: A dynamic predictor is proposed for linear time invariant disturbed systems with delay in the control input. The proposed dynamic predictor has a structure similar to the Luenberger observer. It outperforms the current predictors because it does not need the full state, as long as the plant is observable, it only requires the output from the system. The controller is proposed making use of the state vector from the dynamic predictor. The full control scheme including predictor and control law is applied in simulation. The advantages of the proposed predictor include robustness against external disturbances, and also the fact that the controllers developed making use of predictor are capable of tracking time-variable signals.\",\n \"Title: Robust Observer-Based Sliding Mode Controller For Vehicles With Roll Dynamics\\nYear: 2016\\n\\nAbstract: In this work, an observer-based sliding mode (SM) control scheme for a vehicle considering roll dynamics, is presented. The proposal considers a nonlinear higher order sliding mode reduced observer for the lateral velocity, roll angle and roll velocity. Based on this observer, the controller is designed for the reference tracking of the lateral and yaw velocities of the vehicle, using adaptive super-twisting algorithm. In order to demonstrate the stability of the closed-loop system, Lyapunov functions for each of the algorithms are used. The proposed control scheme has finite-time convergence and the robustness in presence of plant parameters variations. This is demonstrated through the performed simulation.\",\n \"Title: On The Discrete-Time Modeling And Control Of Synchronous Generators By Means Of Variational Integrators And Sliding Modes\\nYear: 2016\\n\\nAbstract: This work deals with the discrete-time modeling and control design for synchronous generators by means of a variational integrator and sliding modes. First, a continuous non-conservative Lagrangian is formulated for the plant, then, the respective discrete Lagrangian is determined. Based on this, discrete-time rules are derived for synchronous generators. Second, a sliding-mode controller is proposed for the velocity stabilization of the system, and a nonlinear observer is designed for the non-measurable states. Simulations show the good performance of the synchronous generator when closed-loop with the novel discrete-time controller.\",\n \"Title: Causal Output Tracking For Discrete-Time Linear Systems\\nYear: 2017\\n\\nAbstract: A new approach to the solution of the causal output tracking problem for discrete-time linear systems is presented. A dynamic steady state estimator is proposed in the form of a state observer taking as output, the reference to be tracked. Once the steady state has been estimated, the output tracking problem is converted into a stabilization problem and a control law is designed such that the output tracking is achieved while the system dynamics keep stable. The main applications of the techniques presented in this work are the non-minimum phase systems, where classical input-output techniques cannot be applied.\",\n \"Title: Robust sensorless observer-based adaptive sliding modes control of synchronous motors.\\nYear: 2018\\n\\nAbstract: •The full order model of the synchronous motor is considered with matched and unmatched perturbations.•Adaptive backstepping and the super-twisting sliding mode control techniques are used to ensure robustness of the system.•A generalized super-twisting algorithm is proposed to design a robust observer for a 3rd order nonlinear system.•The observer proposed is used to estimate the rotor fluxes in presence of plant parameter variations.•The rotor speed estimator proposed is used in the design of the controller and flux observer, avoiding the use of sensor.\",\n \"Title: Sliding mode predictive control of linear uncertain systems with delays.\\nYear: 2018\\n\\nAbstract: In this paper, a continuous control strategy for robust stabilization of a class of uncertain multivariable linear systems with delays in both the state and control variables is proposed. A predictor is designed to compensate the delay effect in the control input, and then an integral sliding mode control technique along with super-twisting algorithm is applied to compensate partially the effect of the perturbation term. Finally, a nominal delay-free part of the control input is designed to stabilize the sliding mode dynamics. The proposed control scheme is extended to the class of systems modeled in Regular form. For this class of perturbed systems with delay in the state, a transformation to the systems with the delay-free state is proposed. The stability conditions of the closed-loop uncertain system are derived, and the results obtained in this work are compared against previous works. To show the effectiveness of the proposed method, simulation results are presented.\",\n \"Title: Higher Order Sliding Mode And Adaptive Backstepping Controllers For A Full-Car Model With Active Suspension\\nYear: 2018\\n\\nAbstract: The robust stabilization of the vertical position, pitch and roll dynamics of the center of gravity for a vehicle with active suspension is addressed in this work. Three different higher order sliding mode (HOSM) control algorithms and an adaptive backstepping controller (ABC) are designed to fulfill the control objective. First, two second order sliding mode controllers, the quasi-continuous and the nested, are designed to stabilize the 2-relative degree vertical position. Then, a super-twisting (ST) algorithm is designed, by means of a 1-relative degree sliding function, obtaining asymptotic stability. For the ABC, adaptive parameters dynamics are proposed, as well as the control law that adds desired linear dynamics. Finally, a performance comparison between the proposed controllers is presented based on simulations results.\",\n \"Title: A second order predefined-time control algorithm\\nYear: 2017\\n\\nAbstract: The predefined-time stabilization of second-order systems, i.e. the fixed-time stabilization with settling time as a function of the controller parameters, is revisited in this paper. The proposed controller is a time-based switched controller which first drive the system trajectories to a linear manifold in predefined time and then uses a nested second-order controller. The application of the results is demonstrated for the trajectory tracking control in fully actuated mechanical systems. An illustrative example of the control of a two-link planar manipulator with predefined-time convergence shows the effectiveness of the proposed algorithm.\",\n \"Title: Semi-global predefined-time stable systems\\nYear: 2017\\n\\nAbstract: A Lyapunov-based construction of a predefined-time stabilizing function (a function that stabilizes a system in fixed-time with settling time as function of the controller parameters) for scalar systems is considered in this paper. The constructed function involves the inverse incomplete gamma function, causing this function to be semi-global, i.e., the domain of definition of the function can be made as large as wanted with an appropriate parameter selection. Finally, the constructed function is used to design predefined-time stabilizing controllers which are robust against vanishing and non-vanishing perturbations.\",\n \"Title: Semi-Global Predefined-Time Stable Vector Systems\\nYear: 2018\\n\\nAbstract: In this paper, we expose a control function which allows the semi-global predefined-time stabilization of first-order vector systems. The predefined-time stability is a stronger class of finite-time stability that has as main advantage the settling time as a tunable parameter of the proposed function. To design that stabilizing function, we use the unit control principle jointly to the inverse incomplete gamma function. For the resulting expression, the domain of definition the inverse incomplete gamma function can be made as large as wanted with an appropriate parameter selection, and, as consequence, the attraction domain of the systems. Therefore, we say that the system exhibits semi-global predefined-time stability. As an essential feature, the parameter which defines the settling time bound and those that tune the attraction domain are independent of each other. Finally, the constructed function is used to design predefined-time stabilizing controllers which are robust against vanishing and non-vanishing perturbations.\",\n \"Title: Robust Discrete-Time Sliding Mode Control With Arbitrarily Small Boundary Layer\\nYear: 2018\\n\\nAbstract: The robust discrete-time sliding mode control problem is addressed. A novel approach to estimate the matched disturbance is presented. This approach is based on the approximation of a discrete-time function actual value by using various previous steps of itself. The matched disturbance previous step is calculated from the system model and then, using the mentioned approach, its actual value is estimated with an error of exponential order w.r.t. the sampling time and the used number of previous steps. Simulations show the effectiveness of the proposed approach to estimate a discrete time function and to robustly stabilize a dynamic system.\",\n \"Title: Robust Discrete-Time Sliding Mode Controller For A Class Of Nonlinear Perturbed Systems\\nYear: 2018\\n\\nAbstract: The output tracking problem for a class of nonlinear systems presented in Nonlinear Block Controllable (NBC) form is addressed. Both matched and unmatched perturbations are considered. First, the Block Control iterative feedback linearization technique combined with a perturbation estimation are employed to design a sliding manifold. With the perturbation estimation, the effect of unmatched perturbation is mitigated. Then, a discrete-time sliding mode controller is synthesized such that the system state is driven toward a vicinity of the designed sliding manifold and stays there for all sampled time instants, avoiding chattering and reducing the matched perturbation effect. The effectiveness of the proposed methodology is confirmed by simulation.\",\n \"Title: Robust Control for Perturbed Linear Systems with Time-Varying Delay via Sliding Mode Control\\nYear: 2018\\n\\nAbstract: In the present paper, a robust predictor-based sliding mode control for perturbed time-varying delay systems is proposed. A predictor is designed to compensate the time-varying delay effect in control input, and then an integral sliding mode control technique along with super-twisting algorithm is applied to compensate partially the effect of the perturbation term. Finally, a nominal delay-free part of the control input is designed to stabilize the sliding mode dynamics. Conditions for the stability of the closed-loop perturbed system are derived. Simulation result is presented.\",\n \"Title: Adaptive back-stepping speed Tracking controller for a Brush-less DC motor with a SM observer for back EMFs\\nYear: 2018\\n\\nAbstract: This paper proposes a robust speed tracking controller design for Brush-less DC Motor (BLDC) under parametric uncertainness based on the combination of a sliding modes (SM) observer and adaptive back-stepping control techniques. The adaptive controller developed in this work in combination with the SM observer allows also a reduction on the torque rippling since electrical torque is controlled in an indirect way based on the obtained estimates of the back EMFs. The effectiveness of the proposed controller is demonstrated by simulations with MATLAB/SIMULINK.\",\n \"Title: Sliding mode identification and control for a class of nonlinear systems\\nYear: 2018\\n\\nAbstract: In this work, the problem of designing a control scheme capable of controlling dynamical systems with unknown time-varying parameters and disturbances, is proposed. In contrast with other works, based on two techniques: adaptive backstepping and sliding modes, an improved method that guaranties the output asymptotic tracking of a smooth reference signal, the stability of the closed-loop system and the identification errors boundedness, is developed. By means of sliding mode observers, the adaptation errors required information is extracted and injected to adaptive laws. The behavior of the proposed control scheme is analyzed by Lyapunov method. The performance of the proposed system is verified with an academic example.\",\n \"Title: Adaptive Discrete Time Sliding Mode Control for a Class of Nonlinear Systems\\nYear: 2018\\n\\nAbstract: In this paper, a robust sliding mode adaptive tracking state feedback control scheme is designed for a class of nonlinear discrete time (DT) perturbed systems in parametric strict feedback form. The design procedure is based on the adaptive backstepping methodology proposed for continuous time systems, using step-by-step a Lyapunov function. First, an adaptive sliding manifold is formulated, and then, a DT sliding mode controller is proposed to achieve the designed manifold be attractive in presence of bounded matched uncertainty. To show the effectiveness of the proposed method, simulation results are performed.\",\n \"Title: Output-Feedback Discrete-Time Sliding Mode Control via Disturbance Estimation\\nYear: 2018\\n\\nAbstract: In this paper, a novel technique to achieve robustness in sliding mode control for discrete-time systems by measuring only its output is presented. The proposed technique is based on the approximation of an arbitrary discrete-time function by using various of its previous steps. Using this approach, the system disturbance is approximated by a dynamical system and as result, the plant model is extended. Based on the extended model, a state observer is designed to estimate both the system state and the disturbance. Using the observer state, a discrete-time sliding mode controller is then designed to robustly stabilize the perturbed system. The simulation results show the effectiveness of the proposed technique.\",\n \"Title: Second Order Sliding Mode Control of a STATCOM with Saturated Inputs\\nYear: 2018\\n\\nAbstract: This paper presents a robust controller for a STATCOM device with saturated inputs. As the primary assumption, the proposed design considers the presence of unknown but bounded external perturbations and parametric variations. This proposal has a cascade structure, where a saturated super twisting control algorithm closes the currents control loop, and a high-gain proportional-integral (PI) algorithm ensures the voltage regulation. Thus, the exposed scheme provides an adequate performance of the STATCOM, considering the saturation of the inputs with the anti-windup feature. Posteriorly, a proper stability analysis presents the conditions for the appropriate operation of the closed-loop system in saturation and non-saturation regimes. Numerical simulations are also included to show the performance of the proposed controller.\",\n \"Title: Predefined-Time Stabilization Of High Order Systems\\nYear: 2017\\n\\nAbstract: The aim of this paper is to introduce a controller that stabilizes a class of arbitrary order systems in predefined-time. The proposed controller is designed with basis on the block-control principle yielding in a nested structure similar to high order sliding mode algorithms and terminal sliding mode algorithms. For this case, it is assumed the availability of the state and the absence of perturbations. Numerical simulations expose the desired performance of this controller.\",\n \"Title: Passivity Sliding Mode Control of Large-Scale Power Systems\\nYear: 2019\\n\\nAbstract: In this brief, a sliding mode (SM) decentralized excitation controller for multimachine power systems is designed to stabilize the speed and regulate the terminal voltage of each generator in the system. To achieve the speed stabilization, a sliding manifold is designed using the passivity technique. An integral SM controller is proposed to reject internal and external perturbations, taking advant...\",\n \"Title: Approximate Causal Output Tracking For Linear Perturbed Systems Via Sliding Mode Control\\nYear: 2020\\n\\nAbstract: A new approach to the solution of what is termed causal output tracking problem for linear time-invariant systems in the presence of both matched and unmatched unmodelled disturbances is presented. The proposed solution is addressed through the design of a dynamic steady state estimator based on the desired system structure and an observer, considering the reference as the system output. The unmatched disturbance is estimated and used in the steady state estimator to achieve invariance, while the matched disturbance robustness is provided via sliding mode control using the super-twisting algorithm. A useful application of the presented techniques is output tracking for non-minimum phase systems; thus, the performed simulation results confirm the effectiveness of the proposed control scheme for this kind of systems.\",\n \"Title: Observer based sliding mode controller for vehicles with roll dynamics.\\nYear: 2019\\n\\nAbstract: In this work, considering the roll dynamics and actuator dynamics, an observer-based control scheme for a vehicle is proposed. The proposal considers a nonlinear higher order sliding mode observer to estimate unmeasurable lateral velocity, roll angle and roll velocity. Using the observer information, a controller based on block control with sliding mode technique is designed for the reference trajectory tracking of the lateral and yaw velocities of the vehicle. The stability of the complete closed-loop system including zero dynamics is analyzed. The effectiveness of the proposed scheme is demonstrated through CarSim simulations.\",\n \"Title: Predefined-time robust contour tracking of robotic manipulators\\nYear: 2019\\n\\nAbstract: Robotic tasks are primarily defined to comply to spatial and time constraints, and in several of such tasks, it is required to track a nominal contour depending just on spatial coordinates, without involving the time variable in the definition of such a contour, in contrast to conventional trajectory tracking schemes. With the aim to satisfy the salient requirements of these robotic tasks, this paper presents an integrated robust controller. The contour is encoded by using a robust velocity field, such that, any particle immerse in such field converges in predefined-time to a predefined vicinity of the given contour. Then, a robust controller is proposed to enforce the velocity field, in turns, in predefined-time. In this sense, a predefined-time-precision contour tracking is assured, by considering a robust controller that does not assume the exact knowledge of the robot parameters for its real-time implementation. Numerical simulations are conducted to highlight the reliability of the proposed scheme.\",\n \"Title: Synchronization of complex networks in discrete-time for trajectory tracking using neural sliding modes with pinning control\\nYear: 2021\\n\\nAbstract: This paper considers a novel solution for synchronization of complex networks in discrete-time, based on Lyapunov theory and passivity degree to define a reliable interval of coupling strength required to achieve synchronization in a directed complex network with non-identical nodes. By means of a discrete-time recurrent high-order neural network trained with an extended Kalman filter algorithm, the unknows pinned nodes are identified. Later, using an adaptive approach of sliding modes in discrete-time to complex networks, a selected pin node is locally controlled to guarantee trajectory tracking. Once the pin node is controlled, network synchronization is achieved, all its nodes follow selected trajectories. The results are illustrated via simulations considering that each node corresponds to a discretized nonlinear chaotic attractor.\",\n \"Title: Adaptive Control Methodology for a Class of Nonlinear Systems With Speed Tracking Implementation for a BLDC Motor\\nYear: 2022\\n\\nAbstract: This work proposesan adaptive control methodology for a class of time-varying uncertain systems. The goal of this work is to address problems in which the model is known, but the parameters involved are either unknown, time-varying, or the system is being affected by disturbances. The proposed methodology relies on Lyapunov control concepts and the bound of the time-varying elements (only the boundedness feature is known) to extend traditional adaptive control techniques restricted to linear constant parameterization. In order to show the efficiency of this methodology, we present a comparison simulation and an implementation for speed tracking in a Brush-Less Direct Current (BLDC) motor under parametric uncertainty.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":99,"string":"99"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13561,"string":"13,561"}}},{"rowIdx":13562,"cells":{"question_id":{"kind":"string","value":"Costas Courcoubetis"},"question":{"kind":"string","value":"Generate an abstract for the title \"Measurement-Based Usage Charges in Comminucations Networks\" using the following items: usage-sensitive charging schemes, effective bandwidth, network resource, usage charge, measurable bounds."},"answer":{"kind":"string","value":"We study usage-sensitive charging schemes for broadband communications networks. We argue that a connection's \"effective bandwidth\" is a good proxy for the quantity of network resource that the connection consumes and can be the basis for a usage charge. The determination of effective bandwidth can be problematic, however, since it involves the moment-generating function of the cell arrival process, which may be difficult to model or measure. This article describes methods of computing usage charges from simple measurements and relating these to bounds on the effective bandwidth. Thus we show that charging for usage on the basis of effective bandwidths can be approximated well by charges based on simple measurements."},"correct_answer":{"kind":"string","value":"We study usage-sensitive charging schemes for broadband communications networks. We argue that a connection's \"effective bandwidth\" is a good proxy for the quantity of network resource that the connection consumes and can be the basis for a usage charge. The determination of effective bandwidth can be problematic, however, since it involves the moment-generating function of the cell arrival process, which may be difficult to model or measure. This article describes methods of computing usage charges from simple measurements and relating these to bounds on the effective bandwidth. Thus we show that charging for usage on the basis of effective bandwidths can be approximated well by charges based on simple measurements."},"documents":{"kind":"list like","value":["Title: Providing Bandwidth Guarantees Over A Best-Effort Network: Call-Admission And Pricing\nYear: 2001\n\nAbstract: This pager introduces a framework for answering questions regarding the conditions on the network load that allow a best-effort network like the Internet to support connections of given duration that require a certain quality of service. Such quality of service is expressed in terms of the percentage of time the bandwidth allocated to a connection may drop below a certain level or the maximum allowable delay in placing the call through the network waiting for more favorable loading conditions. The call-acceptance conditions, which depend on the behavior of the system over the lifetime of accepted calls, are thus based on transient models for the congestion (instead of looking at the average behavior) and attempt to exploit the time-scales of the fluctuations of the number of connections competing for bandwidth. Extensions of the model consider the case of dynamic pricing which allows connections that pay more to get larger shares of the bandwidth, and investigate the trade-off between quality of service, the size of the acceptance region, and the charge to be paid by the connection. Under this framework we introduce an option contract that reduces the risk of quality disruption, if a user has a fixed budget at his disposal, and calculate its price. One potential use of this methodology is towards developing a simple admission control mechanism for placing voice calls through an IP network, where the decisions can be taken by edge devices.","Title: Differentiated Services in the GPRS Wireless Access Environment\nYear: 2001\n\nAbstract: The General Packet Radio Service extends the existing GSM mobile communications technology by providing packet switching and higher data rates in order to efficiently access IP-based services in the Internet. Even though Quality-of-Service notions and parameters are included in the GPRS specification, no realization path for Quality-of-Service support has been proposed. In this paper we adapt the Differentiated Services framework and apply it over the GPRS air interface in order to provide various levels of service differentiation and a true end-to-end application of the Internet Differentiated Services architecture.","Title: Inter-domain coordination models\nYear: 2012\n\nAbstract: In order for the Network Service Providers (NSPs) to provide end-to-end Quality of Service (QoS) at the inter-domain level different coordination models have been proposed by ETICS project. In this work we present and analyse the plausible alternatives of those models and we compare them with each other in terms of information asymmetry issues. We show that different information sets affect the total service provision and we present a basic model analysis on information issues by means of game-theoretic models.","Title: Application and evaluation of large deviation techniques for traffic engineering in broadband networks\nYear: 1998\n\nAbstract: Accurate yet simple methods for traffic engineering are important for efficient dimensioning of broadband networks. The goal of this paper is to apply and evaluate large deviation techniques for traffic engineering. In particular, we employ the recently developed theory of effective bandwidths, where the effective bandwidth depends not only on the statistical characteristics of the traffic stream, but also on a link's operating point through two parameters, the space and time parameters, which are computed using the many sources asymptotic. We show that this effective bandwidth definition can accurately quantify resource usage. Furthermore, we estimate and interpret values of the space and time parameters for various mixes of real traffic demonstrating how these values can be used to clarify the effects on the link performance of the time scales of burstiness of the traffic input, of the link parameters (capacity and buffer), and of traffic control mechanisms, such as traffic shaping. Our approach relies on off-line analysis of traffic traces, the granularity of which is determined by the time parameter of the link, and our experiments involve a large set of MPEG-1 compressed video and Internet Wide Area Network (WAN) traces, as well as modeled voice traffic.","Title: Incentives for content availability in memory-less peer-to-peer file sharing systems\nYear: 2005\n\nAbstract: In this paper we address the issue of content availability in p2p file sharing systems. Content availability is a public good: the copying of a file by one peer does not prevent another peer also from copying it; but contributing files to the common pool is costly. The asymptotic analysis of certain public good models for p2p file sharing suggests that when the aim is to maximize social welfare, a fixed contribution scheme in terms of the number of files shared per unity of time can be asymptotically optimal as the number of participants n grows to infinity. However, the enforcement of such an incentive scheme is not straightforward in a realistic p2p system, where no trusted software or central entity accounting for peers' transactions can be assumed and peers are free to change their identity with no cost. We present a realistic version of the fixed contribution scheme, which does not require the use of system memory but relies only on the time peers are consuming resources to ensure that they contribute adequately. We describe the functionality that should be supported for enforcement and discuss the additional incentive issues that arise in this context, proposing some practical solutions to address them. We also formulate a suitable economic model to estimate the efficiency-loss of the proposed mechanism (compared to the one achieved using the theoretically optimal schemes under complete and incomplete information) and provide some insights for the correct tuning of its basic parameters. Our first results indicate that the proposed mechanism constitutes a good compromise between economic efficiency and implementability and should lead to some interesting and practical solutions for providing incentives for content availability in p2p systems.","Title: A random walk model for studying allocation patterns in auction-based resource allocation\nYear: 2006\n\nAbstract: We consider users bidding in a series of multi-unit sealed-bid auctions, aiming at reserving the same amount of units of the resource auctioned, e.g. transmission slots in a wireless network. Each user attains from each successful allocation of resource units an instant marginal utility that depends on his history of resource allocation. The user’s bid at each auction equals this marginal utility. We introduce a random walk model for transient analysis of this series of auctions, we study the properties of the resulting user resource allocation patterns and we provide a numerical and experimental evaluation of this model.","Title: A new strategy for bidding in the network-wide progressive second price auction for bandwidth\nYear: 2005\n\nAbstract: We revisit the network-wide Progressive Second Price auction (PSP) proposed by Lazar and Semret in cite2. In this mechanism, each bidder submits the same bid in each link of the path he is interested in, taking into account the overall competition in the path. We propose a new strategy in which each bidder apportions his total bid-price in the various links, while taking into account the competition in each link separately. We have carried out a wide variety of experiments and compare our strategy with the original one with respect to efficiency and bidders' net benefit. We show that our strategy yields an outcome closer to the optimal social welfare as well as higher expected net benefit for bidders in most of the cases.","Title: Memory Efficient Algorithms for the Verification of Temporal Properties\nYear: 1990\n\nAbstract: This paper addresses the problem of designing memory efficient algorithms for the verification of temporal properties of finite-state programs. Both the programs and their desired temporal properties are modeled as automata on infinite words (Büchi automata). Verification is then reduced to checking the emptiness of the automation resulting from the product of the program and the property. This problem is usually solved by computing the strongly connected components of the graph representing the product automaton. Here, we present algorithms which solve the emptiness problem without explicitly constructing the strongly connected components of the product graph. By allowing the algorithms to err with a small probability, we can implement them with a randomly accessed memory of size O(n) bits, where n is the number of states of the graph, instead of O(n log n) bits which the presently known algorithms require.","Title: Model-checking for probabilistic real-time systems (extended abstract)\nYear: 1991\n\nAbstract: Model-checking is a method of verifying concurrent systems in which a state-graph model of the system behavior is compared with a temporal logic formula. This paper extends model-checking to stochastic real-time systems, whose behavior depends on probabilistic choice and quantitative time. The specification language is TCTL, a branching-time temporal logic for expressing real-time properties. We interpret the formulas of the logic over generalized semi-Markov processes. Our model can express constraints like “the delay between the request and the response is distributed uniformly between 2 to 4 seconds”.","Title: Tariffs, Mechanisms and Equilibria at a Single Internet Link\nYear: 2009\n\nAbstract: We analyze the interplay between the demand for downloads, choice of congestion control mechanism, and tariff structure at a single link, when users have preferences in terms of average download delay and they are charged according to the number of ECN marked packets they receive. Our model involves a timescale separation approach, where in the fast timescales active flows compete for instantaneous bandwidth share by optimally tuning congestion control parameters in a noncooperative fashion. This is modeled by letting flows choose utility functions within the network utility maximization framework laid down by Kelly [1]. On a slower timescale, users selfishly change their otherwise unrestricted demand for downloads based on the average experienced download delay and charges incurred. We study the equilibrium of this loop of interactions from the point of view of social welfare. For homogeneous users we find that optimal equilibria are induced when they choose among linear utility functions, while this is not the case for logarithmic, i.e., proportionally fair congestion controllers. We next consider two types of users, web-browsing and bittorrent, where the latter are much less sensitive to download delay than the former. If bittorrent users respond to congestion according to proportionally fair utility functions the charge induced by ECN marked packets does not provide the correct signal for service differentiation and the resulting equilibrium is suboptimal. On the other hand, if flows are charged for the volume of bits they transfer as well, then social welfare maximum is attained for some price per bit. This reveals a new role for bit volume pricing: to provide bittorrent users the correct incentives for choosing congestion controllers that effectively give priority to web-browsing users.","Title: Dynamic Bandwidth Pricing: Provision Cost, Market Size, Effective Bandwidths and Price Games\nYear: 2008\n\nAbstract: Nowadays, in the markets of broadband access services, traditional contracts are of \"static\" type. Customers buy the right to use a specific amount of resources for a specific period of time. On the other hand, modern services and applications render the demand for bandwidth highly variable and bursty. New types of contracts emerge (\"dynamic contracts\") which allow customers to dynamically adjust their bandwidth demand. In such an environment, we study the case of a price competition situation between two providers of static and dynamic contracts. We investigate the resulting reaction curves, search for the existence of an equilibrium point and examine if and how the market is segmented between the two providers. Our first model considers simple, constant provision costs. We then extend the model to include costs that depend on the multiplexing capabilities that the contracts offer to the providers, taking into consideration the size of the market. We base our analysis on the theory of effective bandwidths and investigate the new conditions that allow the provider of dynamic contracts to enter the market.","Title: Fair background data transfers of minimal delay impact\nYear: 2012\n\nAbstract: In this paper we present a methodology for the design of congestion control protocols for background data transfers that have a minimal delay impact on short TCP transfers and compete for a target share of the leftover average capacity with other background TCP transfers. We analytically compute the optimal policy and show that it's delay performance can be well approximated by a weighted TCP policy, which maintains a target proportion of TCPs bandwidth at all times in order to achieve the same share of excess capacity. The relative approximation error is always less than 17.2% while it quickly decreases to zero as the number of coexisting background TCP flows increases. Next, we consider a general utility-based fairness criterion for sharing the leftover average capacity, including a penalty term capturing the delay impact to short flows. The criterion is jointly optimized over all allocations of excess capacity to background flows (including TCP ones) on long timescales, and all bandwidth sharing policies on fast time scales. Even though the delay optimal sharing policy that solves the above optimization problem does not lead to distributed congestion control algorithms and more significantly, requires knowledge of the number of competing background TCP flows, both problems disappear under the weighted TCP policy. A distributed weight adjustment policy is considered where, at equilibrium, the overall performance is nearly optimal, with a vanishing relative optimality error as the number of background TCP flows increases. We illustrate the methodology by giving two examples of congestion control algorithms for background transfers. Both achieve low delay for short flows relative to TCP, but at the same time they present strong incentives for adoption against incumbent low priority solutions in public environments.","Title: Extending Promela and Spin for Real Time\nYear: 1996\n\nAbstract: The efficient representation and manipulation of time information is key to any successful implementation of a verification tool. We extend the syntax and semantics of the higher level specification language Promela to include constructs and statements based on the model of timed Büchi automata [2]. We implement these extensions on top of the verification tool Spin.","Title: Auction-based resource reservation in 2.5/3G networks\nYear: 2004\n\nAbstract: We consider UMTS networks in which users request services other than telephony that last for long time intervals: e.g., video clips that last for several minutes. The duration of network time-slots over which resource units are allocated is much shorter. This complicates consistent reservation of resources over longer time scales, where consistent reservation is required to ensure that service quality is constant throughout the entire service session. In this paper, we define an auction-based mechanism for nearly consistent reservation of the resources of a UMTS (or GPRS) network by the users that value them the most, in order to satisfy the longer time scale requirements of their service sessions. Each of these sessions has a fixed target bit-rate. The mechanism is based on a series of Generalized Vickrey Auctions and a set of predefined user utility functions that we propose. Bidding is performed automatically on behalf of the users on the basis of each user's selection of one of these utility functions and his declaration of a total willingness to pay. We argue that under our mechanism the user does not have a clear incentive of not performing a truthful selection of a bidding function according to his own utility. The utility functions we define express appropriately the preferences of the users with respect to the resource allocation pattern in the cases where perfectly consistent allocation cannot be attained. We also provide a mapping of these functions to the UMTS service classes. The effectiveness of our resource reservation mechanism is demonstrated by means of experiments. It appears that most of the users either are served very satisfactorily or essentially are not served at all. The mechanism is implemented at the network base station, and is applicable in practical cases of networks with large numbers of users whose sessions last for many slots.","Title: Pricing differentiated services in the GPRS environment\nYear: 2001\n\nAbstract: The General Packet Radio Service extends the existing GSM mobilecommunications technology by providing packet switching and higherdata rates in order to efficiently access IP-based services in theInternet. Since no realization path for Quality-of-Service supporthas been proposed yet, we adapt the Differentiated Servicesframework and apply it over the GPRS air interface in order toprovide various levels of service differentiation. We also focus onapplying a charging technique so as to publish a unit price foreach service class. These prices are designed to lead to themaximization of Social Welfare and the users' net benefit.","Title: Model-checking in dense real-time\nYear: 1993\n\nAbstract: . Model-checking is a method of verifying concurrent systems in whicha state-transition graph model of the system behavior is compared with a temporallogic formula. This paper extends model-checking for the branching-time logic CTLto the analysis of real-time systems, whose correctness depends on the magnitudesof the timing delays. For specifications, we extend the syntax of CTL to allowquantitative temporal operators such as 93!5 , meaning "possibly within 5 timeunits." The formulas of...","Title: GridEcon - the economic-enhanced next-generation internet\nYear: 2007\n\nAbstract: The major shortcoming of Grid middleware systems is the lack of economic-enhanced Grid services. These new services are necessary in order to let Grid users benefit from the properties of the Grid. Those properties comprise the availability of on-demand computational power, simplicity of access to resources, low cost of ownership, and a pay-for-use pricing model in addition to the already leveraged properties such as cost reduction and aggregated processing power for high-performance computing applications. This paper gives an overview of the EU-funded project GridEcon on Grid economics and business models. It describes its vision of the next generation Grid/Internet, in which individuals, universities, small and medium sized enterprises (SMEs), and large companies have access to the Grid in exactly the same way. Any resource, including servers, storage, software, or data, is accessible as a service. In addition to this, the architecture of an economic-enhanced infrastructure is illustrated and the goal of the project is described.","Title: Resource Control For Loss-Sensitive Traffic In Cdma Networks\nYear: 2004\n\nAbstract: We investigate the problem of efficient resource control for loss-sensitive traffic in CDMA networks, using an economic modelling framework that takes into account the joint control of the transmission rate and the signal quality. Although the corresponding global optimization problem has a non-trivial structure, and we cannot in general guarantee that a solution can be found using the Lagrangian method, we have strong experimental evidence that this is possible for a wide range of user utilities. In this case the global optimum can be achieved in a decentralized manner, using shadow prices to influence the individual user resource requests. Based on this evidence, the main contribution of the paper is to discuss how existing rate control and outer-loop power control procedures can obtain a simple and attractive form that takes into account, through shadow prices, the level of demand and supply in order to achieve efficient resource utilization. Moreover, we describe and evaluate approximations of the proposed resource control model that can simplify its application, and we present extensions of the model for the case where the packet success ratio depends on the transmission rate in addition to the signal quality, and for network paths containing multiple wireless links.","Title: Hybrid Automata: An Algorithmic Approach to the Specification and Verification of Hybrid Systems\nYear: 1992\n\nAbstract: We introduce the framework of hybrid automata as a model and specification language forhybrid systems. Hybrid automata can be viewed as a generalization of timed automata, in whichthe behavior of variables is governed in each state by a set of differential equations. We show thatmany of the examples considered in the workshop can be defined by hybrid automata. Whilethe reachability problem is undecidable even for very restricted classes of hybrid automata, wepresent two semidecision...","Title: Congestion control and pricing of volume intensive applications\nYear: 2009\n\nAbstract: We consider fair sharing of network resources over a period of time between offline and interactive users. The former value total transferred volume while the latter have time-dependent valuations for instantaneous bandwidth. Following the work of Kelly on fair bandwidth sharing, we express the system optimum as an equilibrium of traditional congestion control taking place in a fast timescale and a volume charging mechanism in a slow timescale. The problem is very much related to multipath congestion control where paths are unfolded also in the dimension of time, and our algorithms can be regarded as multipath congestion control algorithms operating on two separate timescales that combine congestion control with scheduling. A novel feature of our algorithms is that, contrary to ordinary dual algorithms, users are not overcharged at any time while approaching the equilibrium. This is a sensible economical property not recognized before in the context of congestion control over long timescales.","Title: Enforcing efficient resource provisioning in peer-to-peer file sharing systems\nYear: 2006\n\nAbstract: In this paper we focus on content availability as the main good provided by a p2p file sharing system and consider peer availability, the amount of time peers stay connected sharing their files, as their main contribution. We wish to study the effectiveness of incentive mechanisms which enforce contribution by somehow relating it with consumption. Towards this end, we propose a mechanism who wishes to regulate the time required for peers to stay on-line sharing their files by dictating a fixed upload throughput used by uploading peers and thus a certain average time for a download to finish. We formulate and analyze a suitable economic model focusing on peer availability in order to understand the role of this important system parameter and provide the means to efficiently tune it.","Title: An efficient auction-based mechanism for hierarchically structured bandwidth markets\nYear: 2006\n\nAbstract: We consider a hierarchical business model for selling bandwidth in a single link. In the top level of the market the social planner allocates bandwidth to intermediate providers, who in turn allocate their assigned shares of bandwidth to the customers in the lower level. We propose an innovative mechanism comprising an appropriate auction in each level; the payment rule of the lower-level auction is imposed by the social planner. We establish that the bidders of all auctions have the incentives both to participate uninhibitively and to bid truthfully, as well as that overall efficiency is ultimately attained. We also investigate the properties of the mechanism in case where each of the competing intermediate providers can select the payment rule on his own, under a variety of assumptions on the information possessed by the various players as well as on their level of freedom to act in the market. For one of the cases analyzed we prove that providers have no incentive to deviate from the original payment rule.","Title: Procedures and tools for analysis of network traffic measurements\nYear: 2002\n\nAbstract: We present procedures and tools for the analysis of network traffic measurements. The tools consist of stand-alone [1] modules that implement advanced statistical analysis procedures and a flexible web-based interface through which a user can create, modify, save, and execute experiments using the statistical analysis modules. The tools do not assume a specific source traffic model, but rather process actual measurements of network traffic. Indeed, the theory that the tools are based on can identify the time-scales that affect a link's performance, hence suggest the appropriate time granularity of traffic measurements. We present and discuss case studies that demonstrate the application of the tools for answering questions related to network management and dimensioning, such as the maximum link utilization when some quality of service is guaranteed, how this utilization is affected by the link buffer and traffic shaping, the acceptance region and the effects of the scheduling discipline, and the token (or leaky) bucket parameters of a traffic stream. The case studies involve actual traffic measurements obtained by a high performance measurement platform that we have deployed at the University of Crete network.","Title: Economic issues in shared infrastructures\nYear: 2012\n\nAbstract: In designing and managing a shared infrastructure, one must take account of the fact that its participants will make self-interested and strategic decisions about the resources that they are willing to contribute to it and/or the share of its cost that they are willing to bear. Taking proper account of the incentive issues that thereby arise, we design mechanisms that, by eliciting appropriate information from the participants, can obtain for them maximal social welfare, subject to charging payments that are sufficient to cover costs. We show that there are incentivizing roles to be played both by the payments that we ask from the participants and the specification of how resources are to be shared. New in this paper is our formulation of models for designing optimal management policies, our analysis that demonstrates the inadequacy of simple sharing policies, and our proposals for some better ones. We learn that simple policies may be far from optimal and that efficient policy design is not trivial. However, we find that optimal policies have simple forms in the limit as the number of participants becomes large.","Title: Adopting the grid for business purposes: the main objectives and the associated economic issues\nYear: 2007\n\nAbstract: Grid technology offers numerous opportunities for the players involved. Despite the fact that the academic community has already exploited many of them, there is an evident reluctance from the business community to act likewise. Recent analysis reveals that the problem lies in overcoming certain business barriers rather than technological ones. At this stage understanding the real-life economic issues from a business perspective is deemed as more important than gaining understanding of complex theoretical economical problems, such as those related to accounting or resource sharing mechanisms especially in cases where the players do not exhibit the required technological expertise. This paper is stimulated from interaction with players from the industry and aims to fill this gap. In particular, we identify and evaluate a number of economic issues that should be taken into consideration by industrial players so that their trust and confidence in the adoption of this promising technology be increased.","Title: Market Mechanisms for Trading Grid Resources\nYear: 2008\n\nAbstract: There has been recently an increasing interest in Grid services and economic-aware Grid systems both in the industry and the academia. In this paper we specify a market for hardware providers and consumers interested in leasing Grid resources for a time period. Our approach comprises a stock-market like mechanism that enables the trading of computational power on the basis of a spot and a futures market. The spot market comprises a pair of bid and ask queues. This grid market is more complicated than the standard spot/futures markets of storable commodities, because the computational service traded in our case comprises of resources that are perishable, and has both quantity and duration specified in terms of a time interval. This is an important feature of our market mechanism, complicating considerably the trading algorithms that we develop and assess in this paper.","Title: Traffic equivalence and substitution in a multiplexer\nYear: 1999\n\nAbstract: For a multiplexer fed with a large number of sources, we derive conditions under which a single source can be substituted for a given subset of the sources while preserving the buffer overflow probability and the dominant time scales of buffer overflows. This equivalence is stronger than simple effective bandwidth equality and takes into account the context in which multiplexing takes place. This allows a substitution to be made for arbitrarily large proportions of the traffic without changing the operating point of the multiplexer as experienced by the rest of the traffic. It corresponds to defining a single source which is equivalent in a sense “local” to a given context, rather than equivalent in a sense which is “universal” to all contexts. The proposed methodology does not rely on traffic models and obtains the necessary information from the actual traffic traces. We study the case of fractional Brownian motion as a single source substitute and provide theoretical and experimental results that validate our approach","Title: On Optimal Packing of Randomly Arriving Objects\nYear: 1991\n\nAbstract: Objects of finitely many types arrive at a facility according to independent stationary Poisson processes. The objects are to be placed in boxes as they arrive. There are finitely many types of boxes and each box type has its own packing configuration. There is an unlimited supply of boxes of each type and once an object is placed in a box it cannot be moved to another box in the future. When a box is completely filled it produces a reward which depends on its configuration. The objective is to select a rule for placing the arriving objects in boxes so as to maximize expected reward. We show how to construct optimal and ε-optimal policies under which the expected number of partially packed boxes does not explode.","Title: Comparing economic incentives in peer-to-peer networks\nYear: 2004\n\nAbstract: Users who join a peer-to-peer network have, in general, suboptimal incentives to contribute to the network, because of the externalities that exist between them. The result is an inefficient network where the overall levels of contribution are less than would be the case if each peer acted in the interests of the entire network of peers. Incentives provided in the form of prices or contribution rules that require no money transfers can play an important role in reducing these inefficiency effects. The problem in designing such incentive schemes is information: Designing an optimal incentive scheme requires complete knowledge of the types and preferences of the individual peers and their identities. In this paper we discuss the above issues in terms of a simple but representative example by introducing the basic economic concepts and models. We then investigate the practical issue of designing several simpler incentive schemes requiring less information and compare their efficiency loss to the optimal. We show using numerical analysis that these schemes converge to a fixed proportion of the full information optimal as the number of peers in the network becomes large. This result means that it is not necessary to collect large amounts of information, or to undertake complicated calculations, in order to implement the correct incentives in a large peer-to-peer network.","Title: An Algebra of Boolean Processes\nYear: 1991\n\nAbstract: This work has been motivated by the study of the S=R models which allow torepresent systems as a set of communicating state machines cooperating through ashared memory.We show that S=R models can be expressed in terms of a process algebra calledBoolean SCCS which is a special case of Milner's SCCS, in the sense that theactions are elements of some boolean algebra. We define for Boolean SCCS anoperational and a symbolic semantics modulo strong bisimulation equivalence. Acomplete...","Title: Computing Accumulated Delays in Real-time Systems\nYear: 1993\n\nAbstract: We present a verification algorithm for duration properties ofreal-time systems. While simple real-time properties constrain thetotal elapsed time between events, duration properties constrain theaccumulated satisfaction time of state predicates. We formalize theconcept of durations by introducing duration measures for timedautomata. A duration measure assigns to each finite run of a timedautomaton a real number —the duration ofthe run— which may be the accumulated satisfaction time of astate predicate along the run. Given a timed automaton with aduration measure, an initial and a final state, and an arithmeticconstraint, the duration-bounded reachability problem asks if there is a run of the automaton from theinitial state to the final state such that the duration of the runsatisfies the constraint. Our main result is an (optimal) PSPACE decision procedure for the duration-bounded reachability problem.","Title: The GridEcon Platform: A Business Scenario Testbed for Commercial Cloud Services\nYear: 2009\n\nAbstract: Within this paper, we present the GridEcon Platform, a testbed for designing and evaluating economics-aware services in a commercial Cloud computing setting. The Platform is based on the idea that the exact working of such services is difficult to predict in the context of a market and, therefore, an environment for evaluating its behavior in an emulated market is needed. To identify the components of the GridEcon Platform, a number of economics-aware services and their interactions have been envisioned. The two most important components of the platform are the Marketplace and the Workflow Engine. The Workflow Engine allows the simple composition of a market environment by describing the service interactions between economics-aware services. The Marketplace allows trading goods using different market mechanisms. The capabilities of these components of the GridEcon Platform in conjunction with the economics-aware services are described in this paper in detail. The validation of an implemented market mechanism and a capacity planning service using the GridEcon Platform also demonstrated the usefulness of the GridEcon Platform.","Title: A marketplace and its market mechanism for trading commoditized computing resources.\nYear: 2010\n\nAbstract: This paper presents the design and implementation of the GridEcon Marketplace. In addition to supporting a market mechanism\n for trading computing resources on a pay-per-use basis, this marketplace also provides an environment for integrating value-added\n support services. These value-added services help consumers to use the utility computing market more efficiently. The GridEcon\n Market Mechanism for virtual machines specifies in detail the unit-of-trade, the bids and asks, as well as the matching algorithm.\n The marketplace and market mechanism are validated by using the GridEcon Platform, which is a service-oriented platform for\n composing market scenarios. Our validation results show that the GridEcon Marketplace fulfills all functional requirements\n and that the GridEcon Market Mechanism is computationally and economically efficient.","Title: Peer-to-Peer Wireless LAN Consortia: Economic Modeling and Architecture\nYear: 2003\n\nAbstract: In this paper we address the incentive issues that arise in a Peer-to-Peer WLAN Consortium [1]. We explore the use of flexible rules on reciprocity to guide domain policies and develop a suitable economic model that demonstrates the basic characteristics of our system.","Title: Minimum and maximum delay problems in real-time systems\nYear: 1992\n\nAbstract: We consider a finite-state system with a finite number of clocks, where the transitions may depend on the values of the clocks, and may reset some of the clocks. We address the complexity and provide algorithms for the following problems. Suppose that the system starts from a given current state with a given assignment of values to the clocks. Can a given target state ever appear in the history of the system? What is the earliest time it can appear? What is the latest time it can appear?","Title: The algorithmic analysis of hybrid systems\nYear: 1995\n\nAbstract: We present a general framework for the formal specification and algorithmic analysis of hybrid systems. A hybrid system consists of a discrete program with an analog environment. We model hybrid systems as finite automata equipped with variables that evolve continuously with time according to dynamical laws. For verification purposes, we restrict ourselves to linear hybrid systems, where all variables follow piecewise-linear trajectories. We provide decidability and undecidability results for classes of linear hybrid systems, and we show that standard program-analysis techniques can be adapted to linear hybrid systems. In particular, we consider symbolic model-checking and minimization procedures that are based on the reachability analysis of an infinite state space. The procedures iteratively compute state sets that are definable as unions of convex polyhedra in multidimensional real space. We also present approximation techniques for dealing with systems for which the iterative procedures do not converge.","Title: Minimization of Timed Transition Systems\nYear: 1992\n\nAbstract: Without Abstract","Title: WOICE: A Decentralized System for Ubiquitous VoIP Services.\nYear: 2009\n\nAbstract: We propose WOICE (Wireless vOICE), a fully distributed architecture capable of supporting and offering ubiquitous voice services over wireless IP-based data networks. Existing WiFi infrastructure - deployed by enterprises or even individuals - will be supplemented with software modules based on open standards in order to support VoIP. Furthermore, a new breed of VoIP providers by aggregating these currently isolated access networks will be able to offer alternative and advanced services to reachable end-users. These services would in certain cases completely bypass traditional telephony and cellular providers and in others smoothly cooperate with them. We will describe the overall architecture and the necessary modules that should be in place for realizating a system that creates a marketplace for communication services.","Title: Markov Decision Processes and Regular Events (Extended Abstract)\nYear: 1990\n\nAbstract: Desirable properties of the infinite histories of a finite state Markov Decision Process are specified in terms of a finite number of events represented as -regular sets. An infinite history of the process produces a reward which depends on the properties it satisfies. We investigate the existence of optimal policies and provide algorithms for the construction of such policies.","Title: Verification of timing Properties of VHDL\nYear: 1993\n\nAbstract: This paper shows how timing properties of VHDL processes can be verified using timed transition systems. The timing model being adopted is the timed automaton model used in the timing extension of Kurshan's COSPAN system. It demonstrates how a VHDL process can be translated into a timed automaton by describing the construction of the corresponding timed process that handles the scheduled signal assignments of the VHDL specification. Verification is performed in the case in which the complement of the timing properties to be verified are provided in terms of a timed automaton. Interestingly enough, this is the case for a large class of hardware properties expressed in terms of timing diagrams.","Title: Verification with Real-Time COSPAN\nYear: 1992\n\nAbstract: We describe some examples using an extension of Kurshan's COSPAN system to verify bounded delay constraints in a dense time model, based on the method proposed by Dill for adding timing constraints to Büchi automata. The S/R model and COSPAN are reviewed as background, then we describe how timing can be incorporated into S/R processes, and briefly describe the modified verification algorithm. The examples consist of several time-dependent versions of the Alternating Bit Protocol and the Fiber Distributed Data Interface (FDDI).","Title: Stabilizing an uncertain production system\nYear: 1989\n\nAbstract: Consider a production system that consists ofm machines each of which can produce parts ofn types. When machinek is used, it produces a part of typei with probabilitypki. Requests arrive for parts, one at a time. With probability ?i an arriving request is for a part of typei. The requests must be served without waiting. Thus, if a requested part is not available, it must be produced. We find necessary and sufficient conditions for the existence of a strategy (a choice of the machines to be used) which makes the inventory of parts stable and we provide such a strategy.","Title: Perfect Packing Theorems and the Average-Case Behavior of Optimal and Online Bin Packing\nYear: 2002\n\nAbstract: We consider the one-dimensional bin packing problem under the discrete uniform distributions $U\\{j,k\\}$, $1 \\leq j \\leq k-1$, in which the bin capacity is $k$ and item sizes are chosen uniformly from the set $\\{1,2,\\ldots,j\\}$. Note that for $0","Title: The Observational Power of Clocks\nYear: 1994\n\nAbstract: We develop a theory of equivalences for timed systems. Two systemsare equivalent iff external observers cannot observe differences in their behavior.The notion of equivalence depends, therefore, on the distinguishingpower of the observers. The power of an observer to measure timeresults in untimed, clock, and timed equivalences: an untimed observercannot measure the time difference between events; a clock observer usesa clock to measure time differences with finite precision; a timed...","Title: The complexity of probabilistic verification\nYear: 1995\n\nAbstract: We determine the complexity of testing whether a finite state, sequential or concurrent probabilistic program satisfies its specification expressed in linear-time temporal logic. For sequential programs, we present an algorithm that runs in time linear in the program and exponential in the specification, and also show that the problem is in PSPACE, matching the known lower bound. For concurrent programs, we show that the problem can be solved in time polynomial in the program and doubly exponential in the specification, and prove that it is complete for double exponential time. We also address these questions for specifications described by &ohgr;-automata or formulas in extended temporal logic.","Title: Service differentiation and performance of weighted window-based congestion control and packet marking algorithms in ECN networks\nYear: 2003\n\nAbstract: We investigate the service differentiation, in terms of average throughput, and the performance achieved using weighted window-based congestion control in networks supporting Explicit Congestion Notification (ECN). Our results show how service differentiation, queueing delay, and average throughput are affected by the increase and decrease rules of the end-system congestion control algorithms, and how they depend on the marking algorithms operating in the routers. The end-system algorithms we investigate include Willingness-To-Pay (willingness-to-pay) and MulTCP congestion control, and the packet marking algorithms include RED, virtual queue marking, and load-based marking. Our investigations consider both single and multiple link topologies, and connections with different round trip times.","Title: GridEcon: A Market Place for Computing Resources\nYear: 2008\n\nAbstract: This paper discusses the rationales for a Grid market and, in particular, the introduction of a market place for trading commoditized computing resources. The market place proposed makes computing resources from different providers substitutable through virtualization. This includes the definition of a spot and future market as well as the parameters that a market mechanism for computing resources should consider. The above market place is complemented by a set of value-added services (e.g. insurance against resource failures, capacity planning, resource quality assurance, stable price offering) that ensure quality for Grid users over time. The market place technology for all of the above services has been designed by the GridEcon project, contributing to a broader adoption of Grid technology and enabling a service-oriented knowledge utility environment.","Title: Application of the many sources asymptotic and effective bandwidths to traffic engineering\nYear: 1999\n\nAbstract: Accurate yet simple methods for traffic engineering are important for efficient management of resources in broadband networks. The goal of this paper is to apply and evaluate large deviation techniques for traffic engineering. In particular, we employ the recently developed theory of effective bandwidths, where the effective bandwidth depends not only on the statistical characteristics of the traffic stream, but also on a link's operating point through two parameters, the space and time parameters, which can be computed using the many sources asymptotic. We show that this effective bandwidth definition can accurately quantify resource usage. Furthermore, we estimate and interpret values of the space and time parameters for various mixes of real traffic demonstrating how these values can be used to clarify the effects on the link performance of the time scales of traffic burstiness, of the link resources (capacity and buffer), and of traffic control mechanisms such as traffic shaping. Our experiments involve a large set of MPEG¿1 compressed video and Internet Wide Area Network (WAN) traces, as well as modeled voice traffic.","Title: A study of simple usage-based charging schemes for broadband networks\nYear: 2000\n\nAbstract: Operators of multi service networks require simple charging schemes with which they can fairly recover costs from their users and effectively allocate network resources. This paper studies an approach for computing such charges from simple measurements (time and volume), and relating these to bounds of the effective bandwidth. To achieve economic efficiency, it is necessary that usage based charging schemes capture the relative amount of resources used by connections. Based on this criteria, we evaluate our approach for real traffic consisting of Internet Wide Area Network traces and MPEG 1 compressed video. Its incentive compatibility is shown with an example involving deterministic multiplexing, and the effect of pricing on a network's equilibrium is investigated for deterministic and statistical multiplexing. Finally, we investigate the incentives for traffic shaping provided by the approach.","Title: Socio-economic challenges for the internet of the future: the case of congestion control\nYear: 2010\n\nAbstract: The Internet is founded on a very simple premise: sharing! Shared communications links are more efficient than dedicated connections that lie idle much of the time. Hence the rules we use for sharing are extremely vital for the healthy operation of the Internet ecosystem and directly affect the value of the network to its users. It becomes a great paradigm of merging the disciplines of computer science and economics, and presents a great number of challenges to the Internet research community. In this talk we will discuss a number of questions like: what is wrong with todays Internet sharing technologies? Are these consistent with economics? More specifically, is TCP (Transmission Control Protocol) sensible from an economic point of view? Which network sharing technologies justify end-toend from an economics perspective? What is required to make P2P (peer-topeer) a blessing instead of a curse? Are there bad applications or just inefficient combinations of sharing technologies and pricing schemes? This keynote will survey some of these systems management issues and the emerging approaches to solve them.","Title: An approach to investigating socio-economic tussles arising from building the future internet\nYear: 2011\n\nAbstract: With the evolution of the Internet from a controlled research network to a worldwide social and economic platform, the initial assumptions regarding stakeholder cooperative behavior are no longer valid. Conflicts have emerged in situations where there are opposing interests. Previous work in the literature has termed these conflicts tussles. This article presents the research of the SESERV project, which develops a methodology to investigate such tussles and is carrying out a survey of tussles identified within the research projects funded under the Future Networks topic of the FP7. Selected tussles covering both social and economic aspects are analyzed also in this article.","Title: An Innovative Decision Support System for Bidding in Simultaneous Ascending Auctions\nYear: 2005\n\nAbstract: We present a powerful decision support system for bidding in simultaneous ascending auctions (SAA). Such auctions are commonly used for the allocation of spectrum licenses and electricity generation contracts. Our system can simulate the SAA under a variety of assumptions on the participating bidders and market conditions while it comprises a multitude of innovative features: a) a formal definition of bidding strategies for the user to choose from, including an original strategy that we propose, ii) means of incorporating a bidder's prior knowledge about market demand, iii) functionality for introducing beliefs about rival bidders' strategies, and iv) provision of on-line recommendation about the bids to be placed in a SAA in progress. We also present a set of experiments that corroborate the usefulness of our system.","Title: The Case of Multi-hop Peer-to-Peer Implementation of Mobile Social Applications\nYear: 2006\n\nAbstract: In this paper we consider issues related to the design of mobile social software in a multi-hop peer-to-peer environment. We believe that by enabling multi-hop peer-topeer communication could in certain cases make social mobile applications more attractive in terms of location privacy or embarassement, but also more effective in terms of the quality and quantity of the matches achieved. Such an approach could thus contribute in the construction of a critical mass which is necessary for the success of these applications and probably one of the reasons they are not yet widely deployed. On the other hand, it introduces significant implemention challenges related to the contruction and management of a dynamic ad-hoc network. We focus on the provision of the suitable incentives for packet forwarding which is a fundamental problem that needs to be addressed in this context since the ad-hoc nature of the mobile social applications constitutes pricing and reputation (or accounting) based approaches of limited applicability. In this paper we argue that suitable memory-less mechanisms should be devised sketching two such mechanisms for different types of packets exchanged in a mobile social application. Interestingly, as we explain, our proposed mechanisms enforcing contribution while consuming have a more general applicability and could also be used in other applications such as the peer-to-peer resource sharing in ubiquitous computing. We finally discuss several strategic issues that should be also taken into account but are ignored by the corresponding literature and more specifically the consideration of mobility and power transimission as part of the rational strategy of peers participaing in such a system.","Title: Economic aspects of building software for service-oriented architectures\nYear: 2010\n\nAbstract: The concept of service-oriented architectures (SOA) has recently emerged as a design principle for the next generation of IT solutions. The main idea behind SOA lies in treating software applications as composed of simpler, cooperating services, implemented by components that communicate over networks through open standards. Such a modular structure is expected to question the effectiveness of current business models for building and distributing software, which generate revenue through license fees. Under SOA, the license fee is replaced by more flexible schemes, whereby users are charged according to their actual usage of services and the corresponding hardware infrastructure. Our work constitutes a first step toward exploring the economic aspects of a market for software components, as well as the incentives of software manufacturers to support this trend. In particular, we examine the factors that affect profitability in an open market for services, we build simple models to predict and explain market growth and we also suggest ways to accelerate this growth, while also achieving a higher level of economic efficiency. Copyright © 2009 John Wiley & Sons, Ltd.","Title: From advertising profits to bandwidth prices: A quantitative methodology for negotiating premium peering.\nYear: 2014\n\nAbstract: We have developed a first of its kind methodology for deriving bandwidth prices for premium direct peering between Access ISPs (A-ISPs) and Content and Service Providers (CSPs) that want to deliver content and services in premium quality. Our methodology establishes a direct link between service profitability, e.g., from advertising, user- and subscriber-loyalty, interconnection costs, and finally bandwidth price for peering. Unlike existing work in both the networking and economics literature, our resulting computational model built around Nash bargaining, can be used for deriving quantitative results comparable to actual market prices. We analyze the US market and derive prices for video that compare favorably with existing prices for transit and paid peering. We also observe that the fair prices returned by the model for high-profit/low-volume services such as search, are orders of magnitude higher than current bandwidth prices. This implies that resolving existing (fierce) interconnection tussles may require per service, instead of wholesale, peering between A-ISPs and CSPs. Our model can be used for deriving initial benchmark prices for such negotiations.","Title: An auction mechanism for allocating the bandwidth of networks to their users\nYear: 2007\n\nAbstract: We present a mechanism for auctioning bandwidth on a network-wide basis to end users or ISPs that will utilize it for the same time period. This mechanism consists of a set of simultaneous multi-unit descending-price (i.e. Dutch) auctions, one per link of the network. The per unit prices of bandwidth at the various links are asymmetric, thus reflecting the asymmetry of demand for these links. A user can be instantly allocated bandwidth over a certain path, by simultaneously bidding for the quantity desired at all relevant auctions. This winner determination rule is complemented by a payment rule of the VCG (Vickrey-Clarke-Groves) type, which provides users with the incentive to bid truthfully, thus simplifying bidding. Also, the mechanism enables the auctioneer to use his prior information on market demand anticipated and its spreading among the various links in order to set effectively the auction's parameters. We argue that our mechanism attains nearly efficient allocation of the network's bandwidth (i.e. the resulting social welfare is close to the respective maximum for the quantity decided to be sold by the auctioneer), while it is simple, scalable and applicable to real networks, even for auctioning the capacity of links owned by multiple providers and then splitting the revenue among them. Alternatively, the mechanism offers the provider the opportunity to optimize his revenue, rather than the social welfare. Since our mechanism's computational complexity is low it can serve as a fast, practical, and near-optimal solution to a generally NP-hard optimization problem.","Title: Simple tariffs based on price multipliers for ATM VBR services.\nYear: 2003\n\nAbstract: We present an approach for constructing simple time-based tariffs for variable bit-rate (VBR) connections from constant bit-rate (CBR) prices, using price multipliers that depend on the traffic parameters of the VBR connection and reflect the resource usage of VBR connections relative to CBR connections. Our approach employs an effective bandwidth bound as a proxy for the maximum amount of traffic that conforms to the connection's traffic contract, and thus for the maximum resource usage. We compare the price multipliers computed by our approach with the multipliers published by an actual ATM service provider, and with those computed using a proxy for resource usage that is based on ATM Forum's generic CAC algorithm. Although our approach is presented in the context of ATM VBR services, it can be applied for creating tariffs for service level agreements (SLAs) where the maximum amount of conforming traffic is given by leaky (or token) bucket constraints. Copyright (C) 2003 AEI.","Title: A Framework for Analyzing the Economics of a Market for Grid Services\nYear: 2009\n\nAbstract: This paper provides a single broad model for the analysis of a range of issues underlying a market for Grid services. The demand and the supply sides of such a market are being treated separately and the relation between the two sides is being studied. We provide numerical results in order to derive conclusions about the viability of a market for Grid services. Underlying our model are parameters such as the cost technologies, the random processes driving demand and supply and the size of the market. We study the effect of the model's parameters, such as risk aversion or the durability of resources, on the system's behavior, eg. on the clearing price or volume of trade.","Title: Resource Control for the EDCA Mechanism in Multi-Rate IEEE 802.11e Networks\nYear: 2006\n\nAbstract: We investigate the problem of efficient resource control for elastic traffic in IEEE 802.11e's Enhanced Distributed Channel Access (EDCA) mechanism. Our approach considers an economic modelling framework based on congestion pricing that captures how various factors, such as the probability of attempting to transmit a frame, the use of the basic CSMA/CA or the RTS/CTS procedure, and the physical layer transmission rate, contribute to congestion. We discuss the application of the framework for achieving classbased throughput differentiation, for performing explicit congestion notification (ECN) marking based on the level of congestion in the wireless channel, and for modelling the performance of TCP congestion control over EDCA. In these application scenarios we discuss how to estimate the optimum minimum contention window and the congestion prices based on the 802.11e parameters and actual measurements, in order to achieve efficient channel utilization.","Title: Service Differentiation in ECN Networks Using Weighted Window-Based Congestion Control for Various Packet Marking Algorithms\nYear: 2001\n\nAbstract: We investigate the service differentiation, in terms of average throughput, and the performance achieved using weighted window-based congestion control in networks supporting Explicit Congestion Notification (ECN). Our results show how service differentiation, queueing delay, and average throughput are affected by the increase and decrease rules of the end-system congestion control algorithms, and how they depend on the marking algorithms operating in the routers. The end-system algorithms we investigate include algorithms that achieve an average throughput proportional to some willingness-to-pay, and the marking algorithms include RED, virtual queue marking, and load-based marking.","Title: An evaluation of pricing schemes that are based on effective usage\nYear: 1998\n\nAbstract: We evaluate usage-based pricing schemes that are based on bounds of the effective bandwidth. The schemes include ones that require simple measurements (time and volume) for the whole duration of a call, and two schemes that require measurements in distinct time intervals, smaller than the duration of a call: pricing: with renegotiation and the virtual bucket scheme. Using MPEG-1 compressed video, the schemes are compared for different link capacities and buffer sizes in terms of their fairness, i.e., the ability to capture the relative amount of resources used by connections.","Title: Design and experimental evaluation of market mechanisms for participatory sensing environments\nYear: 2014\n\nAbstract: Participatory Sensing concerns the sharing of sensor information within user communities, forming a body of knowledge that can be beneficial to the community itself, either directly or through specialized applications. We introduce a framework for a marketplace where such applications can sell and buy sensor information. We focus on the buyers' side and we use various ideas from the cost-sharing literature, to propose three classes of mechanisms. We evaluate them experimentally, comparing their performance according to metrics such as social efficiency, cost coverage and budget deficit, as well as metrics related to encouraging participation.","Title: Designing incentive compatible protocols for background data transfers\nYear: 2014\n\nAbstract: We consider the design of transport protocols for background transfers with the aim of reducing the download delays of interactive (non-background) flows which share the same links. We demonstrate that Lower than Best Effort (LBE) protocols suffer from bandwidth starvation, a fact which not only creates adoption disincentives but can have a harmful effect on the interactive flows as well. Instead we propose a simple access control policy which limits the maximum number of simultaneously transmitting background flows. This maximum limit is dynamically adjusted in order to prevent bandwidth starvation thus eliminating the disincentives present in LBE protocols. Through simulations we show that our policy creates lower transfer delay for the interactive flows, as compared to that under LEDBAT [6] and TCP-LP [8] LBE protocols.","Title: Economic incentives for adopting congestion accountability protocols\nYear: 2014\n\nAbstract: New Internet protocols such as ECN, re-ECN and Conex, provide valuable information to ISPs about the congestion within a network. Such information could be useful in allocating network resources more equitably, as well as to employ new pricing schemes. However, in designing a new protocol one must take proper account of the incentive issues that thereby arise for its adoption. In this paper, we consider a duopoly market consisting of two access ISPs and formulate a noncooperative game over their choices of pricing strategies when one of them uses volume-based and the other congestion-based pricing. We prove, under mild conditions, the existence of a pure strategy Nash equilibrium for the competing providers. We further examine how the different types of transit charges affect the pricing strategies of the access ISPs. We find cases where the volume charging ISP is forced to leave the access market. We conclude that ISPs have economic incentives to adopt congestion accountability mechanisms, since they become more competitive due to the smoother and more predictable traffic they have to accommodate and to the lower transit charges they incur.","Title: Model-checking for real-time systems\nYear: 1990\n\nAbstract: This research extends CTL model-checking to the analysis of real-time systems, whose correctness depends on the magnitudes of the timing delays. For specifications, the syntax of CTL is extended to allow quantitative temporal operators. The formulas of the resulting logic, TCTL, are interpretation over continuous computation trees, trees in which paths are maps from the set of nonnegative reals to system states. To model finite-state systems the notion of timed graphs is introduced-state-transition graphs extended with a mechanism that allows the expression of constant bounds on the delays between the state transition. As the main result, an algorithm is developed for model checking, that is, for determining the truth of a TCTL formula with respect to a timed graph. It is argued that choosing a dense domain, instead of a discrete domain, to model time does not blow up the complexity of the model-checking problem. On the negative side, it is shown that the denseness of the underlying time domain makes TCTL II11-hard. The question of deciding whether a given TCTL formula is implementable by a timed graph is also undecidable","Title: Decoupling bandwidths for networks: a decomposition approach to resource management\nYear: 1994\n\nAbstract: We consider large buffer asymptotics for feed-forward networks of discrete-time queues with deterministic service rate shared by multiple classes of streams subject to work conserving service policies. First we review the concept of effective bandwidths for traffic streams sharing a common buffer subject to subject to tail constraints on the workload distribution. Next, we obtain the effective bandwidth of the departure process from such a queue, proving that in fact the effective bandwidth of the output is at worst equal to that of the input, and depending on the service rate, strictly less than that of the input. We then define the notion of a decoupling bandwidth and the associated constraints, guaranteeing that asymptotics within the network are decoupled. These results provide a framework for call admission schemes which are sensitive to constraints on the tail distri- bution of the workload or approximate cell loss probabilities. Our results require relatively weak assumptions on both the traffic streams and service policies. We consider the problem of \"optimal\" traffic shaping (via buffering) subject to a loss constraint. Finally, we discuss our results in the context of resource management for ATM networks.","Title: Cost-Sharing Models in Participatory Sensing\nYear: 2015\n\nAbstract: In Smart City and Participatory Sensing initiatives the key concept is for user communities to contribute sensor information and form a body of knowledge that can be exploited by innovative applications and data analytics services. A key aspect in all such platforms is that sensor information is not free but comes at a cost. As a result, these platforms may suffer due to insufficient sensor information made publicly available if applications do not share efficiently the cost of the sensor information they consume. We explore the design of specialized market mechanisms that match demand to supply while taking into account important positive demand externalities: sensors are digital goods and their cost can be shared by applications. We focus on the buyer side and define different demand models according to the flexibility in choosing sensor data for satisfying application needs. We then investigate the properties of various costsharing mechanisms with respect to efficiency and budget balance. In doing so, we also propose and study a new mechanism, which although lacks strategyproofness, it exhibits important efficiency improvement along with certain fairness properties.","Title: Negotiating Premium Peering Prices: A Quantitative Model with Applications.\nYear: 2016\n\nAbstract: We have developed a novel methodology for deriving bandwidth prices for premium direct peering between Access ISPs (A-ISPs) and Content and Service Providers (CSPs) that want to deliver content and services in premium quality. Our methodology establishes a direct link between service profitability, for example, from advertising, user and subscriber loyalty, interconnection costs, and finally bandwidth price for peering. Unlike existing work in both the networking and economics literature, our resulting computational model, built around Nash bargaining, can be used for deriving quantitative results comparable to actual market prices. We analyze the U.S. market and derive prices for video, that compare favorably with existing prices for transit and paid peering. We also observe that the fair prices returned by the model for high-profit/low-volume services such as search, are orders of magnitude higher than current bandwidth prices. This implies that resolving existing (fierce) interconnection tussles may require per service, instead of wholesale, peering between A-ISPs and CSPs. Our model can be used for deriving initial benchmark prices for such negotiations.","Title: Pricing The Fast-Lanes: A Qualitative Study On The Implications Of Paid Peering Agreements\nYear: 2016\n\nAbstract: Paid peering is controversial. Generally speaking, large access ISPs favour it, while Content Providers (CPs) do not. But is CP opposition to paid peering irrational, especially if their transferred payments can incentivize the ISP to make greater investment in the common infrastructure? To answer this question we analyze an ecosystem consisting of a single access ISP that directly interconnects with multiple CPs, doing so in two extreme situations: when paid peering is used for all the involved CPs, and when all parties agree on a settlement-free deal. We analyse the equilibrium, finding the total amount of investments and profits of the various stakeholders, when their revenues are affected by the total investment in infrastructure and the charges that result from the different agreements. Interestingly, it turns out whether or not there is benefit to a CP, depends on whether its business model is to sell content, and so it can recover part of the paid peering charge by raising prices to its customers, or if the CP obtains its revenue from ads. It also depends on the volume (in traffic units) per customer transaction, since we find that CPs with high volume per transaction will be charged less per byte, and so will contribute proportionally less to the total paid peering revenue that subsidizes the common ISP infrastructure. We also find a crucial role for the end-users' evaluation of the level of investments of the various market players. Another relevant factor is the access price that an ISP charges its customers.","Title: Dynamic routing for social information sharing.\nYear: 2017\n\nAbstract: Today, mobile users are intensively interconnected thanks to the emerging mobile social networks, where they share location-based information with each other when traveling on different routes and visit different areas of the city. In our model, the information collected is aggregated over all users’ trips and made publicly available as a public good. Due to information overlap, the total useful content amount increases with the diversity in path choices made by the users, and it is crucial to motivate selfish users to choose different paths, despite the potentially higher costs associated with their trips. In this paper, we combine the benefits from social information sharing with the fundamental routing problem where a unit mass of non-atomic selfish users decides their trips in a non-cooperative game by choosing between a high-cost and a low-cost path. To remedy the inefficient low-content equilibrium where all users choose to explore a single path (the low-cost path), we propose and analyze two new incentive mechanisms that can be used by the social network application, one based on side payments and the other on restricting access to content for users that choose the low cost path. Under asymmetric information about user types (their valuations for content quality), both mechanisms efficiently penalize the participants that use the low-cost path and reward the participants that take the high-cost path. They lead to greater path diversity and hence to more total available content at the social cost of reduced user participation or restricted content to part of the users. We show that user heterogeneity can have opposite effects on social efficiency depending on the mechanism used. We also obtain interesting price of anarchy results that show some fundamental tradeoffs between achieving path diversity and maintaining greater user participation, motivating a combined mechanism to further increase the social welfare. Our model extends classical dynamic routing in the case of externalities caused from traffic on different paths of the network.","Title: Special article [paid peering: pricing and adoption incentives].\nYear: 2016\n\nAbstract: Large access providers (ISPs) are seeking for new types of business agreements and pricing models to manage network costs and monetize better the provision of last-mile services. A typical paradigm of such new pricing norms is the proliferation of paid peering deals between ISPs and content providers (CPs), while on top of this, some ISPs are already experimenting with usage-based tariffs, usually through data-plans, instead of the typical fixed-based charging. In this work we define as common platform, the infrastructure in which a single ISP transacts with several CPs through peering agreements. In this context, we examine whether, and under which market conditions, the profitability of the involved stakeholders improves when the establishment of this platform is accompanied by a monetary compensation from the CPs to the ISP (paid peering), v.s. a scenario where their deal is a typical settlement-free one. In both cases, we assume that the ISP implements a usage-based access pricing scheme, implying that end-users will pay more for higher transaction rates with the CPs. Our framework captures some of the most important details of the current market, such as the various business models adopted by the CPs, the end-usersu0027 evaluation towards the ISPu0027s and CPsu0027 level of investments and the traffic rates per transaction for the offered services. By analysing the equilibrium derived by a leader-follower game, it turns out (among other practical takeaways) that whether or not the profitability of a CP improves, it highly depends on whether its business model is to sell content, or if it obtains its revenue from advertisements. Finally, we extract that consumer surplus is considerably higher under paid peering, which in turn implies improved levels of social welfare 1 .","Title: Fair congestion control for long-lived background flows\nYear: 2014\n\nAbstract: We consider a new economic framework for designing congestion control algorithms for long-lived background flows. The algorithms aim at the fair sharing of the network capacity left over by the much shorter (and usually more interactive and delay sensitive) flows, between all competing background flows even if some use TCP in particular. The algorithm acts as a weighted variant of TCP with adaptive weight adjustment based on an estimate of the average throughput that would be obtained by a TCP flow following the same path, with the resulting throughput behavior ranging from low priority to that of TCP. It turns out that this design approximately maximizes the social welfare for the background traffic for appropriately defined utilities for bandwidth averaged over long time scales and a social cost function reflecting the negative externalities on the average download time of interactive flows, over all possible protocol designs. We illustrate our approach by proposing two concrete congestion control algorithms, each of them guaranteeing the throughput of TCP in the case of resource scarcity or resource abundance respectively. Two implementation alternatives are explored; one based on adjusting the congestion window ramp up speed and another based on alternating transmitting and idle periods. We use both analysis and simulations to validate our approach of implementing new background transfer protocols.","Title: Economics In Mobile Data Offloading With Uniform Pricing\nYear: 2017\n\nAbstract: Mobile data offloading is an emerging technology to alleviate cellular network congestion and improve user service quality. In this paper, we investigate the economics of mobile data offloading through access points (APs) deployed by small cell service providers (SSPs), implementing uniform volume prices for all the mobile users (MUs) in each SSP's coverage including the overlapping area. In particular, we consider a data offloading game with a single mobile network operator (MNO) and two SSPs with overlapping coverage areas, where each SSP announces a uniform price for offloading the cellular traffic within its coverage, and the MNO determines the traffic volumes to offload. We show that there is no pure Nash equilibrium (PNE) in the data offloading game, and prove the existence of the mixed strategy Nash equilibrium (MNE) by price randomization. Furthermore, we propose a simple one shot auction mechanism that is easy to implement and has PNEs which yield SSPs the same equilibrium payoffs compared to the expected payoffs obtained at the above MNE.","Title: Drivers, riders and service providers: the impact of the sharing economy on mobility\nYear: 2017\n\nAbstract: Ride sharing denotes the practice of sharing a car such that more than one person travels in a car during a journey. Sharing rides was traditionally restricted to family members and close friends or long-distance journeys scheduled well before the intended time of departure. Only the emergence of mobile computing technologies and GPS location services in combination with electronic payments and online reputation systems provided for the technological cornerstones to make on-demand short-distance ride sharing among strangers viable. Typically individuals enter their trip details on an online platform1 which then facilitates the matching of riders with cars and drivers - and within minutes the individual's trip commences. Such platforms must attract both, supply and demand for rides.","Title: Ex-Post Max-Min Fairness of Generalized AGV Mechanisms.\nYear: 2017\n\nAbstract: We generalize the standard Arrow-d'Aspremont-Gerard-Varet (AGV) mechanism to balance the net payoffs received by all agents, while maintaining Bayesian incentive compatibility, ex-post efficiency, and ex-post budget balance of the standard AGV mechanism. In a private-value setting with independent agents' types and the principal's cost, we formulate a convex optimization problem to find the mechanism (that achieves ex-post max-min fairness) over a set of parameterized generalized AGV mechanisms, through maximizing the expected value of the minimum ex-post net payoff. We reformulate the convex program as a linear program that can be effectively solved when the number of agents is small. When the number of agents is large, we propose to solve the formulated convex program through the incremental subgradient method. Numerical results on two action models show that the proposed mechanism significantly outperforms the standard AGV mechanism in terms of the expected minimum ex-post payoff.","Title: Congestion Control for Background Data Transfers With Minimal Delay Impact.\nYear: 2017\n\nAbstract: Congestion control protocols for background data are commonly conceived and designed to emulate low priority traffic, which yields to transmission control protocol (TCP) flows. In the presence of even a few very long TCP flows, this behavior can cause bandwidth starvation, and hence, the accumulation of large numbers of background data flows for prolonged periods of time, which may ultimately have...","Title: Market mechanisms for cooperative operation of price-maker energy storage in a power network.\nYear: 2017\n\nAbstract: We explore the integration of large-scale, grid-level energy storage into wholesale electricity markets. We conduct a comparative analysis on three natural market mechanisms that have appeared in the literature: 1) the centralized mechanism according to which all batteries are centrally operated to minimize the social cost, 2) the semi-centralized mechanism under which the batteries are centrally ...","Title: Optimal pricing for a peer-to-peer sharing platform under network externalities.\nYear: 2018\n\nAbstract: In this paper, we analyse how a peer-to-peer sharing platform should price its service (when imagined as an excludable public good) to maximize profit, when each user's participation adds value to the platform service by creating a positive externality to other participants. To characterize network externalities as a function of the number of participants, we consider different bounded and unbounded user utility models. The bounded utility model fits many infrastructure sharing applications with bounded network value, in which complete coverage has a finite user valuation (e.g., WiFi or hotspot). The unbounded utility model fits the large scale data sharing and explosion in social media, where it is expected that the network value follows Metcalfe's or Zipf's law. For both models, we analyze the optimal pricing schemes to select heterogeneous users in the platform under complete and incomplete information of users' service valuations. We propose the concept of price of information (PoI) to characterize the profit loss due to lack of information, and present provable PoI bounds for different utility models. We show that the PoI = 2 for the bounded utility model, meaning that just half of profit is lost, whereas the PoI ≥ 2 for the unbounded utility model and increases as for a less concave utility function. We also show that the complicated differentiated pricing scheme which is optimal under incomplete user information, can be replaced by a single uniform price scheme that is asymptotic optimal. Finally, we extend our pricing schemes to a two-sided market by including a new group of 'pure' service users contributing no externalities, and show that the platform may charge zero price to the original group of users in order to attract the pure user group.","Title: A Fair and Budget-Balanced Incentive Mechanism for Energy Management in Buildings.\nYear: 2018\n\nAbstract: We propose a mechanism design based approach for the temperature setting in shared spaces (such as meeting rooms, library, and open office areas) in buildings. The proposed mechanism asks each occupant to report her willingness-to-pay for three possible outcomes: keeping the current temperature, lowering the temperature for one degree, and raising the temperature for one degree. Collecting the occ...","Title: Recommending Paths: Follow or Not Follow?\nYear: 2018\n\nAbstract: Mobile social network applications constitute an important platform for traffic information sharing, helping users collect and share sensor information about the driving conditions they experience on the traveled path in real time. In this paper we analyse the simple but fundamental model of a platform choosing between two paths: one with known deterministic travel cost and the other that alternates over time between a low and a high random cost states, where the low and the high cost states are only partially observable and perform respectively better and worse on average than the fixed cost path. The more users are routed over the stochastic path, the better the platform can infer its actual state and use it efficiently.At the Nash equilibrium, if asked to take the riskier path, in many cases selfish users (that are allowed to have access to the information collected by the platform) will myopically disregard the optimal path suggestions of the platform, leading to a suboptimal system without enough exploration on the stochastic path. We prove the interesting result that if the past collected information is hidden from users, the system becomes incentive compatible and even ‘sophisticated’ users (in the sense that they have full capability to reverse-engineer the platform’s recommendation and derive the path state distribution conditional on the recommendation) prefer to follow the platform’s recommendations. In a more practical setting where the platform implements a model-free Q-learning algorithm to minimise the social travel cost, our analysis suggests that increasing the accuracy of the learning algorithm increases the range of system parameters for which sophisticated users follow the recommendations of the platform, becoming in the limit fully incentive compatible. Finally, we extend the two-path model to include more stochastic paths, and show that incentive compatibility holds under our information restriction mechanism.","Title: Mobile Data Offloading with Uniform Pricing and Overlaps\nYear: 2019\n\nAbstract: Mobile data offloading is an emerging technology to alleviate cellular network congestion and improve user service quality. In this paper, we investigate the\n<italic xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">economics</italic>\nof mobile data offloading through access points (APs) deployed by small cell service providers (SSPs), implementing uniform volume prices for all the mobile users (MUs) in each SSP's coverage including the overlapping area. In particular, we consider a data offloading game with a single mobile network operator (MNO) and two SSPs with overlapping coverage areas, where each SSP announces a uniform price for serving the cellular traffic within its coverage, and the MNO determines the traffic volumes to offload. We show that there is no pure Nash equilibrium (PNE) under such price competition, and determine the corresponding mixed strategy Nash equilibrium (MNE) using price randomization. As a practical solution, we propose a simple one shot auction mechanism that is easy to implement and has PNEs which is payoff equivalent with the MNE under price competition. We believe that this simple mechanism due to its simplicity of determining the equilibrium prices could be used in the negotiation between the SSPs and the MNO to determine the average service prices. Finally, we study the strategic topological infrastructure placement problem using a 1-dimension (1D, linear) user traffic flow model and a 2-dimension (2D) user traffic flow model when SSPs compete assuming uniform price competition as above. We show that the first mover in the placement problem will deploy its APs to cover more than half of the total flow volume and has an advantage to obtain a higher equilibrium payoff.","Title: Verifying Automata Specifications of Probabilistic Real-time Systems\nYear: 1991\n\nAbstract: . We present a model-checking algorithm for a system presented as ageneralized semi-Markov process and a specification given as a deterministic timedautomaton. This leads to a method for automatic verification of timing propertiesof finite-state probabilistic real-time systems.Keywords: Real-time, Probabilistic systems, Automatic verification.1 IntroductionThere is increasing awareness that unexpected behavior from interacting processes cancause serious problems. This observation...","Title: Invited talk II summary: incentive schemes in memory-less P2P systems\nYear: 2005\n\nAbstract: The asymptotic analysis of certain public good models for p2p file sharing systems focusing on content availability, suggests that when the aim is to maximize social welfare, a fixed contribution scheme in terms of the number of files shared per unity of time can be asymptotically optimal as the number of participants n grows to infinity (see [1] and references therein). Such an incentive scheme is very simple and attractive, and is also suitable for other p2p applications with similar public good charasteristics such as WLAN peering. However, its enforcement is not straightforward in cases where no trusted software or central entity accounting for peers' transactions can be assumed and peers are free to change their identity with no cost. That is, when no sort of user memory is available to be able to identify and punish the potential free riders.","Title: Market Mechanisms For Energy Storage Planning And Operation In A Power Network\nYear: 2018\n\nAbstract: We conduct a comparative analysis on three joint market mechanisms for energy storage investment and operation under locational marginal pricing: i) socially optimal storage investment with centralized operation, ii) profit -maximizing storage investment with centralized operation, and iii) profit maximizing storage investment with deregulated operation. For the first mechanism (according to which a social planner centrally optimizes the sitting, sizing, and operation of batteries), we show that the revenue collected from storage operation exactly covers the investment cost at a social optimum. Under the last two mechanisms, a profit maximizing firm strategically determines the sitting and sizing of batteries, and the storage operation is either centrally optimized for cost minimization (mechanism ii) or completely controlled by the storage owner for profit maximization (mechanism iii). Numerical results on the IEEE 57-bus test system reveal that under profit maximizing investment, the centralized and deregulated storage operations usually lead to almost identical outcomes (on installed storage capacity, social cost, and storage profit).","Title: Throughput and Pricing of Ridesharing Systems\nYear: 2019\n\nAbstract: We introduce a queueing model for passengers waiting to get transport by drivers participating in a ridesharing system. The effect of available drivers’ mobility pattern, their willingness to accept rides in a given location, and the incentives offered by the platform, on system throughput is considered. We characterize the largest set of passenger arrival rates which can be served by a fixed number of drivers under any policy dictating the mobility pattern of available drivers. It turns out that any rate in this set can be achieved by offering appropriate constant but region-dependent rewards to drivers for passenger pick up. Moreover, it is shown that dynamic rewards which scale in proportion to the number of passengers waiting for pick up in each region, not only maximize throughput but also significantly decrease pick up delays.","Title: Average-Case Analysis of Greedy Matching for D2D Resource Sharing\nYear: 2021\n\nAbstract: Given the proximity of many wireless users and their diversity in consuming local resources (e.g., data-plans, computation and even energy resources), device-to-device (D2D) resource sharing is a promising approach towards realizing a sharing economy. In the resulting networked economy, n users segment themselves into sellers and buyers that need to be efficiently matched locally. This paper adopt..."],"string":"[\n \"Title: Providing Bandwidth Guarantees Over A Best-Effort Network: Call-Admission And Pricing\\nYear: 2001\\n\\nAbstract: This pager introduces a framework for answering questions regarding the conditions on the network load that allow a best-effort network like the Internet to support connections of given duration that require a certain quality of service. Such quality of service is expressed in terms of the percentage of time the bandwidth allocated to a connection may drop below a certain level or the maximum allowable delay in placing the call through the network waiting for more favorable loading conditions. The call-acceptance conditions, which depend on the behavior of the system over the lifetime of accepted calls, are thus based on transient models for the congestion (instead of looking at the average behavior) and attempt to exploit the time-scales of the fluctuations of the number of connections competing for bandwidth. Extensions of the model consider the case of dynamic pricing which allows connections that pay more to get larger shares of the bandwidth, and investigate the trade-off between quality of service, the size of the acceptance region, and the charge to be paid by the connection. Under this framework we introduce an option contract that reduces the risk of quality disruption, if a user has a fixed budget at his disposal, and calculate its price. One potential use of this methodology is towards developing a simple admission control mechanism for placing voice calls through an IP network, where the decisions can be taken by edge devices.\",\n \"Title: Differentiated Services in the GPRS Wireless Access Environment\\nYear: 2001\\n\\nAbstract: The General Packet Radio Service extends the existing GSM mobile communications technology by providing packet switching and higher data rates in order to efficiently access IP-based services in the Internet. Even though Quality-of-Service notions and parameters are included in the GPRS specification, no realization path for Quality-of-Service support has been proposed. In this paper we adapt the Differentiated Services framework and apply it over the GPRS air interface in order to provide various levels of service differentiation and a true end-to-end application of the Internet Differentiated Services architecture.\",\n \"Title: Inter-domain coordination models\\nYear: 2012\\n\\nAbstract: In order for the Network Service Providers (NSPs) to provide end-to-end Quality of Service (QoS) at the inter-domain level different coordination models have been proposed by ETICS project. In this work we present and analyse the plausible alternatives of those models and we compare them with each other in terms of information asymmetry issues. We show that different information sets affect the total service provision and we present a basic model analysis on information issues by means of game-theoretic models.\",\n \"Title: Application and evaluation of large deviation techniques for traffic engineering in broadband networks\\nYear: 1998\\n\\nAbstract: Accurate yet simple methods for traffic engineering are important for efficient dimensioning of broadband networks. The goal of this paper is to apply and evaluate large deviation techniques for traffic engineering. In particular, we employ the recently developed theory of effective bandwidths, where the effective bandwidth depends not only on the statistical characteristics of the traffic stream, but also on a link's operating point through two parameters, the space and time parameters, which are computed using the many sources asymptotic. We show that this effective bandwidth definition can accurately quantify resource usage. Furthermore, we estimate and interpret values of the space and time parameters for various mixes of real traffic demonstrating how these values can be used to clarify the effects on the link performance of the time scales of burstiness of the traffic input, of the link parameters (capacity and buffer), and of traffic control mechanisms, such as traffic shaping. Our approach relies on off-line analysis of traffic traces, the granularity of which is determined by the time parameter of the link, and our experiments involve a large set of MPEG-1 compressed video and Internet Wide Area Network (WAN) traces, as well as modeled voice traffic.\",\n \"Title: Incentives for content availability in memory-less peer-to-peer file sharing systems\\nYear: 2005\\n\\nAbstract: In this paper we address the issue of content availability in p2p file sharing systems. Content availability is a public good: the copying of a file by one peer does not prevent another peer also from copying it; but contributing files to the common pool is costly. The asymptotic analysis of certain public good models for p2p file sharing suggests that when the aim is to maximize social welfare, a fixed contribution scheme in terms of the number of files shared per unity of time can be asymptotically optimal as the number of participants n grows to infinity. However, the enforcement of such an incentive scheme is not straightforward in a realistic p2p system, where no trusted software or central entity accounting for peers' transactions can be assumed and peers are free to change their identity with no cost. We present a realistic version of the fixed contribution scheme, which does not require the use of system memory but relies only on the time peers are consuming resources to ensure that they contribute adequately. We describe the functionality that should be supported for enforcement and discuss the additional incentive issues that arise in this context, proposing some practical solutions to address them. We also formulate a suitable economic model to estimate the efficiency-loss of the proposed mechanism (compared to the one achieved using the theoretically optimal schemes under complete and incomplete information) and provide some insights for the correct tuning of its basic parameters. Our first results indicate that the proposed mechanism constitutes a good compromise between economic efficiency and implementability and should lead to some interesting and practical solutions for providing incentives for content availability in p2p systems.\",\n \"Title: A random walk model for studying allocation patterns in auction-based resource allocation\\nYear: 2006\\n\\nAbstract: We consider users bidding in a series of multi-unit sealed-bid auctions, aiming at reserving the same amount of units of the resource auctioned, e.g. transmission slots in a wireless network. Each user attains from each successful allocation of resource units an instant marginal utility that depends on his history of resource allocation. The user’s bid at each auction equals this marginal utility. We introduce a random walk model for transient analysis of this series of auctions, we study the properties of the resulting user resource allocation patterns and we provide a numerical and experimental evaluation of this model.\",\n \"Title: A new strategy for bidding in the network-wide progressive second price auction for bandwidth\\nYear: 2005\\n\\nAbstract: We revisit the network-wide Progressive Second Price auction (PSP) proposed by Lazar and Semret in cite2. In this mechanism, each bidder submits the same bid in each link of the path he is interested in, taking into account the overall competition in the path. We propose a new strategy in which each bidder apportions his total bid-price in the various links, while taking into account the competition in each link separately. We have carried out a wide variety of experiments and compare our strategy with the original one with respect to efficiency and bidders' net benefit. We show that our strategy yields an outcome closer to the optimal social welfare as well as higher expected net benefit for bidders in most of the cases.\",\n \"Title: Memory Efficient Algorithms for the Verification of Temporal Properties\\nYear: 1990\\n\\nAbstract: This paper addresses the problem of designing memory efficient algorithms for the verification of temporal properties of finite-state programs. Both the programs and their desired temporal properties are modeled as automata on infinite words (Büchi automata). Verification is then reduced to checking the emptiness of the automation resulting from the product of the program and the property. This problem is usually solved by computing the strongly connected components of the graph representing the product automaton. Here, we present algorithms which solve the emptiness problem without explicitly constructing the strongly connected components of the product graph. By allowing the algorithms to err with a small probability, we can implement them with a randomly accessed memory of size O(n) bits, where n is the number of states of the graph, instead of O(n log n) bits which the presently known algorithms require.\",\n \"Title: Model-checking for probabilistic real-time systems (extended abstract)\\nYear: 1991\\n\\nAbstract: Model-checking is a method of verifying concurrent systems in which a state-graph model of the system behavior is compared with a temporal logic formula. This paper extends model-checking to stochastic real-time systems, whose behavior depends on probabilistic choice and quantitative time. The specification language is TCTL, a branching-time temporal logic for expressing real-time properties. We interpret the formulas of the logic over generalized semi-Markov processes. Our model can express constraints like “the delay between the request and the response is distributed uniformly between 2 to 4 seconds”.\",\n \"Title: Tariffs, Mechanisms and Equilibria at a Single Internet Link\\nYear: 2009\\n\\nAbstract: We analyze the interplay between the demand for downloads, choice of congestion control mechanism, and tariff structure at a single link, when users have preferences in terms of average download delay and they are charged according to the number of ECN marked packets they receive. Our model involves a timescale separation approach, where in the fast timescales active flows compete for instantaneous bandwidth share by optimally tuning congestion control parameters in a noncooperative fashion. This is modeled by letting flows choose utility functions within the network utility maximization framework laid down by Kelly [1]. On a slower timescale, users selfishly change their otherwise unrestricted demand for downloads based on the average experienced download delay and charges incurred. We study the equilibrium of this loop of interactions from the point of view of social welfare. For homogeneous users we find that optimal equilibria are induced when they choose among linear utility functions, while this is not the case for logarithmic, i.e., proportionally fair congestion controllers. We next consider two types of users, web-browsing and bittorrent, where the latter are much less sensitive to download delay than the former. If bittorrent users respond to congestion according to proportionally fair utility functions the charge induced by ECN marked packets does not provide the correct signal for service differentiation and the resulting equilibrium is suboptimal. On the other hand, if flows are charged for the volume of bits they transfer as well, then social welfare maximum is attained for some price per bit. This reveals a new role for bit volume pricing: to provide bittorrent users the correct incentives for choosing congestion controllers that effectively give priority to web-browsing users.\",\n \"Title: Dynamic Bandwidth Pricing: Provision Cost, Market Size, Effective Bandwidths and Price Games\\nYear: 2008\\n\\nAbstract: Nowadays, in the markets of broadband access services, traditional contracts are of \\\"static\\\" type. Customers buy the right to use a specific amount of resources for a specific period of time. On the other hand, modern services and applications render the demand for bandwidth highly variable and bursty. New types of contracts emerge (\\\"dynamic contracts\\\") which allow customers to dynamically adjust their bandwidth demand. In such an environment, we study the case of a price competition situation between two providers of static and dynamic contracts. We investigate the resulting reaction curves, search for the existence of an equilibrium point and examine if and how the market is segmented between the two providers. Our first model considers simple, constant provision costs. We then extend the model to include costs that depend on the multiplexing capabilities that the contracts offer to the providers, taking into consideration the size of the market. We base our analysis on the theory of effective bandwidths and investigate the new conditions that allow the provider of dynamic contracts to enter the market.\",\n \"Title: Fair background data transfers of minimal delay impact\\nYear: 2012\\n\\nAbstract: In this paper we present a methodology for the design of congestion control protocols for background data transfers that have a minimal delay impact on short TCP transfers and compete for a target share of the leftover average capacity with other background TCP transfers. We analytically compute the optimal policy and show that it's delay performance can be well approximated by a weighted TCP policy, which maintains a target proportion of TCPs bandwidth at all times in order to achieve the same share of excess capacity. The relative approximation error is always less than 17.2% while it quickly decreases to zero as the number of coexisting background TCP flows increases. Next, we consider a general utility-based fairness criterion for sharing the leftover average capacity, including a penalty term capturing the delay impact to short flows. The criterion is jointly optimized over all allocations of excess capacity to background flows (including TCP ones) on long timescales, and all bandwidth sharing policies on fast time scales. Even though the delay optimal sharing policy that solves the above optimization problem does not lead to distributed congestion control algorithms and more significantly, requires knowledge of the number of competing background TCP flows, both problems disappear under the weighted TCP policy. A distributed weight adjustment policy is considered where, at equilibrium, the overall performance is nearly optimal, with a vanishing relative optimality error as the number of background TCP flows increases. We illustrate the methodology by giving two examples of congestion control algorithms for background transfers. Both achieve low delay for short flows relative to TCP, but at the same time they present strong incentives for adoption against incumbent low priority solutions in public environments.\",\n \"Title: Extending Promela and Spin for Real Time\\nYear: 1996\\n\\nAbstract: The efficient representation and manipulation of time information is key to any successful implementation of a verification tool. We extend the syntax and semantics of the higher level specification language Promela to include constructs and statements based on the model of timed Büchi automata [2]. We implement these extensions on top of the verification tool Spin.\",\n \"Title: Auction-based resource reservation in 2.5/3G networks\\nYear: 2004\\n\\nAbstract: We consider UMTS networks in which users request services other than telephony that last for long time intervals: e.g., video clips that last for several minutes. The duration of network time-slots over which resource units are allocated is much shorter. This complicates consistent reservation of resources over longer time scales, where consistent reservation is required to ensure that service quality is constant throughout the entire service session. In this paper, we define an auction-based mechanism for nearly consistent reservation of the resources of a UMTS (or GPRS) network by the users that value them the most, in order to satisfy the longer time scale requirements of their service sessions. Each of these sessions has a fixed target bit-rate. The mechanism is based on a series of Generalized Vickrey Auctions and a set of predefined user utility functions that we propose. Bidding is performed automatically on behalf of the users on the basis of each user's selection of one of these utility functions and his declaration of a total willingness to pay. We argue that under our mechanism the user does not have a clear incentive of not performing a truthful selection of a bidding function according to his own utility. The utility functions we define express appropriately the preferences of the users with respect to the resource allocation pattern in the cases where perfectly consistent allocation cannot be attained. We also provide a mapping of these functions to the UMTS service classes. The effectiveness of our resource reservation mechanism is demonstrated by means of experiments. It appears that most of the users either are served very satisfactorily or essentially are not served at all. The mechanism is implemented at the network base station, and is applicable in practical cases of networks with large numbers of users whose sessions last for many slots.\",\n \"Title: Pricing differentiated services in the GPRS environment\\nYear: 2001\\n\\nAbstract: The General Packet Radio Service extends the existing GSM mobilecommunications technology by providing packet switching and higherdata rates in order to efficiently access IP-based services in theInternet. Since no realization path for Quality-of-Service supporthas been proposed yet, we adapt the Differentiated Servicesframework and apply it over the GPRS air interface in order toprovide various levels of service differentiation. We also focus onapplying a charging technique so as to publish a unit price foreach service class. These prices are designed to lead to themaximization of Social Welfare and the users' net benefit.\",\n \"Title: Model-checking in dense real-time\\nYear: 1993\\n\\nAbstract: . Model-checking is a method of verifying concurrent systems in whicha state-transition graph model of the system behavior is compared with a temporallogic formula. This paper extends model-checking for the branching-time logic CTLto the analysis of real-time systems, whose correctness depends on the magnitudesof the timing delays. For specifications, we extend the syntax of CTL to allowquantitative temporal operators such as 93!5 , meaning "possibly within 5 timeunits." The formulas of...\",\n \"Title: GridEcon - the economic-enhanced next-generation internet\\nYear: 2007\\n\\nAbstract: The major shortcoming of Grid middleware systems is the lack of economic-enhanced Grid services. These new services are necessary in order to let Grid users benefit from the properties of the Grid. Those properties comprise the availability of on-demand computational power, simplicity of access to resources, low cost of ownership, and a pay-for-use pricing model in addition to the already leveraged properties such as cost reduction and aggregated processing power for high-performance computing applications. This paper gives an overview of the EU-funded project GridEcon on Grid economics and business models. It describes its vision of the next generation Grid/Internet, in which individuals, universities, small and medium sized enterprises (SMEs), and large companies have access to the Grid in exactly the same way. Any resource, including servers, storage, software, or data, is accessible as a service. In addition to this, the architecture of an economic-enhanced infrastructure is illustrated and the goal of the project is described.\",\n \"Title: Resource Control For Loss-Sensitive Traffic In Cdma Networks\\nYear: 2004\\n\\nAbstract: We investigate the problem of efficient resource control for loss-sensitive traffic in CDMA networks, using an economic modelling framework that takes into account the joint control of the transmission rate and the signal quality. Although the corresponding global optimization problem has a non-trivial structure, and we cannot in general guarantee that a solution can be found using the Lagrangian method, we have strong experimental evidence that this is possible for a wide range of user utilities. In this case the global optimum can be achieved in a decentralized manner, using shadow prices to influence the individual user resource requests. Based on this evidence, the main contribution of the paper is to discuss how existing rate control and outer-loop power control procedures can obtain a simple and attractive form that takes into account, through shadow prices, the level of demand and supply in order to achieve efficient resource utilization. Moreover, we describe and evaluate approximations of the proposed resource control model that can simplify its application, and we present extensions of the model for the case where the packet success ratio depends on the transmission rate in addition to the signal quality, and for network paths containing multiple wireless links.\",\n \"Title: Hybrid Automata: An Algorithmic Approach to the Specification and Verification of Hybrid Systems\\nYear: 1992\\n\\nAbstract: We introduce the framework of hybrid automata as a model and specification language forhybrid systems. Hybrid automata can be viewed as a generalization of timed automata, in whichthe behavior of variables is governed in each state by a set of differential equations. We show thatmany of the examples considered in the workshop can be defined by hybrid automata. Whilethe reachability problem is undecidable even for very restricted classes of hybrid automata, wepresent two semidecision...\",\n \"Title: Congestion control and pricing of volume intensive applications\\nYear: 2009\\n\\nAbstract: We consider fair sharing of network resources over a period of time between offline and interactive users. The former value total transferred volume while the latter have time-dependent valuations for instantaneous bandwidth. Following the work of Kelly on fair bandwidth sharing, we express the system optimum as an equilibrium of traditional congestion control taking place in a fast timescale and a volume charging mechanism in a slow timescale. The problem is very much related to multipath congestion control where paths are unfolded also in the dimension of time, and our algorithms can be regarded as multipath congestion control algorithms operating on two separate timescales that combine congestion control with scheduling. A novel feature of our algorithms is that, contrary to ordinary dual algorithms, users are not overcharged at any time while approaching the equilibrium. This is a sensible economical property not recognized before in the context of congestion control over long timescales.\",\n \"Title: Enforcing efficient resource provisioning in peer-to-peer file sharing systems\\nYear: 2006\\n\\nAbstract: In this paper we focus on content availability as the main good provided by a p2p file sharing system and consider peer availability, the amount of time peers stay connected sharing their files, as their main contribution. We wish to study the effectiveness of incentive mechanisms which enforce contribution by somehow relating it with consumption. Towards this end, we propose a mechanism who wishes to regulate the time required for peers to stay on-line sharing their files by dictating a fixed upload throughput used by uploading peers and thus a certain average time for a download to finish. We formulate and analyze a suitable economic model focusing on peer availability in order to understand the role of this important system parameter and provide the means to efficiently tune it.\",\n \"Title: An efficient auction-based mechanism for hierarchically structured bandwidth markets\\nYear: 2006\\n\\nAbstract: We consider a hierarchical business model for selling bandwidth in a single link. In the top level of the market the social planner allocates bandwidth to intermediate providers, who in turn allocate their assigned shares of bandwidth to the customers in the lower level. We propose an innovative mechanism comprising an appropriate auction in each level; the payment rule of the lower-level auction is imposed by the social planner. We establish that the bidders of all auctions have the incentives both to participate uninhibitively and to bid truthfully, as well as that overall efficiency is ultimately attained. We also investigate the properties of the mechanism in case where each of the competing intermediate providers can select the payment rule on his own, under a variety of assumptions on the information possessed by the various players as well as on their level of freedom to act in the market. For one of the cases analyzed we prove that providers have no incentive to deviate from the original payment rule.\",\n \"Title: Procedures and tools for analysis of network traffic measurements\\nYear: 2002\\n\\nAbstract: We present procedures and tools for the analysis of network traffic measurements. The tools consist of stand-alone [1] modules that implement advanced statistical analysis procedures and a flexible web-based interface through which a user can create, modify, save, and execute experiments using the statistical analysis modules. The tools do not assume a specific source traffic model, but rather process actual measurements of network traffic. Indeed, the theory that the tools are based on can identify the time-scales that affect a link's performance, hence suggest the appropriate time granularity of traffic measurements. We present and discuss case studies that demonstrate the application of the tools for answering questions related to network management and dimensioning, such as the maximum link utilization when some quality of service is guaranteed, how this utilization is affected by the link buffer and traffic shaping, the acceptance region and the effects of the scheduling discipline, and the token (or leaky) bucket parameters of a traffic stream. The case studies involve actual traffic measurements obtained by a high performance measurement platform that we have deployed at the University of Crete network.\",\n \"Title: Economic issues in shared infrastructures\\nYear: 2012\\n\\nAbstract: In designing and managing a shared infrastructure, one must take account of the fact that its participants will make self-interested and strategic decisions about the resources that they are willing to contribute to it and/or the share of its cost that they are willing to bear. Taking proper account of the incentive issues that thereby arise, we design mechanisms that, by eliciting appropriate information from the participants, can obtain for them maximal social welfare, subject to charging payments that are sufficient to cover costs. We show that there are incentivizing roles to be played both by the payments that we ask from the participants and the specification of how resources are to be shared. New in this paper is our formulation of models for designing optimal management policies, our analysis that demonstrates the inadequacy of simple sharing policies, and our proposals for some better ones. We learn that simple policies may be far from optimal and that efficient policy design is not trivial. However, we find that optimal policies have simple forms in the limit as the number of participants becomes large.\",\n \"Title: Adopting the grid for business purposes: the main objectives and the associated economic issues\\nYear: 2007\\n\\nAbstract: Grid technology offers numerous opportunities for the players involved. Despite the fact that the academic community has already exploited many of them, there is an evident reluctance from the business community to act likewise. Recent analysis reveals that the problem lies in overcoming certain business barriers rather than technological ones. At this stage understanding the real-life economic issues from a business perspective is deemed as more important than gaining understanding of complex theoretical economical problems, such as those related to accounting or resource sharing mechanisms especially in cases where the players do not exhibit the required technological expertise. This paper is stimulated from interaction with players from the industry and aims to fill this gap. In particular, we identify and evaluate a number of economic issues that should be taken into consideration by industrial players so that their trust and confidence in the adoption of this promising technology be increased.\",\n \"Title: Market Mechanisms for Trading Grid Resources\\nYear: 2008\\n\\nAbstract: There has been recently an increasing interest in Grid services and economic-aware Grid systems both in the industry and the academia. In this paper we specify a market for hardware providers and consumers interested in leasing Grid resources for a time period. Our approach comprises a stock-market like mechanism that enables the trading of computational power on the basis of a spot and a futures market. The spot market comprises a pair of bid and ask queues. This grid market is more complicated than the standard spot/futures markets of storable commodities, because the computational service traded in our case comprises of resources that are perishable, and has both quantity and duration specified in terms of a time interval. This is an important feature of our market mechanism, complicating considerably the trading algorithms that we develop and assess in this paper.\",\n \"Title: Traffic equivalence and substitution in a multiplexer\\nYear: 1999\\n\\nAbstract: For a multiplexer fed with a large number of sources, we derive conditions under which a single source can be substituted for a given subset of the sources while preserving the buffer overflow probability and the dominant time scales of buffer overflows. This equivalence is stronger than simple effective bandwidth equality and takes into account the context in which multiplexing takes place. This allows a substitution to be made for arbitrarily large proportions of the traffic without changing the operating point of the multiplexer as experienced by the rest of the traffic. It corresponds to defining a single source which is equivalent in a sense “local” to a given context, rather than equivalent in a sense which is “universal” to all contexts. The proposed methodology does not rely on traffic models and obtains the necessary information from the actual traffic traces. We study the case of fractional Brownian motion as a single source substitute and provide theoretical and experimental results that validate our approach\",\n \"Title: On Optimal Packing of Randomly Arriving Objects\\nYear: 1991\\n\\nAbstract: Objects of finitely many types arrive at a facility according to independent stationary Poisson processes. The objects are to be placed in boxes as they arrive. There are finitely many types of boxes and each box type has its own packing configuration. There is an unlimited supply of boxes of each type and once an object is placed in a box it cannot be moved to another box in the future. When a box is completely filled it produces a reward which depends on its configuration. The objective is to select a rule for placing the arriving objects in boxes so as to maximize expected reward. We show how to construct optimal and ε-optimal policies under which the expected number of partially packed boxes does not explode.\",\n \"Title: Comparing economic incentives in peer-to-peer networks\\nYear: 2004\\n\\nAbstract: Users who join a peer-to-peer network have, in general, suboptimal incentives to contribute to the network, because of the externalities that exist between them. The result is an inefficient network where the overall levels of contribution are less than would be the case if each peer acted in the interests of the entire network of peers. Incentives provided in the form of prices or contribution rules that require no money transfers can play an important role in reducing these inefficiency effects. The problem in designing such incentive schemes is information: Designing an optimal incentive scheme requires complete knowledge of the types and preferences of the individual peers and their identities. In this paper we discuss the above issues in terms of a simple but representative example by introducing the basic economic concepts and models. We then investigate the practical issue of designing several simpler incentive schemes requiring less information and compare their efficiency loss to the optimal. We show using numerical analysis that these schemes converge to a fixed proportion of the full information optimal as the number of peers in the network becomes large. This result means that it is not necessary to collect large amounts of information, or to undertake complicated calculations, in order to implement the correct incentives in a large peer-to-peer network.\",\n \"Title: An Algebra of Boolean Processes\\nYear: 1991\\n\\nAbstract: This work has been motivated by the study of the S=R models which allow torepresent systems as a set of communicating state machines cooperating through ashared memory.We show that S=R models can be expressed in terms of a process algebra calledBoolean SCCS which is a special case of Milner's SCCS, in the sense that theactions are elements of some boolean algebra. We define for Boolean SCCS anoperational and a symbolic semantics modulo strong bisimulation equivalence. Acomplete...\",\n \"Title: Computing Accumulated Delays in Real-time Systems\\nYear: 1993\\n\\nAbstract: We present a verification algorithm for duration properties ofreal-time systems. While simple real-time properties constrain thetotal elapsed time between events, duration properties constrain theaccumulated satisfaction time of state predicates. We formalize theconcept of durations by introducing duration measures for timedautomata. A duration measure assigns to each finite run of a timedautomaton a real number —the duration ofthe run— which may be the accumulated satisfaction time of astate predicate along the run. Given a timed automaton with aduration measure, an initial and a final state, and an arithmeticconstraint, the duration-bounded reachability problem asks if there is a run of the automaton from theinitial state to the final state such that the duration of the runsatisfies the constraint. Our main result is an (optimal) PSPACE decision procedure for the duration-bounded reachability problem.\",\n \"Title: The GridEcon Platform: A Business Scenario Testbed for Commercial Cloud Services\\nYear: 2009\\n\\nAbstract: Within this paper, we present the GridEcon Platform, a testbed for designing and evaluating economics-aware services in a commercial Cloud computing setting. The Platform is based on the idea that the exact working of such services is difficult to predict in the context of a market and, therefore, an environment for evaluating its behavior in an emulated market is needed. To identify the components of the GridEcon Platform, a number of economics-aware services and their interactions have been envisioned. The two most important components of the platform are the Marketplace and the Workflow Engine. The Workflow Engine allows the simple composition of a market environment by describing the service interactions between economics-aware services. The Marketplace allows trading goods using different market mechanisms. The capabilities of these components of the GridEcon Platform in conjunction with the economics-aware services are described in this paper in detail. The validation of an implemented market mechanism and a capacity planning service using the GridEcon Platform also demonstrated the usefulness of the GridEcon Platform.\",\n \"Title: A marketplace and its market mechanism for trading commoditized computing resources.\\nYear: 2010\\n\\nAbstract: This paper presents the design and implementation of the GridEcon Marketplace. In addition to supporting a market mechanism\\n for trading computing resources on a pay-per-use basis, this marketplace also provides an environment for integrating value-added\\n support services. These value-added services help consumers to use the utility computing market more efficiently. The GridEcon\\n Market Mechanism for virtual machines specifies in detail the unit-of-trade, the bids and asks, as well as the matching algorithm.\\n The marketplace and market mechanism are validated by using the GridEcon Platform, which is a service-oriented platform for\\n composing market scenarios. Our validation results show that the GridEcon Marketplace fulfills all functional requirements\\n and that the GridEcon Market Mechanism is computationally and economically efficient.\",\n \"Title: Peer-to-Peer Wireless LAN Consortia: Economic Modeling and Architecture\\nYear: 2003\\n\\nAbstract: In this paper we address the incentive issues that arise in a Peer-to-Peer WLAN Consortium [1]. We explore the use of flexible rules on reciprocity to guide domain policies and develop a suitable economic model that demonstrates the basic characteristics of our system.\",\n \"Title: Minimum and maximum delay problems in real-time systems\\nYear: 1992\\n\\nAbstract: We consider a finite-state system with a finite number of clocks, where the transitions may depend on the values of the clocks, and may reset some of the clocks. We address the complexity and provide algorithms for the following problems. Suppose that the system starts from a given current state with a given assignment of values to the clocks. Can a given target state ever appear in the history of the system? What is the earliest time it can appear? What is the latest time it can appear?\",\n \"Title: The algorithmic analysis of hybrid systems\\nYear: 1995\\n\\nAbstract: We present a general framework for the formal specification and algorithmic analysis of hybrid systems. A hybrid system consists of a discrete program with an analog environment. We model hybrid systems as finite automata equipped with variables that evolve continuously with time according to dynamical laws. For verification purposes, we restrict ourselves to linear hybrid systems, where all variables follow piecewise-linear trajectories. We provide decidability and undecidability results for classes of linear hybrid systems, and we show that standard program-analysis techniques can be adapted to linear hybrid systems. In particular, we consider symbolic model-checking and minimization procedures that are based on the reachability analysis of an infinite state space. The procedures iteratively compute state sets that are definable as unions of convex polyhedra in multidimensional real space. We also present approximation techniques for dealing with systems for which the iterative procedures do not converge.\",\n \"Title: Minimization of Timed Transition Systems\\nYear: 1992\\n\\nAbstract: Without Abstract\",\n \"Title: WOICE: A Decentralized System for Ubiquitous VoIP Services.\\nYear: 2009\\n\\nAbstract: We propose WOICE (Wireless vOICE), a fully distributed architecture capable of supporting and offering ubiquitous voice services over wireless IP-based data networks. Existing WiFi infrastructure - deployed by enterprises or even individuals - will be supplemented with software modules based on open standards in order to support VoIP. Furthermore, a new breed of VoIP providers by aggregating these currently isolated access networks will be able to offer alternative and advanced services to reachable end-users. These services would in certain cases completely bypass traditional telephony and cellular providers and in others smoothly cooperate with them. We will describe the overall architecture and the necessary modules that should be in place for realizating a system that creates a marketplace for communication services.\",\n \"Title: Markov Decision Processes and Regular Events (Extended Abstract)\\nYear: 1990\\n\\nAbstract: Desirable properties of the infinite histories of a finite state Markov Decision Process are specified in terms of a finite number of events represented as -regular sets. An infinite history of the process produces a reward which depends on the properties it satisfies. We investigate the existence of optimal policies and provide algorithms for the construction of such policies.\",\n \"Title: Verification of timing Properties of VHDL\\nYear: 1993\\n\\nAbstract: This paper shows how timing properties of VHDL processes can be verified using timed transition systems. The timing model being adopted is the timed automaton model used in the timing extension of Kurshan's COSPAN system. It demonstrates how a VHDL process can be translated into a timed automaton by describing the construction of the corresponding timed process that handles the scheduled signal assignments of the VHDL specification. Verification is performed in the case in which the complement of the timing properties to be verified are provided in terms of a timed automaton. Interestingly enough, this is the case for a large class of hardware properties expressed in terms of timing diagrams.\",\n \"Title: Verification with Real-Time COSPAN\\nYear: 1992\\n\\nAbstract: We describe some examples using an extension of Kurshan's COSPAN system to verify bounded delay constraints in a dense time model, based on the method proposed by Dill for adding timing constraints to Büchi automata. The S/R model and COSPAN are reviewed as background, then we describe how timing can be incorporated into S/R processes, and briefly describe the modified verification algorithm. The examples consist of several time-dependent versions of the Alternating Bit Protocol and the Fiber Distributed Data Interface (FDDI).\",\n \"Title: Stabilizing an uncertain production system\\nYear: 1989\\n\\nAbstract: Consider a production system that consists ofm machines each of which can produce parts ofn types. When machinek is used, it produces a part of typei with probabilitypki. Requests arrive for parts, one at a time. With probability ?i an arriving request is for a part of typei. The requests must be served without waiting. Thus, if a requested part is not available, it must be produced. We find necessary and sufficient conditions for the existence of a strategy (a choice of the machines to be used) which makes the inventory of parts stable and we provide such a strategy.\",\n \"Title: Perfect Packing Theorems and the Average-Case Behavior of Optimal and Online Bin Packing\\nYear: 2002\\n\\nAbstract: We consider the one-dimensional bin packing problem under the discrete uniform distributions $U\\\\{j,k\\\\}$, $1 \\\\leq j \\\\leq k-1$, in which the bin capacity is $k$ and item sizes are chosen uniformly from the set $\\\\{1,2,\\\\ldots,j\\\\}$. Note that for $0\",\n \"Title: The Observational Power of Clocks\\nYear: 1994\\n\\nAbstract: We develop a theory of equivalences for timed systems. Two systemsare equivalent iff external observers cannot observe differences in their behavior.The notion of equivalence depends, therefore, on the distinguishingpower of the observers. The power of an observer to measure timeresults in untimed, clock, and timed equivalences: an untimed observercannot measure the time difference between events; a clock observer usesa clock to measure time differences with finite precision; a timed...\",\n \"Title: The complexity of probabilistic verification\\nYear: 1995\\n\\nAbstract: We determine the complexity of testing whether a finite state, sequential or concurrent probabilistic program satisfies its specification expressed in linear-time temporal logic. For sequential programs, we present an algorithm that runs in time linear in the program and exponential in the specification, and also show that the problem is in PSPACE, matching the known lower bound. For concurrent programs, we show that the problem can be solved in time polynomial in the program and doubly exponential in the specification, and prove that it is complete for double exponential time. We also address these questions for specifications described by &ohgr;-automata or formulas in extended temporal logic.\",\n \"Title: Service differentiation and performance of weighted window-based congestion control and packet marking algorithms in ECN networks\\nYear: 2003\\n\\nAbstract: We investigate the service differentiation, in terms of average throughput, and the performance achieved using weighted window-based congestion control in networks supporting Explicit Congestion Notification (ECN). Our results show how service differentiation, queueing delay, and average throughput are affected by the increase and decrease rules of the end-system congestion control algorithms, and how they depend on the marking algorithms operating in the routers. The end-system algorithms we investigate include Willingness-To-Pay (willingness-to-pay) and MulTCP congestion control, and the packet marking algorithms include RED, virtual queue marking, and load-based marking. Our investigations consider both single and multiple link topologies, and connections with different round trip times.\",\n \"Title: GridEcon: A Market Place for Computing Resources\\nYear: 2008\\n\\nAbstract: This paper discusses the rationales for a Grid market and, in particular, the introduction of a market place for trading commoditized computing resources. The market place proposed makes computing resources from different providers substitutable through virtualization. This includes the definition of a spot and future market as well as the parameters that a market mechanism for computing resources should consider. The above market place is complemented by a set of value-added services (e.g. insurance against resource failures, capacity planning, resource quality assurance, stable price offering) that ensure quality for Grid users over time. The market place technology for all of the above services has been designed by the GridEcon project, contributing to a broader adoption of Grid technology and enabling a service-oriented knowledge utility environment.\",\n \"Title: Application of the many sources asymptotic and effective bandwidths to traffic engineering\\nYear: 1999\\n\\nAbstract: Accurate yet simple methods for traffic engineering are important for efficient management of resources in broadband networks. The goal of this paper is to apply and evaluate large deviation techniques for traffic engineering. In particular, we employ the recently developed theory of effective bandwidths, where the effective bandwidth depends not only on the statistical characteristics of the traffic stream, but also on a link's operating point through two parameters, the space and time parameters, which can be computed using the many sources asymptotic. We show that this effective bandwidth definition can accurately quantify resource usage. Furthermore, we estimate and interpret values of the space and time parameters for various mixes of real traffic demonstrating how these values can be used to clarify the effects on the link performance of the time scales of traffic burstiness, of the link resources (capacity and buffer), and of traffic control mechanisms such as traffic shaping. Our experiments involve a large set of MPEG¿1 compressed video and Internet Wide Area Network (WAN) traces, as well as modeled voice traffic.\",\n \"Title: A study of simple usage-based charging schemes for broadband networks\\nYear: 2000\\n\\nAbstract: Operators of multi service networks require simple charging schemes with which they can fairly recover costs from their users and effectively allocate network resources. This paper studies an approach for computing such charges from simple measurements (time and volume), and relating these to bounds of the effective bandwidth. To achieve economic efficiency, it is necessary that usage based charging schemes capture the relative amount of resources used by connections. Based on this criteria, we evaluate our approach for real traffic consisting of Internet Wide Area Network traces and MPEG 1 compressed video. Its incentive compatibility is shown with an example involving deterministic multiplexing, and the effect of pricing on a network's equilibrium is investigated for deterministic and statistical multiplexing. Finally, we investigate the incentives for traffic shaping provided by the approach.\",\n \"Title: Socio-economic challenges for the internet of the future: the case of congestion control\\nYear: 2010\\n\\nAbstract: The Internet is founded on a very simple premise: sharing! Shared communications links are more efficient than dedicated connections that lie idle much of the time. Hence the rules we use for sharing are extremely vital for the healthy operation of the Internet ecosystem and directly affect the value of the network to its users. It becomes a great paradigm of merging the disciplines of computer science and economics, and presents a great number of challenges to the Internet research community. In this talk we will discuss a number of questions like: what is wrong with todays Internet sharing technologies? Are these consistent with economics? More specifically, is TCP (Transmission Control Protocol) sensible from an economic point of view? Which network sharing technologies justify end-toend from an economics perspective? What is required to make P2P (peer-topeer) a blessing instead of a curse? Are there bad applications or just inefficient combinations of sharing technologies and pricing schemes? This keynote will survey some of these systems management issues and the emerging approaches to solve them.\",\n \"Title: An approach to investigating socio-economic tussles arising from building the future internet\\nYear: 2011\\n\\nAbstract: With the evolution of the Internet from a controlled research network to a worldwide social and economic platform, the initial assumptions regarding stakeholder cooperative behavior are no longer valid. Conflicts have emerged in situations where there are opposing interests. Previous work in the literature has termed these conflicts tussles. This article presents the research of the SESERV project, which develops a methodology to investigate such tussles and is carrying out a survey of tussles identified within the research projects funded under the Future Networks topic of the FP7. Selected tussles covering both social and economic aspects are analyzed also in this article.\",\n \"Title: An Innovative Decision Support System for Bidding in Simultaneous Ascending Auctions\\nYear: 2005\\n\\nAbstract: We present a powerful decision support system for bidding in simultaneous ascending auctions (SAA). Such auctions are commonly used for the allocation of spectrum licenses and electricity generation contracts. Our system can simulate the SAA under a variety of assumptions on the participating bidders and market conditions while it comprises a multitude of innovative features: a) a formal definition of bidding strategies for the user to choose from, including an original strategy that we propose, ii) means of incorporating a bidder's prior knowledge about market demand, iii) functionality for introducing beliefs about rival bidders' strategies, and iv) provision of on-line recommendation about the bids to be placed in a SAA in progress. We also present a set of experiments that corroborate the usefulness of our system.\",\n \"Title: The Case of Multi-hop Peer-to-Peer Implementation of Mobile Social Applications\\nYear: 2006\\n\\nAbstract: In this paper we consider issues related to the design of mobile social software in a multi-hop peer-to-peer environment. We believe that by enabling multi-hop peer-topeer communication could in certain cases make social mobile applications more attractive in terms of location privacy or embarassement, but also more effective in terms of the quality and quantity of the matches achieved. Such an approach could thus contribute in the construction of a critical mass which is necessary for the success of these applications and probably one of the reasons they are not yet widely deployed. On the other hand, it introduces significant implemention challenges related to the contruction and management of a dynamic ad-hoc network. We focus on the provision of the suitable incentives for packet forwarding which is a fundamental problem that needs to be addressed in this context since the ad-hoc nature of the mobile social applications constitutes pricing and reputation (or accounting) based approaches of limited applicability. In this paper we argue that suitable memory-less mechanisms should be devised sketching two such mechanisms for different types of packets exchanged in a mobile social application. Interestingly, as we explain, our proposed mechanisms enforcing contribution while consuming have a more general applicability and could also be used in other applications such as the peer-to-peer resource sharing in ubiquitous computing. We finally discuss several strategic issues that should be also taken into account but are ignored by the corresponding literature and more specifically the consideration of mobility and power transimission as part of the rational strategy of peers participaing in such a system.\",\n \"Title: Economic aspects of building software for service-oriented architectures\\nYear: 2010\\n\\nAbstract: The concept of service-oriented architectures (SOA) has recently emerged as a design principle for the next generation of IT solutions. The main idea behind SOA lies in treating software applications as composed of simpler, cooperating services, implemented by components that communicate over networks through open standards. Such a modular structure is expected to question the effectiveness of current business models for building and distributing software, which generate revenue through license fees. Under SOA, the license fee is replaced by more flexible schemes, whereby users are charged according to their actual usage of services and the corresponding hardware infrastructure. Our work constitutes a first step toward exploring the economic aspects of a market for software components, as well as the incentives of software manufacturers to support this trend. In particular, we examine the factors that affect profitability in an open market for services, we build simple models to predict and explain market growth and we also suggest ways to accelerate this growth, while also achieving a higher level of economic efficiency. Copyright © 2009 John Wiley & Sons, Ltd.\",\n \"Title: From advertising profits to bandwidth prices: A quantitative methodology for negotiating premium peering.\\nYear: 2014\\n\\nAbstract: We have developed a first of its kind methodology for deriving bandwidth prices for premium direct peering between Access ISPs (A-ISPs) and Content and Service Providers (CSPs) that want to deliver content and services in premium quality. Our methodology establishes a direct link between service profitability, e.g., from advertising, user- and subscriber-loyalty, interconnection costs, and finally bandwidth price for peering. Unlike existing work in both the networking and economics literature, our resulting computational model built around Nash bargaining, can be used for deriving quantitative results comparable to actual market prices. We analyze the US market and derive prices for video that compare favorably with existing prices for transit and paid peering. We also observe that the fair prices returned by the model for high-profit/low-volume services such as search, are orders of magnitude higher than current bandwidth prices. This implies that resolving existing (fierce) interconnection tussles may require per service, instead of wholesale, peering between A-ISPs and CSPs. Our model can be used for deriving initial benchmark prices for such negotiations.\",\n \"Title: An auction mechanism for allocating the bandwidth of networks to their users\\nYear: 2007\\n\\nAbstract: We present a mechanism for auctioning bandwidth on a network-wide basis to end users or ISPs that will utilize it for the same time period. This mechanism consists of a set of simultaneous multi-unit descending-price (i.e. Dutch) auctions, one per link of the network. The per unit prices of bandwidth at the various links are asymmetric, thus reflecting the asymmetry of demand for these links. A user can be instantly allocated bandwidth over a certain path, by simultaneously bidding for the quantity desired at all relevant auctions. This winner determination rule is complemented by a payment rule of the VCG (Vickrey-Clarke-Groves) type, which provides users with the incentive to bid truthfully, thus simplifying bidding. Also, the mechanism enables the auctioneer to use his prior information on market demand anticipated and its spreading among the various links in order to set effectively the auction's parameters. We argue that our mechanism attains nearly efficient allocation of the network's bandwidth (i.e. the resulting social welfare is close to the respective maximum for the quantity decided to be sold by the auctioneer), while it is simple, scalable and applicable to real networks, even for auctioning the capacity of links owned by multiple providers and then splitting the revenue among them. Alternatively, the mechanism offers the provider the opportunity to optimize his revenue, rather than the social welfare. Since our mechanism's computational complexity is low it can serve as a fast, practical, and near-optimal solution to a generally NP-hard optimization problem.\",\n \"Title: Simple tariffs based on price multipliers for ATM VBR services.\\nYear: 2003\\n\\nAbstract: We present an approach for constructing simple time-based tariffs for variable bit-rate (VBR) connections from constant bit-rate (CBR) prices, using price multipliers that depend on the traffic parameters of the VBR connection and reflect the resource usage of VBR connections relative to CBR connections. Our approach employs an effective bandwidth bound as a proxy for the maximum amount of traffic that conforms to the connection's traffic contract, and thus for the maximum resource usage. We compare the price multipliers computed by our approach with the multipliers published by an actual ATM service provider, and with those computed using a proxy for resource usage that is based on ATM Forum's generic CAC algorithm. Although our approach is presented in the context of ATM VBR services, it can be applied for creating tariffs for service level agreements (SLAs) where the maximum amount of conforming traffic is given by leaky (or token) bucket constraints. Copyright (C) 2003 AEI.\",\n \"Title: A Framework for Analyzing the Economics of a Market for Grid Services\\nYear: 2009\\n\\nAbstract: This paper provides a single broad model for the analysis of a range of issues underlying a market for Grid services. The demand and the supply sides of such a market are being treated separately and the relation between the two sides is being studied. We provide numerical results in order to derive conclusions about the viability of a market for Grid services. Underlying our model are parameters such as the cost technologies, the random processes driving demand and supply and the size of the market. We study the effect of the model's parameters, such as risk aversion or the durability of resources, on the system's behavior, eg. on the clearing price or volume of trade.\",\n \"Title: Resource Control for the EDCA Mechanism in Multi-Rate IEEE 802.11e Networks\\nYear: 2006\\n\\nAbstract: We investigate the problem of efficient resource control for elastic traffic in IEEE 802.11e's Enhanced Distributed Channel Access (EDCA) mechanism. Our approach considers an economic modelling framework based on congestion pricing that captures how various factors, such as the probability of attempting to transmit a frame, the use of the basic CSMA/CA or the RTS/CTS procedure, and the physical layer transmission rate, contribute to congestion. We discuss the application of the framework for achieving classbased throughput differentiation, for performing explicit congestion notification (ECN) marking based on the level of congestion in the wireless channel, and for modelling the performance of TCP congestion control over EDCA. In these application scenarios we discuss how to estimate the optimum minimum contention window and the congestion prices based on the 802.11e parameters and actual measurements, in order to achieve efficient channel utilization.\",\n \"Title: Service Differentiation in ECN Networks Using Weighted Window-Based Congestion Control for Various Packet Marking Algorithms\\nYear: 2001\\n\\nAbstract: We investigate the service differentiation, in terms of average throughput, and the performance achieved using weighted window-based congestion control in networks supporting Explicit Congestion Notification (ECN). Our results show how service differentiation, queueing delay, and average throughput are affected by the increase and decrease rules of the end-system congestion control algorithms, and how they depend on the marking algorithms operating in the routers. The end-system algorithms we investigate include algorithms that achieve an average throughput proportional to some willingness-to-pay, and the marking algorithms include RED, virtual queue marking, and load-based marking.\",\n \"Title: An evaluation of pricing schemes that are based on effective usage\\nYear: 1998\\n\\nAbstract: We evaluate usage-based pricing schemes that are based on bounds of the effective bandwidth. The schemes include ones that require simple measurements (time and volume) for the whole duration of a call, and two schemes that require measurements in distinct time intervals, smaller than the duration of a call: pricing: with renegotiation and the virtual bucket scheme. Using MPEG-1 compressed video, the schemes are compared for different link capacities and buffer sizes in terms of their fairness, i.e., the ability to capture the relative amount of resources used by connections.\",\n \"Title: Design and experimental evaluation of market mechanisms for participatory sensing environments\\nYear: 2014\\n\\nAbstract: Participatory Sensing concerns the sharing of sensor information within user communities, forming a body of knowledge that can be beneficial to the community itself, either directly or through specialized applications. We introduce a framework for a marketplace where such applications can sell and buy sensor information. We focus on the buyers' side and we use various ideas from the cost-sharing literature, to propose three classes of mechanisms. We evaluate them experimentally, comparing their performance according to metrics such as social efficiency, cost coverage and budget deficit, as well as metrics related to encouraging participation.\",\n \"Title: Designing incentive compatible protocols for background data transfers\\nYear: 2014\\n\\nAbstract: We consider the design of transport protocols for background transfers with the aim of reducing the download delays of interactive (non-background) flows which share the same links. We demonstrate that Lower than Best Effort (LBE) protocols suffer from bandwidth starvation, a fact which not only creates adoption disincentives but can have a harmful effect on the interactive flows as well. Instead we propose a simple access control policy which limits the maximum number of simultaneously transmitting background flows. This maximum limit is dynamically adjusted in order to prevent bandwidth starvation thus eliminating the disincentives present in LBE protocols. Through simulations we show that our policy creates lower transfer delay for the interactive flows, as compared to that under LEDBAT [6] and TCP-LP [8] LBE protocols.\",\n \"Title: Economic incentives for adopting congestion accountability protocols\\nYear: 2014\\n\\nAbstract: New Internet protocols such as ECN, re-ECN and Conex, provide valuable information to ISPs about the congestion within a network. Such information could be useful in allocating network resources more equitably, as well as to employ new pricing schemes. However, in designing a new protocol one must take proper account of the incentive issues that thereby arise for its adoption. In this paper, we consider a duopoly market consisting of two access ISPs and formulate a noncooperative game over their choices of pricing strategies when one of them uses volume-based and the other congestion-based pricing. We prove, under mild conditions, the existence of a pure strategy Nash equilibrium for the competing providers. We further examine how the different types of transit charges affect the pricing strategies of the access ISPs. We find cases where the volume charging ISP is forced to leave the access market. We conclude that ISPs have economic incentives to adopt congestion accountability mechanisms, since they become more competitive due to the smoother and more predictable traffic they have to accommodate and to the lower transit charges they incur.\",\n \"Title: Model-checking for real-time systems\\nYear: 1990\\n\\nAbstract: This research extends CTL model-checking to the analysis of real-time systems, whose correctness depends on the magnitudes of the timing delays. For specifications, the syntax of CTL is extended to allow quantitative temporal operators. The formulas of the resulting logic, TCTL, are interpretation over continuous computation trees, trees in which paths are maps from the set of nonnegative reals to system states. To model finite-state systems the notion of timed graphs is introduced-state-transition graphs extended with a mechanism that allows the expression of constant bounds on the delays between the state transition. As the main result, an algorithm is developed for model checking, that is, for determining the truth of a TCTL formula with respect to a timed graph. It is argued that choosing a dense domain, instead of a discrete domain, to model time does not blow up the complexity of the model-checking problem. On the negative side, it is shown that the denseness of the underlying time domain makes TCTL II11-hard. The question of deciding whether a given TCTL formula is implementable by a timed graph is also undecidable\",\n \"Title: Decoupling bandwidths for networks: a decomposition approach to resource management\\nYear: 1994\\n\\nAbstract: We consider large buffer asymptotics for feed-forward networks of discrete-time queues with deterministic service rate shared by multiple classes of streams subject to work conserving service policies. First we review the concept of effective bandwidths for traffic streams sharing a common buffer subject to subject to tail constraints on the workload distribution. Next, we obtain the effective bandwidth of the departure process from such a queue, proving that in fact the effective bandwidth of the output is at worst equal to that of the input, and depending on the service rate, strictly less than that of the input. We then define the notion of a decoupling bandwidth and the associated constraints, guaranteeing that asymptotics within the network are decoupled. These results provide a framework for call admission schemes which are sensitive to constraints on the tail distri- bution of the workload or approximate cell loss probabilities. Our results require relatively weak assumptions on both the traffic streams and service policies. We consider the problem of \\\"optimal\\\" traffic shaping (via buffering) subject to a loss constraint. Finally, we discuss our results in the context of resource management for ATM networks.\",\n \"Title: Cost-Sharing Models in Participatory Sensing\\nYear: 2015\\n\\nAbstract: In Smart City and Participatory Sensing initiatives the key concept is for user communities to contribute sensor information and form a body of knowledge that can be exploited by innovative applications and data analytics services. A key aspect in all such platforms is that sensor information is not free but comes at a cost. As a result, these platforms may suffer due to insufficient sensor information made publicly available if applications do not share efficiently the cost of the sensor information they consume. We explore the design of specialized market mechanisms that match demand to supply while taking into account important positive demand externalities: sensors are digital goods and their cost can be shared by applications. We focus on the buyer side and define different demand models according to the flexibility in choosing sensor data for satisfying application needs. We then investigate the properties of various costsharing mechanisms with respect to efficiency and budget balance. In doing so, we also propose and study a new mechanism, which although lacks strategyproofness, it exhibits important efficiency improvement along with certain fairness properties.\",\n \"Title: Negotiating Premium Peering Prices: A Quantitative Model with Applications.\\nYear: 2016\\n\\nAbstract: We have developed a novel methodology for deriving bandwidth prices for premium direct peering between Access ISPs (A-ISPs) and Content and Service Providers (CSPs) that want to deliver content and services in premium quality. Our methodology establishes a direct link between service profitability, for example, from advertising, user and subscriber loyalty, interconnection costs, and finally bandwidth price for peering. Unlike existing work in both the networking and economics literature, our resulting computational model, built around Nash bargaining, can be used for deriving quantitative results comparable to actual market prices. We analyze the U.S. market and derive prices for video, that compare favorably with existing prices for transit and paid peering. We also observe that the fair prices returned by the model for high-profit/low-volume services such as search, are orders of magnitude higher than current bandwidth prices. This implies that resolving existing (fierce) interconnection tussles may require per service, instead of wholesale, peering between A-ISPs and CSPs. Our model can be used for deriving initial benchmark prices for such negotiations.\",\n \"Title: Pricing The Fast-Lanes: A Qualitative Study On The Implications Of Paid Peering Agreements\\nYear: 2016\\n\\nAbstract: Paid peering is controversial. Generally speaking, large access ISPs favour it, while Content Providers (CPs) do not. But is CP opposition to paid peering irrational, especially if their transferred payments can incentivize the ISP to make greater investment in the common infrastructure? To answer this question we analyze an ecosystem consisting of a single access ISP that directly interconnects with multiple CPs, doing so in two extreme situations: when paid peering is used for all the involved CPs, and when all parties agree on a settlement-free deal. We analyse the equilibrium, finding the total amount of investments and profits of the various stakeholders, when their revenues are affected by the total investment in infrastructure and the charges that result from the different agreements. Interestingly, it turns out whether or not there is benefit to a CP, depends on whether its business model is to sell content, and so it can recover part of the paid peering charge by raising prices to its customers, or if the CP obtains its revenue from ads. It also depends on the volume (in traffic units) per customer transaction, since we find that CPs with high volume per transaction will be charged less per byte, and so will contribute proportionally less to the total paid peering revenue that subsidizes the common ISP infrastructure. We also find a crucial role for the end-users' evaluation of the level of investments of the various market players. Another relevant factor is the access price that an ISP charges its customers.\",\n \"Title: Dynamic routing for social information sharing.\\nYear: 2017\\n\\nAbstract: Today, mobile users are intensively interconnected thanks to the emerging mobile social networks, where they share location-based information with each other when traveling on different routes and visit different areas of the city. In our model, the information collected is aggregated over all users’ trips and made publicly available as a public good. Due to information overlap, the total useful content amount increases with the diversity in path choices made by the users, and it is crucial to motivate selfish users to choose different paths, despite the potentially higher costs associated with their trips. In this paper, we combine the benefits from social information sharing with the fundamental routing problem where a unit mass of non-atomic selfish users decides their trips in a non-cooperative game by choosing between a high-cost and a low-cost path. To remedy the inefficient low-content equilibrium where all users choose to explore a single path (the low-cost path), we propose and analyze two new incentive mechanisms that can be used by the social network application, one based on side payments and the other on restricting access to content for users that choose the low cost path. Under asymmetric information about user types (their valuations for content quality), both mechanisms efficiently penalize the participants that use the low-cost path and reward the participants that take the high-cost path. They lead to greater path diversity and hence to more total available content at the social cost of reduced user participation or restricted content to part of the users. We show that user heterogeneity can have opposite effects on social efficiency depending on the mechanism used. We also obtain interesting price of anarchy results that show some fundamental tradeoffs between achieving path diversity and maintaining greater user participation, motivating a combined mechanism to further increase the social welfare. Our model extends classical dynamic routing in the case of externalities caused from traffic on different paths of the network.\",\n \"Title: Special article [paid peering: pricing and adoption incentives].\\nYear: 2016\\n\\nAbstract: Large access providers (ISPs) are seeking for new types of business agreements and pricing models to manage network costs and monetize better the provision of last-mile services. A typical paradigm of such new pricing norms is the proliferation of paid peering deals between ISPs and content providers (CPs), while on top of this, some ISPs are already experimenting with usage-based tariffs, usually through data-plans, instead of the typical fixed-based charging. In this work we define as common platform, the infrastructure in which a single ISP transacts with several CPs through peering agreements. In this context, we examine whether, and under which market conditions, the profitability of the involved stakeholders improves when the establishment of this platform is accompanied by a monetary compensation from the CPs to the ISP (paid peering), v.s. a scenario where their deal is a typical settlement-free one. In both cases, we assume that the ISP implements a usage-based access pricing scheme, implying that end-users will pay more for higher transaction rates with the CPs. Our framework captures some of the most important details of the current market, such as the various business models adopted by the CPs, the end-usersu0027 evaluation towards the ISPu0027s and CPsu0027 level of investments and the traffic rates per transaction for the offered services. By analysing the equilibrium derived by a leader-follower game, it turns out (among other practical takeaways) that whether or not the profitability of a CP improves, it highly depends on whether its business model is to sell content, or if it obtains its revenue from advertisements. Finally, we extract that consumer surplus is considerably higher under paid peering, which in turn implies improved levels of social welfare 1 .\",\n \"Title: Fair congestion control for long-lived background flows\\nYear: 2014\\n\\nAbstract: We consider a new economic framework for designing congestion control algorithms for long-lived background flows. The algorithms aim at the fair sharing of the network capacity left over by the much shorter (and usually more interactive and delay sensitive) flows, between all competing background flows even if some use TCP in particular. The algorithm acts as a weighted variant of TCP with adaptive weight adjustment based on an estimate of the average throughput that would be obtained by a TCP flow following the same path, with the resulting throughput behavior ranging from low priority to that of TCP. It turns out that this design approximately maximizes the social welfare for the background traffic for appropriately defined utilities for bandwidth averaged over long time scales and a social cost function reflecting the negative externalities on the average download time of interactive flows, over all possible protocol designs. We illustrate our approach by proposing two concrete congestion control algorithms, each of them guaranteeing the throughput of TCP in the case of resource scarcity or resource abundance respectively. Two implementation alternatives are explored; one based on adjusting the congestion window ramp up speed and another based on alternating transmitting and idle periods. We use both analysis and simulations to validate our approach of implementing new background transfer protocols.\",\n \"Title: Economics In Mobile Data Offloading With Uniform Pricing\\nYear: 2017\\n\\nAbstract: Mobile data offloading is an emerging technology to alleviate cellular network congestion and improve user service quality. In this paper, we investigate the economics of mobile data offloading through access points (APs) deployed by small cell service providers (SSPs), implementing uniform volume prices for all the mobile users (MUs) in each SSP's coverage including the overlapping area. In particular, we consider a data offloading game with a single mobile network operator (MNO) and two SSPs with overlapping coverage areas, where each SSP announces a uniform price for offloading the cellular traffic within its coverage, and the MNO determines the traffic volumes to offload. We show that there is no pure Nash equilibrium (PNE) in the data offloading game, and prove the existence of the mixed strategy Nash equilibrium (MNE) by price randomization. Furthermore, we propose a simple one shot auction mechanism that is easy to implement and has PNEs which yield SSPs the same equilibrium payoffs compared to the expected payoffs obtained at the above MNE.\",\n \"Title: Drivers, riders and service providers: the impact of the sharing economy on mobility\\nYear: 2017\\n\\nAbstract: Ride sharing denotes the practice of sharing a car such that more than one person travels in a car during a journey. Sharing rides was traditionally restricted to family members and close friends or long-distance journeys scheduled well before the intended time of departure. Only the emergence of mobile computing technologies and GPS location services in combination with electronic payments and online reputation systems provided for the technological cornerstones to make on-demand short-distance ride sharing among strangers viable. Typically individuals enter their trip details on an online platform1 which then facilitates the matching of riders with cars and drivers - and within minutes the individual's trip commences. Such platforms must attract both, supply and demand for rides.\",\n \"Title: Ex-Post Max-Min Fairness of Generalized AGV Mechanisms.\\nYear: 2017\\n\\nAbstract: We generalize the standard Arrow-d'Aspremont-Gerard-Varet (AGV) mechanism to balance the net payoffs received by all agents, while maintaining Bayesian incentive compatibility, ex-post efficiency, and ex-post budget balance of the standard AGV mechanism. In a private-value setting with independent agents' types and the principal's cost, we formulate a convex optimization problem to find the mechanism (that achieves ex-post max-min fairness) over a set of parameterized generalized AGV mechanisms, through maximizing the expected value of the minimum ex-post net payoff. We reformulate the convex program as a linear program that can be effectively solved when the number of agents is small. When the number of agents is large, we propose to solve the formulated convex program through the incremental subgradient method. Numerical results on two action models show that the proposed mechanism significantly outperforms the standard AGV mechanism in terms of the expected minimum ex-post payoff.\",\n \"Title: Congestion Control for Background Data Transfers With Minimal Delay Impact.\\nYear: 2017\\n\\nAbstract: Congestion control protocols for background data are commonly conceived and designed to emulate low priority traffic, which yields to transmission control protocol (TCP) flows. In the presence of even a few very long TCP flows, this behavior can cause bandwidth starvation, and hence, the accumulation of large numbers of background data flows for prolonged periods of time, which may ultimately have...\",\n \"Title: Market mechanisms for cooperative operation of price-maker energy storage in a power network.\\nYear: 2017\\n\\nAbstract: We explore the integration of large-scale, grid-level energy storage into wholesale electricity markets. We conduct a comparative analysis on three natural market mechanisms that have appeared in the literature: 1) the centralized mechanism according to which all batteries are centrally operated to minimize the social cost, 2) the semi-centralized mechanism under which the batteries are centrally ...\",\n \"Title: Optimal pricing for a peer-to-peer sharing platform under network externalities.\\nYear: 2018\\n\\nAbstract: In this paper, we analyse how a peer-to-peer sharing platform should price its service (when imagined as an excludable public good) to maximize profit, when each user's participation adds value to the platform service by creating a positive externality to other participants. To characterize network externalities as a function of the number of participants, we consider different bounded and unbounded user utility models. The bounded utility model fits many infrastructure sharing applications with bounded network value, in which complete coverage has a finite user valuation (e.g., WiFi or hotspot). The unbounded utility model fits the large scale data sharing and explosion in social media, where it is expected that the network value follows Metcalfe's or Zipf's law. For both models, we analyze the optimal pricing schemes to select heterogeneous users in the platform under complete and incomplete information of users' service valuations. We propose the concept of price of information (PoI) to characterize the profit loss due to lack of information, and present provable PoI bounds for different utility models. We show that the PoI = 2 for the bounded utility model, meaning that just half of profit is lost, whereas the PoI ≥ 2 for the unbounded utility model and increases as for a less concave utility function. We also show that the complicated differentiated pricing scheme which is optimal under incomplete user information, can be replaced by a single uniform price scheme that is asymptotic optimal. Finally, we extend our pricing schemes to a two-sided market by including a new group of 'pure' service users contributing no externalities, and show that the platform may charge zero price to the original group of users in order to attract the pure user group.\",\n \"Title: A Fair and Budget-Balanced Incentive Mechanism for Energy Management in Buildings.\\nYear: 2018\\n\\nAbstract: We propose a mechanism design based approach for the temperature setting in shared spaces (such as meeting rooms, library, and open office areas) in buildings. The proposed mechanism asks each occupant to report her willingness-to-pay for three possible outcomes: keeping the current temperature, lowering the temperature for one degree, and raising the temperature for one degree. Collecting the occ...\",\n \"Title: Recommending Paths: Follow or Not Follow?\\nYear: 2018\\n\\nAbstract: Mobile social network applications constitute an important platform for traffic information sharing, helping users collect and share sensor information about the driving conditions they experience on the traveled path in real time. In this paper we analyse the simple but fundamental model of a platform choosing between two paths: one with known deterministic travel cost and the other that alternates over time between a low and a high random cost states, where the low and the high cost states are only partially observable and perform respectively better and worse on average than the fixed cost path. The more users are routed over the stochastic path, the better the platform can infer its actual state and use it efficiently.At the Nash equilibrium, if asked to take the riskier path, in many cases selfish users (that are allowed to have access to the information collected by the platform) will myopically disregard the optimal path suggestions of the platform, leading to a suboptimal system without enough exploration on the stochastic path. We prove the interesting result that if the past collected information is hidden from users, the system becomes incentive compatible and even ‘sophisticated’ users (in the sense that they have full capability to reverse-engineer the platform’s recommendation and derive the path state distribution conditional on the recommendation) prefer to follow the platform’s recommendations. In a more practical setting where the platform implements a model-free Q-learning algorithm to minimise the social travel cost, our analysis suggests that increasing the accuracy of the learning algorithm increases the range of system parameters for which sophisticated users follow the recommendations of the platform, becoming in the limit fully incentive compatible. Finally, we extend the two-path model to include more stochastic paths, and show that incentive compatibility holds under our information restriction mechanism.\",\n \"Title: Mobile Data Offloading with Uniform Pricing and Overlaps\\nYear: 2019\\n\\nAbstract: Mobile data offloading is an emerging technology to alleviate cellular network congestion and improve user service quality. In this paper, we investigate the\\n<italic xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">economics</italic>\\nof mobile data offloading through access points (APs) deployed by small cell service providers (SSPs), implementing uniform volume prices for all the mobile users (MUs) in each SSP's coverage including the overlapping area. In particular, we consider a data offloading game with a single mobile network operator (MNO) and two SSPs with overlapping coverage areas, where each SSP announces a uniform price for serving the cellular traffic within its coverage, and the MNO determines the traffic volumes to offload. We show that there is no pure Nash equilibrium (PNE) under such price competition, and determine the corresponding mixed strategy Nash equilibrium (MNE) using price randomization. As a practical solution, we propose a simple one shot auction mechanism that is easy to implement and has PNEs which is payoff equivalent with the MNE under price competition. We believe that this simple mechanism due to its simplicity of determining the equilibrium prices could be used in the negotiation between the SSPs and the MNO to determine the average service prices. Finally, we study the strategic topological infrastructure placement problem using a 1-dimension (1D, linear) user traffic flow model and a 2-dimension (2D) user traffic flow model when SSPs compete assuming uniform price competition as above. We show that the first mover in the placement problem will deploy its APs to cover more than half of the total flow volume and has an advantage to obtain a higher equilibrium payoff.\",\n \"Title: Verifying Automata Specifications of Probabilistic Real-time Systems\\nYear: 1991\\n\\nAbstract: . We present a model-checking algorithm for a system presented as ageneralized semi-Markov process and a specification given as a deterministic timedautomaton. This leads to a method for automatic verification of timing propertiesof finite-state probabilistic real-time systems.Keywords: Real-time, Probabilistic systems, Automatic verification.1 IntroductionThere is increasing awareness that unexpected behavior from interacting processes cancause serious problems. This observation...\",\n \"Title: Invited talk II summary: incentive schemes in memory-less P2P systems\\nYear: 2005\\n\\nAbstract: The asymptotic analysis of certain public good models for p2p file sharing systems focusing on content availability, suggests that when the aim is to maximize social welfare, a fixed contribution scheme in terms of the number of files shared per unity of time can be asymptotically optimal as the number of participants n grows to infinity (see [1] and references therein). Such an incentive scheme is very simple and attractive, and is also suitable for other p2p applications with similar public good charasteristics such as WLAN peering. However, its enforcement is not straightforward in cases where no trusted software or central entity accounting for peers' transactions can be assumed and peers are free to change their identity with no cost. That is, when no sort of user memory is available to be able to identify and punish the potential free riders.\",\n \"Title: Market Mechanisms For Energy Storage Planning And Operation In A Power Network\\nYear: 2018\\n\\nAbstract: We conduct a comparative analysis on three joint market mechanisms for energy storage investment and operation under locational marginal pricing: i) socially optimal storage investment with centralized operation, ii) profit -maximizing storage investment with centralized operation, and iii) profit maximizing storage investment with deregulated operation. For the first mechanism (according to which a social planner centrally optimizes the sitting, sizing, and operation of batteries), we show that the revenue collected from storage operation exactly covers the investment cost at a social optimum. Under the last two mechanisms, a profit maximizing firm strategically determines the sitting and sizing of batteries, and the storage operation is either centrally optimized for cost minimization (mechanism ii) or completely controlled by the storage owner for profit maximization (mechanism iii). Numerical results on the IEEE 57-bus test system reveal that under profit maximizing investment, the centralized and deregulated storage operations usually lead to almost identical outcomes (on installed storage capacity, social cost, and storage profit).\",\n \"Title: Throughput and Pricing of Ridesharing Systems\\nYear: 2019\\n\\nAbstract: We introduce a queueing model for passengers waiting to get transport by drivers participating in a ridesharing system. The effect of available drivers’ mobility pattern, their willingness to accept rides in a given location, and the incentives offered by the platform, on system throughput is considered. We characterize the largest set of passenger arrival rates which can be served by a fixed number of drivers under any policy dictating the mobility pattern of available drivers. It turns out that any rate in this set can be achieved by offering appropriate constant but region-dependent rewards to drivers for passenger pick up. Moreover, it is shown that dynamic rewards which scale in proportion to the number of passengers waiting for pick up in each region, not only maximize throughput but also significantly decrease pick up delays.\",\n \"Title: Average-Case Analysis of Greedy Matching for D2D Resource Sharing\\nYear: 2021\\n\\nAbstract: Given the proximity of many wireless users and their diversity in consuming local resources (e.g., data-plans, computation and even energy resources), device-to-device (D2D) resource sharing is a promising approach towards realizing a sharing economy. In the resulting networked economy, n users segment themselves into sellers and buyers that need to be efficiently matched locally. This paper adopt...\"\n]","isConversational":false},"profile_size":{"kind":"number","value":86,"string":"86"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13562,"string":"13,562"}}},{"rowIdx":13563,"cells":{"question_id":{"kind":"string","value":"Peter Schelkens"},"question":{"kind":"string","value":"Generate an abstract for the title \"Spatio-Temporally Consistent Color and Structure Optimization for Multiview Video Color Correction\" using the following items: Multiview video, Color correction, Spatio-temporal correspondence, Optimization algorithm, Evaluation model."},"answer":{"kind":"string","value":"When compared to conventional 2-D video, multiview video can significantly enhance the visual 3-D experience in 3-D applications by offering horizontal parallax. However, when processing images originating from different views, it is common that the colors between the different cameras are not well- calibrated . To solve this problem, a novel energy function -based color correction method for multiview camera setups is proposed to enforce that colors are as close as possible to those in the reference image but also that the overall structural information is well-preserved. The proposed system introduces a spatio-temporal correspondence matching method to ensure that each pixel in the input image gets bijectively mapped to a reference pixel. By combining this mapping with the original structural information, we construct a global optimization algorithm in a Laplacian matrix formulation and solve it using a sparse matrix solver. We further introduce a novel forward-reverse objective evaluation model to overcome the problem of lack of ground truth in this field. The visual comparisons are shown to outperform state-of-the-art multiview color correction methods, while the objective evaluation reports PSNR gains of up to 1.34 dB and SSIM gains of up to 3.2%, respectively."},"correct_answer":{"kind":"string","value":"When compared to conventional 2-D video, multiview video can significantly enhance the visual 3-D experience in 3-D applications by offering horizontal parallax. However, when processing images originating from different views, it is common that the colors between the different cameras are not well- calibrated . To solve this problem, a novel energy function -based color correction method for multiview camera setups is proposed to enforce that colors are as close as possible to those in the reference image but also that the overall structural information is well-preserved. The proposed system introduces a spatio-temporal correspondence matching method to ensure that each pixel in the input image gets bijectively mapped to a reference pixel. By combining this mapping with the original structural information, we construct a global optimization algorithm in a Laplacian matrix formulation and solve it using a sparse matrix solver. We further introduce a novel forward-reverse objective evaluation model to overcome the problem of lack of ground truth in this field. The visual comparisons are shown to outperform state-of-the-art multiview color correction methods, while the objective evaluation reports PSNR gains of up to 1.34 dB and SSIM gains of up to 3.2%, respectively."},"documents":{"kind":"list like","value":["Title: Wavelet-based scalable L-infinity-oriented compression.\nYear: 2006\n\nAbstract: Among the different classes of coding techniques proposed in literature, predictive schemes have proven their outstanding performance in near-lossless compression. However, these schemes are incapable of providing embedded L(infinity)-oriented compression, or, at most, provide a very limited number of potential L(infinity) bit-stream truncation points. We propose a new multidimensional wavelet-based L(infinity)-constrained scalable coding framework that generates a fully embedded L(infinity)-oriented bit stream and that retains the coding performance and all the scalability options of state-of-the-art L2-oriented wavelet codecs. Moreover, our codec instantiation of the proposed framework clearly outperforms JPEG2000 in L(infinity) coding sense.","Title: Spatio-temporal-SNR scalable wavelet coding with motion-compensated DCT base-layer architectures\nYear: 2003\n\nAbstract: It has been demonstrated recently that 3-D wavelet coding with motion-compensated temporal filtering (MCTF) provides a wide range of spatio-temporal-SNR scalability with state-of-the-art coding performance. However, the coding system is very different from the already standardized motion-compensated DCT (MC-DCT) video coders such as MPEG-2, MPEG-4, or H.26L. Nonetheless, the market acceptance of the new scalable technology will come much easier if backward compatibility to such previous standards would exist. In this paper, we present a new coder architecture where the base layer can be a standard MC-DCT coder, while the enhancement layer is an in-band MCTF codec operating in the overcomplete wavelet domain. First, we propose a simple extension of the scalable 3-D wavelet codec with a standard MC-DCT base layer. As a second step, to improve the performance of the proposed scalable coder over a wide range of bit-rates, we describe several small extensions to the standardized MC-DCT codec structure that can be applied for the base layer coding.","Title: Estimating the Detectability of Small Lesions in High Resolution MR Compressed Images\nYear: 2008\n\nAbstract: Quality degradation in high resolution magnetic resonance (MR) images compressed with JPEG2000 is studied. The present study combines the results of ten quantitative quality criteria with four mathematical observer models to estimate the maximum achievable compression rate (CR) that does not affect an effective diagnosis of specific pathologies related with small lesions detection such as Multiple Sclerosis or Virchow Robins disease. The graphical behavior of metrics employed is presented. At bitrate = 0.062 bpp (CR=160:1) the most compression is achieved while images still preserve the information needed for a safe diagnosis. Images compressed using smaller values than this would be no longer useful for diagnosis tasks. This result facilitates a safer use of the JPEG 2000 codec when compressing high resolution images like the images tested, and assures a safer diagnosis task when dealing with small size lesions that define certain pathology.","Title: Wavelet Coding of Volumetric Medical Datasets.\nYear: 2003\n\nAbstract: Several techniques based on the three-dimensional (3-D) discrete cosine transform (DCT) have been proposed for volumetric data coding. These techniques fail to provide lossless coding coupled with quality and resolution scalability, which is a significant drawback for medical applications. This paper gives an overview of several state-of-the-art 3-D wavelet coders that do meet these requirements and proposes new compression methods exploiting the quadtree and block-based coding concepts, layered zero-coding principles, and context-based arithmetic coding. Additionally, a new 3-D DCT-based coding scheme is designed and used for benchmarking. The proposed wavelet-based coding algorithms produce embedded data streams that can be decoded up to the lossless level and support the desired set of functionality constraints. Moreover, objective and subjective quality evaluation on various medical volumetric datasets shows that the proposed algorithms provide competitive lossy and lossless compression results when compared with the state-of-the-art.","Title: FULLY-SCALABLE WAVELET VIDEO CODING USING IN-BAND MOTION COMPENSATED TEMPORAL FILTERING\nYear: 2003\n\nAbstract: This paper presents a novel fully-scalable wavelet video coding scheme that performs efficient open-loop motion compensated temporal filtering (MCTF) in the wavelet domain (in-band). Unlike the conventional spatial-domain MCTF (SDMCTF) schemes, which apply MCTF on the original image data and then encode the residual image using a critically-sampled wavelet transform, the framework presented here applies the in-band MCTF (IBMCTF) after the discrete wavelet transform (DWT) is performed in the spatial dimensions. To overcome the inefficiency of motion estimation (ME) in the wavelet domain, a complete-to-overcomplete DWT (CODWT) is performed. The proposed framework provides improved quality (SNR) and temporal scalability as compared with existing in-band closed-loop temporal prediction schemes with ODWT and improved spatial scalability as compared to SDMCTF. We present a thorough comparison between SDMCTF and the proposed IBMCTF in terms of coding efficiency and scalability. Furthermore, we describe several extensions that enable the filtering of the various bands to be performed independently, based on the resolution, sequence content, complexity requirements and desired scalability.","Title: Power-Aware Computing Systems\nYear: 2007\n\nAbstract: From April 3rd to April 8th 2005, a scientific workshop on the topic of \"Power-aware Computing Systems\" was held at Schloss Dagstuhl, Germany. The main seminar result is a classification of the obstacles, and therefore research directions, with respect to power consumption seen for different classes of devices, ranging from ultra low-power devices, over handheld devices, to servers and work stations. In a next step the seminar identified the impact different levels of dealing with power concerns can have. This paper summarises the objectives and structure as well as the outcome of that workshop.","Title: Towards Standardized Integration of Images in the Cloud of Linked Data.\nYear: 2013\n\nAbstract: Currently, there are several ways of describing and referring to images in RDF. This ambiguity results in a proliferation of vocabularies for image descriptions, complicating the cross-community data integration of information related to digital images. In addition, there are no standardized guidelines on how to integrate RDF data into the image metadata. Therefore, the JPEG standardization committee has recently initiated some activities to streamline the integration of images in the cloud of Linked Data. One effort is the standardization of an ontology for describing digital images. JPEG aims at providing a technology that enables the uniform description of photos and videos with technical, administrative and semantic metadata compliant with the RDF specification and the principles of Linked Data. Secondly, specifications to integrate RDF data into JPEG images are elaborated. Finally, since descriptions often only apply to a certain part of an image, a last effort is to formalize the specification of regions of interest. In this paper, members of the JPEG committee provide a detailed overview of these ongoing activities.","Title: On hybrid directional transform-based intra-band image coding\nYear: 2007\n\nAbstract: In this paper, we propose a generic hybrid oriented-transform and wavelet-based image representation for intra-band image coding. We instantiate for three popular directional transforms having similar powers of approximation but different redundancy factors. For each transform type, we design a compression scheme wherein we exploit intra-band coefficient dependencies. We show that our schemes outperform alternative approaches reported in literature. Moreover, on some images, we report that two of the proposed codec schemes outperform JPEG2000 by over 1dB. Finally, we investigate the trade-off between oversampling and sparsity and show that, at low rates, hybrid coding schemes with transform redundancy factors as high as 1.25 to 5.8 are capable in fact of outperforming JPEG2000 and its critically-sampled wavelets.","Title: Visually lossless screen content coding using HEVC base-layer.\nYear: 2013\n\nAbstract: This paper presents a novel two-layer coding framework targeting visually loss less compression of screen content video. The proposed framework employs the conventional HEVC standard for the base-layer. For the enhancement layer, a hybrid of spatial and temporal block-prediction mechanism is introduced to guarantee a small energy of the error-residual. Spatial prediction is generally chosen for dynamic areas, while temporal predictions yield better prediction for static areas in a video frame. The prediction residual is quantized based on whether a given block is static or dynamic. Run-length coding, Golomb based binarization and context-based arithmetic coding are employed to efficiently code the quantized residual and form the enhancement-layer. Performance evaluations using 4: 4: 4 screen content sequences show that, for visually lossless video quality, the proposed system significantly saves the bit-rate compared to the two-layer lossless HEVC framework.","Title: Scalable and channel-adaptive unequal error protection of images with LDPC codes\nYear: 2006\n\nAbstract: This paper considers the design of an optimal joint source-channel coding system employing scalable wavelet-based source coders and unequal error protection for error-resilient transmission over binary erasure channels. We theoretically show that the expected average rate-distortion function in the separately encoded subbands is convex with monotonically decreasing slopes and use this fact to propose a new algorithm which minimizes the end-to-end distortion. This algorithm is based on a simplified Viterbi-search followed by Lagrangian-optimization. We show that our proposed solution results in significant complexity reductions while providing very near-to-optimal performance.","Title: Modeling Wavelet Coefficients for Wavelet Subdivision Transforms of 3D Meshes\nYear: 2010\n\nAbstract: In this paper, a Laplacian Mixture (LM) model is proposed to accurately approximate the observed histogram of the wavelet coefficients produced by lifting-based subdivision wavelet transforms. On average, the proposed mixture model gives better histogram fitting for both normal and non-normal meshes compared to the traditionally used Generalized Gaussian (GG) distributions. Exact closed-form expressions for the rate and the distortion of the LM probability density function quantized using generic embedded deadzone scalar quantizer (EDSQ) are derived, without making high-rate assumptions. Experimental evaluations carried out on a set of 3D meshes reveals that, on average, the D-R function for the LM model closely follows and gives a better indication of the experimental D-R compared to the D-R curve of the competing GG model. Optimal embedded quantization for the proposed LM model is experimentally determined. In this sense, it is concluded that the classical Successive Approximation Quantization (SAQ) is an acceptable, but in general, not an optimal embedded quantization solution in wavelet-based scalable coding of 3D meshes.","Title: Scalable wavelet video-coding with in-band prediction - the bottom-up overcomplete discrete wavelet transform\nYear: 2002\n\nAbstract: A new in-band motion compensation algorithm for wavelet-based video coding is proposed: the Bottom-Up Prediction algorithm (BUP). The BUP algorithm overcomes the periodic shift-invariance of the discrete wavelet transform (DWT) and is formalized into new prediction rules using filtering operations. BUP is based on the relationships between the subbands of the shifted input signal and the subbands of the non-shifted reference signal, whereby the number of shifts is limited by the periodic shift-invariance of the DWT. We derive the algorithm for the 1-D DWT with three decomposition levels. The combination of all prediction rules of the BUP algorithm defines a new transform: the Bottom-Up Overcomplete DWT or BUP ODWT, which is shift- invariant. The BUP ODWT calculates the overcomplete subbands by applying the prediction rules to the critically sampled subbands of a wavelet-transformed image. The envisaged application for the BUP algorithm is spatially scalable video coding.","Title: A New Family Of Embedded Multiple Description Scalar Quantizers\nYear: 2004\n\nAbstract: A new family of Embedded Multiple Description Scalar Quantizers (EMDSQ) that Supports the progressive transmission of images over variable-bandwidth error-prone channels is proposed in this paper. A control mechanism that allows for tuning the redundancy between the two descriptions for each quantization level is also designed. The employed mechanism enables the control of the tradeoff between the coding efficiency and error-resilience, and provides an increased robustness by improving the error resilience in the most important layers of the embedded bit-streams. Instantiations of the proposed family are incorporated in a wavelet-based embedded coding system, and the redundancy-control mechanism is practically demonstrated. Experimental results show that the proposed EMDSQ outperform the state-of-the-art Multiple Description Uniform Scalar Quantizers (MDUSQ) previously proposed in literature for error-resilient progressive image transmission.","Title: L-infinite Coding of 3D Representations of Human Affect\nYear: 2012\n\nAbstract: Off-line scanning, coding, transmission and remote animation of the human affect represents a possible processing pipeline for providing 3D immersion in virtual worlds. In this paper we target applications that make use of compact and scalable 3D representations of human affect and require close control over the local error introduced by lossy coding of the mesh geometry. To satisfy this requirement, we propose a novel L-infinite wavelet-based semi-regular mesh coding system. The system lies in contrast with classical mesh coding approaches which make use of the L-2 distortion metric. Specifically, in contrast to an L-2 driven implementation, the proposed system provides a bound on the local error on each vertex resulting from scalar embedded quantization of the wavelet coefficients. The experiments show that the proposed system provides scalability in L-infinite sense and that it outperforms the state-of-the art in L-infinite mesh coding.","Title: Probabilistic motion-compensated prediction in distributed video coding\nYear: 2013\n\nAbstract: Distributed video coding (DVC) constitutes an original coding framework to meet the stringent requirements imposed by uplink-oriented and low-power mobile video applications. The quality of the side information available to the decoder and the efficiency of the employed channel codes are primary factors determining the success of a DVC system. This contribution introduces two novel techniques for probabilistic motion compensation in order to generate side information at the Wyner-Ziv decoder. The employed DVC scheme uses a base layer, serving as a hash to facilitate overlapped block motion estimation at the decoder side. On top of the base layer, a supplementary Wyner-Ziv layer is coded in the DCT domain. Both proposed probabilistic motion compensation techniques are driven by the actual correlation channel statistics and reuse information contained in the hash. Experimental results report significant rate savings caused by the novel side information generation methods compared to previous techniques. Moreover, the compression performance of the presented DVC architecture, featuring the proposed side-information generation techniques, delivers state-of-the-art compression performance.","Title: Complete-To-Overcomplete Discrete Wavelet Transforms For Scalable Video Coding With Mctf\nYear: 2003\n\nAbstract: Techniques for full scalability with motion-compensated temporal filtering (MCTF) in the wavelet-domain (in-band) are presented in this paper. The application of MCTF in the wavelet domain is performed after the production of the overcomplete discrete wavelet transform from the critically-sampled decomposition, a process that occurs at both the encoder and decoder side. This process, which is a complete-to-overcomplete discrete wavelet transform, is critical for the efficiency of the system with respect to scalability, coding performance and complexity. We analyze these aspects of the system and set the necessary constraints for drift-free video coding with in-band MCTF. As a result, the proposed architecture permits the independent operation of MCTF within different resolution levels or even different subbands of the transform and allows the successive refinement of the video information in resolution, frame-rate and quality.","Title: Wyner-Ziv video coding for wireless lightweight multimedia applications.\nYear: 2012\n\nAbstract: Wireless video communications promote promising opportunities involving commercial applications on a grand scale as well as highly specialized niche markets. In this regard, the design of efficient video coding systems, meeting such key requirements as low power, mobility and low complexity, is a challenging problem. The solution can be found in fundamental information theoretic results, which gave rise to the distributed video coding (DVC) paradigm, under which lightweight video encoding schemes can be engineered. This article presents a new hash-based DVC architecture incorporating a novel motion-compensated multi-hypothesis prediction technique. The presented method is able to adapt to the regional variations in temporal correlation in a frame. The proposed codec enables scalable Wyner-Ziv video coding and provides state-of-the-art distributed video compression performance. The key novelty of this article is the expansion of the application domain of DVC from conventional video material to medical imaging. Wireless capsule endoscopy in particular, which is essentially wireless video recording in a pill, is proven to be an important application field. The low complexity encoding characteristics, the ability of the novel motion-compensated multi-hypothesis prediction technique to adapt to regional degrees of temporal correlation (which is of crucial importance in the context of endoscopic video content), and the high compression performance make the proposed distributed video codec a strong candidate for future lightweight (medical) imaging applications.","Title: JPSearch: Metadata Interoperability During Image Exchange [Standards in a Nutshell].\nYear: 2012\n\nAbstract: Billions of new images are generated daily. However, the management and exchange of the associated metadata remains cumbersome. Therefore, the Joint Photographic Experts Group (JPEG) has recently promoted a new set of technologies enabling the interoperability between image repositories and/or their clients to an international standard. JPSearch is a suite of specifications that supports the enric...","Title: Wavelet-based fixed and embedded L-infinite-constrained coding\nYear: 2002\n\nAbstract: A new wavelet-based L-infinite-constrained fixed and embedded coding technique is proposed in this paper. For any desired distortion bound, the embedded bit stream can be truncated at a corresponding bit-rate, for which the required upper bound on the individual elements of the reconstruction error signal is guaranteed. The original signal can also be coded up to a fixed a priori user-defined distortion bound, ranging up to lossless coding. The proposed coding technique supports both 2D (images) and 3D (volumetric) input data.","Title: JPSearch: An answer to the lack of standardization in mobile image retrieval\nYear: 2013\n\nAbstract: The amount of imagery that is available via various content providers grows at a staggering rate. Contemporary handheld devices allow accessing these content providers anytime, anyplace. Moreover, using the build-in camera, one can easily contribute to these online image repositories. The present lack of consistency in how repositories are accessed for retrieval or syncing complicates inter-operability between devices and systems. JPSearch, a recent standardization initiative within the Joint Photographic Experts Group (JPEG) committee, resolves this predicament by defining the interfaces and protocols for data exchange between them, while restricting as little as possible how those devices, systems or components perform their task.","Title: Motion and texture rate-allocation for prediction-based scalable motion-vector coding\nYear: 2005\n\nAbstract: Modern video coding applications require data transmission over variable-bandwidth wired and wireless network channels to a variety of terminals, possibly having different screen resolutions and available computing power. Scalable video coding technology is needed to optimally support these applications. Recently proposed wavelet-based video codecs employing spatial-domain motion-compensated temporal filtering (SDMCTF) provide quality, resolution and frame-rate scalability while delivering compression performance comparable to that of H.264, the state-of-the-art in single-layer video coding. These codecs require quality-scalable coding of the motion vectors to support a large range of bit-rates with optimal compression efficiency. In this paper, the practical use of prediction-based scalable motion-vector coding in the context of scalable SDMCTF-based video coding is investigated. Extensive experimental results demonstrate that, irrespective of the employed motion model, our prediction-based scalable motion-vector codec (MVC) systematically outperforms state-of-the-art wavelet-based solutions for both lossy and lossless compression. A new rate-distortion optimized rate-allocation strategy is proposed, capable of optimally distributing the available bit-budget between the different frames and between the texture and motion information, making the integration of the scalable MVC into a scalable video codec possible. This rate-allocation scheme systematically outperforms heuristic approaches previously employed in the literature. Experiments confirm that by using a scalable MVC, lower bit-rates can be attained without sacrificing motion-estimation efficiency and that the overall coding performance at low rates is significantly improved by a better distribution of the available rate between texture and motion information. The only downside of scalable motion-vector coding is a slight performance loss incurred at high bit-rates.","Title: Wavelet-based compression of medical images: Protocols to improve resolution and quality scalability and region-of-interest coding\nYear: 1999\n\nAbstract: The paper describes a methodology to improve the scalability support of an embedded bit stream, generated with a wavelet-based compression algorithm, and a generic protocol to handle multiple regions-of-interest (ROIs). The generic scheme, illustrated for embedded zero-tree wavelet (EZW) coding, exploits the inherent graceful degradation capabilities of wavelet-based compression methods and ensures an optimal trade-off between the image reconstruction quality and the compression ratio. Additionally, an efficient protocol is proposed to handle multiple ROIs in an interactive client-server set-up for telemedicine applications. The processing of the ROIs takes place in the wavelet domain.","Title: Maximum Likelihood Laplacian Correlation Channel Estimation in Layered Wyner-Ziv Coding\nYear: 2014\n\nAbstract: The distributed source coding theory teaches that efficient compression can be achieved by exploiting the correlation between encoded sources at the decoder only. A critical issue in this coding paradigm is how to accurately estimate the correlation statistics at the decoder. This paper proposes a novel correlation channel estimation method designed for generic layered Wyner-Ziv (WZ) coding. Driven by real-life applications, i.e., video coding, the correlation noise is assumed to be memoryless zero-mean Laplacian with a block-stationary behavior. Unlike existing methods, the proposed technique derives a novel maximum likelihood estimate of the noise scaling parameter per stationarity block. Adhering to layered coding, the derived estimate is successively refined as more stages are decoded. The Crámer-Rao lower bound (CRLB) for our estimator is derived and the mean squared error performance of the proposed successive refinement algorithm is evaluated through simulations. Apart from achieving an accuracy close to the derived CRLB, the proposed algorithm is shown to yield a WZ rate-distortion performance approaching the ideal but unrealistic scenario where the decoder knows a priori the correlation statistics. Moreover, the proposed technique has been integrated into our latest WZ video codec for wireless capsule endoscopy. Experimentation reveals that the proposed method delivers improved compression performance compared to state-of-the-art techniques, while simultaneously reducing the overall decoding complexity.","Title: Constant quality video coding using video content analysis\nYear: 2005\n\nAbstract: In the literature, several rate control techniques have been proposed to aim at the optimal quality of digitally encoded video under given bit budget, channel rate and buffer size constraints. Typically, these approaches are group-of-picture (GOP) based. For longer, heterogeneous sequences, they become unacceptably complex or struggle with model mismatches. In this paper, an off-line segment-based rate control approach is proposed for controlling the distortion variation across successive shots of a video sequence when encoding with single-layer (MPEG-4 baseline, MPEG-4 AVC) and scalable (wavelet) video codecs. Consistent quality is achieved by optimally distributing the available bits among the different segments, based on efficient rate-distortion (R-D) modelling of each segment. The individual segments are defined based on shot segmentation and activity analysis techniques. The algorithm is formulated for three different distribution models: download, progressive download and streaming. The results indicate that the proposed technique improves the quality consistency significantly, while the processing overhead compared to classical two-pass variable bit-rate (VBR) encoding is limited.","Title: Side-information-dependent correlation channel estimation in hash-based distributed video coding.\nYear: 2012\n\nAbstract: In the context of low-cost video encoding, distributed video coding (DVC) has recently emerged as a potential candidate for uplink-oriented applications. This paper builds on a concept of correlation channel (CC) modeling, which expresses the correlation noise as being statistically dependent on the side information (SI). Compared with classical side-information-independent (SII) noise modeling adopted in current DVC solutions, it is theoretically proven that side-information-dependent (SID) modeling improves the Wyner-Ziv coding performance. Anchored in this finding, this paper proposes a novel algorithm for online estimation of the SID CC parameters based on already decoded information. The proposed algorithm enables bit-plane-by-bit-plane successive refinement of the channel estimation leading to progressively improved accuracy. Additionally, the proposed algorithm is included in a novel DVC architecture that employs a competitive hash-based motion estimation technique to generate high-quality SI at the decoder. Experimental results corroborate our theoretical gains and validate the accuracy of the channel estimation algorithm. The performance assessment of the proposed architecture shows remarkable and consistent coding gains over a germane group of state-of-the-art distributed and standard video codecs, even under strenuous conditions, i.e., large groups of pictures and highly irregular motion content.","Title: Semi-Regular Remeshing With Reduced Remeshing Error\nYear: 2010\n\nAbstract: In this paper we present a remeshing algorithm which drastically reduces the aliasing artifacts inherent in regularly-sampled remeshed objects. Starting from a semi-regular mesh, the proposed algorithm reduces the remeshing error and avoids aliasing by displacing vertices such that most samples of the original mesh are present in the remeshed model as well. Computational efficiency is provided by using a search-tree, which efficiently gathers vertices near a given point in a 3D space. Compared to the state-of-the-art semi-regular remesher, the proposed remesher drastically improves the visual quality of the high-frequency regions in remeshed objects. Additionally, the proposed algorithm yields a lower remeshing error, which is reflected by a significantly increased PSNR upper-bound in wavelet-based compression of such meshes.","Title: Platform-scalable Task Partition and Multilevel Buffering in Multi-processor Plessey Corner Detector\nYear: 2007\n\nAbstract: The Plessey corner detector is a key technological component in scene analysis, stereo matching, and object tracking. Due to its high computation complexity, earlier fast implementations mainly focused on hardware implementations. This paper explores the viability of a multi-processor software implementation. A scalable task partitioning for efficiently mapping the Plessey algorithm on a multi-processor platform is proposed. The task partition ensures platform scalability, low inter-processor communication overhead and a well-balanced workload in each task. In addition, a multilevel buffering scheme is presented, minimizing the external memory accesses in each task to one image pixel read per calculated corner response value. The effectiveness of the proposed task partition and buffering scheme has been verified on (i) a cycle accurate simulator with shared memory and (ii) a multiple-TI-C64 DSP board using a message passing paradigm. The proposed solution combines good platform scalability with an additional 30% speedup gain over straightforward parallelization schemes.","Title: Overlapped Block Motion Estimation and Probabilistic Compensation with Application in Distributed Video Coding\nYear: 2009\n\nAbstract: It has been recently demonstrated that, in distributed video coding (DVC) side-information dependent modeling of the correlation channel brings significant performance gains over side-information independent assumptions. In this letter, we present a novel technique enabling advanced side-information dependent estimation of the correlation channel at the decoder starting from a very coarse knowledg...","Title: Scalable wavelet video-coding with in-band prediction - implementation and experimental results\nYear: 2002\n\nAbstract: In this paper we elaborate on a recently proposed approach for scalable video-coding based on in-band prediction in the overcomplete wavelet domain. It is shown that, through a new calculation scheme for the level-by-level complete to overcomplete discrete wavelet transform (DWT) that exploits certain symmetries, important reductions in the multiplication budget are obtained in comparison to the fastest-known algorithm of the literature. Based on the derived overcomplete transform-domain coefficients, a pixel-accurate motion estimation and compensation (ME/MC) algorithm is proposed, which provides a hybrid-coding framework that supports full scalability in resolution, quality and frame rate. To give an indication of the coding performance of such a system, some preliminary results are reported.","Title: Mosaicing of Fibered Fluorescence Microscopy Video\nYear: 2008\n\nAbstract: Fibered fluorescence microscopy is a recent developed image modality using a fiber optic probe connected to a laser scanning unit. It allows for in-vivo scanning of small animal subjects by moving the probe along the tissue surface. During the scans images are continuity captured, allowing to acquire an area larger then the field of view of the probe as a video. But there is still a need to obtain a single static image from the multiple overlapping frames. In this paper we introduce a mosaicing procedure for this kind of video sequence. An additional motivation for the mosaicing is the use of overlapping redundant information to improve the signal to noise level of the acquisition, since the individual frames tend to have both high noise levels and intensity inhomogeneities.","Title: An investigation into the performance of reduction algorithms under load imbalance\nYear: 2012\n\nAbstract: Today, most reduction algorithms are optimized for balanced workloads; they assume all processes will start the reduction at about the same time. However, in practice this is not always the case and significant load imbalances may occur and affect the performance of said algorithms. In this paper we investigate the impact of such imbalances on the most commonly employed reduction algorithms and propose a new algorithm specifically adapted to the presented context. Firstly, we analyze the optimistic case where we have a priori knowledge of all imbalances and propose a near-optimal solution. In the general case, where we do not have any foreknowledge of the imbalances, we propose a dynamically rebalanced tree reduction algorithm. We show experimentally that this algorithm performs better than the default OpenMPI and MVAPICH2 implementations.","Title: An offline bidirectional tracking scheme\nYear: 2005\n\nAbstract: A generic bi-directional scheme is proposed that robustifies the estimation of the maximum-a-posteriori (MAP) sequence of states of a visual object. It enables creative, non-technical users to obtain the path of interesting objects in offline available video material, which can then be used to create interactive movies. To robustify against tracker failure the proposed scheme merges the filtering distributions of a forward tracking particle filter and a backward tracking particle filter at some timesteps, using a reliability-based voting scheme such as in democratic integration. The MAP state sequence is obtained using the Viterbi algorithm on reduced state sets per timestep derived from the merged distributions and is interpolated linearly where tracking failure is suspected. The presented scheme is generic, simple and efficient and shows good results for a color-based particle filter.","Title: Reversible DCT-based lossy-to-lossless still image compression\nYear: 2013\n\nAbstract: This paper presents a novel still image compression scheme that extends the traditional JPEG standard with lossy-to-lossless compression support. The system follows a two-layer design approach which allows for backward compatibility with the conventional JPEG standard for the base layer and provides lossless compression when decoding the enhancement layer. The system employs several coding tools, including quadtree coding, spatial domain prediction, reversible discrete cosine transforms, and context-based arithmetic coding to efficiently encode losslessly the enhancement layer. Performance evaluations using a standard JPEG set of images show that the proposed system yields similar lossless compression performance to the state-of-the-art single-layer JPEG-LS standard while providing quality scalability and a JPEG-compatible base layer at the same time.","Title: Jpeg2000 Part 10 - Volumetric Data Encoding\nYear: 2006\n\nAbstract: The Joint Photographic Experts Group (JPEG) committee (ISO/IEC JTCI/SC29/WGI) is currently pursuing the standardization of a three-dimensional extension of the JPEG-2000 standard (Parts 1 and 2) to support the encoding of volumetric data sets. This extension, Part 10 - Extensions for Three-Dimensional Data (MD), will support functionalities like resolution scalability, quality scalability and region-of-interest coding, while exploiting the entropy in the additional third dimension to improve the rate-distortion performance. In this paper, we give an overview of the markets and application areas targeted by JP3D, the imposed requirements and the algorithm under study.","Title: Digital canvas removal in paintings.\nYear: 2012\n\nAbstract: The periodic structure of the underlying support of paintings on canvas can become quite prominent and disturbing in high resolution digital recordings. In this paper, we construct a new model and method for the digital removal of canvas which is considered as a noise component superimposed on the painting artwork. The high resolution of the images prohibits the efficient application of existing adaptive denoising filters. Hence, a two-step approach is proposed. First a (smoothing) Wiener filter is applied to the complete image. The second step consists of a spatially adaptive extension with low-complexity to obtain a generic digital canvas removal filter.","Title: On the optimality of embedded deadzone scalar-quantizers for wavelet-based L-infinite-constrained image coding\nYear: 2003\n\nAbstract: Summary form only given. Several methods for L∞-distortion constrained compression have been proposed that target a set of fixed reconstruction-error bounds. Recently, a wavelet-based L∞-constrained embedded image-coding technique was proposed that guarantees the required distortion bound while retaining the coding performance and scalability option state-of-the-art wavelet-based L∞-oriented codecs. The optimality of embedded deadzone scalar-quantizers for high-rate L∞-constrained scalable wavelet-based coding of images was analyzed. The optimal quantizers were employed in a coding algorithm that outperforms embedded L∞-oriented wavelet-coders in terms of maximum absolute error. The algorithm was briefly described and coding results were provided.","Title: Special Effects: Efficient and Scalable Encoding of the 3D Metamorphosis Animation with\nYear: 2005\n\nAbstract: Metamorphosis or morphing is a technique to accomplish a gradual shape transformation between a source object and a target\n object. This technique is usually used for animation, i.e., to create some special effects, but it can be also employed as\n a modeling tool where some existing shapes are combined in order to obtain new shapes. Several interactive or automatic morphing\n techniques were designed in the past but always with a specific object representation in mind. In this paper we propose a\n generic framework for 3D morphing, that allows multi-resolution surface extraction and provides a hybrid (surface-volume)\n dynamic representation which is suitable for encoding the static models, for supporting different morphing methods, and for\n encoding the intermediate 3D frames (generated during the morphing process) in a MPEG-4 stream.","Title: In-band motion compensated temporal filtering\nYear: 2004\n\nAbstract: A novel framework for fully scalable video coding that performs open-loop motion-compensated temporal filtering (MCTF) in the wavelet domain (in-band) is presented in this paper. Unlike the conventional spatial-domain MCTF (SDMCTF) schemes, which apply MCTF on the original image data and then encode the residuals using the critically sampled discrete wavelet transform (DWT), the proposed framework applies the in-band MCTF (IBMCTF) after the DWT is performed in the spatial dimensions. To overcome the inefficiency of MCTF in the critically-sampled DWT, a complete-to-overcomplete DWT (CODWT) is performed. Recent theoretical findings on the CODWT are reviewed from the application perspective of fully-scalable IBMCTF, and constraints on the transform calculation that allow for fast and seamless resolution-scalable coding are established. Furthermore, inspired by recent work on advanced prediction techniques, an algorithm for optimized multihypothesis temporal filtering is proposed in this paper. The application of the proposed algorithm in MCTF-based video coding is demonstrated, and similar improvements as for the multihypothesis prediction algorithms employed in closed-loop video coding are experimentally observed. Experimental instantiations of the proposed IBMCTF and SDMCTF coders with multihypothesis prediction produce single embedded bitstreams, from which subsets are extracted to be compared against the current state-of-the-art in video coding.","Title: Wavelet-based fixed and embedded L-infinite-constrained image coding\nYear: 2003\n\nAbstract: A new wavelet-based L-infinity-constrained fixed and embedded coding technique is proposed. The embedded bit stream can be truncated for any desired distortion bound at a corresponding bit rate, so that the target upper bound on the elements of the reconstruction error signal is guaranteed. The original image can also be coded up to a fixed a priori user-defined distortion bound, ranging up to lossless coding. A lifting-based wavelet decorrelating transform is employed on the original image, and exact relations are established between spatial and wavelet domain distortions. The wavelet coefficients are quantized by symmetric uniform quantizers for fixed-distortion coding and by families of embedded uniform deadzone scalar quantizers for embedded coding. The quantized coefficients are finally losslessly encoded using a quadtree-based coding algorithm. Any floating-point lifting-based wavelet transform can be used, and a few of the popular wavelet transforms included in the JPEG2000 verification model are worked out as examples. We compare other L-infinity-constrained coding schemes and show that our proposed coder offers in addition a fully embedded L-infinity-oriented bit stream. We illustrate also that the proposed coder retains the same capabilities as the state-of-the-art embedded wavelet-based co-decs, while providing superior compression results and embeddedness with respect to the L-infinity distortion measure. (C) 2003 SPIE and IST.","Title: Error protection of scalable sources: a comparative analysis of forward error correction and multiple description coding\nYear: 2009\n\nAbstract: In this paper we examine the Distortion-Rate (D-R) performance of two competing approaches for error-resilient transmission of scalable sources over error-prone packet-based memoryless channels. Thereby, we consider scalable Joint Source-Channel Coding (JSCC) based on Forward Error Correction (FEC) and scalable Multiple Description Coding (MDC). Our analysis considers memoryless Gaussian input sources encoded using source coders that can achieve theoretical performance bounds. This input source model is independent of actual implementations of the source coders and therefore allows drawing conclusions on the theoretical D-R performance of both approaches. Our comparative analysis indicates that in matched channel conditions, the use of an optimized FEC-based JSCC approach yields significantly better error resilience compared to the use of scalable MDC Moreover, it is shown that in mismatched conditions a scalable JSCC approach has the potential to outperform a scalable MDC approach.","Title: A Locally Adaptive System for the Fusion of Objective Quality Measures\nYear: 2014\n\nAbstract: Objective measures to automatically predict the perceptual quality of images or videos can reduce the time and cost requirements of end-to-end quality monitoring. For reliable quality predictions, these objective quality measures need to respond consistently with the behavior of the human visual system (HVS). In practice, many important HVS mechanisms are too complex to be modeled directly. Instead, they can be mimicked by machine learning systems, trained on subjective quality assessment databases, and applied on predefined objective quality measures for specific content or distortion classes. On the downside, machine learning systems are often difficult to interpret and may even contradict the input objective quality measures, leading to unreliable quality predictions. To address this problem, we developed an interpretable machine learning system for objective quality assessment, namely the locally adaptive fusion (LAF). This paper describes the LAF system and compares its performance with traditional machine learning. As it turns out, the LAF system is more consistent with the input measures and can better handle heteroscedastic training data.","Title: Single-rate calculation of overcomplete discrete wavelet transforms for scalable coding applications\nYear: 2005\n\nAbstract: A number of emerging resolution-scalable image and video coding algorithms have recently shown very promising performance due to the use of overcomplete wavelet representations. In these applications, the overcomplete discrete wavelet transform (ODWT) is derived starting from the critically-sampled subbands of the DWT (complete representation) of a certain decomposition (resolution) level. This process, which is a complete-to-overcomplete DWT (CODWT), is used for wavelet domain operations that require shift invariance. Specifically, both the encoder and decoder independently construct the overcomplete representation at the best accuracy possible, given the critically-sampled subbands of a certain resolution level. In contrast to the classical approach for performing the CODWT, which is a multi-rate calculation scheme that requires the reconstruction of the input spatial-domain signal, in this paper we propose a new, single-rate calculation scheme, which is formalized for the general case of an arbitrary decomposition (resolution) level. Based on derived symmetry properties, a simple implementation structure of the proposed approach provides interesting tradeoffs for the required multiplication budget in comparison to the conventional approach. This leads to a complexity-scalable solution that fits the versatile requirements of scalable coding environments. The use of the proposed single-rate calculation scheme of the CODWT is demonstrated in several image and video coding systems.","Title: Comparative Study of Wavelet Based Lattice QIM Techniques and Robustness against AWGN and JPEG Attacks\nYear: 2009\n\nAbstract: We study watermarking techniques based on Quantization Index Modulation for which sets of lattice quantizers Q m are used (LQIM). A recipe for constructing such quantizers Q m is proposed, where the size of these sets is variable, so that the payload is easily adaptable. We make a comparative study of 8 dimensional lattices with good quantizer properties, where the embedding is done in the wavelet domain. Along the way, the gap between the theoretical ideas behind QIM and practical systems using lattices is closed by extending techniques, such as dithered quantizers and distortion compensation, from the scalar case to LQIM.","Title: Modeling the Correlation Noise in Spatial Domain Distributed Video Coding\nYear: 2009\n\nAbstract: Conventional models in distributed video coding (DVC) consider the correlation noise to be distributed independently from the realization of the side-information. This paper introduces a novel model, of which the standard deviation depends spatially on the realization of the side-information. The performance penalty in video coding caused by side-information-independency assumptions is theoretical quantified and experimentally confirmed. Furthermore, inspired by the spatial side-information-dependency of the proposed model, a novel approach for estimating the correlation channel from the partial knowledge of it is introduced. The proposed technique is incorporated into a spatial-domain unidirectional DVC system, providing state-of-the-art performance.","Title: A low-complexity UEP methodology demonstrated on a turbo-encoded wavelet image satellite downlink\nYear: 2008\n\nAbstract: Realizing high-quality digital image transmission via a satellite link, while optimizing resource distribution and minimizing battery consumption, is a challenging task. This paper describes a methodology to optimize a turbo-encoded wavelet-based satellite down-link progressive image transmission system with unequal error protection (UEP) techniques. To achieve that goal, we instantiate a generic UEP methodology onto the system, and demonstrate that the proposed solution has little impact on the average performance, while greatly reducing the run-time complexity. Based on a simple design-time distortion model and a low-complexity run-time algorithm, the provided solution can dynamically tune the system's configuration to any bitrate constraint or channel condition. The resulting system outperforms in terms of peak signal-to-noise ratio (PSNR), a state-of-the-art, fine-tuned equal error protection (EEP) solution by as much as 2 dB.","Title: Applying Open-Loop Coding in Predictive Coding Systems\nYear: 2008\n\nAbstract: This paper investigates the application of open-loop coding principles in predictive coding systems. In order to cope with the drift, which is inherent in open-loop predictive coding, a novel rate-distortion (R-D) model is proposed, capturing the propagation of quantization errors in such systems. Additionally, a novel intra-frame video codec employing the transform and spatial prediction modes from H.264 is proposed. The results obtained with the proposed codec show that allocating rate based on the proposed R-D model provides gains of up to 1.9 dB compared to a straightforward rate allocation not accounting for drift. Furthermore, the proposed open-loop predictive codec provides gains of up to 2.3 dB compared to an equivalent closed-loop intra-frame video codec using the transform, prediction modes and rate-allocation from H.264. One concludes that the considered open-loop predictive coding paradigm retains the advantages of open-loop coding, and offers the possibility of further improving the compression performance in predictive coding systems.","Title: Feature point based image watermarking with insertions, deletions, and substitution codes\nYear: 2010\n\nAbstract: In this paper we concentrate on robust image watermarking (i.e. capable of resisting common signal processing operations and intentional attacks to destroy the watermark) based on image features. Kutter et al.(7) motivated that well chosen image features survive admissible image distortions and hence can benefit the watermarking process. These image features are used as location references for the region in which the watermark is embedded. To realize the latter, we make use of previous work(16) where a ring-shaped region, centered around an image feature is determined for watermark embedding. We propose to choose a specific sequence of image features according to strict criteria so that the image features have large distance to other chosen image features so that the ring shaped embedding regions do not overlap. Nevertheless, such a setup remains prone to insertion, deletion and substitution errors. Therefore we applied a two-step coding scheme similar to the one employed by Coumou and Sharma(4) for speech watermarking. Our contribution here lies in extending Coumou and Sharma's one dimensional scheme to the two dimensional setup that is associated with our watermarking technique. The two-step coding scheme concatenates an outer Reed-Solomon error-correction code with an inner, blind, synchronization mechanism.","Title: Encoder-driven rate control and mode decision for distributed video coding.\nYear: 2013\n\nAbstract: To provide low-complexity encoding for video in unidirectional or offline compression scenarios, this paper proposes an efficient feedback-channel-free distributed video coding architecture featuring a novel encoder-driven rate control scheme in tandem with a designated mode selection process. To this end, the encoder features a novel low-complexity motion estimation technique to approximate the side-information (SI) available at the decoder. Then, a SI-dependent correlation channel estimation between the approximated SI and the original frames is used to derive the theoretically required rate for successful Slepian-Wolf (SW) decoding. Based on the evaluation of the expected trade-off between the estimated required coding rate and the estimated distortion outcome, a novel encoder-side mode decision module assigns a different coding mode to distinct portions of the coded frames. In this context, skip, intra and SW coding modes are supported. To reduce the effect of underestimation, the final SW rate is adjusted upwards using a novel rate formula. Additionally, a successive SI refinement technique is exploited at the decoder to decrease the number of SW decoding failures. Experimental results illustrate the benefit of the different coding options and show similar or superior compression performance with respect to the feedback-based DISCOVER benchmark system. Finally, the low-complexity encoding characteristics of the proposed system are confirmed, as well as the beneficial impact of the proposed scheme on the decoding complexity.","Title: On the side-information dependency of the temporal correlation in Wyner-Ziv video coding\nYear: 2009\n\nAbstract: Current models in Wyner-Ziv video coding consider the temporal correlation noise to be side-information independent (SII). This paper goes beyond this assumption and proposes a novel model, of which the parameters are side-information dependent (SID). The proposed model is experimentally validated showing remarkable accuracy improvement over the conventional SII model. Moreover, a novel SID technique for the accurate estimation of the correlation channel in video is introduced. The proposed technique enables the design of a novel pixel-domain Wyner-Ziv video coding system operating without a feedback channel. Preliminary experimental results show that the proposed codec achieves superior performance compared to the state-of-the-art in pixel-domain Wyner-Ziv coding.","Title: Theoretical distortion estimation in L-infinite wavelet-based coding of semi-regular meshes.\nYear: 2012\n\nAbstract: Subdivision-based wavelet coding techniques yield state-of-the-art performance in scalable compression of semi-regular meshes. However, all these codecs make use of the L-2 distortion metric, which gives only a good approximation of the global error produced by lossy coding of the wavelet coefficients. The L-infinite metric has been proven to be a suitable metric for applications where controlling the local, maximum error on each vertex is of critical importance. In this context, an upper bound formulation for the L-infinite distortion for a wavelet-based coding scheme operating on semi-regular meshes is derived. In addition, we propose a rate-distortion optimization algorithm that minimizes the rate for any target L-infinite distortion. It is shown that our L-infinite coding system outperforms the state-of-the-art and that an L-2 driven coding approach for semi-regular meshes loses ground to its L-infinite driven version when the goal is to have a tight control on the local reconstruction error.","Title: Fully scalable intraband coding of wavelet-decomposed 3D meshes\nYear: 2009\n\nAbstract: In this paper, a scalable intraband wavelet coding technique for semi-regular topology surface meshes is proposed. The choice of an intraband codec design is justified through an information theoretic analysis of the statistical dependencies between the wavelet coefficients resulting from a wavelet-based decomposition of the input 3D mesh. The bitstream produced by the proposed codec is both resolution and quality scalable. This lies in contrast with well-known zero-tree based interband coding techniques that only support quality scalability. The experimental results show that, on average, the proposed codec outperforms the state-of-the-art zero-tree based interband codec in rate-distortion sense when using the L2 norm as target distortion metric.","Title: Performing deblocking in video coding based on spatial-domain motion-compensated temporal filtering\nYear: 2006\n\nAbstract: Employing block-based motion models in scalable video coding based on spatial-domain motion-compensated temporal filtering (SDMCTF) introduces artificial block-boundary discontinuities that adversely affect the video compression performance, particularly at low bit-rates. This paper focuses on the problem of deblocking in the context of SDMCTF-based video coding. One possible solution to this problem is the use of overlapped-block motion compensation (OBMC). An alternative solution is applying an adaptive deblocking filter, similar to the one used in H.264. With this respect, a novel adaptive deblocking filter, tailored to SDMCTF video coding is proposed. In terms of visual-quality, both approaches yield similar performance. However, adaptive deblocking is less complex than OBMC, as it requires up to 34% less processing time. Experimental results show that the two techniques significantly improve the subjective and objective quality of the decoded sequences, confirming the expected benefits brought by deblocking in SDMCTF-based video coding.","Title: A Low-Complexity Methodology For Unequal Error Protection Of Scalable Images\nYear: 2006\n\nAbstract: Unequal Error Protection (UEP) is the most practical way to achieve joint source-channel coding while keeping a clean separation between the source and channel coders. In this paper, a generic UEP allocation methodology is proposed. The methodology is based on a run-time algorithm, exploiting design-time models, and is applicable to bit-level transmission scenarios as well as packet-based transmission environments. It works with source-independent modeling and a linear complexity with respect to the number of substreams in the source coded stream and the number of protection levels available, which makes it an attractive solution for practical scenarios. We obtain a substantial complexity reduction compared to a classical Lagrangian optimization while keeping optimality in the allocation of the protection levels. Given a typical source and channel granularity, we obtain a factor 6 complexity reduction.","Title: Robust Motion Vector Coding And Error Concealment In Mctf-Based Video Coding\nYear: 2005\n\nAbstract: Error resilience is of paramount importance in video transmission over variable-bandwidth error-prone channels, such as wireless channels. In this paper, we investigate the influence of corrupted motion vectors in video coding based on motion compensated temporal filtering, and develop various error resilience and concealment mechanisms for this class of codecs. The experimental results show that our proposed motion vector coding technique significantly increases the robustness against transmission errors at the cost of less than 3% in terms of rate. It is also shown that our proposed spatial error-concealment mechanism leads to performance gains of up to 6 dB in comparison to a classical slicing-based approach employing no error concealment.","Title: Scalable Video Coding Based On Motion-Compensated Temporal Filtering: Complexity And Functionality Analysis\nYear: 2004\n\nAbstract: Video coding techniques yielding state-of-the-art compression performance require large amount of computational resources, hence practical implementations, which target a broad market, often tend to trade-off coding efficiency and flexibility for reduced complexity. Scalable video coding instead, not only provides seamless adaptation to bit-rate variation, but also allows the end user to trim down the resources lie needs to perform real-time decoding by limiting the process to a subset of the original content. Hence, by choosing the quality, frame-rate and/or resolution of the reconstructed sequence, each decoder can meet its hardware limitations without affecting, the encoding process of the media provider. This paper proposes a preliminary analysis of the memory-access behavior of a fully scalable video decoder and investigates the capability of selecting the operational settings in order to adapt to the available hardware resources on the target device.","Title: Error-Resilient Video Coding Using Motion Compensated Temporal Filtering And Embedded Multiple Description Scalar Quantizers\nYear: 2005\n\nAbstract: Real time delivery of video over best-effort networks requires compression systems that dynamically adapt the rate to the available channel capacity and exhibit robustness to data loss as retransmission is often impractical. Error-resilience, however, significantly lowers the coding performance when rigid design is performed based on a worst-case scenario. This paper presents an original scalable video coding scheme that couples the compression efficiency of the open-loop architecture with the robustness of multiple description source coding. The use of embedded multiple description quantization and a novel channel-aware rate-allocation allow for shaping on-the-fly the output bit-rate and the degree of resilience. As a result, robustness to data losses is traded for better visual quality when transmission occurs over reliable channels, while error-resilience is introduced when noisy links are involved. The advantage of our proposal is demonstrated in the context of packet-lossy networks.","Title: Analysis of the statistical dependencies in the curvelet domain and applications in image compression\nYear: 2007\n\nAbstract: This paper reports an information-theoretic analysis of the dependencies that exist between curvelet coefficients. We show that strong dependencies exist in local intra-band micro-neighborhoods, and that the shape of these neighborhoods is highly anisotropic. With this respect, it is found that the two immediately adjacent neighbors that lie in a direction orthogonal to the orientation of the subband convey the most information about the coefficient. Moreover, taking into account a larger local neighborhood set than this brings only mild gains with respect to intra-band mutual information estimations. Furthermore, we point out that linear predictors do not represent sufficient statistics, if applied to the entire intra-band neighborhood of a coefficient. We conclude that intra-band dependencies are clearly the strongest, followed by their inter-orientation and inter-scale counterparts; in this respect, the more complex intra-band/inter-scale or intra-band/inter-orientation models bring only mild improvements over intra-band models. Finally, we exploit the coefficient dependencies in a curvelet-based image coding application and show that the scheme is comparable and in some cases even outperforms JPEG2000.","Title: Maximum likelihood motion compensation for distributed video coding\nYear: 2012\n\nAbstract: Aspiring to provide robust low-complexity encoding for video, this work presents a hash-based transform domain distributed video codec, featuring a novel maximum likelihood motion compensation technique to generate high quality side information at the decoder. A simple hash is employed to perform overlapped block motion estimation at the decoder, which produces a set of temporal predictors on a pixel basis. For every pixel position, maximum likelihood motion compensation, based on an online estimate of the conditional dependencies between the temporal predictors and the original frame, combines the cluster of temporal predictors into a single value to serve as decoder-generated side information. Efficient low-density parity-check accumulate channel codes refine the side information in the transform domain. Experimental results demonstrate that the proposed system advances over our previous hash-based distributed video coding architectures, delivering state-of-the-art distributed coding performance, in particular for sequences organized in large groups of pictures or containing highly irregular motion. Notwithstanding the presence of a hash, the presented distributed video codec successfully maintains low encoder complexity.","Title: Spatial-domain unidirectional DVC with side-information dependent correlation channel estimation\nYear: 2009\n\nAbstract: Present distributed video coding (DVC) approaches assume the correlation channel to be side-information independent (SII). We have recently demonstrated that in reality the correlation channel in video is side-information dependent (SID) and that DVC systems which make SII assumptions suffer a significant performance penalty, depending entirely on the correlation statistics of the video data. In this paper, we present a novel spatial domain unidirectional DVC codec which is based on a novel SID technique enabling advanced correlation channel estimation at the decoder starting from a partial knowledge of it. Experimental results show that the proposed codec achieves superior performance compared to the state-of-the-art in Wyner-Ziv video coding while operating without a feedback channel.","Title: Robust Image Content Authentication with Tamper Location\nYear: 2012\n\nAbstract: We propose a novel image authentication system by combining perceptual hashing and robust watermarking. An image is divided into blocks. Each block is represented by a compact hash value. The hash value is embedded in the block. The authenticity of the image can be verified by re-computing hash values and comparing them with the ones extracted from the image. The system can tolerate a wide range of incidental distortion, and locate tampered areas as small as 1/64 of an image. In order to have minimal interference, we design both the hash and the watermark algorithms in the wavelet domain. The hash is formed by the sign bits of wavelet coefficients. The lattice-based QIM watermarking algorithm ensures a high payload while maintaining the image quality. Extensive experiments confirm the good performance of the proposal, and show that our proposal significantly outperforms a state-of-the-art algorithm.","Title: Combined wavelet-domain and motion-compensated video denoising based on video codec motion estimation methods\nYear: 2009\n\nAbstract: Integrating video coding and denoising is a novel processing paradigm, bringing mutual benefits to both video processing tools. In this paper, we propose a novel video denoising approach of which the main idea is reusing motion estimation resources from the video coding module for video denoising. In most cases, the motion fields produced by real-time video codecs cannot be directly employed in video denoising, since they, as opposed to noise filters, tolerate errors in the motion field. In order to solve this problem, we propose a novel motion-field filtering step that refines the accuracy of the motion estimates to a degree that is required for denoising. Additionally, a novel temporal filter is proposed that is robust against errors in the estimated motion field. Numerical results demonstrate that the proposed denoising scheme is of low-complexity and compares favorably to the state-of-the-art video denoising methods.","Title: Progressively refined wyner-ziv video coding for visual sensors\nYear: 2014\n\nAbstract: Wyner-Ziv video coding constitutes an alluring paradigm for visual sensor networks, offering efficient video compression with low complexity encoding characteristics. This work presents a novel hash-driven Wyner-Ziv video coding architecture for visual sensors, implementing the principles of successively refined Wyner-Ziv coding. To this end, so-called side-information refinement levels are constructed for a number of grouped frequency bands of the discrete cosine transform. The proposed codec creates side-information by means of an original overlapped block motion estimation and pixel-based multihypothesis prediction technique, specifically built around the pursued refinement strategy. The quality of the side-information generated at every refinement level is successively improved, leading to gradually enhanced Wyner-Ziv coding performance. Additionally, this work explores several temporal prediction structures, including a new hierarchical unidirectional prediction structure, providing both temporal scalability and low delay coding. Experimental results include a thorough evaluation of our novel Wyner-Ziv codec, assessing the impact of the proposed successive refinement scheme and the supported temporal prediction structures for a wide range of hash configurations and group of pictures sizes. The results report significant compression gains with respect to benchmark systems in Wyner-Ziv video coding (e.g., up to 42.03% over DISCOVER) as well as versus alternative state-of-the-art schemes refining the side-information.","Title: Wavelet based volumetric medical image compression.\nYear: 2015\n\nAbstract: The amount of image data generated each day in health care is ever increasing, especially in combination with the improved scanning resolutions and the importance of volumetric image data sets. Handling these images raises the requirement for efficient compression, archival and transmission techniques. Currently, JPEG 2000׳s core coding system, defined in Part 1, is the default choice for medical images as it is the DICOM-supported compression technique offering the best available performance for this type of data. Yet, JPEG 2000 provides many options that allow for further improving compression performance for which DICOM offers no guidelines. Moreover, over the last years, various studies seem to indicate that performance improvements in wavelet-based image coding are possible when employing directional transforms. In this paper, we thoroughly investigate techniques allowing for improving the performance of JPEG 2000 for volumetric medical image compression. For this purpose, we make use of a newly developed generic codec framework that supports JPEG 2000 with its volumetric extension (JP3D), various directional wavelet transforms as well as a generic intra-band prediction mode. A thorough objective investigation of the performance-complexity trade-offs offered by these techniques on medical data is carried out. Moreover, we provide a comparison of the presented techniques to H.265/MPEG-H HEVC, which is currently the most state-of-the-art video codec available. Additionally, we present results of a first time study on the subjective visual performance when using the aforementioned techniques. This enables us to provide a set of guidelines and settings on how to optimally compress medical volumetric images at an acceptable complexity level.","Title: Scalable multiple-description image coding based on embedded quantization\nYear: 2007\n\nAbstract: Scalable multiple-description (MD) coding allows for fine-grain rate adaptation as well as robust coding of the input source. In this paper, we present a new approach for scalable MD coding of images, which couples the multiresolution nature of the wavelet transform with the robustness and scalability features provided by embedded multiple-description scalar quantization (EMDSQ). Two coding systems are proposed that rely on quadtree coding to compress the side descriptions produced by EMDSQ. The proposed systems are capable of dynamically adapting the bitrate to the available bandwidth while providing robustness to data losses. Experiments performed under different simulated network conditions demonstrate the effectiveness of the proposed scalable MD approach for image streaming over error-prone channels.","Title: Proceedings of the second \"international Traveling Workshop on Interactions between Sparse models and Technology\" (iTWIST'14).\nYear: 2014\n\nAbstract: The implicit objective of the biennial \"international - Traveling Workshop on Interactions between Sparse models and Technology\" (iTWIST) is to foster collaboration between international scientific teams by disseminating ideas through both specific oral/poster presentations and free discussions. For its second edition, the iTWIST workshop took place in the medieval and picturesque town of Namur in Belgium, from Wednesday August 27th till Friday August 29th, 2014. The workshop was conveniently located in \"The Arsenal\" building within walking distance of both hotels and town center. iTWIST'14 has gathered about 70 international participants and has featured 9 invited talks, 10 oral presentations, and 14 posters on the following themes, all related to the theory, application and generalization of the \"sparsity paradigm\": Sparsity-driven data sensing and processing; Union of low dimensional subspaces; Beyond linear and convex inverse problem; Matrix/manifold/graph sensing/processing; Blind inverse problems and dictionary learning; Sparsity and computational neuroscience; Information theory, geometry and randomness; Complexity/accuracy tradeoffs in numerical methods; Sparsity? What's next?; Sparse machine learning and inference.","Title: Heterogeneous Acceleration of Volumetric JPEG 2000\nYear: 2015\n\nAbstract: We present the implementation of a volumetric JPEG 2000 codec as a real-world use case of software acceleration with GPUs and multi-core CPUs. We present a generic methodology to accelerate existing code written in C with OpenCL. Furthermore, we account for the volumetric nature of the processed data and formulate associated optimization guidelines. The resulting software can exploit different accelerator types - GPUs and multi-core CPUs - and delivers a decent speedup on a variety of hardware platforms for a relatively small effort.","Title: Lossy-to-lossless screen content coding using an HEVC base-layer\nYear: 2013\n\nAbstract: This paper presents a novel High Efficiency Video Coding (HEVC) based screen content coding scheme providing lossy-to-lossless compression support. The paper follows a two-layer design approach which allows for backward compatibility with the conventional HEVC standard for the base-layer and yields lossless compression when decoding the enhancement-layer. The system employs Golomb based binarization and context-based arithmetic coding to efficiently encode the enhancement-layer. Performance evaluations using 4:4:4 screen content sequences are carried out. The results show that the proposed system substantially outperforms lossless intra- HEVC and single-layer JPEG-LS.","Title: Efficient intra-frame video coding for low resolution wireless visual sensors\nYear: 2013\n\nAbstract: This paper proposes a low-complexity intra-frame video coding system for very low resolution sequences. The application focuses on power constrained wireless visual sensors, recording extremely low-resolution video, namely, 30×30 pixels. In the proposed framework, high-performance video coding techniques such as block-based transforms, intra prediction and entropy coding are employed. The proposed system features several different configurations, all of which are evaluated both in terms of coding efficiency and computational complexity. Based on the experimental results, the best configuration is identified. Experimental results show that the proposed configuration achieves better coding performance than the H.264/AVC Intra frame codec, at only 23.7% of its encoding complexity.","Title: Image security tools for JPEG standards\nYear: 2014\n\nAbstract: Image privacy and security are receiving increasingly more attention these days though their impact is still underestimated. Every day billions of digital pictures are generated and distributed via social media, news sites and photo sharing applications. These images tend to carry quite some information that can be regarded to be private for the photographer and/or the pictured individuals. Furthermore, this information is not solely constrained to the visible information carried by the picture. For instance, the sensor noise that is embedded in the picture provides information related to the camera used to take the photo. Besides the picture essence also metadata is signaled in the image file containers, providing geospatial information, camera settings, photographer related information, etc. Hence, it is evident to understand that to protect the privacy of the persons that can be associated with the picture, the necessary security precautions need to be taken. Since many pictures are represented in JPEG image formats, it is evident that the JPEG committee is also investigating how it can integrate better support for privacy and security requirements in its suite of standards. In this presentation, an overview will be given of standardized solutions, where we will particularly focus on JPEG 2000 JPSEC framework, but will also discuss how the committee plans to further improve the support for these requirements in the context of the legacy JPEG standard.","Title: Performance optimizations for PatchMatch-based pixel-level multiview inpainting\nYear: 2013\n\nAbstract: As 3D content is becoming ubiquitous in today's media landscape, there is a rising interest for 3D displays that do not demand wearing special headgear in order to experience the 3D effect. Autostereoscopic displays realize this by providing multiple different views of the same scene. It is however unfeasible to record, store or transmit the amount of data that such displays require. Therefore there is a strong need for real-time solutions that can generate multiple extra viewpoints from a limited set of originally recorded views. The main difficulty in current solutions is that the synthesized views contain disocclusion holes where the pixel values are unknown. In order to seamlessly fill-in these holes, inpainting techniques are being used. In this work we consider a depth-based pixel-level inpainting system for multiview video. The employed technique operates in a multi-scale fashion, fills in the disocclusion holes on a pixel-per-pixel basis and computes approximate Nearest Neighbor Fields (NNF) to identify pixel correspondences. To this end, we employ a multi-scale variation on the well-known PatchMatch algorithm followed by a refinement step to escape from local minima in the matching-cost function. In this paper we analyze the performance of different cost functions and search methods within our existing inpainting framework.","Title: Interactive demonstrations of the locally adaptive fusion for combining objective quality measures\nYear: 2014\n\nAbstract: To automate quality monitoring of multimedia applications, objective quality measures for images and video content need to be designed. Objective quality measures that model the Human Visual System (HVS) have a disappointing performance, because the HVS is not sufficiently understood. Integrating machine learning (ML) techniques may increase the performance. Unfortunately, traditional ML is difficult to interpret. To this end, we developed the Locally Adaptive Fusion (LAF), for more flexible and reliable quality predictions. This manuscript proposes six interactive programs and a website that demonstrate the effectiveness of LAF, which complement the technical focus of the corresponding journal paper.","Title: Embedded multiple description scalar quantizers for progressive image transmission\nYear: 2003\n\nAbstract: Robust progressive image transmission over unreliable channels with variable bandwidth requires multiple description coding (MDC) systems that produce highly error-resilient embedded bit-streams. The proposed embedded multiple description scalar quantizers (EMDSQ) meet the desired features consisting of a high redundancy level, fine grain rate adaptation and progressive transmission of each description. Experimental results show that EMDSQ yield better rate-distortion performance in comparison to the multiple description uniform scalar quantizers (MDUSQ) previously proposed in the literature. Moreover, the generalized form of EMDSQ targeting an arbitrary number of channels is proposed, which offers the possibility of designing realistic coders for practical multi-channel communication systems.","Title: Additive Distortion Modeling for Unequal Error Protection of Scalable Multimedia Content\nYear: 2005\n\nAbstract: Due to the improvement of compression technology, limited bandwidth channels can convey more data than before. However, since compressed streams are very sensitive to transmission errors, a joint optimization of the source-channel coding is required. Unequal Error Protection (UEP) is the most pragmatic way to do it, and exploits the embedded nature of bitstreams produced by scalable coders.However, depending on the number of substreams and protection levels, optimization of UEP can be extremely complex. A large simplification comes from considering the contribution to the global distortion of each substream as independent from the others. This paper explores this distortion additivity assumption for JPEG2000, and shows that though the assumption does not fully hold, no visible consequence on UEP performance is introduced, while reducing significantly the complexity of the optimization process.","Title: Intra-frame video coding using an open-loop predictive coding approach\nYear: 2008\n\nAbstract: A novel coding approach, applying open-loop coding principles in predictive coding systems is proposed in this paper. The proposed approach is instantiated with an intra-frame video codec employing the transform and spatial prediction modes from H.264. Additionally, a novel rate-distortion model for open-loop predictive coding is proposed and experimentally validated. Optimally allocating rate based on the proposed model provides significant gains in comparison to a straightforward rate allocation not accounting for drift. Furthermore, the proposed open-loop predictive codec provides gains of up to 2.3 dB in comparison to an equivalent closed-loop intra-frame video codec employing the transform, prediction modes and rate-allocation from H.264. This indicates that, with appropriate drift compensation, open-loop predictive coding offers the possibility for further improving the compression performance in predictive coding systems.","Title: Statistical L-infinite distortion estimation in scalable coding of meshes\nYear: 2008\n\nAbstract: This paper investigates the novel concept of local error control in arbitrary mesh encoding, and proposes a new L-infinite mesh coding approach implementing this concept. In contrast to traditional mesh coding systems that use the mean-square error as distortion measure, the proposed approach employs the L-infinite distortion as target distortion metric. In this context, a novel wavelet-based L-infinite-constrained coding approach for meshes is proposed, which ensures that the maximum local error between the original and decoded meshes is lower than a given upper-bound. Additionally, the proposed system achieves scalability in L-infinite sense, that is, the L-infinite distortion upper-bound can be accurately estimated when decoding any layer from the input stream. Moreover, a distortion estimation approach is proposed, expressing the L-infinite distortion in the spatial domain as a statistical estimate of quantization errors produced in the wavelet domain. An instantiation of the proposed L-infinite coding approach is demonstrated for MESHGRID, which is a scalable 3D object coding system, part of MPEG-4 AFX. The proposed L-infinite coding approach guarantees that the maximum error is upper-bounded, it enables a fast real-time implementation of the rate-allocation, and it preserves all the scalability features and animation capabilities of the employed scalable mesh codec.","Title: Segmentation-based intra coding of depth maps using texture information\nYear: 2013\n\nAbstract: Emerging video technologies, such as 3DTV, increase the demand for high efficiency coding tools. To allow the transmission of a multiview video over the employed transmission channels, a new data format including texture and the corresponding depth information has been proposed. In this paper we present a novel segmentation-based intra codec for depth maps, utilizing the correlation between depth and color data. Our system performs a low-complexity color segmentation on the texture image and encodes the depth values within the segments. The same segmentation is calculated at the decoder, ensuring the closed-loop property of the system. Experimental results show that the proposed system is a viable alternative for 3D-HEVC in low-complexity applications.","Title: Real-time texture sampling and reconstruction with wavelet filters\nYear: 2013\n\nAbstract: Currently, the use of the 2D wavelet transform in texture compression for real-time texture mapping on the GPU is limited. The main cause of this is the lack of real-time texture filtering implementations which do not require specialized hardware. This work proposes a novel system to perform 2D wavelet reconstruction and bilinear texture filtering using a high performance GPU shader. The system is able to generate a performant GLSL shader for arbitrary wavelet filter configurations. This goes beyond earlier works in the literature proposing Haar wavelet and Discrete Cosine Transform (DCT) implementations on the GPU. We analyse the shader performance and run-time complexity for several wavelet filters. The experimental results show that filters longer than Haar are deployable on the GPU while maintaining accurate texture filtering and real-time performance.","Title: Forensic data hiding optimized for JPEG 2000\nYear: 2011\n\nAbstract: This paper presents a novel image adaptive data hiding system using properties of the discrete wavelet transform and which is ready to use in combination with JPEG 2000. Image adaptive watermarking schemes determine the embedding samples and strength from the image statistics. We propose to use the energy of wavelet coefficients at high frequencies to measure the amount of distortion that can be tolerated by a lower frequency coefficient. The watermark decoder in image adaptive data hiding needs to estimate the same parameters used for encoding from a modified source and hence is vulnerable to desynchronization. We present a novel way to resolve these synchronization issues by employing specialized insertion, deletion and substitution codes. Given the low complexity and reduced perceptual impact of the embedding technique, it is suitable for inserting camera and/or projector information to facilitate image forensics.","Title: Optimized directional lifting with reduced complexity\nYear: 2008\n\nAbstract: ABSTRACT The direction-adaptive discrete wavelet,transform,(DADWT) proves to be a very competitive alternative in scalable wavelet-based compression of images, yielding impressive compression perfor- mance gains in comparison to the classical DWT. A major limitation of DADWT though stems from its complexity, requiring an exhaus- tive search for the optimum,prediction direction to be employed in the directional lifting process. This paper proposes a novel al- gorithm to lower the complexity of the DADWT by predicting the optimal prediction direction using a gradient-based technique. The algorithm is developed,based on a mathematical,model of the pre- diction errors generated via directional lifting of an input wedge image. The proposed approach avoids a time-consuming exhaustive search and yet the prediction step remains very simple and fast. It is shown that the proposed algorithm brings a complexity-reduction factor of 11=4 for almost no penalty in the prediction accuracy.","Title: The evolution of multiple description coding based on scalar quantization\nYear: 2007\n\nAbstract: Multiple Description Coding (MDC) allows for error resilient coding of data. This tutorial paper discusses one of the most popular MDC technologies, namely MDC based on scalar quantization, and its application in scalable coding of multimedia. This concept is illustrated by a fully scalable video coder, based on motion compensated temporal filtering, which incorporates the MDC principle for the encoding of both texture and motion vector data. For such system, we derive the principles of channel-aware MDC rate-allocation, which enables adjusting both the bit-rate and the resilience to errors of the output stream. Such adaptation is performed on-the-fly depending on the channel condition. Experimental results prove the advantages of MDC and the effectiveness of MDC system when transmitting video streams over unreliable networks.","Title: A novel MPI reduction algorithm resilient to imbalances in process arrival times.\nYear: 2016\n\nAbstract: Reduction algorithms are optimized only under the assumption that all processes commence the reduction simultaneously. Research on process arrival times has shown that this is rarely the case. Thus, all benchmarking methodologies that take into account only balanced arrival times might not portray a true picture of real-world algorithm performance. In this paper, we select a subset of four reduction algorithms frequently used by library implementations and evaluate their performance for both balanced and imbalanced process arrival times. The main contribution of this paper is a novel imbalance robust algorithm that uses pre-knowledge of process arrival times to construct reduction schedules. The performance of selected algorithms was empirically evaluated on a 128 node subset of the Partnership for Advanced Computing in Europe CURIE supercomputer. The reported results show that the new imbalance robust algorithm universally outperforms all the selected algorithms, whenever the reduction schedule is precomputed. We find that when the cost of schedule construction is included in the total runtime, the new algorithm outperforms the selected algorithms for problem sizes greater than 1 MiB.","Title: Globally optimized multiview video color correction using dense spatio-temporal matching\nYear: 2015\n\nAbstract: Multiview video is becoming increasingly popular as the format for 3D video systems that use autostereoscopic displays or freeviewpoint navigation capabilities. However, the algorithms that drive these applications are not yet mature and can suffer from subtle irregularities such as color imbalances inbetween different cameras. Regarding the problem of color correction, state-of-the-art methods directly apply some form of histogram matching in blocks of pixels between an input frame and a target frame containing the desired color distribution. These methods, however, typically suffer from artifacts in the gradient domain, as they do not take into account local texture information. This paper presents a novel method to correct color differences in multiview video sequences that uses a dense matching-based global optimization framework. The proposed energy function ensures preservation of local structures by regulating deviations from the original image gradients.","Title: Continuous ultrasound speckle tracking with Gaussian mixtures\nYear: 2015\n\nAbstract: Speckle tracking echocardiography (STE) is now widely used for measuring strain, deformations, and motion in cardiology. STE involves three successive steps: acquisition of individual frames, speckle detection, and image registration using speckles as landmarks. This work proposes to avoid explicit detection and registration by representing dynamic ultrasound images as sparse collections of moving Gaussian elements in the continuous joint space-time space. Individual speckles or local clusters of speckles are approximated by a single multivariate Gaussian kernel with associated linear trajectory over a short time span. A hierarchical tree-structured model is fitted to sampled input data such that predicted image estimates can be retrieved by regression after reconstruction, allowing a (bias-variance) trade-off between model complexity and image resolution. The inverse image reconstruction problem is solved with an online Bayesian statistical estimation algorithm. Experiments on clinical data could estimate subtle sub-pixel accurate motion that is difficult to capture with frame-to-frame elastic image registration techniques.","Title: Digital painting analysis, at the cross section of engineering, mathematics and culture\nYear: 2011\n\nAbstract: Museums started digitizing their collections, mainly for preservation, documentation and dissemination purposes, but one can go further than archival: digital acquisition enables digital analysis of art and in particular paintings. This proves to be a useful, non invasive tool, for many applications such as restoration, conservation, art history, material and structure characterization, authentication, dating and even style analysis. Paintings are complex structures. Analysis of the support and all pictorial layers requires multimodal imaging, high resolution and standardized data acquisition. We situate our own research on canvas characterization, craquelure detection, multispectral data analysis, object extraction and image enhancement within the state of the art.","Title: Error-resilient transmission of H.264 SVC streams over DVB-T/H and WiMAX channels with multiple description coding techniques\nYear: 2007\n\nAbstract: The European STREP Project SUIT involves the development of a multiple description scalable video codec for the transmission of video content over DVB-T/H and WiMAX channels. In this paper, we elaborate on the architectural design choices for such a multiple description transmission system, which utilizes H.264 Scalable Video Coding (SVC). We follow a phased approach focusing on the following unbalanced multiple description coding approaches: fully redundant unbalanced multiple description coding, multiple description coding based on redundant slices and multiple description coding by means of embedded multiple description quantisation.","Title: JPEG Pleno: Toward an Efficient Representation of Visual Reality.\nYear: 2016\n\nAbstract: In discussing the rationale behind the vision for JPEG Pleno and how the new standardization initiative aims to reinvent the future of imaging, the authors review plenoptic representation and its underlying practical implications and challenges in implementing real-world applications with an enhanced quality of experience.","Title: Efficient Depth-aware Image Deformation Adaptation for Curved Screen Displays.\nYear: 2017\n\nAbstract: The curved screen has attracted considerable attentions in recent years, since it enables to enlarge the view angle and to enhance the immersive perception for users. However, existing curved surface projections are frequently prone to geometric distortion or loss of content. This paper presents a content-aware and depth-aware image adaptation solution for curved displays. An efficient optimization approach of image deformation is proposed to preserve local scene content and to minimize scene geometry distortion. To follow the original 3D perception of objects in different depth layers, the depth information is re-mapped for individual content scaling. Objective evaluation results reveal that our approach can effectively preserve foreground objects. We also perform a subjective evaluation of the proposed solution, and compare it to two alternative mapping methods, which are tested on different curvatures on both a traditional screen and an ad-hoc curvature-controllable curved display. Experimental results demonstrate that our approach outperforms other existing mapping methods for immersive display of rectangle images on curved screens.","Title: Heterogeneous acceleration of volumetric JPEG 2000 using OpenCL\nYear: 2017\n\nAbstract: AbstractThis paper discusses an OpenCL version of a volumetric JPEG 2000 codec that runs on GPUs, multi-core processors or a combination of both. Since the performance critical part consists of a fine-grained discrete wavelet transform and coarse-grained algorithm Tier-1, the best performance is obtained with a hybrid execution in which the discrete wavelet transform is executed on a GPU and Tier-1 on a multi-core. Using an Intel i7 multi-core in combination with a modest NVIDIA Quadro K620 GPU yields speedups greater than 10 compared with the original sequential code. The performance bottlenecks that arise on GPUs when parallelizing algorithms that are coarse-grained by nature are discussed and also the optimizations that are possible. A performance analysis reveals the inefficiencies and explains the deviations from the GPU peak performance.","Title: JPEG Privacy and Security framework for social networking and GLAM services.\nYear: 2017\n\nAbstract: Current image coding standards provide limited support for privacy and security features. An exception is the JPSEC standard, which defines security extensions in JPEG 2000 specifications (part 8). Notwithstanding this shortcoming, the JPEG committee is currently defining a new JPEG Systems standard, which envisages privacy and security support across JPEG family of standards. In this manuscript, the main philosophy of this emerging specification is outlined along with typical use cases, main requirements as well as examples of potential technological solutions. The upcoming specification guarantees backward and forward compatibility with earlier standards and legacy implementations. Finally, we illustrate the introduced framework by two applications targeting secure photo sharing on social networks and IPR management in the GLAM sector.","Title: Bounds and Conditions for Compressive Digital Holography Using Wavelet Sparsifying Bases.\nYear: 2017\n\nAbstract: Numerous experiments have been conducted with success in the field of compressive digital holography, but the theory to determine optimal measurement conditions is lagging behind. In contrast to a prior study that expects object wavefields to be sparse in the spatial domain, we investigate how the configuration of the interferometer influences the reconstruction of wavefields that are sparse in a ...","Title: From Sparse Coding Significance to Perceptual Quality: A New Approach for Image Quality Assessment.\nYear: 2018\n\nAbstract: An increasing number of image processing applications require an automated quality prediction of the visual content as perceived by humans. Since, sparse coding is suggested to be an underlying strategy of the brain's neural system, it would be logical to assume that specific tasks like quality assessment also attempt to adhere to this strategy. However, existing perceptual quality predictors, oft...","Title: Accurate label-free 3-part leukocyte recognition with single cell lens-free imaging flow cytometry.\nYear: 2018\n\nAbstract: Three-part white blood cell differentials which are key to routine blood workups are typically performed in centralized laboratories on conventional hematology analyzers operated by highly trained staff. With the trend of developing miniaturized blood analysis tool for point-of-need in order to accelerate turnaround times and move routine blood testing away from centralized facilities on the rise, our group has developed a highly miniaturized holographic imaging system for generating lens-free images of white blood cells in suspension. Analysis and classification of its output data, constitutes the final crucial step ensuring appropriate accuracy of the system. In this work, we implement reference holographic images of single white blood cells in suspension, in order to establish an accurate ground truth to increase classification accuracy. We also automate the entire workflow for analyzing the output and demonstrate clear improvement in the accuracy of the 3-part classification. High-dimensional optical and morphological features are extracted from reconstructed digital holograms of single cells using the ground-truth images and advanced machine learning algorithms are investigated and implemented to obtain 99% classification accuracy. Representative features of the three white blood cell subtypes are selected and give comparable results, with a focus on rapid cell recognition and decreased computational cost.","Title: Unitary Transforms Using Time-Frequency Warping for Digital Holograms of Deep Scenes.\nYear: 2018\n\nAbstract: With the advent of ultrahigh-resolution holographic displays, viewing macroscopic deep scenes with large viewing angles becomes a possibility. These deep holograms possess different signal properties in contrast with common applications where the scene content is assumed to lie around a planar slice. Therefore, the conventional approach of refocusing at a fixed depth is ineffective. There is a nee...","Title: Image Reconstruction with Smoothed Mixtures of Regressions\nYear: 2018\n\nAbstract: This work builds upon the kernel regression framework for solving the general image processing problem of denoising, deblurring and interpolating from scattered image samples. A competitive expectation-maximization method estimates globally all parameters of a generative image model, accounting for missing samples. One 2D footprint kernel and a local linear regression plane are estimated per data sample. Kernels can shift and their prior probabilities are estimated as well, unlike in nonparametric models. Missing data yields an underdetermined problem that is regularized by smoothing the marginal mixture density. At each iteration, a balloon estimator computes numerically the spatial “territory” associated to each data samples. Results of these numerical diffusion operations are used to convolve adaptively each kernel in the forward model. Finally, the complete image is reconstructed by smoothing regression for combining conditional means of local linear regressors. Experiments apply this iterative Bayesian technique in image restoration.","Title: A Just Noticeable Difference Subjective Test for High Dynamic Range Images\nYear: 2018\n\nAbstract: High Dynamic Range (HDR) imaging captures a wide range of luminance existing in real-world scenes. Due to large luminance levels and higher brightness of HDR displays, artefacts can be more noticeable to the Human Visual System (HVS). In a first attempt to experimentally quantify those noticeable levels for HDR images, we pioneered in conducting an exhaustive and comprehensive Just Noticeable Difference (JND) subjective experiment of which the outcome is presented in this paper. Six distortions including JPEG, JPEG2000, noise, blur, contrast change, and quantization artefacts have been considered in the test. The distortions were applied to 10 HDR images in 100 distortion levels resulting a database of 6000 HDR test images. The subjects were asked to find the image JND location on each set of 100 images they had the freedom to explore. The effect of content features on the noticeable threshold selection is investigated per distortion type. Our results in some cases show a significant correlation between content features and JNDs. We are hoping that our results can contribute to further exploitation of a precise HVS model for HDR quality assessment and optimization of the coding and bit allocation in HDR compression.","Title: Signal processing challenges for digital holographic video display systems.\nYear: 2019\n\nAbstract: Holography is considered to be the ultimate display technology since it can account for all human visual cues such as stereopsis and eye focusing. Aside from hardware constraints for building holographic displays, there are still many research challenges regarding holographic signal processing that need to be tackled. In this overview, we delineate the steps needed to realize an end-to-end chain from digital content acquisition to display, involving the efficient generation, representation, coding and quality assessment of digital holograms. We discuss the current state-of-the-art and what hurdles remain to be taken to pave the way towards realistic visualization of dynamic holographic content.","Title: Reduced-reference quality assessment of multiply-distorted images based on structural and uncertainty information degradation.\nYear: 2018\n\nAbstract: •A reduced-reference image quality assessment method is proposed for quality prediction of multiply-distorted images.•The structural and uncertainty information changes are measured using shearlets and Rényi entropy analysis.•An improved quality prediction model is obtained by incorporating privileged information in the training stage.","Title: From Adaptive Kernel Density Estimation to Sparse Mixture Models.\nYear: 2018\n\nAbstract: We introduce a balloon estimator in a generalized expectation-maximization method for estimating all parameters of a Gaussian mixture model given one data sample per mixture component. Instead of limiting explicitly the model size, this regularization strategy yields low-complexity sparse models where the number of effective mixture components reduces with an increase of a smoothing probability parameter $mathbf{Pu003e0}$. This semi-parametric method bridges from non-parametric adaptive kernel density estimation (KDE) to parametric ordinary least-squares when $mathbf{P=1}$. Experiments show that simpler sparse mixture models retain the level of details present in the adaptive KDE solution.","Title: Fingerprinting Codes Under the Weak Marking Assumption\nYear: 2018\n\nAbstract: Fingerprinting codes based on the marking assumption provide reliable security against collusion attacks, but lack resilience against channel errors such as symbol erasures or other forms of distortions introduced by transmission over noisy or insecure communication channels. This additional source of distortion errors can be addressed by relaxing the marking assumption. In this paper we examine t...","Title: Integer Fresnel Transform for Lossless Hologram Compression\nYear: 2019\n\nAbstract: Digital holograms fully encode the wavefield of light, thereby having many applications both for 3D object measurements as well as for display purposes, accounting for all human visual cues. Because the statistics and properties of holographic signals differ considerably from natural imagery such as photographs, conventional coding solutions will be sub-optimal. In this paper, we propose the integer Fresnel transform, which is - to our knowledge - the first lossless transform tailored for hologram coding. By combining the proposed transform with JPEG 2000, we report bit-rate savings from 0.12 up to 2.83 bits per channel on a collection of 8 digital holograms obtained from 3 different databases.","Title: Wave Atoms for Lossy Compression of Digital Holograms\nYear: 2019\n\nAbstract: Compression of digital holograms is a major challenge that needs to be resolved to enable the efficient storage, transmission and rendering of macroscopic holographic signals. In this work, we propose to deploy the wave atom transform that has been utilized before for interferometric modalities such as acoustic and seismic signals. This non-adaptive multiresolution transform has good space-frequency localization and its orthonormal basis is suitable for sparsifying holographic signals. By replacing the CDF 9/7 wavelet transform stage in a JPEG 2000 codec with the proposed wave atom transform, we did assess its suitability for coding complex amplitude wavefronts. Experimental results demonstrate improved rate-distortion performance with respect to JPEG 2000 and H.265/HEVC for a set of computer-generated, diffuse, macroscopic holograms.","Title: Validation of dynamic subjective quality assessment methodology for holographic coding solutions\nYear: 2021\n\nAbstract: In preparation of the Call for Proposals on JPEG Pleno Holography, multiple exploration studies are ongoing to define the general procedure for performance assessment of proposed coding solutions. The performance of proposals will be compared against JPEG 2000 and H.265/HEVC intra coding, both serving as anchor codecs. In this paper, we report on the results of a dynamic subjective visual quality ...","Title: Real-Time Computation of 3D Wireframes in Computer-Generated Holography\nYear: 2021\n\nAbstract: Computer-Generated Holography (CGH) algorithms simulate numerical diffraction, being applied in particular for holographic display technology. Due to the wave-based nature of diffraction, CGH is highly computationally intensive, making it especially challenging for driving high-resolution displays in real-time. To this end, we propose a technique for efficiently calculating holograms of 3D line segments. We express the solutions analytically and devise an efficiently computable approximation suitable for massively parallel computing architectures. The algorithms are implemented on a GPU (with CUDA), and we obtain a 70-fold speedup over the reference point-wise algorithm with almost imperceptible quality loss. We report real-time frame rates for CGH of complex 3D line-drawn objects, and validate the algorithm in both a simulation environment as well as on a holographic display setup.","Title: Bitwidth-Optimized Energy-Efficient FFT Design via Scaling Information Propagation\nYear: 2021\n\nAbstract: The Fast Fourier Transform (FFT) is an efficient algorithm widely used in digital signal processing to transform between the time domain and the frequency domain. For fixed-point VLSI implementations, dynamic range growth inevitably occurs at each stage of the FFT operation. However, current methods either waste bitwidth or consume excessive resources when dealing with the dynamic range growth issue. To address this issue, we propose an efficient scaling method called Scaling Information Propagation (SIP) to alleviate the problem of dynamic range growth, which makes full use of bitwidth with much less extra area consumed than the state-of-the-art solutions. In two consecutive transform operations, the SIP method extracts scaling information and makes scaling decisions in the former transform, then executes those in the latter one. We implement the FFT's VLSI architecture in the orthogonal frequency division multiplexing (OFDM) and the holographic video compression (HVC) systems to verify the SIP method. Compared to the state-of-the-art, experimental results after VLSI synthesis show that our method achieves 9.38% energy reduction and 8.36% area savings when requiring 1.02 x 10(-7) bit error ratio (BER) of the OFDM system, and 33.47% energy reduction and 30.98% area savings when requiring 20dB signal-to-noise ratio (SNR) of the HVC system, respectively.","Title: Unifying Structural and Semantic Similarities for Quality Assessment of DIBR-Synthesized Views\nYear: 2022\n\nAbstract: Multi-view 3D content is subject to distortions during the process of depth image-based rendering (DIBR). Studies have shown the unreliable performance of the well-established image quality assessment (IQA) models for evaluation of DIBR-synthesized views which surge the need for more effective IQA methods. Existing objective methods generally rely on the pixel-wise correspondences between the reference and distorted images, while view synthesis can introduce pixel shifts. DIBR distortions such as stretching and local hole-filling errors have different visual impacts from conventional distortions, challenging the existing IQA models. Here, we developed a Full-Reference (FR) objective IQA metric for synthesized views that significantly outperforms 2D IQA and the state-of-the-art DIBR IQA approaches. While the pixel misalignment between the reference and synthesized views is a big challenge for quality assessment, we deployed a Convolutional Neural Network (CNN) model to acquire a feature representation that inherently offers resilience to the imperceptible pixel shift between the compared images. Therefore, our model does not need accurate shift compensation. We deployed a set of quality-aware CNN features representing high-order statistics, to measure the structural similarity which is combined with a semantic similarity measure for accurate quality assessment. Moreover, prediction accuracy is improved by incorporating a visual saliency model acquired using the activations of the higher CNN layers. Experimental results indicate a significant performance gain (14.6% in terms of Spearman's rank-order correlation) compared to the top existing IQA model. The source code of the proposed metric is available at: https://gitlab.com/saeedmp/sequss.","Title: Quality evaluation of the JPEG Pleno Holography Call for Proposals response\nYear: 2022\n\nAbstract: At its 93rd meeting, JPEG received the responses to the Call for Proposals on JPEG Pleno holography. This activity will lead to the first standardized coding solution for digital holograms. This paper describes the quality evaluation process, existing out of a subjective evaluation study and an objective evaluation based on preselected quality metrics. It also includes a critical evaluation of the challenges encountered when performing quality evaluation of holographic content. Based on the evaluation of the response, the JPEG committee was able to select a very reliable coding solution that depicted a performance well beyond that of the selected anchors."],"string":"[\n \"Title: Wavelet-based scalable L-infinity-oriented compression.\\nYear: 2006\\n\\nAbstract: Among the different classes of coding techniques proposed in literature, predictive schemes have proven their outstanding performance in near-lossless compression. However, these schemes are incapable of providing embedded L(infinity)-oriented compression, or, at most, provide a very limited number of potential L(infinity) bit-stream truncation points. We propose a new multidimensional wavelet-based L(infinity)-constrained scalable coding framework that generates a fully embedded L(infinity)-oriented bit stream and that retains the coding performance and all the scalability options of state-of-the-art L2-oriented wavelet codecs. Moreover, our codec instantiation of the proposed framework clearly outperforms JPEG2000 in L(infinity) coding sense.\",\n \"Title: Spatio-temporal-SNR scalable wavelet coding with motion-compensated DCT base-layer architectures\\nYear: 2003\\n\\nAbstract: It has been demonstrated recently that 3-D wavelet coding with motion-compensated temporal filtering (MCTF) provides a wide range of spatio-temporal-SNR scalability with state-of-the-art coding performance. However, the coding system is very different from the already standardized motion-compensated DCT (MC-DCT) video coders such as MPEG-2, MPEG-4, or H.26L. Nonetheless, the market acceptance of the new scalable technology will come much easier if backward compatibility to such previous standards would exist. In this paper, we present a new coder architecture where the base layer can be a standard MC-DCT coder, while the enhancement layer is an in-band MCTF codec operating in the overcomplete wavelet domain. First, we propose a simple extension of the scalable 3-D wavelet codec with a standard MC-DCT base layer. As a second step, to improve the performance of the proposed scalable coder over a wide range of bit-rates, we describe several small extensions to the standardized MC-DCT codec structure that can be applied for the base layer coding.\",\n \"Title: Estimating the Detectability of Small Lesions in High Resolution MR Compressed Images\\nYear: 2008\\n\\nAbstract: Quality degradation in high resolution magnetic resonance (MR) images compressed with JPEG2000 is studied. The present study combines the results of ten quantitative quality criteria with four mathematical observer models to estimate the maximum achievable compression rate (CR) that does not affect an effective diagnosis of specific pathologies related with small lesions detection such as Multiple Sclerosis or Virchow Robins disease. The graphical behavior of metrics employed is presented. At bitrate = 0.062 bpp (CR=160:1) the most compression is achieved while images still preserve the information needed for a safe diagnosis. Images compressed using smaller values than this would be no longer useful for diagnosis tasks. This result facilitates a safer use of the JPEG 2000 codec when compressing high resolution images like the images tested, and assures a safer diagnosis task when dealing with small size lesions that define certain pathology.\",\n \"Title: Wavelet Coding of Volumetric Medical Datasets.\\nYear: 2003\\n\\nAbstract: Several techniques based on the three-dimensional (3-D) discrete cosine transform (DCT) have been proposed for volumetric data coding. These techniques fail to provide lossless coding coupled with quality and resolution scalability, which is a significant drawback for medical applications. This paper gives an overview of several state-of-the-art 3-D wavelet coders that do meet these requirements and proposes new compression methods exploiting the quadtree and block-based coding concepts, layered zero-coding principles, and context-based arithmetic coding. Additionally, a new 3-D DCT-based coding scheme is designed and used for benchmarking. The proposed wavelet-based coding algorithms produce embedded data streams that can be decoded up to the lossless level and support the desired set of functionality constraints. Moreover, objective and subjective quality evaluation on various medical volumetric datasets shows that the proposed algorithms provide competitive lossy and lossless compression results when compared with the state-of-the-art.\",\n \"Title: FULLY-SCALABLE WAVELET VIDEO CODING USING IN-BAND MOTION COMPENSATED TEMPORAL FILTERING\\nYear: 2003\\n\\nAbstract: This paper presents a novel fully-scalable wavelet video coding scheme that performs efficient open-loop motion compensated temporal filtering (MCTF) in the wavelet domain (in-band). Unlike the conventional spatial-domain MCTF (SDMCTF) schemes, which apply MCTF on the original image data and then encode the residual image using a critically-sampled wavelet transform, the framework presented here applies the in-band MCTF (IBMCTF) after the discrete wavelet transform (DWT) is performed in the spatial dimensions. To overcome the inefficiency of motion estimation (ME) in the wavelet domain, a complete-to-overcomplete DWT (CODWT) is performed. The proposed framework provides improved quality (SNR) and temporal scalability as compared with existing in-band closed-loop temporal prediction schemes with ODWT and improved spatial scalability as compared to SDMCTF. We present a thorough comparison between SDMCTF and the proposed IBMCTF in terms of coding efficiency and scalability. Furthermore, we describe several extensions that enable the filtering of the various bands to be performed independently, based on the resolution, sequence content, complexity requirements and desired scalability.\",\n \"Title: Power-Aware Computing Systems\\nYear: 2007\\n\\nAbstract: From April 3rd to April 8th 2005, a scientific workshop on the topic of \\\"Power-aware Computing Systems\\\" was held at Schloss Dagstuhl, Germany. The main seminar result is a classification of the obstacles, and therefore research directions, with respect to power consumption seen for different classes of devices, ranging from ultra low-power devices, over handheld devices, to servers and work stations. In a next step the seminar identified the impact different levels of dealing with power concerns can have. This paper summarises the objectives and structure as well as the outcome of that workshop.\",\n \"Title: Towards Standardized Integration of Images in the Cloud of Linked Data.\\nYear: 2013\\n\\nAbstract: Currently, there are several ways of describing and referring to images in RDF. This ambiguity results in a proliferation of vocabularies for image descriptions, complicating the cross-community data integration of information related to digital images. In addition, there are no standardized guidelines on how to integrate RDF data into the image metadata. Therefore, the JPEG standardization committee has recently initiated some activities to streamline the integration of images in the cloud of Linked Data. One effort is the standardization of an ontology for describing digital images. JPEG aims at providing a technology that enables the uniform description of photos and videos with technical, administrative and semantic metadata compliant with the RDF specification and the principles of Linked Data. Secondly, specifications to integrate RDF data into JPEG images are elaborated. Finally, since descriptions often only apply to a certain part of an image, a last effort is to formalize the specification of regions of interest. In this paper, members of the JPEG committee provide a detailed overview of these ongoing activities.\",\n \"Title: On hybrid directional transform-based intra-band image coding\\nYear: 2007\\n\\nAbstract: In this paper, we propose a generic hybrid oriented-transform and wavelet-based image representation for intra-band image coding. We instantiate for three popular directional transforms having similar powers of approximation but different redundancy factors. For each transform type, we design a compression scheme wherein we exploit intra-band coefficient dependencies. We show that our schemes outperform alternative approaches reported in literature. Moreover, on some images, we report that two of the proposed codec schemes outperform JPEG2000 by over 1dB. Finally, we investigate the trade-off between oversampling and sparsity and show that, at low rates, hybrid coding schemes with transform redundancy factors as high as 1.25 to 5.8 are capable in fact of outperforming JPEG2000 and its critically-sampled wavelets.\",\n \"Title: Visually lossless screen content coding using HEVC base-layer.\\nYear: 2013\\n\\nAbstract: This paper presents a novel two-layer coding framework targeting visually loss less compression of screen content video. The proposed framework employs the conventional HEVC standard for the base-layer. For the enhancement layer, a hybrid of spatial and temporal block-prediction mechanism is introduced to guarantee a small energy of the error-residual. Spatial prediction is generally chosen for dynamic areas, while temporal predictions yield better prediction for static areas in a video frame. The prediction residual is quantized based on whether a given block is static or dynamic. Run-length coding, Golomb based binarization and context-based arithmetic coding are employed to efficiently code the quantized residual and form the enhancement-layer. Performance evaluations using 4: 4: 4 screen content sequences show that, for visually lossless video quality, the proposed system significantly saves the bit-rate compared to the two-layer lossless HEVC framework.\",\n \"Title: Scalable and channel-adaptive unequal error protection of images with LDPC codes\\nYear: 2006\\n\\nAbstract: This paper considers the design of an optimal joint source-channel coding system employing scalable wavelet-based source coders and unequal error protection for error-resilient transmission over binary erasure channels. We theoretically show that the expected average rate-distortion function in the separately encoded subbands is convex with monotonically decreasing slopes and use this fact to propose a new algorithm which minimizes the end-to-end distortion. This algorithm is based on a simplified Viterbi-search followed by Lagrangian-optimization. We show that our proposed solution results in significant complexity reductions while providing very near-to-optimal performance.\",\n \"Title: Modeling Wavelet Coefficients for Wavelet Subdivision Transforms of 3D Meshes\\nYear: 2010\\n\\nAbstract: In this paper, a Laplacian Mixture (LM) model is proposed to accurately approximate the observed histogram of the wavelet coefficients produced by lifting-based subdivision wavelet transforms. On average, the proposed mixture model gives better histogram fitting for both normal and non-normal meshes compared to the traditionally used Generalized Gaussian (GG) distributions. Exact closed-form expressions for the rate and the distortion of the LM probability density function quantized using generic embedded deadzone scalar quantizer (EDSQ) are derived, without making high-rate assumptions. Experimental evaluations carried out on a set of 3D meshes reveals that, on average, the D-R function for the LM model closely follows and gives a better indication of the experimental D-R compared to the D-R curve of the competing GG model. Optimal embedded quantization for the proposed LM model is experimentally determined. In this sense, it is concluded that the classical Successive Approximation Quantization (SAQ) is an acceptable, but in general, not an optimal embedded quantization solution in wavelet-based scalable coding of 3D meshes.\",\n \"Title: Scalable wavelet video-coding with in-band prediction - the bottom-up overcomplete discrete wavelet transform\\nYear: 2002\\n\\nAbstract: A new in-band motion compensation algorithm for wavelet-based video coding is proposed: the Bottom-Up Prediction algorithm (BUP). The BUP algorithm overcomes the periodic shift-invariance of the discrete wavelet transform (DWT) and is formalized into new prediction rules using filtering operations. BUP is based on the relationships between the subbands of the shifted input signal and the subbands of the non-shifted reference signal, whereby the number of shifts is limited by the periodic shift-invariance of the DWT. We derive the algorithm for the 1-D DWT with three decomposition levels. The combination of all prediction rules of the BUP algorithm defines a new transform: the Bottom-Up Overcomplete DWT or BUP ODWT, which is shift- invariant. The BUP ODWT calculates the overcomplete subbands by applying the prediction rules to the critically sampled subbands of a wavelet-transformed image. The envisaged application for the BUP algorithm is spatially scalable video coding.\",\n \"Title: A New Family Of Embedded Multiple Description Scalar Quantizers\\nYear: 2004\\n\\nAbstract: A new family of Embedded Multiple Description Scalar Quantizers (EMDSQ) that Supports the progressive transmission of images over variable-bandwidth error-prone channels is proposed in this paper. A control mechanism that allows for tuning the redundancy between the two descriptions for each quantization level is also designed. The employed mechanism enables the control of the tradeoff between the coding efficiency and error-resilience, and provides an increased robustness by improving the error resilience in the most important layers of the embedded bit-streams. Instantiations of the proposed family are incorporated in a wavelet-based embedded coding system, and the redundancy-control mechanism is practically demonstrated. Experimental results show that the proposed EMDSQ outperform the state-of-the-art Multiple Description Uniform Scalar Quantizers (MDUSQ) previously proposed in literature for error-resilient progressive image transmission.\",\n \"Title: L-infinite Coding of 3D Representations of Human Affect\\nYear: 2012\\n\\nAbstract: Off-line scanning, coding, transmission and remote animation of the human affect represents a possible processing pipeline for providing 3D immersion in virtual worlds. In this paper we target applications that make use of compact and scalable 3D representations of human affect and require close control over the local error introduced by lossy coding of the mesh geometry. To satisfy this requirement, we propose a novel L-infinite wavelet-based semi-regular mesh coding system. The system lies in contrast with classical mesh coding approaches which make use of the L-2 distortion metric. Specifically, in contrast to an L-2 driven implementation, the proposed system provides a bound on the local error on each vertex resulting from scalar embedded quantization of the wavelet coefficients. The experiments show that the proposed system provides scalability in L-infinite sense and that it outperforms the state-of-the art in L-infinite mesh coding.\",\n \"Title: Probabilistic motion-compensated prediction in distributed video coding\\nYear: 2013\\n\\nAbstract: Distributed video coding (DVC) constitutes an original coding framework to meet the stringent requirements imposed by uplink-oriented and low-power mobile video applications. The quality of the side information available to the decoder and the efficiency of the employed channel codes are primary factors determining the success of a DVC system. This contribution introduces two novel techniques for probabilistic motion compensation in order to generate side information at the Wyner-Ziv decoder. The employed DVC scheme uses a base layer, serving as a hash to facilitate overlapped block motion estimation at the decoder side. On top of the base layer, a supplementary Wyner-Ziv layer is coded in the DCT domain. Both proposed probabilistic motion compensation techniques are driven by the actual correlation channel statistics and reuse information contained in the hash. Experimental results report significant rate savings caused by the novel side information generation methods compared to previous techniques. Moreover, the compression performance of the presented DVC architecture, featuring the proposed side-information generation techniques, delivers state-of-the-art compression performance.\",\n \"Title: Complete-To-Overcomplete Discrete Wavelet Transforms For Scalable Video Coding With Mctf\\nYear: 2003\\n\\nAbstract: Techniques for full scalability with motion-compensated temporal filtering (MCTF) in the wavelet-domain (in-band) are presented in this paper. The application of MCTF in the wavelet domain is performed after the production of the overcomplete discrete wavelet transform from the critically-sampled decomposition, a process that occurs at both the encoder and decoder side. This process, which is a complete-to-overcomplete discrete wavelet transform, is critical for the efficiency of the system with respect to scalability, coding performance and complexity. We analyze these aspects of the system and set the necessary constraints for drift-free video coding with in-band MCTF. As a result, the proposed architecture permits the independent operation of MCTF within different resolution levels or even different subbands of the transform and allows the successive refinement of the video information in resolution, frame-rate and quality.\",\n \"Title: Wyner-Ziv video coding for wireless lightweight multimedia applications.\\nYear: 2012\\n\\nAbstract: Wireless video communications promote promising opportunities involving commercial applications on a grand scale as well as highly specialized niche markets. In this regard, the design of efficient video coding systems, meeting such key requirements as low power, mobility and low complexity, is a challenging problem. The solution can be found in fundamental information theoretic results, which gave rise to the distributed video coding (DVC) paradigm, under which lightweight video encoding schemes can be engineered. This article presents a new hash-based DVC architecture incorporating a novel motion-compensated multi-hypothesis prediction technique. The presented method is able to adapt to the regional variations in temporal correlation in a frame. The proposed codec enables scalable Wyner-Ziv video coding and provides state-of-the-art distributed video compression performance. The key novelty of this article is the expansion of the application domain of DVC from conventional video material to medical imaging. Wireless capsule endoscopy in particular, which is essentially wireless video recording in a pill, is proven to be an important application field. The low complexity encoding characteristics, the ability of the novel motion-compensated multi-hypothesis prediction technique to adapt to regional degrees of temporal correlation (which is of crucial importance in the context of endoscopic video content), and the high compression performance make the proposed distributed video codec a strong candidate for future lightweight (medical) imaging applications.\",\n \"Title: JPSearch: Metadata Interoperability During Image Exchange [Standards in a Nutshell].\\nYear: 2012\\n\\nAbstract: Billions of new images are generated daily. However, the management and exchange of the associated metadata remains cumbersome. Therefore, the Joint Photographic Experts Group (JPEG) has recently promoted a new set of technologies enabling the interoperability between image repositories and/or their clients to an international standard. JPSearch is a suite of specifications that supports the enric...\",\n \"Title: Wavelet-based fixed and embedded L-infinite-constrained coding\\nYear: 2002\\n\\nAbstract: A new wavelet-based L-infinite-constrained fixed and embedded coding technique is proposed in this paper. For any desired distortion bound, the embedded bit stream can be truncated at a corresponding bit-rate, for which the required upper bound on the individual elements of the reconstruction error signal is guaranteed. The original signal can also be coded up to a fixed a priori user-defined distortion bound, ranging up to lossless coding. The proposed coding technique supports both 2D (images) and 3D (volumetric) input data.\",\n \"Title: JPSearch: An answer to the lack of standardization in mobile image retrieval\\nYear: 2013\\n\\nAbstract: The amount of imagery that is available via various content providers grows at a staggering rate. Contemporary handheld devices allow accessing these content providers anytime, anyplace. Moreover, using the build-in camera, one can easily contribute to these online image repositories. The present lack of consistency in how repositories are accessed for retrieval or syncing complicates inter-operability between devices and systems. JPSearch, a recent standardization initiative within the Joint Photographic Experts Group (JPEG) committee, resolves this predicament by defining the interfaces and protocols for data exchange between them, while restricting as little as possible how those devices, systems or components perform their task.\",\n \"Title: Motion and texture rate-allocation for prediction-based scalable motion-vector coding\\nYear: 2005\\n\\nAbstract: Modern video coding applications require data transmission over variable-bandwidth wired and wireless network channels to a variety of terminals, possibly having different screen resolutions and available computing power. Scalable video coding technology is needed to optimally support these applications. Recently proposed wavelet-based video codecs employing spatial-domain motion-compensated temporal filtering (SDMCTF) provide quality, resolution and frame-rate scalability while delivering compression performance comparable to that of H.264, the state-of-the-art in single-layer video coding. These codecs require quality-scalable coding of the motion vectors to support a large range of bit-rates with optimal compression efficiency. In this paper, the practical use of prediction-based scalable motion-vector coding in the context of scalable SDMCTF-based video coding is investigated. Extensive experimental results demonstrate that, irrespective of the employed motion model, our prediction-based scalable motion-vector codec (MVC) systematically outperforms state-of-the-art wavelet-based solutions for both lossy and lossless compression. A new rate-distortion optimized rate-allocation strategy is proposed, capable of optimally distributing the available bit-budget between the different frames and between the texture and motion information, making the integration of the scalable MVC into a scalable video codec possible. This rate-allocation scheme systematically outperforms heuristic approaches previously employed in the literature. Experiments confirm that by using a scalable MVC, lower bit-rates can be attained without sacrificing motion-estimation efficiency and that the overall coding performance at low rates is significantly improved by a better distribution of the available rate between texture and motion information. The only downside of scalable motion-vector coding is a slight performance loss incurred at high bit-rates.\",\n \"Title: Wavelet-based compression of medical images: Protocols to improve resolution and quality scalability and region-of-interest coding\\nYear: 1999\\n\\nAbstract: The paper describes a methodology to improve the scalability support of an embedded bit stream, generated with a wavelet-based compression algorithm, and a generic protocol to handle multiple regions-of-interest (ROIs). The generic scheme, illustrated for embedded zero-tree wavelet (EZW) coding, exploits the inherent graceful degradation capabilities of wavelet-based compression methods and ensures an optimal trade-off between the image reconstruction quality and the compression ratio. Additionally, an efficient protocol is proposed to handle multiple ROIs in an interactive client-server set-up for telemedicine applications. The processing of the ROIs takes place in the wavelet domain.\",\n \"Title: Maximum Likelihood Laplacian Correlation Channel Estimation in Layered Wyner-Ziv Coding\\nYear: 2014\\n\\nAbstract: The distributed source coding theory teaches that efficient compression can be achieved by exploiting the correlation between encoded sources at the decoder only. A critical issue in this coding paradigm is how to accurately estimate the correlation statistics at the decoder. This paper proposes a novel correlation channel estimation method designed for generic layered Wyner-Ziv (WZ) coding. Driven by real-life applications, i.e., video coding, the correlation noise is assumed to be memoryless zero-mean Laplacian with a block-stationary behavior. Unlike existing methods, the proposed technique derives a novel maximum likelihood estimate of the noise scaling parameter per stationarity block. Adhering to layered coding, the derived estimate is successively refined as more stages are decoded. The Crámer-Rao lower bound (CRLB) for our estimator is derived and the mean squared error performance of the proposed successive refinement algorithm is evaluated through simulations. Apart from achieving an accuracy close to the derived CRLB, the proposed algorithm is shown to yield a WZ rate-distortion performance approaching the ideal but unrealistic scenario where the decoder knows a priori the correlation statistics. Moreover, the proposed technique has been integrated into our latest WZ video codec for wireless capsule endoscopy. Experimentation reveals that the proposed method delivers improved compression performance compared to state-of-the-art techniques, while simultaneously reducing the overall decoding complexity.\",\n \"Title: Constant quality video coding using video content analysis\\nYear: 2005\\n\\nAbstract: In the literature, several rate control techniques have been proposed to aim at the optimal quality of digitally encoded video under given bit budget, channel rate and buffer size constraints. Typically, these approaches are group-of-picture (GOP) based. For longer, heterogeneous sequences, they become unacceptably complex or struggle with model mismatches. In this paper, an off-line segment-based rate control approach is proposed for controlling the distortion variation across successive shots of a video sequence when encoding with single-layer (MPEG-4 baseline, MPEG-4 AVC) and scalable (wavelet) video codecs. Consistent quality is achieved by optimally distributing the available bits among the different segments, based on efficient rate-distortion (R-D) modelling of each segment. The individual segments are defined based on shot segmentation and activity analysis techniques. The algorithm is formulated for three different distribution models: download, progressive download and streaming. The results indicate that the proposed technique improves the quality consistency significantly, while the processing overhead compared to classical two-pass variable bit-rate (VBR) encoding is limited.\",\n \"Title: Side-information-dependent correlation channel estimation in hash-based distributed video coding.\\nYear: 2012\\n\\nAbstract: In the context of low-cost video encoding, distributed video coding (DVC) has recently emerged as a potential candidate for uplink-oriented applications. This paper builds on a concept of correlation channel (CC) modeling, which expresses the correlation noise as being statistically dependent on the side information (SI). Compared with classical side-information-independent (SII) noise modeling adopted in current DVC solutions, it is theoretically proven that side-information-dependent (SID) modeling improves the Wyner-Ziv coding performance. Anchored in this finding, this paper proposes a novel algorithm for online estimation of the SID CC parameters based on already decoded information. The proposed algorithm enables bit-plane-by-bit-plane successive refinement of the channel estimation leading to progressively improved accuracy. Additionally, the proposed algorithm is included in a novel DVC architecture that employs a competitive hash-based motion estimation technique to generate high-quality SI at the decoder. Experimental results corroborate our theoretical gains and validate the accuracy of the channel estimation algorithm. The performance assessment of the proposed architecture shows remarkable and consistent coding gains over a germane group of state-of-the-art distributed and standard video codecs, even under strenuous conditions, i.e., large groups of pictures and highly irregular motion content.\",\n \"Title: Semi-Regular Remeshing With Reduced Remeshing Error\\nYear: 2010\\n\\nAbstract: In this paper we present a remeshing algorithm which drastically reduces the aliasing artifacts inherent in regularly-sampled remeshed objects. Starting from a semi-regular mesh, the proposed algorithm reduces the remeshing error and avoids aliasing by displacing vertices such that most samples of the original mesh are present in the remeshed model as well. Computational efficiency is provided by using a search-tree, which efficiently gathers vertices near a given point in a 3D space. Compared to the state-of-the-art semi-regular remesher, the proposed remesher drastically improves the visual quality of the high-frequency regions in remeshed objects. Additionally, the proposed algorithm yields a lower remeshing error, which is reflected by a significantly increased PSNR upper-bound in wavelet-based compression of such meshes.\",\n \"Title: Platform-scalable Task Partition and Multilevel Buffering in Multi-processor Plessey Corner Detector\\nYear: 2007\\n\\nAbstract: The Plessey corner detector is a key technological component in scene analysis, stereo matching, and object tracking. Due to its high computation complexity, earlier fast implementations mainly focused on hardware implementations. This paper explores the viability of a multi-processor software implementation. A scalable task partitioning for efficiently mapping the Plessey algorithm on a multi-processor platform is proposed. The task partition ensures platform scalability, low inter-processor communication overhead and a well-balanced workload in each task. In addition, a multilevel buffering scheme is presented, minimizing the external memory accesses in each task to one image pixel read per calculated corner response value. The effectiveness of the proposed task partition and buffering scheme has been verified on (i) a cycle accurate simulator with shared memory and (ii) a multiple-TI-C64 DSP board using a message passing paradigm. The proposed solution combines good platform scalability with an additional 30% speedup gain over straightforward parallelization schemes.\",\n \"Title: Overlapped Block Motion Estimation and Probabilistic Compensation with Application in Distributed Video Coding\\nYear: 2009\\n\\nAbstract: It has been recently demonstrated that, in distributed video coding (DVC) side-information dependent modeling of the correlation channel brings significant performance gains over side-information independent assumptions. In this letter, we present a novel technique enabling advanced side-information dependent estimation of the correlation channel at the decoder starting from a very coarse knowledg...\",\n \"Title: Scalable wavelet video-coding with in-band prediction - implementation and experimental results\\nYear: 2002\\n\\nAbstract: In this paper we elaborate on a recently proposed approach for scalable video-coding based on in-band prediction in the overcomplete wavelet domain. It is shown that, through a new calculation scheme for the level-by-level complete to overcomplete discrete wavelet transform (DWT) that exploits certain symmetries, important reductions in the multiplication budget are obtained in comparison to the fastest-known algorithm of the literature. Based on the derived overcomplete transform-domain coefficients, a pixel-accurate motion estimation and compensation (ME/MC) algorithm is proposed, which provides a hybrid-coding framework that supports full scalability in resolution, quality and frame rate. To give an indication of the coding performance of such a system, some preliminary results are reported.\",\n \"Title: Mosaicing of Fibered Fluorescence Microscopy Video\\nYear: 2008\\n\\nAbstract: Fibered fluorescence microscopy is a recent developed image modality using a fiber optic probe connected to a laser scanning unit. It allows for in-vivo scanning of small animal subjects by moving the probe along the tissue surface. During the scans images are continuity captured, allowing to acquire an area larger then the field of view of the probe as a video. But there is still a need to obtain a single static image from the multiple overlapping frames. In this paper we introduce a mosaicing procedure for this kind of video sequence. An additional motivation for the mosaicing is the use of overlapping redundant information to improve the signal to noise level of the acquisition, since the individual frames tend to have both high noise levels and intensity inhomogeneities.\",\n \"Title: An investigation into the performance of reduction algorithms under load imbalance\\nYear: 2012\\n\\nAbstract: Today, most reduction algorithms are optimized for balanced workloads; they assume all processes will start the reduction at about the same time. However, in practice this is not always the case and significant load imbalances may occur and affect the performance of said algorithms. In this paper we investigate the impact of such imbalances on the most commonly employed reduction algorithms and propose a new algorithm specifically adapted to the presented context. Firstly, we analyze the optimistic case where we have a priori knowledge of all imbalances and propose a near-optimal solution. In the general case, where we do not have any foreknowledge of the imbalances, we propose a dynamically rebalanced tree reduction algorithm. We show experimentally that this algorithm performs better than the default OpenMPI and MVAPICH2 implementations.\",\n \"Title: An offline bidirectional tracking scheme\\nYear: 2005\\n\\nAbstract: A generic bi-directional scheme is proposed that robustifies the estimation of the maximum-a-posteriori (MAP) sequence of states of a visual object. It enables creative, non-technical users to obtain the path of interesting objects in offline available video material, which can then be used to create interactive movies. To robustify against tracker failure the proposed scheme merges the filtering distributions of a forward tracking particle filter and a backward tracking particle filter at some timesteps, using a reliability-based voting scheme such as in democratic integration. The MAP state sequence is obtained using the Viterbi algorithm on reduced state sets per timestep derived from the merged distributions and is interpolated linearly where tracking failure is suspected. The presented scheme is generic, simple and efficient and shows good results for a color-based particle filter.\",\n \"Title: Reversible DCT-based lossy-to-lossless still image compression\\nYear: 2013\\n\\nAbstract: This paper presents a novel still image compression scheme that extends the traditional JPEG standard with lossy-to-lossless compression support. The system follows a two-layer design approach which allows for backward compatibility with the conventional JPEG standard for the base layer and provides lossless compression when decoding the enhancement layer. The system employs several coding tools, including quadtree coding, spatial domain prediction, reversible discrete cosine transforms, and context-based arithmetic coding to efficiently encode losslessly the enhancement layer. Performance evaluations using a standard JPEG set of images show that the proposed system yields similar lossless compression performance to the state-of-the-art single-layer JPEG-LS standard while providing quality scalability and a JPEG-compatible base layer at the same time.\",\n \"Title: Jpeg2000 Part 10 - Volumetric Data Encoding\\nYear: 2006\\n\\nAbstract: The Joint Photographic Experts Group (JPEG) committee (ISO/IEC JTCI/SC29/WGI) is currently pursuing the standardization of a three-dimensional extension of the JPEG-2000 standard (Parts 1 and 2) to support the encoding of volumetric data sets. This extension, Part 10 - Extensions for Three-Dimensional Data (MD), will support functionalities like resolution scalability, quality scalability and region-of-interest coding, while exploiting the entropy in the additional third dimension to improve the rate-distortion performance. In this paper, we give an overview of the markets and application areas targeted by JP3D, the imposed requirements and the algorithm under study.\",\n \"Title: Digital canvas removal in paintings.\\nYear: 2012\\n\\nAbstract: The periodic structure of the underlying support of paintings on canvas can become quite prominent and disturbing in high resolution digital recordings. In this paper, we construct a new model and method for the digital removal of canvas which is considered as a noise component superimposed on the painting artwork. The high resolution of the images prohibits the efficient application of existing adaptive denoising filters. Hence, a two-step approach is proposed. First a (smoothing) Wiener filter is applied to the complete image. The second step consists of a spatially adaptive extension with low-complexity to obtain a generic digital canvas removal filter.\",\n \"Title: On the optimality of embedded deadzone scalar-quantizers for wavelet-based L-infinite-constrained image coding\\nYear: 2003\\n\\nAbstract: Summary form only given. Several methods for L∞-distortion constrained compression have been proposed that target a set of fixed reconstruction-error bounds. Recently, a wavelet-based L∞-constrained embedded image-coding technique was proposed that guarantees the required distortion bound while retaining the coding performance and scalability option state-of-the-art wavelet-based L∞-oriented codecs. The optimality of embedded deadzone scalar-quantizers for high-rate L∞-constrained scalable wavelet-based coding of images was analyzed. The optimal quantizers were employed in a coding algorithm that outperforms embedded L∞-oriented wavelet-coders in terms of maximum absolute error. The algorithm was briefly described and coding results were provided.\",\n \"Title: Special Effects: Efficient and Scalable Encoding of the 3D Metamorphosis Animation with\\nYear: 2005\\n\\nAbstract: Metamorphosis or morphing is a technique to accomplish a gradual shape transformation between a source object and a target\\n object. This technique is usually used for animation, i.e., to create some special effects, but it can be also employed as\\n a modeling tool where some existing shapes are combined in order to obtain new shapes. Several interactive or automatic morphing\\n techniques were designed in the past but always with a specific object representation in mind. In this paper we propose a\\n generic framework for 3D morphing, that allows multi-resolution surface extraction and provides a hybrid (surface-volume)\\n dynamic representation which is suitable for encoding the static models, for supporting different morphing methods, and for\\n encoding the intermediate 3D frames (generated during the morphing process) in a MPEG-4 stream.\",\n \"Title: In-band motion compensated temporal filtering\\nYear: 2004\\n\\nAbstract: A novel framework for fully scalable video coding that performs open-loop motion-compensated temporal filtering (MCTF) in the wavelet domain (in-band) is presented in this paper. Unlike the conventional spatial-domain MCTF (SDMCTF) schemes, which apply MCTF on the original image data and then encode the residuals using the critically sampled discrete wavelet transform (DWT), the proposed framework applies the in-band MCTF (IBMCTF) after the DWT is performed in the spatial dimensions. To overcome the inefficiency of MCTF in the critically-sampled DWT, a complete-to-overcomplete DWT (CODWT) is performed. Recent theoretical findings on the CODWT are reviewed from the application perspective of fully-scalable IBMCTF, and constraints on the transform calculation that allow for fast and seamless resolution-scalable coding are established. Furthermore, inspired by recent work on advanced prediction techniques, an algorithm for optimized multihypothesis temporal filtering is proposed in this paper. The application of the proposed algorithm in MCTF-based video coding is demonstrated, and similar improvements as for the multihypothesis prediction algorithms employed in closed-loop video coding are experimentally observed. Experimental instantiations of the proposed IBMCTF and SDMCTF coders with multihypothesis prediction produce single embedded bitstreams, from which subsets are extracted to be compared against the current state-of-the-art in video coding.\",\n \"Title: Wavelet-based fixed and embedded L-infinite-constrained image coding\\nYear: 2003\\n\\nAbstract: A new wavelet-based L-infinity-constrained fixed and embedded coding technique is proposed. The embedded bit stream can be truncated for any desired distortion bound at a corresponding bit rate, so that the target upper bound on the elements of the reconstruction error signal is guaranteed. The original image can also be coded up to a fixed a priori user-defined distortion bound, ranging up to lossless coding. A lifting-based wavelet decorrelating transform is employed on the original image, and exact relations are established between spatial and wavelet domain distortions. The wavelet coefficients are quantized by symmetric uniform quantizers for fixed-distortion coding and by families of embedded uniform deadzone scalar quantizers for embedded coding. The quantized coefficients are finally losslessly encoded using a quadtree-based coding algorithm. Any floating-point lifting-based wavelet transform can be used, and a few of the popular wavelet transforms included in the JPEG2000 verification model are worked out as examples. We compare other L-infinity-constrained coding schemes and show that our proposed coder offers in addition a fully embedded L-infinity-oriented bit stream. We illustrate also that the proposed coder retains the same capabilities as the state-of-the-art embedded wavelet-based co-decs, while providing superior compression results and embeddedness with respect to the L-infinity distortion measure. (C) 2003 SPIE and IST.\",\n \"Title: Error protection of scalable sources: a comparative analysis of forward error correction and multiple description coding\\nYear: 2009\\n\\nAbstract: In this paper we examine the Distortion-Rate (D-R) performance of two competing approaches for error-resilient transmission of scalable sources over error-prone packet-based memoryless channels. Thereby, we consider scalable Joint Source-Channel Coding (JSCC) based on Forward Error Correction (FEC) and scalable Multiple Description Coding (MDC). Our analysis considers memoryless Gaussian input sources encoded using source coders that can achieve theoretical performance bounds. This input source model is independent of actual implementations of the source coders and therefore allows drawing conclusions on the theoretical D-R performance of both approaches. Our comparative analysis indicates that in matched channel conditions, the use of an optimized FEC-based JSCC approach yields significantly better error resilience compared to the use of scalable MDC Moreover, it is shown that in mismatched conditions a scalable JSCC approach has the potential to outperform a scalable MDC approach.\",\n \"Title: A Locally Adaptive System for the Fusion of Objective Quality Measures\\nYear: 2014\\n\\nAbstract: Objective measures to automatically predict the perceptual quality of images or videos can reduce the time and cost requirements of end-to-end quality monitoring. For reliable quality predictions, these objective quality measures need to respond consistently with the behavior of the human visual system (HVS). In practice, many important HVS mechanisms are too complex to be modeled directly. Instead, they can be mimicked by machine learning systems, trained on subjective quality assessment databases, and applied on predefined objective quality measures for specific content or distortion classes. On the downside, machine learning systems are often difficult to interpret and may even contradict the input objective quality measures, leading to unreliable quality predictions. To address this problem, we developed an interpretable machine learning system for objective quality assessment, namely the locally adaptive fusion (LAF). This paper describes the LAF system and compares its performance with traditional machine learning. As it turns out, the LAF system is more consistent with the input measures and can better handle heteroscedastic training data.\",\n \"Title: Single-rate calculation of overcomplete discrete wavelet transforms for scalable coding applications\\nYear: 2005\\n\\nAbstract: A number of emerging resolution-scalable image and video coding algorithms have recently shown very promising performance due to the use of overcomplete wavelet representations. In these applications, the overcomplete discrete wavelet transform (ODWT) is derived starting from the critically-sampled subbands of the DWT (complete representation) of a certain decomposition (resolution) level. This process, which is a complete-to-overcomplete DWT (CODWT), is used for wavelet domain operations that require shift invariance. Specifically, both the encoder and decoder independently construct the overcomplete representation at the best accuracy possible, given the critically-sampled subbands of a certain resolution level. In contrast to the classical approach for performing the CODWT, which is a multi-rate calculation scheme that requires the reconstruction of the input spatial-domain signal, in this paper we propose a new, single-rate calculation scheme, which is formalized for the general case of an arbitrary decomposition (resolution) level. Based on derived symmetry properties, a simple implementation structure of the proposed approach provides interesting tradeoffs for the required multiplication budget in comparison to the conventional approach. This leads to a complexity-scalable solution that fits the versatile requirements of scalable coding environments. The use of the proposed single-rate calculation scheme of the CODWT is demonstrated in several image and video coding systems.\",\n \"Title: Comparative Study of Wavelet Based Lattice QIM Techniques and Robustness against AWGN and JPEG Attacks\\nYear: 2009\\n\\nAbstract: We study watermarking techniques based on Quantization Index Modulation for which sets of lattice quantizers Q m are used (LQIM). A recipe for constructing such quantizers Q m is proposed, where the size of these sets is variable, so that the payload is easily adaptable. We make a comparative study of 8 dimensional lattices with good quantizer properties, where the embedding is done in the wavelet domain. Along the way, the gap between the theoretical ideas behind QIM and practical systems using lattices is closed by extending techniques, such as dithered quantizers and distortion compensation, from the scalar case to LQIM.\",\n \"Title: Modeling the Correlation Noise in Spatial Domain Distributed Video Coding\\nYear: 2009\\n\\nAbstract: Conventional models in distributed video coding (DVC) consider the correlation noise to be distributed independently from the realization of the side-information. This paper introduces a novel model, of which the standard deviation depends spatially on the realization of the side-information. The performance penalty in video coding caused by side-information-independency assumptions is theoretical quantified and experimentally confirmed. Furthermore, inspired by the spatial side-information-dependency of the proposed model, a novel approach for estimating the correlation channel from the partial knowledge of it is introduced. The proposed technique is incorporated into a spatial-domain unidirectional DVC system, providing state-of-the-art performance.\",\n \"Title: A low-complexity UEP methodology demonstrated on a turbo-encoded wavelet image satellite downlink\\nYear: 2008\\n\\nAbstract: Realizing high-quality digital image transmission via a satellite link, while optimizing resource distribution and minimizing battery consumption, is a challenging task. This paper describes a methodology to optimize a turbo-encoded wavelet-based satellite down-link progressive image transmission system with unequal error protection (UEP) techniques. To achieve that goal, we instantiate a generic UEP methodology onto the system, and demonstrate that the proposed solution has little impact on the average performance, while greatly reducing the run-time complexity. Based on a simple design-time distortion model and a low-complexity run-time algorithm, the provided solution can dynamically tune the system's configuration to any bitrate constraint or channel condition. The resulting system outperforms in terms of peak signal-to-noise ratio (PSNR), a state-of-the-art, fine-tuned equal error protection (EEP) solution by as much as 2 dB.\",\n \"Title: Applying Open-Loop Coding in Predictive Coding Systems\\nYear: 2008\\n\\nAbstract: This paper investigates the application of open-loop coding principles in predictive coding systems. In order to cope with the drift, which is inherent in open-loop predictive coding, a novel rate-distortion (R-D) model is proposed, capturing the propagation of quantization errors in such systems. Additionally, a novel intra-frame video codec employing the transform and spatial prediction modes from H.264 is proposed. The results obtained with the proposed codec show that allocating rate based on the proposed R-D model provides gains of up to 1.9 dB compared to a straightforward rate allocation not accounting for drift. Furthermore, the proposed open-loop predictive codec provides gains of up to 2.3 dB compared to an equivalent closed-loop intra-frame video codec using the transform, prediction modes and rate-allocation from H.264. One concludes that the considered open-loop predictive coding paradigm retains the advantages of open-loop coding, and offers the possibility of further improving the compression performance in predictive coding systems.\",\n \"Title: Feature point based image watermarking with insertions, deletions, and substitution codes\\nYear: 2010\\n\\nAbstract: In this paper we concentrate on robust image watermarking (i.e. capable of resisting common signal processing operations and intentional attacks to destroy the watermark) based on image features. Kutter et al.(7) motivated that well chosen image features survive admissible image distortions and hence can benefit the watermarking process. These image features are used as location references for the region in which the watermark is embedded. To realize the latter, we make use of previous work(16) where a ring-shaped region, centered around an image feature is determined for watermark embedding. We propose to choose a specific sequence of image features according to strict criteria so that the image features have large distance to other chosen image features so that the ring shaped embedding regions do not overlap. Nevertheless, such a setup remains prone to insertion, deletion and substitution errors. Therefore we applied a two-step coding scheme similar to the one employed by Coumou and Sharma(4) for speech watermarking. Our contribution here lies in extending Coumou and Sharma's one dimensional scheme to the two dimensional setup that is associated with our watermarking technique. The two-step coding scheme concatenates an outer Reed-Solomon error-correction code with an inner, blind, synchronization mechanism.\",\n \"Title: Encoder-driven rate control and mode decision for distributed video coding.\\nYear: 2013\\n\\nAbstract: To provide low-complexity encoding for video in unidirectional or offline compression scenarios, this paper proposes an efficient feedback-channel-free distributed video coding architecture featuring a novel encoder-driven rate control scheme in tandem with a designated mode selection process. To this end, the encoder features a novel low-complexity motion estimation technique to approximate the side-information (SI) available at the decoder. Then, a SI-dependent correlation channel estimation between the approximated SI and the original frames is used to derive the theoretically required rate for successful Slepian-Wolf (SW) decoding. Based on the evaluation of the expected trade-off between the estimated required coding rate and the estimated distortion outcome, a novel encoder-side mode decision module assigns a different coding mode to distinct portions of the coded frames. In this context, skip, intra and SW coding modes are supported. To reduce the effect of underestimation, the final SW rate is adjusted upwards using a novel rate formula. Additionally, a successive SI refinement technique is exploited at the decoder to decrease the number of SW decoding failures. Experimental results illustrate the benefit of the different coding options and show similar or superior compression performance with respect to the feedback-based DISCOVER benchmark system. Finally, the low-complexity encoding characteristics of the proposed system are confirmed, as well as the beneficial impact of the proposed scheme on the decoding complexity.\",\n \"Title: On the side-information dependency of the temporal correlation in Wyner-Ziv video coding\\nYear: 2009\\n\\nAbstract: Current models in Wyner-Ziv video coding consider the temporal correlation noise to be side-information independent (SII). This paper goes beyond this assumption and proposes a novel model, of which the parameters are side-information dependent (SID). The proposed model is experimentally validated showing remarkable accuracy improvement over the conventional SII model. Moreover, a novel SID technique for the accurate estimation of the correlation channel in video is introduced. The proposed technique enables the design of a novel pixel-domain Wyner-Ziv video coding system operating without a feedback channel. Preliminary experimental results show that the proposed codec achieves superior performance compared to the state-of-the-art in pixel-domain Wyner-Ziv coding.\",\n \"Title: Theoretical distortion estimation in L-infinite wavelet-based coding of semi-regular meshes.\\nYear: 2012\\n\\nAbstract: Subdivision-based wavelet coding techniques yield state-of-the-art performance in scalable compression of semi-regular meshes. However, all these codecs make use of the L-2 distortion metric, which gives only a good approximation of the global error produced by lossy coding of the wavelet coefficients. The L-infinite metric has been proven to be a suitable metric for applications where controlling the local, maximum error on each vertex is of critical importance. In this context, an upper bound formulation for the L-infinite distortion for a wavelet-based coding scheme operating on semi-regular meshes is derived. In addition, we propose a rate-distortion optimization algorithm that minimizes the rate for any target L-infinite distortion. It is shown that our L-infinite coding system outperforms the state-of-the-art and that an L-2 driven coding approach for semi-regular meshes loses ground to its L-infinite driven version when the goal is to have a tight control on the local reconstruction error.\",\n \"Title: Fully scalable intraband coding of wavelet-decomposed 3D meshes\\nYear: 2009\\n\\nAbstract: In this paper, a scalable intraband wavelet coding technique for semi-regular topology surface meshes is proposed. The choice of an intraband codec design is justified through an information theoretic analysis of the statistical dependencies between the wavelet coefficients resulting from a wavelet-based decomposition of the input 3D mesh. The bitstream produced by the proposed codec is both resolution and quality scalable. This lies in contrast with well-known zero-tree based interband coding techniques that only support quality scalability. The experimental results show that, on average, the proposed codec outperforms the state-of-the-art zero-tree based interband codec in rate-distortion sense when using the L2 norm as target distortion metric.\",\n \"Title: Performing deblocking in video coding based on spatial-domain motion-compensated temporal filtering\\nYear: 2006\\n\\nAbstract: Employing block-based motion models in scalable video coding based on spatial-domain motion-compensated temporal filtering (SDMCTF) introduces artificial block-boundary discontinuities that adversely affect the video compression performance, particularly at low bit-rates. This paper focuses on the problem of deblocking in the context of SDMCTF-based video coding. One possible solution to this problem is the use of overlapped-block motion compensation (OBMC). An alternative solution is applying an adaptive deblocking filter, similar to the one used in H.264. With this respect, a novel adaptive deblocking filter, tailored to SDMCTF video coding is proposed. In terms of visual-quality, both approaches yield similar performance. However, adaptive deblocking is less complex than OBMC, as it requires up to 34% less processing time. Experimental results show that the two techniques significantly improve the subjective and objective quality of the decoded sequences, confirming the expected benefits brought by deblocking in SDMCTF-based video coding.\",\n \"Title: A Low-Complexity Methodology For Unequal Error Protection Of Scalable Images\\nYear: 2006\\n\\nAbstract: Unequal Error Protection (UEP) is the most practical way to achieve joint source-channel coding while keeping a clean separation between the source and channel coders. In this paper, a generic UEP allocation methodology is proposed. The methodology is based on a run-time algorithm, exploiting design-time models, and is applicable to bit-level transmission scenarios as well as packet-based transmission environments. It works with source-independent modeling and a linear complexity with respect to the number of substreams in the source coded stream and the number of protection levels available, which makes it an attractive solution for practical scenarios. We obtain a substantial complexity reduction compared to a classical Lagrangian optimization while keeping optimality in the allocation of the protection levels. Given a typical source and channel granularity, we obtain a factor 6 complexity reduction.\",\n \"Title: Robust Motion Vector Coding And Error Concealment In Mctf-Based Video Coding\\nYear: 2005\\n\\nAbstract: Error resilience is of paramount importance in video transmission over variable-bandwidth error-prone channels, such as wireless channels. In this paper, we investigate the influence of corrupted motion vectors in video coding based on motion compensated temporal filtering, and develop various error resilience and concealment mechanisms for this class of codecs. The experimental results show that our proposed motion vector coding technique significantly increases the robustness against transmission errors at the cost of less than 3% in terms of rate. It is also shown that our proposed spatial error-concealment mechanism leads to performance gains of up to 6 dB in comparison to a classical slicing-based approach employing no error concealment.\",\n \"Title: Scalable Video Coding Based On Motion-Compensated Temporal Filtering: Complexity And Functionality Analysis\\nYear: 2004\\n\\nAbstract: Video coding techniques yielding state-of-the-art compression performance require large amount of computational resources, hence practical implementations, which target a broad market, often tend to trade-off coding efficiency and flexibility for reduced complexity. Scalable video coding instead, not only provides seamless adaptation to bit-rate variation, but also allows the end user to trim down the resources lie needs to perform real-time decoding by limiting the process to a subset of the original content. Hence, by choosing the quality, frame-rate and/or resolution of the reconstructed sequence, each decoder can meet its hardware limitations without affecting, the encoding process of the media provider. This paper proposes a preliminary analysis of the memory-access behavior of a fully scalable video decoder and investigates the capability of selecting the operational settings in order to adapt to the available hardware resources on the target device.\",\n \"Title: Error-Resilient Video Coding Using Motion Compensated Temporal Filtering And Embedded Multiple Description Scalar Quantizers\\nYear: 2005\\n\\nAbstract: Real time delivery of video over best-effort networks requires compression systems that dynamically adapt the rate to the available channel capacity and exhibit robustness to data loss as retransmission is often impractical. Error-resilience, however, significantly lowers the coding performance when rigid design is performed based on a worst-case scenario. This paper presents an original scalable video coding scheme that couples the compression efficiency of the open-loop architecture with the robustness of multiple description source coding. The use of embedded multiple description quantization and a novel channel-aware rate-allocation allow for shaping on-the-fly the output bit-rate and the degree of resilience. As a result, robustness to data losses is traded for better visual quality when transmission occurs over reliable channels, while error-resilience is introduced when noisy links are involved. The advantage of our proposal is demonstrated in the context of packet-lossy networks.\",\n \"Title: Analysis of the statistical dependencies in the curvelet domain and applications in image compression\\nYear: 2007\\n\\nAbstract: This paper reports an information-theoretic analysis of the dependencies that exist between curvelet coefficients. We show that strong dependencies exist in local intra-band micro-neighborhoods, and that the shape of these neighborhoods is highly anisotropic. With this respect, it is found that the two immediately adjacent neighbors that lie in a direction orthogonal to the orientation of the subband convey the most information about the coefficient. Moreover, taking into account a larger local neighborhood set than this brings only mild gains with respect to intra-band mutual information estimations. Furthermore, we point out that linear predictors do not represent sufficient statistics, if applied to the entire intra-band neighborhood of a coefficient. We conclude that intra-band dependencies are clearly the strongest, followed by their inter-orientation and inter-scale counterparts; in this respect, the more complex intra-band/inter-scale or intra-band/inter-orientation models bring only mild improvements over intra-band models. Finally, we exploit the coefficient dependencies in a curvelet-based image coding application and show that the scheme is comparable and in some cases even outperforms JPEG2000.\",\n \"Title: Maximum likelihood motion compensation for distributed video coding\\nYear: 2012\\n\\nAbstract: Aspiring to provide robust low-complexity encoding for video, this work presents a hash-based transform domain distributed video codec, featuring a novel maximum likelihood motion compensation technique to generate high quality side information at the decoder. A simple hash is employed to perform overlapped block motion estimation at the decoder, which produces a set of temporal predictors on a pixel basis. For every pixel position, maximum likelihood motion compensation, based on an online estimate of the conditional dependencies between the temporal predictors and the original frame, combines the cluster of temporal predictors into a single value to serve as decoder-generated side information. Efficient low-density parity-check accumulate channel codes refine the side information in the transform domain. Experimental results demonstrate that the proposed system advances over our previous hash-based distributed video coding architectures, delivering state-of-the-art distributed coding performance, in particular for sequences organized in large groups of pictures or containing highly irregular motion. Notwithstanding the presence of a hash, the presented distributed video codec successfully maintains low encoder complexity.\",\n \"Title: Spatial-domain unidirectional DVC with side-information dependent correlation channel estimation\\nYear: 2009\\n\\nAbstract: Present distributed video coding (DVC) approaches assume the correlation channel to be side-information independent (SII). We have recently demonstrated that in reality the correlation channel in video is side-information dependent (SID) and that DVC systems which make SII assumptions suffer a significant performance penalty, depending entirely on the correlation statistics of the video data. In this paper, we present a novel spatial domain unidirectional DVC codec which is based on a novel SID technique enabling advanced correlation channel estimation at the decoder starting from a partial knowledge of it. Experimental results show that the proposed codec achieves superior performance compared to the state-of-the-art in Wyner-Ziv video coding while operating without a feedback channel.\",\n \"Title: Robust Image Content Authentication with Tamper Location\\nYear: 2012\\n\\nAbstract: We propose a novel image authentication system by combining perceptual hashing and robust watermarking. An image is divided into blocks. Each block is represented by a compact hash value. The hash value is embedded in the block. The authenticity of the image can be verified by re-computing hash values and comparing them with the ones extracted from the image. The system can tolerate a wide range of incidental distortion, and locate tampered areas as small as 1/64 of an image. In order to have minimal interference, we design both the hash and the watermark algorithms in the wavelet domain. The hash is formed by the sign bits of wavelet coefficients. The lattice-based QIM watermarking algorithm ensures a high payload while maintaining the image quality. Extensive experiments confirm the good performance of the proposal, and show that our proposal significantly outperforms a state-of-the-art algorithm.\",\n \"Title: Combined wavelet-domain and motion-compensated video denoising based on video codec motion estimation methods\\nYear: 2009\\n\\nAbstract: Integrating video coding and denoising is a novel processing paradigm, bringing mutual benefits to both video processing tools. In this paper, we propose a novel video denoising approach of which the main idea is reusing motion estimation resources from the video coding module for video denoising. In most cases, the motion fields produced by real-time video codecs cannot be directly employed in video denoising, since they, as opposed to noise filters, tolerate errors in the motion field. In order to solve this problem, we propose a novel motion-field filtering step that refines the accuracy of the motion estimates to a degree that is required for denoising. Additionally, a novel temporal filter is proposed that is robust against errors in the estimated motion field. Numerical results demonstrate that the proposed denoising scheme is of low-complexity and compares favorably to the state-of-the-art video denoising methods.\",\n \"Title: Progressively refined wyner-ziv video coding for visual sensors\\nYear: 2014\\n\\nAbstract: Wyner-Ziv video coding constitutes an alluring paradigm for visual sensor networks, offering efficient video compression with low complexity encoding characteristics. This work presents a novel hash-driven Wyner-Ziv video coding architecture for visual sensors, implementing the principles of successively refined Wyner-Ziv coding. To this end, so-called side-information refinement levels are constructed for a number of grouped frequency bands of the discrete cosine transform. The proposed codec creates side-information by means of an original overlapped block motion estimation and pixel-based multihypothesis prediction technique, specifically built around the pursued refinement strategy. The quality of the side-information generated at every refinement level is successively improved, leading to gradually enhanced Wyner-Ziv coding performance. Additionally, this work explores several temporal prediction structures, including a new hierarchical unidirectional prediction structure, providing both temporal scalability and low delay coding. Experimental results include a thorough evaluation of our novel Wyner-Ziv codec, assessing the impact of the proposed successive refinement scheme and the supported temporal prediction structures for a wide range of hash configurations and group of pictures sizes. The results report significant compression gains with respect to benchmark systems in Wyner-Ziv video coding (e.g., up to 42.03% over DISCOVER) as well as versus alternative state-of-the-art schemes refining the side-information.\",\n \"Title: Wavelet based volumetric medical image compression.\\nYear: 2015\\n\\nAbstract: The amount of image data generated each day in health care is ever increasing, especially in combination with the improved scanning resolutions and the importance of volumetric image data sets. Handling these images raises the requirement for efficient compression, archival and transmission techniques. Currently, JPEG 2000׳s core coding system, defined in Part 1, is the default choice for medical images as it is the DICOM-supported compression technique offering the best available performance for this type of data. Yet, JPEG 2000 provides many options that allow for further improving compression performance for which DICOM offers no guidelines. Moreover, over the last years, various studies seem to indicate that performance improvements in wavelet-based image coding are possible when employing directional transforms. In this paper, we thoroughly investigate techniques allowing for improving the performance of JPEG 2000 for volumetric medical image compression. For this purpose, we make use of a newly developed generic codec framework that supports JPEG 2000 with its volumetric extension (JP3D), various directional wavelet transforms as well as a generic intra-band prediction mode. A thorough objective investigation of the performance-complexity trade-offs offered by these techniques on medical data is carried out. Moreover, we provide a comparison of the presented techniques to H.265/MPEG-H HEVC, which is currently the most state-of-the-art video codec available. Additionally, we present results of a first time study on the subjective visual performance when using the aforementioned techniques. This enables us to provide a set of guidelines and settings on how to optimally compress medical volumetric images at an acceptable complexity level.\",\n \"Title: Scalable multiple-description image coding based on embedded quantization\\nYear: 2007\\n\\nAbstract: Scalable multiple-description (MD) coding allows for fine-grain rate adaptation as well as robust coding of the input source. In this paper, we present a new approach for scalable MD coding of images, which couples the multiresolution nature of the wavelet transform with the robustness and scalability features provided by embedded multiple-description scalar quantization (EMDSQ). Two coding systems are proposed that rely on quadtree coding to compress the side descriptions produced by EMDSQ. The proposed systems are capable of dynamically adapting the bitrate to the available bandwidth while providing robustness to data losses. Experiments performed under different simulated network conditions demonstrate the effectiveness of the proposed scalable MD approach for image streaming over error-prone channels.\",\n \"Title: Proceedings of the second \\\"international Traveling Workshop on Interactions between Sparse models and Technology\\\" (iTWIST'14).\\nYear: 2014\\n\\nAbstract: The implicit objective of the biennial \\\"international - Traveling Workshop on Interactions between Sparse models and Technology\\\" (iTWIST) is to foster collaboration between international scientific teams by disseminating ideas through both specific oral/poster presentations and free discussions. For its second edition, the iTWIST workshop took place in the medieval and picturesque town of Namur in Belgium, from Wednesday August 27th till Friday August 29th, 2014. The workshop was conveniently located in \\\"The Arsenal\\\" building within walking distance of both hotels and town center. iTWIST'14 has gathered about 70 international participants and has featured 9 invited talks, 10 oral presentations, and 14 posters on the following themes, all related to the theory, application and generalization of the \\\"sparsity paradigm\\\": Sparsity-driven data sensing and processing; Union of low dimensional subspaces; Beyond linear and convex inverse problem; Matrix/manifold/graph sensing/processing; Blind inverse problems and dictionary learning; Sparsity and computational neuroscience; Information theory, geometry and randomness; Complexity/accuracy tradeoffs in numerical methods; Sparsity? What's next?; Sparse machine learning and inference.\",\n \"Title: Heterogeneous Acceleration of Volumetric JPEG 2000\\nYear: 2015\\n\\nAbstract: We present the implementation of a volumetric JPEG 2000 codec as a real-world use case of software acceleration with GPUs and multi-core CPUs. We present a generic methodology to accelerate existing code written in C with OpenCL. Furthermore, we account for the volumetric nature of the processed data and formulate associated optimization guidelines. The resulting software can exploit different accelerator types - GPUs and multi-core CPUs - and delivers a decent speedup on a variety of hardware platforms for a relatively small effort.\",\n \"Title: Lossy-to-lossless screen content coding using an HEVC base-layer\\nYear: 2013\\n\\nAbstract: This paper presents a novel High Efficiency Video Coding (HEVC) based screen content coding scheme providing lossy-to-lossless compression support. The paper follows a two-layer design approach which allows for backward compatibility with the conventional HEVC standard for the base-layer and yields lossless compression when decoding the enhancement-layer. The system employs Golomb based binarization and context-based arithmetic coding to efficiently encode the enhancement-layer. Performance evaluations using 4:4:4 screen content sequences are carried out. The results show that the proposed system substantially outperforms lossless intra- HEVC and single-layer JPEG-LS.\",\n \"Title: Efficient intra-frame video coding for low resolution wireless visual sensors\\nYear: 2013\\n\\nAbstract: This paper proposes a low-complexity intra-frame video coding system for very low resolution sequences. The application focuses on power constrained wireless visual sensors, recording extremely low-resolution video, namely, 30×30 pixels. In the proposed framework, high-performance video coding techniques such as block-based transforms, intra prediction and entropy coding are employed. The proposed system features several different configurations, all of which are evaluated both in terms of coding efficiency and computational complexity. Based on the experimental results, the best configuration is identified. Experimental results show that the proposed configuration achieves better coding performance than the H.264/AVC Intra frame codec, at only 23.7% of its encoding complexity.\",\n \"Title: Image security tools for JPEG standards\\nYear: 2014\\n\\nAbstract: Image privacy and security are receiving increasingly more attention these days though their impact is still underestimated. Every day billions of digital pictures are generated and distributed via social media, news sites and photo sharing applications. These images tend to carry quite some information that can be regarded to be private for the photographer and/or the pictured individuals. Furthermore, this information is not solely constrained to the visible information carried by the picture. For instance, the sensor noise that is embedded in the picture provides information related to the camera used to take the photo. Besides the picture essence also metadata is signaled in the image file containers, providing geospatial information, camera settings, photographer related information, etc. Hence, it is evident to understand that to protect the privacy of the persons that can be associated with the picture, the necessary security precautions need to be taken. Since many pictures are represented in JPEG image formats, it is evident that the JPEG committee is also investigating how it can integrate better support for privacy and security requirements in its suite of standards. In this presentation, an overview will be given of standardized solutions, where we will particularly focus on JPEG 2000 JPSEC framework, but will also discuss how the committee plans to further improve the support for these requirements in the context of the legacy JPEG standard.\",\n \"Title: Performance optimizations for PatchMatch-based pixel-level multiview inpainting\\nYear: 2013\\n\\nAbstract: As 3D content is becoming ubiquitous in today's media landscape, there is a rising interest for 3D displays that do not demand wearing special headgear in order to experience the 3D effect. Autostereoscopic displays realize this by providing multiple different views of the same scene. It is however unfeasible to record, store or transmit the amount of data that such displays require. Therefore there is a strong need for real-time solutions that can generate multiple extra viewpoints from a limited set of originally recorded views. The main difficulty in current solutions is that the synthesized views contain disocclusion holes where the pixel values are unknown. In order to seamlessly fill-in these holes, inpainting techniques are being used. In this work we consider a depth-based pixel-level inpainting system for multiview video. The employed technique operates in a multi-scale fashion, fills in the disocclusion holes on a pixel-per-pixel basis and computes approximate Nearest Neighbor Fields (NNF) to identify pixel correspondences. To this end, we employ a multi-scale variation on the well-known PatchMatch algorithm followed by a refinement step to escape from local minima in the matching-cost function. In this paper we analyze the performance of different cost functions and search methods within our existing inpainting framework.\",\n \"Title: Interactive demonstrations of the locally adaptive fusion for combining objective quality measures\\nYear: 2014\\n\\nAbstract: To automate quality monitoring of multimedia applications, objective quality measures for images and video content need to be designed. Objective quality measures that model the Human Visual System (HVS) have a disappointing performance, because the HVS is not sufficiently understood. Integrating machine learning (ML) techniques may increase the performance. Unfortunately, traditional ML is difficult to interpret. To this end, we developed the Locally Adaptive Fusion (LAF), for more flexible and reliable quality predictions. This manuscript proposes six interactive programs and a website that demonstrate the effectiveness of LAF, which complement the technical focus of the corresponding journal paper.\",\n \"Title: Embedded multiple description scalar quantizers for progressive image transmission\\nYear: 2003\\n\\nAbstract: Robust progressive image transmission over unreliable channels with variable bandwidth requires multiple description coding (MDC) systems that produce highly error-resilient embedded bit-streams. The proposed embedded multiple description scalar quantizers (EMDSQ) meet the desired features consisting of a high redundancy level, fine grain rate adaptation and progressive transmission of each description. Experimental results show that EMDSQ yield better rate-distortion performance in comparison to the multiple description uniform scalar quantizers (MDUSQ) previously proposed in the literature. Moreover, the generalized form of EMDSQ targeting an arbitrary number of channels is proposed, which offers the possibility of designing realistic coders for practical multi-channel communication systems.\",\n \"Title: Additive Distortion Modeling for Unequal Error Protection of Scalable Multimedia Content\\nYear: 2005\\n\\nAbstract: Due to the improvement of compression technology, limited bandwidth channels can convey more data than before. However, since compressed streams are very sensitive to transmission errors, a joint optimization of the source-channel coding is required. Unequal Error Protection (UEP) is the most pragmatic way to do it, and exploits the embedded nature of bitstreams produced by scalable coders.However, depending on the number of substreams and protection levels, optimization of UEP can be extremely complex. A large simplification comes from considering the contribution to the global distortion of each substream as independent from the others. This paper explores this distortion additivity assumption for JPEG2000, and shows that though the assumption does not fully hold, no visible consequence on UEP performance is introduced, while reducing significantly the complexity of the optimization process.\",\n \"Title: Intra-frame video coding using an open-loop predictive coding approach\\nYear: 2008\\n\\nAbstract: A novel coding approach, applying open-loop coding principles in predictive coding systems is proposed in this paper. The proposed approach is instantiated with an intra-frame video codec employing the transform and spatial prediction modes from H.264. Additionally, a novel rate-distortion model for open-loop predictive coding is proposed and experimentally validated. Optimally allocating rate based on the proposed model provides significant gains in comparison to a straightforward rate allocation not accounting for drift. Furthermore, the proposed open-loop predictive codec provides gains of up to 2.3 dB in comparison to an equivalent closed-loop intra-frame video codec employing the transform, prediction modes and rate-allocation from H.264. This indicates that, with appropriate drift compensation, open-loop predictive coding offers the possibility for further improving the compression performance in predictive coding systems.\",\n \"Title: Statistical L-infinite distortion estimation in scalable coding of meshes\\nYear: 2008\\n\\nAbstract: This paper investigates the novel concept of local error control in arbitrary mesh encoding, and proposes a new L-infinite mesh coding approach implementing this concept. In contrast to traditional mesh coding systems that use the mean-square error as distortion measure, the proposed approach employs the L-infinite distortion as target distortion metric. In this context, a novel wavelet-based L-infinite-constrained coding approach for meshes is proposed, which ensures that the maximum local error between the original and decoded meshes is lower than a given upper-bound. Additionally, the proposed system achieves scalability in L-infinite sense, that is, the L-infinite distortion upper-bound can be accurately estimated when decoding any layer from the input stream. Moreover, a distortion estimation approach is proposed, expressing the L-infinite distortion in the spatial domain as a statistical estimate of quantization errors produced in the wavelet domain. An instantiation of the proposed L-infinite coding approach is demonstrated for MESHGRID, which is a scalable 3D object coding system, part of MPEG-4 AFX. The proposed L-infinite coding approach guarantees that the maximum error is upper-bounded, it enables a fast real-time implementation of the rate-allocation, and it preserves all the scalability features and animation capabilities of the employed scalable mesh codec.\",\n \"Title: Segmentation-based intra coding of depth maps using texture information\\nYear: 2013\\n\\nAbstract: Emerging video technologies, such as 3DTV, increase the demand for high efficiency coding tools. To allow the transmission of a multiview video over the employed transmission channels, a new data format including texture and the corresponding depth information has been proposed. In this paper we present a novel segmentation-based intra codec for depth maps, utilizing the correlation between depth and color data. Our system performs a low-complexity color segmentation on the texture image and encodes the depth values within the segments. The same segmentation is calculated at the decoder, ensuring the closed-loop property of the system. Experimental results show that the proposed system is a viable alternative for 3D-HEVC in low-complexity applications.\",\n \"Title: Real-time texture sampling and reconstruction with wavelet filters\\nYear: 2013\\n\\nAbstract: Currently, the use of the 2D wavelet transform in texture compression for real-time texture mapping on the GPU is limited. The main cause of this is the lack of real-time texture filtering implementations which do not require specialized hardware. This work proposes a novel system to perform 2D wavelet reconstruction and bilinear texture filtering using a high performance GPU shader. The system is able to generate a performant GLSL shader for arbitrary wavelet filter configurations. This goes beyond earlier works in the literature proposing Haar wavelet and Discrete Cosine Transform (DCT) implementations on the GPU. We analyse the shader performance and run-time complexity for several wavelet filters. The experimental results show that filters longer than Haar are deployable on the GPU while maintaining accurate texture filtering and real-time performance.\",\n \"Title: Forensic data hiding optimized for JPEG 2000\\nYear: 2011\\n\\nAbstract: This paper presents a novel image adaptive data hiding system using properties of the discrete wavelet transform and which is ready to use in combination with JPEG 2000. Image adaptive watermarking schemes determine the embedding samples and strength from the image statistics. We propose to use the energy of wavelet coefficients at high frequencies to measure the amount of distortion that can be tolerated by a lower frequency coefficient. The watermark decoder in image adaptive data hiding needs to estimate the same parameters used for encoding from a modified source and hence is vulnerable to desynchronization. We present a novel way to resolve these synchronization issues by employing specialized insertion, deletion and substitution codes. Given the low complexity and reduced perceptual impact of the embedding technique, it is suitable for inserting camera and/or projector information to facilitate image forensics.\",\n \"Title: Optimized directional lifting with reduced complexity\\nYear: 2008\\n\\nAbstract: ABSTRACT The direction-adaptive discrete wavelet,transform,(DADWT) proves to be a very competitive alternative in scalable wavelet-based compression of images, yielding impressive compression perfor- mance gains in comparison to the classical DWT. A major limitation of DADWT though stems from its complexity, requiring an exhaus- tive search for the optimum,prediction direction to be employed in the directional lifting process. This paper proposes a novel al- gorithm to lower the complexity of the DADWT by predicting the optimal prediction direction using a gradient-based technique. The algorithm is developed,based on a mathematical,model of the pre- diction errors generated via directional lifting of an input wedge image. The proposed approach avoids a time-consuming exhaustive search and yet the prediction step remains very simple and fast. It is shown that the proposed algorithm brings a complexity-reduction factor of 11=4 for almost no penalty in the prediction accuracy.\",\n \"Title: The evolution of multiple description coding based on scalar quantization\\nYear: 2007\\n\\nAbstract: Multiple Description Coding (MDC) allows for error resilient coding of data. This tutorial paper discusses one of the most popular MDC technologies, namely MDC based on scalar quantization, and its application in scalable coding of multimedia. This concept is illustrated by a fully scalable video coder, based on motion compensated temporal filtering, which incorporates the MDC principle for the encoding of both texture and motion vector data. For such system, we derive the principles of channel-aware MDC rate-allocation, which enables adjusting both the bit-rate and the resilience to errors of the output stream. Such adaptation is performed on-the-fly depending on the channel condition. Experimental results prove the advantages of MDC and the effectiveness of MDC system when transmitting video streams over unreliable networks.\",\n \"Title: A novel MPI reduction algorithm resilient to imbalances in process arrival times.\\nYear: 2016\\n\\nAbstract: Reduction algorithms are optimized only under the assumption that all processes commence the reduction simultaneously. Research on process arrival times has shown that this is rarely the case. Thus, all benchmarking methodologies that take into account only balanced arrival times might not portray a true picture of real-world algorithm performance. In this paper, we select a subset of four reduction algorithms frequently used by library implementations and evaluate their performance for both balanced and imbalanced process arrival times. The main contribution of this paper is a novel imbalance robust algorithm that uses pre-knowledge of process arrival times to construct reduction schedules. The performance of selected algorithms was empirically evaluated on a 128 node subset of the Partnership for Advanced Computing in Europe CURIE supercomputer. The reported results show that the new imbalance robust algorithm universally outperforms all the selected algorithms, whenever the reduction schedule is precomputed. We find that when the cost of schedule construction is included in the total runtime, the new algorithm outperforms the selected algorithms for problem sizes greater than 1 MiB.\",\n \"Title: Globally optimized multiview video color correction using dense spatio-temporal matching\\nYear: 2015\\n\\nAbstract: Multiview video is becoming increasingly popular as the format for 3D video systems that use autostereoscopic displays or freeviewpoint navigation capabilities. However, the algorithms that drive these applications are not yet mature and can suffer from subtle irregularities such as color imbalances inbetween different cameras. Regarding the problem of color correction, state-of-the-art methods directly apply some form of histogram matching in blocks of pixels between an input frame and a target frame containing the desired color distribution. These methods, however, typically suffer from artifacts in the gradient domain, as they do not take into account local texture information. This paper presents a novel method to correct color differences in multiview video sequences that uses a dense matching-based global optimization framework. The proposed energy function ensures preservation of local structures by regulating deviations from the original image gradients.\",\n \"Title: Continuous ultrasound speckle tracking with Gaussian mixtures\\nYear: 2015\\n\\nAbstract: Speckle tracking echocardiography (STE) is now widely used for measuring strain, deformations, and motion in cardiology. STE involves three successive steps: acquisition of individual frames, speckle detection, and image registration using speckles as landmarks. This work proposes to avoid explicit detection and registration by representing dynamic ultrasound images as sparse collections of moving Gaussian elements in the continuous joint space-time space. Individual speckles or local clusters of speckles are approximated by a single multivariate Gaussian kernel with associated linear trajectory over a short time span. A hierarchical tree-structured model is fitted to sampled input data such that predicted image estimates can be retrieved by regression after reconstruction, allowing a (bias-variance) trade-off between model complexity and image resolution. The inverse image reconstruction problem is solved with an online Bayesian statistical estimation algorithm. Experiments on clinical data could estimate subtle sub-pixel accurate motion that is difficult to capture with frame-to-frame elastic image registration techniques.\",\n \"Title: Digital painting analysis, at the cross section of engineering, mathematics and culture\\nYear: 2011\\n\\nAbstract: Museums started digitizing their collections, mainly for preservation, documentation and dissemination purposes, but one can go further than archival: digital acquisition enables digital analysis of art and in particular paintings. This proves to be a useful, non invasive tool, for many applications such as restoration, conservation, art history, material and structure characterization, authentication, dating and even style analysis. Paintings are complex structures. Analysis of the support and all pictorial layers requires multimodal imaging, high resolution and standardized data acquisition. We situate our own research on canvas characterization, craquelure detection, multispectral data analysis, object extraction and image enhancement within the state of the art.\",\n \"Title: Error-resilient transmission of H.264 SVC streams over DVB-T/H and WiMAX channels with multiple description coding techniques\\nYear: 2007\\n\\nAbstract: The European STREP Project SUIT involves the development of a multiple description scalable video codec for the transmission of video content over DVB-T/H and WiMAX channels. In this paper, we elaborate on the architectural design choices for such a multiple description transmission system, which utilizes H.264 Scalable Video Coding (SVC). We follow a phased approach focusing on the following unbalanced multiple description coding approaches: fully redundant unbalanced multiple description coding, multiple description coding based on redundant slices and multiple description coding by means of embedded multiple description quantisation.\",\n \"Title: JPEG Pleno: Toward an Efficient Representation of Visual Reality.\\nYear: 2016\\n\\nAbstract: In discussing the rationale behind the vision for JPEG Pleno and how the new standardization initiative aims to reinvent the future of imaging, the authors review plenoptic representation and its underlying practical implications and challenges in implementing real-world applications with an enhanced quality of experience.\",\n \"Title: Efficient Depth-aware Image Deformation Adaptation for Curved Screen Displays.\\nYear: 2017\\n\\nAbstract: The curved screen has attracted considerable attentions in recent years, since it enables to enlarge the view angle and to enhance the immersive perception for users. However, existing curved surface projections are frequently prone to geometric distortion or loss of content. This paper presents a content-aware and depth-aware image adaptation solution for curved displays. An efficient optimization approach of image deformation is proposed to preserve local scene content and to minimize scene geometry distortion. To follow the original 3D perception of objects in different depth layers, the depth information is re-mapped for individual content scaling. Objective evaluation results reveal that our approach can effectively preserve foreground objects. We also perform a subjective evaluation of the proposed solution, and compare it to two alternative mapping methods, which are tested on different curvatures on both a traditional screen and an ad-hoc curvature-controllable curved display. Experimental results demonstrate that our approach outperforms other existing mapping methods for immersive display of rectangle images on curved screens.\",\n \"Title: Heterogeneous acceleration of volumetric JPEG 2000 using OpenCL\\nYear: 2017\\n\\nAbstract: AbstractThis paper discusses an OpenCL version of a volumetric JPEG 2000 codec that runs on GPUs, multi-core processors or a combination of both. Since the performance critical part consists of a fine-grained discrete wavelet transform and coarse-grained algorithm Tier-1, the best performance is obtained with a hybrid execution in which the discrete wavelet transform is executed on a GPU and Tier-1 on a multi-core. Using an Intel i7 multi-core in combination with a modest NVIDIA Quadro K620 GPU yields speedups greater than 10 compared with the original sequential code. The performance bottlenecks that arise on GPUs when parallelizing algorithms that are coarse-grained by nature are discussed and also the optimizations that are possible. A performance analysis reveals the inefficiencies and explains the deviations from the GPU peak performance.\",\n \"Title: JPEG Privacy and Security framework for social networking and GLAM services.\\nYear: 2017\\n\\nAbstract: Current image coding standards provide limited support for privacy and security features. An exception is the JPSEC standard, which defines security extensions in JPEG 2000 specifications (part 8). Notwithstanding this shortcoming, the JPEG committee is currently defining a new JPEG Systems standard, which envisages privacy and security support across JPEG family of standards. In this manuscript, the main philosophy of this emerging specification is outlined along with typical use cases, main requirements as well as examples of potential technological solutions. The upcoming specification guarantees backward and forward compatibility with earlier standards and legacy implementations. Finally, we illustrate the introduced framework by two applications targeting secure photo sharing on social networks and IPR management in the GLAM sector.\",\n \"Title: Bounds and Conditions for Compressive Digital Holography Using Wavelet Sparsifying Bases.\\nYear: 2017\\n\\nAbstract: Numerous experiments have been conducted with success in the field of compressive digital holography, but the theory to determine optimal measurement conditions is lagging behind. In contrast to a prior study that expects object wavefields to be sparse in the spatial domain, we investigate how the configuration of the interferometer influences the reconstruction of wavefields that are sparse in a ...\",\n \"Title: From Sparse Coding Significance to Perceptual Quality: A New Approach for Image Quality Assessment.\\nYear: 2018\\n\\nAbstract: An increasing number of image processing applications require an automated quality prediction of the visual content as perceived by humans. Since, sparse coding is suggested to be an underlying strategy of the brain's neural system, it would be logical to assume that specific tasks like quality assessment also attempt to adhere to this strategy. However, existing perceptual quality predictors, oft...\",\n \"Title: Accurate label-free 3-part leukocyte recognition with single cell lens-free imaging flow cytometry.\\nYear: 2018\\n\\nAbstract: Three-part white blood cell differentials which are key to routine blood workups are typically performed in centralized laboratories on conventional hematology analyzers operated by highly trained staff. With the trend of developing miniaturized blood analysis tool for point-of-need in order to accelerate turnaround times and move routine blood testing away from centralized facilities on the rise, our group has developed a highly miniaturized holographic imaging system for generating lens-free images of white blood cells in suspension. Analysis and classification of its output data, constitutes the final crucial step ensuring appropriate accuracy of the system. In this work, we implement reference holographic images of single white blood cells in suspension, in order to establish an accurate ground truth to increase classification accuracy. We also automate the entire workflow for analyzing the output and demonstrate clear improvement in the accuracy of the 3-part classification. High-dimensional optical and morphological features are extracted from reconstructed digital holograms of single cells using the ground-truth images and advanced machine learning algorithms are investigated and implemented to obtain 99% classification accuracy. Representative features of the three white blood cell subtypes are selected and give comparable results, with a focus on rapid cell recognition and decreased computational cost.\",\n \"Title: Unitary Transforms Using Time-Frequency Warping for Digital Holograms of Deep Scenes.\\nYear: 2018\\n\\nAbstract: With the advent of ultrahigh-resolution holographic displays, viewing macroscopic deep scenes with large viewing angles becomes a possibility. These deep holograms possess different signal properties in contrast with common applications where the scene content is assumed to lie around a planar slice. Therefore, the conventional approach of refocusing at a fixed depth is ineffective. There is a nee...\",\n \"Title: Image Reconstruction with Smoothed Mixtures of Regressions\\nYear: 2018\\n\\nAbstract: This work builds upon the kernel regression framework for solving the general image processing problem of denoising, deblurring and interpolating from scattered image samples. A competitive expectation-maximization method estimates globally all parameters of a generative image model, accounting for missing samples. One 2D footprint kernel and a local linear regression plane are estimated per data sample. Kernels can shift and their prior probabilities are estimated as well, unlike in nonparametric models. Missing data yields an underdetermined problem that is regularized by smoothing the marginal mixture density. At each iteration, a balloon estimator computes numerically the spatial “territory” associated to each data samples. Results of these numerical diffusion operations are used to convolve adaptively each kernel in the forward model. Finally, the complete image is reconstructed by smoothing regression for combining conditional means of local linear regressors. Experiments apply this iterative Bayesian technique in image restoration.\",\n \"Title: A Just Noticeable Difference Subjective Test for High Dynamic Range Images\\nYear: 2018\\n\\nAbstract: High Dynamic Range (HDR) imaging captures a wide range of luminance existing in real-world scenes. Due to large luminance levels and higher brightness of HDR displays, artefacts can be more noticeable to the Human Visual System (HVS). In a first attempt to experimentally quantify those noticeable levels for HDR images, we pioneered in conducting an exhaustive and comprehensive Just Noticeable Difference (JND) subjective experiment of which the outcome is presented in this paper. Six distortions including JPEG, JPEG2000, noise, blur, contrast change, and quantization artefacts have been considered in the test. The distortions were applied to 10 HDR images in 100 distortion levels resulting a database of 6000 HDR test images. The subjects were asked to find the image JND location on each set of 100 images they had the freedom to explore. The effect of content features on the noticeable threshold selection is investigated per distortion type. Our results in some cases show a significant correlation between content features and JNDs. We are hoping that our results can contribute to further exploitation of a precise HVS model for HDR quality assessment and optimization of the coding and bit allocation in HDR compression.\",\n \"Title: Signal processing challenges for digital holographic video display systems.\\nYear: 2019\\n\\nAbstract: Holography is considered to be the ultimate display technology since it can account for all human visual cues such as stereopsis and eye focusing. Aside from hardware constraints for building holographic displays, there are still many research challenges regarding holographic signal processing that need to be tackled. In this overview, we delineate the steps needed to realize an end-to-end chain from digital content acquisition to display, involving the efficient generation, representation, coding and quality assessment of digital holograms. We discuss the current state-of-the-art and what hurdles remain to be taken to pave the way towards realistic visualization of dynamic holographic content.\",\n \"Title: Reduced-reference quality assessment of multiply-distorted images based on structural and uncertainty information degradation.\\nYear: 2018\\n\\nAbstract: •A reduced-reference image quality assessment method is proposed for quality prediction of multiply-distorted images.•The structural and uncertainty information changes are measured using shearlets and Rényi entropy analysis.•An improved quality prediction model is obtained by incorporating privileged information in the training stage.\",\n \"Title: From Adaptive Kernel Density Estimation to Sparse Mixture Models.\\nYear: 2018\\n\\nAbstract: We introduce a balloon estimator in a generalized expectation-maximization method for estimating all parameters of a Gaussian mixture model given one data sample per mixture component. Instead of limiting explicitly the model size, this regularization strategy yields low-complexity sparse models where the number of effective mixture components reduces with an increase of a smoothing probability parameter $mathbf{Pu003e0}$. This semi-parametric method bridges from non-parametric adaptive kernel density estimation (KDE) to parametric ordinary least-squares when $mathbf{P=1}$. Experiments show that simpler sparse mixture models retain the level of details present in the adaptive KDE solution.\",\n \"Title: Fingerprinting Codes Under the Weak Marking Assumption\\nYear: 2018\\n\\nAbstract: Fingerprinting codes based on the marking assumption provide reliable security against collusion attacks, but lack resilience against channel errors such as symbol erasures or other forms of distortions introduced by transmission over noisy or insecure communication channels. This additional source of distortion errors can be addressed by relaxing the marking assumption. In this paper we examine t...\",\n \"Title: Integer Fresnel Transform for Lossless Hologram Compression\\nYear: 2019\\n\\nAbstract: Digital holograms fully encode the wavefield of light, thereby having many applications both for 3D object measurements as well as for display purposes, accounting for all human visual cues. Because the statistics and properties of holographic signals differ considerably from natural imagery such as photographs, conventional coding solutions will be sub-optimal. In this paper, we propose the integer Fresnel transform, which is - to our knowledge - the first lossless transform tailored for hologram coding. By combining the proposed transform with JPEG 2000, we report bit-rate savings from 0.12 up to 2.83 bits per channel on a collection of 8 digital holograms obtained from 3 different databases.\",\n \"Title: Wave Atoms for Lossy Compression of Digital Holograms\\nYear: 2019\\n\\nAbstract: Compression of digital holograms is a major challenge that needs to be resolved to enable the efficient storage, transmission and rendering of macroscopic holographic signals. In this work, we propose to deploy the wave atom transform that has been utilized before for interferometric modalities such as acoustic and seismic signals. This non-adaptive multiresolution transform has good space-frequency localization and its orthonormal basis is suitable for sparsifying holographic signals. By replacing the CDF 9/7 wavelet transform stage in a JPEG 2000 codec with the proposed wave atom transform, we did assess its suitability for coding complex amplitude wavefronts. Experimental results demonstrate improved rate-distortion performance with respect to JPEG 2000 and H.265/HEVC for a set of computer-generated, diffuse, macroscopic holograms.\",\n \"Title: Validation of dynamic subjective quality assessment methodology for holographic coding solutions\\nYear: 2021\\n\\nAbstract: In preparation of the Call for Proposals on JPEG Pleno Holography, multiple exploration studies are ongoing to define the general procedure for performance assessment of proposed coding solutions. The performance of proposals will be compared against JPEG 2000 and H.265/HEVC intra coding, both serving as anchor codecs. In this paper, we report on the results of a dynamic subjective visual quality ...\",\n \"Title: Real-Time Computation of 3D Wireframes in Computer-Generated Holography\\nYear: 2021\\n\\nAbstract: Computer-Generated Holography (CGH) algorithms simulate numerical diffraction, being applied in particular for holographic display technology. Due to the wave-based nature of diffraction, CGH is highly computationally intensive, making it especially challenging for driving high-resolution displays in real-time. To this end, we propose a technique for efficiently calculating holograms of 3D line segments. We express the solutions analytically and devise an efficiently computable approximation suitable for massively parallel computing architectures. The algorithms are implemented on a GPU (with CUDA), and we obtain a 70-fold speedup over the reference point-wise algorithm with almost imperceptible quality loss. We report real-time frame rates for CGH of complex 3D line-drawn objects, and validate the algorithm in both a simulation environment as well as on a holographic display setup.\",\n \"Title: Bitwidth-Optimized Energy-Efficient FFT Design via Scaling Information Propagation\\nYear: 2021\\n\\nAbstract: The Fast Fourier Transform (FFT) is an efficient algorithm widely used in digital signal processing to transform between the time domain and the frequency domain. For fixed-point VLSI implementations, dynamic range growth inevitably occurs at each stage of the FFT operation. However, current methods either waste bitwidth or consume excessive resources when dealing with the dynamic range growth issue. To address this issue, we propose an efficient scaling method called Scaling Information Propagation (SIP) to alleviate the problem of dynamic range growth, which makes full use of bitwidth with much less extra area consumed than the state-of-the-art solutions. In two consecutive transform operations, the SIP method extracts scaling information and makes scaling decisions in the former transform, then executes those in the latter one. We implement the FFT's VLSI architecture in the orthogonal frequency division multiplexing (OFDM) and the holographic video compression (HVC) systems to verify the SIP method. Compared to the state-of-the-art, experimental results after VLSI synthesis show that our method achieves 9.38% energy reduction and 8.36% area savings when requiring 1.02 x 10(-7) bit error ratio (BER) of the OFDM system, and 33.47% energy reduction and 30.98% area savings when requiring 20dB signal-to-noise ratio (SNR) of the HVC system, respectively.\",\n \"Title: Unifying Structural and Semantic Similarities for Quality Assessment of DIBR-Synthesized Views\\nYear: 2022\\n\\nAbstract: Multi-view 3D content is subject to distortions during the process of depth image-based rendering (DIBR). Studies have shown the unreliable performance of the well-established image quality assessment (IQA) models for evaluation of DIBR-synthesized views which surge the need for more effective IQA methods. Existing objective methods generally rely on the pixel-wise correspondences between the reference and distorted images, while view synthesis can introduce pixel shifts. DIBR distortions such as stretching and local hole-filling errors have different visual impacts from conventional distortions, challenging the existing IQA models. Here, we developed a Full-Reference (FR) objective IQA metric for synthesized views that significantly outperforms 2D IQA and the state-of-the-art DIBR IQA approaches. While the pixel misalignment between the reference and synthesized views is a big challenge for quality assessment, we deployed a Convolutional Neural Network (CNN) model to acquire a feature representation that inherently offers resilience to the imperceptible pixel shift between the compared images. Therefore, our model does not need accurate shift compensation. We deployed a set of quality-aware CNN features representing high-order statistics, to measure the structural similarity which is combined with a semantic similarity measure for accurate quality assessment. Moreover, prediction accuracy is improved by incorporating a visual saliency model acquired using the activations of the higher CNN layers. Experimental results indicate a significant performance gain (14.6% in terms of Spearman's rank-order correlation) compared to the top existing IQA model. The source code of the proposed metric is available at: https://gitlab.com/saeedmp/sequss.\",\n \"Title: Quality evaluation of the JPEG Pleno Holography Call for Proposals response\\nYear: 2022\\n\\nAbstract: At its 93rd meeting, JPEG received the responses to the Call for Proposals on JPEG Pleno holography. This activity will lead to the first standardized coding solution for digital holograms. This paper describes the quality evaluation process, existing out of a subjective evaluation study and an objective evaluation based on preselected quality metrics. It also includes a critical evaluation of the challenges encountered when performing quality evaluation of holographic content. Based on the evaluation of the response, the JPEG committee was able to select a very reliable coding solution that depicted a performance well beyond that of the selected anchors.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":106,"string":"106"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13563,"string":"13,563"}}},{"rowIdx":13564,"cells":{"question_id":{"kind":"string","value":"Wei Yi"},"question":{"kind":"string","value":"Generate an abstract for the title \"Multi-Agent Fusion With Different Limited Fields-of-View\" using the following items: multi-agent surveillance systems, information fusion approach, Bayesian-operation InvaRiance on Difference-sets (BIRD) fusion, multi-object tracking, simulation experiments."},"answer":{"kind":"string","value":"A key objective of multi-agent surveillance systems is to monitor a much larger region than the limited field-of-view (FoV) of any individual agent by successfully exploiting cooperation among multiple agents. Whenever either a centralized or a distributed approach is pursued, this goal cannot be achieved unless an appropriately designed fusion strategy is adopted. This paper presents a novel information fusion approach by considering for each agent a known limited, and possibly different, FoV. The proposed method, named Bayesian-operation InvaRiance on Difference-sets (BIRD) fusion, relies on Generalized Covariance Intersection (GCI) and exploits a general and exact decomposition of each multi-object posterior by partitioning the global FoV, i.e. the union of the FoVs of the fusing agents, into common and exclusive FoVs. It is shown how BIRD fusion can be used to perform multi-object estimation based on random finite sets on both a centralized and a distributed peer-to-peer sensor network. Simulation experiments on realistic multi-object tracking scenarios demonstrate the effectiveness of BIRD fusion."},"correct_answer":{"kind":"string","value":"A key objective of multi-agent surveillance systems is to monitor a much larger region than the limited field-of-view (FoV) of any individual agent by successfully exploiting cooperation among multiple agents. Whenever either a centralized or a distributed approach is pursued, this goal cannot be achieved unless an appropriately designed fusion strategy is adopted. This paper presents a novel information fusion approach by considering for each agent a known limited, and possibly different, FoV. The proposed method, named Bayesian-operation InvaRiance on Difference-sets (BIRD) fusion, relies on Generalized Covariance Intersection (GCI) and exploits a general and exact decomposition of each multi-object posterior by partitioning the global FoV, i.e. the union of the FoVs of the fusing agents, into common and exclusive FoVs. It is shown how BIRD fusion can be used to perform multi-object estimation based on random finite sets on both a centralized and a distributed peer-to-peer sensor network. Simulation experiments on realistic multi-object tracking scenarios demonstrate the effectiveness of BIRD fusion."},"documents":{"kind":"list like","value":["Title: Design and Implementation of the Accumulating Algorithm of Digital Signal Averager\nYear: 2011\n\nAbstract: To satisfy the need of on-board accumulation in time-of-fLight mass spectrometer (TOF-MS) appLications, the paper presents an accumulation algorithm in FPGA based on PipeLine and ping-pong operation. The accumulation algorithm is verified on a signal averager hardware platform which takes XILiNX VIRTEX-5 FPGA as the processing core. The caChe module, on-chip accumulation, and USB transmission module is implemented. Testing results show that the range of time measurement is 32us, the minimum time resolution is 333ps and signal to noise ratio (SNR) is improved effectively. The accumulation algorithm is stable and reLiable, so it can be widely used in the measurement of high-resolution time interval and pulse ampLitude.","Title: Greedy Neighborhood Search for Disaster Relief and Evacuation Logistics\nYear: 2008\n\nAbstract: Coordinating relief activities and evacuation in emergency response involves delivering commodities to distribution centers in affected areas and picking up wounded people and transferring them to medical centers. This problem's key features are limited supply and vehicle availability, simultaneous delivery and pickup of commodities and wounded people, and split delivery. The resulting logistics problem is quite complicated. Furthermore, emergency settings are dynamic, and changing the logistics plan at each information update requires an efficient algorithm. A proposed greedy constructive algorithm explores a limited neighborhood around the vehicles' current locations in the partial solution and then appends a two-stop partial itinerary to each vehicle's available route. The algorithm's performance was tested on several networks and compared with that of optimal solutions.","Title: An Efficient Multi-Frame Track-Before-Detect Algorithm for Multi-Target Tracking\nYear: 2013\n\nAbstract: This paper considers the multi-target tracking (MTT) problem through the use of dynamic programming based track-before-detect (DP-TBD) methods. The usual solution of this problem is to adopt a multi-target state, which is the concatenation of individual target states, then search the estimate in the expanded multi-target state space. However, this solution involves a high-dimensional joint maximization which is computationally intractable for most realistic problems. Additionally, the dimension of the multi-target state has to be determined before implementing the DP search. This is problematic when the number of targets is unknown. We make two contributions towards addressing these problems. Firstly, by factorizing the joint posterior density using the structure of MTT, an efficient DP-TBD algorithm is developed to approximately solve the joint maximization in a fast but accurate manner. Secondly, we propose a novel detection procedure such that the dimension of the multi-target state no longer needs be to pre-determined before the DP search. Our analysis indicates that the proposed algorithm could achieve a computational complexity which is almost linear to the number of processed frames and independent of the number of targets. Simulation results show that this algorithm can accurately estimate the number of targets and reliably track multiple targets even when targets are in proximity.","Title: Kernel Fusion: An Effective Method for Better Power Efficiency on Multithreaded GPU.\nYear: 2010\n\nAbstract: As one of the most popular accelerators, Graphics Processing Unit (GPU) has demonstrated high computing power in several application fields. On the other hand, GPU also produces high power consumption and has been one of the most largest power consumers in desktop and supercomputer systems. However, software power optimization method targeted for GPU has not been well studied. In this work, we propose kernel fusion method to reduce energy consumption and improve power efficiency on GPU architecture. Through fusing two or more independent kernels, kernel fusion method achieves higher utilization and much more balanced demand for hardware resources, which provides much more potential for power optimization, such as dynamic voltage and frequency scaling (DVFS). Basing on the CUDA programming model, this paper also gives several different fusion methods targeted for different situations. In order to make judicious fusion strategy, we deduce the process of fusing multiple independent kernels as a dynamic programming problem, which could be well solved with many existing tools and be simply embedded into compiler or runtime system. To reduce the overhead introduced by kernel fusion, we also propose effective method to reduce the usage of shared memory and coordinate the thread space of the kernels to be fused. Detailed experimental evaluation validates that the proposed kernel fusion method could reduce energy consumption without performance loss for several typical kernels.","Title: Target tracking for an unknown and time-varying number of targets via particle filtering\nYear: 2012\n\nAbstract: Track initialization and adapting to time-varying number of targets are challenges for particle filter (PF) based tracking methods, because of the difficulty of monitoring changes in the entire state space with a finite number of particles. Conversely, if the state space only has limited supports (i.e. after thresholding) or state space is well localized (i.e. target prior information is known), PFs can provide remarkable estimation performance via concentrating its computational power on the areas of interest. In this paper, we present a PF to track unknown and time-varying number of targets based on the track-before-detect (TBD) framework. This PF adopts the two-layer structure of the PF in [5] (i.e. one layer for track initiation, the other for track maintenance), but with different track initiation and maintenance importance densities. It initiates new tracks and controls the false alarms based on the information provided by a maximum a posterior (MAP) method. By discretizing the state space, this MAP-based method can successfully localize the posterior density (i.e. find out the parts of the state space which are of interest) with reasonable computational cost. With regard to the track maintenance, it efficiently updates the existing tracks using a joint, measurement-directed sampling method and deletes the false tracks based on the likelihood ratio test (LRT). Numerical examples are also used to illustrate the efficacy of the proposed algorithm.","Title: MPLS-based Multicast Shared Trees\nYear: 2006\n\nAbstract: This paper presents a study of our proposed architecture for the setup of a MultiPoint-to-MultiPoint (MP2MP) Label Switched Path (LSP). This form of LSP is needed for establishing uni-directional multicast shared trees. Such trees are required for information distribution in applications such as video conferencing. The presented architecture is intended for multicast applications within a single autonomous domain and can be extended to cover inter-domain multicast sessions. We propose the use of one (or more) control points in the network called Rendez-vous Points (RP) in a simple extention of the PIM-SM protocol to implement multicast in MPLS networks. This architecture has the advantage of using existing MPLS techniques and existing routing protocols and requires only the addition of more management capabilities at the RPs. The experiments we carried out show that while retaining the advantages of using MPLS over traditional multicast routing, the performance of the new architecture is comparable to that of IP multicast in terms of the volume of control messages and label and memory consumption. Also the architecture scales well with the increase of the number of senders within a multicast group and with the increase of the number of multicast groups.","Title: Thresholding Process Based Dynamic Programming Track-Before-Detect Algorithm\nYear: 2013\n\nAbstract: Dynamic Programming (DP) based Track-Before-Detect (TBD) algorithm is effective in detecting low signal-to-noise ratio (SNR) targets. However, its complexity increases exponentially as the dimension of the target state space increases, so the exact implementation of DP-TBD will become computationally prohibitive if the state dimension is more than two or three, which greatly prevents its applications to many realistic problems. In order to improve the computational efficiency of DP-TBD, a thresholding process based DP-TBD (TP-DP-TBD) is proposed in this paper. In TP-DP-TBD, a low threshold is first used to eliminate the noise-like (with low-amplitude) measurements. Then the DP integration process is modified to only focuses on the thresholded higher-amplitude measurements, thus huge amounts of computation devoted to the less meaningful low-amplitude measurements are saved. Additionally, a merit function transfer process is integrated into DP recursion to guarantee the inheritance and utilization of the target merits. The performance of TP-DP-TBD is investigated under both optical style Cartesian model and surveillance radar model. The results show that substantial computation reduction is achieved with limited performance loss, consequently TP-DP-TBD provides a cost-efficient tradeoff between computational cost and performance. The effect of the merit function transfer on performance is also studied.","Title: A Computationally Efficient Particle Filter for Multitarget Tracking Using an Independence Approximation\nYear: 2013\n\nAbstract: Particle filter (PF) based multi-target tracking (MTT) methods suffer from the curse of dimensionality. Existing strategies to combat this assume posterior independence between target states, in order to then sample targets independently, or to perform joint sampling of closely spaced targets only. When many targets are in proximity, these strategies either perform poorly or are too computationally expensive. We make two contributions towards addressing these limitations. Firstly, we advocate an alternative view of the use of posterior independence which emphasizes the statistical effect of assuming posterior independence on the Monte Carlo (MC) approximation of posterior density. Our analysis suggests that assuming posterior independence can provide a better MC approximation of the prior distribution at the next time, and therefore the posterior at the next time, without regard for how sampling is performed. Secondly, we present a computationally efficient, measurement directed, joint sampling method to cope with the target coupling and measurement ambiguity when targets are near each other. Consequently, we develop a PF which employs posterior independence while sampling targets jointly. This PF is applicable to both the traditional thresholded and track-before-detect style pixelized models. Simulation results for a challenging tracking scenario show that the proposed PF substantially outperforms existing approaches.","Title: A dynamic logistics coordination model for evacuation and support in disaster response activities\nYear: 2007\n\nAbstract: This paper describes an integrated location-distribution model for coordinating logistics support and evacuation operations in disaster response activities. Logistics planning in emergencies involves dispatching commodities (e.g., medical materials and personnel, specialised rescue equipment and rescue teams, food, etc.) to distribution centres in affected areas and evacuation and transfer of wounded people to emergency units. During the initial response time it is also necessary to set up temporary emergency centers and shelters in affected areas to speed up medical care for less heavily wounded survivors. In risk mitigation studies for natural disasters, possible sites where these units can be situated are specified according to risk based urban structural analysis. Logistics coordination in disasters involves the selection of sites that result in maximum coverage of medical need in affected areas. Another important issue that arises in such emergencies is that medical personnel who are on duty in nearby hospitals have to be re-shuffled to serve both temporary and permanent emergency units. Thus, an optimal medical personnel allocation must be determined among these units. The proposed model also considers this issue.","Title: Phase-Modulated Waveform Evaluation and Selection Strategy in Compound-Gaussian Clutter\nYear: 2013\n\nAbstract: This correspondence introduces a waveform evaluation and selection strategy (ESS) for application of optimized phase-modulated (PM) waveforms to mitigate compound-Gaussian clutter in radar. This strategy addresses the trade-off between potential improvement in detection performance from adaptive design of PM waveforms and the cost this entails in terms of system resources. An evaluation factor is introduced to quantify the performance improvement obtained by design and transmission of an optimized PM waveform in the second sub-dwell of a two-stage clutter mitigation scheme versus using linear frequency modulated (LFM) waveform in a sub-dwell. This evaluation factor provides the basis for the proposed ESS, the efficacy of which is demonstrated in simulations.","Title: Automatic Aircraft Recognition Using Maximum Likelihood Ratio Test\nYear: 2001\n\nAbstract: The automatic aircraft recognition in aerial images has been a topic receiving the maximum attention in the literature for more than two decades due to its practical value. Since the original images may be obtained in various uncertain environment, defects such as noise/background clutter, skewness, partial occlusion and blue may affect the image quality which lead to high environment dependence of some existing aircraft detection systems/ algorithms. This paper proposes a novel approach based on the Maximum Likelihood Ratio Test (MLRT) to extract different aircraft from the real aerial images. Objects in the tested images are corrupted randomly by the various defects. Without any extra information, the successful detection rate of the aircraft reaches 96% and that of the nonplane objects reaches 100% when only six aircraft references are chosen. The experiments are carried out to test sensitivity of the proposed method to the noise and partial occlusion. The comparisons with the Moment Fourier Descriptors (MFD) are also given.","Title: Adaptive detection and estimation for an unknown occurring interval signal in correlated Gaussian noise.\nYear: 2015\n\nAbstract: This paper considers the problem of detecting and estimating an unknown occurring interval signal in correlated Gaussian noise, which is often arisen in signal processing society, e.g., identifying the onset times of a seismic wave and detecting a distributed target in unknown occurring range cells. We propose the novel Generalized Likelihood Ratio Test (GLRT) algorithm, where the Maximum Likelihood Estimations (MLEs) of the unknown occurring interval are obtained through a Dynamic Programming (DP) method adaptively without the secondary data. Unlike the classic Sequential Probability Ratio Test (SPRT) methods which consist of an on-line detector before an off-line estimator, the proposed GLRT outputs the decision result and the estimations at the same time. The performances of the proposed algorithm are evaluated by numerical simulations as well as the applications of detecting and suppressing the transient interference in a radar operated on High-Frequency (HF) band with real data. HighlightsWe propose an adaptively GLRT based on DP method without any secondary data and off-line estimator.We evaluate the performance by computer simulations and a practical application in the HF radar with real data.The proposed GLRT shows better performance than the classic methods.The practical process result illustrates the effective performance.","Title: A Fast Maneuvering Target Motion Parameters Estimation Algorithm Based on ACCF\nYear: 2015\n\nAbstract: This letter considers the motion parameters estimation problem for a maneuvering target with arbitrary parameterized motion. The slant range of the target is modeled as a polynomial function in terms of its multiple motion parameters and a fast estimation method based on adjacent cross correlation function (ACCF) is proposed, where the iterative adjacent cross correlation operation is employed to remove the range migration and reduce the order of Doppler frequency migration. Then the motion parameters are estimated via Fourier transform. Compared with the generalized Radon Fourier transform (GRFT), the proposed method can estimate the parameters without searching procedure and acquire close estimation performance at high signal-to-noise ratio (SNR) with a much lower computational cost. Finally, simulations are provided to demonstrate the effectiveness.","Title: Coherent Integration for Maneuvering Target Detection Based on Radon-Lv’s Distribution\nYear: 2015\n\nAbstract: This letter considers the coherent integration problem for a maneuvering target, involving range migration (RM) and Doppler frequency migration (DFM) within one coherent pulse interval. A new coherent integration method, known as Radon-Lv's distribution (RLVD), is proposed. It can not only eliminate the RM effect via jointly searching in the target's motion parameters space, but also remove the DFM and achieve the coherent integration via Lv's distribution (LVD). Finally, several simulations are provided to demonstrate the effectiveness. The results show that for detection ability, the proposed method is superior to the moving target detection (MTD), Radon-Fourier transform (RFT), and Radon-fractional Fourier transform (RFRFT) under low signal-to-noise-ratio (SNR) environment.","Title: Approximation to independent lognormal sum with α-µ distribution and the application\nYear: 2015\n\nAbstract: This paper considers the statistical distribution of the sum of independent and possibly non-identically distributed lognormal random variables (RVs). The α-µ distribution is employed to approximate the probability of density function (PDF) of the lognormal sum. Two approaches, e.g., the moment-based and non-linear least square based (LS-based) methods, are used to obtain the parameters of α-µ distribution. Accounting for maximal-ratio combining (MRC) receiver, an analytical expression of the average bit error probability (ABEP) of the M-ary noncoherent frequency-shift keying (M-NFSK) modulation is derived. Finally, the performances of the α-µ approximation and the ABEP are evaluated via numerical experiments. HighlightsWe adopt the α-µ distribution to approximate the probability density function of the sum of independent and possibly non-identically distributed lognormal variables.We investigate two typical parameter estimation methods, i.e., the moment-based method and the non-linear least square based method.For maximal-ratio combining receiver, we derive an analytical expression of the average bit error probability of the M-ary noncoherent frequency-shift keying modulation resorting to the new α-µ approximation.","Title: A New Differential Raid For High Reliable All Flash Array\nYear: 2014\n\nAbstract: For high I/O performance, Solid State Drive (SSD)-based all Flash arrays (AFAs) are widely employed to larger scale storage systems. However, the aging problem of SSDs limits the lifetime and reliability of AFA. In order to extend the lifetime of AFA, we designed a capacity-based differential RAID (Redundant Array of Independent Disks) scheme (CDiff-RAID). In the proposed scheme, the array is initialized by a group of SSDs with unequal capacities. The smaller SSDs suffer more erase operations and age quickly. To maintain the age differential, the worn SSDs are continually replaced by new SSDs. Compared with parity-based Diff-RAID, the age differential in our scheme is independent of workload and the time-consuming of reconstruction is greatly reduced. We also evaluated the I/O performance of CDiff-RAID using SSD simulators. The proposed scheme outperforms Diff-RAID scheme in sequential and random traces in most cases.","Title: Joint multi-target detection and localization with a noncoherent statistical MIMO radar\nYear: 2014\n\nAbstract: In this paper, the problems of simultaneously detecting and localizing multiple targets are considered for noncoherent multiple-input multiple-output (MIMO) radar with widely separated antennas. By assuming a prior knowledge of target number, an optimal solution to this problem is presented first. It is essentially a maximum-likelihood (ML) estimator searching parameters of interest in a high-dimensional space. However, the complexity of this method increases exponentially with the number of targets. Besides, without the prior information of target number, a multi-hypothesis test strategy on target number is required, which further complicates this method. Therefore, we split the joint maximization into G disjoint optimization problems by clearing the interference from previously declared targets. In this way, we derive two fast but robust suboptimum solutions which allow trading performance for a much lower implementation complexity which is almost independent of the number of targets. In addition, the multi-hypothesis test detector is no longer required when target number is unknown. Simulation results show the proposed suboptimum algorithms can correctly detecting and accurately localizing multiple targets even when targets share common range bin in some paths.","Title: Recursive filtering for target tracking in multi-frame track-before-detect\nYear: 2014\n\nAbstract: This paper addresses the tracking problem with respect to dynamic programming (DP) based track-before-detect (DP-TBD). Basically, DP-TBD is a grid-based method performing search in the discretized state space, therefore the accuracy of its estimation suffers at least half a grid loss. Moreover, different from the conventional detection procedure which provides isolate point plots to a tracker, DP-TBD is a batch processor whose detection outputs are short state sequences in time series. Essentially, these short tracks are neither filtered target trajectories nor conventional point detections. Meantime the false alarms and missing reports of these short tracks also exist especially when target signal-to-noise ratio (SNR) is low. Thus, a tracker which can associate, smooth and finally combine these short tracks into continuous target trajectories is needed. In this work, we propose a two-stage, i.e., prediction and updating stages, recursive filtering algorithm for DP-TBD. The key idea is using the outputted short tracks after every DP-TBD batch processing as the input measurements for the updating stage of filtering. Simulation results show that the proposed algorithm can correctly estimate target trajectories and significantly enhance the tracking accuracy compared with the method without our filtering algorithm.","Title: Coherent Integration Algorithm For A Maneuvering Target With High-Order Range Migration\nYear: 2015\n\nAbstract: This paper considers the coherent integration problem for a maneuvering target with complex motions, where the velocity, acceleration and jerk result in respectively the first-order range migration (FRM), second-order range migration (SRM) and third-order range migration (TRM) within the coherent pulse interval. A new coherent integration algorithm based on generalized keystone transform (KT) and secondorder dechirp process is proposed, which employs the third-order KT, six-order KT, second-order KT and fold factor searching to correct the TRM, SRM and FRM, respectively. The range migration change during each step and computational complexity are also theoretically analyzed. Compared with the generalized Radon Fourier transform (GRFT) algorithm, the presented method can avoid the blind speed sidelobe (BSSL) and acquire close integration performance but with much lower computational cost. Simulations are provided to demonstrate the effectiveness. Finally, a generalized method, named generalized KT and generalized dechirp process (GKTGDP), is also introduced for the maneuvering target with arbitrary high-order range migration.","Title: Track-before-detect strategies for range distributed target detection in compound-Gaussian clutter\nYear: 2016\n\nAbstract: This paper considers range distributed target detection in compound-Gaussian clutter, which is often arisen in high resolution radars. Two dynamic programming based track-before-detect (DP-TBD) strategies, i.e., noncoherent integration based DP-TBD (NCI-DP-TBD) and the generalized likelihood ratio test based DP-TBD (GLRT-DP-TBD) strategies, are proposed to address the detection problem. The detection performances of the proposed strategies are examined via numerical simulations. HighlightsRange distributed targets in compound-Gaussian clutter is considered.Noncoherent integration and GLRT based DP-TBD strategies are proposed.A covariance matrix estimation strategy is proposed.","Title: Distributed Fusion of Labeled Multi-Object Densities Via Label Spaces Matching.\nYear: 2016\n\nAbstract: In this paper, we address the problem of the distributed multi-target tracking with labeled set filters in the framework of Generalized Covariance Intersection (GCI). Our analyses show that the space mismatching phenomenon, which means the same realization drawn from spaces of different sensors does not have the same implication, is quite common in practical scenarios and may bring serious problems. Our contributions are two-fold. Firstly, we provide a principled mathematical definition of label spaces matching (LS-DM) based on information divergence, which is also referred to as LS-M criterion. Then, to handle the LS-DM, we propose a novel two-step distributed fusion algorithm, named as GCI fusion via spaces matching (GCI-LSM). The first step is to match the spaces from different sensors. To this end, we build a ranked assignment problem and design a cost function consistent with LS-M criterion to seek the optimal solution of matching correspondence between spaces of different sensors. The second step is to perform the GCI fusion on the matched space. We also derive the GCI fusion with generic labeled multi-object (LMO) densities based on LS-M, which is the foundation of labeled distributed fusion algorithms. Simulation results for Gaussian mixture implementation highlight the performance of the proposed GCI-LSM algorithm in two different tracking scenarios.","Title: Principled Random Finite Set Approximations of Labeled Multi-Object Densities.\nYear: 2016\n\nAbstract: As a fundamental piece of multi-object Bayesian inference, multi-object density has the ability to describe the uncertainty of the number and values of objects, as well as the statistical correlation between objects, thus perfectly matches the behavior of multi-object system. However, it also makes the set integral suffer from the curse of dimensionality and the inherently combinatorial nature of the problem. In this paper, we study the approximations for the universal labeled multi-object (LMO) density and derive several principled approximations including labeled multi-Bernoulli, labeled Poisson and labeled independent identically clustering process based approximations. Also, a detailed analysis on the characteristics (e.g., approximation error and computational complexity) of the proposed approximations is provided. Then some practical suggestions are made for the applications of these approximations based on the preceding analysis and discussion. Finally, an numerical example is given to support our study.","Title: Distributed Fusion with Multi-Bernoulli Filter based on Generalized Covariance Intersection.\nYear: 2016\n\nAbstract: In this paper, we propose a distributed multiobject tracking algorithm through the use of multi-Bernoulli (MB) filter based on generalized covariance intersection (G-CI). Our analyses show that the G-CI fusion with two MB posterior distributions does not admit an accurate closed-form expression. To solve this problem, we first approximate the fused posterior as the unlabeled version of δ-generaliz...","Title: Fast Non-Searching Method for Maneuvering Target Detection and Motion Parameters Estimation.\nYear: 2016\n\nAbstract: This paper considers the coherent integration problem for maneuvering target detection and motion parameters estimation, involving range migration (RM) and Doppler frequency migration (DFM) within the coherent pulse interval. A fast non-searching method based on adjacent cross correlation function (ACCF) and Lv's distribution (LVD) is proposed, where the adjacent correlation operation is first emp...","Title: Fast Optimal Antenna Placement for Distributed MIMO Radar with Surveillance Performance\nYear: 2015\n\nAbstract: In this paper we demonstrate an optimization problem of antenna placement of distributed multi-input multi-output (MIMO) radar. To evaluate the surveillance performance of the radar system, a coverage ratio is proposed as a criterion. Since the problem is of extremely huge computational complexity due to its complicated objective function and high dimensionality, we propose a solution that contains two parts: 1) a low-complexity method to simplify the objective function; 2) a placement algorithm based on particle swarm optimization (PSO) to deal with the challenge of high dimensionality. We also analyse the computational complexity of our solution. Simulation results verify the validity and advantage in computational complexity of our solution. Our contributions include a novel optimization placement model of distributed MIMO radar and a computational efficient solution to establish the optimal positions of antennas.","Title: Performance prediction of OS-CFAR for generalized swerling-chi fluctuating targets.\nYear: 2016\n\nAbstract: We address the performance prediction of the ordered-statistic constant false alarm rate (OS-CFAR) detector for arbitrarily correlated and possibly nonidentically distributed target echoes in the presence of nonhomogeneous background. Accounting for the generalized Swerling-Chi fluctuating model, including the well-known Swerling models as special cases, we develop the analytic expressions of detection probability in terms of converging series in nonhomogeneous background involving clutter edges and possibly isolated point-like interference. At the analysis stage, we study the truncation error of the derived expressions and then evaluate the impacts on detection performance of the target parameters via numerical simulations.","Title: A Flash-Aware Intra-Disk Redundancy Scheme For High Reliable All Flash Array\nYear: 2015\n\nAbstract: In SSD-based All Flash Array (AFA), Redundant Array of Independent Disks (RAID) technologies are widely used to cope with drive failures and sector errors. However, the redundancy in disk level is either insufficient like RAID-5 or excessive like RAID-6. In order to combat the two types of errors, we combine the RAID-5 scheme with a Flash-aware Intra-disk Redundancy (FAIDR) scheme. For saving parity space, the ratio of redundancy in the FAIDR scheme grows with the increasing of sector errors. However, the parity handling overhead is significant. As NAND Flash blocks can't be updated before erase, we establish all FAIDR stripes on the physical page number of blocks regardless of the page state. Thus, the parity update operations in FAIDR are eliminated. Simulation results show that the proposed scheme has the ability to ensure the AFA's reliability in its all lifetime. It gains 41 percent storage capacity in comparison to RAID-6. And it only leads to very small performance loss compared with pure RAID-5.","Title: A Case Study of SWIM: Optimization of Memory Intensive Application on GPGPU\nYear: 2010\n\nAbstract: Recently, GPGPU has been adopted well in the High Performance Computing (HPC) field. The limited global memory bandwidth poses a great challenge to many GPGPU programmers trying to exploit parallelism within the CPU-GPU heterogeneous platform. In this paper, we choose SWIM, a typical memory intensive application from the SPEC OMP 2001 benchmark suite, for case study. We attempt to optimize the performance and energy consumption of the application utilizing different memory access mechanisms and present optimization methods including matrix transposition and kernel fusion. The experimental results on the Intel Core TM i920 CPU plus GeForce GTX 295 platform shows that, the proposed optimizing methods achieve a speedup of 8.7X over the original OpenMP program and reduce the energy consumption by 83% for the problem size of 2048*2048.","Title: CFAR detector based on clutter partition in heterogeneous background\nYear: 2015\n\nAbstract: The conventional CFAR detectors based on reference window suffer a considerable performance degradation in heterogeneous background. In this paper, a new knowledge based constant false alarm rate (KB-CFAR) detector is proposed to reduce the impact of heterogeneous environment. The proposed KB-CFAR detector exploits clutter partition as auxiliary knowledge to select the reference cells. The clutter partition is obtained by extracting useful knowledge from the massive environment information and it is updated by real-time radar returns. The performance of the new KB-CFAR detector is analyzed by real radar data which is collected by a linear frequency-modulated continuous wave (LFMCW) radar. The results show that the new detector achieves a satisfactory performance and works well under general conditions.","Title: Radar maneuvering target detection and motion parameter estimation based on TRT-SGRFT.\nYear: 2017\n\nAbstract: In this paper, we address the detection and motion parameter estimation problem for a maneuvering target with complex motions, where range migration (RM, including range walk and range curvature) and Doppler frequency migration (DFM) occur during the coherent observation time. A method based on time reversing transform (TRT) and special generalized Radon Fourier transform (SGRFT), i.e., TRT-SGRFT, is proposed to achieve the detection and motion parameters estimation. More specifically, the TRT is firstly applied to separate the motion parameters and reduce the order of RM and DFM. Then the SGRFT operation is employed to estimate part of the motion parameters. After that, the residual motion parameters could be achieved via the second SGRFT operation. The advantage of the TRT-SGRFT algorithm is that it could avoid the blind speed sidelobe (BSSL) effect and obtain a good tradeoff between the computational cost and estimation performance, in comparison with generalized Radon Fourier transform. Several numerical experiments are provided to demonstrate the its effectiveness. HighlightsThe detection and motion estimation problem for a maneuvering target with arbitrary parameterized motion is addressed.A method based on TRT-SGRFT is proposed to achieve the maneuvering target detection and motion parameters estimation.The TRT-SGRFT algorithm can avoid the BSSL and obtain a good tradeoff among computational cost, detection ability and parameter estimation performance.","Title: Joint estimation of location and signal parameters for an LFM emitter.\nYear: 2017\n\nAbstract: This paper considers the localization of an emitter employing linear frequency modulation (LFM) signal with unknown parameters. The proposed approach is based on jointly estimating the emitter position along with the transmitted signal parameters. In the maximum likelihood context, the joint estimation problem is shown to translate into an optimization search in a six-dimensional space. In many practical applications, such a search may be too computationally expensive to run in real-time. We suggest a sub-optimal yet fast solution based on combining the direct position determination (DPD) method with a short time Fourier transform (STFT). Despite the sub-optimality, the proposed DPD-STFT method is able, with low complexity, to achieve higher localization accuracies than methods that are based on estimation of emitter position only (regardless of signal parameters). Indeed, due to joint estimation of emitter position and the transmitted signal parameters, our method harnesses more information for localization purposes. Simulation results show that the proposed method outperforms the existing DPD algorithms, in terms of localization error. HighlightsAn optimal strategy to jointly estimate location and parameters of an LFM emitter.An algorithm to implement localization, based on combining STFT with DPD method.Inclusion of a novel quality factor to asses and improve localization performance.Analysis of computational complexity of the proposed algorithm.","Title: Suboptimum Low Complexity Joint Multi-target Detection and Localization for Noncoherent MIMO Radar with Widely Separated Antennas.\nYear: 2017\n\nAbstract: In this paper, the problems of simultaneously detecting and localizing multiple targets are considered for noncoherent multiple-input multiple-output (MIMO) radar with widely separated antennas. By assuming a prior knowledge of target number, an optimal solution to this problem is presented first. It is essentially a maximum-likelihood (ML) estimator searching parameters of interest in a high dimensional space. However, the complexity of this method increases exponentially with the number G of targets.Besides, without the prior information of the number of targets, a multi-hypothesis testing strategy to determine the number of targets is required, which further complicates this method. Therefore, we split the joint maximization into G disjoint optimization problems by clearing the interference from previously declared targets. In this way, we derive two fast and robust suboptimal solutions which allow trading performance for a much lower implementation complexity which is almost independent of the number of targets. In addition, the multi-hypothesis testing is no longer required when target number is unknown. Simulation results show the proposed algorithms can correctly detect and accurately localize multiple targets even when targets share common range bins in some paths.","Title: Fast coherent integration for maneuvering target with high-order range migration via TRT-SKT-LVD.\nYear: 2016\n\nAbstract: This paper addresses the coherent integration problem for detecting a maneuvering target with high-order range migration (RM), where the velocity, acceleration, and jerk result in, respectively, the first-order range migration (FRM), second-order range migration (SRM), and third-order range migration (TRM) within the coherent pulse interval. A fast coherent integration method based on time reversing transform (TRT), second-order keystone transform (SKT) and Lv's distribution (LVD), i.e., TRT-SKT-LVD, is proposed. In this method, TRT operation is first presented to remove the TRM and FRM simultaneously. After that, the SKT is employed to correct the remaining SRM and then the coherent accumulation could be obtained via LVD. The proposed algorithm is simple and fast in that it can be easily implemented by using complex multiplications, the fast Fourier transform (FFT), and inverse FFT (IFFT). Compared with the existing methods, the presented method can realize the coherent integration without any searching procedure and can achieve a good balance between the computational cost and detection ability. Both simulated and real data demonstrate the effectiveness of the proposed algorithm.","Title: Performance prediction of noncoherent detector for independent and nonidentically distributed NCG fluctuating targets.\nYear: 2016\n\nAbstract: We address the performance prediction of the noncoherent detector (i.e., squared-law plus integrator) for independent but possibly nonidentically distributed (non-id) noncentral Gamma (NCG) fluctuating target model, which is more general and can describe more complex targets. Initially, we obtain the probability density function (pdf) of the sum of independent and non-id NCG random variables in terms of an infinite sum of Gamma pdfs. Furthermore, we develop an analytic expression of the detection probability in terms of converging series based on the generalized Marcum Q function. Finally, at the analysis stage, we first study the truncation error of the derived expression and then evaluate the impacts on the detection performance of the target fluctuating parameters via numerical simulations.","Title: A novel mobile data plan design from the perspective of data period\nYear: 2016\n\nAbstract: In the conventional mobile data plans, the unused data will be cleared at the end of every period (e.g., month). To take advantage of customers' heterogeneous demands across different periods and to provide more flexible data service, some mobile data service providers (SP) have offered data plans with different lengths of period. In this paper, we consider the data plan design problem for a single SP, who provides data plans with different lengths of period to consumers with different characteristics of data demands. We propose a contract-theoretic approach, where the SP offers a period-price contract consisting of a set of period-price combinations. Each period-price combination is designed for a specific type of consumers with a specific data demand distribution. We systematically analyze the incentive compatibility (IC) constraint and individual rationality (IR) constraint which ensure each consumer to purchase the data plan with the period-price combination intended for his type. Based on the IC and IR analysis, we further derive the optimal contract that maximizes the SP's expected revenue. Our numerical results show that our proposed optimal contract can achieve 94% of the maximum social surplus, and can increase the SP's revenue by 25% comparing with the conventional monthly-period data plan.","Title: Detection of weak maneuvering target based on keystone transform and matched filtering process.\nYear: 2017\n\nAbstract: •The problems of range migration and Doppler frequency migration for the weak maneuvering target are addressed.•A method based on keystone transform and matched filtering process, i.e., KT-MFP, is proposed to obtain the coherent integration and achieve the maneuvering target detection.•The KT-MFP can obtain superior coherent integration and detection performance for both the slowmoving and fast-moving target, in comparison with the existing algorithms.•The KT-MFP requires lower computational cost without the BSSL effect, in compared with GRFT. April 5, 2017 DRAFT.","Title: Sequence-Reversing Transform-Based Coherent Integration for High-Speed Target Detection.\nYear: 2017\n\nAbstract: This paper addresses the coherent integration problem for detecting a high-speed target, involving range walk (RW) within the coherent processing time. A fast method based on sequence reversing transform is proposed to eliminate the RW and achieve the coherent accumulation. The proposed method is simple and fast since it can be easily implemented by using complex multiplications, the fast Fourier transform (FFT) and the inverse FFT. Compared with the Radon-Fourier transform and moving target detection methods, the presented algorithm can avoid the blind speed sidelobe and make a good balance between the computational cost and the detection ability. Finally, several simulations are provided to demonstrate its effectiveness.","Title: Multiframe Radar Detection of Fluctuating Targets Using Phase Information.\nYear: 2017\n\nAbstract: This paper considers the detection of fluctuating targets via dynamic-programming-based track-before-detect (DP-TBD) in radar systems. Swerling targets of types 0, 1, and 3 are considered. DP-TBD usually integrates either squared amplitude or the logarithm of the envelope likelihood ratio (LELR) scoring functions. Thus, only amplitude information is taken into account regardless of the fact that the measurements are often complex valued. In this paper, the phase information is used in the integration process of DP-TBD to enhance radar detection performance. More precisely, the logarithm of the complex-measurement-based likelihood ratio (LCLR) is used, taking the place of squared amplitude or the LELR. First, we derive the expressions for the LELR and the LCLR for the three Swerling types. Then, to reduce the complexity of computing LELR and LCLR, we also propose efficient but accurate approximations for the LELR and the LCLR. Simulations are used to assess the performance of different DP-TBD strategies.","Title: The Multiple Model Vo-Vo Filter.\nYear: 2017\n\nAbstract: A jump Markov system (JMS) multiple model solution is presented for tracking highly dynamic targets using the Vo–Vo filter (also known as generalized labeled multi-Bernoulli or GLMB filter). The closed-form solution is derived for Gaussian mixture implementation of the multiple-model Vo–Vo filter. The performance of the proposed method is examined and compared with the state of the art in challenging scenarios involving numerous appearing and disappearing targets with randomly time-varying dynamics. The results demonstrate that in such applications, the proposed method can outperform the competing techniques in terms of tracking accuracy, and is highly robust to variations in application and filter parameters such as clutter rate and model transition probabilities in the JMS.","Title: ISAR imaging of maneuvering target with complex motions based on ACCF-LVD\nYear: 2015\n\nAbstract: This paper considers the inverse synthetic aperture radar (ISAR) imaging problem for a maneuvering target with complex motions, involving range migration (RM) and Doppler frequency migration (DFM) within the coherent integration period of radar imaging, which will degrade the imaging quality. A nonsearching ISAR imaging algorithm based on adjacent cross correlation function (ACCF) and Lv's distribution (LVD), i.e., ACCF-LVD, is proposed, where the ACCF is applied to correct the RM and reduce the order of DFM. Then the motion parameters are estimated via LVD and Fourier transform. With the estimated motion parameters, high quality ISAR images can be achieved. The advantage of the presented method is that it can estimate the motion parameters under low signal-to-noise ratio (SNR) without searching procedures. Finally, several simulation examples are shown to confirm the validity of the proposed algorithm.","Title: Design and Simulation of a Self-adaptive Fuzzy-PID Controller for an Autonomous Underwater Vehicle.\nYear: 2017\n\nAbstract: In recent years, AUV (autonomous underwater vehicle) has been applied to many fields, such as offshore oil exploitation, underwater target detection, military applications and so on, which raises a higher demand for accuracy control of AUV. In this paper, for the rim-drive dead zone and system nonlinearity problem of “Swordfish II” autonomous underwater vehicle in the actual control process, according to the shape and motion characteristics of the AUV, the six DOF (degrees of freedom) dynamic model is established under MATLAB/Simulink. The fuzzy control theory and the traditional PID control algorithm are combined to design a fuzzy-PID controller. Finally, we construct the simulation system. The simulation results show that the designed controller can control the AUV model and solve the dead zone problem well, and the fuzzy PID controller has better control effect than the traditional pid controller.","Title: Robust Distributed Fusion with Labeled Random Finite Sets.\nYear: 2018\n\nAbstract: This paper considers the problem of the distributed fusion of multiobject posteriors in the labeled random finite set filtering framework, using a generalized covariance intersection (GCI) method. Our analysis shows that GCI fusion with labeled multiobject densities strongly relies on label consistencies between local multiobject posteriors at different sensor nodes, and hence suffers from a sever...","Title: Track-Before-Detect Strategies for Radar Detection in G0-Distributed Clutter.\nYear: 2017\n\nAbstract: This paper considers target detection via dynamic-programming based track-before-detect (DP-TBD) for radar systems. The clutter is modeled usingenlr G0 distribution, which is usually used to model clutter received from high-resolution radars and radars working at small grazing angles. Two target models, namely, Swerling 0 and 1 models, are considered to capture the radar cross section changes over time. DP-TBD techniques that integrate amplitude suffer from significant performance loss in this case due to the high likelihood of target-like outliers. In this paper, the log-likelihood ratio (LLR) is used in the integration process of DP-TBD, taking the place of amplitude, to enhance radar detection performance. The expressions for the LLR for the above target models are derived first. However, neither of them has a closed-form solution. In order to reduce the complexity of evaluating the LLR, efficient but accurate approximation methods are proposed. Then the approximated LLR is used in the integration process of DP-TBD. Simulations are used to examine the efficiency of the approximation methods as well as the performances of different DP-TBD strategies.","Title: Joint Node Selection and Power Allocation Strategy for Multitarget Tracking in Decentralized Radar Networks.\nYear: 2018\n\nAbstract: Networked radar systems have been demonstrated to offer enhanced target tracking capabilities. An effective radar resource allocation strategy can efficiently optimize system parameters, leading to performance enhancements. In this paper, two critical but limited system resources are considered for optimization: the number of radar nodes and the transmitted power. In this scenario, a joint node se...","Title: Multiobject Tracking for Generic Observation Model Using Labeled Random Finite Sets.\nYear: 2018\n\nAbstract: This paper presents an exact Bayesian filtering solution for the multiobject tracking problem with the generic observation model. The proposed solution is designed in the labeled random finite set framework, using the product styled representation of labeled multiobject densities, with the standard multiobject transition kernel and no particular simplifying assumptions on the multiobject likelihoo...","Title: Data-Dependent Clustering-CFAR Detector in Heterogeneous Environment.\nYear: 2018\n\nAbstract: This paper devises a new constant false alarm rate (CFAR) detection scheme to deal with the problem of radar target detection in heterogeneous environment. The proposed scheme, called clustering-CFAR detector, is data dependent and composed of three stages: an adaptive clustering procedure that, exploiting the recorded measurements of the clutter environment, divides the detection area into different classes to provide auxiliary information, a dynamic reference cell selector that chooses appropriate secondary data according to the classes, and a conventional CFAR processor to make the final decision about the target presence. The performance of clustering-CFAR detector is analyzed by computer simulation and public radar measured data (IPIX data and MSTAR data), and compared with existing CFAR detectors. The results show that the new detector achieves a better performance in the aspects of terrain classification, control of false alarm points, and probability of detection.","Title: An Efficient Recursive Multiframe Track-Before-Detect Algorithm.\nYear: 2018\n\nAbstract: Multiframe track-before-detect (MF-TBD) usually uses sliding-window-based batch processing, where a number N of the latest data frames are jointly processed at each measurement time. The sliding window mechanism compromises the operating efficiency of MF-TBD by increasing both computational costs and memory requirements, thus, heavily restricting its application in practical problems. In this paper, an improved recursive implementation for MF-TBD is proposed. Unlike the sliding-window-based implementation, the proposed method calculates the merit function, a measure of the possibility that a state is target originated, of the current batch based on an approximated recursive relationship between the merit functions of consecutive batches. As a result, instead having to process the whole batch, at any given time only the latest frame needs to be processed. The recursive relationship is first derived for any arbitrary merit function, and then explored further with several typical merit functions that are used in MF-TBD. Both the theoretical analysis and simulation results demonstrate that the proposed method can achieve almost N times reduction in computational complexity and memory requirements with negligible performance loss.","Title: A Page Lifetime-Aware Scrubbing Scheme For Improving Reliability Of Flash-Based Ssd\nYear: 2017\n\nAbstract: Solid-state drive (SSD) has gain prevalence in consumer and enterprise storage markets. However, its reliability is declining with the wearing of the Flash memory. The error correcting codes (ECCs) are generally applied in SSD to cope with bit errors caused by abrasion, however, the lifetime of pages in Flash is still limited to their fixed correctability. Due to different process and usage, the pages of Flash deteriorate in different speeds, which limited the service time of SSD. In order to prolong the lifetime of SSD, we proposed a Page Lifetime-aware Scrubbing (LaScrub) scheme to find dangerous pages which exist more bit errors. The simulation results show the proposed scheme is able to enhance the reliability of SSD. It gains 86% storage space compared with normal SSD at the same reliability requirement.","Title: Robust position control of hydraulic manipulators using time delay estimation and nonsingular fast terminal sliding mode.\nYear: 2018\n\nAbstract: A simple and robust tracking controller based on time delay estimation and nonsingular fast terminal sliding mode is proposed for the position control of hydraulic manipulator. The proposed controller does not require the mathematical model of hydraulic manipulator dynamics, which ensures that the method is simple and model free. Two elements, time delay estimation and nonsingular fast terminal sliding mode, are implemented in the proposed controller. Time delay estimation estimates the complex dynamics of the hydraulic manipulator, and nonsingular fast terminal sliding mode is implemented as the nonlinear desired error dynamic to ensure fast convergence and high trajectory tracking accuracy. Experiments are performed to verify the tracking performance of the method on a hydraulic manipulator. The results demonstrate that this method achieves faster and higher precision position tracking performance than a conventional time delay control with linear error dynamics.","Title: Event-Triggered Communication Mechanism for Distributed Flocking Control of Nonholonomic Multi-agent System.\nYear: 2018\n\nAbstract: As the scale of multi-agent systems (MAS) increases, communication becomes a bottleneck. In this paper, we propose an event-triggered mechanism to reduce the inter-agent communication cost for the distributed control of MAS. Communication of an agent with others only occurs when event triggering condition (ETC) is met. In the absence of communication, other agents adopt an estimation process to acquire the required information about the agent. Each agent has an above estimation process for itself and another estimation based on Kalman Filter, the latter can represent its actual state considering the measurement value and error from sensors. The error between the two estimators indicates whether the estimator in other agents can maintain a relatively accurate state estimation for this agent, and decides whether the communication is triggered. Simulations demonstrate the effectiveness and advantages of the proposed method for the distributed control of flocking in both Matlab and Gazebo.","Title: Study on HEV's driving condition recognition method based on PSO algorithm.\nYear: 2018\n\nAbstract: In the interest of the hybrid electric vehicle (HEV) real-time road gradient and vehicle load( driving condition) effective identification during the running process, this work takes the series-parallel HEV as the research object and studies on the dynamic identification mechanism of slope and load, based on the analysis of its structural parameters. Firstly, vehicle's driving condition identification model is developed, and the optimization goal function is established using the least square method. Secondly, six different kinds of particle swarm optimization (PSO) algorithm are used for the recognition of vehicle's driving condition, and the results show that hybrid PSO algorithm based on hybrid training algorithm has better calculation accuracy for this problem. Finally, Experiments are carried out to verify the driving condition recognition method based on PSO algorithm. Through the acquisition of a real vehicle data during the running process, road grade and vehicle mass are estimated by using the proposed method, and the effectiveness of the proposed method is proved through comparison of errors between recognition results and true value.","Title: Detection and RM correction approach for manoeuvring target with complex motions.\nYear: 2017\n\nAbstract: This study addresses the coherent accumulation problem for detecting a manoeuvring target with complex motions, where range migration (RM) [i.e. range walk (RW) and range curvature (RC)] and Doppler frequency migration (DFM) occur during the coherent integration time. An efficient approach based on generalised keystone transform (GKT), radon transform (RT) and generalised dechirp process (GDP), i....","Title: Biological Plausibility And Stochasticity In Scalable Vo2 Active Memristor Neurons\nYear: 2018\n\nAbstract: Neuromorphic networks of artificial neurons and synapses can solve computationally hard problems with energy efficiencies unattainable for von Neumann architectures. For image processing, silicon neuromorphic processors outperform graphic processing units in energy efficiency by a large margin, but deliver much lower chip-scale throughput. The performance-efficiency dilemma for silicon processors may not be overcome by Moore's law scaling of silicon transistors. Scalable and biomimetic active memristor neurons and passive memristor synapses form a self-sufficient basis for a transistorless neural network. However, previous demonstrations of memristor neurons only showed simple integrate-and-fire behaviors and did not reveal the rich dynamics and computational complexity of biological neurons. Here we report that neurons built with nanoscale vanadium dioxide active memristors possess all three classes of excitability and most of the known biological neuronal dynamics, and are intrinsically stochastic. With the favorable size and power scaling, there is a path toward an all-memristor neuromorphic cortical computer.","Title: Receive-Beam Resource Allocation for Multiple Target Tracking With Distributed MIMO Radars.\nYear: 2018\n\nAbstract: A distributed multiple-input multiple-output (MIMO) radar system is capable of tracking multiple targets under the “defocused transmit-focused receive” (DTFR) operating mode, in which each transmitter forms a completely defocused beam to illuminate the whole surveillance region and each receiver adopts a focused beam to attain a higher resolution. However, when operating in this mode, there exists...","Title: Computationally efficient coherent detection and parameter estimation algorithm for maneuvering target.\nYear: 2019\n\nAbstract: •A computationally efficient coherent detection and parameter estimation algorithm via SAF-SFT is proposed.•The proposed algorithm could remove the range cell migration and Doppler spread effects without parameter-searching process.•SAF-SFT method could strike a good balance between computational cost and detection probability as well as the estimation performance.","Title: Computationally Efficient Multi-Agent Multi-Object Tracking With Labeled Random Finite Sets.\nYear: 2019\n\nAbstract: This paper addresses multi-agent multi-object tracking with labeled random finite sets via Generalized Covariance Intersection (GCI) fusion. While standard GCI fusion of Labeled Multi-Object (LMO) densities is labelwise and hence fully parallelizable, previous work unfortunately revealed that its fusion performance is highly sensitive to the unavoidable label inconsistencies among different agents...","Title: Fault-Tolerant and Self-Adaptive Market-Based Coordination Using Hoplites Framework for Multi-Robot Patrolling Tasks\nYear: 2018\n\nAbstract: An autonomous robot team can be employed for continuous coverage of a dynamic environment. In this paper, we propose a novel approach for creating multi-robot patrolling policies, which is fault-tolerant and self-adaptive. A dynamic priority queue and time-out replanning mechanism are maintained by each robot to schedule the tasks fault-tolerantly in the context of the market-based method. Hoplites framework is adapted by introducing a self-adaptive threshold adjustment and sharing mechanism to provide a high-level coordination. This work is demonstrated by a multi-robot patrolling task implemented on Robot Operating System (ROS). A flexible tool, Stage, is leveraged to provide the simulated environment. The experimental results validate the effectiveness and availability.","Title: A Coherent Detection and Velocity Estimation Algorithm for the High-Speed Target Based on the Modified Location Rotation Transform.\nYear: 2018\n\nAbstract: High-speed target often suffers from the range walk (RW) problem within the process of long-time coherent integration (CI). The aim of this paper is to present a new coherent detection algorithm, i.e., modified location rotation transform (MLRT), which could effectively correct the RW and estimate the velocity of the high-speed target. In particular, the MLRT algorithm could rotate the coordinate ...","Title: STGRFT for Detection of Maneuvering Weak Target With Multiple Motion Models.\nYear: 2019\n\nAbstract: Long time coherent integration (LTCI) technique enhances radar detection of maneuvering weak targets. Unfortunately, range migration (RM) and Doppler frequency migration (DFM) would occur during the integration period. Existing LTCI methods are all based on the assumption that the target is of single motion model (i.e., the target's motion model is fixed) during the long integration time. However,...","Title: Scaled accuracy based power allocation for multi-target tracking with colocated MIMO radars.\nYear: 2019\n\nAbstract: •A scaled accuracy based power allocation framework is proposed to address multitarget tracking problem in a colocated MIMO radar system.•The proposed strategy can allocate the transmit power adaptively as tracking accuracy requirement change over time and space.•The proposed strategy yields better power allocation options than the existing strategy.•Besides the colocated MIMO radar, the proposed framework can be used in other radar configurations such as the phased array radar and distributed MIMO radar systems.","Title: Asynchronous multi-rate multi-sensor fusion based on random finite set.\nYear: 2019\n\nAbstract: •Three asynchronous fusion algorithms, a centralized fusion algorithm and two distributed fusion algorithms, are proposed.•The proposed SP fusion algorithm could achieve the best performance due to the usage of raw measurements.•Compared with the SP fusion algorithm, the proposed two distributed fusion algorithms based on GCI rule are more robust and their output time is more flexibility.•The proposed S-GCI fusion algorithm needs less computational burden at the cost of a bit less performance compared with the B-GCI fusion algorithm.","Title: Direct position determination of multiple coherent sources using an iterative adaptive approach.\nYear: 2019\n\nAbstract: •The direct localization algorithm combining with the iterative approach is proposed for coherent sources.•The connected domain separation method can efficiently isolate and extract the domains containing different sources.•The proposed algorithm outperforms the existing methods in accuracy and robustness for locating coherent sources.•Increasing antennas can obtain better localization performance from energy gain and spatial filtering effect.","Title: Distributed Multi-sensor Multi-view Fusion based on Generalized Covariance Intersection.\nYear: 2019\n\nAbstract: •Multi-target tracking using multiple sensors having different fields of view.•Fusion of the posterior Probability Hypotheses Densities of the sensors.•Robust distributed fusion based on generalized covariance intersection.•A clustering algorithm is used to derive an efficent parallelized fusion strategy.•Compensation strategy to preserve targets outside the common field of view.","Title: Detecting retail products in situ using CNN without human effort labeling.\nYear: 2019\n\nAbstract: CNN is a powerful tool for many computer vision tasks, achieving much better result than traditional methods. Since CNN has a very large capacity, training such a neural network often requires many data, but it is often expensive to obtain labeled images in real practice, especially for object detection, where collecting bounding box of every object in training set requires many human efforts. This is the case in detection of retail products where there can be many different categories. In this paper, we focus on applying CNN to detect 324-categories products in situ, while requiring no extra effort of labeling bounding box for any image. Our approach is based on an algorithm that extracts bounding box from in-vitro dataset and an algorithm to simulate occlusion. We have successfully shown the effectiveness and usefulness of our methods to build up a Faster RCNN detection model. Similar idea is also applicable in other scenarios.","Title: WRFRFT-based Coherent Detection and Parameter Estimation of Radar Moving Target With Unknown Entry/Departure Time.\nYear: 2020\n\nAbstract: •An efficient coherent detection and parameter estimation method named WRFRFT is proposed for radar moving target with unknown entry/departure time.•The proposed method could effectively estimate the time parameters and motion parameters of the target with across range cell and Doppler spread effects.•WRFRFT-based method could obtain better detection ability and superior estimation performance than the existing popular coherent detection and estimation algorithms.","Title: Comprehensive Evaluation Of The Gf-4 Satellite Image Quality From 2015 To 2020\nYear: 2021\n\nAbstract: GaoFen-4(GF-4) is the highest spatial resolution Earth observation satellite operating in geosynchronous orbit. Its fixed Earth observation location, rapid responsiveness, and wide observation range make it popular in disaster and emergency monitoring. To evaluate the GF-4 image quality in detail on a long-term basis, this study analyzes the image quality after the commissioning phase by focusing on ground sample distance (GSD) and geometric and radiometric quality. The theoretical calculation, geometric and radiometric measurements, and on-site experiments results show that (1) the GSD of the GF-4 image is similar to 50 m at the nadir point and increases gradually with the distance away from the nadir point, (2) most external geometric errors are within the design requirements of 4 km despite some exceeding the limit, and the internal geometric errors are tested within 1 pixel, and (3) image sharpness is generally stable but varies with the atmosphere condition and imaging time, and the radiometric response gradually degrades at the rate of less than 5.5% per year.","Title: A New Flattened Cylinder Specimen For Direct Tensile Test Of Rock\nYear: 2021\n\nAbstract: In recent decades, researchers have paid more attention to the indirect tensile test than to the direct tensile test (DTT) of rocks, mainly due to difficulties in the alignment and the stress concentration at the end of an intact cylindrical specimen. In this paper, a new flattened cylinder specimen and a clamp device were designed to obtain the true tensile strength of the rock in DTT. Stress distributions of the specimen with different lengths (l) and cutting thicknesses (t) were analyzed, and damage processes of the specimen were monitored by the Digital Image Correlation (DIC), the fractured sections were also scanned. Different mechanical parameters were also obtained by the DTT of the flattened cylinder specimens and the intact cylinder specimens, as well as the Brazilian disc. Research results show that the tensile strength obtained by DTT is smaller than the Brazilian disc and is slightly greater than the intact cylindrical specimen. The flattened cylinder specimen with 0.20 <= 2t/D < 0.68 and 0.10 <= l/D <= 0.20 is recommended to measure the true tensile strength of rock material in DTT. This new shape of the specimen is promising to be extended in the uniaxial or triaxial direct tension test.","Title: Event-triggered Neural Network-Based Adaptive Control for a Class of Uncertain Nonlinear Systems\nYear: 2021\n\nAbstract: In the paper, an event triggering adaptive control method based on neural network (NN) is proposed for a class of uncertain nonlinear systems with external disturbances. In order to reduce the network resource utilization, a novel event-triggered condition by the Lyapunov approach is proposed. In addition, the NN controller and adaptive parameters determined by the Lyapunov stability method are updated only at triggered instants to reduce the amount of calculation. Only one NN is used as the controller in the entire system. The stability analysis results of the closed-loop system are obtained by the Lyapunov approach, which shows that all the signals in the systems with bounded disturbance are semi-globally bounded. Zeno behavior is avoided. Finally, the analytical design is confirmed by the simulation results on a two-link robotic manipulator.","Title: A Fast Approach for Detection and Parameter Estimation of Maneuvering Target With Complex Motions in Coherent Radar System\nYear: 2021\n\nAbstract: In coherent radar system, maneuvering target with complex motions usually confronts the issues of range cell migration (RCM) and Doppler frequency spread (DFS) within the long-time coherent integration process. To address these issues and achieve reliable detection and estimation performance, this paper proposes a novel fast approach without motion parameters searching. Particularly, the proposed ...","Title: Trajectory PHD and CPHD Filters With Unknown Detection Profile\nYear: 2022\n\nAbstract: Compared to the probability hypothesis density (PHD) and cardinalized PHD (CPHD) filters, the trajectory PHD (TPHD) and trajectory CPHD (TCPHD) filters are for sets of trajectories, and thus are able to produce trajectory estimates with better performance. In this paper, we develop the TPHD and TCPHD filters which can adaptively learn the history of the unknown target detection probability, and therefore they can perform more robustly in scenarios where targets are with unknown and time-varying detection probabilities. These filters are referred to as the unknown TPHD (U-TPHD) and unknown TCPHD (U-TCPHD) filters. By minimizing the Kullback-Leibler divergence (KLD), the U-TPHD and U-TCPHD filters can obtain, respectively, the best Poisson and independent identically distributed (IID) density approximations over a set of augmented trajectories. For computational efficiency, we also propose the U-TPHD and U-TCPHD filters that only consider the unknown detection profile at the current time. Specifically, the Beta-Gaussian mixture method is adopted for the implementations of proposed filters, which are referred to as the BG-U-TPHD and BG-U-TCPHD filters. The \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$L$</tex-math></inline-formula>\n-scan approximations of these filters with much lower computational burden are also presented. Finally, various simulation results demonstrate that the BG-U-TPHD and BG-U-TCPHD filters can achieve robust tracking performance to adapt to unknown detection profile. Besides, it also shows that usually a small value of \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$L$</tex-math></inline-formula>\n in the \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$L$</tex-math></inline-formula>\n-scan approximation has similar performance than with a large value at a fraction of the computational cost.","Title: Multi-target Joint Tracking and Classification Using the Trajectory PHD Filter.\nYear: 2021\n\nAbstract: To account for joint tracking and classification (JTC) of multiple targets from observation sets in presence of detection uncertainty, noise and clutter, this paper develops a new trajectory probability hypothesis density (TPHD) filter, which is referred to as the JTC-TPHD filter. The JTC-TPHD filter classifies different targets based on their motion models and each target is assigned with multiple class hypotheses. By using this strategy, we can not only obtain the category information of the targets, but also a more accurate trajectory estimation than the traditional TPHD filter. The JTC-TPHD filter is derived by finding the best Poisson posterior approximation over trajectories on an augmented state space using the Kullback-Leibler divergence (KLD) minimization. The Gaussian mixture is adopted for the implementation, which is referred to as the GM-JTC-TPHD filter. The L-scan approximation is also presented for the GM-JTC-TPHD filter, which possesses lower computational burden. Simulation results show that the GM-JTC-TPHD filter can classify targets correctly and obtain accurate trajectory estimation.","Title: Multi-Frame Joint Tracking and Shape Estimation Method for Weak Extended Targets\nYear: 2021\n\nAbstract: This paper addresses the joint tracking and shape estimation (JTSE) problem of elliptical extended targets in low signal-to-noise (SNR) scenarios using multi-frame joint processing. Considering the weak target echoes and unknown parameters of elliptical extended targets, it is challenging to achieve effective detection and tracking. To solve these problems, a multi-frame tracking and shape estimat...","Title: Joint tracking sequence and dwell time allocation for multi-target tracking with phased array radar\nYear: 2022\n\nAbstract: •This work addresses a joint tracking sequence and dwell time allocation (JTSDTA) for multi-target tracking (MTT) based on a phased array radar (PAR) system.•In the proposed model, implementations of target tracking and searching of the PAR are considered to be interleaved for the first time.•A time-aware posterior Cramér–Rao lower bound is derived to quantify the tracking performance of the PAR.•A two steps-based decoupling algorithm is proposed to solve the JTSDTA-related problem efficiently.","Title: Robust Radar Detection and Classification of Traffic Vehicles Based on Anchor-free CenterNet\nYear: 2021\n\nAbstract: Radar has been regarded as the enabling sensor technology to realize the intelligent transportation system. The current works in traffic radar mainly focus on detecting the target with fixed size. However, traffic scenarios consist of various targets with different sizes. Therefore, in this paper, we propose a CenterNet-based radar signal processing framework for detecting and classifying four typ...","Title: Calib-Net: Calibrating the Low-Cost IMU via Deep Convolutional Neural Network\nYear: 2021\n\nAbstract: The low-cost Inertial Measurement Unit (IMU) can provide orientation information and is widely used in our daily life. However, IMUs with bad calibration will provide inaccurate angular velocity and lead to rapid drift of integral orientation in a short time. In this paper, we present the Calib-Net which can achieve the accurate calibration of low-cost IMU via a simple deep convolutional neural network. Following a carefully designed mathematical calibration model, Calib-Net can output compensation components for gyroscope measurements dynamically. Dilation convolution is adopted in Calib-Net for spatio-temporal feature extraction of IMU measurements. We evaluate our proposed system on public datasets quantitively and qualitatively. The experimental results demonstrate that our Calib-Net achieves better calibration performance than other methods, what is more, and the estimated orientation with our Calib-Net is even comparable with the results from visual inertial odometry (VIO) systems.","Title: Semi-Empirical Time-Dependent Parameter of Shear Strength for Traction Force between Deep-Sea Sediment and Tracked Miner\nYear: 2022\n\nAbstract: Based on our direct shear creep experiment and the direct shear rheological constitutive model, a semi-empirical time-dependent parameter of the shear strength is obtained by Mohr-Coulomb shear strength theory, and different time-dependent traction force calculations between deep-sea sediment and a tracked miner are conducted by the work-energy principle. The time-dependent traction force calculation under its influencing factors, including the time, track shoe number, and grounding pressure, are analyzed and proved to be valid by the traction force experiment of a single-track shoe. The results show that the time-dependent cohesion force obtained by a semi-empirical way can be easily used to deduce the time-dependent traction force models under the different grounding pressure distributions and applied into deep-sea engineering application conveniently; the verified traction force models by the traction force experiment of a single-track shoe illustrate that traction force under the decrement grounding pressure distribution is the worst among the four kinds of grounding pressure distributions and suggested for evaluating the most unfavorable traction force and calculating the trafficability and stability of the deep-sea tracked miner.","Title: Exponential Mixture Density Based Approximation to Posterior Cramr-Rao Lower Bound for Distributed Target Tracking\nYear: 2022\n\nAbstract: The posterior Cramr-Rao lower bound (PCRLB) and a number of its extensions including the unconditional PCRLB (U-PCRLB) and the conditional PCRLB (C-PCRLB) have been widely studied in multi-sensor target tracking (MSTT). Previous MSTT works assume that the measurements are conditionally independent, which is often not the case in practice. When the correlations of the measurements are unknown, the standard Bayes update and PCRLB computations cannot be performed. Inspired by geometric average (GA) fusion that implements data fusion with unknown data correlations by employing exponential mixture density (EMD) to compute the global posterior, EMD-based approximations to U-PCRLB and C-PCRLB are derived in this paper for two classic distributed fusion architectures, namely the hierarchical and consensus architectures. The PCRLB can be decomposed into two parts, one coming from the prior information and the other from the information contained in the data. The data information part of PCRLB is approximated via EMD-based posterior, leading to the approximation of the PCRLB. We present a sequential Monte Carlo solution to recursively compute the proposed approximation to the PCRLB for nonlinear non-Gaussian estimation problems. Numerical simulations are provided to show that the proposed approximation to the bound on the estimation mean square error (MSE) is tighter compared to the existing bounds based on likelihood fusion obtained under the measurement independence assumption.","Title: A GDOP-Based Performance Description of TOA Localization with Uncertain Measurements\nYear: 2022\n\nAbstract: In this paper, we study a geometric dilution of a precision (GDOP)-based localization performance metric for multisite radar adopting a time-of-arrival (TOA)-based localization scheme. In contrast to the existing literature, we consider an actual uncertain measurement situation where the detection probabilities of radar nodes are assumed to be less than unity. The aim is to formulate a general signal-decoupled metric to describe the system localization performance while fully considering detection and estimation operations. Specifically, to match the uncertain measurements, we first establish effectively detected time delay measurements (TDMs) for localization and modify the traditional performance bounds for TDM estimation. Then, by combining the localization performance with the effective detection (ED) via probability, we propose a novel geometric dilution of precision with uncertain measurements (GDOP-UM) metric. The proposed metric can truly characterize the localization performance under the uncertain measurement situation. Finally, the simulation results show that the proposed GDOP-UM can describe the actual localization performance regardless of how the detection performance changes.","Title: Subspace-Based Detection and Localization in Distributed MIMO Radars\nYear: 2022\n\nAbstract: In this paper, we consider a distributed multiple-input multiple-output (MIMO) radar which radiates waveforms with non-ideal cross- and auto-correlation functions and derive a novel subspace-based procedure to detect and localize multiple prospective targets. The proposed solution solves a sequence of composite binary hypothesis testing problems by resorting to the generalized information criterion (GIC); in particular, at each step, it aims to detect and localize one additional target, upon removing the interference caused by the previously-detected targets, An illustrative example is provided.","Title: An Efficient Track-Before-Detect for Multi-PRF Radars With Range and Doppler Ambiguities\nYear: 2022\n\nAbstract: This article considers the detection and tracking of weak targets in multiple pulse repetition frequency (multi-PRF) radars using the track-before-detect (TBD) technique. By exploring the measurement independence among different PRFs, we decompose the joint multi-PRF and multiframe optimization problem into several lower dimensional maximizations, each of which corresponds to an intra-PRF multiframe processing. An efficient two-stage TBD procedure with a parallel structure is proposed for the multi-PRF radars. Specifically, in the first stage, a constraint inequality is derived analytically and used to decouple the measurement ambiguities from the nonlinear conversion relationship between polar and Cartesian coordinates. The intra-PRF multiframe integration can then be carried out concurrently using the ambiguous measurements and the target-like measurement plot sequences with different ambiguities are extracted for different PRFs. In the second stage, a covariance combination fusion-based inter-PRF joint disambiguation and estimation algorithm is proposed to solve the ambiguities of the plot sequences and output high-accuracy target tracks. Simulation experiments show that the proposed algorithm can provide a good detection performance and higher tracking accuracy with much lower complexity.","Title: Joint Online Route Planning and Resource Optimization for Multitarget Tracking in Airborne Radar Systems\nYear: 2022\n\nAbstract: Reasonable route planning and resource allocation strategy in the airborne radar systems (ARS), can sufficiently utilize the limited resources and promote the multitarget tracking (MTT) performance. However, using separately the route planning and resource optimization method cannot take full advantage of the airborne platform. Considering this issue, we propose a joint online route planning and resource optimization strategy in the ARS to improve the system capability for MTT. First, a kinematic model of the ARS, including the mathematical expression between the radar states and the system control parameters, is introduced, which integrates the route planning to the radar scheduling scheme. Next, the posterior Cramér–Rao lower bound about route planning and resource optimization variables for the tracking targets is derived. Then, a scaled-based utility function is established to quantify the MTT performance. Hereafter, a nonconvex problem is formulated by minimizing the utility function with route and resource constraints, and then a efficient three-stage partition-based solution is proposed. Finally, simulation experiments demonstrate the effectiveness of the proposed algorithm. Furthermore, over the traditional benchmark algorithm, the tracking performance of the proposed approach improves 30.44%.","Title: Multitarget Detection Strategy for Distributed MIMO Radar With Widely Separated Antennas\nYear: 2022\n\nAbstract: In this article, we propose a novel solution to detect multiple targets using a distributed multiple-input multiple-output (MIMO) radar under the so-called \"defocused transmit-defocused receive\" operating mode. The proposed method employs a grid-based data matching algorithm, aiming to associate the target responses to potential target locations, solving the resulting data puzzle that evaluates the various cells under test (CUTs) in the surveillance area resulting from the intertwined range cells across all transmit-receive channels. Sketchily, the approach divides the surveillance area into identically interlocking and analytically expressible grid cells and then selects the grid cells with the best fitting multichannel data to be equivalently regarded as the CUTs. Next, the generalized likelihood ratio test (GLRT) detector is derived to test for target presence in each of the selected grid cells. A separate procedure is introduced to eliminate the spurious \"shadow targets,\" false alarms occurring in the grid cells without a target while sharing range cells with the targets. The essence of this procedure is to find the source of the observed contributions to the grid cells whose test statistics exceed their thresholds, and simultaneously obtain the positions of the targets. The proposed method is evaluated using both numerical simulations and experimental data recorded by five small radars, demonstrating the effectiveness of the proposed technique.","Title: Adaptive beam burden management for multi-target tracking in distributed radar network\nYear: 2022\n\nAbstract: •The system beam burden with the beam pointing and pulse number matrix is modeled for multi-target tracking (MTT) in distributed radar network (DRN).•This work formulates a novel optimization problem on the basis of beam burden saving with the tracking tasks and the system restrictions.•We design a Lagrangian relaxation-based (LR) fast algorithm to efficiently obtain the solutions.•The ABPNS strategy is proposed by adjusting the beam and pulse number to realize adaptive beam burden management in the DRN.","Title: Strong scattering points-based joint detection and size estimation method for swarm targets\nYear: 2022\n\nAbstract: The paper considers the joint detection and size estimation of swarm targets. The echo energy of swarm targets is doubly-spread in the range-Doppler (RD) image due to spatially dispersed and relative motion between individuals. Therefore, the traditional constant false alarm rate (CFAR) detection strategies of point and range-spread rigid targets may suffer performance degradation because of failing to accumulate all energy of target in the RD image. In this paper, we first developed a swarm targets echo model. Then a novel detector based on the strong scattering points (SPs) estimation is proposed. The detector considers the swarm targets as a whole and estimates and accumulates all the target-like strong SPs to enhance the detection performance. Moreover, the radial size of swarm targets is calculated according to the range of SPs. Finally, the constant false alarm property is derived, and the strategy of detector parameters setting is discussed in detail. A thorough performance assessment is conducted by Monte Carlo simulation. The results show that the proposed method is superior to the modified cell averaging CFAR (MCA-CFAR) and deterministic scatterer model generalized likelihood ratio test (DSM-GLRT) detectors.","Title: Geometric Optimization of Distributed MIMO Radar System for Accurate Localization in Multiple Key Subareas\nYear: 2022\n\nAbstract: •To describe the surveillance and localization performance of the distributed MIMO radar system, we formulate corresponding performance evaluation metrics. Especially in order to uniformly characterize multiple localization performance requirements in multiple subareas, we establish a scaled accuracy-based localization performance metric based on GDOP.•Considering the importance differences of the performance in different subareas and the coupled relationship between different radar performance, we formulate a multi-objective optimization problem with complex coupled constraints for enhancing both the surveillance and localization performance of the system. Furthermore, we simplify the problem by reducing the number of constraints via a scaling process.•To efficiently solve the non-convex multi-objective optimization problem with complex coupled constraint, we improve the traditional MOPSO algorithm by introducing a two-stage update rule of individual best position, and further propose a novel SC-MOPSO algorithm.","Title: Resource Scheduling for Multi-Target Tracking in Multi-Radar Systems With Imperfect Detection\nYear: 2022\n\nAbstract: In this paper, an effective joint radar assignment and power scheduling (JRAPS) scheme is proposed for the multi-target tracking (MTT) task in multi-radar systems (MRSs) under imperfect detection, namely, target related measurements are collected with the probability of detection (P-D) less than 1. The centralized fusion architecture is adopted by the MRS. Specifically, during each sampling time interval, each radar of the MRS is assigned to track certain targets with controllable transmitting power. The measurements collected by all selected local radars, with P-D <= 1, are sent to a central radar to obtain the global MTT results. To maximize the global KIT task performance, the proposed JRAPS scheme implements the online target-to-radar assignment and transmitting power allocation scheme based on the feedback of the MTT results. The posterior Cramer-Rao lower bound (PCRLB) with P-D <= 1 is derived and utilized as the tracking performance metric since it provides a more accurate lower bound on the target state estimates under imperfect detection. Then, an overall cost function (OCT) is formulated based on the derived PCRLB to quantify the global MTT performance. Combined with the practical resource constraints of the MRS, the formulated TRAPS problem is shown to be non-convex. Therefore, we further propose a fast three-stage iterative method to solve this problem efficiently. Simulation results verify the superiority and effectiveness of the proposed JRAPS strategy in terms of both tracking accuracy and target detection performance.","Title: Network architecture optimization for netted MIMO radar systems with surveillance performance\nYear: 2023\n\nAbstract: •The global detection probability based on the clustering framework is derived.•An objective function considering the importance of different surveillance tasks.•Formulating a network architecture optimization problem with several constraints.•A maximum block pattern search algorithm for the nonconvex optimization problem.","Title: Efficient Radar Deep Temporal Detection in Urban Traffic Scenes\nYear: 2022\n\nAbstract: This paper explores object detection on radar range-Doppler map. Most of the radar processing algorithms are proposed for detecting objects without classifying. Meanwhile, these approaches neglect the useful information available in the temporal domain. To address these problems, we propose an online radar deep temporal detection framework by frame-to-frame prediction and association with low computation. The core idea is that once an object is detected, its location and class can be predicted in the future frame to improve detection results. The experiment results illustrate this method achieves better detection and classification performance, and shows the usability of radar data for traffic scenes.","Title: Event-Triggered Neural Network Control for Uncertain Nonlinear Systems Without State Observer*\nYear: 2022\n\nAbstract: A robust approximation-based event-triggered control method is presented for single input single output (SISO) nonlinear continuous-time systems with unmeasurable states and external disturbance. In the whole system, just one neural network (NN) is designed to approximate the unknown part in the controller, and output errors are directly used to construct the system controller and event-triggered mechanism to relieve the burden of system communication. The controller with a simple structure is easier to be realized in practical engineering. The system control signals and adaptive parameters are updated only if the event trigger condition is met, and this way further reduces the waste of network resources caused by frequent system sampling. The application of stability theory of Lyapunov proves that the weight estimation of the NN and tracking errors are ultimate and uniform boundedness, and the efficacy of the proposed scheme is verified with numerical results on a robot.","Title: The PHD Filter for Target Swarms and Its Gaussian Mixture Implementation\nYear: 2022\n\nAbstract: The swarm is used to describe a set of individual targets which are very close together in the measurement space (relative to the resolution of the sensor). In this paper, we consider to estimate the number, shape and density of target swarms in addition to their motion states. In order to deal with the uncertain number of measurement origins, we use amplitude information (AI) to construct hypotheses about the number of individual target source of the merged measurement in the probability hypothesis density (PHD) update step. On this basis, we propose the PHD filter for target swarms (S-PHD filter) and develop Gaussian mixture (GM) implementation to the resulting filter. The performance of this algorithm is demonstrated by a multi swarms scenario with merged measurements, and the results highlight the significant improvement in the estimates of the number of individuals and the density of swarms.","Title: An Improved Two-Stage Based Multi-frame Track-Before-Detect Algorithm in Radar systems\nYear: 2022\n\nAbstract: The detection and tracking of dim targets in complex environments show great importance in radar systems. The track-before-detect (TBD) technology has been widely researched in the scenario where the target signal-to-noise ratio (SNR) is low, however, there has been no suitable solution to the trade-off between tracking accuracy and computational complexity during the process of multi-frame joint detection in radar systems. In this paper, we propose an efficient two-stage based multi-frame detection and tracking algorithm in radar systems. The proposed algorithm consists of a low threshold pre-processing stage, and a TBD processor, which searches possible target tracks from multiple scans and declares the final estimated tracks. The proposed algorithm provides an accurate evolution of target states over time in polar coordinates to avoid the performance loss and model mismatch due to the nonlinear conversion in mixed coordinates. In addition, we further propose a greedy-based recursive algorithm to implement fast track formation from the over-threshold multi-frame measurement points. Simulation results show that the proposed method achieves a better detection and tracking performance with a low computational complexity.","Title: An Integrated Localization Method for Mixed Near-Field and Far-Field Sources Based on Mixed-order Statistic\nYear: 2022\n\nAbstract: This paper considers the integrated localization for the mixed near-field (NF) and far-field (FF) sources using the uniform linear array (ULA). With the help of the polynomial rooting methods and the propagator, an efficient algorithm is proposed to provide an integrated estimation of the direction of arrival (DOA) and the ranges of the sources. It takes low computational burden without the requirements that separating the DOA and range information or pre-classification of the sources. We first construct two special fourth-order cumulant matrices using the received array data, then extract the prior-electrical parameters related to the array elements by its steering matrix, and finally carry out parameter matching and classification. Besides, the proposed algorithm eliminates the need for tedious eigenvalue decomposition and spectral search steps, and has almost no aperture loss. Eventually, several simulation results show that the proposed algorithm has lower computational complexity under an acceptable accuracy, compared to the state-of-the-art methods.","Title: LPI-Based Joint Node Selection and Power Allocation Strategy for Target Tracking in Distributed MIMO Radar\nYear: 2022\n\nAbstract: This paper addresses the problem of the low probability of intercept (LPI) performance optimization for target tracking in the distributed MIMO radar systems through the transmitting resource scheduling (TRS) strategy. The mechanism of the proposed LPI-based TRS strategy is to adopt the optimization technique to collaboratively schedule the transmit radar node and signal power with the target tracking accuracy requirement, which aims to enhance the LPI performance of the overall system. Based on the existing research, we develop an intercept model to describe two stages of the signal intercept process with the specific interceptor equipped on the moving target, i.e., intercept and detection. Under the proposed model, the probability of report (PR) is proposed to evaluate the LPI performance for a single transmitting radar node. Then, we consider the maximum PR to represent the LPI performance metric for the overall system. Hence, using it as the objective function, the optimization problem is established with the constraints of the target tracking requirement and the system resource. By introducing the two-step partition-based solution, the proposed non-convex problem is solved efficiently. Finally, several numerical simulations demonstrate the theoretical calculations and validate the effectiveness of the proposed LPI-based TRS strategy.","Title: Multi-Frame Track-Before-Detect for Scalable Extended Target Tracking\nYear: 2022\n\nAbstract: This paper mainly addresses the scalable detection and tracking of the extended target in the low signal-to-noise(SNR) environment. As the appearance and shape of the extended target are constantly varied, it is challenging to achieve robust detection and tracking. For this, a novel adaptive scale (AS) kernelized correlation filter (KCF) based on multi-frame track-before-detect (MF-TBD) framework is proposed. By embedding scaling pools into the response map to handle the scale variation and accumulating target energy overall feasible trajectories, AS-MF-TBD estimates the kinematic state and geometric shapes simultaneously. Both simulation data and real radar data are used to demonstrate the superiority of the proposed method in terms of detection performance and estimation accuracy.","Title: A Fast and Robust Maneuvering Target Tracking Method without Markov Assumption\nYear: 2022\n\nAbstract: In the traditional methods of maneuvering target tracking, it is necessary to adjust the state transition model in time to match the maneuvering target, which will cause the problems of model decision delay and competition. Besides, the commonly adopted first-order Markov assumption can lead to the loss of information when motion modes are relevant to time. In order to solve these problems, a data-driven algorithm based on LightGBM is proposed in this paper. Maneuvering target tracking is modeled as a non-probabilistic method of direct mapping from sensor measurement to target state, track samples of different motion modes are used for training, and fast online tracking is realized. Comparing it with interacting multiple model (IMM) algorithm in a variety of different scenarios, simulation results show that the proposed method has advantages in accuracy and speed. Finally, the robustness of the algorithm is verified under the compound noise of Cauchy and Gaussian distributions.","Title: Arithmetic Average Based Multi-sensor TPHD Filter for Distributed Multi-target Tracking\nYear: 2022\n\nAbstract: Compared with the probability hypothesis density (PHD) filter for sets of targets, the trajectory probability hypothesis density (TPHD) filter can estimate the sets of trajectories in a principle way and has better target tracking performance. This paper aims at extending the TPHD filter to distributed multitarget tracking (MTT) for the multi-sensor system. However, in the trajectory set based distributed fusion implementation, the trajectory state difference phenomenon makes the clustering and merging techniques unfeasible in trajectory state space. To address this problem, this paper studies the space decomposition of the TPHD and proposes a distributed MTT method based on the TPHD filter with the weighted arithmetic average (WAA) fusion rule. First, we prove the rationality of the space decomposition in the posterior density of the TPHD filter. Then, based on the proposed property, we derive the WAA fusion formulation of the TPHD filter by minimizing the weighted sum of Kullback-Leibler divergences (KLD) from local posterior densities, and develop the analytical Gaussian mixture (GM) implementation with the L-scan approximation. Numerical results demonstrate the efficacy of the proposed fusion method.","Title: Modeling and Statistical Analysis of Vision Distance Measurement Uncertainty for Multi-Automotive Sensor Target Tracking\nYear: 2022\n\nAbstract: Among all multi-automotive sensor target tracking algorithms, statistical sequential estimation techniques are in the dominant position. For these techniques, statistical sensor measurement uncertainty model is essential. However, for camera, which is a primary automotive sensor, there exists a problem that vision measurements are usually not statistical model-based. In this paper, to open up a possibility of establishing an access between the vision measurements and statistics-based methods, we try to describe vision distance measurement (VDM) uncertainty in a statistical way. As the foundation for statistically analysing VDM uncertainty, we first derive the expression of VDM uncertainty. Based on the expression, we then model the probability distribution of VDM uncertainty, thus the statistical VDM uncertainty model is obtained and VDM is connected with the statistics-based fusion methods. The proposed statistical model is validated by experimental data."],"string":"[\n \"Title: Design and Implementation of the Accumulating Algorithm of Digital Signal Averager\\nYear: 2011\\n\\nAbstract: To satisfy the need of on-board accumulation in time-of-fLight mass spectrometer (TOF-MS) appLications, the paper presents an accumulation algorithm in FPGA based on PipeLine and ping-pong operation. The accumulation algorithm is verified on a signal averager hardware platform which takes XILiNX VIRTEX-5 FPGA as the processing core. The caChe module, on-chip accumulation, and USB transmission module is implemented. Testing results show that the range of time measurement is 32us, the minimum time resolution is 333ps and signal to noise ratio (SNR) is improved effectively. The accumulation algorithm is stable and reLiable, so it can be widely used in the measurement of high-resolution time interval and pulse ampLitude.\",\n \"Title: Greedy Neighborhood Search for Disaster Relief and Evacuation Logistics\\nYear: 2008\\n\\nAbstract: Coordinating relief activities and evacuation in emergency response involves delivering commodities to distribution centers in affected areas and picking up wounded people and transferring them to medical centers. This problem's key features are limited supply and vehicle availability, simultaneous delivery and pickup of commodities and wounded people, and split delivery. The resulting logistics problem is quite complicated. Furthermore, emergency settings are dynamic, and changing the logistics plan at each information update requires an efficient algorithm. A proposed greedy constructive algorithm explores a limited neighborhood around the vehicles' current locations in the partial solution and then appends a two-stop partial itinerary to each vehicle's available route. The algorithm's performance was tested on several networks and compared with that of optimal solutions.\",\n \"Title: An Efficient Multi-Frame Track-Before-Detect Algorithm for Multi-Target Tracking\\nYear: 2013\\n\\nAbstract: This paper considers the multi-target tracking (MTT) problem through the use of dynamic programming based track-before-detect (DP-TBD) methods. The usual solution of this problem is to adopt a multi-target state, which is the concatenation of individual target states, then search the estimate in the expanded multi-target state space. However, this solution involves a high-dimensional joint maximization which is computationally intractable for most realistic problems. Additionally, the dimension of the multi-target state has to be determined before implementing the DP search. This is problematic when the number of targets is unknown. We make two contributions towards addressing these problems. Firstly, by factorizing the joint posterior density using the structure of MTT, an efficient DP-TBD algorithm is developed to approximately solve the joint maximization in a fast but accurate manner. Secondly, we propose a novel detection procedure such that the dimension of the multi-target state no longer needs be to pre-determined before the DP search. Our analysis indicates that the proposed algorithm could achieve a computational complexity which is almost linear to the number of processed frames and independent of the number of targets. Simulation results show that this algorithm can accurately estimate the number of targets and reliably track multiple targets even when targets are in proximity.\",\n \"Title: Kernel Fusion: An Effective Method for Better Power Efficiency on Multithreaded GPU.\\nYear: 2010\\n\\nAbstract: As one of the most popular accelerators, Graphics Processing Unit (GPU) has demonstrated high computing power in several application fields. On the other hand, GPU also produces high power consumption and has been one of the most largest power consumers in desktop and supercomputer systems. However, software power optimization method targeted for GPU has not been well studied. In this work, we propose kernel fusion method to reduce energy consumption and improve power efficiency on GPU architecture. Through fusing two or more independent kernels, kernel fusion method achieves higher utilization and much more balanced demand for hardware resources, which provides much more potential for power optimization, such as dynamic voltage and frequency scaling (DVFS). Basing on the CUDA programming model, this paper also gives several different fusion methods targeted for different situations. In order to make judicious fusion strategy, we deduce the process of fusing multiple independent kernels as a dynamic programming problem, which could be well solved with many existing tools and be simply embedded into compiler or runtime system. To reduce the overhead introduced by kernel fusion, we also propose effective method to reduce the usage of shared memory and coordinate the thread space of the kernels to be fused. Detailed experimental evaluation validates that the proposed kernel fusion method could reduce energy consumption without performance loss for several typical kernels.\",\n \"Title: Target tracking for an unknown and time-varying number of targets via particle filtering\\nYear: 2012\\n\\nAbstract: Track initialization and adapting to time-varying number of targets are challenges for particle filter (PF) based tracking methods, because of the difficulty of monitoring changes in the entire state space with a finite number of particles. Conversely, if the state space only has limited supports (i.e. after thresholding) or state space is well localized (i.e. target prior information is known), PFs can provide remarkable estimation performance via concentrating its computational power on the areas of interest. In this paper, we present a PF to track unknown and time-varying number of targets based on the track-before-detect (TBD) framework. This PF adopts the two-layer structure of the PF in [5] (i.e. one layer for track initiation, the other for track maintenance), but with different track initiation and maintenance importance densities. It initiates new tracks and controls the false alarms based on the information provided by a maximum a posterior (MAP) method. By discretizing the state space, this MAP-based method can successfully localize the posterior density (i.e. find out the parts of the state space which are of interest) with reasonable computational cost. With regard to the track maintenance, it efficiently updates the existing tracks using a joint, measurement-directed sampling method and deletes the false tracks based on the likelihood ratio test (LRT). Numerical examples are also used to illustrate the efficacy of the proposed algorithm.\",\n \"Title: MPLS-based Multicast Shared Trees\\nYear: 2006\\n\\nAbstract: This paper presents a study of our proposed architecture for the setup of a MultiPoint-to-MultiPoint (MP2MP) Label Switched Path (LSP). This form of LSP is needed for establishing uni-directional multicast shared trees. Such trees are required for information distribution in applications such as video conferencing. The presented architecture is intended for multicast applications within a single autonomous domain and can be extended to cover inter-domain multicast sessions. We propose the use of one (or more) control points in the network called Rendez-vous Points (RP) in a simple extention of the PIM-SM protocol to implement multicast in MPLS networks. This architecture has the advantage of using existing MPLS techniques and existing routing protocols and requires only the addition of more management capabilities at the RPs. The experiments we carried out show that while retaining the advantages of using MPLS over traditional multicast routing, the performance of the new architecture is comparable to that of IP multicast in terms of the volume of control messages and label and memory consumption. Also the architecture scales well with the increase of the number of senders within a multicast group and with the increase of the number of multicast groups.\",\n \"Title: Thresholding Process Based Dynamic Programming Track-Before-Detect Algorithm\\nYear: 2013\\n\\nAbstract: Dynamic Programming (DP) based Track-Before-Detect (TBD) algorithm is effective in detecting low signal-to-noise ratio (SNR) targets. However, its complexity increases exponentially as the dimension of the target state space increases, so the exact implementation of DP-TBD will become computationally prohibitive if the state dimension is more than two or three, which greatly prevents its applications to many realistic problems. In order to improve the computational efficiency of DP-TBD, a thresholding process based DP-TBD (TP-DP-TBD) is proposed in this paper. In TP-DP-TBD, a low threshold is first used to eliminate the noise-like (with low-amplitude) measurements. Then the DP integration process is modified to only focuses on the thresholded higher-amplitude measurements, thus huge amounts of computation devoted to the less meaningful low-amplitude measurements are saved. Additionally, a merit function transfer process is integrated into DP recursion to guarantee the inheritance and utilization of the target merits. The performance of TP-DP-TBD is investigated under both optical style Cartesian model and surveillance radar model. The results show that substantial computation reduction is achieved with limited performance loss, consequently TP-DP-TBD provides a cost-efficient tradeoff between computational cost and performance. The effect of the merit function transfer on performance is also studied.\",\n \"Title: A Computationally Efficient Particle Filter for Multitarget Tracking Using an Independence Approximation\\nYear: 2013\\n\\nAbstract: Particle filter (PF) based multi-target tracking (MTT) methods suffer from the curse of dimensionality. Existing strategies to combat this assume posterior independence between target states, in order to then sample targets independently, or to perform joint sampling of closely spaced targets only. When many targets are in proximity, these strategies either perform poorly or are too computationally expensive. We make two contributions towards addressing these limitations. Firstly, we advocate an alternative view of the use of posterior independence which emphasizes the statistical effect of assuming posterior independence on the Monte Carlo (MC) approximation of posterior density. Our analysis suggests that assuming posterior independence can provide a better MC approximation of the prior distribution at the next time, and therefore the posterior at the next time, without regard for how sampling is performed. Secondly, we present a computationally efficient, measurement directed, joint sampling method to cope with the target coupling and measurement ambiguity when targets are near each other. Consequently, we develop a PF which employs posterior independence while sampling targets jointly. This PF is applicable to both the traditional thresholded and track-before-detect style pixelized models. Simulation results for a challenging tracking scenario show that the proposed PF substantially outperforms existing approaches.\",\n \"Title: A dynamic logistics coordination model for evacuation and support in disaster response activities\\nYear: 2007\\n\\nAbstract: This paper describes an integrated location-distribution model for coordinating logistics support and evacuation operations in disaster response activities. Logistics planning in emergencies involves dispatching commodities (e.g., medical materials and personnel, specialised rescue equipment and rescue teams, food, etc.) to distribution centres in affected areas and evacuation and transfer of wounded people to emergency units. During the initial response time it is also necessary to set up temporary emergency centers and shelters in affected areas to speed up medical care for less heavily wounded survivors. In risk mitigation studies for natural disasters, possible sites where these units can be situated are specified according to risk based urban structural analysis. Logistics coordination in disasters involves the selection of sites that result in maximum coverage of medical need in affected areas. Another important issue that arises in such emergencies is that medical personnel who are on duty in nearby hospitals have to be re-shuffled to serve both temporary and permanent emergency units. Thus, an optimal medical personnel allocation must be determined among these units. The proposed model also considers this issue.\",\n \"Title: Phase-Modulated Waveform Evaluation and Selection Strategy in Compound-Gaussian Clutter\\nYear: 2013\\n\\nAbstract: This correspondence introduces a waveform evaluation and selection strategy (ESS) for application of optimized phase-modulated (PM) waveforms to mitigate compound-Gaussian clutter in radar. This strategy addresses the trade-off between potential improvement in detection performance from adaptive design of PM waveforms and the cost this entails in terms of system resources. An evaluation factor is introduced to quantify the performance improvement obtained by design and transmission of an optimized PM waveform in the second sub-dwell of a two-stage clutter mitigation scheme versus using linear frequency modulated (LFM) waveform in a sub-dwell. This evaluation factor provides the basis for the proposed ESS, the efficacy of which is demonstrated in simulations.\",\n \"Title: Automatic Aircraft Recognition Using Maximum Likelihood Ratio Test\\nYear: 2001\\n\\nAbstract: The automatic aircraft recognition in aerial images has been a topic receiving the maximum attention in the literature for more than two decades due to its practical value. Since the original images may be obtained in various uncertain environment, defects such as noise/background clutter, skewness, partial occlusion and blue may affect the image quality which lead to high environment dependence of some existing aircraft detection systems/ algorithms. This paper proposes a novel approach based on the Maximum Likelihood Ratio Test (MLRT) to extract different aircraft from the real aerial images. Objects in the tested images are corrupted randomly by the various defects. Without any extra information, the successful detection rate of the aircraft reaches 96% and that of the nonplane objects reaches 100% when only six aircraft references are chosen. The experiments are carried out to test sensitivity of the proposed method to the noise and partial occlusion. The comparisons with the Moment Fourier Descriptors (MFD) are also given.\",\n \"Title: Adaptive detection and estimation for an unknown occurring interval signal in correlated Gaussian noise.\\nYear: 2015\\n\\nAbstract: This paper considers the problem of detecting and estimating an unknown occurring interval signal in correlated Gaussian noise, which is often arisen in signal processing society, e.g., identifying the onset times of a seismic wave and detecting a distributed target in unknown occurring range cells. We propose the novel Generalized Likelihood Ratio Test (GLRT) algorithm, where the Maximum Likelihood Estimations (MLEs) of the unknown occurring interval are obtained through a Dynamic Programming (DP) method adaptively without the secondary data. Unlike the classic Sequential Probability Ratio Test (SPRT) methods which consist of an on-line detector before an off-line estimator, the proposed GLRT outputs the decision result and the estimations at the same time. The performances of the proposed algorithm are evaluated by numerical simulations as well as the applications of detecting and suppressing the transient interference in a radar operated on High-Frequency (HF) band with real data. HighlightsWe propose an adaptively GLRT based on DP method without any secondary data and off-line estimator.We evaluate the performance by computer simulations and a practical application in the HF radar with real data.The proposed GLRT shows better performance than the classic methods.The practical process result illustrates the effective performance.\",\n \"Title: A Fast Maneuvering Target Motion Parameters Estimation Algorithm Based on ACCF\\nYear: 2015\\n\\nAbstract: This letter considers the motion parameters estimation problem for a maneuvering target with arbitrary parameterized motion. The slant range of the target is modeled as a polynomial function in terms of its multiple motion parameters and a fast estimation method based on adjacent cross correlation function (ACCF) is proposed, where the iterative adjacent cross correlation operation is employed to remove the range migration and reduce the order of Doppler frequency migration. Then the motion parameters are estimated via Fourier transform. Compared with the generalized Radon Fourier transform (GRFT), the proposed method can estimate the parameters without searching procedure and acquire close estimation performance at high signal-to-noise ratio (SNR) with a much lower computational cost. Finally, simulations are provided to demonstrate the effectiveness.\",\n \"Title: Coherent Integration for Maneuvering Target Detection Based on Radon-Lv’s Distribution\\nYear: 2015\\n\\nAbstract: This letter considers the coherent integration problem for a maneuvering target, involving range migration (RM) and Doppler frequency migration (DFM) within one coherent pulse interval. A new coherent integration method, known as Radon-Lv's distribution (RLVD), is proposed. It can not only eliminate the RM effect via jointly searching in the target's motion parameters space, but also remove the DFM and achieve the coherent integration via Lv's distribution (LVD). Finally, several simulations are provided to demonstrate the effectiveness. The results show that for detection ability, the proposed method is superior to the moving target detection (MTD), Radon-Fourier transform (RFT), and Radon-fractional Fourier transform (RFRFT) under low signal-to-noise-ratio (SNR) environment.\",\n \"Title: Approximation to independent lognormal sum with α-µ distribution and the application\\nYear: 2015\\n\\nAbstract: This paper considers the statistical distribution of the sum of independent and possibly non-identically distributed lognormal random variables (RVs). The α-µ distribution is employed to approximate the probability of density function (PDF) of the lognormal sum. Two approaches, e.g., the moment-based and non-linear least square based (LS-based) methods, are used to obtain the parameters of α-µ distribution. Accounting for maximal-ratio combining (MRC) receiver, an analytical expression of the average bit error probability (ABEP) of the M-ary noncoherent frequency-shift keying (M-NFSK) modulation is derived. Finally, the performances of the α-µ approximation and the ABEP are evaluated via numerical experiments. HighlightsWe adopt the α-µ distribution to approximate the probability density function of the sum of independent and possibly non-identically distributed lognormal variables.We investigate two typical parameter estimation methods, i.e., the moment-based method and the non-linear least square based method.For maximal-ratio combining receiver, we derive an analytical expression of the average bit error probability of the M-ary noncoherent frequency-shift keying modulation resorting to the new α-µ approximation.\",\n \"Title: A New Differential Raid For High Reliable All Flash Array\\nYear: 2014\\n\\nAbstract: For high I/O performance, Solid State Drive (SSD)-based all Flash arrays (AFAs) are widely employed to larger scale storage systems. However, the aging problem of SSDs limits the lifetime and reliability of AFA. In order to extend the lifetime of AFA, we designed a capacity-based differential RAID (Redundant Array of Independent Disks) scheme (CDiff-RAID). In the proposed scheme, the array is initialized by a group of SSDs with unequal capacities. The smaller SSDs suffer more erase operations and age quickly. To maintain the age differential, the worn SSDs are continually replaced by new SSDs. Compared with parity-based Diff-RAID, the age differential in our scheme is independent of workload and the time-consuming of reconstruction is greatly reduced. We also evaluated the I/O performance of CDiff-RAID using SSD simulators. The proposed scheme outperforms Diff-RAID scheme in sequential and random traces in most cases.\",\n \"Title: Joint multi-target detection and localization with a noncoherent statistical MIMO radar\\nYear: 2014\\n\\nAbstract: In this paper, the problems of simultaneously detecting and localizing multiple targets are considered for noncoherent multiple-input multiple-output (MIMO) radar with widely separated antennas. By assuming a prior knowledge of target number, an optimal solution to this problem is presented first. It is essentially a maximum-likelihood (ML) estimator searching parameters of interest in a high-dimensional space. However, the complexity of this method increases exponentially with the number of targets. Besides, without the prior information of target number, a multi-hypothesis test strategy on target number is required, which further complicates this method. Therefore, we split the joint maximization into G disjoint optimization problems by clearing the interference from previously declared targets. In this way, we derive two fast but robust suboptimum solutions which allow trading performance for a much lower implementation complexity which is almost independent of the number of targets. In addition, the multi-hypothesis test detector is no longer required when target number is unknown. Simulation results show the proposed suboptimum algorithms can correctly detecting and accurately localizing multiple targets even when targets share common range bin in some paths.\",\n \"Title: Recursive filtering for target tracking in multi-frame track-before-detect\\nYear: 2014\\n\\nAbstract: This paper addresses the tracking problem with respect to dynamic programming (DP) based track-before-detect (DP-TBD). Basically, DP-TBD is a grid-based method performing search in the discretized state space, therefore the accuracy of its estimation suffers at least half a grid loss. Moreover, different from the conventional detection procedure which provides isolate point plots to a tracker, DP-TBD is a batch processor whose detection outputs are short state sequences in time series. Essentially, these short tracks are neither filtered target trajectories nor conventional point detections. Meantime the false alarms and missing reports of these short tracks also exist especially when target signal-to-noise ratio (SNR) is low. Thus, a tracker which can associate, smooth and finally combine these short tracks into continuous target trajectories is needed. In this work, we propose a two-stage, i.e., prediction and updating stages, recursive filtering algorithm for DP-TBD. The key idea is using the outputted short tracks after every DP-TBD batch processing as the input measurements for the updating stage of filtering. Simulation results show that the proposed algorithm can correctly estimate target trajectories and significantly enhance the tracking accuracy compared with the method without our filtering algorithm.\",\n \"Title: Coherent Integration Algorithm For A Maneuvering Target With High-Order Range Migration\\nYear: 2015\\n\\nAbstract: This paper considers the coherent integration problem for a maneuvering target with complex motions, where the velocity, acceleration and jerk result in respectively the first-order range migration (FRM), second-order range migration (SRM) and third-order range migration (TRM) within the coherent pulse interval. A new coherent integration algorithm based on generalized keystone transform (KT) and secondorder dechirp process is proposed, which employs the third-order KT, six-order KT, second-order KT and fold factor searching to correct the TRM, SRM and FRM, respectively. The range migration change during each step and computational complexity are also theoretically analyzed. Compared with the generalized Radon Fourier transform (GRFT) algorithm, the presented method can avoid the blind speed sidelobe (BSSL) and acquire close integration performance but with much lower computational cost. Simulations are provided to demonstrate the effectiveness. Finally, a generalized method, named generalized KT and generalized dechirp process (GKTGDP), is also introduced for the maneuvering target with arbitrary high-order range migration.\",\n \"Title: Track-before-detect strategies for range distributed target detection in compound-Gaussian clutter\\nYear: 2016\\n\\nAbstract: This paper considers range distributed target detection in compound-Gaussian clutter, which is often arisen in high resolution radars. Two dynamic programming based track-before-detect (DP-TBD) strategies, i.e., noncoherent integration based DP-TBD (NCI-DP-TBD) and the generalized likelihood ratio test based DP-TBD (GLRT-DP-TBD) strategies, are proposed to address the detection problem. The detection performances of the proposed strategies are examined via numerical simulations. HighlightsRange distributed targets in compound-Gaussian clutter is considered.Noncoherent integration and GLRT based DP-TBD strategies are proposed.A covariance matrix estimation strategy is proposed.\",\n \"Title: Distributed Fusion of Labeled Multi-Object Densities Via Label Spaces Matching.\\nYear: 2016\\n\\nAbstract: In this paper, we address the problem of the distributed multi-target tracking with labeled set filters in the framework of Generalized Covariance Intersection (GCI). Our analyses show that the space mismatching phenomenon, which means the same realization drawn from spaces of different sensors does not have the same implication, is quite common in practical scenarios and may bring serious problems. Our contributions are two-fold. Firstly, we provide a principled mathematical definition of label spaces matching (LS-DM) based on information divergence, which is also referred to as LS-M criterion. Then, to handle the LS-DM, we propose a novel two-step distributed fusion algorithm, named as GCI fusion via spaces matching (GCI-LSM). The first step is to match the spaces from different sensors. To this end, we build a ranked assignment problem and design a cost function consistent with LS-M criterion to seek the optimal solution of matching correspondence between spaces of different sensors. The second step is to perform the GCI fusion on the matched space. We also derive the GCI fusion with generic labeled multi-object (LMO) densities based on LS-M, which is the foundation of labeled distributed fusion algorithms. Simulation results for Gaussian mixture implementation highlight the performance of the proposed GCI-LSM algorithm in two different tracking scenarios.\",\n \"Title: Principled Random Finite Set Approximations of Labeled Multi-Object Densities.\\nYear: 2016\\n\\nAbstract: As a fundamental piece of multi-object Bayesian inference, multi-object density has the ability to describe the uncertainty of the number and values of objects, as well as the statistical correlation between objects, thus perfectly matches the behavior of multi-object system. However, it also makes the set integral suffer from the curse of dimensionality and the inherently combinatorial nature of the problem. In this paper, we study the approximations for the universal labeled multi-object (LMO) density and derive several principled approximations including labeled multi-Bernoulli, labeled Poisson and labeled independent identically clustering process based approximations. Also, a detailed analysis on the characteristics (e.g., approximation error and computational complexity) of the proposed approximations is provided. Then some practical suggestions are made for the applications of these approximations based on the preceding analysis and discussion. Finally, an numerical example is given to support our study.\",\n \"Title: Distributed Fusion with Multi-Bernoulli Filter based on Generalized Covariance Intersection.\\nYear: 2016\\n\\nAbstract: In this paper, we propose a distributed multiobject tracking algorithm through the use of multi-Bernoulli (MB) filter based on generalized covariance intersection (G-CI). Our analyses show that the G-CI fusion with two MB posterior distributions does not admit an accurate closed-form expression. To solve this problem, we first approximate the fused posterior as the unlabeled version of δ-generaliz...\",\n \"Title: Fast Non-Searching Method for Maneuvering Target Detection and Motion Parameters Estimation.\\nYear: 2016\\n\\nAbstract: This paper considers the coherent integration problem for maneuvering target detection and motion parameters estimation, involving range migration (RM) and Doppler frequency migration (DFM) within the coherent pulse interval. A fast non-searching method based on adjacent cross correlation function (ACCF) and Lv's distribution (LVD) is proposed, where the adjacent correlation operation is first emp...\",\n \"Title: Fast Optimal Antenna Placement for Distributed MIMO Radar with Surveillance Performance\\nYear: 2015\\n\\nAbstract: In this paper we demonstrate an optimization problem of antenna placement of distributed multi-input multi-output (MIMO) radar. To evaluate the surveillance performance of the radar system, a coverage ratio is proposed as a criterion. Since the problem is of extremely huge computational complexity due to its complicated objective function and high dimensionality, we propose a solution that contains two parts: 1) a low-complexity method to simplify the objective function; 2) a placement algorithm based on particle swarm optimization (PSO) to deal with the challenge of high dimensionality. We also analyse the computational complexity of our solution. Simulation results verify the validity and advantage in computational complexity of our solution. Our contributions include a novel optimization placement model of distributed MIMO radar and a computational efficient solution to establish the optimal positions of antennas.\",\n \"Title: Performance prediction of OS-CFAR for generalized swerling-chi fluctuating targets.\\nYear: 2016\\n\\nAbstract: We address the performance prediction of the ordered-statistic constant false alarm rate (OS-CFAR) detector for arbitrarily correlated and possibly nonidentically distributed target echoes in the presence of nonhomogeneous background. Accounting for the generalized Swerling-Chi fluctuating model, including the well-known Swerling models as special cases, we develop the analytic expressions of detection probability in terms of converging series in nonhomogeneous background involving clutter edges and possibly isolated point-like interference. At the analysis stage, we study the truncation error of the derived expressions and then evaluate the impacts on detection performance of the target parameters via numerical simulations.\",\n \"Title: A Flash-Aware Intra-Disk Redundancy Scheme For High Reliable All Flash Array\\nYear: 2015\\n\\nAbstract: In SSD-based All Flash Array (AFA), Redundant Array of Independent Disks (RAID) technologies are widely used to cope with drive failures and sector errors. However, the redundancy in disk level is either insufficient like RAID-5 or excessive like RAID-6. In order to combat the two types of errors, we combine the RAID-5 scheme with a Flash-aware Intra-disk Redundancy (FAIDR) scheme. For saving parity space, the ratio of redundancy in the FAIDR scheme grows with the increasing of sector errors. However, the parity handling overhead is significant. As NAND Flash blocks can't be updated before erase, we establish all FAIDR stripes on the physical page number of blocks regardless of the page state. Thus, the parity update operations in FAIDR are eliminated. Simulation results show that the proposed scheme has the ability to ensure the AFA's reliability in its all lifetime. It gains 41 percent storage capacity in comparison to RAID-6. And it only leads to very small performance loss compared with pure RAID-5.\",\n \"Title: A Case Study of SWIM: Optimization of Memory Intensive Application on GPGPU\\nYear: 2010\\n\\nAbstract: Recently, GPGPU has been adopted well in the High Performance Computing (HPC) field. The limited global memory bandwidth poses a great challenge to many GPGPU programmers trying to exploit parallelism within the CPU-GPU heterogeneous platform. In this paper, we choose SWIM, a typical memory intensive application from the SPEC OMP 2001 benchmark suite, for case study. We attempt to optimize the performance and energy consumption of the application utilizing different memory access mechanisms and present optimization methods including matrix transposition and kernel fusion. The experimental results on the Intel Core TM i920 CPU plus GeForce GTX 295 platform shows that, the proposed optimizing methods achieve a speedup of 8.7X over the original OpenMP program and reduce the energy consumption by 83% for the problem size of 2048*2048.\",\n \"Title: CFAR detector based on clutter partition in heterogeneous background\\nYear: 2015\\n\\nAbstract: The conventional CFAR detectors based on reference window suffer a considerable performance degradation in heterogeneous background. In this paper, a new knowledge based constant false alarm rate (KB-CFAR) detector is proposed to reduce the impact of heterogeneous environment. The proposed KB-CFAR detector exploits clutter partition as auxiliary knowledge to select the reference cells. The clutter partition is obtained by extracting useful knowledge from the massive environment information and it is updated by real-time radar returns. The performance of the new KB-CFAR detector is analyzed by real radar data which is collected by a linear frequency-modulated continuous wave (LFMCW) radar. The results show that the new detector achieves a satisfactory performance and works well under general conditions.\",\n \"Title: Radar maneuvering target detection and motion parameter estimation based on TRT-SGRFT.\\nYear: 2017\\n\\nAbstract: In this paper, we address the detection and motion parameter estimation problem for a maneuvering target with complex motions, where range migration (RM, including range walk and range curvature) and Doppler frequency migration (DFM) occur during the coherent observation time. A method based on time reversing transform (TRT) and special generalized Radon Fourier transform (SGRFT), i.e., TRT-SGRFT, is proposed to achieve the detection and motion parameters estimation. More specifically, the TRT is firstly applied to separate the motion parameters and reduce the order of RM and DFM. Then the SGRFT operation is employed to estimate part of the motion parameters. After that, the residual motion parameters could be achieved via the second SGRFT operation. The advantage of the TRT-SGRFT algorithm is that it could avoid the blind speed sidelobe (BSSL) effect and obtain a good tradeoff between the computational cost and estimation performance, in comparison with generalized Radon Fourier transform. Several numerical experiments are provided to demonstrate the its effectiveness. HighlightsThe detection and motion estimation problem for a maneuvering target with arbitrary parameterized motion is addressed.A method based on TRT-SGRFT is proposed to achieve the maneuvering target detection and motion parameters estimation.The TRT-SGRFT algorithm can avoid the BSSL and obtain a good tradeoff among computational cost, detection ability and parameter estimation performance.\",\n \"Title: Joint estimation of location and signal parameters for an LFM emitter.\\nYear: 2017\\n\\nAbstract: This paper considers the localization of an emitter employing linear frequency modulation (LFM) signal with unknown parameters. The proposed approach is based on jointly estimating the emitter position along with the transmitted signal parameters. In the maximum likelihood context, the joint estimation problem is shown to translate into an optimization search in a six-dimensional space. In many practical applications, such a search may be too computationally expensive to run in real-time. We suggest a sub-optimal yet fast solution based on combining the direct position determination (DPD) method with a short time Fourier transform (STFT). Despite the sub-optimality, the proposed DPD-STFT method is able, with low complexity, to achieve higher localization accuracies than methods that are based on estimation of emitter position only (regardless of signal parameters). Indeed, due to joint estimation of emitter position and the transmitted signal parameters, our method harnesses more information for localization purposes. Simulation results show that the proposed method outperforms the existing DPD algorithms, in terms of localization error. HighlightsAn optimal strategy to jointly estimate location and parameters of an LFM emitter.An algorithm to implement localization, based on combining STFT with DPD method.Inclusion of a novel quality factor to asses and improve localization performance.Analysis of computational complexity of the proposed algorithm.\",\n \"Title: Suboptimum Low Complexity Joint Multi-target Detection and Localization for Noncoherent MIMO Radar with Widely Separated Antennas.\\nYear: 2017\\n\\nAbstract: In this paper, the problems of simultaneously detecting and localizing multiple targets are considered for noncoherent multiple-input multiple-output (MIMO) radar with widely separated antennas. By assuming a prior knowledge of target number, an optimal solution to this problem is presented first. It is essentially a maximum-likelihood (ML) estimator searching parameters of interest in a high dimensional space. However, the complexity of this method increases exponentially with the number G of targets.Besides, without the prior information of the number of targets, a multi-hypothesis testing strategy to determine the number of targets is required, which further complicates this method. Therefore, we split the joint maximization into G disjoint optimization problems by clearing the interference from previously declared targets. In this way, we derive two fast and robust suboptimal solutions which allow trading performance for a much lower implementation complexity which is almost independent of the number of targets. In addition, the multi-hypothesis testing is no longer required when target number is unknown. Simulation results show the proposed algorithms can correctly detect and accurately localize multiple targets even when targets share common range bins in some paths.\",\n \"Title: Fast coherent integration for maneuvering target with high-order range migration via TRT-SKT-LVD.\\nYear: 2016\\n\\nAbstract: This paper addresses the coherent integration problem for detecting a maneuvering target with high-order range migration (RM), where the velocity, acceleration, and jerk result in, respectively, the first-order range migration (FRM), second-order range migration (SRM), and third-order range migration (TRM) within the coherent pulse interval. A fast coherent integration method based on time reversing transform (TRT), second-order keystone transform (SKT) and Lv's distribution (LVD), i.e., TRT-SKT-LVD, is proposed. In this method, TRT operation is first presented to remove the TRM and FRM simultaneously. After that, the SKT is employed to correct the remaining SRM and then the coherent accumulation could be obtained via LVD. The proposed algorithm is simple and fast in that it can be easily implemented by using complex multiplications, the fast Fourier transform (FFT), and inverse FFT (IFFT). Compared with the existing methods, the presented method can realize the coherent integration without any searching procedure and can achieve a good balance between the computational cost and detection ability. Both simulated and real data demonstrate the effectiveness of the proposed algorithm.\",\n \"Title: Performance prediction of noncoherent detector for independent and nonidentically distributed NCG fluctuating targets.\\nYear: 2016\\n\\nAbstract: We address the performance prediction of the noncoherent detector (i.e., squared-law plus integrator) for independent but possibly nonidentically distributed (non-id) noncentral Gamma (NCG) fluctuating target model, which is more general and can describe more complex targets. Initially, we obtain the probability density function (pdf) of the sum of independent and non-id NCG random variables in terms of an infinite sum of Gamma pdfs. Furthermore, we develop an analytic expression of the detection probability in terms of converging series based on the generalized Marcum Q function. Finally, at the analysis stage, we first study the truncation error of the derived expression and then evaluate the impacts on the detection performance of the target fluctuating parameters via numerical simulations.\",\n \"Title: A novel mobile data plan design from the perspective of data period\\nYear: 2016\\n\\nAbstract: In the conventional mobile data plans, the unused data will be cleared at the end of every period (e.g., month). To take advantage of customers' heterogeneous demands across different periods and to provide more flexible data service, some mobile data service providers (SP) have offered data plans with different lengths of period. In this paper, we consider the data plan design problem for a single SP, who provides data plans with different lengths of period to consumers with different characteristics of data demands. We propose a contract-theoretic approach, where the SP offers a period-price contract consisting of a set of period-price combinations. Each period-price combination is designed for a specific type of consumers with a specific data demand distribution. We systematically analyze the incentive compatibility (IC) constraint and individual rationality (IR) constraint which ensure each consumer to purchase the data plan with the period-price combination intended for his type. Based on the IC and IR analysis, we further derive the optimal contract that maximizes the SP's expected revenue. Our numerical results show that our proposed optimal contract can achieve 94% of the maximum social surplus, and can increase the SP's revenue by 25% comparing with the conventional monthly-period data plan.\",\n \"Title: Detection of weak maneuvering target based on keystone transform and matched filtering process.\\nYear: 2017\\n\\nAbstract: •The problems of range migration and Doppler frequency migration for the weak maneuvering target are addressed.•A method based on keystone transform and matched filtering process, i.e., KT-MFP, is proposed to obtain the coherent integration and achieve the maneuvering target detection.•The KT-MFP can obtain superior coherent integration and detection performance for both the slowmoving and fast-moving target, in comparison with the existing algorithms.•The KT-MFP requires lower computational cost without the BSSL effect, in compared with GRFT. April 5, 2017 DRAFT.\",\n \"Title: Sequence-Reversing Transform-Based Coherent Integration for High-Speed Target Detection.\\nYear: 2017\\n\\nAbstract: This paper addresses the coherent integration problem for detecting a high-speed target, involving range walk (RW) within the coherent processing time. A fast method based on sequence reversing transform is proposed to eliminate the RW and achieve the coherent accumulation. The proposed method is simple and fast since it can be easily implemented by using complex multiplications, the fast Fourier transform (FFT) and the inverse FFT. Compared with the Radon-Fourier transform and moving target detection methods, the presented algorithm can avoid the blind speed sidelobe and make a good balance between the computational cost and the detection ability. Finally, several simulations are provided to demonstrate its effectiveness.\",\n \"Title: Multiframe Radar Detection of Fluctuating Targets Using Phase Information.\\nYear: 2017\\n\\nAbstract: This paper considers the detection of fluctuating targets via dynamic-programming-based track-before-detect (DP-TBD) in radar systems. Swerling targets of types 0, 1, and 3 are considered. DP-TBD usually integrates either squared amplitude or the logarithm of the envelope likelihood ratio (LELR) scoring functions. Thus, only amplitude information is taken into account regardless of the fact that the measurements are often complex valued. In this paper, the phase information is used in the integration process of DP-TBD to enhance radar detection performance. More precisely, the logarithm of the complex-measurement-based likelihood ratio (LCLR) is used, taking the place of squared amplitude or the LELR. First, we derive the expressions for the LELR and the LCLR for the three Swerling types. Then, to reduce the complexity of computing LELR and LCLR, we also propose efficient but accurate approximations for the LELR and the LCLR. Simulations are used to assess the performance of different DP-TBD strategies.\",\n \"Title: The Multiple Model Vo-Vo Filter.\\nYear: 2017\\n\\nAbstract: A jump Markov system (JMS) multiple model solution is presented for tracking highly dynamic targets using the Vo–Vo filter (also known as generalized labeled multi-Bernoulli or GLMB filter). The closed-form solution is derived for Gaussian mixture implementation of the multiple-model Vo–Vo filter. The performance of the proposed method is examined and compared with the state of the art in challenging scenarios involving numerous appearing and disappearing targets with randomly time-varying dynamics. The results demonstrate that in such applications, the proposed method can outperform the competing techniques in terms of tracking accuracy, and is highly robust to variations in application and filter parameters such as clutter rate and model transition probabilities in the JMS.\",\n \"Title: ISAR imaging of maneuvering target with complex motions based on ACCF-LVD\\nYear: 2015\\n\\nAbstract: This paper considers the inverse synthetic aperture radar (ISAR) imaging problem for a maneuvering target with complex motions, involving range migration (RM) and Doppler frequency migration (DFM) within the coherent integration period of radar imaging, which will degrade the imaging quality. A nonsearching ISAR imaging algorithm based on adjacent cross correlation function (ACCF) and Lv's distribution (LVD), i.e., ACCF-LVD, is proposed, where the ACCF is applied to correct the RM and reduce the order of DFM. Then the motion parameters are estimated via LVD and Fourier transform. With the estimated motion parameters, high quality ISAR images can be achieved. The advantage of the presented method is that it can estimate the motion parameters under low signal-to-noise ratio (SNR) without searching procedures. Finally, several simulation examples are shown to confirm the validity of the proposed algorithm.\",\n \"Title: Design and Simulation of a Self-adaptive Fuzzy-PID Controller for an Autonomous Underwater Vehicle.\\nYear: 2017\\n\\nAbstract: In recent years, AUV (autonomous underwater vehicle) has been applied to many fields, such as offshore oil exploitation, underwater target detection, military applications and so on, which raises a higher demand for accuracy control of AUV. In this paper, for the rim-drive dead zone and system nonlinearity problem of “Swordfish II” autonomous underwater vehicle in the actual control process, according to the shape and motion characteristics of the AUV, the six DOF (degrees of freedom) dynamic model is established under MATLAB/Simulink. The fuzzy control theory and the traditional PID control algorithm are combined to design a fuzzy-PID controller. Finally, we construct the simulation system. The simulation results show that the designed controller can control the AUV model and solve the dead zone problem well, and the fuzzy PID controller has better control effect than the traditional pid controller.\",\n \"Title: Robust Distributed Fusion with Labeled Random Finite Sets.\\nYear: 2018\\n\\nAbstract: This paper considers the problem of the distributed fusion of multiobject posteriors in the labeled random finite set filtering framework, using a generalized covariance intersection (GCI) method. Our analysis shows that GCI fusion with labeled multiobject densities strongly relies on label consistencies between local multiobject posteriors at different sensor nodes, and hence suffers from a sever...\",\n \"Title: Track-Before-Detect Strategies for Radar Detection in G0-Distributed Clutter.\\nYear: 2017\\n\\nAbstract: This paper considers target detection via dynamic-programming based track-before-detect (DP-TBD) for radar systems. The clutter is modeled usingenlr G0 distribution, which is usually used to model clutter received from high-resolution radars and radars working at small grazing angles. Two target models, namely, Swerling 0 and 1 models, are considered to capture the radar cross section changes over time. DP-TBD techniques that integrate amplitude suffer from significant performance loss in this case due to the high likelihood of target-like outliers. In this paper, the log-likelihood ratio (LLR) is used in the integration process of DP-TBD, taking the place of amplitude, to enhance radar detection performance. The expressions for the LLR for the above target models are derived first. However, neither of them has a closed-form solution. In order to reduce the complexity of evaluating the LLR, efficient but accurate approximation methods are proposed. Then the approximated LLR is used in the integration process of DP-TBD. Simulations are used to examine the efficiency of the approximation methods as well as the performances of different DP-TBD strategies.\",\n \"Title: Joint Node Selection and Power Allocation Strategy for Multitarget Tracking in Decentralized Radar Networks.\\nYear: 2018\\n\\nAbstract: Networked radar systems have been demonstrated to offer enhanced target tracking capabilities. An effective radar resource allocation strategy can efficiently optimize system parameters, leading to performance enhancements. In this paper, two critical but limited system resources are considered for optimization: the number of radar nodes and the transmitted power. In this scenario, a joint node se...\",\n \"Title: Multiobject Tracking for Generic Observation Model Using Labeled Random Finite Sets.\\nYear: 2018\\n\\nAbstract: This paper presents an exact Bayesian filtering solution for the multiobject tracking problem with the generic observation model. The proposed solution is designed in the labeled random finite set framework, using the product styled representation of labeled multiobject densities, with the standard multiobject transition kernel and no particular simplifying assumptions on the multiobject likelihoo...\",\n \"Title: Data-Dependent Clustering-CFAR Detector in Heterogeneous Environment.\\nYear: 2018\\n\\nAbstract: This paper devises a new constant false alarm rate (CFAR) detection scheme to deal with the problem of radar target detection in heterogeneous environment. The proposed scheme, called clustering-CFAR detector, is data dependent and composed of three stages: an adaptive clustering procedure that, exploiting the recorded measurements of the clutter environment, divides the detection area into different classes to provide auxiliary information, a dynamic reference cell selector that chooses appropriate secondary data according to the classes, and a conventional CFAR processor to make the final decision about the target presence. The performance of clustering-CFAR detector is analyzed by computer simulation and public radar measured data (IPIX data and MSTAR data), and compared with existing CFAR detectors. The results show that the new detector achieves a better performance in the aspects of terrain classification, control of false alarm points, and probability of detection.\",\n \"Title: An Efficient Recursive Multiframe Track-Before-Detect Algorithm.\\nYear: 2018\\n\\nAbstract: Multiframe track-before-detect (MF-TBD) usually uses sliding-window-based batch processing, where a number N of the latest data frames are jointly processed at each measurement time. The sliding window mechanism compromises the operating efficiency of MF-TBD by increasing both computational costs and memory requirements, thus, heavily restricting its application in practical problems. In this paper, an improved recursive implementation for MF-TBD is proposed. Unlike the sliding-window-based implementation, the proposed method calculates the merit function, a measure of the possibility that a state is target originated, of the current batch based on an approximated recursive relationship between the merit functions of consecutive batches. As a result, instead having to process the whole batch, at any given time only the latest frame needs to be processed. The recursive relationship is first derived for any arbitrary merit function, and then explored further with several typical merit functions that are used in MF-TBD. Both the theoretical analysis and simulation results demonstrate that the proposed method can achieve almost N times reduction in computational complexity and memory requirements with negligible performance loss.\",\n \"Title: A Page Lifetime-Aware Scrubbing Scheme For Improving Reliability Of Flash-Based Ssd\\nYear: 2017\\n\\nAbstract: Solid-state drive (SSD) has gain prevalence in consumer and enterprise storage markets. However, its reliability is declining with the wearing of the Flash memory. The error correcting codes (ECCs) are generally applied in SSD to cope with bit errors caused by abrasion, however, the lifetime of pages in Flash is still limited to their fixed correctability. Due to different process and usage, the pages of Flash deteriorate in different speeds, which limited the service time of SSD. In order to prolong the lifetime of SSD, we proposed a Page Lifetime-aware Scrubbing (LaScrub) scheme to find dangerous pages which exist more bit errors. The simulation results show the proposed scheme is able to enhance the reliability of SSD. It gains 86% storage space compared with normal SSD at the same reliability requirement.\",\n \"Title: Robust position control of hydraulic manipulators using time delay estimation and nonsingular fast terminal sliding mode.\\nYear: 2018\\n\\nAbstract: A simple and robust tracking controller based on time delay estimation and nonsingular fast terminal sliding mode is proposed for the position control of hydraulic manipulator. The proposed controller does not require the mathematical model of hydraulic manipulator dynamics, which ensures that the method is simple and model free. Two elements, time delay estimation and nonsingular fast terminal sliding mode, are implemented in the proposed controller. Time delay estimation estimates the complex dynamics of the hydraulic manipulator, and nonsingular fast terminal sliding mode is implemented as the nonlinear desired error dynamic to ensure fast convergence and high trajectory tracking accuracy. Experiments are performed to verify the tracking performance of the method on a hydraulic manipulator. The results demonstrate that this method achieves faster and higher precision position tracking performance than a conventional time delay control with linear error dynamics.\",\n \"Title: Event-Triggered Communication Mechanism for Distributed Flocking Control of Nonholonomic Multi-agent System.\\nYear: 2018\\n\\nAbstract: As the scale of multi-agent systems (MAS) increases, communication becomes a bottleneck. In this paper, we propose an event-triggered mechanism to reduce the inter-agent communication cost for the distributed control of MAS. Communication of an agent with others only occurs when event triggering condition (ETC) is met. In the absence of communication, other agents adopt an estimation process to acquire the required information about the agent. Each agent has an above estimation process for itself and another estimation based on Kalman Filter, the latter can represent its actual state considering the measurement value and error from sensors. The error between the two estimators indicates whether the estimator in other agents can maintain a relatively accurate state estimation for this agent, and decides whether the communication is triggered. Simulations demonstrate the effectiveness and advantages of the proposed method for the distributed control of flocking in both Matlab and Gazebo.\",\n \"Title: Study on HEV's driving condition recognition method based on PSO algorithm.\\nYear: 2018\\n\\nAbstract: In the interest of the hybrid electric vehicle (HEV) real-time road gradient and vehicle load( driving condition) effective identification during the running process, this work takes the series-parallel HEV as the research object and studies on the dynamic identification mechanism of slope and load, based on the analysis of its structural parameters. Firstly, vehicle's driving condition identification model is developed, and the optimization goal function is established using the least square method. Secondly, six different kinds of particle swarm optimization (PSO) algorithm are used for the recognition of vehicle's driving condition, and the results show that hybrid PSO algorithm based on hybrid training algorithm has better calculation accuracy for this problem. Finally, Experiments are carried out to verify the driving condition recognition method based on PSO algorithm. Through the acquisition of a real vehicle data during the running process, road grade and vehicle mass are estimated by using the proposed method, and the effectiveness of the proposed method is proved through comparison of errors between recognition results and true value.\",\n \"Title: Detection and RM correction approach for manoeuvring target with complex motions.\\nYear: 2017\\n\\nAbstract: This study addresses the coherent accumulation problem for detecting a manoeuvring target with complex motions, where range migration (RM) [i.e. range walk (RW) and range curvature (RC)] and Doppler frequency migration (DFM) occur during the coherent integration time. An efficient approach based on generalised keystone transform (GKT), radon transform (RT) and generalised dechirp process (GDP), i....\",\n \"Title: Biological Plausibility And Stochasticity In Scalable Vo2 Active Memristor Neurons\\nYear: 2018\\n\\nAbstract: Neuromorphic networks of artificial neurons and synapses can solve computationally hard problems with energy efficiencies unattainable for von Neumann architectures. For image processing, silicon neuromorphic processors outperform graphic processing units in energy efficiency by a large margin, but deliver much lower chip-scale throughput. The performance-efficiency dilemma for silicon processors may not be overcome by Moore's law scaling of silicon transistors. Scalable and biomimetic active memristor neurons and passive memristor synapses form a self-sufficient basis for a transistorless neural network. However, previous demonstrations of memristor neurons only showed simple integrate-and-fire behaviors and did not reveal the rich dynamics and computational complexity of biological neurons. Here we report that neurons built with nanoscale vanadium dioxide active memristors possess all three classes of excitability and most of the known biological neuronal dynamics, and are intrinsically stochastic. With the favorable size and power scaling, there is a path toward an all-memristor neuromorphic cortical computer.\",\n \"Title: Receive-Beam Resource Allocation for Multiple Target Tracking With Distributed MIMO Radars.\\nYear: 2018\\n\\nAbstract: A distributed multiple-input multiple-output (MIMO) radar system is capable of tracking multiple targets under the “defocused transmit-focused receive” (DTFR) operating mode, in which each transmitter forms a completely defocused beam to illuminate the whole surveillance region and each receiver adopts a focused beam to attain a higher resolution. However, when operating in this mode, there exists...\",\n \"Title: Computationally efficient coherent detection and parameter estimation algorithm for maneuvering target.\\nYear: 2019\\n\\nAbstract: •A computationally efficient coherent detection and parameter estimation algorithm via SAF-SFT is proposed.•The proposed algorithm could remove the range cell migration and Doppler spread effects without parameter-searching process.•SAF-SFT method could strike a good balance between computational cost and detection probability as well as the estimation performance.\",\n \"Title: Computationally Efficient Multi-Agent Multi-Object Tracking With Labeled Random Finite Sets.\\nYear: 2019\\n\\nAbstract: This paper addresses multi-agent multi-object tracking with labeled random finite sets via Generalized Covariance Intersection (GCI) fusion. While standard GCI fusion of Labeled Multi-Object (LMO) densities is labelwise and hence fully parallelizable, previous work unfortunately revealed that its fusion performance is highly sensitive to the unavoidable label inconsistencies among different agents...\",\n \"Title: Fault-Tolerant and Self-Adaptive Market-Based Coordination Using Hoplites Framework for Multi-Robot Patrolling Tasks\\nYear: 2018\\n\\nAbstract: An autonomous robot team can be employed for continuous coverage of a dynamic environment. In this paper, we propose a novel approach for creating multi-robot patrolling policies, which is fault-tolerant and self-adaptive. A dynamic priority queue and time-out replanning mechanism are maintained by each robot to schedule the tasks fault-tolerantly in the context of the market-based method. Hoplites framework is adapted by introducing a self-adaptive threshold adjustment and sharing mechanism to provide a high-level coordination. This work is demonstrated by a multi-robot patrolling task implemented on Robot Operating System (ROS). A flexible tool, Stage, is leveraged to provide the simulated environment. The experimental results validate the effectiveness and availability.\",\n \"Title: A Coherent Detection and Velocity Estimation Algorithm for the High-Speed Target Based on the Modified Location Rotation Transform.\\nYear: 2018\\n\\nAbstract: High-speed target often suffers from the range walk (RW) problem within the process of long-time coherent integration (CI). The aim of this paper is to present a new coherent detection algorithm, i.e., modified location rotation transform (MLRT), which could effectively correct the RW and estimate the velocity of the high-speed target. In particular, the MLRT algorithm could rotate the coordinate ...\",\n \"Title: STGRFT for Detection of Maneuvering Weak Target With Multiple Motion Models.\\nYear: 2019\\n\\nAbstract: Long time coherent integration (LTCI) technique enhances radar detection of maneuvering weak targets. Unfortunately, range migration (RM) and Doppler frequency migration (DFM) would occur during the integration period. Existing LTCI methods are all based on the assumption that the target is of single motion model (i.e., the target's motion model is fixed) during the long integration time. However,...\",\n \"Title: Scaled accuracy based power allocation for multi-target tracking with colocated MIMO radars.\\nYear: 2019\\n\\nAbstract: •A scaled accuracy based power allocation framework is proposed to address multitarget tracking problem in a colocated MIMO radar system.•The proposed strategy can allocate the transmit power adaptively as tracking accuracy requirement change over time and space.•The proposed strategy yields better power allocation options than the existing strategy.•Besides the colocated MIMO radar, the proposed framework can be used in other radar configurations such as the phased array radar and distributed MIMO radar systems.\",\n \"Title: Asynchronous multi-rate multi-sensor fusion based on random finite set.\\nYear: 2019\\n\\nAbstract: •Three asynchronous fusion algorithms, a centralized fusion algorithm and two distributed fusion algorithms, are proposed.•The proposed SP fusion algorithm could achieve the best performance due to the usage of raw measurements.•Compared with the SP fusion algorithm, the proposed two distributed fusion algorithms based on GCI rule are more robust and their output time is more flexibility.•The proposed S-GCI fusion algorithm needs less computational burden at the cost of a bit less performance compared with the B-GCI fusion algorithm.\",\n \"Title: Direct position determination of multiple coherent sources using an iterative adaptive approach.\\nYear: 2019\\n\\nAbstract: •The direct localization algorithm combining with the iterative approach is proposed for coherent sources.•The connected domain separation method can efficiently isolate and extract the domains containing different sources.•The proposed algorithm outperforms the existing methods in accuracy and robustness for locating coherent sources.•Increasing antennas can obtain better localization performance from energy gain and spatial filtering effect.\",\n \"Title: Distributed Multi-sensor Multi-view Fusion based on Generalized Covariance Intersection.\\nYear: 2019\\n\\nAbstract: •Multi-target tracking using multiple sensors having different fields of view.•Fusion of the posterior Probability Hypotheses Densities of the sensors.•Robust distributed fusion based on generalized covariance intersection.•A clustering algorithm is used to derive an efficent parallelized fusion strategy.•Compensation strategy to preserve targets outside the common field of view.\",\n \"Title: Detecting retail products in situ using CNN without human effort labeling.\\nYear: 2019\\n\\nAbstract: CNN is a powerful tool for many computer vision tasks, achieving much better result than traditional methods. Since CNN has a very large capacity, training such a neural network often requires many data, but it is often expensive to obtain labeled images in real practice, especially for object detection, where collecting bounding box of every object in training set requires many human efforts. This is the case in detection of retail products where there can be many different categories. In this paper, we focus on applying CNN to detect 324-categories products in situ, while requiring no extra effort of labeling bounding box for any image. Our approach is based on an algorithm that extracts bounding box from in-vitro dataset and an algorithm to simulate occlusion. We have successfully shown the effectiveness and usefulness of our methods to build up a Faster RCNN detection model. Similar idea is also applicable in other scenarios.\",\n \"Title: WRFRFT-based Coherent Detection and Parameter Estimation of Radar Moving Target With Unknown Entry/Departure Time.\\nYear: 2020\\n\\nAbstract: •An efficient coherent detection and parameter estimation method named WRFRFT is proposed for radar moving target with unknown entry/departure time.•The proposed method could effectively estimate the time parameters and motion parameters of the target with across range cell and Doppler spread effects.•WRFRFT-based method could obtain better detection ability and superior estimation performance than the existing popular coherent detection and estimation algorithms.\",\n \"Title: Comprehensive Evaluation Of The Gf-4 Satellite Image Quality From 2015 To 2020\\nYear: 2021\\n\\nAbstract: GaoFen-4(GF-4) is the highest spatial resolution Earth observation satellite operating in geosynchronous orbit. Its fixed Earth observation location, rapid responsiveness, and wide observation range make it popular in disaster and emergency monitoring. To evaluate the GF-4 image quality in detail on a long-term basis, this study analyzes the image quality after the commissioning phase by focusing on ground sample distance (GSD) and geometric and radiometric quality. The theoretical calculation, geometric and radiometric measurements, and on-site experiments results show that (1) the GSD of the GF-4 image is similar to 50 m at the nadir point and increases gradually with the distance away from the nadir point, (2) most external geometric errors are within the design requirements of 4 km despite some exceeding the limit, and the internal geometric errors are tested within 1 pixel, and (3) image sharpness is generally stable but varies with the atmosphere condition and imaging time, and the radiometric response gradually degrades at the rate of less than 5.5% per year.\",\n \"Title: A New Flattened Cylinder Specimen For Direct Tensile Test Of Rock\\nYear: 2021\\n\\nAbstract: In recent decades, researchers have paid more attention to the indirect tensile test than to the direct tensile test (DTT) of rocks, mainly due to difficulties in the alignment and the stress concentration at the end of an intact cylindrical specimen. In this paper, a new flattened cylinder specimen and a clamp device were designed to obtain the true tensile strength of the rock in DTT. Stress distributions of the specimen with different lengths (l) and cutting thicknesses (t) were analyzed, and damage processes of the specimen were monitored by the Digital Image Correlation (DIC), the fractured sections were also scanned. Different mechanical parameters were also obtained by the DTT of the flattened cylinder specimens and the intact cylinder specimens, as well as the Brazilian disc. Research results show that the tensile strength obtained by DTT is smaller than the Brazilian disc and is slightly greater than the intact cylindrical specimen. The flattened cylinder specimen with 0.20 <= 2t/D < 0.68 and 0.10 <= l/D <= 0.20 is recommended to measure the true tensile strength of rock material in DTT. This new shape of the specimen is promising to be extended in the uniaxial or triaxial direct tension test.\",\n \"Title: Event-triggered Neural Network-Based Adaptive Control for a Class of Uncertain Nonlinear Systems\\nYear: 2021\\n\\nAbstract: In the paper, an event triggering adaptive control method based on neural network (NN) is proposed for a class of uncertain nonlinear systems with external disturbances. In order to reduce the network resource utilization, a novel event-triggered condition by the Lyapunov approach is proposed. In addition, the NN controller and adaptive parameters determined by the Lyapunov stability method are updated only at triggered instants to reduce the amount of calculation. Only one NN is used as the controller in the entire system. The stability analysis results of the closed-loop system are obtained by the Lyapunov approach, which shows that all the signals in the systems with bounded disturbance are semi-globally bounded. Zeno behavior is avoided. Finally, the analytical design is confirmed by the simulation results on a two-link robotic manipulator.\",\n \"Title: A Fast Approach for Detection and Parameter Estimation of Maneuvering Target With Complex Motions in Coherent Radar System\\nYear: 2021\\n\\nAbstract: In coherent radar system, maneuvering target with complex motions usually confronts the issues of range cell migration (RCM) and Doppler frequency spread (DFS) within the long-time coherent integration process. To address these issues and achieve reliable detection and estimation performance, this paper proposes a novel fast approach without motion parameters searching. Particularly, the proposed ...\",\n \"Title: Trajectory PHD and CPHD Filters With Unknown Detection Profile\\nYear: 2022\\n\\nAbstract: Compared to the probability hypothesis density (PHD) and cardinalized PHD (CPHD) filters, the trajectory PHD (TPHD) and trajectory CPHD (TCPHD) filters are for sets of trajectories, and thus are able to produce trajectory estimates with better performance. In this paper, we develop the TPHD and TCPHD filters which can adaptively learn the history of the unknown target detection probability, and therefore they can perform more robustly in scenarios where targets are with unknown and time-varying detection probabilities. These filters are referred to as the unknown TPHD (U-TPHD) and unknown TCPHD (U-TCPHD) filters. By minimizing the Kullback-Leibler divergence (KLD), the U-TPHD and U-TCPHD filters can obtain, respectively, the best Poisson and independent identically distributed (IID) density approximations over a set of augmented trajectories. For computational efficiency, we also propose the U-TPHD and U-TCPHD filters that only consider the unknown detection profile at the current time. Specifically, the Beta-Gaussian mixture method is adopted for the implementations of proposed filters, which are referred to as the BG-U-TPHD and BG-U-TCPHD filters. The \\n<inline-formula xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\"><tex-math notation=\\\"LaTeX\\\">$L$</tex-math></inline-formula>\\n-scan approximations of these filters with much lower computational burden are also presented. Finally, various simulation results demonstrate that the BG-U-TPHD and BG-U-TCPHD filters can achieve robust tracking performance to adapt to unknown detection profile. Besides, it also shows that usually a small value of \\n<inline-formula xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\"><tex-math notation=\\\"LaTeX\\\">$L$</tex-math></inline-formula>\\n in the \\n<inline-formula xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\"><tex-math notation=\\\"LaTeX\\\">$L$</tex-math></inline-formula>\\n-scan approximation has similar performance than with a large value at a fraction of the computational cost.\",\n \"Title: Multi-target Joint Tracking and Classification Using the Trajectory PHD Filter.\\nYear: 2021\\n\\nAbstract: To account for joint tracking and classification (JTC) of multiple targets from observation sets in presence of detection uncertainty, noise and clutter, this paper develops a new trajectory probability hypothesis density (TPHD) filter, which is referred to as the JTC-TPHD filter. The JTC-TPHD filter classifies different targets based on their motion models and each target is assigned with multiple class hypotheses. By using this strategy, we can not only obtain the category information of the targets, but also a more accurate trajectory estimation than the traditional TPHD filter. The JTC-TPHD filter is derived by finding the best Poisson posterior approximation over trajectories on an augmented state space using the Kullback-Leibler divergence (KLD) minimization. The Gaussian mixture is adopted for the implementation, which is referred to as the GM-JTC-TPHD filter. The L-scan approximation is also presented for the GM-JTC-TPHD filter, which possesses lower computational burden. Simulation results show that the GM-JTC-TPHD filter can classify targets correctly and obtain accurate trajectory estimation.\",\n \"Title: Multi-Frame Joint Tracking and Shape Estimation Method for Weak Extended Targets\\nYear: 2021\\n\\nAbstract: This paper addresses the joint tracking and shape estimation (JTSE) problem of elliptical extended targets in low signal-to-noise (SNR) scenarios using multi-frame joint processing. Considering the weak target echoes and unknown parameters of elliptical extended targets, it is challenging to achieve effective detection and tracking. To solve these problems, a multi-frame tracking and shape estimat...\",\n \"Title: Joint tracking sequence and dwell time allocation for multi-target tracking with phased array radar\\nYear: 2022\\n\\nAbstract: •This work addresses a joint tracking sequence and dwell time allocation (JTSDTA) for multi-target tracking (MTT) based on a phased array radar (PAR) system.•In the proposed model, implementations of target tracking and searching of the PAR are considered to be interleaved for the first time.•A time-aware posterior Cramér–Rao lower bound is derived to quantify the tracking performance of the PAR.•A two steps-based decoupling algorithm is proposed to solve the JTSDTA-related problem efficiently.\",\n \"Title: Robust Radar Detection and Classification of Traffic Vehicles Based on Anchor-free CenterNet\\nYear: 2021\\n\\nAbstract: Radar has been regarded as the enabling sensor technology to realize the intelligent transportation system. The current works in traffic radar mainly focus on detecting the target with fixed size. However, traffic scenarios consist of various targets with different sizes. Therefore, in this paper, we propose a CenterNet-based radar signal processing framework for detecting and classifying four typ...\",\n \"Title: Calib-Net: Calibrating the Low-Cost IMU via Deep Convolutional Neural Network\\nYear: 2021\\n\\nAbstract: The low-cost Inertial Measurement Unit (IMU) can provide orientation information and is widely used in our daily life. However, IMUs with bad calibration will provide inaccurate angular velocity and lead to rapid drift of integral orientation in a short time. In this paper, we present the Calib-Net which can achieve the accurate calibration of low-cost IMU via a simple deep convolutional neural network. Following a carefully designed mathematical calibration model, Calib-Net can output compensation components for gyroscope measurements dynamically. Dilation convolution is adopted in Calib-Net for spatio-temporal feature extraction of IMU measurements. We evaluate our proposed system on public datasets quantitively and qualitatively. The experimental results demonstrate that our Calib-Net achieves better calibration performance than other methods, what is more, and the estimated orientation with our Calib-Net is even comparable with the results from visual inertial odometry (VIO) systems.\",\n \"Title: Semi-Empirical Time-Dependent Parameter of Shear Strength for Traction Force between Deep-Sea Sediment and Tracked Miner\\nYear: 2022\\n\\nAbstract: Based on our direct shear creep experiment and the direct shear rheological constitutive model, a semi-empirical time-dependent parameter of the shear strength is obtained by Mohr-Coulomb shear strength theory, and different time-dependent traction force calculations between deep-sea sediment and a tracked miner are conducted by the work-energy principle. The time-dependent traction force calculation under its influencing factors, including the time, track shoe number, and grounding pressure, are analyzed and proved to be valid by the traction force experiment of a single-track shoe. The results show that the time-dependent cohesion force obtained by a semi-empirical way can be easily used to deduce the time-dependent traction force models under the different grounding pressure distributions and applied into deep-sea engineering application conveniently; the verified traction force models by the traction force experiment of a single-track shoe illustrate that traction force under the decrement grounding pressure distribution is the worst among the four kinds of grounding pressure distributions and suggested for evaluating the most unfavorable traction force and calculating the trafficability and stability of the deep-sea tracked miner.\",\n \"Title: Exponential Mixture Density Based Approximation to Posterior Cramr-Rao Lower Bound for Distributed Target Tracking\\nYear: 2022\\n\\nAbstract: The posterior Cramr-Rao lower bound (PCRLB) and a number of its extensions including the unconditional PCRLB (U-PCRLB) and the conditional PCRLB (C-PCRLB) have been widely studied in multi-sensor target tracking (MSTT). Previous MSTT works assume that the measurements are conditionally independent, which is often not the case in practice. When the correlations of the measurements are unknown, the standard Bayes update and PCRLB computations cannot be performed. Inspired by geometric average (GA) fusion that implements data fusion with unknown data correlations by employing exponential mixture density (EMD) to compute the global posterior, EMD-based approximations to U-PCRLB and C-PCRLB are derived in this paper for two classic distributed fusion architectures, namely the hierarchical and consensus architectures. The PCRLB can be decomposed into two parts, one coming from the prior information and the other from the information contained in the data. The data information part of PCRLB is approximated via EMD-based posterior, leading to the approximation of the PCRLB. We present a sequential Monte Carlo solution to recursively compute the proposed approximation to the PCRLB for nonlinear non-Gaussian estimation problems. Numerical simulations are provided to show that the proposed approximation to the bound on the estimation mean square error (MSE) is tighter compared to the existing bounds based on likelihood fusion obtained under the measurement independence assumption.\",\n \"Title: A GDOP-Based Performance Description of TOA Localization with Uncertain Measurements\\nYear: 2022\\n\\nAbstract: In this paper, we study a geometric dilution of a precision (GDOP)-based localization performance metric for multisite radar adopting a time-of-arrival (TOA)-based localization scheme. In contrast to the existing literature, we consider an actual uncertain measurement situation where the detection probabilities of radar nodes are assumed to be less than unity. The aim is to formulate a general signal-decoupled metric to describe the system localization performance while fully considering detection and estimation operations. Specifically, to match the uncertain measurements, we first establish effectively detected time delay measurements (TDMs) for localization and modify the traditional performance bounds for TDM estimation. Then, by combining the localization performance with the effective detection (ED) via probability, we propose a novel geometric dilution of precision with uncertain measurements (GDOP-UM) metric. The proposed metric can truly characterize the localization performance under the uncertain measurement situation. Finally, the simulation results show that the proposed GDOP-UM can describe the actual localization performance regardless of how the detection performance changes.\",\n \"Title: Subspace-Based Detection and Localization in Distributed MIMO Radars\\nYear: 2022\\n\\nAbstract: In this paper, we consider a distributed multiple-input multiple-output (MIMO) radar which radiates waveforms with non-ideal cross- and auto-correlation functions and derive a novel subspace-based procedure to detect and localize multiple prospective targets. The proposed solution solves a sequence of composite binary hypothesis testing problems by resorting to the generalized information criterion (GIC); in particular, at each step, it aims to detect and localize one additional target, upon removing the interference caused by the previously-detected targets, An illustrative example is provided.\",\n \"Title: An Efficient Track-Before-Detect for Multi-PRF Radars With Range and Doppler Ambiguities\\nYear: 2022\\n\\nAbstract: This article considers the detection and tracking of weak targets in multiple pulse repetition frequency (multi-PRF) radars using the track-before-detect (TBD) technique. By exploring the measurement independence among different PRFs, we decompose the joint multi-PRF and multiframe optimization problem into several lower dimensional maximizations, each of which corresponds to an intra-PRF multiframe processing. An efficient two-stage TBD procedure with a parallel structure is proposed for the multi-PRF radars. Specifically, in the first stage, a constraint inequality is derived analytically and used to decouple the measurement ambiguities from the nonlinear conversion relationship between polar and Cartesian coordinates. The intra-PRF multiframe integration can then be carried out concurrently using the ambiguous measurements and the target-like measurement plot sequences with different ambiguities are extracted for different PRFs. In the second stage, a covariance combination fusion-based inter-PRF joint disambiguation and estimation algorithm is proposed to solve the ambiguities of the plot sequences and output high-accuracy target tracks. Simulation experiments show that the proposed algorithm can provide a good detection performance and higher tracking accuracy with much lower complexity.\",\n \"Title: Joint Online Route Planning and Resource Optimization for Multitarget Tracking in Airborne Radar Systems\\nYear: 2022\\n\\nAbstract: Reasonable route planning and resource allocation strategy in the airborne radar systems (ARS), can sufficiently utilize the limited resources and promote the multitarget tracking (MTT) performance. However, using separately the route planning and resource optimization method cannot take full advantage of the airborne platform. Considering this issue, we propose a joint online route planning and resource optimization strategy in the ARS to improve the system capability for MTT. First, a kinematic model of the ARS, including the mathematical expression between the radar states and the system control parameters, is introduced, which integrates the route planning to the radar scheduling scheme. Next, the posterior Cramér–Rao lower bound about route planning and resource optimization variables for the tracking targets is derived. Then, a scaled-based utility function is established to quantify the MTT performance. Hereafter, a nonconvex problem is formulated by minimizing the utility function with route and resource constraints, and then a efficient three-stage partition-based solution is proposed. Finally, simulation experiments demonstrate the effectiveness of the proposed algorithm. Furthermore, over the traditional benchmark algorithm, the tracking performance of the proposed approach improves 30.44%.\",\n \"Title: Multitarget Detection Strategy for Distributed MIMO Radar With Widely Separated Antennas\\nYear: 2022\\n\\nAbstract: In this article, we propose a novel solution to detect multiple targets using a distributed multiple-input multiple-output (MIMO) radar under the so-called \\\"defocused transmit-defocused receive\\\" operating mode. The proposed method employs a grid-based data matching algorithm, aiming to associate the target responses to potential target locations, solving the resulting data puzzle that evaluates the various cells under test (CUTs) in the surveillance area resulting from the intertwined range cells across all transmit-receive channels. Sketchily, the approach divides the surveillance area into identically interlocking and analytically expressible grid cells and then selects the grid cells with the best fitting multichannel data to be equivalently regarded as the CUTs. Next, the generalized likelihood ratio test (GLRT) detector is derived to test for target presence in each of the selected grid cells. A separate procedure is introduced to eliminate the spurious \\\"shadow targets,\\\" false alarms occurring in the grid cells without a target while sharing range cells with the targets. The essence of this procedure is to find the source of the observed contributions to the grid cells whose test statistics exceed their thresholds, and simultaneously obtain the positions of the targets. The proposed method is evaluated using both numerical simulations and experimental data recorded by five small radars, demonstrating the effectiveness of the proposed technique.\",\n \"Title: Adaptive beam burden management for multi-target tracking in distributed radar network\\nYear: 2022\\n\\nAbstract: •The system beam burden with the beam pointing and pulse number matrix is modeled for multi-target tracking (MTT) in distributed radar network (DRN).•This work formulates a novel optimization problem on the basis of beam burden saving with the tracking tasks and the system restrictions.•We design a Lagrangian relaxation-based (LR) fast algorithm to efficiently obtain the solutions.•The ABPNS strategy is proposed by adjusting the beam and pulse number to realize adaptive beam burden management in the DRN.\",\n \"Title: Strong scattering points-based joint detection and size estimation method for swarm targets\\nYear: 2022\\n\\nAbstract: The paper considers the joint detection and size estimation of swarm targets. The echo energy of swarm targets is doubly-spread in the range-Doppler (RD) image due to spatially dispersed and relative motion between individuals. Therefore, the traditional constant false alarm rate (CFAR) detection strategies of point and range-spread rigid targets may suffer performance degradation because of failing to accumulate all energy of target in the RD image. In this paper, we first developed a swarm targets echo model. Then a novel detector based on the strong scattering points (SPs) estimation is proposed. The detector considers the swarm targets as a whole and estimates and accumulates all the target-like strong SPs to enhance the detection performance. Moreover, the radial size of swarm targets is calculated according to the range of SPs. Finally, the constant false alarm property is derived, and the strategy of detector parameters setting is discussed in detail. A thorough performance assessment is conducted by Monte Carlo simulation. The results show that the proposed method is superior to the modified cell averaging CFAR (MCA-CFAR) and deterministic scatterer model generalized likelihood ratio test (DSM-GLRT) detectors.\",\n \"Title: Geometric Optimization of Distributed MIMO Radar System for Accurate Localization in Multiple Key Subareas\\nYear: 2022\\n\\nAbstract: •To describe the surveillance and localization performance of the distributed MIMO radar system, we formulate corresponding performance evaluation metrics. Especially in order to uniformly characterize multiple localization performance requirements in multiple subareas, we establish a scaled accuracy-based localization performance metric based on GDOP.•Considering the importance differences of the performance in different subareas and the coupled relationship between different radar performance, we formulate a multi-objective optimization problem with complex coupled constraints for enhancing both the surveillance and localization performance of the system. Furthermore, we simplify the problem by reducing the number of constraints via a scaling process.•To efficiently solve the non-convex multi-objective optimization problem with complex coupled constraint, we improve the traditional MOPSO algorithm by introducing a two-stage update rule of individual best position, and further propose a novel SC-MOPSO algorithm.\",\n \"Title: Resource Scheduling for Multi-Target Tracking in Multi-Radar Systems With Imperfect Detection\\nYear: 2022\\n\\nAbstract: In this paper, an effective joint radar assignment and power scheduling (JRAPS) scheme is proposed for the multi-target tracking (MTT) task in multi-radar systems (MRSs) under imperfect detection, namely, target related measurements are collected with the probability of detection (P-D) less than 1. The centralized fusion architecture is adopted by the MRS. Specifically, during each sampling time interval, each radar of the MRS is assigned to track certain targets with controllable transmitting power. The measurements collected by all selected local radars, with P-D <= 1, are sent to a central radar to obtain the global MTT results. To maximize the global KIT task performance, the proposed JRAPS scheme implements the online target-to-radar assignment and transmitting power allocation scheme based on the feedback of the MTT results. The posterior Cramer-Rao lower bound (PCRLB) with P-D <= 1 is derived and utilized as the tracking performance metric since it provides a more accurate lower bound on the target state estimates under imperfect detection. Then, an overall cost function (OCT) is formulated based on the derived PCRLB to quantify the global MTT performance. Combined with the practical resource constraints of the MRS, the formulated TRAPS problem is shown to be non-convex. Therefore, we further propose a fast three-stage iterative method to solve this problem efficiently. Simulation results verify the superiority and effectiveness of the proposed JRAPS strategy in terms of both tracking accuracy and target detection performance.\",\n \"Title: Network architecture optimization for netted MIMO radar systems with surveillance performance\\nYear: 2023\\n\\nAbstract: •The global detection probability based on the clustering framework is derived.•An objective function considering the importance of different surveillance tasks.•Formulating a network architecture optimization problem with several constraints.•A maximum block pattern search algorithm for the nonconvex optimization problem.\",\n \"Title: Efficient Radar Deep Temporal Detection in Urban Traffic Scenes\\nYear: 2022\\n\\nAbstract: This paper explores object detection on radar range-Doppler map. Most of the radar processing algorithms are proposed for detecting objects without classifying. Meanwhile, these approaches neglect the useful information available in the temporal domain. To address these problems, we propose an online radar deep temporal detection framework by frame-to-frame prediction and association with low computation. The core idea is that once an object is detected, its location and class can be predicted in the future frame to improve detection results. The experiment results illustrate this method achieves better detection and classification performance, and shows the usability of radar data for traffic scenes.\",\n \"Title: Event-Triggered Neural Network Control for Uncertain Nonlinear Systems Without State Observer*\\nYear: 2022\\n\\nAbstract: A robust approximation-based event-triggered control method is presented for single input single output (SISO) nonlinear continuous-time systems with unmeasurable states and external disturbance. In the whole system, just one neural network (NN) is designed to approximate the unknown part in the controller, and output errors are directly used to construct the system controller and event-triggered mechanism to relieve the burden of system communication. The controller with a simple structure is easier to be realized in practical engineering. The system control signals and adaptive parameters are updated only if the event trigger condition is met, and this way further reduces the waste of network resources caused by frequent system sampling. The application of stability theory of Lyapunov proves that the weight estimation of the NN and tracking errors are ultimate and uniform boundedness, and the efficacy of the proposed scheme is verified with numerical results on a robot.\",\n \"Title: The PHD Filter for Target Swarms and Its Gaussian Mixture Implementation\\nYear: 2022\\n\\nAbstract: The swarm is used to describe a set of individual targets which are very close together in the measurement space (relative to the resolution of the sensor). In this paper, we consider to estimate the number, shape and density of target swarms in addition to their motion states. In order to deal with the uncertain number of measurement origins, we use amplitude information (AI) to construct hypotheses about the number of individual target source of the merged measurement in the probability hypothesis density (PHD) update step. On this basis, we propose the PHD filter for target swarms (S-PHD filter) and develop Gaussian mixture (GM) implementation to the resulting filter. The performance of this algorithm is demonstrated by a multi swarms scenario with merged measurements, and the results highlight the significant improvement in the estimates of the number of individuals and the density of swarms.\",\n \"Title: An Improved Two-Stage Based Multi-frame Track-Before-Detect Algorithm in Radar systems\\nYear: 2022\\n\\nAbstract: The detection and tracking of dim targets in complex environments show great importance in radar systems. The track-before-detect (TBD) technology has been widely researched in the scenario where the target signal-to-noise ratio (SNR) is low, however, there has been no suitable solution to the trade-off between tracking accuracy and computational complexity during the process of multi-frame joint detection in radar systems. In this paper, we propose an efficient two-stage based multi-frame detection and tracking algorithm in radar systems. The proposed algorithm consists of a low threshold pre-processing stage, and a TBD processor, which searches possible target tracks from multiple scans and declares the final estimated tracks. The proposed algorithm provides an accurate evolution of target states over time in polar coordinates to avoid the performance loss and model mismatch due to the nonlinear conversion in mixed coordinates. In addition, we further propose a greedy-based recursive algorithm to implement fast track formation from the over-threshold multi-frame measurement points. Simulation results show that the proposed method achieves a better detection and tracking performance with a low computational complexity.\",\n \"Title: An Integrated Localization Method for Mixed Near-Field and Far-Field Sources Based on Mixed-order Statistic\\nYear: 2022\\n\\nAbstract: This paper considers the integrated localization for the mixed near-field (NF) and far-field (FF) sources using the uniform linear array (ULA). With the help of the polynomial rooting methods and the propagator, an efficient algorithm is proposed to provide an integrated estimation of the direction of arrival (DOA) and the ranges of the sources. It takes low computational burden without the requirements that separating the DOA and range information or pre-classification of the sources. We first construct two special fourth-order cumulant matrices using the received array data, then extract the prior-electrical parameters related to the array elements by its steering matrix, and finally carry out parameter matching and classification. Besides, the proposed algorithm eliminates the need for tedious eigenvalue decomposition and spectral search steps, and has almost no aperture loss. Eventually, several simulation results show that the proposed algorithm has lower computational complexity under an acceptable accuracy, compared to the state-of-the-art methods.\",\n \"Title: LPI-Based Joint Node Selection and Power Allocation Strategy for Target Tracking in Distributed MIMO Radar\\nYear: 2022\\n\\nAbstract: This paper addresses the problem of the low probability of intercept (LPI) performance optimization for target tracking in the distributed MIMO radar systems through the transmitting resource scheduling (TRS) strategy. The mechanism of the proposed LPI-based TRS strategy is to adopt the optimization technique to collaboratively schedule the transmit radar node and signal power with the target tracking accuracy requirement, which aims to enhance the LPI performance of the overall system. Based on the existing research, we develop an intercept model to describe two stages of the signal intercept process with the specific interceptor equipped on the moving target, i.e., intercept and detection. Under the proposed model, the probability of report (PR) is proposed to evaluate the LPI performance for a single transmitting radar node. Then, we consider the maximum PR to represent the LPI performance metric for the overall system. Hence, using it as the objective function, the optimization problem is established with the constraints of the target tracking requirement and the system resource. By introducing the two-step partition-based solution, the proposed non-convex problem is solved efficiently. Finally, several numerical simulations demonstrate the theoretical calculations and validate the effectiveness of the proposed LPI-based TRS strategy.\",\n \"Title: Multi-Frame Track-Before-Detect for Scalable Extended Target Tracking\\nYear: 2022\\n\\nAbstract: This paper mainly addresses the scalable detection and tracking of the extended target in the low signal-to-noise(SNR) environment. As the appearance and shape of the extended target are constantly varied, it is challenging to achieve robust detection and tracking. For this, a novel adaptive scale (AS) kernelized correlation filter (KCF) based on multi-frame track-before-detect (MF-TBD) framework is proposed. By embedding scaling pools into the response map to handle the scale variation and accumulating target energy overall feasible trajectories, AS-MF-TBD estimates the kinematic state and geometric shapes simultaneously. Both simulation data and real radar data are used to demonstrate the superiority of the proposed method in terms of detection performance and estimation accuracy.\",\n \"Title: A Fast and Robust Maneuvering Target Tracking Method without Markov Assumption\\nYear: 2022\\n\\nAbstract: In the traditional methods of maneuvering target tracking, it is necessary to adjust the state transition model in time to match the maneuvering target, which will cause the problems of model decision delay and competition. Besides, the commonly adopted first-order Markov assumption can lead to the loss of information when motion modes are relevant to time. In order to solve these problems, a data-driven algorithm based on LightGBM is proposed in this paper. Maneuvering target tracking is modeled as a non-probabilistic method of direct mapping from sensor measurement to target state, track samples of different motion modes are used for training, and fast online tracking is realized. Comparing it with interacting multiple model (IMM) algorithm in a variety of different scenarios, simulation results show that the proposed method has advantages in accuracy and speed. Finally, the robustness of the algorithm is verified under the compound noise of Cauchy and Gaussian distributions.\",\n \"Title: Arithmetic Average Based Multi-sensor TPHD Filter for Distributed Multi-target Tracking\\nYear: 2022\\n\\nAbstract: Compared with the probability hypothesis density (PHD) filter for sets of targets, the trajectory probability hypothesis density (TPHD) filter can estimate the sets of trajectories in a principle way and has better target tracking performance. This paper aims at extending the TPHD filter to distributed multitarget tracking (MTT) for the multi-sensor system. However, in the trajectory set based distributed fusion implementation, the trajectory state difference phenomenon makes the clustering and merging techniques unfeasible in trajectory state space. To address this problem, this paper studies the space decomposition of the TPHD and proposes a distributed MTT method based on the TPHD filter with the weighted arithmetic average (WAA) fusion rule. First, we prove the rationality of the space decomposition in the posterior density of the TPHD filter. Then, based on the proposed property, we derive the WAA fusion formulation of the TPHD filter by minimizing the weighted sum of Kullback-Leibler divergences (KLD) from local posterior densities, and develop the analytical Gaussian mixture (GM) implementation with the L-scan approximation. Numerical results demonstrate the efficacy of the proposed fusion method.\",\n \"Title: Modeling and Statistical Analysis of Vision Distance Measurement Uncertainty for Multi-Automotive Sensor Target Tracking\\nYear: 2022\\n\\nAbstract: Among all multi-automotive sensor target tracking algorithms, statistical sequential estimation techniques are in the dominant position. For these techniques, statistical sensor measurement uncertainty model is essential. However, for camera, which is a primary automotive sensor, there exists a problem that vision measurements are usually not statistical model-based. In this paper, to open up a possibility of establishing an access between the vision measurements and statistics-based methods, we try to describe vision distance measurement (VDM) uncertainty in a statistical way. As the foundation for statistically analysing VDM uncertainty, we first derive the expression of VDM uncertainty. Based on the expression, we then model the probability distribution of VDM uncertainty, thus the statistical VDM uncertainty model is obtained and VDM is connected with the statistics-based fusion methods. The proposed statistical model is validated by experimental data.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":97,"string":"97"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13564,"string":"13,564"}}},{"rowIdx":13565,"cells":{"question_id":{"kind":"string","value":"Weiwei Li"},"question":{"kind":"string","value":"Generate an abstract for the title \"A Better Translation From Ltl To Transition-Based Generalized Buchi Automata\" using the following items: 1. Linear Temporal Logic (LTL), 2. Buchi Automata, 3. Translation Algorithm, 4. Performance Analysis, 5. Simplification"},"answer":{"kind":"string","value":"Translating linear temporal logic (LTL) formulas into Buchi automata is one of the most important aspects of LTL model checking. Certain successful algorithms, such as LTL2BA and SPOT, first translate an LTL formula into a transition-based generalized Buchi automaton (TGBA) and then degeneralize it into a Buchi automaton. This paper focuses on achieving a better translation from LTL to TGBA and analyzing the performance of every step of the algorithm. We decompose the translation into three steps to give a step-wise description and improve all three steps. The first step is the basic translation without acceptance conditions and simplifications, which combines the advantages of both LTL2BA and SPOT. Second, we introduce a new definition of acceptance conditions. Our proofs and experiments have shown that our technique is more efficient and improves the degeneralization ability. Finally, we introduce the simplifications of our algorithm. We focus on not only producing better final Buchi automata but also minimizing intermediate automata, which can reduce the execution time."},"correct_answer":{"kind":"string","value":"Translating linear temporal logic (LTL) formulas into Buchi automata is one of the most important aspects of LTL model checking. Certain successful algorithms, such as LTL2BA and SPOT, first translate an LTL formula into a transition-based generalized Buchi automaton (TGBA) and then degeneralize it into a Buchi automaton. This paper focuses on achieving a better translation from LTL to TGBA and analyzing the performance of every step of the algorithm. We decompose the translation into three steps to give a step-wise description and improve all three steps. The first step is the basic translation without acceptance conditions and simplifications, which combines the advantages of both LTL2BA and SPOT. Second, we introduce a new definition of acceptance conditions. Our proofs and experiments have shown that our technique is more efficient and improves the degeneralization ability. Finally, we introduce the simplifications of our algorithm. We focus on not only producing better final Buchi automata but also minimizing intermediate automata, which can reduce the execution time."},"documents":{"kind":"list like","value":["Title: A complete development process of finite element software for body-in-white structure with semi-rigid beams in .NET framework\nYear: 2012\n\nAbstract: At the conceptual design stage, simplified finite element (FE) model of body-in-white (BIW) structure focuses on its specific merit to provide early-stage predictions for detailed FE model of that. This paper exploits a semi-rigid beam element (SRBE) that consists of a beam element with two semi-rigid connections at the ends to simulate the flexibility of joint. Guyan reduction method condenses the SRBE as a super element. A special finite element software for structural modeling and analysis of BIW (SMAB) is developed in .NET framework. The Unified Modeling Language is employed to depict the classes and their relationship. The design patterns are identified and applied in the framework design to facilitate communication and system expansion. Microsoft DirectX and GDI+ implement graphics display of spatial BIW frame and planar thin-walled cross section. Based on multi-threaded technology in .NET, subspace iteration method is paralleled to speed up the mode analysis. As a result, the efficiency of the SRBE is demonstrated by a benchmarking automotive body. Multi-threaded parallel is effective and useful, especially for frequency optimization.","Title: From task parameters to motor synergies: A hierarchical framework for approximately-optimal control of redundant manipulators.\nYear: 2005\n\nAbstract: We present a hierarchical framework for approximately-optimal control of redundant manipulators. The plant is augmented with a low-level feedback controller, designed to yield input-output behavior that captures the task-relevant aspects of plant dynamics but has reduced dimensionality. This makes it possible to reformulate the optimal control problem in terms of the augmented dynamics, and optimize a high-level feedback controller without running into the curse of dimensionality. The resulting control hierarchy compares favorably to existing methods in robotics. Furthermore we demonstrate a number of similarities to (non-hierarchical) optimal feedback control. Besides its engineering applications, the new framework addresses a key unresolved problem in the neural control of movement. It has long been hypothesized that coordination involves selective control of task parameters via muscle synergies, but the link between these parameters and the synergies capable of controlling them has remained elusive. Our framework provides this missing link.","Title: Estimation and Control of Systems with Multiplicative Noise via Linear Matrix Inequalities\nYear: 2005\n\nAbstract: This paper provides estimation and control de- sign methods for linear discrete-time systems with multi- plicative noise. First, we present the design of the state feedback controller, such that the closed loop system is mean square stable. Second, we provide sufficient conditions for the existence of the state estimator; these conditions are expressed in terms of linear matrix inequalities (LMIs), and the parametrization of all admissible solutions is addressed. Finally, an estimator design approach is formulated using LMIs, and the performance of the estimator is examined by means of numerical examples.","Title: Iterative Linear Quadratic Regulator Design for Nonlinear Biological Movement Systems\nYear: 2004\n\nAbstract: This paper presents an Iterative Linear Quadratic Regulator (ILQR) method for locally-optimal feedback con- trol of nonlinear dynamical systems. The method is applied to a musculo-skeletal arm model with 10 state dimensions and 6 controls, and is used to compute energy-optimal reaching movements. Numerical compar- isons with three existing methods demonstrate that the new method converges substantially faster and finds slightly better solutions.","Title: Synthetic Fault Diagnosis Method of Power Transformer Based on Rough Set Theory and Improved Artificial Immune Network Classification Algorithm\nYear: 2008\n\nAbstract: According to complementary strategy, this paper presents a new power transformer fault diagnosis method based on rough sets theory (RST)and improved artificial immune network classification algorithm. Through reduction approach of RST information table to simplify expert knowledge and reduce fault symptoms, the minimal diagnostic rules can be obtained. An improved artificial immune network classification algorithm is proposed on the base of them. At first the artificial immune network which both antigens and memory antibodies with class information has been added into are trained to learn the features of fault samples. In this way, the memory antibody cells pool which can represent the fault samples better than those without class information can be obtained. Then the k-nearest neighbor method is used to classify the fault samples. Compared with the IEC three-ratio method and BP neural network (BPNN), the proposed algorithm has better capability to classify single-fault and multiple-fault samples as well as higher diagnosis precision.","Title: An approach for testing safety-critical software\nYear: 1999\n\nAbstract: A novel approach for testing the effectiveness, efficiency, safety and relative appropriateness of Computer Interlocking Software (CIS)-a kind of safety-critical software is presented with a software platform developed to support this approach. A brief description of the proposed approach is also included","Title: An optimization viewpoint of decision-theoretic rough set model\nYear: 2011\n\nAbstract: This paper considers an optimization viewpoint of decisiontheoretic rough set model. An optimization problem is proposed by considering the minimization of the decision cost. Based on the optimization problem, cost functions and thresholds used in decision-theoretic rough set model can be learned from the given data automatically. An adaptive learning algorithm Alcofa is proposed. Another significant inference drawn from the solution of the optimization problem is a minimum cost based attribute reduction. The attribute reduction can be interpreted as finding the minimal attribute set to make the decision cost minimum. The optimization viewpoint can bring some new insights into the research on decision-theoretic rough set model.","Title: Fault Diagnosis of Power Transformers Based on BP Network with Clonal Selection Algorithm\nYear: 2007\n\nAbstract: In this paper, a novel approach based on BP network (BPN) for fault diagnosis of power transformers is proposed. Optimization of BPN weights is achieved by using clonal selection algorithms (CSA). In addition, the mutation probability is adjusted adaptively according to the affinity of antibody. Compared with previous approaches, this one can avoid prematurity effectively, with good self-learning and self-memory ability. The experiment results show that the presented approach outperforms previous ones in both classification accuracy and computational efficiency.","Title: Checkpoint scheduling model for optimality\nYear: 2011\n\nAbstract: To minimize the expected execution time, a general checkpoint scheduling algorithm is proposed to determine the near optimal checkpointing time sequence. More precisely, based on a simple timing policy, an execution analytical model is introduced and the expected effective ratio is derived. By maximizing the expected effective ratio, the optimal checkpoint period for the exponential failure distribution can be obtained directly, and a general checkpoint scheduling algorithm is developed to perform the near optimal checkpointing time sequence for an arbitrary failure distribution. Experimental results reveal that the proposal can perform varying checkpoint interval according to the failure distribution and the expected effective ratio of the execution is considerable for the long-running application in term of reliability.","Title: Grid Synchronization Systems Of Three-Phase Grid-Connected Power Converters: A Complex-Vector-Filter Perspective\nYear: 2014\n\nAbstract: Due to the significance of extracting the grid voltage information, the grid synchronization system plays an important role in the control of grid-connected power converters, and various grid voltage synchronization schemes have been proposed. This paper adopts the complex-vector-filter method (CVFM) to analyze the grid synchronization systems. With this method, the pairs of scalar signals, for example, the alpha- and beta-axis components in the stationary alpha - beta frame, are combined into one complex vector. As a consequence, the grid synchronization systems can be described with the complex transfer functions, which is very convenient to evaluate the steady-state performance, for example, the fundamental and harmonic sequence decoupling/cancellation, and dynamic performance of these systems. Moreover, the CVFM also provides a more generalized perspective to understand and develop the grid synchronization systems. Therefore, some of the representative systems are reanalyzed with the CVFM in this paper. A generalized second-order complex-vector filter and a third-order complex-vector filter are proposed with the CVFM to achieve better dynamic performance or higher harmonic attenuation. Moreover, a brief comparison of the complex-vector filters analyzed in this paper is presented. The effectiveness of the CVFM and the proposed two complex-vector filters are verified by the simulation and experimental results.","Title: Linear quadratic optimal control of contact transition with fingertip\nYear: 2009\n\nAbstract: This paper proposes an optimal control methodology that addresses the problem of control of fingertip during a general class of task that requires the fingertip to make a transition from non-contact motion to contact motion. Specifically, the task that the fingertip makes and transitions from motion to static well-directed force production. Here we present a mathematical framework for controlling the contact transition, while switching between non-contact and contact controller is needed and handled by the optimal control strategy. The non-linear differential algebraic equation that describes the dynamics of the index finger is linearized, and then a modified linear quadratic optimal control problem is solved. The resulting optimal feedback control law guarantees good regulation of contact force, velocity and position. Simulation results are presented to demonstrate the effectiveness of the new approach.","Title: Network criticality in vehicular networks\nYear: 2012\n\nAbstract: Network criticality (resistance distance) is a graph-theoretic metric that quantifies network robustness, and that was originally designed to capture the effect of environmental changes in core communication networks. This paper establishes a relationship between information centrality and network criticality and provides a justification for using the average network criticality of a node to quantify the nodes relative importance in a graph.This results provides a basis for designing robust clustering algorithms for vehicular networks.","Title: An Iterative Optimal Control and Estimation Design for Nonlinear Stochastic System\nYear: 2006\n\nAbstract: This paper presents an iterative linear-quadratic-Gaussian method for locally-optimal control and estimation of nonlinear stochastic systems. The new method constructs an affine feedback control law obtained by minimizing a novel quadratic approximation to the optimal cost-to-go function. It also constructs a non-adaptive filter optimized with respect to the current control law. The control law and filter are iteratively improved until convergence. The performance of the algorithm is illustrated on a complex biomechanical control problem involving a stochastic model of the human arm","Title: A Method for Improving Concurrent Write Performance by Dynamic Mapping Virtual Storage System Combined with Cache Management\nYear: 2011\n\nAbstract: The paper presents a new dynamic mapping virtual storage system, which is characterized by: (a) distribute storage resources according to need, thus improving storage resource utilization, (b) more fundamentally, through write-anywhere pattern to sequentialize concurrent write requests by dynamic address mapping mechanism, thus improving concurrent write performance for applications. Meanwhile, taking into account imperfection small size write requests handling and the negative impact on read performance, we combine the system with CBD system which is responsible for optimizing cache management. We have implemented our virtual storage system in Linux kernel 2.6.18 as a pseudo device driver. The experiment results show that concurrent write bandwidth approaches raw disk write performance. And in 64-stream concurrent write case, E-ASD can outperform concurrent write bandwidth of LVM by up to 160%, and its max single-stream read loss is less than 25% compared to LVM, with concur-rent read performance close to each other. Read performance still has space to be improved given larger cache and proper prefetching.","Title: Barriers to mobile commerce adoption: an analysis framework for a country-level perspective\nYear: 2008\n\nAbstract: This research examines and categorises the country-level adoption barriers of mobile commerce services, and tests those barriers in a case study of one country. The outcomes of the research presented in this paper are a broad framework of mobile commerce barriers that might be applicable when undertaking a country-level analysis, and a case study which tests the value of this framework in helping to uncover the specific barriers of mobile commerce adoption in one example country, New Zealand.","Title: Full-Feedforward Schemes Of Grid Voltages For A Three-Phase Lcl-Type Grid-Connected Inverter\nYear: 2013\n\nAbstract: This paper investigates the feedforward schemes of grid voltages for a three-phase LCL-type grid-connected inverter. The full-feedforward functions of grid voltages are derived for the stationary alpha - beta frame, synchronous d - q frame, and decoupled synchronous d - q frame-controlled three-phase LCL-type grid-connected inverters. The derived full-feedforward functions mainly consist of three parts which are proportional, derivative, and second derivative parts. The use of the traditional proportional feedforward function in the three-phase LCL-type grid-connected inverter will result in the amplification of the high-frequency injected grid current harmonics. With the proposed full-feedforward schemes, the injected grid current harmonics and unbalance caused by grid voltages can be greatly reduced. The effectiveness of the proposed feedforward schemes is verified by the experimental results.","Title: Robust clustering for connected vehicles using local network criticality\nYear: 2012\n\nAbstract: This paper proposes a robust Criticality-based Clustering Algorithm (CCA) for Vehicular Ad Hoc NETworks (VANETs) based on the concept of network criticality. Network criticality is a global metric on an undirected graph, that quantifies the robustness of the graph against environmental changes such as topology. In this paper, we localize the notion of network criticality and apply it to control cluster formation in the vehicular wireless network. We use the localized notion of node criticality together with a universal link measure, Link Expiration Time (LET), to derive a distributed multi-hop clustering algorithm for VANETs. Simulation results show that the proposed CCA forms robust cluster structures.","Title: Discovering structural regularity in facade image\nYear: 2010\n\nAbstract: 3D reconstruction from single view is a very difficult problem in computer vision and computer graphics. Few works have been done in this area. In this paper, we proposed a method to discover structural regularity in single facade images on the procedural modeling of buildings by using SIFT algorithm. Firstly, symmetry information is detected from the image using SIFT algorithm for image analysis. Next, the symmetry information will be removed from the facade image. Finally, get the SIFT feature of widows and doors. Experiments show that our method can successfully discover regular structures with a higher recognition rate than existing method especially amidst shadowing and the changing illumination environment.","Title: Automatic segmentation of focused objects from images with low depth of field\nYear: 2010\n\nAbstract: In this paper, we propose an automatic segmentation approach to extract focused objects from images with low depth of field (DOF). A focus energy map is first estimated based on the difference of high-frequency components between focused region and defocused background, and is exploited to construct region/boundary saliency maps on the basis of a pre-segmentation result by watershed transform. Then region/boundary masks for focused object are generated by entropy thresholding and flood filling, and an efficient boundary linking method is proposed to obtain closed region/boundary masks, which are exploited to reasonably generate a trimap containing seed regions for focused object and defocused background, and uncertain regions, respectively. Finally, the trimap is used as the input to an image matting model, which is utilized to classify the pixels in the uncertain regions to obtain an accurate focused object segmentation result based on the estimated alpha matte. Experimental results for a variety of low DOF images demonstrate the good segmentation performance of the proposed approach.","Title: Robust discovery of partial rigid symmetries on 3D models\nYear: 2010\n\nAbstract: The ubiquity of symmetry in nature and man-made artifacts has made symmetry discovery an important tool for numerous applications. The most fundamental and visually prominent kind of symmetry is partial rigid symmetry, which is explained as the invariance between parts of a 3D model under a set of translation, rotation, reflection, and uniform scaling generators. Thus automatic discovery of partial rigid symmetry on general 3D models, with no assumption on the size, shape or location of the symmetric parts, keeps to be a hot topic during recent years. Among such kind of works, the transformation voting technique [Mitra et al. 2006] is most widely used, due to its high efficiency and easiness for understanding and implementation.","Title: Salient Object Extraction Based on Region Saliency Ratio\nYear: 2009\n\nAbstract: This paper proposes an efficient approach to extract salient objects in an image. A scale-invariant saliency map is first constructed based on a multi-resolution feature contrast calculation, meanwhile the image is segmented into homogenous regions using nonparametric kernel density estimation (NKDE). Then the region saliency ratio of each region combination to its complement is calculated in turn. Finally, salient objects are extracted by maximizing the region saliency ratio. Experimental results demonstrate the effectiveness of the proposed approach.","Title: The e-derivative of Boolean functions and its application in the fault detection and cryptographic system.\nYear: 2011\n\nAbstract: Purpose - The purpose of this paper is to show that the e-derivative is of great significance when revealing cryptographic characteristics of Boolean functions and assurance of the safety of the cryptographic system. Design/methodology/approach - A new definition is introduced in this paper. Findings - After introducing the e-derivative, the authors obtain some relevant cryptographic properties and theorems which can fully reveal the relationships between characteristics and structures of Boolean functions. In logic circuit fields, the authors conduct the test to all the signals with numerical value 1 in the transmission line using the e-derivative, which improves the ability of the test. In addition, when there is no expression of the transmission line the authors will obtain it through the vector, and then do troubleshooting. Research limitations/implications - Logic circuit is the main limitation in the fault detection. Practical implications - A very useful method for assurance of the safety of cryptographic systems. Originality/value - This paper is the first time, in cryptology fields, that the e-derivative of Boolean functions as a new definition is introduced to fully reveal the relationships between characteristics and structures of Boolean functions. In addition, they can also be applied to the locating test of the breakdown, which improves the ability of the test.","Title: Optimized Controller Design for -Type Grid-Connected Inverter to Achieve High Robustness Against Grid-Impedance Variation\nYear: 2015\n\nAbstract: Capacitor-current-feedback active damping is an effective method to suppress the LCL-filter resonance in grid-connected inverters. However, due to the variation of grid impedance, the LCL-filter resonance frequency will vary in a wide range, which challenges the design of the capacitor-current-feedback coefficient. Moreover, if the resonance frequency is equal to one-sixth of the sampling frequency (fs/6), the digitally controlled LCL-type grid-connected inverter can be hardly stable no matter how much the capacitor-current-feedback coefficient is. In this paper, the optimal design of the capacitor-current-feedback coefficient is presented to deal with the wide-range variation of grid impedance. First, the gain margin requirements for system stability are derived under various resonance frequencies. By evaluating the effect of grid impedance on gain margins, an optimal capacitor-current-feedback coefficient is obtained. With this feedback coefficient, stable operations will be retained for all resonance frequencies except (fs/6). Second, in order to improve system stability for a resonance frequency of (fs/6), a phase-lag compensation for the loop gain is proposed. Finally, a 6-kW prototype is tested to verify the proposed design procedure.","Title: Service Outsourcing Character Oriented Privacy Conflict Detection Method in Cloud Computing.\nYear: 2014\n\nAbstract: Cloud computing has provided services for users as a software paradigm. However, it is difficult to ensure privacy information security because of its opening, virtualization, and service outsourcing features. Therefore how to protect user privacy information has become a research focus. In this paper, firstly, we model service privacy policy and user privacy preference with description logic. Secondly, we use the pellet reasonor to verify the consistency and satisfiability, so as to detect the privacy conflict between services and user. Thirdly, we present the algorithm of detecting privacy conflict in the process of cloud service composition and prove the correctness and feasibility of this method by case study and experiment analysis. Our method can reduce the risk of user sensitive privacy information being illegally used and propagated by outsourcing services. In the meantime, the method avoids the exception in the process of service composition by the privacy conflict, and improves the trust degree of cloud service providers.","Title: Video quality of experience in the presence of accessibility and retainability failures\nYear: 2014\n\nAbstract: Accurate Quality of Experience measurement for streaming video has become more crucial with the increase in demand for online video viewing. Quantifying video Quality of Experience is a challenging task. Significant efforts to quantify video Quality of Experience have primarily focused on the measurement of Quality of Experience for videos with network and compression related impairments. These impairments, however, may not always be the only main factors affecting Quality of Experience in an entire video viewing session. In this paper, we evaluate Quality of Experience for entire video viewing sessions, from the beginning to the end. In doing so, we evaluate videos with temporary interruptions as well as those with permanent interruptions or failures. We consider two types of failures, namely Accessibility and Retainability failures, and present the results of two subjective studies. These results indicate: (a) Accessibility and Retainability failures are rated lower compared to temporary interruption impairments; (b) Accessibility failures are rated close to the lowest value on the rating scale; and (c) the traditionally used 5-point scale to measure video Quality of Experience is not sufficient in the presence of Accessibility and Retainability failures.","Title: Impact of Retainability Failures on Video Quality of Experience\nYear: 2014\n\nAbstract: Measurement of video Quality of Experience is needed to enable telecommunication service operators to provide acceptable video viewing experience to their customers. Without accurate Quality of Experience measurement, it is hard to predict customer satisfaction. Accessibility and Retain ability failures are important factors that can affect the experience of viewing video on the Internet. Most of the published work in the assessment of Video Quality of Experience has focused on the quality of video viewing sessions without considering the impact of Accessibility and Retain ability failures. In this paper, we focus on the Quality of Experience for video sessions in the presence of Retain ability failures. We primarily evaluate video Quality of Experience for videos with temporary interruptions, called Integrity impairments, as well as with permanent interruptions called Retain ability failures. A subjective study was conducted to observe the effect of Retain ability failure on video Quality of Experience. The results of the subjective study suggest that videos with Retain ability failures are rated much lower than videos without any failure. An initial linear model for QoE of videos subject to Retain ability failures is introduced.","Title: Preservation of P-Positions between a Game and Its Restricted Game\nYear: 2014\n\nAbstract: E. Duchene et al investigate games whose sets of allowed moves are subsets of Wythoff's one, and whose set of P-positions is exactly Wythoff's sequence. Their results show that such a game does not exist. In this paper, we find a pair of games, one is the restricted version of the other, such that they have the same set of P-positions. We also find another pair of games, one is the restricted version of the other, such that they do not have the same set of P-positions. This means that the preservation of P-positions between a game and its restricted game depends on the game itself.","Title: Compressed sensing-based unequal error protection by linear codes\nYear: 2014\n\nAbstract: In many wireless communication systems, data can be divided into different importance levels. For these systems, unequal error protection (UEP) techniques are used to ensure lower bit error rate for the more important classes. Moreover, if the precise characteristics of the channel are known, UEP can be used to correctly recover the more important classes even under severe receiving conditions. In this study, a UEP scheme based on compressed sensing via a linear program is proposed. Discrete wavelet transform (DWT) is chosen as the sparsifying basis, and then DWT-coded information is divided into two-layered coded streams, each of which is transmitted differentially by applying an unequal number of information bits in linear codes according to the time-varying characteristic of the corrupted channel. In this proposed transmission scheme, the more important information is to guarantee error-free transmission. At the decoder, one can simply reconstruct the signal via the l1-minimisation algorithm. Simulation results show that the proposed scheme can achieve a higher peak signal-to-noise ratio (PSNR) and obviously improve the error resilience compared to the equal error protection scheme and other UEP methods. More importantly, with the increase of channel corrupted ratio, the drop rate of PSNR is much slower than other solutions. It indicates that the proposed method has better robustness for severe channel conditions.","Title: A generalized iterative LQG method for locally-optimal feedback control of constrained nonlinear stochastic systems\nYear: 2005\n\nAbstract: We present an iterative Linear-Quadratic- Gaussian method for locally-optimal feedback control of nonlinear stochastic systems subject to control constraints. Previously, similar methods have been restricted to determin- istic unconstrained problems with quadratic costs. The new method constructs an afÞne feedback control law, obtained by minimizing a novel quadratic approximation to the optimal cost-to-go function. Global convergence is guaranteed through a Levenberg-Marquardt method; convergence in the vicinity of a local minimum is quadratic. Performance is illustrated on a limited-torque inverted pendulum problem, as well as a complex biomechanical control problem involving a stochastic model of the human arm, with 10 state dimensions and 6 muscle actuators. A Matlab implementation of the new algorithm is availabe at www.cogsci.ucsd.edu/~todorov.","Title: Three-way decisions based software defect prediction\nYear: 2016\n\nAbstract: Based on a two-stage classification method and a two-stage ranking method on three-way decisions, this paper introduces a three-way decisions framework for cost-sensitive software defect prediction. For the classification problem in software defect prediction, traditional two-way decisions methods usually generate a higher classification error and more decision cost. Here, a two-stage classification method that integrates three-way decisions and ensemble learning to predict software defect is proposed. Experimental results on NASA data sets show that our method can obtain a higher accuracy and a lower decision cost. For the ranking problem in software defect prediction, a two-stage ranking method is introduced. In the first stage, all software modules are classified into three different regions based on three-way decisions. A dominance relation rough set based ranking algorithm is next applied to rank the modules in each region. Comparison experiments with 6 other ranking methods present that our proposed method can obtain a better result on FPA measure.","Title: Iterative Linearization Methods For Approximately Optimal Control And Estimation Of Non-Linear Stochastic System\nYear: 2007\n\nAbstract: This paper presents an iterative Linear-Quadratic-Gaussian method for locally-optimal control and estimation of non-linear stochastic systems. The new method constructs an affine feedback control law, obtained by minimizing a novel quadratic approximation to the optimal cost-to-go function, and a non-adaptive estimator optimized with respect to the current control law. The control law and filter are iteratively improved until convergence. The performance of the algorithm is illustrated on a complex biomechanical control problem involving a stochastic model of the human arm.","Title: Prediction of facial deformation after complete denture prosthesis using BP neural network\nYear: 2015\n\nAbstract: With the accelerated aging of world population, complete denture prosthesis plays an increasingly important role in mouth rehabilitation. In addition to recovering stomatognathic system function, restoring the appearance of a third of the area under the face has become a great challenge in complete denture prosthesis. This study analyzes the interactive relationship between the appearance of a third of the area under the face and complete denture, and proposes a new method to predict facial deformation after complete denture prosthesis. Firstly, to improve computational efficiency, the feature template is constructed to replace the deformed facial region. Secondly, a forecast model of elastic deformation is constructed using BP neural network and predicts elastic deformation amount because of the inhomogeneous, anisotropic and nonlinear material properties of soft tissue. Finally, a new feature template is calculated using deformation amount, and the deformation of preoperative model is simulated using Laplacian deformation technique. The average error rates of different hidden layer nodes in the neural network are analysed. Deformation and postoperative models are superimposed for match analysis. Experimental results show that this method can predict facial soft tissue deformation quickly and accurately. BP neural network is used to predict feature positions on facial model.The deformation of facial model is simulated using Laplacian deformation technique.This method can predict facial deformation quickly and accurately.CT scan is not necessary.","Title: Neighborhood based decision-theoretic rough set models\nYear: 2016\n\nAbstract: As an extension of Pawlak rough set model, decision-theoretic rough set model (DTRS) adopts the Bayesian decision theory to compute the required thresholds in probabilistic rough set models. It gives a new semantic interpretation of the positive, boundary and negative regions by using three-way decisions. DTRS has been widely discussed and applied in data mining and decision making. However, one limitation of DTRS is its lack of ability to deal with numerical data directly. In order to overcome this disadvantage and extend the theory of DTRS, this paper proposes a neighborhood based decision-theoretic rough set model (NDTRS) under the framework of DTRS. Basic concepts of NDTRS are introduced. A positive region related attribute reduct and a minimum cost attribute reduct in the proposed model are defined and analyzed. Experimental results show that our methods can get a short reduct. Furthermore, a new neighborhood classifier based on three-way decisions is constructed and compared with other classifiers. Comparison experiments show that the proposed classifier can get a high accuracy and a low misclassification cost.","Title: DRUGS System Improving the Effects of Clinical Pathways: A Systematic Study\nYear: 2016\n\nAbstract: Abstract The aim of the study is to assess the feasibility of Drugs Rational Usage Guideline System (DRUGS)-supported clinical pathway (CP) for breast carcinoma, cataract, inguinal hernia and 2-diabetes mellitus whether the application of such a system could improve work efficiency, medical safety, and decrease hospital cost. Four kinds of diseases which included 1773 cases (where 901 cases using paper-based clinical pathways and 872 cases using DRUGS-supported clinical pathways) were selected and their demographic and clinical data were collected. The evaluation criteria were length of stay, preoperative length of stay, hospital cost, antibiotics prescribed during hospitalization, unscheduled surgery, complications and prognosis. The median total LOS was 1 to 3 days shorter in the DRUGS-supported CP group as compared to the Paper-based CP group for all types (p < 0.05). Totel hospital cost decreased significantly in the DRUGS-supported CP group than that in Paper-based CP group. About antibiotics prescribed during hospitalization, there were no statistically differences in the time of initial dose of antibiotic and the duration of administration except the choice of antibiotic categories. The proportion of DRUGS-supported clinical pathway conditions where a broad-spectrum antibiotic was prescribed decreased from 63.6 to 34.5 % (p < 0.01) in the Paper-based group. While after the intervention, the differences were statistically not significant in unscheduled surgery, complications and prognosis. In this study, DRUGS-supported clinical pathway for breast carcinoma, cataract, inguinal hernia, 2-diabetes mellitus was smoothly shifted from a paper-based to an electronic system, and confer benefits at the hospital level.","Title: Impact Of Technical And Content Quality On Overall Experience Of Ott Video\nYear: 2016\n\nAbstract: Quality of Experience (QoE) is a crucial guiding factor for network management of an end-to-end service session. The network provider can control the resources allocated to sessions and in doing so, influence the Technical Quality (TQ), which covers the technical aspects of signal quality during the session. On the other hand, the network provider has no control over the Content Quality (CQ), which pertains to the user's level of interest in a particular video. Together TQ and CQ influence the Overall eXperience (OX) in a session. In this paper, we present results from a user subjective study in which the impact of TQ and CQ on OX was investigated for Over-The-Top (OTT) video sessions from the perspective of a network provider. This perspective places a focus on those elements of QoE that can be controlled by the provider. Various studies have shown that very high interest in a content can strongly influence QoE independent of other factors, so our study uses videos that are neutral with respect to content. We assess the TQ, CQ and OX for video sessions that contain Integrity impairments (in the form of image freezing) and failures in terms of session Accessibility and Retainability. Our findings indicate that TQ and CQ have a strong impact on OX in the presence of impairments, but no failures. On the other hand, TQ is the main determinant of OX when failures are present.","Title: Privacy Perceptive Wireless Service Game Model\nYear: 2014\n\nAbstract: Wireless applications are taking an important role in our lives. More and more merchants begin to provide wireless connection so as to attract customers. For business purpose, they often ask customers to provide their personal information for using wireless, which may threat customer. So, it is desired to make a trade off between the competitive profits of both service providers and user privacy. In this paper, we introduce a privacy perceptive service model based on game theory, which takes into account the interests of all stakeholders. Considering the characteristics of customer behavior in practice, we design the complete information and sequential game. Users are allowed to provide different levels of personal information as their preferences, while a merchant assigns different wireless bandwidth based on customer provided information. The Nash Equilibriums are discussed so as to find the acceptable strategy. Quantitative study is performed on the impact of each parameter in the game, which could help merchants to design appropriate policies.","Title: Research on Value Evaluation of E-commerce Business Model\nYear: 2015\n\nAbstract: The study of business models is an important topic for entrepreneurship and organizational purpose because business models affect firms' possibilities for value creation and value capture. The present study leaves a gap to be bridged, and the gap reveals a critically important research problem to be resolved, that is, how to visualize these transaction activities and calculate revenue respectively. This paper, therefore, adopts e3 model for investigations of E-commerce enterprise value generating in the process, and can be used to select or optimize value activities that advance both research and practice in the field of E-commerce enterprises. The result indicates that: Profitability sheet is able to capture customers' behaviors, which gives them insights to better understanding online customers' experiences and expectations, and make reasonable decisions. This model illustrates the main activities of business model and exploits the ways to optimize value activities. This modeling approach provides pro-activities way to actually experiment with alternative business models, by making enterprise to simulate various possibilities before committing to specific investments in reality. (C) 2015 The Authors. Published by Elsevier B.V.","Title: Acceptability And Quality Of Experience In Over The Top Video\nYear: 2015\n\nAbstract: Consumer acceptance is of great interest in the adoption of novel multimedia products and services. A number of factors can greatly influence the customer experience during a video session, impacting the acceptability of the product or service. Factors such as the Technical Quality (TQ) which covers the technical aspects of the signal quality can be controlled from the network provider. On the other hand, the network provider has no control on the subject's level of interest in a video, the Content Quality (CQ). Together TQ and CQ influence the Overall eXperience (OX) of the subject and, in the case of Over The Top (OTT) video, in a video session. In this paper, we present results from a user subjective study in which the impact of TQ and OX on the acceptability was investigated for OTT video sessions. To minimize the CQ impact on OX, the videos were carefully selected to have relatively neutral content. We assess the impact of TQ and OX on acceptability for videos containing impairment and failure events during lifecycle, thus affecting Accessibility, Retainability and Integrity. Our experimental results indicate that TQ and OX have a strong impact on acceptability.","Title: A Multicriteria Interval-Valued Intuitionistic Fuzzy Set TOPSIS Decision-Making Approach Based on the Improved Score Function.\nYear: 2016\n\nAbstract: In this paper, an improved technique for order preference by similarity to an ideal solution (TOPSIS) method was proposed based on completely unknown attribute weight information, as well as taking multiple criteria decision making of interval-valued intuitionistic fuzzy number as the evaluation information. First of all, a cumulative interval score function considering the influence of hesitation as well as a cumulative score function containing the risk preference of the decision maker were constructed. Then, a new information entropy function was constructed by using the newly defined score function, which presents a new method that directly utilizes evaluation information to obtain criterion weight. Next, all schemes were sequenced by virtue of relative closeness and the criterion weight of each alternative and ideal scheme. Finally, the effectiveness of the proposed method was illustrated by comparison with examples.","Title: Singular boundary method for acoustic eigenanalysis.\nYear: 2016\n\nAbstract: This paper applies the singular boundary method (SBM) to two- (2D) and three-dimensional (3D) acoustics eigenproblems in simply- and multiply-connected domains. The SBM is a strong-form boundary discretization numerical method and is meshless, integration-free, and easy-to-implement. By introducing the concept of the source intensity factors, the singularity of fundamental solutions can be isolated to avoid the singular numerical integrals in the boundary element method (BEM). Similar to the BEM, the spurious eigenvalues may arise in the SBM computation. In order to extract out spurious eigenvalues from SBM results, the singular value decomposition updating techniques and the Burton-Miller method are implemented. Several 2D and 3D benchmark examples subjected to Dirichlet and Neumann boundaries are tested to examine the accuracy and stability of the present SBM strategy.","Title: The Interaction Design Of Household Intelligent Breathing Training System\nYear: 2016\n\nAbstract: This paper achieved the acquisition of the breathing signal and identification of different respiratory modes by several attempts. On this basis, we designed an intelligent breathing-exercise system to meet the household needs. The intelligent system consists of three parts: the first part is the user interface of the breathing-exercise system, the second part is the background database and the last part is the internet community platform. The user interface of the breathing-exercise system is used for displaying user's breathing curve and other visual feedback. All the breathing training data will be recorded in the second part: database. Those breathing exercisers can share their results on the internet community platform. The intelligent breathing-exercise system will provide a memorable user experience for users by blending these three parts.","Title: Understanding The Relationships Between Performance Metrics And Qoe For Over-The-Top Video\nYear: 2016\n\nAbstract: In this paper, we study the relationships between Quality of Service (QoS) and Quality of Experience (QoE) in a session-based Over-The-Top (OTT) video service. A number of Performance Metrics (PMs) with and without the existence of failures during a video are examined. As QoE factors, Technical Quality (TQ) and Acceptability are used. We analyze the correlation between QoS performance metrics and QoE factors, and find new PMs should be employed because failures are included in QoE evaluation. We also summarize the relationships between QoS metrics and QoE factors through machine learning approaches. Using decision tree, we have a general idea about the relationships between PMs and QoE factors. We also understand the impact caused by failures and the value of rating scales.","Title: Risk measurement and optimization model of coal generation contracts for the difference between prospect M-V and normal triangular fuzzy stochastic variables.\nYear: 2016\n\nAbstract: Purpose - Coal and power generation are related upstream and downstream industries. Coal price marketization and electricity price regulation have caused the price of coal to be sensitive to the benefits of generators. The paper aims to discuss these issues. Design/methodology/approach - As a financial tool, contracts for differences can both help balance interests and reduce risks caused by spot price fluctuation. This thesis regards coal demand as a triangular fuzzy stochastic variable while directing a levelling consideration towards risk returns for coal and power enterprises that are involved in coal generation contracts for differences. Risk and benefit measurement models were established between coal suppliers and power generators, and risk and benefit balance optimization models for contract negotiation were constructed. Findings - A numerical example showed that the above models can be effectively used to avoid the risks of coal-electricity parties. Originality/value - This thesis regards coal demand as a triangular fuzzy random variable while directing a levelling consideration towards the risk return to coal and power enterprises that are involved with coal generation contracts for differences. The features of this thesis are the following: demand information is regarded as a fuzzy random variable instead of a random variable. With historical data, sales experience and increasingly clear macro-economic conditions, coal and power enterprises are able to make a fuzzy decision - to a certain extent - when the transaction approaches. Accurate market information enables the supply chain system to satisfy the clients' needs better, improve the profit level or avoid severe financial damages; by developing a feasible set of contracts for different parameters, it is possible to estimate whether the price difference enables supply chain coordination, requires changes or gives accounts to all involved parties of the supply chain; and without the assumption that the traditional M-V rule is unfavourable to decision makers, this thesis proposes the prospect M-V rule, which involves decision makers' projections of future coal generation prices and enables wide applicability of the response method to contracts for differences.","Title: Chinese Microblog Sentiment Analysis Based on Sentiment Features.\nYear: 2016\n\nAbstract: As the microblog has increasingly become an information platform for netizens to share their ideas, the study on the sentiment analysis of microblog has got scholars’ wide attention both at home and abroad. The primary goal of this research is to improve the accuracy of microblog sentiment polarity classification. With a view to the characteristics of microblog, a new method of semantically related feature extraction is proposed. Firstly, the Chinese word features are selected by text presentation in VSM and computing the weight by TF*IDF. Secondly, the proposed eight microblog semantic features are extracted, including sentence sentiment judgment based on emotional dictionary. Finally, three kinds of machine learning methods are used to classify the Chinese microblog under the feature vector combining the two methods. The experimental results indicate that the proposed feature extraction method outperforms the state-of-the-art approaches, and for this feature extraction algorithm, the classification performance is best when using the Naive Bayes algorithm.","Title: Band gap calculations of photonic crystals by singular boundary method.\nYear: 2017\n\nAbstract: This paper applies the singular boundary method (SBM) to compute band gap structures of two-dimensional (2D) photonic crystals. By using the fundamental solutions, a linear eigenvalue equation is established to describe the relationship between the Block wave and frequency. For given frequencies, the solution of the equation yields the related Block wave vectors. The convergence, accuracy and efficiency are carefully examined through several benchmark cases.","Title: Capturing User Behavior in Subjective Quality Assessment of OTT Video Service.\nYear: 2016\n\nAbstract: Customer satisfaction is an important factor governing adoption and retention of multimedia products and services, such as Over-The-Top (OTT) video transmission. Quality of Experience involves user-centric evaluation of various services. However, users differ in terms of their ratings of service quality. Some rating differences are due to unreliability(outlierusers who are not motivated, or are not sensitive to differences in quality), but others are systematic differences in rating that may reflect different perspectives on quality. In this paper, we explore the use of outlier analysis and clustering as tools for interpreting QoE data. We report on experimental results demonstrating the use of outlier analysis and clustering. In interpreting the clusters, we examine users' opinions on different types of video disruption, and their ability to distinguish the different levels of impairments/failures.","Title: A Cross-Layer Approach To Message Authentication Based On Sparse Representation For Wireless Body Area Networks\nYear: 2017\n\nAbstract: Due to personal privacy, information security is a particularly important issue in wireless body area networks. Although current communication protocol has considered the security, there is still a risk that lapses in physical layer security may be a potential threat in wireless body area networks. In order to achieve the comprehensive security protection, in this article, we propose a cross-layer authentication approach, where the higher layer and physical layer are simultaneously considered. More specifically, different from current literatures, we exploit the sparse representation of the original signal as the basis of physical layer authentication, and a representative selection algorithm is presented to achieve the signal detection. Moreover, based on IEEE Standard 802.15.6, we give a feasible unite authentication scheme, in which physical layer strategy and higher layer strategy are combined to enhance the security of authentication. To verify the effectiveness of the proposed physical layer strategy, we conducted experiments and numerical simulations. Furthermore, the computational complexity of the proposed algorithm and the efficiency of the cross-layer approach are analyzed, respectively. Results showed that the proposed approach can significantly enhance the authentication and the complexity is acceptable (polynomial complexity).","Title: Physical-layer security in Internet of Things based on compressed sensing and frequency selection.\nYear: 2017\n\nAbstract: Information security is a vital concern in Internet of Things (IoT). Traditional security method based on public or private key encryption scheme is limited by the trade-off between low cost and high level of security. Among different security solutions, utilising compressed sensing (CS) in combination with the physical-layer security to achieve the security is a remarkable method. However, in the...","Title: The Impact of Estimating High-Resolution Tropospheric Gradients on Multi-GNSS Precise Positioning.\nYear: 2017\n\nAbstract: Benefits from the modernized US Global Positioning System (GPS), the revitalized Russian GLObal NAvigation Satellite System (GLONASS), and the newly-developed Chinese BeiDou Navigation Satellite System (BDS) and European Galileo, multi-constellation Global Navigation Satellite System (GNSS) has emerged as a powerful tool not only in positioning, navigation, and timing (PNT), but also in remote sensing of the atmosphere and ionosphere. Both precise positioning and the derivation of atmospheric parameters can benefit from multi-GNSS observations. In this contribution, extensive evaluations are conducted with multi-GNSS datasets collected from 134 globally-distributed ground stations of the International GNSS Service (IGS) Multi-GNSS Experiment (MGEX) network in July 2016. The datasets are processed in six different constellation combinations, i.e., GPS-, GLONASS-, BDS-only, GPS + GLONASS, GPS + BDS, and GPS + GLONASS + BDS + Galileo precise point positioning (PPP). Tropospheric gradients are estimated with eight different temporal resolutions, from 1 h to 24 h, to investigate the impact of estimating high-resolution gradients on position estimates. The standard deviation (STD) is used as an indicator of positioning repeatability. The results show that estimating tropospheric gradients with high temporal resolution can achieve better positioning performance than the traditional strategy in which tropospheric gradients are estimated on a daily basis. Moreover, the impact of estimating tropospheric gradients with different temporal resolutions at various elevation cutoff angles (from 3 degrees to 20 degrees) is investigated. It can be observed that with increasing elevation cutoff angles, the improvement in positioning repeatability is decreased.","Title: The design and implement of China-Japan-South Korea logistics information platform based on information service exchange\nYear: 2016\n\nAbstract: With the rapid development of economy, trade and logistics among the three countries, the demand for establishment of the China-Japan-South Korea logistics information platform (CJK-LIP) is more and more urgent. This paper analyzes the requirements of the platform, proposes the overall design idea, and discusses critical technique problems in the platform development and implement. The proposed design method was successfully applied in the construction of CJK-LIP. Application effect of CJK-LIP show that CJP-LIP based on information service exchange will be a valuable model for international logistics information platform construction.","Title: The Optimal Checkpoint Interval for the Long-Running Application\nYear: 2017\n\nAbstract: For the distributed computing system, excessive or deficient checkpointing operations would result in severe performance degradation. To minimize the expected computation execution of the long-running application with a general failure distribution, the optimal equidistant checkpoint interval for fault tolerant performance optimization is analyzed and derived in this paper. More precisely, the optimal checkpointing period to determine the proper checkpoint sequence is proposed, and the derivation of the expected effective rate of the defined computation cycle is introduced. Corresponding to the maximal expected effective rate, the constraint of the optimal checkpoint sequence can be obtained. From the constraint of optimality, the optimal equidistant checkpoint interval can be obtained according to the minimal fault tolerant overhead ratio. By the numerical results, the proposal is practical to determine a proper equidistant checkpoint interval for fault tolerant performance optimization.","Title: Deformation of facial model for complete denture prosthesis using ARAP group method and elastic properties.\nYear: 2017\n\nAbstract: With the development of 3D printing and computer graphics technology, mouth rehabilitation has increasingly adopted digital methods. This research proposes a new method to transform the appearance of facial model after complete denture prosthesis. A feature template with few feature points is first constructed according to the facial muscle anatomy and facial deformation after complete denture prosthesis. Next, the traditional as-rigid-as-possible (ARAP) method is optimised by clustering based on facial muscles. The optimised ARAP method is then used for real-time and interactive simulations. Finally, by classifying the degrees of elasticity in the model with additional weights, the simulation can be customised to the skin of individual patients. Different degrees of elastic deformation and post-operative models are superimposed for match analysis. Compared with our previous study, the error is reduced by 24.05%. Results show that our method can deform facial models accurately and efficiently.","Title: A Multi-objective Attribute Reduction Method in Decision-Theoretic Rough Set Model.\nYear: 2017\n\nAbstract: Many attribute reduction methods have been proposed for decision-theoretic rough set model based on different definitions of attribute reduct, while an attribute reduct can be seen as an attribute subset that satisfies specific criteria. Most reducts are defined on the basis of a single criterion, which may result in the difficulty for users to choose appropriate reduct to design related reduction algorithm. To address this problem, we propose a multi-objective attribute reduction method based on NSGA-II for decision-theoretic rough set model. Three different definitions of attribute reduct based on positive region, decision cost and mutual information are considered and transferred to a multi-objective optimization problem. Experimental results show that the multi-objective reduction method can obtain a robust and better classification performance.","Title: A Wearable Autonomous Heart Rate Sensor Based On Piezoelectric-Charge-Gated Thin-Film Transistor For Continuous Multi-Point Monitoring\nYear: 2017\n\nAbstract: Autonomous wearable biomedical sensors enable continuous human body vital signs monitoring with features such as being conformable, mobile, cost-effective and self-powered. In this work, we report on an autonomous and multi-positional sensor capable of heart rate and blood pressure monitoring. The device concept is based on a piezoelectric-charge-gated thin-film transistor (PCGTFT) where a polyvinylidene fluoride (PVDF) piezoelectric sandwich structure is incorporated with an amorphous silicon (aSi:H) dual-gate TFT (DGTFT). An analytical model and preliminary experimental results will be presented along with a demonstration of a proof-of-concept sensing system for continuous multi-point heart rate monitoring.","Title: An empirical study of touch-based authentication methods on smartwatches.\nYear: 2017\n\nAbstract: The emergence of smartwatches poses new challenges to information security. Although there are mature touch-based authentication methods for smartphones, the effectiveness of using these methods on smartwatches is still unclear. We conducted a user study (n=16) to evaluate how authentication methods (PIN and Pattern), UIs (Square and Circular), and display sizes (38mm and 42mm) affect authentication accuracy, speed, and security. Circular UIs are tailored to smartwatches with fewer UI elements. Results show that 1) PIN is more accurate and secure than Pattern; 2) Pattern is much faster than PIN; 3) Square UIs are more secure but less accurate than Circular UIs; 4) display size does not affect accuracy or speed, but security; 5) Square PIN is the most secure method of all. The study also reveals a security concern that participants' favorite method is not the best in any of the measures. We finally discuss implications for future touch-based smartwatch authentication design.","Title: Partial order reduction for checking LTL formulae with the next-time operator.\nYear: 2017\n\nAbstract: Partial order reduction (POR) has been one of the most successful techniques for reducing the state spaces in model checking concurrent asynchronous systems with respect to linear-time temporal logic (LTL) formulae. Conventional POR only support checking formulae without the next-time operator because these formulae are invariant under stuttering, thus are suitable for POR. In this article we introduce a novel POR technique for checking formulae with the next-time operator. This new technique enables model checking algorithms to use POR even if the formulae are not stutter invariant (SI). More precisely, it supports the full class of LTL formulae with the next-time operator. The core of our approach is a heuristic algorithm for detecting and extending SI components of Buchi automata (BA) during the translation from LTL formulae into BAs. Then, the detected component is used to direct the reduction of the state space of a system. In addition, an SI-component of a BA enables an improvement of LTL model checking algorithms based on the nested depth-first search. We show that our technique is efficient when the synchronous product of an SI-component and a system is large.","Title: Quantile Induced Heavy Ordered Weighted Averaging Operators And Its Application In Incentive Decision Making\nYear: 2018\n\nAbstract: To integrate incentives into the information aggregation process in decision making, we propose a new type of aggregation operator, denominated as the quantile induced heavy ordered weighted averaging (QI-HOWA) operator in this paper. A primary characteristic of this type of operator is that the quantile variable can be used not only to measure the relative performance of alternatives but also to facilitate the incentive preference expression of the decision maker. We further provide a calculation technology of the QI-HOWA weights, in which various incentive preferences of the decision maker can be considered through parameter adjustment. In addition, we discuss certain properties of the QI-HOWA operator and note the extent to which they are effective. Finally, a numerical example regarding the selection of optimal alternatives by incentive measures is provided, and the aggregations are compared with those of ordered weighted averaging and unweighted averaging operators to illustrate the validity of the QI-HOWA operator.","Title: The Study Of The Mid-Term Screening Of Graduate Students Based On The Attribute Reduction Algorithm\nYear: 2018\n\nAbstract: In this paper, we use attribute reduction algorithm and attribute value reduction algorithm based on rough set theory to analyze the sample data of the mid-term screening of graduate students in a higher university. Through the analysis of the sample data, we get the key factors influencing the comprehensive evaluation of midterm screening of graduate students. Factors for this the best way to develop graduate students teaching management, to the midterm screening effective guidance later stage of learning, to raise the level of our graduate students teaching, management, training, in order to raise the high quality for the state and society of outstanding graduate students.","Title: Label Distribution Learning by Exploiting Sample Correlations Locally.\nYear: 2018\n\nAbstract: Label distribution learning (LDL) is a novel multi-label learning paradigm proposed in recent years for solving label ambiguity. Existing approaches typically exploit label correlations globally to improve the effectiveness of label distribution learning, by assuming that the label correlations are shared by all instances. However, different instances may share different label correlations, and few correlations are globally applicable in real-world applications. In this paper, we propose a new label distribution learning algorithm by exploiting sample correlations locally (LDL-SCL). To encode the influence of local samples, we design a local correlation vector for each instance based on the clustered local samples. Then we predict the label distribution for an unseen instance based on the original features and the local correlation vector simultaneously. Experimental results demonstrate that LDL-SCL can effectively deal with the label distribution problems and perform remarkably better than the state-of-the-art LDL methods.","Title: Label Distribution Learning by Exploiting Label Correlations.\nYear: 2018\n\nAbstract: Label distribution learning (LDL) is a newly arisen machine learning method that has been increasingly studied in recent years. In theory, LDL can be seen as a generalization of multi-label learning. Previous studies have shown that LDL is an effective approach to solve the label ambiguity problem. However, the dramatic increase in the number of possible label sets brings a challenge in performance to LDL. In this paper, we propose a novel label distribution learning algorithm to address the above issue. The key idea is to exploit correlations between different labels. We encode the label correlation into a distance to measure the similarity of any two labels. Moreover, we construct a distance-mapping function from the label set to the parameter matrix. Experimental results on eight real label distributed data sets demonstrate that the proposed algorithm performs remarkably better than both the state-of-the-art LDL methods and multi-label learning methods.","Title: The Consideration of Formal Errors in Spatiotemporal Filtering Using Principal Component Analysis for Regional GNSS Position Time Series.\nYear: 2018\n\nAbstract: In the daily operation of regional GNSS (Global Navigation Satellite System) networks, the formal errors of all stations' coordinate components are calculated. However, spatiotemporal filtering based on traditional Principal Component Analysis (PCA) for regional GNSS position time series does not take these formal errors into account. This paper developed a PCA-based approach to extract Common Mode Error (CME) from the position time series of a regional GNSS station network, where formal errors were applied to construct a weight factor. Because coordinate components with larger errors have smaller weight factors in extracting CME, the performance of our proposed approach was anticipated to be better than the traditional PCA approach. The position time series of 25 stations in the Yunnan Province, China, were analyzed using our approach, as well as the traditional PCA approach. The average errors of the residual time series after removing the CMEs with our approach were 1.30 mm, 1.72 mm, and 4.62 mm for North, East and Up components, and the reductions with respect to those of the original time series were 18.23%, 15.42%, and 18.25%, respectively. If CMEs were removed from the traditional PCA approach, the corresponding average errors were 1.34 mm, 1.81 mm, and 4.84 mm, with reductions of 15.84%, 10.86%, and 14.32%, respectively. Compared to the traditional PCA approach, the average errors of our approach were reduced by about 2.39%, 4.56%, and 3.93% in the North, East and Up components, respectively. Analysis of CME indicated that it mainly contained white and flicker noise. In the synthetic position time series with 500 repeated simulations, the CME extracted by our approach was closer to the true simulated values than those extracted by the traditional PCA approach, regardless of whether local effects were considered or not. Specifically, the mean root mean square (RMS) reduction of our approach, relative to PCA, ranged from 1.35% to 3.93%. Our simulations illustrated that the RMS reductions depended not only on the magnitude, but also the variation of the formal error series, which further highlights the necessity of considering formal errors in spatiotemporal filtering.","Title: Application Of Attribute Reduction Algorithm Based On Rough Set In Financial Management Of Commercial Banks\nYear: 2017\n\nAbstract: In this paper, we use the attribute reduction algorithm of rough set to analyze the sample data of the annual balance of the residents in commercial banks. Through the analysis of the sample data, we get different customer categories. For different customers, we develop different financial products. On the one hand, it meets the needs of customers; on the other hand, it increases the profitability of commercial banks, so as to realize the new idea of 'guided by data'.","Title: Bug Patterns Localization Based on Topic Model for Bugs in Program Loop\nYear: 2018\n\nAbstract: Statistic fault localization technique is a popular technique to perform fault localization via comparing the different statistics of failed executions and successful executions. They measure suspiciousness for each program component and rank them descending by their suspiciousness to assist programmers to identify root cause of failures. Unfortunately, since these techniques based on hit-spectrum information, which use binary flags to register whether a program component is covered or not during a program execution, their results are inherently limited, especially when the bug(s) occur in complex program loops. To resolve this issue, we propose a bug patterns localization technique based on topic model for the bugs in complex loops. We also conduct experiments involving 28 program versions contained bugs in program loops in Siemens program suite to compare the effectiveness of the proposed approach against Dstar. The results show that our approach is promising.","Title: A quantitative relationship between Application Performance Metrics and Quality of Experience for Over-The-Top video.\nYear: 2018\n\nAbstract: Quality of Experience (QoE) is a measure of the overall level of customer satisfaction with a vendor. In telecommunications, consumer satisfaction is of great interest in the adoption of novel multimedia products and services. A number of factors can greatly influence the customer experience during a video session. Factors such as user perception, experience, and expectations are expressed by QoE while factors such as application and network performance are expressed by Quality of Service (QoS) parameters. This paper studies the relationship between QoS and QoE in a session-based mobile video streaming. Specific QoS Application Performance Metrics (APMs) are examined based on a QoE assessment database which is built for experimentation and contains 108 subjects. It is shown that these APMs are highly related to two QoE factors, Technical Quality (TQ) and Acceptability. Furthermore, Viewing Ration (VR) parameter and the corresponding Kendall correlation between VR and QoE factors proves that VR is a valuable metric for mapping QoS to QoE. We further generated the compacted decision tree to predict QoE factors through Rebuffering Ratio (RR), Non-interruption Content Viewing Ratio (VRc), and Non-interruption Viewing Ratio (VRs). Through extensive experimentation, a general relationship between APMs and QoE factors has been examined and a primary QoE model is proposed based on this relationship.","Title: Mobile physical layer spoofing detection based on sparse representation.\nYear: 2018\n\nAbstract: Recently, physical layer authentication techniques are emerging to detect spoofing attacks in wireless networks based on high consistency between two legitimate successive channel information. However, this high consistency is not always available, especially in dynamic networks, such as mobile scenarios. In this study, the authors propose a physical layer spoofing detection scheme based on sparse...","Title: Design And Contact Analysis Of Gripper For Tube Inspection Robots\nYear: 2018\n\nAbstract: Automatic robots have been extensively employed to inspect steam generator (SG) tubes. Positioning and fixation of robots underneath tube-sheet plates are enabled by tube grippers. In this paper, configuration design and performance analysis of a new tube gripper were carried out. To calculate the gripper's load capacity, numerical analysis was used to establish finite element (FE) model of the gripper. Furthermore, orthogonal experiment and statistical methods combined with FE analysis were adopted to explore the influence on contact between the gripper and the tube's inner wall by manufacturing tolerance. Three factors of dimension and form tolerance were selected in sensitivity analysis. The influencing factors of contact stress were analysed by analysis of variance (ANOVA). Physical prototype experiment results were provided to validate the design rationality. Results of analysis and actual tests proved that the gripper with inelastic fingers exhibited excellent load capacity, superior contact performance and longer service life time.","Title: Physical layer spoofing detection based on sparse signal processing and fuzzy recognition.\nYear: 2017\n\nAbstract: Spoofing attacks is one of the most critical attacks in wireless communication security. Traditional solutions are based on cryptology which is performed in the upper layers, and face many challenges especially in resource-limited application. To overcome this hurdle, physical-layer security has been received a lot of attention recently. In this study, the authors propose a physical-layer spoofing...","Title: Transforming the SMV model into MMV model based on the characteristics of wavelet coefficients.\nYear: 2017\n\nAbstract: Sparse signal recovery or compressed sensing (CS) theory has recently received attention in the image compression field. CS has proven that the successful recovery rate of the multiple measurement vectors (MMVs) model is higher than that of the single measurement vector (SMV) case. Most existing algorithms have focused on sparse signal recovery using the MMV model, without considering converting t...","Title: Subspace Clustering Through Sub-Clusters\nYear: 2018\n\nAbstract: The problem of dimension reduction is of increasing importance in modern data analysis. In this paper, we consider modeling the collection of points in a high dimensional space as a union of low dimensional subspaces. In particular we propose a highly scalable sampling based algorithm that clusters the entire data via first spectral clustering of a small random sample followed by classifying or labeling the remaining out-of-sample points. The key idea is that this random subset borrows information across the entire dataset and that the problem of clustering points can be replaced with the more efficient problem of \"clustering sub-clusters\". We provide theoretical guarantees for our procedure. The numerical results indicate that for large datasets the proposed algorithm outperforms other state-of-the-art subspace clustering algorithms with respect to accuracy and speed.","Title: Multi-objective attribute reduction in three-way decision-theoretic rough set model.\nYear: 2019\n\nAbstract: Attribute reduction plays an important role in rough set theory. Many attribute reduction methods have been proposed based on different definitions of attribute reduct, while an attribute reduct can be regarded as a minimal attribute subset that satisfies specific criteria. Most reducts are defined on the basis of a single criterion, which can only affect one specific characteristic of the data or one preference of users. However, the single criterion-based attribute reduct may not meet the requirement of complex problems. To address this problem, based on three-way decision-theoretic rough set model, this paper defines a multi-objective attribute reduct. Three types of criteria for defining attribute reduct including the positive region, decision cost and mutual information are considered and combined to a multi-objective optimization problem. Based on the proposed multi-objective attribute reduct, we also introduce a multi-objective optimization-based attribute reduction method and an ensemble learning-based attribute reduction method. Experimental results on several datasets show that the proposed attribute reduction methods can obtain a robust and better classification performance.","Title: A fast singular boundary method for 3D Helmholtz equation.\nYear: 2019\n\nAbstract: This paper presents a fast singular boundary method (SBM) for three-dimensional (3D) Helmholtz equation. The SBM is a boundary-type meshless method which incorporates the advantages of the boundary element method (BEM) and the method of fundamental solutions (MFS). It is easy-to-program, and attractive to the problems with complex geometries. However, the SBM is usually limited to small-scale problems, because of the operation count of O(N3) with direct solvers or O(N2) with iterative solvers, as well as the memory requirement of O(N2). To overcome this drawback, this study makes the first attempt to employ the precorrected-FFT (PFFT) to accelerate the SBM matrix–vector multiplication at each iteration step of the GMRES for 3D Helmholtz equation. Consequently, the computational complexity can be reduced from O(N2) to O(NlogN) or O(N). Three numerical examples are successfully tested on a desktop computer. The results clearly demonstrate the accuracy and efficiency of the developed fast PFFT-SBM strategy.","Title: Behavioral Ordered Weighted Averaging Operator And The Application In Multiattribute Decision Making\nYear: 2019\n\nAbstract: This paper introduces a new type of behavioral ordered weighted averaging (BOWA) operator, to incorporate decision maker's gains and losses behavior tendency into the information aggregation process. The main characteristic of this BOWA operator is that it considers behavioral weights and ordered weights in the same formulation. We further provide a calculation method of the behavioral weights, in which various psychological preferences of different attribute types of the decision maker can be expressed intuitively. In addition, we discuss some particular cases of BOWA operator and its main properties. Finally, a numerical example is used to illustrate the use of the proposed method.","Title: Anomalous Path Detection for Spatial Crowdsourcing-Based Indoor Navigation System.\nYear: 2018\n\nAbstract: Indoor navigation system provides customized path planning for requesters who are unfamiliar with the indoor environment, such as shopping mall and airport. Spatial crowdsourcing technology can be applied to indoor navigation to offer fundamental services related to location. However, spatial crowdsourcing-based indoor navigation is vulnerable to the intrusion of injected anomalous paths from attackers. In this paper, we propose an anomalous path detection (APD) scheme to classify attackers according to their reputation management and abnormal trajectory sequence. Specifically, we first develop a crowdsourcing system to support the indoor location service using the fog as the spatial crowdsourcing server. Then, we identify two levels of attackers, i.e., the malicious responders and the semi-honest responders in the indoor environment according to their attacking purposes. Through the responders' historical records from the fog server, we analyze a series of trajectory sequences consisting of the distance between the current position and the destination to distinguish the semi-honest responders from the normal. In addition, we propose a semi-supervised learning with hidden Markov model (HMM) to detect the semi-honest responders. Finally, the extensive simulations show that the APD scheme can achieve higher accuracy with the acceptable false rate.","Title: Experimental study of human body effect on temperature sensor integrated RFID tag\nYear: 2017\n\nAbstract: An ultra-high frequency (UHF) passive radio frequency identification (RFID) tag consisting of flexible antenna and built-in temperature sensor chip is utilized for temperature monitoring of human body. A series of experiments were performed to study the body effects on temperature measurements. The tags were mounted at different positions of the body and in the different tag orientations as well. The experimental study of sensor integrated RFID tag shows that performance of the tag is more stable, yielding good signal strength and more consistent temperature measurement given the small form factor and form attachment. Minimizing the deteriorating effect of human body on tag's performance and the accuracy of measurements in different scenerios gives the perspective for implementation in realtime physiological monitoring as per the medical requirements.","Title: A Natural Language Processing Method Of Chinese Instruction For Multi-Legged Manipulating Robot\nYear: 2018\n\nAbstract: In this paper, a natural language processing (NLP) framework is proposed to build the Chinese natural language interaction between human and the multi-legged manipulating robot which has both locomotion and manipulation functions. The Chinese natural language instruction is transformed into formal representations by this method. Firstly, Cascaded Conditional Random Fields (CCRFs) is employed to analyze the syntax of the instruction. Letter-based features are used in each layer of CCRFs to solve problem caused by word segmentation. In order to understand the modification relationship between entity classes in chunk, one judgment method based on Support Vector Machine (SVM) is proposed. To solve the problem of the large number of core verbs in instructions generated by the multiple robot motion types, a classification of verbs based on Naive Bayes classifier was presented. And the semantic framework of each type of verbs is established to determine the necessary and unnecessary roles for each kind of verbs. At last, several experiments are carried out and the results of each step of framework are presented to demonstrate the effectiveness of the method. It is instructive to understand Chinese natural language instructions for robots with complex motion.","Title: A multiphase cost-sensitive learning method based on the multiclass three-way decision-theoretic rough set model.\nYear: 2019\n\nAbstract: The decision-theoretic rough set (DTRS) can be regarded as a type of cost-sensitive learning method that incorporates cost functions. Three-way decisions for an object are made in the DTRS model to obtain the minimum Bayesian decision cost, including acceptance, rejection and deferment decisions. However, because the cost matrix in traditional cost-sensitive learning problems does not include the cost for deferment decisions, DTRS cannot be directly used to solve traditional cost-sensitive learning problems. In this paper, we present a multiclass three-way decision-theoretic rough set model that can be directly applied to traditional cost-sensitive learning problems. This model can calculate the corresponding cost for deferment decisions based on the traditional cost matrix. In addition, we propose a multiphase cost-sensitive learning method based on the multiclass three-way decision-theoretic rough set model. The deferment and rejection samples are reduced and eliminated through multiphase iterative learning. Compared with the existing multiclass cost-sensitive learning methods, our proposed method can achieve higher classification accuracy and lower misclassification cost.","Title: Rank-Based Analysis Method To Determine Owa Weights And Its Application In Group Decision Making\nYear: 2019\n\nAbstract: To determine the weights of the ordered weighted averaging (OWA) operator, a new approach that is termed as rank-based analysis method with subjective information is developed. Available information called importance ratio that is obtained by subjective judgment of expert(s) is used in the OWA weights determination. The proposed approach is further extended to the situation of group decision making and is also used to obtain the weights of experts in this situation. In this case, the final OWA weights are obtained through a complete consistent group judgment matrix. Finally, a numerical example regarding the selection of the optimal R&D personnel is given to illustrate the application of the new method in group decision making.","Title: Induced Cluster-Based Owa Operators With Reliability Measures And The Application In Group Decision-Making\nYear: 2019\n\nAbstract: Considering the distributed structural characteristics of arguments to be aggregated, we propose a new type of aggregation operator, called induced cluster-based ordered weighted averaging (OWA; abbreviated as cluster-IOWA) operator, in this article. The main characteristic of the cluster-IOWA operator is that the arguments are aggregated by local clusters, and the order-inducing variable is used for representing a particular characteristic with respect to a local cluster. The cluster-OWA operator is commutativity, idempotence, and boundedness. We then discuss two important issues with respect to the cluster-IOWA operator. The order-inducing variables are determined by considering the overall reliability of the local cluster. Based on this, the position weighting vector of the local clusters is designed by taking into account both the reliability measures and the decision maker's preference. Finally, a numerical example, regarding the performance evaluation of middle managers carried out by a group of participants, is developed to illustrate the application and validity of the cluster-IOWA operator.","Title: Calibration of the Copolarized Backscattering Measurements From Gaofen-3 Synthetic Aperture Radar Wave Mode Imagery\nYear: 2019\n\nAbstract: Accurate absolute radiometric calibration of spaceborne synthetic aperture radar (SAR) sensor is of great importance in quantitative oceanic monitoring. Traditionally, the calibration constant is determined by analyzing measurements of man-made calibrators deployed on land at a high expense. In this paper, a technique called numerical weather prediction (NWP)-based ocean calibration method for estimating the calibration constant on the basis of SAR observations over ocean is applied to the first Chinese C-band SAR satellite Gaofen-3 (GF-3). The ocean calibration is performed on GF-3 wave mode SAR images at VV and HH polarizations over a period of one year from September 2017 to August 2018. Verification against the independent scatterometer-derived winds shows that after NWP-based ocean recalibration, the accuracy of wind speed retrieval from GF-3 wave mode imagery could reach the wind speed estimation performance using the state-of-the-art of SAR data. These results indicate that GF-3 wind retrievals are promising for operational oceanic products and applications if the SAR data are appropriately calibrated by the proposed NWP-based ocean calibration approach.","Title: Intelligent diagnosis of cardiac valve calcification in ESRD patients with peritoneal dialysis based on improved Takagi-Sugeno-Kang fuzzy system\nYear: 2019\n\nAbstract: AbstractWithout a B-ultrasound result, if a doctor diagnoses a suspected patient using only the basic clinical features, such as age, gender, serum calcium, and urea clearance index (KT/V), the diagnostic accuracy will be very low (even less than 50%). To solve this problem, a machine learning technology is proposed to intelligently diagnose cardiac valve calcification in end-stage renal disease (ESRD) patients with peritoneal dialysis. Compared with classical classification technologies, the proposed method aims to develop a model that has both medical interpretability and high recognition performance. In terms of interpretability, the Takagi-Sugeno-Kang fuzzy system is considered a basic model due to its built-in interpretable ability. In addition, because the distribution of the positive class (cardiac valve calcification is positive) and negative class (cardiac valve calcification is negative) in the peritoneal dialysis patient dataset is unbalanced, a novel unbalanced TSK (Takagi-Sugeno-Kang) fuzzy system (B-TSK-FS) is developed using a novel unbalanced fuzzy learning mechanism. The corresponding results reveal that the B-TSK-FS method obtains promising results (the max testing accuracy is over 98%) compared with classical machine learning methods for intelligently diagnosing cardiac valve calcification in ESRD patients with peritoneal dialysis.","Title: Competitive Behavior Induced Owa Operator And The Weighting Method\nYear: 2021\n\nAbstract: In view of the decision-making problem that the alternatives actively participate in the decision-making process and present competitive behavior, this paper proposes a new type of aggregation operator called the competitive behavior induced ordered weighted averaging (CB-IOWA) operator. The primary characteristic of the CB-IOWA operator is that the alternative is permitted to highlight its competitive advantage through the order-inducing variables and the weighting vector. A method is introduced to determine the final weighting vector based on the weights provided by different alternatives that can reflect their competitive advantages. The final weights are calculated by maximizing the closeness degree between the final weights and the ones provided the alternatives. The main properties of the CB-IOWA operator are discussed, such as commutativity, boundness, and monotonicity under certain condition. Finally, an application example regarding the competitive development of the economy-environment of four cities in China is provided to illustrate the usage of the CB-IOWA operator. The validity of the operator is checked by comparison with the weighted arithmetic averaging operator representing noncompetitive behavior.","Title: A System-Level Approach Towards A Hybrid Energy Harvesting Glove\nYear: 2021\n\nAbstract: This paper presents a novel wearable hybrid harvester system as a glove that contains four distinct scavenging modules of flexible transducer film, photosensitive 3D dual-gate thin-film transistor, and a particular power management box. Each single module is formed by a piezoelectric-charge-gated TFT (PCGTFT). The reported system is capable of scavenging energy from two various free of charge energy sources (Piezoelectric plus Photoelectric). Aforesaid system unlike other state-of-the-arts overcomes several key challenges in interfacing, storage and power management. Harvested energy which is administered through power and storage management system ultimately lightens a typical light emitting diode (LED), testifies capability of such glove to power up some low-power electronic devices.","Title: Reprogrammable 3d Liquid-Crystalline Actuators With Precisely Controllable Stepwise Actuation\nYear: 2021\n\nAbstract: Liquid-crystalline elastomers (LCEs) are considered ideal soft actuator materials for a wide range of applications, especially the thriving soft robotics. However, 3D LCE actuators capable of precisely controllable stepwise actuation, which can enhance functionality and versatility of LCE robots for multifarious complicated applications, are still in urgent need for the reported LCE actuators mainly exploit the one-step actuation upon the liquid-crystallin (LC)-isotropic phase transition temperature (T-i). Herein, a catalyst-free LC-vitrimer actuator with supercritical behavior is designed, which can perform precisely controllable stepwise actuation with extraordinary shape stability over a broad temperature range of about 70 degrees C. Moreover, supercritical behavior enables the actuator to be used in nematic phase, imparting the actuator with some extra advantages, such as higher mechanical strength and actuation stability, over the one used above T-i. Furthermore, the LCE can be reprogrammable into arbitrary 3D actuators, which can further be integrated into single-material actuators with complex stepwise actuation, offering a generalized strategy of LCE actuators for sophisticated practical soft robots.","Title: Semi-supervised label distribution learning via projection graph embedding\nYear: 2021\n\nAbstract: Label distribution learning (LDL) is a new machine learning paradigm that addresses label ambiguity by emphasizing the relevance of each label to a particular instance. In supervised learning, many LDL algorithms have been proposed, which often require a large amount of well-annotated training data to achieve good performance. However, annotating a label distribution is more complicated and expensive than annotating a single label or multiple labels with logical values of 0 and 1. Thus, we propose a projection graph embedding algorithm for semi-supervised label distribution learning (PGE-SLDL). Specifically, we seek a potential space by orthogonal neighborhood preserving projections, named capture space. This capture space is used to select more valuable features and construct a graph that contains more accurate data structure information. We utilize the sample correlation information contained between graph nodes to recover the unknown label distribution. In addition, compared with fixed graphs in traditional semi-supervised learning, we carry out projection and graph construction simultaneously to obtain a self-updating projection graph, which is more helpful to learn label distribution. The experimental results validate the effectiveness of the proposed algorithm.","Title: Integrated Modulation of Dual-Parallel NPC Inverters With Eliminated CMV\nYear: 2022\n\nAbstract: An integrated zero common-mode voltage (ZCMV) modulation algorithm is developed in this article for dual-parallel neutral-point-clamped (NPC) inverters treated as an integrated system. While the conventional methods achieve CMV elimination with the cost of worse harmonic performance, the proposed method guarantees a ZCMV and better current quality simultaneously by taking advantage of the extra sp...","Title: Unpaired Underwater Image Enhancement Based on CycleGAN\nYear: 2022\n\nAbstract: Underwater image enhancement recovers degraded underwater images to produce corresponding clear images. Image enhancement methods based on deep learning usually use paired data to train the model, while such paired data, e.g., the degraded images and the corresponding clear images, are difficult to capture simultaneously in the underwater environment. In addition, how to retain the detailed information well in the enhanced image is another critical problem. To solve such issues, we propose a novel unpaired underwater image enhancement method via a cycle generative adversarial network (UW-CycleGAN) to recover the degraded underwater images. Our proposed UW-CycleGAN model includes three main modules: (1) A content loss regularizer is adopted into the generator in CycleGAN, which constrains the detailed information existing in one degraded image to remain in the corresponding generated clear image; (2) A blur-promoting adversarial loss regularizer is introduced into the discriminator to reduce the blur and noise in the generated clear images; (3) We add the DenseNet block to the generator to retain more information of each feature map in the training stage. Finally, experimental results on two unpaired underwater image datasets produced satisfactory performance compared to the state-of-the-art image enhancement methods, which proves the effectiveness of the proposed model.","Title: Security Solutions for IoT-Enabled Applications Against the Disease Pandemic\nYear: 2021\n\nAbstract: Amid the COVID-19 pandemic, the Internet of Things (IoT) has rapidly taken on a powerful role to combat it with the emerging smart medicine, smart community, smart city, smart home, and so on. IoT devices are widely deployed in various places to provide feasible solutions for personnel temperature detection, medical treatment, and quarantine control during the pandemic. However, IoT-enabled applications inevitably face serious security challenges due to the massive amount of sensitive data. In this article, we first investigate several strategies of pandemic control based on IoT and the critical supporting technologies, 5G, artificial intelligence, and blockchain. Then we present the security issues in the process of data collection to usage in the applications. Through the analysis of specific paradigms, the Internet of Medical Things (IoMT) and a location check-in system, we exploit the potential security challenges, including data leakage and false sensing. Countermeasures are proposed to address these issues and are effectively validated by simulation results. Finally, some open issues instead of data security threats of IoT are outlined.","Title: Skeleton-Based Spatio-Temporal U-Network for 3D Human Pose Estimation in Video\nYear: 2022\n\nAbstract: Despite the great progress in 3D pose estimation from videos, there is still a lack of effective means to extract spatio-temporal features of different granularity from complex dynamic skeleton sequences. To tackle this problem, we propose a novel, skeleton-based spatio-temporal U-Net(STUNet) scheme to deal with spatio-temporal features in multiple scales for 3D human pose estimation in video. The proposed STUNet architecture consists of a cascade structure of semantic graph convolution layers and structural temporal dilated convolution layers, progressively extracting and fusing the spatio-temporal semantic features from fine-grained to coarse-grained. This U-shaped network achieves scale compression and feature squeezing by downscaling and upscaling, while abstracting multi-resolution spatio-temporal dependencies through skip connections. Experiments demonstrate that our model effectively captures comprehensive spatio-temporal features in multiple scales and achieves substantial improvements over mainstream methods on real-world datasets.","Title: Hyperspectral Classification Based on Coupling Multiscale Super-Pixels and Spatial Spectral Features\nYear: 2022\n\nAbstract: Hyperspectral remotely sensed images contain not only the spatial information of ground objects, but also their rich spectral information. Effectively improve the classification accuracy of hyperspectral images (HSIs), the purpose is to accurately grasp the current land resource utilization information, which is of great significance to the formulation and implementation of future land and space planning. Existing research on the classification of hyperspectral remotely sensed images uses a single-scale superpixel method for image segmentation. The optimal number of superpixels cannot be determined, image details may get missed, and a single kernel matrix cannot represent the information from multiple features, which results in reduced classification accuracy. Therefore, this study intends to use the superpixel segmentation method to perform multiscale superpixel segmentation on the principal components of HSIs at multiple scales, and to couple the multiscale superpixel spatial spectral kernel (SSK) with the original spectral kernel to form a synthetic kernel using weights for HSI classification. The three HSIs of Pavia University, Pavia Center, and Washington DC Mall are used as the experimental data to test and analyze this method. The experimental results show that the effective classification accuracies for the three datasets obtained by this method are 5.28%, 5.90%, and 7.71% higher than the five comparison methods, at best. The results prove that this method can effectively solve the problem of imprecise image feature extraction and an unknown number of initial superpixels and can significantly improve the classification accuracy of HSIs.","Title: Semantic-Structural Graph Convolutional Networks for Whole-Body Human Pose Estimation\nYear: 2022\n\nAbstract: Existing whole-body human pose estimation methods mostly segment the parts of the body's hands and feet for specific processing, which not only splits the overall semantics of the body, but also increases the amount of calculation and the complexity of the model. To address these drawbacks, we designed a novel semantic-structural graph convolutional network (SSGCN) for whole-body human pose estimation tasks, which leverages the whole-body graph structure to analyze the semantics of the whole-body keypoints through a graph convolutional network and improves the accuracy of pose estimation. Firstly, we introduced a novel heat-map-based keypoint embedding, which encodes the position information and feature information of the keypoints of the human body. Secondly, we propose a novel semantic-structural graph convolutional network consisting of several sets of cascaded structure-based graph layers and data-dependent whole-body non-local layers. Specifically, the proposed method extracts groups of keypoints and constructs a high-level abstract body graph to process the high-level semantic information of the whole-body keypoints. The experimental results showed that our method achieved very promising results on the challenging COCO whole-body dataset.","Title: Orientation and Channel-Independent RF Fingerprinting for 5G IEEE 802.11ad Devices\nYear: 2022\n\nAbstract: Physical-layer fingerprinting is a promising technique to identify Internet of Things (IoT) devices. In this article, we investigate a new radio-frequency (RF) fingerprinting based on the distinctive signal-to-noise-ratio (SNR) trace in the sector-level sweep (SLS) procedure of 5G IEEE 802.11ad devices. This SLS SNR trace-based fingerprinting can directly apply to off-the-shelf devices without any extra hardware requirements and be independent of the wireless channel and environment. To tackle the impact of orientation on the RF fingerprinting, we propose a novel fingerprinting framework, involving correlation analysis, surface fitting, curve pursuing, and binary classification, named the CSCB framework. Using this framework, the proposed SLS SNR trace-based fingerprinting can achieve device authentication at any orientation with one receiver under line-of-sight (LOS) or non-LOS (NLOS) scenarios. We conduct proof-of-concept experiments using off-the-shelf IEEE 802.11ad devices (Talon AD7200 and MG360 WiGig) to evaluate the performance of the proposed fingerprinting schemes. Experimental results show the effectiveness of the proposed fingerprinting schemes where the verification accuracy of the proposed scheme can reach 99% with only 200 training samples.","Title: Retrospective quality by design r(QbD) for lactose production using historical process data and design of experiments\nYear: 2022\n\nAbstract: Quality by Design (QbD) is a popular formal approach for designing, upscaling and optimizing industrial production facilities towards guaranteed quality. To avoid the many costly experiments required for QbD, historical production data may be exploited for optimization instead in what is known as a retrospective QbD (rQbD) study. Current rQbD literature does limitedly discuss data-driven identification of Critical Process Parameters (CPPs) to optimize limited process knowledge availability, and does not cover situations where technical operation limits have not yet been fully explored and/or where parallel equipment (lines) are used. This work presents a new rQbD strategy that addresses these challenges by balancing knowledge that can be obtained from statistical analysis of historical data, together with process experts with a carefully designed set of plant-scale experiments within current operational limits. This novel strategy is demonstrated on a long-running industrial lactose production facility. By digitally and experimentally exploring historical operation variability, we found new operational regimes for this production that may lead to up to 7% product quality improvement, reduced energy consumption and increased process under-standing. Although optimizing a specific process by necessity requires a process-specific approach, the way in which we systematically optimize the current process with Hybrid AI (combining available knowledge with new insights from historical data) shows that approaches that are currently used in prospective process upscaling may be modified to be invaluable for optimization of full-scale processes with a long operational history. (C) 2022 The Authors. Published by Elsevier B.V. CC_BY_4.0","Title: Interpretable Personalized Experimentation\nYear: 2022\n\nAbstract: Black-box heterogeneous treatment effect (HTE) models are increasingly being used to create personalized policies that assign individuals to their optimal treatments. However, they are difficult to understand, and can be burdensome to maintain in a production environment. In this paper, we present a scalable, interpretable personalized experimentation system, implemented and deployed in production at Meta. The system works in a multiple treatment, multiple outcome setting typical at Meta to: (1) learn explanations for black-box HTE models; (2) generate interpretable personalized policies. We evaluate the methods used in the system on publicly available data and Meta use cases, and discuss lessons learnt during the development of the system.","Title: Label enhancement with label-specific feature learning\nYear: 2022\n\nAbstract: Label distribution learning (LDL) is a novel machine learning paradigm. It addresses the problem of label ambiguity by emphasizing the relevance of each label to a particular instance. Unlike the simple logic vectors in single label learning (SLL) and multi-label learning (MLL), LDL assigns descriptive labels to each instance. Since it is often difficult to collect training sets with precise labels, label enhancement (LE) can be used to recover label distributions from logical labels. Most existing LE models construct label distributions based on a mapping from the feature space to the label space and include two main assumptions: (1) the relationship between features and labels is linear, and (2) all features are shared by all labels. However, in reality, the relationship between features and labels is not purely linear, and different labels may be determined by different features. To solve this problem, in this paper, we propose a new algorithm, that first maps features to a high-dimensional space to explore the nonlinear mapping relationship between features and labels, and then takes full advantage of the labels’ properties determined by specific features, while considering the correlation between labels. The experimental comparison with the existing algorithms verifies the effectiveness of the algorithm proposed in this paper.","Title: Label distribution learning with noisy labels via three-way decisions\nYear: 2022\n\nAbstract: Label distribution learning (LDL) as a soft-labeling paradigm is allowed to learn single or multi-labeled information distribution. Overwhelmingly, in the open world, the distribution of labels is usually disturbed by the noise (such as the man-made induction bias, shake of hardware devices), which in turn affects the decision of downstream tasks. To address this problem, we propose a novel LDL approach by using the three-way decisions theory to clear the amplified noise in this paper. First, we evaluate the confidence of each training sample and use a three-way decisions-based method to identify the trustworthy samples and the noisy samples. Second, we apply the sample correlation between the trustworthy samples and the noisy samples to correct the noisy labels. Finally, we re-weight every sample based on the learned confidences to train the robust LDL model. Experiments show that our approach has better performance in handling noisy data compared to existing algorithms."],"string":"[\n \"Title: A complete development process of finite element software for body-in-white structure with semi-rigid beams in .NET framework\\nYear: 2012\\n\\nAbstract: At the conceptual design stage, simplified finite element (FE) model of body-in-white (BIW) structure focuses on its specific merit to provide early-stage predictions for detailed FE model of that. This paper exploits a semi-rigid beam element (SRBE) that consists of a beam element with two semi-rigid connections at the ends to simulate the flexibility of joint. Guyan reduction method condenses the SRBE as a super element. A special finite element software for structural modeling and analysis of BIW (SMAB) is developed in .NET framework. The Unified Modeling Language is employed to depict the classes and their relationship. The design patterns are identified and applied in the framework design to facilitate communication and system expansion. Microsoft DirectX and GDI+ implement graphics display of spatial BIW frame and planar thin-walled cross section. Based on multi-threaded technology in .NET, subspace iteration method is paralleled to speed up the mode analysis. As a result, the efficiency of the SRBE is demonstrated by a benchmarking automotive body. Multi-threaded parallel is effective and useful, especially for frequency optimization.\",\n \"Title: From task parameters to motor synergies: A hierarchical framework for approximately-optimal control of redundant manipulators.\\nYear: 2005\\n\\nAbstract: We present a hierarchical framework for approximately-optimal control of redundant manipulators. The plant is augmented with a low-level feedback controller, designed to yield input-output behavior that captures the task-relevant aspects of plant dynamics but has reduced dimensionality. This makes it possible to reformulate the optimal control problem in terms of the augmented dynamics, and optimize a high-level feedback controller without running into the curse of dimensionality. The resulting control hierarchy compares favorably to existing methods in robotics. Furthermore we demonstrate a number of similarities to (non-hierarchical) optimal feedback control. Besides its engineering applications, the new framework addresses a key unresolved problem in the neural control of movement. It has long been hypothesized that coordination involves selective control of task parameters via muscle synergies, but the link between these parameters and the synergies capable of controlling them has remained elusive. Our framework provides this missing link.\",\n \"Title: Estimation and Control of Systems with Multiplicative Noise via Linear Matrix Inequalities\\nYear: 2005\\n\\nAbstract: This paper provides estimation and control de- sign methods for linear discrete-time systems with multi- plicative noise. First, we present the design of the state feedback controller, such that the closed loop system is mean square stable. Second, we provide sufficient conditions for the existence of the state estimator; these conditions are expressed in terms of linear matrix inequalities (LMIs), and the parametrization of all admissible solutions is addressed. Finally, an estimator design approach is formulated using LMIs, and the performance of the estimator is examined by means of numerical examples.\",\n \"Title: Iterative Linear Quadratic Regulator Design for Nonlinear Biological Movement Systems\\nYear: 2004\\n\\nAbstract: This paper presents an Iterative Linear Quadratic Regulator (ILQR) method for locally-optimal feedback con- trol of nonlinear dynamical systems. The method is applied to a musculo-skeletal arm model with 10 state dimensions and 6 controls, and is used to compute energy-optimal reaching movements. Numerical compar- isons with three existing methods demonstrate that the new method converges substantially faster and finds slightly better solutions.\",\n \"Title: Synthetic Fault Diagnosis Method of Power Transformer Based on Rough Set Theory and Improved Artificial Immune Network Classification Algorithm\\nYear: 2008\\n\\nAbstract: According to complementary strategy, this paper presents a new power transformer fault diagnosis method based on rough sets theory (RST)and improved artificial immune network classification algorithm. Through reduction approach of RST information table to simplify expert knowledge and reduce fault symptoms, the minimal diagnostic rules can be obtained. An improved artificial immune network classification algorithm is proposed on the base of them. At first the artificial immune network which both antigens and memory antibodies with class information has been added into are trained to learn the features of fault samples. In this way, the memory antibody cells pool which can represent the fault samples better than those without class information can be obtained. Then the k-nearest neighbor method is used to classify the fault samples. Compared with the IEC three-ratio method and BP neural network (BPNN), the proposed algorithm has better capability to classify single-fault and multiple-fault samples as well as higher diagnosis precision.\",\n \"Title: An approach for testing safety-critical software\\nYear: 1999\\n\\nAbstract: A novel approach for testing the effectiveness, efficiency, safety and relative appropriateness of Computer Interlocking Software (CIS)-a kind of safety-critical software is presented with a software platform developed to support this approach. A brief description of the proposed approach is also included\",\n \"Title: An optimization viewpoint of decision-theoretic rough set model\\nYear: 2011\\n\\nAbstract: This paper considers an optimization viewpoint of decisiontheoretic rough set model. An optimization problem is proposed by considering the minimization of the decision cost. Based on the optimization problem, cost functions and thresholds used in decision-theoretic rough set model can be learned from the given data automatically. An adaptive learning algorithm Alcofa is proposed. Another significant inference drawn from the solution of the optimization problem is a minimum cost based attribute reduction. The attribute reduction can be interpreted as finding the minimal attribute set to make the decision cost minimum. The optimization viewpoint can bring some new insights into the research on decision-theoretic rough set model.\",\n \"Title: Fault Diagnosis of Power Transformers Based on BP Network with Clonal Selection Algorithm\\nYear: 2007\\n\\nAbstract: In this paper, a novel approach based on BP network (BPN) for fault diagnosis of power transformers is proposed. Optimization of BPN weights is achieved by using clonal selection algorithms (CSA). In addition, the mutation probability is adjusted adaptively according to the affinity of antibody. Compared with previous approaches, this one can avoid prematurity effectively, with good self-learning and self-memory ability. The experiment results show that the presented approach outperforms previous ones in both classification accuracy and computational efficiency.\",\n \"Title: Checkpoint scheduling model for optimality\\nYear: 2011\\n\\nAbstract: To minimize the expected execution time, a general checkpoint scheduling algorithm is proposed to determine the near optimal checkpointing time sequence. More precisely, based on a simple timing policy, an execution analytical model is introduced and the expected effective ratio is derived. By maximizing the expected effective ratio, the optimal checkpoint period for the exponential failure distribution can be obtained directly, and a general checkpoint scheduling algorithm is developed to perform the near optimal checkpointing time sequence for an arbitrary failure distribution. Experimental results reveal that the proposal can perform varying checkpoint interval according to the failure distribution and the expected effective ratio of the execution is considerable for the long-running application in term of reliability.\",\n \"Title: Grid Synchronization Systems Of Three-Phase Grid-Connected Power Converters: A Complex-Vector-Filter Perspective\\nYear: 2014\\n\\nAbstract: Due to the significance of extracting the grid voltage information, the grid synchronization system plays an important role in the control of grid-connected power converters, and various grid voltage synchronization schemes have been proposed. This paper adopts the complex-vector-filter method (CVFM) to analyze the grid synchronization systems. With this method, the pairs of scalar signals, for example, the alpha- and beta-axis components in the stationary alpha - beta frame, are combined into one complex vector. As a consequence, the grid synchronization systems can be described with the complex transfer functions, which is very convenient to evaluate the steady-state performance, for example, the fundamental and harmonic sequence decoupling/cancellation, and dynamic performance of these systems. Moreover, the CVFM also provides a more generalized perspective to understand and develop the grid synchronization systems. Therefore, some of the representative systems are reanalyzed with the CVFM in this paper. A generalized second-order complex-vector filter and a third-order complex-vector filter are proposed with the CVFM to achieve better dynamic performance or higher harmonic attenuation. Moreover, a brief comparison of the complex-vector filters analyzed in this paper is presented. The effectiveness of the CVFM and the proposed two complex-vector filters are verified by the simulation and experimental results.\",\n \"Title: Linear quadratic optimal control of contact transition with fingertip\\nYear: 2009\\n\\nAbstract: This paper proposes an optimal control methodology that addresses the problem of control of fingertip during a general class of task that requires the fingertip to make a transition from non-contact motion to contact motion. Specifically, the task that the fingertip makes and transitions from motion to static well-directed force production. Here we present a mathematical framework for controlling the contact transition, while switching between non-contact and contact controller is needed and handled by the optimal control strategy. The non-linear differential algebraic equation that describes the dynamics of the index finger is linearized, and then a modified linear quadratic optimal control problem is solved. The resulting optimal feedback control law guarantees good regulation of contact force, velocity and position. Simulation results are presented to demonstrate the effectiveness of the new approach.\",\n \"Title: Network criticality in vehicular networks\\nYear: 2012\\n\\nAbstract: Network criticality (resistance distance) is a graph-theoretic metric that quantifies network robustness, and that was originally designed to capture the effect of environmental changes in core communication networks. This paper establishes a relationship between information centrality and network criticality and provides a justification for using the average network criticality of a node to quantify the nodes relative importance in a graph.This results provides a basis for designing robust clustering algorithms for vehicular networks.\",\n \"Title: An Iterative Optimal Control and Estimation Design for Nonlinear Stochastic System\\nYear: 2006\\n\\nAbstract: This paper presents an iterative linear-quadratic-Gaussian method for locally-optimal control and estimation of nonlinear stochastic systems. The new method constructs an affine feedback control law obtained by minimizing a novel quadratic approximation to the optimal cost-to-go function. It also constructs a non-adaptive filter optimized with respect to the current control law. The control law and filter are iteratively improved until convergence. The performance of the algorithm is illustrated on a complex biomechanical control problem involving a stochastic model of the human arm\",\n \"Title: A Method for Improving Concurrent Write Performance by Dynamic Mapping Virtual Storage System Combined with Cache Management\\nYear: 2011\\n\\nAbstract: The paper presents a new dynamic mapping virtual storage system, which is characterized by: (a) distribute storage resources according to need, thus improving storage resource utilization, (b) more fundamentally, through write-anywhere pattern to sequentialize concurrent write requests by dynamic address mapping mechanism, thus improving concurrent write performance for applications. Meanwhile, taking into account imperfection small size write requests handling and the negative impact on read performance, we combine the system with CBD system which is responsible for optimizing cache management. We have implemented our virtual storage system in Linux kernel 2.6.18 as a pseudo device driver. The experiment results show that concurrent write bandwidth approaches raw disk write performance. And in 64-stream concurrent write case, E-ASD can outperform concurrent write bandwidth of LVM by up to 160%, and its max single-stream read loss is less than 25% compared to LVM, with concur-rent read performance close to each other. Read performance still has space to be improved given larger cache and proper prefetching.\",\n \"Title: Barriers to mobile commerce adoption: an analysis framework for a country-level perspective\\nYear: 2008\\n\\nAbstract: This research examines and categorises the country-level adoption barriers of mobile commerce services, and tests those barriers in a case study of one country. The outcomes of the research presented in this paper are a broad framework of mobile commerce barriers that might be applicable when undertaking a country-level analysis, and a case study which tests the value of this framework in helping to uncover the specific barriers of mobile commerce adoption in one example country, New Zealand.\",\n \"Title: Full-Feedforward Schemes Of Grid Voltages For A Three-Phase Lcl-Type Grid-Connected Inverter\\nYear: 2013\\n\\nAbstract: This paper investigates the feedforward schemes of grid voltages for a three-phase LCL-type grid-connected inverter. The full-feedforward functions of grid voltages are derived for the stationary alpha - beta frame, synchronous d - q frame, and decoupled synchronous d - q frame-controlled three-phase LCL-type grid-connected inverters. The derived full-feedforward functions mainly consist of three parts which are proportional, derivative, and second derivative parts. The use of the traditional proportional feedforward function in the three-phase LCL-type grid-connected inverter will result in the amplification of the high-frequency injected grid current harmonics. With the proposed full-feedforward schemes, the injected grid current harmonics and unbalance caused by grid voltages can be greatly reduced. The effectiveness of the proposed feedforward schemes is verified by the experimental results.\",\n \"Title: Robust clustering for connected vehicles using local network criticality\\nYear: 2012\\n\\nAbstract: This paper proposes a robust Criticality-based Clustering Algorithm (CCA) for Vehicular Ad Hoc NETworks (VANETs) based on the concept of network criticality. Network criticality is a global metric on an undirected graph, that quantifies the robustness of the graph against environmental changes such as topology. In this paper, we localize the notion of network criticality and apply it to control cluster formation in the vehicular wireless network. We use the localized notion of node criticality together with a universal link measure, Link Expiration Time (LET), to derive a distributed multi-hop clustering algorithm for VANETs. Simulation results show that the proposed CCA forms robust cluster structures.\",\n \"Title: Discovering structural regularity in facade image\\nYear: 2010\\n\\nAbstract: 3D reconstruction from single view is a very difficult problem in computer vision and computer graphics. Few works have been done in this area. In this paper, we proposed a method to discover structural regularity in single facade images on the procedural modeling of buildings by using SIFT algorithm. Firstly, symmetry information is detected from the image using SIFT algorithm for image analysis. Next, the symmetry information will be removed from the facade image. Finally, get the SIFT feature of widows and doors. Experiments show that our method can successfully discover regular structures with a higher recognition rate than existing method especially amidst shadowing and the changing illumination environment.\",\n \"Title: Automatic segmentation of focused objects from images with low depth of field\\nYear: 2010\\n\\nAbstract: In this paper, we propose an automatic segmentation approach to extract focused objects from images with low depth of field (DOF). A focus energy map is first estimated based on the difference of high-frequency components between focused region and defocused background, and is exploited to construct region/boundary saliency maps on the basis of a pre-segmentation result by watershed transform. Then region/boundary masks for focused object are generated by entropy thresholding and flood filling, and an efficient boundary linking method is proposed to obtain closed region/boundary masks, which are exploited to reasonably generate a trimap containing seed regions for focused object and defocused background, and uncertain regions, respectively. Finally, the trimap is used as the input to an image matting model, which is utilized to classify the pixels in the uncertain regions to obtain an accurate focused object segmentation result based on the estimated alpha matte. Experimental results for a variety of low DOF images demonstrate the good segmentation performance of the proposed approach.\",\n \"Title: Robust discovery of partial rigid symmetries on 3D models\\nYear: 2010\\n\\nAbstract: The ubiquity of symmetry in nature and man-made artifacts has made symmetry discovery an important tool for numerous applications. The most fundamental and visually prominent kind of symmetry is partial rigid symmetry, which is explained as the invariance between parts of a 3D model under a set of translation, rotation, reflection, and uniform scaling generators. Thus automatic discovery of partial rigid symmetry on general 3D models, with no assumption on the size, shape or location of the symmetric parts, keeps to be a hot topic during recent years. Among such kind of works, the transformation voting technique [Mitra et al. 2006] is most widely used, due to its high efficiency and easiness for understanding and implementation.\",\n \"Title: Salient Object Extraction Based on Region Saliency Ratio\\nYear: 2009\\n\\nAbstract: This paper proposes an efficient approach to extract salient objects in an image. A scale-invariant saliency map is first constructed based on a multi-resolution feature contrast calculation, meanwhile the image is segmented into homogenous regions using nonparametric kernel density estimation (NKDE). Then the region saliency ratio of each region combination to its complement is calculated in turn. Finally, salient objects are extracted by maximizing the region saliency ratio. Experimental results demonstrate the effectiveness of the proposed approach.\",\n \"Title: The e-derivative of Boolean functions and its application in the fault detection and cryptographic system.\\nYear: 2011\\n\\nAbstract: Purpose - The purpose of this paper is to show that the e-derivative is of great significance when revealing cryptographic characteristics of Boolean functions and assurance of the safety of the cryptographic system. Design/methodology/approach - A new definition is introduced in this paper. Findings - After introducing the e-derivative, the authors obtain some relevant cryptographic properties and theorems which can fully reveal the relationships between characteristics and structures of Boolean functions. In logic circuit fields, the authors conduct the test to all the signals with numerical value 1 in the transmission line using the e-derivative, which improves the ability of the test. In addition, when there is no expression of the transmission line the authors will obtain it through the vector, and then do troubleshooting. Research limitations/implications - Logic circuit is the main limitation in the fault detection. Practical implications - A very useful method for assurance of the safety of cryptographic systems. Originality/value - This paper is the first time, in cryptology fields, that the e-derivative of Boolean functions as a new definition is introduced to fully reveal the relationships between characteristics and structures of Boolean functions. In addition, they can also be applied to the locating test of the breakdown, which improves the ability of the test.\",\n \"Title: Optimized Controller Design for -Type Grid-Connected Inverter to Achieve High Robustness Against Grid-Impedance Variation\\nYear: 2015\\n\\nAbstract: Capacitor-current-feedback active damping is an effective method to suppress the LCL-filter resonance in grid-connected inverters. However, due to the variation of grid impedance, the LCL-filter resonance frequency will vary in a wide range, which challenges the design of the capacitor-current-feedback coefficient. Moreover, if the resonance frequency is equal to one-sixth of the sampling frequency (fs/6), the digitally controlled LCL-type grid-connected inverter can be hardly stable no matter how much the capacitor-current-feedback coefficient is. In this paper, the optimal design of the capacitor-current-feedback coefficient is presented to deal with the wide-range variation of grid impedance. First, the gain margin requirements for system stability are derived under various resonance frequencies. By evaluating the effect of grid impedance on gain margins, an optimal capacitor-current-feedback coefficient is obtained. With this feedback coefficient, stable operations will be retained for all resonance frequencies except (fs/6). Second, in order to improve system stability for a resonance frequency of (fs/6), a phase-lag compensation for the loop gain is proposed. Finally, a 6-kW prototype is tested to verify the proposed design procedure.\",\n \"Title: Service Outsourcing Character Oriented Privacy Conflict Detection Method in Cloud Computing.\\nYear: 2014\\n\\nAbstract: Cloud computing has provided services for users as a software paradigm. However, it is difficult to ensure privacy information security because of its opening, virtualization, and service outsourcing features. Therefore how to protect user privacy information has become a research focus. In this paper, firstly, we model service privacy policy and user privacy preference with description logic. Secondly, we use the pellet reasonor to verify the consistency and satisfiability, so as to detect the privacy conflict between services and user. Thirdly, we present the algorithm of detecting privacy conflict in the process of cloud service composition and prove the correctness and feasibility of this method by case study and experiment analysis. Our method can reduce the risk of user sensitive privacy information being illegally used and propagated by outsourcing services. In the meantime, the method avoids the exception in the process of service composition by the privacy conflict, and improves the trust degree of cloud service providers.\",\n \"Title: Video quality of experience in the presence of accessibility and retainability failures\\nYear: 2014\\n\\nAbstract: Accurate Quality of Experience measurement for streaming video has become more crucial with the increase in demand for online video viewing. Quantifying video Quality of Experience is a challenging task. Significant efforts to quantify video Quality of Experience have primarily focused on the measurement of Quality of Experience for videos with network and compression related impairments. These impairments, however, may not always be the only main factors affecting Quality of Experience in an entire video viewing session. In this paper, we evaluate Quality of Experience for entire video viewing sessions, from the beginning to the end. In doing so, we evaluate videos with temporary interruptions as well as those with permanent interruptions or failures. We consider two types of failures, namely Accessibility and Retainability failures, and present the results of two subjective studies. These results indicate: (a) Accessibility and Retainability failures are rated lower compared to temporary interruption impairments; (b) Accessibility failures are rated close to the lowest value on the rating scale; and (c) the traditionally used 5-point scale to measure video Quality of Experience is not sufficient in the presence of Accessibility and Retainability failures.\",\n \"Title: Impact of Retainability Failures on Video Quality of Experience\\nYear: 2014\\n\\nAbstract: Measurement of video Quality of Experience is needed to enable telecommunication service operators to provide acceptable video viewing experience to their customers. Without accurate Quality of Experience measurement, it is hard to predict customer satisfaction. Accessibility and Retain ability failures are important factors that can affect the experience of viewing video on the Internet. Most of the published work in the assessment of Video Quality of Experience has focused on the quality of video viewing sessions without considering the impact of Accessibility and Retain ability failures. In this paper, we focus on the Quality of Experience for video sessions in the presence of Retain ability failures. We primarily evaluate video Quality of Experience for videos with temporary interruptions, called Integrity impairments, as well as with permanent interruptions called Retain ability failures. A subjective study was conducted to observe the effect of Retain ability failure on video Quality of Experience. The results of the subjective study suggest that videos with Retain ability failures are rated much lower than videos without any failure. An initial linear model for QoE of videos subject to Retain ability failures is introduced.\",\n \"Title: Preservation of P-Positions between a Game and Its Restricted Game\\nYear: 2014\\n\\nAbstract: E. Duchene et al investigate games whose sets of allowed moves are subsets of Wythoff's one, and whose set of P-positions is exactly Wythoff's sequence. Their results show that such a game does not exist. In this paper, we find a pair of games, one is the restricted version of the other, such that they have the same set of P-positions. We also find another pair of games, one is the restricted version of the other, such that they do not have the same set of P-positions. This means that the preservation of P-positions between a game and its restricted game depends on the game itself.\",\n \"Title: Compressed sensing-based unequal error protection by linear codes\\nYear: 2014\\n\\nAbstract: In many wireless communication systems, data can be divided into different importance levels. For these systems, unequal error protection (UEP) techniques are used to ensure lower bit error rate for the more important classes. Moreover, if the precise characteristics of the channel are known, UEP can be used to correctly recover the more important classes even under severe receiving conditions. In this study, a UEP scheme based on compressed sensing via a linear program is proposed. Discrete wavelet transform (DWT) is chosen as the sparsifying basis, and then DWT-coded information is divided into two-layered coded streams, each of which is transmitted differentially by applying an unequal number of information bits in linear codes according to the time-varying characteristic of the corrupted channel. In this proposed transmission scheme, the more important information is to guarantee error-free transmission. At the decoder, one can simply reconstruct the signal via the l1-minimisation algorithm. Simulation results show that the proposed scheme can achieve a higher peak signal-to-noise ratio (PSNR) and obviously improve the error resilience compared to the equal error protection scheme and other UEP methods. More importantly, with the increase of channel corrupted ratio, the drop rate of PSNR is much slower than other solutions. It indicates that the proposed method has better robustness for severe channel conditions.\",\n \"Title: A generalized iterative LQG method for locally-optimal feedback control of constrained nonlinear stochastic systems\\nYear: 2005\\n\\nAbstract: We present an iterative Linear-Quadratic- Gaussian method for locally-optimal feedback control of nonlinear stochastic systems subject to control constraints. Previously, similar methods have been restricted to determin- istic unconstrained problems with quadratic costs. The new method constructs an afÞne feedback control law, obtained by minimizing a novel quadratic approximation to the optimal cost-to-go function. Global convergence is guaranteed through a Levenberg-Marquardt method; convergence in the vicinity of a local minimum is quadratic. Performance is illustrated on a limited-torque inverted pendulum problem, as well as a complex biomechanical control problem involving a stochastic model of the human arm, with 10 state dimensions and 6 muscle actuators. A Matlab implementation of the new algorithm is availabe at www.cogsci.ucsd.edu/~todorov.\",\n \"Title: Three-way decisions based software defect prediction\\nYear: 2016\\n\\nAbstract: Based on a two-stage classification method and a two-stage ranking method on three-way decisions, this paper introduces a three-way decisions framework for cost-sensitive software defect prediction. For the classification problem in software defect prediction, traditional two-way decisions methods usually generate a higher classification error and more decision cost. Here, a two-stage classification method that integrates three-way decisions and ensemble learning to predict software defect is proposed. Experimental results on NASA data sets show that our method can obtain a higher accuracy and a lower decision cost. For the ranking problem in software defect prediction, a two-stage ranking method is introduced. In the first stage, all software modules are classified into three different regions based on three-way decisions. A dominance relation rough set based ranking algorithm is next applied to rank the modules in each region. Comparison experiments with 6 other ranking methods present that our proposed method can obtain a better result on FPA measure.\",\n \"Title: Iterative Linearization Methods For Approximately Optimal Control And Estimation Of Non-Linear Stochastic System\\nYear: 2007\\n\\nAbstract: This paper presents an iterative Linear-Quadratic-Gaussian method for locally-optimal control and estimation of non-linear stochastic systems. The new method constructs an affine feedback control law, obtained by minimizing a novel quadratic approximation to the optimal cost-to-go function, and a non-adaptive estimator optimized with respect to the current control law. The control law and filter are iteratively improved until convergence. The performance of the algorithm is illustrated on a complex biomechanical control problem involving a stochastic model of the human arm.\",\n \"Title: Prediction of facial deformation after complete denture prosthesis using BP neural network\\nYear: 2015\\n\\nAbstract: With the accelerated aging of world population, complete denture prosthesis plays an increasingly important role in mouth rehabilitation. In addition to recovering stomatognathic system function, restoring the appearance of a third of the area under the face has become a great challenge in complete denture prosthesis. This study analyzes the interactive relationship between the appearance of a third of the area under the face and complete denture, and proposes a new method to predict facial deformation after complete denture prosthesis. Firstly, to improve computational efficiency, the feature template is constructed to replace the deformed facial region. Secondly, a forecast model of elastic deformation is constructed using BP neural network and predicts elastic deformation amount because of the inhomogeneous, anisotropic and nonlinear material properties of soft tissue. Finally, a new feature template is calculated using deformation amount, and the deformation of preoperative model is simulated using Laplacian deformation technique. The average error rates of different hidden layer nodes in the neural network are analysed. Deformation and postoperative models are superimposed for match analysis. Experimental results show that this method can predict facial soft tissue deformation quickly and accurately. BP neural network is used to predict feature positions on facial model.The deformation of facial model is simulated using Laplacian deformation technique.This method can predict facial deformation quickly and accurately.CT scan is not necessary.\",\n \"Title: Neighborhood based decision-theoretic rough set models\\nYear: 2016\\n\\nAbstract: As an extension of Pawlak rough set model, decision-theoretic rough set model (DTRS) adopts the Bayesian decision theory to compute the required thresholds in probabilistic rough set models. It gives a new semantic interpretation of the positive, boundary and negative regions by using three-way decisions. DTRS has been widely discussed and applied in data mining and decision making. However, one limitation of DTRS is its lack of ability to deal with numerical data directly. In order to overcome this disadvantage and extend the theory of DTRS, this paper proposes a neighborhood based decision-theoretic rough set model (NDTRS) under the framework of DTRS. Basic concepts of NDTRS are introduced. A positive region related attribute reduct and a minimum cost attribute reduct in the proposed model are defined and analyzed. Experimental results show that our methods can get a short reduct. Furthermore, a new neighborhood classifier based on three-way decisions is constructed and compared with other classifiers. Comparison experiments show that the proposed classifier can get a high accuracy and a low misclassification cost.\",\n \"Title: DRUGS System Improving the Effects of Clinical Pathways: A Systematic Study\\nYear: 2016\\n\\nAbstract: Abstract The aim of the study is to assess the feasibility of Drugs Rational Usage Guideline System (DRUGS)-supported clinical pathway (CP) for breast carcinoma, cataract, inguinal hernia and 2-diabetes mellitus whether the application of such a system could improve work efficiency, medical safety, and decrease hospital cost. Four kinds of diseases which included 1773 cases (where 901 cases using paper-based clinical pathways and 872 cases using DRUGS-supported clinical pathways) were selected and their demographic and clinical data were collected. The evaluation criteria were length of stay, preoperative length of stay, hospital cost, antibiotics prescribed during hospitalization, unscheduled surgery, complications and prognosis. The median total LOS was 1 to 3 days shorter in the DRUGS-supported CP group as compared to the Paper-based CP group for all types (p < 0.05). Totel hospital cost decreased significantly in the DRUGS-supported CP group than that in Paper-based CP group. About antibiotics prescribed during hospitalization, there were no statistically differences in the time of initial dose of antibiotic and the duration of administration except the choice of antibiotic categories. The proportion of DRUGS-supported clinical pathway conditions where a broad-spectrum antibiotic was prescribed decreased from 63.6 to 34.5 % (p < 0.01) in the Paper-based group. While after the intervention, the differences were statistically not significant in unscheduled surgery, complications and prognosis. In this study, DRUGS-supported clinical pathway for breast carcinoma, cataract, inguinal hernia, 2-diabetes mellitus was smoothly shifted from a paper-based to an electronic system, and confer benefits at the hospital level.\",\n \"Title: Impact Of Technical And Content Quality On Overall Experience Of Ott Video\\nYear: 2016\\n\\nAbstract: Quality of Experience (QoE) is a crucial guiding factor for network management of an end-to-end service session. The network provider can control the resources allocated to sessions and in doing so, influence the Technical Quality (TQ), which covers the technical aspects of signal quality during the session. On the other hand, the network provider has no control over the Content Quality (CQ), which pertains to the user's level of interest in a particular video. Together TQ and CQ influence the Overall eXperience (OX) in a session. In this paper, we present results from a user subjective study in which the impact of TQ and CQ on OX was investigated for Over-The-Top (OTT) video sessions from the perspective of a network provider. This perspective places a focus on those elements of QoE that can be controlled by the provider. Various studies have shown that very high interest in a content can strongly influence QoE independent of other factors, so our study uses videos that are neutral with respect to content. We assess the TQ, CQ and OX for video sessions that contain Integrity impairments (in the form of image freezing) and failures in terms of session Accessibility and Retainability. Our findings indicate that TQ and CQ have a strong impact on OX in the presence of impairments, but no failures. On the other hand, TQ is the main determinant of OX when failures are present.\",\n \"Title: Privacy Perceptive Wireless Service Game Model\\nYear: 2014\\n\\nAbstract: Wireless applications are taking an important role in our lives. More and more merchants begin to provide wireless connection so as to attract customers. For business purpose, they often ask customers to provide their personal information for using wireless, which may threat customer. So, it is desired to make a trade off between the competitive profits of both service providers and user privacy. In this paper, we introduce a privacy perceptive service model based on game theory, which takes into account the interests of all stakeholders. Considering the characteristics of customer behavior in practice, we design the complete information and sequential game. Users are allowed to provide different levels of personal information as their preferences, while a merchant assigns different wireless bandwidth based on customer provided information. The Nash Equilibriums are discussed so as to find the acceptable strategy. Quantitative study is performed on the impact of each parameter in the game, which could help merchants to design appropriate policies.\",\n \"Title: Research on Value Evaluation of E-commerce Business Model\\nYear: 2015\\n\\nAbstract: The study of business models is an important topic for entrepreneurship and organizational purpose because business models affect firms' possibilities for value creation and value capture. The present study leaves a gap to be bridged, and the gap reveals a critically important research problem to be resolved, that is, how to visualize these transaction activities and calculate revenue respectively. This paper, therefore, adopts e3 model for investigations of E-commerce enterprise value generating in the process, and can be used to select or optimize value activities that advance both research and practice in the field of E-commerce enterprises. The result indicates that: Profitability sheet is able to capture customers' behaviors, which gives them insights to better understanding online customers' experiences and expectations, and make reasonable decisions. This model illustrates the main activities of business model and exploits the ways to optimize value activities. This modeling approach provides pro-activities way to actually experiment with alternative business models, by making enterprise to simulate various possibilities before committing to specific investments in reality. (C) 2015 The Authors. Published by Elsevier B.V.\",\n \"Title: Acceptability And Quality Of Experience In Over The Top Video\\nYear: 2015\\n\\nAbstract: Consumer acceptance is of great interest in the adoption of novel multimedia products and services. A number of factors can greatly influence the customer experience during a video session, impacting the acceptability of the product or service. Factors such as the Technical Quality (TQ) which covers the technical aspects of the signal quality can be controlled from the network provider. On the other hand, the network provider has no control on the subject's level of interest in a video, the Content Quality (CQ). Together TQ and CQ influence the Overall eXperience (OX) of the subject and, in the case of Over The Top (OTT) video, in a video session. In this paper, we present results from a user subjective study in which the impact of TQ and OX on the acceptability was investigated for OTT video sessions. To minimize the CQ impact on OX, the videos were carefully selected to have relatively neutral content. We assess the impact of TQ and OX on acceptability for videos containing impairment and failure events during lifecycle, thus affecting Accessibility, Retainability and Integrity. Our experimental results indicate that TQ and OX have a strong impact on acceptability.\",\n \"Title: A Multicriteria Interval-Valued Intuitionistic Fuzzy Set TOPSIS Decision-Making Approach Based on the Improved Score Function.\\nYear: 2016\\n\\nAbstract: In this paper, an improved technique for order preference by similarity to an ideal solution (TOPSIS) method was proposed based on completely unknown attribute weight information, as well as taking multiple criteria decision making of interval-valued intuitionistic fuzzy number as the evaluation information. First of all, a cumulative interval score function considering the influence of hesitation as well as a cumulative score function containing the risk preference of the decision maker were constructed. Then, a new information entropy function was constructed by using the newly defined score function, which presents a new method that directly utilizes evaluation information to obtain criterion weight. Next, all schemes were sequenced by virtue of relative closeness and the criterion weight of each alternative and ideal scheme. Finally, the effectiveness of the proposed method was illustrated by comparison with examples.\",\n \"Title: Singular boundary method for acoustic eigenanalysis.\\nYear: 2016\\n\\nAbstract: This paper applies the singular boundary method (SBM) to two- (2D) and three-dimensional (3D) acoustics eigenproblems in simply- and multiply-connected domains. The SBM is a strong-form boundary discretization numerical method and is meshless, integration-free, and easy-to-implement. By introducing the concept of the source intensity factors, the singularity of fundamental solutions can be isolated to avoid the singular numerical integrals in the boundary element method (BEM). Similar to the BEM, the spurious eigenvalues may arise in the SBM computation. In order to extract out spurious eigenvalues from SBM results, the singular value decomposition updating techniques and the Burton-Miller method are implemented. Several 2D and 3D benchmark examples subjected to Dirichlet and Neumann boundaries are tested to examine the accuracy and stability of the present SBM strategy.\",\n \"Title: The Interaction Design Of Household Intelligent Breathing Training System\\nYear: 2016\\n\\nAbstract: This paper achieved the acquisition of the breathing signal and identification of different respiratory modes by several attempts. On this basis, we designed an intelligent breathing-exercise system to meet the household needs. The intelligent system consists of three parts: the first part is the user interface of the breathing-exercise system, the second part is the background database and the last part is the internet community platform. The user interface of the breathing-exercise system is used for displaying user's breathing curve and other visual feedback. All the breathing training data will be recorded in the second part: database. Those breathing exercisers can share their results on the internet community platform. The intelligent breathing-exercise system will provide a memorable user experience for users by blending these three parts.\",\n \"Title: Understanding The Relationships Between Performance Metrics And Qoe For Over-The-Top Video\\nYear: 2016\\n\\nAbstract: In this paper, we study the relationships between Quality of Service (QoS) and Quality of Experience (QoE) in a session-based Over-The-Top (OTT) video service. A number of Performance Metrics (PMs) with and without the existence of failures during a video are examined. As QoE factors, Technical Quality (TQ) and Acceptability are used. We analyze the correlation between QoS performance metrics and QoE factors, and find new PMs should be employed because failures are included in QoE evaluation. We also summarize the relationships between QoS metrics and QoE factors through machine learning approaches. Using decision tree, we have a general idea about the relationships between PMs and QoE factors. We also understand the impact caused by failures and the value of rating scales.\",\n \"Title: Risk measurement and optimization model of coal generation contracts for the difference between prospect M-V and normal triangular fuzzy stochastic variables.\\nYear: 2016\\n\\nAbstract: Purpose - Coal and power generation are related upstream and downstream industries. Coal price marketization and electricity price regulation have caused the price of coal to be sensitive to the benefits of generators. The paper aims to discuss these issues. Design/methodology/approach - As a financial tool, contracts for differences can both help balance interests and reduce risks caused by spot price fluctuation. This thesis regards coal demand as a triangular fuzzy stochastic variable while directing a levelling consideration towards risk returns for coal and power enterprises that are involved in coal generation contracts for differences. Risk and benefit measurement models were established between coal suppliers and power generators, and risk and benefit balance optimization models for contract negotiation were constructed. Findings - A numerical example showed that the above models can be effectively used to avoid the risks of coal-electricity parties. Originality/value - This thesis regards coal demand as a triangular fuzzy random variable while directing a levelling consideration towards the risk return to coal and power enterprises that are involved with coal generation contracts for differences. The features of this thesis are the following: demand information is regarded as a fuzzy random variable instead of a random variable. With historical data, sales experience and increasingly clear macro-economic conditions, coal and power enterprises are able to make a fuzzy decision - to a certain extent - when the transaction approaches. Accurate market information enables the supply chain system to satisfy the clients' needs better, improve the profit level or avoid severe financial damages; by developing a feasible set of contracts for different parameters, it is possible to estimate whether the price difference enables supply chain coordination, requires changes or gives accounts to all involved parties of the supply chain; and without the assumption that the traditional M-V rule is unfavourable to decision makers, this thesis proposes the prospect M-V rule, which involves decision makers' projections of future coal generation prices and enables wide applicability of the response method to contracts for differences.\",\n \"Title: Chinese Microblog Sentiment Analysis Based on Sentiment Features.\\nYear: 2016\\n\\nAbstract: As the microblog has increasingly become an information platform for netizens to share their ideas, the study on the sentiment analysis of microblog has got scholars’ wide attention both at home and abroad. The primary goal of this research is to improve the accuracy of microblog sentiment polarity classification. With a view to the characteristics of microblog, a new method of semantically related feature extraction is proposed. Firstly, the Chinese word features are selected by text presentation in VSM and computing the weight by TF*IDF. Secondly, the proposed eight microblog semantic features are extracted, including sentence sentiment judgment based on emotional dictionary. Finally, three kinds of machine learning methods are used to classify the Chinese microblog under the feature vector combining the two methods. The experimental results indicate that the proposed feature extraction method outperforms the state-of-the-art approaches, and for this feature extraction algorithm, the classification performance is best when using the Naive Bayes algorithm.\",\n \"Title: Band gap calculations of photonic crystals by singular boundary method.\\nYear: 2017\\n\\nAbstract: This paper applies the singular boundary method (SBM) to compute band gap structures of two-dimensional (2D) photonic crystals. By using the fundamental solutions, a linear eigenvalue equation is established to describe the relationship between the Block wave and frequency. For given frequencies, the solution of the equation yields the related Block wave vectors. The convergence, accuracy and efficiency are carefully examined through several benchmark cases.\",\n \"Title: Capturing User Behavior in Subjective Quality Assessment of OTT Video Service.\\nYear: 2016\\n\\nAbstract: Customer satisfaction is an important factor governing adoption and retention of multimedia products and services, such as Over-The-Top (OTT) video transmission. Quality of Experience involves user-centric evaluation of various services. However, users differ in terms of their ratings of service quality. Some rating differences are due to unreliability(outlierusers who are not motivated, or are not sensitive to differences in quality), but others are systematic differences in rating that may reflect different perspectives on quality. In this paper, we explore the use of outlier analysis and clustering as tools for interpreting QoE data. We report on experimental results demonstrating the use of outlier analysis and clustering. In interpreting the clusters, we examine users' opinions on different types of video disruption, and their ability to distinguish the different levels of impairments/failures.\",\n \"Title: A Cross-Layer Approach To Message Authentication Based On Sparse Representation For Wireless Body Area Networks\\nYear: 2017\\n\\nAbstract: Due to personal privacy, information security is a particularly important issue in wireless body area networks. Although current communication protocol has considered the security, there is still a risk that lapses in physical layer security may be a potential threat in wireless body area networks. In order to achieve the comprehensive security protection, in this article, we propose a cross-layer authentication approach, where the higher layer and physical layer are simultaneously considered. More specifically, different from current literatures, we exploit the sparse representation of the original signal as the basis of physical layer authentication, and a representative selection algorithm is presented to achieve the signal detection. Moreover, based on IEEE Standard 802.15.6, we give a feasible unite authentication scheme, in which physical layer strategy and higher layer strategy are combined to enhance the security of authentication. To verify the effectiveness of the proposed physical layer strategy, we conducted experiments and numerical simulations. Furthermore, the computational complexity of the proposed algorithm and the efficiency of the cross-layer approach are analyzed, respectively. Results showed that the proposed approach can significantly enhance the authentication and the complexity is acceptable (polynomial complexity).\",\n \"Title: Physical-layer security in Internet of Things based on compressed sensing and frequency selection.\\nYear: 2017\\n\\nAbstract: Information security is a vital concern in Internet of Things (IoT). Traditional security method based on public or private key encryption scheme is limited by the trade-off between low cost and high level of security. Among different security solutions, utilising compressed sensing (CS) in combination with the physical-layer security to achieve the security is a remarkable method. However, in the...\",\n \"Title: The Impact of Estimating High-Resolution Tropospheric Gradients on Multi-GNSS Precise Positioning.\\nYear: 2017\\n\\nAbstract: Benefits from the modernized US Global Positioning System (GPS), the revitalized Russian GLObal NAvigation Satellite System (GLONASS), and the newly-developed Chinese BeiDou Navigation Satellite System (BDS) and European Galileo, multi-constellation Global Navigation Satellite System (GNSS) has emerged as a powerful tool not only in positioning, navigation, and timing (PNT), but also in remote sensing of the atmosphere and ionosphere. Both precise positioning and the derivation of atmospheric parameters can benefit from multi-GNSS observations. In this contribution, extensive evaluations are conducted with multi-GNSS datasets collected from 134 globally-distributed ground stations of the International GNSS Service (IGS) Multi-GNSS Experiment (MGEX) network in July 2016. The datasets are processed in six different constellation combinations, i.e., GPS-, GLONASS-, BDS-only, GPS + GLONASS, GPS + BDS, and GPS + GLONASS + BDS + Galileo precise point positioning (PPP). Tropospheric gradients are estimated with eight different temporal resolutions, from 1 h to 24 h, to investigate the impact of estimating high-resolution gradients on position estimates. The standard deviation (STD) is used as an indicator of positioning repeatability. The results show that estimating tropospheric gradients with high temporal resolution can achieve better positioning performance than the traditional strategy in which tropospheric gradients are estimated on a daily basis. Moreover, the impact of estimating tropospheric gradients with different temporal resolutions at various elevation cutoff angles (from 3 degrees to 20 degrees) is investigated. It can be observed that with increasing elevation cutoff angles, the improvement in positioning repeatability is decreased.\",\n \"Title: The design and implement of China-Japan-South Korea logistics information platform based on information service exchange\\nYear: 2016\\n\\nAbstract: With the rapid development of economy, trade and logistics among the three countries, the demand for establishment of the China-Japan-South Korea logistics information platform (CJK-LIP) is more and more urgent. This paper analyzes the requirements of the platform, proposes the overall design idea, and discusses critical technique problems in the platform development and implement. The proposed design method was successfully applied in the construction of CJK-LIP. Application effect of CJK-LIP show that CJP-LIP based on information service exchange will be a valuable model for international logistics information platform construction.\",\n \"Title: The Optimal Checkpoint Interval for the Long-Running Application\\nYear: 2017\\n\\nAbstract: For the distributed computing system, excessive or deficient checkpointing operations would result in severe performance degradation. To minimize the expected computation execution of the long-running application with a general failure distribution, the optimal equidistant checkpoint interval for fault tolerant performance optimization is analyzed and derived in this paper. More precisely, the optimal checkpointing period to determine the proper checkpoint sequence is proposed, and the derivation of the expected effective rate of the defined computation cycle is introduced. Corresponding to the maximal expected effective rate, the constraint of the optimal checkpoint sequence can be obtained. From the constraint of optimality, the optimal equidistant checkpoint interval can be obtained according to the minimal fault tolerant overhead ratio. By the numerical results, the proposal is practical to determine a proper equidistant checkpoint interval for fault tolerant performance optimization.\",\n \"Title: Deformation of facial model for complete denture prosthesis using ARAP group method and elastic properties.\\nYear: 2017\\n\\nAbstract: With the development of 3D printing and computer graphics technology, mouth rehabilitation has increasingly adopted digital methods. This research proposes a new method to transform the appearance of facial model after complete denture prosthesis. A feature template with few feature points is first constructed according to the facial muscle anatomy and facial deformation after complete denture prosthesis. Next, the traditional as-rigid-as-possible (ARAP) method is optimised by clustering based on facial muscles. The optimised ARAP method is then used for real-time and interactive simulations. Finally, by classifying the degrees of elasticity in the model with additional weights, the simulation can be customised to the skin of individual patients. Different degrees of elastic deformation and post-operative models are superimposed for match analysis. Compared with our previous study, the error is reduced by 24.05%. Results show that our method can deform facial models accurately and efficiently.\",\n \"Title: A Multi-objective Attribute Reduction Method in Decision-Theoretic Rough Set Model.\\nYear: 2017\\n\\nAbstract: Many attribute reduction methods have been proposed for decision-theoretic rough set model based on different definitions of attribute reduct, while an attribute reduct can be seen as an attribute subset that satisfies specific criteria. Most reducts are defined on the basis of a single criterion, which may result in the difficulty for users to choose appropriate reduct to design related reduction algorithm. To address this problem, we propose a multi-objective attribute reduction method based on NSGA-II for decision-theoretic rough set model. Three different definitions of attribute reduct based on positive region, decision cost and mutual information are considered and transferred to a multi-objective optimization problem. Experimental results show that the multi-objective reduction method can obtain a robust and better classification performance.\",\n \"Title: A Wearable Autonomous Heart Rate Sensor Based On Piezoelectric-Charge-Gated Thin-Film Transistor For Continuous Multi-Point Monitoring\\nYear: 2017\\n\\nAbstract: Autonomous wearable biomedical sensors enable continuous human body vital signs monitoring with features such as being conformable, mobile, cost-effective and self-powered. In this work, we report on an autonomous and multi-positional sensor capable of heart rate and blood pressure monitoring. The device concept is based on a piezoelectric-charge-gated thin-film transistor (PCGTFT) where a polyvinylidene fluoride (PVDF) piezoelectric sandwich structure is incorporated with an amorphous silicon (aSi:H) dual-gate TFT (DGTFT). An analytical model and preliminary experimental results will be presented along with a demonstration of a proof-of-concept sensing system for continuous multi-point heart rate monitoring.\",\n \"Title: An empirical study of touch-based authentication methods on smartwatches.\\nYear: 2017\\n\\nAbstract: The emergence of smartwatches poses new challenges to information security. Although there are mature touch-based authentication methods for smartphones, the effectiveness of using these methods on smartwatches is still unclear. We conducted a user study (n=16) to evaluate how authentication methods (PIN and Pattern), UIs (Square and Circular), and display sizes (38mm and 42mm) affect authentication accuracy, speed, and security. Circular UIs are tailored to smartwatches with fewer UI elements. Results show that 1) PIN is more accurate and secure than Pattern; 2) Pattern is much faster than PIN; 3) Square UIs are more secure but less accurate than Circular UIs; 4) display size does not affect accuracy or speed, but security; 5) Square PIN is the most secure method of all. The study also reveals a security concern that participants' favorite method is not the best in any of the measures. We finally discuss implications for future touch-based smartwatch authentication design.\",\n \"Title: Partial order reduction for checking LTL formulae with the next-time operator.\\nYear: 2017\\n\\nAbstract: Partial order reduction (POR) has been one of the most successful techniques for reducing the state spaces in model checking concurrent asynchronous systems with respect to linear-time temporal logic (LTL) formulae. Conventional POR only support checking formulae without the next-time operator because these formulae are invariant under stuttering, thus are suitable for POR. In this article we introduce a novel POR technique for checking formulae with the next-time operator. This new technique enables model checking algorithms to use POR even if the formulae are not stutter invariant (SI). More precisely, it supports the full class of LTL formulae with the next-time operator. The core of our approach is a heuristic algorithm for detecting and extending SI components of Buchi automata (BA) during the translation from LTL formulae into BAs. Then, the detected component is used to direct the reduction of the state space of a system. In addition, an SI-component of a BA enables an improvement of LTL model checking algorithms based on the nested depth-first search. We show that our technique is efficient when the synchronous product of an SI-component and a system is large.\",\n \"Title: Quantile Induced Heavy Ordered Weighted Averaging Operators And Its Application In Incentive Decision Making\\nYear: 2018\\n\\nAbstract: To integrate incentives into the information aggregation process in decision making, we propose a new type of aggregation operator, denominated as the quantile induced heavy ordered weighted averaging (QI-HOWA) operator in this paper. A primary characteristic of this type of operator is that the quantile variable can be used not only to measure the relative performance of alternatives but also to facilitate the incentive preference expression of the decision maker. We further provide a calculation technology of the QI-HOWA weights, in which various incentive preferences of the decision maker can be considered through parameter adjustment. In addition, we discuss certain properties of the QI-HOWA operator and note the extent to which they are effective. Finally, a numerical example regarding the selection of optimal alternatives by incentive measures is provided, and the aggregations are compared with those of ordered weighted averaging and unweighted averaging operators to illustrate the validity of the QI-HOWA operator.\",\n \"Title: The Study Of The Mid-Term Screening Of Graduate Students Based On The Attribute Reduction Algorithm\\nYear: 2018\\n\\nAbstract: In this paper, we use attribute reduction algorithm and attribute value reduction algorithm based on rough set theory to analyze the sample data of the mid-term screening of graduate students in a higher university. Through the analysis of the sample data, we get the key factors influencing the comprehensive evaluation of midterm screening of graduate students. Factors for this the best way to develop graduate students teaching management, to the midterm screening effective guidance later stage of learning, to raise the level of our graduate students teaching, management, training, in order to raise the high quality for the state and society of outstanding graduate students.\",\n \"Title: Label Distribution Learning by Exploiting Sample Correlations Locally.\\nYear: 2018\\n\\nAbstract: Label distribution learning (LDL) is a novel multi-label learning paradigm proposed in recent years for solving label ambiguity. Existing approaches typically exploit label correlations globally to improve the effectiveness of label distribution learning, by assuming that the label correlations are shared by all instances. However, different instances may share different label correlations, and few correlations are globally applicable in real-world applications. In this paper, we propose a new label distribution learning algorithm by exploiting sample correlations locally (LDL-SCL). To encode the influence of local samples, we design a local correlation vector for each instance based on the clustered local samples. Then we predict the label distribution for an unseen instance based on the original features and the local correlation vector simultaneously. Experimental results demonstrate that LDL-SCL can effectively deal with the label distribution problems and perform remarkably better than the state-of-the-art LDL methods.\",\n \"Title: Label Distribution Learning by Exploiting Label Correlations.\\nYear: 2018\\n\\nAbstract: Label distribution learning (LDL) is a newly arisen machine learning method that has been increasingly studied in recent years. In theory, LDL can be seen as a generalization of multi-label learning. Previous studies have shown that LDL is an effective approach to solve the label ambiguity problem. However, the dramatic increase in the number of possible label sets brings a challenge in performance to LDL. In this paper, we propose a novel label distribution learning algorithm to address the above issue. The key idea is to exploit correlations between different labels. We encode the label correlation into a distance to measure the similarity of any two labels. Moreover, we construct a distance-mapping function from the label set to the parameter matrix. Experimental results on eight real label distributed data sets demonstrate that the proposed algorithm performs remarkably better than both the state-of-the-art LDL methods and multi-label learning methods.\",\n \"Title: The Consideration of Formal Errors in Spatiotemporal Filtering Using Principal Component Analysis for Regional GNSS Position Time Series.\\nYear: 2018\\n\\nAbstract: In the daily operation of regional GNSS (Global Navigation Satellite System) networks, the formal errors of all stations' coordinate components are calculated. However, spatiotemporal filtering based on traditional Principal Component Analysis (PCA) for regional GNSS position time series does not take these formal errors into account. This paper developed a PCA-based approach to extract Common Mode Error (CME) from the position time series of a regional GNSS station network, where formal errors were applied to construct a weight factor. Because coordinate components with larger errors have smaller weight factors in extracting CME, the performance of our proposed approach was anticipated to be better than the traditional PCA approach. The position time series of 25 stations in the Yunnan Province, China, were analyzed using our approach, as well as the traditional PCA approach. The average errors of the residual time series after removing the CMEs with our approach were 1.30 mm, 1.72 mm, and 4.62 mm for North, East and Up components, and the reductions with respect to those of the original time series were 18.23%, 15.42%, and 18.25%, respectively. If CMEs were removed from the traditional PCA approach, the corresponding average errors were 1.34 mm, 1.81 mm, and 4.84 mm, with reductions of 15.84%, 10.86%, and 14.32%, respectively. Compared to the traditional PCA approach, the average errors of our approach were reduced by about 2.39%, 4.56%, and 3.93% in the North, East and Up components, respectively. Analysis of CME indicated that it mainly contained white and flicker noise. In the synthetic position time series with 500 repeated simulations, the CME extracted by our approach was closer to the true simulated values than those extracted by the traditional PCA approach, regardless of whether local effects were considered or not. Specifically, the mean root mean square (RMS) reduction of our approach, relative to PCA, ranged from 1.35% to 3.93%. Our simulations illustrated that the RMS reductions depended not only on the magnitude, but also the variation of the formal error series, which further highlights the necessity of considering formal errors in spatiotemporal filtering.\",\n \"Title: Application Of Attribute Reduction Algorithm Based On Rough Set In Financial Management Of Commercial Banks\\nYear: 2017\\n\\nAbstract: In this paper, we use the attribute reduction algorithm of rough set to analyze the sample data of the annual balance of the residents in commercial banks. Through the analysis of the sample data, we get different customer categories. For different customers, we develop different financial products. On the one hand, it meets the needs of customers; on the other hand, it increases the profitability of commercial banks, so as to realize the new idea of 'guided by data'.\",\n \"Title: Bug Patterns Localization Based on Topic Model for Bugs in Program Loop\\nYear: 2018\\n\\nAbstract: Statistic fault localization technique is a popular technique to perform fault localization via comparing the different statistics of failed executions and successful executions. They measure suspiciousness for each program component and rank them descending by their suspiciousness to assist programmers to identify root cause of failures. Unfortunately, since these techniques based on hit-spectrum information, which use binary flags to register whether a program component is covered or not during a program execution, their results are inherently limited, especially when the bug(s) occur in complex program loops. To resolve this issue, we propose a bug patterns localization technique based on topic model for the bugs in complex loops. We also conduct experiments involving 28 program versions contained bugs in program loops in Siemens program suite to compare the effectiveness of the proposed approach against Dstar. The results show that our approach is promising.\",\n \"Title: A quantitative relationship between Application Performance Metrics and Quality of Experience for Over-The-Top video.\\nYear: 2018\\n\\nAbstract: Quality of Experience (QoE) is a measure of the overall level of customer satisfaction with a vendor. In telecommunications, consumer satisfaction is of great interest in the adoption of novel multimedia products and services. A number of factors can greatly influence the customer experience during a video session. Factors such as user perception, experience, and expectations are expressed by QoE while factors such as application and network performance are expressed by Quality of Service (QoS) parameters. This paper studies the relationship between QoS and QoE in a session-based mobile video streaming. Specific QoS Application Performance Metrics (APMs) are examined based on a QoE assessment database which is built for experimentation and contains 108 subjects. It is shown that these APMs are highly related to two QoE factors, Technical Quality (TQ) and Acceptability. Furthermore, Viewing Ration (VR) parameter and the corresponding Kendall correlation between VR and QoE factors proves that VR is a valuable metric for mapping QoS to QoE. We further generated the compacted decision tree to predict QoE factors through Rebuffering Ratio (RR), Non-interruption Content Viewing Ratio (VRc), and Non-interruption Viewing Ratio (VRs). Through extensive experimentation, a general relationship between APMs and QoE factors has been examined and a primary QoE model is proposed based on this relationship.\",\n \"Title: Mobile physical layer spoofing detection based on sparse representation.\\nYear: 2018\\n\\nAbstract: Recently, physical layer authentication techniques are emerging to detect spoofing attacks in wireless networks based on high consistency between two legitimate successive channel information. However, this high consistency is not always available, especially in dynamic networks, such as mobile scenarios. In this study, the authors propose a physical layer spoofing detection scheme based on sparse...\",\n \"Title: Design And Contact Analysis Of Gripper For Tube Inspection Robots\\nYear: 2018\\n\\nAbstract: Automatic robots have been extensively employed to inspect steam generator (SG) tubes. Positioning and fixation of robots underneath tube-sheet plates are enabled by tube grippers. In this paper, configuration design and performance analysis of a new tube gripper were carried out. To calculate the gripper's load capacity, numerical analysis was used to establish finite element (FE) model of the gripper. Furthermore, orthogonal experiment and statistical methods combined with FE analysis were adopted to explore the influence on contact between the gripper and the tube's inner wall by manufacturing tolerance. Three factors of dimension and form tolerance were selected in sensitivity analysis. The influencing factors of contact stress were analysed by analysis of variance (ANOVA). Physical prototype experiment results were provided to validate the design rationality. Results of analysis and actual tests proved that the gripper with inelastic fingers exhibited excellent load capacity, superior contact performance and longer service life time.\",\n \"Title: Physical layer spoofing detection based on sparse signal processing and fuzzy recognition.\\nYear: 2017\\n\\nAbstract: Spoofing attacks is one of the most critical attacks in wireless communication security. Traditional solutions are based on cryptology which is performed in the upper layers, and face many challenges especially in resource-limited application. To overcome this hurdle, physical-layer security has been received a lot of attention recently. In this study, the authors propose a physical-layer spoofing...\",\n \"Title: Transforming the SMV model into MMV model based on the characteristics of wavelet coefficients.\\nYear: 2017\\n\\nAbstract: Sparse signal recovery or compressed sensing (CS) theory has recently received attention in the image compression field. CS has proven that the successful recovery rate of the multiple measurement vectors (MMVs) model is higher than that of the single measurement vector (SMV) case. Most existing algorithms have focused on sparse signal recovery using the MMV model, without considering converting t...\",\n \"Title: Subspace Clustering Through Sub-Clusters\\nYear: 2018\\n\\nAbstract: The problem of dimension reduction is of increasing importance in modern data analysis. In this paper, we consider modeling the collection of points in a high dimensional space as a union of low dimensional subspaces. In particular we propose a highly scalable sampling based algorithm that clusters the entire data via first spectral clustering of a small random sample followed by classifying or labeling the remaining out-of-sample points. The key idea is that this random subset borrows information across the entire dataset and that the problem of clustering points can be replaced with the more efficient problem of \\\"clustering sub-clusters\\\". We provide theoretical guarantees for our procedure. The numerical results indicate that for large datasets the proposed algorithm outperforms other state-of-the-art subspace clustering algorithms with respect to accuracy and speed.\",\n \"Title: Multi-objective attribute reduction in three-way decision-theoretic rough set model.\\nYear: 2019\\n\\nAbstract: Attribute reduction plays an important role in rough set theory. Many attribute reduction methods have been proposed based on different definitions of attribute reduct, while an attribute reduct can be regarded as a minimal attribute subset that satisfies specific criteria. Most reducts are defined on the basis of a single criterion, which can only affect one specific characteristic of the data or one preference of users. However, the single criterion-based attribute reduct may not meet the requirement of complex problems. To address this problem, based on three-way decision-theoretic rough set model, this paper defines a multi-objective attribute reduct. Three types of criteria for defining attribute reduct including the positive region, decision cost and mutual information are considered and combined to a multi-objective optimization problem. Based on the proposed multi-objective attribute reduct, we also introduce a multi-objective optimization-based attribute reduction method and an ensemble learning-based attribute reduction method. Experimental results on several datasets show that the proposed attribute reduction methods can obtain a robust and better classification performance.\",\n \"Title: A fast singular boundary method for 3D Helmholtz equation.\\nYear: 2019\\n\\nAbstract: This paper presents a fast singular boundary method (SBM) for three-dimensional (3D) Helmholtz equation. The SBM is a boundary-type meshless method which incorporates the advantages of the boundary element method (BEM) and the method of fundamental solutions (MFS). It is easy-to-program, and attractive to the problems with complex geometries. However, the SBM is usually limited to small-scale problems, because of the operation count of O(N3) with direct solvers or O(N2) with iterative solvers, as well as the memory requirement of O(N2). To overcome this drawback, this study makes the first attempt to employ the precorrected-FFT (PFFT) to accelerate the SBM matrix–vector multiplication at each iteration step of the GMRES for 3D Helmholtz equation. Consequently, the computational complexity can be reduced from O(N2) to O(NlogN) or O(N). Three numerical examples are successfully tested on a desktop computer. The results clearly demonstrate the accuracy and efficiency of the developed fast PFFT-SBM strategy.\",\n \"Title: Behavioral Ordered Weighted Averaging Operator And The Application In Multiattribute Decision Making\\nYear: 2019\\n\\nAbstract: This paper introduces a new type of behavioral ordered weighted averaging (BOWA) operator, to incorporate decision maker's gains and losses behavior tendency into the information aggregation process. The main characteristic of this BOWA operator is that it considers behavioral weights and ordered weights in the same formulation. We further provide a calculation method of the behavioral weights, in which various psychological preferences of different attribute types of the decision maker can be expressed intuitively. In addition, we discuss some particular cases of BOWA operator and its main properties. Finally, a numerical example is used to illustrate the use of the proposed method.\",\n \"Title: Anomalous Path Detection for Spatial Crowdsourcing-Based Indoor Navigation System.\\nYear: 2018\\n\\nAbstract: Indoor navigation system provides customized path planning for requesters who are unfamiliar with the indoor environment, such as shopping mall and airport. Spatial crowdsourcing technology can be applied to indoor navigation to offer fundamental services related to location. However, spatial crowdsourcing-based indoor navigation is vulnerable to the intrusion of injected anomalous paths from attackers. In this paper, we propose an anomalous path detection (APD) scheme to classify attackers according to their reputation management and abnormal trajectory sequence. Specifically, we first develop a crowdsourcing system to support the indoor location service using the fog as the spatial crowdsourcing server. Then, we identify two levels of attackers, i.e., the malicious responders and the semi-honest responders in the indoor environment according to their attacking purposes. Through the responders' historical records from the fog server, we analyze a series of trajectory sequences consisting of the distance between the current position and the destination to distinguish the semi-honest responders from the normal. In addition, we propose a semi-supervised learning with hidden Markov model (HMM) to detect the semi-honest responders. Finally, the extensive simulations show that the APD scheme can achieve higher accuracy with the acceptable false rate.\",\n \"Title: Experimental study of human body effect on temperature sensor integrated RFID tag\\nYear: 2017\\n\\nAbstract: An ultra-high frequency (UHF) passive radio frequency identification (RFID) tag consisting of flexible antenna and built-in temperature sensor chip is utilized for temperature monitoring of human body. A series of experiments were performed to study the body effects on temperature measurements. The tags were mounted at different positions of the body and in the different tag orientations as well. The experimental study of sensor integrated RFID tag shows that performance of the tag is more stable, yielding good signal strength and more consistent temperature measurement given the small form factor and form attachment. Minimizing the deteriorating effect of human body on tag's performance and the accuracy of measurements in different scenerios gives the perspective for implementation in realtime physiological monitoring as per the medical requirements.\",\n \"Title: A Natural Language Processing Method Of Chinese Instruction For Multi-Legged Manipulating Robot\\nYear: 2018\\n\\nAbstract: In this paper, a natural language processing (NLP) framework is proposed to build the Chinese natural language interaction between human and the multi-legged manipulating robot which has both locomotion and manipulation functions. The Chinese natural language instruction is transformed into formal representations by this method. Firstly, Cascaded Conditional Random Fields (CCRFs) is employed to analyze the syntax of the instruction. Letter-based features are used in each layer of CCRFs to solve problem caused by word segmentation. In order to understand the modification relationship between entity classes in chunk, one judgment method based on Support Vector Machine (SVM) is proposed. To solve the problem of the large number of core verbs in instructions generated by the multiple robot motion types, a classification of verbs based on Naive Bayes classifier was presented. And the semantic framework of each type of verbs is established to determine the necessary and unnecessary roles for each kind of verbs. At last, several experiments are carried out and the results of each step of framework are presented to demonstrate the effectiveness of the method. It is instructive to understand Chinese natural language instructions for robots with complex motion.\",\n \"Title: A multiphase cost-sensitive learning method based on the multiclass three-way decision-theoretic rough set model.\\nYear: 2019\\n\\nAbstract: The decision-theoretic rough set (DTRS) can be regarded as a type of cost-sensitive learning method that incorporates cost functions. Three-way decisions for an object are made in the DTRS model to obtain the minimum Bayesian decision cost, including acceptance, rejection and deferment decisions. However, because the cost matrix in traditional cost-sensitive learning problems does not include the cost for deferment decisions, DTRS cannot be directly used to solve traditional cost-sensitive learning problems. In this paper, we present a multiclass three-way decision-theoretic rough set model that can be directly applied to traditional cost-sensitive learning problems. This model can calculate the corresponding cost for deferment decisions based on the traditional cost matrix. In addition, we propose a multiphase cost-sensitive learning method based on the multiclass three-way decision-theoretic rough set model. The deferment and rejection samples are reduced and eliminated through multiphase iterative learning. Compared with the existing multiclass cost-sensitive learning methods, our proposed method can achieve higher classification accuracy and lower misclassification cost.\",\n \"Title: Rank-Based Analysis Method To Determine Owa Weights And Its Application In Group Decision Making\\nYear: 2019\\n\\nAbstract: To determine the weights of the ordered weighted averaging (OWA) operator, a new approach that is termed as rank-based analysis method with subjective information is developed. Available information called importance ratio that is obtained by subjective judgment of expert(s) is used in the OWA weights determination. The proposed approach is further extended to the situation of group decision making and is also used to obtain the weights of experts in this situation. In this case, the final OWA weights are obtained through a complete consistent group judgment matrix. Finally, a numerical example regarding the selection of the optimal R&D personnel is given to illustrate the application of the new method in group decision making.\",\n \"Title: Induced Cluster-Based Owa Operators With Reliability Measures And The Application In Group Decision-Making\\nYear: 2019\\n\\nAbstract: Considering the distributed structural characteristics of arguments to be aggregated, we propose a new type of aggregation operator, called induced cluster-based ordered weighted averaging (OWA; abbreviated as cluster-IOWA) operator, in this article. The main characteristic of the cluster-IOWA operator is that the arguments are aggregated by local clusters, and the order-inducing variable is used for representing a particular characteristic with respect to a local cluster. The cluster-OWA operator is commutativity, idempotence, and boundedness. We then discuss two important issues with respect to the cluster-IOWA operator. The order-inducing variables are determined by considering the overall reliability of the local cluster. Based on this, the position weighting vector of the local clusters is designed by taking into account both the reliability measures and the decision maker's preference. Finally, a numerical example, regarding the performance evaluation of middle managers carried out by a group of participants, is developed to illustrate the application and validity of the cluster-IOWA operator.\",\n \"Title: Calibration of the Copolarized Backscattering Measurements From Gaofen-3 Synthetic Aperture Radar Wave Mode Imagery\\nYear: 2019\\n\\nAbstract: Accurate absolute radiometric calibration of spaceborne synthetic aperture radar (SAR) sensor is of great importance in quantitative oceanic monitoring. Traditionally, the calibration constant is determined by analyzing measurements of man-made calibrators deployed on land at a high expense. In this paper, a technique called numerical weather prediction (NWP)-based ocean calibration method for estimating the calibration constant on the basis of SAR observations over ocean is applied to the first Chinese C-band SAR satellite Gaofen-3 (GF-3). The ocean calibration is performed on GF-3 wave mode SAR images at VV and HH polarizations over a period of one year from September 2017 to August 2018. Verification against the independent scatterometer-derived winds shows that after NWP-based ocean recalibration, the accuracy of wind speed retrieval from GF-3 wave mode imagery could reach the wind speed estimation performance using the state-of-the-art of SAR data. These results indicate that GF-3 wind retrievals are promising for operational oceanic products and applications if the SAR data are appropriately calibrated by the proposed NWP-based ocean calibration approach.\",\n \"Title: Intelligent diagnosis of cardiac valve calcification in ESRD patients with peritoneal dialysis based on improved Takagi-Sugeno-Kang fuzzy system\\nYear: 2019\\n\\nAbstract: AbstractWithout a B-ultrasound result, if a doctor diagnoses a suspected patient using only the basic clinical features, such as age, gender, serum calcium, and urea clearance index (KT/V), the diagnostic accuracy will be very low (even less than 50%). To solve this problem, a machine learning technology is proposed to intelligently diagnose cardiac valve calcification in end-stage renal disease (ESRD) patients with peritoneal dialysis. Compared with classical classification technologies, the proposed method aims to develop a model that has both medical interpretability and high recognition performance. In terms of interpretability, the Takagi-Sugeno-Kang fuzzy system is considered a basic model due to its built-in interpretable ability. In addition, because the distribution of the positive class (cardiac valve calcification is positive) and negative class (cardiac valve calcification is negative) in the peritoneal dialysis patient dataset is unbalanced, a novel unbalanced TSK (Takagi-Sugeno-Kang) fuzzy system (B-TSK-FS) is developed using a novel unbalanced fuzzy learning mechanism. The corresponding results reveal that the B-TSK-FS method obtains promising results (the max testing accuracy is over 98%) compared with classical machine learning methods for intelligently diagnosing cardiac valve calcification in ESRD patients with peritoneal dialysis.\",\n \"Title: Competitive Behavior Induced Owa Operator And The Weighting Method\\nYear: 2021\\n\\nAbstract: In view of the decision-making problem that the alternatives actively participate in the decision-making process and present competitive behavior, this paper proposes a new type of aggregation operator called the competitive behavior induced ordered weighted averaging (CB-IOWA) operator. The primary characteristic of the CB-IOWA operator is that the alternative is permitted to highlight its competitive advantage through the order-inducing variables and the weighting vector. A method is introduced to determine the final weighting vector based on the weights provided by different alternatives that can reflect their competitive advantages. The final weights are calculated by maximizing the closeness degree between the final weights and the ones provided the alternatives. The main properties of the CB-IOWA operator are discussed, such as commutativity, boundness, and monotonicity under certain condition. Finally, an application example regarding the competitive development of the economy-environment of four cities in China is provided to illustrate the usage of the CB-IOWA operator. The validity of the operator is checked by comparison with the weighted arithmetic averaging operator representing noncompetitive behavior.\",\n \"Title: A System-Level Approach Towards A Hybrid Energy Harvesting Glove\\nYear: 2021\\n\\nAbstract: This paper presents a novel wearable hybrid harvester system as a glove that contains four distinct scavenging modules of flexible transducer film, photosensitive 3D dual-gate thin-film transistor, and a particular power management box. Each single module is formed by a piezoelectric-charge-gated TFT (PCGTFT). The reported system is capable of scavenging energy from two various free of charge energy sources (Piezoelectric plus Photoelectric). Aforesaid system unlike other state-of-the-arts overcomes several key challenges in interfacing, storage and power management. Harvested energy which is administered through power and storage management system ultimately lightens a typical light emitting diode (LED), testifies capability of such glove to power up some low-power electronic devices.\",\n \"Title: Reprogrammable 3d Liquid-Crystalline Actuators With Precisely Controllable Stepwise Actuation\\nYear: 2021\\n\\nAbstract: Liquid-crystalline elastomers (LCEs) are considered ideal soft actuator materials for a wide range of applications, especially the thriving soft robotics. However, 3D LCE actuators capable of precisely controllable stepwise actuation, which can enhance functionality and versatility of LCE robots for multifarious complicated applications, are still in urgent need for the reported LCE actuators mainly exploit the one-step actuation upon the liquid-crystallin (LC)-isotropic phase transition temperature (T-i). Herein, a catalyst-free LC-vitrimer actuator with supercritical behavior is designed, which can perform precisely controllable stepwise actuation with extraordinary shape stability over a broad temperature range of about 70 degrees C. Moreover, supercritical behavior enables the actuator to be used in nematic phase, imparting the actuator with some extra advantages, such as higher mechanical strength and actuation stability, over the one used above T-i. Furthermore, the LCE can be reprogrammable into arbitrary 3D actuators, which can further be integrated into single-material actuators with complex stepwise actuation, offering a generalized strategy of LCE actuators for sophisticated practical soft robots.\",\n \"Title: Semi-supervised label distribution learning via projection graph embedding\\nYear: 2021\\n\\nAbstract: Label distribution learning (LDL) is a new machine learning paradigm that addresses label ambiguity by emphasizing the relevance of each label to a particular instance. In supervised learning, many LDL algorithms have been proposed, which often require a large amount of well-annotated training data to achieve good performance. However, annotating a label distribution is more complicated and expensive than annotating a single label or multiple labels with logical values of 0 and 1. Thus, we propose a projection graph embedding algorithm for semi-supervised label distribution learning (PGE-SLDL). Specifically, we seek a potential space by orthogonal neighborhood preserving projections, named capture space. This capture space is used to select more valuable features and construct a graph that contains more accurate data structure information. We utilize the sample correlation information contained between graph nodes to recover the unknown label distribution. In addition, compared with fixed graphs in traditional semi-supervised learning, we carry out projection and graph construction simultaneously to obtain a self-updating projection graph, which is more helpful to learn label distribution. The experimental results validate the effectiveness of the proposed algorithm.\",\n \"Title: Integrated Modulation of Dual-Parallel NPC Inverters With Eliminated CMV\\nYear: 2022\\n\\nAbstract: An integrated zero common-mode voltage (ZCMV) modulation algorithm is developed in this article for dual-parallel neutral-point-clamped (NPC) inverters treated as an integrated system. While the conventional methods achieve CMV elimination with the cost of worse harmonic performance, the proposed method guarantees a ZCMV and better current quality simultaneously by taking advantage of the extra sp...\",\n \"Title: Unpaired Underwater Image Enhancement Based on CycleGAN\\nYear: 2022\\n\\nAbstract: Underwater image enhancement recovers degraded underwater images to produce corresponding clear images. Image enhancement methods based on deep learning usually use paired data to train the model, while such paired data, e.g., the degraded images and the corresponding clear images, are difficult to capture simultaneously in the underwater environment. In addition, how to retain the detailed information well in the enhanced image is another critical problem. To solve such issues, we propose a novel unpaired underwater image enhancement method via a cycle generative adversarial network (UW-CycleGAN) to recover the degraded underwater images. Our proposed UW-CycleGAN model includes three main modules: (1) A content loss regularizer is adopted into the generator in CycleGAN, which constrains the detailed information existing in one degraded image to remain in the corresponding generated clear image; (2) A blur-promoting adversarial loss regularizer is introduced into the discriminator to reduce the blur and noise in the generated clear images; (3) We add the DenseNet block to the generator to retain more information of each feature map in the training stage. Finally, experimental results on two unpaired underwater image datasets produced satisfactory performance compared to the state-of-the-art image enhancement methods, which proves the effectiveness of the proposed model.\",\n \"Title: Security Solutions for IoT-Enabled Applications Against the Disease Pandemic\\nYear: 2021\\n\\nAbstract: Amid the COVID-19 pandemic, the Internet of Things (IoT) has rapidly taken on a powerful role to combat it with the emerging smart medicine, smart community, smart city, smart home, and so on. IoT devices are widely deployed in various places to provide feasible solutions for personnel temperature detection, medical treatment, and quarantine control during the pandemic. However, IoT-enabled applications inevitably face serious security challenges due to the massive amount of sensitive data. In this article, we first investigate several strategies of pandemic control based on IoT and the critical supporting technologies, 5G, artificial intelligence, and blockchain. Then we present the security issues in the process of data collection to usage in the applications. Through the analysis of specific paradigms, the Internet of Medical Things (IoMT) and a location check-in system, we exploit the potential security challenges, including data leakage and false sensing. Countermeasures are proposed to address these issues and are effectively validated by simulation results. Finally, some open issues instead of data security threats of IoT are outlined.\",\n \"Title: Skeleton-Based Spatio-Temporal U-Network for 3D Human Pose Estimation in Video\\nYear: 2022\\n\\nAbstract: Despite the great progress in 3D pose estimation from videos, there is still a lack of effective means to extract spatio-temporal features of different granularity from complex dynamic skeleton sequences. To tackle this problem, we propose a novel, skeleton-based spatio-temporal U-Net(STUNet) scheme to deal with spatio-temporal features in multiple scales for 3D human pose estimation in video. The proposed STUNet architecture consists of a cascade structure of semantic graph convolution layers and structural temporal dilated convolution layers, progressively extracting and fusing the spatio-temporal semantic features from fine-grained to coarse-grained. This U-shaped network achieves scale compression and feature squeezing by downscaling and upscaling, while abstracting multi-resolution spatio-temporal dependencies through skip connections. Experiments demonstrate that our model effectively captures comprehensive spatio-temporal features in multiple scales and achieves substantial improvements over mainstream methods on real-world datasets.\",\n \"Title: Hyperspectral Classification Based on Coupling Multiscale Super-Pixels and Spatial Spectral Features\\nYear: 2022\\n\\nAbstract: Hyperspectral remotely sensed images contain not only the spatial information of ground objects, but also their rich spectral information. Effectively improve the classification accuracy of hyperspectral images (HSIs), the purpose is to accurately grasp the current land resource utilization information, which is of great significance to the formulation and implementation of future land and space planning. Existing research on the classification of hyperspectral remotely sensed images uses a single-scale superpixel method for image segmentation. The optimal number of superpixels cannot be determined, image details may get missed, and a single kernel matrix cannot represent the information from multiple features, which results in reduced classification accuracy. Therefore, this study intends to use the superpixel segmentation method to perform multiscale superpixel segmentation on the principal components of HSIs at multiple scales, and to couple the multiscale superpixel spatial spectral kernel (SSK) with the original spectral kernel to form a synthetic kernel using weights for HSI classification. The three HSIs of Pavia University, Pavia Center, and Washington DC Mall are used as the experimental data to test and analyze this method. The experimental results show that the effective classification accuracies for the three datasets obtained by this method are 5.28%, 5.90%, and 7.71% higher than the five comparison methods, at best. The results prove that this method can effectively solve the problem of imprecise image feature extraction and an unknown number of initial superpixels and can significantly improve the classification accuracy of HSIs.\",\n \"Title: Semantic-Structural Graph Convolutional Networks for Whole-Body Human Pose Estimation\\nYear: 2022\\n\\nAbstract: Existing whole-body human pose estimation methods mostly segment the parts of the body's hands and feet for specific processing, which not only splits the overall semantics of the body, but also increases the amount of calculation and the complexity of the model. To address these drawbacks, we designed a novel semantic-structural graph convolutional network (SSGCN) for whole-body human pose estimation tasks, which leverages the whole-body graph structure to analyze the semantics of the whole-body keypoints through a graph convolutional network and improves the accuracy of pose estimation. Firstly, we introduced a novel heat-map-based keypoint embedding, which encodes the position information and feature information of the keypoints of the human body. Secondly, we propose a novel semantic-structural graph convolutional network consisting of several sets of cascaded structure-based graph layers and data-dependent whole-body non-local layers. Specifically, the proposed method extracts groups of keypoints and constructs a high-level abstract body graph to process the high-level semantic information of the whole-body keypoints. The experimental results showed that our method achieved very promising results on the challenging COCO whole-body dataset.\",\n \"Title: Orientation and Channel-Independent RF Fingerprinting for 5G IEEE 802.11ad Devices\\nYear: 2022\\n\\nAbstract: Physical-layer fingerprinting is a promising technique to identify Internet of Things (IoT) devices. In this article, we investigate a new radio-frequency (RF) fingerprinting based on the distinctive signal-to-noise-ratio (SNR) trace in the sector-level sweep (SLS) procedure of 5G IEEE 802.11ad devices. This SLS SNR trace-based fingerprinting can directly apply to off-the-shelf devices without any extra hardware requirements and be independent of the wireless channel and environment. To tackle the impact of orientation on the RF fingerprinting, we propose a novel fingerprinting framework, involving correlation analysis, surface fitting, curve pursuing, and binary classification, named the CSCB framework. Using this framework, the proposed SLS SNR trace-based fingerprinting can achieve device authentication at any orientation with one receiver under line-of-sight (LOS) or non-LOS (NLOS) scenarios. We conduct proof-of-concept experiments using off-the-shelf IEEE 802.11ad devices (Talon AD7200 and MG360 WiGig) to evaluate the performance of the proposed fingerprinting schemes. Experimental results show the effectiveness of the proposed fingerprinting schemes where the verification accuracy of the proposed scheme can reach 99% with only 200 training samples.\",\n \"Title: Retrospective quality by design r(QbD) for lactose production using historical process data and design of experiments\\nYear: 2022\\n\\nAbstract: Quality by Design (QbD) is a popular formal approach for designing, upscaling and optimizing industrial production facilities towards guaranteed quality. To avoid the many costly experiments required for QbD, historical production data may be exploited for optimization instead in what is known as a retrospective QbD (rQbD) study. Current rQbD literature does limitedly discuss data-driven identification of Critical Process Parameters (CPPs) to optimize limited process knowledge availability, and does not cover situations where technical operation limits have not yet been fully explored and/or where parallel equipment (lines) are used. This work presents a new rQbD strategy that addresses these challenges by balancing knowledge that can be obtained from statistical analysis of historical data, together with process experts with a carefully designed set of plant-scale experiments within current operational limits. This novel strategy is demonstrated on a long-running industrial lactose production facility. By digitally and experimentally exploring historical operation variability, we found new operational regimes for this production that may lead to up to 7% product quality improvement, reduced energy consumption and increased process under-standing. Although optimizing a specific process by necessity requires a process-specific approach, the way in which we systematically optimize the current process with Hybrid AI (combining available knowledge with new insights from historical data) shows that approaches that are currently used in prospective process upscaling may be modified to be invaluable for optimization of full-scale processes with a long operational history. (C) 2022 The Authors. Published by Elsevier B.V. CC_BY_4.0\",\n \"Title: Interpretable Personalized Experimentation\\nYear: 2022\\n\\nAbstract: Black-box heterogeneous treatment effect (HTE) models are increasingly being used to create personalized policies that assign individuals to their optimal treatments. However, they are difficult to understand, and can be burdensome to maintain in a production environment. In this paper, we present a scalable, interpretable personalized experimentation system, implemented and deployed in production at Meta. The system works in a multiple treatment, multiple outcome setting typical at Meta to: (1) learn explanations for black-box HTE models; (2) generate interpretable personalized policies. We evaluate the methods used in the system on publicly available data and Meta use cases, and discuss lessons learnt during the development of the system.\",\n \"Title: Label enhancement with label-specific feature learning\\nYear: 2022\\n\\nAbstract: Label distribution learning (LDL) is a novel machine learning paradigm. It addresses the problem of label ambiguity by emphasizing the relevance of each label to a particular instance. Unlike the simple logic vectors in single label learning (SLL) and multi-label learning (MLL), LDL assigns descriptive labels to each instance. Since it is often difficult to collect training sets with precise labels, label enhancement (LE) can be used to recover label distributions from logical labels. Most existing LE models construct label distributions based on a mapping from the feature space to the label space and include two main assumptions: (1) the relationship between features and labels is linear, and (2) all features are shared by all labels. However, in reality, the relationship between features and labels is not purely linear, and different labels may be determined by different features. To solve this problem, in this paper, we propose a new algorithm, that first maps features to a high-dimensional space to explore the nonlinear mapping relationship between features and labels, and then takes full advantage of the labels’ properties determined by specific features, while considering the correlation between labels. The experimental comparison with the existing algorithms verifies the effectiveness of the algorithm proposed in this paper.\",\n \"Title: Label distribution learning with noisy labels via three-way decisions\\nYear: 2022\\n\\nAbstract: Label distribution learning (LDL) as a soft-labeling paradigm is allowed to learn single or multi-labeled information distribution. Overwhelmingly, in the open world, the distribution of labels is usually disturbed by the noise (such as the man-made induction bias, shake of hardware devices), which in turn affects the decision of downstream tasks. To address this problem, we propose a novel LDL approach by using the three-way decisions theory to clear the amplified noise in this paper. First, we evaluate the confidence of each training sample and use a three-way decisions-based method to identify the trustworthy samples and the noisy samples. Second, we apply the sample correlation between the trustworthy samples and the noisy samples to correct the noisy labels. Finally, we re-weight every sample based on the learned confidences to train the robust LDL model. Experiments show that our approach has better performance in handling noisy data compared to existing algorithms.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":95,"string":"95"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13565,"string":"13,565"}}},{"rowIdx":13566,"cells":{"question_id":{"kind":"string","value":"David E Keyes"},"question":{"kind":"string","value":"Generate an abstract for the title \"Batched QR and SVD Algorithms on GPUs with Applications in Hierarchical Matrix Compression.\" using the following items: 1. GPU \n2. QR decomposition \n3. Singular value decomposition \n4. Batched computations \n5. Matrix compression."},"answer":{"kind":"string","value":"•High performance GPU hosted batched QR decomposition kernels are developed and outperform current implementations for small and rectangular matrices.•Various GPU hosted batched singular value decomposition kernels are developed and used as building blocks of a batched randomized SVD kernel for numerically low rank matrix blocks.•Batched QR, SVD, and GEMM kernels are used to compress hierarchical matrices entirely on the GPU."},"correct_answer":{"kind":"string","value":"•High performance GPU hosted batched QR decomposition kernels are developed and outperform current implementations for small and rectangular matrices.•Various GPU hosted batched singular value decomposition kernels are developed and used as building blocks of a batched randomized SVD kernel for numerically low rank matrix blocks.•Batched QR, SVD, and GEMM kernels are used to compress hierarchical matrices entirely on the GPU."},"documents":{"kind":"list like","value":["Title: Petaflop/s, seriously\nYear: 2007\n\nAbstract: Sustained floating-point rates on real applications, as tracked by the Gordon Bell Prize, have increased by over five orders of magnitude from 1988, when 1 Gigaflop/s was reported on a structural simulation, to 2006, when 200 Teraflop/s were reported on a molecular dynamics simulation. Various versions of Moore's Law over the same interval provide only two to three orders of magnitude of improvement for an individual processor; the remaining factor comes from concurrency, which is of order 100,000 for the BlueGene/L computer, the platform of choice for the majority of recent Bell Prize finalists. As the semiconductor industry begins to slip relative to its own roadmap for silicon-based logic and memory, concurrency will play an increasing role in attaining the next order of magnitude, to arrive at the long-awaited milepost of 1 Petaflop/s sustained on a practical application, which should occur around 2009. Simulations based on Eulerian formulations of partial differential equations can be among the first applications to take advantage of petascale capabilities, but not the way most are presently being pursued. Only weak scaling can get around the fundamental limitation expressed in Amdahl's Law and only optimal implicit formulations can get around another limitation on scaling that is an immediate consequence of Courant-Friedrichs-Lewy stability theory under weak scaling of a PDE. Many PDE-based applications and other lattice-based applications with petascale roadmaps, such as quantum chromodynamics, will likely be forced to adopt optimal implicit solvers. However, even this narrow path to petascale simulation is made treacherous by the imperative of dynamic adaptivity, which drives us to consider algorithms and queuing policies that are less synchronous than those in common use today. Drawing on the SCaLeS report (2003-2004), the latest ITRS roadmap, some back-of-the-envelope estimates, and numerical experiences with PDE-based codes on recently available platforms, we will attempt to project the pathway to Petaflop/s for representative applications.","Title: M06 - Issues for the future of supercomputing: impact of Moore's law and architecture on application performance\nYear: 2006\n\nAbstract: This tutorial will explain technologies driving supercomputer speed increases and supercomputers' resulting ability to solve increasingly important problems. In guided session, participants will walk through future supercomputer performance projection on a representative application.Specific topics will include:A review of scientific problems amenable to supercomputers based on the broad-based SCaLeS study. The editor of the SCaLeS report presents this session.The International Technology Roadmap for Semiconductors (ITRS) and its implications for the size, speed, power, and architecture of supercomputers.Current microprocessor-based and emerging \"advanced\" architectures and their implications on applications performance. There will be special emphasis on the multi-core trend and how applications can use multiple cores effectively.We will discuss the physics issues (speed and power) that will define the end of the current evolutionary trend. We will show how nanotech, reversible logic, and quantum computing may become the basis of a revolutionary change.","Title: Four Horizons for Enhancing the Performance of Parallel Simulations Based on Partial Differential Equations\nYear: 2000\n\nAbstract: Simulations of PDE-based systems, such as flight vehicles, the global climate, petroleum reservoirs, semiconductor devices, and nuclear weapons, typically perform an order of magnitude or more below other scientific simulations (e.g., from chemistry and physics) with dense linear algebra or N-body kernels at their core. In this presentation, we briefly review the algorithmic structure of typical PDE solvers that is responsible for this situation and consider possible architectural and algorithmic sources for performance improvement. Some of these improvements are also applicable to other types of simulations, but we examine their consequences for PDEs: potential to exploit orders of magnitude more processor-memory units, better organization of the simulation for today's and likely near-future hierarchical memories, alternative formulations of the discrete systems to be solved, and new horizons in adaptivity. Each category is motivated by recent experiences in computational aerodynamics at the 1 Teraflop/s scale.","Title: Analyzing the Parallel Scalability of an Implicit Unstructured Mesh CFD Code\nYear: 2000\n\nAbstract: In this paper, we identify the scalability bottlenecks of an unstructured grid CFD code (PETSc-FUN3D) by studying the impact of several algorithmic and architectural parameters and by examining different programming models. We discuss the basic performance characteristics of this PDE code with the help of simple performance models developed in our earlier work, presenting primarily experimental results. In addition to achievingg ood per-processor performance (which has been addressed in our cited work and without which scalability claims are suspect) we strive to improve the implementation and convergence scalability of PETSc-FUN3D on thousands of processors.","Title: Three parallel programming paradigms: comparisons on an archetypal PDE computation\nYear: 1999\n\nAbstract: Three paradigms for distributed-memory parallel computation that free the application programmer from the details of message passing are compared for an archetypal structured scientific computation -- a nonlinear, structured-grid partial differential equation boundary value problem -- using the same algorithm on the same hardware. All of the paradigms -- parallel languages represented by the Portland Group's HPF, (semi-)automated serial-to-parallel source-to-source translation represented by CAP-Tools from the University of Greenwich, and parallel libraries represented by Argonne's PETSc -- are found to be easy to use for this problem class, and all are reasonably effective in exploiting concurrency after a short learning curve. The level of involvement required by the application programmer under any paradigm includes specification of the data partitioning, corresponding to a geometrically simple decomposition of the domain of the PDE. Programming in SPMD style for the PETSc library requires writing only the routines that discretize the PDE and its Jacobian, managing subdomain-to-processor mappings (affine global-to-local index mappings), and interfacing to library solver routines. Programming for HPF requires a complete sequential implementation of the same algorithm as a starting point, introduction of concurrency through subdomain blocking (a task similar to the index mapping), and modest experimentation with rewriting loops to elucidate to the compiler the latent concurrency. Programming with CAPTools involves feeding the same sequential implementation to the CAPTools interactive parallelization system, and guiding the source-to-source code transformation by responding to various queries about quantities knowable only at runtime. Results representative of \"the state of the practice\" for a scaled sequence of structured grid problems are given on three of the most important contemporary high-performance platforms: the IBM SP, the SGI Origin 2000, and the CRAYY T3E.","Title: Parallel Newton--Krylov--Schwarz Algorithms for the Transonic Full Potential Equation\nYear: 1998\n\nAbstract: We study parallel two-level overlapping Schwarz algorithms for solving nonlinear finite element problems, in particular, for the full potential equation of aerodynamics discretized in two dimensions with bilinear elements. The overall algorithm, Newton--Krylov--Schwarz (NKS), employs an inexact finite difference Newton method and a Krylov space iterative method, with a two-level overlapping Schwarz method as a preconditioner. We demonstrate that NKS, combined with a density upwinding continuation strategy for problems with weak shocks, is robust and economical for this class of mixed elliptic-hyperbolic nonlinear partial differential equations, with proper specification of several parameters. We study upwinding parameters, inner convergence tolerance, coarse grid density, subdomain overlap, and the level of fill-in in the incomplete factorization, and report their effect on numerical convergence rate, overall execution time, and parallel efficiency on a distributed-memory parallel computer.","Title: Moving grids for magnetic reconnection via Newton–Krylov methods\nYear: 2011\n\nAbstract: This paper presents a set of computationally efficient, adaptive grids for magnetic reconnection phenomenon where the current density can develop large gradients in the reconnection region. Four-field extended MagnetoHydroDynamics (MHD) equations with hyperviscosity terms are transformed so that the curvilinear coordinates replace the Cartesian coordinates as the independent variables, and moving grids' velocities are also considered in this transformed system as a part of interpolating the physical solutions from the old grid to the new grid as time advances. The curvilinear coordinates derived from the current density through the Monge–Kantorovich (MK) optimization approach help to reduce the resolution requirements during the computation.","Title: High-performacne parallel implicit CFD\nYear: 2001\n\nAbstract: Fluid dynamical simulations based on finite discretizations on (quasi-)static grids scale well in parallel, but execute at a disappointing percentage of per-processor peak floating point operation rates without special attention to layout and access ordering of data. We document both claims from our experience with an unstructured grid CFD code that is typical of the state of the practice at NASA. These basic performance characteristics of PDE-based codes can be understood with surprisingly simple models, for which we quote earlier work, presenting primarily experimental results. The performance models and experimental results motivate algorithmic and software practices that lead to improvements in both parallel scalability and per node performance. This snapshot of ongoing work updates our 1999 Bell Prize-winning simulation on ASCI computers.","Title: Domain decomposition on parallel computers\nYear: 1989\n\nAbstract: We consider the application of domain decomposition techniques to the solution of sparse linear systems arising from implicit PDE discretizations on parallel computers. Representatives of two popular MIMD architectures, message passing (the Intel iPSC/2-SX) and shared memory (the Encore Multimax 320), are employed. We run the same numerical experiments on each, namely stripwise and boxwise decompositions of the unit square, using up to 64 subdomains and containing up to 64K degrees of freedom. We produce a tight-fitting complexity model for the former and discuss the difficulty of doing so for the latter. We also evaluate which of three types of domain decomposition preconditioners that have appeared in the literature of self-adjoint elliptic problems are most efficient in different regions of machine-problem parameter space. Some form of global sharing of information in the preconditioner is required for efficient overall parallel implementation in the region of most practical interest (large problem sizes and large numbers of processors); otherwise, an increasing iteration count inveighs against the gains of concurrency. Our results on a per iteration basis also hold for sparse discrete systems arising from other types of partial differential equations, but in the absence of a theory for the dependence of the convergence rate upon the granularity of the decomposition, the overall results are only suggestive for more general systems.","Title: Multiphysics simulations: Challenges and opportunities.\nYear: 2013\n\nAbstract: We consider multiphysics applications from algorithmic and architectural perspectives, where â聙聹algorithmicâ聙聺 includes both mathematical analysis and computational complexity, and â聙聹architecturalâ聙聺 includes both software and hardware environments. Many diverse multiphysics applications can be reduced, en route to their computational simulation, to a common algebraic coupling paradigm. Mathematical analysis of multiphysics coupling in this form is not always practical for realistic applications, but model problems representative of applications discussed herein can provide insight. A variety of software frameworks for multiphysics applications have been constructed and refined within disciplinary communities and executed on leading-edge computer systems. We examine several of these, expose some commonalities among them, and attempt to extrapolate best practices to future systems. From our study, we summarize challenges and forecast opportunities.","Title: Numerical simulation of four-field extended magnetohydrodynamics in dynamically adaptive curvilinear coordinates via Newton-Krylov-Schwarz\nYear: 2012\n\nAbstract: Numerical simulations of the four-field extended magnetohydrodynamics (MHD) equations with hyper-resistivity terms present a difficult challenge because of demanding spatial resolution requirements. A time-dependent sequence of r-refinement adaptive grids obtained from solving a single Monge-Ampere (MA) equation addresses the high-resolution requirements near the x-point for numerical simulation of the magnetic reconnection problem. The MHD equations are transformed from Cartesian coordinates to solution-defined curvilinear coordinates. After the application of an implicit scheme to the time-dependent problem, the parallel Newton-Krylov-Schwarz (NKS) algorithm is used to solve the system at each time step. Convergence and accuracy studies show that the curvilinear solution requires less computational effort than a pure Cartesian treatment. This is due both to the more optimal placement of the grid points and to the improved convergence of the implicit solver, nonlinearly and linearly. The latter effect, which is significant (more than an order of magnitude in number of inner linear iterations for equivalent accuracy), does not yet seem to be widely appreciated.","Title: Optimizing the performance of streaming numerical kernels on the IBM Blue Gene/P PowerPC 450 processor\nYear: 2013\n\nAbstract: Several emerging petascale architectures use energy-efficient processors with vectorized computational units and in-order thread processing. On these architectures the sustained performance of streaming numerical kernels, ubiquitous in the solution of partial differential equations, represents a challenge despite the regularity of memory access. Sophisticated optimization techniques are required to fully utilize the CPU. We propose a new method for constructing streaming numerical kernels using a high-level assembly synthesis and optimization framework. We describe an implementation of this method in Python targeting the IBM脗庐 Blue Gene脗庐/P supercomputer's PowerPC脗庐 450 core. This paper details the high-level design, construction, simulation, verification, and analysis of these kernels utilizing a subset of the CPU's instruction set. We demonstrate the effectiveness of our approach by implementing several three-dimensional stencil kernels over a variety of cached memory scenarios and analyzing the mechanically scheduled variants, including a 27-point stencil achieving a 1.7脙聴 speedup over the best previously published results.","Title: Optimizing Memory-Bound Symv Kernel On Gpu Hardware Accelerators\nYear: 2012\n\nAbstract: Hardware accelerators are becoming ubiquitous high performance scientific computing. They are capable of delivering an unprecedented level of concurrent execution contexts. High-level programming language extensions (e. g., CUDA), profiling tools (e. g., PAPI-CUDA, CUDA Profiler) are paramount to improve productivity, while effectively exploiting the underlying hardware. We present an optimized numerical kernel for computing the symmetric matrix-vector product on nVidia Fermi GPUs. Due to its inherent memory-bound nature, this kernel is very critical in the tridiagonalization of a symmetric dense matrix, which is a preprocessing step to calculate the eigenpairs. Using a novel design to address the irregular memory accesses by hiding latency and increasing bandwidth, our preliminary asymptotic results show 3.5x and 2.5x fold speedups over the similar CUBLAS 4.0 kernel, and 7-8% and 30% fold improvement over the Matrix Algebra on GPU and Multicore Architectures (MAGMA) library in single and double precision arithmetics, respectively.","Title: Multi-core issues - Multi-Core for HPC: breakthrough or breakdown?\nYear: 2006\n\nAbstract: A dramatic trend in computing is the adoption of multi-core technology by the vendors from which our current and future HPC systems are being derived. Multi-core is offered as a path to continued reliance and benefits of Moore's Law while reining in the previously unfettered growth of power consumption and design complexity. Are we saved? or is it but a fools mission, trapping us in a technical cul de sac with no long term direction and no way to reinvent an alternative future. The panel will consider the following questions:* Can multi-core span the next decade of Moore's Law progression?* Are the pins and caches a strangle hold on the future effectiveness of multi-core?* Can innovative algorithmic techniques exploit the opportunities and address the challenges of multi-core?* How will programming models and supporting system software change to accommodate the unique properties and peculiarities of multi-core structures?","Title: Towards energy efficiency and maximum computational intensity for stencil algorithms using wavefront diamond temporal blocking.\nYear: 2014\n\nAbstract: We study the impact of tunable parameters on computational intensity (i.e., inverse code balance) and energy consumption of multicore-optimized wavefront diamond temporal blocking (MWD) applied to different stencil-based update schemes. MWD combines the concepts of diamond tiling and multicore-aware wavefront blocking in order to achieve lower cache size requirements than standard single-core wavefront temporal blocking. We analyze the impact of the cache block size on the theoretical and observed code balance, introduce loop tiling in the leading dimension to widen the range of applicable diamond sizes, and show performance results on a contemporary Intel CPU. The impact of code balance on power dissipation on the CPU and in the DRAM is investigated and shows that DRAM power is a decisive factor for energy consumption, which is strongly influenced by the code balance. Furthermore we show that highest performance does not necessarily lead to lowest energy even if the clock speed is fixed.","Title: KBLAS: An Optimized Library for Dense Matrix-Vector Multiplication on GPU Accelerators.\nYear: 2014\n\nAbstract: KBLAS is an open-source, high-performance library that provides optimized kernels for a subset of Level 2 BLAS functionalities on CUDA-enabled GPUs. Since performance of dense matrix-vector multiplication is hindered by the overhead of memory accesses, a double-buffering optimization technique is employed to overlap data motion with computation. After identifying a proper set of tuning parameters, KBLAS efficiently runs on various GPU architectures while avoiding code rewriting and retaining compliance with the standard BLAS API. Another optimization technique allows ensuring coalesced memory access when dealing with submatrices, especially for high-level dense linear algebra algorithms. All KBLAS kernels have been leveraged to a multi-GPU environment, which requires the introduction of new APIs. Considering general matrices, KBLAS is very competitive with existing state-of-the-art kernels and provides a smoother performance across a wide range of matrix dimensions. Considering symmetric and Hermitian matrices, the KBLAS performance outperforms existing state-of-the-art implementations on all matrix sizes and achieves asymptotically up to 50% and 60% speedup against the best competitor on single GPU and multi-GPUs systems, respectively. Performance results also validate our performance model. A subset of KBLAS high-performance kernels have been integrated into NVIDIA's standard BLAS implementation (cuBLAS) for larger dissemination, starting from version 6.0.","Title: Smooth and robust solutions for Dirichlet boundary control of fluid-solid conjugate heat transfer problems.\nYear: 2015\n\nAbstract: We study a new optimization scheme that generates smooth and robust solutions for Dirichlet velocity boundary control (DVBC) of conjugate heat transfer (CHT) processes. The solutions to the DVBC of the incompressible Navier–Stokes equations are typically nonsmooth, due to the regularity degradation of the boundary stress in the adjoint Navier–Stokes equations. This nonsmoothness is inherited by the solutions to the DVBC of CHT processes, since the CHT process couples the Navier–Stokes equations of fluid motion with the convection–diffusion equations of fluid–solid thermal interaction. Our objective in the CHT boundary control problem is to select optimally the fluid inflow profile that minimizes an objective function that involves the sum of the mismatch between the temperature distribution in the fluid system and a prescribed temperature profile and the cost of the control.","Title: High Performance Pseudo-analytical Simulation of Multi-Object Adaptive Optics over Multi-GPU Systems.\nYear: 2014\n\nAbstract: Multi-object adaptive optics (MOAO) is a novel adaptive optics (AO) technique dedicated to the special case of wide-field multi-object spectrographs (MOS). It applies dedicated wavefront corrections to numerous independent tiny patches spread over a large field of view (FOV). The control of each deformable mirror (DM) is done individually using a tomographic reconstruction of the phase based on measurements from a number of wavefront sensors (WFS) pointing at natural and artificial guide stars in the field. The output of this study helps the design of a new instrument called MOSAIC, a multi-object spectrograph proposed for the European Extremely Large Telescope (E-ELT)(1). We have developed a novel hybrid pseudo-analytical simulation scheme that allows us to accurately simulate in detail the tomographic problem. The main challenge resides in the computation of the tomographic reconstructor, which involves pseudo-inversion of a large dense symmetric matrix. The pseudo-inverse is computed using an eigenvalue decomposition, based on the divide and conquer algorithm, on multicore systems with multi-GPUs. Thanks to a new symmetric matrix-vector product (SYMV) multi-GPU kernel, our overall implementation scores significant speedups over standard numerical libraries on multicore, like Intel MKL, and up to 60% speedups over the standard MAGMA implementation on 8 Kepler K20c GPUs. At 40,000 unknowns, this appears to be the largest-scale tomographic AO matrix solver submitted to computation, to date, to our knowledge and opens new research directions for extreme scale AO simulations.","Title: Performance Modeling and Tuning of an Unstructured Mesh CFD Application\nYear: 2000\n\nAbstract: This paper describes performance tuning experiences with a three-dimensional unstructured grid Euler flow code from NASA, which we have reimplemented in the PETSc framework and ported to several large-scale machines, including the ASCI Red and Blue Pacific machines, the SGI Origin, the Cray T3E, and Beowulf clusters. The code achieves a respectable level of performance for sparse problems, typical of scientific and engineering codes based on partial differential equations, and scales well up to thousands of processors. Since the gap between CPU speed and memory access rate is widening, the code is analyzed from a memory-centric perspective (in contrast to traditional flop-orientation) to understand its sequential and parallel performance. Performance tuning is approached on three fronts: data layouts to enhance locality of reference, algorithmic parameters, and parallel programming model. This effort was guided partly by some simple performance models developed for the sparse matrix-vector product operation.","Title: Multicore-Optimized Wavefront Diamond Blocking for Optimizing Stencil Updates.\nYear: 2015\n\nAbstract: The importance of stencil-based algorithms in computational science has focused attention on optimized parallel implementations for multilevel cache-based processors. Temporal blocking schemes leverage the large bandwidth and low latency of caches to accelerate stencil updates and approach theoretical peak performance. A key ingredient is the reduction of data traffic across slow data paths, especially the main memory interface. In this work we combine the ideas of multicore wavefront temporal blocking and diamond tiling to arrive at stencil update schemes that show large reductions in memory pressure compared to existing approaches. The resulting schemes show performance advantages in bandwidth-starved situations, which are exacerbated by the high bytes per lattice update case of variable coefficients. Our thread groups concept provides a controllable trade-off between concurrency and memory usage, shifting the pressure between the memory interface and the CPU. We present performance results on a contemporary Intel processor.","Title: Nonlinearly Preconditioned Inexact Newton Algorithms\nYear: 2002\n\nAbstract: Inexact Newton algorithms are commonly used for solving large sparse nonlinear system of equations $F(u^{\\ast})=0$ arising, for example, from the discretization of partial differential equations. Even with global strategies such as linesearch or trust region, the methods often stagnate at local minima of $\\|F\\|$, especially for problems with unbalanced nonlinearities, because the methods do not have built-in machinery to deal with the unbalanced nonlinearities. To find the same solution $u^{\\ast}$, one may want to solve instead an equivalent nonlinearly preconditioned system ${\\cal F}(u^{\\ast})=0$ whose nonlinearities are more balanced. In this paper, we propose and study a nonlinear additive Schwarz-based parallel nonlinear preconditioner and show numerically that the new method converges well even for some difficult problems, such as high Reynolds number flows, where a traditional inexact Newton method fails.","Title: Towards polyalgorithmic linear system solvers for nonlinear elliptic problems\nYear: 1994\n\nAbstract: The authors investigate the performance of several preconditioned conjugate gradient-like algorithms and a standard stationary iterative method (block-line successive overrelaxation (SOR)) on linear systems of equations that arise from a nonlinear elliptic flame sheet problem simulation. The nonlinearity forces a pseudotransient continuation process that makes the problem parabolic and thus compacts the spectrum of the Jacobian matrix so that simple relaxation methods are viable in the initial stages of the solution process. However, because of the transition from parabolic to elliptic character as the timestep is increased in pursuit of the steady-state solution, the performance of the candidate linear solvers spreads as the domain of convergence of Newton's method is approached. In numerical experiments over the course of a full nonlinear solution trajectory, short recurrence or optimal Krylov algorithms combined with a Gauss-Seidel (GS) preconditioning yield better execution times with respect to the standard block-line SOR techniques, but SOR performs competitively at a smaller storage cost until the final stages. Block-incomplete factorization preconditioned methods, on the other hand, require nearly a factor of two more storage than SOR and are uniformly less effective during the pseudotransient stages. The advantage of GS preconditioning is partly attributable to die exploitation of a dominant convection direction in the examples; nevertheless, a multidomain version of GS with streamwise coupling lagged at rows between adjacent subdomains incurs only a modest penalty.","Title: Field-Split Preconditioned Inexact Newton Algorithms\nYear: 2015\n\nAbstract: The multiplicative Schwarz preconditioned inexact Newton (MSPIN) algorithm is presented as a complement to additive Schwarz preconditioned inexact Newton (ASPIN). At an algebraic level, ASPIN and MSPIN are variants of the same strategy to improve the convergence of systems with unbalanced nonlinearities; however, they have natural complementarity in practice. MSPIN is naturally based on partitioning of degrees of freedom in a nonlinear PDE system by field type rather than by subdomain, where a modest factor of concurrency can be sacrificed for physically motivated convergence robustness. ASPIN, originally introduced for decompositions into subdomains, is natural for high concurrency and reduction of global synchronization. We consider both types of inexact Newton algorithms in the field-split context, and we augment the classical convergence theory of ASPIN for the multiplicative case. Numerical experiments show that MSPIN can be significantly more robust than Newton methods based on global linearizations, and that MSPIN can be more robust than ASPIN and maintain fast convergence even for challenging problems, such as high Reynolds number Navier-Stokes equations.","Title: Parallel performance of domain-decomposed preconditioned Krylov methods for PDEs with locally uniform refinement\nYear: 1992\n\nAbstract: Preconditioners based on domain decomposition appear natural for the Krylov solution of implicitly discretized partial differential equations (PDEs) on parallel computers. Two-scale preconditioners (involving a global coarse-grid solve, independent solves over interfaces connecting the coarse-grid points, and independent subdomain solves) have been known since the early 1980s to be \"near optimal\" in the sense of ensuring a bounded, or at most logarithmically growing, iteration count as the mesh is refined. As a result, the refinement of the mesh can be chosen locally on the basis of truncation error, and the granularity of the domain decomposition can be chosen globally on the basis of parallel computing considerations with only mild effects on the convergence rate of the algorithm. However, overall computational complexity depends not only on the algebraic convergence rate, but also on the operation counts of the components of the preconditioner that must be applied at each iteration. The costs of solving the subdomain systems and the crosspoint system show superlinear growth in their respective (and inversely related) sizes. On the subdomains, the superlinear terms arise from arithmetic only; in the crosspoint system the cost of nonlocal data exchange is also superlinear. These factors make the preconditioner granularity and the choice of its components problem- and machine-dependent compromises. The tradeoffs involved are illustrated through numerical experiments on both shared- and distributed-memory computers for convection-diffusion problems. Because of the development of boundary layers, these problems benefit from local mesh refinement, which is straightforward to accommodate within the domain decomposition framework in a locally uniform sense, but which introduces load balancing as a further consideration in selecting the granularity of the preconditioner. In spite of the tradeoffs, cumulative speedups are obtainable out to at least medium-scale granularity (up to 64 processors in our tests). The largest problems involve O(10(5)) unknowns partitioned into O(10(3)) subdomains and converge in O(10) iterations requiring O(1) seconds on the Intel iPSC/860.","Title: Domain decomposition with local mesh refinement\nYear: 1992\n\nAbstract: A preconditioned Krylov iterative algorithm based on domain decomposition for linear systems arising from implicit finite-difference or finite-element discretizations of partial differential equation problems requiring local mesh refinement is described. To keep data structures as simple as possible for parallel computing applications, the fundamental computational unit in the algorithm is defined as a subregion of the domain spanned by a locally uniform tensor-product grid, called a tile. In the tile-based domain decomposition approach, two levels of discretization are considered at each point of the domain: a global coarse grid defined by tile vertices only, and a local fine grid where the degree of resolution can vary from tile to tile. One global level and one local level provide the flexibility required to adaptively discretize a diverse collection of problems on irregular regions and solve them at convergence rates that deteriorate only logarithmically in the finest mesh parameter, with the coarse tessellation held fixed. A logarithmic departure from optimality seems to be a reasonable compromise for the simplicity of the composite grid data structure and concomitant regular data exchange patterns in a multiprocessor environment. Some experiments with up to 1024 tiles are reported, and the evolution of the algorithm is commented on and contrasted with optimal nonrefining two-level algorithms and optimal refining multilevel algorithms. Computational comparisons with some other popular methods are presented.","Title: Multi-dimensional intra-tile parallelization for memory-starved stencil computations\nYear: 2015\n\nAbstract: Optimizing the performance of stencil algorithms has been the subject of intense research over the last two decades. Since many stencil schemes have low arithmetic intensity, most optimizations focus on increasing the temporal data access locality, thus reducing the data traffic through the main memory interface with the ultimate goal of decoupling from this bottleneck. There are, however, only a few approaches that explicitly leverage the shared cache feature of modern multicore chips. If every thread works on its private, separate cache block, the available cache space can become too small, and sufficient temporal locality may not be achieved. We propose a flexible multidimensional intratile parallelization method for stencil algorithms on multicore CPUs with a shared outer-level cache. This method leads to a significant reduction in the required cache space without adverse effects from hardware prefetching or TLB shortage. Our Girih framework includes an autotuner to select optimal parameter configurations on the target hardware. We conduct performance experiments on two contemporary Intel processors and compare with the state-of-the-art stencil frameworks Pluto and Pochoir, using four corner-case stencil schemes and a wide range of problem sizes. Girih shows substantial performance advantages and best arithmetic intensity at almost all problem sizes, especially on low-intensity stencils with variable coefficients. We study in detail the performance behavior at varying grid sizes using phenomenological performance modeling. Our analysis of energy consumption reveals that our method can save energy through reduced DRAM bandwidth usage even at a marginal performance gain. It is thus well suited for future architectures that will be strongly challenged by the cost of data movement, be it in terms of performance or energy consumption.","Title: Optimization of an Electromagnetics Code with Multicore Wavefront Diamond Blocking and Multi-dimensional Intra-Tile Parallelization\nYear: 2015\n\nAbstract: Understanding and optimizing the properties of solar cells is becoming a key issue in the search for alternatives to nuclear and fossil energy sources. A theoretical analysis via numerical simulations involves solving Maxwell's Equations in discretized form and typically requires substantial computing effort. We start from a hybrid-parallel (MPI+OpenMP) production code that implements the Time Harmonic Inverse Iteration Method (THIIM) with Finite-Difference Frequency Domain (FDFD) discretization. Although this algorithm has the characteristics of a strongly bandwidth-bound stencil update scheme, it is significantly different from the popular stencil types that have been exhaustively studied in the high performance computing literature to date. We apply a recently developed stencil optimization technique, multicore wavefront diamond tiling with multi-dimensional cache block sharing, and describe in detail the peculiarities that need to be considered due to the special stencil structure. Concurrency in updating the components of the electric and magnetic fields provides an additional level of parallelism. The dependence of the cache size requirement of the optimized code on the blocking parameters is modeled accurately, and an auto-tuner searches for optimal configurations in the remaining parameter space. We were able to completely decouple the execution from the memory bandwidth bottleneck, accelerating the implementation by a factor of three to four compared to an optimal implementation with pure spatial blocking on an 18-core Intel Haswell CPU.","Title: A Comparison Of Some Domain Decomposition And Ilu Preconditioned Iterative Methods For Nonsymmetric Elliptic Problems\nYear: 1994\n\nAbstract: In recent years, competitive domain-decomposed preconditioned iterative techniques of Krylov-Schwarz type have been developed for nonsymmetric linear elliptic systems. Such systems arise when convection-diffusion-reaction problems from computational fluid dynamics or heat and mass transfer are linearized for iterative solution. Through domain decomposition, a large problem is divided into many smaller problems whose requirements for coordination can be controlled to allow effective solution on parallel machines. A central question is how to choose these small problems and how to arrange the order of their solution. Different specifications of decomposition and solution order lead to a plethora of algorithms possessing complementary advantages and disadvantages. In this report we compare several methods, including the additive Schwarz algorithm, the classical multiplicative Schwarz algorithm, an accelerated multiplicative Schwarz algorithm, the tile algorithm, the CGK algorithm, the CSPD algorithm, and also the popular global ILU-family of preconditioners, on some nonsymmetric or indefinite elliptic model problems discretized by finite difference methods. The preconditioned problems are solved by the unrestarted GMRES method. A version of the accelerated multiplicative Schwarz method is a consistently good performer.","Title: A scalable community detection algorithm for large graphs using stochastic block models\nYear: 2015\n\nAbstract: Community detection in graphs is widely used in social and biological networks, and the stochastic block model is a powerful probabilistic tool for describing graphs with community structures. However, in the era of \\\"big data,\\\" traditional inference algorithms for such a model are increasingly limited due to their high time complexity and poor scalability. In this paper, we propose a multi-stage maximum likelihood approach to recover the latent parameters of the stochastic block model, in time linear with respect to the number of edges. We also propose a parallel algorithm based on message passing. Our algorithm can overlap communication and computation, providing speedup without compromising accuracy as the number of processors grows. For example, to process a real-world graph with about 1.3 million nodes and 10 million edges, our algorithm requires about 6 seconds on 64 cores of a contemporary commodity Linux cluster. Experiments demonstrate that the algorithm can produce high quality results on both benchmark and real-world graphs. An example of finding more meaningful communities is illustrated consequently in comparison with a popular modularity maximization algorithm.","Title: Exploring Shared-Memory Optimizations for an Unstructured Mesh CFD Application on Modern Parallel Systems\nYear: 2015\n\nAbstract: In this work, we revisit the 1999 Gordon Bell Prize winning PETSc-FUN3D aerodynamics code, extending it with highly-tuned shared-memory parallelization and detailed performance analysis on modern highly parallel architectures. An unstructured-grid implicit flow solver, which forms the backbone of computational aerodynamics, poses particular challenges due to its large irregular working sets, unstructured memory accesses, and variable/limited amount of parallelism. This code, based on a domain decomposition approach, exposes tradeoffs between the number of threads assigned to each MPI-rank sub domain, and the total number of domains. By applying several algorithm- and architecture-aware optimization techniques for unstructured grids, we show a 6.9X speed-up in performance on a single-node Intel® XeonTM1 E5 2690 v2 processor relative to the out-of-the-box compilation. Our scaling studies on TACC Stampede supercomputer show that our optimizations continue to provide performance benefits over baseline implementation as we scale up to 256 nodes.","Title: High Performance Multi-GPU SpMV for Multi-component PDE-Based Applications.\nYear: 2015\n\nAbstract: Leveraging optimization techniques (e.g., register blocking and double buffering) introduced in the context of KBLAS, a Level 2 BLAS high performance library on GPUs, the authors implement dense matrix-vector multiplications within a sparse-block structure. While these optimizations are important for high performance dense kernel executions, they are even more critical when dealing with sparse linear algebra operations. The most time-consuming phase of many multicomponent applications, such as models of reacting flows or petroleum reservoirs, is the solution at each implicit time step of large, sparse spatially structured or unstructured linear systems. The standard method is a preconditioned Krylov solver. The Sparse Matrix-Vector multiplication (SpMV) is, in turn, one of the most time-consuming operations in such solvers. Because there is no data reuse of the elements of the matrix within a single SpMV, kernel performance is limited by the speed at which data can be transferred from memory to registers, making the bus bandwidth the major bottleneck. On the other hand, in case of a multi-species model, the resulting Jacobian has a dense block structure. For contemporary petroleum reservoir simulations, the block size typically ranges from three to a few dozen among different models, and still larger blocks are relevant within adaptively model-refined regions of the domain, though generally the size of the blocks, related to the number of conserved species, is constant over large regions within a given model. This structure can be exploited beyond the convenience of a block compressed row data format, because it offers opportunities to hide the data motion with useful computations. The new SpMV kernel outperforms existing state-of-the-art implementations on single and multi-GPUs using matrices with dense block structure representative of porous media applications with both structured and unstructured multi-component grids.","Title: Performance optimization of Sparse Matrix-Vector Multiplication for multi-component PDE-based applications using GPUs.\nYear: 2016\n\nAbstract: Simulations of many multi-component PDE-based applications, such as petroleum reservoirs or reacting flows, are dominated by the solution, on each time step and within each Newton step, of large sparse linear systems. The standard solver is a preconditioned Krylov method. Along with application of the preconditioner, memory-bound Sparse Matrix-Vector Multiplication SpMV is the most time-consuming operation in such solvers. Multi-species models produce Jacobians with a dense block structure, where the block size can be as large as a few dozen. Failing to exploit this dense block structure vastly underutilizes hardware capable of delivering high performance on dense BLAS operations. This paper presents a GPU-accelerated SpMV kernel for block-sparse matrices. Dense matrix-vector multiplications within the sparse-block structure leverage optimization techniques from the KBLAS library, a high performance library for dense BLAS kernels. The design ideas of KBLAS can be applied to block-sparse matrices. Furthermore, a technique is proposed to balance the workload among thread blocks when there are large variations in the lengths of nonzero rows. Multi-GPU performance is highlighted. The proposed SpMV kernel outperforms existing state-of-the-art implementations using matrices with real structures from different applications. Copyright © 2016 John Wiley & Sons, Ltd.","Title: Efficient Sphere Detector Algorithm for Massive MIMO using GPU Hardware Accelerator.\nYear: 2016\n\nAbstract: To further enhance the capacity of next generation wireless communication systems, massive multiple-input multiple-output (MIMO) has recently appeared as a necessary enabling technology to achieve high performance signal processing for large-scale multiple antennas. However, massive MIMO systems inevitably generate signal processing overheads, which translate into ever-increasing rate of complexity, and therefore, such systems may not maintain the inherent real-time requirement of wireless systems. We redesign the nonlinear sphere decoder method to increase the performance of the system, cast most memory-bound computations into compute-bound operations to reduce the overall complexity, and maintain the real-time processing thanks to the graphics processing unit (GPU) computational power. We show a comprehensive complexity and performance analysis on an unprecedented MIMO system scale, which can ease the design phase toward simulating future massive MIMO wireless systems.","Title: High Performance Polar Decomposition on Distributed Memory Systems.\nYear: 2016\n\nAbstract: The polar decomposition of a dense matrix is an important operation in linear algebra. It can be directly calculated through the singular value decomposition SVD or iteratively using the QR dynamically-weighted Halley algorithm QDWH. The former is difficult to parallelize due to the preponderant number of memory-bound operations during the bidiagonal reduction. We investigate the latter scenario, which performs more floating-point operations but exposes at the same time more parallelism, and therefore, runs closer to the theoretical peak performance of the system, thanks to more compute-bound matrix operations. Profiling results show the performance scalability of QDWH for calculating the polar decomposition using around 9200 MPI processes on well and ill-conditioned matrices of 100ï¾źKï¾ź$$\\\\times $$ï¾ź100ï¾źK problem size. We study then the performance impact of the QDWH-based polar decomposition as a pre-processing step toward calculating the SVD itself. The new distributed-memory implementation of the QDWH-SVD solver achieves up to five-fold speedup against current state-of-the-art vendor SVD implementations.","Title: Redesigning Triangular Dense Matrix Computations on GPUs.\nYear: 2016\n\nAbstract: A new implementation of the triangular matrix-matrix multiplication TRMM and the triangular solve TRSM kernels are described on GPU hardware accelerators. Although part of the Level 3 BLAS family, these highly computationally intensive kernels fail to achieve the percentage of the theoretical peak performance on GPUs that one would expect when running kernels with similar surface-to-volume ratio on hardware accelerators, i.e., the standard matrix-matrix multiplication GEMM. The authors propose adopting a recursive formulation, which enriches the TRMM and TRSM inner structures with GEMM calls and, therefore, reduces memory traffic while increasing the level of concurrency. The new implementation enables efficient use of the GPU memory hierarchy and mitigates the latency overhead, to run at the speed of the higher cache levels. Performance comparisons show upï¾źto eightfold and twofold speedups for large dense matrix sizes, against the existing state-of-the-art TRMM and TRSM implementations from NVIDIA cuBLAS, respectively, across various GPU generations. Once integrated into high-level Cholesky-based dense linear algebra algorithms, the performance impact on the overall applications demonstrates upï¾źto fourfold and twofold speedups, against the equivalent native implementations, linked with cuBLAS TRMM and TRSM kernels, respectively. The new TRMM/TRSM kernel implementations are part of the open-source KBLAS software library http://ecrc.kaust.edu.sa/Pages/Res-kblas.aspx and are lined up for integration into the NVIDIA cuBLAS library in the upcoming v8.0 release.","Title: Research and Education in Computational Science and Engineering.\nYear: 2018\n\nAbstract: This report presents challenges, opportunities, and directions for computational science and engineering (CSE) research and education for the next decade. Over the past two decades the field of CSE has penetrated both basic and applied research in academia, industry, and laboratories to advance discovery, optimize systems, support decision-makers, and educate the scientific and engineering workforce. Informed by centuries of theory and experiment, CSE performs computational experiments to answer questions that neither theory nor experiment, alone is equipped to answer. CSE provides scientists and engineers with algorithmic inventions and software systems that, transcend disciplines and scales. CSE brings the power of parallelism to bear on troves of data. Mathematics-based advanced computing has become a prevalent means of discovery and innovation in essentially all areas of science, engineering, technology, and society, and the CSE community is at the core of this transformation. However, a combination of disruptive developments-including the architectural complexity of extreme-scale computing, the data revolution and increased attention to data-driven discovery, and the specialization required to follow the applications to new frontiers-is redefining the scope and reach of the CSE endeavor. With these many current and expanding opportunities for the CSE field, there is a growing demand for CSE graduates and a need to expand CSE educational offerings. This need includes CSE programs at both the undergraduate and graduate levels, as well as continuing education and professional development programs, exploiting the synergy between computational science and data science. Yet, as institutions consider new and evolving educational programs, it is essential to consider the broader research challenges and opportunities that provide the context for CSE education and workforce development.","Title: Convergence Analysis for the Multiplicative Schwarz Preconditioned Inexact Newton Algorithm.\nYear: 2016\n\nAbstract: The multiplicative Schwarz preconditioned inexact Newton (MSPIN) algorithm, based on decomposition by field type rather than by subdomain, was recently introduced to improve the convergence of systems with unbalanced nonlinearities. This paper provides a convergence analysis of the MSPIN algorithm. Under reasonable assumptions, it is shown that MSPIN is locally convergent, and desired superlinear or even quadratic convergence can be obtained when the forcing terms are picked suitably.","Title: Unstructured computational aerodynamics on many integrated core architecture.\nYear: 2016\n\nAbstract: Porting implicit unstructured mesh code onto Intel Xeon Phi Knights Corner.Employing aggressive optimizations to improve the CFD flux kernel on Phi hardware.Exploring different thread affinity modes with different programming paradigms.Achieving 3.8x speedup using the offload mode compare to the baseline.Achieving 5x speedup using the native mode compare to the baseline. Shared memory parallelization of the flux kernel of PETSc-FUN3D, an unstructured tetrahedral mesh Euler flow code previously studied for distributed memory and multi-core shared memory, is evaluated on up to 61 cores per node and up to 4 threads per core. We explore several thread-level optimizations to improve flux kernel performance on the state-of-the-art many integrated core (MIC) Intel processor Xeon Phi Knights Corner, with a focus on strong thread scaling. While the linear algebraic kernel is bottlenecked by memory bandwidth for even modest numbers of cores sharing a common memory, the flux kernel, which arises in the control volume discretization of the conservation law residuals and in the formation of the preconditioner for the Jacobian by finite-differencing the conservation law residuals, is compute-intensive and is known to exploit effectively contemporary multi-core hardware. We extend study of the performance of the flux kernel to the Xeon Phi in three thread affinity modes, namely scatter, compact, and balanced, in both offload and native mode, with and without various code optimizations to improve alignment and reduce cache coherency penalties. Relative to baseline out-of-the-box optimized compilation, code restructuring optimizations provide about 3.8x speedup using the offload mode and about 5x speedup using the native mode. Even with these gains for the flux kernel, with respect to execution time the MIC simply achieves par with optimized compilation on a contemporary multi-core Intel CPU, the 16-core Sandy Bridge E5 2670. Nevertheless, the optimizations employed to reduce the data motion and cache coherency protocol penalties of the MIC are expected to be of value for CFD and many other unstructured applications as many-core architecture evolves. We explore large-scale distributed-shared memory performance on the Cray XC40 supercomputer, to demonstrate that optimizations employed on Phi hybridize to this context, where each of thousands of nodes are comprised of two sockets of Intel Xeon Haswell CPUs with 32 cores per node.","Title: A framework for dense triangular matrix kernels on various manycore architectures.\nYear: 2017\n\nAbstract: We present a new high-performance framework for dense triangular Basic Linear Algebra Subroutines (BLAS) kernels, ie, triangular matrix-matrix multiplication (TRMM) and triangular solve (TRSM), on various manycore architectures. This is an extension of a previous work on a single GPU by the same authors, presented at the EuroPar'16 conference, in which we demonstrated the effectiveness of recursive formulations in enhancing the performance of these kernels. In this paper, the performance of triangular BLAS kernels on a single GPU is further enhanced by implementing customized in-place CUDA kernels for TRMM and TRSM, which are called at the bottom of the recursion. In addition, a multi-GPU implementation of TRMM and TRSM is proposed and we show an almost linear performance scaling, as the number of GPUs increases. Finally, the algorithmic recursive formulation of these triangular BLAS kernels is in fact oblivious to the targeted hardware architecture. We, therefore, port these recursive kernels to homogeneous x86 hardware architectures by relying on the vendor optimized BLAS implementations. Results reported on various hardware architectures highlight a significant performance improvement against state-of-the-art implementations. These new kernels are freely available in the KAUST BLAS (KBLAS) open-source library at .","Title: On the Robustness and Prospects of Adaptive BDDC Methods for Finite Element Discretizations of Elliptic PDEs with High-Contrast Coefficients.\nYear: 2016\n\nAbstract: Balancing Domain Decomposition by Constraints (BDDC) methods have proven to be powerful preconditioners for large and sparse linear systems arising from the finite element discretization of elliptic PDEs. Condition number bounds can be theoretically established that are independent of the number of subdomains of the decomposition. The core of the methods resides in the design of a larger and partially discontinuous finite element space that allows for fast application of the preconditioner, where Cholesky factorizations of the subdomain finite element problems are additively combined with a coarse, global solver. Multilevel and highly-scalable algorithms can be obtained by replacing the coarse Cholesky solver with a coarse BDDC preconditioner. BDDC methods have the remarkable ability to control the condition number, since the coarse space of the preconditioner can be adaptively enriched at the cost of solving local eigenproblems. The proper identification of these eigenproblems extends the robustness of the methods to any heterogeneity in the distribution of the coefficients of the PDEs, not only when the coefficients jumps align with the subdomain boundaries or when the high contrast regions are confined to lie in the interior of the subdomains. The specific adaptive technique considered in this paper does not depend upon any interaction of discretization and partition; it relies purely on algebraic operations. Coarse space adaptation in BDDC methods has attractive algorithmic properties, since the technique enhances the concurrency and the arithmetic intensity of the preconditioning step of the sparse implicit solver with the aim of controlling the number of iterations of the Krylov method in a black-box fashion, thus reducing the number of global synchronization steps and matrix vector multiplications needed by the iterative solver; data movement and memory bound kernels in the solve phase can be thus limited at the expense of extra local flops during the setup of the preconditioner. This paper presents an exposition of the BDDC algorithm that identifies the current computational bottlenecks that could prevent it from being competitive with other solvers, and proposes solutions in anticipation of exascale architectures. Furthermore, the discussion aims to give interested practitioners sufficient insights to decide whether or not to pursue BDDC in their applications. In addition, the work presents novel numerical results using the distributed memory implementation of BDDC in the PETSc library for vector field problems arising in the context of porous media flows and electromagnetic modeling; the results provide evidence of the robustness of these methods for highly heterogenous problems and non-conforming discretizations.","Title: ExaGeoStat: A High Performance Unified Framework for Geostatistics on Manycore Systems.\nYear: 2017\n\nAbstract: We present ExaGeoStat , a high performance software for geospatial statistics in climate and environment modeling. In contrast to simulation based on partial differential equations derived from first-principles modeling, ExaGeoStat employs a statistical model based on the evaluation of the Gaussian log-likelihood function, which operates on a large dense covariance matrix. Generated by the parametrizable Matern covariance function, the resulting matrix is symmetric and positive definite. The computational tasks involved during the evaluation of the Gaussian log-likelihood function become daunting as the number $n$ of geographical locations grows, as ${mathcal O}(n^2)$ storage and ${mathcal O}(n^3)$ operations are required. While many approximation methods have been devised from the side of statistical modeling to ameliorate these polynomial complexities, we are interested here in the complementary approach of evaluating the exact algebraic result by exploiting advances in solution algorithms and many-core computer architectures. Using state-of-the-art high performance dense linear algebra libraries associated with various leading edge parallel architectures (Intel KNLs, NVIDIA GPUs, and distributed-memory systems), ExaGeoStat raises the game for statistical applications from climate and environmental science. ExaGeoStat provides a reference evaluation of statistical parameters, with which to assess the validity of the various approaches based on approximation. The software takes a first step in the merger of large-scale data analytics and extreme computing for geospatial statistical applications, to be followed by additional complexity reducing improvements from the solver side that can be implemented under the same interface. Thus, a single uncompromised statistical model can ultimately be executed in a wide variety of emerging exascale environments.","Title: Asynchronous Task-Based Polar Decomposition on Single Node Manycore Architectures.\nYear: 2018\n\nAbstract: This paper introduces the first asynchronous, task-based formulation of the polar decomposition and its corresponding implementation on manycore architectures. Based on a new formulation of the iterative QR dynamically-weighted Halley algorithm (QDWH) for the calculation of the polar decomposition, the proposed implementation replaces the original and hostile LU factorization for the condition num...","Title: Fast multipole preconditioners for sparse matrices arising from elliptic equations.\nYear: 2018\n\nAbstract: Among optimal hierarchical algorithms for the computational solution of elliptic problems, the fast multipole method (FMM) stands out for its adaptability to emerging architectures, having high arithmetic intensity, tunable accuracy, and relaxable global synchronization requirements. We demonstrate that, beyond its traditional use as a solver in problems for which explicit free-space kernel representations are available, the FMM has applicability as a preconditioner in finite domain elliptic boundary value problems, by equipping it with boundary integral capability for satisfying conditions at finite boundaries and by wrapping it in a Krylov method for extensibility to more general operators. Here, we do not discuss the well developed applications of FMM to implement matrix-vector multiplications within Krylov solvers of boundary element methods. Instead, we propose using FMM for the volume-to-volume contribution of inhomogeneous Poisson-like problems, where the boundary integral is a small part of the overall computation. Our method may be used to precondition sparse matrices arising from finite difference/element discretizations, and can handle a broader range of scientific applications. It is capable of algebraic convergence rates down to the truncation error of the discretized PDE comparable to those of multigrid methods, and it offers potentially superior multicore and distributed memory scalability properties on commodity architecture supercomputers. Compared with other methods exploiting the low-rank character of off-diagonal blocks of the dense resolvent operator, FMM-preconditioned Krylov iteration may reduce the amount of communication because it is matrix-free and exploits the tree structure of FMM. We describe our tests in reproducible detail with freely available codes and outline directions for further extensibility.","Title: Extreme Scale FMM-Accelerated Boundary Integral Equation Solver for Wave Scattering.\nYear: 2018\n\nAbstract: We present a boundary integral equation solver for wave scattering suited for many-core processors, which are expected to be the building blocks of energy-austere exascale systems, and on which algorithmic and architecture-oriented optimizations are essential for achieving worthy performance. We use the GMRES iterative method and the fast multipole method (FMM) to implement the matrix-vector product (MatVec) kernel. We implement highly optimized kernels for both shared- and distributed-memory architectures and provide various performance models for tuning the task-based tree traversal implementation of FMM. We develop optimal architecture-specific and algorithm-aware partitioning, load balancing, and communication reducing mechanisms to scale up to 6,144 compute nodes of a Cray XC40 with 196,608 hardware cores. Task-based traversal achieves roughly 77% and 60% of the peak single precision floating point performance of a 56-core Skylake Scalable processor and a 72-core Knights Landing processor, respectively. These represent approximately sevenfold speedups compared to the baseline scalar code. We report weak scalability in accordance with the best possible O(log P) communication complexity and the theoretical scaling complexity of FMM. These result in up to 85% efficiency in strong scaling while computing in excess of 2 billion degrees of freedom (DoFs) on the full-scale Cray XC40. To demonstrate the applicability of the proposed implementation of FMM, we use it in the solution of the acoustic scattering problem involving soft targets. This calls for the discretization of an integral equation with an oscillatory kernel and (weak) 1/R singularity. Applications of FMM to the problems involving hard targets, which call for implementation of more complicated singularity extraction/cancellation schemes to account for 1/R-2 singularity, or to Burton-Miller surface integral formulation, which results in a well-conditioned matrix system, are not considered in this work, but the efficiency and scalability results demonstrated here are expected to be observed for these cases as well. The numerical results are converged to within 1.0e-4 relative 2-norm residual accuracy of the analytical solution for the sphere, and a multiple flower-shaped scatterer displays comparable algebraic convergence in a more general setting.","Title: Accelerated Cyclic Reduction: A distributed-memory fast solver for structured linear systems.\nYear: 2018\n\nAbstract: We present Accelerated Cyclic Reduction (ACR), a distributed-memory fast solver for rank-compressible block tridiagonal linear systems arising from the discretization of elliptic operators, developed here for three dimensions. Algorithmic synergies between Cyclic Reduction and hierarchical matrix arithmetic operations result in a solver that has O(kNlogN(logN+k2)) arithmetic complexity and O(k Nlog N) memory footprint, where N is the number of degrees of freedom and k is the rank of a block in the hierarchical approximation, and which exhibits substantial concurrency. We provide a baseline for performance and applicability by comparing with the multifrontal method with and without hierarchical semi-separable matrices, with algebraic multigrid and with the classic cyclic reduction method. Over a set of large-scale elliptic systems with features of nonsymmetry and indefiniteness, the robustness of the direct solvers extends beyond that of the multigrid solver, and relative to the multifrontal approach ACR has lower or comparable execution time and size of the factors, with substantially lower numerical ranks. ACR exhibits good strong and weak scaling in a distributed context and, as with any direct solver, is advantageous for problems that require the solution of multiple right-hand sides. Numerical experiments show that the rank k patterns are of O(1) for the Poisson equation and of O(n) for the indefinite Helmholtz equation. The solver is ideal in situations where low-accuracy solutions are sufficient, or otherwise as a preconditioner within an iterative method.","Title: Tile Low-Rank Approximation of Large-Scale Maximum Likelihood Estimation on Manycore Architectures.\nYear: 2018\n\nAbstract: Maximum likelihood estimation is an important statistical technique for estimating missing data, for example in climate and environmental applications, which are usually large and feature data points that are irregularly spaced. In particular, the Gaussian log-likelihood function is the emph{de facto} model, which operates on the resulting sizable dense covariance matrix. The advent of high performance systems with advanced computing power and memory capacity have enabled full simulations only for rather small dimensional climate problems, solved at the machine precision accuracy. The challenge for high dimensional problems lies in the computation requirements of the log-likelihood function, which necessitates ${mathcal O}(n^2)$ storage and ${mathcal O}(n^3)$ operations, where $n$ represents the number of given spatial locations. This prohibitive computational cost may be reduced by using approximation techniques that not only enable large-scale simulations otherwise intractable but also maintain the accuracy and the fidelity of the spatial statistics model. In this paper, we extend the Exascale GeoStatistics software framework (i.e., ExaGeoStat) to support the Tile Low-Rank (TLR) approximation technique, which exploits the data sparsity of the dense covariance matrix by compressing the off-diagonal tiles up to a user-defined accuracy threshold. The underlying linear algebra operations may then be carried out on this data compression format, which may ultimately reduce the arithmetic complexity of the maximum likelihood estimation and the corresponding memory footprint. Performance results of TLR-based computations on shared and distributed-memory systems attain up to 13X and 5X speedups, respectively, compared to full accuracy simulations using synthetic and real datasets (up to 2M), while ensuring adequate prediction accuracy.","Title: Fast parallel multidimensional FFT using advanced MPI.\nYear: 2018\n\nAbstract: We present a new method for performing global redistributions of multidimensional arrays essential to parallel fast Fourier (or similar) transforms. Traditional methods use standard all-to-all collective communication of contiguous memory buffers, thus necessarily requiring local data realignment steps intermixed in-between redistribution and transform steps. Instead, our method takes advantage of subarray datatypes and generalized all-to-all scatter/gather from the MPI-2 standard to communicate discontiguous memory buffers, effectively eliminating the need for local data realignments. Despite generalized all-to-all communication of discontiguous data being generally slower, our proposal economizes in local work. For a range of strong and weak scaling tests, we found the overall performance of our method to be on par and often better than well-established libraries like MPI-FFTW, P3DFFT, and 2DECOMP&FFT. We provide compact routines implemented at the highest possible level using the MPI bindings for the C programming language. These routines apply to any global redistribution, over any two directions of a multidimensional array, decomposed on arbitrary Cartesian processor grids (1D slabs, 2D pencils, or even higher-dimensional decompositions). The high level implementation makes the code easy to read, maintain, and eventually extend. Our approach enables for future speedups from optimizations in the internal datatype handling engines within MPI implementations.","Title: Reconstructing parameters of the FitzHugh–Nagumo system from boundary potential measurements\nYear: 2007\n\nAbstract: We consider distributed parameter identification problems for the FitzHugh–Nagumo model of electrocardiology. The model describes the evolution of electrical potentials in heart tissues. The mathematical problem is to reconstruct physical parameters of the system through partial knowledge of its solutions on the boundary of the domain. We present a parallel algorithm of Newton–Krylov type that combines Newton’s method for numerical optimization with Krylov subspace solvers for the resulting Karush–Kuhn–Tucker system. We show by numerical simulations that parameter reconstruction can be performed from measurements taken on the boundary of the domain only. We discuss the effects of various model parameters on the quality of reconstructions.","Title: Parallel accelerated cyclic reduction preconditioner for three-dimensional elliptic PDEs with variable coefficients.\nYear: 2018\n\nAbstract: We present a robust and scalable preconditioner for the solution of large-scale linear systems that arise from the discretization of elliptic PDEs amenable to rank compression. The preconditioner is based on hierarchical low-rank approximations and the cyclic reduction method. The setup and application phases of the preconditioner achieve log-linear complexity in memory footprint and number of operations, and numerical experiments exhibit good weak and strong scalability at large processor counts in a distributed memory environment. Numerical experiments with linear systems that feature symmetry and nonsymmetry, definiteness and indefiniteness, constant and variable coefficients demonstrate the preconditioner applicability and robustness. Furthermore, it is possible to control the number of iterations via the accuracy threshold of the hierarchical matrix approximations and their arithmetic operations, and the tuning of the admissibility condition parameter. Together, these parameters allow for optimization of the memory requirements and performance of the preconditioner.","Title: Real-Time Massively Distributed Multi-object Adaptive Optics Simulations for the European Extremely Large Telescope\nYear: 2018\n\nAbstract: The European Extremely Large Telescope (E-ELT) is one of today's most challenging projects in ground-based astronomy. Addressing one of the key science cases for the E-ELT, the study of the early Universe, requires the implementation of multi-object adaptive optics (MOAO), a dedicated concept relying on turbulence tomography. We use a novel pseudo-analytical approach to simulate the performance of tomographic reconstruction of the atmospheric turbulence in an MOAO system on real datasets. We simulate simultaneously 4K galaxies in a common field of view on massively parallel supercomputers during a single night of observations. We are able to generate a first-ever high-resolution galaxy map at almost a real-time throughput. This simulation scale opens new research horizons in numerical methods for experimental astronomy, some core components of the pipeline standing as pathfinders toward actual operations and future astronomic discoveries on the E-ELT.","Title: Tile Low-Rank GEMM Using Batched Operations on GPUs.\nYear: 2018\n\nAbstract: Dense General Matrix-Matrix (GEMM) multiplication is a core operation of the Basic Linear Algebra Subroutines (BLAS) library, and therefore, often resides at the bottom of the traditional software stack for many scientific applications. In fact, chip manufacturers give a special attention to the GEMM kernel implementation since this is exactly where most of the high-performance software libraries extract hardware performance. With the emergence of big data applications involving large data-sparse, hierarchically low-rank matrices, the off-diagonal tiles can be compressed to reduce the algorithmic complexity and the memory footprint. The resulting tile low-rank (TLR) data format is composed of small data structures, which retain the most significant information for each tile. However, to operate on low-rank tiles, a new GEMM operation and its corresponding API have to be designed on GPUs so that the data sparsity structure of the matrix can be exploited while leveraging the underlying TLR compression format. The main idea consists of aggregating all operations into a single kernel launch to compensate for their low arithmetic intensities and to mitigate the data transfer overhead on GPUs. The new TLR-GEMM kernel outperforms the cuBLAS dense batched GEMM by more than an order of magnitude and creates new opportunities for TLR advanced algorithms.","Title: Exploiting Data Sparsity for Large-Scale Matrix Computations.\nYear: 2018\n\nAbstract: Exploiting data sparsity in dense matrices is an algorithmic bridge between architectures that are increasingly memory-austere on a per-core basis and extreme-scale applications. In this work, we leverage the Hierarchical matrix Computations on Manycore Architectures (HiCMA) library in order to tackle this challenging problem by achieving significant reductions in time to solution and memory footprint, while preserving a specified accuracy requirement of the application. We have extended HiCMA to provide a high-performance implementation on distributed-memory systems of one of the most widely used matrix factorization in large-scale scientific applications, i.e., the Cholesky factorization. It employs the tile low-rank data format to compress the dense data-sparse off-diagonal tiles of the matrix. It then decomposes the matrix computations into interdependent tasks and relies on the dynamic runtime system StarPU for asynchronous out-of-order scheduling, while allowing high user productivity. Performance comparisons and memory footprint on matrix dimensions up to eleven million show a performance gain and memory saving of more than an order of magnitude for both metrics on thousands of cores, against state-of-the-art open-source and vendor optimized numerical libraries. This represents an important milestone in enabling large-scale matrix computations toward solving big data problems in geospatial statistics for climate/weather forecasting applications.","Title: Optimizations of Unstructured Aerodynamics Computations for Many-core Architectures.\nYear: 2018\n\nAbstract: We investigate several state-of-the-practice shared-memory optimization techniques applied to key routines of an unstructured computational aerodynamics application with irregular memory accesses. We illustrate for the Intel Knights Landing processor, as a representative of the processors in contemporary leading supercomputers, identifying and addressing performance challenges without compromising...","Title: Parallel Approximation of the Maximum Likelihood Estimation for the Prediction of Large-Scale Geostatistics Simulations\nYear: 2018\n\nAbstract: Maximum likelihood estimation is an important statistical technique for estimating missing data, for example in climate and environmental applications, which are usually large and feature data points that are irregularly spaced. In particular, the Gaussian log-likelihood function is the de facto model, which operates on the resulting sizable dense covariance matrix. The advent of high performance systems with advanced computing power and memory capacity have enabled full simulations only for rather small dimensional climate problems, solved at the machine precision accuracy. The challenge for high dimensional problems lies in the computation requirements of the log-likelihood function, which necessitates O(n\n<sup xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">2</sup>\n) storage and O(n\n<sup xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">3</sup>\n) operations, where n represents the number of given spatial locations. This prohibitive computational cost may be reduced by using approximation techniques that not only enable large-scale simulations otherwise intractable, but also maintain the accuracy and the fidelity of the spatial statistics model. In this paper, we extend the Exascale GeoStatistics software framework (i.e., ExaGeoStat\n<sup xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">1</sup>\n) to support the Tile Low-Rank (TLR) approximation technique, which exploits the data sparsity of the dense covariance matrix by compressing the off-diagonal tiles up to a user-defined accuracy threshold. The underlying linear algebra operations may then be carried out on this data compression format, which may ultimately reduce the arithmetic complexity of the maximum likelihood estimation and the corresponding memory footprint. Performance results of TLR-based computations on shared and distributed-memory systems attain up to 13X and 5X speedups, respectively, compared to full accuracy simulations using synthetic and real datasets (up to 2M), while ensuring adequate prediction accuracy.","Title: ExaGeoStat: A High Performance Unified Software for Geostatistics on Manycore Systems.\nYear: 2018\n\nAbstract: We present ExaGeoStat, a high performance software for geospatial statistics in climate and environment modeling. In contrast to simulation based on partial differential equations derived from first-principles modeling, ExaGeoStat employs a statistical model based on the evaluation of the Gaussian log-likelihood function, which operates on a large dense covariance matrix. Generated by the parametr...","Title: SBP-SAT finite difference discretization of acoustic wave equations on staggered block-wise uniform grids.\nYear: 2019\n\nAbstract: We consider the numerical simulation of the acoustic wave equations arising from seismic applications, for which staggered grid finite difference methods are popular choices due to their simplicity and efficiency. We relax the uniform grid restriction on finite difference methods and allow the grids to be block-wise uniform with nonconforming interfaces. In doing so, variations in the wave speeds of the subterranean media can be accounted for more efficiently. Staggered grid finite difference operators satisfying the summation-by-parts (SBP) property are devised to approximate the spatial derivatives appearing in the acoustic wave equation. These operators are applied within each block independently. The coupling between blocks is achieved through simultaneous approximation terms (SATs), which impose the interface conditions weakly, i.e., by penalty. Ratio of the grid spacing of neighboring blocks is allowed to be rational number, for which specially designed interpolation formulas are presented. These interpolation formulas constitute key pieces of the simultaneous approximation terms. The overall discretization is shown to be energy-conserving and examined on test cases of both theoretical and practical interests, delivering accurate and stable simulation results.","Title: Performance Assessment of Hybrid Parallelism for Large-Scale Reservoir Simulation on Multi- and Many-core Architectures\nYear: 2018\n\nAbstract: Two trends are reshaping the landscape of petroleum reservoir simulators, one architecturally and one application driven: an increasing number of cores per node and increasing computational intensity arising from higher fidelity physics at each cell. Implicit algebraic solvers being the dominant kernels, we present hybrid MPI and OpenMP implementations of the linear solver of GigaPOWERS, a full-scale real-world massively parallel simulator for black oil and composition models. We also evaluate the impact of explicit communication and computation overlap by including the halo exchange in the task-dependency graph. We analyze the performance of these modifications across multi- and many-core architectures, i.e., Intel Haswell, Skylake, and Knights Landing, using a variety of synthetic and real-world models. The hybrid approach results in up to 50% reduction of time to solution on a 16 million-cell SPE10-like model on Skylake whereas on a smaller, 1 million-cell, model on Haswell and Knights Landing both implementations achieve very similar performance. In the real-world reservoir simulations, the hybrid parallelism has reduced communication volume, memory consumption, and improved load balancing.","Title: Combining finite element and finite difference methods for isotropic elastic wave simulations in an energy-conserving manner.\nYear: 2019\n\nAbstract: We consider numerical simulation of the isotropic elastic wave equations arising from seismic applications with non-trivial land topography. The more flexible finite element method is applied to the shallow region of the simulation domain to account for the topography, and combined with the more efficient finite difference method that is applied to the deep region of the simulation domain. We demonstrate that these two discretization methods, albeit starting from different formulations of the elastic wave equation, can be joined together smoothly via weakly imposed interface conditions. Discrete energy analysis is employed to derive the proper interface treatment, leading to an overall discretization that is energy-conserving. Numerical examples are presented to demonstrate the efficacy of the proposed interface treatment.","Title: Tucker Tensor Analysis of Matérn Functions in Spatial Statistics.\nYear: 2019\n\nAbstract: In this work, we describe advanced numerical tools for working with multivariate functions and for the analysis of large data sets. These tools will drastically reduce the required computing time and the storage cost, and, therefore, will allow us to consider much larger data sets or finer meshes. Covariance matrices are crucial in spatio-temporal statistical tasks, but are often very expensive to compute and store, especially in three dimensions. Therefore, we approximate covariance functions by cheap surrogates in a low-rank tensor format. We apply the Tucker and canonical tensor decompositions to a family of Matern- and Slater-type functions with varying parameters and demonstrate numerically that their approximations exhibit exponentially fast convergence. We prove the exponential convergence of the Tucker and canonical approximations in tensor rank parameters. Several statistical operations are performed in this low-rank tensor format, including evaluating the conditional covariance matrix, spatially averaged estimation variance, computing a quadratic form, determinant, trace, loglikelihood, inverse, and Cholesky decomposition of a large covariance matrix. Low-rank tensor approximations reduce the computing and storage costs essentially. For example, the storage cost is reduced from an exponential O(n(d)) to a linear scaling O(drn), where d is the spatial dimension, n is the number of mesh points in one direction, and r is the tensor rank. Prerequisites for applicability of the proposed techniques are the assumptions that the data, locations, and measurements lie on a tensor (axes-parallel) grid and that the covariance function depends on a distance, parallel to x - y parallel to.","Title: Batched Triangular Dense Linear Algebra Kernels for Very Small Matrix Sizes on GPUs\nYear: 2019\n\nAbstract: Batched dense linear algebra kernels are becoming ubiquitous in scientific applications, ranging from tensor contractions in deep learning to data compression in hierarchical low-rank matrix approximation. Within a single API call, these kernels are capable of simultaneously launching up to thousands of similar matrix computations, removing the expensive overhead of multiple API calls while increasing the occupancy of the underlying hardware. A challenge is that for the existing hardware landscape (x86, GPUs, etc.), only a subset of the required batched operations is implemented by the vendors, with limited support for very small problem sizes. We describe the design and performance of a new class of batched triangular dense linear algebra kernels on very small data sizes (up to 256) using single and multiple GPUs. By deploying recursive formulations, stressing the register usage, maintaining data locality, reducing threads synchronization, and fusing successive kernel calls, the new batched kernels outperform existing state-of-the-art implementations.","Title: A Note on Adaptive Nonlinear Preconditioning Techniques.\nYear: 2018\n\nAbstract: Nonlinear preconditioning is a globalization technique for Newton's method applied to systems of equations with unbalanced nonlinearities, in which nonlinear residual norm reduction stagnates due to slowly evolving subsets of the degrees of freedom. Even though the Newton corrections may effectively be sparse, a standard Newton method still requires large ill-conditioned linear systems resulting from global linearizations of the nonlinear residual to be solved at each step. Nonlinear preconditioners may enable faster global convergence by shifting work to where it is most strategic, on subsets of the original system. They require additional computation per outer iteration while aiming for many fewer outer iterations and correspondingly fewer global synchronizations. In this work, we improve upon previous nonlinear preconditioning implementations by introducing parameters that allow turning off nonlinear preconditioning during outer Newton iterations where it is not needed. Numerical experiments show that the adaptive nonlinear preconditioning algorithm has performance similar to monolithically applied nonlinear preconditioning, preserving robustness for some challenging problems representative of several PDE-based applications while saving work on nonlinear subproblems.","Title: Hierarchical-block conditioning approximations for high-dimensional multivariate normal probabilities\nYear: 2019\n\nAbstract: This paper presents a new method to estimate large-scale multivariate normal probabilities. The approach combines a hierarchical representation with processing of the covariance matrix that decomposes the n-dimensional problem into a sequence of smaller m-dimensional ones. It also includes a d-dimensional conditioning method that further decomposes the m-dimensional problems into smaller d-dimensional problems. The resulting two-level hierarchical-block conditioning method requires Monte Carlo simulations to be performed only in d dimensions, with \\(d \\ll n\\), and allows the complexity of the algorithm’s major cost to be \\(O(n \\log n)\\). The run-time cost of the method depends on two parameters, m and d, where m represents the diagonal block size and controls the sizes of the blocks of the covariance matrix that are replaced by low-rank approximations, and d allows a trade-off of accuracy for expensive computations in the evaluation of the probabilities of m-dimensional blocks. We also introduce an inexpensive block reordering strategy to provide improved accuracy in the overall probability computation. The downside of this method, as with other such conditioning approximations, is the absence of an internal estimate of its error to use in tuning the approximation. Numerical simulations on problems from 2D spatial statistics with dimensions up to 16,384 indicate that the algorithm achieves a \\(1\\%\\) error level and improves the run time over a one-level hierarchical Quasi-Monte Carlo method by a factor between 10 and 15.","Title: Outsmarting the Atmospheric Turbulence for Ground-Based Telescopes Using the Stochastic Levenberg-Marquardt Method\nYear: 2021\n\nAbstract: One of the main challenges for ground-based optical astronomy is to compensate for atmospheric turbulence in near real-time. The goal is to obtain images as close as possible to the diffraction limit of the telescope. This challenge is addressed on the latest generation of giant optical telescopes by deploying multi-conjugate adaptive optics (MCAO) systems performing predictive tomography of the turbulence and multi-layer compensation. Such complex systems require a high fidelity estimate of the turbulence profile above the telescope, to be updated regularly during operations as turbulence conditions evolve. In this paper, we modify the traditional Levenberg-Marquardt (LM) algorithm by considering stochastically chosen subsystems of the full problem to identify the required parameters efficiently, while coping with the real-time challenge. While LM operates on the full set data samples, the resulting Stochastic LM (SLM) method randomly selects subsamples to compute corresponding approximate gradients and Hessians. Hence, SLM reduces the algorithmic complexity per iteration and shortens the overall time to solution, while maintaining LM's numerical robustness. We present a new convergence analysis for SLM, implement the algorithm with optimized GPU kernels, and deploy it on shared-memory systems with multiple GPU accelerators. We assess SLM in the adaptive optics system configurations in the context of the MCAO-Assisted Visible Imager & Spectrograph (MAVIS) instrument for the Very Large Telescope (VLT). We demonstrate performance superiority of SLM over the traditional LM algorithm and the classical stochastic first-order methods. At the scale of VLT AO, SLM finishes the optimization process and accurately retrieves the parameters (e.g., turbulence strength and wind speed profiles) in less than a second using up to eight NVIDIA A100 GPUs, which permits high acuity real-time throughput over a night of observations.","Title: H2Opus: a distributed-memory multi-GPU software package for non-local operators\nYear: 2022\n\nAbstract: Hierarchical \n \n \n \n \n ${\\mathscr{H}}^{2}$\n -matrices are asymptotically optimal representations for the discretizations of non-local operators such as those arising in integral equations or from kernel functions. Their O(N) complexity in both memory and operator application makes them particularly suited for large-scale problems. As a result, there is a need for software that provides support for distributed operations on these matrices to allow large-scale problems to be represented. In this paper, we present high-performance, distributed-memory GPU-accelerated algorithms and implementations for matrix-vector multiplication and matrix recompression of hierarchical matrices in the \n \n \n \n \n ${\\mathscr{H}}^{2}$\n format. The algorithms are a new module of H2Opus, a performance-oriented package that supports a broad variety of \n \n \n \n \n ${\\mathscr{H}}^{2}$\n matrix operations on CPUs and GPUs. Performance in the distributed GPU setting is achieved by marshaling the tree data of the hierarchical matrix representation to allow batched kernels to be executed on the individual GPUs. MPI is used for inter-process communication. We optimize the communication data volume and hide much of the communication cost with local compute phases of the algorithms. Results show near-ideal scalability up to 1024 NVIDIA V100 GPUs on Summit, with performance exceeding 2.3 Tflop/s/GPU for the matrix-vector multiplication, and 670 Gflop/s/GPU for matrix compression, which involves batched QR and SVD operations. We illustrate the flexibility and efficiency of the library by solving a 2D variable diffusivity integral fractional diffusion problem with an algebraic multigrid-preconditioned Krylov solver and demonstrate scalability up to 16M degrees of freedom problems on 64 GPUs.","Title: Approximate Error Bounds On Solutions Of Nonlinearly Preconditioned Pdes\nYear: 2021\n\nAbstract: In many multiphysics applications, an ultimate quantity of interest can be written as a linear functional of the solution to the discretized governing nonlinear partial differential equations and finding a sufficiently accurate pointwise solution may be regarded as a step toward that end. In this paper, we derive a posteriori approximate error bounds for linear functionals corresponding to quantities of interest using two kinds of nonlinear preconditioning techniques. Nonlinear preconditioning, such as the inexact Newton with backtracking and nonlinear elimination algorithm and the multiplicative Schwarz preconditioned inexact Newton algorithm, may be effective in improving global convergence for Newton's method. It may prevent stagnation of the nonlinear residual norm and reduce the number of solutions of large ill-conditioned linear systems involving a global Jacobian required at each nonlinear iteration. We illustrate the effectiveness of the new bounds using canonical nonlinear PDE models: a flame sheet model and a nonlinear coupled lid-driven cavity problem.","Title: An O(N) Algorithm For Computing Expectation Of N-Dimensional Truncated Multi-Variate Normal Distribution I: Fundamentals\nYear: 2021\n\nAbstract: In this paper, we present the fundamentals of a hierarchical algorithm for computing the N-dimensional integral phi(a, b; A) = integral(b)(a) H(x) f (x vertical bar A)dx representing the expectation of a function H(X) where f (x vertical bar A) is the truncated multi-variate normal (TMVN) distribution with zero mean, x is the vector of integration variables for the N-dimensional random vector X, A is the inverse of the covariance matrix Sigma, and a and b are constant vectors. The algorithm assumes that H(x) is \"low-rank\" and is designed for properly clustered X so that the matrix A has \"low-rank\" blocks and \"low-dimensional\" features. We demonstrate the divide-and-conquer idea when A is a symmetric positive definite tridiagonal matrix and present the necessary building blocks and rigorous potential theory-based algorithm analysis when A is given by the exponential covariance model. The algorithm overall complexity is O(N) for N-dimensional problems, with a prefactor determined by the rank of the off-diagonal matrix blocks and number of effective variables. Very high accuracy results for N as large as 2048 are obtained on a desktop computer with 16G memory using the fast Fourier transform (FFT) and non-uniform FFT to validate the analysis. The current paper focuses on the ideas using the simple yet representative examples where the off-diagonal matrix blocks are rank 1 and the number of effective variables is bounded by 2, to allow concise notations and easier explanation. In a subsequent paper, we discuss the generalization of current scheme using the sparse grid technique for higher rank problems and demonstrate how all the moments of k(th) order or less (a total of O(N-k) integrals) can be computed using O(N-k) operations for k >= 2 and O(N log N) operations for k = 1.","Title: Accelerating Geostatistical Modeling and Prediction With Mixed-Precision Computations: A High-Productivity Approach With PaRSEC\nYear: 2022\n\nAbstract: Geostatistical modeling, one of the prime motivating applications for exascale computing, is a technique for predicting desired quantities from geographically distributed data, based on statistical models and optimization of parameters. Spatial data are assumed to possess properties of stationarity or non-stationarity via a kernel fitted to a covariance matrix. A primary workhorse of stationary sp...","Title: Meeting the real-time challenges of ground-based telescopes using low-rank matrix computations\nYear: 2021\n\nAbstract: ABSTRACTAdaptive Optics (AO) is a technology that permits to measure and mitigate the distortion effects of atmospheric turbulence on optical beams. AO must operate in real-time by controlling thousands of actuators to shape the surface of deformable mirrors deployed on ground-based telescopes to compensate for these distortions. The command vectors that trigger how each individual actuator should act to bend a portion of the mirror are obtained from Matrix-Vector Multiplications (MVM). We identify and leverage the data sparsity structure of these control matrices coming from the MAVIS instruments for the European Southern Observatory's Very Large Telescope. We provide performance evaluation on x86 and accelerator-based systems. We present the impact of tile low-rank (TLR) matrix approximations on time-to-solution for the MVM and assess the produced image quality. We achieve performance improvement up to two orders of magnitude for TLR-MVM compared to regular dense MVM, while maintaining the image quality.","Title: Implications of Reduced Communication Precision in a Collocated Discontinuous Galerkin Finite Element Framework\nYear: 2021\n\nAbstract: Compute capability of high-performance hardware has been growing at immense rates, increasing over 130x in the last decade. Communication bandwidth, however, only grew by a factor of 6x in the same time, leading to a significant decrease in the byte-to-flop metric. This trend leads us to the situation where, in many cases, computation is virtually free, and the dominant cost of a parallel applicat...","Title: NONLINEAR PRECONDITIONING STRATEGIES FOR TWO-PHASE FLOWS IN POROUS MEDIA DISCRETIZED BY A FULLY IMPLICIT DISCONTINUOUS GALERKIN METHOD\\ast\nYear: 2021\n\nAbstract: We consider numerical simulation of two-phase flows in porous media using implicit methods. Because of the complex features involving heterogeneous permeability and nonlinear capillary effects, the nonlinear algebraic systems arising from the discretization are very difficult to solve. The traditional Newton method suffers from slow convergence in the form of a long stagnation or sometimes does not converge at all. In this paper, we develop nonlinear preconditioning strategies for the system of two-phase flows discretized by a fully implicit discontinuous Galerkin method. The preconditioners identify and approximately eliminate the local high nonlinearities that cause the Newton method to take small updates. Specifically, we propose two elimination strategies: one is based on exploring the unbalanced nonlinearities of the pressure and the saturation fields, and the other is based on identifying certain elements of the finite element space that have much higher nonlinearities than the rest of the elements. We compare the performance and robustness of the proposed algorithms with an existing single-field elimination approach and the classical inexact Newton method with respect to some physical and numerical parameters. Experiments on three-dimensional porous media applications show that the proposed algorithms are superior to other methods in terms of robustness and parallel efficiency.","Title: tlrmvnmvt: Computing High-Dimensional Multivariate Normal and Student-t Probabilities with Low-Rank Methods in R\nYear: 2022\n\nAbstract: This paper introduces the usage and performance of the R package tlrmvnmvt, aimed at computing high-dimensional multivariate normal and Student-t probabilities. The package implements the tile-low-rank methods with block reordering and the separation of-variable methods with univariate reordering. The performance is compared with two other state-of-the-art R packages, namely the mvtnorm and the TruncatedNormal packages. Our package has the best scalability and is likely to be the only option in thousands of dimensions. However, for applications with high accuracy requirements, the Truncated Normal package is more suitable. As an application example, we show that the excursion sets of a latent Gaussian random field can be computed with the tlrmvnmvt package without any model approximation and hence, the accuracy of the produced excursion sets is improved.","Title: Performance analysis of relaxation Runge-Kutta methods\nYear: 2022\n\nAbstract: Recently, global and local relaxation Runge-Kutta methods have been developed for guaranteeing the conservation, dissipation, or other solution properties for general convex functionals whose dynamics are crucial for an ordinary differential equation solution. These novel time integration procedures have an application in a wide range of problems that require dynamics-consistent and stable numerical methods. The application of a relaxation scheme involves solving scalar nonlinear algebraic equations to find the relaxation parameter. Even though root-finding may seem to be a problem technically straightforward and computationally insignificant, we address the problem at scale as we solve full-scale industrial problems on a CPU-powered supercomputer and show its cost to be considerable. In particular, we apply the relaxation schemes in the context of the compressible Navier-Stokes equations and use them to enforce the correct entropy evolution. We use seven different algorithms to solve for the global and local relaxation parameters and analyze their strong scalability. As a result of this analysis, within the global relaxation scheme, we recommend using Brent's method for problems with a low polynomial degree and of small sizes for the global relaxation scheme, while secant proves to be the best choice for higher polynomial degree solutions and large problem sizes. For the local relaxation scheme, we recommend secant. Further, we compare the schemes' performance using their most efficient implementations, where we look at their effect on the timestep size, overhead, and weak scalability. We show the global relaxation scheme to be always more expensive than the local approach-typically 1.1-1.5 times the cost. At the same time, we highlight scenarios where the global relaxation scheme might underperform due to its increased communication requirements. Finally, we present an analysis that sets expectations on the computational overhead anticipated based on the system properties.","Title: Parallel Approximations of the Tukey g-and-h Likelihoods and Predictions for Non-Gaussian Geostatistics\nYear: 2022\n\nAbstract: Maximum likelihood estimation is an essential tool in the procedure to impute missing data in climate/weather applications. By defining a particular statistical model, the maximum likelihood estimation can be used to understand the underlying structure of given geospatial data. The Gaussian random field has been widely used to describe geospatial data, as one of the most popular models under the hood of maximum likelihood estimation. Computation of Gaussian log-likelihood demands operations on a dense symmetric positive definite matrix, often parameterized by the Matérn correlation function. This computation of the log-likelihood requires \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">$\\mathcal{O}(n^{2})$</tex>\n storage and \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">$\\mathcal{O}(n^{3})$</tex>\n operations, which can be a huge task considering that the number of geographical locations, \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">$n$</tex>\n, now commonly reaches into the millions. However, despite its appealing theoretical properties, the assumptions of Gaussianity may be unrealistic since real data often show signs of skewness or have some extreme values. Herein, we consider the Tukey \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">${g-}$</tex>\n and \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">$-h$</tex>\n (TGH) random field as an example of a non-Gaussian random field that shows more robustness in modeling geospatial data by including two more parameters to incorporate skewness and heavy tail features in the model. This work provides the first HPC implementation of the TGH random field's inference on parallel hardware architectures. Using task-based programming models associated with dynamic runtime systems, our implementation leverages the high concurrency of current parallel systems. This permits to run the exact log-likelihood evaluation of the Tukey g-and-h (TGH) random fields for a decent number of geospatial locations. To tackle large-scale problems, we provide additionally an implementation of the given model using two different low-rank approximations. We compress the aforementioned positive-definite symmetric matrix for computing the log-likelihood and rely on the Tile Low-Rank (TLR) and the Hierarchical Off-Diagonal Low-Rank (HODLR) matrix approximations. We assess the performance and accuracy of the proposed implementations using synthetic datasets up to \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">$800K$</tex>\n and a \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">$300K$</tex>\n precipitation data of Germany to demonstrate the advantage of using non-Gaussian over Gaussian random fields. Moreover, by relying on TLR/HODLR matrix computations, we can now solve for larger matrix sizes while preserving the required accuracy for prediction. We show the performance superiority of TLR over HODLR matrix computations when calculating the TGH likelihoods and predictions. Our TLR-based approximation shows a speedup up to \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">$7.29X$</tex>\n and \n<tex xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">$2.96X$</tex>\n on shared-memory and distributed-memory systems, respectively, compared to the exact implementation.","Title: A NONLINEAR ELIMINATION PRECONDITIONED INEXACT NEWTON ALGORITHM\nYear: 2022\n\nAbstract: A nonlinear elimination preconditioned inexact Newton (NEPIN) algorithm is proposed for problems with localized strong nonlinearities. Due to unbalanced nonlinearities (\"nonlinear stiffness\"\"), the traditional inexact Newton method often exhibits a long plateau in the norm of the nonlinear residual or even fails to converge. NEPIN implicitly removes the components causing trouble for the global convergence through a correction based on nonlinear elimination within a subspace that provides a modified direction for the global Newton iteration. Numerical experiments show that NEPIN can be more robust than global inexact Newton algorithms and maintain fast convergence even for challenging problems, such as full potential transonic flows. NEPIN complements several previously studied nonlinear preconditioners with which it compares favorably experimentally on a classic shocked duct flow problem considered herein. NEPIN is shown to be fairly insensitive to mesh resolution and \"bad\"\" subproblem identification based on the local Mach number or the local nonlinear residual for transonic flow over a wing.","Title: High-Performance Spatial Data Compression for Scientific Applications\nYear: 2022\n\nAbstract: We implement an efficient data compression algorithm that reduces the memory footprint of spatial datasets generated during scientific simulations. Storing regularly these datasets is typically needed for checkpoint/restart or for post-processing purposes. Our lossy compression approach, codenamed HLRcompress (https://gitlab.mis.mpg. de/rok/HLRcompress), combines a hierarchical low-rank approximation technique with binary compression. This novel hybrid method is agnostic to the particular domain of application. We study the impact of HLRcompress on accuracy using synthetic datasets to demonstrate the software capabilities, including robustness and versatility. We assess different algebraic compression methods and report performance results on various parallel architectures. We then integrate it into a workflow of a direct numerical simulation solver for turbulent combustion on distributed-memory systems. We compress the generated snapshots during time integration using accuracy thresholds for each individual chemical species, without degrading the practical accuracy of the overall pressure and temperature. We eventually compare against state-of-the-art compression software. Our implementation achieves on average greater than 100-fold compression of the original size of the datasets."],"string":"[\n \"Title: Petaflop/s, seriously\\nYear: 2007\\n\\nAbstract: Sustained floating-point rates on real applications, as tracked by the Gordon Bell Prize, have increased by over five orders of magnitude from 1988, when 1 Gigaflop/s was reported on a structural simulation, to 2006, when 200 Teraflop/s were reported on a molecular dynamics simulation. Various versions of Moore's Law over the same interval provide only two to three orders of magnitude of improvement for an individual processor; the remaining factor comes from concurrency, which is of order 100,000 for the BlueGene/L computer, the platform of choice for the majority of recent Bell Prize finalists. As the semiconductor industry begins to slip relative to its own roadmap for silicon-based logic and memory, concurrency will play an increasing role in attaining the next order of magnitude, to arrive at the long-awaited milepost of 1 Petaflop/s sustained on a practical application, which should occur around 2009. Simulations based on Eulerian formulations of partial differential equations can be among the first applications to take advantage of petascale capabilities, but not the way most are presently being pursued. Only weak scaling can get around the fundamental limitation expressed in Amdahl's Law and only optimal implicit formulations can get around another limitation on scaling that is an immediate consequence of Courant-Friedrichs-Lewy stability theory under weak scaling of a PDE. Many PDE-based applications and other lattice-based applications with petascale roadmaps, such as quantum chromodynamics, will likely be forced to adopt optimal implicit solvers. However, even this narrow path to petascale simulation is made treacherous by the imperative of dynamic adaptivity, which drives us to consider algorithms and queuing policies that are less synchronous than those in common use today. Drawing on the SCaLeS report (2003-2004), the latest ITRS roadmap, some back-of-the-envelope estimates, and numerical experiences with PDE-based codes on recently available platforms, we will attempt to project the pathway to Petaflop/s for representative applications.\",\n \"Title: M06 - Issues for the future of supercomputing: impact of Moore's law and architecture on application performance\\nYear: 2006\\n\\nAbstract: This tutorial will explain technologies driving supercomputer speed increases and supercomputers' resulting ability to solve increasingly important problems. In guided session, participants will walk through future supercomputer performance projection on a representative application.Specific topics will include:A review of scientific problems amenable to supercomputers based on the broad-based SCaLeS study. The editor of the SCaLeS report presents this session.The International Technology Roadmap for Semiconductors (ITRS) and its implications for the size, speed, power, and architecture of supercomputers.Current microprocessor-based and emerging \\\"advanced\\\" architectures and their implications on applications performance. There will be special emphasis on the multi-core trend and how applications can use multiple cores effectively.We will discuss the physics issues (speed and power) that will define the end of the current evolutionary trend. We will show how nanotech, reversible logic, and quantum computing may become the basis of a revolutionary change.\",\n \"Title: Four Horizons for Enhancing the Performance of Parallel Simulations Based on Partial Differential Equations\\nYear: 2000\\n\\nAbstract: Simulations of PDE-based systems, such as flight vehicles, the global climate, petroleum reservoirs, semiconductor devices, and nuclear weapons, typically perform an order of magnitude or more below other scientific simulations (e.g., from chemistry and physics) with dense linear algebra or N-body kernels at their core. In this presentation, we briefly review the algorithmic structure of typical PDE solvers that is responsible for this situation and consider possible architectural and algorithmic sources for performance improvement. Some of these improvements are also applicable to other types of simulations, but we examine their consequences for PDEs: potential to exploit orders of magnitude more processor-memory units, better organization of the simulation for today's and likely near-future hierarchical memories, alternative formulations of the discrete systems to be solved, and new horizons in adaptivity. Each category is motivated by recent experiences in computational aerodynamics at the 1 Teraflop/s scale.\",\n \"Title: Analyzing the Parallel Scalability of an Implicit Unstructured Mesh CFD Code\\nYear: 2000\\n\\nAbstract: In this paper, we identify the scalability bottlenecks of an unstructured grid CFD code (PETSc-FUN3D) by studying the impact of several algorithmic and architectural parameters and by examining different programming models. We discuss the basic performance characteristics of this PDE code with the help of simple performance models developed in our earlier work, presenting primarily experimental results. In addition to achievingg ood per-processor performance (which has been addressed in our cited work and without which scalability claims are suspect) we strive to improve the implementation and convergence scalability of PETSc-FUN3D on thousands of processors.\",\n \"Title: Three parallel programming paradigms: comparisons on an archetypal PDE computation\\nYear: 1999\\n\\nAbstract: Three paradigms for distributed-memory parallel computation that free the application programmer from the details of message passing are compared for an archetypal structured scientific computation -- a nonlinear, structured-grid partial differential equation boundary value problem -- using the same algorithm on the same hardware. All of the paradigms -- parallel languages represented by the Portland Group's HPF, (semi-)automated serial-to-parallel source-to-source translation represented by CAP-Tools from the University of Greenwich, and parallel libraries represented by Argonne's PETSc -- are found to be easy to use for this problem class, and all are reasonably effective in exploiting concurrency after a short learning curve. The level of involvement required by the application programmer under any paradigm includes specification of the data partitioning, corresponding to a geometrically simple decomposition of the domain of the PDE. Programming in SPMD style for the PETSc library requires writing only the routines that discretize the PDE and its Jacobian, managing subdomain-to-processor mappings (affine global-to-local index mappings), and interfacing to library solver routines. Programming for HPF requires a complete sequential implementation of the same algorithm as a starting point, introduction of concurrency through subdomain blocking (a task similar to the index mapping), and modest experimentation with rewriting loops to elucidate to the compiler the latent concurrency. Programming with CAPTools involves feeding the same sequential implementation to the CAPTools interactive parallelization system, and guiding the source-to-source code transformation by responding to various queries about quantities knowable only at runtime. Results representative of \\\"the state of the practice\\\" for a scaled sequence of structured grid problems are given on three of the most important contemporary high-performance platforms: the IBM SP, the SGI Origin 2000, and the CRAYY T3E.\",\n \"Title: Parallel Newton--Krylov--Schwarz Algorithms for the Transonic Full Potential Equation\\nYear: 1998\\n\\nAbstract: We study parallel two-level overlapping Schwarz algorithms for solving nonlinear finite element problems, in particular, for the full potential equation of aerodynamics discretized in two dimensions with bilinear elements. The overall algorithm, Newton--Krylov--Schwarz (NKS), employs an inexact finite difference Newton method and a Krylov space iterative method, with a two-level overlapping Schwarz method as a preconditioner. We demonstrate that NKS, combined with a density upwinding continuation strategy for problems with weak shocks, is robust and economical for this class of mixed elliptic-hyperbolic nonlinear partial differential equations, with proper specification of several parameters. We study upwinding parameters, inner convergence tolerance, coarse grid density, subdomain overlap, and the level of fill-in in the incomplete factorization, and report their effect on numerical convergence rate, overall execution time, and parallel efficiency on a distributed-memory parallel computer.\",\n \"Title: Moving grids for magnetic reconnection via Newton–Krylov methods\\nYear: 2011\\n\\nAbstract: This paper presents a set of computationally efficient, adaptive grids for magnetic reconnection phenomenon where the current density can develop large gradients in the reconnection region. Four-field extended MagnetoHydroDynamics (MHD) equations with hyperviscosity terms are transformed so that the curvilinear coordinates replace the Cartesian coordinates as the independent variables, and moving grids' velocities are also considered in this transformed system as a part of interpolating the physical solutions from the old grid to the new grid as time advances. The curvilinear coordinates derived from the current density through the Monge–Kantorovich (MK) optimization approach help to reduce the resolution requirements during the computation.\",\n \"Title: High-performacne parallel implicit CFD\\nYear: 2001\\n\\nAbstract: Fluid dynamical simulations based on finite discretizations on (quasi-)static grids scale well in parallel, but execute at a disappointing percentage of per-processor peak floating point operation rates without special attention to layout and access ordering of data. We document both claims from our experience with an unstructured grid CFD code that is typical of the state of the practice at NASA. These basic performance characteristics of PDE-based codes can be understood with surprisingly simple models, for which we quote earlier work, presenting primarily experimental results. The performance models and experimental results motivate algorithmic and software practices that lead to improvements in both parallel scalability and per node performance. This snapshot of ongoing work updates our 1999 Bell Prize-winning simulation on ASCI computers.\",\n \"Title: Domain decomposition on parallel computers\\nYear: 1989\\n\\nAbstract: We consider the application of domain decomposition techniques to the solution of sparse linear systems arising from implicit PDE discretizations on parallel computers. Representatives of two popular MIMD architectures, message passing (the Intel iPSC/2-SX) and shared memory (the Encore Multimax 320), are employed. We run the same numerical experiments on each, namely stripwise and boxwise decompositions of the unit square, using up to 64 subdomains and containing up to 64K degrees of freedom. We produce a tight-fitting complexity model for the former and discuss the difficulty of doing so for the latter. We also evaluate which of three types of domain decomposition preconditioners that have appeared in the literature of self-adjoint elliptic problems are most efficient in different regions of machine-problem parameter space. Some form of global sharing of information in the preconditioner is required for efficient overall parallel implementation in the region of most practical interest (large problem sizes and large numbers of processors); otherwise, an increasing iteration count inveighs against the gains of concurrency. Our results on a per iteration basis also hold for sparse discrete systems arising from other types of partial differential equations, but in the absence of a theory for the dependence of the convergence rate upon the granularity of the decomposition, the overall results are only suggestive for more general systems.\",\n \"Title: Multiphysics simulations: Challenges and opportunities.\\nYear: 2013\\n\\nAbstract: We consider multiphysics applications from algorithmic and architectural perspectives, where â聙聹algorithmicâ聙聺 includes both mathematical analysis and computational complexity, and â聙聹architecturalâ聙聺 includes both software and hardware environments. Many diverse multiphysics applications can be reduced, en route to their computational simulation, to a common algebraic coupling paradigm. Mathematical analysis of multiphysics coupling in this form is not always practical for realistic applications, but model problems representative of applications discussed herein can provide insight. A variety of software frameworks for multiphysics applications have been constructed and refined within disciplinary communities and executed on leading-edge computer systems. We examine several of these, expose some commonalities among them, and attempt to extrapolate best practices to future systems. From our study, we summarize challenges and forecast opportunities.\",\n \"Title: Numerical simulation of four-field extended magnetohydrodynamics in dynamically adaptive curvilinear coordinates via Newton-Krylov-Schwarz\\nYear: 2012\\n\\nAbstract: Numerical simulations of the four-field extended magnetohydrodynamics (MHD) equations with hyper-resistivity terms present a difficult challenge because of demanding spatial resolution requirements. A time-dependent sequence of r-refinement adaptive grids obtained from solving a single Monge-Ampere (MA) equation addresses the high-resolution requirements near the x-point for numerical simulation of the magnetic reconnection problem. The MHD equations are transformed from Cartesian coordinates to solution-defined curvilinear coordinates. After the application of an implicit scheme to the time-dependent problem, the parallel Newton-Krylov-Schwarz (NKS) algorithm is used to solve the system at each time step. Convergence and accuracy studies show that the curvilinear solution requires less computational effort than a pure Cartesian treatment. This is due both to the more optimal placement of the grid points and to the improved convergence of the implicit solver, nonlinearly and linearly. The latter effect, which is significant (more than an order of magnitude in number of inner linear iterations for equivalent accuracy), does not yet seem to be widely appreciated.\",\n \"Title: Optimizing the performance of streaming numerical kernels on the IBM Blue Gene/P PowerPC 450 processor\\nYear: 2013\\n\\nAbstract: Several emerging petascale architectures use energy-efficient processors with vectorized computational units and in-order thread processing. On these architectures the sustained performance of streaming numerical kernels, ubiquitous in the solution of partial differential equations, represents a challenge despite the regularity of memory access. Sophisticated optimization techniques are required to fully utilize the CPU. We propose a new method for constructing streaming numerical kernels using a high-level assembly synthesis and optimization framework. We describe an implementation of this method in Python targeting the IBM脗庐 Blue Gene脗庐/P supercomputer's PowerPC脗庐 450 core. This paper details the high-level design, construction, simulation, verification, and analysis of these kernels utilizing a subset of the CPU's instruction set. We demonstrate the effectiveness of our approach by implementing several three-dimensional stencil kernels over a variety of cached memory scenarios and analyzing the mechanically scheduled variants, including a 27-point stencil achieving a 1.7脙聴 speedup over the best previously published results.\",\n \"Title: Optimizing Memory-Bound Symv Kernel On Gpu Hardware Accelerators\\nYear: 2012\\n\\nAbstract: Hardware accelerators are becoming ubiquitous high performance scientific computing. They are capable of delivering an unprecedented level of concurrent execution contexts. High-level programming language extensions (e. g., CUDA), profiling tools (e. g., PAPI-CUDA, CUDA Profiler) are paramount to improve productivity, while effectively exploiting the underlying hardware. We present an optimized numerical kernel for computing the symmetric matrix-vector product on nVidia Fermi GPUs. Due to its inherent memory-bound nature, this kernel is very critical in the tridiagonalization of a symmetric dense matrix, which is a preprocessing step to calculate the eigenpairs. Using a novel design to address the irregular memory accesses by hiding latency and increasing bandwidth, our preliminary asymptotic results show 3.5x and 2.5x fold speedups over the similar CUBLAS 4.0 kernel, and 7-8% and 30% fold improvement over the Matrix Algebra on GPU and Multicore Architectures (MAGMA) library in single and double precision arithmetics, respectively.\",\n \"Title: Multi-core issues - Multi-Core for HPC: breakthrough or breakdown?\\nYear: 2006\\n\\nAbstract: A dramatic trend in computing is the adoption of multi-core technology by the vendors from which our current and future HPC systems are being derived. Multi-core is offered as a path to continued reliance and benefits of Moore's Law while reining in the previously unfettered growth of power consumption and design complexity. Are we saved? or is it but a fools mission, trapping us in a technical cul de sac with no long term direction and no way to reinvent an alternative future. The panel will consider the following questions:* Can multi-core span the next decade of Moore's Law progression?* Are the pins and caches a strangle hold on the future effectiveness of multi-core?* Can innovative algorithmic techniques exploit the opportunities and address the challenges of multi-core?* How will programming models and supporting system software change to accommodate the unique properties and peculiarities of multi-core structures?\",\n \"Title: Towards energy efficiency and maximum computational intensity for stencil algorithms using wavefront diamond temporal blocking.\\nYear: 2014\\n\\nAbstract: We study the impact of tunable parameters on computational intensity (i.e., inverse code balance) and energy consumption of multicore-optimized wavefront diamond temporal blocking (MWD) applied to different stencil-based update schemes. MWD combines the concepts of diamond tiling and multicore-aware wavefront blocking in order to achieve lower cache size requirements than standard single-core wavefront temporal blocking. We analyze the impact of the cache block size on the theoretical and observed code balance, introduce loop tiling in the leading dimension to widen the range of applicable diamond sizes, and show performance results on a contemporary Intel CPU. The impact of code balance on power dissipation on the CPU and in the DRAM is investigated and shows that DRAM power is a decisive factor for energy consumption, which is strongly influenced by the code balance. Furthermore we show that highest performance does not necessarily lead to lowest energy even if the clock speed is fixed.\",\n \"Title: KBLAS: An Optimized Library for Dense Matrix-Vector Multiplication on GPU Accelerators.\\nYear: 2014\\n\\nAbstract: KBLAS is an open-source, high-performance library that provides optimized kernels for a subset of Level 2 BLAS functionalities on CUDA-enabled GPUs. Since performance of dense matrix-vector multiplication is hindered by the overhead of memory accesses, a double-buffering optimization technique is employed to overlap data motion with computation. After identifying a proper set of tuning parameters, KBLAS efficiently runs on various GPU architectures while avoiding code rewriting and retaining compliance with the standard BLAS API. Another optimization technique allows ensuring coalesced memory access when dealing with submatrices, especially for high-level dense linear algebra algorithms. All KBLAS kernels have been leveraged to a multi-GPU environment, which requires the introduction of new APIs. Considering general matrices, KBLAS is very competitive with existing state-of-the-art kernels and provides a smoother performance across a wide range of matrix dimensions. Considering symmetric and Hermitian matrices, the KBLAS performance outperforms existing state-of-the-art implementations on all matrix sizes and achieves asymptotically up to 50% and 60% speedup against the best competitor on single GPU and multi-GPUs systems, respectively. Performance results also validate our performance model. A subset of KBLAS high-performance kernels have been integrated into NVIDIA's standard BLAS implementation (cuBLAS) for larger dissemination, starting from version 6.0.\",\n \"Title: Smooth and robust solutions for Dirichlet boundary control of fluid-solid conjugate heat transfer problems.\\nYear: 2015\\n\\nAbstract: We study a new optimization scheme that generates smooth and robust solutions for Dirichlet velocity boundary control (DVBC) of conjugate heat transfer (CHT) processes. The solutions to the DVBC of the incompressible Navier–Stokes equations are typically nonsmooth, due to the regularity degradation of the boundary stress in the adjoint Navier–Stokes equations. This nonsmoothness is inherited by the solutions to the DVBC of CHT processes, since the CHT process couples the Navier–Stokes equations of fluid motion with the convection–diffusion equations of fluid–solid thermal interaction. Our objective in the CHT boundary control problem is to select optimally the fluid inflow profile that minimizes an objective function that involves the sum of the mismatch between the temperature distribution in the fluid system and a prescribed temperature profile and the cost of the control.\",\n \"Title: High Performance Pseudo-analytical Simulation of Multi-Object Adaptive Optics over Multi-GPU Systems.\\nYear: 2014\\n\\nAbstract: Multi-object adaptive optics (MOAO) is a novel adaptive optics (AO) technique dedicated to the special case of wide-field multi-object spectrographs (MOS). It applies dedicated wavefront corrections to numerous independent tiny patches spread over a large field of view (FOV). The control of each deformable mirror (DM) is done individually using a tomographic reconstruction of the phase based on measurements from a number of wavefront sensors (WFS) pointing at natural and artificial guide stars in the field. The output of this study helps the design of a new instrument called MOSAIC, a multi-object spectrograph proposed for the European Extremely Large Telescope (E-ELT)(1). We have developed a novel hybrid pseudo-analytical simulation scheme that allows us to accurately simulate in detail the tomographic problem. The main challenge resides in the computation of the tomographic reconstructor, which involves pseudo-inversion of a large dense symmetric matrix. The pseudo-inverse is computed using an eigenvalue decomposition, based on the divide and conquer algorithm, on multicore systems with multi-GPUs. Thanks to a new symmetric matrix-vector product (SYMV) multi-GPU kernel, our overall implementation scores significant speedups over standard numerical libraries on multicore, like Intel MKL, and up to 60% speedups over the standard MAGMA implementation on 8 Kepler K20c GPUs. At 40,000 unknowns, this appears to be the largest-scale tomographic AO matrix solver submitted to computation, to date, to our knowledge and opens new research directions for extreme scale AO simulations.\",\n \"Title: Performance Modeling and Tuning of an Unstructured Mesh CFD Application\\nYear: 2000\\n\\nAbstract: This paper describes performance tuning experiences with a three-dimensional unstructured grid Euler flow code from NASA, which we have reimplemented in the PETSc framework and ported to several large-scale machines, including the ASCI Red and Blue Pacific machines, the SGI Origin, the Cray T3E, and Beowulf clusters. The code achieves a respectable level of performance for sparse problems, typical of scientific and engineering codes based on partial differential equations, and scales well up to thousands of processors. Since the gap between CPU speed and memory access rate is widening, the code is analyzed from a memory-centric perspective (in contrast to traditional flop-orientation) to understand its sequential and parallel performance. Performance tuning is approached on three fronts: data layouts to enhance locality of reference, algorithmic parameters, and parallel programming model. This effort was guided partly by some simple performance models developed for the sparse matrix-vector product operation.\",\n \"Title: Multicore-Optimized Wavefront Diamond Blocking for Optimizing Stencil Updates.\\nYear: 2015\\n\\nAbstract: The importance of stencil-based algorithms in computational science has focused attention on optimized parallel implementations for multilevel cache-based processors. Temporal blocking schemes leverage the large bandwidth and low latency of caches to accelerate stencil updates and approach theoretical peak performance. A key ingredient is the reduction of data traffic across slow data paths, especially the main memory interface. In this work we combine the ideas of multicore wavefront temporal blocking and diamond tiling to arrive at stencil update schemes that show large reductions in memory pressure compared to existing approaches. The resulting schemes show performance advantages in bandwidth-starved situations, which are exacerbated by the high bytes per lattice update case of variable coefficients. Our thread groups concept provides a controllable trade-off between concurrency and memory usage, shifting the pressure between the memory interface and the CPU. We present performance results on a contemporary Intel processor.\",\n \"Title: Nonlinearly Preconditioned Inexact Newton Algorithms\\nYear: 2002\\n\\nAbstract: Inexact Newton algorithms are commonly used for solving large sparse nonlinear system of equations $F(u^{\\\\ast})=0$ arising, for example, from the discretization of partial differential equations. Even with global strategies such as linesearch or trust region, the methods often stagnate at local minima of $\\\\|F\\\\|$, especially for problems with unbalanced nonlinearities, because the methods do not have built-in machinery to deal with the unbalanced nonlinearities. To find the same solution $u^{\\\\ast}$, one may want to solve instead an equivalent nonlinearly preconditioned system ${\\\\cal F}(u^{\\\\ast})=0$ whose nonlinearities are more balanced. In this paper, we propose and study a nonlinear additive Schwarz-based parallel nonlinear preconditioner and show numerically that the new method converges well even for some difficult problems, such as high Reynolds number flows, where a traditional inexact Newton method fails.\",\n \"Title: Towards polyalgorithmic linear system solvers for nonlinear elliptic problems\\nYear: 1994\\n\\nAbstract: The authors investigate the performance of several preconditioned conjugate gradient-like algorithms and a standard stationary iterative method (block-line successive overrelaxation (SOR)) on linear systems of equations that arise from a nonlinear elliptic flame sheet problem simulation. The nonlinearity forces a pseudotransient continuation process that makes the problem parabolic and thus compacts the spectrum of the Jacobian matrix so that simple relaxation methods are viable in the initial stages of the solution process. However, because of the transition from parabolic to elliptic character as the timestep is increased in pursuit of the steady-state solution, the performance of the candidate linear solvers spreads as the domain of convergence of Newton's method is approached. In numerical experiments over the course of a full nonlinear solution trajectory, short recurrence or optimal Krylov algorithms combined with a Gauss-Seidel (GS) preconditioning yield better execution times with respect to the standard block-line SOR techniques, but SOR performs competitively at a smaller storage cost until the final stages. Block-incomplete factorization preconditioned methods, on the other hand, require nearly a factor of two more storage than SOR and are uniformly less effective during the pseudotransient stages. The advantage of GS preconditioning is partly attributable to die exploitation of a dominant convection direction in the examples; nevertheless, a multidomain version of GS with streamwise coupling lagged at rows between adjacent subdomains incurs only a modest penalty.\",\n \"Title: Field-Split Preconditioned Inexact Newton Algorithms\\nYear: 2015\\n\\nAbstract: The multiplicative Schwarz preconditioned inexact Newton (MSPIN) algorithm is presented as a complement to additive Schwarz preconditioned inexact Newton (ASPIN). At an algebraic level, ASPIN and MSPIN are variants of the same strategy to improve the convergence of systems with unbalanced nonlinearities; however, they have natural complementarity in practice. MSPIN is naturally based on partitioning of degrees of freedom in a nonlinear PDE system by field type rather than by subdomain, where a modest factor of concurrency can be sacrificed for physically motivated convergence robustness. ASPIN, originally introduced for decompositions into subdomains, is natural for high concurrency and reduction of global synchronization. We consider both types of inexact Newton algorithms in the field-split context, and we augment the classical convergence theory of ASPIN for the multiplicative case. Numerical experiments show that MSPIN can be significantly more robust than Newton methods based on global linearizations, and that MSPIN can be more robust than ASPIN and maintain fast convergence even for challenging problems, such as high Reynolds number Navier-Stokes equations.\",\n \"Title: Parallel performance of domain-decomposed preconditioned Krylov methods for PDEs with locally uniform refinement\\nYear: 1992\\n\\nAbstract: Preconditioners based on domain decomposition appear natural for the Krylov solution of implicitly discretized partial differential equations (PDEs) on parallel computers. Two-scale preconditioners (involving a global coarse-grid solve, independent solves over interfaces connecting the coarse-grid points, and independent subdomain solves) have been known since the early 1980s to be \\\"near optimal\\\" in the sense of ensuring a bounded, or at most logarithmically growing, iteration count as the mesh is refined. As a result, the refinement of the mesh can be chosen locally on the basis of truncation error, and the granularity of the domain decomposition can be chosen globally on the basis of parallel computing considerations with only mild effects on the convergence rate of the algorithm. However, overall computational complexity depends not only on the algebraic convergence rate, but also on the operation counts of the components of the preconditioner that must be applied at each iteration. The costs of solving the subdomain systems and the crosspoint system show superlinear growth in their respective (and inversely related) sizes. On the subdomains, the superlinear terms arise from arithmetic only; in the crosspoint system the cost of nonlocal data exchange is also superlinear. These factors make the preconditioner granularity and the choice of its components problem- and machine-dependent compromises. The tradeoffs involved are illustrated through numerical experiments on both shared- and distributed-memory computers for convection-diffusion problems. Because of the development of boundary layers, these problems benefit from local mesh refinement, which is straightforward to accommodate within the domain decomposition framework in a locally uniform sense, but which introduces load balancing as a further consideration in selecting the granularity of the preconditioner. In spite of the tradeoffs, cumulative speedups are obtainable out to at least medium-scale granularity (up to 64 processors in our tests). The largest problems involve O(10(5)) unknowns partitioned into O(10(3)) subdomains and converge in O(10) iterations requiring O(1) seconds on the Intel iPSC/860.\",\n \"Title: Domain decomposition with local mesh refinement\\nYear: 1992\\n\\nAbstract: A preconditioned Krylov iterative algorithm based on domain decomposition for linear systems arising from implicit finite-difference or finite-element discretizations of partial differential equation problems requiring local mesh refinement is described. To keep data structures as simple as possible for parallel computing applications, the fundamental computational unit in the algorithm is defined as a subregion of the domain spanned by a locally uniform tensor-product grid, called a tile. In the tile-based domain decomposition approach, two levels of discretization are considered at each point of the domain: a global coarse grid defined by tile vertices only, and a local fine grid where the degree of resolution can vary from tile to tile. One global level and one local level provide the flexibility required to adaptively discretize a diverse collection of problems on irregular regions and solve them at convergence rates that deteriorate only logarithmically in the finest mesh parameter, with the coarse tessellation held fixed. A logarithmic departure from optimality seems to be a reasonable compromise for the simplicity of the composite grid data structure and concomitant regular data exchange patterns in a multiprocessor environment. Some experiments with up to 1024 tiles are reported, and the evolution of the algorithm is commented on and contrasted with optimal nonrefining two-level algorithms and optimal refining multilevel algorithms. Computational comparisons with some other popular methods are presented.\",\n \"Title: Multi-dimensional intra-tile parallelization for memory-starved stencil computations\\nYear: 2015\\n\\nAbstract: Optimizing the performance of stencil algorithms has been the subject of intense research over the last two decades. Since many stencil schemes have low arithmetic intensity, most optimizations focus on increasing the temporal data access locality, thus reducing the data traffic through the main memory interface with the ultimate goal of decoupling from this bottleneck. There are, however, only a few approaches that explicitly leverage the shared cache feature of modern multicore chips. If every thread works on its private, separate cache block, the available cache space can become too small, and sufficient temporal locality may not be achieved. We propose a flexible multidimensional intratile parallelization method for stencil algorithms on multicore CPUs with a shared outer-level cache. This method leads to a significant reduction in the required cache space without adverse effects from hardware prefetching or TLB shortage. Our Girih framework includes an autotuner to select optimal parameter configurations on the target hardware. We conduct performance experiments on two contemporary Intel processors and compare with the state-of-the-art stencil frameworks Pluto and Pochoir, using four corner-case stencil schemes and a wide range of problem sizes. Girih shows substantial performance advantages and best arithmetic intensity at almost all problem sizes, especially on low-intensity stencils with variable coefficients. We study in detail the performance behavior at varying grid sizes using phenomenological performance modeling. Our analysis of energy consumption reveals that our method can save energy through reduced DRAM bandwidth usage even at a marginal performance gain. It is thus well suited for future architectures that will be strongly challenged by the cost of data movement, be it in terms of performance or energy consumption.\",\n \"Title: Optimization of an Electromagnetics Code with Multicore Wavefront Diamond Blocking and Multi-dimensional Intra-Tile Parallelization\\nYear: 2015\\n\\nAbstract: Understanding and optimizing the properties of solar cells is becoming a key issue in the search for alternatives to nuclear and fossil energy sources. A theoretical analysis via numerical simulations involves solving Maxwell's Equations in discretized form and typically requires substantial computing effort. We start from a hybrid-parallel (MPI+OpenMP) production code that implements the Time Harmonic Inverse Iteration Method (THIIM) with Finite-Difference Frequency Domain (FDFD) discretization. Although this algorithm has the characteristics of a strongly bandwidth-bound stencil update scheme, it is significantly different from the popular stencil types that have been exhaustively studied in the high performance computing literature to date. We apply a recently developed stencil optimization technique, multicore wavefront diamond tiling with multi-dimensional cache block sharing, and describe in detail the peculiarities that need to be considered due to the special stencil structure. Concurrency in updating the components of the electric and magnetic fields provides an additional level of parallelism. The dependence of the cache size requirement of the optimized code on the blocking parameters is modeled accurately, and an auto-tuner searches for optimal configurations in the remaining parameter space. We were able to completely decouple the execution from the memory bandwidth bottleneck, accelerating the implementation by a factor of three to four compared to an optimal implementation with pure spatial blocking on an 18-core Intel Haswell CPU.\",\n \"Title: A Comparison Of Some Domain Decomposition And Ilu Preconditioned Iterative Methods For Nonsymmetric Elliptic Problems\\nYear: 1994\\n\\nAbstract: In recent years, competitive domain-decomposed preconditioned iterative techniques of Krylov-Schwarz type have been developed for nonsymmetric linear elliptic systems. Such systems arise when convection-diffusion-reaction problems from computational fluid dynamics or heat and mass transfer are linearized for iterative solution. Through domain decomposition, a large problem is divided into many smaller problems whose requirements for coordination can be controlled to allow effective solution on parallel machines. A central question is how to choose these small problems and how to arrange the order of their solution. Different specifications of decomposition and solution order lead to a plethora of algorithms possessing complementary advantages and disadvantages. In this report we compare several methods, including the additive Schwarz algorithm, the classical multiplicative Schwarz algorithm, an accelerated multiplicative Schwarz algorithm, the tile algorithm, the CGK algorithm, the CSPD algorithm, and also the popular global ILU-family of preconditioners, on some nonsymmetric or indefinite elliptic model problems discretized by finite difference methods. The preconditioned problems are solved by the unrestarted GMRES method. A version of the accelerated multiplicative Schwarz method is a consistently good performer.\",\n \"Title: A scalable community detection algorithm for large graphs using stochastic block models\\nYear: 2015\\n\\nAbstract: Community detection in graphs is widely used in social and biological networks, and the stochastic block model is a powerful probabilistic tool for describing graphs with community structures. However, in the era of \\\\\\\"big data,\\\\\\\" traditional inference algorithms for such a model are increasingly limited due to their high time complexity and poor scalability. In this paper, we propose a multi-stage maximum likelihood approach to recover the latent parameters of the stochastic block model, in time linear with respect to the number of edges. We also propose a parallel algorithm based on message passing. Our algorithm can overlap communication and computation, providing speedup without compromising accuracy as the number of processors grows. For example, to process a real-world graph with about 1.3 million nodes and 10 million edges, our algorithm requires about 6 seconds on 64 cores of a contemporary commodity Linux cluster. Experiments demonstrate that the algorithm can produce high quality results on both benchmark and real-world graphs. An example of finding more meaningful communities is illustrated consequently in comparison with a popular modularity maximization algorithm.\",\n \"Title: Exploring Shared-Memory Optimizations for an Unstructured Mesh CFD Application on Modern Parallel Systems\\nYear: 2015\\n\\nAbstract: In this work, we revisit the 1999 Gordon Bell Prize winning PETSc-FUN3D aerodynamics code, extending it with highly-tuned shared-memory parallelization and detailed performance analysis on modern highly parallel architectures. An unstructured-grid implicit flow solver, which forms the backbone of computational aerodynamics, poses particular challenges due to its large irregular working sets, unstructured memory accesses, and variable/limited amount of parallelism. This code, based on a domain decomposition approach, exposes tradeoffs between the number of threads assigned to each MPI-rank sub domain, and the total number of domains. By applying several algorithm- and architecture-aware optimization techniques for unstructured grids, we show a 6.9X speed-up in performance on a single-node Intel® XeonTM1 E5 2690 v2 processor relative to the out-of-the-box compilation. Our scaling studies on TACC Stampede supercomputer show that our optimizations continue to provide performance benefits over baseline implementation as we scale up to 256 nodes.\",\n \"Title: High Performance Multi-GPU SpMV for Multi-component PDE-Based Applications.\\nYear: 2015\\n\\nAbstract: Leveraging optimization techniques (e.g., register blocking and double buffering) introduced in the context of KBLAS, a Level 2 BLAS high performance library on GPUs, the authors implement dense matrix-vector multiplications within a sparse-block structure. While these optimizations are important for high performance dense kernel executions, they are even more critical when dealing with sparse linear algebra operations. The most time-consuming phase of many multicomponent applications, such as models of reacting flows or petroleum reservoirs, is the solution at each implicit time step of large, sparse spatially structured or unstructured linear systems. The standard method is a preconditioned Krylov solver. The Sparse Matrix-Vector multiplication (SpMV) is, in turn, one of the most time-consuming operations in such solvers. Because there is no data reuse of the elements of the matrix within a single SpMV, kernel performance is limited by the speed at which data can be transferred from memory to registers, making the bus bandwidth the major bottleneck. On the other hand, in case of a multi-species model, the resulting Jacobian has a dense block structure. For contemporary petroleum reservoir simulations, the block size typically ranges from three to a few dozen among different models, and still larger blocks are relevant within adaptively model-refined regions of the domain, though generally the size of the blocks, related to the number of conserved species, is constant over large regions within a given model. This structure can be exploited beyond the convenience of a block compressed row data format, because it offers opportunities to hide the data motion with useful computations. The new SpMV kernel outperforms existing state-of-the-art implementations on single and multi-GPUs using matrices with dense block structure representative of porous media applications with both structured and unstructured multi-component grids.\",\n \"Title: Performance optimization of Sparse Matrix-Vector Multiplication for multi-component PDE-based applications using GPUs.\\nYear: 2016\\n\\nAbstract: Simulations of many multi-component PDE-based applications, such as petroleum reservoirs or reacting flows, are dominated by the solution, on each time step and within each Newton step, of large sparse linear systems. The standard solver is a preconditioned Krylov method. Along with application of the preconditioner, memory-bound Sparse Matrix-Vector Multiplication SpMV is the most time-consuming operation in such solvers. Multi-species models produce Jacobians with a dense block structure, where the block size can be as large as a few dozen. Failing to exploit this dense block structure vastly underutilizes hardware capable of delivering high performance on dense BLAS operations. This paper presents a GPU-accelerated SpMV kernel for block-sparse matrices. Dense matrix-vector multiplications within the sparse-block structure leverage optimization techniques from the KBLAS library, a high performance library for dense BLAS kernels. The design ideas of KBLAS can be applied to block-sparse matrices. Furthermore, a technique is proposed to balance the workload among thread blocks when there are large variations in the lengths of nonzero rows. Multi-GPU performance is highlighted. The proposed SpMV kernel outperforms existing state-of-the-art implementations using matrices with real structures from different applications. Copyright © 2016 John Wiley & Sons, Ltd.\",\n \"Title: Efficient Sphere Detector Algorithm for Massive MIMO using GPU Hardware Accelerator.\\nYear: 2016\\n\\nAbstract: To further enhance the capacity of next generation wireless communication systems, massive multiple-input multiple-output (MIMO) has recently appeared as a necessary enabling technology to achieve high performance signal processing for large-scale multiple antennas. However, massive MIMO systems inevitably generate signal processing overheads, which translate into ever-increasing rate of complexity, and therefore, such systems may not maintain the inherent real-time requirement of wireless systems. We redesign the nonlinear sphere decoder method to increase the performance of the system, cast most memory-bound computations into compute-bound operations to reduce the overall complexity, and maintain the real-time processing thanks to the graphics processing unit (GPU) computational power. We show a comprehensive complexity and performance analysis on an unprecedented MIMO system scale, which can ease the design phase toward simulating future massive MIMO wireless systems.\",\n \"Title: High Performance Polar Decomposition on Distributed Memory Systems.\\nYear: 2016\\n\\nAbstract: The polar decomposition of a dense matrix is an important operation in linear algebra. It can be directly calculated through the singular value decomposition SVD or iteratively using the QR dynamically-weighted Halley algorithm QDWH. The former is difficult to parallelize due to the preponderant number of memory-bound operations during the bidiagonal reduction. We investigate the latter scenario, which performs more floating-point operations but exposes at the same time more parallelism, and therefore, runs closer to the theoretical peak performance of the system, thanks to more compute-bound matrix operations. Profiling results show the performance scalability of QDWH for calculating the polar decomposition using around 9200 MPI processes on well and ill-conditioned matrices of 100ï¾źKï¾ź$$\\\\\\\\times $$ï¾ź100ï¾źK problem size. We study then the performance impact of the QDWH-based polar decomposition as a pre-processing step toward calculating the SVD itself. The new distributed-memory implementation of the QDWH-SVD solver achieves up to five-fold speedup against current state-of-the-art vendor SVD implementations.\",\n \"Title: Redesigning Triangular Dense Matrix Computations on GPUs.\\nYear: 2016\\n\\nAbstract: A new implementation of the triangular matrix-matrix multiplication TRMM and the triangular solve TRSM kernels are described on GPU hardware accelerators. Although part of the Level 3 BLAS family, these highly computationally intensive kernels fail to achieve the percentage of the theoretical peak performance on GPUs that one would expect when running kernels with similar surface-to-volume ratio on hardware accelerators, i.e., the standard matrix-matrix multiplication GEMM. The authors propose adopting a recursive formulation, which enriches the TRMM and TRSM inner structures with GEMM calls and, therefore, reduces memory traffic while increasing the level of concurrency. The new implementation enables efficient use of the GPU memory hierarchy and mitigates the latency overhead, to run at the speed of the higher cache levels. Performance comparisons show upï¾źto eightfold and twofold speedups for large dense matrix sizes, against the existing state-of-the-art TRMM and TRSM implementations from NVIDIA cuBLAS, respectively, across various GPU generations. Once integrated into high-level Cholesky-based dense linear algebra algorithms, the performance impact on the overall applications demonstrates upï¾źto fourfold and twofold speedups, against the equivalent native implementations, linked with cuBLAS TRMM and TRSM kernels, respectively. The new TRMM/TRSM kernel implementations are part of the open-source KBLAS software library http://ecrc.kaust.edu.sa/Pages/Res-kblas.aspx and are lined up for integration into the NVIDIA cuBLAS library in the upcoming v8.0 release.\",\n \"Title: Research and Education in Computational Science and Engineering.\\nYear: 2018\\n\\nAbstract: This report presents challenges, opportunities, and directions for computational science and engineering (CSE) research and education for the next decade. Over the past two decades the field of CSE has penetrated both basic and applied research in academia, industry, and laboratories to advance discovery, optimize systems, support decision-makers, and educate the scientific and engineering workforce. Informed by centuries of theory and experiment, CSE performs computational experiments to answer questions that neither theory nor experiment, alone is equipped to answer. CSE provides scientists and engineers with algorithmic inventions and software systems that, transcend disciplines and scales. CSE brings the power of parallelism to bear on troves of data. Mathematics-based advanced computing has become a prevalent means of discovery and innovation in essentially all areas of science, engineering, technology, and society, and the CSE community is at the core of this transformation. However, a combination of disruptive developments-including the architectural complexity of extreme-scale computing, the data revolution and increased attention to data-driven discovery, and the specialization required to follow the applications to new frontiers-is redefining the scope and reach of the CSE endeavor. With these many current and expanding opportunities for the CSE field, there is a growing demand for CSE graduates and a need to expand CSE educational offerings. This need includes CSE programs at both the undergraduate and graduate levels, as well as continuing education and professional development programs, exploiting the synergy between computational science and data science. Yet, as institutions consider new and evolving educational programs, it is essential to consider the broader research challenges and opportunities that provide the context for CSE education and workforce development.\",\n \"Title: Convergence Analysis for the Multiplicative Schwarz Preconditioned Inexact Newton Algorithm.\\nYear: 2016\\n\\nAbstract: The multiplicative Schwarz preconditioned inexact Newton (MSPIN) algorithm, based on decomposition by field type rather than by subdomain, was recently introduced to improve the convergence of systems with unbalanced nonlinearities. This paper provides a convergence analysis of the MSPIN algorithm. Under reasonable assumptions, it is shown that MSPIN is locally convergent, and desired superlinear or even quadratic convergence can be obtained when the forcing terms are picked suitably.\",\n \"Title: Unstructured computational aerodynamics on many integrated core architecture.\\nYear: 2016\\n\\nAbstract: Porting implicit unstructured mesh code onto Intel Xeon Phi Knights Corner.Employing aggressive optimizations to improve the CFD flux kernel on Phi hardware.Exploring different thread affinity modes with different programming paradigms.Achieving 3.8x speedup using the offload mode compare to the baseline.Achieving 5x speedup using the native mode compare to the baseline. Shared memory parallelization of the flux kernel of PETSc-FUN3D, an unstructured tetrahedral mesh Euler flow code previously studied for distributed memory and multi-core shared memory, is evaluated on up to 61 cores per node and up to 4 threads per core. We explore several thread-level optimizations to improve flux kernel performance on the state-of-the-art many integrated core (MIC) Intel processor Xeon Phi Knights Corner, with a focus on strong thread scaling. While the linear algebraic kernel is bottlenecked by memory bandwidth for even modest numbers of cores sharing a common memory, the flux kernel, which arises in the control volume discretization of the conservation law residuals and in the formation of the preconditioner for the Jacobian by finite-differencing the conservation law residuals, is compute-intensive and is known to exploit effectively contemporary multi-core hardware. We extend study of the performance of the flux kernel to the Xeon Phi in three thread affinity modes, namely scatter, compact, and balanced, in both offload and native mode, with and without various code optimizations to improve alignment and reduce cache coherency penalties. Relative to baseline out-of-the-box optimized compilation, code restructuring optimizations provide about 3.8x speedup using the offload mode and about 5x speedup using the native mode. Even with these gains for the flux kernel, with respect to execution time the MIC simply achieves par with optimized compilation on a contemporary multi-core Intel CPU, the 16-core Sandy Bridge E5 2670. Nevertheless, the optimizations employed to reduce the data motion and cache coherency protocol penalties of the MIC are expected to be of value for CFD and many other unstructured applications as many-core architecture evolves. We explore large-scale distributed-shared memory performance on the Cray XC40 supercomputer, to demonstrate that optimizations employed on Phi hybridize to this context, where each of thousands of nodes are comprised of two sockets of Intel Xeon Haswell CPUs with 32 cores per node.\",\n \"Title: A framework for dense triangular matrix kernels on various manycore architectures.\\nYear: 2017\\n\\nAbstract: We present a new high-performance framework for dense triangular Basic Linear Algebra Subroutines (BLAS) kernels, ie, triangular matrix-matrix multiplication (TRMM) and triangular solve (TRSM), on various manycore architectures. This is an extension of a previous work on a single GPU by the same authors, presented at the EuroPar'16 conference, in which we demonstrated the effectiveness of recursive formulations in enhancing the performance of these kernels. In this paper, the performance of triangular BLAS kernels on a single GPU is further enhanced by implementing customized in-place CUDA kernels for TRMM and TRSM, which are called at the bottom of the recursion. In addition, a multi-GPU implementation of TRMM and TRSM is proposed and we show an almost linear performance scaling, as the number of GPUs increases. Finally, the algorithmic recursive formulation of these triangular BLAS kernels is in fact oblivious to the targeted hardware architecture. We, therefore, port these recursive kernels to homogeneous x86 hardware architectures by relying on the vendor optimized BLAS implementations. Results reported on various hardware architectures highlight a significant performance improvement against state-of-the-art implementations. These new kernels are freely available in the KAUST BLAS (KBLAS) open-source library at .\",\n \"Title: On the Robustness and Prospects of Adaptive BDDC Methods for Finite Element Discretizations of Elliptic PDEs with High-Contrast Coefficients.\\nYear: 2016\\n\\nAbstract: Balancing Domain Decomposition by Constraints (BDDC) methods have proven to be powerful preconditioners for large and sparse linear systems arising from the finite element discretization of elliptic PDEs. Condition number bounds can be theoretically established that are independent of the number of subdomains of the decomposition. The core of the methods resides in the design of a larger and partially discontinuous finite element space that allows for fast application of the preconditioner, where Cholesky factorizations of the subdomain finite element problems are additively combined with a coarse, global solver. Multilevel and highly-scalable algorithms can be obtained by replacing the coarse Cholesky solver with a coarse BDDC preconditioner. BDDC methods have the remarkable ability to control the condition number, since the coarse space of the preconditioner can be adaptively enriched at the cost of solving local eigenproblems. The proper identification of these eigenproblems extends the robustness of the methods to any heterogeneity in the distribution of the coefficients of the PDEs, not only when the coefficients jumps align with the subdomain boundaries or when the high contrast regions are confined to lie in the interior of the subdomains. The specific adaptive technique considered in this paper does not depend upon any interaction of discretization and partition; it relies purely on algebraic operations. Coarse space adaptation in BDDC methods has attractive algorithmic properties, since the technique enhances the concurrency and the arithmetic intensity of the preconditioning step of the sparse implicit solver with the aim of controlling the number of iterations of the Krylov method in a black-box fashion, thus reducing the number of global synchronization steps and matrix vector multiplications needed by the iterative solver; data movement and memory bound kernels in the solve phase can be thus limited at the expense of extra local flops during the setup of the preconditioner. This paper presents an exposition of the BDDC algorithm that identifies the current computational bottlenecks that could prevent it from being competitive with other solvers, and proposes solutions in anticipation of exascale architectures. Furthermore, the discussion aims to give interested practitioners sufficient insights to decide whether or not to pursue BDDC in their applications. In addition, the work presents novel numerical results using the distributed memory implementation of BDDC in the PETSc library for vector field problems arising in the context of porous media flows and electromagnetic modeling; the results provide evidence of the robustness of these methods for highly heterogenous problems and non-conforming discretizations.\",\n \"Title: ExaGeoStat: A High Performance Unified Framework for Geostatistics on Manycore Systems.\\nYear: 2017\\n\\nAbstract: We present ExaGeoStat , a high performance software for geospatial statistics in climate and environment modeling. In contrast to simulation based on partial differential equations derived from first-principles modeling, ExaGeoStat employs a statistical model based on the evaluation of the Gaussian log-likelihood function, which operates on a large dense covariance matrix. Generated by the parametrizable Matern covariance function, the resulting matrix is symmetric and positive definite. The computational tasks involved during the evaluation of the Gaussian log-likelihood function become daunting as the number $n$ of geographical locations grows, as ${mathcal O}(n^2)$ storage and ${mathcal O}(n^3)$ operations are required. While many approximation methods have been devised from the side of statistical modeling to ameliorate these polynomial complexities, we are interested here in the complementary approach of evaluating the exact algebraic result by exploiting advances in solution algorithms and many-core computer architectures. Using state-of-the-art high performance dense linear algebra libraries associated with various leading edge parallel architectures (Intel KNLs, NVIDIA GPUs, and distributed-memory systems), ExaGeoStat raises the game for statistical applications from climate and environmental science. ExaGeoStat provides a reference evaluation of statistical parameters, with which to assess the validity of the various approaches based on approximation. The software takes a first step in the merger of large-scale data analytics and extreme computing for geospatial statistical applications, to be followed by additional complexity reducing improvements from the solver side that can be implemented under the same interface. Thus, a single uncompromised statistical model can ultimately be executed in a wide variety of emerging exascale environments.\",\n \"Title: Asynchronous Task-Based Polar Decomposition on Single Node Manycore Architectures.\\nYear: 2018\\n\\nAbstract: This paper introduces the first asynchronous, task-based formulation of the polar decomposition and its corresponding implementation on manycore architectures. Based on a new formulation of the iterative QR dynamically-weighted Halley algorithm (QDWH) for the calculation of the polar decomposition, the proposed implementation replaces the original and hostile LU factorization for the condition num...\",\n \"Title: Fast multipole preconditioners for sparse matrices arising from elliptic equations.\\nYear: 2018\\n\\nAbstract: Among optimal hierarchical algorithms for the computational solution of elliptic problems, the fast multipole method (FMM) stands out for its adaptability to emerging architectures, having high arithmetic intensity, tunable accuracy, and relaxable global synchronization requirements. We demonstrate that, beyond its traditional use as a solver in problems for which explicit free-space kernel representations are available, the FMM has applicability as a preconditioner in finite domain elliptic boundary value problems, by equipping it with boundary integral capability for satisfying conditions at finite boundaries and by wrapping it in a Krylov method for extensibility to more general operators. Here, we do not discuss the well developed applications of FMM to implement matrix-vector multiplications within Krylov solvers of boundary element methods. Instead, we propose using FMM for the volume-to-volume contribution of inhomogeneous Poisson-like problems, where the boundary integral is a small part of the overall computation. Our method may be used to precondition sparse matrices arising from finite difference/element discretizations, and can handle a broader range of scientific applications. It is capable of algebraic convergence rates down to the truncation error of the discretized PDE comparable to those of multigrid methods, and it offers potentially superior multicore and distributed memory scalability properties on commodity architecture supercomputers. Compared with other methods exploiting the low-rank character of off-diagonal blocks of the dense resolvent operator, FMM-preconditioned Krylov iteration may reduce the amount of communication because it is matrix-free and exploits the tree structure of FMM. We describe our tests in reproducible detail with freely available codes and outline directions for further extensibility.\",\n \"Title: Extreme Scale FMM-Accelerated Boundary Integral Equation Solver for Wave Scattering.\\nYear: 2018\\n\\nAbstract: We present a boundary integral equation solver for wave scattering suited for many-core processors, which are expected to be the building blocks of energy-austere exascale systems, and on which algorithmic and architecture-oriented optimizations are essential for achieving worthy performance. We use the GMRES iterative method and the fast multipole method (FMM) to implement the matrix-vector product (MatVec) kernel. We implement highly optimized kernels for both shared- and distributed-memory architectures and provide various performance models for tuning the task-based tree traversal implementation of FMM. We develop optimal architecture-specific and algorithm-aware partitioning, load balancing, and communication reducing mechanisms to scale up to 6,144 compute nodes of a Cray XC40 with 196,608 hardware cores. Task-based traversal achieves roughly 77% and 60% of the peak single precision floating point performance of a 56-core Skylake Scalable processor and a 72-core Knights Landing processor, respectively. These represent approximately sevenfold speedups compared to the baseline scalar code. We report weak scalability in accordance with the best possible O(log P) communication complexity and the theoretical scaling complexity of FMM. These result in up to 85% efficiency in strong scaling while computing in excess of 2 billion degrees of freedom (DoFs) on the full-scale Cray XC40. To demonstrate the applicability of the proposed implementation of FMM, we use it in the solution of the acoustic scattering problem involving soft targets. This calls for the discretization of an integral equation with an oscillatory kernel and (weak) 1/R singularity. Applications of FMM to the problems involving hard targets, which call for implementation of more complicated singularity extraction/cancellation schemes to account for 1/R-2 singularity, or to Burton-Miller surface integral formulation, which results in a well-conditioned matrix system, are not considered in this work, but the efficiency and scalability results demonstrated here are expected to be observed for these cases as well. The numerical results are converged to within 1.0e-4 relative 2-norm residual accuracy of the analytical solution for the sphere, and a multiple flower-shaped scatterer displays comparable algebraic convergence in a more general setting.\",\n \"Title: Accelerated Cyclic Reduction: A distributed-memory fast solver for structured linear systems.\\nYear: 2018\\n\\nAbstract: We present Accelerated Cyclic Reduction (ACR), a distributed-memory fast solver for rank-compressible block tridiagonal linear systems arising from the discretization of elliptic operators, developed here for three dimensions. Algorithmic synergies between Cyclic Reduction and hierarchical matrix arithmetic operations result in a solver that has O(kNlogN(logN+k2)) arithmetic complexity and O(k Nlog N) memory footprint, where N is the number of degrees of freedom and k is the rank of a block in the hierarchical approximation, and which exhibits substantial concurrency. We provide a baseline for performance and applicability by comparing with the multifrontal method with and without hierarchical semi-separable matrices, with algebraic multigrid and with the classic cyclic reduction method. Over a set of large-scale elliptic systems with features of nonsymmetry and indefiniteness, the robustness of the direct solvers extends beyond that of the multigrid solver, and relative to the multifrontal approach ACR has lower or comparable execution time and size of the factors, with substantially lower numerical ranks. ACR exhibits good strong and weak scaling in a distributed context and, as with any direct solver, is advantageous for problems that require the solution of multiple right-hand sides. Numerical experiments show that the rank k patterns are of O(1) for the Poisson equation and of O(n) for the indefinite Helmholtz equation. The solver is ideal in situations where low-accuracy solutions are sufficient, or otherwise as a preconditioner within an iterative method.\",\n \"Title: Tile Low-Rank Approximation of Large-Scale Maximum Likelihood Estimation on Manycore Architectures.\\nYear: 2018\\n\\nAbstract: Maximum likelihood estimation is an important statistical technique for estimating missing data, for example in climate and environmental applications, which are usually large and feature data points that are irregularly spaced. In particular, the Gaussian log-likelihood function is the emph{de facto} model, which operates on the resulting sizable dense covariance matrix. The advent of high performance systems with advanced computing power and memory capacity have enabled full simulations only for rather small dimensional climate problems, solved at the machine precision accuracy. The challenge for high dimensional problems lies in the computation requirements of the log-likelihood function, which necessitates ${mathcal O}(n^2)$ storage and ${mathcal O}(n^3)$ operations, where $n$ represents the number of given spatial locations. This prohibitive computational cost may be reduced by using approximation techniques that not only enable large-scale simulations otherwise intractable but also maintain the accuracy and the fidelity of the spatial statistics model. In this paper, we extend the Exascale GeoStatistics software framework (i.e., ExaGeoStat) to support the Tile Low-Rank (TLR) approximation technique, which exploits the data sparsity of the dense covariance matrix by compressing the off-diagonal tiles up to a user-defined accuracy threshold. The underlying linear algebra operations may then be carried out on this data compression format, which may ultimately reduce the arithmetic complexity of the maximum likelihood estimation and the corresponding memory footprint. Performance results of TLR-based computations on shared and distributed-memory systems attain up to 13X and 5X speedups, respectively, compared to full accuracy simulations using synthetic and real datasets (up to 2M), while ensuring adequate prediction accuracy.\",\n \"Title: Fast parallel multidimensional FFT using advanced MPI.\\nYear: 2018\\n\\nAbstract: We present a new method for performing global redistributions of multidimensional arrays essential to parallel fast Fourier (or similar) transforms. Traditional methods use standard all-to-all collective communication of contiguous memory buffers, thus necessarily requiring local data realignment steps intermixed in-between redistribution and transform steps. Instead, our method takes advantage of subarray datatypes and generalized all-to-all scatter/gather from the MPI-2 standard to communicate discontiguous memory buffers, effectively eliminating the need for local data realignments. Despite generalized all-to-all communication of discontiguous data being generally slower, our proposal economizes in local work. For a range of strong and weak scaling tests, we found the overall performance of our method to be on par and often better than well-established libraries like MPI-FFTW, P3DFFT, and 2DECOMP&FFT. We provide compact routines implemented at the highest possible level using the MPI bindings for the C programming language. These routines apply to any global redistribution, over any two directions of a multidimensional array, decomposed on arbitrary Cartesian processor grids (1D slabs, 2D pencils, or even higher-dimensional decompositions). The high level implementation makes the code easy to read, maintain, and eventually extend. Our approach enables for future speedups from optimizations in the internal datatype handling engines within MPI implementations.\",\n \"Title: Reconstructing parameters of the FitzHugh–Nagumo system from boundary potential measurements\\nYear: 2007\\n\\nAbstract: We consider distributed parameter identification problems for the FitzHugh–Nagumo model of electrocardiology. The model describes the evolution of electrical potentials in heart tissues. The mathematical problem is to reconstruct physical parameters of the system through partial knowledge of its solutions on the boundary of the domain. We present a parallel algorithm of Newton–Krylov type that combines Newton’s method for numerical optimization with Krylov subspace solvers for the resulting Karush–Kuhn–Tucker system. We show by numerical simulations that parameter reconstruction can be performed from measurements taken on the boundary of the domain only. We discuss the effects of various model parameters on the quality of reconstructions.\",\n \"Title: Parallel accelerated cyclic reduction preconditioner for three-dimensional elliptic PDEs with variable coefficients.\\nYear: 2018\\n\\nAbstract: We present a robust and scalable preconditioner for the solution of large-scale linear systems that arise from the discretization of elliptic PDEs amenable to rank compression. The preconditioner is based on hierarchical low-rank approximations and the cyclic reduction method. The setup and application phases of the preconditioner achieve log-linear complexity in memory footprint and number of operations, and numerical experiments exhibit good weak and strong scalability at large processor counts in a distributed memory environment. Numerical experiments with linear systems that feature symmetry and nonsymmetry, definiteness and indefiniteness, constant and variable coefficients demonstrate the preconditioner applicability and robustness. Furthermore, it is possible to control the number of iterations via the accuracy threshold of the hierarchical matrix approximations and their arithmetic operations, and the tuning of the admissibility condition parameter. Together, these parameters allow for optimization of the memory requirements and performance of the preconditioner.\",\n \"Title: Real-Time Massively Distributed Multi-object Adaptive Optics Simulations for the European Extremely Large Telescope\\nYear: 2018\\n\\nAbstract: The European Extremely Large Telescope (E-ELT) is one of today's most challenging projects in ground-based astronomy. Addressing one of the key science cases for the E-ELT, the study of the early Universe, requires the implementation of multi-object adaptive optics (MOAO), a dedicated concept relying on turbulence tomography. We use a novel pseudo-analytical approach to simulate the performance of tomographic reconstruction of the atmospheric turbulence in an MOAO system on real datasets. We simulate simultaneously 4K galaxies in a common field of view on massively parallel supercomputers during a single night of observations. We are able to generate a first-ever high-resolution galaxy map at almost a real-time throughput. This simulation scale opens new research horizons in numerical methods for experimental astronomy, some core components of the pipeline standing as pathfinders toward actual operations and future astronomic discoveries on the E-ELT.\",\n \"Title: Tile Low-Rank GEMM Using Batched Operations on GPUs.\\nYear: 2018\\n\\nAbstract: Dense General Matrix-Matrix (GEMM) multiplication is a core operation of the Basic Linear Algebra Subroutines (BLAS) library, and therefore, often resides at the bottom of the traditional software stack for many scientific applications. In fact, chip manufacturers give a special attention to the GEMM kernel implementation since this is exactly where most of the high-performance software libraries extract hardware performance. With the emergence of big data applications involving large data-sparse, hierarchically low-rank matrices, the off-diagonal tiles can be compressed to reduce the algorithmic complexity and the memory footprint. The resulting tile low-rank (TLR) data format is composed of small data structures, which retain the most significant information for each tile. However, to operate on low-rank tiles, a new GEMM operation and its corresponding API have to be designed on GPUs so that the data sparsity structure of the matrix can be exploited while leveraging the underlying TLR compression format. The main idea consists of aggregating all operations into a single kernel launch to compensate for their low arithmetic intensities and to mitigate the data transfer overhead on GPUs. The new TLR-GEMM kernel outperforms the cuBLAS dense batched GEMM by more than an order of magnitude and creates new opportunities for TLR advanced algorithms.\",\n \"Title: Exploiting Data Sparsity for Large-Scale Matrix Computations.\\nYear: 2018\\n\\nAbstract: Exploiting data sparsity in dense matrices is an algorithmic bridge between architectures that are increasingly memory-austere on a per-core basis and extreme-scale applications. In this work, we leverage the Hierarchical matrix Computations on Manycore Architectures (HiCMA) library in order to tackle this challenging problem by achieving significant reductions in time to solution and memory footprint, while preserving a specified accuracy requirement of the application. We have extended HiCMA to provide a high-performance implementation on distributed-memory systems of one of the most widely used matrix factorization in large-scale scientific applications, i.e., the Cholesky factorization. It employs the tile low-rank data format to compress the dense data-sparse off-diagonal tiles of the matrix. It then decomposes the matrix computations into interdependent tasks and relies on the dynamic runtime system StarPU for asynchronous out-of-order scheduling, while allowing high user productivity. Performance comparisons and memory footprint on matrix dimensions up to eleven million show a performance gain and memory saving of more than an order of magnitude for both metrics on thousands of cores, against state-of-the-art open-source and vendor optimized numerical libraries. This represents an important milestone in enabling large-scale matrix computations toward solving big data problems in geospatial statistics for climate/weather forecasting applications.\",\n \"Title: Optimizations of Unstructured Aerodynamics Computations for Many-core Architectures.\\nYear: 2018\\n\\nAbstract: We investigate several state-of-the-practice shared-memory optimization techniques applied to key routines of an unstructured computational aerodynamics application with irregular memory accesses. We illustrate for the Intel Knights Landing processor, as a representative of the processors in contemporary leading supercomputers, identifying and addressing performance challenges without compromising...\",\n \"Title: Parallel Approximation of the Maximum Likelihood Estimation for the Prediction of Large-Scale Geostatistics Simulations\\nYear: 2018\\n\\nAbstract: Maximum likelihood estimation is an important statistical technique for estimating missing data, for example in climate and environmental applications, which are usually large and feature data points that are irregularly spaced. In particular, the Gaussian log-likelihood function is the de facto model, which operates on the resulting sizable dense covariance matrix. The advent of high performance systems with advanced computing power and memory capacity have enabled full simulations only for rather small dimensional climate problems, solved at the machine precision accuracy. The challenge for high dimensional problems lies in the computation requirements of the log-likelihood function, which necessitates O(n\\n<sup xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">2</sup>\\n) storage and O(n\\n<sup xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">3</sup>\\n) operations, where n represents the number of given spatial locations. This prohibitive computational cost may be reduced by using approximation techniques that not only enable large-scale simulations otherwise intractable, but also maintain the accuracy and the fidelity of the spatial statistics model. In this paper, we extend the Exascale GeoStatistics software framework (i.e., ExaGeoStat\\n<sup xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">1</sup>\\n) to support the Tile Low-Rank (TLR) approximation technique, which exploits the data sparsity of the dense covariance matrix by compressing the off-diagonal tiles up to a user-defined accuracy threshold. The underlying linear algebra operations may then be carried out on this data compression format, which may ultimately reduce the arithmetic complexity of the maximum likelihood estimation and the corresponding memory footprint. Performance results of TLR-based computations on shared and distributed-memory systems attain up to 13X and 5X speedups, respectively, compared to full accuracy simulations using synthetic and real datasets (up to 2M), while ensuring adequate prediction accuracy.\",\n \"Title: ExaGeoStat: A High Performance Unified Software for Geostatistics on Manycore Systems.\\nYear: 2018\\n\\nAbstract: We present ExaGeoStat, a high performance software for geospatial statistics in climate and environment modeling. In contrast to simulation based on partial differential equations derived from first-principles modeling, ExaGeoStat employs a statistical model based on the evaluation of the Gaussian log-likelihood function, which operates on a large dense covariance matrix. Generated by the parametr...\",\n \"Title: SBP-SAT finite difference discretization of acoustic wave equations on staggered block-wise uniform grids.\\nYear: 2019\\n\\nAbstract: We consider the numerical simulation of the acoustic wave equations arising from seismic applications, for which staggered grid finite difference methods are popular choices due to their simplicity and efficiency. We relax the uniform grid restriction on finite difference methods and allow the grids to be block-wise uniform with nonconforming interfaces. In doing so, variations in the wave speeds of the subterranean media can be accounted for more efficiently. Staggered grid finite difference operators satisfying the summation-by-parts (SBP) property are devised to approximate the spatial derivatives appearing in the acoustic wave equation. These operators are applied within each block independently. The coupling between blocks is achieved through simultaneous approximation terms (SATs), which impose the interface conditions weakly, i.e., by penalty. Ratio of the grid spacing of neighboring blocks is allowed to be rational number, for which specially designed interpolation formulas are presented. These interpolation formulas constitute key pieces of the simultaneous approximation terms. The overall discretization is shown to be energy-conserving and examined on test cases of both theoretical and practical interests, delivering accurate and stable simulation results.\",\n \"Title: Performance Assessment of Hybrid Parallelism for Large-Scale Reservoir Simulation on Multi- and Many-core Architectures\\nYear: 2018\\n\\nAbstract: Two trends are reshaping the landscape of petroleum reservoir simulators, one architecturally and one application driven: an increasing number of cores per node and increasing computational intensity arising from higher fidelity physics at each cell. Implicit algebraic solvers being the dominant kernels, we present hybrid MPI and OpenMP implementations of the linear solver of GigaPOWERS, a full-scale real-world massively parallel simulator for black oil and composition models. We also evaluate the impact of explicit communication and computation overlap by including the halo exchange in the task-dependency graph. We analyze the performance of these modifications across multi- and many-core architectures, i.e., Intel Haswell, Skylake, and Knights Landing, using a variety of synthetic and real-world models. The hybrid approach results in up to 50% reduction of time to solution on a 16 million-cell SPE10-like model on Skylake whereas on a smaller, 1 million-cell, model on Haswell and Knights Landing both implementations achieve very similar performance. In the real-world reservoir simulations, the hybrid parallelism has reduced communication volume, memory consumption, and improved load balancing.\",\n \"Title: Combining finite element and finite difference methods for isotropic elastic wave simulations in an energy-conserving manner.\\nYear: 2019\\n\\nAbstract: We consider numerical simulation of the isotropic elastic wave equations arising from seismic applications with non-trivial land topography. The more flexible finite element method is applied to the shallow region of the simulation domain to account for the topography, and combined with the more efficient finite difference method that is applied to the deep region of the simulation domain. We demonstrate that these two discretization methods, albeit starting from different formulations of the elastic wave equation, can be joined together smoothly via weakly imposed interface conditions. Discrete energy analysis is employed to derive the proper interface treatment, leading to an overall discretization that is energy-conserving. Numerical examples are presented to demonstrate the efficacy of the proposed interface treatment.\",\n \"Title: Tucker Tensor Analysis of Matérn Functions in Spatial Statistics.\\nYear: 2019\\n\\nAbstract: In this work, we describe advanced numerical tools for working with multivariate functions and for the analysis of large data sets. These tools will drastically reduce the required computing time and the storage cost, and, therefore, will allow us to consider much larger data sets or finer meshes. Covariance matrices are crucial in spatio-temporal statistical tasks, but are often very expensive to compute and store, especially in three dimensions. Therefore, we approximate covariance functions by cheap surrogates in a low-rank tensor format. We apply the Tucker and canonical tensor decompositions to a family of Matern- and Slater-type functions with varying parameters and demonstrate numerically that their approximations exhibit exponentially fast convergence. We prove the exponential convergence of the Tucker and canonical approximations in tensor rank parameters. Several statistical operations are performed in this low-rank tensor format, including evaluating the conditional covariance matrix, spatially averaged estimation variance, computing a quadratic form, determinant, trace, loglikelihood, inverse, and Cholesky decomposition of a large covariance matrix. Low-rank tensor approximations reduce the computing and storage costs essentially. For example, the storage cost is reduced from an exponential O(n(d)) to a linear scaling O(drn), where d is the spatial dimension, n is the number of mesh points in one direction, and r is the tensor rank. Prerequisites for applicability of the proposed techniques are the assumptions that the data, locations, and measurements lie on a tensor (axes-parallel) grid and that the covariance function depends on a distance, parallel to x - y parallel to.\",\n \"Title: Batched Triangular Dense Linear Algebra Kernels for Very Small Matrix Sizes on GPUs\\nYear: 2019\\n\\nAbstract: Batched dense linear algebra kernels are becoming ubiquitous in scientific applications, ranging from tensor contractions in deep learning to data compression in hierarchical low-rank matrix approximation. Within a single API call, these kernels are capable of simultaneously launching up to thousands of similar matrix computations, removing the expensive overhead of multiple API calls while increasing the occupancy of the underlying hardware. A challenge is that for the existing hardware landscape (x86, GPUs, etc.), only a subset of the required batched operations is implemented by the vendors, with limited support for very small problem sizes. We describe the design and performance of a new class of batched triangular dense linear algebra kernels on very small data sizes (up to 256) using single and multiple GPUs. By deploying recursive formulations, stressing the register usage, maintaining data locality, reducing threads synchronization, and fusing successive kernel calls, the new batched kernels outperform existing state-of-the-art implementations.\",\n \"Title: A Note on Adaptive Nonlinear Preconditioning Techniques.\\nYear: 2018\\n\\nAbstract: Nonlinear preconditioning is a globalization technique for Newton's method applied to systems of equations with unbalanced nonlinearities, in which nonlinear residual norm reduction stagnates due to slowly evolving subsets of the degrees of freedom. Even though the Newton corrections may effectively be sparse, a standard Newton method still requires large ill-conditioned linear systems resulting from global linearizations of the nonlinear residual to be solved at each step. Nonlinear preconditioners may enable faster global convergence by shifting work to where it is most strategic, on subsets of the original system. They require additional computation per outer iteration while aiming for many fewer outer iterations and correspondingly fewer global synchronizations. In this work, we improve upon previous nonlinear preconditioning implementations by introducing parameters that allow turning off nonlinear preconditioning during outer Newton iterations where it is not needed. Numerical experiments show that the adaptive nonlinear preconditioning algorithm has performance similar to monolithically applied nonlinear preconditioning, preserving robustness for some challenging problems representative of several PDE-based applications while saving work on nonlinear subproblems.\",\n \"Title: Hierarchical-block conditioning approximations for high-dimensional multivariate normal probabilities\\nYear: 2019\\n\\nAbstract: This paper presents a new method to estimate large-scale multivariate normal probabilities. The approach combines a hierarchical representation with processing of the covariance matrix that decomposes the n-dimensional problem into a sequence of smaller m-dimensional ones. It also includes a d-dimensional conditioning method that further decomposes the m-dimensional problems into smaller d-dimensional problems. The resulting two-level hierarchical-block conditioning method requires Monte Carlo simulations to be performed only in d dimensions, with \\\\(d \\\\ll n\\\\), and allows the complexity of the algorithm’s major cost to be \\\\(O(n \\\\log n)\\\\). The run-time cost of the method depends on two parameters, m and d, where m represents the diagonal block size and controls the sizes of the blocks of the covariance matrix that are replaced by low-rank approximations, and d allows a trade-off of accuracy for expensive computations in the evaluation of the probabilities of m-dimensional blocks. We also introduce an inexpensive block reordering strategy to provide improved accuracy in the overall probability computation. The downside of this method, as with other such conditioning approximations, is the absence of an internal estimate of its error to use in tuning the approximation. Numerical simulations on problems from 2D spatial statistics with dimensions up to 16,384 indicate that the algorithm achieves a \\\\(1\\\\%\\\\) error level and improves the run time over a one-level hierarchical Quasi-Monte Carlo method by a factor between 10 and 15.\",\n \"Title: Outsmarting the Atmospheric Turbulence for Ground-Based Telescopes Using the Stochastic Levenberg-Marquardt Method\\nYear: 2021\\n\\nAbstract: One of the main challenges for ground-based optical astronomy is to compensate for atmospheric turbulence in near real-time. The goal is to obtain images as close as possible to the diffraction limit of the telescope. This challenge is addressed on the latest generation of giant optical telescopes by deploying multi-conjugate adaptive optics (MCAO) systems performing predictive tomography of the turbulence and multi-layer compensation. Such complex systems require a high fidelity estimate of the turbulence profile above the telescope, to be updated regularly during operations as turbulence conditions evolve. In this paper, we modify the traditional Levenberg-Marquardt (LM) algorithm by considering stochastically chosen subsystems of the full problem to identify the required parameters efficiently, while coping with the real-time challenge. While LM operates on the full set data samples, the resulting Stochastic LM (SLM) method randomly selects subsamples to compute corresponding approximate gradients and Hessians. Hence, SLM reduces the algorithmic complexity per iteration and shortens the overall time to solution, while maintaining LM's numerical robustness. We present a new convergence analysis for SLM, implement the algorithm with optimized GPU kernels, and deploy it on shared-memory systems with multiple GPU accelerators. We assess SLM in the adaptive optics system configurations in the context of the MCAO-Assisted Visible Imager & Spectrograph (MAVIS) instrument for the Very Large Telescope (VLT). We demonstrate performance superiority of SLM over the traditional LM algorithm and the classical stochastic first-order methods. At the scale of VLT AO, SLM finishes the optimization process and accurately retrieves the parameters (e.g., turbulence strength and wind speed profiles) in less than a second using up to eight NVIDIA A100 GPUs, which permits high acuity real-time throughput over a night of observations.\",\n \"Title: H2Opus: a distributed-memory multi-GPU software package for non-local operators\\nYear: 2022\\n\\nAbstract: Hierarchical \\n \\n \\n \\n \\n ${\\\\mathscr{H}}^{2}$\\n -matrices are asymptotically optimal representations for the discretizations of non-local operators such as those arising in integral equations or from kernel functions. Their O(N) complexity in both memory and operator application makes them particularly suited for large-scale problems. As a result, there is a need for software that provides support for distributed operations on these matrices to allow large-scale problems to be represented. In this paper, we present high-performance, distributed-memory GPU-accelerated algorithms and implementations for matrix-vector multiplication and matrix recompression of hierarchical matrices in the \\n \\n \\n \\n \\n ${\\\\mathscr{H}}^{2}$\\n format. The algorithms are a new module of H2Opus, a performance-oriented package that supports a broad variety of \\n \\n \\n \\n \\n ${\\\\mathscr{H}}^{2}$\\n matrix operations on CPUs and GPUs. Performance in the distributed GPU setting is achieved by marshaling the tree data of the hierarchical matrix representation to allow batched kernels to be executed on the individual GPUs. MPI is used for inter-process communication. We optimize the communication data volume and hide much of the communication cost with local compute phases of the algorithms. Results show near-ideal scalability up to 1024 NVIDIA V100 GPUs on Summit, with performance exceeding 2.3 Tflop/s/GPU for the matrix-vector multiplication, and 670 Gflop/s/GPU for matrix compression, which involves batched QR and SVD operations. We illustrate the flexibility and efficiency of the library by solving a 2D variable diffusivity integral fractional diffusion problem with an algebraic multigrid-preconditioned Krylov solver and demonstrate scalability up to 16M degrees of freedom problems on 64 GPUs.\",\n \"Title: Approximate Error Bounds On Solutions Of Nonlinearly Preconditioned Pdes\\nYear: 2021\\n\\nAbstract: In many multiphysics applications, an ultimate quantity of interest can be written as a linear functional of the solution to the discretized governing nonlinear partial differential equations and finding a sufficiently accurate pointwise solution may be regarded as a step toward that end. In this paper, we derive a posteriori approximate error bounds for linear functionals corresponding to quantities of interest using two kinds of nonlinear preconditioning techniques. Nonlinear preconditioning, such as the inexact Newton with backtracking and nonlinear elimination algorithm and the multiplicative Schwarz preconditioned inexact Newton algorithm, may be effective in improving global convergence for Newton's method. It may prevent stagnation of the nonlinear residual norm and reduce the number of solutions of large ill-conditioned linear systems involving a global Jacobian required at each nonlinear iteration. We illustrate the effectiveness of the new bounds using canonical nonlinear PDE models: a flame sheet model and a nonlinear coupled lid-driven cavity problem.\",\n \"Title: An O(N) Algorithm For Computing Expectation Of N-Dimensional Truncated Multi-Variate Normal Distribution I: Fundamentals\\nYear: 2021\\n\\nAbstract: In this paper, we present the fundamentals of a hierarchical algorithm for computing the N-dimensional integral phi(a, b; A) = integral(b)(a) H(x) f (x vertical bar A)dx representing the expectation of a function H(X) where f (x vertical bar A) is the truncated multi-variate normal (TMVN) distribution with zero mean, x is the vector of integration variables for the N-dimensional random vector X, A is the inverse of the covariance matrix Sigma, and a and b are constant vectors. The algorithm assumes that H(x) is \\\"low-rank\\\" and is designed for properly clustered X so that the matrix A has \\\"low-rank\\\" blocks and \\\"low-dimensional\\\" features. We demonstrate the divide-and-conquer idea when A is a symmetric positive definite tridiagonal matrix and present the necessary building blocks and rigorous potential theory-based algorithm analysis when A is given by the exponential covariance model. The algorithm overall complexity is O(N) for N-dimensional problems, with a prefactor determined by the rank of the off-diagonal matrix blocks and number of effective variables. Very high accuracy results for N as large as 2048 are obtained on a desktop computer with 16G memory using the fast Fourier transform (FFT) and non-uniform FFT to validate the analysis. The current paper focuses on the ideas using the simple yet representative examples where the off-diagonal matrix blocks are rank 1 and the number of effective variables is bounded by 2, to allow concise notations and easier explanation. In a subsequent paper, we discuss the generalization of current scheme using the sparse grid technique for higher rank problems and demonstrate how all the moments of k(th) order or less (a total of O(N-k) integrals) can be computed using O(N-k) operations for k >= 2 and O(N log N) operations for k = 1.\",\n \"Title: Accelerating Geostatistical Modeling and Prediction With Mixed-Precision Computations: A High-Productivity Approach With PaRSEC\\nYear: 2022\\n\\nAbstract: Geostatistical modeling, one of the prime motivating applications for exascale computing, is a technique for predicting desired quantities from geographically distributed data, based on statistical models and optimization of parameters. Spatial data are assumed to possess properties of stationarity or non-stationarity via a kernel fitted to a covariance matrix. A primary workhorse of stationary sp...\",\n \"Title: Meeting the real-time challenges of ground-based telescopes using low-rank matrix computations\\nYear: 2021\\n\\nAbstract: ABSTRACTAdaptive Optics (AO) is a technology that permits to measure and mitigate the distortion effects of atmospheric turbulence on optical beams. AO must operate in real-time by controlling thousands of actuators to shape the surface of deformable mirrors deployed on ground-based telescopes to compensate for these distortions. The command vectors that trigger how each individual actuator should act to bend a portion of the mirror are obtained from Matrix-Vector Multiplications (MVM). We identify and leverage the data sparsity structure of these control matrices coming from the MAVIS instruments for the European Southern Observatory's Very Large Telescope. We provide performance evaluation on x86 and accelerator-based systems. We present the impact of tile low-rank (TLR) matrix approximations on time-to-solution for the MVM and assess the produced image quality. We achieve performance improvement up to two orders of magnitude for TLR-MVM compared to regular dense MVM, while maintaining the image quality.\",\n \"Title: Implications of Reduced Communication Precision in a Collocated Discontinuous Galerkin Finite Element Framework\\nYear: 2021\\n\\nAbstract: Compute capability of high-performance hardware has been growing at immense rates, increasing over 130x in the last decade. Communication bandwidth, however, only grew by a factor of 6x in the same time, leading to a significant decrease in the byte-to-flop metric. This trend leads us to the situation where, in many cases, computation is virtually free, and the dominant cost of a parallel applicat...\",\n \"Title: NONLINEAR PRECONDITIONING STRATEGIES FOR TWO-PHASE FLOWS IN POROUS MEDIA DISCRETIZED BY A FULLY IMPLICIT DISCONTINUOUS GALERKIN METHOD\\\\ast\\nYear: 2021\\n\\nAbstract: We consider numerical simulation of two-phase flows in porous media using implicit methods. Because of the complex features involving heterogeneous permeability and nonlinear capillary effects, the nonlinear algebraic systems arising from the discretization are very difficult to solve. The traditional Newton method suffers from slow convergence in the form of a long stagnation or sometimes does not converge at all. In this paper, we develop nonlinear preconditioning strategies for the system of two-phase flows discretized by a fully implicit discontinuous Galerkin method. The preconditioners identify and approximately eliminate the local high nonlinearities that cause the Newton method to take small updates. Specifically, we propose two elimination strategies: one is based on exploring the unbalanced nonlinearities of the pressure and the saturation fields, and the other is based on identifying certain elements of the finite element space that have much higher nonlinearities than the rest of the elements. We compare the performance and robustness of the proposed algorithms with an existing single-field elimination approach and the classical inexact Newton method with respect to some physical and numerical parameters. Experiments on three-dimensional porous media applications show that the proposed algorithms are superior to other methods in terms of robustness and parallel efficiency.\",\n \"Title: tlrmvnmvt: Computing High-Dimensional Multivariate Normal and Student-t Probabilities with Low-Rank Methods in R\\nYear: 2022\\n\\nAbstract: This paper introduces the usage and performance of the R package tlrmvnmvt, aimed at computing high-dimensional multivariate normal and Student-t probabilities. The package implements the tile-low-rank methods with block reordering and the separation of-variable methods with univariate reordering. The performance is compared with two other state-of-the-art R packages, namely the mvtnorm and the TruncatedNormal packages. Our package has the best scalability and is likely to be the only option in thousands of dimensions. However, for applications with high accuracy requirements, the Truncated Normal package is more suitable. As an application example, we show that the excursion sets of a latent Gaussian random field can be computed with the tlrmvnmvt package without any model approximation and hence, the accuracy of the produced excursion sets is improved.\",\n \"Title: Performance analysis of relaxation Runge-Kutta methods\\nYear: 2022\\n\\nAbstract: Recently, global and local relaxation Runge-Kutta methods have been developed for guaranteeing the conservation, dissipation, or other solution properties for general convex functionals whose dynamics are crucial for an ordinary differential equation solution. These novel time integration procedures have an application in a wide range of problems that require dynamics-consistent and stable numerical methods. The application of a relaxation scheme involves solving scalar nonlinear algebraic equations to find the relaxation parameter. Even though root-finding may seem to be a problem technically straightforward and computationally insignificant, we address the problem at scale as we solve full-scale industrial problems on a CPU-powered supercomputer and show its cost to be considerable. In particular, we apply the relaxation schemes in the context of the compressible Navier-Stokes equations and use them to enforce the correct entropy evolution. We use seven different algorithms to solve for the global and local relaxation parameters and analyze their strong scalability. As a result of this analysis, within the global relaxation scheme, we recommend using Brent's method for problems with a low polynomial degree and of small sizes for the global relaxation scheme, while secant proves to be the best choice for higher polynomial degree solutions and large problem sizes. For the local relaxation scheme, we recommend secant. Further, we compare the schemes' performance using their most efficient implementations, where we look at their effect on the timestep size, overhead, and weak scalability. We show the global relaxation scheme to be always more expensive than the local approach-typically 1.1-1.5 times the cost. At the same time, we highlight scenarios where the global relaxation scheme might underperform due to its increased communication requirements. Finally, we present an analysis that sets expectations on the computational overhead anticipated based on the system properties.\",\n \"Title: Parallel Approximations of the Tukey g-and-h Likelihoods and Predictions for Non-Gaussian Geostatistics\\nYear: 2022\\n\\nAbstract: Maximum likelihood estimation is an essential tool in the procedure to impute missing data in climate/weather applications. By defining a particular statistical model, the maximum likelihood estimation can be used to understand the underlying structure of given geospatial data. The Gaussian random field has been widely used to describe geospatial data, as one of the most popular models under the hood of maximum likelihood estimation. Computation of Gaussian log-likelihood demands operations on a dense symmetric positive definite matrix, often parameterized by the Matérn correlation function. This computation of the log-likelihood requires \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">$\\\\mathcal{O}(n^{2})$</tex>\\n storage and \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">$\\\\mathcal{O}(n^{3})$</tex>\\n operations, which can be a huge task considering that the number of geographical locations, \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">$n$</tex>\\n, now commonly reaches into the millions. However, despite its appealing theoretical properties, the assumptions of Gaussianity may be unrealistic since real data often show signs of skewness or have some extreme values. Herein, we consider the Tukey \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">${g-}$</tex>\\n and \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">$-h$</tex>\\n (TGH) random field as an example of a non-Gaussian random field that shows more robustness in modeling geospatial data by including two more parameters to incorporate skewness and heavy tail features in the model. This work provides the first HPC implementation of the TGH random field's inference on parallel hardware architectures. Using task-based programming models associated with dynamic runtime systems, our implementation leverages the high concurrency of current parallel systems. This permits to run the exact log-likelihood evaluation of the Tukey g-and-h (TGH) random fields for a decent number of geospatial locations. To tackle large-scale problems, we provide additionally an implementation of the given model using two different low-rank approximations. We compress the aforementioned positive-definite symmetric matrix for computing the log-likelihood and rely on the Tile Low-Rank (TLR) and the Hierarchical Off-Diagonal Low-Rank (HODLR) matrix approximations. We assess the performance and accuracy of the proposed implementations using synthetic datasets up to \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">$800K$</tex>\\n and a \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">$300K$</tex>\\n precipitation data of Germany to demonstrate the advantage of using non-Gaussian over Gaussian random fields. Moreover, by relying on TLR/HODLR matrix computations, we can now solve for larger matrix sizes while preserving the required accuracy for prediction. We show the performance superiority of TLR over HODLR matrix computations when calculating the TGH likelihoods and predictions. Our TLR-based approximation shows a speedup up to \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">$7.29X$</tex>\\n and \\n<tex xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">$2.96X$</tex>\\n on shared-memory and distributed-memory systems, respectively, compared to the exact implementation.\",\n \"Title: A NONLINEAR ELIMINATION PRECONDITIONED INEXACT NEWTON ALGORITHM\\nYear: 2022\\n\\nAbstract: A nonlinear elimination preconditioned inexact Newton (NEPIN) algorithm is proposed for problems with localized strong nonlinearities. Due to unbalanced nonlinearities (\\\"nonlinear stiffness\\\"\\\"), the traditional inexact Newton method often exhibits a long plateau in the norm of the nonlinear residual or even fails to converge. NEPIN implicitly removes the components causing trouble for the global convergence through a correction based on nonlinear elimination within a subspace that provides a modified direction for the global Newton iteration. Numerical experiments show that NEPIN can be more robust than global inexact Newton algorithms and maintain fast convergence even for challenging problems, such as full potential transonic flows. NEPIN complements several previously studied nonlinear preconditioners with which it compares favorably experimentally on a classic shocked duct flow problem considered herein. NEPIN is shown to be fairly insensitive to mesh resolution and \\\"bad\\\"\\\" subproblem identification based on the local Mach number or the local nonlinear residual for transonic flow over a wing.\",\n \"Title: High-Performance Spatial Data Compression for Scientific Applications\\nYear: 2022\\n\\nAbstract: We implement an efficient data compression algorithm that reduces the memory footprint of spatial datasets generated during scientific simulations. Storing regularly these datasets is typically needed for checkpoint/restart or for post-processing purposes. Our lossy compression approach, codenamed HLRcompress (https://gitlab.mis.mpg. de/rok/HLRcompress), combines a hierarchical low-rank approximation technique with binary compression. This novel hybrid method is agnostic to the particular domain of application. We study the impact of HLRcompress on accuracy using synthetic datasets to demonstrate the software capabilities, including robustness and versatility. We assess different algebraic compression methods and report performance results on various parallel architectures. We then integrate it into a workflow of a direct numerical simulation solver for turbulent combustion on distributed-memory systems. We compress the generated snapshots during time integration using accuracy thresholds for each individual chemical species, without degrading the practical accuracy of the overall pressure and temperature. We eventually compare against state-of-the-art compression software. Our implementation achieves on average greater than 100-fold compression of the original size of the datasets.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":75,"string":"75"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13566,"string":"13,566"}}},{"rowIdx":13567,"cells":{"question_id":{"kind":"string","value":"Hongen Liao"},"question":{"kind":"string","value":"Generate an abstract for the title \"Real-time epicardial excitation time map overlay\" using the following items: 1. Epicardial Excitation Time (ECET)\n2. Conditional Density Propagation\n3. Electrophysiological mapping\n4. Surgeon's endoscope stream\n5. Augument"},"answer":{"kind":"string","value":"This paper presents a method to display Epicardial Excitation Time (ECET) map information intuitively to a surgeon. The method uses Conditional Density Propagation to track the epicardium during the complete cardiac cycle. Then, combining this algorithm with electrophysiological mapping techniques, pre-recorded ECET maps are overlay onto the surgeons endoscope stream to give direct feedback on cardiac excitation propagation. To the best of our knowledge, this is the first report of electrophysiologic data being augumented with a surgeons endoscope stream."},"correct_answer":{"kind":"string","value":"This paper presents a method to display Epicardial Excitation Time (ECET) map information intuitively to a surgeon. The method uses Conditional Density Propagation to track the epicardium during the complete cardiac cycle. Then, combining this algorithm with electrophysiological mapping techniques, pre-recorded ECET maps are overlay onto the surgeons endoscope stream to give direct feedback on cardiac excitation propagation. To the best of our knowledge, this is the first report of electrophysiologic data being augumented with a surgeons endoscope stream."},"documents":{"kind":"list like","value":["Title: Long Visualization Depth Autostereoscopic Display using Light Field Rendering based Integral Videography\nYear: 2006\n\nAbstract: three-dimensional (3D) computer graphics (CG) image without using special viewing glasses or tracking devices. The images are created by employing a light field rendering and pixel distribution algorithm for integral photography (IP) / integral videography (IV) imaging. How to enhance the image depth, especial the long visualization depth, is a challenge work. In this study, the images are rendered from a referential viewing area for each viewpoint and the elemental images are reconstructed by pixel redistribution method. The corresponding result images are projected to a screen that is separated from the lens array by a referential viewing distance as the setting of image rendering. A photographic film is used to record the elemental image through each lens with a photograph-taken method. Our photograph-taken based IV display enables precise 3D images with long visualization depth to be displayed at long viewing distances without any influence from deviated or distorted lenses in a lens array. We created an autostereoscopic display that appears to have three-dimensionality even when viewed from a distance, with an image depth of 5.7 m or more in front of the display. To the best of our knowledge, the presented long-distance IV display is technically unique as it is the first report of generating an autostereoscopic image with such a long viewing distance in the field of computer graphics. CR Categories and Subject Descriptors: B.4.2 Input/Output Devices - Image Display","Title: A novel high intensity focused ultrasound robotic system for breast cancer treatment.\nYear: 2013\n\nAbstract: High intensity focused ultrasound (HIFU) is a promising technique for cancer treatment owing to its minimal invasiveness and safety. However, skin burn, long treatment time and incomplete ablation are main shortcomings of this method. This paper presents a novel HIFU robotic system for breast cancer treatment. The robot has 4 rotational degrees of freedom with the workspace located in a water tank for HIFU beam imaging and ablation treatment. The HIFU transducer combined with a diagnostic 2D linear ultrasound probe is mounted on the robot end- effector, which is rotated around the HIFU focus when ablating the tumor. HIFU beams are visualized by the 2D probe using beam imaging. Skin burn can be prevented or alleviated by avoiding long time insonification towards the same skin area. The time cost could be significantly reduced, as there is no need to interrupt the ablation procedure for cooling the skin. In addition, our proposed robot control strategies can avoid incomplete ablation. Experiments were carried out and the results showed the effectiveness of our proposed system.","Title: Pulsative flow segmentation in MRA image series by AR modeling and EM algorithm\nYear: 2006\n\nAbstract: Segmentation of CSF and pulsative blood flow, based on a single phase contrast MRA (PC-MRA) image can lead to imperfect classifications. In this paper, we present a novel automated flow segmentation method by using PC-MRA image series. The intensity time series of each pixel is modeled as an autoregressive (AR) process and features including the Linear Prediction Coefficients (LPC), covariance matrix of LPC and variance of prediction error are extracted from each profile. Bayesian classification of the feature space is then achieved using a non-Gaussian likelihood probability function and unknown parameters of the likelihood function are estimated by a generalized Expectation-Maximization (EM) algorithm. The efficiency of the method evaluated on both synthetic and real retrospective gated PC-MRA images indicate that robust segmentation of CSF and vessels can be achieved by using this method.","Title: High-Resolution Stereoscopic Surgical Display Using Parallel Integral Videography and Multi-projector\nYear: 2002\n\nAbstract: This paper proposes a high-resolution stereoscopic surgical display using the integral videography (IV) and multiple projectors. IV projects a computer generated graphical object by multiple rays through \"fly's eye lens\", which can display geometrically accurate autostereoscopic images and reproduce motion parallax in three-dimensional (3-D) space. This technique requires neither special glasses nor sensing device to track viewer's eyes, thus being suitable for pre-operative diagnosis and intra-operative use. This paper reports the use of multiple reduction projection display system and a parallel-calculation to achieve high-resolution IV image. We evaluate the feasibility of this display by developing a 3-D CT stereoscopic image and applying it to surgical planning and intra-operative guidance. The main contribution of this paper is application and modification of medical stereoscopic technique originally developed in high-resolution multi-projector stereoscopic display system.","Title: Nonmagnetic rigid and flexible outer sheath with pneumatic interlocking mechanism for minimally invasive surgical approach.\nYear: 2009\n\nAbstract: We developed a nonmagnetic rigid and flexible outer sheath with pneumatic interlocking mechanism using flexible toothed links and a wire-driven bending distal end. The outer sheath can be switched between rigid and flexible modes easily depending on surgical scenes, and the angle of its distal end can be controlled by three nylon wires. All components of flexible parts are made of MRI-compatible nonmagnetic plastics. We manufactured the device with 300-mm long, 16-mm outer diameter, 7-mm inner diameter and 90-mm bending distal end. Holding power of the device in rigid mode was maximum 3.6 N, which was sufficient for surgical tasks in body cavity. In vivo experiment using a swine, our device performed smooth insertion of a flexible endoscope and a biopsy forceps into reverse side of the liver, intestines and spleen with a curved path. In conclusion, our device shows availability of secure approach of surgical instruments into deep cavity.","Title: MRI-Compatible Rigid and Flexible Outer Sheath Device with Pneumatic Locking Mechanism for Minimally Invasive Surgery\nYear: 2008\n\nAbstract: To reduce the invasiveness of surgery, we developed an outer sheath device using a flexible toothed link and pneumatic locking mechanism that works with flexible devices used in minimally invasive surgery. The outer sheath can be switched between flexible and rigid modes, and the angle of its tip can be controlled by a nylon wire. All parts of this device are made of plastic and are MRI-compatible. We manufactured a sheath prototype, 300 mm long, with a 20-mm outer diameter, and an 8-mm inner diameter. Experiment results showed that the outer sheath can protect tissues from high insertion force and secure the path for flexible devices. It can follow a curved path with a reasonable radius.","Title: An image restoration method for extracting features from three-dimensional auto-stereoscopic integral photography images\nYear: 2011\n\nAbstract: Integral Photography (IP) [Hongen Liao 2011] is a visualization technique that produce true three-dimensional images in the real space that can be viewed without supplementary glasses. Using a micro lens array, different aspects of the object can be viewed from different directions, giving observers a sense of depth. Utilizing IP images into augmented reality systems (where images are merged into a real scene) requires an automated alignment between virtual IP images and real objects, leading to the need of understanding the displayed IP images. A natural approach would be capturing multiple views of the IP images and then extracting necessary information from those views. However, since IP images are viewed through a lens array, a lens-like pattern is produced in the observed image(Fig.1(a)). When captured by a camera, simply applying existing image processing methods such as feature extraction would result in unexpected behaviors (e.g. detection of fault features). The purpose of this study is to develop an image restoration method by removing those lens-like pattern from the IP image captured by a camera.","Title: Interactive 3D Navigation System for Image-guided Surgery\nYear: 2009\n\nAbstract: This paper presents a novel surgery navigation system based on a three-dimensional (3D) imaging technique, integral videography (IV). In our system, the 3D structure of the object of interest is reconstructed using surface rendering and corresponding pixel distribution methods. We developed a high-speed algorithm that renders high-quality IV images from the surface model in real time and allows interactions like rotating and scaling to be done smoothly. Using the patient-image registration method, IV images can be displayed with the correct size and relative position with respect to the surgical instruments. Experiments were carried out with various anatomical models, and the results show that our system could be useful in many clinical situations such as orthopedic surgery and neurosurgery.","Title: Image-based electrode array tracking for epicardial electrophysiological mapping in minimally invasive arrhythmia surgery.\nYear: 2011\n\nAbstract: Electrophysiological mapping is effective in realizing a precise minimally invasive arrhythmia surgery. Recently, an epicardial electrophysiological mapping system for minimally invasive arrhythmia surgery was reported. The system requires a small electrode array, a tracking system and a global mapping algorithm. The optical tracking system employed in the research requires line of sight and complicated configuration. This paper proposes a new tracking method for locating an electrode array.We developed a small electrode array and optical markers. Center points of respective optical markers and the electrode array are tracked via an endoscopic stream and calculated in image space. The orientation of the electrode array is calculated using the dot product between the vector joining two center points of two upper optical markers and the vector joining two end points of the longest edge of the electrode array.Mean tracking errors of position and orientation of the electrode array were 0.51 mm and 0.64°, respectively. And the processing time was constant at 46 ms per frame. Our method could successfully track the electrode array on the epicardium during in vivo experiment and a global epicardial electrophysiological map was reconstructed from separately measured epicardial electrograms by the small electrode array.An image-based tracking method for locating an electrode array was proposed. Tracking accuracy, processing time and applicability to surgical environment of our method proved to be acceptable. Consequently, our method enables the electrode array tracking system to be simplified with no separate tracking system.","Title: A Bayesian Approach for Construction of Sparse Statistical Shape Models Using Dirichlet Distribution.\nYear: 2013\n\nAbstract: Statistical shape models (SSMs) made using point sets are important tools to capture the variations within shape populations. One popular method for construction of SSMs is based on the Expectation-Maximization (EM) algorithm which establishes probabilistic matches between the model and training points. In this paper, we propose a novel Bayesian framework to automatically determine the optimal number of the model points. We use a Dirichlet distribution as a prior to enforce sparsity on the mixture weights of Gaussians. Insignificant model points are determined and pruned out using a quadratic programming technique. We apply our method to learn a sparse SSM from 15 manually segmented caudate nuclei data sets. The generalization ability of the proposed model compares favorably to a traditional EM based model.","Title: Ultrasound imaging method for internal jugular vein measurement and estimation of circulating blood volume.\nYear: 2014\n\nAbstract: Evaluation of circulating blood volume is important in assessing the status of patients. Although some studies have suggested that ultrasound images of the internal jugular vein (IJV) can be used for the analysis of circulating blood volume, accurate extraction of IJV is necessary to reduce errors. Therefore, this study was designed to develop a new algorithm for dynamic segmentation of IJV and determine appropriate indicators to evaluate the circulating blood volume.Our algorithm is based on snake and speckle tracking models. As the region of interest (ROI) of the control points of the snake tracking algorithm was dynamically moved using speckle tracking, ROI size can be decreased leading to a reduction in the tracking error. Some experiments were performed to validate our algorithm. Subsequently, the algorithm was used for the experiment simulating dehydration state among 11 subjects.Results of the validation experiment suggest that our algorithm showed higher performance for IJV extraction compared with standard methods. Furthermore, it was revealed that some indices such as the average area of IJV were related to the dehydration state of subjects.This study proposed a new algorithm, which was based on snake and speckle tracking models, for dynamic extraction of IJV in ultrasound images. In addition to algorithm validation, it was suggested that some indices the ultrasound image of IJV could be used for the evaluation of the circulating blood volume.","Title: A variational method for geometric regularization of vascular segmentation in medical images.\nYear: 2008\n\nAbstract: In this paper, a level-set-based geometric regularization method is proposed which has the ability to estimate the local orientation of the evolving front and utilize it as shape induced information for anisotropic propagation. We show that preserving anisotropic fronts can improve elongations of the extracted structures, while minimizing the risk of leakage. To that end, for an evolving front using its shape-offset level-set representation, a novel energy functional is defined. It is shown that constrained optimization of this functional results in an anisotropic expansion flow which is usefull for vessel segmentation. We have validated our method using synthetic data sets, 2-D retinal angiogram images and magnetic resonance angiography volumetric data sets. A comparison has been made with two state-of-the-art vessel segmentation methods. Quantitative results, as well as qualitative comparisons of segmentations, indicate that our regularization method is a promising tool to improve the efficiency of both techniques.","Title: Real-Time Marker-Free Patient Registration and Image-Based Navigation Using Stereovision for Dental Surgery.\nYear: 2013\n\nAbstract: Surgical navigation techniques have been evolving rapidly in the field of oral and maxillofacial surgery (OMS). However, challenges still exist in the current state of the art of computer-assisted OMS especially from the viewpoint of dental surgery. The challenges include the invasive patient registration procedure, the difficulty of reference marker attachment, navigation error caused by patient movement, bulky optical markers and maintenance of line of sight for commercial optical tracking devices, inaccuracy and susceptibility of electromagnetic (EM) sensors to magnetic interference for EM tracking devices. In this paper, a new solution is proposed to overcome the mentioned challenges. A stereo camera is designed as a tracking device for both instrument tracking and patient tracking, which is customized optimally for the limited surgical space of dental surgery. A small dot pattern is mounted to the surgical tool for instrument tracking, which can be seen by the camera at all times during the operation. The patient registration is achieved by patient tracking and 3D contour matching with the preoperative patient model, requiring no fiducial marker and reference marker. In addition, the registration is updated in real-time. Experiments were performed to evaluate our method and an average overall error of 0.71 mm was achieved.","Title: New 4-D imaging for real-time intraoperative MRI: adaptive 4-D scan.\nYear: 2006\n\nAbstract: Aiming at real-time 3-D visualization of organ motion to navigate surgical procedures in MRI-guided surgery, a new 4-D MR imaging technique called \"Adaptive 4-D Scan\" has been proposed. The technique is designed to acquire a time series of volumetric 3-D images (4-D image) of cyclically moving organ, even in a low-field open-configuration MR scanner. A pre-operative 4-D image is acquired with respiratory phase parameter, which is monitored by using navigator-echo-based real-time tracking of the liver and diaphragm. During operation, the respiratory phase is again monitored in real-time, and a 3-D image, reflecting the current state of the target organ, is extracted from the pre-operative 4-D image and provided to physicians as a pseudo real-time 3-D image. We implemented Adaptive 4-D Scan into a 0.5 Tesla open-configuration clinical MRI system for intervention. Phantom and volunteer studies were performed to assess feasibility of this technique, in terms of image quality, imaging time and position accuracy of the imaged subject. A 4-D image (matrix: 256 x 128 x 10 x 8) of cyclically moving phantom was acquired in 719 s, and RMS position error between the imaged subject and the real subject was 2.3 mm, where the range of motion was 50 mm. 4-D image of the moving liver was also successfully acquired under near clinical condition. In conclusion, the study shows that the proposed method is feasible and has capability to provide real-time dynamic 3-D atlas for surgical navigation.","Title: Robotic hair harvesting system: a new proposal.\nYear: 2011\n\nAbstract: Follicular Unit Extraction (FUE) has become a popular hair transplanting method for solving male-pattern baldness problem. Manually harvesting hairs one by one, however, is a tedious and time-consuming job to doctors. We design an accurate hair harvesting robot with a novel and efficient end-effector which consists of one digital microscope and a punch device. The microscope is first employed to automatically localize target hairs and then guides the punch device for harvesting after shifting. The end-effector shows average bias and precision of 0.014 mm by virtue of a rotary guidance design for the motorized shifting mechanism.","Title: A fully automatic 2D ultrasound probe calibration method using a simple phantom.\nYear: 2011\n\nAbstract: Probe calibration should be performed for free-hand tracking and ultrasound (US) fusion with a position sensor-attached US probe in US guided interventions. Difficulties in locating the probe imaging plane and adjusting the alignment of the probe with the calibration phantom call for more complicated and precise phantoms. However, the manufacture of such phantoms is time-consuming and expensive. Furthermore, the segmentation of the phantom features in US images cannot be reliably automated, which requires manual intervention. This paper presents a fully automatic 2D probe calibration method using a rather simple phantom. By detecting and analyzing the temporal signal arising from the ultrasound artifacts when sweeping the 2D probe over the phantom, the imaging plane of the probe and the image features can be reliably identified. Then the features are fed to a nonlinear optimizer to estimate the calibration parameters. The whole procedure including the setup can be done within 2 minutes with the fiducial registration error (FRE) of 0.16 mm. A US-guided biopsy experiment has also been conducted which shows the target registration error (TRE) of less than 1 mm. © 2011 IEEE.","Title: Autostereoscopic 3D Display with Long Visualization Depth using Referential Viewing Area based Integral Photography.\nYear: 2011\n\nAbstract: We developed an autostereoscopic display for distant viewing of three-dimensional (3D) computer graphics (CG) images without using special viewing glasses or tracking devices. The images are created by employing referential viewing area based CG image generation and pixel distribution algorithm for integral photography (IP) and integral videography (IV) imaging. CG image rendering is used to generate IP/IV elemental images. The images can be viewed from each viewpoint within a referential viewing area and the elemental images are reconstructed from rendered CG images by pixel redistribution and compensation method. The elemental images are projected onto a screen that is placed at the same referential viewing distance from the lens array as in the image rendering. Photographic film is used to record the elemental images through each lens. The method enables 3D images with a long visualization depth to be viewed from relatively long distances without any apparent influence from deviated or distorted lenses in the array. We succeeded in creating an actual autostereoscopic images with an image depth of several meters in front of and behind the display that appear to have three-dimensionality even when viewed from a distance.","Title: Augmented reality navigation with automatic marker-free image registration using 3-D image overlay for dental surgery.\nYear: 2014\n\nAbstract: Computer-assisted oral and maxillofacial surgery (OMS) has been rapidly evolving since the last decade. State-of-the-art surgical navigation in OMS still suffers from bulky tracking sensors, troublesome image registration procedures, patient movement, loss of depth perception in visual guidance, and low navigation accuracy. We present an augmented reality navigation system with automatic marker-free image registration using 3-D image overlay and stereo tracking for dental surgery. A customized stereo camera is designed to track both the patient and instrument. Image registration is performed by patient tracking and real-time 3-D contour matching, without requiring any fiducial and reference markers. Real-time autostereoscopic 3-D imaging is implemented with the help of a consumer-level graphics processing unit. The resulting 3-D image of the patient's anatomy is overlaid on the surgical site by a half-silvered mirror using image registration and IP-camera registration to guide the surgeon by exposing hidden critical structures. The 3-D image of the surgical instrument is also overlaid over the real one for an augmented display. The 3-D images present both stereo and motion parallax from which depth perception can be obtained. Experiments were performed to evaluate various aspects of the system; the overall image overlay error of the proposed system was 0.71 mm.","Title: Precision-guided surgical navigation system using laser guidance and 3D autostereoscopic image overlay.\nYear: 2010\n\nAbstract: This paper describes a precision-guided surgical navigation system for minimally invasive surgery. The system combines a laser guidance technique with a three-dimensional (3D) autostereoscopic image overlay technique. Images of surgical anatomic structures superimposed onto the patient are created by employing an animated imaging method called integral videography (IV), which can display geometrically accurate 3D autostereoscopic images and reproduce motion parallax without the need for special viewing or tracking devices. To improve the placement accuracy of surgical instruments, we integrated an image overlay system with a laser guidance system for alignment of the surgical instrument and better visualization of patient's internal structure. We fabricated a laser guidance device and mounted it on an IV image overlay device. Experimental evaluations showed that the system could guide a linear surgical instrument toward a target with an average error of 2.48mm and standard deviation of 1.76mm. Further improvement to the design of the laser guidance device and the patient-image registration procedure of the IV image overlay will make this system practical; its use would increase surgical accuracy and reduce invasiveness.","Title: Real-Time Autostereoscopic Visualization of Registration-Generated 4D MR Image of Beating Heart\nYear: 2008\n\nAbstract: This paper presents a real-time autostereoscopic visualization system using the principle of Integral Videography(IV). We develop MIP and composite volume ray casting method for IV volume rendering, and implemented the algorithm on GPU to achieve real-time rendering. The system was used to visualize 4D MR image that was generated from registration of 3D MR image and 4D ultrasound image. The registration scheme consists of inter-modality rigid registration between 3D MR image and 3D ultrasound image and intra-modality non-rigid registration between 3D ultrasound images. Registration processes were also implemented on GPU. Evaluation of processing speed showed that GPU processing time was 48x, 13x, 21x faster than CPU processing time for IV volume rendering, rigid registration, and non-rigid registration respectively. We also enabled real-time user interactivity for IV visualization system. In the future, We plan to use this system to develop intra-operative surgery navigation system for intra-cardiac surgery on beating heart.","Title: PHYSICALLY ACCURATE B-SPLINE BASED NON-RIGID REGISTRATION USING VARIABLE SPRING MODEL\nYear: 2007\n\nAbstract: This paper presents a physically accurate fast parametric non-rigid registration method based on B-spline deformation model. Our method is based on using variable spring model to add a physical constraint to ensure that the deformation does not violate elasticity characteristic of the organs. Using the assumption of human organs as elastic bodies, we modeled the whole volume as 3-D springs connecting each neighboring voxels. Registration was performed on B-spline deformation model with the combination of SSD and total spring energy as optimization criterion. We also added the concept of elasticity map to made it possible to model a more specific elasticity characteristic such as hardness/softness of the organ. We showed that compared to the original algorithm, our algorithm was able to realize a more realistic deformation. We also applied our method to register 3D CT images of human heart from different heart beat phases.","Title: An integrated diagnosis and therapeutic system using intra-operative 5-aminolevulinic-acid-induced fluorescence guided robotic laser ablation for precision neurosurgery.\nYear: 2012\n\nAbstract: We have developed an integrated diagnosis and therapeutic system for precision malignant gliomas resection during neurosurgery. A combination of three-dimensional (3-D) magnetic resonance imaging (MRI) navigation and 5-aminolevulinic acid (5-ALA)-induced fluorescence based intra-operative tumor diagnosis technique has been incorporated into a robotic laser ablation neurosurgery system with an automatic focusing and robotic scanning mechanism. 5-ALA is a non-fluorescent prodrug that leads to intracellular accumulation of fluorescent protoporphyrins IX (PpIX) in malignant glioma. The PpIX tends to accumulate in pathological lesions, and emits red fluorescence when excited by blue light. This fluorescence is illuminated with laser excitation, enables intra-operative identification of the position of a tumor and provides guidance for resection with laser photocoagulation. The information provided by the MRI is enhanced by the intra-operative 5-ALA fluorescence data, and this enhanced information is integrated into a robotic laser ablation system. The accuracy of the fluorescent measurement of the tumor is improved using high-precision spectral analysis. The fluorescence assists in the detection of malignant brain tumors intra-operatively and improves their removal rate.","Title: Vessel bifurcation localization based on intraoperative three-dimensional ultrasound and catheter path for image-guided catheter intervention of oral cancers.\nYear: 2013\n\nAbstract: We present a method to localize intraoperative target vessel bifurcations under bones for ultrasound (US) image-guided catheter interventions. A catheter path is recorded to acquire skeletons for the target vessel bifurcations that cannot be imaged by intraoperative US. The catheter path is combined with the centerlines of the three-dimensional (3D) US image to construct a preliminary skeleton. Based on the preliminary skeleton, the orientations of target vessels are determined by registration with the preoperative image and the bifurcations were localized by computing the vessel length. An accurate intraoperative vessel skeleton is obtained for correcting the preoperative image to compensate for vessel deformation. A reality check of the proposed method was performed in a phantom experiment. Reasonable results were obtained. The in vivo experiment verified the clinical workflow of the proposed method in an in vivo environment. The accuracy of the centerline length of the vessel for localizing the target artery bifurcation was 2.4mm. These results suggest that the proposed method can allow the catheter tip to stop at the target artery bifurcations and enter into the target arteries. This method can be applied for virtual reality-enhanced image-guided catheter intervention of oral cancers.","Title: Surgical Manipulator with Balloon for Stabilizing Fetus in Utero under Ultrasound Guidance\nYear: 2008\n\nAbstract: This paper describes a surgical manipulator to stabilize intrauterine fetus with ultrasound image guidance. The manipulator includes an outline of 4 mm in diameter, a mechanism with 7 joints, and a set of balloons. The manipulator is arranged as straight form and the balloon is fold to be a minimum size before the insertion. Accuracy evaluation of bending performance showed that the standard deviations were ±3.6 degrees on wired-driven mechanism and ±1.6 degrees on linkage-driven mechanism. Experimental results also demonstrated high repeatability of the mechanisms. In feasibility experiments, ultrasound images in 2D and 3D modes were examined for guiding the manipulator. The 2D images provided wide view and easily viewable display of the balloon inflation. The 3D images provided easily viewable display of bending motion of the arm and the relative position of a phantom. It is possible to operate the manipulator in utero under the ultrasound guidance in 2D and 3D by switching them in each procedural stage.","Title: Intra-operative Real-Time 3-D Information Display System Based on Integral Videography\nYear: 2001\n\nAbstract: A real-time 3-D surgical navigation system that superimposes the real, intuitive 3-D image for medical diagnosis and operation was developed in this paper. This system creates 3-D image based on the principle of integral photography (IP), named \"Integral Videography (IV)\", which can display geometrically accurate 3-D image and reproduce motion parallax without any need of special devices. 3-D image was supenmposed on the surgical fields in the patient via a half-silvered mirror as if they could be seen through the body. In addition, a real-time IV algorithm for calculating the 3-D image of surgical instruments was used for registration between the location of surgical instruments and the organ during the operation. The experimental results of puncturing a point location and avoiding critical area showed the errors of this navigation system were in the range of 2-3mm. By introducing a display device with higher pixel density, accuracy of the system can be improved.","Title: Novel endoscope system with plasma flushing for off-pump cardiac surgery.\nYear: 2009\n\nAbstract: The purpose of this study is to develop a new endoscope for performing simple surgical tasks inside a cardiac atrium/chamber filled with blood, i.e., for performing \"off-pump\" cardiac surgeries. In general, it is very difficult to observe the inner wall of the vessels containing circulating blood because the light from the endoscope is scattered by the red blood cells. \"Plasma flushing\" performed using the separator system is developed to observe the inner side of the heart filled with blood and to remove blood cells from the front of the endoscope tip. The system was used in in vitro quantitative measurement of the device performance and in vivo experiments on a swine. In these experiments, we successfully obtained high-resolution images of the interior of the heart during off-pump surgery.","Title: Combination of intraoperative 5-aminolevulinic acid-induced fluorescence and 3-D MR imaging for guidance of robotic laser ablation for precision neurosurgery.\nYear: 2008\n\nAbstract: A combination of 5-aminolevulinic acid (5-ALA)-induced fluorescence and three-dimensional (3-D) magnetic resonance imaging (MRI) of a brain tumor has been incorporated into a robotic laser ablation neurosurgery system. 5-ALA is a non-fluorescent prodrug that leads to intracellular accumulation of fluorescent protoporphyrins IX (PpIX) in malignant glioma. The PpIX tends to accumulate in pathological lesions, and emits red fluorescence when excited by blue light. This fluorescence is illuminated with laser excitation, enables intra-operative identification of the position of a tumor and provides guidance for resection with laser photocoagulation. The information provided by the MRI is enhanced by the 5-ALA fluorescence data, and this enhanced information is integrated into a robotic laser ablation system. The accuracy of the fluorescent measurement of the tumor is improved using high-precision spectral analysis. The fluorescence assists in the detection of malignant brain tumors intraoperatively and improves their removal rate.","Title: Generalization of geometrical flux maximizing flow on Riemannian manifolds for improved volumetric blood vessel segmentation.\nYear: 2012\n\nAbstract: Geometric flux maximizing flow (FLUX) is an active contour based method which evolves an initial surface to maximize the flux of a vector field on the surface. For blood vessel segmentation, the vector field is defined as the vectors specified by vascular edge strengths and orientations. Hence, the segmentation performance depends on the quality of the detected edge vector field. In this paper, we propose a new method for level set based segmentation of blood vessels by generalizing the FLUX on a Riemannian manifold (R-FLUX). We consider a 3D scalar image I(x) as a manifold embedded in the 4D space (x, I(x)) and compute the image metric by pullback from the 4D space, whose metric tensor depends on the vessel enhancing diffusion (VED) tensor. This allows us to devise a non-linear filter which both projects and normalizes the original image gradient vectors under the inverse of local VED tensors. The filtered gradient vectors pertaining to the vessels are less sensitive to the local image contrast and more coherent with the local vessel orientation. The method has been applied to both synthetic and real TOF MRA data sets. Comparisons are made with the FLUX and vesselsness response based segmentations, indicating that the R-FLUX outperforms both methods in terms of leakage minimization and thiner vessel delineation.","Title: Fusion of Laser Guidance and 3-D Autostereoscopic Image Overlay for Precision-Guided Surgery\nYear: 2008\n\nAbstract: This paper describes a precision-guided surgical navigation system for minimally invasive surgery using fusion of laser guidance technique and three-dimensional (3-D) autostereoscopic image overlay technique. The images superimposed onto the patient are created by employing an animated autostereoscopic image called integral videography (IV), which display geometrically accurate 3-D autostereoscopic images and reproduce motion parallax without the need for special viewing or tracking devices. To improve the insertion accuracy of surgical instrument, we integrated the image overlay system with laser guidance for visualization of insertion point and orientation of the surgical instrument. We designed and manufactured a laser guidance device and mounted it to the IV image overlay device. Accuracy evaluations showed that the system could guide a linear surgical instrument toward a target with an average error of 2.48 mm and standard deviation of 1.76 mm. Improvement in the design of the laser guidance device and the patient-image registration of the IV image overlay will make this system practical and its use will increase surgical accuracy and reduce invasiveness.","Title: Image-guidance for cardiac surgery using dynamic autostereoscopic display system\nYear: 2004\n\nAbstract: An autostereoscopic display and analysis system has been developed to aid the development of three-dimensional (3-D) cardiac imaging for use during minimally invasive cardiac surgery. We have previously developed a dynamic autostereoscopic display device by integral videography autostereoscopic imaging method with high-resolution multi-projection technique. This paper integrated the high-quality autostereoscopic display system into image-guided cardiac surgery. The human heart was scanned in a sequence of volumetric images representing the motion of the human heart throughout the cardiac cycle, based on dynamic information of heart from CT scanning. Time-sequence 3-D image analysis can be particularly useful in cardiac applications. This approach will allow us to acquire the optimum method to produce 3-D cardiac image for planning and guidance of minimally invasive cardiac surgery.","Title: R-PLUS: a Riemannian anisotropic edge detection scheme for vascular segmentation.\nYear: 2008\n\nAbstract: In this paper, detection of edges in oriented fields is addressed. In some applications such as vessel segmentation because of the intrinsic orientation of the structures, edge detection is only demanded in a particular subspace. This is specially usefull when a curve evolution is chosen for segmentation since gradients in parallel to vessel orientation may stop the contour. An anisotropic edge detection scheme is generalized on a Riemannian manifold using the local structure tensor. The method is the generalization of the PLUS operator proposed for accurate curved edge detection. Examples are given and the comparison is made with the state-of-the-art flux maximizing flow which indicates that significant improvements in terms of leakage minimization and thiner vessel delineation is achievable using our methodology.","Title: Non-metal Slice Image Overlay Display System Used Inside the Open Type MRI\nYear: 2008\n\nAbstract: MRI is now utilized not only for diagnosis but also for intraoperative surgical treatment. In the MRI environment, the surgeon imagines the position of target region by observing pre-acquired images and patient body during the operation, therefore the spatial position of the diseased part depends on the surgeon's knowledge and experience. Thus information from the MRI image is not fully utilized. In order to solve this problem, we developed the prototypes of slice image overlay system for open-type MRI, which comprises the MRI compatible display device, image communication, registration system and visualization software, which is adopted to clinical 0.2T Open MRI. Phantom experiments were performed and the TRE of the overlay device was less than 1.0mm at the center of the display.","Title: Ultrasonic Motor Driving Method For Emi-Free Image In Mr Image-Guided Surgical Robotic System\nYear: 2007\n\nAbstract: Electromagnetic interference (EMI) between magnetic resonance (MR) imager and surgical manipulator is a severe problem, that degrades the image quality, in MR image-guided surgical robotic systems. We propose a novel motor driving method to acquire noise-free image. Noise generation accompanied by motor actuation is permitted only during the \"dead time\" when the MR imager stops signal acquisition to wait for relaxation of protons. For the synchronized control between MR imager and motor driving system, we adopted a radio-frequency pulse signal detected by a special antenna as a synchronous trigger. This method can be applied widely because it only senses a part of the scanning signal and requires neither hardware nor software changes to the MR imager. The evaluation results showed the feasibility of RF pulse as a synchronous trigger and the availability of sequence-based noise reduction method.","Title: Fast and accurate ultrasonography for visceral fat measurement.\nYear: 2010\n\nAbstract: Visceral fat area (VFA) has close relationship with hypertension, diabetes and cardiovascular disease, and therefore serve as a reliable indicator of these diseases. Abdominal computed tomography (CT) enables precise quantification of the VFA and has been considered as the gold standard for VFA assessment. In this paper, we develope a novel method to quickly and accurately measure the VFA with ultrasonography (US). We evaluated the novel method on five volunteers and the diagnosis procedures lasted less than 30 seconds averagely. The simulation results by our method were compared with VFA estimated by abdominal CT. The correlation coefficient between them was 0.913 for men and 0.858 for women. And the mean deviation of between VFA by CT and by our method was 19.8 cm2 for men and 13.3 cm2 for women.","Title: 3-D augmented reality for MRI-guided surgery using integral videography autostereoscopic image overlay.\nYear: 2010\n\nAbstract: A 3-D augmented reality navigation system using autostereoscopic images was developed for MRI-guided surgery. The 3-D images are created by employing an animated autostereoscopic image, integral videography (IV), which provides geometrically accurate 3-D spatial images and reproduces motion parallax without using any supplementary eyeglasses or tracking devices. The spatially projected 3-D images are superimposed onto the surgical area and viewed via a half-silvered mirror. A fast and accurate spatial image registration method was developed for intraoperative i.v. image-guided therapy. Preliminary experiments showed that the total system error in patient-to-image registration was 0.90 +/- 0.21 mm, and the procedure time for guiding a needle toward a target was shortened by 75%. An animal experiment was also conducted to evaluate the performance of the system. The feasibility studies showed that augmented reality of the image overlay system could increase the surgical instrument placement accuracy and reduce the procedure time as a result of intuitive 3-D viewing.","Title: Automatic Focusing and Robotic Scanning Mechanism for Precision Laser Ablation in Neurosurgery\nYear: 2010\n\nAbstract: We have developed an laser ablation system with an automatic focusing (AF) and robotic scanning mechanism for precision malignant gliomas resection in neurosurgery. A 5-aminolevulinic acid (5-ALA)-induced fluorescence based intra-operative tumor diagnosis technique has been incorporated into the robotic laser ablation system. The system enables an intra-operative identification of the position of a tumor with fluorescence illuminated by a laser excitation. The AF and robotic scanning mechanism assists in tracking the surface of the malignant brain tumors, and provides position information for both laser scanning in 5-ALA fluorescence identification and laser ablation in tumor resection. Experimental results showed the automatic focus device had a precision of 0.5 mm and the mechanism could track the brain surface with a movement caused by pulsation and respiration.","Title: A robot assisted hip fracture reduction with a navigation system.\nYear: 2008\n\nAbstract: A fracture reduction robot is described as assisting in safe and precise fracture reduction. The robot is connected with pins that are inserted into the patient's bone fragments, together with a customized jig. The robot has six degrees of freedom with high precision, so that precise fracture reduction can be conducted. The failsafe unit of the fracture reduction robot can mitigate excessive reduction force that may cause complications such as avascular necrosis. We have integrated the fracture reduction robot with a navigation system that tracks the relative position of the bone fragments and generates the reduction path. The integrated system is evaluated with the simulated fracture reduction of a hip fracture model (n = 8). Three-dimensional parameters related to the mechanical axis--the proximal femur angle, the distal femur angle, and the length of the mechanical axis--were evaluated by comparing the normal values with those after reduction; these average differences are 1.76 degrees , 0.28 degrees and 0.76mm, respectively. The automated fracture reduction feature makes it possible for medical staff to work at a distance from radiation sources; for patients, the integrated fracture reduction system has the potential to reduce fractures with high precision.","Title: Surgical navigation of integral videography image overlay for open MRI-Guided glioma surgery\nYear: 2006\n\nAbstract: A three-dimensional (3D) image overlay navigation system has been developed for open MRI guided surgery. The images are created by employing animated autostereoscopic image of integral videography (IV), which provides geometrically accurate 3D images and reproduce motion parallax without using any supplementary glasses or tracking devices. The spatially projected 3D images are superimposed onto the patient and are viewed via a half-slivered mirror. We developed a spatial image registration method for intraoperative IV image-guided diagnosis and therapy. Two sets of preliminary experiments showed that the total system error was 0.90 ± 0.21mm in patient-to-image registration and the procedure time of guiding a needle toward a target was shortened to 24%. The augmented reality of the image overlay system is possible to increase the surgical instrument placement accuracy and also to reduce the procedure time as the result of intuitive 3D viewing.","Title: Fast Image Mapping of Endoscopic Image Mosaics with Three-Dimensional Ultrasound Image for Intrauterine Treatment of Twin-to-Twin Transfusion Syndrome\nYear: 2008\n\nAbstract: This paper describes a fast image mapping system that integrates endoscopic image mosaics with three-dimensional (3-D) ultrasound images for assisting intrauterine treatment of twin-to-twin transfusion syndrome (TTTS) by laser photocoagulation. Endoscopic laser photocoagulation treatment has a good survival rate and a low complication rate for twins. However, the small field of view and lack of surrounding information makes the identification of vessels anastomosis difficult. We have developed an extended placenta visualization system with the fusion of endoscopic image mosaics with a 3-D ultrasound-image placenta model. Fully automatic and fast calibration is used for endoscope calibration in fluid. The 3-D spatial position of the endoscopic images and the ultrasound image are tracked by a 3-D position tracking device. The mosaiced endoscope images are registered to the surface of the 3-D ultrasound placenta model by using a fast GPU-based image rendering method. Experimental results show that the system may provide an improved and efficient way of planning and guidance in laser photocoagulation TTTS treatment.","Title: Fetus support manipulator with flexible balloon-based stabilizer for endoscopic intrauterine surgery.\nYear: 2006\n\nAbstract: A novel manipulator is described for stabilizing fetus and preventing it from free-floating during the endoscopic intrauterine surgery. Minimally invasive endoscopic fetal surgery enables intrauterine intervention with reduced risk to the mother and fetus. We designed and fabricated a prototype of a fetus supporting manipulator equipped with a flexible bending/curving mechanisms and a soft balloon-based stabilizer. The flexible bending and curving mechanisms enable the stabilizer to reach the target sites within the uterus under an ultrasound-guidance. The balloon-based stabilizer could be inserted into the uterus with a small incision for entry. The accuracy evaluation showed that the maximum error of the curving mechanism was as small as 7 mm and the standard deviation of the bending mechanism was just 1.6 degrees. In the experiments using a fetus model, the manipulator could be well controlled under ultrasound guidance and its curving mechanism with the balloon-based stabilizer could be clearly visualized during the implementation of fetus model supporting. The manipulator has the potential to be used in minimally invasive intrauterine surgery, though further improvements and experiments remain to be carried out.","Title: A coaxial laser endoscope with arbitrary spots in endoscopic view for fetal surgery.\nYear: 2009\n\nAbstract: In this paper, we describe a rigid endoscope that transmits a laser beam coaxially to arbitrary points in the endoscopic view, mainly for treatment of twin-to-twin transfusion syndrome. The endoscope consists of a hotmirror for coaxial transmission of visible light and a Nd:YAG laser beam, and galvanometers for controlling the beam irradiation angle. We evaluated the transmission efficiency of the laser power, the spot size through the endoscope and accuracy in positioning the beam. The maximum laser transmission efficiency was 39% and the spot diameter was 2.2-3.2 mm at a distance of 10-20 mm. The positioning accuracy was mostly within 1.0 mm in the endoscopic view at the distance. The average laser power density on the spot was estimated to be 170-370 W/cm2, and a chicken liver was successfully coagulated by changing the laser beam irradiation angle.","Title: An Autostereoscopic Display System for Image-Guided Surgery Using High-Quality Integral Videography with High Performance Computing\nYear: 2003\n\nAbstract: This paper presents a high-quality glass-less autostereoscopic display system that is integrated into image-guided surgery. The glass-less autostereoscopic image was created by employing a modification of Integral Videography (IV), which is an animated extension of Integral Photography. IV records and reproduces 3-D images using a micro convex lens array and plat display, which can display geometrically accurate 3-D autostereoscopic images and reproduce motion parallax without any need for special viewing devices. This paper reports the use of high-resolution IV display and high performance computing for producing medical 3-D image. We evaluate the feasibility of this display by using a set of medical image with segmentation and reproducing 3-D CT and MRI autostereoscopic image for surgical planning and intra-operative guidance. The main contribution of this paper is application and modification of medical 3-D display technique originally developed in high-resolution autostereoscopic display and high performance computing.","Title: High performance computing for parallel rendering in surgical autostereoscopic display and navigation\nYear: 2003\n\nAbstract: The three-dimensional medical information obtained from Magnetic Resonance Image (MRI), X-ray Computed Tomography (CT), etc. is used for an operation-supporting image, and operations under image guidance have also been conducted. One of the stereoscopic methods is Integral Videography (IV), which is an animated extension of Integral Photography, reproducing a computer-generated graphical object. Though the advantage of IV has been proven in both feasibility studies and clinical applications, one of the issues still unsolved is the notable quantity of calculation that causes significant delay in rendering. Then using parallel processing method integrating Message Passing Interface (MPI) on High Performance Computer (HPC), we shortened the calculating time of IV picture by the shortest at about 0.2 s. Furthermore, by using socket communication, it transmitted to another portable note PC, which is for a display. Then, we achieved the fast presentation of autostereoscopic images seen from an arbitrary direction that was specified with the mouse from the note PC side at anywhere we can access to the network.","Title: High-sensitive fluorescence endoscope using electrocardiograph-synchronized multiple exposure.\nYear: 2011\n\nAbstract: Fluorescence-based measurement of cardiac disease, using autofluorescent substances that already exist in the heart, has not been used for endoscopic surgery because the endoscopic lenses cannot transmit sufficient light. A highly sensitive fluorescence endoscope using an electrocardiograph (ECG)-synchronized multiple exposure (ESME) approach was developed that provides a bright fluorescent image.A system was developed consisting of an endoscope, an excitation light, an ECG amplifier, a trigger and delay unit, and a computer. This system is based on periodic motion of the heart. Since the shape of the heart can be photographed by ECG triggering in a similar manner, a bright image can be synthesized by accumulating multiple trigger-captured images. Laboratory and in vivo experiments were performed to confirm the effectiveness of ESME.The experimental results revealed that the trigger unit generated the synchronization signals required to produce high-quality images of the heart depending on heart rate. The difference among trigger-captured images from the actual organ, which affects the quality of ESME images, was estimated at 0.65 mm from the calculated displacement of a marker on the heart. The results also revealed that a bright fluorescent image can be captured by ESME.A highly sensitive fluorescence endoscope using ESME was developed and successfully tested. The experimental results indicated that the method enabled high-quality image acquisition in a very low illumination environment. This system is effective for the observation of faint fluorescence in the heart and is useful for the intraoperative examination of the heart status.","Title: Rigid-flexible outer sheath model using slider linkage locking mechanism and air pressure for endoscopic surgery.\nYear: 2006\n\nAbstract: The objective of this paper is to develop an outer sheath for flexible endoscopic manipulators. This sheath can switch two states including flexible and rigid, and make a rigid curved path for inserting manipulators. The flexible mode can be curved into a required shape. The rigid mode can hold the shape of the sheath, and then keep the path for instruments. Through the managed path, the flexible manipulators become easy to reach the target. We proposed a serial multi joint model to realize the flexible mechanism. This model is composed of a set of frame units which are connected serially. Each unit can be rotated to a given angle around the center of the joint. We developed a slider-link mechanism and a gear stopper controlled by air pressure for rigid mode. We designed and fabricated the prototype with a diameter of 16 mm and length of 290 mm. The experiment showed that the device could be switched from the flexible mode to the rigid mode when the air pressure was over 150 kPa, and each joint could hold its angle against the maximum 400 mNm. The phantom experiment showed that the flexible devices are possible to transmit the wire tension to the endpoint of the manipulator without changing the curving shape with by the developed outer sheath device.","Title: Integral videography autostereoscopic display using multiprojection\nYear: 2006\n\nAbstract: Various techniques of autostereoscopic three-dimensional (3D) display for displaying real 3D information have been investigated in recent years. We have performed research and application of integral videography (IV), which is an animated display technique based on the principle of integral photography (IP). Although IV and IP are ideal ways to display 3D images, their spatial viewing resolution and high-speed image rendering algorithm must be improved. The pixel pitch of the display and the lens pitch are the main factors affecting IV image quality. We improved the image quality by increasing the number and density of the pixels. Parallel image processing and displaying method were also developed for animated IV imaging. We have created a scalable high-resolution image using multiple projection and project it onto a small screen to achieve high-density pixels using long-focal-length projection optics. Furthermore, an image calibration method was developed for geometrical correction and color modulation to generate seamless IV images. We evaluated the developed device by displaying an X-ray computed tomography (CT) image. Measurement of the geometrical accuracy of the reproduced IV images demonstrated that the spatial resolution of the display system matched that by theoretical analysis. Experimental results showed that our technique is very effective in obtaining 3D animated images with practical spatial resolution. © 2006 Wiley Periodicals, Inc. Syst Comp Jpn, 37(5): 34–45, 2006; Published online in Wiley InterScience (). DOI 10.1002/scj.20424","Title: Surgical Manipulator With Linkage Mechanism For Anterior Cruciate Ligament Reconstruction\nYear: 2007\n\nAbstract: This paper describes a surgical manipulator designed with regard to its safety and sterilization in medical setting of anterior cruciate ligament (ACL) reconstruction. The manipulator system consists of two five-bar linkage mechanisms, an optical tracking device, and a user interface of surgical navigation. We designed and fabricated a prototype manipulator and evaluated the performance of the system by a set of experiments. The accuracy evaluation showed that the average backlash of the 4 link axes was 4.14 degrees and the accuracy of the pitch and yaw angle of the K-wire are 1.55 degrees +/- 2.74 degrees and -2.48 degrees +/- 2.92 degrees. This system enables intrinsic safety and ease of sterilization due to limitations in the range of movement. Furthermore, the system achieves reliabitity because of the simplicity of both its mechanism and its user interface. The manipulator has the potential to be used in minimally invasive ACL reconstruction surgery, although further improvements and experiments remain to be carried out.","Title: Augmented reality system for oral surgery using 3D auto stereoscopic visualization.\nYear: 2011\n\nAbstract: We present an augmented reality system for oral and maxillofacial surgery in this paper. Instead of being displayed on a separated screen, three-dimensional (3D) virtual presentations of osseous structures and soft tissues are projected onto the patient's body, providing surgeons with exact knowledge of depth information of high risk tissues inside the bone. We employ a 3D integral imaging technique which produce motion parallax in both horizontal and vertical direction over a wide viewing area in this study. In addition, surgeons are able to check the progress of the operation in real-time through an intuitive 3D based interface which is content-rich, hardware accelerated. These features prevent surgeons from penetrating into high risk areas and thus help improve the quality of the operation. Operational tasks such as hole drilling, screw fixation were performed using our system and showed an overall positional error of less than 1 mm. Feasibility of our system was also verified with a human volunteer experiment.","Title: Integral Videography Overlay Navigation System Using Mutual Information-Based Registration\nYear: 2004\n\nAbstract: We propose an automatic markerless registration method for Integral Videography (IV) overlay navigation system that directly matches a video image of the patient to a 2-D rendering of a surface model extracted from the 3-D surface model image. IV overlay system is integrated into a surgical navigation system to superimpose a real three-dimensional (3-D) auto stereoscopic image onto the patient via a half-silvered mirror. This registration technique doesn't require the setting of markers, which is invasive and extends the duration of the procedure. Accuracy measurements were performed using a triangular-prism-shaped model. The mean Target Registration Error (TRE) was 1.6 +/- 0.6 mm. We also applied our method to a human face registration and obtained successful results, although some noises were added to the video image. The presented method is more suitable for clinical applications of IV image overlay than previous methods since the presented method is sufficiently accurate, robust and more convenient for surgeon.","Title: Three-dimensional auto-stereoscopic animated image with a long viewing distance using high-precision image correction\nYear: 2009\n\nAbstract: Three-dimensional (3-D) displays have got a lot of attention because they have a much higher sense of realism and are more intuitive than 2-D displays. In particular, autostereoscopic displays are suitable for everyday use because they can be observed from an arbitrary viewpoint without supplementary glasses or tracking devices. Integral Videography (IV) is one of the methods for autostereoscopic animated images that extends Integral Photography (IP) to animation. IP/IV uses a combination of a lens array and a number of calculated elemental images with different perspectives. In particular, IV with a depth of several meters can be applied in many areas. However, most IV reports have an image depth of only several centimeters. Only a small deviation of a lens from its designed position would result in several degrees of deviation of the light ray from the back of the lens. We have developed the static autostereoscopic image by projecting the light sources from an object onto a photographic film through the lens-array [Liao et al. 2005]. In this study, we obtained animated IV of 1 m image depth with less distortion using our method to correct IV images. The method is technically unique.","Title: Effective statistical edge integration using a flux maximizing scheme for volumetric vascular segmentation in MRA.\nYear: 2007\n\nAbstract: Evolutionary schemes based on the level set theory are effective tools for medical image segmentation. In this paper, a new variational technique for edge integration is presented. Region statistical measures and orientation information from ramp-like edges, are fused within an energy minimization scheme that is based on a new interpretation of edge concept. A region driven advection term simulating the edge strength effect is directly obtained from this minimization strategy. We have applied our method to several real Magnetic Resonance Angiography data sets and comparison has been made with a state-of-the-art vessel segmentation method. Presented results indicate that using this method a significant improvement is achievable and the method can be an effective tool to extract vessels in MRA intracranial images.","Title: Development and evaluation of a novel actuator using MR magnetic field\nYear: 2007\n\nAbstract: Recently, Intra-Operative Magnetic Resonance Imaging (IO-MRI) has attracted attention for the reason of operating quality. In the other hand, surgical assist robot with high accuracy is widely researched. By combining IO-MRI and surgical assist robot to an MR-compatible surgical assist robot, higher quality of operation is expected. In this paper, we introduce a novel actuator for MR-compatible Surgical Assist Robot, which actively uses MR magnetic field and is drive by regulating pulse current. Use of ratchet mechanism enables step wise motion of the rotor similar to stepping motors. The position of the rotor can be controlled without an encoder. Clock-wise rotation, counter clock-wise rotation, constrained and non-constrained conditions were achieved by using this mechanism. Evaluation of the first prototype (84times76times58(mm3), 380 (gram)) shows that it is MR-compatible and has a very high positioning accuracy (step angle CW: 12.04plusmn0.28 (deg), CCW : 12.22plusmn0.36(deg); backlash < 1 (deg)). No-load maximum rotational speed is 35.71(rpm) and maximum torque 15.1times10-3 (Nm).","Title: Hazard analysis of fracture-reduction robot and its application to safety design of fracture-reduction assisting robotic system\nYear: 2010\n\nAbstract: In this paper, we discuss the issue of safety in robot-assisted fracture reduction. We define the hazards of robot-assisted fracture reduction and design safety control methods. Although a large reduction force is required to reduce femoral neck fractures, an unexpectedly large force produced by a robot may cause injury to the patient. We have designed two mechanical failsafe units and a software force limiter; this along with velocity control can guarantee a safe operation in reduction force. In addition, to reduce the movement of bone fragments as much as possible, we devised spatially constrained control methods for fracture-reduction robots. The fracture-reduction system was evaluated using simulated fracture reductions.","Title: High Quality Autostereoscopic Surgical Display Using Anti-aliased Integral Videography Imaging\nYear: 2004\n\nAbstract: This paper presents an autostereoscopic three-dimensional (3-D) surgical display with high quality integral videography (IV) rendering algorithm. IV is an animated extension of integral photography, which provides 3-D images without using any supplementary glasses or tracking devices. Despite IV's many advantages, the quality of its spatial image has thus far been poor. We developed a high quality image rendering method with oversampling technique for enhancing the resolution of elemental IV image and low-pass-filter for smoothing the image. Furthermore, we manufactured a high-resolution IV display for evaluating the feasibility of proposed method. The experimental results show the quality of anti-aliased IV image is improved. We also integrated the developed IV image into image-guided surgery display system. This approach will allow us to acquire the optimum process to produce high quality 3-D image for planning and guidance of minimally invasive surgery.","Title: Ultrasound Image-Guided Mapping of Endoscopic Views on a 3D Placenta Model: A Tracker-Less Approach.\nYear: 2013\n\nAbstract: This work presents a framework for mapping of free-hand endoscopic views onto 3D anatomical model constructed from ultrasound images without the use of external trackers. It is non-disruptive in terms of existing surgical workflow as surgeons do not need to accommodate operational constraints associated with the use of additionally installed motion sensors or tracking systems. A passive fiducial marker is attached to the tip of the endoscope to create a geometric eccentricity that encodes the position and orientation of the camera. The relative position between the endoscope and the anatomical model under the ultrasound image reference frame is used to establish a texture map that overlays endoscopic views onto the surface of the model. This addresses operational challenges including the limited field-of-view (FOV) and the lack of 3D perspective associated with minimally invasive procedure. Experimental results show that average tool position and orientation errors are 1.32 mm and 1.6 degrees respectively. The R.M.S. error of the overall image mapping obtained based on comparison of dimension of landmarks is 3.30 mm with standard deviation of 2.14 mm. The feasibility of the framework is also demonstrated through implementations on a phantom model.","Title: Anisotropic Haralick Edge Detection Scheme with Application to Vessel Segmentation\nYear: 2008\n\nAbstract: In this paper, detection of edges in oriented fields is addressed. Haralick edge detection is an accurate scheme for estimation of the edge in a Euclidean space. However, in some applications such as edge detection for vessel segmentation because of the intrinsic orientation of structures, accuracy is only demanded in a particular subspace. This is specially usefull when a curve evolution is chosen for segmentation since gradients in parallel to vessel orientation stops evolution. Haralick edge detection is generalized on a Riemannian space using the inner product of the vectors under a space metric tensor. This eliminates the spurious gradients and preserves the accuracy on the vessel border. Examples are given and the comparison is made with the state-of-the-art flux maximizing flow indicating that significant improvements in terms of leakage minimization and thiner vessel delineation is achievable using our methodology.","Title: Surgical navigation by autostereoscopic image overlay of integral videography.\nYear: 2004\n\nAbstract: This paper describes an autostereoscopic image overlay technique that is integrated into a surgical navigation system to superimpose a real three-dimensional (3-D) image onto the patient via a half-silvered mirror. The images are created by employing a modified version of integral videography (IV), which is an animated extension of integral photography. IV records and reproduces 3-D images using a microconvex lens array and flat display; it can display geometrically accurate 3-D autostereoscopic images and reproduce motion parallax without the need for special devices. The use of semitransparent display devices makes it appear that the 3-D image is inside the patient's body. This is the first report of applying an autostereoscopic display with an image overlay system in surgical navigation. Experiments demonstrated that the fast IV rendering technique and patient-image registration method produce an average registration accuracy of 1.13 mm. Experiments using a target in phantom agar showed that the system can guide a needle toward a target with an average error of 2.6 mm. Improvement in the quality of the IV display will make this system practical and its use will increase surgical accuracy and reduce invasiveness.","Title: Real-time computer-generated integral imaging and 3D image calibration for augmented reality surgical navigation.\nYear: 2015\n\nAbstract: Autostereoscopic 3D image overlay for augmented reality (AR) based surgical navigation has been studied and reported many times. For the purpose of surgical overlay, the 3D image is expected to have the same geometric shape as the original organ, and can be transformed to a specified location for image overlay. However, how to generate a 3D image with high geometric fidelity and quantitative evaluation of 3D image's geometric accuracy have not been addressed. This paper proposes a graphics processing unit (GPU) based computer-generated integral imaging pipeline for real-time autostereoscopic 3D display, and an automatic closed-loop 3D image calibration paradigm for displaying undistorted 3D images. Based on the proposed methods, a novel AR device for 3D image surgical overlay is presented, which mainly consists of a 3D display, an AR window, a stereo camera for 3D measurement, and a workstation for information processing. The evaluation on the 3D image rendering performance with 2560×1600 elemental image resolution shows the rendering speeds of 50-60 frames per second (fps) for surface models, and 5-8 fps for large medical volumes. The evaluation of the undistorted 3D image after the calibration yields sub-millimeter geometric accuracy. A phantom experiment simulating oral and maxillofacial surgery was also performed to evaluate the proposed AR overlay device in terms of the image registration accuracy, 3D image overlay accuracy, and the visual effects of the overlay. The experimental results show satisfactory image registration and image overlay accuracy, and confirm the system usability.","Title: Micro laser ablation system integrated with image sensor for minimally invasive surgery\nYear: 2014\n\nAbstract: A novel micro surgical system with a micro laser ablation module integrated with an imaging sensor is designed and developed for minimally invasive surgery. The system consists of a Nd:YAG laser source with the wavelength of 1064nm to remove lesion, a CCD camera to guide laser, a parallel four-bar mechanism, an endoscope bending section, a rigid steel tube, an aluminum-alloy supporting table and a personal computer for sending control commands. The advantage of the system is the small size of its distal module, i.e., the diameter and the length of the module are 3.5 millimeters and 15.0 millimeters, respectively. A methodological frame for ablating lesion using the system proposed is developed based on the system configuration. Experiments on ex-vivo swine tissue are implemented to evaluate laser ablation performance of the prototyped system.","Title: Floating autostereoscopic 3D display with multidimensional images for telesurgical visualization.\nYear: 2016\n\nAbstract: PurposeWe propose a combined floating autostereoscopic three-dimensional (3D) display approach for telesurgical visualization, which could reproduce live surgical scene in a realistic and intuitive manner.","Title: Vision-based markerless registration using stereo vision and an augmented reality surgical navigation system: a pilot study.\nYear: 2015\n\nAbstract: This study evaluated the use of an augmented reality navigation system that provides a markerless registration system using stereo vision in oral and maxillofacial surgery.A feasibility study was performed on a subject, wherein a stereo camera was used for tracking and markerless registration. The computed tomography data obtained from the volunteer was used to create an integral videography image and a 3-dimensional rapid prototype model of the jaw. The overlay of the subject's anatomic site and its 3D-IV image were displayed in real space using a 3D-AR display. Extraction of characteristic points and teeth matching were done using parallax images from two stereo cameras for patient-image registration.Accurate registration of the volunteer's anatomy with IV stereoscopic images via image matching was done using the fully automated markerless system, which recognized the incisal edges of the teeth and captured information pertaining to their position with an average target registration error of < 1 mm. These 3D-CT images were then displayed in real space with high accuracy using AR. Even when the viewing position was changed, the 3D images could be observed as if they were floating in real space without using special glasses.Teeth were successfully used for registration via 3D image (contour) matching. This system, without using references or fiducial markers, displayed 3D-CT images in real space with high accuracy. The system provided real-time markerless registration and 3D image matching via stereo vision, which, combined with AR, could have significant clinical applications.","Title: 3D Surgical Overlay with Markerless Image Registration Using a Single Camera.\nYear: 2015\n\nAbstract: Minimum invasive surgery can benefit from surgical visualization, which is achieved by either virtual reality or augmented reality. We previously proposed an integrated 3D image overlay based surgical visualization solution including 3D image rendering, distortion correction, and spatial projection. For correct spatial projection of the 3D image, image registration is necessary. In this paper we present a 3D image overlay based augmented reality surgical navigation system with markerless image registration using a single camera. The innovation compared with our previous work lies in the single camera based image registration method for 3D image overlay. The 3D mesh model of patient's teeth which is created from the preoperative CT data is matched with the intraoperative image captured by a single optical camera to determine the six-degree-of-freedom pose of the model with respect to the camera. The obtained pose is used to superimpose the 3D image of critical hidden tissues on patient's body directly via a translucent mirror for surgical visualization. The image registration performs automatically within approximate 0.2 s, which enables real-time update to tackle patient's movement. Experimental results show that the registration accuracy is about 1mm and confirm the feasibility of the 3D surgical overlay system.","Title: Registration of CT and Ultrasound Images of the Spine with Neural Network and Orientation Code Mutual Information.\nYear: 2016\n\nAbstract: Pairwise registration of 2D ultrasound (US) and 3D computed tomography (CT) images can improve the efficiency and safety of image-guided anesthesia in spine surgery. However, accurate 2D US and 3D CT registration for multiple vertebras without an appropriate initial registration position is still a challenge, due to the difference of image modalities and missing bone structures in US image. This paper proposes a novel 2D US and 3D CT registration method, in which convolutional neural network (CNN) classification of US images is reported for the first time to achieve rough image registration. And a new orientation code mutual information metric is further applied to finish local registration refinement. By combining automatic rough registration with fine registration refinement, our algorithm achieves 2D US and 3D CT registration for multiple vertebras (L2-L4) without the requirement of an appropriate initial alignment. The accuracy of our algorithm is validated on 50 in vivo clinical US images dataset of multiple vertebras. And a mean target registration error of 2.3 mm is acquired, which is lower than the clinically acceptable accuracy 3.5 mm.","Title: Radiation-Free 3D Navigation and Vascular Reconstruction for Aortic Stent Graft Deployment.\nYear: 2016\n\nAbstract: We propose a radiation-free three dimensional (3D) navigation and vasculature reconstruction method to reduce aortic stent graft deployment time and repeated exposures to high doses of X-ray radiation. 2D intraoperative US images are fused with 3D preoperative magnetic resonance (MR) image to provide intuitive 3D navigation for deployment of the stent-graft in the proposed system. On the other hand, calibrated 2D US images track catheter’s tip and construct intraoperative 3D aortic model by combining segmented intravascular ultrasound (IVUS) images. The constructed 3D aortic models can quantitatively assess morphological characteristics of aortic to assist deployment of stent graft. This system was validated by using in vitro cardiac and aorta phantom. The mean target registration error of 2D US-3D MR registration was 2.70 mm and the average tracking error of position of IVUS catheter’s tip was 1.12 mm. Accurate contours detection results in IVUS images were acquired. Meanwhile, Hausdorff distances are 0.78 mm and 0.59 mm for outer and inner contours, respectively; Dice coefficients are 90.21 % and 89.96 %. Experiment results demonstrate that our radiation-free navigation and 3D vasculature reconstruction method is promising for deployment of stent graft in vivo studies.","Title: Automatic Fast-Registration Surgical Navigation System Using Depth Camera and Integral Videography 3D Image Overlay.\nYear: 2016\n\nAbstract: We propose an automatic fast-registration augmented reality (AR) surgical navigation system for minimally invasive surgery. The system integrates a fast-registration technique with three-dimensional (3D) integral videography (IV) image overlay. The detailed anatomic information generated by IV technique is superimposed to the patient using 3D autostereoscopic images, which reproduce motion parallax with naked eyes. To reduce the patient-3D overlay image registration time and achieve automatic execution, we integrate a 3D image overlay system with a real-time patient tracking system, which utilizes particle filter algorithm and depth camera. Experimental results showed that the system can lower the registration time and the can reach up to 10 frames per second (fps). The 3D overlay image to the patient registration average error is 1.88 mm, with standard deviation of 0.72 mm. Further work for improvement of the depth camera acquisition accuracy and tracking algorithm makes this system more feasible and practical.","Title: 3D Catheter Shape Determination for Endovascular Navigation Using a Two-Step Particle Filter and Ultrasound Scanning\nYear: 2017\n\nAbstract: In endovascular catheter interventions, the determination of the three-dimensional (3D) catheter shape can increase navigation information and help reduce trauma. This study describes a shape determination method for a flexible interventional catheter using ultrasound scanning and a two-step particle filter without X-ray fluoroscopy. First, we propose a multi-feature, multi-template particle filte...","Title: Augmented reality surgical navigation with ultrasound-assisted registration for pedicle screw placement: a pilot study.\nYear: 2017\n\nAbstract: We present a novel augmented reality (AR) surgical navigation system based on ultrasound-assisted registration for pedicle screw placement. This system provides the clinically desired targeting accuracy and reduces radiation exposure.Ultrasound (US) is used to perform registration between preoperative computed tomography (CT) images and patient, and the registration is performed by least-squares fitting of these two three-dimensional (3D) point sets of anatomical landmarks taken from US and CT images. An integral videography overlay device is calibrated to accurately display naked-eye 3D images for surgical navigation. We use a 3.0-mm Kirschner wire (K-wire) instead of a pedicle screw in this study, and the K-wire is calibrated to obtain its orientation and tip location. Based on the above registration and calibration, naked-eye 3D images of the planning path and the spine are superimposed onto patient in situ using our AR navigation system. Simultaneously, a 3D image of the K-wire is overlaid accurately on the real one to guide the insertion procedure. The targeting accuracy is evaluated postoperatively by performing a CT scan.An agar phantom experiment was performed. Eight K-wires were inserted successfully after US-assisted registration, and the mean targeting error and angle error were 3.35 mm and [Formula: see text], respectively. Furthermore, an additional sheep cadaver experiment was performed. Four K-wires were inserted successfully. The mean targeting error was 3.79 mm and the mean angle error was [Formula: see text], and US-assisted registration yielded better targeting results than skin markers-based registration (targeting errors: 2.41 vs. 5.18 mm, angle errors: [Formula: see text] vs. [Formula: see text].Experimental outcomes demonstrate that the proposed navigation system has acceptable targeting accuracy. In particular, the proposed navigation method reduces repeated radiation exposure to the patient and surgeons. Therefore, it has promising prospects for clinical use.","Title: 3D interactive surgical visualization system using mobile spatial information acquisition and autostereoscopic display.\nYear: 2017\n\nAbstract: Work flow of 3D interactive surgical visualization system using mobile spatial information acquisition and autostereoscopic display.","Title: High-Quality See-Through Surgical Guidance System Using Enhanced 3-D Autostereoscopic Augmented Reality.\nYear: 2017\n\nAbstract: Objective: Precise minimally invasive surgery (MIS) has significant advantages over traditional open surgery in clinic. Although pre-/intraoperative diagnosis images can provide necessary guidance for therapy, hand-eye discoordination occurs when guidance information is displayed away from the surgical area. In this study, we introduce a real three-dimensional (3-D) see-through guidance system for...","Title: Real-time Lens based Rendering Algorithm for Super-multiview Integral Photography without Image Resampling.\nYear: 2018\n\nAbstract: We propose a computer generated integral photography (CGIP) method that employs a lens based rendering (LBR) algorithm for super-multiview displays to achieve higher frame rates and better image quality without pixel resampling or view interpolation. The algorithm can utilize both fixed and programmable graphics pipelines to accelerate CGIP rendering and inter-perspective antialiasing. Two hardwar...","Title: Visible Forceps Manipulator With Novel Linkage Bending Mechanism For Neurosurgery\nYear: 2017\n\nAbstract: In minimally invasive surgery (MIS), especially the neurosurgery, surgeons often suffer from occlusion region problem. Common surgical instruments, like endoscope and corresponding operating tools, are hard to solve it due to their size and rigid mechanical structure. In this paper, we present a visible forceps manipulator with novel linkage bending mechanism, which realizes the flexible bending capability and high output force, as well as the integrated endoscopic function. We present the simplified experiment to evaluate the results of mechanical performance and brain phantom test to evaluate feasibility and usefulness in neurosurgery. Preliminary results show that phantom experiments using the brain phantom verify the feasibility of the novel manipulator.","Title: Spatial Position Measurement System for Surgical Navigation Using 3-D Image Marker-Based Tracking Tools With Compact Volume.\nYear: 2018\n\nAbstract: We develop a spatial position measurement system using three-dimensional (3-D) image marker-based tracking tools targeted at surgical navigation in minimally invasive surgery. We generate 3-D image markers with spatial information encoded to 2-D images, design tracking tools with the 3-D image markers, and analyze the tracking tools' theoretical spatial errors, which are primarily limited by the s...","Title: A novel integration of spectral-domain optical-coherence-tomography and laser-ablation system for precision treatment.\nYear: 2018\n\nAbstract: We propose an optical integrated diagnosis and therapeutic system. The primary experimental results show the high efficiency and feasibility of our theranostics system, which is promising for realizing accurate resection of tumors in vivo and in situ in the future.","Title: Lumen and media-adventitia border detection in IVUS images using texture enhanced deformable model.\nYear: 2018\n\nAbstract: •Propose a texture enhanced deformable model (TEDM) for accurate detection of lumen and media–adventitia (MA) borders in intravascular ultrasound (IVUS) images.•TEDM has scientific importance related to the quantitative plaque information and 3D vessel visualization.•The accuracy of the TEDM is validated by using 1500 IVUS images from 15 in vivo patient sequences.•TEDM can detect lumen and MA surfaces precisely with small surface distance errors of 0.17 mm and 0.19 mm, respectively.","Title: Optimized Optical Coherence Tomography Imaging With Hough Transform-Based Fixed-Pattern Noise Reduction.\nYear: 2018\n\nAbstract: Fixed-pattern noise seriously affects the clinical application of optical coherence tomography (OCT), especially, in the imaging of tumorous tissue. We propose a Plough transform-based fixed-pattern noise reduction (HTFPNR) method to reduce the fixed-pattern noise for optimizing imaging of tumorous tissue with OCT system. Using by the HTFPNR method, we detect and map the outline of fixed-pattern noise in the OCT images, and finally efficiently reduce the fixed-pattern noise by the longitudinal and horizontal intelligent processing procedure. We adopt the image-to-noise ratio with full information (INRfi) and the noise reduction ratio (NRR) to evaluate the outcome of fixed-pattern noise reduction ratio, respectively. The INRfi of OCT image's noise reduction of ex vivo brainstem tumor is approximate 21.92 dB. Six groups of OCT images, including three types of fixed-pattern noises, have been validated via experimental evaluation of the ex vivo gastric tumor. In the different types of fixed-pattern noise, the mean INR(fi)s are 25.24, 23.04, and 19.35 dB, respectively. This result demonstrates that it is highly efficient and useful in fixed-pattern noise reduction. The fluctuating range of the NRR is 0.84-0.88 for three types of added noise in the OCT images. This result demonstrates that the HTFPNR method can as possible as save useful information by comparing to previous research. This proposed HTFPNR method can be used into the fixed-pattern noise reduction of OCT images in other soft biological tissue in the future.","Title: Clustering of Morphological Features for Identifying Femur Cavity Subtypes With Difficulties of Intramedullary Nail Implantation.\nYear: 2018\n\nAbstract: Intramedullary (IM) nail implantation is currently the standard treatment for femoral intertrochanteric fractures. However, individual differences in femur cavity bring a challenge in designing well-matched IM nails and cause difficulties in IM nail implantation. Therefore, there is an intense need to analyze femur cavities to predict difficulties in IM nail implantation to assist the design of IM...","Title: A Cascaded Deep Convolutional Neural Network for Joint Segmentation and Genotype Prediction of Brainstem Gliomas.\nYear: 2018\n\nAbstract: Goal: Automatic segmentation of brainstem gliomas and prediction of genotype (H3 K27M) mutation status based on magnetic resonance (MR) images are crucial but challenging tasks for computer-aided diagnosis in neurosurgery. In this paper, we present a novel cascaded deep convolutional neural network (CNN) to address these two challenging tasks simultaneously. Methods: Our novel segmentation task co...","Title: Reconstruction and Registration of Large-Scale Medical Scene Using Point Clouds Data from Different Modalities.\nYear: 2018\n\nAbstract: Sensing the medical scenario can ensure the safety during the surgical operations. So, in this regard, a monitor platform which can obtain the accurate location information of the surgery room is desperately needed. Compared to 2D camera image, 3D data contains more information of distance and direction. Therefore, 3D sensors are more suitable to be used in surgical scene monitoring. However, each 3D sensor has its own limitations. For example, Lidar (Light Detection and Ranging) can detect large-scale environment with high precision, but the point clouds or depth maps are very sparse. As for commodity RGBD sensors, such as Kinect, can accurately capture denser data, but limited to a small range from 0.5 to 4.5m. So, a proper method which can address these problems for fusing different modalities data is important. In this paper, we proposed a method which can fuse different modalities 3D data to get a large-scale and dense point cloud. The key contributions of our work are as follows. First, we proposed a 3D data collecting system to reconstruct the medical scenes. By fusing the Lidar and Kinect data, a large-scale medical scene with more details can be reconstructed. Second, we proposed a location-based fast point clouds registration algorithm to deal with different modality datasets.","Title: An Intelligent Theranostics Method Using Optical Coherence Tomography Guided Automatic Laser Ablation for Neurosurgery.\nYear: 2018\n\nAbstract: Methods to reduce the high disability and fatality rate of neurosurgery caused by surgeon tremor and uncertainty of intraoperative tumor boundary have long been greatly concerned. We have proposed an intelligent theranostics method for a compact integrated diagnosis and therapeutic platform to automatically excise the brain tumor with high precision. To perform this operation, hardware combination of a benchtop spectral domain optical coherence tomography (SD-OCT) and a high-power laser has been implemented by a customized fiber combiner. By the double-clad fiber out of the combiner which works as both the OCT sample arm fiber and the ablation laser output fiber, an in situ theranostics process without registration can be conducted. The SD-OCT image analysis from both en face and depth perspectives shows the structural information intuitively and accordingly generate the ablation-guiding grid. Based on the grid, the continuous wave high-power laser ablates ex vivo porcine brain models in submillimeter automatically. The feasibility of this theranostics method has been verified by the comparison of attenuation maps before and after ablation, and the contrast of its pathology.","Title: Augmented reality surgical navigation with accurate CBCT-patient registration for dental implant placement.\nYear: 2019\n\nAbstract: It is challenging to achieve high implant accuracy in dental implant placement, because high risk tissues need to be avoided. In this study, we present an augmented reality (AR) surgical navigation with an accurate cone beam computed tomography (CBCT)-patient registration method to provide clinically desired dental implant accuracy. A registration device is used for registration between preoperative data and patient outside the patient's mouth. After registration, the registration device is worn on the patient's teeth for tracking the patient. Naked-eye 3D images of the planning path and the mandibular nerve are superimposed onto the patient in situ to form an AR scene. Simultaneously, a 3D image of the drill is overlaid accurately on the real one to guide the implant procedure. Finally, implant accuracy is evaluated postoperatively. A model experiment was performed by an experienced dentist. Totally, ten parallel pins were inserted into five 3D-printed mandible models guided by our AR navigation method and through the dentist's experience, respectively. AR-guided dental implant placement showed better results than the dentist's experience (mean target error = 1.25 mm vs. 1.63 mm; mean angle error = 4.03° vs. 6.10°). Experimental results indicate that the proposed method is expected to be applied in the clinic. Graphical abstract ᅟ.","Title: Unified Mathematical Model for Multilayer-Multiframe Compressive Light Field Displays Using LCDs.\nYear: 2019\n\nAbstract: We propose a unified mathematical model for multilayer-multiframe compressive light field displays that supports both attenuation-based and polarization-based architectures. We show that the light field decomposition of such a display can be cast as a bound constrained nonlinear matrix optimization problem. Efficient light field decomposition algorithms are developed using the limited-memory BFGS (L-BFGS) method for automultiscopic displays with high resolution and high image fidelity. In addition, this framework is the first to support multilayer polarization-based compressive light field displays with time multiplexing. This new architecture significantly reduces artifacts compared with attenuation-based multilayer-multiframe displays; thus, it can allow the requirements regarding the number of layers or the refresh rate to be relaxed. We verify the proposed methods by constructing two 3-layer prototypes using high-speed LCDs, one based on the attenuation architecture and one based on the polarization architecture. Moreover, an efficient CUDA-based program is implemented. Our displays can produce images with higher spatial resolution with thinner form factors compared with traditional automultiscopic displays in both simulations and experiments.","Title: Three-Dimensional Feature-Enhanced Network for Automatic Femur Segmentation.\nYear: 2019\n\nAbstract: Automatic femur segmentation from computed tomography volume is a crucial but challenging task for computer-aided diagnosis in orthopedic surgeries. The main obstacles are weak bone boundaries, narrowness of joint space, variations in femur density and shape, as well as diverse leg postures. In this paper, we presented a novel 3-D feature-enhanced network to address these challenges. The novelty o...","Title: Automatic distance measurement of abdominal aorta for ultrasonography-based visceral fat estimation.\nYear: 2013\n\nAbstract: Ultrasonography-based visceral fat estimation is a promising method to assess central obesity, which is associated with metabolic syndrome. The key to this method is to measure three types of distance in the ultrasound image. The most important one is the distance from the skin surface to the posterior wall of the abdominal aorta. We present a novel automatic measurement method to calculate this distance using 1D ultrasound signal processing. It is different from the conventional 2D image processing based methods which have high failure rate when the target is blurred or partially imaged. The proposed method identifies the waveforms of the aorta along a group of ultrasound scan lines and a rating mechanism is introduced to choose the best waveform for distance calculation. The robustness and accuracy of the method were evaluated by experiments based on clinical data.","Title: Design and Modularization of Multi-DoF Soft Robotic Actuators\nYear: 2019\n\nAbstract: Soft robotic technologies offer a promising solution for unconstructed applications due to their complete softness, robustness, and flexibility. In this letter, we explore the design and modularization of the flexible pneumatic robotic actuators for applying in minimally invasive surgery or other constrained spaces. We first identify the basic multi-cavity structure and subsequently explore three ...","Title: Augmented reality-based autostereoscopic surgical visualization system for telesurgery\nYear: 2021\n\nAbstract: Purpose The visualization of remote surgical scenes is the key to realizing the remote operation of surgical robots. However, current non-endoscopic surgical robot systems lack an effective visualization tool to offer sufficient surgical scene information and depth perception. Methods We propose a novel autostereoscopic surgical visualization system integrating 3D intraoperative scene reconstruction, autostereoscopic 3D display, and augmented reality-based image fusion. The preoperative organ structure and the intraoperative surface point cloud are obtained from medical imaging and the RGB-D camera, respectively, and aligned by an automatic marker-free intraoperative registration algorithm. After registration, preoperative meshes with precalculated illumination and intraoperative textured point cloud are blended in real time. Finally, the fused image is shown on a 3D autostereoscopic display device to achieve depth perception. Results A prototype of the autostereoscopic surgical visualization system was built. The system had a horizontal image resolution of 1.31 mm, a vertical image resolution of 0.82 mm, an average rendering rate of 33.1 FPS, an average registration rate of 20.5 FPS, and average registration errors of approximately 3 mm. A telesurgical robot prototype based on 3D autostereoscopic display was built. The quantitative evaluation experiments showed that our system achieved similar operational accuracy (1.79 +/- 0.87 mm) as the conventional system (1.95 +/- 0.71 mm), while having advantages in terms of completion time (with 34.11% reduction) and path length (with 35.87% reduction). Post-experimental questionnaires indicated that the system was user-friendly for novices and experts. Conclusion We propose a 3D surgical visualization system with augmented instruction and depth perception for telesurgery. The qualitative and quantitative evaluation results illustrate the accuracy and efficiency of the proposed system. Therefore, it shows great prospects in robotic surgery and telesurgery.","Title: Intelligent optical diagnosis and treatment system for automated image-guided laser ablation of tumors\nYear: 2021\n\nAbstract: Purpose For tumor resections near critical structures, accurate identification of tumor boundaries and maximum removal are the keys to improve surgical outcome and patient survival rate, especially in neurosurgery. In this paper, we propose an intelligent optical diagnosis and treatment system for tumor removal, with automated lesion localization and laser ablation. Methods The proposed system contains a laser ablation module, an optical coherence tomography (OCT) unit, and a robotic arm along with a stereo camera. The robotic arm can move the OCT sample arm and the laser ablation front-end to the suspected lesion area. The corresponding diagnosis and treatment procedures include computer-aided lesion segmentation using OCT, automated ablation planning, and laser control. The ablation process is controlled by a deflectable mirror, and a non-common-path ablation planning algorithm based on the transformation from lesion positions to mirror deflection angles is presented. Results Phantom and animal experiments are carried out for system verification. The robot could reach the planned position with high precision, which is approximately 1.16 mm. Tissue classification with OCT images achieves 91.7% accuracy. The error of OCT-guided automated laser ablation is approximately 0.74 mm. Experiments on mouse brain tumors show that the proposed system is capable of clearing lesions efficiently and precisely. We also conducted an ex vivo porcine brain experiment to verify the whole process of the system. Conclusion An intelligent optical diagnosis and treatment system is proposed for tumor removal. Experimental results show that the proposed system and method are promising for precise and intelligent theranostics. Compared to conventional cancer diagnosis and treatment, the proposed system allows for automated operations monitored in real-time, with higher precision and efficiency.","Title: Force-guided autonomous robotic ultrasound scanning control method for soft uncertain environment\nYear: 2021\n\nAbstract: Purpose Autonomous ultrasound imaging by robotic ultrasound scanning systems in complex soft uncertain clinical environments is important and challenging to assist in therapy. To cope with the complex environment faced by the ultrasound probe during the scanning process, we propose an autonomous robotic ultrasound (US) control method based on reinforcement learning (RL) model to build the relationship between the environment and the system. The proposed method requires only contact force as input information to achieve robot control of the posture and contact force of the probe without any a priori information about the target and the environment. Methods First, an RL agent is proposed and trained by a policy gradient theorem-based RL model with the 6-degree-of-freedom (DOF) contact force of the US probe to learn the relationship between contact force and output force directly. Then, a force control strategy based on the admittance controller is proposed for synchronous force, orientation and position control by defining the desired contact force as the action space. Results The proposed method was evaluated via collected US images, contact force and scan trajectories by scanning an unknown soft phantom. The experimental results indicated that the proposed method differs from the free-hand scanned approach in the US images within 3 +/- 0.4%. The analysis results of contact forces and trajectories indicated that our method could make stable scanning processes on a soft uncertain skin surface and obtained US images. Conclusion We propose a concise and efficient force-guided US robot scanning control method for soft uncertain environment based on reinforcement learning. Experimental results validated our method's feasibility and validity for complex skin surface scanning, and the volunteer experiments indicated the potential application value in the complex clinical environment of robotic US imaging system especially with limited visual information.","Title: H3K27M Mutations Prediction for Brainstem Gliomas Based on Diffusion Radiomics Learning\nYear: 2021\n\nAbstract: H3K27M mutation is the most common mutation in brainstem gliomas (BSGs), which is related with highly invasive neoplasms and poor prognosis. Accurate presurgical and noninvasive prediction of H3K27M mutations based on preoperative multi-modal neuroimaging is of great clinical value in the diagnosis, prognosis and therapeutic selection of BSGs. Traditional BSG radiomics models usually only focus on tumor local morphometric characteristics. However, given that highly invasive BSGs may significantly affect large-scale brain network connectivity, we reasonably infer that local radiomics and global connectomics may provide different perspectives for H3K27M genotype prediction. Therefore, we define a graph-based diffusion radiomics learning model to integrate these two kinds of features seamlessly. Specifically, edges of the defined brain network are determined by neural fiber connections, while node features of brainstem are governed by local tumor radiomics. Upon this model, we further propose a multi-mechanism diffusion convolutional network to couple multi-modal information and generate a joint representation for brain disease diagnosis. By graph diffusion convolution, the local radiomics information spread along the brain network structure to enhance graph representation learning, and eventually the learned diffusion radiomics features contribute to disease prediction. Experiments on a real BSG dataset demonstrate the effectiveness and advantages of our proposed method for preoperative prediction of H3K27M statuses.","Title: 3D Volume Visualization and Screen-based Interaction with Dynamic Ray Casting on Autostereoscopic Display\nYear: 2021\n\nAbstract: Augmented reality (AR) is an emerging technology to improve visualization experiences. However, visualizing volume data is limited in existing AR systems due to the lack of an intuitive and precise exploration scheme. In this paper, we present a 3D augmented volume visualization and screen-based interaction method. An autostereoscopic handheld display is utilized to achieve naked-eye 3D perception...","Title: An Unsupervised Convolution Neural Network for Deformable Registration of Mono/Multi-Modality Medical Images\nYear: 2021\n\nAbstract: Image registration is a fundamental and crucial step in medical image analysis. However, due to the differences between mono-mode and multi-mode registration tasks and the complexity of the corresponding relationship between multi-mode image intensity, the existing unsupervised methods based on deep learning can hardly achieve the two registration tasks simultaneously. In this paper, we proposed a novel approach to register both mono- and multi-mode images in a same framework. By approximately calculating the mutual information in a differentiable form and combining it with CNN, the deformation field can be predicted quickly and accurately without any prior information about the image intensity relationship. The registration process is implemented in an unsupervised manner, avoiding the need for the ground truth of the deformation field. We utilize two public datasets to evaluate the performance of the algorithm for mono-mode and multi-mode image registration, which confirms the effectiveness and feasibility of our method. In addition, the experiments on patient data also demonstrate the practicability and robustness of the proposed method.","Title: Image-Based 3D Ultrasound Reconstruction with Optical Flow via Pyramid Warping Network\nYear: 2021\n\nAbstract: 3D Ultrasound (US) contains rich spatial information which is helpful for medical diagnosis. However, current reconstruction methods with tracking devices are not suitable for clinical application. The sensorless freehand methods reconstruct based on US images which is less accuracy. In this paper, we proposed a network which reconstructs the US volume based on US images features and optical flow features. We proposed the pyramid warping layer which merges the image features and optical flow features with warping operation. To fuse the warped features of different scales in different pyramid levels, we adopted the fusion module using the attention mechanism. Meanwhile, we adopted the channel attention and spatial attention to our network. Our method was evaluated in 100 freehand US sweeps of human forearms which exhibits the efficient performance on volume reconstruction compared with other methods.","Title: A Smoke Removal Method Based on Combined Data and Modified U-Net for Endoscopic Images\nYear: 2021\n\nAbstract: In minimally invasive surgery, the ablation of human tissue will produce a lot of smoke, which will interfere with the surgeon's operation. We propose a smoke removal method based on combined data and modified U-net for endoscopic images. The real dataset and the synthetic dataset are built using a small amount of images with smoke. The real dataset is combined with the synthetic dataset successively. Qualitative evaluation shows that the quality of the output smoke-free image is the best when training using the combined data, compared to using only either the real dataset or the synthetic dataset above. Quantitative evaluation shows that the effect of smoke removal is still the best when training using the combined data in our method.","Title: TypeSeg: A type-aware encoder-decoder network for multi-type ultrasound images co-segmentation\nYear: 2022\n\nAbstract: Purpose: As a portable and radiation-free imaging modality, ultrasound can be easily used to image various types of tissue structures. It is important to develop a method which supports the multi-type ultrasound images co-segmentation. However, state-of-the-art ultrasound segmentation methods commonly only focus on the single type images or ignore the type-aware information. Methods: To solve the above problem, this work proposes a novel type-aware encoder-decoder network (TypeSeg) for the multi-type ultrasound images co-segmentation. First, we develop a type-aware metric learning module to find an optimum latent feature space where the ultrasound images of the same types are close and that of the different types are separated by a certain margin. Second, depending on the extracted features, a decision module decides whether the input ultrasound images have the common tissue type or not, and the encoder-decoder network produces a segmentation mask accordingly. Results: We evaluate the performance of the proposed TypeSeg model on the ultrasound dataset that contains four types of tissues. The proposed TypeSeg model achieves the overall best results with the mean IOU score of 87.51% +/- 3.93% for the multi-type ultrasound images. Conclusion: The experimental results indicate that the proposed method outperforms all the compared state-of-the-art algorithms for the multi-type ultrasound images co-segmentation task. (C) 2021 Elsevier B.V. All rights reserved.","Title: Ultrasound Speckle Reduction Using Wavelet-Based Generative Adversarial Network\nYear: 2022\n\nAbstract: The visual quality of ultrasound (US) images is crucial for clinical diagnosis and treatment. The main source of image quality degradation is the inherent speckle noise generated during US image acquisition. Current deep learning-based methods cannot preserve the maximum boundary contrast when removing noise and speckle. In this paper, we address the issue by proposing a novel wavelet-based generative adversarial network (GAN) for real-time high-quality US image reconstruction, viz. WGAN-DUS. First, we propose a batch normalization module (BNM) to balance the importance of each sub-band image and fuse sub-band features simultaneously. Then, a wavelet reconstruction module (WRM) integrated with a cascade of wavelet residual channel attention block (WRCAB) is proposed to extract distinctive sub-band features used to reconstruct denoised images. A gradual tuning strategy is proposed to fine-tune our generator for better despeckling performance. We further propose a wavelet-based discriminator and a comprehensive loss function to effectively suppress speckle noise and preserve the image features. Besides, we have designed an algorithm to estimate the noise levels during despeckling of real US images. The performance of our network was then evaluated on natural, synthetic, simulated and clinical US images and compared against various despeckling methods. To verify the feasibility of WGAN-DUS, we further extend our work to uterine fibroid segmentation with the denoised US image of the proposed approach. Experimental result demonstrates that our proposed method is feasible and can be generalized to clinical applications for despeckling of US images in real-time without losing its fine details.","Title: Rib Suppression in Digital Chest Tomosynthesis\nYear: 2022\n\nAbstract: Digital chest tomosynthesis (DCT) is a technique to produce sectional 3D images of a human chest for pulmonary disease screening, with 2D X-ray projections taken within an extremely limited range of angles. However, under the limited angle scenario, DCT contains strong artifacts caused by the presence of ribs, jamming the imaging quality of the lung area. Recently, great progress has been achieved for rib suppression in a single X-ray image, to reveal a clearer lung texture. We firstly extend the rib suppression problem to the 3D case at the software level. We propose a Tomosynthesis RIb SuPpression and Lung Enhancement Network (TRIPLE-Net) to model the 3D rib component and provide a rib-free DCT. TRIPLE-Net takes the advantages from both 2D and 3D domains, which models the ribs in DCT with the exact FBP procedure and 3D depth information, respectively. The experiments on simulated datasets and clinical data have shown the effectiveness of TRIPLE-Net to preserve lung details as well as improve the imaging quality of pulmonary diseases. Finally, an expert user study confirms our findings. Our code is available at https://github.com/sunyhl/Rib-Suppressionin-Digital-Chest-Tomosynthesis.","Title: Statistical and individual characteristics-based reconstruction for craniomaxillofacial surgery\nYear: 2022\n\nAbstract: In craniomaxillofacial (CMF) surgery planning, a preoperative reconstruction of the CMF reference model is crucial for surgical restoration, especially the reconstruction of bilateral defects. Current reconstruction algorithms mainly generate reference models from the image analysis aspect, however, clinical indicators of the CMF reference model mostly consider the distribution of anatomical landmarks. Generating a reference model with optimal clinical evaluation helps promote the feasibility of an algorithm. We first build a dataset with 100 normal skull models and then calculate a statistical shape model (SSM) and the distribution of normal cephalometric values, which indicate the statistical features of a population. To further generate personalized reference models, we apply non-rigid registration to align the SSM with the defect skull model. An evaluation standard to select the optimal reference model considers both global performance and anatomical evaluation. Moreover, we develop a landmark detection network to improve the automatic level of the algorithm. The proposed method performs better than methods including Iterative Closest Point and SSM. From a global evaluation aspect, the results show that the RMSE between the reference model and the ground truth is \n \n \n \n $$1.71 \\pm 0.23$$\n \n mm, the percentage of vertices with error below 2 mm is \n \n \n \n $$85 \\pm 4$$\n \n % and the average faces distance is \n \n \n \n $$1.38 \\pm 0.20$$\n \n mm (better than the state-of-the-art method). From the anatomical evaluation aspect, the target registration error between the landmark pairs is \n \n \n \n $$4.12 \\pm 2.27$$\n \n mm. In addition, the clinical application confirms that the reference model can meet clinical requirements. To the best of our knowledge, we propose the first CMF reconstruction method considering the global performance of reconstruction and anatomically local evaluation from clinical experience. Simulated experiments and clinical cases prove the general applicability and strength of the method.","Title: Efficient lower-limb segmentation for large-scale volumetric CT by using projection view and voxel group attention\nYear: 2022\n\nAbstract: 3D segmentation of lower-limb bones such as tibia and fibula from volumetric computed tomography (CT) is very important for surgical planning and navigation. However, the input data of the large-scale 3D volumetric CT will increase the computation cost, and meantime, the spatial connections of the long-range voxels will be easily ignored. To solve these problems, this paper proposes an accurate and efficient 3D bone segmentation approach based on the distribution characteristics of the lower-limb bones in volumetric CT. On one hand, the efficient segmentation framework was built for the large-scale volumetric CT using 2D projection view-based slice filtering and parameter-reduced separable convolution. On the other hand, we developed a new voxel group attention mechanism to emphasize the connection of the long-range voxel groups and improve the representational capability of the segmentation network. The experimental results showed that the proposed 3D bone segmentation approach achieved high segmentation accuracy under the conditions of limited computations. It was additionally shown that the proposed approach outperformed the state-of-the-art 3D models for the segmentation of lower-limb bones.","Title: Spatial Position Estimation Method for 3D Ultrasound Reconstruction Based on Hybrid Transfomers\nYear: 2022\n\nAbstract: 3D ultrasound reconstruction has important application value in clinical diagnosis. To obtain the spatial relationships between ultrasound images in reconstruction tasks, especially the contextual information in long-range sequences, this work proposes a spatial position estimation method for 3D ultrasound image reconstruction based on a hybrid transformer encoder. Inspired by the transformer sequ...","Title: Spatiotemporal reconstruction method of carotid artery ultrasound from freehand sonography\nYear: 2022\n\nAbstract: 4D reconstruction based on radiation-free ultrasound can provide valuable information about the anatomy. Current 4D US technologies are either faced with limited field-of-view (FoV), technical complications, or cumbersome setups. This paper proposes a spatiotemporal US reconstruction framework to enhance its ability to provide dynamic structure information. We propose a spatiotemporal US reconstruction framework based on freehand sonography. First, a collecting strategy is presented to acquire 2D US images in multiple spatial and temporal positions. A morphology-based phase extraction method after pose correction is presented to decouple the compounding image variations. For temporal alignment and reconstruction, a robust kernel regression model is established to reconstruct images in arbitrary phases. Finally, the spatiotemporal reconstruction is demonstrated in the form of 4D movies by integrating the US images according to the tracked poses and estimated phases. Quantitative and qualitative experiments were conducted on the carotid US to validate the feasibility of the proposed pipeline. The mean phase localization and heart rate estimation errors were 0.07 ± 0.04 s and 0.83 ± 3.35 bpm, respectively, compared with cardiac gating signals. The assessment of reconstruction quality showed a low RMSE (<0.06) between consecutive images. Quantitative comparisons of anatomy reconstruction from the generated US volumes and MRI showed an average surface distance of 0.39 ± 0.09 mm on the common carotid artery and 0.53 ± 0.05 mm with a landmark localization error of 0.60 ± 0.18 mm on carotid bifurcation. A novel spatiotemporal US reconstruction framework based on freehand sonography is proposed that preserves the utility nature of conventional freehand US. Evaluations on in vivo datasets indicated that our framework could achieve acceptable reconstruction performance and show potential application value in the US examination of dynamic anatomy.","Title: Augmented reality navigation with ultrasound-assisted point cloud registration for percutaneous ablation of liver tumors\nYear: 2022\n\nAbstract: We present a novel augmented reality (AR) surgical navigation method with ultrasound-assisted point cloud registration for percutaneous ablation of liver tumors. A preliminary study is carried out to verify its feasibility. Two three-dimensional (3D) point clouds of the liver surface are derived from the preoperative images and intraoperative tracked US images, respectively. To compensate for the soft tissue deformation, the point cloud registration between the preoperative images and the liver is performed using the non-rigid iterative closest point (ICP) algorithm. A 3D AR device based on integral videography technology is designed to accurately display naked-eye 3D images for surgical navigation. Based on the above registration, naked-eye 3D images of the liver surface, planning path, entry points, and tumor can be overlaid in situ through our 3D AR device. Finally, the AR-guided targeting accuracy is evaluated through entry point positioning. Experiments on both the liver phantom and in vitro pork liver were conducted. Several entry points on the liver surface were used to evaluate the targeting accuracy. The preliminary validation on the liver phantom showed average entry-point errors (EPEs) of 2.34 ± 0.45 mm, 2.25 ± 0.72 mm, 2.71 ± 0.82 mm, and 2.50 ± 1.11 mm at distinct US point cloud coverage rates of 100%, 75%, 50%, and 25%, respectively. The average EPEs of the deformed pork liver were 4.49 ± 1.88 mm and 5.02 ± 2.03 mm at the coverage rates of 100% and 75%, and the average covered-entry-point errors (CEPEs) were 4.96 ± 2.05 mm and 2.97 ± 1.37 mm at 50% and 25%, respectively. Experimental outcomes demonstrate that the proposed AR navigation method based on US-assisted point cloud registration has achieved an acceptable targeting accuracy on the liver surface even in the case of liver deformation.","Title: Anatomically constrained deformable 3D reconstruction of intraoperative uterus from preoperative MRI data on uterine fibroid treatment\nYear: 2022\n\nAbstract: Detections of the uterus using preoperative magnetic resonance imaging (MRI) data are required for intraoperative navigation in High-Intensity Focused Ultrasound (HIFU) treatment of uterine fibroids. The modelling of the intraoperative uterus between empty and full bladder anatomy deems to be a crucial step to quantify uterus deformation caused by the bladder-and-rectum-filling (BRF) technique. In this paper, a novel two-stage conditional generative adversarial network (cGAN) is proposed to perform intraoperative uterus deformable reconstruction with anatomical constraints using only single preoperative MRI data, viz. ACDeformRec. The highly constrained anatomy properties are further captured by a novel attention network that formulates high-level uterus segmentation task by incorporating correlations of surrounding organs for more anatomically plausible reconstruction. The experimental result demonstrates the robustness of ACDeformRec by evaluating its performance on 181 clinical datasets and has achieved the lowest reconstruction error of 0.735 +/- 0.045 mm with Dice Similarity Coefficient of 94.23% and Normalised Cross Correlation of 97.23%.","Title: Annotation-guided encoder-decoder network for bone extraction in ultrasound-assisted orthopedic surgery\nYear: 2022\n\nAbstract: The patients and surgeons are usually exposed in massive ionizing radiation during fluoroscopy-based navigation orthopedic surgery. Comparatively, ultrasound-assisted orthopedic surgery could not only decrease the risk of radiation but also provide rich navigation information. However, due to the artifacts in ultrasound images, the extraction of bone structure from ultrasound sequences can be a particularly difficult task, which leads to some major challenges in ultrasound-assisted orthopedic navigation. In this paper, we propose an annotation-guided encoder-decoder network (AGN) to extract bone structure from the radiation-free ultrasound sequences. Specifically, the variability of the ultrasound probe's pose leads to the change of the ultrasound frame during the acquisition of ultrasound sequences. Therefore, a feature alignment module deployed in the AGN model is used to achieve reliable matching across ultrasound frames. Moreover, inspired by the interactive ultrasound analysis, where user annotated foreground information can help target extraction, our AGN model incorporates the annotation information obtained by Siamese networks. Experimental results validated that the AGN model not only produced better bone surface extraction than state-of-the-art methods (IOU: 0.92 versus. 0.88), but also achieved almost real-time extraction with the speed about 15 frames per second. In addition, the acquired bone surface further provided radiation-free 3D intraoperative bone structure for the intuitive navigation of orthopedic surgery.","Title: Deformable Torso Anatomy Education with Three-Dimensional Autostereoscopic Visualization and Free-Hand Interaction\nYear: 2022\n\nAbstract: Virtual reality/augmented reality has advantages in immersive learning and conveying meta-information in anatomy education. In this study, we present an interactive naked-eye 3D torso anatomy education environment based on population anatomical information. In particular, we utilize the deformable anatomy models, constructed from big data of healthy adults, to convey the anatomy knowledge of truthfulness organ shapes and population anatomical variances. In addition, the proposed anatomy education system is conveyed by free-hand 3D autostereoscopic visualization, which supports multi-user and glass-free. A user study with fourteen students was implemented and demonstrates that the system is appropriate and useful for torso anatomy education.","Title: A LESS SUPERVISED AUTOMATIC DELINEATION METHOD FOR INTRACRANIAL GERM CELL TUMOR RADIOTHERAPY TARGETS\nYear: 2022\n\nAbstract: Central nervous system germ cell tumor is one of the most common intracranial malignant tumors in children. Currently, radiotherapy is the main treatment method. Local tumor and whole ventricular system irradiation can ensure the curative effect while reducing radiation damage in the late stage of radiotherapy. However, it is time-consuming and laborious for doctors to manually delineate the ventricular system. Therefore, this research proposed a less supervised automatic delineation method for intracranial germ cell tumor radiotherapy targets. By integrating the idea of unsupervised image registration, it has fast segmentation speed and avoids the dependence of supervised learning methods on extensive labeled data. The average target delineation accuracy can reach 80.5% and the average time is 153.42s when only one set of annotated data is required. The proposed algorithm can assist doctors in radiotherapy target delineation and has the potential to be extended to actual clinical applications.","Title: Uncertainty-aware Cascade Network for Ultrasound Image Segmentation with Ambiguous Boundary\nYear: 2022\n\nAbstract: Ultrasound image segmentation plays an essential role in automatic disease diagnosis. However, to achieve precise ultrasound segmentation is still a challenge caused by the ambiguous lesion boundary and imaging artifacts such as speckles and shadowing noise. Considering that the pixels with high uncertainty generally distributing in the boundary regions of prediction maps, are likely to overlap with the confused regions of ultrasound, we proposed an uncertainty-aware cascade network. Our network uses the confidence map to evaluate the uncertainty of each pixel to enhance the segmentation of ambiguous boundary. On the one hand, the confidence map fuses with the ultrasound features and predicted mask using the adaptive fusion module (AFM) which enriches the context features from different modalities. In addition, the uncertainty attention module (UAM) is proposed based on the confidence map. This module focuses on the influential features with cross attention constrained by the uncertainty of pixels which can extract the localized features of confused ultrasound regions. On the other hand, the recurrent edge correction module (RECM) further improves the segmentation of ambiguous boundary. This module increases the weights of confident features neighboring the uncertainty boundaries in order to refine the predictions of edge pixels with low confidence. We evaluated the proposed method on three public ultrasound datasets and the segmentation results show that our method achieved higher Dice scores and lower Hausdorff distance with more precise boundary details compared with state-of-the-art methods."],"string":"[\n \"Title: Long Visualization Depth Autostereoscopic Display using Light Field Rendering based Integral Videography\\nYear: 2006\\n\\nAbstract: three-dimensional (3D) computer graphics (CG) image without using special viewing glasses or tracking devices. The images are created by employing a light field rendering and pixel distribution algorithm for integral photography (IP) / integral videography (IV) imaging. How to enhance the image depth, especial the long visualization depth, is a challenge work. In this study, the images are rendered from a referential viewing area for each viewpoint and the elemental images are reconstructed by pixel redistribution method. The corresponding result images are projected to a screen that is separated from the lens array by a referential viewing distance as the setting of image rendering. A photographic film is used to record the elemental image through each lens with a photograph-taken method. Our photograph-taken based IV display enables precise 3D images with long visualization depth to be displayed at long viewing distances without any influence from deviated or distorted lenses in a lens array. We created an autostereoscopic display that appears to have three-dimensionality even when viewed from a distance, with an image depth of 5.7 m or more in front of the display. To the best of our knowledge, the presented long-distance IV display is technically unique as it is the first report of generating an autostereoscopic image with such a long viewing distance in the field of computer graphics. CR Categories and Subject Descriptors: B.4.2 Input/Output Devices - Image Display\",\n \"Title: A novel high intensity focused ultrasound robotic system for breast cancer treatment.\\nYear: 2013\\n\\nAbstract: High intensity focused ultrasound (HIFU) is a promising technique for cancer treatment owing to its minimal invasiveness and safety. However, skin burn, long treatment time and incomplete ablation are main shortcomings of this method. This paper presents a novel HIFU robotic system for breast cancer treatment. The robot has 4 rotational degrees of freedom with the workspace located in a water tank for HIFU beam imaging and ablation treatment. The HIFU transducer combined with a diagnostic 2D linear ultrasound probe is mounted on the robot end- effector, which is rotated around the HIFU focus when ablating the tumor. HIFU beams are visualized by the 2D probe using beam imaging. Skin burn can be prevented or alleviated by avoiding long time insonification towards the same skin area. The time cost could be significantly reduced, as there is no need to interrupt the ablation procedure for cooling the skin. In addition, our proposed robot control strategies can avoid incomplete ablation. Experiments were carried out and the results showed the effectiveness of our proposed system.\",\n \"Title: Pulsative flow segmentation in MRA image series by AR modeling and EM algorithm\\nYear: 2006\\n\\nAbstract: Segmentation of CSF and pulsative blood flow, based on a single phase contrast MRA (PC-MRA) image can lead to imperfect classifications. In this paper, we present a novel automated flow segmentation method by using PC-MRA image series. The intensity time series of each pixel is modeled as an autoregressive (AR) process and features including the Linear Prediction Coefficients (LPC), covariance matrix of LPC and variance of prediction error are extracted from each profile. Bayesian classification of the feature space is then achieved using a non-Gaussian likelihood probability function and unknown parameters of the likelihood function are estimated by a generalized Expectation-Maximization (EM) algorithm. The efficiency of the method evaluated on both synthetic and real retrospective gated PC-MRA images indicate that robust segmentation of CSF and vessels can be achieved by using this method.\",\n \"Title: High-Resolution Stereoscopic Surgical Display Using Parallel Integral Videography and Multi-projector\\nYear: 2002\\n\\nAbstract: This paper proposes a high-resolution stereoscopic surgical display using the integral videography (IV) and multiple projectors. IV projects a computer generated graphical object by multiple rays through \\\"fly's eye lens\\\", which can display geometrically accurate autostereoscopic images and reproduce motion parallax in three-dimensional (3-D) space. This technique requires neither special glasses nor sensing device to track viewer's eyes, thus being suitable for pre-operative diagnosis and intra-operative use. This paper reports the use of multiple reduction projection display system and a parallel-calculation to achieve high-resolution IV image. We evaluate the feasibility of this display by developing a 3-D CT stereoscopic image and applying it to surgical planning and intra-operative guidance. The main contribution of this paper is application and modification of medical stereoscopic technique originally developed in high-resolution multi-projector stereoscopic display system.\",\n \"Title: Nonmagnetic rigid and flexible outer sheath with pneumatic interlocking mechanism for minimally invasive surgical approach.\\nYear: 2009\\n\\nAbstract: We developed a nonmagnetic rigid and flexible outer sheath with pneumatic interlocking mechanism using flexible toothed links and a wire-driven bending distal end. The outer sheath can be switched between rigid and flexible modes easily depending on surgical scenes, and the angle of its distal end can be controlled by three nylon wires. All components of flexible parts are made of MRI-compatible nonmagnetic plastics. We manufactured the device with 300-mm long, 16-mm outer diameter, 7-mm inner diameter and 90-mm bending distal end. Holding power of the device in rigid mode was maximum 3.6 N, which was sufficient for surgical tasks in body cavity. In vivo experiment using a swine, our device performed smooth insertion of a flexible endoscope and a biopsy forceps into reverse side of the liver, intestines and spleen with a curved path. In conclusion, our device shows availability of secure approach of surgical instruments into deep cavity.\",\n \"Title: MRI-Compatible Rigid and Flexible Outer Sheath Device with Pneumatic Locking Mechanism for Minimally Invasive Surgery\\nYear: 2008\\n\\nAbstract: To reduce the invasiveness of surgery, we developed an outer sheath device using a flexible toothed link and pneumatic locking mechanism that works with flexible devices used in minimally invasive surgery. The outer sheath can be switched between flexible and rigid modes, and the angle of its tip can be controlled by a nylon wire. All parts of this device are made of plastic and are MRI-compatible. We manufactured a sheath prototype, 300 mm long, with a 20-mm outer diameter, and an 8-mm inner diameter. Experiment results showed that the outer sheath can protect tissues from high insertion force and secure the path for flexible devices. It can follow a curved path with a reasonable radius.\",\n \"Title: An image restoration method for extracting features from three-dimensional auto-stereoscopic integral photography images\\nYear: 2011\\n\\nAbstract: Integral Photography (IP) [Hongen Liao 2011] is a visualization technique that produce true three-dimensional images in the real space that can be viewed without supplementary glasses. Using a micro lens array, different aspects of the object can be viewed from different directions, giving observers a sense of depth. Utilizing IP images into augmented reality systems (where images are merged into a real scene) requires an automated alignment between virtual IP images and real objects, leading to the need of understanding the displayed IP images. A natural approach would be capturing multiple views of the IP images and then extracting necessary information from those views. However, since IP images are viewed through a lens array, a lens-like pattern is produced in the observed image(Fig.1(a)). When captured by a camera, simply applying existing image processing methods such as feature extraction would result in unexpected behaviors (e.g. detection of fault features). The purpose of this study is to develop an image restoration method by removing those lens-like pattern from the IP image captured by a camera.\",\n \"Title: Interactive 3D Navigation System for Image-guided Surgery\\nYear: 2009\\n\\nAbstract: This paper presents a novel surgery navigation system based on a three-dimensional (3D) imaging technique, integral videography (IV). In our system, the 3D structure of the object of interest is reconstructed using surface rendering and corresponding pixel distribution methods. We developed a high-speed algorithm that renders high-quality IV images from the surface model in real time and allows interactions like rotating and scaling to be done smoothly. Using the patient-image registration method, IV images can be displayed with the correct size and relative position with respect to the surgical instruments. Experiments were carried out with various anatomical models, and the results show that our system could be useful in many clinical situations such as orthopedic surgery and neurosurgery.\",\n \"Title: Image-based electrode array tracking for epicardial electrophysiological mapping in minimally invasive arrhythmia surgery.\\nYear: 2011\\n\\nAbstract: Electrophysiological mapping is effective in realizing a precise minimally invasive arrhythmia surgery. Recently, an epicardial electrophysiological mapping system for minimally invasive arrhythmia surgery was reported. The system requires a small electrode array, a tracking system and a global mapping algorithm. The optical tracking system employed in the research requires line of sight and complicated configuration. This paper proposes a new tracking method for locating an electrode array.We developed a small electrode array and optical markers. Center points of respective optical markers and the electrode array are tracked via an endoscopic stream and calculated in image space. The orientation of the electrode array is calculated using the dot product between the vector joining two center points of two upper optical markers and the vector joining two end points of the longest edge of the electrode array.Mean tracking errors of position and orientation of the electrode array were 0.51 mm and 0.64°, respectively. And the processing time was constant at 46 ms per frame. Our method could successfully track the electrode array on the epicardium during in vivo experiment and a global epicardial electrophysiological map was reconstructed from separately measured epicardial electrograms by the small electrode array.An image-based tracking method for locating an electrode array was proposed. Tracking accuracy, processing time and applicability to surgical environment of our method proved to be acceptable. Consequently, our method enables the electrode array tracking system to be simplified with no separate tracking system.\",\n \"Title: A Bayesian Approach for Construction of Sparse Statistical Shape Models Using Dirichlet Distribution.\\nYear: 2013\\n\\nAbstract: Statistical shape models (SSMs) made using point sets are important tools to capture the variations within shape populations. One popular method for construction of SSMs is based on the Expectation-Maximization (EM) algorithm which establishes probabilistic matches between the model and training points. In this paper, we propose a novel Bayesian framework to automatically determine the optimal number of the model points. We use a Dirichlet distribution as a prior to enforce sparsity on the mixture weights of Gaussians. Insignificant model points are determined and pruned out using a quadratic programming technique. We apply our method to learn a sparse SSM from 15 manually segmented caudate nuclei data sets. The generalization ability of the proposed model compares favorably to a traditional EM based model.\",\n \"Title: Ultrasound imaging method for internal jugular vein measurement and estimation of circulating blood volume.\\nYear: 2014\\n\\nAbstract: Evaluation of circulating blood volume is important in assessing the status of patients. Although some studies have suggested that ultrasound images of the internal jugular vein (IJV) can be used for the analysis of circulating blood volume, accurate extraction of IJV is necessary to reduce errors. Therefore, this study was designed to develop a new algorithm for dynamic segmentation of IJV and determine appropriate indicators to evaluate the circulating blood volume.Our algorithm is based on snake and speckle tracking models. As the region of interest (ROI) of the control points of the snake tracking algorithm was dynamically moved using speckle tracking, ROI size can be decreased leading to a reduction in the tracking error. Some experiments were performed to validate our algorithm. Subsequently, the algorithm was used for the experiment simulating dehydration state among 11 subjects.Results of the validation experiment suggest that our algorithm showed higher performance for IJV extraction compared with standard methods. Furthermore, it was revealed that some indices such as the average area of IJV were related to the dehydration state of subjects.This study proposed a new algorithm, which was based on snake and speckle tracking models, for dynamic extraction of IJV in ultrasound images. In addition to algorithm validation, it was suggested that some indices the ultrasound image of IJV could be used for the evaluation of the circulating blood volume.\",\n \"Title: A variational method for geometric regularization of vascular segmentation in medical images.\\nYear: 2008\\n\\nAbstract: In this paper, a level-set-based geometric regularization method is proposed which has the ability to estimate the local orientation of the evolving front and utilize it as shape induced information for anisotropic propagation. We show that preserving anisotropic fronts can improve elongations of the extracted structures, while minimizing the risk of leakage. To that end, for an evolving front using its shape-offset level-set representation, a novel energy functional is defined. It is shown that constrained optimization of this functional results in an anisotropic expansion flow which is usefull for vessel segmentation. We have validated our method using synthetic data sets, 2-D retinal angiogram images and magnetic resonance angiography volumetric data sets. A comparison has been made with two state-of-the-art vessel segmentation methods. Quantitative results, as well as qualitative comparisons of segmentations, indicate that our regularization method is a promising tool to improve the efficiency of both techniques.\",\n \"Title: Real-Time Marker-Free Patient Registration and Image-Based Navigation Using Stereovision for Dental Surgery.\\nYear: 2013\\n\\nAbstract: Surgical navigation techniques have been evolving rapidly in the field of oral and maxillofacial surgery (OMS). However, challenges still exist in the current state of the art of computer-assisted OMS especially from the viewpoint of dental surgery. The challenges include the invasive patient registration procedure, the difficulty of reference marker attachment, navigation error caused by patient movement, bulky optical markers and maintenance of line of sight for commercial optical tracking devices, inaccuracy and susceptibility of electromagnetic (EM) sensors to magnetic interference for EM tracking devices. In this paper, a new solution is proposed to overcome the mentioned challenges. A stereo camera is designed as a tracking device for both instrument tracking and patient tracking, which is customized optimally for the limited surgical space of dental surgery. A small dot pattern is mounted to the surgical tool for instrument tracking, which can be seen by the camera at all times during the operation. The patient registration is achieved by patient tracking and 3D contour matching with the preoperative patient model, requiring no fiducial marker and reference marker. In addition, the registration is updated in real-time. Experiments were performed to evaluate our method and an average overall error of 0.71 mm was achieved.\",\n \"Title: New 4-D imaging for real-time intraoperative MRI: adaptive 4-D scan.\\nYear: 2006\\n\\nAbstract: Aiming at real-time 3-D visualization of organ motion to navigate surgical procedures in MRI-guided surgery, a new 4-D MR imaging technique called \\\"Adaptive 4-D Scan\\\" has been proposed. The technique is designed to acquire a time series of volumetric 3-D images (4-D image) of cyclically moving organ, even in a low-field open-configuration MR scanner. A pre-operative 4-D image is acquired with respiratory phase parameter, which is monitored by using navigator-echo-based real-time tracking of the liver and diaphragm. During operation, the respiratory phase is again monitored in real-time, and a 3-D image, reflecting the current state of the target organ, is extracted from the pre-operative 4-D image and provided to physicians as a pseudo real-time 3-D image. We implemented Adaptive 4-D Scan into a 0.5 Tesla open-configuration clinical MRI system for intervention. Phantom and volunteer studies were performed to assess feasibility of this technique, in terms of image quality, imaging time and position accuracy of the imaged subject. A 4-D image (matrix: 256 x 128 x 10 x 8) of cyclically moving phantom was acquired in 719 s, and RMS position error between the imaged subject and the real subject was 2.3 mm, where the range of motion was 50 mm. 4-D image of the moving liver was also successfully acquired under near clinical condition. In conclusion, the study shows that the proposed method is feasible and has capability to provide real-time dynamic 3-D atlas for surgical navigation.\",\n \"Title: Robotic hair harvesting system: a new proposal.\\nYear: 2011\\n\\nAbstract: Follicular Unit Extraction (FUE) has become a popular hair transplanting method for solving male-pattern baldness problem. Manually harvesting hairs one by one, however, is a tedious and time-consuming job to doctors. We design an accurate hair harvesting robot with a novel and efficient end-effector which consists of one digital microscope and a punch device. The microscope is first employed to automatically localize target hairs and then guides the punch device for harvesting after shifting. The end-effector shows average bias and precision of 0.014 mm by virtue of a rotary guidance design for the motorized shifting mechanism.\",\n \"Title: A fully automatic 2D ultrasound probe calibration method using a simple phantom.\\nYear: 2011\\n\\nAbstract: Probe calibration should be performed for free-hand tracking and ultrasound (US) fusion with a position sensor-attached US probe in US guided interventions. Difficulties in locating the probe imaging plane and adjusting the alignment of the probe with the calibration phantom call for more complicated and precise phantoms. However, the manufacture of such phantoms is time-consuming and expensive. Furthermore, the segmentation of the phantom features in US images cannot be reliably automated, which requires manual intervention. This paper presents a fully automatic 2D probe calibration method using a rather simple phantom. By detecting and analyzing the temporal signal arising from the ultrasound artifacts when sweeping the 2D probe over the phantom, the imaging plane of the probe and the image features can be reliably identified. Then the features are fed to a nonlinear optimizer to estimate the calibration parameters. The whole procedure including the setup can be done within 2 minutes with the fiducial registration error (FRE) of 0.16 mm. A US-guided biopsy experiment has also been conducted which shows the target registration error (TRE) of less than 1 mm. © 2011 IEEE.\",\n \"Title: Autostereoscopic 3D Display with Long Visualization Depth using Referential Viewing Area based Integral Photography.\\nYear: 2011\\n\\nAbstract: We developed an autostereoscopic display for distant viewing of three-dimensional (3D) computer graphics (CG) images without using special viewing glasses or tracking devices. The images are created by employing referential viewing area based CG image generation and pixel distribution algorithm for integral photography (IP) and integral videography (IV) imaging. CG image rendering is used to generate IP/IV elemental images. The images can be viewed from each viewpoint within a referential viewing area and the elemental images are reconstructed from rendered CG images by pixel redistribution and compensation method. The elemental images are projected onto a screen that is placed at the same referential viewing distance from the lens array as in the image rendering. Photographic film is used to record the elemental images through each lens. The method enables 3D images with a long visualization depth to be viewed from relatively long distances without any apparent influence from deviated or distorted lenses in the array. We succeeded in creating an actual autostereoscopic images with an image depth of several meters in front of and behind the display that appear to have three-dimensionality even when viewed from a distance.\",\n \"Title: Augmented reality navigation with automatic marker-free image registration using 3-D image overlay for dental surgery.\\nYear: 2014\\n\\nAbstract: Computer-assisted oral and maxillofacial surgery (OMS) has been rapidly evolving since the last decade. State-of-the-art surgical navigation in OMS still suffers from bulky tracking sensors, troublesome image registration procedures, patient movement, loss of depth perception in visual guidance, and low navigation accuracy. We present an augmented reality navigation system with automatic marker-free image registration using 3-D image overlay and stereo tracking for dental surgery. A customized stereo camera is designed to track both the patient and instrument. Image registration is performed by patient tracking and real-time 3-D contour matching, without requiring any fiducial and reference markers. Real-time autostereoscopic 3-D imaging is implemented with the help of a consumer-level graphics processing unit. The resulting 3-D image of the patient's anatomy is overlaid on the surgical site by a half-silvered mirror using image registration and IP-camera registration to guide the surgeon by exposing hidden critical structures. The 3-D image of the surgical instrument is also overlaid over the real one for an augmented display. The 3-D images present both stereo and motion parallax from which depth perception can be obtained. Experiments were performed to evaluate various aspects of the system; the overall image overlay error of the proposed system was 0.71 mm.\",\n \"Title: Precision-guided surgical navigation system using laser guidance and 3D autostereoscopic image overlay.\\nYear: 2010\\n\\nAbstract: This paper describes a precision-guided surgical navigation system for minimally invasive surgery. The system combines a laser guidance technique with a three-dimensional (3D) autostereoscopic image overlay technique. Images of surgical anatomic structures superimposed onto the patient are created by employing an animated imaging method called integral videography (IV), which can display geometrically accurate 3D autostereoscopic images and reproduce motion parallax without the need for special viewing or tracking devices. To improve the placement accuracy of surgical instruments, we integrated an image overlay system with a laser guidance system for alignment of the surgical instrument and better visualization of patient's internal structure. We fabricated a laser guidance device and mounted it on an IV image overlay device. Experimental evaluations showed that the system could guide a linear surgical instrument toward a target with an average error of 2.48mm and standard deviation of 1.76mm. Further improvement to the design of the laser guidance device and the patient-image registration procedure of the IV image overlay will make this system practical; its use would increase surgical accuracy and reduce invasiveness.\",\n \"Title: Real-Time Autostereoscopic Visualization of Registration-Generated 4D MR Image of Beating Heart\\nYear: 2008\\n\\nAbstract: This paper presents a real-time autostereoscopic visualization system using the principle of Integral Videography(IV). We develop MIP and composite volume ray casting method for IV volume rendering, and implemented the algorithm on GPU to achieve real-time rendering. The system was used to visualize 4D MR image that was generated from registration of 3D MR image and 4D ultrasound image. The registration scheme consists of inter-modality rigid registration between 3D MR image and 3D ultrasound image and intra-modality non-rigid registration between 3D ultrasound images. Registration processes were also implemented on GPU. Evaluation of processing speed showed that GPU processing time was 48x, 13x, 21x faster than CPU processing time for IV volume rendering, rigid registration, and non-rigid registration respectively. We also enabled real-time user interactivity for IV visualization system. In the future, We plan to use this system to develop intra-operative surgery navigation system for intra-cardiac surgery on beating heart.\",\n \"Title: PHYSICALLY ACCURATE B-SPLINE BASED NON-RIGID REGISTRATION USING VARIABLE SPRING MODEL\\nYear: 2007\\n\\nAbstract: This paper presents a physically accurate fast parametric non-rigid registration method based on B-spline deformation model. Our method is based on using variable spring model to add a physical constraint to ensure that the deformation does not violate elasticity characteristic of the organs. Using the assumption of human organs as elastic bodies, we modeled the whole volume as 3-D springs connecting each neighboring voxels. Registration was performed on B-spline deformation model with the combination of SSD and total spring energy as optimization criterion. We also added the concept of elasticity map to made it possible to model a more specific elasticity characteristic such as hardness/softness of the organ. We showed that compared to the original algorithm, our algorithm was able to realize a more realistic deformation. We also applied our method to register 3D CT images of human heart from different heart beat phases.\",\n \"Title: An integrated diagnosis and therapeutic system using intra-operative 5-aminolevulinic-acid-induced fluorescence guided robotic laser ablation for precision neurosurgery.\\nYear: 2012\\n\\nAbstract: We have developed an integrated diagnosis and therapeutic system for precision malignant gliomas resection during neurosurgery. A combination of three-dimensional (3-D) magnetic resonance imaging (MRI) navigation and 5-aminolevulinic acid (5-ALA)-induced fluorescence based intra-operative tumor diagnosis technique has been incorporated into a robotic laser ablation neurosurgery system with an automatic focusing and robotic scanning mechanism. 5-ALA is a non-fluorescent prodrug that leads to intracellular accumulation of fluorescent protoporphyrins IX (PpIX) in malignant glioma. The PpIX tends to accumulate in pathological lesions, and emits red fluorescence when excited by blue light. This fluorescence is illuminated with laser excitation, enables intra-operative identification of the position of a tumor and provides guidance for resection with laser photocoagulation. The information provided by the MRI is enhanced by the intra-operative 5-ALA fluorescence data, and this enhanced information is integrated into a robotic laser ablation system. The accuracy of the fluorescent measurement of the tumor is improved using high-precision spectral analysis. The fluorescence assists in the detection of malignant brain tumors intra-operatively and improves their removal rate.\",\n \"Title: Vessel bifurcation localization based on intraoperative three-dimensional ultrasound and catheter path for image-guided catheter intervention of oral cancers.\\nYear: 2013\\n\\nAbstract: We present a method to localize intraoperative target vessel bifurcations under bones for ultrasound (US) image-guided catheter interventions. A catheter path is recorded to acquire skeletons for the target vessel bifurcations that cannot be imaged by intraoperative US. The catheter path is combined with the centerlines of the three-dimensional (3D) US image to construct a preliminary skeleton. Based on the preliminary skeleton, the orientations of target vessels are determined by registration with the preoperative image and the bifurcations were localized by computing the vessel length. An accurate intraoperative vessel skeleton is obtained for correcting the preoperative image to compensate for vessel deformation. A reality check of the proposed method was performed in a phantom experiment. Reasonable results were obtained. The in vivo experiment verified the clinical workflow of the proposed method in an in vivo environment. The accuracy of the centerline length of the vessel for localizing the target artery bifurcation was 2.4mm. These results suggest that the proposed method can allow the catheter tip to stop at the target artery bifurcations and enter into the target arteries. This method can be applied for virtual reality-enhanced image-guided catheter intervention of oral cancers.\",\n \"Title: Surgical Manipulator with Balloon for Stabilizing Fetus in Utero under Ultrasound Guidance\\nYear: 2008\\n\\nAbstract: This paper describes a surgical manipulator to stabilize intrauterine fetus with ultrasound image guidance. The manipulator includes an outline of 4 mm in diameter, a mechanism with 7 joints, and a set of balloons. The manipulator is arranged as straight form and the balloon is fold to be a minimum size before the insertion. Accuracy evaluation of bending performance showed that the standard deviations were ±3.6 degrees on wired-driven mechanism and ±1.6 degrees on linkage-driven mechanism. Experimental results also demonstrated high repeatability of the mechanisms. In feasibility experiments, ultrasound images in 2D and 3D modes were examined for guiding the manipulator. The 2D images provided wide view and easily viewable display of the balloon inflation. The 3D images provided easily viewable display of bending motion of the arm and the relative position of a phantom. It is possible to operate the manipulator in utero under the ultrasound guidance in 2D and 3D by switching them in each procedural stage.\",\n \"Title: Intra-operative Real-Time 3-D Information Display System Based on Integral Videography\\nYear: 2001\\n\\nAbstract: A real-time 3-D surgical navigation system that superimposes the real, intuitive 3-D image for medical diagnosis and operation was developed in this paper. This system creates 3-D image based on the principle of integral photography (IP), named \\\"Integral Videography (IV)\\\", which can display geometrically accurate 3-D image and reproduce motion parallax without any need of special devices. 3-D image was supenmposed on the surgical fields in the patient via a half-silvered mirror as if they could be seen through the body. In addition, a real-time IV algorithm for calculating the 3-D image of surgical instruments was used for registration between the location of surgical instruments and the organ during the operation. The experimental results of puncturing a point location and avoiding critical area showed the errors of this navigation system were in the range of 2-3mm. By introducing a display device with higher pixel density, accuracy of the system can be improved.\",\n \"Title: Novel endoscope system with plasma flushing for off-pump cardiac surgery.\\nYear: 2009\\n\\nAbstract: The purpose of this study is to develop a new endoscope for performing simple surgical tasks inside a cardiac atrium/chamber filled with blood, i.e., for performing \\\"off-pump\\\" cardiac surgeries. In general, it is very difficult to observe the inner wall of the vessels containing circulating blood because the light from the endoscope is scattered by the red blood cells. \\\"Plasma flushing\\\" performed using the separator system is developed to observe the inner side of the heart filled with blood and to remove blood cells from the front of the endoscope tip. The system was used in in vitro quantitative measurement of the device performance and in vivo experiments on a swine. In these experiments, we successfully obtained high-resolution images of the interior of the heart during off-pump surgery.\",\n \"Title: Combination of intraoperative 5-aminolevulinic acid-induced fluorescence and 3-D MR imaging for guidance of robotic laser ablation for precision neurosurgery.\\nYear: 2008\\n\\nAbstract: A combination of 5-aminolevulinic acid (5-ALA)-induced fluorescence and three-dimensional (3-D) magnetic resonance imaging (MRI) of a brain tumor has been incorporated into a robotic laser ablation neurosurgery system. 5-ALA is a non-fluorescent prodrug that leads to intracellular accumulation of fluorescent protoporphyrins IX (PpIX) in malignant glioma. The PpIX tends to accumulate in pathological lesions, and emits red fluorescence when excited by blue light. This fluorescence is illuminated with laser excitation, enables intra-operative identification of the position of a tumor and provides guidance for resection with laser photocoagulation. The information provided by the MRI is enhanced by the 5-ALA fluorescence data, and this enhanced information is integrated into a robotic laser ablation system. The accuracy of the fluorescent measurement of the tumor is improved using high-precision spectral analysis. The fluorescence assists in the detection of malignant brain tumors intraoperatively and improves their removal rate.\",\n \"Title: Generalization of geometrical flux maximizing flow on Riemannian manifolds for improved volumetric blood vessel segmentation.\\nYear: 2012\\n\\nAbstract: Geometric flux maximizing flow (FLUX) is an active contour based method which evolves an initial surface to maximize the flux of a vector field on the surface. For blood vessel segmentation, the vector field is defined as the vectors specified by vascular edge strengths and orientations. Hence, the segmentation performance depends on the quality of the detected edge vector field. In this paper, we propose a new method for level set based segmentation of blood vessels by generalizing the FLUX on a Riemannian manifold (R-FLUX). We consider a 3D scalar image I(x) as a manifold embedded in the 4D space (x, I(x)) and compute the image metric by pullback from the 4D space, whose metric tensor depends on the vessel enhancing diffusion (VED) tensor. This allows us to devise a non-linear filter which both projects and normalizes the original image gradient vectors under the inverse of local VED tensors. The filtered gradient vectors pertaining to the vessels are less sensitive to the local image contrast and more coherent with the local vessel orientation. The method has been applied to both synthetic and real TOF MRA data sets. Comparisons are made with the FLUX and vesselsness response based segmentations, indicating that the R-FLUX outperforms both methods in terms of leakage minimization and thiner vessel delineation.\",\n \"Title: Fusion of Laser Guidance and 3-D Autostereoscopic Image Overlay for Precision-Guided Surgery\\nYear: 2008\\n\\nAbstract: This paper describes a precision-guided surgical navigation system for minimally invasive surgery using fusion of laser guidance technique and three-dimensional (3-D) autostereoscopic image overlay technique. The images superimposed onto the patient are created by employing an animated autostereoscopic image called integral videography (IV), which display geometrically accurate 3-D autostereoscopic images and reproduce motion parallax without the need for special viewing or tracking devices. To improve the insertion accuracy of surgical instrument, we integrated the image overlay system with laser guidance for visualization of insertion point and orientation of the surgical instrument. We designed and manufactured a laser guidance device and mounted it to the IV image overlay device. Accuracy evaluations showed that the system could guide a linear surgical instrument toward a target with an average error of 2.48 mm and standard deviation of 1.76 mm. Improvement in the design of the laser guidance device and the patient-image registration of the IV image overlay will make this system practical and its use will increase surgical accuracy and reduce invasiveness.\",\n \"Title: Image-guidance for cardiac surgery using dynamic autostereoscopic display system\\nYear: 2004\\n\\nAbstract: An autostereoscopic display and analysis system has been developed to aid the development of three-dimensional (3-D) cardiac imaging for use during minimally invasive cardiac surgery. We have previously developed a dynamic autostereoscopic display device by integral videography autostereoscopic imaging method with high-resolution multi-projection technique. This paper integrated the high-quality autostereoscopic display system into image-guided cardiac surgery. The human heart was scanned in a sequence of volumetric images representing the motion of the human heart throughout the cardiac cycle, based on dynamic information of heart from CT scanning. Time-sequence 3-D image analysis can be particularly useful in cardiac applications. This approach will allow us to acquire the optimum method to produce 3-D cardiac image for planning and guidance of minimally invasive cardiac surgery.\",\n \"Title: R-PLUS: a Riemannian anisotropic edge detection scheme for vascular segmentation.\\nYear: 2008\\n\\nAbstract: In this paper, detection of edges in oriented fields is addressed. In some applications such as vessel segmentation because of the intrinsic orientation of the structures, edge detection is only demanded in a particular subspace. This is specially usefull when a curve evolution is chosen for segmentation since gradients in parallel to vessel orientation may stop the contour. An anisotropic edge detection scheme is generalized on a Riemannian manifold using the local structure tensor. The method is the generalization of the PLUS operator proposed for accurate curved edge detection. Examples are given and the comparison is made with the state-of-the-art flux maximizing flow which indicates that significant improvements in terms of leakage minimization and thiner vessel delineation is achievable using our methodology.\",\n \"Title: Non-metal Slice Image Overlay Display System Used Inside the Open Type MRI\\nYear: 2008\\n\\nAbstract: MRI is now utilized not only for diagnosis but also for intraoperative surgical treatment. In the MRI environment, the surgeon imagines the position of target region by observing pre-acquired images and patient body during the operation, therefore the spatial position of the diseased part depends on the surgeon's knowledge and experience. Thus information from the MRI image is not fully utilized. In order to solve this problem, we developed the prototypes of slice image overlay system for open-type MRI, which comprises the MRI compatible display device, image communication, registration system and visualization software, which is adopted to clinical 0.2T Open MRI. Phantom experiments were performed and the TRE of the overlay device was less than 1.0mm at the center of the display.\",\n \"Title: Ultrasonic Motor Driving Method For Emi-Free Image In Mr Image-Guided Surgical Robotic System\\nYear: 2007\\n\\nAbstract: Electromagnetic interference (EMI) between magnetic resonance (MR) imager and surgical manipulator is a severe problem, that degrades the image quality, in MR image-guided surgical robotic systems. We propose a novel motor driving method to acquire noise-free image. Noise generation accompanied by motor actuation is permitted only during the \\\"dead time\\\" when the MR imager stops signal acquisition to wait for relaxation of protons. For the synchronized control between MR imager and motor driving system, we adopted a radio-frequency pulse signal detected by a special antenna as a synchronous trigger. This method can be applied widely because it only senses a part of the scanning signal and requires neither hardware nor software changes to the MR imager. The evaluation results showed the feasibility of RF pulse as a synchronous trigger and the availability of sequence-based noise reduction method.\",\n \"Title: Fast and accurate ultrasonography for visceral fat measurement.\\nYear: 2010\\n\\nAbstract: Visceral fat area (VFA) has close relationship with hypertension, diabetes and cardiovascular disease, and therefore serve as a reliable indicator of these diseases. Abdominal computed tomography (CT) enables precise quantification of the VFA and has been considered as the gold standard for VFA assessment. In this paper, we develope a novel method to quickly and accurately measure the VFA with ultrasonography (US). We evaluated the novel method on five volunteers and the diagnosis procedures lasted less than 30 seconds averagely. The simulation results by our method were compared with VFA estimated by abdominal CT. The correlation coefficient between them was 0.913 for men and 0.858 for women. And the mean deviation of between VFA by CT and by our method was 19.8 cm2 for men and 13.3 cm2 for women.\",\n \"Title: 3-D augmented reality for MRI-guided surgery using integral videography autostereoscopic image overlay.\\nYear: 2010\\n\\nAbstract: A 3-D augmented reality navigation system using autostereoscopic images was developed for MRI-guided surgery. The 3-D images are created by employing an animated autostereoscopic image, integral videography (IV), which provides geometrically accurate 3-D spatial images and reproduces motion parallax without using any supplementary eyeglasses or tracking devices. The spatially projected 3-D images are superimposed onto the surgical area and viewed via a half-silvered mirror. A fast and accurate spatial image registration method was developed for intraoperative i.v. image-guided therapy. Preliminary experiments showed that the total system error in patient-to-image registration was 0.90 +/- 0.21 mm, and the procedure time for guiding a needle toward a target was shortened by 75%. An animal experiment was also conducted to evaluate the performance of the system. The feasibility studies showed that augmented reality of the image overlay system could increase the surgical instrument placement accuracy and reduce the procedure time as a result of intuitive 3-D viewing.\",\n \"Title: Automatic Focusing and Robotic Scanning Mechanism for Precision Laser Ablation in Neurosurgery\\nYear: 2010\\n\\nAbstract: We have developed an laser ablation system with an automatic focusing (AF) and robotic scanning mechanism for precision malignant gliomas resection in neurosurgery. A 5-aminolevulinic acid (5-ALA)-induced fluorescence based intra-operative tumor diagnosis technique has been incorporated into the robotic laser ablation system. The system enables an intra-operative identification of the position of a tumor with fluorescence illuminated by a laser excitation. The AF and robotic scanning mechanism assists in tracking the surface of the malignant brain tumors, and provides position information for both laser scanning in 5-ALA fluorescence identification and laser ablation in tumor resection. Experimental results showed the automatic focus device had a precision of 0.5 mm and the mechanism could track the brain surface with a movement caused by pulsation and respiration.\",\n \"Title: A robot assisted hip fracture reduction with a navigation system.\\nYear: 2008\\n\\nAbstract: A fracture reduction robot is described as assisting in safe and precise fracture reduction. The robot is connected with pins that are inserted into the patient's bone fragments, together with a customized jig. The robot has six degrees of freedom with high precision, so that precise fracture reduction can be conducted. The failsafe unit of the fracture reduction robot can mitigate excessive reduction force that may cause complications such as avascular necrosis. We have integrated the fracture reduction robot with a navigation system that tracks the relative position of the bone fragments and generates the reduction path. The integrated system is evaluated with the simulated fracture reduction of a hip fracture model (n = 8). Three-dimensional parameters related to the mechanical axis--the proximal femur angle, the distal femur angle, and the length of the mechanical axis--were evaluated by comparing the normal values with those after reduction; these average differences are 1.76 degrees , 0.28 degrees and 0.76mm, respectively. The automated fracture reduction feature makes it possible for medical staff to work at a distance from radiation sources; for patients, the integrated fracture reduction system has the potential to reduce fractures with high precision.\",\n \"Title: Surgical navigation of integral videography image overlay for open MRI-Guided glioma surgery\\nYear: 2006\\n\\nAbstract: A three-dimensional (3D) image overlay navigation system has been developed for open MRI guided surgery. The images are created by employing animated autostereoscopic image of integral videography (IV), which provides geometrically accurate 3D images and reproduce motion parallax without using any supplementary glasses or tracking devices. The spatially projected 3D images are superimposed onto the patient and are viewed via a half-slivered mirror. We developed a spatial image registration method for intraoperative IV image-guided diagnosis and therapy. Two sets of preliminary experiments showed that the total system error was 0.90 ± 0.21mm in patient-to-image registration and the procedure time of guiding a needle toward a target was shortened to 24%. The augmented reality of the image overlay system is possible to increase the surgical instrument placement accuracy and also to reduce the procedure time as the result of intuitive 3D viewing.\",\n \"Title: Fast Image Mapping of Endoscopic Image Mosaics with Three-Dimensional Ultrasound Image for Intrauterine Treatment of Twin-to-Twin Transfusion Syndrome\\nYear: 2008\\n\\nAbstract: This paper describes a fast image mapping system that integrates endoscopic image mosaics with three-dimensional (3-D) ultrasound images for assisting intrauterine treatment of twin-to-twin transfusion syndrome (TTTS) by laser photocoagulation. Endoscopic laser photocoagulation treatment has a good survival rate and a low complication rate for twins. However, the small field of view and lack of surrounding information makes the identification of vessels anastomosis difficult. We have developed an extended placenta visualization system with the fusion of endoscopic image mosaics with a 3-D ultrasound-image placenta model. Fully automatic and fast calibration is used for endoscope calibration in fluid. The 3-D spatial position of the endoscopic images and the ultrasound image are tracked by a 3-D position tracking device. The mosaiced endoscope images are registered to the surface of the 3-D ultrasound placenta model by using a fast GPU-based image rendering method. Experimental results show that the system may provide an improved and efficient way of planning and guidance in laser photocoagulation TTTS treatment.\",\n \"Title: Fetus support manipulator with flexible balloon-based stabilizer for endoscopic intrauterine surgery.\\nYear: 2006\\n\\nAbstract: A novel manipulator is described for stabilizing fetus and preventing it from free-floating during the endoscopic intrauterine surgery. Minimally invasive endoscopic fetal surgery enables intrauterine intervention with reduced risk to the mother and fetus. We designed and fabricated a prototype of a fetus supporting manipulator equipped with a flexible bending/curving mechanisms and a soft balloon-based stabilizer. The flexible bending and curving mechanisms enable the stabilizer to reach the target sites within the uterus under an ultrasound-guidance. The balloon-based stabilizer could be inserted into the uterus with a small incision for entry. The accuracy evaluation showed that the maximum error of the curving mechanism was as small as 7 mm and the standard deviation of the bending mechanism was just 1.6 degrees. In the experiments using a fetus model, the manipulator could be well controlled under ultrasound guidance and its curving mechanism with the balloon-based stabilizer could be clearly visualized during the implementation of fetus model supporting. The manipulator has the potential to be used in minimally invasive intrauterine surgery, though further improvements and experiments remain to be carried out.\",\n \"Title: A coaxial laser endoscope with arbitrary spots in endoscopic view for fetal surgery.\\nYear: 2009\\n\\nAbstract: In this paper, we describe a rigid endoscope that transmits a laser beam coaxially to arbitrary points in the endoscopic view, mainly for treatment of twin-to-twin transfusion syndrome. The endoscope consists of a hotmirror for coaxial transmission of visible light and a Nd:YAG laser beam, and galvanometers for controlling the beam irradiation angle. We evaluated the transmission efficiency of the laser power, the spot size through the endoscope and accuracy in positioning the beam. The maximum laser transmission efficiency was 39% and the spot diameter was 2.2-3.2 mm at a distance of 10-20 mm. The positioning accuracy was mostly within 1.0 mm in the endoscopic view at the distance. The average laser power density on the spot was estimated to be 170-370 W/cm2, and a chicken liver was successfully coagulated by changing the laser beam irradiation angle.\",\n \"Title: An Autostereoscopic Display System for Image-Guided Surgery Using High-Quality Integral Videography with High Performance Computing\\nYear: 2003\\n\\nAbstract: This paper presents a high-quality glass-less autostereoscopic display system that is integrated into image-guided surgery. The glass-less autostereoscopic image was created by employing a modification of Integral Videography (IV), which is an animated extension of Integral Photography. IV records and reproduces 3-D images using a micro convex lens array and plat display, which can display geometrically accurate 3-D autostereoscopic images and reproduce motion parallax without any need for special viewing devices. This paper reports the use of high-resolution IV display and high performance computing for producing medical 3-D image. We evaluate the feasibility of this display by using a set of medical image with segmentation and reproducing 3-D CT and MRI autostereoscopic image for surgical planning and intra-operative guidance. The main contribution of this paper is application and modification of medical 3-D display technique originally developed in high-resolution autostereoscopic display and high performance computing.\",\n \"Title: High performance computing for parallel rendering in surgical autostereoscopic display and navigation\\nYear: 2003\\n\\nAbstract: The three-dimensional medical information obtained from Magnetic Resonance Image (MRI), X-ray Computed Tomography (CT), etc. is used for an operation-supporting image, and operations under image guidance have also been conducted. One of the stereoscopic methods is Integral Videography (IV), which is an animated extension of Integral Photography, reproducing a computer-generated graphical object. Though the advantage of IV has been proven in both feasibility studies and clinical applications, one of the issues still unsolved is the notable quantity of calculation that causes significant delay in rendering. Then using parallel processing method integrating Message Passing Interface (MPI) on High Performance Computer (HPC), we shortened the calculating time of IV picture by the shortest at about 0.2 s. Furthermore, by using socket communication, it transmitted to another portable note PC, which is for a display. Then, we achieved the fast presentation of autostereoscopic images seen from an arbitrary direction that was specified with the mouse from the note PC side at anywhere we can access to the network.\",\n \"Title: High-sensitive fluorescence endoscope using electrocardiograph-synchronized multiple exposure.\\nYear: 2011\\n\\nAbstract: Fluorescence-based measurement of cardiac disease, using autofluorescent substances that already exist in the heart, has not been used for endoscopic surgery because the endoscopic lenses cannot transmit sufficient light. A highly sensitive fluorescence endoscope using an electrocardiograph (ECG)-synchronized multiple exposure (ESME) approach was developed that provides a bright fluorescent image.A system was developed consisting of an endoscope, an excitation light, an ECG amplifier, a trigger and delay unit, and a computer. This system is based on periodic motion of the heart. Since the shape of the heart can be photographed by ECG triggering in a similar manner, a bright image can be synthesized by accumulating multiple trigger-captured images. Laboratory and in vivo experiments were performed to confirm the effectiveness of ESME.The experimental results revealed that the trigger unit generated the synchronization signals required to produce high-quality images of the heart depending on heart rate. The difference among trigger-captured images from the actual organ, which affects the quality of ESME images, was estimated at 0.65 mm from the calculated displacement of a marker on the heart. The results also revealed that a bright fluorescent image can be captured by ESME.A highly sensitive fluorescence endoscope using ESME was developed and successfully tested. The experimental results indicated that the method enabled high-quality image acquisition in a very low illumination environment. This system is effective for the observation of faint fluorescence in the heart and is useful for the intraoperative examination of the heart status.\",\n \"Title: Rigid-flexible outer sheath model using slider linkage locking mechanism and air pressure for endoscopic surgery.\\nYear: 2006\\n\\nAbstract: The objective of this paper is to develop an outer sheath for flexible endoscopic manipulators. This sheath can switch two states including flexible and rigid, and make a rigid curved path for inserting manipulators. The flexible mode can be curved into a required shape. The rigid mode can hold the shape of the sheath, and then keep the path for instruments. Through the managed path, the flexible manipulators become easy to reach the target. We proposed a serial multi joint model to realize the flexible mechanism. This model is composed of a set of frame units which are connected serially. Each unit can be rotated to a given angle around the center of the joint. We developed a slider-link mechanism and a gear stopper controlled by air pressure for rigid mode. We designed and fabricated the prototype with a diameter of 16 mm and length of 290 mm. The experiment showed that the device could be switched from the flexible mode to the rigid mode when the air pressure was over 150 kPa, and each joint could hold its angle against the maximum 400 mNm. The phantom experiment showed that the flexible devices are possible to transmit the wire tension to the endpoint of the manipulator without changing the curving shape with by the developed outer sheath device.\",\n \"Title: Integral videography autostereoscopic display using multiprojection\\nYear: 2006\\n\\nAbstract: Various techniques of autostereoscopic three-dimensional (3D) display for displaying real 3D information have been investigated in recent years. We have performed research and application of integral videography (IV), which is an animated display technique based on the principle of integral photography (IP). Although IV and IP are ideal ways to display 3D images, their spatial viewing resolution and high-speed image rendering algorithm must be improved. The pixel pitch of the display and the lens pitch are the main factors affecting IV image quality. We improved the image quality by increasing the number and density of the pixels. Parallel image processing and displaying method were also developed for animated IV imaging. We have created a scalable high-resolution image using multiple projection and project it onto a small screen to achieve high-density pixels using long-focal-length projection optics. Furthermore, an image calibration method was developed for geometrical correction and color modulation to generate seamless IV images. We evaluated the developed device by displaying an X-ray computed tomography (CT) image. Measurement of the geometrical accuracy of the reproduced IV images demonstrated that the spatial resolution of the display system matched that by theoretical analysis. Experimental results showed that our technique is very effective in obtaining 3D animated images with practical spatial resolution. © 2006 Wiley Periodicals, Inc. Syst Comp Jpn, 37(5): 34–45, 2006; Published online in Wiley InterScience (). DOI 10.1002/scj.20424\",\n \"Title: Surgical Manipulator With Linkage Mechanism For Anterior Cruciate Ligament Reconstruction\\nYear: 2007\\n\\nAbstract: This paper describes a surgical manipulator designed with regard to its safety and sterilization in medical setting of anterior cruciate ligament (ACL) reconstruction. The manipulator system consists of two five-bar linkage mechanisms, an optical tracking device, and a user interface of surgical navigation. We designed and fabricated a prototype manipulator and evaluated the performance of the system by a set of experiments. The accuracy evaluation showed that the average backlash of the 4 link axes was 4.14 degrees and the accuracy of the pitch and yaw angle of the K-wire are 1.55 degrees +/- 2.74 degrees and -2.48 degrees +/- 2.92 degrees. This system enables intrinsic safety and ease of sterilization due to limitations in the range of movement. Furthermore, the system achieves reliabitity because of the simplicity of both its mechanism and its user interface. The manipulator has the potential to be used in minimally invasive ACL reconstruction surgery, although further improvements and experiments remain to be carried out.\",\n \"Title: Augmented reality system for oral surgery using 3D auto stereoscopic visualization.\\nYear: 2011\\n\\nAbstract: We present an augmented reality system for oral and maxillofacial surgery in this paper. Instead of being displayed on a separated screen, three-dimensional (3D) virtual presentations of osseous structures and soft tissues are projected onto the patient's body, providing surgeons with exact knowledge of depth information of high risk tissues inside the bone. We employ a 3D integral imaging technique which produce motion parallax in both horizontal and vertical direction over a wide viewing area in this study. In addition, surgeons are able to check the progress of the operation in real-time through an intuitive 3D based interface which is content-rich, hardware accelerated. These features prevent surgeons from penetrating into high risk areas and thus help improve the quality of the operation. Operational tasks such as hole drilling, screw fixation were performed using our system and showed an overall positional error of less than 1 mm. Feasibility of our system was also verified with a human volunteer experiment.\",\n \"Title: Integral Videography Overlay Navigation System Using Mutual Information-Based Registration\\nYear: 2004\\n\\nAbstract: We propose an automatic markerless registration method for Integral Videography (IV) overlay navigation system that directly matches a video image of the patient to a 2-D rendering of a surface model extracted from the 3-D surface model image. IV overlay system is integrated into a surgical navigation system to superimpose a real three-dimensional (3-D) auto stereoscopic image onto the patient via a half-silvered mirror. This registration technique doesn't require the setting of markers, which is invasive and extends the duration of the procedure. Accuracy measurements were performed using a triangular-prism-shaped model. The mean Target Registration Error (TRE) was 1.6 +/- 0.6 mm. We also applied our method to a human face registration and obtained successful results, although some noises were added to the video image. The presented method is more suitable for clinical applications of IV image overlay than previous methods since the presented method is sufficiently accurate, robust and more convenient for surgeon.\",\n \"Title: Three-dimensional auto-stereoscopic animated image with a long viewing distance using high-precision image correction\\nYear: 2009\\n\\nAbstract: Three-dimensional (3-D) displays have got a lot of attention because they have a much higher sense of realism and are more intuitive than 2-D displays. In particular, autostereoscopic displays are suitable for everyday use because they can be observed from an arbitrary viewpoint without supplementary glasses or tracking devices. Integral Videography (IV) is one of the methods for autostereoscopic animated images that extends Integral Photography (IP) to animation. IP/IV uses a combination of a lens array and a number of calculated elemental images with different perspectives. In particular, IV with a depth of several meters can be applied in many areas. However, most IV reports have an image depth of only several centimeters. Only a small deviation of a lens from its designed position would result in several degrees of deviation of the light ray from the back of the lens. We have developed the static autostereoscopic image by projecting the light sources from an object onto a photographic film through the lens-array [Liao et al. 2005]. In this study, we obtained animated IV of 1 m image depth with less distortion using our method to correct IV images. The method is technically unique.\",\n \"Title: Effective statistical edge integration using a flux maximizing scheme for volumetric vascular segmentation in MRA.\\nYear: 2007\\n\\nAbstract: Evolutionary schemes based on the level set theory are effective tools for medical image segmentation. In this paper, a new variational technique for edge integration is presented. Region statistical measures and orientation information from ramp-like edges, are fused within an energy minimization scheme that is based on a new interpretation of edge concept. A region driven advection term simulating the edge strength effect is directly obtained from this minimization strategy. We have applied our method to several real Magnetic Resonance Angiography data sets and comparison has been made with a state-of-the-art vessel segmentation method. Presented results indicate that using this method a significant improvement is achievable and the method can be an effective tool to extract vessels in MRA intracranial images.\",\n \"Title: Development and evaluation of a novel actuator using MR magnetic field\\nYear: 2007\\n\\nAbstract: Recently, Intra-Operative Magnetic Resonance Imaging (IO-MRI) has attracted attention for the reason of operating quality. In the other hand, surgical assist robot with high accuracy is widely researched. By combining IO-MRI and surgical assist robot to an MR-compatible surgical assist robot, higher quality of operation is expected. In this paper, we introduce a novel actuator for MR-compatible Surgical Assist Robot, which actively uses MR magnetic field and is drive by regulating pulse current. Use of ratchet mechanism enables step wise motion of the rotor similar to stepping motors. The position of the rotor can be controlled without an encoder. Clock-wise rotation, counter clock-wise rotation, constrained and non-constrained conditions were achieved by using this mechanism. Evaluation of the first prototype (84times76times58(mm3), 380 (gram)) shows that it is MR-compatible and has a very high positioning accuracy (step angle CW: 12.04plusmn0.28 (deg), CCW : 12.22plusmn0.36(deg); backlash < 1 (deg)). No-load maximum rotational speed is 35.71(rpm) and maximum torque 15.1times10-3 (Nm).\",\n \"Title: Hazard analysis of fracture-reduction robot and its application to safety design of fracture-reduction assisting robotic system\\nYear: 2010\\n\\nAbstract: In this paper, we discuss the issue of safety in robot-assisted fracture reduction. We define the hazards of robot-assisted fracture reduction and design safety control methods. Although a large reduction force is required to reduce femoral neck fractures, an unexpectedly large force produced by a robot may cause injury to the patient. We have designed two mechanical failsafe units and a software force limiter; this along with velocity control can guarantee a safe operation in reduction force. In addition, to reduce the movement of bone fragments as much as possible, we devised spatially constrained control methods for fracture-reduction robots. The fracture-reduction system was evaluated using simulated fracture reductions.\",\n \"Title: High Quality Autostereoscopic Surgical Display Using Anti-aliased Integral Videography Imaging\\nYear: 2004\\n\\nAbstract: This paper presents an autostereoscopic three-dimensional (3-D) surgical display with high quality integral videography (IV) rendering algorithm. IV is an animated extension of integral photography, which provides 3-D images without using any supplementary glasses or tracking devices. Despite IV's many advantages, the quality of its spatial image has thus far been poor. We developed a high quality image rendering method with oversampling technique for enhancing the resolution of elemental IV image and low-pass-filter for smoothing the image. Furthermore, we manufactured a high-resolution IV display for evaluating the feasibility of proposed method. The experimental results show the quality of anti-aliased IV image is improved. We also integrated the developed IV image into image-guided surgery display system. This approach will allow us to acquire the optimum process to produce high quality 3-D image for planning and guidance of minimally invasive surgery.\",\n \"Title: Ultrasound Image-Guided Mapping of Endoscopic Views on a 3D Placenta Model: A Tracker-Less Approach.\\nYear: 2013\\n\\nAbstract: This work presents a framework for mapping of free-hand endoscopic views onto 3D anatomical model constructed from ultrasound images without the use of external trackers. It is non-disruptive in terms of existing surgical workflow as surgeons do not need to accommodate operational constraints associated with the use of additionally installed motion sensors or tracking systems. A passive fiducial marker is attached to the tip of the endoscope to create a geometric eccentricity that encodes the position and orientation of the camera. The relative position between the endoscope and the anatomical model under the ultrasound image reference frame is used to establish a texture map that overlays endoscopic views onto the surface of the model. This addresses operational challenges including the limited field-of-view (FOV) and the lack of 3D perspective associated with minimally invasive procedure. Experimental results show that average tool position and orientation errors are 1.32 mm and 1.6 degrees respectively. The R.M.S. error of the overall image mapping obtained based on comparison of dimension of landmarks is 3.30 mm with standard deviation of 2.14 mm. The feasibility of the framework is also demonstrated through implementations on a phantom model.\",\n \"Title: Anisotropic Haralick Edge Detection Scheme with Application to Vessel Segmentation\\nYear: 2008\\n\\nAbstract: In this paper, detection of edges in oriented fields is addressed. Haralick edge detection is an accurate scheme for estimation of the edge in a Euclidean space. However, in some applications such as edge detection for vessel segmentation because of the intrinsic orientation of structures, accuracy is only demanded in a particular subspace. This is specially usefull when a curve evolution is chosen for segmentation since gradients in parallel to vessel orientation stops evolution. Haralick edge detection is generalized on a Riemannian space using the inner product of the vectors under a space metric tensor. This eliminates the spurious gradients and preserves the accuracy on the vessel border. Examples are given and the comparison is made with the state-of-the-art flux maximizing flow indicating that significant improvements in terms of leakage minimization and thiner vessel delineation is achievable using our methodology.\",\n \"Title: Surgical navigation by autostereoscopic image overlay of integral videography.\\nYear: 2004\\n\\nAbstract: This paper describes an autostereoscopic image overlay technique that is integrated into a surgical navigation system to superimpose a real three-dimensional (3-D) image onto the patient via a half-silvered mirror. The images are created by employing a modified version of integral videography (IV), which is an animated extension of integral photography. IV records and reproduces 3-D images using a microconvex lens array and flat display; it can display geometrically accurate 3-D autostereoscopic images and reproduce motion parallax without the need for special devices. The use of semitransparent display devices makes it appear that the 3-D image is inside the patient's body. This is the first report of applying an autostereoscopic display with an image overlay system in surgical navigation. Experiments demonstrated that the fast IV rendering technique and patient-image registration method produce an average registration accuracy of 1.13 mm. Experiments using a target in phantom agar showed that the system can guide a needle toward a target with an average error of 2.6 mm. Improvement in the quality of the IV display will make this system practical and its use will increase surgical accuracy and reduce invasiveness.\",\n \"Title: Real-time computer-generated integral imaging and 3D image calibration for augmented reality surgical navigation.\\nYear: 2015\\n\\nAbstract: Autostereoscopic 3D image overlay for augmented reality (AR) based surgical navigation has been studied and reported many times. For the purpose of surgical overlay, the 3D image is expected to have the same geometric shape as the original organ, and can be transformed to a specified location for image overlay. However, how to generate a 3D image with high geometric fidelity and quantitative evaluation of 3D image's geometric accuracy have not been addressed. This paper proposes a graphics processing unit (GPU) based computer-generated integral imaging pipeline for real-time autostereoscopic 3D display, and an automatic closed-loop 3D image calibration paradigm for displaying undistorted 3D images. Based on the proposed methods, a novel AR device for 3D image surgical overlay is presented, which mainly consists of a 3D display, an AR window, a stereo camera for 3D measurement, and a workstation for information processing. The evaluation on the 3D image rendering performance with 2560×1600 elemental image resolution shows the rendering speeds of 50-60 frames per second (fps) for surface models, and 5-8 fps for large medical volumes. The evaluation of the undistorted 3D image after the calibration yields sub-millimeter geometric accuracy. A phantom experiment simulating oral and maxillofacial surgery was also performed to evaluate the proposed AR overlay device in terms of the image registration accuracy, 3D image overlay accuracy, and the visual effects of the overlay. The experimental results show satisfactory image registration and image overlay accuracy, and confirm the system usability.\",\n \"Title: Micro laser ablation system integrated with image sensor for minimally invasive surgery\\nYear: 2014\\n\\nAbstract: A novel micro surgical system with a micro laser ablation module integrated with an imaging sensor is designed and developed for minimally invasive surgery. The system consists of a Nd:YAG laser source with the wavelength of 1064nm to remove lesion, a CCD camera to guide laser, a parallel four-bar mechanism, an endoscope bending section, a rigid steel tube, an aluminum-alloy supporting table and a personal computer for sending control commands. The advantage of the system is the small size of its distal module, i.e., the diameter and the length of the module are 3.5 millimeters and 15.0 millimeters, respectively. A methodological frame for ablating lesion using the system proposed is developed based on the system configuration. Experiments on ex-vivo swine tissue are implemented to evaluate laser ablation performance of the prototyped system.\",\n \"Title: Floating autostereoscopic 3D display with multidimensional images for telesurgical visualization.\\nYear: 2016\\n\\nAbstract: PurposeWe propose a combined floating autostereoscopic three-dimensional (3D) display approach for telesurgical visualization, which could reproduce live surgical scene in a realistic and intuitive manner.\",\n \"Title: Vision-based markerless registration using stereo vision and an augmented reality surgical navigation system: a pilot study.\\nYear: 2015\\n\\nAbstract: This study evaluated the use of an augmented reality navigation system that provides a markerless registration system using stereo vision in oral and maxillofacial surgery.A feasibility study was performed on a subject, wherein a stereo camera was used for tracking and markerless registration. The computed tomography data obtained from the volunteer was used to create an integral videography image and a 3-dimensional rapid prototype model of the jaw. The overlay of the subject's anatomic site and its 3D-IV image were displayed in real space using a 3D-AR display. Extraction of characteristic points and teeth matching were done using parallax images from two stereo cameras for patient-image registration.Accurate registration of the volunteer's anatomy with IV stereoscopic images via image matching was done using the fully automated markerless system, which recognized the incisal edges of the teeth and captured information pertaining to their position with an average target registration error of < 1 mm. These 3D-CT images were then displayed in real space with high accuracy using AR. Even when the viewing position was changed, the 3D images could be observed as if they were floating in real space without using special glasses.Teeth were successfully used for registration via 3D image (contour) matching. This system, without using references or fiducial markers, displayed 3D-CT images in real space with high accuracy. The system provided real-time markerless registration and 3D image matching via stereo vision, which, combined with AR, could have significant clinical applications.\",\n \"Title: 3D Surgical Overlay with Markerless Image Registration Using a Single Camera.\\nYear: 2015\\n\\nAbstract: Minimum invasive surgery can benefit from surgical visualization, which is achieved by either virtual reality or augmented reality. We previously proposed an integrated 3D image overlay based surgical visualization solution including 3D image rendering, distortion correction, and spatial projection. For correct spatial projection of the 3D image, image registration is necessary. In this paper we present a 3D image overlay based augmented reality surgical navigation system with markerless image registration using a single camera. The innovation compared with our previous work lies in the single camera based image registration method for 3D image overlay. The 3D mesh model of patient's teeth which is created from the preoperative CT data is matched with the intraoperative image captured by a single optical camera to determine the six-degree-of-freedom pose of the model with respect to the camera. The obtained pose is used to superimpose the 3D image of critical hidden tissues on patient's body directly via a translucent mirror for surgical visualization. The image registration performs automatically within approximate 0.2 s, which enables real-time update to tackle patient's movement. Experimental results show that the registration accuracy is about 1mm and confirm the feasibility of the 3D surgical overlay system.\",\n \"Title: Registration of CT and Ultrasound Images of the Spine with Neural Network and Orientation Code Mutual Information.\\nYear: 2016\\n\\nAbstract: Pairwise registration of 2D ultrasound (US) and 3D computed tomography (CT) images can improve the efficiency and safety of image-guided anesthesia in spine surgery. However, accurate 2D US and 3D CT registration for multiple vertebras without an appropriate initial registration position is still a challenge, due to the difference of image modalities and missing bone structures in US image. This paper proposes a novel 2D US and 3D CT registration method, in which convolutional neural network (CNN) classification of US images is reported for the first time to achieve rough image registration. And a new orientation code mutual information metric is further applied to finish local registration refinement. By combining automatic rough registration with fine registration refinement, our algorithm achieves 2D US and 3D CT registration for multiple vertebras (L2-L4) without the requirement of an appropriate initial alignment. The accuracy of our algorithm is validated on 50 in vivo clinical US images dataset of multiple vertebras. And a mean target registration error of 2.3 mm is acquired, which is lower than the clinically acceptable accuracy 3.5 mm.\",\n \"Title: Radiation-Free 3D Navigation and Vascular Reconstruction for Aortic Stent Graft Deployment.\\nYear: 2016\\n\\nAbstract: We propose a radiation-free three dimensional (3D) navigation and vasculature reconstruction method to reduce aortic stent graft deployment time and repeated exposures to high doses of X-ray radiation. 2D intraoperative US images are fused with 3D preoperative magnetic resonance (MR) image to provide intuitive 3D navigation for deployment of the stent-graft in the proposed system. On the other hand, calibrated 2D US images track catheter’s tip and construct intraoperative 3D aortic model by combining segmented intravascular ultrasound (IVUS) images. The constructed 3D aortic models can quantitatively assess morphological characteristics of aortic to assist deployment of stent graft. This system was validated by using in vitro cardiac and aorta phantom. The mean target registration error of 2D US-3D MR registration was 2.70 mm and the average tracking error of position of IVUS catheter’s tip was 1.12 mm. Accurate contours detection results in IVUS images were acquired. Meanwhile, Hausdorff distances are 0.78 mm and 0.59 mm for outer and inner contours, respectively; Dice coefficients are 90.21 % and 89.96 %. Experiment results demonstrate that our radiation-free navigation and 3D vasculature reconstruction method is promising for deployment of stent graft in vivo studies.\",\n \"Title: Automatic Fast-Registration Surgical Navigation System Using Depth Camera and Integral Videography 3D Image Overlay.\\nYear: 2016\\n\\nAbstract: We propose an automatic fast-registration augmented reality (AR) surgical navigation system for minimally invasive surgery. The system integrates a fast-registration technique with three-dimensional (3D) integral videography (IV) image overlay. The detailed anatomic information generated by IV technique is superimposed to the patient using 3D autostereoscopic images, which reproduce motion parallax with naked eyes. To reduce the patient-3D overlay image registration time and achieve automatic execution, we integrate a 3D image overlay system with a real-time patient tracking system, which utilizes particle filter algorithm and depth camera. Experimental results showed that the system can lower the registration time and the can reach up to 10 frames per second (fps). The 3D overlay image to the patient registration average error is 1.88 mm, with standard deviation of 0.72 mm. Further work for improvement of the depth camera acquisition accuracy and tracking algorithm makes this system more feasible and practical.\",\n \"Title: 3D Catheter Shape Determination for Endovascular Navigation Using a Two-Step Particle Filter and Ultrasound Scanning\\nYear: 2017\\n\\nAbstract: In endovascular catheter interventions, the determination of the three-dimensional (3D) catheter shape can increase navigation information and help reduce trauma. This study describes a shape determination method for a flexible interventional catheter using ultrasound scanning and a two-step particle filter without X-ray fluoroscopy. First, we propose a multi-feature, multi-template particle filte...\",\n \"Title: Augmented reality surgical navigation with ultrasound-assisted registration for pedicle screw placement: a pilot study.\\nYear: 2017\\n\\nAbstract: We present a novel augmented reality (AR) surgical navigation system based on ultrasound-assisted registration for pedicle screw placement. This system provides the clinically desired targeting accuracy and reduces radiation exposure.Ultrasound (US) is used to perform registration between preoperative computed tomography (CT) images and patient, and the registration is performed by least-squares fitting of these two three-dimensional (3D) point sets of anatomical landmarks taken from US and CT images. An integral videography overlay device is calibrated to accurately display naked-eye 3D images for surgical navigation. We use a 3.0-mm Kirschner wire (K-wire) instead of a pedicle screw in this study, and the K-wire is calibrated to obtain its orientation and tip location. Based on the above registration and calibration, naked-eye 3D images of the planning path and the spine are superimposed onto patient in situ using our AR navigation system. Simultaneously, a 3D image of the K-wire is overlaid accurately on the real one to guide the insertion procedure. The targeting accuracy is evaluated postoperatively by performing a CT scan.An agar phantom experiment was performed. Eight K-wires were inserted successfully after US-assisted registration, and the mean targeting error and angle error were 3.35 mm and [Formula: see text], respectively. Furthermore, an additional sheep cadaver experiment was performed. Four K-wires were inserted successfully. The mean targeting error was 3.79 mm and the mean angle error was [Formula: see text], and US-assisted registration yielded better targeting results than skin markers-based registration (targeting errors: 2.41 vs. 5.18 mm, angle errors: [Formula: see text] vs. [Formula: see text].Experimental outcomes demonstrate that the proposed navigation system has acceptable targeting accuracy. In particular, the proposed navigation method reduces repeated radiation exposure to the patient and surgeons. Therefore, it has promising prospects for clinical use.\",\n \"Title: 3D interactive surgical visualization system using mobile spatial information acquisition and autostereoscopic display.\\nYear: 2017\\n\\nAbstract: Work flow of 3D interactive surgical visualization system using mobile spatial information acquisition and autostereoscopic display.\",\n \"Title: High-Quality See-Through Surgical Guidance System Using Enhanced 3-D Autostereoscopic Augmented Reality.\\nYear: 2017\\n\\nAbstract: Objective: Precise minimally invasive surgery (MIS) has significant advantages over traditional open surgery in clinic. Although pre-/intraoperative diagnosis images can provide necessary guidance for therapy, hand-eye discoordination occurs when guidance information is displayed away from the surgical area. In this study, we introduce a real three-dimensional (3-D) see-through guidance system for...\",\n \"Title: Real-time Lens based Rendering Algorithm for Super-multiview Integral Photography without Image Resampling.\\nYear: 2018\\n\\nAbstract: We propose a computer generated integral photography (CGIP) method that employs a lens based rendering (LBR) algorithm for super-multiview displays to achieve higher frame rates and better image quality without pixel resampling or view interpolation. The algorithm can utilize both fixed and programmable graphics pipelines to accelerate CGIP rendering and inter-perspective antialiasing. Two hardwar...\",\n \"Title: Visible Forceps Manipulator With Novel Linkage Bending Mechanism For Neurosurgery\\nYear: 2017\\n\\nAbstract: In minimally invasive surgery (MIS), especially the neurosurgery, surgeons often suffer from occlusion region problem. Common surgical instruments, like endoscope and corresponding operating tools, are hard to solve it due to their size and rigid mechanical structure. In this paper, we present a visible forceps manipulator with novel linkage bending mechanism, which realizes the flexible bending capability and high output force, as well as the integrated endoscopic function. We present the simplified experiment to evaluate the results of mechanical performance and brain phantom test to evaluate feasibility and usefulness in neurosurgery. Preliminary results show that phantom experiments using the brain phantom verify the feasibility of the novel manipulator.\",\n \"Title: Spatial Position Measurement System for Surgical Navigation Using 3-D Image Marker-Based Tracking Tools With Compact Volume.\\nYear: 2018\\n\\nAbstract: We develop a spatial position measurement system using three-dimensional (3-D) image marker-based tracking tools targeted at surgical navigation in minimally invasive surgery. We generate 3-D image markers with spatial information encoded to 2-D images, design tracking tools with the 3-D image markers, and analyze the tracking tools' theoretical spatial errors, which are primarily limited by the s...\",\n \"Title: A novel integration of spectral-domain optical-coherence-tomography and laser-ablation system for precision treatment.\\nYear: 2018\\n\\nAbstract: We propose an optical integrated diagnosis and therapeutic system. The primary experimental results show the high efficiency and feasibility of our theranostics system, which is promising for realizing accurate resection of tumors in vivo and in situ in the future.\",\n \"Title: Lumen and media-adventitia border detection in IVUS images using texture enhanced deformable model.\\nYear: 2018\\n\\nAbstract: •Propose a texture enhanced deformable model (TEDM) for accurate detection of lumen and media–adventitia (MA) borders in intravascular ultrasound (IVUS) images.•TEDM has scientific importance related to the quantitative plaque information and 3D vessel visualization.•The accuracy of the TEDM is validated by using 1500 IVUS images from 15 in vivo patient sequences.•TEDM can detect lumen and MA surfaces precisely with small surface distance errors of 0.17 mm and 0.19 mm, respectively.\",\n \"Title: Optimized Optical Coherence Tomography Imaging With Hough Transform-Based Fixed-Pattern Noise Reduction.\\nYear: 2018\\n\\nAbstract: Fixed-pattern noise seriously affects the clinical application of optical coherence tomography (OCT), especially, in the imaging of tumorous tissue. We propose a Plough transform-based fixed-pattern noise reduction (HTFPNR) method to reduce the fixed-pattern noise for optimizing imaging of tumorous tissue with OCT system. Using by the HTFPNR method, we detect and map the outline of fixed-pattern noise in the OCT images, and finally efficiently reduce the fixed-pattern noise by the longitudinal and horizontal intelligent processing procedure. We adopt the image-to-noise ratio with full information (INRfi) and the noise reduction ratio (NRR) to evaluate the outcome of fixed-pattern noise reduction ratio, respectively. The INRfi of OCT image's noise reduction of ex vivo brainstem tumor is approximate 21.92 dB. Six groups of OCT images, including three types of fixed-pattern noises, have been validated via experimental evaluation of the ex vivo gastric tumor. In the different types of fixed-pattern noise, the mean INR(fi)s are 25.24, 23.04, and 19.35 dB, respectively. This result demonstrates that it is highly efficient and useful in fixed-pattern noise reduction. The fluctuating range of the NRR is 0.84-0.88 for three types of added noise in the OCT images. This result demonstrates that the HTFPNR method can as possible as save useful information by comparing to previous research. This proposed HTFPNR method can be used into the fixed-pattern noise reduction of OCT images in other soft biological tissue in the future.\",\n \"Title: Clustering of Morphological Features for Identifying Femur Cavity Subtypes With Difficulties of Intramedullary Nail Implantation.\\nYear: 2018\\n\\nAbstract: Intramedullary (IM) nail implantation is currently the standard treatment for femoral intertrochanteric fractures. However, individual differences in femur cavity bring a challenge in designing well-matched IM nails and cause difficulties in IM nail implantation. Therefore, there is an intense need to analyze femur cavities to predict difficulties in IM nail implantation to assist the design of IM...\",\n \"Title: A Cascaded Deep Convolutional Neural Network for Joint Segmentation and Genotype Prediction of Brainstem Gliomas.\\nYear: 2018\\n\\nAbstract: Goal: Automatic segmentation of brainstem gliomas and prediction of genotype (H3 K27M) mutation status based on magnetic resonance (MR) images are crucial but challenging tasks for computer-aided diagnosis in neurosurgery. In this paper, we present a novel cascaded deep convolutional neural network (CNN) to address these two challenging tasks simultaneously. Methods: Our novel segmentation task co...\",\n \"Title: Reconstruction and Registration of Large-Scale Medical Scene Using Point Clouds Data from Different Modalities.\\nYear: 2018\\n\\nAbstract: Sensing the medical scenario can ensure the safety during the surgical operations. So, in this regard, a monitor platform which can obtain the accurate location information of the surgery room is desperately needed. Compared to 2D camera image, 3D data contains more information of distance and direction. Therefore, 3D sensors are more suitable to be used in surgical scene monitoring. However, each 3D sensor has its own limitations. For example, Lidar (Light Detection and Ranging) can detect large-scale environment with high precision, but the point clouds or depth maps are very sparse. As for commodity RGBD sensors, such as Kinect, can accurately capture denser data, but limited to a small range from 0.5 to 4.5m. So, a proper method which can address these problems for fusing different modalities data is important. In this paper, we proposed a method which can fuse different modalities 3D data to get a large-scale and dense point cloud. The key contributions of our work are as follows. First, we proposed a 3D data collecting system to reconstruct the medical scenes. By fusing the Lidar and Kinect data, a large-scale medical scene with more details can be reconstructed. Second, we proposed a location-based fast point clouds registration algorithm to deal with different modality datasets.\",\n \"Title: An Intelligent Theranostics Method Using Optical Coherence Tomography Guided Automatic Laser Ablation for Neurosurgery.\\nYear: 2018\\n\\nAbstract: Methods to reduce the high disability and fatality rate of neurosurgery caused by surgeon tremor and uncertainty of intraoperative tumor boundary have long been greatly concerned. We have proposed an intelligent theranostics method for a compact integrated diagnosis and therapeutic platform to automatically excise the brain tumor with high precision. To perform this operation, hardware combination of a benchtop spectral domain optical coherence tomography (SD-OCT) and a high-power laser has been implemented by a customized fiber combiner. By the double-clad fiber out of the combiner which works as both the OCT sample arm fiber and the ablation laser output fiber, an in situ theranostics process without registration can be conducted. The SD-OCT image analysis from both en face and depth perspectives shows the structural information intuitively and accordingly generate the ablation-guiding grid. Based on the grid, the continuous wave high-power laser ablates ex vivo porcine brain models in submillimeter automatically. The feasibility of this theranostics method has been verified by the comparison of attenuation maps before and after ablation, and the contrast of its pathology.\",\n \"Title: Augmented reality surgical navigation with accurate CBCT-patient registration for dental implant placement.\\nYear: 2019\\n\\nAbstract: It is challenging to achieve high implant accuracy in dental implant placement, because high risk tissues need to be avoided. In this study, we present an augmented reality (AR) surgical navigation with an accurate cone beam computed tomography (CBCT)-patient registration method to provide clinically desired dental implant accuracy. A registration device is used for registration between preoperative data and patient outside the patient's mouth. After registration, the registration device is worn on the patient's teeth for tracking the patient. Naked-eye 3D images of the planning path and the mandibular nerve are superimposed onto the patient in situ to form an AR scene. Simultaneously, a 3D image of the drill is overlaid accurately on the real one to guide the implant procedure. Finally, implant accuracy is evaluated postoperatively. A model experiment was performed by an experienced dentist. Totally, ten parallel pins were inserted into five 3D-printed mandible models guided by our AR navigation method and through the dentist's experience, respectively. AR-guided dental implant placement showed better results than the dentist's experience (mean target error = 1.25 mm vs. 1.63 mm; mean angle error = 4.03° vs. 6.10°). Experimental results indicate that the proposed method is expected to be applied in the clinic. Graphical abstract ᅟ.\",\n \"Title: Unified Mathematical Model for Multilayer-Multiframe Compressive Light Field Displays Using LCDs.\\nYear: 2019\\n\\nAbstract: We propose a unified mathematical model for multilayer-multiframe compressive light field displays that supports both attenuation-based and polarization-based architectures. We show that the light field decomposition of such a display can be cast as a bound constrained nonlinear matrix optimization problem. Efficient light field decomposition algorithms are developed using the limited-memory BFGS (L-BFGS) method for automultiscopic displays with high resolution and high image fidelity. In addition, this framework is the first to support multilayer polarization-based compressive light field displays with time multiplexing. This new architecture significantly reduces artifacts compared with attenuation-based multilayer-multiframe displays; thus, it can allow the requirements regarding the number of layers or the refresh rate to be relaxed. We verify the proposed methods by constructing two 3-layer prototypes using high-speed LCDs, one based on the attenuation architecture and one based on the polarization architecture. Moreover, an efficient CUDA-based program is implemented. Our displays can produce images with higher spatial resolution with thinner form factors compared with traditional automultiscopic displays in both simulations and experiments.\",\n \"Title: Three-Dimensional Feature-Enhanced Network for Automatic Femur Segmentation.\\nYear: 2019\\n\\nAbstract: Automatic femur segmentation from computed tomography volume is a crucial but challenging task for computer-aided diagnosis in orthopedic surgeries. The main obstacles are weak bone boundaries, narrowness of joint space, variations in femur density and shape, as well as diverse leg postures. In this paper, we presented a novel 3-D feature-enhanced network to address these challenges. The novelty o...\",\n \"Title: Automatic distance measurement of abdominal aorta for ultrasonography-based visceral fat estimation.\\nYear: 2013\\n\\nAbstract: Ultrasonography-based visceral fat estimation is a promising method to assess central obesity, which is associated with metabolic syndrome. The key to this method is to measure three types of distance in the ultrasound image. The most important one is the distance from the skin surface to the posterior wall of the abdominal aorta. We present a novel automatic measurement method to calculate this distance using 1D ultrasound signal processing. It is different from the conventional 2D image processing based methods which have high failure rate when the target is blurred or partially imaged. The proposed method identifies the waveforms of the aorta along a group of ultrasound scan lines and a rating mechanism is introduced to choose the best waveform for distance calculation. The robustness and accuracy of the method were evaluated by experiments based on clinical data.\",\n \"Title: Design and Modularization of Multi-DoF Soft Robotic Actuators\\nYear: 2019\\n\\nAbstract: Soft robotic technologies offer a promising solution for unconstructed applications due to their complete softness, robustness, and flexibility. In this letter, we explore the design and modularization of the flexible pneumatic robotic actuators for applying in minimally invasive surgery or other constrained spaces. We first identify the basic multi-cavity structure and subsequently explore three ...\",\n \"Title: Augmented reality-based autostereoscopic surgical visualization system for telesurgery\\nYear: 2021\\n\\nAbstract: Purpose The visualization of remote surgical scenes is the key to realizing the remote operation of surgical robots. However, current non-endoscopic surgical robot systems lack an effective visualization tool to offer sufficient surgical scene information and depth perception. Methods We propose a novel autostereoscopic surgical visualization system integrating 3D intraoperative scene reconstruction, autostereoscopic 3D display, and augmented reality-based image fusion. The preoperative organ structure and the intraoperative surface point cloud are obtained from medical imaging and the RGB-D camera, respectively, and aligned by an automatic marker-free intraoperative registration algorithm. After registration, preoperative meshes with precalculated illumination and intraoperative textured point cloud are blended in real time. Finally, the fused image is shown on a 3D autostereoscopic display device to achieve depth perception. Results A prototype of the autostereoscopic surgical visualization system was built. The system had a horizontal image resolution of 1.31 mm, a vertical image resolution of 0.82 mm, an average rendering rate of 33.1 FPS, an average registration rate of 20.5 FPS, and average registration errors of approximately 3 mm. A telesurgical robot prototype based on 3D autostereoscopic display was built. The quantitative evaluation experiments showed that our system achieved similar operational accuracy (1.79 +/- 0.87 mm) as the conventional system (1.95 +/- 0.71 mm), while having advantages in terms of completion time (with 34.11% reduction) and path length (with 35.87% reduction). Post-experimental questionnaires indicated that the system was user-friendly for novices and experts. Conclusion We propose a 3D surgical visualization system with augmented instruction and depth perception for telesurgery. The qualitative and quantitative evaluation results illustrate the accuracy and efficiency of the proposed system. Therefore, it shows great prospects in robotic surgery and telesurgery.\",\n \"Title: Intelligent optical diagnosis and treatment system for automated image-guided laser ablation of tumors\\nYear: 2021\\n\\nAbstract: Purpose For tumor resections near critical structures, accurate identification of tumor boundaries and maximum removal are the keys to improve surgical outcome and patient survival rate, especially in neurosurgery. In this paper, we propose an intelligent optical diagnosis and treatment system for tumor removal, with automated lesion localization and laser ablation. Methods The proposed system contains a laser ablation module, an optical coherence tomography (OCT) unit, and a robotic arm along with a stereo camera. The robotic arm can move the OCT sample arm and the laser ablation front-end to the suspected lesion area. The corresponding diagnosis and treatment procedures include computer-aided lesion segmentation using OCT, automated ablation planning, and laser control. The ablation process is controlled by a deflectable mirror, and a non-common-path ablation planning algorithm based on the transformation from lesion positions to mirror deflection angles is presented. Results Phantom and animal experiments are carried out for system verification. The robot could reach the planned position with high precision, which is approximately 1.16 mm. Tissue classification with OCT images achieves 91.7% accuracy. The error of OCT-guided automated laser ablation is approximately 0.74 mm. Experiments on mouse brain tumors show that the proposed system is capable of clearing lesions efficiently and precisely. We also conducted an ex vivo porcine brain experiment to verify the whole process of the system. Conclusion An intelligent optical diagnosis and treatment system is proposed for tumor removal. Experimental results show that the proposed system and method are promising for precise and intelligent theranostics. Compared to conventional cancer diagnosis and treatment, the proposed system allows for automated operations monitored in real-time, with higher precision and efficiency.\",\n \"Title: Force-guided autonomous robotic ultrasound scanning control method for soft uncertain environment\\nYear: 2021\\n\\nAbstract: Purpose Autonomous ultrasound imaging by robotic ultrasound scanning systems in complex soft uncertain clinical environments is important and challenging to assist in therapy. To cope with the complex environment faced by the ultrasound probe during the scanning process, we propose an autonomous robotic ultrasound (US) control method based on reinforcement learning (RL) model to build the relationship between the environment and the system. The proposed method requires only contact force as input information to achieve robot control of the posture and contact force of the probe without any a priori information about the target and the environment. Methods First, an RL agent is proposed and trained by a policy gradient theorem-based RL model with the 6-degree-of-freedom (DOF) contact force of the US probe to learn the relationship between contact force and output force directly. Then, a force control strategy based on the admittance controller is proposed for synchronous force, orientation and position control by defining the desired contact force as the action space. Results The proposed method was evaluated via collected US images, contact force and scan trajectories by scanning an unknown soft phantom. The experimental results indicated that the proposed method differs from the free-hand scanned approach in the US images within 3 +/- 0.4%. The analysis results of contact forces and trajectories indicated that our method could make stable scanning processes on a soft uncertain skin surface and obtained US images. Conclusion We propose a concise and efficient force-guided US robot scanning control method for soft uncertain environment based on reinforcement learning. Experimental results validated our method's feasibility and validity for complex skin surface scanning, and the volunteer experiments indicated the potential application value in the complex clinical environment of robotic US imaging system especially with limited visual information.\",\n \"Title: H3K27M Mutations Prediction for Brainstem Gliomas Based on Diffusion Radiomics Learning\\nYear: 2021\\n\\nAbstract: H3K27M mutation is the most common mutation in brainstem gliomas (BSGs), which is related with highly invasive neoplasms and poor prognosis. Accurate presurgical and noninvasive prediction of H3K27M mutations based on preoperative multi-modal neuroimaging is of great clinical value in the diagnosis, prognosis and therapeutic selection of BSGs. Traditional BSG radiomics models usually only focus on tumor local morphometric characteristics. However, given that highly invasive BSGs may significantly affect large-scale brain network connectivity, we reasonably infer that local radiomics and global connectomics may provide different perspectives for H3K27M genotype prediction. Therefore, we define a graph-based diffusion radiomics learning model to integrate these two kinds of features seamlessly. Specifically, edges of the defined brain network are determined by neural fiber connections, while node features of brainstem are governed by local tumor radiomics. Upon this model, we further propose a multi-mechanism diffusion convolutional network to couple multi-modal information and generate a joint representation for brain disease diagnosis. By graph diffusion convolution, the local radiomics information spread along the brain network structure to enhance graph representation learning, and eventually the learned diffusion radiomics features contribute to disease prediction. Experiments on a real BSG dataset demonstrate the effectiveness and advantages of our proposed method for preoperative prediction of H3K27M statuses.\",\n \"Title: 3D Volume Visualization and Screen-based Interaction with Dynamic Ray Casting on Autostereoscopic Display\\nYear: 2021\\n\\nAbstract: Augmented reality (AR) is an emerging technology to improve visualization experiences. However, visualizing volume data is limited in existing AR systems due to the lack of an intuitive and precise exploration scheme. In this paper, we present a 3D augmented volume visualization and screen-based interaction method. An autostereoscopic handheld display is utilized to achieve naked-eye 3D perception...\",\n \"Title: An Unsupervised Convolution Neural Network for Deformable Registration of Mono/Multi-Modality Medical Images\\nYear: 2021\\n\\nAbstract: Image registration is a fundamental and crucial step in medical image analysis. However, due to the differences between mono-mode and multi-mode registration tasks and the complexity of the corresponding relationship between multi-mode image intensity, the existing unsupervised methods based on deep learning can hardly achieve the two registration tasks simultaneously. In this paper, we proposed a novel approach to register both mono- and multi-mode images in a same framework. By approximately calculating the mutual information in a differentiable form and combining it with CNN, the deformation field can be predicted quickly and accurately without any prior information about the image intensity relationship. The registration process is implemented in an unsupervised manner, avoiding the need for the ground truth of the deformation field. We utilize two public datasets to evaluate the performance of the algorithm for mono-mode and multi-mode image registration, which confirms the effectiveness and feasibility of our method. In addition, the experiments on patient data also demonstrate the practicability and robustness of the proposed method.\",\n \"Title: Image-Based 3D Ultrasound Reconstruction with Optical Flow via Pyramid Warping Network\\nYear: 2021\\n\\nAbstract: 3D Ultrasound (US) contains rich spatial information which is helpful for medical diagnosis. However, current reconstruction methods with tracking devices are not suitable for clinical application. The sensorless freehand methods reconstruct based on US images which is less accuracy. In this paper, we proposed a network which reconstructs the US volume based on US images features and optical flow features. We proposed the pyramid warping layer which merges the image features and optical flow features with warping operation. To fuse the warped features of different scales in different pyramid levels, we adopted the fusion module using the attention mechanism. Meanwhile, we adopted the channel attention and spatial attention to our network. Our method was evaluated in 100 freehand US sweeps of human forearms which exhibits the efficient performance on volume reconstruction compared with other methods.\",\n \"Title: A Smoke Removal Method Based on Combined Data and Modified U-Net for Endoscopic Images\\nYear: 2021\\n\\nAbstract: In minimally invasive surgery, the ablation of human tissue will produce a lot of smoke, which will interfere with the surgeon's operation. We propose a smoke removal method based on combined data and modified U-net for endoscopic images. The real dataset and the synthetic dataset are built using a small amount of images with smoke. The real dataset is combined with the synthetic dataset successively. Qualitative evaluation shows that the quality of the output smoke-free image is the best when training using the combined data, compared to using only either the real dataset or the synthetic dataset above. Quantitative evaluation shows that the effect of smoke removal is still the best when training using the combined data in our method.\",\n \"Title: TypeSeg: A type-aware encoder-decoder network for multi-type ultrasound images co-segmentation\\nYear: 2022\\n\\nAbstract: Purpose: As a portable and radiation-free imaging modality, ultrasound can be easily used to image various types of tissue structures. It is important to develop a method which supports the multi-type ultrasound images co-segmentation. However, state-of-the-art ultrasound segmentation methods commonly only focus on the single type images or ignore the type-aware information. Methods: To solve the above problem, this work proposes a novel type-aware encoder-decoder network (TypeSeg) for the multi-type ultrasound images co-segmentation. First, we develop a type-aware metric learning module to find an optimum latent feature space where the ultrasound images of the same types are close and that of the different types are separated by a certain margin. Second, depending on the extracted features, a decision module decides whether the input ultrasound images have the common tissue type or not, and the encoder-decoder network produces a segmentation mask accordingly. Results: We evaluate the performance of the proposed TypeSeg model on the ultrasound dataset that contains four types of tissues. The proposed TypeSeg model achieves the overall best results with the mean IOU score of 87.51% +/- 3.93% for the multi-type ultrasound images. Conclusion: The experimental results indicate that the proposed method outperforms all the compared state-of-the-art algorithms for the multi-type ultrasound images co-segmentation task. (C) 2021 Elsevier B.V. All rights reserved.\",\n \"Title: Ultrasound Speckle Reduction Using Wavelet-Based Generative Adversarial Network\\nYear: 2022\\n\\nAbstract: The visual quality of ultrasound (US) images is crucial for clinical diagnosis and treatment. The main source of image quality degradation is the inherent speckle noise generated during US image acquisition. Current deep learning-based methods cannot preserve the maximum boundary contrast when removing noise and speckle. In this paper, we address the issue by proposing a novel wavelet-based generative adversarial network (GAN) for real-time high-quality US image reconstruction, viz. WGAN-DUS. First, we propose a batch normalization module (BNM) to balance the importance of each sub-band image and fuse sub-band features simultaneously. Then, a wavelet reconstruction module (WRM) integrated with a cascade of wavelet residual channel attention block (WRCAB) is proposed to extract distinctive sub-band features used to reconstruct denoised images. A gradual tuning strategy is proposed to fine-tune our generator for better despeckling performance. We further propose a wavelet-based discriminator and a comprehensive loss function to effectively suppress speckle noise and preserve the image features. Besides, we have designed an algorithm to estimate the noise levels during despeckling of real US images. The performance of our network was then evaluated on natural, synthetic, simulated and clinical US images and compared against various despeckling methods. To verify the feasibility of WGAN-DUS, we further extend our work to uterine fibroid segmentation with the denoised US image of the proposed approach. Experimental result demonstrates that our proposed method is feasible and can be generalized to clinical applications for despeckling of US images in real-time without losing its fine details.\",\n \"Title: Rib Suppression in Digital Chest Tomosynthesis\\nYear: 2022\\n\\nAbstract: Digital chest tomosynthesis (DCT) is a technique to produce sectional 3D images of a human chest for pulmonary disease screening, with 2D X-ray projections taken within an extremely limited range of angles. However, under the limited angle scenario, DCT contains strong artifacts caused by the presence of ribs, jamming the imaging quality of the lung area. Recently, great progress has been achieved for rib suppression in a single X-ray image, to reveal a clearer lung texture. We firstly extend the rib suppression problem to the 3D case at the software level. We propose a Tomosynthesis RIb SuPpression and Lung Enhancement Network (TRIPLE-Net) to model the 3D rib component and provide a rib-free DCT. TRIPLE-Net takes the advantages from both 2D and 3D domains, which models the ribs in DCT with the exact FBP procedure and 3D depth information, respectively. The experiments on simulated datasets and clinical data have shown the effectiveness of TRIPLE-Net to preserve lung details as well as improve the imaging quality of pulmonary diseases. Finally, an expert user study confirms our findings. Our code is available at https://github.com/sunyhl/Rib-Suppressionin-Digital-Chest-Tomosynthesis.\",\n \"Title: Statistical and individual characteristics-based reconstruction for craniomaxillofacial surgery\\nYear: 2022\\n\\nAbstract: In craniomaxillofacial (CMF) surgery planning, a preoperative reconstruction of the CMF reference model is crucial for surgical restoration, especially the reconstruction of bilateral defects. Current reconstruction algorithms mainly generate reference models from the image analysis aspect, however, clinical indicators of the CMF reference model mostly consider the distribution of anatomical landmarks. Generating a reference model with optimal clinical evaluation helps promote the feasibility of an algorithm. We first build a dataset with 100 normal skull models and then calculate a statistical shape model (SSM) and the distribution of normal cephalometric values, which indicate the statistical features of a population. To further generate personalized reference models, we apply non-rigid registration to align the SSM with the defect skull model. An evaluation standard to select the optimal reference model considers both global performance and anatomical evaluation. Moreover, we develop a landmark detection network to improve the automatic level of the algorithm. The proposed method performs better than methods including Iterative Closest Point and SSM. From a global evaluation aspect, the results show that the RMSE between the reference model and the ground truth is \\n \\n \\n \\n $$1.71 \\\\pm 0.23$$\\n \\n mm, the percentage of vertices with error below 2 mm is \\n \\n \\n \\n $$85 \\\\pm 4$$\\n \\n % and the average faces distance is \\n \\n \\n \\n $$1.38 \\\\pm 0.20$$\\n \\n mm (better than the state-of-the-art method). From the anatomical evaluation aspect, the target registration error between the landmark pairs is \\n \\n \\n \\n $$4.12 \\\\pm 2.27$$\\n \\n mm. In addition, the clinical application confirms that the reference model can meet clinical requirements. To the best of our knowledge, we propose the first CMF reconstruction method considering the global performance of reconstruction and anatomically local evaluation from clinical experience. Simulated experiments and clinical cases prove the general applicability and strength of the method.\",\n \"Title: Efficient lower-limb segmentation for large-scale volumetric CT by using projection view and voxel group attention\\nYear: 2022\\n\\nAbstract: 3D segmentation of lower-limb bones such as tibia and fibula from volumetric computed tomography (CT) is very important for surgical planning and navigation. However, the input data of the large-scale 3D volumetric CT will increase the computation cost, and meantime, the spatial connections of the long-range voxels will be easily ignored. To solve these problems, this paper proposes an accurate and efficient 3D bone segmentation approach based on the distribution characteristics of the lower-limb bones in volumetric CT. On one hand, the efficient segmentation framework was built for the large-scale volumetric CT using 2D projection view-based slice filtering and parameter-reduced separable convolution. On the other hand, we developed a new voxel group attention mechanism to emphasize the connection of the long-range voxel groups and improve the representational capability of the segmentation network. The experimental results showed that the proposed 3D bone segmentation approach achieved high segmentation accuracy under the conditions of limited computations. It was additionally shown that the proposed approach outperformed the state-of-the-art 3D models for the segmentation of lower-limb bones.\",\n \"Title: Spatial Position Estimation Method for 3D Ultrasound Reconstruction Based on Hybrid Transfomers\\nYear: 2022\\n\\nAbstract: 3D ultrasound reconstruction has important application value in clinical diagnosis. To obtain the spatial relationships between ultrasound images in reconstruction tasks, especially the contextual information in long-range sequences, this work proposes a spatial position estimation method for 3D ultrasound image reconstruction based on a hybrid transformer encoder. Inspired by the transformer sequ...\",\n \"Title: Spatiotemporal reconstruction method of carotid artery ultrasound from freehand sonography\\nYear: 2022\\n\\nAbstract: 4D reconstruction based on radiation-free ultrasound can provide valuable information about the anatomy. Current 4D US technologies are either faced with limited field-of-view (FoV), technical complications, or cumbersome setups. This paper proposes a spatiotemporal US reconstruction framework to enhance its ability to provide dynamic structure information. We propose a spatiotemporal US reconstruction framework based on freehand sonography. First, a collecting strategy is presented to acquire 2D US images in multiple spatial and temporal positions. A morphology-based phase extraction method after pose correction is presented to decouple the compounding image variations. For temporal alignment and reconstruction, a robust kernel regression model is established to reconstruct images in arbitrary phases. Finally, the spatiotemporal reconstruction is demonstrated in the form of 4D movies by integrating the US images according to the tracked poses and estimated phases. Quantitative and qualitative experiments were conducted on the carotid US to validate the feasibility of the proposed pipeline. The mean phase localization and heart rate estimation errors were 0.07 ± 0.04 s and 0.83 ± 3.35 bpm, respectively, compared with cardiac gating signals. The assessment of reconstruction quality showed a low RMSE (<0.06) between consecutive images. Quantitative comparisons of anatomy reconstruction from the generated US volumes and MRI showed an average surface distance of 0.39 ± 0.09 mm on the common carotid artery and 0.53 ± 0.05 mm with a landmark localization error of 0.60 ± 0.18 mm on carotid bifurcation. A novel spatiotemporal US reconstruction framework based on freehand sonography is proposed that preserves the utility nature of conventional freehand US. Evaluations on in vivo datasets indicated that our framework could achieve acceptable reconstruction performance and show potential application value in the US examination of dynamic anatomy.\",\n \"Title: Augmented reality navigation with ultrasound-assisted point cloud registration for percutaneous ablation of liver tumors\\nYear: 2022\\n\\nAbstract: We present a novel augmented reality (AR) surgical navigation method with ultrasound-assisted point cloud registration for percutaneous ablation of liver tumors. A preliminary study is carried out to verify its feasibility. Two three-dimensional (3D) point clouds of the liver surface are derived from the preoperative images and intraoperative tracked US images, respectively. To compensate for the soft tissue deformation, the point cloud registration between the preoperative images and the liver is performed using the non-rigid iterative closest point (ICP) algorithm. A 3D AR device based on integral videography technology is designed to accurately display naked-eye 3D images for surgical navigation. Based on the above registration, naked-eye 3D images of the liver surface, planning path, entry points, and tumor can be overlaid in situ through our 3D AR device. Finally, the AR-guided targeting accuracy is evaluated through entry point positioning. Experiments on both the liver phantom and in vitro pork liver were conducted. Several entry points on the liver surface were used to evaluate the targeting accuracy. The preliminary validation on the liver phantom showed average entry-point errors (EPEs) of 2.34 ± 0.45 mm, 2.25 ± 0.72 mm, 2.71 ± 0.82 mm, and 2.50 ± 1.11 mm at distinct US point cloud coverage rates of 100%, 75%, 50%, and 25%, respectively. The average EPEs of the deformed pork liver were 4.49 ± 1.88 mm and 5.02 ± 2.03 mm at the coverage rates of 100% and 75%, and the average covered-entry-point errors (CEPEs) were 4.96 ± 2.05 mm and 2.97 ± 1.37 mm at 50% and 25%, respectively. Experimental outcomes demonstrate that the proposed AR navigation method based on US-assisted point cloud registration has achieved an acceptable targeting accuracy on the liver surface even in the case of liver deformation.\",\n \"Title: Anatomically constrained deformable 3D reconstruction of intraoperative uterus from preoperative MRI data on uterine fibroid treatment\\nYear: 2022\\n\\nAbstract: Detections of the uterus using preoperative magnetic resonance imaging (MRI) data are required for intraoperative navigation in High-Intensity Focused Ultrasound (HIFU) treatment of uterine fibroids. The modelling of the intraoperative uterus between empty and full bladder anatomy deems to be a crucial step to quantify uterus deformation caused by the bladder-and-rectum-filling (BRF) technique. In this paper, a novel two-stage conditional generative adversarial network (cGAN) is proposed to perform intraoperative uterus deformable reconstruction with anatomical constraints using only single preoperative MRI data, viz. ACDeformRec. The highly constrained anatomy properties are further captured by a novel attention network that formulates high-level uterus segmentation task by incorporating correlations of surrounding organs for more anatomically plausible reconstruction. The experimental result demonstrates the robustness of ACDeformRec by evaluating its performance on 181 clinical datasets and has achieved the lowest reconstruction error of 0.735 +/- 0.045 mm with Dice Similarity Coefficient of 94.23% and Normalised Cross Correlation of 97.23%.\",\n \"Title: Annotation-guided encoder-decoder network for bone extraction in ultrasound-assisted orthopedic surgery\\nYear: 2022\\n\\nAbstract: The patients and surgeons are usually exposed in massive ionizing radiation during fluoroscopy-based navigation orthopedic surgery. Comparatively, ultrasound-assisted orthopedic surgery could not only decrease the risk of radiation but also provide rich navigation information. However, due to the artifacts in ultrasound images, the extraction of bone structure from ultrasound sequences can be a particularly difficult task, which leads to some major challenges in ultrasound-assisted orthopedic navigation. In this paper, we propose an annotation-guided encoder-decoder network (AGN) to extract bone structure from the radiation-free ultrasound sequences. Specifically, the variability of the ultrasound probe's pose leads to the change of the ultrasound frame during the acquisition of ultrasound sequences. Therefore, a feature alignment module deployed in the AGN model is used to achieve reliable matching across ultrasound frames. Moreover, inspired by the interactive ultrasound analysis, where user annotated foreground information can help target extraction, our AGN model incorporates the annotation information obtained by Siamese networks. Experimental results validated that the AGN model not only produced better bone surface extraction than state-of-the-art methods (IOU: 0.92 versus. 0.88), but also achieved almost real-time extraction with the speed about 15 frames per second. In addition, the acquired bone surface further provided radiation-free 3D intraoperative bone structure for the intuitive navigation of orthopedic surgery.\",\n \"Title: Deformable Torso Anatomy Education with Three-Dimensional Autostereoscopic Visualization and Free-Hand Interaction\\nYear: 2022\\n\\nAbstract: Virtual reality/augmented reality has advantages in immersive learning and conveying meta-information in anatomy education. In this study, we present an interactive naked-eye 3D torso anatomy education environment based on population anatomical information. In particular, we utilize the deformable anatomy models, constructed from big data of healthy adults, to convey the anatomy knowledge of truthfulness organ shapes and population anatomical variances. In addition, the proposed anatomy education system is conveyed by free-hand 3D autostereoscopic visualization, which supports multi-user and glass-free. A user study with fourteen students was implemented and demonstrates that the system is appropriate and useful for torso anatomy education.\",\n \"Title: A LESS SUPERVISED AUTOMATIC DELINEATION METHOD FOR INTRACRANIAL GERM CELL TUMOR RADIOTHERAPY TARGETS\\nYear: 2022\\n\\nAbstract: Central nervous system germ cell tumor is one of the most common intracranial malignant tumors in children. Currently, radiotherapy is the main treatment method. Local tumor and whole ventricular system irradiation can ensure the curative effect while reducing radiation damage in the late stage of radiotherapy. However, it is time-consuming and laborious for doctors to manually delineate the ventricular system. Therefore, this research proposed a less supervised automatic delineation method for intracranial germ cell tumor radiotherapy targets. By integrating the idea of unsupervised image registration, it has fast segmentation speed and avoids the dependence of supervised learning methods on extensive labeled data. The average target delineation accuracy can reach 80.5% and the average time is 153.42s when only one set of annotated data is required. The proposed algorithm can assist doctors in radiotherapy target delineation and has the potential to be extended to actual clinical applications.\",\n \"Title: Uncertainty-aware Cascade Network for Ultrasound Image Segmentation with Ambiguous Boundary\\nYear: 2022\\n\\nAbstract: Ultrasound image segmentation plays an essential role in automatic disease diagnosis. However, to achieve precise ultrasound segmentation is still a challenge caused by the ambiguous lesion boundary and imaging artifacts such as speckles and shadowing noise. Considering that the pixels with high uncertainty generally distributing in the boundary regions of prediction maps, are likely to overlap with the confused regions of ultrasound, we proposed an uncertainty-aware cascade network. Our network uses the confidence map to evaluate the uncertainty of each pixel to enhance the segmentation of ambiguous boundary. On the one hand, the confidence map fuses with the ultrasound features and predicted mask using the adaptive fusion module (AFM) which enriches the context features from different modalities. In addition, the uncertainty attention module (UAM) is proposed based on the confidence map. This module focuses on the influential features with cross attention constrained by the uncertainty of pixels which can extract the localized features of confused ultrasound regions. On the other hand, the recurrent edge correction module (RECM) further improves the segmentation of ambiguous boundary. This module increases the weights of confident features neighboring the uncertainty boundaries in order to refine the predictions of edge pixels with low confidence. We evaluated the proposed method on three public ultrasound datasets and the segmentation results show that our method achieved higher Dice scores and lower Hausdorff distance with more precise boundary details compared with state-of-the-art methods.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":105,"string":"105"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13567,"string":"13,567"}}},{"rowIdx":13568,"cells":{"question_id":{"kind":"string","value":"Ignas G Niemegeers"},"question":{"kind":"string","value":"Generate an abstract for the title \"Ad Hoc Federation of Networks (FedNets) - Mechanisms and Requirements\" using the following items: FedNets, ad hoc wireless networks, user-centric, context aware, ambient assisted networks."},"answer":{"kind":"string","value":"We have been witnessing developments of many innovative applications in the field of ad hoc wireless networks as well as many projects in the area of user-centric, context aware and ambient assisted networks. This paper introduces a novel framework called FedNets that bridges user centric networks to form an ad hoc federation to achieve group cooperation. Our preliminary study of all the mechanisms, requirements and potential techniques that need to be addressed in this regard is presented. We describe a roadmap towards a fully networked cooperative interaction. We believe that this paper will serve as the basis for our next stages of development, as well as for other similar works in the field of wireless ad hoc networks."},"correct_answer":{"kind":"string","value":"We have been witnessing developments of many innovative applications in the field of ad hoc wireless networks as well as many projects in the area of user-centric, context aware and ambient assisted networks. This paper introduces a novel framework called FedNets that bridges user centric networks to form an ad hoc federation to achieve group cooperation. Our preliminary study of all the mechanisms, requirements and potential techniques that need to be addressed in this regard is presented. We describe a roadmap towards a fully networked cooperative interaction. We believe that this paper will serve as the basis for our next stages of development, as well as for other similar works in the field of wireless ad hoc networks."},"documents":{"kind":"list like","value":["Title: Performance Study Of Mimo-Ofdm Platform In Narrow-Band Sub-1 Ghz Wireless Lans\nYear: 2013\n\nAbstract: Using multiple antennas at the transceivers has become a necessity in high data rate wireless communication systems. Multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) has been adopted as a mandatory modulation scheme in the upcoming IEEE 802.11ah standard. IEEE 802.11ah will specify carrier frequencies for Wireless Local Area Networks (WLANs) operating in the sub-1 GHz band. This is challenging for MIMO-OFDM in sub-1 GHz because of limited bandwidth. Thus building a prototype will provide necessary information to understand this new scenario. We build a prototype in order to test and validate modifications to physical layer (PHY) and media access control (MAC) for narrow-band data transmissions. We present here, in detail, the steps to build a real-time MIMO-OFDM testing platform that is useful for evaluating narrow-band sub-1GHz transmission characteristics. We analyzed potential MIMO-OFDM implementations and conducted extensive measurements on our platform in order to verify the required system enhancements. This paper gives a hands-on account of designing and testing a sub-1 GHz WLAN platform.","Title: A Survey on Handoffs - Lessons for 60 GHz Based Wireless Systems.\nYear: 2012\n\nAbstract: Wireless communication is playing an important role in our daily life since it offers flexibility and mobility. New multimedia services demand data-rates of up to hundreds of Mbps and thus higher frequency bands are being explored to support these new high data rate services. However, to support mobility, handoff is a must in many of these networks and systems. In the mean time the 60 GHz band has...","Title: Network Resource Awareness and Control in Mobile Applications\nYear: 2007\n\nAbstract: As mobile devices become increasingly capable of simultaneously using various kinds of wireless and fixed networks, developers must match mobile application needs to network resource dynamics. The authors' network resource model and network abstraction layer (NAL) software offer an extensible network resource abstraction to applications running on mobile devices. The NAL service can be used in conjunction with well-known APIs, such as the Socket API, which provide access to these resources; it also accommodates support for mobility-management facilities, such as Mobile IP, if available.","Title: A Performance Modeling and Evaluation of the Cambridge Fast Ring\nYear: 1992\n\nAbstract: Performance of the Cambridge Fast Ring (CFR), a high-speed slotted ring with normal slots, is studied. It is shown that the CFR can be represented by a multiqueue multiple cyclic server model with a 1-limited service discipline and with a restriction that only one server at a time can be serving a queue. Exact necessary and sufficient stability conditions are stated. An approximate analytic M/G/1 vacation model in which analysis concentrates on one station while the others are represented by a vacation period is developed to estimate the expected message waiting times. It is shown that the model is accurate and usable over a wide range of parameters. A performance evaluation of the CFR based on this model is presented. The performance is compared to that of a variant which does not restrict the number of slots a station may simultaneously use.","Title: Experience-based network resource usage on mobile hosts\nYear: 2006\n\nAbstract: The network resources available on a mobile device may be better employed when taking into account the history of mobility patterns of the user. We propose an experience-based mechanism for network resource usage by applications on mobile hosts. It incorporates a machine learning component that adapts the network resource usage given a high-level metric, typically consisting of system as well as application related sub-metrics.","Title: Solving The Incertitude Of Vertical Handovers In Heterogeneous Mobile Wireless Network Using Mdp\nYear: 2008\n\nAbstract: In this paper we introduce and examine via simulations, a Markov Decision Process (MDP) based approach for vertical handover decision-making. We consider here the case of a heterogeneous network environment where both the legacy WLAN and the novel line-of-sight (LOS)-dependent 60GHz radio systems are deployed to support indoor high quality multi-media applications. The MDP decision approach takes into account multiple factors such as user preference, network situation, device capability and effects of environment. We show its ability to effectively make handover decision in uncertain situations. The method and results herein are applicable, in general, to any other situation where such a decision is to be made.","Title: Smart Network Interface Selection For E-Dtns\nYear: 2012\n\nAbstract: We implement two energy models that accurately and comprehensively estimates the system energy cost and communication energy cost for using Bluetooth and Wi-Fi interfaces. The energy models running on a system is used to smartly pick the most energy optimal network interface so that data transfer between two end points is maximized.","Title: Link quality-based transmission power adaptation for energy saving in IEEE 802.11\nYear: 2009\n\nAbstract: Today, many wireless devices are mobile and battery powered, which means that the service duration is highly dependent on energy consumptions and battery capacities. Based on the fact that battery capacity is still limited, energy saving is an important issue in wireless communication. Many methods have been proposed to save energy in wireless communication, such as power-aware routing and sleep mode. However, another important approach, transmission power adaptation, is not yet well discussed. In this paper, we propose a mechanism that selects the appropriate transmission power level to transmit packets instead of using a fixed default power level for all packets, which is common in IEEE 802.11. Lower transmission power levels may result in more retransmissions, but still consumes less energy. In total, energy can be saved and interference can be reduced by selecting a good transmission power level. We propose a two phase power saving mechanism and implemented it in a real test-bed. To evaluate the performance of our mechanism, we used an indoor environment to carry out measurements. The measurement results indicate that a significant amount of energy can be saved for the transmitter in most scenarios compared to using a fixed transmission power level for all packets.","Title: A System Perspective on Cognition for Autonomic Computing and Communication\nYear: 2005\n\nAbstract: In this paper we present a conceptual view on the incorporation of cognitive processing capabilities in future generation computer systems. We argue that cognition is at the heart of autonomic behavior, and therefore a necessary ingredient for autonomic computing and communication. We introduce a bio- inspired cognitive engine that interacts with and has control over major operating system components, and showcase, based on scenario descriptions, how communicating applications take advantage of this setup by adapting and autonomously reacting to changes in heterogeneous and widely varying network resources. As a conceptual work, this paper does not cover experiences with the implementation of the introduced concepts, nor does it describe experimental results.","Title: Event-Driven MAC Protocol for Dual-Radio Cooperation.\nYear: 2012\n\nAbstract: One of the sources of the energy waste in wireless sensor networks is idle listening, the time in which a node monitors the free channel. In applications where the events occur sporadically, energy consumption due to idle listening can be further reduced by dual-radio cooperation. In dual-radio cooperation, nodes in the network have two stacks. One stack makes use of a low-power wake-up radio for event-driven communication over the main radio. The other stack may employ any sensor networking medium access control protocol only over the main radio. One of the two stacks can be dynamically operational depending on the rate of the events or the packet arrival rate. When the event rate becomes small, the event-driven stack takes over the operation. If the event rate increases, it could be more efficient to operate the legacy single radio stack. In this paper, we investigate the performance of the dual-radio cooperation in wireless sensor networks. A medium access control protocol is proposed for the dual-radio cooperation to maximize the energy efficiency of the wireless network. We define the critical event rate as the event rate threshold above which the single-radio stack performs better than the dual-radio stack. We analyzed and validated the critical event rate by simulations. We show that 70-97% energy conservation is possible by employing the dual-radio cooperation.","Title: Energy efficiency and channel allocation in P2PWRAN\nYear: 2013\n\nAbstract: IEEE 802.22 is proposed for reusing the TV channels with cognitive radio technology to build Wireless Regional Area Networks (WRANs). The cellular topology of a WRAN makes the spectrum management easy. However, this topology reduces the network capacity significantly, because in one slot only one user can be allocated with a channel. Therefore, Peer-toPeer WRAN (P2PWRAN) is proposed, in which direct CPE(Consumer Premises Equipment)-to-CPE intra-cell communication is supported. While increasing the network capacity significantly, reducing the total energy consumption is a challenge.The energy efficiency of nodes in the standard for WRAN is difficult to improve, since the transmission distance of nodes to the BS is relatively static. P2PWRAN enables to achieve higher energy efficiency through better possibilities of channel allocation. Channel allocation influences energy consumption and network capacity significantly. Channel allocation also affects the upper layers, for example the routing protocol. Therefore, guaranteeing the network performance and fairness amongst CPEs, while achieving green networks, needs cross-layer design. In this paper, we define the channel allocation problem in P2PWRAN and we recognize it as a Multi-objective Quadratic Programming (MOQP) problem. An Energy-aware Channel Allocation (ECA) algorithm is proposed for exploiting the P2PWRAN paradigm to achieve the lower communication energy consumption and higher network energy efficiency and throughput than simple greedy policies. ECA jointly considers the network layer and Media Access Control (MAC) layer. By controlling the parameters in ECA, the fairness of channel allocation can also be guaranteed.","Title: Evaluation of MPEG-4 Video Streaming over UMTS/WCDMA Dedicated Channels\nYear: 2005\n\nAbstract: Universal Mobile Telecommunications System (UMTS) is athird-generation mobile communication system that supportswireless wideband multimedia communications. This paperinvestigates the performance of streaming MPEG-4 videoover UMTS dedicated channels with varying channel conditions.In order to evaluate the performance, simulation models for UMTSand MPEG-4 were developed. Simulation results show that UMTSRadio Control Link (RLC) protocol operating in the acknowledgedmode outperforms the unacknowledged mode. The latter modeprovides timely delivery but no error recovery. The acknowledgedmode can deliver acceptable video quality for RLC block error ratesup to 40% utilizing a pre-decoder buffer. We also discover thatRLC block size can affect the video quality with a larger RLC block size attaining better quality. However, the optimum RLC blocksize is dependent on the encoded video frame size and the channelbit rate. Therefore, the RLC block size can be tuned to a proper value given the encoded video frame size and channel bit rate.","Title: Predicting mobility events on personal devices.\nYear: 2010\n\nAbstract: High-end mobile phones are quickly becoming versatile sensing platforms, capable of continuously capturing the dynamic context of their owners through various sensors. A change in this context is often caused by the fact that owners-and therefore the devices they carry-are moving from one place to another. In this paper, we model the sensed environment as a stream of events, and assume, given that people are creatures of habit, that time correlations exists between successive events. We propose a method for the prediction in time of the next occurrence of an event of interest, such as 'arriving at a certain location' or 'meeting with another person', with a focus on the prediction of network visibility events as observed through the wireless network interfaces of the device. Our approach is based on using other events in the stream as predictors for the event we are interested in, and, in the case of multiple predictors, applying different strategies for the selection of the best predictor. Using two real-world data sets, we found that including predictors of infrequently occurring events results in better predictions using the best selection strategy. Also, we found that cross-sensor (cross-interface) information in most cases improves the prediction performance.","Title: Fairness in Wireless Networks:Issues, Measures and Challenges\nYear: 2014\n\nAbstract: The pervasiveness of wireless technology has indeed created massive opportunity to integrate almost everything into the Internet fabric. This can be seen with the advent of Internet of Things and Cyber Physical Systems, which involves cooperation of massive number of intelligent devices to provide intelligent services. Fairness amongst these devices is an important issue that can be analysed from several dimensions, e.g., energy usage, achieving required quality of services, spectrum sharing, and so on. This article focusses on these viewpoints while looking at fairness research. To generalize, mainly wireless networks are considered. First, we present a general view of fairness studies, and pose three core questions that help us delineate the nuances in defining fairness. Then, the existing fairness models are summarized and compared. We also look into the major fairness research domains in wireless networks such as fair energy consumption control, power control, topology control, link and flow scheduling, channel assignment, rate allocation, congestion control and routing protocols. We make a distinction amongst fairness, utility and resource allocation to begin with. Later, we present their inter-relation. At the end of this article, we list the common properties of fairness and give an example of fairness management. Several open research challenges that point to further work on fairness in wireless networks are also discussed. Indeed, the research on fairness is entangled with many other aspects such as performance, utility, optimization and throughput at the network and node levels. While consolidating the contributions in the literature, this article tries to explain the niceties of all these aspects in the domain of wireless networking.","Title: A Link Stability Model for Indoor 60GHz Radio Wireless Networks\nYear: 2009\n\nAbstract: With the multitude of mobile communication devices supporting users, it is natural to expect that these devices would work ubiquitously anywhere and any time with high data rate support. 60 GHz is an obvious choice for the high data rate indoor communication. In this paper we present an holistic analytical model of 60 GHz link stability for indoor wireless networks. The model is built considering the indoor channel characteristics of 60 GHz radio, the antenna configuration and the mobility of persons. Also, we take an example to show the relevance of our model and its applicability in numerical evaluation.","Title: Impact of Antenna Pattern and Link Model on Directional Neighbor Discovery in 60 GHz Networks\nYear: 2011\n\nAbstract: 60 GHz radio is a very attractive technology for short-range wireless communication due to its capability to provide Gbps data rate. To address the neighbor discovery (ND) problem in 60 GHz networks, we propose a novel analytical framework to investigate the ND performance. The main difficulty in modeling the ND process in a 60 GHz network is the involvement of the directional antennas with gain differences between the antenna's main lobe and side lobes. Different antenna modes - directional or omni-directional - coupled with different ND mechanisms make the analytical study demanding. In this article, we propose a comprehensive theoretical model to demonstrate the performance of ND processes using one-way ND and handshake-based ND mechanisms. Moreover, we combine them with different antenna modes, i.e., directional transmitting with omni-directional listening (DO) and directional transmitting with directional listening (DD) modes. The impact of antenna modes on the ND process is analyzed. Since 60 GHz radio is prone to co-channel interference, we examine a realistic interference-aware link model and antenna pattern via simulation studies. Our work is specifically beneficial to provide guidelines for applying directional neighbor discovery process within 60 GHz wireless networks.","Title: Performance analysis of the link layer protocol for UWB impulse radio networks\nYear: 2006\n\nAbstract: The Self-organizing Device discovery and Data transmission protocol (SDD) is a novel self-organizing link layer protocol, that was proposed for UWB impulse radio based networks. By means of simulations, it has been indicated that the SDD protocol enables a node to discover most of its neighboring nodes in short discovery time, in the mean time, appropriately avoids collisions. Further evaluations are carried out in this paper by creating a new analytical model to investigate the theoretical relations among the discovery time, the discovery ratio and the system parameters. It can be concluded that the SDD protocol is an efficient, scalable and well performed protocol","Title: Toward A Seamless Communication Architecture For In-Building Networks At The 60 Ghz Band\nYear: 2006\n\nAbstract: This paper addresses the issues of designing an infrastructure to support seamless in-building communication at the 60 GHz band. Recently, the 60 GHz band has received much attention due to its 5 GHz of available spectrum. However the propagation of signals in this band is strongly hindered by attenuation and line-of-sight requirements. The situation gets worse in the in-building environment where signal propagation is obstructed by physical objects such as walls, furniture etc. In this paper, we present a novel Radio over Fiber (RoF) architecture that is cost-effective and is able to deliver high data-rate of the order of gigabits. To ensure a seamless communication environment at the 60 GHz band, we propose the concept of Extended Cells (EC) in order to create sufficient overlap areas. Finally, we illustrate the effectiveness of the proposed architecture by simulating an in-building network at the 60 GHz band employing the RoF and EC concepts.","Title: Communication Context for Adaptive Mobile Applications\nYear: 2005\n\nAbstract: This paper describes our ongoing work on a system that provides information about network resources and related communication facilities to applications running on a mobile device. It defines and elaborates on the concept of communication context and describes the architecture of this core low-level enabling system for ubiquitous and pervasive applications operating in a mobile environment.","Title: Exploring Multipath Capacity for Indoor 60 GHz Radio Networks\nYear: 2010\n\nAbstract: We study the application of multipath routing to enhance reliability of high quality video in 60 GHz radio indoor networks. We propose a cross-layer approach taking into account the possibility of layered-coding for video stream transportation and the characteristics of 60 GHz radio links based on our earlier studies on the stability of 60 GHz radio links. Further three traffic allocation schemes have been investigated and compared for the best use of multipath capacity.","Title: Architectural and QoS Aspects of Personal Networks\nYear: 2006\n\nAbstract: Personal networks (PNs) are future communication systems that combine wireless and infrastructure based networks to provide users a variety of services anywhere and anytime. PNs introduce new design challenges due to the heterogeneity of the involved technologies, the need for self-organization, the dynamics of the PN composition, the application-driven nature, the co-operation with infrastructure-based networks, and the security hazards. This paper discusses the challenges of security, service discovery and QoS provisioning in designing self-organized PNs and combines them all into an integrated architectural framework","Title: A framework for gateway and access network selection in ad-hoc distributed personal networks\nYear: 2010\n\nAbstract: A personal network (PN) is a network of personal devices regardless of their geographical locations. A PN may consist of several ad-hoc sub-networks, each of which is referred to as a cluster. Different clusters are connected to each other through their gateways and using interconnecting structures, e.g., the Internet. Inside each cluster, there might be several nodes that are able to act as gateway. Gateway nodes may support different access technologies, such as WiFi, Ethernet, WiMax, etc. It also may happen that a single gateway, which supports several access technologies, is able to connect to different access networks. To benefit from the availability of different gateways and access networks with different characteristics for a PN user, an efficient methodology for gateway and access network selection is needed. In this work, we are motivated to design a generic architectural framework for selecting the best gateway(s) and access network(s) in a PN. Issues, such as characteristics of the gateway nodes and access networks, QoS requirements of the applications, mobility of gateway nodes, dynamism of the PN topology, etc., have been considered in designing the proposed framework. The main advantages of this framework is leveraging the benefits of different alternative solutions, in terms of overhead, efficiency, complexity, and user satisfaction.","Title: Analysis of the Performance Boundaries of Sub-1 GHz WLANs in the 920MHz ISM-Band.\nYear: 2013\n\nAbstract: We explore the performance boundaries of sub-1 GHz Wireless Local Area Networks (WLANs) and summarize the limitations of our proposed narrow-band WLAN system that uses carrier frequencies at 920 MHz with 1 MHz channel bandwidth. While there are several evaluation reports on WLAN testbeds available, no report is available that discusses the boundaries of narrow-band multiple-input multiple-output Orthogonal Frequency Division Multiplexing (MIMO-OFDM) WLANs in sub-1 GHz. Even though 2.4 GHz/5 GHz MIMO-OFDM WLANs have been widely investigated, the characteristics of narrow-band WLAN is different from that of existing studies. This is mainly the result of limited channel bandwidth at lower frequency bands. This paper covers both, the fundamental performance boundaries for sub-1 GHz WLANs, and measurement results of our proposed software-based 2x2 MIMO-OFDM sub-1 GHz WLAN system operating at 920 MHz in order to identify the performance regimes, when using transmission protocols, including ICMP, UDP, and TCP.","Title: Adaptive CCA for IEEE 802.15.4 Wireless Sensor Networks to Mitigate Interference\nYear: 2010\n\nAbstract: IEEE 802.15.4 Wireless Sensor Networks (WSNs) share the 2.4 GHz Industrial, Scientific, and Medical (ISM) license-free band with many other wireless technologies such as IEEE 802.11b/g Wireless Local Area Networks (WLANs). Because of the low-power, however, IEEE 802.15.4 WSNs are potentially vulnerable to the interference introduced by the other wireless technologies such as IEEE 802.11b/g WLANs, which have much higher power. Particularly, in the presence of heavy interference, IEEE 802.15.4 nodes can hardly get a chance to access the channel, which could result in discarding a large amount of packets. In this paper, we propose a decentralized approach to help IEEE 802.15.4 nodes mitigate interference. By adaptively and distributively adjusting Clear Channel Assessment (CCA) thresholds of IEEE 802.15.4 nodes in the presence of heavy interference, the approach can substantially reduce the amount of discarded packets due to channel access failures, and therefore significantly enhance the performance of IEEE 802.15.4 WSNs. The approach is robust, responsive and easy to be implemented at a low cost. The effectiveness of the approach is validated by OPNET simulation.","Title: Fairness and network capacity trade-off in P2P IEEE 802.22 networks\nYear: 2012\n\nAbstract: Wireless Regional Area Networks(WRAN), IEEE 802.22 standard, adopts a centralized cellular topology including base station (BS) and customer-premise equipments (CPEs). However this may lead to limited network capacity, since every CPE needs to communicate to the BS and only one channel can be used to communicate in the whole cell per time slot with one BS antenna. Peer to Peer WRAN (P2PWRAN) was proposed [3] to circumvent this, where CPEs can communicate with each other directly. P2PWRAN increases network capacity compared to the standard WRAN because of spatial reuse. Multiple communication channels can be simultaneously allocated and reused in one time slot by the BS and multiple CPEs. In IEEE 802.22b [4], a standard in progress, peer to peer communication plays an important role for enabling smart grids. Channel allocation influences the network performance significantly, therefore, we formulate the spectrum allocation problem in P2PWRAN as a quadratically constrained programming (QCP) problem. It is also proved as a computationally hard problem. The Greedy Coloring Algorithm (GCA) is examined in P2PWRAN channel allocation, and the results show that it may cause severe unfairness in allocation of channels. Thus a Fair Greedy Coloring Algorithm (FGCA) is proposed to guarantee a fair allocation by queuing flows considering previous allocations. However, FGCA guarantees fairness but leads to decrease in performance of P2PWRAN. Therefore, a Trade-off FGCA (TFGCA) is proposed considering fairness and network performance at the same time during allocation. Simulation results show that with the adjustment of two factors in TFGCA, network performance and fairness can be balanced.","Title: Coexistence Performance of IEEE 802.15.4 Wireless Sensor Networks Under IEEE 802.11b/g Interference\nYear: 2013\n\nAbstract: IEEE 802.15.4 Wireless Sensor Networks (WSNs) and IEEE 802.11b/g Wireless Local Area Networks (WLANs) are often collocated, causing a coexistence issue since these networks share the same 2.4GHz Industrial, Scientific, and Medical band. In our previous work, we built a coexistence model of IEEE 802.15.4 WSNs and IEEE 802.11b/g WLANs. By identifying three distinct coexistence regions, the model explained the coexistence behavior of IEEE 802.15.4 WSNs and IEEE 802.11b/g WLANs, and the model was experimentally validated. In this paper, we improve the model by introducing two important implementation factors: the transceiver's Rx-to-Tx turnaround time and the Clear Channel Assessment partial detection effect. The enhanced model significantly improves the accuracy on explaining and predicting the coexistence performance of IEEE 802.15.4 WSNs in the real-life environment. Furthermore, under the guidance of the model, the coexistence performance of IEEE 802.15.4 WSNs is extensively investigated in various coexistence scenarios by analysis, simulation and experiments, respectively. The simulation and experimental results agree with our analysis. The coexistence model is believed to be helpful in resolving the coexistence issue.","Title: Maximizing the Fair Allocation of Opportunistic Spectrum for CR Ad Hoc Networks.\nYear: 2011\n\nAbstract: Cognitive Radios (CRs) address spectrum scarcity and under-utilization of the spectrum. Allocating spectrum to CR nodes in a network is neither easy nor straight-forward since spectrum availability is space and time-varying. The Link layer always sets up communication between nodes in a CR ad hoc network by allocating the available spectrum suitably. In a time-slotted model, allocation of channels amongst CR nodes is a NP complete problem. Moreover, since the availability of spectrum is time varying, all the nodes in the network may not be allocated sufficient resources. Thus there is always a possibility that a node may starve compared to its counterparts. In this article, we address the issue of fairness in a CR ad hoc network and propose novel distributed heuristics to allocate spectrum fairly. In literature, max-min, max-sum and max-proportional fairness based schemes are proposed to provide fairness in these CR ad hoc networks. We compare our heuristic results with these schemes, and discuss why our method is significantly better.","Title: Beam switching support to resolve link-blockage problem in 60 GHz WPANs\nYear: 2009\n\nAbstract: In this paper, we propose a solution to resolve link blockage problem in 60 GHz WPANs. Line-of-Sight (LOS) link is easily blocked by a moving person, which is concerned as one of the severe problems in 60 GHz systems. Beamforming is a feasible technique to resolve link blockage by switching the beam path from LOS link to a Non-LOS (NLOS) link. We propose and evaluate two kinds of Beam Switching (BS) mechanisms: instant decision based BS and environment learning based BS. We examine these mechanisms in a typical indoor WPAN scenario. Extensive simulations have been carried out, and our results reveal that combining angle-of-arrival with the received signal to noise ratio could make better decision for beam switching. Our work provides valuable observations for beam switching during point-to-point communication using 60 GHz radio.","Title: Successive interference cancellation in the uplink of Radio-over-Fiber systems\nYear: 2013\n\nAbstract: Successive Interference Cancellation (SIC) is a well-known technique for enhancing wireless channel efficiency by allowing concurrent transmissions from multiple users. This multiuser detection technique is usually used in cellular systems. In this paper we adapt it to Radio-over-Fiber (RoF) system for the same purpose. Due to the similarity to using SIC in typical cellular networks, we mainly focus on exploring the uniqueness of applying SIC in a distributed antenna application of RoF for uplink communication. We find that the physical channel capacity is decreased by additional noise in distributed antennas, but the MAC layer throughput could still be increased due to simultaneous transmission. We show that the performance can be further improved by employing transmit power control at the terminals and amplification control at the remote access units. All these findings make us believe that SIC is a promising technique in RoF system.","Title: Quantitative evaluation of scalability in broadband intelligent networks\nYear: 1998\n\nAbstract: Scalability is the ability of a network to maintain the quality of service while increasing certain parameters relating to the size of the network, such as the number of users, the number of network nodes, the number of services provided, geographical spread, etc. In the design of a B-IN signalling system, network scalability is an important issue that must be taken into account. In this paper we use simulation to investigate scalability issues related to a Broadband Intelligent Network (B-IN), such as that being considered in the ACTS project INSIGNIA. In particular, we study the impact of processor speed and configuration (in B-IN physical entities) on signalling performance. As signalling performance measures we consider the mean call setup delay of a B-IN service request and the network throughput. For Broadband Virtual Private Network (B-VPN) service, we perform scalability experiments by increasing some of the network parameters such as the number of users and the number of nodes.","Title: A Novel Approach for the Link Layer in Impulse-based UWB Ad Hoc Networks\nYear: 2007\n\nAbstract: Ultra Wide Band (UWB) impulse radio, promises to be suitable for short-range, low-power, low cost and high data rate applications. While most UWB research is concentrating on the physical layer, little research has been published on the link layer. The fundamental operations in the link layer need to take into account the specifics of impulse radio and also the particular features of ad hoc networks. A novel self-organizing link layer protocol based on time hopping impulse radio, called SDD, was proposed by the authors. This protocol is a collision-free mechanism that enables the devices to discover neighboring nodes and arrange the access to communication resources shared among the nodes. In this paper, some issues related to the self-organizing link layer based on UWB impulse radio are investigated and addressed. The SDD protocol is further developed and specified in detail. The simulations are carried out using GloMoSim simulation environment. An SDD module has been developed and embedded in the simulator. Results show that the SDD protocol can work properly and efficiently in a single-hop ad hoc network.","Title: Mobility modeling for personal networks: properties and its impact\nYear: 2010\n\nAbstract: User centered Personal Network (PN) interconnects various devices belonging to a user in different geographic locations, such as home, office, car, etc., to form one network to improve the quality of life of the user. In PNs, the user's devices move in different groups, where these groups merge and split. A PN Mobility Model (PNMM) has been proposed to model the devices' movements in PNs. In this paper, PNMM has been evaluated: stability evaluation proves that PNMM maintains a long-term steady state. The impact of PNMM on the performance of protocols designed for PNs has been discussed in this paper. Simulation results give insights of the performance of the PN Clustering Protocol, and provide guidelines for future design of clustering in PNs.","Title: An Intra-Cell Peer To Peer Protocol In Ieee 802.22 Networks\nYear: 2011\n\nAbstract: IEEE 802.22 mostly considers a typical cellular network with centralized control. Therefore, all intra-cell communications have to be via base stations. It not only leads to extra power consumption for intra-cell communications, but also only one channel can be used per time slot in the whole cell. This may lead to decrease in the network performance. Therefore, we propose a protocol called intra-cell peer to peer communication protocol (ICP2P) based on IEEE 802.22 protocol. The main accomplishments of ICP2P are to support direct intra-cell peer to peer communications between Customer Premises Equipments (CPEs), to allocate multiple channels per time slot in a cell, and to re-allocate same channels in one time slot. The network performance, such as capacity, channel usage and energy consumption, are significantly improved by ICP2P. Additionally, a weighed demand queue algorithm (WDQA) is proposed to guarantee the fairness in channel access opportunities for CPEs.","Title: Performance modelling of the Orwell basic access mechanism\nYear: 1987\n\nAbstract: Orwell is a high speed slotted ring. Its protocol uses destination release of the slots. Because of this the carried load can be much larger than the transmission rate. A new analytical model of the Orwell basic access mechanism is presented in this paper. The model shows to be accurate and usable over a wide range of parameters. The performance analysis of the Orwell basic access mechanism is presented.","Title: Performability modelling tools and techniques\nYear: 1996\n\nAbstract: Over the last decade considerable effort has been put in the development of techniques to assess the performance and the dependability of computer and communication systems in an integrated way. This so-called performability modelling becomes especially useful when the system under study can operate partially, which is for instance the case for fault-tolerant computer systems and distributed systems. Modelling techniques are a fundamental prerequisite for actually doing performability analysis. A prerequisite of a more practical but not less important nature is the availability of software tools to support the modelling techniques and to allow system designers to incorporate the new techniques in the design process of systems. Since performability modelling requires many aspects of a system to be specified, high requirements should be posed on performability modelling tools. Moreover, these tools should be structured such that the models can be specified at a level that is easy to understand for a system designer, and that the mathematical aspects are hidden as much as possible. The output of the tool should also be such that it can be understood with only limited knowledge of the underlying mathematical model. We have developed a new, fairly general modelling tool framework that can be used as a guide to assess the usability and structure of performability modelling tools. After briefly reviewing the mathematical aspects of performability modelling we discuss this framework. We then discuss 12 recently developed tools (Metaphor, Numas, Metasan, Metfac, Save, Sharpe, SPNP, Tangram, Penpet, UltraSAN, Surf-2, DyQNtool + ) that can all be used for some aspects of performability modelling and analysis. We assess among other things their structure, their capabilities in terms of measures that can be obtained, and the used modelling formalism. We also discuss directions for future work in the field of performability modelling tools.","Title: A study on wireless sensor network based indoor positioning systems for context-aware applications.\nYear: 2012\n\nAbstract: The newer context-aware applications require many inputs and amongst them the location information is one of the most important. In the near future, we see the potential in using wireless sensors deployed inside buildings to support in generating the location information besides their other routine tasks. This paper records the efforts involved in designing and prototyping a centralized indoor positioning system for tracking a target's position. We present an in-depth discussion on the RSSI based positioning algorithms both, range-based and range-free. Considering that the signal strength can be distorted heavily due to multi-path and shadowing, we have proposed the weighing schemes to leverage the credibility of the measured RSSI values. For online tracking, we have also proposed boundary selection and local grid scanning to lower the searching time, and the RSSI data dissemination and collection schemes to reduce traffic overhead. Evaluations have been done in both the field measurements and with an RSSI generator. The generator has been developed to simulate and replace the real measurements for the ease of algorithm design and testing, thus avoiding repetitive field measurements. The results show that positioning systems with adequate accuracy can be built with our proposed schemes. We expect that these proposed schemes to be integrated in multitude of systems with context-aware applications, which use location information. Copyright (C) 2010 John Wiley & Sons, Ltd.","Title: The concept of fairness and a GFC protocol\nYear: 1992\n\nAbstract: Multiple access to a single user-network interface (UNI) is a cost-effective solution for the provision of B-ISDN services in customer premises. In a multiple access configuration, a fairly shared access must ensure that each terminal will get a guaranteed bandwidth to provide an acceptable quality of service (QoS) and that the allocation of the remaining bandwidth will result in the same improvement of QoS for all the terminals. An approach to allocating the remaining bandwidth is proposed. It is based on both the guaranteed bandwidth required by the terminal and the probability that the terminal's bit rate exceeds its guaranteed bandwidth. The basic generic flow control (GFC) protocol mechanism and the GFC field settings for a dual-bus customer premises network (CPN) are described","Title: Distributed Heuristics for Allocating Spectrum in CR Ad hoc Networks\nYear: 2010\n\nAbstract: Cognitive Radios (CRs) address spectrum scarcity and under-utilization of the spectrum. However, realizing a CR network is neither easy nor straight-forward. In particular the link layer should facilitate setting up of communication between nodes enabling sharing of channels, while maximizing the spectrum utilization. We address the problem of allocation of channels to CR nodes. For time-slotted model, allocating the channels among CR nodes is an NP complete problem. We propose novel distributed heuristics, one based on clique sets called Clique Based Heuristic and the second one is a low complexity, locally adapting Localized Heuristic; both use local control channel. We compare the results of these heuristics with the optimal and a distributed graph coloring algorithm proposed in the literature. We also compare the spectrum utilization and overheads for all these algorithms.","Title: Density estimation for out-of-range events on personal mobile devices\nYear: 2008\n\nAbstract: Over the years, personal mobile devices have obtained increasing capabilities to concurrently connect to surrounding networks and nearby other devices. Good knowledge on the dynamics in the availability of these heterogeneous entities constitutes essential input for various data communication problems, ranging from the adaptation of applications on mobile devices to multi-homed situations, to the optimization of routing protocols for delay tolerant networks. In this paper, we focus on a method for the prediction in time of the loss in visibility of currently in-range network entities, as observed on a personal mobile device. We are interested in estimating the full probability density function of the time of these out-of-range events, as this allows us to ask arbitrary questions such as: what is the probability of losing connection X in the next Y minutes? To do so, we model the mobility of the user by applying kernel density estimation on previously observed mobility traces collected during a user experiment we ran with 12 participants in a six week period, logging cellular, 802.11 wireless LAN, Bluetooth, and various other events on the participant's mobile device.","Title: Collaborative and Cognitive Network Platforms: Vision and Research Challenges\nYear: 2011\n\nAbstract: In this paper, we present a visionary concept referred to as Collaborative and Cognitive Network Platforms (CCNPs) as a future-proof solution for creating a dependable, self-organizing and self-managing communication substrate for effective ICT solutions to societal problems. CCNP creates a cooperative communication platform to support critical services across a range of business sectors. CCNP is based on the personal network (PN) technology which is an inherently cooperative environment prototyped in the Dutch Freeband PNP2008 and the European Union IST MAGNET projects. In CCNP, the cognitive control plane strives to exploit the resources to better satisfy the requirements of networked applications. CCNP facilitates collaboration inherently. Through cognition in the cognitive control plane, CCNP becomes a self-managed substrate. The self-managed substrate, in this paper, is defined as cognitive and collaborative middleware on which future applications run without user intervention. Endemic sensor networks may be incorporated into the CCNP concept to feed its cognitive control plane. In this paper, we present the CCNP concept and discuss the research challenges related to collaboration and cognition.","Title: FEW-PNets - A framework for emulation of wireless personal networks\nYear: 2008\n\nAbstract: With the advent of miniaturization and higher computing capabilities of wireless devices, there is an exponential rise in the number of devices such as personal digital assistants, etc. Application development for these devices is the main focus for many device manufacturers and software developers. The applications developed need to be tested and tried before releasing it to the market. There are cross-compilers to produce the binary files of the software applications that can run on these devices. However, it is rather difficult for developers to test it on the actual devices every time a change is made in the program. Further, a wireless network infrastructure should also be in place for testing. To expedite application development FEW-PNets — a Framework for Emulation of Wireless Personal Networks is proposed. It is a networked application to emulate a Personal Network (PN) owned by a person. It supports emulation of the most functions of PNs. We explain our model by emulating simple PN devices. In this article we elucidate the concepts, advantages, application domains and limitations of FEW-PNets.","Title: A novel link quality assessment method for mobile multi-rate multi-hop wireless networks\nYear: 2010\n\nAbstract: Accurate and fast wireless link quality assessment (LQA) for wireless channels would bring huge benefits for mobile multi-hop and multi-rate wireless ad hoc and sensor networks in the form of improved end-to-end performance. In this paper, we propose a novel LQA method based on cross-layer information. The method is implemented in a real test-bed, which is based on IEEE 802.11b/g, and achieved a significant LQA improvement up to 50% in mobile scenarios without introducing overhead. The effectiveness of accurate and fast LQA is demonstrated by feeding it into the routing layer to enable the route decisions to adapt faster during changing situations, especially in mobile scenarios. The experiment results show that the proposed LQA method can lead to faster and smarter routing decisions and higher end-to-end throughput compared to traditional methods, which only use hello packets to determine the link quality in mobile scenario.","Title: A multipath approach for reliable high quality video transport in indoor 60 GHz radio networks\nYear: 2010\n\nAbstract: The emerging high definition in-home multimedia applications require higher capacity from future wireless networks in the indoor environments. In this paper, we consider a novel mesh-like architecture for indoor wireless networks using 60 GHz radio for supporting multimedia applications with sufficient data rate and stringent quality-of-service (QoS) guarantees. We concentrate on studying the application of multipath routing to enhance reliability of transporting high quality video since 60 GHz transmission is highly dependent on line-of-sight (LOS). Based on our previous study on the stability of 60 GHz radios, we propose a cross-layer approach taking into account layered coding technique for video stream transportation. We present our model along with an analysis to assess multipath routing in this scenario. Further, we give an in-depth discussion on two traffic allocation schemes to harness multipath capability.","Title: Performance analysis of the frame aggregation mechanisms in IEEE 802.15.3c\nYear: 2009\n\nAbstract: To achieve ultra high data rate communications in 60 GHz Wireless Personal Area Networks (WPANs), frame aggregation is devised in IEEE 802.15.3c to increase the MAC transmission efficiency. In this work, we propose a novel statistical model to analyze the behavior of two types of frame aggregation mechanisms: standard frame aggregation and low latency frame aggregation. Through our model, we could fully understand the performance of the frame aggregation mechanisms in terms of transmission delay and system capacity. The accuracy of our model is validated by extensive simulations. Our work provides valuable insights for ultra high performance transmission using 60 GHz radio.","Title: Procedure to build interference map in peer to peer IEEE 802.22 networks\nYear: 2012\n\nAbstract: Peer to peer wireless regional area network (P2PWRAN) is proposed as an extension to support peer to peer communication based on IEEE 802.22 [1]. Multiple channel allocation and reuse of channels in the same time slot in P2PWRAN significantly increase the network capacity compared to standard IEEE 802.22 networks. One of the key issues in bringing P2PWRAN into reality is building the interference map. The concept of interference map/mapping has been mentioned in literature however, how to build an interference map is still an open issue. Therefore, we propose a simple and self-adapting interference map building protocol (SIMBP) for P2PWRAN, which can also be used in other multi-channel wireless networks with minor modifications. The simulation results show that SIMBP converges under the P2PWRAN setting, and with number of available channels and growing number of nodes, the capacity of the network is reached eventually.","Title: Ad hoc networking in future wireless communications\nYear: 2003\n\nAbstract: This paper contains an overview of the discussions on future research directions within the subject of ad hoc networking, held at the Wireless World Research Forum meetings during 2001. Ad hoc networking is an emerging research field where ad hoc networks are no longer viewed as stand-alone groups of wireless terminals. On the contrary, ad hoc networks are expected to become fundamental in the future development of infrastructure networks and they will be the basis for enabling ubiquitous communications. Ad hoc networking involves new research issues at all layers.","Title: MAC Protocol Design for Impulse Radio UWB Based WPANs\nYear: 2007\n\nAbstract: Ultra Wide Band (UWB) has recently gained great interest for high data rate short-range WPANs. Designing a MAC protocol for UWB WPANs must take into account the physical layer characteristics. Due to large bandwidth and spreading spectrum, a time hopping impulse radio UWB (IR-UWB) system can accomplish multiple access by using code division multiplexing. Unlike most existing MAC protocols relying on contention access to limit transmissions, the MAC design for impulse radio intends to accommodate simultaneous transmissions. In this paper, we newly propose a MAC scheme for IR-UWB WPANs. This scheme is a hybrid CDMA and TDMA approach based on time hopping spread spectrum and the timing format of IEEE 802.15.3 MAC. The simulation results show that the proposed approach outperforms the existing protocol under our specifications and WPAN scenarios.","Title: A Self-Organizing Link Layer Protocol for UWB Ad Hoc Networks\nYear: 2004\n\nAbstract: Ultra Wide Band (UWB) impulse radio is a promising technology for future short-range, low-power, low cost and high data rate ad hoe networks. The technology is being explored in a number of research projects. While most UWB research for this class of networks is concentrating on the physical layer, little research has been published on link layer protocols which exploit the specifics of UWB impulse radio. In this paper, we focus on the self-organization concept and the peculiarities of UWB technology from a physical and a link layer point of view. A novel self-organizing link layer protocol based on time hopping spread spectrum is proposed in this paper. This protocol promises to be an efficient and collision-free mechanism that enables the devices to discover neighbor nodes and arrange the access to communication resources shared among the nodes. The adjustable parameters of the protocol enable the network to adapt to a dynamic environment.","Title: Analysing IEEE 802.15.3c protocol in Fi-Wi hybrid networks.\nYear: 2013\n\nAbstract: Even though 60 GHz frequency band has limited coverage, due to the availability of higher bandwidth worldwide (unlicensed, approximately 5 GHz between 57-62 GHz), it is one of the promising candidates for future broadband employing both Fiber and Wireless (Fi-Wi) technology. In this article we briefly explain an in-home radio over fiber (RoF) architecture employing 60 GHz spectrum band. The main challenge for designers of such networks is the additional delay introduced by the optical distribution network. Later in the light of this increased delay, we provide an investigation on the feasibility of applying the IEEE 802.15.3c MAC protocol in such a RoF system at 60 GHz. Mainly, we analyse various ACK mechanisms in the IEEE 802.15.3c when it is used in conjunction with RoF network. The analytical results show that the fiber optic network causes a slight drop in the throughput but it could still be easily used in home networks. This gives us an opportunity to sustain high bandwidth at low cost for home networking, also supporting mobility. Modelling of a RoF system with IEEE 802.15.3c and numerical results thereof are the contributions of this article.","Title: Voice Activity Detection For Voip-An Information Theoretic Approach\nYear: 2006\n\nAbstract: Voice enabled applications over the Internet are rapidly gaining popularity. Reducing the total bandwidth requirement can make a non-trivial difference for the subscribers having low speed connectivity. Voice Activity Detection algorithms for VoIP applications can save bandwidth by filtering the frames that do not contain speech. In this paper we introduce a novel technique to identify the voice and silent regions of a speech stream very much suitable for VoIP calls. We use an information theoretic measure, called spectral entropy, for differentiating the silence from the speech zones. Specifically we developed a heuristic approach that uses an adaptive threshold to minimize the miss detection in the presence of noise. The performance of our approach is compared with the relatively new 3GPP TS 26.194 (AMR-WB) standard, along with the listeners' intelligibility rating. Our algorithm yields comparatively better saving in bandwidth, yet maintaining good quality of the speech streams.","Title: Adaptable Link Quality Estimation for Multi Data Rate Communication Networks\nYear: 2009\n\nAbstract: QoS-sensitive applications transmitted over wireless links require precise knowledge of the wireless environment. However, the dynamic nature of wireless channel, together with its different configurations and types, makes so called link quality estimation (LQE) a difficult task. This paper looks into the design of an accurate and fast LQE method for a multi data rate environment. We investigate the impact of various conditions on the LQE accuracy. In result, two different link quality estimation sources, i.e., based on hello packet delivery ratio and signal strength, are measured and their performances are compared. We find that these two methods are not always accurate. As an improvement we propose an adaptive LQE method that chooses different LQE indicators depending on the wireless environment. The performance of the proposed method is verified via an extensive measurement campaign.","Title: On the Performance of TCP Vegas over UMTS/WCDMA Channels with Large Round-Trip Time Variations\nYear: 2004\n\nAbstract: Universal Mobile Telecommunications System (UMTS) is a third-generation cellular network that enables high-speed mobile Internet access. This paper evaluates and compares the performance of two well-known versions of Transmission Control Protocol (TCP), namely, Vegas and Reno, in a UMTS environment. Bulk data transfer was considered in the simulation with varying radio channel conditions. We assume that data losses are only due to the radio channel. Simulation results show that the performance of Vegas is worse than Reno even though data losses incurred by the radio channel are completely recovered by the UMTS radio link control layer. This has led us to conduct a thorough investigation on the behavior of Vegas in order to identify the cause of performance degradation in Vegas. The poor performance of Vegas is attributed to the UMTS radio interface characteristics which resulted in large and highly variable TCP round-trip times. Vegas would interpret the round-trip time variation as a sign of congestion, and consequently, shrink its window size which reduces the transmission rate. Furthermore, a sudden increase in the instantaneous round-trip time can trigger spurious timeouts at the TCP sender using Vegas which performs unnecessary retransmissions. Spurious timeouts can lead to significant throughput reduction. Reno, on the other hand, does not show any abnormality and delivers the expected performance.","Title: An Analytical Energy Consumption Model for Packet Transfer over Wireless Links.\nYear: 2012\n\nAbstract: We provide a detailed analytical model for estimating the total energy consumed to exchange a packet over a wireless link. Our model improves many of the current models by considering details such as consumed energy by processing elements of transceivers, packet retransmission, reliability of links, size of data packets and acknowledgments, and also the data rate of wireless links. To develop the model, we use experimental results based on IEEE 802.15.4 devices to show that consumed energy for receiving erroneous packets is comparable to the consumed energy for receiving error-free packets.","Title: A Link Layer Protocol For Selforganizing Ultra Wide Band Impulse Radio Networks\nYear: 2005\n\nAbstract: Ultra Wide Band (UWB) impulse radio, promises to be suitable for short-range, low-power, low cost and high data rate applications. While most UWB research is concentrating on the physical layer, little research has been published on the link layer. A novel self-organizing link layer protocol (SDD) based on time hopping impulse radio was proposed by the authors. In this paper, the SDD protocol is further developed and specified in detail. The simulations are carried out using the GloMoSim simulation environment.","Title: A Framework For Massive Access And Radio Resource Management In Urban Wlans\nYear: 2013\n\nAbstract: The challenge in future unlicensed wireless networks will be the coordination of spectrum access among wireless LAN (WLAN) nodes in high dense areas. New WLAN standards, such as IEEE 802.11ah, will provide optimized outdoor WLAN coverage with a range of 1 km, thus exceeding the number of associated WLAN nodes. Hence, massive access schemes are required in order to coordinate the wireless access among WLAN nodes and to mitigate mutual interference. In addition, various wireless networks will co-exist, such as WLANs and sensor networks, competing for channel access. Hence, a framework in order to coordinate WLANs and sensor networks in dense areas is essential. This paper is intended to outline the challenges of wireless access in urban WLANs and proposes a framework which considers the challenges of dense wireless access and discusses potential solutions. The contributions are twofold. A clustering framework is proposed to deal with the massive access of wireless nodes, in particular of IEEE 802.11ah WLAN nodes. Further, a radio resource monitoring and management (RRMM) is proposed to manage the co-existence among wireless networks. A novel system architecture is proposed and discussed in order to monitor and manage the WLAN access in highly dense urban areas.","Title: Home Networking at 60 GHz: Challenges and Research Issues.\nYear: 2008\n\nAbstract: We highly believe that 60 GHz wireless technology is an ideal candidate to be the future high-speed WLAN standard for home\n networks thanks to its unique characteristics. In this paper, we review and present the main aspects of the 60 GHz radio regarding\n future home networks. We introduce the fundamental challenges with the research issues and the considerations on some promising\n solutions.","Title: An implementation study of relay selection schemes for energy harvesting WSNs\nYear: 2013\n\nAbstract: We propose energy harvesting technologies and cooperative relaying techniques to power the devices and improve reliability. We propose schemes to (a) maximize the packet reception ratio (PRR) by cooperation and (b) minimize the average packet delay (APD) by cooperation amongst nodes. Our key result and insight from the testbed implementation is about total data transmitted by each relay. A greedy policy that relays more data under a good harvesting condition turns out to be a sub optimal policy. This is because, energy replenishment is a slow process. The optimal scheme offers a low APD and also improves PRR.","Title: Throughput schemes for energy harvesting sensor networks\nYear: 2012\n\nAbstract: With the growing dependence on the wireless sensor devices, it is imperative that power supply to them and their energy usage has acquired centre stage. Energy harvesting is one possible remedy. To enhance the use of devices that are using harvested energy, many novel schemes need to be devised. In this paper we propose four throughput enhancement schemes with a wide range of applicability to energy harvesting wireless sensor networks. The schemes are designed by taking into consideration available resources in the nodes as well as the energy harvesters that are used to drive the sensor nodes. All the four schemes are implemented, evaluated and compared for their performances. We present schemes from simple naive scheme which is of low complexity, to probabilistic probing scheme which incorporates advanced methods to appropriately use the harvested energy. All the schemes require minimum system resources and sufficient harvested energy before starting communication. Moreover, when the harvested energy varies due to the changes in the environment and characteristics of wireless channel varies, we found that truncated channel inversion with power control is the appropriate scheme. This article provides a thorough step by step discussion on the implementation of maximizing communication throughput in energy harvesting sensor nodes under varying environment. We also provide some insights into energy requirements for a multi-node network.","Title: Spectrum- and energy-efficient D2DWRAN\nYear: 2014\n\nAbstract: IEEE 802.22 is the first world-wide standard for wireless regional area networks using cognitive radio technologies. Major drawbacks of these networks are limited network capacity and low energy efficiency. One potential solution for these drawbacks is device-to-device communication, which can be incorporated as D2DWRAN. D2DWRAN features direct intracell communication, reuse of channels, and use of multiple operating channels. In this article, we mainly discuss energy-efficient spectrum sharing in D2DWRAN. We explicate a mechanism to increase energy efficiency for long distance links using relays. Along with energy, channel utilization and fairness are also considered while choosing relays. Thus, significant improvement in both network capacity and energy efficiency can be achieved. The schemes discussed here can also be used in other D2D communication settings.","Title: Comparing network performance for constant and variable bit rate sources.\nYear: 1993\n\nAbstract: The token ring FDDI and the slotted bus DQDB are currently the two most important standards for high-speed networks. Each of these networks has its own specific advantages and disadvantages, and its own specific type of load for which it performs best. We propose a new network design, called Universal Channel Network (UCN), which is adaptive and can be made to behave like a token ring, a slotted ring, or anything in between. The aim is to get a network that combines the advantages of both the token ring and the slotted ring. In this paper, UCN is explained and its performance is compared with that of DQDB and FDDI. The workload model used is a detailed representation of today's traffic, extrapolated to the future. Because of the uncertainty concerning future traffic, two versions of the workload model are used. In the first version, voice and video sources are modelled as constant bit rate sources. In the second version, variable bit rate sources are used to represent voice and video sources. The network performance for both coding techniques is compared.","Title: Downlink throughput characterization of RoF based hybrid 802.11 Fi-Wi\nYear: 2014\n\nAbstract: Radio over Fiber (RoF) based IEEE 802.11x (where x can be a, b, g, n, ac) Fiber Wireless (Fi-Wi) LANs are being thought of as a better solution towards deployment of Enterprise WLAN. It provides optimized load distribution among Access Points (APs) and also better efficient power management of the overall network. In most of the existing works about WLAN deployments, the classical models characterizing the throughput achievable discusses about either single collision solutions where simultaneous transmissions are not allowed or multi cell Enterprise WLANs where the simultaneous transmission are the most predominant feature. But RoF based WLANs because of its inherent architectural design has a different structure. In this paper we want to characterize the downlink throughput that is achievable in such a RoF based Fi-Wi network. The placement of the Cell Access Nodes (CANs) which are simple antennas located indoor to radiate the radio signals is an important factor as well as the connectivity between different CANs and the APs located at the Home Communication Controller (HCC). We analytically provide an upper bound on the difference between the maximum and minimum downlink throughput achievable by an AP in such RoF based Fi-Wi network.","Title: Green hybrid Fi-Wi LAN\nYear: 2014\n\nAbstract: Enterprise WLANs encompassing supernumerary Access Points (APs) are being used to provide capacity to indoor environments. The approach of over dimensioned deployment of APs to meet the capacity needs of users at critical time points are advocated without little concern towards the important issue of power wastage. For a sustainable greener environment the issue of power saving in such enterprise WLANs has to be integrated in our present day design. Hybrid Fiber-Wireless (Fi-Wi) LAN is envisioned as the evolution of enterprise WLAN which can help us enormously to reduce the power wastage while meeting the demands of the indoor users. In this paper, we propose a hybrid Fi-Wi LAN architecture based On Demand Strategy which results in massive power saving (approximately 60%) by switching of APs or Cell Access Nodes (CANs) based on user demand estimation while maintaining coverage throughout the whole cell area.","Title: Measurements and performance of large MIMO systems at 2.4 GHz for indoor WLAN\nYear: 2014\n\nAbstract: Large MIMO is a new and promising technique for boosting wireless link capacity. Currently, the research of large MIMO focuses on the theoretical aspects while work on the practical implementations and analysis are very limited. In this paper, we present some critical investigations on the behaviors of large MIMO in real propagation environment based on extensive channel measurements. On the one hand, our analysis has verified the advantages that the MIMO channel vectors between users tend to be orthogonal and the optimal capacity approximates the theoretical upper bound as the antenna array size grows. On the other hand, we observe that the practical limitations of allowable antenna array size and achievable spatial degrees of freedom in the physical channel have significant impacts. Furthermore, in order to enable the utilization of linear precoding techniques in such high-dimensional MIMO systems to achieve close-to-optimal capacity, a large ratio between the antenna array size of the large MIMO system and the number of simultaneously served users should be maintained.","Title: Codebook Selection Strategies in Long-range Sub-1 GHz WLANs.\nYear: 2014\n\nAbstract: The new long-range WLANs will enable a simple and cost-efficient setup of wireless ambient systems, including smart city and smart grid communication networks. The emerging IEEE 802.11ah WLAN standard utilizes carrier frequencies at 900MHz in the ISM-band and facilitates multi-antenna MIMO-OFDM. In order to further increase the wireless coverage, precoding selection strategies are proposed and analyzed. The Grassmannian and the Kerdock manifold are in the focus of this study due to their simplicity and effectiveness. Whereas the Grassmannian manifold may provide an optimal codeword distance, the Kerdock man- ifold has been found to be more effective due to its reduced quantized codewords combined with an increased SNR. Simulation studies as well as real-world evaluation results indicate a 2-3dB increase of precoding gains when the proposed Kerdock manifold is applied, thus increasing the coverage. The codebooks have been evaluated in a SDR-based IEEE 802.11ah WLAN prototype.","Title: An Ultra-Wide Band Based Ad Hoc Networking Scheme For Personnel Tracking In Emergencies\nYear: 2008\n\nAbstract: This paper describes an ultra wideband (UWB) based ad hoc networking scheme which enables a control center to track all emergency personnel deployed in buildings in an incident area, and also enables communication of other types of data directed from emergency personnel, including status information and alarm in particular. In this network, GPS-enabled vehicles form a coordination reference framework. They also serve as egress points to collect data generated by emergency personnel. Additional dropped terminals are used to expand the network coverage. A Time Division Multiple Access (TDMA) based medium access protocol is used by terminals to access the radio channel. An anycast routing protocol is proposed and designed. This protocol enables each ad hoc device attached to emergency personnel to discover a communication path to one of the egress points. The network management related information is included in packet headers. Thus terminals can select free time slots and construct routes by overhearing neighbor's transmissions. Taking the advantage of the periodically nature of position calculation and reporting, each terminal is maintained with up-to-date network status.","Title: Architectures for Communication in Personal Networks\nYear: 2006\n\nAbstract: Personal networks (PN) is a new concept related to pervasive computing with a strong user-focus view. The key to a successful PN realization is a general network architecture that is capable of bridging different current and future technologies and offers a homogeneous and clear view to the end-user. In this paper, we focus on forming a PN by connecting remote personal devices using infrastructure-based IP networks, including 3G networks and WLAN hotspots. One way is to upgrade the current access networks with new functionality to support PNs. Since many devices in PNs are mobile and battery powered, this may help them to achieve a faster service and to save energy. However, to deploy such functionality is not easy and may hamper the adoption of PNs altogether. Therefore, in this paper we study three possible inter-cluster communication architectures that can use current IP networks. To discern the above proposal we also give a detailed picture of PN network architecture supported by infrastructure. We believe that this detailed discussion will help the success of PNs","Title: Adapting IEEE 802.22 OFDMA system for P2PWRANs.\nYear: 2013\n\nAbstract: IEEE 802.22 is the first standard to uitilize cognitive radio technology for wireless regional area networks (WRANs). It has adopted a cellular topology containing one base station (BS) and multiple customer premises equipments (CPEs) in a cell. It covers a very large area with radius ranging up to 100 km. Orthogonal frequency-division multiple access (OFDMA) system is employed and it is further slotted in time-domain. In one slot only one CPE can be allocated, and every intra-cell packet needs to be routed through the BS. This point-to-multi-point structure significantly limits the network capacity in the case of CPE-CPE communication, which is predicted to happen, and even the use of OFDMA does not help much. Therefore, peer-to-peer WRAN (P2PWRAN) has been proposed to support direct intra-cell communication in order to extend the network capacity. In this paper, the IEEE 802.22 OFDMA system is adapted to P2PWRAN to support direct CPE-CPE communications. The downstream burst allocation in P2PWRAN is similar to the OFDMA system of IEEE 802.16, which has been widely studied in the literature. Therefore, in this paper we look at the unexplored upstream burst allocation where slots can be reused among CPEs and the BS with power control mechanisms. We propose a burst allocation algorithm that maximizes the network capacity greedily. The algorithm is examined under various conditions such as, number of CPEs, size of requests, length of US subframes and type of requests (CPE-BS or CPE-BS). We also show that our algorithm performs better than the existing solutions.","Title: Solving incertitude of vertical handovers in heterogeneous mobile wireless network\nYear: 2007\n\nAbstract: Wireless communication network is evolving towards more heterogeneous with rapid emergence of new radio technologies. 60GHz radio relying on line-of-sight (LOS) transmission is one of the promising solutions to support in-door high quality multimedia applications. Due to the vulnerable nature of LOS links, vertical handover, by switching the mobile device to the backup WLAN system from 60GHz radio, is critical to support a seamless session. However vertical handover for short LOS blocking events degrade perceived quality of the media due to frequent switching. In this paper, we propose a cognitive approach to solve incertitude during vertical handover. We first present an overview of our proposed cognitive network architecture, and further under this framework we introduce and examine via simulations a decision algorithm based on decision theory. Our algorithm takes into account multiple factors such as user preference, network situation, device capability and effect of environment. We show its ability to effectively make handover decision in uncertain situations. The method and results herein are applicable, in general, to any other situation where such a decision is to be made.","Title: Self-coexistence and spectrum sharing in device-to-device WRANs\nYear: 2014\n\nAbstract: IEEE 802.22 wireless regional area network (WRAN) standard employing cognitive radios is gaining much attention recently. The WRAN standard targets reuse of unused TV channels. We propose a device-to-device wireless regional area network (D2DWRAN) to extend the capacity of IEEE 802.22. Network capacity can be increased by supporting direct intra-cell device to device (D2D) communication through channel reuse and also with aggregation of nonadjacent multiple operating channels. Self-coexistence of neighboring IEEE 802.22 cells is a major challenge in WRANs, since the availability of channels varies frequently and channels are reused by every cell as much as possible. Currently, IEEE 802.22 does not consider D2D communication. We propose a two-tier spectrum sharing mechanism for channel allocation in D2DWRANs - both at intra-cell and inter-cell levels. Algorithms and a thorough analysis are presented. We examine our proposal through simulations. Results show significant performance improvement compared to IEEE 802.22. However, there are many hurdles to cross, such as, coexistence with other cognitive radio networks (CRNs), the intra-cell routing, allocation of other network resources, etc. We believe that further work in this domain can lead to increase in capacity not only in WRANs but also in other cellular networks.","Title: Multi-channel management for D2D communications in IEEE 802.22 WRANs\nYear: 2014\n\nAbstract: IEEE 802.22 standard is the first worldwide standard using cognitive radio (CR) technology for wireless regional area networks (WRANs). Orthogonal frequency-division multiple access (OFDMA) is employed for point-to-multi-point (P2M) communication between base stations (BS) and customer premises equipments (CPEs) in WRANs. The large coverage of a single cell (up to 100 km of radius) is the unique feature of WRANs. However, channel reuse is not considered, i.e., in one slot the channel can be allocated to only one CPE, and every intra-cell packet needs to be routed through the BS. Therefore, device-to-device WRAN (D2DWRAN) has been proposed to support direct intra-cell CPE-CPE communication. Power control strategies are used to achieve higher reuse of OFDMA slots. In this paper, we take one step further to enhance the network capacity by adopting multichannel management (MCM) with infrastructure support. This enables multiple operating channels to be used simultaneously in D2DWRANs. New channel switching strategies are also studied. We implement a MCM scheme in an OFDMA setting for D2DWRANs, and examine it in the upstream (US). For downstream (DS), there are already many proposals in the literature, which can be adopted directly by D2DWRAN. The simulation results show that MCM scheme significantly improves the network performance compared to single-channel D2DWRANs and WRANs.","Title: On systems generating context triggers through energy harvesting\nYear: 2014\n\nAbstract: Context awareness is one of the building blocks of smart applications that constitutes smart spaces. With the emergence of cyber physical systems, it is now possible to create spaces that are truly adaptable and smart. In these spaces, contextual parameters are captured by many wireless sensor nodes. This collected data is processed to understand the context in real time. Since a large number of sensors are deployed, processing all the data is a big task. Moreover, since the sensors are powered by batteries, they have a limited lifetime. Sensing only when there is a context event can save energy as well as reduce data processing. To make sensor nodes operate perpetually, ambient energy harvesters can be used. Typically, the energy harvesting source of a sensor is related to the physical parameters the sensor is measuring. We propose exploiting this property to develop a context-event triggering mechanism in this article. We also adapt the context-aware application framework for incorporating context-event triggers by harvesters. A Smart-M3-based architecture is also proposed. Through a real-world use case, we illustrate significant energy savings and reduced data processing in our proposed approach.","Title: A simulation study on a high-speed slotted-ring access mechanism with dynamically adaptive slot sizes\nYear: 1991\n\nAbstract: Two access mechanisms complementary in performance are the token ring and the slotted ring. While the token ring outperforms the slotted ring for long messages, the latter performs significantly better for short messages. The paper shows that the factor that causes this difference is the number of tokens on the ring. It proposes a new network design in which the number of tokens has been made adaptive, resulting in a network that can be made to behave like both the token ring and the slotted ring, or anything in between, i.e. it performs optimally for any given message length. The paper presents first simulation results that show the performance gain that is achieved by using this principle. Further, it introduces a new priority mechanism in which an upper bound for the access delay of high priority messages is guaranteed","Title: Cooperative Density Estimation in Random Wireless Ad Hoc Networks\nYear: 2012\n\nAbstract: Density estimation is crucial for wireless ad hoc networks for adequate capacity planning. Protocols have to adapt their operation to the density since the throughput in an ad hoc network approaches asymptotically to zero as the density increases. A wireless node can estimate the global density by using local information such as the received power from neighbors. In this paper, we propose a cross layer protocol to compute the density estimate. The accuracy of the estimate can be enhanced and its variance can be reduced through cooperation among the nodes. Nodes share the received power measurements with each other. Based on the collected observations, the maximum likelihood estimate is computed. It is shown that cooperative density estimation has better accuracy with less variance than the individual estimation. When nodes share received power measurements from further away neighbors, the variance of the estimate is further reduced.","Title: Impact of antenna directionality and energy harvesting rate on neighbor discovery in EH-IoTs\nYear: 2013\n\nAbstract: Homes, offices and vehicles are getting networked. This will enable context aware, autonomous operation of many support systems that could be controlled remotely. To achieve this there would be a large number of tiny devices - sensors and actuators - which are networked and they are termed generally as Internet of Things (IoT) devices. In future, they will be powered through harvested energy from the ambience to enable perennial lifetime and minimal manual maintenance. Some examples of energy sources are photovoltaic panels and piezoelectric crystals. Several challenges arise due to the nature of sources of energy. One of these challenges is that the devices (nodes) leave and re-enter networks due to fluctuating availability of harvested energy. As a result, the neighbor table maintained at each node changes quite often leading to complications in forming and maintaining routes. In fact initial neighbor discovery (ND) itself is a difficult task. Further, usage of directional antennas would increase the time taken to complete ND. We study such a network through exhaustive simulation study considering various parameters. We demonstrate the benefits and challenges of using directional antennas for ND. We present a scheme that nodes could use to discover their neighbor during initial deployment and another scheme that could be used for subsequent discovery on re-entry into the network. We show that a dedicated ND protocol is necessary for energy harvesting networks and that directional ND is beneficial in these networks under some circumstances.","Title: Performance evaluation of power-aware routing algorithms in Personal Networks\nYear: 2009\n\nAbstract: Personal networks (PN) is a concept in the area of modern communication networks in which devices belonging to a person cooperate to offer services to the person regardless of their geographical locations. A PN may consist of several ad-hoc sub-networks (or clusters) which are interconnected through infrastructure networks such as the Internet. Failure of personal devices due to battery shortage is one of the main threads for service availability in PNs. Power-aware routing (PAR) has been proposed as an effective scheme to prolong the lifetime of battery-operated nodes in wireless ad-hoc networks. This scheme, however, could be utilized in PNs as well, since a PN cluster can be modeled as a wireless ad-hoc network. Aiming at prolonging the lifetime of nodes in PNs, in this paper we propose a PAR algorithm for PNs and we evaluate and compare its performance with the performance of other PAR algorithms proposed so-far for wireless ad-hoc networks. We analyze the effect of node density, routing overhead, heterogeneity of PN nodes in terms of their power supplies, and communication pattern within PN clusters on the performance of the PAR algorithms. Taking into account various parameters, we show that in a PN environment, our proposed scheme can profoundly extend the lifetime of PN nodes compared to the other algorithms. We also show that there can be an optimal value for route refresh interval, as a parameter which affects the generated routing overhead, and the node density of PN clusters, in such a way that the network lifetime is maximized.","Title: Robust 60 GHz Indoor Connectivity: Is It Possible with Reflections?\nYear: 2010\n\nAbstract: In this paper, we investigate the robustness of the 60 GHz connectivity in typical indoor environments by analyzing the outage probability. We define three realistic indoor scenarios which may host the 60 GHz networks in the future and perform simulations with a verified 3D ray tracing tool on them. In the first set of simulations, we show the impact of access unit position on the connectivity. In the next step, we show that increasing the obstacle density linearly decreases the outage probability of the 60 GHz network and this decrease is found to be sharper for certain positions of the access point. In the last step, we demonstrate the direct effect of reflective surface availability on the connectivity. A 60 GHz indoor network relying solely on the reflections in the absence of line-of-sight path is very vulnerable to outages even in moderately populated indoor environments.","Title: CogCell: cognitive interplay between 60 GHz picocells and 2.4/5 GHz hotspots in the 5G era.\nYear: 2015\n\nAbstract: The rapid proliferation of wireless communication devices and the emergence of a variety of new applications have triggered investigations into next-generation mobile broadband systems, i.e. 5G. Legacy 2G-4G systems covering large areas were envisioned to serve both indoor and outdoor environments. However, in the 5G era, 80 percent of all traffic is expected to be generated indoors. Hence, the cu...","Title: An architectural framework for 5G indoor communications\nYear: 2015\n\nAbstract: In this paper, we emphasize on indoor networks in 5G era. We explore the possible technologies and architectural solutions for 5G indoor communications. Owing to the fact that requirements for indoor and outdoor communications will be quite different in the next generation networks, we try to define an architectural framework for 5G indoor communications. Our proposed architecture focuses on three aspects of indoor network architecture, namely; air interface, backhaul connectivity and indoor signal distribution.","Title: Choose Wisely: Topology Control In Energy-Harvesting Wireless Sensor Networks\nYear: 2016\n\nAbstract: Ambient energy-harvesting technology is a promising approach to keep wireless sensor networks (WSNs) operating perpetually. Depending on the harvesting source nodes can either be active (alive) or inactive (dead) at any instant in such Energy-Harvesting WSNs (EH-WSNs). Thus, even in a static deployment of EH-WSNs, the network topology is no longer static. A popular method to increase energy-efficiency in WSNs is by employing topology control algorithms. Most of the topology control algorithms in the literature cannot handle the situation when nodes have different energy-levels, and when number of active nodes varies with time in EH-WSN. To address this issue, we present two localized energy based topology control algorithms, viz., EBTC-1 and EBTC-2. EBTC-1 is for convergecast applications of WSNs and EBTC-2 is for a generic scenario where all nodes are required to be connected. While typical topology control algorithms select a particular number of neighbors, the distinguishing feature of both these algorithms is that they select neighbors based on energy-levels, and render the global topology strongly-connected. Simulation results confirm that EBTC-1 and EBTC-2 reduce the transmission power and they let nodes have neighbors with high remaining energy. Results show that our proposed algorithms increase at least 33% in the remaining energy per neighbor. In addition, in terms of energy consumption and fault-tolerance, our proposed algorithms typically achieve 1-connected topology using 74% less energy compared to K-Neigh.","Title: 60 GHz MAC Standardization: Progress and Way Forward.\nYear: 2015\n\nAbstract: Communication at mmWave frequencies has been the focus in the recent years. In this paper, we discuss standardization efforts in 60 GHz short range communication and the progress therein. We compare the available standards in terms of network architecture, medium access control mechanisms, physical layer techniques and several other features. Comparative analysis indicates that IEEE 802.11ad is likely to lead the short-range indoor communication at 60 GHz. We bring to the fore resolved and unresolved issues pertaining to robust WLAN connectivity at 60 GHz. Further, we discuss the role of mmWave bands in 5G communication scenarios and highlight the further efforts required in terms of research and standardization.","Title: Murphy Loves Ci: Unfolding And Improving Constructive Interference In Wsns\nYear: 2016\n\nAbstract: Constructive Interference (CI) phenomenon has been exploited by Glossy, a mechanism for low-latency and reliable network flooding and time synchronization for wireless sensor networks. Recently, CI has also been used for other applications such as data collection and multicasting in static and mobile WSNs. These applications base their working on the high reliability promised by Glossy regardless of the physical conditions of deployment, number of nodes in the network, and unreliable wireless channels that may be detrimental for CI. There are several works that study the working of CI, but they present inconsistent views. We study CI from a receiver's viewpoint, list factors that affect CI and also specify how and why they affect. We validate our arguments with results from extensive and rigorous experimentation in real-world settings. This paper presents comprehensive insights into CI phenomenon.With this understanding, we improve the performance of CI through an energy-efficient and distributed algorithm. We cause destructive interference on a designated byte to provide negative feedback. We leverage this to adapt transmission powers. Compared to Glossy, we achieve 25% lesser packet losses while using only half of its transmission power.","Title: V2X Application-Reliability Analysis of Data-Rate and Message-Rate Congestion Control Algorithms.\nYear: 2017\n\nAbstract: Intelligent transportation systems (ITS) require vehicle-to-everything communication. In dense traffic, the communication channel may become congested, impairing the reliability of the ITS safety applications. Therefore, the European Telecommunications Standard Institute demands decentralized congestion control (DCC) to control the channel load. Our objective is to investigate whether message-rate or data-rate congestion control provides better application reliability. We compare LIMERIC and PDR-DCC as representatives of the two principles. We analyzed the application reliability of LIMERIC with different data-rates and PDR-DCC with different message-rates for varying traffic densities and application requirements. We observed that, for applications with demanding requirements and in a large variety of vehicular densities, PDR-DCC (data-rate) provides more reliable communication support than LIMERIC (message-rate). Furthermore, this letter hints that a combined message-rate and data-rate congestion control can improve reliability further.","Title: Performance Analysis of IEEE 802.11ad MAC Protocol.\nYear: 2017\n\nAbstract: IEEE 802.11ad specifies a hybrid medium access control (MAC) protocol consisting of contention as well as non-contention-based channel access mechanisms. It also employs directional antennas to compensate for the high free-space path loss observed in 60 GHz frequency band. Therefore, it significantly differs from other IEEE 802.11(b/g/n/ac) MAC protocols and thus requires new methods to analyze its performance. We propose a new analytical model for performance analysis of IEEE 802.11ad employing a 3-D Markov chain considering all the features of IEEE 802.11ad medium access mechanisms including the presence of non-contention access and the different number of sectors due to the use of directional antennas. We show that the number of sectors has a high impact on the network throughput. We show that the MAC packet delay is significantly affected by the duration of the contention period. Our results indicate that a suitable choice of the number of sectors and contention period can improve the channel utilization and MAC delay performance.","Title: Distributed Antenna System for Mitigating Shadowing Effect in 60 GHz WLAN\nYear: 2015\n\nAbstract: The 60 GHz unlicensed frequency band has been adopted to support high data rate WLAN applications. The problem of this frequency band is that the wireless signal is vulnerable to the shadowing of objects. Especially in indoor scenarios, humans may frequently block the signal paths that cause disconnections of devices. This problem can be mitigated by using multiple antennas that can create rich signal paths between the devices. The idea presented in this paper is to employ a distributed antenna system (DAS) architecture enabled by radio-over-fiber technology to alleviate the shadowing problem. In the DAS architecture, multiple remote antenna units (RAUs), each of them equipped with an antenna array, are connected with the same WLAN access points via optical fibers. But there are questions that need to be answered, including the beamforming strategy with multiple RAUs, and the placement and the required number of the RAUs. These questions are related to factors like room size, population density, etc. We discuss these questions in this paper and present our findings through theoretical models and extensive simulations, which can be referred to for practical implementations. We find that the proposed 60 GHz DAS can significantly improve the link connectivity. Even when only two RAUs are used, the signal coverage can be improved to an acceptable percentage in heavy shadowing scenarios.","Title: Coexistence Of Decentralized Congestion Control Algorithms For V2v Communication\nYear: 2018\n\nAbstract: Channel congestion is one of the most critical issues in IEEE 802.11p-based vehicular communications as it leads to the unreliability of safety applications. As a countermeasure, many Decentralized Congestion Control (DCC) algorithms have been proposed. One of the most prominent DCC algorithms is the message-rate based LIMERIC. Recently, algorithms have also been proposed to support higher vehicle density (better scalability). One of such algorithms is the combined message-rate and data-rate based congestion control algorithm (MD-DCC). MD-DCC can support around 2.7 times higher vehicular density than LIMERIC. However, if LIMERIC has been deployed, can MD-DCC be introduced and coexist well with LIMERIC. The objective of this paper is to investigate how MD-DCC coexists with LIMERIC. Given a scenario where vehicles may use either LIMERIC or MD-DCC, we study the impact of coexistence on channel load, fairness and reliability of vehicles at different densities. Simulation studies show that there is no significant degradation of reliability both for LIMERIC and MD-DCC at different densities. On the contrary, coexistence can improve the reliability of LIMERIC vehicles. Furthermore, MD-DCC can support vehicles at large densities even when it coexists with LIMERIC retaining its scalability. However, fair allocation of resources is not guaranteed when LIMERIC and MD-DCC coexist.","Title: Deep Q-network based dynamic power allocation for cell-free massive MIMO\nYear: 2021\n\nAbstract: Numerical optimization has been investigated for decades to address complex problems. Many effective methods, e.g., the weighted minimum mean square error (WMMSE) algorithm, have been developed for a large variety of applications in wireless communication systems. However, these methods often require high computational cost creating a serious gap between theoretical analysis and real-time processi..."],"string":"[\n \"Title: Performance Study Of Mimo-Ofdm Platform In Narrow-Band Sub-1 Ghz Wireless Lans\\nYear: 2013\\n\\nAbstract: Using multiple antennas at the transceivers has become a necessity in high data rate wireless communication systems. Multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) has been adopted as a mandatory modulation scheme in the upcoming IEEE 802.11ah standard. IEEE 802.11ah will specify carrier frequencies for Wireless Local Area Networks (WLANs) operating in the sub-1 GHz band. This is challenging for MIMO-OFDM in sub-1 GHz because of limited bandwidth. Thus building a prototype will provide necessary information to understand this new scenario. We build a prototype in order to test and validate modifications to physical layer (PHY) and media access control (MAC) for narrow-band data transmissions. We present here, in detail, the steps to build a real-time MIMO-OFDM testing platform that is useful for evaluating narrow-band sub-1GHz transmission characteristics. We analyzed potential MIMO-OFDM implementations and conducted extensive measurements on our platform in order to verify the required system enhancements. This paper gives a hands-on account of designing and testing a sub-1 GHz WLAN platform.\",\n \"Title: A Survey on Handoffs - Lessons for 60 GHz Based Wireless Systems.\\nYear: 2012\\n\\nAbstract: Wireless communication is playing an important role in our daily life since it offers flexibility and mobility. New multimedia services demand data-rates of up to hundreds of Mbps and thus higher frequency bands are being explored to support these new high data rate services. However, to support mobility, handoff is a must in many of these networks and systems. In the mean time the 60 GHz band has...\",\n \"Title: Network Resource Awareness and Control in Mobile Applications\\nYear: 2007\\n\\nAbstract: As mobile devices become increasingly capable of simultaneously using various kinds of wireless and fixed networks, developers must match mobile application needs to network resource dynamics. The authors' network resource model and network abstraction layer (NAL) software offer an extensible network resource abstraction to applications running on mobile devices. The NAL service can be used in conjunction with well-known APIs, such as the Socket API, which provide access to these resources; it also accommodates support for mobility-management facilities, such as Mobile IP, if available.\",\n \"Title: A Performance Modeling and Evaluation of the Cambridge Fast Ring\\nYear: 1992\\n\\nAbstract: Performance of the Cambridge Fast Ring (CFR), a high-speed slotted ring with normal slots, is studied. It is shown that the CFR can be represented by a multiqueue multiple cyclic server model with a 1-limited service discipline and with a restriction that only one server at a time can be serving a queue. Exact necessary and sufficient stability conditions are stated. An approximate analytic M/G/1 vacation model in which analysis concentrates on one station while the others are represented by a vacation period is developed to estimate the expected message waiting times. It is shown that the model is accurate and usable over a wide range of parameters. A performance evaluation of the CFR based on this model is presented. The performance is compared to that of a variant which does not restrict the number of slots a station may simultaneously use.\",\n \"Title: Experience-based network resource usage on mobile hosts\\nYear: 2006\\n\\nAbstract: The network resources available on a mobile device may be better employed when taking into account the history of mobility patterns of the user. We propose an experience-based mechanism for network resource usage by applications on mobile hosts. It incorporates a machine learning component that adapts the network resource usage given a high-level metric, typically consisting of system as well as application related sub-metrics.\",\n \"Title: Solving The Incertitude Of Vertical Handovers In Heterogeneous Mobile Wireless Network Using Mdp\\nYear: 2008\\n\\nAbstract: In this paper we introduce and examine via simulations, a Markov Decision Process (MDP) based approach for vertical handover decision-making. We consider here the case of a heterogeneous network environment where both the legacy WLAN and the novel line-of-sight (LOS)-dependent 60GHz radio systems are deployed to support indoor high quality multi-media applications. The MDP decision approach takes into account multiple factors such as user preference, network situation, device capability and effects of environment. We show its ability to effectively make handover decision in uncertain situations. The method and results herein are applicable, in general, to any other situation where such a decision is to be made.\",\n \"Title: Smart Network Interface Selection For E-Dtns\\nYear: 2012\\n\\nAbstract: We implement two energy models that accurately and comprehensively estimates the system energy cost and communication energy cost for using Bluetooth and Wi-Fi interfaces. The energy models running on a system is used to smartly pick the most energy optimal network interface so that data transfer between two end points is maximized.\",\n \"Title: Link quality-based transmission power adaptation for energy saving in IEEE 802.11\\nYear: 2009\\n\\nAbstract: Today, many wireless devices are mobile and battery powered, which means that the service duration is highly dependent on energy consumptions and battery capacities. Based on the fact that battery capacity is still limited, energy saving is an important issue in wireless communication. Many methods have been proposed to save energy in wireless communication, such as power-aware routing and sleep mode. However, another important approach, transmission power adaptation, is not yet well discussed. In this paper, we propose a mechanism that selects the appropriate transmission power level to transmit packets instead of using a fixed default power level for all packets, which is common in IEEE 802.11. Lower transmission power levels may result in more retransmissions, but still consumes less energy. In total, energy can be saved and interference can be reduced by selecting a good transmission power level. We propose a two phase power saving mechanism and implemented it in a real test-bed. To evaluate the performance of our mechanism, we used an indoor environment to carry out measurements. The measurement results indicate that a significant amount of energy can be saved for the transmitter in most scenarios compared to using a fixed transmission power level for all packets.\",\n \"Title: A System Perspective on Cognition for Autonomic Computing and Communication\\nYear: 2005\\n\\nAbstract: In this paper we present a conceptual view on the incorporation of cognitive processing capabilities in future generation computer systems. We argue that cognition is at the heart of autonomic behavior, and therefore a necessary ingredient for autonomic computing and communication. We introduce a bio- inspired cognitive engine that interacts with and has control over major operating system components, and showcase, based on scenario descriptions, how communicating applications take advantage of this setup by adapting and autonomously reacting to changes in heterogeneous and widely varying network resources. As a conceptual work, this paper does not cover experiences with the implementation of the introduced concepts, nor does it describe experimental results.\",\n \"Title: Event-Driven MAC Protocol for Dual-Radio Cooperation.\\nYear: 2012\\n\\nAbstract: One of the sources of the energy waste in wireless sensor networks is idle listening, the time in which a node monitors the free channel. In applications where the events occur sporadically, energy consumption due to idle listening can be further reduced by dual-radio cooperation. In dual-radio cooperation, nodes in the network have two stacks. One stack makes use of a low-power wake-up radio for event-driven communication over the main radio. The other stack may employ any sensor networking medium access control protocol only over the main radio. One of the two stacks can be dynamically operational depending on the rate of the events or the packet arrival rate. When the event rate becomes small, the event-driven stack takes over the operation. If the event rate increases, it could be more efficient to operate the legacy single radio stack. In this paper, we investigate the performance of the dual-radio cooperation in wireless sensor networks. A medium access control protocol is proposed for the dual-radio cooperation to maximize the energy efficiency of the wireless network. We define the critical event rate as the event rate threshold above which the single-radio stack performs better than the dual-radio stack. We analyzed and validated the critical event rate by simulations. We show that 70-97% energy conservation is possible by employing the dual-radio cooperation.\",\n \"Title: Energy efficiency and channel allocation in P2PWRAN\\nYear: 2013\\n\\nAbstract: IEEE 802.22 is proposed for reusing the TV channels with cognitive radio technology to build Wireless Regional Area Networks (WRANs). The cellular topology of a WRAN makes the spectrum management easy. However, this topology reduces the network capacity significantly, because in one slot only one user can be allocated with a channel. Therefore, Peer-toPeer WRAN (P2PWRAN) is proposed, in which direct CPE(Consumer Premises Equipment)-to-CPE intra-cell communication is supported. While increasing the network capacity significantly, reducing the total energy consumption is a challenge.The energy efficiency of nodes in the standard for WRAN is difficult to improve, since the transmission distance of nodes to the BS is relatively static. P2PWRAN enables to achieve higher energy efficiency through better possibilities of channel allocation. Channel allocation influences energy consumption and network capacity significantly. Channel allocation also affects the upper layers, for example the routing protocol. Therefore, guaranteeing the network performance and fairness amongst CPEs, while achieving green networks, needs cross-layer design. In this paper, we define the channel allocation problem in P2PWRAN and we recognize it as a Multi-objective Quadratic Programming (MOQP) problem. An Energy-aware Channel Allocation (ECA) algorithm is proposed for exploiting the P2PWRAN paradigm to achieve the lower communication energy consumption and higher network energy efficiency and throughput than simple greedy policies. ECA jointly considers the network layer and Media Access Control (MAC) layer. By controlling the parameters in ECA, the fairness of channel allocation can also be guaranteed.\",\n \"Title: Evaluation of MPEG-4 Video Streaming over UMTS/WCDMA Dedicated Channels\\nYear: 2005\\n\\nAbstract: Universal Mobile Telecommunications System (UMTS) is athird-generation mobile communication system that supportswireless wideband multimedia communications. This paperinvestigates the performance of streaming MPEG-4 videoover UMTS dedicated channels with varying channel conditions.In order to evaluate the performance, simulation models for UMTSand MPEG-4 were developed. Simulation results show that UMTSRadio Control Link (RLC) protocol operating in the acknowledgedmode outperforms the unacknowledged mode. The latter modeprovides timely delivery but no error recovery. The acknowledgedmode can deliver acceptable video quality for RLC block error ratesup to 40% utilizing a pre-decoder buffer. We also discover thatRLC block size can affect the video quality with a larger RLC block size attaining better quality. However, the optimum RLC blocksize is dependent on the encoded video frame size and the channelbit rate. Therefore, the RLC block size can be tuned to a proper value given the encoded video frame size and channel bit rate.\",\n \"Title: Predicting mobility events on personal devices.\\nYear: 2010\\n\\nAbstract: High-end mobile phones are quickly becoming versatile sensing platforms, capable of continuously capturing the dynamic context of their owners through various sensors. A change in this context is often caused by the fact that owners-and therefore the devices they carry-are moving from one place to another. In this paper, we model the sensed environment as a stream of events, and assume, given that people are creatures of habit, that time correlations exists between successive events. We propose a method for the prediction in time of the next occurrence of an event of interest, such as 'arriving at a certain location' or 'meeting with another person', with a focus on the prediction of network visibility events as observed through the wireless network interfaces of the device. Our approach is based on using other events in the stream as predictors for the event we are interested in, and, in the case of multiple predictors, applying different strategies for the selection of the best predictor. Using two real-world data sets, we found that including predictors of infrequently occurring events results in better predictions using the best selection strategy. Also, we found that cross-sensor (cross-interface) information in most cases improves the prediction performance.\",\n \"Title: Fairness in Wireless Networks:Issues, Measures and Challenges\\nYear: 2014\\n\\nAbstract: The pervasiveness of wireless technology has indeed created massive opportunity to integrate almost everything into the Internet fabric. This can be seen with the advent of Internet of Things and Cyber Physical Systems, which involves cooperation of massive number of intelligent devices to provide intelligent services. Fairness amongst these devices is an important issue that can be analysed from several dimensions, e.g., energy usage, achieving required quality of services, spectrum sharing, and so on. This article focusses on these viewpoints while looking at fairness research. To generalize, mainly wireless networks are considered. First, we present a general view of fairness studies, and pose three core questions that help us delineate the nuances in defining fairness. Then, the existing fairness models are summarized and compared. We also look into the major fairness research domains in wireless networks such as fair energy consumption control, power control, topology control, link and flow scheduling, channel assignment, rate allocation, congestion control and routing protocols. We make a distinction amongst fairness, utility and resource allocation to begin with. Later, we present their inter-relation. At the end of this article, we list the common properties of fairness and give an example of fairness management. Several open research challenges that point to further work on fairness in wireless networks are also discussed. Indeed, the research on fairness is entangled with many other aspects such as performance, utility, optimization and throughput at the network and node levels. While consolidating the contributions in the literature, this article tries to explain the niceties of all these aspects in the domain of wireless networking.\",\n \"Title: A Link Stability Model for Indoor 60GHz Radio Wireless Networks\\nYear: 2009\\n\\nAbstract: With the multitude of mobile communication devices supporting users, it is natural to expect that these devices would work ubiquitously anywhere and any time with high data rate support. 60 GHz is an obvious choice for the high data rate indoor communication. In this paper we present an holistic analytical model of 60 GHz link stability for indoor wireless networks. The model is built considering the indoor channel characteristics of 60 GHz radio, the antenna configuration and the mobility of persons. Also, we take an example to show the relevance of our model and its applicability in numerical evaluation.\",\n \"Title: Impact of Antenna Pattern and Link Model on Directional Neighbor Discovery in 60 GHz Networks\\nYear: 2011\\n\\nAbstract: 60 GHz radio is a very attractive technology for short-range wireless communication due to its capability to provide Gbps data rate. To address the neighbor discovery (ND) problem in 60 GHz networks, we propose a novel analytical framework to investigate the ND performance. The main difficulty in modeling the ND process in a 60 GHz network is the involvement of the directional antennas with gain differences between the antenna's main lobe and side lobes. Different antenna modes - directional or omni-directional - coupled with different ND mechanisms make the analytical study demanding. In this article, we propose a comprehensive theoretical model to demonstrate the performance of ND processes using one-way ND and handshake-based ND mechanisms. Moreover, we combine them with different antenna modes, i.e., directional transmitting with omni-directional listening (DO) and directional transmitting with directional listening (DD) modes. The impact of antenna modes on the ND process is analyzed. Since 60 GHz radio is prone to co-channel interference, we examine a realistic interference-aware link model and antenna pattern via simulation studies. Our work is specifically beneficial to provide guidelines for applying directional neighbor discovery process within 60 GHz wireless networks.\",\n \"Title: Performance analysis of the link layer protocol for UWB impulse radio networks\\nYear: 2006\\n\\nAbstract: The Self-organizing Device discovery and Data transmission protocol (SDD) is a novel self-organizing link layer protocol, that was proposed for UWB impulse radio based networks. By means of simulations, it has been indicated that the SDD protocol enables a node to discover most of its neighboring nodes in short discovery time, in the mean time, appropriately avoids collisions. Further evaluations are carried out in this paper by creating a new analytical model to investigate the theoretical relations among the discovery time, the discovery ratio and the system parameters. It can be concluded that the SDD protocol is an efficient, scalable and well performed protocol\",\n \"Title: Toward A Seamless Communication Architecture For In-Building Networks At The 60 Ghz Band\\nYear: 2006\\n\\nAbstract: This paper addresses the issues of designing an infrastructure to support seamless in-building communication at the 60 GHz band. Recently, the 60 GHz band has received much attention due to its 5 GHz of available spectrum. However the propagation of signals in this band is strongly hindered by attenuation and line-of-sight requirements. The situation gets worse in the in-building environment where signal propagation is obstructed by physical objects such as walls, furniture etc. In this paper, we present a novel Radio over Fiber (RoF) architecture that is cost-effective and is able to deliver high data-rate of the order of gigabits. To ensure a seamless communication environment at the 60 GHz band, we propose the concept of Extended Cells (EC) in order to create sufficient overlap areas. Finally, we illustrate the effectiveness of the proposed architecture by simulating an in-building network at the 60 GHz band employing the RoF and EC concepts.\",\n \"Title: Communication Context for Adaptive Mobile Applications\\nYear: 2005\\n\\nAbstract: This paper describes our ongoing work on a system that provides information about network resources and related communication facilities to applications running on a mobile device. It defines and elaborates on the concept of communication context and describes the architecture of this core low-level enabling system for ubiquitous and pervasive applications operating in a mobile environment.\",\n \"Title: Exploring Multipath Capacity for Indoor 60 GHz Radio Networks\\nYear: 2010\\n\\nAbstract: We study the application of multipath routing to enhance reliability of high quality video in 60 GHz radio indoor networks. We propose a cross-layer approach taking into account the possibility of layered-coding for video stream transportation and the characteristics of 60 GHz radio links based on our earlier studies on the stability of 60 GHz radio links. Further three traffic allocation schemes have been investigated and compared for the best use of multipath capacity.\",\n \"Title: Architectural and QoS Aspects of Personal Networks\\nYear: 2006\\n\\nAbstract: Personal networks (PNs) are future communication systems that combine wireless and infrastructure based networks to provide users a variety of services anywhere and anytime. PNs introduce new design challenges due to the heterogeneity of the involved technologies, the need for self-organization, the dynamics of the PN composition, the application-driven nature, the co-operation with infrastructure-based networks, and the security hazards. This paper discusses the challenges of security, service discovery and QoS provisioning in designing self-organized PNs and combines them all into an integrated architectural framework\",\n \"Title: A framework for gateway and access network selection in ad-hoc distributed personal networks\\nYear: 2010\\n\\nAbstract: A personal network (PN) is a network of personal devices regardless of their geographical locations. A PN may consist of several ad-hoc sub-networks, each of which is referred to as a cluster. Different clusters are connected to each other through their gateways and using interconnecting structures, e.g., the Internet. Inside each cluster, there might be several nodes that are able to act as gateway. Gateway nodes may support different access technologies, such as WiFi, Ethernet, WiMax, etc. It also may happen that a single gateway, which supports several access technologies, is able to connect to different access networks. To benefit from the availability of different gateways and access networks with different characteristics for a PN user, an efficient methodology for gateway and access network selection is needed. In this work, we are motivated to design a generic architectural framework for selecting the best gateway(s) and access network(s) in a PN. Issues, such as characteristics of the gateway nodes and access networks, QoS requirements of the applications, mobility of gateway nodes, dynamism of the PN topology, etc., have been considered in designing the proposed framework. The main advantages of this framework is leveraging the benefits of different alternative solutions, in terms of overhead, efficiency, complexity, and user satisfaction.\",\n \"Title: Analysis of the Performance Boundaries of Sub-1 GHz WLANs in the 920MHz ISM-Band.\\nYear: 2013\\n\\nAbstract: We explore the performance boundaries of sub-1 GHz Wireless Local Area Networks (WLANs) and summarize the limitations of our proposed narrow-band WLAN system that uses carrier frequencies at 920 MHz with 1 MHz channel bandwidth. While there are several evaluation reports on WLAN testbeds available, no report is available that discusses the boundaries of narrow-band multiple-input multiple-output Orthogonal Frequency Division Multiplexing (MIMO-OFDM) WLANs in sub-1 GHz. Even though 2.4 GHz/5 GHz MIMO-OFDM WLANs have been widely investigated, the characteristics of narrow-band WLAN is different from that of existing studies. This is mainly the result of limited channel bandwidth at lower frequency bands. This paper covers both, the fundamental performance boundaries for sub-1 GHz WLANs, and measurement results of our proposed software-based 2x2 MIMO-OFDM sub-1 GHz WLAN system operating at 920 MHz in order to identify the performance regimes, when using transmission protocols, including ICMP, UDP, and TCP.\",\n \"Title: Adaptive CCA for IEEE 802.15.4 Wireless Sensor Networks to Mitigate Interference\\nYear: 2010\\n\\nAbstract: IEEE 802.15.4 Wireless Sensor Networks (WSNs) share the 2.4 GHz Industrial, Scientific, and Medical (ISM) license-free band with many other wireless technologies such as IEEE 802.11b/g Wireless Local Area Networks (WLANs). Because of the low-power, however, IEEE 802.15.4 WSNs are potentially vulnerable to the interference introduced by the other wireless technologies such as IEEE 802.11b/g WLANs, which have much higher power. Particularly, in the presence of heavy interference, IEEE 802.15.4 nodes can hardly get a chance to access the channel, which could result in discarding a large amount of packets. In this paper, we propose a decentralized approach to help IEEE 802.15.4 nodes mitigate interference. By adaptively and distributively adjusting Clear Channel Assessment (CCA) thresholds of IEEE 802.15.4 nodes in the presence of heavy interference, the approach can substantially reduce the amount of discarded packets due to channel access failures, and therefore significantly enhance the performance of IEEE 802.15.4 WSNs. The approach is robust, responsive and easy to be implemented at a low cost. The effectiveness of the approach is validated by OPNET simulation.\",\n \"Title: Fairness and network capacity trade-off in P2P IEEE 802.22 networks\\nYear: 2012\\n\\nAbstract: Wireless Regional Area Networks(WRAN), IEEE 802.22 standard, adopts a centralized cellular topology including base station (BS) and customer-premise equipments (CPEs). However this may lead to limited network capacity, since every CPE needs to communicate to the BS and only one channel can be used to communicate in the whole cell per time slot with one BS antenna. Peer to Peer WRAN (P2PWRAN) was proposed [3] to circumvent this, where CPEs can communicate with each other directly. P2PWRAN increases network capacity compared to the standard WRAN because of spatial reuse. Multiple communication channels can be simultaneously allocated and reused in one time slot by the BS and multiple CPEs. In IEEE 802.22b [4], a standard in progress, peer to peer communication plays an important role for enabling smart grids. Channel allocation influences the network performance significantly, therefore, we formulate the spectrum allocation problem in P2PWRAN as a quadratically constrained programming (QCP) problem. It is also proved as a computationally hard problem. The Greedy Coloring Algorithm (GCA) is examined in P2PWRAN channel allocation, and the results show that it may cause severe unfairness in allocation of channels. Thus a Fair Greedy Coloring Algorithm (FGCA) is proposed to guarantee a fair allocation by queuing flows considering previous allocations. However, FGCA guarantees fairness but leads to decrease in performance of P2PWRAN. Therefore, a Trade-off FGCA (TFGCA) is proposed considering fairness and network performance at the same time during allocation. Simulation results show that with the adjustment of two factors in TFGCA, network performance and fairness can be balanced.\",\n \"Title: Coexistence Performance of IEEE 802.15.4 Wireless Sensor Networks Under IEEE 802.11b/g Interference\\nYear: 2013\\n\\nAbstract: IEEE 802.15.4 Wireless Sensor Networks (WSNs) and IEEE 802.11b/g Wireless Local Area Networks (WLANs) are often collocated, causing a coexistence issue since these networks share the same 2.4GHz Industrial, Scientific, and Medical band. In our previous work, we built a coexistence model of IEEE 802.15.4 WSNs and IEEE 802.11b/g WLANs. By identifying three distinct coexistence regions, the model explained the coexistence behavior of IEEE 802.15.4 WSNs and IEEE 802.11b/g WLANs, and the model was experimentally validated. In this paper, we improve the model by introducing two important implementation factors: the transceiver's Rx-to-Tx turnaround time and the Clear Channel Assessment partial detection effect. The enhanced model significantly improves the accuracy on explaining and predicting the coexistence performance of IEEE 802.15.4 WSNs in the real-life environment. Furthermore, under the guidance of the model, the coexistence performance of IEEE 802.15.4 WSNs is extensively investigated in various coexistence scenarios by analysis, simulation and experiments, respectively. The simulation and experimental results agree with our analysis. The coexistence model is believed to be helpful in resolving the coexistence issue.\",\n \"Title: Maximizing the Fair Allocation of Opportunistic Spectrum for CR Ad Hoc Networks.\\nYear: 2011\\n\\nAbstract: Cognitive Radios (CRs) address spectrum scarcity and under-utilization of the spectrum. Allocating spectrum to CR nodes in a network is neither easy nor straight-forward since spectrum availability is space and time-varying. The Link layer always sets up communication between nodes in a CR ad hoc network by allocating the available spectrum suitably. In a time-slotted model, allocation of channels amongst CR nodes is a NP complete problem. Moreover, since the availability of spectrum is time varying, all the nodes in the network may not be allocated sufficient resources. Thus there is always a possibility that a node may starve compared to its counterparts. In this article, we address the issue of fairness in a CR ad hoc network and propose novel distributed heuristics to allocate spectrum fairly. In literature, max-min, max-sum and max-proportional fairness based schemes are proposed to provide fairness in these CR ad hoc networks. We compare our heuristic results with these schemes, and discuss why our method is significantly better.\",\n \"Title: Beam switching support to resolve link-blockage problem in 60 GHz WPANs\\nYear: 2009\\n\\nAbstract: In this paper, we propose a solution to resolve link blockage problem in 60 GHz WPANs. Line-of-Sight (LOS) link is easily blocked by a moving person, which is concerned as one of the severe problems in 60 GHz systems. Beamforming is a feasible technique to resolve link blockage by switching the beam path from LOS link to a Non-LOS (NLOS) link. We propose and evaluate two kinds of Beam Switching (BS) mechanisms: instant decision based BS and environment learning based BS. We examine these mechanisms in a typical indoor WPAN scenario. Extensive simulations have been carried out, and our results reveal that combining angle-of-arrival with the received signal to noise ratio could make better decision for beam switching. Our work provides valuable observations for beam switching during point-to-point communication using 60 GHz radio.\",\n \"Title: Successive interference cancellation in the uplink of Radio-over-Fiber systems\\nYear: 2013\\n\\nAbstract: Successive Interference Cancellation (SIC) is a well-known technique for enhancing wireless channel efficiency by allowing concurrent transmissions from multiple users. This multiuser detection technique is usually used in cellular systems. In this paper we adapt it to Radio-over-Fiber (RoF) system for the same purpose. Due to the similarity to using SIC in typical cellular networks, we mainly focus on exploring the uniqueness of applying SIC in a distributed antenna application of RoF for uplink communication. We find that the physical channel capacity is decreased by additional noise in distributed antennas, but the MAC layer throughput could still be increased due to simultaneous transmission. We show that the performance can be further improved by employing transmit power control at the terminals and amplification control at the remote access units. All these findings make us believe that SIC is a promising technique in RoF system.\",\n \"Title: Quantitative evaluation of scalability in broadband intelligent networks\\nYear: 1998\\n\\nAbstract: Scalability is the ability of a network to maintain the quality of service while increasing certain parameters relating to the size of the network, such as the number of users, the number of network nodes, the number of services provided, geographical spread, etc. In the design of a B-IN signalling system, network scalability is an important issue that must be taken into account. In this paper we use simulation to investigate scalability issues related to a Broadband Intelligent Network (B-IN), such as that being considered in the ACTS project INSIGNIA. In particular, we study the impact of processor speed and configuration (in B-IN physical entities) on signalling performance. As signalling performance measures we consider the mean call setup delay of a B-IN service request and the network throughput. For Broadband Virtual Private Network (B-VPN) service, we perform scalability experiments by increasing some of the network parameters such as the number of users and the number of nodes.\",\n \"Title: A Novel Approach for the Link Layer in Impulse-based UWB Ad Hoc Networks\\nYear: 2007\\n\\nAbstract: Ultra Wide Band (UWB) impulse radio, promises to be suitable for short-range, low-power, low cost and high data rate applications. While most UWB research is concentrating on the physical layer, little research has been published on the link layer. The fundamental operations in the link layer need to take into account the specifics of impulse radio and also the particular features of ad hoc networks. A novel self-organizing link layer protocol based on time hopping impulse radio, called SDD, was proposed by the authors. This protocol is a collision-free mechanism that enables the devices to discover neighboring nodes and arrange the access to communication resources shared among the nodes. In this paper, some issues related to the self-organizing link layer based on UWB impulse radio are investigated and addressed. The SDD protocol is further developed and specified in detail. The simulations are carried out using GloMoSim simulation environment. An SDD module has been developed and embedded in the simulator. Results show that the SDD protocol can work properly and efficiently in a single-hop ad hoc network.\",\n \"Title: Mobility modeling for personal networks: properties and its impact\\nYear: 2010\\n\\nAbstract: User centered Personal Network (PN) interconnects various devices belonging to a user in different geographic locations, such as home, office, car, etc., to form one network to improve the quality of life of the user. In PNs, the user's devices move in different groups, where these groups merge and split. A PN Mobility Model (PNMM) has been proposed to model the devices' movements in PNs. In this paper, PNMM has been evaluated: stability evaluation proves that PNMM maintains a long-term steady state. The impact of PNMM on the performance of protocols designed for PNs has been discussed in this paper. Simulation results give insights of the performance of the PN Clustering Protocol, and provide guidelines for future design of clustering in PNs.\",\n \"Title: An Intra-Cell Peer To Peer Protocol In Ieee 802.22 Networks\\nYear: 2011\\n\\nAbstract: IEEE 802.22 mostly considers a typical cellular network with centralized control. Therefore, all intra-cell communications have to be via base stations. It not only leads to extra power consumption for intra-cell communications, but also only one channel can be used per time slot in the whole cell. This may lead to decrease in the network performance. Therefore, we propose a protocol called intra-cell peer to peer communication protocol (ICP2P) based on IEEE 802.22 protocol. The main accomplishments of ICP2P are to support direct intra-cell peer to peer communications between Customer Premises Equipments (CPEs), to allocate multiple channels per time slot in a cell, and to re-allocate same channels in one time slot. The network performance, such as capacity, channel usage and energy consumption, are significantly improved by ICP2P. Additionally, a weighed demand queue algorithm (WDQA) is proposed to guarantee the fairness in channel access opportunities for CPEs.\",\n \"Title: Performance modelling of the Orwell basic access mechanism\\nYear: 1987\\n\\nAbstract: Orwell is a high speed slotted ring. Its protocol uses destination release of the slots. Because of this the carried load can be much larger than the transmission rate. A new analytical model of the Orwell basic access mechanism is presented in this paper. The model shows to be accurate and usable over a wide range of parameters. The performance analysis of the Orwell basic access mechanism is presented.\",\n \"Title: Performability modelling tools and techniques\\nYear: 1996\\n\\nAbstract: Over the last decade considerable effort has been put in the development of techniques to assess the performance and the dependability of computer and communication systems in an integrated way. This so-called performability modelling becomes especially useful when the system under study can operate partially, which is for instance the case for fault-tolerant computer systems and distributed systems. Modelling techniques are a fundamental prerequisite for actually doing performability analysis. A prerequisite of a more practical but not less important nature is the availability of software tools to support the modelling techniques and to allow system designers to incorporate the new techniques in the design process of systems. Since performability modelling requires many aspects of a system to be specified, high requirements should be posed on performability modelling tools. Moreover, these tools should be structured such that the models can be specified at a level that is easy to understand for a system designer, and that the mathematical aspects are hidden as much as possible. The output of the tool should also be such that it can be understood with only limited knowledge of the underlying mathematical model. We have developed a new, fairly general modelling tool framework that can be used as a guide to assess the usability and structure of performability modelling tools. After briefly reviewing the mathematical aspects of performability modelling we discuss this framework. We then discuss 12 recently developed tools (Metaphor, Numas, Metasan, Metfac, Save, Sharpe, SPNP, Tangram, Penpet, UltraSAN, Surf-2, DyQNtool + ) that can all be used for some aspects of performability modelling and analysis. We assess among other things their structure, their capabilities in terms of measures that can be obtained, and the used modelling formalism. We also discuss directions for future work in the field of performability modelling tools.\",\n \"Title: A study on wireless sensor network based indoor positioning systems for context-aware applications.\\nYear: 2012\\n\\nAbstract: The newer context-aware applications require many inputs and amongst them the location information is one of the most important. In the near future, we see the potential in using wireless sensors deployed inside buildings to support in generating the location information besides their other routine tasks. This paper records the efforts involved in designing and prototyping a centralized indoor positioning system for tracking a target's position. We present an in-depth discussion on the RSSI based positioning algorithms both, range-based and range-free. Considering that the signal strength can be distorted heavily due to multi-path and shadowing, we have proposed the weighing schemes to leverage the credibility of the measured RSSI values. For online tracking, we have also proposed boundary selection and local grid scanning to lower the searching time, and the RSSI data dissemination and collection schemes to reduce traffic overhead. Evaluations have been done in both the field measurements and with an RSSI generator. The generator has been developed to simulate and replace the real measurements for the ease of algorithm design and testing, thus avoiding repetitive field measurements. The results show that positioning systems with adequate accuracy can be built with our proposed schemes. We expect that these proposed schemes to be integrated in multitude of systems with context-aware applications, which use location information. Copyright (C) 2010 John Wiley & Sons, Ltd.\",\n \"Title: The concept of fairness and a GFC protocol\\nYear: 1992\\n\\nAbstract: Multiple access to a single user-network interface (UNI) is a cost-effective solution for the provision of B-ISDN services in customer premises. In a multiple access configuration, a fairly shared access must ensure that each terminal will get a guaranteed bandwidth to provide an acceptable quality of service (QoS) and that the allocation of the remaining bandwidth will result in the same improvement of QoS for all the terminals. An approach to allocating the remaining bandwidth is proposed. It is based on both the guaranteed bandwidth required by the terminal and the probability that the terminal's bit rate exceeds its guaranteed bandwidth. The basic generic flow control (GFC) protocol mechanism and the GFC field settings for a dual-bus customer premises network (CPN) are described\",\n \"Title: Distributed Heuristics for Allocating Spectrum in CR Ad hoc Networks\\nYear: 2010\\n\\nAbstract: Cognitive Radios (CRs) address spectrum scarcity and under-utilization of the spectrum. However, realizing a CR network is neither easy nor straight-forward. In particular the link layer should facilitate setting up of communication between nodes enabling sharing of channels, while maximizing the spectrum utilization. We address the problem of allocation of channels to CR nodes. For time-slotted model, allocating the channels among CR nodes is an NP complete problem. We propose novel distributed heuristics, one based on clique sets called Clique Based Heuristic and the second one is a low complexity, locally adapting Localized Heuristic; both use local control channel. We compare the results of these heuristics with the optimal and a distributed graph coloring algorithm proposed in the literature. We also compare the spectrum utilization and overheads for all these algorithms.\",\n \"Title: Density estimation for out-of-range events on personal mobile devices\\nYear: 2008\\n\\nAbstract: Over the years, personal mobile devices have obtained increasing capabilities to concurrently connect to surrounding networks and nearby other devices. Good knowledge on the dynamics in the availability of these heterogeneous entities constitutes essential input for various data communication problems, ranging from the adaptation of applications on mobile devices to multi-homed situations, to the optimization of routing protocols for delay tolerant networks. In this paper, we focus on a method for the prediction in time of the loss in visibility of currently in-range network entities, as observed on a personal mobile device. We are interested in estimating the full probability density function of the time of these out-of-range events, as this allows us to ask arbitrary questions such as: what is the probability of losing connection X in the next Y minutes? To do so, we model the mobility of the user by applying kernel density estimation on previously observed mobility traces collected during a user experiment we ran with 12 participants in a six week period, logging cellular, 802.11 wireless LAN, Bluetooth, and various other events on the participant's mobile device.\",\n \"Title: Collaborative and Cognitive Network Platforms: Vision and Research Challenges\\nYear: 2011\\n\\nAbstract: In this paper, we present a visionary concept referred to as Collaborative and Cognitive Network Platforms (CCNPs) as a future-proof solution for creating a dependable, self-organizing and self-managing communication substrate for effective ICT solutions to societal problems. CCNP creates a cooperative communication platform to support critical services across a range of business sectors. CCNP is based on the personal network (PN) technology which is an inherently cooperative environment prototyped in the Dutch Freeband PNP2008 and the European Union IST MAGNET projects. In CCNP, the cognitive control plane strives to exploit the resources to better satisfy the requirements of networked applications. CCNP facilitates collaboration inherently. Through cognition in the cognitive control plane, CCNP becomes a self-managed substrate. The self-managed substrate, in this paper, is defined as cognitive and collaborative middleware on which future applications run without user intervention. Endemic sensor networks may be incorporated into the CCNP concept to feed its cognitive control plane. In this paper, we present the CCNP concept and discuss the research challenges related to collaboration and cognition.\",\n \"Title: FEW-PNets - A framework for emulation of wireless personal networks\\nYear: 2008\\n\\nAbstract: With the advent of miniaturization and higher computing capabilities of wireless devices, there is an exponential rise in the number of devices such as personal digital assistants, etc. Application development for these devices is the main focus for many device manufacturers and software developers. The applications developed need to be tested and tried before releasing it to the market. There are cross-compilers to produce the binary files of the software applications that can run on these devices. However, it is rather difficult for developers to test it on the actual devices every time a change is made in the program. Further, a wireless network infrastructure should also be in place for testing. To expedite application development FEW-PNets — a Framework for Emulation of Wireless Personal Networks is proposed. It is a networked application to emulate a Personal Network (PN) owned by a person. It supports emulation of the most functions of PNs. We explain our model by emulating simple PN devices. In this article we elucidate the concepts, advantages, application domains and limitations of FEW-PNets.\",\n \"Title: A novel link quality assessment method for mobile multi-rate multi-hop wireless networks\\nYear: 2010\\n\\nAbstract: Accurate and fast wireless link quality assessment (LQA) for wireless channels would bring huge benefits for mobile multi-hop and multi-rate wireless ad hoc and sensor networks in the form of improved end-to-end performance. In this paper, we propose a novel LQA method based on cross-layer information. The method is implemented in a real test-bed, which is based on IEEE 802.11b/g, and achieved a significant LQA improvement up to 50% in mobile scenarios without introducing overhead. The effectiveness of accurate and fast LQA is demonstrated by feeding it into the routing layer to enable the route decisions to adapt faster during changing situations, especially in mobile scenarios. The experiment results show that the proposed LQA method can lead to faster and smarter routing decisions and higher end-to-end throughput compared to traditional methods, which only use hello packets to determine the link quality in mobile scenario.\",\n \"Title: A multipath approach for reliable high quality video transport in indoor 60 GHz radio networks\\nYear: 2010\\n\\nAbstract: The emerging high definition in-home multimedia applications require higher capacity from future wireless networks in the indoor environments. In this paper, we consider a novel mesh-like architecture for indoor wireless networks using 60 GHz radio for supporting multimedia applications with sufficient data rate and stringent quality-of-service (QoS) guarantees. We concentrate on studying the application of multipath routing to enhance reliability of transporting high quality video since 60 GHz transmission is highly dependent on line-of-sight (LOS). Based on our previous study on the stability of 60 GHz radios, we propose a cross-layer approach taking into account layered coding technique for video stream transportation. We present our model along with an analysis to assess multipath routing in this scenario. Further, we give an in-depth discussion on two traffic allocation schemes to harness multipath capability.\",\n \"Title: Performance analysis of the frame aggregation mechanisms in IEEE 802.15.3c\\nYear: 2009\\n\\nAbstract: To achieve ultra high data rate communications in 60 GHz Wireless Personal Area Networks (WPANs), frame aggregation is devised in IEEE 802.15.3c to increase the MAC transmission efficiency. In this work, we propose a novel statistical model to analyze the behavior of two types of frame aggregation mechanisms: standard frame aggregation and low latency frame aggregation. Through our model, we could fully understand the performance of the frame aggregation mechanisms in terms of transmission delay and system capacity. The accuracy of our model is validated by extensive simulations. Our work provides valuable insights for ultra high performance transmission using 60 GHz radio.\",\n \"Title: Procedure to build interference map in peer to peer IEEE 802.22 networks\\nYear: 2012\\n\\nAbstract: Peer to peer wireless regional area network (P2PWRAN) is proposed as an extension to support peer to peer communication based on IEEE 802.22 [1]. Multiple channel allocation and reuse of channels in the same time slot in P2PWRAN significantly increase the network capacity compared to standard IEEE 802.22 networks. One of the key issues in bringing P2PWRAN into reality is building the interference map. The concept of interference map/mapping has been mentioned in literature however, how to build an interference map is still an open issue. Therefore, we propose a simple and self-adapting interference map building protocol (SIMBP) for P2PWRAN, which can also be used in other multi-channel wireless networks with minor modifications. The simulation results show that SIMBP converges under the P2PWRAN setting, and with number of available channels and growing number of nodes, the capacity of the network is reached eventually.\",\n \"Title: Ad hoc networking in future wireless communications\\nYear: 2003\\n\\nAbstract: This paper contains an overview of the discussions on future research directions within the subject of ad hoc networking, held at the Wireless World Research Forum meetings during 2001. Ad hoc networking is an emerging research field where ad hoc networks are no longer viewed as stand-alone groups of wireless terminals. On the contrary, ad hoc networks are expected to become fundamental in the future development of infrastructure networks and they will be the basis for enabling ubiquitous communications. Ad hoc networking involves new research issues at all layers.\",\n \"Title: MAC Protocol Design for Impulse Radio UWB Based WPANs\\nYear: 2007\\n\\nAbstract: Ultra Wide Band (UWB) has recently gained great interest for high data rate short-range WPANs. Designing a MAC protocol for UWB WPANs must take into account the physical layer characteristics. Due to large bandwidth and spreading spectrum, a time hopping impulse radio UWB (IR-UWB) system can accomplish multiple access by using code division multiplexing. Unlike most existing MAC protocols relying on contention access to limit transmissions, the MAC design for impulse radio intends to accommodate simultaneous transmissions. In this paper, we newly propose a MAC scheme for IR-UWB WPANs. This scheme is a hybrid CDMA and TDMA approach based on time hopping spread spectrum and the timing format of IEEE 802.15.3 MAC. The simulation results show that the proposed approach outperforms the existing protocol under our specifications and WPAN scenarios.\",\n \"Title: A Self-Organizing Link Layer Protocol for UWB Ad Hoc Networks\\nYear: 2004\\n\\nAbstract: Ultra Wide Band (UWB) impulse radio is a promising technology for future short-range, low-power, low cost and high data rate ad hoe networks. The technology is being explored in a number of research projects. While most UWB research for this class of networks is concentrating on the physical layer, little research has been published on link layer protocols which exploit the specifics of UWB impulse radio. In this paper, we focus on the self-organization concept and the peculiarities of UWB technology from a physical and a link layer point of view. A novel self-organizing link layer protocol based on time hopping spread spectrum is proposed in this paper. This protocol promises to be an efficient and collision-free mechanism that enables the devices to discover neighbor nodes and arrange the access to communication resources shared among the nodes. The adjustable parameters of the protocol enable the network to adapt to a dynamic environment.\",\n \"Title: Analysing IEEE 802.15.3c protocol in Fi-Wi hybrid networks.\\nYear: 2013\\n\\nAbstract: Even though 60 GHz frequency band has limited coverage, due to the availability of higher bandwidth worldwide (unlicensed, approximately 5 GHz between 57-62 GHz), it is one of the promising candidates for future broadband employing both Fiber and Wireless (Fi-Wi) technology. In this article we briefly explain an in-home radio over fiber (RoF) architecture employing 60 GHz spectrum band. The main challenge for designers of such networks is the additional delay introduced by the optical distribution network. Later in the light of this increased delay, we provide an investigation on the feasibility of applying the IEEE 802.15.3c MAC protocol in such a RoF system at 60 GHz. Mainly, we analyse various ACK mechanisms in the IEEE 802.15.3c when it is used in conjunction with RoF network. The analytical results show that the fiber optic network causes a slight drop in the throughput but it could still be easily used in home networks. This gives us an opportunity to sustain high bandwidth at low cost for home networking, also supporting mobility. Modelling of a RoF system with IEEE 802.15.3c and numerical results thereof are the contributions of this article.\",\n \"Title: Voice Activity Detection For Voip-An Information Theoretic Approach\\nYear: 2006\\n\\nAbstract: Voice enabled applications over the Internet are rapidly gaining popularity. Reducing the total bandwidth requirement can make a non-trivial difference for the subscribers having low speed connectivity. Voice Activity Detection algorithms for VoIP applications can save bandwidth by filtering the frames that do not contain speech. In this paper we introduce a novel technique to identify the voice and silent regions of a speech stream very much suitable for VoIP calls. We use an information theoretic measure, called spectral entropy, for differentiating the silence from the speech zones. Specifically we developed a heuristic approach that uses an adaptive threshold to minimize the miss detection in the presence of noise. The performance of our approach is compared with the relatively new 3GPP TS 26.194 (AMR-WB) standard, along with the listeners' intelligibility rating. Our algorithm yields comparatively better saving in bandwidth, yet maintaining good quality of the speech streams.\",\n \"Title: Adaptable Link Quality Estimation for Multi Data Rate Communication Networks\\nYear: 2009\\n\\nAbstract: QoS-sensitive applications transmitted over wireless links require precise knowledge of the wireless environment. However, the dynamic nature of wireless channel, together with its different configurations and types, makes so called link quality estimation (LQE) a difficult task. This paper looks into the design of an accurate and fast LQE method for a multi data rate environment. We investigate the impact of various conditions on the LQE accuracy. In result, two different link quality estimation sources, i.e., based on hello packet delivery ratio and signal strength, are measured and their performances are compared. We find that these two methods are not always accurate. As an improvement we propose an adaptive LQE method that chooses different LQE indicators depending on the wireless environment. The performance of the proposed method is verified via an extensive measurement campaign.\",\n \"Title: On the Performance of TCP Vegas over UMTS/WCDMA Channels with Large Round-Trip Time Variations\\nYear: 2004\\n\\nAbstract: Universal Mobile Telecommunications System (UMTS) is a third-generation cellular network that enables high-speed mobile Internet access. This paper evaluates and compares the performance of two well-known versions of Transmission Control Protocol (TCP), namely, Vegas and Reno, in a UMTS environment. Bulk data transfer was considered in the simulation with varying radio channel conditions. We assume that data losses are only due to the radio channel. Simulation results show that the performance of Vegas is worse than Reno even though data losses incurred by the radio channel are completely recovered by the UMTS radio link control layer. This has led us to conduct a thorough investigation on the behavior of Vegas in order to identify the cause of performance degradation in Vegas. The poor performance of Vegas is attributed to the UMTS radio interface characteristics which resulted in large and highly variable TCP round-trip times. Vegas would interpret the round-trip time variation as a sign of congestion, and consequently, shrink its window size which reduces the transmission rate. Furthermore, a sudden increase in the instantaneous round-trip time can trigger spurious timeouts at the TCP sender using Vegas which performs unnecessary retransmissions. Spurious timeouts can lead to significant throughput reduction. Reno, on the other hand, does not show any abnormality and delivers the expected performance.\",\n \"Title: An Analytical Energy Consumption Model for Packet Transfer over Wireless Links.\\nYear: 2012\\n\\nAbstract: We provide a detailed analytical model for estimating the total energy consumed to exchange a packet over a wireless link. Our model improves many of the current models by considering details such as consumed energy by processing elements of transceivers, packet retransmission, reliability of links, size of data packets and acknowledgments, and also the data rate of wireless links. To develop the model, we use experimental results based on IEEE 802.15.4 devices to show that consumed energy for receiving erroneous packets is comparable to the consumed energy for receiving error-free packets.\",\n \"Title: A Link Layer Protocol For Selforganizing Ultra Wide Band Impulse Radio Networks\\nYear: 2005\\n\\nAbstract: Ultra Wide Band (UWB) impulse radio, promises to be suitable for short-range, low-power, low cost and high data rate applications. While most UWB research is concentrating on the physical layer, little research has been published on the link layer. A novel self-organizing link layer protocol (SDD) based on time hopping impulse radio was proposed by the authors. In this paper, the SDD protocol is further developed and specified in detail. The simulations are carried out using the GloMoSim simulation environment.\",\n \"Title: A Framework For Massive Access And Radio Resource Management In Urban Wlans\\nYear: 2013\\n\\nAbstract: The challenge in future unlicensed wireless networks will be the coordination of spectrum access among wireless LAN (WLAN) nodes in high dense areas. New WLAN standards, such as IEEE 802.11ah, will provide optimized outdoor WLAN coverage with a range of 1 km, thus exceeding the number of associated WLAN nodes. Hence, massive access schemes are required in order to coordinate the wireless access among WLAN nodes and to mitigate mutual interference. In addition, various wireless networks will co-exist, such as WLANs and sensor networks, competing for channel access. Hence, a framework in order to coordinate WLANs and sensor networks in dense areas is essential. This paper is intended to outline the challenges of wireless access in urban WLANs and proposes a framework which considers the challenges of dense wireless access and discusses potential solutions. The contributions are twofold. A clustering framework is proposed to deal with the massive access of wireless nodes, in particular of IEEE 802.11ah WLAN nodes. Further, a radio resource monitoring and management (RRMM) is proposed to manage the co-existence among wireless networks. A novel system architecture is proposed and discussed in order to monitor and manage the WLAN access in highly dense urban areas.\",\n \"Title: Home Networking at 60 GHz: Challenges and Research Issues.\\nYear: 2008\\n\\nAbstract: We highly believe that 60 GHz wireless technology is an ideal candidate to be the future high-speed WLAN standard for home\\n networks thanks to its unique characteristics. In this paper, we review and present the main aspects of the 60 GHz radio regarding\\n future home networks. We introduce the fundamental challenges with the research issues and the considerations on some promising\\n solutions.\",\n \"Title: An implementation study of relay selection schemes for energy harvesting WSNs\\nYear: 2013\\n\\nAbstract: We propose energy harvesting technologies and cooperative relaying techniques to power the devices and improve reliability. We propose schemes to (a) maximize the packet reception ratio (PRR) by cooperation and (b) minimize the average packet delay (APD) by cooperation amongst nodes. Our key result and insight from the testbed implementation is about total data transmitted by each relay. A greedy policy that relays more data under a good harvesting condition turns out to be a sub optimal policy. This is because, energy replenishment is a slow process. The optimal scheme offers a low APD and also improves PRR.\",\n \"Title: Throughput schemes for energy harvesting sensor networks\\nYear: 2012\\n\\nAbstract: With the growing dependence on the wireless sensor devices, it is imperative that power supply to them and their energy usage has acquired centre stage. Energy harvesting is one possible remedy. To enhance the use of devices that are using harvested energy, many novel schemes need to be devised. In this paper we propose four throughput enhancement schemes with a wide range of applicability to energy harvesting wireless sensor networks. The schemes are designed by taking into consideration available resources in the nodes as well as the energy harvesters that are used to drive the sensor nodes. All the four schemes are implemented, evaluated and compared for their performances. We present schemes from simple naive scheme which is of low complexity, to probabilistic probing scheme which incorporates advanced methods to appropriately use the harvested energy. All the schemes require minimum system resources and sufficient harvested energy before starting communication. Moreover, when the harvested energy varies due to the changes in the environment and characteristics of wireless channel varies, we found that truncated channel inversion with power control is the appropriate scheme. This article provides a thorough step by step discussion on the implementation of maximizing communication throughput in energy harvesting sensor nodes under varying environment. We also provide some insights into energy requirements for a multi-node network.\",\n \"Title: Spectrum- and energy-efficient D2DWRAN\\nYear: 2014\\n\\nAbstract: IEEE 802.22 is the first world-wide standard for wireless regional area networks using cognitive radio technologies. Major drawbacks of these networks are limited network capacity and low energy efficiency. One potential solution for these drawbacks is device-to-device communication, which can be incorporated as D2DWRAN. D2DWRAN features direct intracell communication, reuse of channels, and use of multiple operating channels. In this article, we mainly discuss energy-efficient spectrum sharing in D2DWRAN. We explicate a mechanism to increase energy efficiency for long distance links using relays. Along with energy, channel utilization and fairness are also considered while choosing relays. Thus, significant improvement in both network capacity and energy efficiency can be achieved. The schemes discussed here can also be used in other D2D communication settings.\",\n \"Title: Comparing network performance for constant and variable bit rate sources.\\nYear: 1993\\n\\nAbstract: The token ring FDDI and the slotted bus DQDB are currently the two most important standards for high-speed networks. Each of these networks has its own specific advantages and disadvantages, and its own specific type of load for which it performs best. We propose a new network design, called Universal Channel Network (UCN), which is adaptive and can be made to behave like a token ring, a slotted ring, or anything in between. The aim is to get a network that combines the advantages of both the token ring and the slotted ring. In this paper, UCN is explained and its performance is compared with that of DQDB and FDDI. The workload model used is a detailed representation of today's traffic, extrapolated to the future. Because of the uncertainty concerning future traffic, two versions of the workload model are used. In the first version, voice and video sources are modelled as constant bit rate sources. In the second version, variable bit rate sources are used to represent voice and video sources. The network performance for both coding techniques is compared.\",\n \"Title: Downlink throughput characterization of RoF based hybrid 802.11 Fi-Wi\\nYear: 2014\\n\\nAbstract: Radio over Fiber (RoF) based IEEE 802.11x (where x can be a, b, g, n, ac) Fiber Wireless (Fi-Wi) LANs are being thought of as a better solution towards deployment of Enterprise WLAN. It provides optimized load distribution among Access Points (APs) and also better efficient power management of the overall network. In most of the existing works about WLAN deployments, the classical models characterizing the throughput achievable discusses about either single collision solutions where simultaneous transmissions are not allowed or multi cell Enterprise WLANs where the simultaneous transmission are the most predominant feature. But RoF based WLANs because of its inherent architectural design has a different structure. In this paper we want to characterize the downlink throughput that is achievable in such a RoF based Fi-Wi network. The placement of the Cell Access Nodes (CANs) which are simple antennas located indoor to radiate the radio signals is an important factor as well as the connectivity between different CANs and the APs located at the Home Communication Controller (HCC). We analytically provide an upper bound on the difference between the maximum and minimum downlink throughput achievable by an AP in such RoF based Fi-Wi network.\",\n \"Title: Green hybrid Fi-Wi LAN\\nYear: 2014\\n\\nAbstract: Enterprise WLANs encompassing supernumerary Access Points (APs) are being used to provide capacity to indoor environments. The approach of over dimensioned deployment of APs to meet the capacity needs of users at critical time points are advocated without little concern towards the important issue of power wastage. For a sustainable greener environment the issue of power saving in such enterprise WLANs has to be integrated in our present day design. Hybrid Fiber-Wireless (Fi-Wi) LAN is envisioned as the evolution of enterprise WLAN which can help us enormously to reduce the power wastage while meeting the demands of the indoor users. In this paper, we propose a hybrid Fi-Wi LAN architecture based On Demand Strategy which results in massive power saving (approximately 60%) by switching of APs or Cell Access Nodes (CANs) based on user demand estimation while maintaining coverage throughout the whole cell area.\",\n \"Title: Measurements and performance of large MIMO systems at 2.4 GHz for indoor WLAN\\nYear: 2014\\n\\nAbstract: Large MIMO is a new and promising technique for boosting wireless link capacity. Currently, the research of large MIMO focuses on the theoretical aspects while work on the practical implementations and analysis are very limited. In this paper, we present some critical investigations on the behaviors of large MIMO in real propagation environment based on extensive channel measurements. On the one hand, our analysis has verified the advantages that the MIMO channel vectors between users tend to be orthogonal and the optimal capacity approximates the theoretical upper bound as the antenna array size grows. On the other hand, we observe that the practical limitations of allowable antenna array size and achievable spatial degrees of freedom in the physical channel have significant impacts. Furthermore, in order to enable the utilization of linear precoding techniques in such high-dimensional MIMO systems to achieve close-to-optimal capacity, a large ratio between the antenna array size of the large MIMO system and the number of simultaneously served users should be maintained.\",\n \"Title: Codebook Selection Strategies in Long-range Sub-1 GHz WLANs.\\nYear: 2014\\n\\nAbstract: The new long-range WLANs will enable a simple and cost-efficient setup of wireless ambient systems, including smart city and smart grid communication networks. The emerging IEEE 802.11ah WLAN standard utilizes carrier frequencies at 900MHz in the ISM-band and facilitates multi-antenna MIMO-OFDM. In order to further increase the wireless coverage, precoding selection strategies are proposed and analyzed. The Grassmannian and the Kerdock manifold are in the focus of this study due to their simplicity and effectiveness. Whereas the Grassmannian manifold may provide an optimal codeword distance, the Kerdock man- ifold has been found to be more effective due to its reduced quantized codewords combined with an increased SNR. Simulation studies as well as real-world evaluation results indicate a 2-3dB increase of precoding gains when the proposed Kerdock manifold is applied, thus increasing the coverage. The codebooks have been evaluated in a SDR-based IEEE 802.11ah WLAN prototype.\",\n \"Title: An Ultra-Wide Band Based Ad Hoc Networking Scheme For Personnel Tracking In Emergencies\\nYear: 2008\\n\\nAbstract: This paper describes an ultra wideband (UWB) based ad hoc networking scheme which enables a control center to track all emergency personnel deployed in buildings in an incident area, and also enables communication of other types of data directed from emergency personnel, including status information and alarm in particular. In this network, GPS-enabled vehicles form a coordination reference framework. They also serve as egress points to collect data generated by emergency personnel. Additional dropped terminals are used to expand the network coverage. A Time Division Multiple Access (TDMA) based medium access protocol is used by terminals to access the radio channel. An anycast routing protocol is proposed and designed. This protocol enables each ad hoc device attached to emergency personnel to discover a communication path to one of the egress points. The network management related information is included in packet headers. Thus terminals can select free time slots and construct routes by overhearing neighbor's transmissions. Taking the advantage of the periodically nature of position calculation and reporting, each terminal is maintained with up-to-date network status.\",\n \"Title: Architectures for Communication in Personal Networks\\nYear: 2006\\n\\nAbstract: Personal networks (PN) is a new concept related to pervasive computing with a strong user-focus view. The key to a successful PN realization is a general network architecture that is capable of bridging different current and future technologies and offers a homogeneous and clear view to the end-user. In this paper, we focus on forming a PN by connecting remote personal devices using infrastructure-based IP networks, including 3G networks and WLAN hotspots. One way is to upgrade the current access networks with new functionality to support PNs. Since many devices in PNs are mobile and battery powered, this may help them to achieve a faster service and to save energy. However, to deploy such functionality is not easy and may hamper the adoption of PNs altogether. Therefore, in this paper we study three possible inter-cluster communication architectures that can use current IP networks. To discern the above proposal we also give a detailed picture of PN network architecture supported by infrastructure. We believe that this detailed discussion will help the success of PNs\",\n \"Title: Adapting IEEE 802.22 OFDMA system for P2PWRANs.\\nYear: 2013\\n\\nAbstract: IEEE 802.22 is the first standard to uitilize cognitive radio technology for wireless regional area networks (WRANs). It has adopted a cellular topology containing one base station (BS) and multiple customer premises equipments (CPEs) in a cell. It covers a very large area with radius ranging up to 100 km. Orthogonal frequency-division multiple access (OFDMA) system is employed and it is further slotted in time-domain. In one slot only one CPE can be allocated, and every intra-cell packet needs to be routed through the BS. This point-to-multi-point structure significantly limits the network capacity in the case of CPE-CPE communication, which is predicted to happen, and even the use of OFDMA does not help much. Therefore, peer-to-peer WRAN (P2PWRAN) has been proposed to support direct intra-cell communication in order to extend the network capacity. In this paper, the IEEE 802.22 OFDMA system is adapted to P2PWRAN to support direct CPE-CPE communications. The downstream burst allocation in P2PWRAN is similar to the OFDMA system of IEEE 802.16, which has been widely studied in the literature. Therefore, in this paper we look at the unexplored upstream burst allocation where slots can be reused among CPEs and the BS with power control mechanisms. We propose a burst allocation algorithm that maximizes the network capacity greedily. The algorithm is examined under various conditions such as, number of CPEs, size of requests, length of US subframes and type of requests (CPE-BS or CPE-BS). We also show that our algorithm performs better than the existing solutions.\",\n \"Title: Solving incertitude of vertical handovers in heterogeneous mobile wireless network\\nYear: 2007\\n\\nAbstract: Wireless communication network is evolving towards more heterogeneous with rapid emergence of new radio technologies. 60GHz radio relying on line-of-sight (LOS) transmission is one of the promising solutions to support in-door high quality multimedia applications. Due to the vulnerable nature of LOS links, vertical handover, by switching the mobile device to the backup WLAN system from 60GHz radio, is critical to support a seamless session. However vertical handover for short LOS blocking events degrade perceived quality of the media due to frequent switching. In this paper, we propose a cognitive approach to solve incertitude during vertical handover. We first present an overview of our proposed cognitive network architecture, and further under this framework we introduce and examine via simulations a decision algorithm based on decision theory. Our algorithm takes into account multiple factors such as user preference, network situation, device capability and effect of environment. We show its ability to effectively make handover decision in uncertain situations. The method and results herein are applicable, in general, to any other situation where such a decision is to be made.\",\n \"Title: Self-coexistence and spectrum sharing in device-to-device WRANs\\nYear: 2014\\n\\nAbstract: IEEE 802.22 wireless regional area network (WRAN) standard employing cognitive radios is gaining much attention recently. The WRAN standard targets reuse of unused TV channels. We propose a device-to-device wireless regional area network (D2DWRAN) to extend the capacity of IEEE 802.22. Network capacity can be increased by supporting direct intra-cell device to device (D2D) communication through channel reuse and also with aggregation of nonadjacent multiple operating channels. Self-coexistence of neighboring IEEE 802.22 cells is a major challenge in WRANs, since the availability of channels varies frequently and channels are reused by every cell as much as possible. Currently, IEEE 802.22 does not consider D2D communication. We propose a two-tier spectrum sharing mechanism for channel allocation in D2DWRANs - both at intra-cell and inter-cell levels. Algorithms and a thorough analysis are presented. We examine our proposal through simulations. Results show significant performance improvement compared to IEEE 802.22. However, there are many hurdles to cross, such as, coexistence with other cognitive radio networks (CRNs), the intra-cell routing, allocation of other network resources, etc. We believe that further work in this domain can lead to increase in capacity not only in WRANs but also in other cellular networks.\",\n \"Title: Multi-channel management for D2D communications in IEEE 802.22 WRANs\\nYear: 2014\\n\\nAbstract: IEEE 802.22 standard is the first worldwide standard using cognitive radio (CR) technology for wireless regional area networks (WRANs). Orthogonal frequency-division multiple access (OFDMA) is employed for point-to-multi-point (P2M) communication between base stations (BS) and customer premises equipments (CPEs) in WRANs. The large coverage of a single cell (up to 100 km of radius) is the unique feature of WRANs. However, channel reuse is not considered, i.e., in one slot the channel can be allocated to only one CPE, and every intra-cell packet needs to be routed through the BS. Therefore, device-to-device WRAN (D2DWRAN) has been proposed to support direct intra-cell CPE-CPE communication. Power control strategies are used to achieve higher reuse of OFDMA slots. In this paper, we take one step further to enhance the network capacity by adopting multichannel management (MCM) with infrastructure support. This enables multiple operating channels to be used simultaneously in D2DWRANs. New channel switching strategies are also studied. We implement a MCM scheme in an OFDMA setting for D2DWRANs, and examine it in the upstream (US). For downstream (DS), there are already many proposals in the literature, which can be adopted directly by D2DWRAN. The simulation results show that MCM scheme significantly improves the network performance compared to single-channel D2DWRANs and WRANs.\",\n \"Title: On systems generating context triggers through energy harvesting\\nYear: 2014\\n\\nAbstract: Context awareness is one of the building blocks of smart applications that constitutes smart spaces. With the emergence of cyber physical systems, it is now possible to create spaces that are truly adaptable and smart. In these spaces, contextual parameters are captured by many wireless sensor nodes. This collected data is processed to understand the context in real time. Since a large number of sensors are deployed, processing all the data is a big task. Moreover, since the sensors are powered by batteries, they have a limited lifetime. Sensing only when there is a context event can save energy as well as reduce data processing. To make sensor nodes operate perpetually, ambient energy harvesters can be used. Typically, the energy harvesting source of a sensor is related to the physical parameters the sensor is measuring. We propose exploiting this property to develop a context-event triggering mechanism in this article. We also adapt the context-aware application framework for incorporating context-event triggers by harvesters. A Smart-M3-based architecture is also proposed. Through a real-world use case, we illustrate significant energy savings and reduced data processing in our proposed approach.\",\n \"Title: A simulation study on a high-speed slotted-ring access mechanism with dynamically adaptive slot sizes\\nYear: 1991\\n\\nAbstract: Two access mechanisms complementary in performance are the token ring and the slotted ring. While the token ring outperforms the slotted ring for long messages, the latter performs significantly better for short messages. The paper shows that the factor that causes this difference is the number of tokens on the ring. It proposes a new network design in which the number of tokens has been made adaptive, resulting in a network that can be made to behave like both the token ring and the slotted ring, or anything in between, i.e. it performs optimally for any given message length. The paper presents first simulation results that show the performance gain that is achieved by using this principle. Further, it introduces a new priority mechanism in which an upper bound for the access delay of high priority messages is guaranteed\",\n \"Title: Cooperative Density Estimation in Random Wireless Ad Hoc Networks\\nYear: 2012\\n\\nAbstract: Density estimation is crucial for wireless ad hoc networks for adequate capacity planning. Protocols have to adapt their operation to the density since the throughput in an ad hoc network approaches asymptotically to zero as the density increases. A wireless node can estimate the global density by using local information such as the received power from neighbors. In this paper, we propose a cross layer protocol to compute the density estimate. The accuracy of the estimate can be enhanced and its variance can be reduced through cooperation among the nodes. Nodes share the received power measurements with each other. Based on the collected observations, the maximum likelihood estimate is computed. It is shown that cooperative density estimation has better accuracy with less variance than the individual estimation. When nodes share received power measurements from further away neighbors, the variance of the estimate is further reduced.\",\n \"Title: Impact of antenna directionality and energy harvesting rate on neighbor discovery in EH-IoTs\\nYear: 2013\\n\\nAbstract: Homes, offices and vehicles are getting networked. This will enable context aware, autonomous operation of many support systems that could be controlled remotely. To achieve this there would be a large number of tiny devices - sensors and actuators - which are networked and they are termed generally as Internet of Things (IoT) devices. In future, they will be powered through harvested energy from the ambience to enable perennial lifetime and minimal manual maintenance. Some examples of energy sources are photovoltaic panels and piezoelectric crystals. Several challenges arise due to the nature of sources of energy. One of these challenges is that the devices (nodes) leave and re-enter networks due to fluctuating availability of harvested energy. As a result, the neighbor table maintained at each node changes quite often leading to complications in forming and maintaining routes. In fact initial neighbor discovery (ND) itself is a difficult task. Further, usage of directional antennas would increase the time taken to complete ND. We study such a network through exhaustive simulation study considering various parameters. We demonstrate the benefits and challenges of using directional antennas for ND. We present a scheme that nodes could use to discover their neighbor during initial deployment and another scheme that could be used for subsequent discovery on re-entry into the network. We show that a dedicated ND protocol is necessary for energy harvesting networks and that directional ND is beneficial in these networks under some circumstances.\",\n \"Title: Performance evaluation of power-aware routing algorithms in Personal Networks\\nYear: 2009\\n\\nAbstract: Personal networks (PN) is a concept in the area of modern communication networks in which devices belonging to a person cooperate to offer services to the person regardless of their geographical locations. A PN may consist of several ad-hoc sub-networks (or clusters) which are interconnected through infrastructure networks such as the Internet. Failure of personal devices due to battery shortage is one of the main threads for service availability in PNs. Power-aware routing (PAR) has been proposed as an effective scheme to prolong the lifetime of battery-operated nodes in wireless ad-hoc networks. This scheme, however, could be utilized in PNs as well, since a PN cluster can be modeled as a wireless ad-hoc network. Aiming at prolonging the lifetime of nodes in PNs, in this paper we propose a PAR algorithm for PNs and we evaluate and compare its performance with the performance of other PAR algorithms proposed so-far for wireless ad-hoc networks. We analyze the effect of node density, routing overhead, heterogeneity of PN nodes in terms of their power supplies, and communication pattern within PN clusters on the performance of the PAR algorithms. Taking into account various parameters, we show that in a PN environment, our proposed scheme can profoundly extend the lifetime of PN nodes compared to the other algorithms. We also show that there can be an optimal value for route refresh interval, as a parameter which affects the generated routing overhead, and the node density of PN clusters, in such a way that the network lifetime is maximized.\",\n \"Title: Robust 60 GHz Indoor Connectivity: Is It Possible with Reflections?\\nYear: 2010\\n\\nAbstract: In this paper, we investigate the robustness of the 60 GHz connectivity in typical indoor environments by analyzing the outage probability. We define three realistic indoor scenarios which may host the 60 GHz networks in the future and perform simulations with a verified 3D ray tracing tool on them. In the first set of simulations, we show the impact of access unit position on the connectivity. In the next step, we show that increasing the obstacle density linearly decreases the outage probability of the 60 GHz network and this decrease is found to be sharper for certain positions of the access point. In the last step, we demonstrate the direct effect of reflective surface availability on the connectivity. A 60 GHz indoor network relying solely on the reflections in the absence of line-of-sight path is very vulnerable to outages even in moderately populated indoor environments.\",\n \"Title: CogCell: cognitive interplay between 60 GHz picocells and 2.4/5 GHz hotspots in the 5G era.\\nYear: 2015\\n\\nAbstract: The rapid proliferation of wireless communication devices and the emergence of a variety of new applications have triggered investigations into next-generation mobile broadband systems, i.e. 5G. Legacy 2G-4G systems covering large areas were envisioned to serve both indoor and outdoor environments. However, in the 5G era, 80 percent of all traffic is expected to be generated indoors. Hence, the cu...\",\n \"Title: An architectural framework for 5G indoor communications\\nYear: 2015\\n\\nAbstract: In this paper, we emphasize on indoor networks in 5G era. We explore the possible technologies and architectural solutions for 5G indoor communications. Owing to the fact that requirements for indoor and outdoor communications will be quite different in the next generation networks, we try to define an architectural framework for 5G indoor communications. Our proposed architecture focuses on three aspects of indoor network architecture, namely; air interface, backhaul connectivity and indoor signal distribution.\",\n \"Title: Choose Wisely: Topology Control In Energy-Harvesting Wireless Sensor Networks\\nYear: 2016\\n\\nAbstract: Ambient energy-harvesting technology is a promising approach to keep wireless sensor networks (WSNs) operating perpetually. Depending on the harvesting source nodes can either be active (alive) or inactive (dead) at any instant in such Energy-Harvesting WSNs (EH-WSNs). Thus, even in a static deployment of EH-WSNs, the network topology is no longer static. A popular method to increase energy-efficiency in WSNs is by employing topology control algorithms. Most of the topology control algorithms in the literature cannot handle the situation when nodes have different energy-levels, and when number of active nodes varies with time in EH-WSN. To address this issue, we present two localized energy based topology control algorithms, viz., EBTC-1 and EBTC-2. EBTC-1 is for convergecast applications of WSNs and EBTC-2 is for a generic scenario where all nodes are required to be connected. While typical topology control algorithms select a particular number of neighbors, the distinguishing feature of both these algorithms is that they select neighbors based on energy-levels, and render the global topology strongly-connected. Simulation results confirm that EBTC-1 and EBTC-2 reduce the transmission power and they let nodes have neighbors with high remaining energy. Results show that our proposed algorithms increase at least 33% in the remaining energy per neighbor. In addition, in terms of energy consumption and fault-tolerance, our proposed algorithms typically achieve 1-connected topology using 74% less energy compared to K-Neigh.\",\n \"Title: 60 GHz MAC Standardization: Progress and Way Forward.\\nYear: 2015\\n\\nAbstract: Communication at mmWave frequencies has been the focus in the recent years. In this paper, we discuss standardization efforts in 60 GHz short range communication and the progress therein. We compare the available standards in terms of network architecture, medium access control mechanisms, physical layer techniques and several other features. Comparative analysis indicates that IEEE 802.11ad is likely to lead the short-range indoor communication at 60 GHz. We bring to the fore resolved and unresolved issues pertaining to robust WLAN connectivity at 60 GHz. Further, we discuss the role of mmWave bands in 5G communication scenarios and highlight the further efforts required in terms of research and standardization.\",\n \"Title: Murphy Loves Ci: Unfolding And Improving Constructive Interference In Wsns\\nYear: 2016\\n\\nAbstract: Constructive Interference (CI) phenomenon has been exploited by Glossy, a mechanism for low-latency and reliable network flooding and time synchronization for wireless sensor networks. Recently, CI has also been used for other applications such as data collection and multicasting in static and mobile WSNs. These applications base their working on the high reliability promised by Glossy regardless of the physical conditions of deployment, number of nodes in the network, and unreliable wireless channels that may be detrimental for CI. There are several works that study the working of CI, but they present inconsistent views. We study CI from a receiver's viewpoint, list factors that affect CI and also specify how and why they affect. We validate our arguments with results from extensive and rigorous experimentation in real-world settings. This paper presents comprehensive insights into CI phenomenon.With this understanding, we improve the performance of CI through an energy-efficient and distributed algorithm. We cause destructive interference on a designated byte to provide negative feedback. We leverage this to adapt transmission powers. Compared to Glossy, we achieve 25% lesser packet losses while using only half of its transmission power.\",\n \"Title: V2X Application-Reliability Analysis of Data-Rate and Message-Rate Congestion Control Algorithms.\\nYear: 2017\\n\\nAbstract: Intelligent transportation systems (ITS) require vehicle-to-everything communication. In dense traffic, the communication channel may become congested, impairing the reliability of the ITS safety applications. Therefore, the European Telecommunications Standard Institute demands decentralized congestion control (DCC) to control the channel load. Our objective is to investigate whether message-rate or data-rate congestion control provides better application reliability. We compare LIMERIC and PDR-DCC as representatives of the two principles. We analyzed the application reliability of LIMERIC with different data-rates and PDR-DCC with different message-rates for varying traffic densities and application requirements. We observed that, for applications with demanding requirements and in a large variety of vehicular densities, PDR-DCC (data-rate) provides more reliable communication support than LIMERIC (message-rate). Furthermore, this letter hints that a combined message-rate and data-rate congestion control can improve reliability further.\",\n \"Title: Performance Analysis of IEEE 802.11ad MAC Protocol.\\nYear: 2017\\n\\nAbstract: IEEE 802.11ad specifies a hybrid medium access control (MAC) protocol consisting of contention as well as non-contention-based channel access mechanisms. It also employs directional antennas to compensate for the high free-space path loss observed in 60 GHz frequency band. Therefore, it significantly differs from other IEEE 802.11(b/g/n/ac) MAC protocols and thus requires new methods to analyze its performance. We propose a new analytical model for performance analysis of IEEE 802.11ad employing a 3-D Markov chain considering all the features of IEEE 802.11ad medium access mechanisms including the presence of non-contention access and the different number of sectors due to the use of directional antennas. We show that the number of sectors has a high impact on the network throughput. We show that the MAC packet delay is significantly affected by the duration of the contention period. Our results indicate that a suitable choice of the number of sectors and contention period can improve the channel utilization and MAC delay performance.\",\n \"Title: Distributed Antenna System for Mitigating Shadowing Effect in 60 GHz WLAN\\nYear: 2015\\n\\nAbstract: The 60 GHz unlicensed frequency band has been adopted to support high data rate WLAN applications. The problem of this frequency band is that the wireless signal is vulnerable to the shadowing of objects. Especially in indoor scenarios, humans may frequently block the signal paths that cause disconnections of devices. This problem can be mitigated by using multiple antennas that can create rich signal paths between the devices. The idea presented in this paper is to employ a distributed antenna system (DAS) architecture enabled by radio-over-fiber technology to alleviate the shadowing problem. In the DAS architecture, multiple remote antenna units (RAUs), each of them equipped with an antenna array, are connected with the same WLAN access points via optical fibers. But there are questions that need to be answered, including the beamforming strategy with multiple RAUs, and the placement and the required number of the RAUs. These questions are related to factors like room size, population density, etc. We discuss these questions in this paper and present our findings through theoretical models and extensive simulations, which can be referred to for practical implementations. We find that the proposed 60 GHz DAS can significantly improve the link connectivity. Even when only two RAUs are used, the signal coverage can be improved to an acceptable percentage in heavy shadowing scenarios.\",\n \"Title: Coexistence Of Decentralized Congestion Control Algorithms For V2v Communication\\nYear: 2018\\n\\nAbstract: Channel congestion is one of the most critical issues in IEEE 802.11p-based vehicular communications as it leads to the unreliability of safety applications. As a countermeasure, many Decentralized Congestion Control (DCC) algorithms have been proposed. One of the most prominent DCC algorithms is the message-rate based LIMERIC. Recently, algorithms have also been proposed to support higher vehicle density (better scalability). One of such algorithms is the combined message-rate and data-rate based congestion control algorithm (MD-DCC). MD-DCC can support around 2.7 times higher vehicular density than LIMERIC. However, if LIMERIC has been deployed, can MD-DCC be introduced and coexist well with LIMERIC. The objective of this paper is to investigate how MD-DCC coexists with LIMERIC. Given a scenario where vehicles may use either LIMERIC or MD-DCC, we study the impact of coexistence on channel load, fairness and reliability of vehicles at different densities. Simulation studies show that there is no significant degradation of reliability both for LIMERIC and MD-DCC at different densities. On the contrary, coexistence can improve the reliability of LIMERIC vehicles. Furthermore, MD-DCC can support vehicles at large densities even when it coexists with LIMERIC retaining its scalability. However, fair allocation of resources is not guaranteed when LIMERIC and MD-DCC coexist.\",\n \"Title: Deep Q-network based dynamic power allocation for cell-free massive MIMO\\nYear: 2021\\n\\nAbstract: Numerical optimization has been investigated for decades to address complex problems. Many effective methods, e.g., the weighted minimum mean square error (WMMSE) algorithm, have been developed for a large variety of applications in wireless communication systems. However, these methods often require high computational cost creating a serious gap between theoretical analysis and real-time processi...\"\n]","isConversational":false},"profile_size":{"kind":"number","value":86,"string":"86"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13568,"string":"13,568"}}},{"rowIdx":13569,"cells":{"question_id":{"kind":"string","value":"Paul M B Vit nyi"},"question":{"kind":"string","value":"Generate an abstract for the title \"How well can a graph be n-colored?\" using the following items: Graph coloring, Edge minimization, NP-complete, Algorithm, Complete graphs."},"answer":{"kind":"string","value":"The problem of how ''near'' we can come to a n-coloring of a given graph is investigated. I.e., what is the minimum possible number of edges joining equicolored vertices if we color the vertices of a given graph with n colors. In its generality the problem of finding such an optimal color assignment to the vertices (given the graph and the number of colors) is NP-complete. For each graph G, however, colors can be assigned to the vertices in such a way that the number of offending edges is less than the total number of edges divided by the number of colors. Furthermore, an @W(epn) deterministic algorithm for finding such an n-color assignment is exhibited where e is the number of edges and p is the number of vertices of the graph (e=p=n). A priori solutions for the minimal number of offending edges are given for complete graphs; similarly for equicolored K\\\"m in K\\\"p and equicolored graphs in K\\\"p."},"correct_answer":{"kind":"string","value":"The problem of how ''near'' we can come to a n-coloring of a given graph is investigated. I.e., what is the minimum possible number of edges joining equicolored vertices if we color the vertices of a given graph with n colors. In its generality the problem of finding such an optimal color assignment to the vertices (given the graph and the number of colors) is NP-complete. For each graph G, however, colors can be assigned to the vertices in such a way that the number of offending edges is less than the total number of edges divided by the number of colors. Furthermore, an @W(epn) deterministic algorithm for finding such an n-color assignment is exhibited where e is the number of edges and p is the number of vertices of the graph (e=p=n). A priori solutions for the minimal number of offending edges are given for complete graphs; similarly for equicolored K\\\"m in K\\\"p and equicolored graphs in K\\\"p."},"documents":{"kind":"list like","value":["Title: Registers\nYear: 2006\n\nAbstract: Entry in: Encyclopedia of Algorithms, Ming-Yang Kao, Ed., Springer, Toappear. Synonyms: Wait-free registers, wait-free shared variables, asynchronouscommunication hardware. Problem Definition: Consider a system of asynchronousprocesses that communicate among themselves by only executing read and writeoperations on a set of shared variables (also known as shared registers). Thesystem has no global clock or other synchronization primitives.","Title: Randomness\nYear: 2001\n\nAbstract: Here we present in a single essay a combination and completion of the severalaspects of the problem of randomness of individual objects which of necessityoccur scattered in our texbook \"An Introduction to Kolmogorov Complexity andIts Applications\" (M. Li and P. Vitanyi), 2nd Ed., Springer-Verlag, 1997.","Title: Algorithmic Chaos\nYear: 2003\n\nAbstract: Many physical theories like chaos theory are fundamentally concerned with the\nconceptual tension between determinism and randomness. Kolmogorov complexity\ncan express randomness in determinism and gives an approach to formulate\nchaotic behavior.","Title: Universal Similarity\nYear: 2005\n\nAbstract: We survey a new area of parameter-free simi- larity distance measures useful in data-mining, pattern recog- nition, learning and automatic semantics extraction. Given a family of distances on a set of objects, a distance is universal up to a certain precision for that family if it minorizes every distance in the family between every two objects in the set, up to the stated precision (we do not require the universal dis- tance to be an element of the family). We consider similarity distances for two types of objects: literal objects that as such contain all of their meaning, like genomes or books, and names for objects. The latter may have literal embodyments like the first type, but may also be abstract like \"red\" or \"christian- ity.\" For the first type we consider a family of computable dis- tance measures corresponding to parameters expressing simi- larity according to particular features between pairs of literal objects. For the second type we consider similarity distances generated by web users corresponding to particular semantic relations between the (names for) the designated objects. For both families we give universal similarity distance measures, incorporating all particular distance measures in the family. In the first case the universal distance is based on compression and in the second case it is based on Google page counts re- lated to search terms. In both cases experiments on a massive scale give evidence of the viability of the approaches.","Title: Relativized Obliviousness\nYear: 1980\n\nAbstract: Relativized obliviousness is introduced to capture the intuitive idea, that some problems allow fastest computations which are more oblivious than do other problems, without any of such computations being oblivious in the standard sense. It is shown that each increase in the obliviousness of an algorithm (in several different well-defined meanings), for the solution of some problems, may necessarily require an increase in computation time from T(n) steps to T(n) log T(n) steps. There is, however, no problem for which a total oblivious algorithm requires more than order T(n) log T(n) steps, if the best algorithm for it runs in T(n) steps. We use on-line Turing machines as model of computation.","Title: Compression-Based Similarity\nYear: 2011\n\nAbstract: First we consider pair-wise distances for literal objects consisting of finite binary files. These files are taken to contain all of their meaning, like genomes or books. The distances are based on compression of the objects concerned, normalized, and can be viewed as similarity distances. Second, we consider pair-wise distances between names of objects, like \"red\" or \"christianity.\" In this case the distances are based on searches of the Internet. Such a search can be performed by any search engine that returns aggregate page counts. We can extract a code length from the numbers returned, use the same formula as before, and derive a similarity or relative semantics between names for objects. The theory is based on Kolmogorov complexity. We test both similarities extensively experimentally.","Title: Atomic Multireader Register\nYear: 1987\n\nAbstract: We give implementations for atomic, shared, asynchronous, wait-free registers: (i) A new implemenfation of an atomic, 1-writer, 1-reader, b-bit register from O (b) safe, boolean registers (i.e., from scratch). The solution uses neither repeated writing of the input nor repeated reading of the output. (ii) An implementation of an atomic, 1-writer, n-reader, multibit register from O(n\n2) atomic, 1-writer, 1-reader, multibit registers. Both constructions rely on the same idea. In a sense (ii) is a generalization of (i). These results show how to construct atomic, multireader registers from — basically — elementary hardware like flip-flops.","Title: Individual communication complexity\nYear: 2007\n\nAbstract: We initiate the theory of communication complexity of individual inputs held by the agents. This contrasts with the usual communication complexity model, where one counts the amount of communication for the worst-case or the average-case inputs. The individual communication complexity gives more information (the worst-case and the average-case can be derived from it but not vice versa) and may in some cases be of more interest. It is given in terms of the Kolmogorov complexities of the individual inputs. There are different measures of communication complexity depending on whether the protocol is guaranteed to be correct for all inputs or not, and whether there's one-way or two-way communication. Bounds are provided for the communication of specific functions and connections between the different communication measures are shown. Some counter-intuitive results: for deterministic protocols that need to communicate Bob's input to Alice they need to communicate all of Bob's input (rather than the information difference with Alice's input), and there are so-called ''non-communicable'' inputs.","Title: Counting is easy\nYear: 2001\n\nAbstract: For any fixed k, a remarkably simple single-tape Turing machine can simulate k independent counters in real time.","Title: Distributed Match-Making\nYear: 1988\n\nAbstract: In many distributed computing environments, processes are concurrently executed by nodes in a store- and-forward communication\n network. Distributed control issues as diverse as name server, mutual exclusion, and replicated data management involve making\n matches between such processes. We propose a formal problem called “distributed match-making” as the generic paradigm. Algorithms\n for distributed match-making are developed and the complexity is investigated in terms of messages and in terms of storage\n needed. Lower bounds on the complexity of distributed match-making are established. Optimal algorithms, or nearly optimal\n algorithms, are given for particular network topologies.","Title: Normalized Information Distance\nYear: 2008\n\nAbstract: The normalized information distance is a universal distance measure for objects of all kinds. It is based on Kolmogorov complexity and thus uncomputable, but there are ways to utilize it. First, compression algorithms can be used to approximate the Kolmogorov complexity if the objects have a string representation. Second, for names and abstract concepts, page count statistics from the World Wide Web can be used. These practical realizations of the normalized information distance can then be applied to machine learning tasks, expecially clustering, to perform feature-free and parameter-free data mining. This chapter discusses the theoretical foundations of the normalized information distance and both practical realizations. It presents numerous examples of successful real-world applications based on these distance measures, ranging from bioinformatics to music clustering to machine translation.","Title: On Empirical Entropy\nYear: 2011\n\nAbstract: We propose a compression-based version of the empirical entropy of a finite string over a finite alphabet. Whereas previously one considers the naked entropy of (possibly higher order) Markov processes, we consider the sum of the description of the random variable involved plus the entropy it induces. We assume only that the distribution involved is computable. To test the new notion we compare the Normalized Information Distance (the similarity metric) with a related measure based on Mutual Information in Shannon's framework. This way the similarities and differences of the last two concepts are exposed.","Title: Optimal routing tables\nYear: 1996\n\nAbstract: The optimal space used to represent routing schemes incommunication networks is established, both for worst-casestatic networks and on the average for all static networks.Several factors may influence the cost of representing a routingscheme for a particular network. It is therefore unavoidablethat we first describe several reasonable models in whichto measure this cost. Failure to do so in the past has obfuscatedprevious results.We show that, in most models, for almost all graphs...","Title: The similarity metric\nYear: 2004\n\nAbstract: A new class of distances appropriate for measuring similarity relations between sequences, say one type of similarity per distance, is studied. We propose a new "normalized information distance," based on the noncomputable notion of Kolmogorov complexity, and show that it is in this class and it minorizes every computable distance in the class (that is, it is universal in that it discovers all com...","Title: Tales of Huffman\nYear: 2006\n\nAbstract: We study the new problem of Huffman-like codes subject to individual restrictions on the code-word lengths of a subset of the source words. These are prefix codes with min imal expected code-word length for a random source where additionally the code-word lengths of a subset of the source words is prescribed, possibly differently for every such source word. Based on a structural analysis of properties of optimal solutions, we construct an efficient dynamic programming algorithm for this problem, a nd for an integer programming problem that may be of independent interest.","Title: The Incompressibility Method\nYear: 2000\n\nAbstract: . Kolmogorov complexity is a modern notion of randomnessdealing with the quantity of information in individual objects; that is,pointwise randomness rather than average randomness as produced bya random source. It was proposed by A.N. Kolmogorov in 1965 to quantifythe randomness of individual objects in an objective and absolutemanner. This is impossible for classical probability theory. Kolmogorovcomplexity is known variously as `algorithmic information\", `algorithmicentropy\",...","Title: A Proof Technique for Register Automicity\nYear: 1988\n\nAbstract: An implementation of a concurrent data object is wait-free if any process can complete any operation in a bounded number of steps, independently of the execution speeds of the programs. Much recent work has been done on concurrent access of shared variables by asynchronous processes. That work shows that implementing such shared variables does not require synchronization (by e.g. mutual exclusion), but can be solved in a wait-free manner. A fruitful paradigm in this context is the notion of a shared register satisfying a niceness condition called atomicity. Recent proposed solutions have led to the realization that: (1) neither the problem to be solved nor the model required were rigorously defined, (2) there was no clear insight in what constitutes a good proof of correctness in the area, and (3) the proposed protocols are so complicated that although correctness may be possible in a \"platonic fashion\", verifiability seems impossible to attain by human beings. A lot of controversy and allegations about constructions and proofs have arisen. Consequently, we have spent great effort to put the area on a rigorous basis. The thrust of this paper is to provide a new proof technique, and demonstrate its applicability by a non-trivial example. In other words, a new model is rigorously presented for the first time, and then a new method is given for proving register atomicity. It is then used to give a simple proof of the atomicity of the first and only direct construction of a multireader multiwriter register from atomic 1-reader 1-writer registers. (This construction was given in [8], by two of the present authors, with a completely different proof.)","Title: A Note on Weighted Distributed Match-Making\nYear: 1992\n\nAbstract: ABSTRACT This is a preliminary draft version of [Mathematical Systems Theory, 25(1992), 123 - 140] In many distributed computing environments, processes are concurrently executed by nodes in a store - and - forward network Distributed control issues as diverse as name - server, mutual exclusion and replicated data management, involve making matches between processes The generic paradigm is a formal problem called \"\"distributed match - making\"\" We define multi - dimensional and weighted versions, and the relations between the two, and develop a very general method to prove lower bounds on the complexity as a trade - off between number of messages and \"\"distributed - ness\"\" The resulting lower bounds are tight in all cases we have exam - ined We present a success - stop version of distributed match - making that is analysed in terms of a weight distribution that in all cases results in approximately halving the (expected) number of messages required in the corresponding strategy that does not use these weights","Title: Reversible Simulation of Irreversible Computation by Pebble Games\nYear: 1997\n\nAbstract: Reversible simulation of irreversible algorithms is analyzed in the stylized form of a 'reversible' pebble game. The reacheable reversible simulation instantaneous descriptions (pebble configurations) are characterized completely. As a corollary we obtain the reversible simulation by Bennett and that among all simulations that can be modelled by the pebble game, Bennett's simulation is optimal in that it uses the least auxiliary space for the greatest number of simulated steps. One can reduce the auxiliary storage overhead incurred by the reversible simulation at the cost of allowing limited erasing leading to an irreversibility-space tradeoff. We show that in this resource-bounded setting the limited erasing needs to be performed at precise instants during the simulation. We show that the reversible simulation can be modified so that it is applicable also when the simulated computation time is unknown.","Title: Genetic Fitness Optimization Using Rapidly Mixing Markov Chains\nYear: 1996\n\nAbstract: A notion of highly probable fitness optimization through evolutionary computing runs on small size populations in a very general setting is proposed. This has applications to evolutionary learning. Based on rapidly mixing Markov chains, the approach pertains to most types of evolutionary genetic algorithms, genetic programming and the like. For systems having associated rapidly mixing Markov chains and appropriate stationary distributions the new method finds optimal programs (individuals) with probability almost 1. Algorithmically, the novel approach prescribes a strategy of executing many short computation runs, rather than one long computation run. Given an arbitrary evolutionary program it may be infeasible to determine whether its associated matrix is rapidly mixing. In our proposed structured evolutionary program discipline, the development of the program and the guaranty of the rapidly mixing property go hand in hand. We conclude with a tentative toy example.","Title: Depth as Randomness Deficiency\nYear: 2009\n\nAbstract: Depth of an object concerns a tradeoff between computation time and excess of program length over the shortest program length\n required to obtain the object. It gives an unconditional lower bound on the computation time from a given program in absence\n of auxiliary information. Variants known as logical depth and computational depth are expressed in Kolmogorov complexity theory.\n \n We derive quantitative relation between logical depth and computational depth and unify the different depth notions by relating them to A. Kolmogorov and L. Levin’s fruitful notion of randomness deficiency. Subsequently, we revisit\n the computational depth of infinite strings, study the notion of super deep sequences and relate it with other approaches.","Title: Combinatorics and Kolmogorov complexity\nYear: 1991\n\nAbstract: The authors investigate combinatorial properties of finite sequences with high Kolmogorov complexity. They also demonstrate the utility of a Kolmogorov complexity method in combinatorial theory by several examples (such as the coin-weighing problem)","Title: Analysis of Sorting Algorithms by Kolmogorov Complexity (A Survey)\nYear: 2009\n\nAbstract: Recently, many results on the computational complexity of sorting algorithms were obtained using Kolmogorov complexity (the\n incompressibility method). Especially, the usually hard average-case analysis is ammenable to this method. Here we survey\n such results about Bubblesort, Heapsort, Shellsort, Dobosiewiczsort, Shakersort, and sorting with stacks and queues in sequential\n or parallel mode. Especially in the case of Shellsort the uses of Kolmogorov complexity surprisingly easily resolved problems\n that had stayed open for a long time despite strenuous attacks.","Title: Algorithmic Identification of Probabilities.\nYear: 2013\n\nAbstract: TThe problem is to identify a probability associated with a set of natural numbers, given an infinite data sequence of elements from the set. If the given sequence is drawn i.i.d. and the probability mass function involved (the target) belongs to a computably enumerable (c.e.) or co-computably enumerable (co-c.e.) set of computable probability mass functions, then there is an algorithm to almost surely identify the target in the limit. The technical tool is the strong law of large numbers. If the set is finite and the elements of the sequence are dependent while the sequence is typical in the sense of Martin-L\\\"of for at least one measure belonging to a c.e. or co-c.e. set of computable measures, then there is an algorithm to identify in the limit a computable measure for which the sequence is typical (there may be more than one such measure). The technical tool is the theory of Kolmogorov complexity. We give the algorithms and consider the associated predictions.","Title: Information Distance: New Developments\nYear: 2012\n\nAbstract: In pattern recognition, learning, and data mining one obtains information from information-carrying objects. This involves an objective definition of the information in a single object, the information to go from one object to another object in a pair of objects, the information to go from one object to any other object in a multiple of objects, and the shared information between objects. This is called “information distance.” We survey a selection of new developments in information distance.","Title: Weighted Distributed Match-Making\nYear: 1988\n\nAbstract: In many distributed computing environments, processes are concurrently executed by nodes in a store-and-forward network. Distributed\n control issues as diverse as name-server, mutual exclusion and replicated data management, involve making matches between\n processes. The generic paradigm is a formal problem called “distributed match-making”. The applications require solutions\n to weighted versions of the problem. We define new multi-dimensional and weighted versions, and the relations between the\n two, and develop a very general method to prove lower bounds on the complexity as a trade-off between number of messages and\n “distributedness”. The resulting lower bounds are tight in all cases we have examined.","Title: The Google Similarity Distance\nYear: 2007\n\nAbstract: Words and phrases acquire meaning from the way they are used in society, from their relative semantics to other words and phrases. For computers, the equivalent of \"society” is \"database,” and the equivalent of \"use” is \"a way to search the database.” We present a new theory of similarity between words and phrases based on information distance and Kolmogorov complexity. To fix thoughts, we use the World Wide Web (WWW) as the database, and Google as the search engine. The method is also applicable to other search engines and databases. This theory is then applied to construct a method to automatically extract similarity, the Google similarity distance, of words and phrases from the WWW using Google page counts. The WWW is the largest database on earth, and the context information entered by millions of independent users averages out to provide automatic semantics of useful quality. We give applications in hierarchical clustering, classification, and language translation. We give examples to distinguish between colors and numbers, cluster names of paintings by 17th century Dutch masters and names of books by English novelists, the ability to understand emergencies and primes, and we demonstrate the ability to do a simple automatic English-Spanish translation. Finally, we use the WordNet database as an objective baseline against which to judge the performance of our method. We conduct a massive randomized trial in binary classification using support vector machines to learn categories based on our Google distance, resulting in an a mean agreement of 87 percent with the expert crafted WordNet categories.","Title: The Quantum Computing Challenge\nYear: 2001\n\nAbstract: The laws ofp hysics imposes limits on increases in computing power. Two ofthe se limits are interconnect wires in multicomputers and thermodynamic limits to energy dissipation in conventional irreversible technology. Quantum computing is a new computational technology that promises to eliminate problems ofl atency and wiring associated with parallel computers and the rapidly approaching ultimate limits to computing power imposed by the fundamental thermodynamics. Moreover, a quantum computer will be able to exponentially improve known classical algorithms for factoring, and quadratically improve every classical algorithm for searching an unstructured list, as well as give various speedups in communication complexity, by exploiting unique quantum mechanical features. Finally, a quantum computer may be able to simulate quantum mechanical systems, something which seems out ofthe question for classical computers, thus reaching the ultimate goal ofrep lacing actual quantum mechanical experiments with simulated ones. On the downside, for some problems quantum mechanical computers cannot significantly improve the performance of classical computers.","Title: Randomized Wait-Free Naming\nYear: 1994\n\nAbstract: We present new distributed randomized naming protocols improving previous results in renaming and uniqueprocessor identity protocols. They are wait-free (which implies maximal fault-tolerance) and allow strongeradversaries. They also have low complexity. We give the first wait-free protocol achieving optimal key spacerange. (This is impossible for deterministic wait-free methods, so we use randomization.) We also introducea novel wait-free object, a test-and-set object which upon invocation ...","Title: Normalized Information Distance is Not Semicomputable\nYear: 2010\n\nAbstract: Normalized information distance (NID) uses the theoretical notion of\nKolmogorov complexity, which for practical purposes is approximated by the\nlength of the compressed version of the file involved, using a real-world\ncompression program. This practical application is called 'normalized\ncompression distance' and it is trivially computable. It is a parameter-free\nsimilarity measure based on compression, and is used in pattern recognition,\ndata mining, phylogeny, clustering, and classification. The complexity\nproperties of its theoretical precursor, the NID, have been open. We show that\nthe NID is neither upper semicomputable nor lower semicomputable.","Title: The probabilistic analysis of language acquisition: Theoretical, computational, and experimental analysis\nYear: 2010\n\nAbstract: There is much debate over the degree to which language learning is governed by innate language-specific biases, or acquired through cognition-general principles. Here we examine the probabilistic language acquisition hypothesis on three levels: We outline a novel theoretical result showing that it is possible to learn the exact generative model underlying a wide class of languages, purely from observing samples of the language. We then describe a recently proposed practical framework, which quantifies natural language learnability, allowing specific learnability predictions to be made for the first time. In previous work, this framework was used to make learnability predictions for a wide variety of linguistic constructions, for which learnability has been much debated. Here, we present a new experiment which tests these learnability predictions. We find that our experimental results support the possibility that these linguistic constructions are acquired probabilistically from cognition-general principles.","Title: Minimum description length induction, Bayesianism, and Kolmogorov complexity\nYear: 1999\n\nAbstract: The relationship between the Bayesian approach and the minimum description length approach is established. We sharpen and clarify the general modeling principles minimum description length (MDL) and minimum message length (MML), abstracted as the ideal MDL principle and defined from Bayes's rule by means of Kolmogorov complexity. The basic condition under which the ideal principle should be applied is encapsulated as the fundamental inequality, which in broad terms states that the principle is valid when the data are random, relative to every contemplated hypothesis and also these hypotheses are random relative to the (universal) prior. The ideal principle states that the prior probability associated with the hypothesis should be given by the algorithmic universal probability, and the sum of the log universal probability of the model plus the log of the probability of the data given the model should be minimized. If we restrict the model class to finite sets then application of the ideal principle turns into Kolmogorov's minimal sufficient statistic. In general, we show that data compression is almost always the best strategy, both in model selection and prediction","Title: A Fast Quartet tree heuristic for hierarchical clustering\nYear: 2011\n\nAbstract: The Minimum Quartet Tree Cost problem is to construct an optimal weight tree from the 3(n4) weighted quartet topologies on n objects, where optimality means that the summed weight of the embedded quartet topologies is optimal (so it can be the case that the optimal tree embeds all quartets as nonoptimal topologies). We present a Monte Carlo heuristic, based on randomized hill-climbing, for approximating the optimal weight tree, given the quartet topology weights. The method repeatedly transforms a dendrogram, with all objects involved as leaves, achieving a monotonic approximation to the exact single globally optimal tree. The problem and the solution heuristic has been extensively used for general hierarchical clustering of nontree-like (non-phylogeny) data in various domains and across domains with heterogeneous data. We also present a greatly improved heuristic, reducing the running time by a factor of order a thousand to ten thousand. All this is implemented and available, as part of the CompLearn package. We compare performance and running time of the original and improved versions with those of UPGMA, BioNJ, and NJ, as implemented in the SplitsTree package on genomic data for which the latter are optimized.","Title: On the Simulation of Many Storage Heads by a Single One (Extended Abstract)\nYear: 1983\n\nAbstract: Each multitape Turing machine, of which the storage heads scan O(log n) distinct squares in each interval of n steps, for all n 1, can be real- time simulated by an oblivious one-head tape unit. There exist multitape Turing machines, e.g. the normal pushdown store, for which the fastest on- line simulation by an oblivious one- head tape unit requires (n n) time.","Title: Simple Wait-Free Multireader Registers\nYear: 2002\n\nAbstract: Multireader shared registers are basic objects used as communication medium in asynchronous concurrent computation. We propose a surprisingly simple and natural scheme to obtain several waitfree constructions of bounded 1-writer multireader registers from atomic 1-writer 1-reader registers, that is easier to prove correct than any previous construction. Our main construction is the first symmetric pure timestamp one that is optimal with respect to the worst-case local use of control bits; the other one is optimal with respect to global use of control bits; both are optimal in time.","Title: Conditional Kolmogorov complexity and universal probability\nYear: 2013\n\nAbstract: The Coding Theorem of L.A. Levin connects unconditional prefix Kolmogorov complexity with the universal semiprobability mass function. There are conditional versions referred to in several publications but as yet there exist no written proofs. Under the classic definition of conditional probability, there is no conditional version of the Coding Theorem. We give the appropriate definitions and provide complete proofs of the conditional version of the Coding Theorem.","Title: A lower bound on the average-case complexity of shellsort\nYear: 2000\n\nAbstract: We demonstrate an &Ohgr;(pn1+1/p ) lower bound on the average-case running time (uniform distribution) of p-pass Shellsort. This is the first nontrivial general lower bound for average-case Shellsort.","Title: Average-Case Analysis Using Kolmogorov Complexity\nYear: 1997\n\nAbstract: This expository paper demonstrates how to use Kolmogorov complexityto do the average-case analysis via four examples, and exhibits a surprisingproperty of the celebrated associated universal distribution. The four examplesare: average case analysis of Heapsort [17, 15], average nni-distance betweentwo binary rooted leave-labeled trees [20], compact routing in computernetworks [3], average-case analysis of an adder algorithm [4]. The propertyis that the average-case complexity of any...","Title: Nonsequentail Computation and Laws of Nature\nYear: 1986\n\nAbstract: Traditionally, computational complexity theory deals with sequential computations. In the computational models the underlying physics is hardly accounted for. This attitude has persisted in common models for parallel computations. Wrongly, as we shall argue, since the laws of physics intrude forcefully when we want to obtain realistic estimates of the performance of parallel or distributed algorithms. First, we shall explain why it is reasonable to abstract away from the physical details in sequential computations. Second, we show why certain common approaches in the theory of parallel complexity do not give useful information about the actual complexity of the parallel computation. Third, we give some examples of the interplay between physical considerations and actual complexity of distributed computations.","Title: New Applications of the Incompressibility Method.\nYear: 1999\n\nAbstract: The incompressibility method is an elementary yet powerful proof technique. It has been used successfully in many areas. To further demonstrate its power and elegance we exhibit new simple proofs using the incompressibility method.","Title: Algorithmic Arguments in Physics of Computation\nYear: 1995\n\nAbstract: . We show the usefulness of incompressibility arguments basedon Kolmogorov complexity in physics of computation by several examples.These include analysis of energy parsimonious `adiabatic\" computation,and scalability of network architectures.1 IntroductionIn [Shannon, 1948] C. Shannon formulated information theory dealing with theaverage number of bits required to communicate a message produced by a randomsource from a sender to a receiver who both agree on the ensemble ofpossible...","Title: Growth of Strings in Context Dependent Lindenmayer Systems\nYear: 1974\n\nAbstract: Growth functions of context dependent Lindenmayer systems are investigated. Bounds on the fastest and slowest growth in such systems are derived, and a method to obtain (P)D1L growth functions from (P)D2L growth functions is given. Closure of context dependent growth functions under several operations is studied with special emphasis on an application of the firing squad synchronization problem. It is shown that, although all growth functions of DILs using a one letter alphabet are DOL growth functions, there are growth functions of PDILs using a two letter alphabet which are not. Several open problems concerning the decidability of growth equivalence, growth type classification etc. of context dependent growth are shown to be undecidable. As a byproduct we obtain that the language equivalence of PDILs is undecidable and that a problem proposed by Varshavsky has a negative solution.","Title: The Power and Perils of MDL\nYear: 2006\n\nAbstract: We point out a potential weakness in the application of the celebrated minimum description length (MDL) principle for model selection. Specifically, it is shown that (although the index of the model class which actually minimizes a two-part code has many desirable properties) a model which has a shorter two- part code-length than another is not necessarily better (unless of course it achieves the global minimum). This is illustrated by an application to infer a grammar (DFA) from positive examples. We also analyze computability issues, and robustness under receding of the data. Generally, the classical approach is inadequate to express the goodness-of-fit of individual models for individual data sets. In practice however, this is precisely what we are interested in: both to express the goodness of a procedure and where and how it can fail. To achieve this practical goal, we paradoxically have to use the, supposedly impractical, vehicle of Kolmogorov complexity.","Title: On The Simulation Of Many Storage Heads By One\nYear: 1984\n\nAbstract: Each multitape Turing machine, of which the storage heads scan O ( log n ) distinct squares in each interval of n steps, for all n ⩾ 1, can be real-time simulated by an oblivious `one-head tape unit. To simulate but the normal pushdown store, on-line by an oblivious one-head tape unit, requires ω ( n √ n ) time. (This improves the known ω ( n log n ) lower bound for the on-line simulation of multitape Turing machines by oblivious one-tape Turing machines.)","Title: On a problem in the collective behavior of automata\nYear: 1976\n\nAbstract: Varshavsky defines the function L(n) as the maximum finite length of a configuration which can be grown from one activated automation in a linear cell space of identical finite state automata having n internal states. It is shown that L increases faster than any computable function, even if the flow of information in the linear cell space is restricted to one direction.","Title: On the Importance of Having an Identity or is Consensus Really Universal?\nYear: 2006\n\nAbstract: We show that Naming-the existence of distinct IDs known to all- is a necessary assumption of Herlihy's universality result for Consensus. We then show in a very precise sense that Naming is harder than Consensus and bring to the surface some important differences existing between popular shared memory models which usually remain unnoticed.","Title: Kolmogorov's structure functions and model selection\nYear: 2004\n\nAbstract: In 1974, Kolmogorov proposed a nonprobabilistic approach to statistics and model selection. Let data be finite binary strings and models be finite sets of binary strings. Consider model classes consisting of models of given maximal (Kolmogorov) complexity. The \"structure function\" of the given data expresses the relation between the complexity level constraint on a model class and the least log-cardinality of a model in the class containing the data. We show that the structure function determines all stochastic properties of the data: for every constrained model class it determines the individual best fitting model in the class irrespective of whether the \"true\" model is in the model class considered or not. In this setting, this happens with certainty, rather than with high probability as is in the classical case. We precisely quantify the goodness-of-fit of an individual model with respect to individual data. We show that-within the obvious constraints-every graph is realized by the structure function of some data. We determine the (un)computability properties of the various functions contemplated and of the \"algorithmic minimal sufficient statistic.\".","Title: Thermodynamics of computation and information distance\nYear: 1993\n\nAbstract: Intuitively, the minimal information distance between z and g is the length of the shortest program for a universal computer to transform z into v and y into z. This measure will be shown to be, up to a logarithmic additive term, equal to the maximum of the conditional Kolmogorov complexities El(z, y) = max{~(ylz), K(zly)}. Any reasonable distance to measure similarity of pictures should be an effectively approximable, symmetric, positive function of z and y satisfying a reasonable normalization condition and obeying the triangle inequal","Title: Turing Machines and Understanding Computational Complexity\nYear: 2012\n\nAbstract: We describe the Turing Machine, list some of its many influences on the theory of computation and complexity of computations, and illustrate its importance.","Title: A note on the recursive enumerability of some classes of recursively enumerable languages\nYear: 1978\n\nAbstract: An elementary proof is presented for the fact that the class of infinite recursive languages is not recursively enumerable. Its relevance for contemporary linguistics and computer science is explained.","Title: Efficient Simulations of Multicounter Machines\nYear: 1982\n\nAbstract: An oblivious 1-tape Turing machine can on-line simulate a multicounter machine in linear time and logarithmic space. This leads to a linear cost combinational logic network implementing the first n steps of a multicounter machine and also to a linear time/logarithmic space on-line simulation by an oblivious logarithmic cost RAM. An oblivious log *n-head tape unit can simulate the first n steps of a multicounter machine in real-time, which leads to a linear cost combinational logic network with a constant data rate.","Title: Distributed elections in an archimedean ring of processors\nYear: 2009\n\nAbstract: sphere has no magnitude,we cannot conceive it to bear any ratio whateverto the surface of the sphere.Archimedes, The Sand-Reckoner1. IntroductionWe address the issue of time in distributed systems. Undergenuinely reasonable assumptions about time in distributed systemsthere exist, for some problems, logically time-independentsolutions which are more efficient than achievable with################* This work was supported by the Stichting Mathematisch Centrum.+ Full Address: Centre ...","Title: Individual Communication Complexity: Extended Abstract\nYear: 2004\n\nAbstract: We initiate the theory of communication complexity of indi- vidual inputs held by the agents, rather than worst-case or average-case. We consider total, partial, and partially correct protocols, one-way versus two-way, with (not in this version) and without help bits.","Title: Optimality of wait-free atomic multiwriter variables\nYear: 1992\n\nAbstract: Known implementations of concurrent wait-free atomic shared multiwritervariables use Q(n) control bits per subvariable. It has been shownthat implementations of sequential time-stamp systems require W(n) controlbits per subvariable. We exhibit a sequential wait-free atomic sharedmultiwriter variable construction using log n control bits per subvariable.There arises the question of the optimality of concurrent implementationsof the same, and of weak time-stamp systems. We also show that...","Title: Space-efficient Routing Tables for Almost All Networks and the Incompressibility Method\nYear: 1999\n\nAbstract: We use the incompressibility method based on Kolmogorov complexity to determine the total number of bits of routing information for almost all network topologies. In most models for routing, for almost all labeled graphs, $\\Theta (n^2)$ bits are necessary and sufficient for shortest path routing. By \"almost all graphs\" we mean the Kolmogorov random graphs which constitute a fraction of 1 - 1/nc of all graphs on n nodes, where c 0 is an arbitrary fixed constant. There is a model for which the average case lower bound rises to $\\Omega(n^2 \\log n )$ and another model where the average case upper bound drops to $O(n \\log^2 n)$. This clearly exposes the sensitivity of such bounds to the model under consideration. If paths have to be short, but need not be shortest (if the stretch factor may be larger than 1), then much less space is needed on average, even in the more demanding models. Full-information routing requires $\\Theta (n^3)$ bits on average. For worst-case static networks we prove an $\\Omega(n^2 \\log n )$ lower bound for shortest path routing and all stretch factors","Title: Identification of Probabilities of Languages\nYear: 2012\n\nAbstract: We consider the problem of inferring the probability distribution associated with a language, given data consisting of an infinite sequence of elements of the languge. We do this under two assumptions on the algorithms concerned: (i) like a real-life algorothm it has round-off errors, and (ii) it has no round-off errors. Assuming (i) we (a) consider a probability mass function of the elements of the language if the data are drawn independent identically distributed (i.i.d.), provided the probability mass function is computable and has a finite expectation. We give an effective procedure to almost surely identify in the limit the target probability mass function using the Strong Law of Large Numbers. Second (b) we treat the case of possibly incomputable probabilistic mass functions in the above setting. In this case we can only pointswize converge to the target probability mass function almost surely. Third (c) we consider the case where the data are dependent assuming they are typical for at least one computable measure and the language is finite. There is an effective procedure to identify by infinite recurrence a nonempty subset of the computable measures according to which the data is typical. Here we use the theory of Kolmogorov complexity. Assuming (ii) we obtain the weaker result for (a) that the target distribution is identified by infinite recurrence almost surely; (b) stays the same as under assumption (i). We consider the associated predictions.","Title: A New Approach to Formal Language Theory by Kolmogorov Complexity (Preliminary Version)\nYear: 1989\n\nAbstract: We introduce Kolmogorov complexity as a new technique in Formal Language Theory. We give an alternative for pumping lemma(s) and a new characterization for regular languages. For the separation of deterministic contextfree languages and contextfree languages no pumping lemmas or any other general method was known. We give a first general technique to separate these classes, and illustrate its use on four examples previously requiring labourous ad hoc methods. The approach is also successful at the high end of the Chomsky hierarchy since one can quantify nonrecursiveness in terms of Kolmogorov complexity. We also give a new proof, using Kolmogorov complexity, of Yao and Rivest's result that k + 1 heads are better than k heads.","Title: On the Size of D0L Languages\nYear: 1974\n\nAbstract: Languages generated by monogenic (i.e. deterministic) context independent Lindenmayer systems (DOL systems) are investigated. Necessary and sufficient conditions are established under which the language generated by a DOL system is finite. Thus, sharp bounds on the cardinality of such a language are obtained. A feasible solution for the membership problem is given. The problems are solved of what is the minimum sized alphabet over which there is a DOL language of cardinality n and, conversely, what is the maximum sized finite DOL language over an alphabet of m letters. This in turn provides us with some number theoretic functions, interesting in their own right, of which several properties, interrelations and asymptotic approximations are derived.","Title: Average-Case Analysis via Incompressibility\nYear: 1997\n\nAbstract: We will demonstrate how to use Kolmogorov complexity to do the average-case analysis via some examples. These examples include: longest common subsequence problem and shortest common supersequence problem [9, 11], problems in computational geometry [14], average case analysis of Heapsort [19, 17], average nni-distance between two binary rooted leave-labeled trees [23], compact routing in computer networks.","Title: Average-Case Complexity of Shellsort\nYear: 1999\n\nAbstract: We prove a general lower bound on the average-case complexity of Shellsort: the average number of data-movements (and comparisons) made by a p-pass Shellsort for any incremental sequence is Ω(pn1+1=p) for every p. The proof method is an incompressibility argument based on Kolmogorov complexity. Using similar techniques, the average-case complexity of several other sorting algorithms is analyzed.","Title: Automatic Meaning Discovery Using Google\nYear: 2006\n\nAbstract: Abstract We present a new theory of relative semantics between objects, based on information distance and Kolmogorov,complexity. This theory is then applied to construct a method,to automatically extract the meaning,of words and phrases from the world-wide-web using Google page counts. The approach is novel in its unrestricted problem domain, simplicity of implementation, and manifestly ontological underpinnings. The world-wide-web is the largest database on earth, and the latent semantic context information entered by millions of independent users averages out to provide automatic meaning,of useful quality. We give examples to distinguish between,colors and numbers, cluster names of paintings by 17th century Dutch masters and names of books by English novelists, the ability to understand emergencies, and primes, and we demonstrate the ability to do a simple automatic English-Spanish translation. Finally, we use the WordNet database as an objective baseline against which to judge the performance,of our method. We conduct a massive randomized trial in binary classification using support vector machines to learn categories based on our Google distance, resulting in an a mean agreement of 87% with the expert crafted WordNet categories.","Title: How to share concurrent wait-free variables\nYear: 1996\n\nAbstract: Sharing data between multiple asynchronous users—each of which can atomically read and write the data—is a feature that may help to increase the amount of parallelism in distributed systems. An algorithm implementing this feature is presented. The main construction of an n-user atomic variable directly from single-writer, single-reader atomic variables uses O(n) control bits and O(n) accesses per Read/Write running in O(1) parallel time.","Title: On two-tape real-time computation and queues\nYear: 1984\n\nAbstract: A Turing machine with two storage tapes cannot simulate a queue in both real-time and with at least one storage tape head always within o(n) squares from the start square. This fact may be useful for showing that a two-head tape unit is more powerful in real-time than two one-head tape units, as is commonly conjectured.","Title: Average-Case Analysis of Algorithms Using Kolmogorov Complexity\nYear: 2000\n\nAbstract: Abstract: Analyzing the average-case complexity of algorithms is a very practical but verydifficult problem in computer science. In the past few years, we have demonstrated thatKolmogorov complexity is an important tool for analyzing the average-case complexityof algorithms. We have developed the incompressibility method [7]. In this paper, weuse several simple examples to further demonstrate the power and simplicity of suchmethod. We prove bounds on the average-case number of stacks (queues)...","Title: Big omega versus the wild functions\nYear: 1984\n\nAbstract: The question of the desirable properties and proper definitions of the Order-of-Magnitude symbols, in particular O and Ø, is addressed once more. The definitions proposed are chosen for complementary mathematical properties, rather than for similarity of form.","Title: On a Generalized Ruin Problem\nYear: 2001\n\nAbstract: We consider a natural generalization of the classical ruin problem to more than two parties. Our \"ruin\" problem, which we will call the (k, I)-game, starts with k players each having I units as its initial capital. At each round of the game, all remaining k′ players pay 1/k′th unit as game fee, play the game, and one of the players wins and receives the combined game fees of 1 unit. A player who cannot pay the next game fee goes bankrupt, and the game terminates when all players but one are bankrupt.We analyze the length of the game, that is, the number of rounds executed until the game terminates, and give upper and lower bounds for the expected game length.","Title: On inverse deterministic pushdown transductions\nYear: 1978\n\nAbstract: Classes of source languages which can be mapped by a deterministic pushdown (DPDA) transduction into a given object language (while their complement is mapped into the complement of the object language) are studied. Such classes of source languages are inverse DPDA transductions of the given object language. Similarly for classes of object languages. The inverse DPDA transductions of the Dyck sets are studied in greater detail: they can be recognized in deterministic storage (log n)' but do not comprise all context free languages; their emptiness problem is unsolvable and their closure under homomorphism constitutes the r.e. sets. For each object language L we can exhibit a storage hardest language for the class of inverse DPDA transductions of L; similarly for the classes of regular, deterministic context free, and context free object languages. Last, we classify the classes of inverse DPDA transductions of the regular, deterministic context free, context free and deterministic context sensitive languages.","Title: Physics and the New Computation\nYear: 1995\n\nAbstract: New computation devices increasingly depend on particular physical properties rather than on logical organization alone as used to be the case in conventional technologies. This has impact on the synthesis and analysis of algorithms and the computation models on which they are to run. Therefore, scientists working in these areas will have to understand and apply physical law in their considerations. We discuss three cases in some detail: interconnect length and communication in massive multicomputers which depend on the geometry of space and speed of light; energy dissipation and reversible (adiabatic) computation which depend on thermodynamics; and quantum coherent parallel computing which depends on quantum mechanics.","Title: Philosophical Issues in Kolmogorov Complexity\nYear: 1992\n\nAbstract: this article at a conceptuallevel, it is sufficient to know that Kolmogorov complexity of a finite string xis simply the length of the shortest program, say in FORTRAN","Title: On Prediction by Data Compression\nYear: 1997\n\nAbstract: . Traditional wisdom has it that the better a theory compressesthe learning data concerning some phenomenon under investigation,the better we learn, generalize, and the better the theory predictsunknown data. This belief is vindicated in practice but apparentlyhas not been rigorously proved in a general setting. Making these ideasrigorous involves the length of the shortest effective description of anindividual object: its Kolmogorov complexity. In a previous paper wehave shown that...","Title: On the logical depth function\nYear: 2013\n\nAbstract: For a finite binary string $x$ its logical depth $d$ for significance $b$ is the shortest running time of a program for $x$ of length $K(x)+b$. There is another definition of logical depth. We give a new proof that the two versions are close. There is an infinite sequence of strings of consecutive lengths such that for every string there is a $b$ such that incrementing $b$ by 1 makes the associated depths go from incomputable to computable. The maximal gap between depths resulting from incrementing appropriate $b$'s by 1 is incomputable. The size of this gap is upper bounded by the Busy Beaver function. Both the upper and the lower bound hold for the depth with significance 0. As a consequence, the minimal computation time of the associated shortest programs rises faster than any computable function but not so fast as the Busy Beaver function.","Title: Rate distortion and denoising of individual data using Kolmogorov complexity\nYear: 2010\n\nAbstract: We examine the structure of families of distortion balls from the perspective of Kolmogorov complexity. Special attention is paid to the canonical rate-distortion function of a source word which returns the minimal Kolmogorov complexity of all distortion balls containing that word subject to a bound on their cardinality. This canonical rate-distortion function is related to the more standard algorithmic rate-distortion function for the given distortion measure. Examples are given of list distortion, Hamming distortion, and Euclidean distortion. The algorithmic rate-distortion function can behave differently from Shannon's rate-distortion function. To this end, we show that the canonical rate-distortion function can and does assume a wide class of shapes (unlike Shannon's); we relate low algorithmic mutual information to low Kolmogorov complexity (and consequently suggest that certain aspects of the mutual information formulation of Shannon's rate-distortion function behave differently than would an analogous formulation using algorithmic mutual information); we explore the notion that low Kolmogorov complexity distortion balls containing a given word capture the interesting properties of that word (which is hard to formalize in Shannon's theory) and this suggests an approach to denoising.","Title: Normalized Compression Distance of Multiples\nYear: 2012\n\nAbstract: Normalized compression distance (NCD) is a parameter-free similarity measure based on compression. The NCD between pairs of objects is not sufficient for al l applications. We propose an NCD of finite multisets (multiples) of objacts that is metric and is bette r for many applications. Previously, attempts to obtain such an NCD failed. We use the theoretical notion of Kolmogorov complexity that for practical purposes is approximated from above by the length of the compressed version of the file involved, using a real-world compression program. We applied the new NCD for multiples to retinal progenitor cell questions that were earlier treated with the pairwise NCD. Here we get significantly better results. We also applied the NCD for multiples to synthetic time sequence data. The preliminary results are as good as nearest neighbor Euclidean classifier. Index Terms— Normalized compression distance, multisets or multiples, pattern recognition, data mining, similarity, Kolmogorov complexity, retinal progenitor cell classification, synthetic data classification","Title: Kolmogorov's Structure Functions with an Application to the Foundations of Model Selection\nYear: 2002\n\nAbstract: In 1974 Kolmogorov proposed a non-probabilistic approach to statistics, an individual combinatorial relation between the data and its model. We vindicate, for the first time, the rightness of the original \"structure function\", proposed by Kolmogorov: minimizing the data-to-model code length (finding the ML estimator or MDL estimator), in a class of contemplated models of prescribed maximal (Kolmogorov) complexity, always results in a model of best fit (expressed as minimal randomness deficiency). We show that both the structure function and the minimum randomness deficiency function can assume all shapes over their full domain (improving an old result of L.A. Levin and both an old and a recent one of V.V. Vyugin). We determine the (un)computability properties of the various functions and \"algorithmic sufficient statistic.\"","Title: Approximation of the two-part MDL code\nYear: 2009\n\nAbstract: Approximation of the optimal two-part minimum description length (MDL) code for given data, through successive monotonically length-decreasing two-part MDL codes, has the following properties: i) computation of each step may take arbitrarily long; ii) we may not know when we reach the optimum, or whether we will reach the optimum at all; iii) the sequence of models generated may not monotonically improve the goodness of fit; but iv) the model associated with the optimum has (almost) the best goodness of fit. To express the practically interesting goodness of fit of individual models for individual data sets we have to rely on Kolmogorov complexity.","Title: Average case complexity under the universal distribution equals worst-case complexity\nYear: 1992\n\nAbstract: The average complexity of any algorithm whatsoever under the universal distribution is of the same order of magnitude as the worst-case complexity. This holds both for time complexity and for space complexity. To focus our discussion, we use as illustrations the particular case of sorting algorithms, and the general case of the average case complexity of NP-complete problems.","Title: Kolmogorov complexity arguments in combinatorics\nYear: 1994\n\nAbstract: The utility of a Kolmogorov complexity method in combinatorialtheory is demonstrated by several examples.1 IntroductionProbabilistic arguments in combinatorial theory, as used by Erdos and Spencer[5], are usually aimed at establishing the existence of an object, in a nonconstructivesense. It is ascertained that a certain member of a class hasa certain property, without actually exhibiting that object. Usually, themethod proceeds by exhibiting a random process which produces the...","Title: Deterministic Lindenmayer languages, nonterminals and homomorphisms\nYear: 1976\n\nAbstract: Lindenmayer systems are a class of parallel rewriting systems originally introduced to model the growth and development of filamentous organisms. Families of languages generated by deterministic Lindenmayer systems (i.e., those in which each string has a unique successor) are investigated. In particular, the use of nonterminals, homomorphisms, and the combination of these are studied for deterministic Lindenmayer systems using one-sided context (D1Ls) and two-sided context (D2Ls). Languages obtained from Lindenmayer systems by the use of nonterminals are called extensions. Typical results are: The closure under letter-to-letter homomorphism of the family of extensions of D1L languages is equal to the family of recursively enumerable languages, although the family of extensions of D1L languages does not even contain all regular languages. Let P denote the restriction that the system does not rewrite a letter as the empty word. The family of extensions of PD2L languages is equal to the family of languages accepted by deterministic linear bounded automata. The closure under nonerasing homomorphism of the family of extensions of PD1L languages does not even contain languages like { a 1 , a 2 ,⋯, a n } ∗ --{λ}, n ⩾2 . The closure of the family of PD1L languages under homomorphisms which map a letter either to itself or to the empty word is equal to the family of recursively enumerable languages. Strict inclusion results follow from necessary conditions for a language to be in one of the considered families. By stating the results in their strongest form, the paper contains a systematic classification of the effect of nonterminals, letter-to-letter homomorphisms, nonerasing homomorphisms and homomorphisms for all the basic types of deterministic Lindenmayer systems using context.","Title: An n1.618 lower bound on the time to simulate one queue or two pushdown stores by one tape\nYear: 1985\n\nAbstract: To simulate two pushdown stores, or one queue, on-line by a one-head tape unit requires Ω(n1.618) time.","Title: Distributed match-making for processes in computer networks (preliminary version)\nYear: 1986\n\nAbstract: In the very large multiprocessor systems and, on a grander scale, computer networks now emerging, processes are not tied to fixed processors but run on processors taken from a pool of processors. Processors are released when a process dies, migrates or when the process crashes. In distributed operating systems using the service concept, processes can be clients asking for a service, servers giving a service or both. Establishing communication between a process asking for a service and a process giving that service, without centralized control in a distributed environment with mobile processes, constitutes the problem of distributed match-making. Logically, such a match-making phase precedes routing in store-and-forward computer networks of this type. Algorithms for distributed match-making are developed and their complexity is investigated in terms of message passes and in terms of storage needed. The theoretical limitations of distributed match-making are established, and the techniques are applied to several network topologies.","Title: The Generalized Universal Law of Generalization\nYear: 2001\n\nAbstract: It has been argued by Shepard that there is a robust psychological law that relates the distance between a pair of items in psychological space and the probability that they will be perceived as similar. Specifically, this probability is a negative exponential function of the distance between the pair of items. In experimental contexts, distance is typically defined in terms of a multidimensional space—but this assumption seems unlikely to hold for complex stimuli. We show that, nonetheless, the Universal Law of Generalization can be derived in the more complex setting of arbitrary stimuli, using a much more universal measure of distance. This universal distance is defined as the length of the shortest program that transforms the representations of the two items of interest into one another: The algorithmic information distance. It is universal in the sense that it minorizes every computable distance: It is the smallest computable distance. We show that the Universal Law of Generalization holds with probability going to one—provided the probabilities concerned are computable. We also give a mathematically more appealing form of the Universal Law.","Title: A New Quartet Tree Heuristic for Hierarchical Clustering\nYear: 2006\n\nAbstract: We consider the problem of constructing an an optimal-weight tree from the 3*(n choose 4) weighted quartet topologies on n objects, where optimality means that the summed weight of the embedded quartet topologiesis optimal (so it can be the case that the optimal tree embeds all quartets as non-optimal topologies). We present a heuristic for reconstructing the optimal-weight tree, and a canonical manner to derive the quartet-topology weights from a given distance matrix. The method repeatedly transforms a bifurcating tree, with all objects involved as leaves, achieving a monotonic approximation to the exact single globally optimal tree. This contrasts to other heuristic search methods from biological phylogeny, like DNAML or quartet puzzling, which, repeatedly, incrementally construct a solution from a random order of objects, and subsequently add agreement values.","Title: Time-bounded incompressibility of compressible strings and sequences\nYear: 2009\n\nAbstract: For every total recursive time bound t, a constant fraction of all compressible (low Kolmogorov complexity) strings is t-bounded incompressible (high time-bounded Kolmogorov complexity); there are uncountably many infinite sequences of which every initial segment of length n is compressible to logn yet t-bounded incompressible below 14n-logn; and there is a countably infinite number of recursive infinite sequences of which every initial segment is similarly t-bounded incompressible. These results and their proofs are related to, but different from, Barzdins's lemma.","Title: Average-Case Complexity of Shellsort (Preliminary version)\nYear: 1999\n\nAbstract: We prove a general lower bound on the average-case com- plexity of Shellsort: the average number of data-movements (and com- parisons) made by a p-pass Shellsort for any incremental sequence is (pn1+1/p) for every p. The proof method is an incompressibility ar- gument based on Kolmogorov complexity. Using similar techniques, the average-case complexity of several other sorting algorithms is analyzed.","Title: Two decades of applied Kolmogorov complexity: in memoriam Andrei Nikolaevich Kolmogorov 1903-87\nYear: 1988\n\nAbstract: The authors provide an introduction to the main ideas of Kolmogorov complexity and survey the wealth of useful applications of this notion. It is based on a theory of information content of strings, intuitively, that the amount of information in a finite string is the size (i.e. number of bits) of the smallest program that, started with a blank memory, computes the string and then terminates. The following are treated: (1) application of the fact that some strings are compressible; this includes a strong version of Godel's incompleteness theorem; (2) lower-bound arguments that rest on application of the fact that certain strings cannot be compressed at all; applications range from Turing machines to electronic chips; (3) probability theory and a priori probability; applications range from foundational issues to the theory of learning and inductive inference in artificial intelligence. (4) Resource-bounded Kolmogorov complexity; applications range from NP-completeness and the P versus NP question to cryptography","Title: Linear Time Simulation of Multihead Turing Machines with Head-to-Head Jumps\nYear: 1977\n\nAbstract: The main result of this paper shows that, given a Turing machine with several read-write heads per tape and which has the additional one move shift operation \"shift a given head to the position of some other given head\", one can effectively construct a multitape Turing machine with a single read-write head per tape which simulates it in linear time; i.e. if the original machine operates in time T(n), then the simulating machine will operate in time cT(n), for some constant c.","Title: Time and Space Bounds for Reversible Simulation\nYear: 2001\n\nAbstract: We prove a general upper bound on the tradeoff between time and space that suffices for the reversible simulation of irreversible computation. Previously, only simulations using exponential time or quadratic space were known. The tradeoff shows for the first time that we can simultaneously achieve subexponential time and subquadratic space. The boundary values are the exponential time with hardly any extra space required by the Lange-McKenzie-Tapp method and the (log 3)th power time with square space required by the Bennett method. We also give the first general lower bound on the extra storage space required by general reversible simulation. This lower bound is optimal in that it is achieved by some reversible simulations.","Title: Normalized Google Distance of Multisets with Applications.\nYear: 2013\n\nAbstract: Normalized Google distance (NGD) is a relative semantic distance based on the World Wide Web (or any other large electronic database, for instance Wikipedia) and a search engine that returns aggregate page counts. The earlier NGD between pairs of search terms (including phrases) is not sufficient for all applications. We propose an NGD of finite multisets of search terms that is better for many applications. This gives a relative semantics shared by a multiset of search terms. We give applications and compare the results with those obtained using the pairwise NGD. The derivation of NGD method is based on Kolmogorov complexity.","Title: Wait-free Test-and-Set (Extended Abstract)\nYear: 1992\n\nAbstract: )Yehuda Afek1;4, Eli Gafni2?, John Tromp3, and Paul M.B. Vitanyi3;51Computer Science Department, Tel-Aviv University, Ramat-Aviv 69978 Israel.2Department of Computer Science, University of California, Los Angeles CA 90024.3Centrum voor Wiskunde en Informatica, Kruislaan 413, 1098 SJ Amsterdam, TheNetherlands.4AT&T Bell Laboratories, 600 Mountain Avenue, Murray Hill, NJ 07974.5Faculteit Wiskunde en Informatica, Universiteit van Amsterdam.Abstract. This...","Title: Nonapproximability of the normalized information distance\nYear: 2009\n\nAbstract: Normalized information distance (NID) uses the theoretical notion of Kolmogorov complexity, which for practical purposes is approximated by the length of the compressed version of the file involved, using a real-world compression program. This practical application is called 'normalized compression distance' and it is trivially computable. It is a parameter-free similarity measure based on compression, and is used in pattern recognition, data mining, phylogeny, clustering, and classification. The complexity properties of its theoretical precursor, the NID, have been open. We show that the NID is neither upper semicomputable nor lower semicomputable.","Title: The Simple Roots of Real-Time Computation Hierarchies (Preliminary Version)\nYear: 1984\n\nAbstract: A BLAH machine is any memory device that can be simulated in real-time by a multitape Turing machine and such that a multiBLAH machine can real-time simulate a pushdown store. A multiBLAH machine consists of a finite control connected to an input terminal and an output terminal and one or more copies of the BLAH memory unit. It is shown that a (k+1)-BLAH machine is more powerful in real-time than a k-BLAH machine, for each k. Thus the hierarchies, within the real-time definable computations, are proper and smooth, that is, adding a device always increases power. It also turns out that all real-time hierarchy results in this vein are simple corollaries of a single root: the real-time hierarchy of multipushdown store machines. As examples of such new results we mention that in real-time, k+1 tape-units with a fast rewind square are more powerful than k such units; that (k+1)-head tape-units with fast rewind squares are more powerful than k-head tape-units with fast rewind squares; that (k+1)-dequeue machines are more powerful than k-dequeue machines; and that (k+1)-concatenable-dequeue machines are more powerful than k-concatenable-dequeue machines.","Title: On Logical Depth and the Running Time of Shortest Programs.\nYear: 2013\n\nAbstract: The logical depth with significance $b$ of a finite binary string $x$ is the shortest running time of a binary program for $x$ that can be compressed by at most $b$ bits. There is another definition of logical depth. We give two theorems about the quantitative relation between these versions: the first theorem concerns a variation of a known fact with a new proof, the second theorem and its proof are new. We select the above version of logical depth and show the following. There is an infinite sequence of strings of increasing length such that for each $j$ there is a $b$ such that the logical depth of the $j$th string as a function of $j$ is incomputable (it rises faster than any computable function) but with $b$ replaced by $b+1$ the resuling function is computable. Hence the maximal gap between the logical depths resulting from incrementing appropriate $b$'s by 1 rises faster than any computable function. All functions mentioned are upper bounded by the Busy Beaver function. Since for every string its logical depth is nonincreasing in $b$, the minimal computation time of the shortest programs for the sequence of strings as a function of $j$ rises faster than any computable function but not so fast as the Busy Beaver function.","Title: Statistical properties of finite sequences with high Kolmogorov complexity\nYear: 1994\n\nAbstract: We investigate to what extent finite binary sequences with highKolmogorov complexity are normal (all blocks of equal length occurequally frequent), and the maximal length of all-zero or all-one runswhich occur with certainty.1 IntroductionEach individual infinite sequence generated by a (12;12 ) Bernoulli process (flippinga fair coin) has (with probability 1) the property that the relative frequencyof zeros in an initial n-length segment goes to12 for n goes to...","Title: Normalized Web Distance and Word Similarity\nYear: 2009\n\nAbstract: There is a great deal of work in cognitive psychology, linguistics, and\ncomputer science, about using word (or phrase) frequencies in context in text\ncorpora to develop measures for word similarity or word association, going back\nto at least the 1960s. The goal of this chapter is to introduce the\nnormalizedis a general way to tap the amorphous low-grade knowledge available\nfor free on the Internet, typed in by local users aiming at personal\ngratification of diverse objectives, and yet globally achieving what is\neffectively the largest semantic electronic database in the world. Moreover,\nthis database is available for all by using any search engine that can return\naggregate page-count estimates for a large range of search-queries. In the\npaper introducing the NWD it was called `normalized Google distance (NGD),' but\nsince Google doesn't allow computer searches anymore, we opt for the more\nneutral and descriptive NWD. web distance (NWD) method to determine similarity\nbetween words and phrases. It","Title: Square time is optimal for simulation of one pushdown store or one queue by an oblivious one-head tape unit\nYear: 1985\n\nAbstract: To simulate a pushdown store or queue on-line by an oblivious one-head tape unit takes at least square time. Since each multitape Turing machine can be trivially simulated by an oblivious one-head tape unit in square time, this result its optimal.","Title: On The Average-Case Complexity of Shellsort.\nYear: 2018\n\nAbstract: We prove a lower bound expressed in the increment sequence on the average-case complexity of the number of inversions of Shellsort. This lower bound is sharp in every case where it could be checked. A special case of this lower bound yields the general Jiang-Li-Vitanyi lower bound. We obtain new results, for example, determining the average-case complexity precisely in the Yao-Janson-Knuth 3-pass case.","Title: Exact Expression For Information Distance.\nYear: 2014\n\nAbstract: Information distance can be defined not only between two strings but also in a finite multiset of strings of cardinality greater than two. We determine a best upper bound on the information distance. It is exact, since the upper bound on the information distance for all multisets is the same as the lower bound for infinitely many multisets of each of infinitely many cardinalities, up to a constant...","Title: Web Similarity.\nYear: 2015\n\nAbstract: Normalized web distance (NWD) is a similarity or normalized semantic distance based on the World Wide Web or any other large electronic database, for instance Wikipedia, and a search engine that returns reliable aggregate page counts. For sets of search terms the NWD gives a similarity on a scale from 0 (identical) to 1 (completely different). The NWD approximates the similarity according to all (upper semi)computable properties. We develop the theory and give applications. The derivation of the NWD method is based on Kolmogorov complexity.","Title: Three approaches to the quantitative definition of information in an individual pure quantum state\nYear: 2000\n\nAbstract: In analogy of classical Kolmogorov complexity we de- velop a theory of the algorithmic information in bits con- tained in any one of continuously many pure quantum states: quantum Kolmogorov complexity. Classical Kol- mogorov complexity coincides with the new quantum Kol- mogorov complexity restricted to the classical domain. Quantum Kolmogorov complexity is upper bounded and can be effectively approximated from above. With high probability a quantum object is incompressible. There are two alternative approaches possible: to define the complex- ity as the length of the shortest qubit program that effec- tively describes the object, and to use classical descriptions with computable real parameters.","Title: Language Learning From Positive Evidence, Reconsidered: A Simplicity-Based Approach.\nYear: 2013\n\nAbstract: Children learn their native language by exposure to their linguistic and communicative environment, but apparently without requiring that their mistakes be corrected. Such learning from positive evidence has been viewed as raising logical problems for language acquisition. In particular, without correction, how is the child to recover from conjecturing an over-general grammar, which will be consistent with any sentence that the child hears? There have been many proposals concerning how this logical problem can be dissolved. In this study, we review recent formal results showing that the learner has sufficient data to learn successfully from positive evidence, if it favors the simplest encoding of the linguistic input. Results include the learnability of linguistic prediction, grammaticality judgments, language production, and form-meaning mappings. The simplicity approach can also be scaled down to analyze the learnability of specific linguistic constructions, and it is amenable to empirical testing as a framework for describing human language acquisition.","Title: Normalized Compression Distance of Multisets with Applications\nYear: 2015\n\nAbstract: Pairwise normalized compression distance (NCD) is a parameter-free, feature-free, alignment-free, similarity metric based on compression. We propose an NCD of multisets that is also metric. Previously, attempts to obtain such an NCD failed. For classification purposes it is superior to the pairwise NCD in accuracy and implementation complexity. We cover the entire trajectory from theoretical under...","Title: On the rate of decrease in logical depth.\nYear: 2017\n\nAbstract: The logical depth with significance b of a string x is the shortest running time of a program for x that can be compressed by at most b bits. Another definition is based on algorithmic probability. We give a simple new proof for the known relation between the two definitions. We also prove the following: Given a string we can consider the maximal decrease in logical depth when the significance parameter increases by 1. There exists a sequence of strings of lengths n=1,2,…, such that this maximal decrease as a function of n rises faster than any computable function but not as fast as the Busy Beaver function. This holds also for the computation times of the shortest programs of these strings.","Title: Identification of Probabilities.\nYear: 2017\n\nAbstract: Within psychology, neuroscience and artificial intelligence, there has been increasing interest in the proposal that the brain builds probabilistic models of sensory and linguistic input: that is, to infer a probabilistic model from a sample. The practical problems of such inference are substantial: the brain has limited data and restricted computational resources. But there is a more fundamental question: is the problem of inferring a probabilistic model from a sample possible even in principle? We explore this question and find some surprisingly positive and general results. First, for a broad class of probability distributions characterized by computability restrictions, we specify a learning algorithm that will almost surely identify a probability distribution in the limit given a finite i.i.d. sample of sufficient but unknown length. This is similarly shown to hold for sequences generated by a broad class of Markov chains, subject to computability assumptions. The technical tool is the strong law of large numbers. Second, for a large class of dependent sequences, we specify an algorithm which identifies in the limit a computable measure for which the sequence is typical, in the sense of Martin-Löf (there may be more than one such measure). The technical tool is the theory of Kolmogorov complexity. We analyze the associated predictions in both cases. We also briefly consider special cases, including language learning, and wider theoretical implications for psychology.","Title: Inductive reasoning and Kolmogorov complexity\nYear: 1989\n\nAbstract: The inductive reasoning concepts of R.J. Solomonoff (Inf. Control. vol.7, p.1-22, 224-254, 1964) are considered. The thesis is developed that Solomonoff's method is fundamental in the sense that many other induction principles can be viewed as particular ways to obtain computable approximations to it. This is demonstrated explicitly in the cases of E.M. Gold's (1967, 1978) paradigm for inductive inference, L.G. Valiant's (1984) learning (by adding computational requirements), J. Rissanen's (1982) minimum description length principle, Fisher's maximum-likelihood principle (J. Rissanen, 1982), and E.T. Jayne's (1968, 1982) maximum entropy principle. Several new theorems and derivations to this effect are presented. What can and cannot be learned in terms of Kolmogorov complexity is delimited, and an experiment in machine learning of handwritten characters is described","Title: Logical depth for reversible Turing machines with an application to the rate of decrease in logical depth for general Turing machines\nYear: 2019\n\nAbstract: The logical depth of a reversible Turing machine equals the shortest running time of a shortest program for it. This is applied to show that the result in [1] is valid notwithstanding the error noted in Corrigendum [7].","Title: Fast Phylogeny of SARS-CoV-2 by Compression\nYear: 2022\n\nAbstract: The compression method to assess similarity, in the sense of having a small normalized compression distance (NCD), was developed based on algorithmic information theory to quantify the similarity in files ranging from words and languages to genomes and music pieces. It has been validated on objects from different domains always using essentially the same software. We analyze the whole-genome phylogeny and taxonomy of the SARS-CoV-2 virus, which is responsible for causing the COVID-19 disease, using the alignment-free compression method to assess similarity. We compare the SARS-CoV-2 virus with a database of over 6500 viruses. The results suggest that the SARS-CoV-2 virus is closest in that database to the RaTG13 virus and rather close to the bat SARS-like coronaviruses bat-SL-CoVZXC21 and bat-SL-CoVZC45. Over 6500 viruses are identified (given by their registration code) with larger NCDs. The NCDs are compared with the NCDs between the mtDNA of familiar species. We address the question of whether pangolins are involved in the SARS-CoV-2 virus. The compression method is simpler and possibly faster than any other whole-genome method, which makes it the ideal tool to explore phylogeny. Here, we use it for the complex case of determining this similarity between the COVID-19 virus, SARS-CoV-2 and many other viruses. The resulting phylogeny and taxonomy closely resemble earlier results from by alignment-based methods and a machine-learning method, providing the most compelling evidence to date for the compression method, showing that one can achieve equivalent results both simply and quickly."],"string":"[\n \"Title: Registers\\nYear: 2006\\n\\nAbstract: Entry in: Encyclopedia of Algorithms, Ming-Yang Kao, Ed., Springer, Toappear. Synonyms: Wait-free registers, wait-free shared variables, asynchronouscommunication hardware. Problem Definition: Consider a system of asynchronousprocesses that communicate among themselves by only executing read and writeoperations on a set of shared variables (also known as shared registers). Thesystem has no global clock or other synchronization primitives.\",\n \"Title: Randomness\\nYear: 2001\\n\\nAbstract: Here we present in a single essay a combination and completion of the severalaspects of the problem of randomness of individual objects which of necessityoccur scattered in our texbook \\\"An Introduction to Kolmogorov Complexity andIts Applications\\\" (M. Li and P. Vitanyi), 2nd Ed., Springer-Verlag, 1997.\",\n \"Title: Algorithmic Chaos\\nYear: 2003\\n\\nAbstract: Many physical theories like chaos theory are fundamentally concerned with the\\nconceptual tension between determinism and randomness. Kolmogorov complexity\\ncan express randomness in determinism and gives an approach to formulate\\nchaotic behavior.\",\n \"Title: Universal Similarity\\nYear: 2005\\n\\nAbstract: We survey a new area of parameter-free simi- larity distance measures useful in data-mining, pattern recog- nition, learning and automatic semantics extraction. Given a family of distances on a set of objects, a distance is universal up to a certain precision for that family if it minorizes every distance in the family between every two objects in the set, up to the stated precision (we do not require the universal dis- tance to be an element of the family). We consider similarity distances for two types of objects: literal objects that as such contain all of their meaning, like genomes or books, and names for objects. The latter may have literal embodyments like the first type, but may also be abstract like \\\"red\\\" or \\\"christian- ity.\\\" For the first type we consider a family of computable dis- tance measures corresponding to parameters expressing simi- larity according to particular features between pairs of literal objects. For the second type we consider similarity distances generated by web users corresponding to particular semantic relations between the (names for) the designated objects. For both families we give universal similarity distance measures, incorporating all particular distance measures in the family. In the first case the universal distance is based on compression and in the second case it is based on Google page counts re- lated to search terms. In both cases experiments on a massive scale give evidence of the viability of the approaches.\",\n \"Title: Relativized Obliviousness\\nYear: 1980\\n\\nAbstract: Relativized obliviousness is introduced to capture the intuitive idea, that some problems allow fastest computations which are more oblivious than do other problems, without any of such computations being oblivious in the standard sense. It is shown that each increase in the obliviousness of an algorithm (in several different well-defined meanings), for the solution of some problems, may necessarily require an increase in computation time from T(n) steps to T(n) log T(n) steps. There is, however, no problem for which a total oblivious algorithm requires more than order T(n) log T(n) steps, if the best algorithm for it runs in T(n) steps. We use on-line Turing machines as model of computation.\",\n \"Title: Compression-Based Similarity\\nYear: 2011\\n\\nAbstract: First we consider pair-wise distances for literal objects consisting of finite binary files. These files are taken to contain all of their meaning, like genomes or books. The distances are based on compression of the objects concerned, normalized, and can be viewed as similarity distances. Second, we consider pair-wise distances between names of objects, like \\\"red\\\" or \\\"christianity.\\\" In this case the distances are based on searches of the Internet. Such a search can be performed by any search engine that returns aggregate page counts. We can extract a code length from the numbers returned, use the same formula as before, and derive a similarity or relative semantics between names for objects. The theory is based on Kolmogorov complexity. We test both similarities extensively experimentally.\",\n \"Title: Atomic Multireader Register\\nYear: 1987\\n\\nAbstract: We give implementations for atomic, shared, asynchronous, wait-free registers: (i) A new implemenfation of an atomic, 1-writer, 1-reader, b-bit register from O (b) safe, boolean registers (i.e., from scratch). The solution uses neither repeated writing of the input nor repeated reading of the output. (ii) An implementation of an atomic, 1-writer, n-reader, multibit register from O(n\\n2) atomic, 1-writer, 1-reader, multibit registers. Both constructions rely on the same idea. In a sense (ii) is a generalization of (i). These results show how to construct atomic, multireader registers from — basically — elementary hardware like flip-flops.\",\n \"Title: Individual communication complexity\\nYear: 2007\\n\\nAbstract: We initiate the theory of communication complexity of individual inputs held by the agents. This contrasts with the usual communication complexity model, where one counts the amount of communication for the worst-case or the average-case inputs. The individual communication complexity gives more information (the worst-case and the average-case can be derived from it but not vice versa) and may in some cases be of more interest. It is given in terms of the Kolmogorov complexities of the individual inputs. There are different measures of communication complexity depending on whether the protocol is guaranteed to be correct for all inputs or not, and whether there's one-way or two-way communication. Bounds are provided for the communication of specific functions and connections between the different communication measures are shown. Some counter-intuitive results: for deterministic protocols that need to communicate Bob's input to Alice they need to communicate all of Bob's input (rather than the information difference with Alice's input), and there are so-called ''non-communicable'' inputs.\",\n \"Title: Counting is easy\\nYear: 2001\\n\\nAbstract: For any fixed k, a remarkably simple single-tape Turing machine can simulate k independent counters in real time.\",\n \"Title: Distributed Match-Making\\nYear: 1988\\n\\nAbstract: In many distributed computing environments, processes are concurrently executed by nodes in a store- and-forward communication\\n network. Distributed control issues as diverse as name server, mutual exclusion, and replicated data management involve making\\n matches between such processes. We propose a formal problem called “distributed match-making” as the generic paradigm. Algorithms\\n for distributed match-making are developed and the complexity is investigated in terms of messages and in terms of storage\\n needed. Lower bounds on the complexity of distributed match-making are established. Optimal algorithms, or nearly optimal\\n algorithms, are given for particular network topologies.\",\n \"Title: Normalized Information Distance\\nYear: 2008\\n\\nAbstract: The normalized information distance is a universal distance measure for objects of all kinds. It is based on Kolmogorov complexity and thus uncomputable, but there are ways to utilize it. First, compression algorithms can be used to approximate the Kolmogorov complexity if the objects have a string representation. Second, for names and abstract concepts, page count statistics from the World Wide Web can be used. These practical realizations of the normalized information distance can then be applied to machine learning tasks, expecially clustering, to perform feature-free and parameter-free data mining. This chapter discusses the theoretical foundations of the normalized information distance and both practical realizations. It presents numerous examples of successful real-world applications based on these distance measures, ranging from bioinformatics to music clustering to machine translation.\",\n \"Title: On Empirical Entropy\\nYear: 2011\\n\\nAbstract: We propose a compression-based version of the empirical entropy of a finite string over a finite alphabet. Whereas previously one considers the naked entropy of (possibly higher order) Markov processes, we consider the sum of the description of the random variable involved plus the entropy it induces. We assume only that the distribution involved is computable. To test the new notion we compare the Normalized Information Distance (the similarity metric) with a related measure based on Mutual Information in Shannon's framework. This way the similarities and differences of the last two concepts are exposed.\",\n \"Title: Optimal routing tables\\nYear: 1996\\n\\nAbstract: The optimal space used to represent routing schemes incommunication networks is established, both for worst-casestatic networks and on the average for all static networks.Several factors may influence the cost of representing a routingscheme for a particular network. It is therefore unavoidablethat we first describe several reasonable models in whichto measure this cost. Failure to do so in the past has obfuscatedprevious results.We show that, in most models, for almost all graphs...\",\n \"Title: The similarity metric\\nYear: 2004\\n\\nAbstract: A new class of distances appropriate for measuring similarity relations between sequences, say one type of similarity per distance, is studied. We propose a new "normalized information distance," based on the noncomputable notion of Kolmogorov complexity, and show that it is in this class and it minorizes every computable distance in the class (that is, it is universal in that it discovers all com...\",\n \"Title: Tales of Huffman\\nYear: 2006\\n\\nAbstract: We study the new problem of Huffman-like codes subject to individual restrictions on the code-word lengths of a subset of the source words. These are prefix codes with min imal expected code-word length for a random source where additionally the code-word lengths of a subset of the source words is prescribed, possibly differently for every such source word. Based on a structural analysis of properties of optimal solutions, we construct an efficient dynamic programming algorithm for this problem, a nd for an integer programming problem that may be of independent interest.\",\n \"Title: The Incompressibility Method\\nYear: 2000\\n\\nAbstract: . Kolmogorov complexity is a modern notion of randomnessdealing with the quantity of information in individual objects; that is,pointwise randomness rather than average randomness as produced bya random source. It was proposed by A.N. Kolmogorov in 1965 to quantifythe randomness of individual objects in an objective and absolutemanner. This is impossible for classical probability theory. Kolmogorovcomplexity is known variously as `algorithmic information\\\", `algorithmicentropy\\\",...\",\n \"Title: A Proof Technique for Register Automicity\\nYear: 1988\\n\\nAbstract: An implementation of a concurrent data object is wait-free if any process can complete any operation in a bounded number of steps, independently of the execution speeds of the programs. Much recent work has been done on concurrent access of shared variables by asynchronous processes. That work shows that implementing such shared variables does not require synchronization (by e.g. mutual exclusion), but can be solved in a wait-free manner. A fruitful paradigm in this context is the notion of a shared register satisfying a niceness condition called atomicity. Recent proposed solutions have led to the realization that: (1) neither the problem to be solved nor the model required were rigorously defined, (2) there was no clear insight in what constitutes a good proof of correctness in the area, and (3) the proposed protocols are so complicated that although correctness may be possible in a \\\"platonic fashion\\\", verifiability seems impossible to attain by human beings. A lot of controversy and allegations about constructions and proofs have arisen. Consequently, we have spent great effort to put the area on a rigorous basis. The thrust of this paper is to provide a new proof technique, and demonstrate its applicability by a non-trivial example. In other words, a new model is rigorously presented for the first time, and then a new method is given for proving register atomicity. It is then used to give a simple proof of the atomicity of the first and only direct construction of a multireader multiwriter register from atomic 1-reader 1-writer registers. (This construction was given in [8], by two of the present authors, with a completely different proof.)\",\n \"Title: A Note on Weighted Distributed Match-Making\\nYear: 1992\\n\\nAbstract: ABSTRACT This is a preliminary draft version of [Mathematical Systems Theory, 25(1992), 123 - 140] In many distributed computing environments, processes are concurrently executed by nodes in a store - and - forward network Distributed control issues as diverse as name - server, mutual exclusion and replicated data management, involve making matches between processes The generic paradigm is a formal problem called \\\"\\\"distributed match - making\\\"\\\" We define multi - dimensional and weighted versions, and the relations between the two, and develop a very general method to prove lower bounds on the complexity as a trade - off between number of messages and \\\"\\\"distributed - ness\\\"\\\" The resulting lower bounds are tight in all cases we have exam - ined We present a success - stop version of distributed match - making that is analysed in terms of a weight distribution that in all cases results in approximately halving the (expected) number of messages required in the corresponding strategy that does not use these weights\",\n \"Title: Reversible Simulation of Irreversible Computation by Pebble Games\\nYear: 1997\\n\\nAbstract: Reversible simulation of irreversible algorithms is analyzed in the stylized form of a 'reversible' pebble game. The reacheable reversible simulation instantaneous descriptions (pebble configurations) are characterized completely. As a corollary we obtain the reversible simulation by Bennett and that among all simulations that can be modelled by the pebble game, Bennett's simulation is optimal in that it uses the least auxiliary space for the greatest number of simulated steps. One can reduce the auxiliary storage overhead incurred by the reversible simulation at the cost of allowing limited erasing leading to an irreversibility-space tradeoff. We show that in this resource-bounded setting the limited erasing needs to be performed at precise instants during the simulation. We show that the reversible simulation can be modified so that it is applicable also when the simulated computation time is unknown.\",\n \"Title: Genetic Fitness Optimization Using Rapidly Mixing Markov Chains\\nYear: 1996\\n\\nAbstract: A notion of highly probable fitness optimization through evolutionary computing runs on small size populations in a very general setting is proposed. This has applications to evolutionary learning. Based on rapidly mixing Markov chains, the approach pertains to most types of evolutionary genetic algorithms, genetic programming and the like. For systems having associated rapidly mixing Markov chains and appropriate stationary distributions the new method finds optimal programs (individuals) with probability almost 1. Algorithmically, the novel approach prescribes a strategy of executing many short computation runs, rather than one long computation run. Given an arbitrary evolutionary program it may be infeasible to determine whether its associated matrix is rapidly mixing. In our proposed structured evolutionary program discipline, the development of the program and the guaranty of the rapidly mixing property go hand in hand. We conclude with a tentative toy example.\",\n \"Title: Depth as Randomness Deficiency\\nYear: 2009\\n\\nAbstract: Depth of an object concerns a tradeoff between computation time and excess of program length over the shortest program length\\n required to obtain the object. It gives an unconditional lower bound on the computation time from a given program in absence\\n of auxiliary information. Variants known as logical depth and computational depth are expressed in Kolmogorov complexity theory.\\n \\n We derive quantitative relation between logical depth and computational depth and unify the different depth notions by relating them to A. Kolmogorov and L. Levin’s fruitful notion of randomness deficiency. Subsequently, we revisit\\n the computational depth of infinite strings, study the notion of super deep sequences and relate it with other approaches.\",\n \"Title: Combinatorics and Kolmogorov complexity\\nYear: 1991\\n\\nAbstract: The authors investigate combinatorial properties of finite sequences with high Kolmogorov complexity. They also demonstrate the utility of a Kolmogorov complexity method in combinatorial theory by several examples (such as the coin-weighing problem)\",\n \"Title: Analysis of Sorting Algorithms by Kolmogorov Complexity (A Survey)\\nYear: 2009\\n\\nAbstract: Recently, many results on the computational complexity of sorting algorithms were obtained using Kolmogorov complexity (the\\n incompressibility method). Especially, the usually hard average-case analysis is ammenable to this method. Here we survey\\n such results about Bubblesort, Heapsort, Shellsort, Dobosiewiczsort, Shakersort, and sorting with stacks and queues in sequential\\n or parallel mode. Especially in the case of Shellsort the uses of Kolmogorov complexity surprisingly easily resolved problems\\n that had stayed open for a long time despite strenuous attacks.\",\n \"Title: Algorithmic Identification of Probabilities.\\nYear: 2013\\n\\nAbstract: TThe problem is to identify a probability associated with a set of natural numbers, given an infinite data sequence of elements from the set. If the given sequence is drawn i.i.d. and the probability mass function involved (the target) belongs to a computably enumerable (c.e.) or co-computably enumerable (co-c.e.) set of computable probability mass functions, then there is an algorithm to almost surely identify the target in the limit. The technical tool is the strong law of large numbers. If the set is finite and the elements of the sequence are dependent while the sequence is typical in the sense of Martin-L\\\\\\\"of for at least one measure belonging to a c.e. or co-c.e. set of computable measures, then there is an algorithm to identify in the limit a computable measure for which the sequence is typical (there may be more than one such measure). The technical tool is the theory of Kolmogorov complexity. We give the algorithms and consider the associated predictions.\",\n \"Title: Information Distance: New Developments\\nYear: 2012\\n\\nAbstract: In pattern recognition, learning, and data mining one obtains information from information-carrying objects. This involves an objective definition of the information in a single object, the information to go from one object to another object in a pair of objects, the information to go from one object to any other object in a multiple of objects, and the shared information between objects. This is called “information distance.” We survey a selection of new developments in information distance.\",\n \"Title: Weighted Distributed Match-Making\\nYear: 1988\\n\\nAbstract: In many distributed computing environments, processes are concurrently executed by nodes in a store-and-forward network. Distributed\\n control issues as diverse as name-server, mutual exclusion and replicated data management, involve making matches between\\n processes. The generic paradigm is a formal problem called “distributed match-making”. The applications require solutions\\n to weighted versions of the problem. We define new multi-dimensional and weighted versions, and the relations between the\\n two, and develop a very general method to prove lower bounds on the complexity as a trade-off between number of messages and\\n “distributedness”. The resulting lower bounds are tight in all cases we have examined.\",\n \"Title: The Google Similarity Distance\\nYear: 2007\\n\\nAbstract: Words and phrases acquire meaning from the way they are used in society, from their relative semantics to other words and phrases. For computers, the equivalent of \\\"society” is \\\"database,” and the equivalent of \\\"use” is \\\"a way to search the database.” We present a new theory of similarity between words and phrases based on information distance and Kolmogorov complexity. To fix thoughts, we use the World Wide Web (WWW) as the database, and Google as the search engine. The method is also applicable to other search engines and databases. This theory is then applied to construct a method to automatically extract similarity, the Google similarity distance, of words and phrases from the WWW using Google page counts. The WWW is the largest database on earth, and the context information entered by millions of independent users averages out to provide automatic semantics of useful quality. We give applications in hierarchical clustering, classification, and language translation. We give examples to distinguish between colors and numbers, cluster names of paintings by 17th century Dutch masters and names of books by English novelists, the ability to understand emergencies and primes, and we demonstrate the ability to do a simple automatic English-Spanish translation. Finally, we use the WordNet database as an objective baseline against which to judge the performance of our method. We conduct a massive randomized trial in binary classification using support vector machines to learn categories based on our Google distance, resulting in an a mean agreement of 87 percent with the expert crafted WordNet categories.\",\n \"Title: The Quantum Computing Challenge\\nYear: 2001\\n\\nAbstract: The laws ofp hysics imposes limits on increases in computing power. Two ofthe se limits are interconnect wires in multicomputers and thermodynamic limits to energy dissipation in conventional irreversible technology. Quantum computing is a new computational technology that promises to eliminate problems ofl atency and wiring associated with parallel computers and the rapidly approaching ultimate limits to computing power imposed by the fundamental thermodynamics. Moreover, a quantum computer will be able to exponentially improve known classical algorithms for factoring, and quadratically improve every classical algorithm for searching an unstructured list, as well as give various speedups in communication complexity, by exploiting unique quantum mechanical features. Finally, a quantum computer may be able to simulate quantum mechanical systems, something which seems out ofthe question for classical computers, thus reaching the ultimate goal ofrep lacing actual quantum mechanical experiments with simulated ones. On the downside, for some problems quantum mechanical computers cannot significantly improve the performance of classical computers.\",\n \"Title: Randomized Wait-Free Naming\\nYear: 1994\\n\\nAbstract: We present new distributed randomized naming protocols improving previous results in renaming and uniqueprocessor identity protocols. They are wait-free (which implies maximal fault-tolerance) and allow strongeradversaries. They also have low complexity. We give the first wait-free protocol achieving optimal key spacerange. (This is impossible for deterministic wait-free methods, so we use randomization.) We also introducea novel wait-free object, a test-and-set object which upon invocation ...\",\n \"Title: Normalized Information Distance is Not Semicomputable\\nYear: 2010\\n\\nAbstract: Normalized information distance (NID) uses the theoretical notion of\\nKolmogorov complexity, which for practical purposes is approximated by the\\nlength of the compressed version of the file involved, using a real-world\\ncompression program. This practical application is called 'normalized\\ncompression distance' and it is trivially computable. It is a parameter-free\\nsimilarity measure based on compression, and is used in pattern recognition,\\ndata mining, phylogeny, clustering, and classification. The complexity\\nproperties of its theoretical precursor, the NID, have been open. We show that\\nthe NID is neither upper semicomputable nor lower semicomputable.\",\n \"Title: The probabilistic analysis of language acquisition: Theoretical, computational, and experimental analysis\\nYear: 2010\\n\\nAbstract: There is much debate over the degree to which language learning is governed by innate language-specific biases, or acquired through cognition-general principles. Here we examine the probabilistic language acquisition hypothesis on three levels: We outline a novel theoretical result showing that it is possible to learn the exact generative model underlying a wide class of languages, purely from observing samples of the language. We then describe a recently proposed practical framework, which quantifies natural language learnability, allowing specific learnability predictions to be made for the first time. In previous work, this framework was used to make learnability predictions for a wide variety of linguistic constructions, for which learnability has been much debated. Here, we present a new experiment which tests these learnability predictions. We find that our experimental results support the possibility that these linguistic constructions are acquired probabilistically from cognition-general principles.\",\n \"Title: Minimum description length induction, Bayesianism, and Kolmogorov complexity\\nYear: 1999\\n\\nAbstract: The relationship between the Bayesian approach and the minimum description length approach is established. We sharpen and clarify the general modeling principles minimum description length (MDL) and minimum message length (MML), abstracted as the ideal MDL principle and defined from Bayes's rule by means of Kolmogorov complexity. The basic condition under which the ideal principle should be applied is encapsulated as the fundamental inequality, which in broad terms states that the principle is valid when the data are random, relative to every contemplated hypothesis and also these hypotheses are random relative to the (universal) prior. The ideal principle states that the prior probability associated with the hypothesis should be given by the algorithmic universal probability, and the sum of the log universal probability of the model plus the log of the probability of the data given the model should be minimized. If we restrict the model class to finite sets then application of the ideal principle turns into Kolmogorov's minimal sufficient statistic. In general, we show that data compression is almost always the best strategy, both in model selection and prediction\",\n \"Title: A Fast Quartet tree heuristic for hierarchical clustering\\nYear: 2011\\n\\nAbstract: The Minimum Quartet Tree Cost problem is to construct an optimal weight tree from the 3(n4) weighted quartet topologies on n objects, where optimality means that the summed weight of the embedded quartet topologies is optimal (so it can be the case that the optimal tree embeds all quartets as nonoptimal topologies). We present a Monte Carlo heuristic, based on randomized hill-climbing, for approximating the optimal weight tree, given the quartet topology weights. The method repeatedly transforms a dendrogram, with all objects involved as leaves, achieving a monotonic approximation to the exact single globally optimal tree. The problem and the solution heuristic has been extensively used for general hierarchical clustering of nontree-like (non-phylogeny) data in various domains and across domains with heterogeneous data. We also present a greatly improved heuristic, reducing the running time by a factor of order a thousand to ten thousand. All this is implemented and available, as part of the CompLearn package. We compare performance and running time of the original and improved versions with those of UPGMA, BioNJ, and NJ, as implemented in the SplitsTree package on genomic data for which the latter are optimized.\",\n \"Title: On the Simulation of Many Storage Heads by a Single One (Extended Abstract)\\nYear: 1983\\n\\nAbstract: Each multitape Turing machine, of which the storage heads scan O(log n) distinct squares in each interval of n steps, for all n 1, can be real- time simulated by an oblivious one-head tape unit. There exist multitape Turing machines, e.g. the normal pushdown store, for which the fastest on- line simulation by an oblivious one- head tape unit requires (n n) time.\",\n \"Title: Simple Wait-Free Multireader Registers\\nYear: 2002\\n\\nAbstract: Multireader shared registers are basic objects used as communication medium in asynchronous concurrent computation. We propose a surprisingly simple and natural scheme to obtain several waitfree constructions of bounded 1-writer multireader registers from atomic 1-writer 1-reader registers, that is easier to prove correct than any previous construction. Our main construction is the first symmetric pure timestamp one that is optimal with respect to the worst-case local use of control bits; the other one is optimal with respect to global use of control bits; both are optimal in time.\",\n \"Title: Conditional Kolmogorov complexity and universal probability\\nYear: 2013\\n\\nAbstract: The Coding Theorem of L.A. Levin connects unconditional prefix Kolmogorov complexity with the universal semiprobability mass function. There are conditional versions referred to in several publications but as yet there exist no written proofs. Under the classic definition of conditional probability, there is no conditional version of the Coding Theorem. We give the appropriate definitions and provide complete proofs of the conditional version of the Coding Theorem.\",\n \"Title: A lower bound on the average-case complexity of shellsort\\nYear: 2000\\n\\nAbstract: We demonstrate an &Ohgr;(pn1+1/p ) lower bound on the average-case running time (uniform distribution) of p-pass Shellsort. This is the first nontrivial general lower bound for average-case Shellsort.\",\n \"Title: Average-Case Analysis Using Kolmogorov Complexity\\nYear: 1997\\n\\nAbstract: This expository paper demonstrates how to use Kolmogorov complexityto do the average-case analysis via four examples, and exhibits a surprisingproperty of the celebrated associated universal distribution. The four examplesare: average case analysis of Heapsort [17, 15], average nni-distance betweentwo binary rooted leave-labeled trees [20], compact routing in computernetworks [3], average-case analysis of an adder algorithm [4]. The propertyis that the average-case complexity of any...\",\n \"Title: Nonsequentail Computation and Laws of Nature\\nYear: 1986\\n\\nAbstract: Traditionally, computational complexity theory deals with sequential computations. In the computational models the underlying physics is hardly accounted for. This attitude has persisted in common models for parallel computations. Wrongly, as we shall argue, since the laws of physics intrude forcefully when we want to obtain realistic estimates of the performance of parallel or distributed algorithms. First, we shall explain why it is reasonable to abstract away from the physical details in sequential computations. Second, we show why certain common approaches in the theory of parallel complexity do not give useful information about the actual complexity of the parallel computation. Third, we give some examples of the interplay between physical considerations and actual complexity of distributed computations.\",\n \"Title: New Applications of the Incompressibility Method.\\nYear: 1999\\n\\nAbstract: The incompressibility method is an elementary yet powerful proof technique. It has been used successfully in many areas. To further demonstrate its power and elegance we exhibit new simple proofs using the incompressibility method.\",\n \"Title: Algorithmic Arguments in Physics of Computation\\nYear: 1995\\n\\nAbstract: . We show the usefulness of incompressibility arguments basedon Kolmogorov complexity in physics of computation by several examples.These include analysis of energy parsimonious `adiabatic\\\" computation,and scalability of network architectures.1 IntroductionIn [Shannon, 1948] C. Shannon formulated information theory dealing with theaverage number of bits required to communicate a message produced by a randomsource from a sender to a receiver who both agree on the ensemble ofpossible...\",\n \"Title: Growth of Strings in Context Dependent Lindenmayer Systems\\nYear: 1974\\n\\nAbstract: Growth functions of context dependent Lindenmayer systems are investigated. Bounds on the fastest and slowest growth in such systems are derived, and a method to obtain (P)D1L growth functions from (P)D2L growth functions is given. Closure of context dependent growth functions under several operations is studied with special emphasis on an application of the firing squad synchronization problem. It is shown that, although all growth functions of DILs using a one letter alphabet are DOL growth functions, there are growth functions of PDILs using a two letter alphabet which are not. Several open problems concerning the decidability of growth equivalence, growth type classification etc. of context dependent growth are shown to be undecidable. As a byproduct we obtain that the language equivalence of PDILs is undecidable and that a problem proposed by Varshavsky has a negative solution.\",\n \"Title: The Power and Perils of MDL\\nYear: 2006\\n\\nAbstract: We point out a potential weakness in the application of the celebrated minimum description length (MDL) principle for model selection. Specifically, it is shown that (although the index of the model class which actually minimizes a two-part code has many desirable properties) a model which has a shorter two- part code-length than another is not necessarily better (unless of course it achieves the global minimum). This is illustrated by an application to infer a grammar (DFA) from positive examples. We also analyze computability issues, and robustness under receding of the data. Generally, the classical approach is inadequate to express the goodness-of-fit of individual models for individual data sets. In practice however, this is precisely what we are interested in: both to express the goodness of a procedure and where and how it can fail. To achieve this practical goal, we paradoxically have to use the, supposedly impractical, vehicle of Kolmogorov complexity.\",\n \"Title: On The Simulation Of Many Storage Heads By One\\nYear: 1984\\n\\nAbstract: Each multitape Turing machine, of which the storage heads scan O ( log n ) distinct squares in each interval of n steps, for all n ⩾ 1, can be real-time simulated by an oblivious `one-head tape unit. To simulate but the normal pushdown store, on-line by an oblivious one-head tape unit, requires ω ( n √ n ) time. (This improves the known ω ( n log n ) lower bound for the on-line simulation of multitape Turing machines by oblivious one-tape Turing machines.)\",\n \"Title: On a problem in the collective behavior of automata\\nYear: 1976\\n\\nAbstract: Varshavsky defines the function L(n) as the maximum finite length of a configuration which can be grown from one activated automation in a linear cell space of identical finite state automata having n internal states. It is shown that L increases faster than any computable function, even if the flow of information in the linear cell space is restricted to one direction.\",\n \"Title: On the Importance of Having an Identity or is Consensus Really Universal?\\nYear: 2006\\n\\nAbstract: We show that Naming-the existence of distinct IDs known to all- is a necessary assumption of Herlihy's universality result for Consensus. We then show in a very precise sense that Naming is harder than Consensus and bring to the surface some important differences existing between popular shared memory models which usually remain unnoticed.\",\n \"Title: Kolmogorov's structure functions and model selection\\nYear: 2004\\n\\nAbstract: In 1974, Kolmogorov proposed a nonprobabilistic approach to statistics and model selection. Let data be finite binary strings and models be finite sets of binary strings. Consider model classes consisting of models of given maximal (Kolmogorov) complexity. The \\\"structure function\\\" of the given data expresses the relation between the complexity level constraint on a model class and the least log-cardinality of a model in the class containing the data. We show that the structure function determines all stochastic properties of the data: for every constrained model class it determines the individual best fitting model in the class irrespective of whether the \\\"true\\\" model is in the model class considered or not. In this setting, this happens with certainty, rather than with high probability as is in the classical case. We precisely quantify the goodness-of-fit of an individual model with respect to individual data. We show that-within the obvious constraints-every graph is realized by the structure function of some data. We determine the (un)computability properties of the various functions contemplated and of the \\\"algorithmic minimal sufficient statistic.\\\".\",\n \"Title: Thermodynamics of computation and information distance\\nYear: 1993\\n\\nAbstract: Intuitively, the minimal information distance between z and g is the length of the shortest program for a universal computer to transform z into v and y into z. This measure will be shown to be, up to a logarithmic additive term, equal to the maximum of the conditional Kolmogorov complexities El(z, y) = max{~(ylz), K(zly)}. Any reasonable distance to measure similarity of pictures should be an effectively approximable, symmetric, positive function of z and y satisfying a reasonable normalization condition and obeying the triangle inequal\",\n \"Title: Turing Machines and Understanding Computational Complexity\\nYear: 2012\\n\\nAbstract: We describe the Turing Machine, list some of its many influences on the theory of computation and complexity of computations, and illustrate its importance.\",\n \"Title: A note on the recursive enumerability of some classes of recursively enumerable languages\\nYear: 1978\\n\\nAbstract: An elementary proof is presented for the fact that the class of infinite recursive languages is not recursively enumerable. Its relevance for contemporary linguistics and computer science is explained.\",\n \"Title: Efficient Simulations of Multicounter Machines\\nYear: 1982\\n\\nAbstract: An oblivious 1-tape Turing machine can on-line simulate a multicounter machine in linear time and logarithmic space. This leads to a linear cost combinational logic network implementing the first n steps of a multicounter machine and also to a linear time/logarithmic space on-line simulation by an oblivious logarithmic cost RAM. An oblivious log *n-head tape unit can simulate the first n steps of a multicounter machine in real-time, which leads to a linear cost combinational logic network with a constant data rate.\",\n \"Title: Distributed elections in an archimedean ring of processors\\nYear: 2009\\n\\nAbstract: sphere has no magnitude,we cannot conceive it to bear any ratio whateverto the surface of the sphere.Archimedes, The Sand-Reckoner1. IntroductionWe address the issue of time in distributed systems. Undergenuinely reasonable assumptions about time in distributed systemsthere exist, for some problems, logically time-independentsolutions which are more efficient than achievable with################* This work was supported by the Stichting Mathematisch Centrum.+ Full Address: Centre ...\",\n \"Title: Individual Communication Complexity: Extended Abstract\\nYear: 2004\\n\\nAbstract: We initiate the theory of communication complexity of indi- vidual inputs held by the agents, rather than worst-case or average-case. We consider total, partial, and partially correct protocols, one-way versus two-way, with (not in this version) and without help bits.\",\n \"Title: Optimality of wait-free atomic multiwriter variables\\nYear: 1992\\n\\nAbstract: Known implementations of concurrent wait-free atomic shared multiwritervariables use Q(n) control bits per subvariable. It has been shownthat implementations of sequential time-stamp systems require W(n) controlbits per subvariable. We exhibit a sequential wait-free atomic sharedmultiwriter variable construction using log n control bits per subvariable.There arises the question of the optimality of concurrent implementationsof the same, and of weak time-stamp systems. We also show that...\",\n \"Title: Space-efficient Routing Tables for Almost All Networks and the Incompressibility Method\\nYear: 1999\\n\\nAbstract: We use the incompressibility method based on Kolmogorov complexity to determine the total number of bits of routing information for almost all network topologies. In most models for routing, for almost all labeled graphs, $\\\\Theta (n^2)$ bits are necessary and sufficient for shortest path routing. By \\\"almost all graphs\\\" we mean the Kolmogorov random graphs which constitute a fraction of 1 - 1/nc of all graphs on n nodes, where c 0 is an arbitrary fixed constant. There is a model for which the average case lower bound rises to $\\\\Omega(n^2 \\\\log n )$ and another model where the average case upper bound drops to $O(n \\\\log^2 n)$. This clearly exposes the sensitivity of such bounds to the model under consideration. If paths have to be short, but need not be shortest (if the stretch factor may be larger than 1), then much less space is needed on average, even in the more demanding models. Full-information routing requires $\\\\Theta (n^3)$ bits on average. For worst-case static networks we prove an $\\\\Omega(n^2 \\\\log n )$ lower bound for shortest path routing and all stretch factors\",\n \"Title: Identification of Probabilities of Languages\\nYear: 2012\\n\\nAbstract: We consider the problem of inferring the probability distribution associated with a language, given data consisting of an infinite sequence of elements of the languge. We do this under two assumptions on the algorithms concerned: (i) like a real-life algorothm it has round-off errors, and (ii) it has no round-off errors. Assuming (i) we (a) consider a probability mass function of the elements of the language if the data are drawn independent identically distributed (i.i.d.), provided the probability mass function is computable and has a finite expectation. We give an effective procedure to almost surely identify in the limit the target probability mass function using the Strong Law of Large Numbers. Second (b) we treat the case of possibly incomputable probabilistic mass functions in the above setting. In this case we can only pointswize converge to the target probability mass function almost surely. Third (c) we consider the case where the data are dependent assuming they are typical for at least one computable measure and the language is finite. There is an effective procedure to identify by infinite recurrence a nonempty subset of the computable measures according to which the data is typical. Here we use the theory of Kolmogorov complexity. Assuming (ii) we obtain the weaker result for (a) that the target distribution is identified by infinite recurrence almost surely; (b) stays the same as under assumption (i). We consider the associated predictions.\",\n \"Title: A New Approach to Formal Language Theory by Kolmogorov Complexity (Preliminary Version)\\nYear: 1989\\n\\nAbstract: We introduce Kolmogorov complexity as a new technique in Formal Language Theory. We give an alternative for pumping lemma(s) and a new characterization for regular languages. For the separation of deterministic contextfree languages and contextfree languages no pumping lemmas or any other general method was known. We give a first general technique to separate these classes, and illustrate its use on four examples previously requiring labourous ad hoc methods. The approach is also successful at the high end of the Chomsky hierarchy since one can quantify nonrecursiveness in terms of Kolmogorov complexity. We also give a new proof, using Kolmogorov complexity, of Yao and Rivest's result that k + 1 heads are better than k heads.\",\n \"Title: On the Size of D0L Languages\\nYear: 1974\\n\\nAbstract: Languages generated by monogenic (i.e. deterministic) context independent Lindenmayer systems (DOL systems) are investigated. Necessary and sufficient conditions are established under which the language generated by a DOL system is finite. Thus, sharp bounds on the cardinality of such a language are obtained. A feasible solution for the membership problem is given. The problems are solved of what is the minimum sized alphabet over which there is a DOL language of cardinality n and, conversely, what is the maximum sized finite DOL language over an alphabet of m letters. This in turn provides us with some number theoretic functions, interesting in their own right, of which several properties, interrelations and asymptotic approximations are derived.\",\n \"Title: Average-Case Analysis via Incompressibility\\nYear: 1997\\n\\nAbstract: We will demonstrate how to use Kolmogorov complexity to do the average-case analysis via some examples. These examples include: longest common subsequence problem and shortest common supersequence problem [9, 11], problems in computational geometry [14], average case analysis of Heapsort [19, 17], average nni-distance between two binary rooted leave-labeled trees [23], compact routing in computer networks.\",\n \"Title: Average-Case Complexity of Shellsort\\nYear: 1999\\n\\nAbstract: We prove a general lower bound on the average-case complexity of Shellsort: the average number of data-movements (and comparisons) made by a p-pass Shellsort for any incremental sequence is Ω(pn1+1=p) for every p. The proof method is an incompressibility argument based on Kolmogorov complexity. Using similar techniques, the average-case complexity of several other sorting algorithms is analyzed.\",\n \"Title: Automatic Meaning Discovery Using Google\\nYear: 2006\\n\\nAbstract: Abstract We present a new theory of relative semantics between objects, based on information distance and Kolmogorov,complexity. This theory is then applied to construct a method,to automatically extract the meaning,of words and phrases from the world-wide-web using Google page counts. The approach is novel in its unrestricted problem domain, simplicity of implementation, and manifestly ontological underpinnings. The world-wide-web is the largest database on earth, and the latent semantic context information entered by millions of independent users averages out to provide automatic meaning,of useful quality. We give examples to distinguish between,colors and numbers, cluster names of paintings by 17th century Dutch masters and names of books by English novelists, the ability to understand emergencies, and primes, and we demonstrate the ability to do a simple automatic English-Spanish translation. Finally, we use the WordNet database as an objective baseline against which to judge the performance,of our method. We conduct a massive randomized trial in binary classification using support vector machines to learn categories based on our Google distance, resulting in an a mean agreement of 87% with the expert crafted WordNet categories.\",\n \"Title: How to share concurrent wait-free variables\\nYear: 1996\\n\\nAbstract: Sharing data between multiple asynchronous users—each of which can atomically read and write the data—is a feature that may help to increase the amount of parallelism in distributed systems. An algorithm implementing this feature is presented. The main construction of an n-user atomic variable directly from single-writer, single-reader atomic variables uses O(n) control bits and O(n) accesses per Read/Write running in O(1) parallel time.\",\n \"Title: On two-tape real-time computation and queues\\nYear: 1984\\n\\nAbstract: A Turing machine with two storage tapes cannot simulate a queue in both real-time and with at least one storage tape head always within o(n) squares from the start square. This fact may be useful for showing that a two-head tape unit is more powerful in real-time than two one-head tape units, as is commonly conjectured.\",\n \"Title: Average-Case Analysis of Algorithms Using Kolmogorov Complexity\\nYear: 2000\\n\\nAbstract: Abstract: Analyzing the average-case complexity of algorithms is a very practical but verydifficult problem in computer science. In the past few years, we have demonstrated thatKolmogorov complexity is an important tool for analyzing the average-case complexityof algorithms. We have developed the incompressibility method [7]. In this paper, weuse several simple examples to further demonstrate the power and simplicity of suchmethod. We prove bounds on the average-case number of stacks (queues)...\",\n \"Title: Big omega versus the wild functions\\nYear: 1984\\n\\nAbstract: The question of the desirable properties and proper definitions of the Order-of-Magnitude symbols, in particular O and Ø, is addressed once more. The definitions proposed are chosen for complementary mathematical properties, rather than for similarity of form.\",\n \"Title: On a Generalized Ruin Problem\\nYear: 2001\\n\\nAbstract: We consider a natural generalization of the classical ruin problem to more than two parties. Our \\\"ruin\\\" problem, which we will call the (k, I)-game, starts with k players each having I units as its initial capital. At each round of the game, all remaining k′ players pay 1/k′th unit as game fee, play the game, and one of the players wins and receives the combined game fees of 1 unit. A player who cannot pay the next game fee goes bankrupt, and the game terminates when all players but one are bankrupt.We analyze the length of the game, that is, the number of rounds executed until the game terminates, and give upper and lower bounds for the expected game length.\",\n \"Title: On inverse deterministic pushdown transductions\\nYear: 1978\\n\\nAbstract: Classes of source languages which can be mapped by a deterministic pushdown (DPDA) transduction into a given object language (while their complement is mapped into the complement of the object language) are studied. Such classes of source languages are inverse DPDA transductions of the given object language. Similarly for classes of object languages. The inverse DPDA transductions of the Dyck sets are studied in greater detail: they can be recognized in deterministic storage (log n)' but do not comprise all context free languages; their emptiness problem is unsolvable and their closure under homomorphism constitutes the r.e. sets. For each object language L we can exhibit a storage hardest language for the class of inverse DPDA transductions of L; similarly for the classes of regular, deterministic context free, and context free object languages. Last, we classify the classes of inverse DPDA transductions of the regular, deterministic context free, context free and deterministic context sensitive languages.\",\n \"Title: Physics and the New Computation\\nYear: 1995\\n\\nAbstract: New computation devices increasingly depend on particular physical properties rather than on logical organization alone as used to be the case in conventional technologies. This has impact on the synthesis and analysis of algorithms and the computation models on which they are to run. Therefore, scientists working in these areas will have to understand and apply physical law in their considerations. We discuss three cases in some detail: interconnect length and communication in massive multicomputers which depend on the geometry of space and speed of light; energy dissipation and reversible (adiabatic) computation which depend on thermodynamics; and quantum coherent parallel computing which depends on quantum mechanics.\",\n \"Title: Philosophical Issues in Kolmogorov Complexity\\nYear: 1992\\n\\nAbstract: this article at a conceptuallevel, it is sufficient to know that Kolmogorov complexity of a finite string xis simply the length of the shortest program, say in FORTRAN\",\n \"Title: On Prediction by Data Compression\\nYear: 1997\\n\\nAbstract: . Traditional wisdom has it that the better a theory compressesthe learning data concerning some phenomenon under investigation,the better we learn, generalize, and the better the theory predictsunknown data. This belief is vindicated in practice but apparentlyhas not been rigorously proved in a general setting. Making these ideasrigorous involves the length of the shortest effective description of anindividual object: its Kolmogorov complexity. In a previous paper wehave shown that...\",\n \"Title: On the logical depth function\\nYear: 2013\\n\\nAbstract: For a finite binary string $x$ its logical depth $d$ for significance $b$ is the shortest running time of a program for $x$ of length $K(x)+b$. There is another definition of logical depth. We give a new proof that the two versions are close. There is an infinite sequence of strings of consecutive lengths such that for every string there is a $b$ such that incrementing $b$ by 1 makes the associated depths go from incomputable to computable. The maximal gap between depths resulting from incrementing appropriate $b$'s by 1 is incomputable. The size of this gap is upper bounded by the Busy Beaver function. Both the upper and the lower bound hold for the depth with significance 0. As a consequence, the minimal computation time of the associated shortest programs rises faster than any computable function but not so fast as the Busy Beaver function.\",\n \"Title: Rate distortion and denoising of individual data using Kolmogorov complexity\\nYear: 2010\\n\\nAbstract: We examine the structure of families of distortion balls from the perspective of Kolmogorov complexity. Special attention is paid to the canonical rate-distortion function of a source word which returns the minimal Kolmogorov complexity of all distortion balls containing that word subject to a bound on their cardinality. This canonical rate-distortion function is related to the more standard algorithmic rate-distortion function for the given distortion measure. Examples are given of list distortion, Hamming distortion, and Euclidean distortion. The algorithmic rate-distortion function can behave differently from Shannon's rate-distortion function. To this end, we show that the canonical rate-distortion function can and does assume a wide class of shapes (unlike Shannon's); we relate low algorithmic mutual information to low Kolmogorov complexity (and consequently suggest that certain aspects of the mutual information formulation of Shannon's rate-distortion function behave differently than would an analogous formulation using algorithmic mutual information); we explore the notion that low Kolmogorov complexity distortion balls containing a given word capture the interesting properties of that word (which is hard to formalize in Shannon's theory) and this suggests an approach to denoising.\",\n \"Title: Normalized Compression Distance of Multiples\\nYear: 2012\\n\\nAbstract: Normalized compression distance (NCD) is a parameter-free similarity measure based on compression. The NCD between pairs of objects is not sufficient for al l applications. We propose an NCD of finite multisets (multiples) of objacts that is metric and is bette r for many applications. Previously, attempts to obtain such an NCD failed. We use the theoretical notion of Kolmogorov complexity that for practical purposes is approximated from above by the length of the compressed version of the file involved, using a real-world compression program. We applied the new NCD for multiples to retinal progenitor cell questions that were earlier treated with the pairwise NCD. Here we get significantly better results. We also applied the NCD for multiples to synthetic time sequence data. The preliminary results are as good as nearest neighbor Euclidean classifier. Index Terms— Normalized compression distance, multisets or multiples, pattern recognition, data mining, similarity, Kolmogorov complexity, retinal progenitor cell classification, synthetic data classification\",\n \"Title: Kolmogorov's Structure Functions with an Application to the Foundations of Model Selection\\nYear: 2002\\n\\nAbstract: In 1974 Kolmogorov proposed a non-probabilistic approach to statistics, an individual combinatorial relation between the data and its model. We vindicate, for the first time, the rightness of the original \\\"structure function\\\", proposed by Kolmogorov: minimizing the data-to-model code length (finding the ML estimator or MDL estimator), in a class of contemplated models of prescribed maximal (Kolmogorov) complexity, always results in a model of best fit (expressed as minimal randomness deficiency). We show that both the structure function and the minimum randomness deficiency function can assume all shapes over their full domain (improving an old result of L.A. Levin and both an old and a recent one of V.V. Vyugin). We determine the (un)computability properties of the various functions and \\\"algorithmic sufficient statistic.\\\"\",\n \"Title: Approximation of the two-part MDL code\\nYear: 2009\\n\\nAbstract: Approximation of the optimal two-part minimum description length (MDL) code for given data, through successive monotonically length-decreasing two-part MDL codes, has the following properties: i) computation of each step may take arbitrarily long; ii) we may not know when we reach the optimum, or whether we will reach the optimum at all; iii) the sequence of models generated may not monotonically improve the goodness of fit; but iv) the model associated with the optimum has (almost) the best goodness of fit. To express the practically interesting goodness of fit of individual models for individual data sets we have to rely on Kolmogorov complexity.\",\n \"Title: Average case complexity under the universal distribution equals worst-case complexity\\nYear: 1992\\n\\nAbstract: The average complexity of any algorithm whatsoever under the universal distribution is of the same order of magnitude as the worst-case complexity. This holds both for time complexity and for space complexity. To focus our discussion, we use as illustrations the particular case of sorting algorithms, and the general case of the average case complexity of NP-complete problems.\",\n \"Title: Kolmogorov complexity arguments in combinatorics\\nYear: 1994\\n\\nAbstract: The utility of a Kolmogorov complexity method in combinatorialtheory is demonstrated by several examples.1 IntroductionProbabilistic arguments in combinatorial theory, as used by Erdos and Spencer[5], are usually aimed at establishing the existence of an object, in a nonconstructivesense. It is ascertained that a certain member of a class hasa certain property, without actually exhibiting that object. Usually, themethod proceeds by exhibiting a random process which produces the...\",\n \"Title: Deterministic Lindenmayer languages, nonterminals and homomorphisms\\nYear: 1976\\n\\nAbstract: Lindenmayer systems are a class of parallel rewriting systems originally introduced to model the growth and development of filamentous organisms. Families of languages generated by deterministic Lindenmayer systems (i.e., those in which each string has a unique successor) are investigated. In particular, the use of nonterminals, homomorphisms, and the combination of these are studied for deterministic Lindenmayer systems using one-sided context (D1Ls) and two-sided context (D2Ls). Languages obtained from Lindenmayer systems by the use of nonterminals are called extensions. Typical results are: The closure under letter-to-letter homomorphism of the family of extensions of D1L languages is equal to the family of recursively enumerable languages, although the family of extensions of D1L languages does not even contain all regular languages. Let P denote the restriction that the system does not rewrite a letter as the empty word. The family of extensions of PD2L languages is equal to the family of languages accepted by deterministic linear bounded automata. The closure under nonerasing homomorphism of the family of extensions of PD1L languages does not even contain languages like { a 1 , a 2 ,⋯, a n } ∗ --{λ}, n ⩾2 . The closure of the family of PD1L languages under homomorphisms which map a letter either to itself or to the empty word is equal to the family of recursively enumerable languages. Strict inclusion results follow from necessary conditions for a language to be in one of the considered families. By stating the results in their strongest form, the paper contains a systematic classification of the effect of nonterminals, letter-to-letter homomorphisms, nonerasing homomorphisms and homomorphisms for all the basic types of deterministic Lindenmayer systems using context.\",\n \"Title: An n1.618 lower bound on the time to simulate one queue or two pushdown stores by one tape\\nYear: 1985\\n\\nAbstract: To simulate two pushdown stores, or one queue, on-line by a one-head tape unit requires Ω(n1.618) time.\",\n \"Title: Distributed match-making for processes in computer networks (preliminary version)\\nYear: 1986\\n\\nAbstract: In the very large multiprocessor systems and, on a grander scale, computer networks now emerging, processes are not tied to fixed processors but run on processors taken from a pool of processors. Processors are released when a process dies, migrates or when the process crashes. In distributed operating systems using the service concept, processes can be clients asking for a service, servers giving a service or both. Establishing communication between a process asking for a service and a process giving that service, without centralized control in a distributed environment with mobile processes, constitutes the problem of distributed match-making. Logically, such a match-making phase precedes routing in store-and-forward computer networks of this type. Algorithms for distributed match-making are developed and their complexity is investigated in terms of message passes and in terms of storage needed. The theoretical limitations of distributed match-making are established, and the techniques are applied to several network topologies.\",\n \"Title: The Generalized Universal Law of Generalization\\nYear: 2001\\n\\nAbstract: It has been argued by Shepard that there is a robust psychological law that relates the distance between a pair of items in psychological space and the probability that they will be perceived as similar. Specifically, this probability is a negative exponential function of the distance between the pair of items. In experimental contexts, distance is typically defined in terms of a multidimensional space—but this assumption seems unlikely to hold for complex stimuli. We show that, nonetheless, the Universal Law of Generalization can be derived in the more complex setting of arbitrary stimuli, using a much more universal measure of distance. This universal distance is defined as the length of the shortest program that transforms the representations of the two items of interest into one another: The algorithmic information distance. It is universal in the sense that it minorizes every computable distance: It is the smallest computable distance. We show that the Universal Law of Generalization holds with probability going to one—provided the probabilities concerned are computable. We also give a mathematically more appealing form of the Universal Law.\",\n \"Title: A New Quartet Tree Heuristic for Hierarchical Clustering\\nYear: 2006\\n\\nAbstract: We consider the problem of constructing an an optimal-weight tree from the 3*(n choose 4) weighted quartet topologies on n objects, where optimality means that the summed weight of the embedded quartet topologiesis optimal (so it can be the case that the optimal tree embeds all quartets as non-optimal topologies). We present a heuristic for reconstructing the optimal-weight tree, and a canonical manner to derive the quartet-topology weights from a given distance matrix. The method repeatedly transforms a bifurcating tree, with all objects involved as leaves, achieving a monotonic approximation to the exact single globally optimal tree. This contrasts to other heuristic search methods from biological phylogeny, like DNAML or quartet puzzling, which, repeatedly, incrementally construct a solution from a random order of objects, and subsequently add agreement values.\",\n \"Title: Time-bounded incompressibility of compressible strings and sequences\\nYear: 2009\\n\\nAbstract: For every total recursive time bound t, a constant fraction of all compressible (low Kolmogorov complexity) strings is t-bounded incompressible (high time-bounded Kolmogorov complexity); there are uncountably many infinite sequences of which every initial segment of length n is compressible to logn yet t-bounded incompressible below 14n-logn; and there is a countably infinite number of recursive infinite sequences of which every initial segment is similarly t-bounded incompressible. These results and their proofs are related to, but different from, Barzdins's lemma.\",\n \"Title: Average-Case Complexity of Shellsort (Preliminary version)\\nYear: 1999\\n\\nAbstract: We prove a general lower bound on the average-case com- plexity of Shellsort: the average number of data-movements (and com- parisons) made by a p-pass Shellsort for any incremental sequence is (pn1+1/p) for every p. The proof method is an incompressibility ar- gument based on Kolmogorov complexity. Using similar techniques, the average-case complexity of several other sorting algorithms is analyzed.\",\n \"Title: Two decades of applied Kolmogorov complexity: in memoriam Andrei Nikolaevich Kolmogorov 1903-87\\nYear: 1988\\n\\nAbstract: The authors provide an introduction to the main ideas of Kolmogorov complexity and survey the wealth of useful applications of this notion. It is based on a theory of information content of strings, intuitively, that the amount of information in a finite string is the size (i.e. number of bits) of the smallest program that, started with a blank memory, computes the string and then terminates. The following are treated: (1) application of the fact that some strings are compressible; this includes a strong version of Godel's incompleteness theorem; (2) lower-bound arguments that rest on application of the fact that certain strings cannot be compressed at all; applications range from Turing machines to electronic chips; (3) probability theory and a priori probability; applications range from foundational issues to the theory of learning and inductive inference in artificial intelligence. (4) Resource-bounded Kolmogorov complexity; applications range from NP-completeness and the P versus NP question to cryptography\",\n \"Title: Linear Time Simulation of Multihead Turing Machines with Head-to-Head Jumps\\nYear: 1977\\n\\nAbstract: The main result of this paper shows that, given a Turing machine with several read-write heads per tape and which has the additional one move shift operation \\\"shift a given head to the position of some other given head\\\", one can effectively construct a multitape Turing machine with a single read-write head per tape which simulates it in linear time; i.e. if the original machine operates in time T(n), then the simulating machine will operate in time cT(n), for some constant c.\",\n \"Title: Time and Space Bounds for Reversible Simulation\\nYear: 2001\\n\\nAbstract: We prove a general upper bound on the tradeoff between time and space that suffices for the reversible simulation of irreversible computation. Previously, only simulations using exponential time or quadratic space were known. The tradeoff shows for the first time that we can simultaneously achieve subexponential time and subquadratic space. The boundary values are the exponential time with hardly any extra space required by the Lange-McKenzie-Tapp method and the (log 3)th power time with square space required by the Bennett method. We also give the first general lower bound on the extra storage space required by general reversible simulation. This lower bound is optimal in that it is achieved by some reversible simulations.\",\n \"Title: Normalized Google Distance of Multisets with Applications.\\nYear: 2013\\n\\nAbstract: Normalized Google distance (NGD) is a relative semantic distance based on the World Wide Web (or any other large electronic database, for instance Wikipedia) and a search engine that returns aggregate page counts. The earlier NGD between pairs of search terms (including phrases) is not sufficient for all applications. We propose an NGD of finite multisets of search terms that is better for many applications. This gives a relative semantics shared by a multiset of search terms. We give applications and compare the results with those obtained using the pairwise NGD. The derivation of NGD method is based on Kolmogorov complexity.\",\n \"Title: Wait-free Test-and-Set (Extended Abstract)\\nYear: 1992\\n\\nAbstract: )Yehuda Afek1;4, Eli Gafni2?, John Tromp3, and Paul M.B. Vitanyi3;51Computer Science Department, Tel-Aviv University, Ramat-Aviv 69978 Israel.2Department of Computer Science, University of California, Los Angeles CA 90024.3Centrum voor Wiskunde en Informatica, Kruislaan 413, 1098 SJ Amsterdam, TheNetherlands.4AT&T Bell Laboratories, 600 Mountain Avenue, Murray Hill, NJ 07974.5Faculteit Wiskunde en Informatica, Universiteit van Amsterdam.Abstract. This...\",\n \"Title: Nonapproximability of the normalized information distance\\nYear: 2009\\n\\nAbstract: Normalized information distance (NID) uses the theoretical notion of Kolmogorov complexity, which for practical purposes is approximated by the length of the compressed version of the file involved, using a real-world compression program. This practical application is called 'normalized compression distance' and it is trivially computable. It is a parameter-free similarity measure based on compression, and is used in pattern recognition, data mining, phylogeny, clustering, and classification. The complexity properties of its theoretical precursor, the NID, have been open. We show that the NID is neither upper semicomputable nor lower semicomputable.\",\n \"Title: The Simple Roots of Real-Time Computation Hierarchies (Preliminary Version)\\nYear: 1984\\n\\nAbstract: A BLAH machine is any memory device that can be simulated in real-time by a multitape Turing machine and such that a multiBLAH machine can real-time simulate a pushdown store. A multiBLAH machine consists of a finite control connected to an input terminal and an output terminal and one or more copies of the BLAH memory unit. It is shown that a (k+1)-BLAH machine is more powerful in real-time than a k-BLAH machine, for each k. Thus the hierarchies, within the real-time definable computations, are proper and smooth, that is, adding a device always increases power. It also turns out that all real-time hierarchy results in this vein are simple corollaries of a single root: the real-time hierarchy of multipushdown store machines. As examples of such new results we mention that in real-time, k+1 tape-units with a fast rewind square are more powerful than k such units; that (k+1)-head tape-units with fast rewind squares are more powerful than k-head tape-units with fast rewind squares; that (k+1)-dequeue machines are more powerful than k-dequeue machines; and that (k+1)-concatenable-dequeue machines are more powerful than k-concatenable-dequeue machines.\",\n \"Title: On Logical Depth and the Running Time of Shortest Programs.\\nYear: 2013\\n\\nAbstract: The logical depth with significance $b$ of a finite binary string $x$ is the shortest running time of a binary program for $x$ that can be compressed by at most $b$ bits. There is another definition of logical depth. We give two theorems about the quantitative relation between these versions: the first theorem concerns a variation of a known fact with a new proof, the second theorem and its proof are new. We select the above version of logical depth and show the following. There is an infinite sequence of strings of increasing length such that for each $j$ there is a $b$ such that the logical depth of the $j$th string as a function of $j$ is incomputable (it rises faster than any computable function) but with $b$ replaced by $b+1$ the resuling function is computable. Hence the maximal gap between the logical depths resulting from incrementing appropriate $b$'s by 1 rises faster than any computable function. All functions mentioned are upper bounded by the Busy Beaver function. Since for every string its logical depth is nonincreasing in $b$, the minimal computation time of the shortest programs for the sequence of strings as a function of $j$ rises faster than any computable function but not so fast as the Busy Beaver function.\",\n \"Title: Statistical properties of finite sequences with high Kolmogorov complexity\\nYear: 1994\\n\\nAbstract: We investigate to what extent finite binary sequences with highKolmogorov complexity are normal (all blocks of equal length occurequally frequent), and the maximal length of all-zero or all-one runswhich occur with certainty.1 IntroductionEach individual infinite sequence generated by a (12;12 ) Bernoulli process (flippinga fair coin) has (with probability 1) the property that the relative frequencyof zeros in an initial n-length segment goes to12 for n goes to...\",\n \"Title: Normalized Web Distance and Word Similarity\\nYear: 2009\\n\\nAbstract: There is a great deal of work in cognitive psychology, linguistics, and\\ncomputer science, about using word (or phrase) frequencies in context in text\\ncorpora to develop measures for word similarity or word association, going back\\nto at least the 1960s. The goal of this chapter is to introduce the\\nnormalizedis a general way to tap the amorphous low-grade knowledge available\\nfor free on the Internet, typed in by local users aiming at personal\\ngratification of diverse objectives, and yet globally achieving what is\\neffectively the largest semantic electronic database in the world. Moreover,\\nthis database is available for all by using any search engine that can return\\naggregate page-count estimates for a large range of search-queries. In the\\npaper introducing the NWD it was called `normalized Google distance (NGD),' but\\nsince Google doesn't allow computer searches anymore, we opt for the more\\nneutral and descriptive NWD. web distance (NWD) method to determine similarity\\nbetween words and phrases. It\",\n \"Title: Square time is optimal for simulation of one pushdown store or one queue by an oblivious one-head tape unit\\nYear: 1985\\n\\nAbstract: To simulate a pushdown store or queue on-line by an oblivious one-head tape unit takes at least square time. Since each multitape Turing machine can be trivially simulated by an oblivious one-head tape unit in square time, this result its optimal.\",\n \"Title: On The Average-Case Complexity of Shellsort.\\nYear: 2018\\n\\nAbstract: We prove a lower bound expressed in the increment sequence on the average-case complexity of the number of inversions of Shellsort. This lower bound is sharp in every case where it could be checked. A special case of this lower bound yields the general Jiang-Li-Vitanyi lower bound. We obtain new results, for example, determining the average-case complexity precisely in the Yao-Janson-Knuth 3-pass case.\",\n \"Title: Exact Expression For Information Distance.\\nYear: 2014\\n\\nAbstract: Information distance can be defined not only between two strings but also in a finite multiset of strings of cardinality greater than two. We determine a best upper bound on the information distance. It is exact, since the upper bound on the information distance for all multisets is the same as the lower bound for infinitely many multisets of each of infinitely many cardinalities, up to a constant...\",\n \"Title: Web Similarity.\\nYear: 2015\\n\\nAbstract: Normalized web distance (NWD) is a similarity or normalized semantic distance based on the World Wide Web or any other large electronic database, for instance Wikipedia, and a search engine that returns reliable aggregate page counts. For sets of search terms the NWD gives a similarity on a scale from 0 (identical) to 1 (completely different). The NWD approximates the similarity according to all (upper semi)computable properties. We develop the theory and give applications. The derivation of the NWD method is based on Kolmogorov complexity.\",\n \"Title: Three approaches to the quantitative definition of information in an individual pure quantum state\\nYear: 2000\\n\\nAbstract: In analogy of classical Kolmogorov complexity we de- velop a theory of the algorithmic information in bits con- tained in any one of continuously many pure quantum states: quantum Kolmogorov complexity. Classical Kol- mogorov complexity coincides with the new quantum Kol- mogorov complexity restricted to the classical domain. Quantum Kolmogorov complexity is upper bounded and can be effectively approximated from above. With high probability a quantum object is incompressible. There are two alternative approaches possible: to define the complex- ity as the length of the shortest qubit program that effec- tively describes the object, and to use classical descriptions with computable real parameters.\",\n \"Title: Language Learning From Positive Evidence, Reconsidered: A Simplicity-Based Approach.\\nYear: 2013\\n\\nAbstract: Children learn their native language by exposure to their linguistic and communicative environment, but apparently without requiring that their mistakes be corrected. Such learning from positive evidence has been viewed as raising logical problems for language acquisition. In particular, without correction, how is the child to recover from conjecturing an over-general grammar, which will be consistent with any sentence that the child hears? There have been many proposals concerning how this logical problem can be dissolved. In this study, we review recent formal results showing that the learner has sufficient data to learn successfully from positive evidence, if it favors the simplest encoding of the linguistic input. Results include the learnability of linguistic prediction, grammaticality judgments, language production, and form-meaning mappings. The simplicity approach can also be scaled down to analyze the learnability of specific linguistic constructions, and it is amenable to empirical testing as a framework for describing human language acquisition.\",\n \"Title: Normalized Compression Distance of Multisets with Applications\\nYear: 2015\\n\\nAbstract: Pairwise normalized compression distance (NCD) is a parameter-free, feature-free, alignment-free, similarity metric based on compression. We propose an NCD of multisets that is also metric. Previously, attempts to obtain such an NCD failed. For classification purposes it is superior to the pairwise NCD in accuracy and implementation complexity. We cover the entire trajectory from theoretical under...\",\n \"Title: On the rate of decrease in logical depth.\\nYear: 2017\\n\\nAbstract: The logical depth with significance b of a string x is the shortest running time of a program for x that can be compressed by at most b bits. Another definition is based on algorithmic probability. We give a simple new proof for the known relation between the two definitions. We also prove the following: Given a string we can consider the maximal decrease in logical depth when the significance parameter increases by 1. There exists a sequence of strings of lengths n=1,2,…, such that this maximal decrease as a function of n rises faster than any computable function but not as fast as the Busy Beaver function. This holds also for the computation times of the shortest programs of these strings.\",\n \"Title: Identification of Probabilities.\\nYear: 2017\\n\\nAbstract: Within psychology, neuroscience and artificial intelligence, there has been increasing interest in the proposal that the brain builds probabilistic models of sensory and linguistic input: that is, to infer a probabilistic model from a sample. The practical problems of such inference are substantial: the brain has limited data and restricted computational resources. But there is a more fundamental question: is the problem of inferring a probabilistic model from a sample possible even in principle? We explore this question and find some surprisingly positive and general results. First, for a broad class of probability distributions characterized by computability restrictions, we specify a learning algorithm that will almost surely identify a probability distribution in the limit given a finite i.i.d. sample of sufficient but unknown length. This is similarly shown to hold for sequences generated by a broad class of Markov chains, subject to computability assumptions. The technical tool is the strong law of large numbers. Second, for a large class of dependent sequences, we specify an algorithm which identifies in the limit a computable measure for which the sequence is typical, in the sense of Martin-Löf (there may be more than one such measure). The technical tool is the theory of Kolmogorov complexity. We analyze the associated predictions in both cases. We also briefly consider special cases, including language learning, and wider theoretical implications for psychology.\",\n \"Title: Inductive reasoning and Kolmogorov complexity\\nYear: 1989\\n\\nAbstract: The inductive reasoning concepts of R.J. Solomonoff (Inf. Control. vol.7, p.1-22, 224-254, 1964) are considered. The thesis is developed that Solomonoff's method is fundamental in the sense that many other induction principles can be viewed as particular ways to obtain computable approximations to it. This is demonstrated explicitly in the cases of E.M. Gold's (1967, 1978) paradigm for inductive inference, L.G. Valiant's (1984) learning (by adding computational requirements), J. Rissanen's (1982) minimum description length principle, Fisher's maximum-likelihood principle (J. Rissanen, 1982), and E.T. Jayne's (1968, 1982) maximum entropy principle. Several new theorems and derivations to this effect are presented. What can and cannot be learned in terms of Kolmogorov complexity is delimited, and an experiment in machine learning of handwritten characters is described\",\n \"Title: Logical depth for reversible Turing machines with an application to the rate of decrease in logical depth for general Turing machines\\nYear: 2019\\n\\nAbstract: The logical depth of a reversible Turing machine equals the shortest running time of a shortest program for it. This is applied to show that the result in [1] is valid notwithstanding the error noted in Corrigendum [7].\",\n \"Title: Fast Phylogeny of SARS-CoV-2 by Compression\\nYear: 2022\\n\\nAbstract: The compression method to assess similarity, in the sense of having a small normalized compression distance (NCD), was developed based on algorithmic information theory to quantify the similarity in files ranging from words and languages to genomes and music pieces. It has been validated on objects from different domains always using essentially the same software. We analyze the whole-genome phylogeny and taxonomy of the SARS-CoV-2 virus, which is responsible for causing the COVID-19 disease, using the alignment-free compression method to assess similarity. We compare the SARS-CoV-2 virus with a database of over 6500 viruses. The results suggest that the SARS-CoV-2 virus is closest in that database to the RaTG13 virus and rather close to the bat SARS-like coronaviruses bat-SL-CoVZXC21 and bat-SL-CoVZC45. Over 6500 viruses are identified (given by their registration code) with larger NCDs. The NCDs are compared with the NCDs between the mtDNA of familiar species. We address the question of whether pangolins are involved in the SARS-CoV-2 virus. The compression method is simpler and possibly faster than any other whole-genome method, which makes it the ideal tool to explore phylogeny. Here, we use it for the complex case of determining this similarity between the COVID-19 virus, SARS-CoV-2 and many other viruses. The resulting phylogeny and taxonomy closely resemble earlier results from by alignment-based methods and a machine-learning method, providing the most compelling evidence to date for the compression method, showing that one can achieve equivalent results both simply and quickly.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":106,"string":"106"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13569,"string":"13,569"}}},{"rowIdx":13570,"cells":{"question_id":{"kind":"string","value":"Lorenza Saitta"},"question":{"kind":"string","value":"Generate an abstract for the title \"A Monte Carlo analysis of ensemble classification\" using the following items: 1. Monte Carlo ensemble classifier\n2. Theoretical analysis\n3. Conditions for outperformance\n4. Closed form expression\n5. Expected error behavior."},"answer":{"kind":"string","value":"In this paper we extend previous results providing a theoretical analysis of a new Monte Carlo ensemble classifier. The framework allows us to characterize the conditions under which the ensemble approach can be expected to outperform the single hypothesis classifier. Moreover, we provide a closed form expression for the distribution of the true ensemble accuracy, as well as of its mean and variance. We then exploit this result in order to analyze the expected error behavior in a particularly interesting case."},"correct_answer":{"kind":"string","value":"In this paper we extend previous results providing a theoretical analysis of a new Monte Carlo ensemble classifier. The framework allows us to characterize the conditions under which the ensemble approach can be expected to outperform the single hypothesis classifier. Moreover, we provide a closed form expression for the distribution of the true ensemble accuracy, as well as of its mean and variance. We then exploit this result in order to analyze the expected error behavior in a particularly interesting case."},"documents":{"kind":"list like","value":["Title: Learning on the phase transition edge\nYear: 2001\n\nAbstract: Extracting rules from RBFs is not a trivial task because of nonlinear functions or high input dimensionality. In such cases, some of the hidden units of the RBF network have a tendency to be \"shared\" across several output classes or even may not contribute ...","Title: Coevolutionary, Distributed Search for Inducing Concept Description\nYear: 1998\n\nAbstract: This paper presents a highly parallel genetic algorithm, designed for concept induction in propositional and first order logics. The parallel architecture is an adaptation for set covering problems, of the diffusion model developed for optimization.","Title: Feature construction from synergic pairs to improve microarray-based classification.\nYear: 2007\n\nAbstract: Microarray experiments that allow simultaneous expression profiling of thousands of genes in various conditions (tissues, cells or time) generate data whose analysis raises difficult problems. In particular, there is a vast disproportion between the number of attributes (tens of thousands) and the number of examples (several tens). Dimension reduction is therefore a key step before applying classification approaches. Many methods have been proposed to this purpose, but only a few of them considered a direct quantification of transcriptional interactions. We describe and experimentally validate a new dimension reduction and feature construction method, which assesses interactions between expression profiles to improve microarray-based classification accuracy.Our approach relies on a mutual information measure that exposes some elementary constituents of the information contained in a pair of gene expression profiles. We show that their analysis implies a term that represents the information of the interaction between the two genes. The principle of our method, called FeatKNN, is to exploit the information provided by highly synergic gene pairs to improve classification accuracy. First, a heuristic search selects the most informative gene pairs. Then, for each selected pair, a new feature, representing the classification margin of a KNN classifier in the gene pairs space, is constructed. We show experimentally that the interactional information has a degree of significance comparable to that of the gene expression profiles considered separately. Our method has been tested with different classifiers and yielded significant improvements in accuracy on several public microarray databases. Moreover, a synthetic assessment of the biological significance of the concept of synergic gene pairs suggested its ability to uncover relevant mechanisms underlying interactions among various cellular processes.","Title: Monte Carlo theory as an explanation of bagging and boosting\nYear: 2003\n\nAbstract: In this paper we propose the framework of Monte Carlo algorithms as a useful one to analyze ensemble learning. In particular, this framework allows one to guess when bagging will be useful, explains why increasing the margin improves performances, and suggests a new way of performing ensemble learning and error estimation.","Title: Abstraction and Complexity Measures\nYear: 2007\n\nAbstract: ion is fundamental for both human and artificial reasoning. The word denotes different activities and process, but\n all are intuitively related to the notion of complexity/simplicity, which is as elusive a notion as abstraction. From an analysis\n of the literature on abstraction and complexity it clearly appears that it is unrealistic to find definitions valid in all\n disciplines and for all tasks. Hence, we consider a particular model of abstraction, and try to investigate how complexity\n measures could be mapped to it. Preliminary results show that abstraction and complexity are not monotonically coupled notions,\n and that complexity may either increase or decrease with abstraction according to the definition of both and to the specificities\n of the considered domain.","Title: Abstracting Complex Interaction Networks\nYear: 2009\n\nAbstract: The exploration of complex interaction networks has attracted considerable interest in various fields, rang- ing from fundamental biology and medicine to statis- tical physics and information technology. In \"-omics\" disciplines, significant progresses have been made in understanding the large-scale properties and the bio- logical relevance of these interactions. Some properties such as \"scale-free\" distribution of nodes connectivity or \"centrality\" are aspects commonly described in such complex interaction systems. In many of these studies the analysis of network topology is complemented by a semantic analysis that may rely on dierent labels asso- ciated to the interacting entities. One of the bottleneck of these semantic analysis is that they are computa- tionally costly. In this paper we present a framework to explore abstraction of networks useful to speedup the computation of ground network measures. Such abstraction mechanisms may be used to eciently pro- vide accurate approximations of ground network mea- sures.","Title: Learning Simple Recursive Theories\nYear: 1993\n\nAbstract: The task of learning relations has been concerned, so far, with the acquisition of intensional descriptions of unrelated concepts. However, in many real domains concepts are strictly related to each other and the instances of one of them cannot possibly be recognised without previous recognition of other objects as instances of related concepts. A typical case is the problem of labelling parts of a scene in order to interpret it. This paper extends in several ways the learning relations paradigm; in particular, a new methodology, allowing a recursive theory to be inferred from a set of examples, is presented. The learning algorithm works bottom-up, creating first an acyclic graph that classifies all the instances in the training set. Afterwards, a recursive theory is synthesised from the graph.","Title: Learning Mutually Dependent Relations\nYear: 1992\n\nAbstract: So far, the task of learning relations has been concerned with the acquisition of intensional descriptions of unrelated concepts. However, in many real domains concepts are strictly related to each other and the instances of one of them cannot possibly be recognized without previous recognition of other objects as instances of related concepts. A typical case is the problem of labeling parts of a scene in order to interprete it. This paper extends in several ways the learning relations paradigm; in particular, a new methodology, allowing a recursive theory to be inferred from a set of examples, is presented. Each example may contain instances of several relations at the same time. The learning algorithm works bottom up, creating first an acyclic graph that classifies all the instances in the training set. Afterward, a recursive theory is synthesized from the graph. A prototype implementation of the algorithm has been realized by extending the program ML-SMART. The results obtained for an artificial labeling problem are described and discussed.","Title: Investigating the Relationships between Abduction and Inverse Resolution in Proposition Calculus\nYear: 1991\n\nAbstract: In systems devoted to diagnostic tasks, the term abduction usually denotes the process of explaining a set of observed manifestations, in the light of an existing domain theory. This amounts to find a set of hypotheses such that the manifestations logically follow from them and from the theory. Then, abduction corresponds to reasoning from consequences to possible premises. On the other hand, an analogous line of reasoning has been independently followed in constructive induction, where new predicates are introduced and theories are completed by trying to invert the resolution mechanism. In this paper, links are established between these two approaches by showing that the inverse resolution operators are sound abductive inference rules, in the case of propositional calculus. Some relationships with deduction in a closed world assumption are also investigated.","Title: An Expert System for Interpreting Speech Patterns\nYear: 1982\n\nAbstract: Efficient syllabic hypothesization in continuous speech has been so far an unsolved problem. A novel solution based on the extraction of acoustic cues is proposed in this paper. This extraction is performed by parallel processes implementing an expert system represented by a grammar of frames.","Title: Multi Level Knowledge in Modeling Qualitative Physics Learning\nYear: 2000\n\nAbstract: The goal of the reported research is the development of a computational approach that could help a cognitive scientist to interactively represent a learner's mental models, and to automatically v alidate their coherence with respect to the a vailable e xperimental data. In a reported case-study, the student's mental models are inferred from questionnaires and interviews collected during a sequence of teaching sessions. These putative c ognitive models are based o n a theory o f knowledge representation, derived from psychological results and educational studies, which accounts for the e volution of the student's knowledge over a learning period. The learning system WHY, able to handle (causal) domain knowledge, shows how to model the answers and the causal explanations given by the learner.","Title: Extending Inverse Resolution to Build Up Abstractions\nYear: 1991\n\nAbstract: ion is a conceptual framework potentially unifying and integrating different methodologies developed in machine learning and problem solving. According to the common understanding of the term, abstraction is a mapping between languages, and, then, in many learning tasks it can be so formulated. On the contrary, in this paper abstraction is defined as a mapping between models, implicitly extending Tenenberg's restricted predicate mapping in order to include more complex abstraction schemes and to cope with the problem of inconsistency in the abstract model.","Title: Modeling production rules by Means of Predicate Transition networks\nYear: 1985\n\nAbstract: A formal tool, the PREDICATE TRANSITION networks, is proposed for representing rules in production systems. These networks have been introduced by Genrich and Lautenbach as a kind of high-level Petri nets. The peculiarity of such a formal tool lies in its well-formalized semantics, which can be described by means of an algebraic formalism and can be analyzed using the S-invariant method in order to prove properties of the network. A brief review of PREDICATE TRANSITION networks is followed by a description of their use in AI applications. In particular the characteristics which make them suitable for representing knowledge and control strategies in expert systems are analyzed.","Title: A Coevolutionary Approach to Concept Learning\nYear: 1997\n\nAbstract: This paper presents a highly parallel genetic algorithm, designed for concept induction in propositional and First Order logics. The parallel architecture is an adaptation to set covering problems of the diffusion model developed for optimization problems.","Title: Statistical Characterization of a Computer Grid\nYear: 2009\n\nAbstract: Large-scale statistical analysis of more than 28 million jobs collected during 20 months of grid activity was undertaken in\n order to examine the relations between users, computing elements and jobs in the network. The results give insight into the\n global system behaviour and can be used to build models applicable in various contexts of grid computing. As an example, we\n here construct probabilistic models that prove to be able to accurately predict job abortion.","Title: Formal Models of Selection in Genetic Algorithms\nYear: 1994\n\nAbstract: In this paper three formal models of selection operators (two known from the literature and one newly porposed) for genetic algorithms, used to learn structured concepts descriptions containing small disjuncts, are presented. The evolution of a population, according to these operators, with a generation gap equal to or less than one, is investigated in order to assess their ability to let different species emerge and coexist; finally, the average asymptotic behaviour of the population is determined. Experimental results confirm both bevahiours previously reported in the literature and the prediction of the new model.","Title: Relational Learning: Hard Problems and Phase Transitions\nYear: 1999\n\nAbstract: This paper focuses on a major step of machine learning, namely checking whether an example matches a candidate hypothesis. In relational learning, matching can be viewed as a Constraint Satisfaction Problem (CSP). The complexity of the task is analyzed in the Phase Transition framework, investigating the impact on the effectiveness of two relational learners: FOIL and G-NET. The critical factors of complexity, and their critical values, are experimentally investigated on artificial problems. This leads to distinguish several kinds of learning domains, depending on whether the target concept lies in the \"mushy\" region or not. Interestingly, experiments done with FOIL and G-NET show that both learners tend to induce hypotheses generating matching problems located inside the phase transition region, even if the constructed target concept lies far outside. Moreover, target concepts constructed too close to the phase transition are hard and both learners fail. The paper offers an explanation for this fact, and proposes a classification of learning domains and their hardness.","Title: Relational learning as search in a critical region\nYear: 2003\n\nAbstract: Machine learning strongly relies on the covering test to assess whether a candidate hypothesis covers training examples. The present paper investigates learning relational concepts from examples, termed relational learning or inductive logic programming. In particular, it investigates the chances of success and the computational cost of relational learning, which appears to be severely affected by the presence of a phase transition in the covering test. To this aim, three up-to-date relational learners have been applied to a wide range of artificial, fully relational learning problems. A first experimental observation is that the phase transition behaves as an attractor for relational learning; no matter which region the learning problem belongs to, all three learners produce hypotheses lying within or close to the phase transition region. Second, a failure region appears. All three learners fail to learn any accurate hypothesis in this region. Quite surprisingly, the probability of failure does not systematically increase with the size of the underlying target concept: under some circumstances, longer concepts may be easier to accurately approximate than shorter ones. Some interpretations for these findings are proposed and discussed.","Title: Modelling Conceptual Change: An Interdisciplinary Approach\nYear: 1997\n\nAbstract: A computational approach to the simulation of cognitive models of conceptual change in children learning elementary physics is presented. The student's mental model is inferred from a sequence of interviews collected along a period of eleven teaching sessions. Goal of the simulation is to support the cognitive scientist's investigation of learning in humans. The hypothesized cognitive models are based on a theory of conceptual change, derived from psychology results and educational experiences, which accounts for the evolution of the student's knowledge over a learning period. A Machine Learning (ML) system, able to handle domain knowledge (including a causal model of the domain), has been chosen as tool for the simulation of the cognitive models evolution. The system performs knowledge revision and provides causal explanation for its conclusions.","Title: Transition de phase et apprentissage relationnel : une analyse critique\nYear: 2003\n\nAbstract: Ce papier presente une revue et une analyse des resultats obtenus par les auteurs et par leurs collaborateurs en ce qui concerne lu0027effet de la transition de phase dans le probleme de lu0027appartement. Ce qui a ete mis en evidence est que la recherche du0027hypotheses, guidee par les heuristiques plus connues (gain du0027information ou simplicite), finissent inevitablement dans la region TP: donc, on ne peut pas eviter cette region a haute complexite. Les conclusions quu0027on peut en tirer pour lu0027apprentissage relationnel est quu0027il semble difficile de pouvoir approcher des problemes plus importants que ceux que lu0027on a actuellement.","Title: Experimental evaluation of hierarchical hidden markov models\nYear: 2005\n\nAbstract: Building profiles for processes and for interactive users is a important task in intrusion detection. This paper presents the results obtained with a Hierarchical Hidden Markov Model. The algorithm discovers typical ”motives” of a process behavior, and correlates them into a hierarchical model. Motives can be interleaved with possibly long gaps where no regular behavior is detectable. We assume that motives could be affected by noise,modeled as insertion, deletion and substitution errors. In this paper the learning algorithm is briefly recalled and then it is experimentally evaluated on three profiling case studies. The first case is built on a suite of artificial traces automatically generated by a set of given HHMMs. The challenge for the algorithm is to reconstruct the original model from the traces. It will be shown that the algorithm is able to learn HHMMs very similar to the original ones, in presence of noise and distractors. The second and third case studies refer to the problem of constructing a discriminant model for a user typing on a keyboard.","Title: Perception-Based Granularity Levels in Concept Representation\nYear: 2000\n\nAbstract: In this paper we propose a perception-based view of abstraction, which originates from the observation that conceptualization of a domain involves entities belonging to several episteological levels. The fundamental level corresponds to the preception of a world. For memorization purposes, some kind of structure is needed, in order to organize objects and relations perceived in the world into coherent ensembles. To communicate with others, a language must be invented, and, finally, a theory makes it possible to reason about the world. After invented, suitable properties abstraction should have to be useful for concept representation, examples of abstraction operators, designed to perform the abstraction process in practice, will be introduced.","Title: Learning disjunctive concepts with distributed genetic algorithms\nYear: 1994\n\nAbstract: Inducing concept descriptions from examples is a machine learning task that can be formulated as a search problem. Recently, genetic algorithms proved to be an appealing alternative for searching concept description spaces, because of their great exploration power and their suitability to exploit massive parallelism. In order to exploit genetic algorithms for searching the space of formulas φ (the elements of the population), we must reformulate the set covering problem by suitably defining the representation scheme for the formulas, the genetic operators and the fitness function. The system described in this paper, REGAL (Giordana and Saitta, 1993), encodes first order logic (FOL) formulas φ into bit strings of fixed length. In order to make this possible, REGAL's representation language, L, has been constrained to be finite, by defining a language template, Λ, which represents the maximally complex formula in L. The resulting language contains FOL formulas built up with conjunction, disjunction, negation, internal disjunction and existential and universal quantification","Title: An analysis of the “universal suffrage” selection operator\nYear: 1996\n\nAbstract: The “universal suffrage” selection operator, designed primarily for concept learning inside the system REGAL, is discussed for both overlapping and nonoverlapping populations. Analysis of its behavior is performed by using the “virtual average population” method, a new tool for investigating asymptotic properties of convergence of macroscopic quantities related to the population of a genetic algorithm.","Title: Scheduling of processes in a speech understanding system based on approximate reasoning\nYear: 1979\n\nAbstract: A method for scheduling knowledge source instantiations in a Speech Understanding System is presented. It is based on the evaluation of linguistic probabilities and on approximate reasoning.","Title: Is a Greedy Covering Strategy an Extreme Boosting?\nYear: 2002\n\nAbstract: A new view of majority voting as a Monte Carlo stochastic algorithm is presented in this paper. Relation between the two approaches allows Adaboost's example weighting strategy to be compared with the greedy covering strategy used for a long time in Machine Learning. The greedy covering strategy does not clearly show overfitting, it runs in at least one order of magnitude less time, it reaches zero error on the training set in few trials, and the error on the test set is most of the time comparable to that exhibited by AdaBoost.","Title: World Model Construction in Children during Physics Learning\nYear: 1997\n\nAbstract: A computational approach to the simulation of cognitive modelling of children learning elementary physics is presented. Goal of the simulation is to support the cognitive scientist's investigation of learning in humans. The Machine Learning system WHY, able to handle a causal model of the domain, has been chosen as tool for the simulation. In this paper the focus will be on the knowledge representation schemes, useful to support such a simulation.","Title: A Network Genetic Algorithm for Concept Learning\nYear: 1997\n\nAbstract: This paper presents a highly parallel genetic algo- rithm, designed for concept induction in proposi- tional and first order logics. The system exploits niches and species for learning multimodal con- cepts; it deeply differs from other systems be- cause of the distributed architecture, which to- tally eliminates the concept of common memory. A first implementation of the system, designed for checking the possibility of exploiting paral- lel processing in network computer, is evaluated on standard benchmarks. The experimental re- sults show that the system reaches good perfor- mances both with respect to the quality of the learned knowledge and with respect to the speed up on a workstation cluster.","Title: Learning Disjunctive Concepts by Means of Genetic Algorithms\nYear: 1994\n\nAbstract: REGAL is a Distributed Genetic Algorithm designed for learning concept descriptions from examples, in First Order Logic. In particular, each individual in the population represents a conjunctive formula in VL2 language. In order to increase the efficiency of the generalization process, REGAL has been provided with a new selection operator, called Universal Suffrage operator, which guarantees (in probability) to maintain a population covering all the learning events. As generalization mostly takes place when two individuals covering different sets of examples are crossed, the global generalization capability of the system is increased. Moreover, in the case of disjunctive or multiple concepts, the universal suffrage algorithm allows the formation of different species, each one corresponding to a different disjunct. In this way, all the disjuncts can be learned in parallel obtaining, in average, more general solutions than by learning them one at a time. A formal analysis of the universal suffrage operator is presented, providing theoretical explanations of the experimentally observed behaviour. A comparison with the classical selection algorithm and with the sharing function method is also made. Finally, a long term control strategy, called “Tories and Whigs”, is proposed in order to overcome the problem of lethal matings between uncompatible disjuncts. The effectiveness of REGAL is demonstrated on several learning problems.","Title: Integrated architectures for machine learning\nYear: 2001\n\nAbstract: With the growing complexity of Machine Learning applications, the need for using integrated or hybrid (or multistrategy) approaches\n becomes more and more imperative, and an increasing amount of research effort is devoted to this issue. The increasing complexity\n of applications is not the only reason making multistrategic approaches appealing: as it is well known, no single approach/system\n can claim to be uniformly superior to any other, so that hybridisation seems a natural and viable way of compensating drawbacks\n and enhancing advantages. Even though there is no common agreement on what integration exactly means in Machine Learning, in a broad sense an integrated architecture can be defined as one which is organised or structured so that its constituent units function co-operatively.","Title: Italian Machine Learning And Data Mining Research: The Last Years\nYear: 2013\n\nAbstract: With the increasing amount of information in electronic form the fields of Machine Learning and Data Mining continue to grow by providing new advances in theory, applications and systems. The aim of this paper is to consider some recent theoretical aspects and approaches to ML and DM with an emphasis on the Italian research.","Title: Hypothesis diversity in ensemble classification\nYear: 2006\n\nAbstract: The paper discusses the issue of hypothesis diversity in ensemble classifiers. The measures of diversity previously proposed in the literature are analyzed inside a unifying framework based on Monte Carlo stochastic algorithms. The paper shows that no measure is useful to predict ensemble performance, because all of them have only a very loose relation with the expected accuracy of the classifier.","Title: Abstracting Concepts with Inverse Resolution\nYear: 1991\n\nAbstract: Usually, abstraction is defined as a mapping between languages. In this paper it is, instead, defined as a mapping between models and extends Tenenberg's restricted predicate mapping. The mapping is axiomatized by means of a theory TA defining the semantics of the relations in the abstract model from those existing in the ground model. Therefore, this form of abstraction is semantic and must be evaluated using a deductive mechanism. A restricted class of semantic abstraction (CP-abstraction) is characterized, having the property of preserving the concept instances with respect to a given model. CP-abstraction fits into the paradigm of inverse resolution, which has been proposed as a framework for constructive learning. A restricted form of absorption rule is also introduced to compute it. Finally, a new operation, called term abstraction, is defined; it allows abstraction mechanisms, similar to the ones used to handle abstract data types, to be implemented.","Title: Use of a Causal Model to Learn Diagnostic Knowledge in a Real Domain\nYear: 1991\n\nAbstract: This paper presents a system which learns a diagnostic knowledge base using a-priori knowledge and a set of examples. The a-priori knowledge consists of a causal model of the domain and a body of phenomenological theory, describing the links between abstract concepts and their possible manifestations in the world. The phenomenological knowledge is used deductively, the causal model is used abductively and the examples are used inductively. The system has been applied to learn the knowledge base of a diagnostic expert system for mechanical trouble-shooting.","Title: An Experimental Study of Phase Transitions in Matching\nYear: 1999\n\nAbstract: Finding models of a predicate logic formula is a well-known hard problem, whose complexity is exponential in the number of variables. However, even though this number is kept constant, substantial differences in complexity arise when searching for solutions in different problem instances. Such a behavior appears to be quite general, according to recent results reported in the literature; in fact, several classes of hard problems exhibit a narrow phase transition with respect to some order parameter, in correspondence of which the complexity dramatically rises up, still remaining tractable elsewhere. In this paper we provide an extensive experimental study on the emergence of a phase transition in the problem of matching a Horn clause to a universe, searching for a model of the clause or for a proof that no such model exists. As it turns out, phase transition in the matching problem depends in an essential way on two order parameters, one capturing syntactic aspects of the clause structure (intensional aspect), while the other related to the structure of the universe (extensional aspect).","Title: Evidence Combination in Expert Systems\nYear: 1985\n\nAbstract: This paper discusses some of the problems related to the representation of uncertain knowledge and to the combination of evidence degrees in rule-based expert systems. Some of the methods proposed in the literature are briefly analysed with particular attention to the Subjective Bayesian Probability (used in PROSPECTOR) and the Confirmation Theory adopted in MYCIN.","Title: An expert system for mapping acoustic cues into phonetic features\nYear: 1984\n\nAbstract: The paper describes the conception of the sublexical levels of a speech-understanding system as a society of experts. Experts cooperate in extracting and describing acoustic cues, generating and verifying phonetic hypotheses, and accessing a large lexicon. The knowledge of each expert is described by a frame language which allows integration between structural and procedural knowledge. Structural knowledge deals with relations between facts like acoustic-cue descriptions and phonetic-feature hypotheses. Procedural knowledge deals with rules for the use of relations, for the generation of contextual constraints for relation application, and for the extraction of new cues in specified contexts. The main purpose of the research proposed here is that of providing at the same time a model for computer perception and algorithms useful for designing complex systems operating in real time. Some experimental results on the performance of the proposed system are reported.","Title: Can Relational Learning Scale Up?\nYear: 2000\n\nAbstract: A key step of supervised learning is testing whether a candidate hypothesis covers a given example. When learning in first order logic languages, the covering test is equivalent to a Constraint Satisfaction Problem (CSP). For critical values of some order parameters, CSPs present a phase pransition, that is, the probability of finding a solution abruptly drops from almost 1 to almost 0, and the complexity dramatically increases. This paper analyzes the complexity and feasibility of learning in first order logic languages with respect to the phase transition of the covering test.","Title: Automatic learning of fuzzy naming relations over finite languages\nYear: 1980\n\nAbstract: A method for learning the membership of strings belonging to finite languages is proposed. The learning is based on sets of strings of fuzzy linguistic variables. These strings belong to languages, each one of which describes a class of phenomena. The learning algorithm attempts to maximize the number of times a string reaches the highest possibility value for the language representing the class of phenomena containing the sample described by the string. Application to automatic speech recognition is described, and experimental results are presented showing the benefits of the proposed method.","Title: Phase Transitions in Relational Learning\nYear: 2000\n\nAbstract: One of the major limitations of relational learning is due to the complexity of verifying hypotheses on examples. In this paper we investigate this task in light of recent published results, which show that many hard problems exhibit a narrow “phase transition” with respect to some order parameter, coupled with a large increase in computational complexity. First we show that matching a class of artificially generated Horn clauses on ground instances presents a typical phase transition in solvability with respect to both the number of literals in the clause and the number of constants occurring in the instance to match. Then, we demonstrate that phase transitions also appear in real-world learning problems, and that learners tend to generate inductive hypotheses lying exactly on the phase transition. On the other hand, an extensive experimenting revealed that not every matching problem inside the phase transition region is intractable. However, unfortunately, identifying those that are feasible cannot be done solely on the basis of the order parameters. To face this problem, we propose a method, based on a Monte Carlo algorithm, to estimate on-line the likelihood that the current matching problem will exceed a given amount of computational resources. The impact of the above findings on relational learning is discussed.","Title: On-Line Estimation of Matching Complexity in First Order Logic\nYear: 1999\n\nAbstract: The expressiveness of First Order Logic (FOL) languages is severely counterbalanced by the complexity of matching formulas\n on a universe. Matching is an instance of the class of Constraint Satisfaction Problems (CSP), which have shown to undergo\n a phase transition with respect to two order parameters: constraint density and constraint tightness. This paper analyzes the problem of satisfying FOL Horn clauses in the light of these recent results. By means of an extensive\n experimental analysis, we show how Horn clause verification exhibits a typical phase transition with respect to the number\n of binary (or greater arity) predicates, and with respect to the ratio between the number of constants in the universe and\n the cardinality of the basic predicates extension.","Title: Rigel: An Inductive Learning System\nYear: 1991\n\nAbstract: This paper is aimed at showing the benefits obtained by explicitly introducing a priori control knowledge into the inductive process. The starting point is Michalski's Induce system, which has been modified and augmented. Although the basic philosophy has been changed as little as possible, Induce has been radically modified from the algorithmic point of view, resulting in the new learning system Rigel. The main ideas taken from Induce are the sequential learning of descriptions of each concept against all the others, the Covering algorithm, the Star definition, and the VL2 representation language. The modifications consist of a new way of computing the Star, the use of a separate body of heuristic knowledge to strongly direct the search, the implementation of a larger subset of the VL2 language, a reasoned way of selecting the seed, and the use of rules to evaluate the worthiness of the inductive assertions. The effectiveness of Rigel has been tested both on artificial and on real-world case studies.","Title: Image databases browsing by unsupervised learning\nYear: 2008\n\nAbstract: Content-Based Image Retrieval systems provide a variety of usages. The most common one is target search, in which a user is trying to find a specific target image. Instead, we present, in this paper, a flexible image dataset browsing system. The user can browse the whole dataset looking for any \"interesting\" image. To this aim, images are first abstracted through a set of signatures describing their color and texture composition. Afterwards, unsupervised clustering is performed to split the image set into several clusters of \"similar\" images. Every cluster is represented by its centroid as an icon. The set of icons is presented to the user, who can pick one in order to see the images belonging to the cluster. Multi-dimensional scaling is used to visualize images in the same cluster by mapping the images onto a two-dimensional space. The experiments performed with a general-purpose image dataset consisting of one thousand images, categorized into ten classes, show the usefulness of the system.","Title: Learning Relations Using Genetic Algorithms\nYear: 1993\n\nAbstract: Inducing concept descriptions from examples requires a large space of hypotheses to be explored. Genetic algorithms offer an appealing alternative to traditional search algorithms, because of their multi-point search strategy. In this paper, the new system REGAL is described: it uses genetic algorithms to learn first order logic concept descriptions. Moreover, it can be easily integrated with a deductive component, in order to exploit a domain theory. Two approaches to learning disjunctive concept descriptions are presented: the first one is a modification of the classical method of learning one disjunct at a time, whereas the second one is based on the idea of fitness sharing and tries to let subpopulations be spontaneously formed, according to the theory of the niches and species. The approaches have been compared on an artificial domain.","Title: Learning in the “Real World”\nYear: 1998\n\nAbstract: In this paper we define and characterize the process of developing a \"real-world\" Machine Learning application, with its difficulties and relevant i ssues, distinguishing it from t he popular practice of exploiting ready-to-use data sets. To this aim, we a nalyze a nd summarize the lessons learned from applying Machine Learning techniques to a variety of problems. We believe that these lessons, though primarily based on our personal experience, can be generalized to a wider r ange of situations and are supported by the reported experiences of other researchers.","Title: Learning profiles based on hierarchical hidden markov model\nYear: 2005\n\nAbstract: This paper presents a method for automatically constructing a sophisticated user/process profile from traces of user/process behavior. User profile is encoded by means of a Hierarchical Hidden Markov Model (HHMM). The HHMM is a well formalized tool suitable to model complex patterns in long temporal or spatial sequences. The method described here is based on a recent algorithm, which is able to synthesize the HHMM structure from a set of logs of the user activity. The algorithm follows a bottom-up strategy, in which elementary facts in the sequences (motives) are progressively grouped, thus building the abstraction hierarchy of a HHMM, layer after layer. The method is firstly evaluated on artificial data. Then a user identification task, from real traces, is considered. A preliminary experimentation with several different users produced encouraging results.","Title: Machine Learning for Information Extraction\nYear: 1997\n\nAbstract: This paper presents a brief overview of the history and trends of Machine Learning, organized according to its goal and principal methodologies. More details are given for the concept learning task, one of the most mature in the field. Applications to Information Extraction tasks are discussed.","Title: Multistrategy Learning and Theory Revision\nYear: 1993\n\nAbstract: This article presents the system WHY, which learns and updates a diagnostic knowledge base using domain knowledge and a set of examples. The a priori knowledge consists of a causal model of the domain that states the relationships among basic phenomena, and a body of phenomenological theory that describes the links between abstract concepts and their possible manifestations in the world. The phenomenological knowledge is used deductively, the causal model is used abductively, and the examples are used inductively. The problems of imperfection and intractability of the theory are handled by allowing the system to make assumptions during its reasoning. In this way, robust knowledge can be learned with limited complexity and a small number of examples. The system works in a first-order logic environment and has been applied in a real domain.","Title: Knowledge representation in machine learning\nYear: 1994\n\nAbstract: This paper investigates the influence of knowledge representation languages on the complexity of the learning process. However, the aim of the paper is not to give a state-of-the-art account of the involved issues, but to survey the underlying ideas. Then, references will be provided only occasionally and all the specific quantitative results are left to the presentation. Finally, the paper is intentionally unbalanced, because a larger space is given to those issues that are more novel or less investigated in the literature.","Title: A Model of Abstraction in Visual Perception\nYear: 2001\n\nAbstract: In artificial intelligence, abstraction has been mainly studied as a mapping between languages in relation to problem-solving, with the aim of reducing the complexity of the task. However, abstraction has a much larger scope in reasoning; we are investigating, in this article, how abstraction can be used in concept representation. To this aim, we propose a novel, perception-based model of abstraction, which originates from the observation that conceptualization of a domain, even though involving entities belonging to several epistemological levels, is nevertheless primarily based on perception. This view has been recently advocated by Goldstone and Barsalou in cognitive science. A model of representation/abstraction is then proposed and its application to a real-world problem of robot visual perception and categorization is presented.","Title: Structured Hidden Markov Model: A General Framework for Modeling Complex Sequences\nYear: 2007\n\nAbstract: Structured Hidden Markov Model (S-HMM) is a variant of Hierarchical Hidden Markov Model that shows interesting capabilities of extracting knowledge from symbolic sequences. In fact, the S-HMM structure provides an abstraction mechanism allowing a high level symbolic description of the knowledge embedded in S-HMM to be easily obtained. The paper provides a theoretical analysis of the complexity of the matching and training algorithms on S-HMMs. More specifically, it is shown that Baum-Welch algorithm benefits from the so called locality property, which allows specific components to be modified and retrained, without doing so for the full model. The problem of modeling duration and of extracting (embedding) readable knowledge from (into) a S-HMM is also discussed.","Title: A Comparison between Two Statistical Relational Models\nYear: 2008\n\nAbstract: Statistical Relational Learning has received much attention this last decade. In the ILP community, several models have emerged for modelling and learning uncertain knowledge, expressed in subset of first order logics. Nevertheless, no deep comparisons have been made among them and, given an application, determining which model must be chosen is difficult. In this paper, we compare two of them, namely Markov Logic Networks and Bayesian Programs, especially with respect to their representation ability and inference methods. The comparison shows that the two models are substantially different, from the point of view of the user, so that choosing one really means choosing a different philosophy to look at the problem. In order to make the comparison more concrete, we have used a running example, which shows most of the interesting points of the approaches, yet remaining exactly tractable.","Title: Boosting as a Monte Carlo Algorithm\nYear: 2001\n\nAbstract: A new view of majority voting as a Monte Carlo stochastic algorithm is presented in this paper. The relation between the two approches allows Adaboost?s example weighting strategy to be compared with the greedy covering strategy used for a long time in Machine Learning. Even though one may expect that the greedy strategy is very much prone to overfitting, extensive experimental results do not support this guess. The greedy strategy does not clearly show overfitting, it runs in at least one order of magnitude less time, it reaches zero error on the training set in few trials, and the error on the test set is most of the time comparable, if not lower, than that exhibited by Adaboost.","Title: An Abstraction-based Machine Learning Approach to Cartographic Generalisation\nYear: 2005\n\nAbstract: This article proposes a machine learning approach to overcome the knowledge acquisition bottleneck limiting the automation of cartographic generalisation. It first explains why this automation must be guided by a differentiation of two main types of knowledge involved in this process. More precisely, it shows that cartographic generalisation is best viewed as a combination of two processes: representing (formulating, renaming knowledge) and abstracting (simplifying a given representation). The whole process of creating maps fits into an abstraction framework developed in Artificial Intelligence to account for the difference between knowledge abstraction and knowledge representation. The utility of this framework lies in its efficiency to support the automation of knowledge acquisition for cartographic generalisation as a combined learning of both abstraction and representation knowledge. The results of Machine Learning experiments show the advantages of this approach.","Title: Learning complex event descriptions by abstraction\nYear: 2005\n\nAbstract: The presence of long gaps dramatically increases the difficulty of detecting and characterizing complex events hidden in long sequences. In order to cope with this problem, a learning algorithm based on an abstraction mechanism is proposed: it can infer a Hierarchical Hidden Markov Model, from a learning set of sequences. The induction algorithm proceeds bottom-up, progressively coarsening the sequence granularity, and letting correlations between subsequences, separated by long gaps, naturally emerge. As a case study, the method is evaluated on an application of user profiling. The results show that the proposed algorithm is suitable for developing real applications in network security and monitoring.","Title: An expert system oriented to complex pattern recognition problems\nYear: 1985\n\nAbstract: The general framework bimbo, designed for acquiring knowledge on expert systems oriented to pattern analysis, is presented. bimbo, which is itself an expert system, works in a largely automated mode and makes use of a large body of general knowledge and of powerful heuristics. The learning methodology explicitly addresses the problem of the presence of errors and noise in the data and utilizes an approximate reasoning scheme based on fuzzy sets. bimbo has been applied in a test case taken from the field of continuous speech recognition, with very good results in increasing the system accuracy and decreasing the demands upon human time.","Title: Incremental construction of structured hidden Markov models\nYear: 2007\n\nAbstract: This paper presents an algorithm for inferring a Structured Hidden Markov Model (S-HMM) from a set of sequences. The S-HMMs are a subclass of the Hierarchical Hidden Markov Models and are well suited to problems of process/user profiling. The learning algorithm is unsupervised, and follows a mixed bottom-up/top-down strategy, in which elementary facts in the sequences (motifs) are progressively grouped, thus building up the abstraction hierarchy of a S-HMM, layer after layer. The algorithm is validated on a suite of artificial datasets, where the challenge for the learning algorithm is to reconstruct the model that generated the data. Then, an application to a real problem of molecular biology is briefly described.","Title: Dealing with uncertain knowledge in medical decision-making: A case study in hepatology\nYear: 1989\n\nAbstract: It has widely been recognized that knowledge-based expert systems need efficient mechanisms to model the uncertainty associated with many decision-making activities. Such a need is particularly urgent in medicine. In this paper, we present an approach based on fuzzy logic to give a possible solution to this problem; its pros and cons are discussed by taking into account the experience gained in developing LITO1 and LITO2, two expert systems devoted to the assessment of the liver function and to the diagnosis of hepatic diseases. The advantages of mixing fuzzy production rules with frame-like structures (introduced for representing the clinical data) are discussed. In particular, the use of fuzzy linguistic variables for modeling the possible values of the clinical data is described: this allows, for a clear and perspicuous description of the correspondence between quantitative and qualitative expressions. Furthermore, different alternatives for evaluating and combining evidence are reviewed. Finally, the need of introducing frame structures also for representing diagnostic hypotheses is discussed, together with the problem of evaluating the fuzzy match between the prototypical description of a diagnostic hypothesis and the data describing the status of the specific patient under examination.","Title: Explaining Bagging with Monte Carlo Theory\nYear: 2003\n\nAbstract: In this paper we propose the use of the framework of Monte Carlo stochastic algorithms to analyze ensemble learning, specifically, bagging. In particular, this framework allows one to explain bagging's behavior and also why increasing the margin improves performances. Experimental results support the theoretical analysis.","Title: Formalizing the abstraction process in model-based diagnosis\nYear: 2007\n\nAbstract: Several theories have been proposed to capture the essence of abstraction. Among these, the KRA model offers a framework where a set of generic abstraction operators allows abstraction to be automated. In this paper we show how to describe component-based abstraction for the Model-Based Diagnosis task within the KRA framework, and we discuss the benefits of such a formalization. The clear and explicit partition of the system model into different levels required by KRA (going from the perception level up to the theory level) opens the way to explore richer and better founded kinds of abstraction to apply to the MBD task. Another noticeable advantage is that, by suitably personalizing the generic abstraction operators of KRA, the whole abstraction process, from the definition of abstract (macro)components to the computation of their behaviors starting from those of the ground components, can be performed automatically in such a way that important relationships between ground and abstract diagnoses are guaranteed.","Title: Experimental comparison between bagging and Monte Carlo ensemble classification\nYear: 2005\n\nAbstract: Properties of ensemble classification can be studied using the framework of Monte Carlo stochastic algorithms. Within this framework it is also possible to define a new ensemble classifier, whose accuracy probability distribution can be computed exactly. This paper has two goals: first, an experimental comparison between the theoretical predictions and experimental results; second, a systematic comparison between bagging and Monte Carlo ensemble classification.","Title: Abstracting Visual Percepts to Learn Concepts\nYear: 2002\n\nAbstract: To efficiently identify properties from its environment is an essential ability of a mobile robot who needs to interact with humans. Successful approaches to provide robots with such ability are based on ad-hoc perceptual representation provided by AI designers. Instead, our goal is to endow autonomous mobile robots (in our experiments a Pioneer 2DX) with a perceptual system that can efficiently adapt itself to ease the learning task required to anchor symbols. Our approach is in the line of meta-learning algorithms that iteratively change representations so as to discover one that is well fitted for the task. The architecture we propose may be seen as a combination of the two widely used approach in feature selection: the Wrapper-model and the Filter-model. Experiments using the PLIC system to identify the presence of Humans and Fire Extinguishers show the interest of such an approach, which dynamically abstracts a well fitted image description depending on the concept to learn.","Title: Improving Learning Using Causality and Abduction\nYear: 1991\n\nAbstract: This paper presents a system which learns and maintains a diagnostic knowledge base using a causal model of the domain, a body of phenomenological theory and a set of examples. The phenomenological knowledge is used deductively, the causal model is used abductively and the examples are used inductively. The problems of imperfection and intractability of the theory are handled by allowing the system to make assumptions during its reasoning. In this way, robust knowledge can be learned with limited complexity and limited number of examples. An example in the domain of heat transfer is presented.","Title: A Fine-Grained Parallel Evolutionary Program for Concept Induction\nYear: 1996\n\nAbstract: This paper presents a highly parallel genetic algorithm, called G-NET, designed for concept induction in propositional and first order logics. As well as other systems oriented to the same task, G-NET exploits niches and species for learning multimodal concepts; on the other hand it deeply differs from other systems because of the distributed architecture, which totally eliminates the concept of common memory. A simulator of the system, designed in order to check the possibility of exploiting parallel processing, is evaluated on a standard benchmark. The experimental results show that a multi-processor implemented with standard technology could reach speed-up of thousands of times.","Title: Integrating Multiple Learning Strategies in First Order Logics\nYear: 1997\n\nAbstract: This paper describes a representation framework that offers aunifying platform for alternative systems, which learn concepts inFirst Order Logics. The main aspects of this framework arediscussed. First of all, the separation between the hypothesislogical language (a version of the VL21 language) and therepresentation of data by means of a relational database ismotivated. Then, the functional layer between data and hypotheses,which makes the data accessible by the logical level through a set ofabstract properties is described. A novelty, in the hypothesisrepresentation language, is the introduction of the construct ofinternal disjunction; such a construct, first used by the AQ andInduce systems, is here made operational via a set of algorithms,capable to learn it, for both the discrete and the continuous-valuedattributes case. These algorithms are embedded in learning systems(SMART+, REGAL, SNAP, WHY, RTL) using different paradigms (symbolic,genetic or connectionist), thus realizing an effective integrationamong them; in fact, categorical and numerical attributes can behandled in a uniform way. In order to exemplify the effectivenessof the representation framework and of the multistrategy integration,the results obtained by the above systems in some application domainsare summarized.","Title: Classification: From Expert To Ensemble To Crowd\nYear: 2014\n\nAbstract: In this paper a model of the classification task, based on the Monte Carlo theory of stochastic algorithms, is presented. The model allows different classification paradigms to be described, such as the single classifier, the ensemble, and the crowdsourcing ones. Some experimental work validates the theoretical predictions about the relatioship between the mean and variance of the classification accuracy and the number of classifiers.","Title: Automated concept acquisition in noisy environments\nYear: 1988\n\nAbstract: A system that performs automated concept acquisition from examples and has been specially designed to work in noisy environments is presented. The learning methodology is aimed at the target problem of finding discriminant descriptions of a given set of concepts and uses both examples and counterexamples. The learned knowledge is expressed in the form of production rules, organized into separate clusters, linked together in a graph structure. Knowledge extraction is guided by a top-down control strategy, through a process of specialization. The system also utilizes a technique of problem reduction to contain the computational complexity. Several criteria are proposed for evaluating the acquired knowledge. The methodology has been tested on a problem in the field of speech recognition and the experimental results obtained are reported and discussed.","Title: Pattern recognition and Valiant's learning framework\nYear: 1993\n\nAbstract: The computational learning approach shows that the concept descriptions acquired from examples are approximately correct with a degree of probability that grows with the size of the training sample. The same problem has also been widely investigated in the field of pattern recognition under a variety of problem settings. Some of the results obtained in both fields are surveyed and compared, and the limits of their applicability are analyzed. Moreover, new and tighter bounds for the growth function of some classes of Boolean formulas are presented.","Title: Fuzzy semantic network for a speech understanding system - an experimental study\nYear: 1978\n\nAbstract: The relevance of the Semantics for a Speech Understanding System (SUS) was pointed out by many authors and is by now a classic topic. The present work contains an implementation of a semantic analysis for the SUS of the CENS, in Turin. The semantic knowledge is contained in a graph on which a set of fuzzy relations is defined. The semantic algorithm accepts, as input, a table of alternative weighted words, found in the processed spectrogram at phonological and lexical level. The output is a sentence assumed as the \"best\" interpretation of the whole spectrogram. The algorithm is implemented in FORTRAN and PL1 on a IBM 370/158 and it utilizes a recursive evaluation function for the concurrent hypotheses. We compare experimentally the efficiency of the algorithm, by varying the number of alternative input words and their weight, the number of vertices and arcs of the semantic graph and the evaluation function. The results show a good performance of this pure semantic analysis.","Title: Exploring the power of genetic search in learning symbolic classifiers\nYear: 1996\n\nAbstract: In this paper we show, in a constructive way, that there are problems for which the use of genetic algorithm based learning systems can be at least as effective as traditional symbolic or connectionist approaches. To this aim, the system REGAL* is briefly described, and its application to two classical benchmarks for Machine Learning is discussed, by comparing the results with the best ones published in the literature.","Title: A motion detection system for video surveillance\nYear: 2013\n\nAbstract: In this paper we describe a motion detection module, which is part of Horus, a video surveillance system already deployed in field. The module integrates and improves upon existing algorithms, whose combination proves to be particularly effective in automatically detecting motion sequences in very noisy outdoor environments, with both diurnal and nocturnal illumination conditions. The system has been tested on some public-domain benchmarks, and also on a real application.","Title: Piero Torasso'S Early Contributions\nYear: 2018\n\nAbstract: In this paper a brief overview of Pietro Torasso's academic career and early scientific work is provided. Particular attention is given to his interests in Artificial Intelligence topics."],"string":"[\n \"Title: Learning on the phase transition edge\\nYear: 2001\\n\\nAbstract: Extracting rules from RBFs is not a trivial task because of nonlinear functions or high input dimensionality. In such cases, some of the hidden units of the RBF network have a tendency to be \\\"shared\\\" across several output classes or even may not contribute ...\",\n \"Title: Coevolutionary, Distributed Search for Inducing Concept Description\\nYear: 1998\\n\\nAbstract: This paper presents a highly parallel genetic algorithm, designed for concept induction in propositional and first order logics. The parallel architecture is an adaptation for set covering problems, of the diffusion model developed for optimization.\",\n \"Title: Feature construction from synergic pairs to improve microarray-based classification.\\nYear: 2007\\n\\nAbstract: Microarray experiments that allow simultaneous expression profiling of thousands of genes in various conditions (tissues, cells or time) generate data whose analysis raises difficult problems. In particular, there is a vast disproportion between the number of attributes (tens of thousands) and the number of examples (several tens). Dimension reduction is therefore a key step before applying classification approaches. Many methods have been proposed to this purpose, but only a few of them considered a direct quantification of transcriptional interactions. We describe and experimentally validate a new dimension reduction and feature construction method, which assesses interactions between expression profiles to improve microarray-based classification accuracy.Our approach relies on a mutual information measure that exposes some elementary constituents of the information contained in a pair of gene expression profiles. We show that their analysis implies a term that represents the information of the interaction between the two genes. The principle of our method, called FeatKNN, is to exploit the information provided by highly synergic gene pairs to improve classification accuracy. First, a heuristic search selects the most informative gene pairs. Then, for each selected pair, a new feature, representing the classification margin of a KNN classifier in the gene pairs space, is constructed. We show experimentally that the interactional information has a degree of significance comparable to that of the gene expression profiles considered separately. Our method has been tested with different classifiers and yielded significant improvements in accuracy on several public microarray databases. Moreover, a synthetic assessment of the biological significance of the concept of synergic gene pairs suggested its ability to uncover relevant mechanisms underlying interactions among various cellular processes.\",\n \"Title: Monte Carlo theory as an explanation of bagging and boosting\\nYear: 2003\\n\\nAbstract: In this paper we propose the framework of Monte Carlo algorithms as a useful one to analyze ensemble learning. In particular, this framework allows one to guess when bagging will be useful, explains why increasing the margin improves performances, and suggests a new way of performing ensemble learning and error estimation.\",\n \"Title: Abstraction and Complexity Measures\\nYear: 2007\\n\\nAbstract: ion is fundamental for both human and artificial reasoning. The word denotes different activities and process, but\\n all are intuitively related to the notion of complexity/simplicity, which is as elusive a notion as abstraction. From an analysis\\n of the literature on abstraction and complexity it clearly appears that it is unrealistic to find definitions valid in all\\n disciplines and for all tasks. Hence, we consider a particular model of abstraction, and try to investigate how complexity\\n measures could be mapped to it. Preliminary results show that abstraction and complexity are not monotonically coupled notions,\\n and that complexity may either increase or decrease with abstraction according to the definition of both and to the specificities\\n of the considered domain.\",\n \"Title: Abstracting Complex Interaction Networks\\nYear: 2009\\n\\nAbstract: The exploration of complex interaction networks has attracted considerable interest in various fields, rang- ing from fundamental biology and medicine to statis- tical physics and information technology. In \\\"-omics\\\" disciplines, significant progresses have been made in understanding the large-scale properties and the bio- logical relevance of these interactions. Some properties such as \\\"scale-free\\\" distribution of nodes connectivity or \\\"centrality\\\" are aspects commonly described in such complex interaction systems. In many of these studies the analysis of network topology is complemented by a semantic analysis that may rely on dierent labels asso- ciated to the interacting entities. One of the bottleneck of these semantic analysis is that they are computa- tionally costly. In this paper we present a framework to explore abstraction of networks useful to speedup the computation of ground network measures. Such abstraction mechanisms may be used to eciently pro- vide accurate approximations of ground network mea- sures.\",\n \"Title: Learning Simple Recursive Theories\\nYear: 1993\\n\\nAbstract: The task of learning relations has been concerned, so far, with the acquisition of intensional descriptions of unrelated concepts. However, in many real domains concepts are strictly related to each other and the instances of one of them cannot possibly be recognised without previous recognition of other objects as instances of related concepts. A typical case is the problem of labelling parts of a scene in order to interpret it. This paper extends in several ways the learning relations paradigm; in particular, a new methodology, allowing a recursive theory to be inferred from a set of examples, is presented. The learning algorithm works bottom-up, creating first an acyclic graph that classifies all the instances in the training set. Afterwards, a recursive theory is synthesised from the graph.\",\n \"Title: Learning Mutually Dependent Relations\\nYear: 1992\\n\\nAbstract: So far, the task of learning relations has been concerned with the acquisition of intensional descriptions of unrelated concepts. However, in many real domains concepts are strictly related to each other and the instances of one of them cannot possibly be recognized without previous recognition of other objects as instances of related concepts. A typical case is the problem of labeling parts of a scene in order to interprete it. This paper extends in several ways the learning relations paradigm; in particular, a new methodology, allowing a recursive theory to be inferred from a set of examples, is presented. Each example may contain instances of several relations at the same time. The learning algorithm works bottom up, creating first an acyclic graph that classifies all the instances in the training set. Afterward, a recursive theory is synthesized from the graph. A prototype implementation of the algorithm has been realized by extending the program ML-SMART. The results obtained for an artificial labeling problem are described and discussed.\",\n \"Title: Investigating the Relationships between Abduction and Inverse Resolution in Proposition Calculus\\nYear: 1991\\n\\nAbstract: In systems devoted to diagnostic tasks, the term abduction usually denotes the process of explaining a set of observed manifestations, in the light of an existing domain theory. This amounts to find a set of hypotheses such that the manifestations logically follow from them and from the theory. Then, abduction corresponds to reasoning from consequences to possible premises. On the other hand, an analogous line of reasoning has been independently followed in constructive induction, where new predicates are introduced and theories are completed by trying to invert the resolution mechanism. In this paper, links are established between these two approaches by showing that the inverse resolution operators are sound abductive inference rules, in the case of propositional calculus. Some relationships with deduction in a closed world assumption are also investigated.\",\n \"Title: An Expert System for Interpreting Speech Patterns\\nYear: 1982\\n\\nAbstract: Efficient syllabic hypothesization in continuous speech has been so far an unsolved problem. A novel solution based on the extraction of acoustic cues is proposed in this paper. This extraction is performed by parallel processes implementing an expert system represented by a grammar of frames.\",\n \"Title: Multi Level Knowledge in Modeling Qualitative Physics Learning\\nYear: 2000\\n\\nAbstract: The goal of the reported research is the development of a computational approach that could help a cognitive scientist to interactively represent a learner's mental models, and to automatically v alidate their coherence with respect to the a vailable e xperimental data. In a reported case-study, the student's mental models are inferred from questionnaires and interviews collected during a sequence of teaching sessions. These putative c ognitive models are based o n a theory o f knowledge representation, derived from psychological results and educational studies, which accounts for the e volution of the student's knowledge over a learning period. The learning system WHY, able to handle (causal) domain knowledge, shows how to model the answers and the causal explanations given by the learner.\",\n \"Title: Extending Inverse Resolution to Build Up Abstractions\\nYear: 1991\\n\\nAbstract: ion is a conceptual framework potentially unifying and integrating different methodologies developed in machine learning and problem solving. According to the common understanding of the term, abstraction is a mapping between languages, and, then, in many learning tasks it can be so formulated. On the contrary, in this paper abstraction is defined as a mapping between models, implicitly extending Tenenberg's restricted predicate mapping in order to include more complex abstraction schemes and to cope with the problem of inconsistency in the abstract model.\",\n \"Title: Modeling production rules by Means of Predicate Transition networks\\nYear: 1985\\n\\nAbstract: A formal tool, the PREDICATE TRANSITION networks, is proposed for representing rules in production systems. These networks have been introduced by Genrich and Lautenbach as a kind of high-level Petri nets. The peculiarity of such a formal tool lies in its well-formalized semantics, which can be described by means of an algebraic formalism and can be analyzed using the S-invariant method in order to prove properties of the network. A brief review of PREDICATE TRANSITION networks is followed by a description of their use in AI applications. In particular the characteristics which make them suitable for representing knowledge and control strategies in expert systems are analyzed.\",\n \"Title: A Coevolutionary Approach to Concept Learning\\nYear: 1997\\n\\nAbstract: This paper presents a highly parallel genetic algorithm, designed for concept induction in propositional and First Order logics. The parallel architecture is an adaptation to set covering problems of the diffusion model developed for optimization problems.\",\n \"Title: Statistical Characterization of a Computer Grid\\nYear: 2009\\n\\nAbstract: Large-scale statistical analysis of more than 28 million jobs collected during 20 months of grid activity was undertaken in\\n order to examine the relations between users, computing elements and jobs in the network. The results give insight into the\\n global system behaviour and can be used to build models applicable in various contexts of grid computing. As an example, we\\n here construct probabilistic models that prove to be able to accurately predict job abortion.\",\n \"Title: Formal Models of Selection in Genetic Algorithms\\nYear: 1994\\n\\nAbstract: In this paper three formal models of selection operators (two known from the literature and one newly porposed) for genetic algorithms, used to learn structured concepts descriptions containing small disjuncts, are presented. The evolution of a population, according to these operators, with a generation gap equal to or less than one, is investigated in order to assess their ability to let different species emerge and coexist; finally, the average asymptotic behaviour of the population is determined. Experimental results confirm both bevahiours previously reported in the literature and the prediction of the new model.\",\n \"Title: Relational Learning: Hard Problems and Phase Transitions\\nYear: 1999\\n\\nAbstract: This paper focuses on a major step of machine learning, namely checking whether an example matches a candidate hypothesis. In relational learning, matching can be viewed as a Constraint Satisfaction Problem (CSP). The complexity of the task is analyzed in the Phase Transition framework, investigating the impact on the effectiveness of two relational learners: FOIL and G-NET. The critical factors of complexity, and their critical values, are experimentally investigated on artificial problems. This leads to distinguish several kinds of learning domains, depending on whether the target concept lies in the \\\"mushy\\\" region or not. Interestingly, experiments done with FOIL and G-NET show that both learners tend to induce hypotheses generating matching problems located inside the phase transition region, even if the constructed target concept lies far outside. Moreover, target concepts constructed too close to the phase transition are hard and both learners fail. The paper offers an explanation for this fact, and proposes a classification of learning domains and their hardness.\",\n \"Title: Relational learning as search in a critical region\\nYear: 2003\\n\\nAbstract: Machine learning strongly relies on the covering test to assess whether a candidate hypothesis covers training examples. The present paper investigates learning relational concepts from examples, termed relational learning or inductive logic programming. In particular, it investigates the chances of success and the computational cost of relational learning, which appears to be severely affected by the presence of a phase transition in the covering test. To this aim, three up-to-date relational learners have been applied to a wide range of artificial, fully relational learning problems. A first experimental observation is that the phase transition behaves as an attractor for relational learning; no matter which region the learning problem belongs to, all three learners produce hypotheses lying within or close to the phase transition region. Second, a failure region appears. All three learners fail to learn any accurate hypothesis in this region. Quite surprisingly, the probability of failure does not systematically increase with the size of the underlying target concept: under some circumstances, longer concepts may be easier to accurately approximate than shorter ones. Some interpretations for these findings are proposed and discussed.\",\n \"Title: Modelling Conceptual Change: An Interdisciplinary Approach\\nYear: 1997\\n\\nAbstract: A computational approach to the simulation of cognitive models of conceptual change in children learning elementary physics is presented. The student's mental model is inferred from a sequence of interviews collected along a period of eleven teaching sessions. Goal of the simulation is to support the cognitive scientist's investigation of learning in humans. The hypothesized cognitive models are based on a theory of conceptual change, derived from psychology results and educational experiences, which accounts for the evolution of the student's knowledge over a learning period. A Machine Learning (ML) system, able to handle domain knowledge (including a causal model of the domain), has been chosen as tool for the simulation of the cognitive models evolution. The system performs knowledge revision and provides causal explanation for its conclusions.\",\n \"Title: Transition de phase et apprentissage relationnel : une analyse critique\\nYear: 2003\\n\\nAbstract: Ce papier presente une revue et une analyse des resultats obtenus par les auteurs et par leurs collaborateurs en ce qui concerne lu0027effet de la transition de phase dans le probleme de lu0027appartement. Ce qui a ete mis en evidence est que la recherche du0027hypotheses, guidee par les heuristiques plus connues (gain du0027information ou simplicite), finissent inevitablement dans la region TP: donc, on ne peut pas eviter cette region a haute complexite. Les conclusions quu0027on peut en tirer pour lu0027apprentissage relationnel est quu0027il semble difficile de pouvoir approcher des problemes plus importants que ceux que lu0027on a actuellement.\",\n \"Title: Experimental evaluation of hierarchical hidden markov models\\nYear: 2005\\n\\nAbstract: Building profiles for processes and for interactive users is a important task in intrusion detection. This paper presents the results obtained with a Hierarchical Hidden Markov Model. The algorithm discovers typical ”motives” of a process behavior, and correlates them into a hierarchical model. Motives can be interleaved with possibly long gaps where no regular behavior is detectable. We assume that motives could be affected by noise,modeled as insertion, deletion and substitution errors. In this paper the learning algorithm is briefly recalled and then it is experimentally evaluated on three profiling case studies. The first case is built on a suite of artificial traces automatically generated by a set of given HHMMs. The challenge for the algorithm is to reconstruct the original model from the traces. It will be shown that the algorithm is able to learn HHMMs very similar to the original ones, in presence of noise and distractors. The second and third case studies refer to the problem of constructing a discriminant model for a user typing on a keyboard.\",\n \"Title: Perception-Based Granularity Levels in Concept Representation\\nYear: 2000\\n\\nAbstract: In this paper we propose a perception-based view of abstraction, which originates from the observation that conceptualization of a domain involves entities belonging to several episteological levels. The fundamental level corresponds to the preception of a world. For memorization purposes, some kind of structure is needed, in order to organize objects and relations perceived in the world into coherent ensembles. To communicate with others, a language must be invented, and, finally, a theory makes it possible to reason about the world. After invented, suitable properties abstraction should have to be useful for concept representation, examples of abstraction operators, designed to perform the abstraction process in practice, will be introduced.\",\n \"Title: Learning disjunctive concepts with distributed genetic algorithms\\nYear: 1994\\n\\nAbstract: Inducing concept descriptions from examples is a machine learning task that can be formulated as a search problem. Recently, genetic algorithms proved to be an appealing alternative for searching concept description spaces, because of their great exploration power and their suitability to exploit massive parallelism. In order to exploit genetic algorithms for searching the space of formulas φ (the elements of the population), we must reformulate the set covering problem by suitably defining the representation scheme for the formulas, the genetic operators and the fitness function. The system described in this paper, REGAL (Giordana and Saitta, 1993), encodes first order logic (FOL) formulas φ into bit strings of fixed length. In order to make this possible, REGAL's representation language, L, has been constrained to be finite, by defining a language template, Λ, which represents the maximally complex formula in L. The resulting language contains FOL formulas built up with conjunction, disjunction, negation, internal disjunction and existential and universal quantification\",\n \"Title: An analysis of the “universal suffrage” selection operator\\nYear: 1996\\n\\nAbstract: The “universal suffrage” selection operator, designed primarily for concept learning inside the system REGAL, is discussed for both overlapping and nonoverlapping populations. Analysis of its behavior is performed by using the “virtual average population” method, a new tool for investigating asymptotic properties of convergence of macroscopic quantities related to the population of a genetic algorithm.\",\n \"Title: Scheduling of processes in a speech understanding system based on approximate reasoning\\nYear: 1979\\n\\nAbstract: A method for scheduling knowledge source instantiations in a Speech Understanding System is presented. It is based on the evaluation of linguistic probabilities and on approximate reasoning.\",\n \"Title: Is a Greedy Covering Strategy an Extreme Boosting?\\nYear: 2002\\n\\nAbstract: A new view of majority voting as a Monte Carlo stochastic algorithm is presented in this paper. Relation between the two approaches allows Adaboost's example weighting strategy to be compared with the greedy covering strategy used for a long time in Machine Learning. The greedy covering strategy does not clearly show overfitting, it runs in at least one order of magnitude less time, it reaches zero error on the training set in few trials, and the error on the test set is most of the time comparable to that exhibited by AdaBoost.\",\n \"Title: World Model Construction in Children during Physics Learning\\nYear: 1997\\n\\nAbstract: A computational approach to the simulation of cognitive modelling of children learning elementary physics is presented. Goal of the simulation is to support the cognitive scientist's investigation of learning in humans. The Machine Learning system WHY, able to handle a causal model of the domain, has been chosen as tool for the simulation. In this paper the focus will be on the knowledge representation schemes, useful to support such a simulation.\",\n \"Title: A Network Genetic Algorithm for Concept Learning\\nYear: 1997\\n\\nAbstract: This paper presents a highly parallel genetic algo- rithm, designed for concept induction in proposi- tional and first order logics. The system exploits niches and species for learning multimodal con- cepts; it deeply differs from other systems be- cause of the distributed architecture, which to- tally eliminates the concept of common memory. A first implementation of the system, designed for checking the possibility of exploiting paral- lel processing in network computer, is evaluated on standard benchmarks. The experimental re- sults show that the system reaches good perfor- mances both with respect to the quality of the learned knowledge and with respect to the speed up on a workstation cluster.\",\n \"Title: Learning Disjunctive Concepts by Means of Genetic Algorithms\\nYear: 1994\\n\\nAbstract: REGAL is a Distributed Genetic Algorithm designed for learning concept descriptions from examples, in First Order Logic. In particular, each individual in the population represents a conjunctive formula in VL2 language. In order to increase the efficiency of the generalization process, REGAL has been provided with a new selection operator, called Universal Suffrage operator, which guarantees (in probability) to maintain a population covering all the learning events. As generalization mostly takes place when two individuals covering different sets of examples are crossed, the global generalization capability of the system is increased. Moreover, in the case of disjunctive or multiple concepts, the universal suffrage algorithm allows the formation of different species, each one corresponding to a different disjunct. In this way, all the disjuncts can be learned in parallel obtaining, in average, more general solutions than by learning them one at a time. A formal analysis of the universal suffrage operator is presented, providing theoretical explanations of the experimentally observed behaviour. A comparison with the classical selection algorithm and with the sharing function method is also made. Finally, a long term control strategy, called “Tories and Whigs”, is proposed in order to overcome the problem of lethal matings between uncompatible disjuncts. The effectiveness of REGAL is demonstrated on several learning problems.\",\n \"Title: Integrated architectures for machine learning\\nYear: 2001\\n\\nAbstract: With the growing complexity of Machine Learning applications, the need for using integrated or hybrid (or multistrategy) approaches\\n becomes more and more imperative, and an increasing amount of research effort is devoted to this issue. The increasing complexity\\n of applications is not the only reason making multistrategic approaches appealing: as it is well known, no single approach/system\\n can claim to be uniformly superior to any other, so that hybridisation seems a natural and viable way of compensating drawbacks\\n and enhancing advantages. Even though there is no common agreement on what integration exactly means in Machine Learning, in a broad sense an integrated architecture can be defined as one which is organised or structured so that its constituent units function co-operatively.\",\n \"Title: Italian Machine Learning And Data Mining Research: The Last Years\\nYear: 2013\\n\\nAbstract: With the increasing amount of information in electronic form the fields of Machine Learning and Data Mining continue to grow by providing new advances in theory, applications and systems. The aim of this paper is to consider some recent theoretical aspects and approaches to ML and DM with an emphasis on the Italian research.\",\n \"Title: Hypothesis diversity in ensemble classification\\nYear: 2006\\n\\nAbstract: The paper discusses the issue of hypothesis diversity in ensemble classifiers. The measures of diversity previously proposed in the literature are analyzed inside a unifying framework based on Monte Carlo stochastic algorithms. The paper shows that no measure is useful to predict ensemble performance, because all of them have only a very loose relation with the expected accuracy of the classifier.\",\n \"Title: Abstracting Concepts with Inverse Resolution\\nYear: 1991\\n\\nAbstract: Usually, abstraction is defined as a mapping between languages. In this paper it is, instead, defined as a mapping between models and extends Tenenberg's restricted predicate mapping. The mapping is axiomatized by means of a theory TA defining the semantics of the relations in the abstract model from those existing in the ground model. Therefore, this form of abstraction is semantic and must be evaluated using a deductive mechanism. A restricted class of semantic abstraction (CP-abstraction) is characterized, having the property of preserving the concept instances with respect to a given model. CP-abstraction fits into the paradigm of inverse resolution, which has been proposed as a framework for constructive learning. A restricted form of absorption rule is also introduced to compute it. Finally, a new operation, called term abstraction, is defined; it allows abstraction mechanisms, similar to the ones used to handle abstract data types, to be implemented.\",\n \"Title: Use of a Causal Model to Learn Diagnostic Knowledge in a Real Domain\\nYear: 1991\\n\\nAbstract: This paper presents a system which learns a diagnostic knowledge base using a-priori knowledge and a set of examples. The a-priori knowledge consists of a causal model of the domain and a body of phenomenological theory, describing the links between abstract concepts and their possible manifestations in the world. The phenomenological knowledge is used deductively, the causal model is used abductively and the examples are used inductively. The system has been applied to learn the knowledge base of a diagnostic expert system for mechanical trouble-shooting.\",\n \"Title: An Experimental Study of Phase Transitions in Matching\\nYear: 1999\\n\\nAbstract: Finding models of a predicate logic formula is a well-known hard problem, whose complexity is exponential in the number of variables. However, even though this number is kept constant, substantial differences in complexity arise when searching for solutions in different problem instances. Such a behavior appears to be quite general, according to recent results reported in the literature; in fact, several classes of hard problems exhibit a narrow phase transition with respect to some order parameter, in correspondence of which the complexity dramatically rises up, still remaining tractable elsewhere. In this paper we provide an extensive experimental study on the emergence of a phase transition in the problem of matching a Horn clause to a universe, searching for a model of the clause or for a proof that no such model exists. As it turns out, phase transition in the matching problem depends in an essential way on two order parameters, one capturing syntactic aspects of the clause structure (intensional aspect), while the other related to the structure of the universe (extensional aspect).\",\n \"Title: Evidence Combination in Expert Systems\\nYear: 1985\\n\\nAbstract: This paper discusses some of the problems related to the representation of uncertain knowledge and to the combination of evidence degrees in rule-based expert systems. Some of the methods proposed in the literature are briefly analysed with particular attention to the Subjective Bayesian Probability (used in PROSPECTOR) and the Confirmation Theory adopted in MYCIN.\",\n \"Title: An expert system for mapping acoustic cues into phonetic features\\nYear: 1984\\n\\nAbstract: The paper describes the conception of the sublexical levels of a speech-understanding system as a society of experts. Experts cooperate in extracting and describing acoustic cues, generating and verifying phonetic hypotheses, and accessing a large lexicon. The knowledge of each expert is described by a frame language which allows integration between structural and procedural knowledge. Structural knowledge deals with relations between facts like acoustic-cue descriptions and phonetic-feature hypotheses. Procedural knowledge deals with rules for the use of relations, for the generation of contextual constraints for relation application, and for the extraction of new cues in specified contexts. The main purpose of the research proposed here is that of providing at the same time a model for computer perception and algorithms useful for designing complex systems operating in real time. Some experimental results on the performance of the proposed system are reported.\",\n \"Title: Can Relational Learning Scale Up?\\nYear: 2000\\n\\nAbstract: A key step of supervised learning is testing whether a candidate hypothesis covers a given example. When learning in first order logic languages, the covering test is equivalent to a Constraint Satisfaction Problem (CSP). For critical values of some order parameters, CSPs present a phase pransition, that is, the probability of finding a solution abruptly drops from almost 1 to almost 0, and the complexity dramatically increases. This paper analyzes the complexity and feasibility of learning in first order logic languages with respect to the phase transition of the covering test.\",\n \"Title: Automatic learning of fuzzy naming relations over finite languages\\nYear: 1980\\n\\nAbstract: A method for learning the membership of strings belonging to finite languages is proposed. The learning is based on sets of strings of fuzzy linguistic variables. These strings belong to languages, each one of which describes a class of phenomena. The learning algorithm attempts to maximize the number of times a string reaches the highest possibility value for the language representing the class of phenomena containing the sample described by the string. Application to automatic speech recognition is described, and experimental results are presented showing the benefits of the proposed method.\",\n \"Title: Phase Transitions in Relational Learning\\nYear: 2000\\n\\nAbstract: One of the major limitations of relational learning is due to the complexity of verifying hypotheses on examples. In this paper we investigate this task in light of recent published results, which show that many hard problems exhibit a narrow “phase transition” with respect to some order parameter, coupled with a large increase in computational complexity. First we show that matching a class of artificially generated Horn clauses on ground instances presents a typical phase transition in solvability with respect to both the number of literals in the clause and the number of constants occurring in the instance to match. Then, we demonstrate that phase transitions also appear in real-world learning problems, and that learners tend to generate inductive hypotheses lying exactly on the phase transition. On the other hand, an extensive experimenting revealed that not every matching problem inside the phase transition region is intractable. However, unfortunately, identifying those that are feasible cannot be done solely on the basis of the order parameters. To face this problem, we propose a method, based on a Monte Carlo algorithm, to estimate on-line the likelihood that the current matching problem will exceed a given amount of computational resources. The impact of the above findings on relational learning is discussed.\",\n \"Title: On-Line Estimation of Matching Complexity in First Order Logic\\nYear: 1999\\n\\nAbstract: The expressiveness of First Order Logic (FOL) languages is severely counterbalanced by the complexity of matching formulas\\n on a universe. Matching is an instance of the class of Constraint Satisfaction Problems (CSP), which have shown to undergo\\n a phase transition with respect to two order parameters: constraint density and constraint tightness. This paper analyzes the problem of satisfying FOL Horn clauses in the light of these recent results. By means of an extensive\\n experimental analysis, we show how Horn clause verification exhibits a typical phase transition with respect to the number\\n of binary (or greater arity) predicates, and with respect to the ratio between the number of constants in the universe and\\n the cardinality of the basic predicates extension.\",\n \"Title: Rigel: An Inductive Learning System\\nYear: 1991\\n\\nAbstract: This paper is aimed at showing the benefits obtained by explicitly introducing a priori control knowledge into the inductive process. The starting point is Michalski's Induce system, which has been modified and augmented. Although the basic philosophy has been changed as little as possible, Induce has been radically modified from the algorithmic point of view, resulting in the new learning system Rigel. The main ideas taken from Induce are the sequential learning of descriptions of each concept against all the others, the Covering algorithm, the Star definition, and the VL2 representation language. The modifications consist of a new way of computing the Star, the use of a separate body of heuristic knowledge to strongly direct the search, the implementation of a larger subset of the VL2 language, a reasoned way of selecting the seed, and the use of rules to evaluate the worthiness of the inductive assertions. The effectiveness of Rigel has been tested both on artificial and on real-world case studies.\",\n \"Title: Image databases browsing by unsupervised learning\\nYear: 2008\\n\\nAbstract: Content-Based Image Retrieval systems provide a variety of usages. The most common one is target search, in which a user is trying to find a specific target image. Instead, we present, in this paper, a flexible image dataset browsing system. The user can browse the whole dataset looking for any \\\"interesting\\\" image. To this aim, images are first abstracted through a set of signatures describing their color and texture composition. Afterwards, unsupervised clustering is performed to split the image set into several clusters of \\\"similar\\\" images. Every cluster is represented by its centroid as an icon. The set of icons is presented to the user, who can pick one in order to see the images belonging to the cluster. Multi-dimensional scaling is used to visualize images in the same cluster by mapping the images onto a two-dimensional space. The experiments performed with a general-purpose image dataset consisting of one thousand images, categorized into ten classes, show the usefulness of the system.\",\n \"Title: Learning Relations Using Genetic Algorithms\\nYear: 1993\\n\\nAbstract: Inducing concept descriptions from examples requires a large space of hypotheses to be explored. Genetic algorithms offer an appealing alternative to traditional search algorithms, because of their multi-point search strategy. In this paper, the new system REGAL is described: it uses genetic algorithms to learn first order logic concept descriptions. Moreover, it can be easily integrated with a deductive component, in order to exploit a domain theory. Two approaches to learning disjunctive concept descriptions are presented: the first one is a modification of the classical method of learning one disjunct at a time, whereas the second one is based on the idea of fitness sharing and tries to let subpopulations be spontaneously formed, according to the theory of the niches and species. The approaches have been compared on an artificial domain.\",\n \"Title: Learning in the “Real World”\\nYear: 1998\\n\\nAbstract: In this paper we define and characterize the process of developing a \\\"real-world\\\" Machine Learning application, with its difficulties and relevant i ssues, distinguishing it from t he popular practice of exploiting ready-to-use data sets. To this aim, we a nalyze a nd summarize the lessons learned from applying Machine Learning techniques to a variety of problems. We believe that these lessons, though primarily based on our personal experience, can be generalized to a wider r ange of situations and are supported by the reported experiences of other researchers.\",\n \"Title: Learning profiles based on hierarchical hidden markov model\\nYear: 2005\\n\\nAbstract: This paper presents a method for automatically constructing a sophisticated user/process profile from traces of user/process behavior. User profile is encoded by means of a Hierarchical Hidden Markov Model (HHMM). The HHMM is a well formalized tool suitable to model complex patterns in long temporal or spatial sequences. The method described here is based on a recent algorithm, which is able to synthesize the HHMM structure from a set of logs of the user activity. The algorithm follows a bottom-up strategy, in which elementary facts in the sequences (motives) are progressively grouped, thus building the abstraction hierarchy of a HHMM, layer after layer. The method is firstly evaluated on artificial data. Then a user identification task, from real traces, is considered. A preliminary experimentation with several different users produced encouraging results.\",\n \"Title: Machine Learning for Information Extraction\\nYear: 1997\\n\\nAbstract: This paper presents a brief overview of the history and trends of Machine Learning, organized according to its goal and principal methodologies. More details are given for the concept learning task, one of the most mature in the field. Applications to Information Extraction tasks are discussed.\",\n \"Title: Multistrategy Learning and Theory Revision\\nYear: 1993\\n\\nAbstract: This article presents the system WHY, which learns and updates a diagnostic knowledge base using domain knowledge and a set of examples. The a priori knowledge consists of a causal model of the domain that states the relationships among basic phenomena, and a body of phenomenological theory that describes the links between abstract concepts and their possible manifestations in the world. The phenomenological knowledge is used deductively, the causal model is used abductively, and the examples are used inductively. The problems of imperfection and intractability of the theory are handled by allowing the system to make assumptions during its reasoning. In this way, robust knowledge can be learned with limited complexity and a small number of examples. The system works in a first-order logic environment and has been applied in a real domain.\",\n \"Title: Knowledge representation in machine learning\\nYear: 1994\\n\\nAbstract: This paper investigates the influence of knowledge representation languages on the complexity of the learning process. However, the aim of the paper is not to give a state-of-the-art account of the involved issues, but to survey the underlying ideas. Then, references will be provided only occasionally and all the specific quantitative results are left to the presentation. Finally, the paper is intentionally unbalanced, because a larger space is given to those issues that are more novel or less investigated in the literature.\",\n \"Title: A Model of Abstraction in Visual Perception\\nYear: 2001\\n\\nAbstract: In artificial intelligence, abstraction has been mainly studied as a mapping between languages in relation to problem-solving, with the aim of reducing the complexity of the task. However, abstraction has a much larger scope in reasoning; we are investigating, in this article, how abstraction can be used in concept representation. To this aim, we propose a novel, perception-based model of abstraction, which originates from the observation that conceptualization of a domain, even though involving entities belonging to several epistemological levels, is nevertheless primarily based on perception. This view has been recently advocated by Goldstone and Barsalou in cognitive science. A model of representation/abstraction is then proposed and its application to a real-world problem of robot visual perception and categorization is presented.\",\n \"Title: Structured Hidden Markov Model: A General Framework for Modeling Complex Sequences\\nYear: 2007\\n\\nAbstract: Structured Hidden Markov Model (S-HMM) is a variant of Hierarchical Hidden Markov Model that shows interesting capabilities of extracting knowledge from symbolic sequences. In fact, the S-HMM structure provides an abstraction mechanism allowing a high level symbolic description of the knowledge embedded in S-HMM to be easily obtained. The paper provides a theoretical analysis of the complexity of the matching and training algorithms on S-HMMs. More specifically, it is shown that Baum-Welch algorithm benefits from the so called locality property, which allows specific components to be modified and retrained, without doing so for the full model. The problem of modeling duration and of extracting (embedding) readable knowledge from (into) a S-HMM is also discussed.\",\n \"Title: A Comparison between Two Statistical Relational Models\\nYear: 2008\\n\\nAbstract: Statistical Relational Learning has received much attention this last decade. In the ILP community, several models have emerged for modelling and learning uncertain knowledge, expressed in subset of first order logics. Nevertheless, no deep comparisons have been made among them and, given an application, determining which model must be chosen is difficult. In this paper, we compare two of them, namely Markov Logic Networks and Bayesian Programs, especially with respect to their representation ability and inference methods. The comparison shows that the two models are substantially different, from the point of view of the user, so that choosing one really means choosing a different philosophy to look at the problem. In order to make the comparison more concrete, we have used a running example, which shows most of the interesting points of the approaches, yet remaining exactly tractable.\",\n \"Title: Boosting as a Monte Carlo Algorithm\\nYear: 2001\\n\\nAbstract: A new view of majority voting as a Monte Carlo stochastic algorithm is presented in this paper. The relation between the two approches allows Adaboost?s example weighting strategy to be compared with the greedy covering strategy used for a long time in Machine Learning. Even though one may expect that the greedy strategy is very much prone to overfitting, extensive experimental results do not support this guess. The greedy strategy does not clearly show overfitting, it runs in at least one order of magnitude less time, it reaches zero error on the training set in few trials, and the error on the test set is most of the time comparable, if not lower, than that exhibited by Adaboost.\",\n \"Title: An Abstraction-based Machine Learning Approach to Cartographic Generalisation\\nYear: 2005\\n\\nAbstract: This article proposes a machine learning approach to overcome the knowledge acquisition bottleneck limiting the automation of cartographic generalisation. It first explains why this automation must be guided by a differentiation of two main types of knowledge involved in this process. More precisely, it shows that cartographic generalisation is best viewed as a combination of two processes: representing (formulating, renaming knowledge) and abstracting (simplifying a given representation). The whole process of creating maps fits into an abstraction framework developed in Artificial Intelligence to account for the difference between knowledge abstraction and knowledge representation. The utility of this framework lies in its efficiency to support the automation of knowledge acquisition for cartographic generalisation as a combined learning of both abstraction and representation knowledge. The results of Machine Learning experiments show the advantages of this approach.\",\n \"Title: Learning complex event descriptions by abstraction\\nYear: 2005\\n\\nAbstract: The presence of long gaps dramatically increases the difficulty of detecting and characterizing complex events hidden in long sequences. In order to cope with this problem, a learning algorithm based on an abstraction mechanism is proposed: it can infer a Hierarchical Hidden Markov Model, from a learning set of sequences. The induction algorithm proceeds bottom-up, progressively coarsening the sequence granularity, and letting correlations between subsequences, separated by long gaps, naturally emerge. As a case study, the method is evaluated on an application of user profiling. The results show that the proposed algorithm is suitable for developing real applications in network security and monitoring.\",\n \"Title: An expert system oriented to complex pattern recognition problems\\nYear: 1985\\n\\nAbstract: The general framework bimbo, designed for acquiring knowledge on expert systems oriented to pattern analysis, is presented. bimbo, which is itself an expert system, works in a largely automated mode and makes use of a large body of general knowledge and of powerful heuristics. The learning methodology explicitly addresses the problem of the presence of errors and noise in the data and utilizes an approximate reasoning scheme based on fuzzy sets. bimbo has been applied in a test case taken from the field of continuous speech recognition, with very good results in increasing the system accuracy and decreasing the demands upon human time.\",\n \"Title: Incremental construction of structured hidden Markov models\\nYear: 2007\\n\\nAbstract: This paper presents an algorithm for inferring a Structured Hidden Markov Model (S-HMM) from a set of sequences. The S-HMMs are a subclass of the Hierarchical Hidden Markov Models and are well suited to problems of process/user profiling. The learning algorithm is unsupervised, and follows a mixed bottom-up/top-down strategy, in which elementary facts in the sequences (motifs) are progressively grouped, thus building up the abstraction hierarchy of a S-HMM, layer after layer. The algorithm is validated on a suite of artificial datasets, where the challenge for the learning algorithm is to reconstruct the model that generated the data. Then, an application to a real problem of molecular biology is briefly described.\",\n \"Title: Dealing with uncertain knowledge in medical decision-making: A case study in hepatology\\nYear: 1989\\n\\nAbstract: It has widely been recognized that knowledge-based expert systems need efficient mechanisms to model the uncertainty associated with many decision-making activities. Such a need is particularly urgent in medicine. In this paper, we present an approach based on fuzzy logic to give a possible solution to this problem; its pros and cons are discussed by taking into account the experience gained in developing LITO1 and LITO2, two expert systems devoted to the assessment of the liver function and to the diagnosis of hepatic diseases. The advantages of mixing fuzzy production rules with frame-like structures (introduced for representing the clinical data) are discussed. In particular, the use of fuzzy linguistic variables for modeling the possible values of the clinical data is described: this allows, for a clear and perspicuous description of the correspondence between quantitative and qualitative expressions. Furthermore, different alternatives for evaluating and combining evidence are reviewed. Finally, the need of introducing frame structures also for representing diagnostic hypotheses is discussed, together with the problem of evaluating the fuzzy match between the prototypical description of a diagnostic hypothesis and the data describing the status of the specific patient under examination.\",\n \"Title: Explaining Bagging with Monte Carlo Theory\\nYear: 2003\\n\\nAbstract: In this paper we propose the use of the framework of Monte Carlo stochastic algorithms to analyze ensemble learning, specifically, bagging. In particular, this framework allows one to explain bagging's behavior and also why increasing the margin improves performances. Experimental results support the theoretical analysis.\",\n \"Title: Formalizing the abstraction process in model-based diagnosis\\nYear: 2007\\n\\nAbstract: Several theories have been proposed to capture the essence of abstraction. Among these, the KRA model offers a framework where a set of generic abstraction operators allows abstraction to be automated. In this paper we show how to describe component-based abstraction for the Model-Based Diagnosis task within the KRA framework, and we discuss the benefits of such a formalization. The clear and explicit partition of the system model into different levels required by KRA (going from the perception level up to the theory level) opens the way to explore richer and better founded kinds of abstraction to apply to the MBD task. Another noticeable advantage is that, by suitably personalizing the generic abstraction operators of KRA, the whole abstraction process, from the definition of abstract (macro)components to the computation of their behaviors starting from those of the ground components, can be performed automatically in such a way that important relationships between ground and abstract diagnoses are guaranteed.\",\n \"Title: Experimental comparison between bagging and Monte Carlo ensemble classification\\nYear: 2005\\n\\nAbstract: Properties of ensemble classification can be studied using the framework of Monte Carlo stochastic algorithms. Within this framework it is also possible to define a new ensemble classifier, whose accuracy probability distribution can be computed exactly. This paper has two goals: first, an experimental comparison between the theoretical predictions and experimental results; second, a systematic comparison between bagging and Monte Carlo ensemble classification.\",\n \"Title: Abstracting Visual Percepts to Learn Concepts\\nYear: 2002\\n\\nAbstract: To efficiently identify properties from its environment is an essential ability of a mobile robot who needs to interact with humans. Successful approaches to provide robots with such ability are based on ad-hoc perceptual representation provided by AI designers. Instead, our goal is to endow autonomous mobile robots (in our experiments a Pioneer 2DX) with a perceptual system that can efficiently adapt itself to ease the learning task required to anchor symbols. Our approach is in the line of meta-learning algorithms that iteratively change representations so as to discover one that is well fitted for the task. The architecture we propose may be seen as a combination of the two widely used approach in feature selection: the Wrapper-model and the Filter-model. Experiments using the PLIC system to identify the presence of Humans and Fire Extinguishers show the interest of such an approach, which dynamically abstracts a well fitted image description depending on the concept to learn.\",\n \"Title: Improving Learning Using Causality and Abduction\\nYear: 1991\\n\\nAbstract: This paper presents a system which learns and maintains a diagnostic knowledge base using a causal model of the domain, a body of phenomenological theory and a set of examples. The phenomenological knowledge is used deductively, the causal model is used abductively and the examples are used inductively. The problems of imperfection and intractability of the theory are handled by allowing the system to make assumptions during its reasoning. In this way, robust knowledge can be learned with limited complexity and limited number of examples. An example in the domain of heat transfer is presented.\",\n \"Title: A Fine-Grained Parallel Evolutionary Program for Concept Induction\\nYear: 1996\\n\\nAbstract: This paper presents a highly parallel genetic algorithm, called G-NET, designed for concept induction in propositional and first order logics. As well as other systems oriented to the same task, G-NET exploits niches and species for learning multimodal concepts; on the other hand it deeply differs from other systems because of the distributed architecture, which totally eliminates the concept of common memory. A simulator of the system, designed in order to check the possibility of exploiting parallel processing, is evaluated on a standard benchmark. The experimental results show that a multi-processor implemented with standard technology could reach speed-up of thousands of times.\",\n \"Title: Integrating Multiple Learning Strategies in First Order Logics\\nYear: 1997\\n\\nAbstract: This paper describes a representation framework that offers aunifying platform for alternative systems, which learn concepts inFirst Order Logics. The main aspects of this framework arediscussed. First of all, the separation between the hypothesislogical language (a version of the VL21 language) and therepresentation of data by means of a relational database ismotivated. Then, the functional layer between data and hypotheses,which makes the data accessible by the logical level through a set ofabstract properties is described. A novelty, in the hypothesisrepresentation language, is the introduction of the construct ofinternal disjunction; such a construct, first used by the AQ andInduce systems, is here made operational via a set of algorithms,capable to learn it, for both the discrete and the continuous-valuedattributes case. These algorithms are embedded in learning systems(SMART+, REGAL, SNAP, WHY, RTL) using different paradigms (symbolic,genetic or connectionist), thus realizing an effective integrationamong them; in fact, categorical and numerical attributes can behandled in a uniform way. In order to exemplify the effectivenessof the representation framework and of the multistrategy integration,the results obtained by the above systems in some application domainsare summarized.\",\n \"Title: Classification: From Expert To Ensemble To Crowd\\nYear: 2014\\n\\nAbstract: In this paper a model of the classification task, based on the Monte Carlo theory of stochastic algorithms, is presented. The model allows different classification paradigms to be described, such as the single classifier, the ensemble, and the crowdsourcing ones. Some experimental work validates the theoretical predictions about the relatioship between the mean and variance of the classification accuracy and the number of classifiers.\",\n \"Title: Automated concept acquisition in noisy environments\\nYear: 1988\\n\\nAbstract: A system that performs automated concept acquisition from examples and has been specially designed to work in noisy environments is presented. The learning methodology is aimed at the target problem of finding discriminant descriptions of a given set of concepts and uses both examples and counterexamples. The learned knowledge is expressed in the form of production rules, organized into separate clusters, linked together in a graph structure. Knowledge extraction is guided by a top-down control strategy, through a process of specialization. The system also utilizes a technique of problem reduction to contain the computational complexity. Several criteria are proposed for evaluating the acquired knowledge. The methodology has been tested on a problem in the field of speech recognition and the experimental results obtained are reported and discussed.\",\n \"Title: Pattern recognition and Valiant's learning framework\\nYear: 1993\\n\\nAbstract: The computational learning approach shows that the concept descriptions acquired from examples are approximately correct with a degree of probability that grows with the size of the training sample. The same problem has also been widely investigated in the field of pattern recognition under a variety of problem settings. Some of the results obtained in both fields are surveyed and compared, and the limits of their applicability are analyzed. Moreover, new and tighter bounds for the growth function of some classes of Boolean formulas are presented.\",\n \"Title: Fuzzy semantic network for a speech understanding system - an experimental study\\nYear: 1978\\n\\nAbstract: The relevance of the Semantics for a Speech Understanding System (SUS) was pointed out by many authors and is by now a classic topic. The present work contains an implementation of a semantic analysis for the SUS of the CENS, in Turin. The semantic knowledge is contained in a graph on which a set of fuzzy relations is defined. The semantic algorithm accepts, as input, a table of alternative weighted words, found in the processed spectrogram at phonological and lexical level. The output is a sentence assumed as the \\\"best\\\" interpretation of the whole spectrogram. The algorithm is implemented in FORTRAN and PL1 on a IBM 370/158 and it utilizes a recursive evaluation function for the concurrent hypotheses. We compare experimentally the efficiency of the algorithm, by varying the number of alternative input words and their weight, the number of vertices and arcs of the semantic graph and the evaluation function. The results show a good performance of this pure semantic analysis.\",\n \"Title: Exploring the power of genetic search in learning symbolic classifiers\\nYear: 1996\\n\\nAbstract: In this paper we show, in a constructive way, that there are problems for which the use of genetic algorithm based learning systems can be at least as effective as traditional symbolic or connectionist approaches. To this aim, the system REGAL* is briefly described, and its application to two classical benchmarks for Machine Learning is discussed, by comparing the results with the best ones published in the literature.\",\n \"Title: A motion detection system for video surveillance\\nYear: 2013\\n\\nAbstract: In this paper we describe a motion detection module, which is part of Horus, a video surveillance system already deployed in field. The module integrates and improves upon existing algorithms, whose combination proves to be particularly effective in automatically detecting motion sequences in very noisy outdoor environments, with both diurnal and nocturnal illumination conditions. The system has been tested on some public-domain benchmarks, and also on a real application.\",\n \"Title: Piero Torasso'S Early Contributions\\nYear: 2018\\n\\nAbstract: In this paper a brief overview of Pietro Torasso's academic career and early scientific work is provided. Particular attention is given to his interests in Artificial Intelligence topics.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":72,"string":"72"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13570,"string":"13,570"}}},{"rowIdx":13571,"cells":{"question_id":{"kind":"string","value":"Sherry Y Chen"},"question":{"kind":"string","value":"Generate an abstract for the title \"Digital libraries: what do users want?\" using the following items: digital libraries, user suggestions, functionality, features, IT abilities"},"answer":{"kind":"string","value":"Purpose - The purpose of this study is to determine user suggestions for digital libraries, functionality and features. Design/methodology/approach - A survey was conducted as part of this study, in which users' suggestions for digital libraries were solicited, as well as their ranking opinions on a range of suggested digital library features. Findings - The study revealed that, regardless of users' information technology (IT) backgrounds, their expectations of digital libraries' functionality are the same. However, based on users' previous experiences with digital libraries, their requirements with respect to specific features may change. Practical implications - Involving users in digital library design should be an integral step in the process of building a digital library - in addition to the classic roles of evaluation and testing. Originality/value - In previous digital library user studies, users were involved implicitly (e.g. observed) or explicitly (e.g. diary notes). However, they were never asked to suggest digital library features or functionalities, as this was left to usability and domain experts. This study approached digital library design from a new perspective, giving users an opportunity to express their suggestions on future functionality and features of digital libraries. Moreover, in contrast to previous work, this study has explicitly taken into account the IT abilities of those interacting with a digital library."},"correct_answer":{"kind":"string","value":"Purpose - The purpose of this study is to determine user suggestions for digital libraries, functionality and features. Design/methodology/approach - A survey was conducted as part of this study, in which users' suggestions for digital libraries were solicited, as well as their ranking opinions on a range of suggested digital library features. Findings - The study revealed that, regardless of users' information technology (IT) backgrounds, their expectations of digital libraries' functionality are the same. However, based on users' previous experiences with digital libraries, their requirements with respect to specific features may change. Practical implications - Involving users in digital library design should be an integral step in the process of building a digital library - in addition to the classic roles of evaluation and testing. Originality/value - In previous digital library user studies, users were involved implicitly (e.g. observed) or explicitly (e.g. diary notes). However, they were never asked to suggest digital library features or functionalities, as this was left to usability and domain experts. This study approached digital library design from a new perspective, giving users an opportunity to express their suggestions on future functionality and features of digital libraries. Moreover, in contrast to previous work, this study has explicitly taken into account the IT abilities of those interacting with a digital library."},"documents":{"kind":"list like","value":["Title: Navigation in hypermedia learning systems: experts vs. novices\nYear: 2006\n\nAbstract: With the advancement of Web technology, hypermedia learning systems are becoming more widespread in educational settings. Hypermedia learning systems present course content with non-sequential formats, so students are required to develop learning paths by themselves. Yet, empirical evidence indicates that not all students can benefit from hypermedia learning. Research into individual differences suggests that prior knowledge has significant effects on student learning in hypermedia systems, with experts and novices showing different preferences to the use of hypermedia learning systems and requiring different levels of navigation support. It is therefore essential to develop a mechanism to help designers understand the needs of experts and novices. To address this issue, this paper presents a framework to illustrate the needs of students with different levels of prior knowledge by analyzing the findings of previous research. The overall aim of this framework is to integrate students' prior knowledge into the design of hypermedia learning systems. Finally, implications for the design of hypermedia learning systems are discussed.","Title: Effects of gender differences and spatial abilities within a digital pentominoes game\nYear: 2010\n\nAbstract: Spatial ability is a critical skill in geometric learning. Several studies investigate how to use digital games to improve spatial abilities. However, not every learner favors this kind of support. To this end, there is a need to examine how human factors affect learners' reactions to the use of a digital game to support geometric learning. In this vein, this paper addresses this issue by developing a digital pentominoes game and examining the effects of two essential human factors, especially gender differences and spatial abilities, on students' performance. The results demonstrate that students' spatial abilities were significantly improved after they took the digital pentominoes game. The results also demonstrate that the digital game can reasonably reduce the differences between boys and girls. Moreover, the major gender differences lie within mental rotation among the three types of spatial ability and also mainly exist in the low spatial ability group. Finally, the findings are applied to develop a framework that can be used to enhance the understanding of gender differences and spatial abilities within the digital pentominoes game.","Title: Evaluation of a personalized digital library based on cognitive styles: Adaptivity vs. adaptability.\nYear: 2009\n\nAbstract: Personalization can be addressed by adaptability and adaptivity, which have different advantages and disadvantages. This study investigates how digital library (DL) users react to these two techniques. More specifically, we develop a personalized DL to suit the needs of different cognitive styles based on the findings of our previous work [Frias-Martinez, E., Chen, S. Y., & Liu, X. (2008) Investigation of behavior and perception of digital library users: A cognitive style perspective. International Journal of Information Management]. The personalized DL includes two versions: adaptive version and adaptable version. The results showed that users not only performed better in the adaptive version, but also they perceived more positively to the adaptive version. In addition, cognitive styles have great effects on users' responses to adaptability and adaptivity. These results provide guidance for designers to select suitable techniques to develop personalized DLs.","Title: Mining students' behavior in web-based learning programs\nYear: 2009\n\nAbstract: There has been a proliferation of web-based learning programs (WBLPs). Unlike traditional computer-based learning programs, WBLPs are used by a population of learners who have diverse background. How different learners access the WBLPs has been investigated by several studies, which indicate that cognitive style is an important factor that influences learners' preferences. However, these studies mainly use statistical methods to analyze learners' preferences. In this paper, we propose to analyze learners' preferences with a data mining technique. Findings in our study show that Field Independent learners frequently use backward/forward buttons and spent less time for navigation. On the other hand, Field Dependent learners often use main menu and have more repeated visiting. Implications for these findings are discussed.","Title: User perceptions of online public library catalogues.\nYear: 2008\n\nAbstract: Online Public Library Catalogues (OPACs) are widely used electronic library catalogues giving a wealth of remote access to library information resources. Users should be involved early in the OPAC development cycle process in order to ensure a usable and functional interface, as the integration of user-defined requirements of OPACs, along with the other human-computer interaction considerations, offer a better understanding of user perceptions and expectations in respect of OPACs, ultimately resulting in truly user-centred OPACs. Accordingly, the purpose of this study was to establish user suggestions for a typical OPAC application's functionality and features. To this end, an experiment was undertaken to find out the type of interaction features that users prefer to have in an OPAC. The study revealed that regardless of users' Information Technology (IT) backgrounds, their functionality expectations of OPACs are the same as users are expecting OPACs to facilitate easier ways to achieve their tasks. However, based on users' previous experiences with OPACs, their requirements with respect to specific features may change.","Title: Matching and Mismatching in Web-based Learning: A Human Centered Apprach\nYear: 2011\n\nAbstract: This study examines the effects of matching and mismatching on web-based learning with a human-centered approach. In particular, we emphasize on how prior knowledge and gender differences affect students' reactions to a matching/mismatching environment. In general, the results demonstrate that learners in a matching condition perform better than those in a mismatching condition. Furthermore, males can overcome the mismatching condition when they acquire declarative knowledge. In contrast, females can positively face the mismatching condition when they obtain procedural knowledge. In summary, the matching/mismatching is not only related to gender differences but also is linked to declarative/procedural knowledge.","Title: Perceived quality of multimedia educational content: A cognitive style approach\nYear: 2006\n\nAbstract: Cognitive styles influence the way how humans process information, with previous research demonstrating that they have significant effects on student learning in multimedia environments. On the other hand, the perceptual quality of the human multimedia experience is notoriously difficult to measure. In this paper, we report the results of an empirical study, which investigated the relationship between user cognitive styles and perceptual multimedia quality, in which users had the possibility to specify their desired Quality of Service settings -- in terms of frame rates and color depth. Results show that whilst color choice is impacted by a participant's cognitive style, such Quality of Service parameters do not significantly affect perceived multimedia quality, and that users do not necessarily choose optimum presentation settings to enhance their perceived enjoyment and assimilation of multimedia informational content.","Title: Hypermedia learning and prior knowledge: domain expertise vs. system expertise\nYear: 2005\n\nAbstract: Prior knowledge is often argued to be an important determinant in hypermedia learning, and may be thought of as including two important elements: domain expertise and system expertise. However, there has been a lack of research considering these issues together. In an attempt to address this shortcoming, this paper presents a study that examines how domain expertise and system expertise influence students' learning performance in, and perceptions of, a hypermedia system. The results indicate that participants with lower domain knowledge show a greater improvement in their learning performance than those with higher domain knowledge. Furthermore, those who enjoy using the Web more are likely to have positive perceptions of non-linear interaction. Discussions on how to accommodate the different needs of students with varying levels of prior knowledge are provided based on the results.","Title: Mining Students' Learning Patterns And Performance In Web-Based Instruction: A Cognitive Style Approach\nYear: 2011\n\nAbstract: Personalization has been widely used in Web-based instruction (WBI). To deliver effective personalization, there is a need to understand different preferences of each student. Cognitive style has been identified as one of the most pertinent factors that affect students' learning preferences. Therefore, it is essential to investigate how learners with different cognitive styles interact with WBI programs. This paper presents an empirical study, which examines the effects of cognitive styles on students' learning patterns and the effects of learning patterns on their learning performances. Riding's cognitive style analysis was used to identify the students' cognitive styles. Data mining, especially a clustering technique, was used to analyze the results. It was found that field independent students frequently used an alphabetical index whereas field dependent students often chose a hierarchical map. Such learning patterns also have great effects on their performance, especially for field dependent students.","Title: Measuring quality of perception in distributed multimedia: Verbalizers vs. imagers\nYear: 2008\n\nAbstract: This paper presents the results of a study which investigated the impact of cognitive styles on perceptual multimedia quality. More specifically, we examine the different preferences demonstrated by verbalizers and imagers when viewing multimedia content presented with different quality of service (QoS) levels pertaining to frame rates and color depth. Recognizing multimedia's infotainment duality, we used the quality of perception (QoP) metric to characterize perceived quality. Results showed that in terms of low and high dynamisms clips, the frame rate at which multimedia content is displayed influences the levels of information assimilated by Imagers. Whilst black and white presentations are shown to be beneficial for both Biomodals and Imagers in order to experience enhanced levels of information assimilation, Imagers were shown to enjoy presentations in full 24-bit colour.","Title: Mining Learning Preferences in Web-based Instruction: Holists vs. Serialists.\nYear: 2011\n\nAbstract: Web-based instruction programs are used by learners with diverse knowledge, skills and needs. These differences determine their preferences for the design of Web-based instruction programs and ultimately influence learners' success in using them. Cognitive style has been found to significantly affect learners' preferences of web-based instruction programs. However, the majority of previous studies focus on Field Dependence/Independence. Pask's Holist/Serialist dimension has conceptual links with Field Dependence/Independence but it is left mostly unstudied. Therefore, this study focuses on identifying how this dimension of cognitive style affects learner preferences of Web-based instruction programs. A data mining approach is used to illustrate the difference in preferences between Holists and Serialists. The findings show that there are clear differences in regard to content presentation and navigation support. A set of design features were then produced to help designers incorporate cognitive styles into the development of Web-based instruction programs to ensure that they can accommodate learners' different preferences.","Title: An Integrated Approach for Modeling Learning Patterns of Students in Web-Based Instruction: A Cognitive Style Perspective\nYear: 2008\n\nAbstract: Web-based instruction (WBI) programs, which have been increasingly developed in educational settings, are used by diverse learners. Therefore, individual differences are key factors for the development of WBI programs. Among various dimensions of individual differences, the study presented in this article focuses on cognitive styles. More specifically, this study investigates how cognitive styles affect students' learning patterns in a WBI program with an integrated approach, utilizing both traditional statistical and data-mining techniques. The former are applied to determine whether cognitive styles significantly affected students' learning patterns. The latter use clustering and classification methods. In terms of clustering, the K-means algorithm has been employed to produce groups of students that share similar learning patterns, and subsequently the corresponding cognitive style for each group is identified. As far as classification is concerned, the students' learning patterns are analyzed using a decision tree with which eight rules are produced for the automatic identification of students' cognitive styles based on their learning patterns. The results from these techniques appear to be consistent and the overall findings suggest that cognitive styles have important effects on students' learning patterns within WBI. The findings are applied to develop a model that can support the development of WBI programs.","Title: Web-based interaction: A review of three important human factors.\nYear: 2010\n\nAbstract: With the rapid development of information technology, the World Wide Web has been widely used in various applications, such as search engines, online learning and electronic commerce. These applications are used by a diverse population of users with heterogeneous backgrounds, in terms of their knowledge, skills, and needs. Therefore, human factors are key issues for the development of Web-based applications, leading research into human factors to grow significantly in the past decade. This paper identifies and reviews three important human factors that have been examined in existing empirical studies, including gender differences, prior knowledge, and cognitive styles. The main results from the analysis include that: (a) females have more disorientation problems than males; (b) flexible paths are more beneficial to experts while structured content is more useful to novices; and (c) Field Dependent and Field Independent users prefer to employ different search strategies. In addition to reviewing the existing empirical studies, this paper also highlights areas of future research.","Title: The Influences of Number and Nature of Classifiers on Consensus Feature Selection\nYear: 2008\n\nAbstract: Abstract - Wrapper feature selection approaches are widely used to select a small subset of relevant features from a dataset. However, Wrappers suffer from the fact that they only use a single classifier when selecting the features. The downside to this approach is that each classifier will have its own biases and will therefore select very different features. In order to overcome the biases of individual classifiers, we propose Consensus Feature Selection (CFS), which combines different classifiers for feature selection. In this way, selecting classifiers for use in the combinations is very important. Therefore, we investigate how the number and nature of classifiers influence the number of features selected and the classification accuracies that these features generate. In terms of number of classifiers, results showed that few selected more relevant features whereas many selected few features. In addition, 3-classifier combinations selected features that led to highest accuracies. In terms of nature of classifiers, decision trees identified most number of features whereas Bayesian classifiers identified least number of features. However, features selected by Bayesian classifiers led to accuracies higher than the other classifiers","Title: Discussion Models in Seamless Online Learning Integrated Discussion (SOLID) Environment: Selection vs. Performance\nYear: 2012\n\nAbstract: Synchronous discussion and asynchronous discussion have different advantages and disadvantages. However, it is unclear how students react to these two models. Thus, we developed a learning environment that combines these two models, and then examined how students selected them. The findings show that students selected different models at different discussion stages. Additionally, the mixed discussion model, which combines synchronous and asynchronous models, can achieve better performance in discussion, regardless of the types of discussion activity.","Title: Associating learners' cognitive style with their navigation behaviors: a data-mining approach\nYear: 2011\n\nAbstract: Investigating how learners' individual differences affect their navigation behavior can help us understand learners' preferences and can be used to develop the Web-based learning system that can meet learners' needs. Among various individual differences, learners' cognitive styles need to be considered because it is concerned how users process information. There is also a need to examine the relationship between their navigation behavior and performance because such findings can be used to suggest suitable navigation tools to meet learners' real needs. To this end, the aims of this study tend to use a data mining approach to not only investigate the relationship between learners' cognitive styles and their navigation behavior but also analyze how their navigation behavior influence performance. The results indicate that holists and serialists have different navigation behavior. However, there is no direct relationship between learning performance and navigation behavior.","Title: Automatic cognitive style identification of digital library users for personalization\nYear: 2007\n\nAbstract: Digital libraries have become one of the most important Web services for information seeking. One of their main drawbacks is their global approach: In general, there is just one interface for all users. One of the key elements in improving user satisfaction in digital libraries is personalization. When considering personalizing factors, cognitive styles have been proved to be one of the relevant parameters that affect information seeking. This justifies the introduction of cognitive style as one of the parameters of a Web personalized service. Nevertheless, this approach has one major drawback: Each user has to run a time-consuming test that determines his or her cognitive style. In this article, we present a study of how different classification systems can be used to automatically identify the cognitive style of a user using the set of interactions with a digital library. These classification systems can be used to automatically personalize, from a cognitive-style point of view, the interaction of the digital library and each of its users. © 2007 Wiley Periodicals, Inc.","Title: Investigation Of The Use Of Navigation Tools In Web-Based Learning: A Data Mining Approach\nYear: 2008\n\nAbstract: Web-based learning is widespread in educational settings. The popularity of Web-based learning is in great measure because of its flexibility. Multiple navigation tools provided some of this flexibility. Different navigation tools offer different functions. Therefore, it is important to understand how the navigation tools are used by learners with different backgrounds, knowledge, and skills. This article presents two empirical studies in which data-mining approaches were used to analyze learners' navigation behavior. The results indicate that prior knowledge and subject content are two potential factors influencing the use of navigation tools. In addition, the lack of appropriate use of navigation tools may adversely influence learning performance. The results have been integrated into a model that can help designers develop Web-based learning programs and other Web-based applications that can be tailored to learners' needs.","Title: Cognitive styles and hypermedia navigation: development of a learning model\nYear: 2002\n\nAbstract: There has been an increased growth in the use of hypermedia to deliver learning and teaching material. However, much remains to be learned about how different learners perceive such systems. Therefore, it is essential to build robust learning models to illustrate how hypermedia features are experienced by different learners. Research into individual differences suggests cognitive styles have a significant effect on student learning in hypermedia systems. In particular, Witkin's Field Dependence has been extensively examined in previous studies. This article reviews the published findings from empirical studies of hypermedia learning. Specifically, the review classifies the research into five themes: nonlinear learning, learner control, navigation in hyperspace, matching and mismatching, and learning effectiveness. A learning model, developed from an analysis of findings of the previous studies, is presented. Finally, implications for the design of hypermedia learning systems are discussed.","Title: Cognitive styles and users' responses to structured information representation\nYear: 2004\n\nAbstract: Structured information repositories, such as digital libraries, Web directories, and subject gateways, require effective ways to organise and manage information. This paper focuses on Web directories and investigates the relationships between users' cognitive styles and information representation. The results indicate that cognitive style influences participants' reactions to the organisation of subject categories, presentation of the results, and screen layout. The findings are applied to develop a design framework that can support the improvement of Web directories and other structured digital information resources. Finally, implications for information representation are discussed.","Title: Recent Soft Computing Approaches to User Modeling in Adaptive Hypermedia\nYear: 2004\n\nAbstract: The ability of an adaptive hypermedia system to create tailored environments depends mainly on the amount and accuracy of information stored in each user model. One of the difficulties that user modeling faces is the necessity of capturing the imprecise nature of human behavior. Soft Computing has the ability to handle and process uncertainty which makes it possible to model and simulate human decision-making. This paper surveys different soft computing techniques that can be used to efficiently and accurately capture user behavior. The paper also presents guidelines that show which techniques should be used according to the task implemented by the application.","Title: Mammogram retrieval on similar mass lesions.\nYear: 2012\n\nAbstract: Enormous numbers of digital mammograms have been produced in hospitals and breast screening centers. To exploit those valuable resources in aiding diagnoses and research, content-based mammogram retrieval systems are required to effectively access the mammogram databases. This paper presents a content-based mammogram retrieval system, which allows medical professionals to seek mass lesions that are pathologically similar to a given example. In this retrieval system, shape and margin features of mass lesions are extracted to represent the characteristics of mammographic lesions. To compare the similarity between the query example and any lesion within the databases, this study proposes a similarity measure scheme which involves the hierarchical arrangement of mammographic features and a weighting distance measure. This makes similarity measure of the retrieval system consistent with the way radiologists observe mass lesions. This study used the DDSM dataset to evaluate the effectiveness of the extracted shape feature and margin feature, respectively. Experimental results demonstrate that, when Zernike moments are used, round-shape masses are the most discriminative among four types of shape; the circumscribed-margin masses can be effectively discriminated among the four types of margins. Moreover, the result also shows that, when retrieving round-shape and circumscribed margin masses, this retrieval system can achieve the highest precision among all mass lesion types.","Title: Web-based Assessment and Test Analyses (WATA) system: development and evaluation\nYear: 2004\n\nAbstract: This study introduces the development of a Web-based assessment system, the Web-based Assessment and Test Analyses (WATA) system, and examines its impacts on teacher education. The WATA system is a follow-on system, which applies the Triple-A Model (assembling, administering, and appraising). Its functions include (1) an engine for teachers to administer and manage testing, (2) an engine for students to apply tests, and (3) an engine for generating test results and analyses for teachers. Two studies were undertaken to assess the usefulness and potential benefits of the WATA system for teacher education. In the first study, 47 in-service teachers were asked to assess the functions of the WATA system. The results indicated that they were satisfied with the Triple-A Model of the WATA system. In the second study, 30 pre-service teachers were required to use the WATA system during the teacher-training program. After 4 months of experience in using the WATA system, the pre-service teachers' perspectives of assessment have been changed significantly. The findings of these two studies might provide some guidance to help those who are interested in the development of Web-based assessment and intend to infuse information technology into teacher education.","Title: Satellite-based delivery of educational content to geographically isolated communities: a service based approach\nYear: 2009\n\nAbstract: Enabling learning for members of geographically isolated communities presents benefits in terms of promoting regional development and cost savings for governments and companies. However, notwithstanding recent advances in e-Learning, from both technological and pedagogical perspectives, there are very few, if any, recognised methodologies for user-led design of satellite-based e-learning infrastructures. In this paper, we present a methodology for designing a satellite and wireless based network infrastructure and learning services to support distance learning for such isolated communities. This methodology entails (a) the involvement of community members in the development of targeted learning services from an early stage, and (b) a service-oriented approach to learning solution deployment. Results show, that, while the technological premises of distance learning can be accommodated by hybrid satellite/wireless infrastructures, this has to be complemented with (a) high-quality audio---visual educational material, and (b) the opportunity for community members to interact with other community members either as groups (common-room oriented scenarios) or individuals (home-based scenarios), thus providing an impetus for learner engagement in both formal and informal activities.","Title: Evaluating The Learning Effectiveness Of Using Web-Based Instruction: An Individual Differences Approach\nYear: 2005\n\nAbstract: The use of Web-based instruction is becoming widespread in higher education; however, much remains to be learned about how different learners react to such instructional programs. The study presented in this article evaluates students' learning performance and their perceptions in a Web-based instructional program, which was applied to teach students how to use HTML in Brunel University's Department of Information Systems and Computing. Sixty-one master's students participated in this study. A number of interesting interactions were found. Students' task achievements were affected by the levels of their previous system experience. On the other hand, the post-test and gain scores were positively influenced by their perceptions and attitudes toward the Web-based instructional program. The implications of these findings are discussed.","Title: The effects of a Creative Commons approach on collaborative learning.\nYear: 2013\n\nAbstract: Social media on the World Wide Web, such as Wiki, are increasingly applied to support collaborative learning for students to conduct a project together. However, recent studies indicated that students, learning in the collaborative project, may not actively contribute to the collaborative work and are involved only in a limited level of positive interdependence due to the ambiguous ownership on the collaborative work. To this end, this study proposes an approach to support collaborative learning based on the notion of Creative Commons (CC). CC may be helpful to enhance positive interdependency because it provides a mechanism which protects individual creations while encouraging remixing and deriving new creations from them. This study conducted an empirical evaluation to examine how students react to the collaborative learning with CC notions. The results showed that CC can significantly improve participants' attitude to the derivative works, the satisfaction level of remix outcomes, perception to the peer interaction and the sense of work ownership. Such results suggest that CC may be a potential avenue for increasing individuals' responsibilities and motivating them to participate in the collaborative learning activity.","Title: The contribution of data mining to information science\nYear: 2004\n\nAbstract: The information explosion is a serious challenge for current information institutions. On the other hand, data mining, which is the search for valuable information in large volumes of data, is one of the solutions to face this challenge. In the past several years, data mining has made a significant contribution to the field of information science. This paper examines the impact of data mining by reviewing existing applications, including personalized environments, electronic commerce, and search engines. For these three types of application, how data mining can enhance their functions is discussed. The reader of this paper is expected to get an overview of the state of the art research associated with these applications. Furthermore, we identify the limitations of current work and raise several directions for future research.","Title: The assessment of usability of electronic shopping: A heuristic evaluation.\nYear: 2005\n\nAbstract: Today there are thousands of electronic shops accessible via the Web. Some provide user-friendly features whilst others seem not to consider usability factors at all. Yet, it is critical that the electronic shopping interface is user-friendly so as to help users to obtain their desired results. This study applied heuristic evaluation to examine the usability of current electronic shopping. In particular, it focused on four UK-based supermarkets offering electronic services: including ASDA, Iceland, Sainsbury, and Tesco. The evaluation consists of two stages: a free-flow inspection and a task-based inspection. The results indicate that the most significant and common usability problems have been found to lie within the areas of 'User Control and Freedom' and 'Help and Documentation'. The findings of this study are applied to develop a set of usability guidelines to support the future design of effective interfaces for electronic shopping.","Title: Electronic trombone: an interactive tool to promote musical learning and performance creativity\nYear: 2011\n\nAbstract: A lot of musical toys (e.g., toy guitars, toy drums and percussion) can provide a good introduction to musical instruments and give children high entertainment. Via playing an instrument toy, children can sense and practice what kinds of physical movements (e.g., fingers and/or mouths) they need to play its corresponding actual instrument to produce sounds. In this paper, we present an electronic trombone with three different playing modes for children. The research mission is to offer a simple and joyful approach to engaging children in learning to play a trombone.","Title: Investigation Of Learners' Perceptions For Video Summarization And Recommendation\nYear: 2012\n\nAbstract: Recently, multimedia-based learning is widespread in educational settings. A number of studies investigate how to develop effective techniques to manage a huge volume of video sources, such as summarization and recommendation. However, few studies examine how these techniques affect learners' perceptions in multimedia learning systems. This article aims to examine learners' perceptions for summarization and recommendation, with an emphasis on the perspective of prior experience. In this study, we developed a multimedia content summarization and recommendation system, which can automatically extract summaries from raw video sources and recommend suitable video content to learners through emails. The results demonstrate that learners' prior experience and preferences for the presentation of document types affect their perceptions, including the enhancement of interests, the ease of information acquisition and the intention for the further use of the system. Finally, the findings are applied to develop a framework that can support for the design of multimedia learning systems.","Title: Investigation of behavior and perception of digital library users: A cognitive style perspective.\nYear: 2008\n\nAbstract: Cognitive style is an influential factor in users' information seeking. The study presented in this paper examines how users' cognitive styles affect their behavior and perception in digital libraries. Fifty participants took part in this study. Two dimensions of cognitive styles were considered: (a) Field Dependence/Independence; (2) Verbalizer/Imager. The results showed that Intermediate users and Verbalizers have not only more positive perception, but they also complete the tasks in effective ways. Implications for the design of personalized digital libraries are also discussed.","Title: The comparisons of the influences of prior knowledge on two game-based learning systems\nYear: 2013\n\nAbstract: Game-based learning provides many benefits, such as enhancing the interaction with students and stimulating their learning motivation. Thus, it is popular to learners who have diverse characteristics. To this end, individual differences play an essential role. Among various individual differences, previous studies demonstrated that prior knowledge has great effects on game-based learning. However, such studies mainly considered a single game-based learning system. To address this issue, this study examined how prior knowledge affects students' reactions to two different types of game based learning systems, i.e., the Machinarium and the CSI: Web Adventures. The former delivers procedural knowledge while the latter provides declarative knowledge. The results from this study indicate that prior knowledge has positive impacts in the CSI group while it has negative impacts in the Machinarium group. These findings imply that prior knowledge is useful for the context of declarative knowledge while it may not be helpful for the context of procedural knowledge.","Title: A flexible interface design for Web directories to accommodate different cognitive styles\nYear: 2005\n\nAbstract: Search engines are very popular tools for collecting information from distributed resources. They provide not only search facilities, but they also offer directories for users to browse content divided into groups. In this paper, we've adopted an individual differences approach to explore user's attitudes towards various interface features provided by existing Web directories. Among a variety of individual differences, cognitive style is a particularly important characteristic that influences the effectiveness of information seeking. Empirical results indicate that users' cognitive styles influence their reactions to the organization of subject categories, presentation of the results, and screen layout. We developed a set of design guidelines on the basis of these results, and propose a flexible interface that adopts these guidelines to accommodate the preferences of different cognitive style groups.","Title: Mining User Preferences Of Multimedia Interfaces With K-Modes\nYear: 2006\n\nAbstract: Interactive multimedia learning systems use sophisticated techniques to present advanced interface features. However, not all users appreciate the strengths of such interface features because of the variations of user backgrounds and skills. In this context, human factors are important issues in deciding user preferences. This study applies a data mining approach to examine user preferences of interface features and to identify the influence of human factors on this issue. K-modes, a data mining technique extensively applied to user modeling, was used to group users' preferences. The results indicated that users' preferences could be divided into eight groups where gender and computer experience significantly influenced the choices made by users.","Title: Grouping Users' Communities in an Interactive Web-Based Learning System: A Data Mining Approach\nYear: 2005\n\nAbstract: Discovery of user communities can help designers to develop learning systems that are more suitable to the needs of different individuals. The study builds user communities based on their common navigation behavior using the K-means algorithm. The results indicated that three user groups could be identified and cognitive style is a potential factor that may influence the patterns of each group.","Title: Data mining from 1994 to 2004: an application-orientated review\nYear: 2005\n\nAbstract: Data mining, which is also known as knowledge discovery, is one of the most popular topics in information technology. It concerns the process of automatically extracting useful information and has the promise of discovering hidden relationships that exist in large databases. These relationships represent valuable knowledge that is crucial for many applications. This paper presents a review of works on current applications of data mining, which focus on four main application areas, including bioinformatics data, information retrieval, adaptive hypermedia and electronic commerce. How data mining can enhance functions for these four areas is described. The reader of this paper is expected to get an overview of the state-of-the-art research associated with these applications. Furthermore, we identify the limitations of current works and raise several directions for future research.","Title: The Clusters of Gaming Behavior in MMORPG: A Case Study in Taiwan\nYear: 2012\n\nAbstract: The aim of this study is to investigate the relationship between online gaming behavior and special underlying social group structures through the use of player information of a certain MMORPG. Data were collected from the logs of an MMORPG in 2011, recording five types of gaming behaviors, player gender, avatar gender and avatar occupation types to investigate whether different avatars affect gaming behavior. This study includes 1,727 male players and 656 female players. The results indicate that the six-cluster model has the goodness of model fit. Different clusters differ in the five different types of gaming behaviors and the choice of avatar gender and occupations. This result shows that different avatar roles will influence players' gaming behaviors.","Title: A Personalised Interface for Web Directories Based on Cognitive Styles\nYear: 2004\n\nAbstract: Implementing personalisation in Web Directories depends not only on developing appropriate architectures and equipping Web Directories with adaptation techniques, but also on incorporating human factors considerations at an early design stage. Among a range of human factors this paper explores cognitive styles and their influence on users' preferences. Preferences with respect to the organisation and presentation of the content, and the navigation paths are identified through a small-scale study. The findings are analysed and used to implement a prototype Web Directory Browser, gearing interface features to cognitive style-related preferences.","Title: Mining learners' behavior in accessing web-based interface\nYear: 2007\n\nAbstract: Web-based technology has already been adopted as a tool to support teaching and learning in higher education. One criterion affecting the usability of such a technology is the design of web-based interface (WBI) within web-based learning programs. How different users access the WBIs has been investigated by several studies, which mainly analyze the collected data using statistical methods. In this paper, we propose to analyze users' learning behavior using Data Mining (DM) techniques. Findings in our study show that learners with different cognitive styles seem to have various learning preferences, and DM is an efficient tool to analyze the behavior of different cognitive style groups.","Title: Evaluating Webct Use In Relation To Students' Attitude And Performance\nYear: 2008\n\nAbstract: Teaching and learning are no longer limited by place or time. Large numbers of educational institutions are now using course management systems such as WebCT. This study examined students' attitude and achievement in relation to their use of WebCT. Statistical data was gathered weekly from the WebCT tracking system log files. Students' achievement was measured by the grades they obtained in the observed module. Students' attitudes towards WebCT were measured using a Likert-type attitude scale. The results of this study indicated that students have a positive attitude towards WebCT. The module leader valued WebCT as a supporting tool to traditional face-to-face learning but not as an effective tool for helping students' learning. The use of WebCT was found to have positive influence on students' achievement as statistically significant relationships were found between students' use of WebCT and their grades.","Title: Knowledge exploration with concept association techniques\nYear: 2010\n\nAbstract: Purpose - Exploratory learning is regarded as an important ability for developing knowledge from open environments. During the exploration, learners not only need to acquire new information based on their current interests, but also they need to form new perspectives by incorporating new knowledge into their previous knowledge. This paper seeks to address these issues. Design/methodology/approach - To this end, this paper proposes an approach that includes a concept association bank to recommend related concepts in a domain based on the goal of an exploration. By doing so, learners' knowledge can be expanded beyond their current understanding. An experiment was conducted to investigate how the proposed approach facilitated the learners' exploration. Findings - The results indicated that the concept association bank is a useful mechanism to help learners gain new understanding, including providing exploration directions, reducing complexity and cognitive load, facilitating data- and goal-driven exploration strategies, and commenting on new understanding. The implications of these results are discussed. Originality/value - Current recommendation systems emphasise a data-driven strategy, which seeks isolated pieces of information, instead of suggesting directions related to their exploration goal. The problem with such an approach is that learners' exploration will be limited by their existing knowledge. Thus, this paper presents an approach to support both data- and goal-driven strategies.","Title: A cognitive approach to user perception of multimedia quality: An empirical investigation\nYear: 2006\n\nAbstract: Whilst multimedia technology has been one of the main contributing factors behind the Web's success, delivery of personalized multimedia content has been a desire seldom achieved in practice. Moreover, the perspective adopted is rarely viewed from a cognitive styles standpoint, notwithstanding the fact that they have significant effects on users' preferences with respect to the presentation of multimedia content. Indeed, research has thus far neglected to examine the effect of cognitive styles on users' subjective perceptions of multimedia quality. This paper aims to examine the relationships between users' cognitive styles, the multimedia quality of service delivered by the underlying network, and users' quality of perception (understood as both enjoyment and informational assimilation) associated with the viewed multimedia content. Results from the empirical study reported here show that all users, regardless of cognitive style, have higher levels of understanding of informational content in multimedia video clips (represented in our study by excerpts from television programmes) with weak dynamism, but that they enjoy moderately dynamic clips most. Additionally, multimedia content was found to significantly influence users' levels of understanding and enjoyment. Surprisingly, our study highlighted the fact that Bimodal users prefer to draw on visual sources for informational purposes, and that the presence of text in multimedia clips has a detrimental effect on the knowledge acquisition of all three cognitive style groups.","Title: A Probabilistic SVM Approach to Annotation of Calcification Mammograms\nYear: 2010\n\nAbstract: Due to the increasing use of digital medical images, a need exists to develop an approach for automatic image annotation, which provides textual labels for images. Thus added labels can be used to access images using textual queries. Automatic image annotation can be separated into two individual tasks: feature extraction and image classification. In this paper, the authors present feature extraction methods for calcification mammograms. The resultant features, based on BI-RADS standards, make annotated image contents represent the correct medical meaning and tag correspondent terms. Furthermore, this paper also proposes a probabilistic SVM approach to image classification. Finally, the experimental results indicate that the probabilistic SVM approach to image annotation can achieve 79.5% in the average accuracy rate.","Title: A surrogate competition approach to enhancing game-based learning\nYear: 2013\n\nAbstract: Competition is useful in game-based learning, although it can also generate negative influences. To expand the potential for competition models in game-based learning, this study proposes the notion of surrogate competition, which eliminates direct competition between students. Such surrogates could be employed as buffers so that the competition between students is more relaxed. To explore the possible benefits of a surrogate approach to competition, the My-Pet-My-Arena system has been developed and evaluated. Two empirical studies were conducted to examine the effects of the surrogate competition. The results revealed that surrogate competition enhanced students’ learning achievement as well as increased their motivation. Furthermore, the surrogate competition might also assist students in attributing competitive failures to a lack of effort. Working from the results obtained in these two studies, a general model of surrogate competition is proposed to help designers implement forms of surrogate competition in other systems for game-based learning.","Title: Cognitive styles and search engine preferences: Field dependence/independence vs holism/serialism\nYear: 2010\n\nAbstract: Purpose - Cognitive style has been identified to be significantly influential in deciding users' preferences of search engines. In particular, Witkin's field dependence/independence has been widely studied in the area of web searching. It has been suggested that this cognitive style has conceptual links with the holism/serialism. This study aims to investigate the differences between the field dependence/independence and holism/serialism. Design/methodology/approach - An empirical study was conducted with 120 students from a UK university. Riding's cognitive style analysis (CSA) and Ford's study preference questionnaire (SPQ) were used to identify the students' cognitive styles. A questionnaire was designed to identify users' preferences for the design of search engines. Data mining techniques were applied to analyse the data obtained from the empirical study. Findings - The results highlight three findings. First, a fundamental link is confirmed between the two cognitive styles. Second, the relationship between field dependent users and holists is suggested to be more prominent than that of field independent users and serialists. Third, the interface design preferences of field dependent and field independent users can be split more clearly than those of holists and serialists. Originality/value - The contributions of this study include a deeper understanding of the similarities and differences between field dependence/independence and holists/serialists as well as proposing a novel methodology for data analyses.","Title: Cognitive styles and web-based instruction: field dependent/independent vs. Holist/Serialist\nYear: 2009\n\nAbstract: Evolutionary multiobjective optimization (EMO) is an active research area in the field of evolutionary computation. EMO algorithms are designed to find a non-dominated solution set that approximates the entire Pareto front of a multiobjective optimization ...","Title: Incorporating Usability Criteria into the Development of Animated Hierarchical Maps.\nYear: 2013\n\nAbstract: Nowadays, Web-based learning systems have become popular because they can provide multiple tools, among which hierarchical maps are widely used to support teaching and learning. However, traditional hierarchical maps may let learners easily get lost within large information space. This study proposes an animated hierarchical map to address this problem. To consider the effects of different instructional approaches, we also conducted an empirical study to examine how learners in collaborative learning and those in individual learning reacted differently to the animated hierarchical map and traditional hierarchical map. A questionnaire, which was designed based on Nielsen's heuristics, was applied to examine learners' reactions to the proposed animated hierarchical map. The results show that the proposed animated hierarchical map is generally better than the traditional hierarchical map but the effects on individual learning is limited.","Title: When educational agents meet surrogate competition: Impacts of competitive educational agents on students' motivation and performance.\nYear: 2014\n\nAbstract: The design of educational agents increasingly attracts researchers' attention recently. One of major reasons is that educational agents could enhance student learning in various aspects. Although research in this area has mushroomed, such research mainly emphasizes on students in higher education. It is still unclear how educational agents influence young student learning. In addition, competition is a significant element, but fewer studies take competition into account while designing educational agents. Although some studies have indicated that educational agents in competitive environments has positive effects on students' perception and attribution, its impacts on students' motivation and performance are unclear. Thus, the study develops an integrative agent that combines educational and competitive elements for young students, and further examines its influences, in terms of motivation and learning performance. The results revealed that such competitive educational agents could enhance students' motivation and learning performance.","Title: Design of adaptive hypermedia learning systems: A cognitive style approach\nYear: 2011\n\nAbstract: In the past decade, a number of adaptive hypermedia learning systems have been developed. However, most of these systems tailor presentation content and navigational support solely according to students' prior knowledge. On the other hand, previous research suggested that cognitive styles significantly affect student learning because they refer to how learners process and organize information. To this end, the study presented in this paper developed an adaptive hypermedia learning system tailored to students' cognitive styles, with an emphasis on Pask's Holist-Serialist dimension. How students react to this adaptive hypermedia learning system, including both learning performance and perceptions, was examined in this study. Forty-four undergraduate and postgraduate students participated in the study. The findings indicated that, in general, adapting to cognitive styles improves student learning. The results also showed that the adaptive hypermedia learning system have more effects on students' perceptions than performance. The implications of these results for the design of adaptive hypermedia learning systems are discussed.","Title: The role of human factors in stereotyping behavior and perception of digital library users: a robust clustering approach\nYear: 2007\n\nAbstract: To deliver effective personalization for digital library users, it is necessary to identify which human factors are most relevant in determining the behavior and perception of these users. This paper examines three key human factors: cognitive styles, levels of expertise and gender differences, and utilizes three individual clustering techniques: k-means, hierarchical clustering and fuzzy clustering to understand user behavior and perception. Moreover, robust clustering, capable of correcting the bias of individual clustering techniques, is used to obtain a deeper understanding. The robust clustering approach produced results that highlighted the relevance of cognitive style for user behavior, i.e., cognitive style dominates and justifies each of the robust clusters created. We also found that perception was mainly determined by the level of expertise of a user. We conclude that robust clustering is an effective technique to analyze user behavior and perception.","Title: Identifying user preferences with Wrapper-based Decision Trees\nYear: 2011\n\nAbstract: Wrapper feature selection approaches are widely used to choose a small subset of relevant features from a dataset. However, Wrappers suffer from the fact that they only use a single classifier. The downside to this is that each classifier will have its own biases and will therefore select very different features. To overcome the biases of individual classifiers, we propose a new data mining method called Wrapper-based Decision Trees (WDT). The WDT method uses multiple classifiers for selecting relevant features and decision trees to visualize relationships among the selected features. We use the WDT to investigate the influences of the levels of computer experience on users' preferences for the design of search engines. The benefit of using WDT lies within the fact that it can uncover the most accurate set of relevant features to help differentiate the preferences of users with diverse levels of computer experience. The results indicate that the users with varied levels of computer experiences have different preferences regarding the following features: the number of icons, the arrangement of search results, and the presentation of error messages. Such findings can be used to develop personalized search engines to accommodate users' different levels of computer experience.","Title: Evaluation of the credibility of internet shopping in the UK\nYear: 2009\n\nAbstract: Purpose - The purpose of this paper is to examine the credibility of internet shopping. Credibility, 20 March 2009 which refers to the believability of information, is an important consideration of internet shopping. Design/methodology/approach - The evaluation is conducted by incorporating Fogg's 10 Stanford Guidelines for Web Credibility into Nielsen's heuristic evaluation. Furthermore, security and individualisation are considered as additional heuristics. Evaluation criteria are developed based on these 12 heuristics. Three UK car insurance web sites are selected for evaluation, including the AA, Norwich Union and Tesco. Findings - The results show that the Norwich Union site seems to be the most credible while the Tesco site appears to be the least credible. The most significant credibility problems are found to lie in the areas of \"trustworthiness\", \"expertise\" and \"real-world feel\". In other words, these three areas are key issues for future improvement of these sites. Originality/value - This paper contributes to the literature by providing a set of credibility design guidelines, which can be used to support the improved development of future internet shopping designs, especially car insurance web sites.","Title: Game Poetry: Promotion of Poetry through a Game-Based Learning System\nYear: 2013\n\nAbstract: Poetry is a creative language which enhances imagination and self-reflection. However, the ambiguity of poetry increased the difficulty in interpretation, which builds students' negative perception because they consider that reading poetry is joyless and hard-understanding. Furthermore, the negative perception demotivates most students to read poetry. Therefore, to motivate students to read poetry, there is a need to improve the negative perception of poetry. Meanwhile, previous research shows games are able to avoid ambiguity by providing clear feedback and guidance. Besides, many studies also suggest game-based learning positively improves students' learning perception and motivation. To the end, this study develops a game-based learning system to motivate students to read poetry. The results show this game-based learning system significantly improves students' interpretation and perception. However, the long-term negative perception and the lack of self-efficacy affect students' motivation to read poetry after gaming.","Title: Experiences with Developing a User-Centered Digital Library\nYear: 2010\n\nAbstract: The purpose of this study was to find out whether users should be involved in the design stage of a Digital Library DL. Hence an experiment has been undertaken to find out the level of user satisfaction with a user-centred DL UDL compared to a non-user-centred DL NDL. In this experiment, the UDL prototype has been compared with the NDL prototype. The two prototypes were then evaluated by separate users with various Information Technology IT backgrounds. Results show that users' task performance was better in the UDL version of the prototype.","Title: Cognitive Modeling Of Student Learning In Web-Based Instructional Programs\nYear: 2004\n\nAbstract: There has been tremendous growth in Web-based instruction over the past few years. Because the user group of Web-based instruction includes learners from heterogeneous backgrounds, individual differences become an important issue in the development of Web-based instructional programs. Among a variety of individual differences, cognitive style is a particularly important characteristic. This study aims to determine the relationships between learners' cognitive styles and their perceptions and attitudes toward the features of a Web-based instructional program. The results indicate that cognitive styles influence students' reactions to nonlinear interaction, independent learning, and navigation tools and the difficulties and problems that they encounter. The findings are applied to develop a learning model that can support the design of Web-based instructional programs.","Title: A Cognitive Apprenticeship Approach to Facilitating Web-based Collaborative Problem Solving.\nYear: 2012\n\nAbstract: Enhancing students' problem-solving abilities has been recognized as an important and challenging issue for technology-enhanced learning. Thus, previous research has attempted to address this issue by developing various mechanisms, among which a cognitive apprenticeship model can particularly enhance students' abilities. However, it is not clear whether such a mechanism is suitable for every learner. Thus, this study examines the effects of human factors on problem-solving effectiveness in the cognitive apprenticeship model. Among various human factors, this study focuses on cognitive styles, with an emphasis on Witken's Field Dependence. The results indicate that Field Dependent learners can get great benefits from the cognitive apprenticeship model via collaborative learning. Implications for how to accommodate the needs of different cognitive style groups are discussed.","Title: The influences of a two-tier test strategy on student learning: A lag sequential analysis approach.\nYear: 2015\n\nAbstract: Recently, programming skills have become a core competence. Many teaching strategies were developed to improve programming skills. Among them, online tests were widely applied to enhance students learning. Nonetheless, they may not be able to engage students in deep thinking and reflections. Thus, a two-tier test strategy was proposed to address this issue. However, previous research mainly focused on investigating the effectiveness of the two-tier test strategy but there is a lack of studies that investigate why the two-tier test approach is effective. To this end, we developed an online test, where the two-tier test strategy was implemented. Additionally, an empirical study was conducted to explore the influences of the two-tier test approach on students' learning performance and behavior patterns. Pre-test and post-test scores were applied to assess students' learning performance while a lag sequential analysis was used to analyze behavior patterns. Regarding learning performance, the proposed two-tier test can improve students' programming skills. Regarding behavior patterns, the two-tier test approach facilitates students to develop a learning by reviewing strategy, which is useful to improve their programming skills.","Title: The Development Of A Robot-Based Learning Companion: A User-Centered Design Approach\nYear: 2015\n\nAbstract: A computer-vision-based method is widely employed to support the development of a variety of applications. In this vein, this study uses a computer-vision-based method to develop a playful learning system, which is a robot-based learning companion named RobotTell. Unlike existing playful learning systems, a user-centered design (UCD) approach is applied to assess the interface design of the RobotTell System so that the human-robot interaction can be enhanced during the design process. More specifically, the interface of the RobotTell system is assessed based on Nielsen's heuristics. Based on the results of Nielsen's heuristic evaluation, three versions of the RobotTell system were produced. The implications for the design of these three versions of the RobotTell system are discussed.","Title: A Study Of Users' Reactions To A Mixed Online Discussion Model: A Lag Sequential Analysis Approach\nYear: 2015\n\nAbstract: There is an increasing growth of Social Network Services (SNSs). A variety of SNSs are applied in online interpersonal platforms. Among them, asynchronous and synchronous discussions are widely examined. However, there is a lack of research into the effects of integrated discussion services that combines asynchronous and synchronous discussions. Thus, the study investigates users' performances and behavior patterns in a mixed discussion model that integrates asynchronous and synchronous discussions with a lag sequential analysis. The results showed that most groups chose to adopt the mixed discussion model (i.e., using both synchronous and asynchronous discussions), and only one group totally adopted the asynchronous discussion model. The study further analyzed the learners' learning effectiveness and behavioral patterns, and the results indicated that the groups using the mixed model had a positive performance to a certain extent in terms of learning effectiveness and knowledge construction. In addition, users with the mixed discussion model demonstrated diverse behaviors, which were more complex than that of those with a single-way discussion model. Furthermore, regarding the groups using \"balanced synchronous and asynchronous discussions\" and the groups \"mainly using synchronous discussion supplemented by less asynchronous discussion\" in the mixed model (simultaneously using synchronous and asynchronous discussions), those users who mainly used synchronous discussion supplemented by asynchronous discussion had more diverse behaviors of knowledge construction.","Title: The Effects of Game-Based Learning on Mathematical Confidence and Performance: High Ability vs. Low Ability.\nYear: 2014\n\nAbstract: Many students possess low confidence toward learning mathematics, which, in turn, may lead them to give up pursuing more mathematics knowledge. Recently, game-based learning (GBL) is regarded as a potential means in improving students' confidence. Thus, this study tried to promote students' confidence toward mathematics by using GBL. In addition, this study also investigated whether GBL is beneficial to all students with various abilities. The results demonstrated that this approach yielded better outcomes than the paper-based setting in both students' confidence and students' performance. The students with high and low levels of ability in the GBL group gained a significant improvement on the confidence toward mathematics. Additionally, low-ability students in the GBL group attained better mathematics performance than those in the paper-based setting.","Title: Individual Differences and Joyful Assessment-Based Learning\nYear: 2014\n\nAbstract: E-assessment learning, which combines the advantages of e-assessment and learning, was proposed in educational settings. However, it still makes most of learners feel anxious. Thus, this study develops a Joyful Assessment-based Learning System (JALS), which incorporates game-based learning into e-assessment learning. In order to get a complete understanding of how individual differences affect learners' reactions, an empirical study was conducted. Among various individual differences, cognitive styles and prior knowledge were considered as targets to investigate learners' reactions to the JALS. Our results suggest that FD learners preferred to use the game-based learning tool and demonstrated good performance. Conversely, the different level of prior knowledge may affect FI learners' learning preferences and learning performance.","Title: Cognitive styles and the use of electronic journals in a mobile context.\nYear: 2014\n\nAbstract: Purpose - The purpose of this paper is to examine how different cognitive style groups use electronic journals via mobile devices, with an emphasis on the use of search mechanisms, the approaches to judge the relevance and to refine the results, the use of horizontal screens, and on the influence of negative emotions. Design/methodology/approach - Participants' cognitive styles were identified as Holists and Serialists with Ford's Study Preference Questionnaire. The behaviour of the participants interacting with the ScienceDirect was observed and was analysed with both traditional statistics and data mining. Findings - Holists prefer to use multiple searches when undertaking the Expert Search only, which is a complex search mechanism. Conversely, Serialists have a tendency to use multiple searches, irrespective of the complexity of search mechanisms. Additionally, the Expert Search may be helpful for Serialists while the Basic Search seems beneficial to Holists. Regarding the judgement of relevance, Holists tend to use a variety of approaches while Serialists prefer to use a single approach. In general, the major results of this study confirm those of Vilar and Z. umer (2008). Originality/value - The novelty of this study covers two aspects. One is to examine the behaviour of Holists and Serialists in the mobile context of electronic journals and the other is to analyse data with an integrative approach that combines both traditional statistics and data mining. The former contributes to the understandings of how to design mobile electronic journals that can accommodate the needs of different cognitive style groups. The latter can not only strengthen the reliability of the results obtained from traditional statistics, but also discover findings neglected by traditional statistics.","Title: The effects of gender differences on the use of annotatable multimedia e-readers\nYear: 2015\n\nAbstract: AbstractPurpose - This paper aims to examine how gender differences influence students' reactions to the use of the annotatable multimedia e-reader AME. To reach this aim, we develop an AME where various annotation tools are provided to help students learn English in-class and after-class. Design/methodology/approach - An empirical study was conducted with 63 fifth-grade students from an elementary school. A pre-test and post-test were used to identify their prior knowledge and learning achievement, respectively. A questionnaire was applied to identify participants' perceptions towards the AME. Findings - The results show that students' post-test scores are significantly related to after-class behaviour, instead of in-class behaviour. Females prefer to use the text annotation and teachers' voice, but it is voice annotation that is beneficial to improve their learning achievement. Conversely, males prefer to use the text-to-speech only, but it is text annotation that is helpful to improve their learning achievement. Additionally, the ease of use affects males' intention to use the AME to learn English after-class while it has no effects on females. Originality/value - This study not only shows the importance of gender differences but also demonstrates the essence of after-class learning behaviour. More importantly, a framework is proposed to support designers to develop e-readers that can accommodate the preferences of females and males.","Title: The influences of an interactive group-based videogame: Cognitive styles vs. prior ability\nYear: 2015\n\nAbstract: This study aims to investigate how cognitive styles and the levels of prior abilities influence frustration tolerance and learning achievement in an interactive group-based videogame named “Multiple-Choice Practice Island.” In the aspect of cognitive styles, differences between field dependent and field independent students were explored. The results show that field independent students make better learning achievement but field dependent students demonstrate higher frustration tolerance. Besides, the low-ability students significantly made more improvement than the high-ability students. The low-ability students also demonstrate higher frustration tolerance in this study. The findings can guide designers how to develop adaptive group-based videogames that match the needs of each individual.","Title: Investigation Of Multiple Human Factors In Personalized Learning\nYear: 2016\n\nAbstract: In the past decade, a number of personalized learning systems have been developed and they mainly focus on learners' prior knowledge. On the other hand, previous research suggested that gender differences and cognitive styles have great effects on student learning. To this end, this study examines how human factors, especially gender differences and cognitive styles, affect learners' reactions to a personalized and non-personalized learning systems based on learners' prior knowledge. Forty-four university students participated in this study. The results show that females and Serialists show positive reactions to the personalized learning system, while males and Holists demonstrate similar reactions to the personalized learning system and the non-personalized learning system. The implications of these results for the design of personalized learning systems are discussed.","Title: The impact of cognitive styles on perceptual distributed multimedia quality\nYear: 2003\n\nAbstract: Multimedia technology has been widely used in web-based instruction, but previous studies have indicated that individual differences, especially cognitive styles, have significant effects on users' preferences with respect to presentation of multimedia content. However, such research has thus far neglected to examine the effect of cognitive styles on users' subjective perceptions of multimedia quality. This study aims to examine the relationships among users' cognitive styles, the multimedia Quality of Service (QoS) delivered by the underlying network, and Quality of Perception (QoP), which encompasses user levels of enjoyment and understanding of the informational content provided by multimedia material. Accordingly, 132 users took part in an experiment in which they were shown multimedia video clips presented with different values of two QoS parameters (frame rate and colour depth). Results show that, whilst the two QoS parameters do not impact user QoP multimedia content and dynamism levels significantly influence the user understanding and enjoyment component of QoP.","Title: Matching/mismatching revisited: an empirical study of learning and teaching styles\nYear: 2001\n\nAbstract: This paper presents results of a research project that explored the relationship between matching and mismatching instructional presentation style (breadth-first and depth-first) with students' cognitive style (field-dependence/-independence) in a computer-based learning environment, 73 postgraduate students were asked to create Web pages using HTML, using instructional materials that were either matched or mismatched with their cognitive styles. Significant differences in performance on a multiple choice test of conceptual knowledge were found for students learning in matched and mismatched conditions. Performance in matched conditions was significantly superior to that in mismatched conditions. However, significant effects were found for gender, matching mainly affecting male students. Performance on a practical test of Web page creation was not linked to matching or mismatching, but was linked to an interaction between gender and instructional presentation style, The findings provide support for the notion that matching and mismatching can have significant effects on learning outcomes. The paper concludes with suggestions for further research.","Title: An Investigation of a Game-Based Anti-Drug System: Addictive Learners vs. Non-addictive Learners\nYear: 2015\n\nAbstract: The spread of drugs causes serious problems recently. To enhance people awareness about drugs, several approaches were adopted to deliver anti-drug materials. However, limitations were observed in these approaches. Thus, this study develops a game-based anti-drug system (GADS), which incorporates game-based learning into anti-drug materials. In order to get a complete understanding of whether the GADS is suitable to both non-addictive users and addictive users, an empirical study was conducted. Drug experience was considered as a target to investigate learners' reactions to the GADS. The results indicated that non-addictive learners had better game performance and learning performance than addictive learners because addictive learners' working memory was negatively affected by drugs.","Title: An Investigation of a Two-Tier Test Strategy in a University Calculus Course: Causes versus Consequences.\nYear: 2016\n\nAbstract: Online tests have been identified as a core learning activity in higher education. Unlike conventional online tests, which cannot completely reflect students' learning status, two-tier tests not only consider students' answers, but also take into account reasons for their answers. Due to such significance, research into the two-tier tests had mushroomed but few studies examined why the two-tier te...","Title: An Investigation Of A Joyful Peer Response System: High Ability Vs. Low Ability\nYear: 2016\n\nAbstract: Existing peer response approaches majorly take place in traditional educational contexts, where students may not be highly motivated. On the other hand, game-based learning takes a joyful way to enhance students' motivation. Accordingly, we propose a joyful peer response (JPR), where game-based learning and peer response are integrated together. Furthermore, two empirical studies were conducted to investigate how high-and low-ability students perform in the JPR and how their perceptions are associated with their performance. Regarding the former, the findings suggest that the JPR is beneficial to high-and low-ability students. Regarding the latter, the findings imply that the writing performance of the high-ability students is majorly related to their perceptions of game elements. On the other hand, the writing performance of the low-ability students is related to their perceptions of both game elements and peer response.","Title: Incorporating customization and personalization into game-based learning: A cognitive style perspective.\nYear: 2016\n\nAbstract: In recent years, customization and personalization were widely applied to accommodate the needs of different cognitive style groups. Such two approaches have different advantages and disadvantages but there is a lack of studies to compare these two approaches from the perspective of game-based learning, which is currently popular in educational settings. To this end, we developed a customized game-based learning system and a personalized game-based learning system and conducted two empirical studies to examine how cognitive styles affected learner's reactions to these two game-based learning systems. The results from the customized game-based learning system showed that Holists might not always favor to listen to music because they frequently switched on/off music. On the other hand, Serialists did not prefer to use hints. In addition, learners with the customized game-based learning system had more positive perceptions while learners with the personalized game-based learning system had more negative perceptions though both systems were useful to enhance learners' learning performance, regardless of their cognitive styles.","Title: The effects of task activities and gaming scales on eye reading and visual search performance.\nYear: 2017\n\nAbstract: The effects of video gaming have been explored extensively in many learning-related fields. However, limited research has investigated whether the frequencies of exposing to gaming environments influence students' Chinese reading comprehension, arithmetic word problem test, spatial recognition, and graphical patterns recognition. To address this issue, this study adopted eye-tracking technology to explore learners' reading and visual search responses on various reading materials. Task effects and correlation analyses were conducted on a sample of 19 sophomores with varying gaming frequencies, eye reading fixations, and regressions. The results revealed that practice makes users have fewer fixation frequencies on a similar task. Meanwhile, the results demonstrated a positive gaming frequency correlation on eye fixations and regressions for Chinese reading comprehension and arithmetic word problem test, and a negative gaming frequency correlation for spatial and graphical patterns recognition tasks.","Title: Exploring The Interactive Patterns Of Concept Map-Based Online Discussion: A Sequential Analysis Of Users' Operations, Cognitive Processing, And Knowledge Construction\nYear: 2016\n\nAbstract: Concept maps can be used as a cognitive tool to assist learners' knowledge construction. However, in a concept map-based online discussion environment, studies that take into consideration learners' manipulative actions of composing concept maps, cognitive process among learners' discussion, and social knowledge construction at the same time are still lacking. Therefore, this study employed lag sequential analysis to empirically explore learners' behavioral patterns in a concept map-based online discussion environment. We explored learners' operations in the concept map tool, and their discussion behavioral patterns of cognitive processing phases and social knowledge construction. The results showed that it was easier for learners to define a node than to define a relationship when they used the concept map. Additionally, the concept map can enhance higher level cognitive processing phases and lead learners to return to the main discussions from off-topic discussions significantly.","Title: The Influences of the 2D Image-Based Augmented Reality and Virtual Reality on Student Learning.\nYear: 2017\n\nAbstract: Virtual reality (VR) learning environments can provide students with concepts of the simulated phenomena, but users are not allowed to interact with real elements. Conversely, augmented reality (AR) learning environments blend real-world environments so AR could enhance the effects of computer simulation and promote students' realistic experience. However, AR-based learning environments had a lot of dynamic real objects which may increase learners' mental effort. Moreover, paucity of research compared AR with VR and other mature technologies. Thus, the aim of this study is to compare the influence of the 2D imagebased VR and AR in an inquiry-based astronomy course. The findings of this study suggested that the real objects presented in the AR system could reduce the mental load because students could take the real objects of the AR system as the reference objects of the movement of the moon. Furthermore, the sense of the immediacy is increased due to the fact that peers appear on the AR system. Accordingly, the real objects and the sense of the immediacy not only enhance the learning motivations, but also encourage the students to keep conducting the tasks.","Title: Developing a multimedia encyclopedia for little people and their families.\nYear: 2017\n\nAbstract: Among a variety of disabilities, persons with achondroplasia, i.e., little people, have relatively minor symptoms though they have small bodies. Thus, they can learn how to lead an independent life. This paper presents a study on the development of a multimedia encyclopedia for little people and their families. Due to their different roles, this study examines how little people and their families reacted to the multimedia encyclopedia. In total, 18 little people and 15 family members participated in this study. Their interaction behavior was recorded in log files and their perceptions were identified by a questionnaire. The results indicated that little people preferred to use a search mechanism while their family members favored to use pop-up windows. Either little people or their families thought that the multimedia encyclopedia provided some new information that they did not know. However, family members’ attention, confidence and satisfaction with the multimedia encyclopedia were higher than those of the little people. This study can contribute to the knowledge of how to develop individualized multimedia encyclopedias that can accommodate the different preferences of little people and their family members.","Title: Students' Reactions to Different Levels of Game Scenarios: A Cognitive Style Approach.\nYear: 2017\n\nAbstract: Game-based learning comprises of a set of concrete scenarios, which can be presented with various presentation modalities (e.g., text, text with graphic, and context). Such presentation modalities disseminate information in different ways. On the other hand, cognitive styles affect how people process information. Accordingly, the study presented in this paper investigates the influences of cognitive styles on students' performance within different levels of game-based scenarios. More specifically, this study examines how students with different cognitive styles (i.e., Holist/Serialist) react to three presentation modalities (i.e., text, text with graphic, and context) in game-based scenarios. The results of an empirical study with 96 students revealed that all students made improvement, regardless of any presentation modalities. However, Holists significantly performed better than Serialists in the context version. According to the results, further suggestions for system design are discussed.","Title: An investigation of the development of an animated E-book: A gender difference approach.\nYear: 2018\n\nAbstract: Two empirical studies, i.e., preliminary study and main study, were included in this research. In the preliminary study, we developed a prototyped animated e-book system (PAES) that incorporated animation films into an e-book and examined how females and males reacted to design features provided by the PAES. The results from the preliminary study showed that gender differences significantly affected students' reactions to the PAES. Accordingly, the main study attempted to develop an animated e-book system (AES) that could accommodate gender differences based on the results from the preliminary study. The results from our main study showed that gender differences had no significant effects on students' learning performance and learning perceptions though females and males demonstrated different learning behavior. In other words, both females and males could equally benefit from the AES. Briefly, the major contribution of this research was to provide guidance on how to develop an animated e-book from a gender difference perspective. Such guidance can help designers to develop a versatile animated e-book that can support the needs and preferences of both females and males.","Title: The influences of self-regulated learning support and prior knowledge on improving learning performance.\nYear: 2018\n\nAbstract: Self-regulated learning (SRL) is helpful to students. On the other hand, prior knowledge has great effects on students' self-regulation and learning performance. To this end, this study aimed to examine how high prior knowledge students (HPKs) and low prior knowledge students (LPKs) behaved differently when interacting with a SRL environment. To achieve this aim, we proposed a self-regulated learning support system (SRLSS) for a mathematical course. The results showed that the gap of learning performance between the HPKs and LPKs was removed after a long-term learning process. Moreover, the LPKs and HPKs behaved similarly in the forethought and self-reflection phases but some behavior differences were found in the performance phase, where the LPKs relied on the notes and sought support the dashboard and quiz records while the HPKs did not demonstrate such a tendency. Our results' theoretical and methodological implications and possible applications for further research are also discussed.","Title: Incorporating digital games into anti-drug material: non-drug-addicted learners vs. drug-addicted learners\nYear: 2020\n\nAbstract: Drug abuse causes serious problems around the world. Thus, anti-drug material has been increasingly developed and used. However, conventional anti-drug material has until today shown insufficient effects on preventing drug abuse, especially for drug-addicted users. To this end, this study proposes a game-based anti-drug material and examines how drug-addicted people and non-drug-addicted people interact with the game-based anti-drug material. A lag sequential analysis approach was applied to analyze their different behavior patterns. The results revealed that the experience of drug addiction influenced the way individuals access drug information, complete tasks and reach goals during their gaming process. Based on the findings, some suggestions are proposed for the development of adaptive game-based anti-drug material that can accommodate the needs of drug-addicted and non-drug-addicted people to achieve the goal of universal access.","Title: The Effects of Cognitive Styles on the Use of Hints in Academic English: A Learning Analytics Approach.\nYear: 2017\n\nAbstract: Various hints have been proposed to scaffold students. Such hints can be broadly divided into direct hints and indirect hints. On the other hand, individual differences existed among learners. In particular, cognitive styles greatly affect student learning. However, there is a lack of studies to investigate how cognitive styles affect students' reactions to the use of hints in the context of academic English. To address this issue, this study aimed to examine the effects of cognitive styles on the use of hints in such a context. To obtain deep understandings, a learning-analytics approach was applied in this study, including quantitative measurement, qualitative evaluation and lag sequential analyses. The results from the quantitative measurement indicated that the frequencies of using Chinese hints and synonym hints were significantly associated with Serialists' task scores while such significant correlations were not found for Holists. The results from the qualitative evaluation showed that Holists favored to use synonym hints while Serialists preferred to use Chinese hints. The results from the lag sequential analyses suggested that Holists demonstrated iterative behavior while Serialists showed sequential behavior. In summary, cognitive styles have considerable influences on students' learning patterns in the context of Academic English.","Title: An Investigation of the Effects of Scaffolding Task-Based English Logic Learning: A Cognitive Style Approach\nYear: 2022\n\nAbstract: Task-based learning can be used to promote language learning. On the other hand, scaffolding instruction can help students without sufficient abilities to undertake tasks. Accordingly, we developed the Scaffolding Task-based English Logic Learning (S-TELL) by incorporating scaffolding instruction into Task-based Learning. The S-TELL was applied to facilitate research students to learn how to present clear logical relationships in academic writing. Additionally, an empirical study was conducted to investigate how different cognitive style groups (i.e., Serialists and Holists) reacted to the S-TELL, including learning performance, learning perception and learning behavior. Regarding learning performance, the findings showed that either Holists or Serialists made significant improvement via interacting with the S-TELL. Moreover, Holists significantly obtained higher post-scores than Serialists though they had similar task scores. Regarding learning perception, few perception differences existed between Serialists and Holists. Regarding learning behavior, Holists demonstrated more behavior sequences than Serialists. This was due to the fact that Holists often checked the current status and used the English hint while Serialists did not show such behavior. According to these findings, we proposed a framework, which can be employed to personalize the S-TELL in our future work so that the preferences of Holists and Serialists can be accommodated."],"string":"[\n \"Title: Navigation in hypermedia learning systems: experts vs. novices\\nYear: 2006\\n\\nAbstract: With the advancement of Web technology, hypermedia learning systems are becoming more widespread in educational settings. Hypermedia learning systems present course content with non-sequential formats, so students are required to develop learning paths by themselves. Yet, empirical evidence indicates that not all students can benefit from hypermedia learning. Research into individual differences suggests that prior knowledge has significant effects on student learning in hypermedia systems, with experts and novices showing different preferences to the use of hypermedia learning systems and requiring different levels of navigation support. It is therefore essential to develop a mechanism to help designers understand the needs of experts and novices. To address this issue, this paper presents a framework to illustrate the needs of students with different levels of prior knowledge by analyzing the findings of previous research. The overall aim of this framework is to integrate students' prior knowledge into the design of hypermedia learning systems. Finally, implications for the design of hypermedia learning systems are discussed.\",\n \"Title: Effects of gender differences and spatial abilities within a digital pentominoes game\\nYear: 2010\\n\\nAbstract: Spatial ability is a critical skill in geometric learning. Several studies investigate how to use digital games to improve spatial abilities. However, not every learner favors this kind of support. To this end, there is a need to examine how human factors affect learners' reactions to the use of a digital game to support geometric learning. In this vein, this paper addresses this issue by developing a digital pentominoes game and examining the effects of two essential human factors, especially gender differences and spatial abilities, on students' performance. The results demonstrate that students' spatial abilities were significantly improved after they took the digital pentominoes game. The results also demonstrate that the digital game can reasonably reduce the differences between boys and girls. Moreover, the major gender differences lie within mental rotation among the three types of spatial ability and also mainly exist in the low spatial ability group. Finally, the findings are applied to develop a framework that can be used to enhance the understanding of gender differences and spatial abilities within the digital pentominoes game.\",\n \"Title: Evaluation of a personalized digital library based on cognitive styles: Adaptivity vs. adaptability.\\nYear: 2009\\n\\nAbstract: Personalization can be addressed by adaptability and adaptivity, which have different advantages and disadvantages. This study investigates how digital library (DL) users react to these two techniques. More specifically, we develop a personalized DL to suit the needs of different cognitive styles based on the findings of our previous work [Frias-Martinez, E., Chen, S. Y., & Liu, X. (2008) Investigation of behavior and perception of digital library users: A cognitive style perspective. International Journal of Information Management]. The personalized DL includes two versions: adaptive version and adaptable version. The results showed that users not only performed better in the adaptive version, but also they perceived more positively to the adaptive version. In addition, cognitive styles have great effects on users' responses to adaptability and adaptivity. These results provide guidance for designers to select suitable techniques to develop personalized DLs.\",\n \"Title: Mining students' behavior in web-based learning programs\\nYear: 2009\\n\\nAbstract: There has been a proliferation of web-based learning programs (WBLPs). Unlike traditional computer-based learning programs, WBLPs are used by a population of learners who have diverse background. How different learners access the WBLPs has been investigated by several studies, which indicate that cognitive style is an important factor that influences learners' preferences. However, these studies mainly use statistical methods to analyze learners' preferences. In this paper, we propose to analyze learners' preferences with a data mining technique. Findings in our study show that Field Independent learners frequently use backward/forward buttons and spent less time for navigation. On the other hand, Field Dependent learners often use main menu and have more repeated visiting. Implications for these findings are discussed.\",\n \"Title: User perceptions of online public library catalogues.\\nYear: 2008\\n\\nAbstract: Online Public Library Catalogues (OPACs) are widely used electronic library catalogues giving a wealth of remote access to library information resources. Users should be involved early in the OPAC development cycle process in order to ensure a usable and functional interface, as the integration of user-defined requirements of OPACs, along with the other human-computer interaction considerations, offer a better understanding of user perceptions and expectations in respect of OPACs, ultimately resulting in truly user-centred OPACs. Accordingly, the purpose of this study was to establish user suggestions for a typical OPAC application's functionality and features. To this end, an experiment was undertaken to find out the type of interaction features that users prefer to have in an OPAC. The study revealed that regardless of users' Information Technology (IT) backgrounds, their functionality expectations of OPACs are the same as users are expecting OPACs to facilitate easier ways to achieve their tasks. However, based on users' previous experiences with OPACs, their requirements with respect to specific features may change.\",\n \"Title: Matching and Mismatching in Web-based Learning: A Human Centered Apprach\\nYear: 2011\\n\\nAbstract: This study examines the effects of matching and mismatching on web-based learning with a human-centered approach. In particular, we emphasize on how prior knowledge and gender differences affect students' reactions to a matching/mismatching environment. In general, the results demonstrate that learners in a matching condition perform better than those in a mismatching condition. Furthermore, males can overcome the mismatching condition when they acquire declarative knowledge. In contrast, females can positively face the mismatching condition when they obtain procedural knowledge. In summary, the matching/mismatching is not only related to gender differences but also is linked to declarative/procedural knowledge.\",\n \"Title: Perceived quality of multimedia educational content: A cognitive style approach\\nYear: 2006\\n\\nAbstract: Cognitive styles influence the way how humans process information, with previous research demonstrating that they have significant effects on student learning in multimedia environments. On the other hand, the perceptual quality of the human multimedia experience is notoriously difficult to measure. In this paper, we report the results of an empirical study, which investigated the relationship between user cognitive styles and perceptual multimedia quality, in which users had the possibility to specify their desired Quality of Service settings -- in terms of frame rates and color depth. Results show that whilst color choice is impacted by a participant's cognitive style, such Quality of Service parameters do not significantly affect perceived multimedia quality, and that users do not necessarily choose optimum presentation settings to enhance their perceived enjoyment and assimilation of multimedia informational content.\",\n \"Title: Hypermedia learning and prior knowledge: domain expertise vs. system expertise\\nYear: 2005\\n\\nAbstract: Prior knowledge is often argued to be an important determinant in hypermedia learning, and may be thought of as including two important elements: domain expertise and system expertise. However, there has been a lack of research considering these issues together. In an attempt to address this shortcoming, this paper presents a study that examines how domain expertise and system expertise influence students' learning performance in, and perceptions of, a hypermedia system. The results indicate that participants with lower domain knowledge show a greater improvement in their learning performance than those with higher domain knowledge. Furthermore, those who enjoy using the Web more are likely to have positive perceptions of non-linear interaction. Discussions on how to accommodate the different needs of students with varying levels of prior knowledge are provided based on the results.\",\n \"Title: Mining Students' Learning Patterns And Performance In Web-Based Instruction: A Cognitive Style Approach\\nYear: 2011\\n\\nAbstract: Personalization has been widely used in Web-based instruction (WBI). To deliver effective personalization, there is a need to understand different preferences of each student. Cognitive style has been identified as one of the most pertinent factors that affect students' learning preferences. Therefore, it is essential to investigate how learners with different cognitive styles interact with WBI programs. This paper presents an empirical study, which examines the effects of cognitive styles on students' learning patterns and the effects of learning patterns on their learning performances. Riding's cognitive style analysis was used to identify the students' cognitive styles. Data mining, especially a clustering technique, was used to analyze the results. It was found that field independent students frequently used an alphabetical index whereas field dependent students often chose a hierarchical map. Such learning patterns also have great effects on their performance, especially for field dependent students.\",\n \"Title: Measuring quality of perception in distributed multimedia: Verbalizers vs. imagers\\nYear: 2008\\n\\nAbstract: This paper presents the results of a study which investigated the impact of cognitive styles on perceptual multimedia quality. More specifically, we examine the different preferences demonstrated by verbalizers and imagers when viewing multimedia content presented with different quality of service (QoS) levels pertaining to frame rates and color depth. Recognizing multimedia's infotainment duality, we used the quality of perception (QoP) metric to characterize perceived quality. Results showed that in terms of low and high dynamisms clips, the frame rate at which multimedia content is displayed influences the levels of information assimilated by Imagers. Whilst black and white presentations are shown to be beneficial for both Biomodals and Imagers in order to experience enhanced levels of information assimilation, Imagers were shown to enjoy presentations in full 24-bit colour.\",\n \"Title: Mining Learning Preferences in Web-based Instruction: Holists vs. Serialists.\\nYear: 2011\\n\\nAbstract: Web-based instruction programs are used by learners with diverse knowledge, skills and needs. These differences determine their preferences for the design of Web-based instruction programs and ultimately influence learners' success in using them. Cognitive style has been found to significantly affect learners' preferences of web-based instruction programs. However, the majority of previous studies focus on Field Dependence/Independence. Pask's Holist/Serialist dimension has conceptual links with Field Dependence/Independence but it is left mostly unstudied. Therefore, this study focuses on identifying how this dimension of cognitive style affects learner preferences of Web-based instruction programs. A data mining approach is used to illustrate the difference in preferences between Holists and Serialists. The findings show that there are clear differences in regard to content presentation and navigation support. A set of design features were then produced to help designers incorporate cognitive styles into the development of Web-based instruction programs to ensure that they can accommodate learners' different preferences.\",\n \"Title: An Integrated Approach for Modeling Learning Patterns of Students in Web-Based Instruction: A Cognitive Style Perspective\\nYear: 2008\\n\\nAbstract: Web-based instruction (WBI) programs, which have been increasingly developed in educational settings, are used by diverse learners. Therefore, individual differences are key factors for the development of WBI programs. Among various dimensions of individual differences, the study presented in this article focuses on cognitive styles. More specifically, this study investigates how cognitive styles affect students' learning patterns in a WBI program with an integrated approach, utilizing both traditional statistical and data-mining techniques. The former are applied to determine whether cognitive styles significantly affected students' learning patterns. The latter use clustering and classification methods. In terms of clustering, the K-means algorithm has been employed to produce groups of students that share similar learning patterns, and subsequently the corresponding cognitive style for each group is identified. As far as classification is concerned, the students' learning patterns are analyzed using a decision tree with which eight rules are produced for the automatic identification of students' cognitive styles based on their learning patterns. The results from these techniques appear to be consistent and the overall findings suggest that cognitive styles have important effects on students' learning patterns within WBI. The findings are applied to develop a model that can support the development of WBI programs.\",\n \"Title: Web-based interaction: A review of three important human factors.\\nYear: 2010\\n\\nAbstract: With the rapid development of information technology, the World Wide Web has been widely used in various applications, such as search engines, online learning and electronic commerce. These applications are used by a diverse population of users with heterogeneous backgrounds, in terms of their knowledge, skills, and needs. Therefore, human factors are key issues for the development of Web-based applications, leading research into human factors to grow significantly in the past decade. This paper identifies and reviews three important human factors that have been examined in existing empirical studies, including gender differences, prior knowledge, and cognitive styles. The main results from the analysis include that: (a) females have more disorientation problems than males; (b) flexible paths are more beneficial to experts while structured content is more useful to novices; and (c) Field Dependent and Field Independent users prefer to employ different search strategies. In addition to reviewing the existing empirical studies, this paper also highlights areas of future research.\",\n \"Title: The Influences of Number and Nature of Classifiers on Consensus Feature Selection\\nYear: 2008\\n\\nAbstract: Abstract - Wrapper feature selection approaches are widely used to select a small subset of relevant features from a dataset. However, Wrappers suffer from the fact that they only use a single classifier when selecting the features. The downside to this approach is that each classifier will have its own biases and will therefore select very different features. In order to overcome the biases of individual classifiers, we propose Consensus Feature Selection (CFS), which combines different classifiers for feature selection. In this way, selecting classifiers for use in the combinations is very important. Therefore, we investigate how the number and nature of classifiers influence the number of features selected and the classification accuracies that these features generate. In terms of number of classifiers, results showed that few selected more relevant features whereas many selected few features. In addition, 3-classifier combinations selected features that led to highest accuracies. In terms of nature of classifiers, decision trees identified most number of features whereas Bayesian classifiers identified least number of features. However, features selected by Bayesian classifiers led to accuracies higher than the other classifiers\",\n \"Title: Discussion Models in Seamless Online Learning Integrated Discussion (SOLID) Environment: Selection vs. Performance\\nYear: 2012\\n\\nAbstract: Synchronous discussion and asynchronous discussion have different advantages and disadvantages. However, it is unclear how students react to these two models. Thus, we developed a learning environment that combines these two models, and then examined how students selected them. The findings show that students selected different models at different discussion stages. Additionally, the mixed discussion model, which combines synchronous and asynchronous models, can achieve better performance in discussion, regardless of the types of discussion activity.\",\n \"Title: Associating learners' cognitive style with their navigation behaviors: a data-mining approach\\nYear: 2011\\n\\nAbstract: Investigating how learners' individual differences affect their navigation behavior can help us understand learners' preferences and can be used to develop the Web-based learning system that can meet learners' needs. Among various individual differences, learners' cognitive styles need to be considered because it is concerned how users process information. There is also a need to examine the relationship between their navigation behavior and performance because such findings can be used to suggest suitable navigation tools to meet learners' real needs. To this end, the aims of this study tend to use a data mining approach to not only investigate the relationship between learners' cognitive styles and their navigation behavior but also analyze how their navigation behavior influence performance. The results indicate that holists and serialists have different navigation behavior. However, there is no direct relationship between learning performance and navigation behavior.\",\n \"Title: Automatic cognitive style identification of digital library users for personalization\\nYear: 2007\\n\\nAbstract: Digital libraries have become one of the most important Web services for information seeking. One of their main drawbacks is their global approach: In general, there is just one interface for all users. One of the key elements in improving user satisfaction in digital libraries is personalization. When considering personalizing factors, cognitive styles have been proved to be one of the relevant parameters that affect information seeking. This justifies the introduction of cognitive style as one of the parameters of a Web personalized service. Nevertheless, this approach has one major drawback: Each user has to run a time-consuming test that determines his or her cognitive style. In this article, we present a study of how different classification systems can be used to automatically identify the cognitive style of a user using the set of interactions with a digital library. These classification systems can be used to automatically personalize, from a cognitive-style point of view, the interaction of the digital library and each of its users. © 2007 Wiley Periodicals, Inc.\",\n \"Title: Investigation Of The Use Of Navigation Tools In Web-Based Learning: A Data Mining Approach\\nYear: 2008\\n\\nAbstract: Web-based learning is widespread in educational settings. The popularity of Web-based learning is in great measure because of its flexibility. Multiple navigation tools provided some of this flexibility. Different navigation tools offer different functions. Therefore, it is important to understand how the navigation tools are used by learners with different backgrounds, knowledge, and skills. This article presents two empirical studies in which data-mining approaches were used to analyze learners' navigation behavior. The results indicate that prior knowledge and subject content are two potential factors influencing the use of navigation tools. In addition, the lack of appropriate use of navigation tools may adversely influence learning performance. The results have been integrated into a model that can help designers develop Web-based learning programs and other Web-based applications that can be tailored to learners' needs.\",\n \"Title: Cognitive styles and hypermedia navigation: development of a learning model\\nYear: 2002\\n\\nAbstract: There has been an increased growth in the use of hypermedia to deliver learning and teaching material. However, much remains to be learned about how different learners perceive such systems. Therefore, it is essential to build robust learning models to illustrate how hypermedia features are experienced by different learners. Research into individual differences suggests cognitive styles have a significant effect on student learning in hypermedia systems. In particular, Witkin's Field Dependence has been extensively examined in previous studies. This article reviews the published findings from empirical studies of hypermedia learning. Specifically, the review classifies the research into five themes: nonlinear learning, learner control, navigation in hyperspace, matching and mismatching, and learning effectiveness. A learning model, developed from an analysis of findings of the previous studies, is presented. Finally, implications for the design of hypermedia learning systems are discussed.\",\n \"Title: Cognitive styles and users' responses to structured information representation\\nYear: 2004\\n\\nAbstract: Structured information repositories, such as digital libraries, Web directories, and subject gateways, require effective ways to organise and manage information. This paper focuses on Web directories and investigates the relationships between users' cognitive styles and information representation. The results indicate that cognitive style influences participants' reactions to the organisation of subject categories, presentation of the results, and screen layout. The findings are applied to develop a design framework that can support the improvement of Web directories and other structured digital information resources. Finally, implications for information representation are discussed.\",\n \"Title: Recent Soft Computing Approaches to User Modeling in Adaptive Hypermedia\\nYear: 2004\\n\\nAbstract: The ability of an adaptive hypermedia system to create tailored environments depends mainly on the amount and accuracy of information stored in each user model. One of the difficulties that user modeling faces is the necessity of capturing the imprecise nature of human behavior. Soft Computing has the ability to handle and process uncertainty which makes it possible to model and simulate human decision-making. This paper surveys different soft computing techniques that can be used to efficiently and accurately capture user behavior. The paper also presents guidelines that show which techniques should be used according to the task implemented by the application.\",\n \"Title: Mammogram retrieval on similar mass lesions.\\nYear: 2012\\n\\nAbstract: Enormous numbers of digital mammograms have been produced in hospitals and breast screening centers. To exploit those valuable resources in aiding diagnoses and research, content-based mammogram retrieval systems are required to effectively access the mammogram databases. This paper presents a content-based mammogram retrieval system, which allows medical professionals to seek mass lesions that are pathologically similar to a given example. In this retrieval system, shape and margin features of mass lesions are extracted to represent the characteristics of mammographic lesions. To compare the similarity between the query example and any lesion within the databases, this study proposes a similarity measure scheme which involves the hierarchical arrangement of mammographic features and a weighting distance measure. This makes similarity measure of the retrieval system consistent with the way radiologists observe mass lesions. This study used the DDSM dataset to evaluate the effectiveness of the extracted shape feature and margin feature, respectively. Experimental results demonstrate that, when Zernike moments are used, round-shape masses are the most discriminative among four types of shape; the circumscribed-margin masses can be effectively discriminated among the four types of margins. Moreover, the result also shows that, when retrieving round-shape and circumscribed margin masses, this retrieval system can achieve the highest precision among all mass lesion types.\",\n \"Title: Web-based Assessment and Test Analyses (WATA) system: development and evaluation\\nYear: 2004\\n\\nAbstract: This study introduces the development of a Web-based assessment system, the Web-based Assessment and Test Analyses (WATA) system, and examines its impacts on teacher education. The WATA system is a follow-on system, which applies the Triple-A Model (assembling, administering, and appraising). Its functions include (1) an engine for teachers to administer and manage testing, (2) an engine for students to apply tests, and (3) an engine for generating test results and analyses for teachers. Two studies were undertaken to assess the usefulness and potential benefits of the WATA system for teacher education. In the first study, 47 in-service teachers were asked to assess the functions of the WATA system. The results indicated that they were satisfied with the Triple-A Model of the WATA system. In the second study, 30 pre-service teachers were required to use the WATA system during the teacher-training program. After 4 months of experience in using the WATA system, the pre-service teachers' perspectives of assessment have been changed significantly. The findings of these two studies might provide some guidance to help those who are interested in the development of Web-based assessment and intend to infuse information technology into teacher education.\",\n \"Title: Satellite-based delivery of educational content to geographically isolated communities: a service based approach\\nYear: 2009\\n\\nAbstract: Enabling learning for members of geographically isolated communities presents benefits in terms of promoting regional development and cost savings for governments and companies. However, notwithstanding recent advances in e-Learning, from both technological and pedagogical perspectives, there are very few, if any, recognised methodologies for user-led design of satellite-based e-learning infrastructures. In this paper, we present a methodology for designing a satellite and wireless based network infrastructure and learning services to support distance learning for such isolated communities. This methodology entails (a) the involvement of community members in the development of targeted learning services from an early stage, and (b) a service-oriented approach to learning solution deployment. Results show, that, while the technological premises of distance learning can be accommodated by hybrid satellite/wireless infrastructures, this has to be complemented with (a) high-quality audio---visual educational material, and (b) the opportunity for community members to interact with other community members either as groups (common-room oriented scenarios) or individuals (home-based scenarios), thus providing an impetus for learner engagement in both formal and informal activities.\",\n \"Title: Evaluating The Learning Effectiveness Of Using Web-Based Instruction: An Individual Differences Approach\\nYear: 2005\\n\\nAbstract: The use of Web-based instruction is becoming widespread in higher education; however, much remains to be learned about how different learners react to such instructional programs. The study presented in this article evaluates students' learning performance and their perceptions in a Web-based instructional program, which was applied to teach students how to use HTML in Brunel University's Department of Information Systems and Computing. Sixty-one master's students participated in this study. A number of interesting interactions were found. Students' task achievements were affected by the levels of their previous system experience. On the other hand, the post-test and gain scores were positively influenced by their perceptions and attitudes toward the Web-based instructional program. The implications of these findings are discussed.\",\n \"Title: The effects of a Creative Commons approach on collaborative learning.\\nYear: 2013\\n\\nAbstract: Social media on the World Wide Web, such as Wiki, are increasingly applied to support collaborative learning for students to conduct a project together. However, recent studies indicated that students, learning in the collaborative project, may not actively contribute to the collaborative work and are involved only in a limited level of positive interdependence due to the ambiguous ownership on the collaborative work. To this end, this study proposes an approach to support collaborative learning based on the notion of Creative Commons (CC). CC may be helpful to enhance positive interdependency because it provides a mechanism which protects individual creations while encouraging remixing and deriving new creations from them. This study conducted an empirical evaluation to examine how students react to the collaborative learning with CC notions. The results showed that CC can significantly improve participants' attitude to the derivative works, the satisfaction level of remix outcomes, perception to the peer interaction and the sense of work ownership. Such results suggest that CC may be a potential avenue for increasing individuals' responsibilities and motivating them to participate in the collaborative learning activity.\",\n \"Title: The contribution of data mining to information science\\nYear: 2004\\n\\nAbstract: The information explosion is a serious challenge for current information institutions. On the other hand, data mining, which is the search for valuable information in large volumes of data, is one of the solutions to face this challenge. In the past several years, data mining has made a significant contribution to the field of information science. This paper examines the impact of data mining by reviewing existing applications, including personalized environments, electronic commerce, and search engines. For these three types of application, how data mining can enhance their functions is discussed. The reader of this paper is expected to get an overview of the state of the art research associated with these applications. Furthermore, we identify the limitations of current work and raise several directions for future research.\",\n \"Title: The assessment of usability of electronic shopping: A heuristic evaluation.\\nYear: 2005\\n\\nAbstract: Today there are thousands of electronic shops accessible via the Web. Some provide user-friendly features whilst others seem not to consider usability factors at all. Yet, it is critical that the electronic shopping interface is user-friendly so as to help users to obtain their desired results. This study applied heuristic evaluation to examine the usability of current electronic shopping. In particular, it focused on four UK-based supermarkets offering electronic services: including ASDA, Iceland, Sainsbury, and Tesco. The evaluation consists of two stages: a free-flow inspection and a task-based inspection. The results indicate that the most significant and common usability problems have been found to lie within the areas of 'User Control and Freedom' and 'Help and Documentation'. The findings of this study are applied to develop a set of usability guidelines to support the future design of effective interfaces for electronic shopping.\",\n \"Title: Electronic trombone: an interactive tool to promote musical learning and performance creativity\\nYear: 2011\\n\\nAbstract: A lot of musical toys (e.g., toy guitars, toy drums and percussion) can provide a good introduction to musical instruments and give children high entertainment. Via playing an instrument toy, children can sense and practice what kinds of physical movements (e.g., fingers and/or mouths) they need to play its corresponding actual instrument to produce sounds. In this paper, we present an electronic trombone with three different playing modes for children. The research mission is to offer a simple and joyful approach to engaging children in learning to play a trombone.\",\n \"Title: Investigation Of Learners' Perceptions For Video Summarization And Recommendation\\nYear: 2012\\n\\nAbstract: Recently, multimedia-based learning is widespread in educational settings. A number of studies investigate how to develop effective techniques to manage a huge volume of video sources, such as summarization and recommendation. However, few studies examine how these techniques affect learners' perceptions in multimedia learning systems. This article aims to examine learners' perceptions for summarization and recommendation, with an emphasis on the perspective of prior experience. In this study, we developed a multimedia content summarization and recommendation system, which can automatically extract summaries from raw video sources and recommend suitable video content to learners through emails. The results demonstrate that learners' prior experience and preferences for the presentation of document types affect their perceptions, including the enhancement of interests, the ease of information acquisition and the intention for the further use of the system. Finally, the findings are applied to develop a framework that can support for the design of multimedia learning systems.\",\n \"Title: Investigation of behavior and perception of digital library users: A cognitive style perspective.\\nYear: 2008\\n\\nAbstract: Cognitive style is an influential factor in users' information seeking. The study presented in this paper examines how users' cognitive styles affect their behavior and perception in digital libraries. Fifty participants took part in this study. Two dimensions of cognitive styles were considered: (a) Field Dependence/Independence; (2) Verbalizer/Imager. The results showed that Intermediate users and Verbalizers have not only more positive perception, but they also complete the tasks in effective ways. Implications for the design of personalized digital libraries are also discussed.\",\n \"Title: The comparisons of the influences of prior knowledge on two game-based learning systems\\nYear: 2013\\n\\nAbstract: Game-based learning provides many benefits, such as enhancing the interaction with students and stimulating their learning motivation. Thus, it is popular to learners who have diverse characteristics. To this end, individual differences play an essential role. Among various individual differences, previous studies demonstrated that prior knowledge has great effects on game-based learning. However, such studies mainly considered a single game-based learning system. To address this issue, this study examined how prior knowledge affects students' reactions to two different types of game based learning systems, i.e., the Machinarium and the CSI: Web Adventures. The former delivers procedural knowledge while the latter provides declarative knowledge. The results from this study indicate that prior knowledge has positive impacts in the CSI group while it has negative impacts in the Machinarium group. These findings imply that prior knowledge is useful for the context of declarative knowledge while it may not be helpful for the context of procedural knowledge.\",\n \"Title: A flexible interface design for Web directories to accommodate different cognitive styles\\nYear: 2005\\n\\nAbstract: Search engines are very popular tools for collecting information from distributed resources. They provide not only search facilities, but they also offer directories for users to browse content divided into groups. In this paper, we've adopted an individual differences approach to explore user's attitudes towards various interface features provided by existing Web directories. Among a variety of individual differences, cognitive style is a particularly important characteristic that influences the effectiveness of information seeking. Empirical results indicate that users' cognitive styles influence their reactions to the organization of subject categories, presentation of the results, and screen layout. We developed a set of design guidelines on the basis of these results, and propose a flexible interface that adopts these guidelines to accommodate the preferences of different cognitive style groups.\",\n \"Title: Mining User Preferences Of Multimedia Interfaces With K-Modes\\nYear: 2006\\n\\nAbstract: Interactive multimedia learning systems use sophisticated techniques to present advanced interface features. However, not all users appreciate the strengths of such interface features because of the variations of user backgrounds and skills. In this context, human factors are important issues in deciding user preferences. This study applies a data mining approach to examine user preferences of interface features and to identify the influence of human factors on this issue. K-modes, a data mining technique extensively applied to user modeling, was used to group users' preferences. The results indicated that users' preferences could be divided into eight groups where gender and computer experience significantly influenced the choices made by users.\",\n \"Title: Grouping Users' Communities in an Interactive Web-Based Learning System: A Data Mining Approach\\nYear: 2005\\n\\nAbstract: Discovery of user communities can help designers to develop learning systems that are more suitable to the needs of different individuals. The study builds user communities based on their common navigation behavior using the K-means algorithm. The results indicated that three user groups could be identified and cognitive style is a potential factor that may influence the patterns of each group.\",\n \"Title: Data mining from 1994 to 2004: an application-orientated review\\nYear: 2005\\n\\nAbstract: Data mining, which is also known as knowledge discovery, is one of the most popular topics in information technology. It concerns the process of automatically extracting useful information and has the promise of discovering hidden relationships that exist in large databases. These relationships represent valuable knowledge that is crucial for many applications. This paper presents a review of works on current applications of data mining, which focus on four main application areas, including bioinformatics data, information retrieval, adaptive hypermedia and electronic commerce. How data mining can enhance functions for these four areas is described. The reader of this paper is expected to get an overview of the state-of-the-art research associated with these applications. Furthermore, we identify the limitations of current works and raise several directions for future research.\",\n \"Title: The Clusters of Gaming Behavior in MMORPG: A Case Study in Taiwan\\nYear: 2012\\n\\nAbstract: The aim of this study is to investigate the relationship between online gaming behavior and special underlying social group structures through the use of player information of a certain MMORPG. Data were collected from the logs of an MMORPG in 2011, recording five types of gaming behaviors, player gender, avatar gender and avatar occupation types to investigate whether different avatars affect gaming behavior. This study includes 1,727 male players and 656 female players. The results indicate that the six-cluster model has the goodness of model fit. Different clusters differ in the five different types of gaming behaviors and the choice of avatar gender and occupations. This result shows that different avatar roles will influence players' gaming behaviors.\",\n \"Title: A Personalised Interface for Web Directories Based on Cognitive Styles\\nYear: 2004\\n\\nAbstract: Implementing personalisation in Web Directories depends not only on developing appropriate architectures and equipping Web Directories with adaptation techniques, but also on incorporating human factors considerations at an early design stage. Among a range of human factors this paper explores cognitive styles and their influence on users' preferences. Preferences with respect to the organisation and presentation of the content, and the navigation paths are identified through a small-scale study. The findings are analysed and used to implement a prototype Web Directory Browser, gearing interface features to cognitive style-related preferences.\",\n \"Title: Mining learners' behavior in accessing web-based interface\\nYear: 2007\\n\\nAbstract: Web-based technology has already been adopted as a tool to support teaching and learning in higher education. One criterion affecting the usability of such a technology is the design of web-based interface (WBI) within web-based learning programs. How different users access the WBIs has been investigated by several studies, which mainly analyze the collected data using statistical methods. In this paper, we propose to analyze users' learning behavior using Data Mining (DM) techniques. Findings in our study show that learners with different cognitive styles seem to have various learning preferences, and DM is an efficient tool to analyze the behavior of different cognitive style groups.\",\n \"Title: Evaluating Webct Use In Relation To Students' Attitude And Performance\\nYear: 2008\\n\\nAbstract: Teaching and learning are no longer limited by place or time. Large numbers of educational institutions are now using course management systems such as WebCT. This study examined students' attitude and achievement in relation to their use of WebCT. Statistical data was gathered weekly from the WebCT tracking system log files. Students' achievement was measured by the grades they obtained in the observed module. Students' attitudes towards WebCT were measured using a Likert-type attitude scale. The results of this study indicated that students have a positive attitude towards WebCT. The module leader valued WebCT as a supporting tool to traditional face-to-face learning but not as an effective tool for helping students' learning. The use of WebCT was found to have positive influence on students' achievement as statistically significant relationships were found between students' use of WebCT and their grades.\",\n \"Title: Knowledge exploration with concept association techniques\\nYear: 2010\\n\\nAbstract: Purpose - Exploratory learning is regarded as an important ability for developing knowledge from open environments. During the exploration, learners not only need to acquire new information based on their current interests, but also they need to form new perspectives by incorporating new knowledge into their previous knowledge. This paper seeks to address these issues. Design/methodology/approach - To this end, this paper proposes an approach that includes a concept association bank to recommend related concepts in a domain based on the goal of an exploration. By doing so, learners' knowledge can be expanded beyond their current understanding. An experiment was conducted to investigate how the proposed approach facilitated the learners' exploration. Findings - The results indicated that the concept association bank is a useful mechanism to help learners gain new understanding, including providing exploration directions, reducing complexity and cognitive load, facilitating data- and goal-driven exploration strategies, and commenting on new understanding. The implications of these results are discussed. Originality/value - Current recommendation systems emphasise a data-driven strategy, which seeks isolated pieces of information, instead of suggesting directions related to their exploration goal. The problem with such an approach is that learners' exploration will be limited by their existing knowledge. Thus, this paper presents an approach to support both data- and goal-driven strategies.\",\n \"Title: A cognitive approach to user perception of multimedia quality: An empirical investigation\\nYear: 2006\\n\\nAbstract: Whilst multimedia technology has been one of the main contributing factors behind the Web's success, delivery of personalized multimedia content has been a desire seldom achieved in practice. Moreover, the perspective adopted is rarely viewed from a cognitive styles standpoint, notwithstanding the fact that they have significant effects on users' preferences with respect to the presentation of multimedia content. Indeed, research has thus far neglected to examine the effect of cognitive styles on users' subjective perceptions of multimedia quality. This paper aims to examine the relationships between users' cognitive styles, the multimedia quality of service delivered by the underlying network, and users' quality of perception (understood as both enjoyment and informational assimilation) associated with the viewed multimedia content. Results from the empirical study reported here show that all users, regardless of cognitive style, have higher levels of understanding of informational content in multimedia video clips (represented in our study by excerpts from television programmes) with weak dynamism, but that they enjoy moderately dynamic clips most. Additionally, multimedia content was found to significantly influence users' levels of understanding and enjoyment. Surprisingly, our study highlighted the fact that Bimodal users prefer to draw on visual sources for informational purposes, and that the presence of text in multimedia clips has a detrimental effect on the knowledge acquisition of all three cognitive style groups.\",\n \"Title: A Probabilistic SVM Approach to Annotation of Calcification Mammograms\\nYear: 2010\\n\\nAbstract: Due to the increasing use of digital medical images, a need exists to develop an approach for automatic image annotation, which provides textual labels for images. Thus added labels can be used to access images using textual queries. Automatic image annotation can be separated into two individual tasks: feature extraction and image classification. In this paper, the authors present feature extraction methods for calcification mammograms. The resultant features, based on BI-RADS standards, make annotated image contents represent the correct medical meaning and tag correspondent terms. Furthermore, this paper also proposes a probabilistic SVM approach to image classification. Finally, the experimental results indicate that the probabilistic SVM approach to image annotation can achieve 79.5% in the average accuracy rate.\",\n \"Title: A surrogate competition approach to enhancing game-based learning\\nYear: 2013\\n\\nAbstract: Competition is useful in game-based learning, although it can also generate negative influences. To expand the potential for competition models in game-based learning, this study proposes the notion of surrogate competition, which eliminates direct competition between students. Such surrogates could be employed as buffers so that the competition between students is more relaxed. To explore the possible benefits of a surrogate approach to competition, the My-Pet-My-Arena system has been developed and evaluated. Two empirical studies were conducted to examine the effects of the surrogate competition. The results revealed that surrogate competition enhanced students’ learning achievement as well as increased their motivation. Furthermore, the surrogate competition might also assist students in attributing competitive failures to a lack of effort. Working from the results obtained in these two studies, a general model of surrogate competition is proposed to help designers implement forms of surrogate competition in other systems for game-based learning.\",\n \"Title: Cognitive styles and search engine preferences: Field dependence/independence vs holism/serialism\\nYear: 2010\\n\\nAbstract: Purpose - Cognitive style has been identified to be significantly influential in deciding users' preferences of search engines. In particular, Witkin's field dependence/independence has been widely studied in the area of web searching. It has been suggested that this cognitive style has conceptual links with the holism/serialism. This study aims to investigate the differences between the field dependence/independence and holism/serialism. Design/methodology/approach - An empirical study was conducted with 120 students from a UK university. Riding's cognitive style analysis (CSA) and Ford's study preference questionnaire (SPQ) were used to identify the students' cognitive styles. A questionnaire was designed to identify users' preferences for the design of search engines. Data mining techniques were applied to analyse the data obtained from the empirical study. Findings - The results highlight three findings. First, a fundamental link is confirmed between the two cognitive styles. Second, the relationship between field dependent users and holists is suggested to be more prominent than that of field independent users and serialists. Third, the interface design preferences of field dependent and field independent users can be split more clearly than those of holists and serialists. Originality/value - The contributions of this study include a deeper understanding of the similarities and differences between field dependence/independence and holists/serialists as well as proposing a novel methodology for data analyses.\",\n \"Title: Cognitive styles and web-based instruction: field dependent/independent vs. Holist/Serialist\\nYear: 2009\\n\\nAbstract: Evolutionary multiobjective optimization (EMO) is an active research area in the field of evolutionary computation. EMO algorithms are designed to find a non-dominated solution set that approximates the entire Pareto front of a multiobjective optimization ...\",\n \"Title: Incorporating Usability Criteria into the Development of Animated Hierarchical Maps.\\nYear: 2013\\n\\nAbstract: Nowadays, Web-based learning systems have become popular because they can provide multiple tools, among which hierarchical maps are widely used to support teaching and learning. However, traditional hierarchical maps may let learners easily get lost within large information space. This study proposes an animated hierarchical map to address this problem. To consider the effects of different instructional approaches, we also conducted an empirical study to examine how learners in collaborative learning and those in individual learning reacted differently to the animated hierarchical map and traditional hierarchical map. A questionnaire, which was designed based on Nielsen's heuristics, was applied to examine learners' reactions to the proposed animated hierarchical map. The results show that the proposed animated hierarchical map is generally better than the traditional hierarchical map but the effects on individual learning is limited.\",\n \"Title: When educational agents meet surrogate competition: Impacts of competitive educational agents on students' motivation and performance.\\nYear: 2014\\n\\nAbstract: The design of educational agents increasingly attracts researchers' attention recently. One of major reasons is that educational agents could enhance student learning in various aspects. Although research in this area has mushroomed, such research mainly emphasizes on students in higher education. It is still unclear how educational agents influence young student learning. In addition, competition is a significant element, but fewer studies take competition into account while designing educational agents. Although some studies have indicated that educational agents in competitive environments has positive effects on students' perception and attribution, its impacts on students' motivation and performance are unclear. Thus, the study develops an integrative agent that combines educational and competitive elements for young students, and further examines its influences, in terms of motivation and learning performance. The results revealed that such competitive educational agents could enhance students' motivation and learning performance.\",\n \"Title: Design of adaptive hypermedia learning systems: A cognitive style approach\\nYear: 2011\\n\\nAbstract: In the past decade, a number of adaptive hypermedia learning systems have been developed. However, most of these systems tailor presentation content and navigational support solely according to students' prior knowledge. On the other hand, previous research suggested that cognitive styles significantly affect student learning because they refer to how learners process and organize information. To this end, the study presented in this paper developed an adaptive hypermedia learning system tailored to students' cognitive styles, with an emphasis on Pask's Holist-Serialist dimension. How students react to this adaptive hypermedia learning system, including both learning performance and perceptions, was examined in this study. Forty-four undergraduate and postgraduate students participated in the study. The findings indicated that, in general, adapting to cognitive styles improves student learning. The results also showed that the adaptive hypermedia learning system have more effects on students' perceptions than performance. The implications of these results for the design of adaptive hypermedia learning systems are discussed.\",\n \"Title: The role of human factors in stereotyping behavior and perception of digital library users: a robust clustering approach\\nYear: 2007\\n\\nAbstract: To deliver effective personalization for digital library users, it is necessary to identify which human factors are most relevant in determining the behavior and perception of these users. This paper examines three key human factors: cognitive styles, levels of expertise and gender differences, and utilizes three individual clustering techniques: k-means, hierarchical clustering and fuzzy clustering to understand user behavior and perception. Moreover, robust clustering, capable of correcting the bias of individual clustering techniques, is used to obtain a deeper understanding. The robust clustering approach produced results that highlighted the relevance of cognitive style for user behavior, i.e., cognitive style dominates and justifies each of the robust clusters created. We also found that perception was mainly determined by the level of expertise of a user. We conclude that robust clustering is an effective technique to analyze user behavior and perception.\",\n \"Title: Identifying user preferences with Wrapper-based Decision Trees\\nYear: 2011\\n\\nAbstract: Wrapper feature selection approaches are widely used to choose a small subset of relevant features from a dataset. However, Wrappers suffer from the fact that they only use a single classifier. The downside to this is that each classifier will have its own biases and will therefore select very different features. To overcome the biases of individual classifiers, we propose a new data mining method called Wrapper-based Decision Trees (WDT). The WDT method uses multiple classifiers for selecting relevant features and decision trees to visualize relationships among the selected features. We use the WDT to investigate the influences of the levels of computer experience on users' preferences for the design of search engines. The benefit of using WDT lies within the fact that it can uncover the most accurate set of relevant features to help differentiate the preferences of users with diverse levels of computer experience. The results indicate that the users with varied levels of computer experiences have different preferences regarding the following features: the number of icons, the arrangement of search results, and the presentation of error messages. Such findings can be used to develop personalized search engines to accommodate users' different levels of computer experience.\",\n \"Title: Evaluation of the credibility of internet shopping in the UK\\nYear: 2009\\n\\nAbstract: Purpose - The purpose of this paper is to examine the credibility of internet shopping. Credibility, 20 March 2009 which refers to the believability of information, is an important consideration of internet shopping. Design/methodology/approach - The evaluation is conducted by incorporating Fogg's 10 Stanford Guidelines for Web Credibility into Nielsen's heuristic evaluation. Furthermore, security and individualisation are considered as additional heuristics. Evaluation criteria are developed based on these 12 heuristics. Three UK car insurance web sites are selected for evaluation, including the AA, Norwich Union and Tesco. Findings - The results show that the Norwich Union site seems to be the most credible while the Tesco site appears to be the least credible. The most significant credibility problems are found to lie in the areas of \\\"trustworthiness\\\", \\\"expertise\\\" and \\\"real-world feel\\\". In other words, these three areas are key issues for future improvement of these sites. Originality/value - This paper contributes to the literature by providing a set of credibility design guidelines, which can be used to support the improved development of future internet shopping designs, especially car insurance web sites.\",\n \"Title: Game Poetry: Promotion of Poetry through a Game-Based Learning System\\nYear: 2013\\n\\nAbstract: Poetry is a creative language which enhances imagination and self-reflection. However, the ambiguity of poetry increased the difficulty in interpretation, which builds students' negative perception because they consider that reading poetry is joyless and hard-understanding. Furthermore, the negative perception demotivates most students to read poetry. Therefore, to motivate students to read poetry, there is a need to improve the negative perception of poetry. Meanwhile, previous research shows games are able to avoid ambiguity by providing clear feedback and guidance. Besides, many studies also suggest game-based learning positively improves students' learning perception and motivation. To the end, this study develops a game-based learning system to motivate students to read poetry. The results show this game-based learning system significantly improves students' interpretation and perception. However, the long-term negative perception and the lack of self-efficacy affect students' motivation to read poetry after gaming.\",\n \"Title: Experiences with Developing a User-Centered Digital Library\\nYear: 2010\\n\\nAbstract: The purpose of this study was to find out whether users should be involved in the design stage of a Digital Library DL. Hence an experiment has been undertaken to find out the level of user satisfaction with a user-centred DL UDL compared to a non-user-centred DL NDL. In this experiment, the UDL prototype has been compared with the NDL prototype. The two prototypes were then evaluated by separate users with various Information Technology IT backgrounds. Results show that users' task performance was better in the UDL version of the prototype.\",\n \"Title: Cognitive Modeling Of Student Learning In Web-Based Instructional Programs\\nYear: 2004\\n\\nAbstract: There has been tremendous growth in Web-based instruction over the past few years. Because the user group of Web-based instruction includes learners from heterogeneous backgrounds, individual differences become an important issue in the development of Web-based instructional programs. Among a variety of individual differences, cognitive style is a particularly important characteristic. This study aims to determine the relationships between learners' cognitive styles and their perceptions and attitudes toward the features of a Web-based instructional program. The results indicate that cognitive styles influence students' reactions to nonlinear interaction, independent learning, and navigation tools and the difficulties and problems that they encounter. The findings are applied to develop a learning model that can support the design of Web-based instructional programs.\",\n \"Title: A Cognitive Apprenticeship Approach to Facilitating Web-based Collaborative Problem Solving.\\nYear: 2012\\n\\nAbstract: Enhancing students' problem-solving abilities has been recognized as an important and challenging issue for technology-enhanced learning. Thus, previous research has attempted to address this issue by developing various mechanisms, among which a cognitive apprenticeship model can particularly enhance students' abilities. However, it is not clear whether such a mechanism is suitable for every learner. Thus, this study examines the effects of human factors on problem-solving effectiveness in the cognitive apprenticeship model. Among various human factors, this study focuses on cognitive styles, with an emphasis on Witken's Field Dependence. The results indicate that Field Dependent learners can get great benefits from the cognitive apprenticeship model via collaborative learning. Implications for how to accommodate the needs of different cognitive style groups are discussed.\",\n \"Title: The influences of a two-tier test strategy on student learning: A lag sequential analysis approach.\\nYear: 2015\\n\\nAbstract: Recently, programming skills have become a core competence. Many teaching strategies were developed to improve programming skills. Among them, online tests were widely applied to enhance students learning. Nonetheless, they may not be able to engage students in deep thinking and reflections. Thus, a two-tier test strategy was proposed to address this issue. However, previous research mainly focused on investigating the effectiveness of the two-tier test strategy but there is a lack of studies that investigate why the two-tier test approach is effective. To this end, we developed an online test, where the two-tier test strategy was implemented. Additionally, an empirical study was conducted to explore the influences of the two-tier test approach on students' learning performance and behavior patterns. Pre-test and post-test scores were applied to assess students' learning performance while a lag sequential analysis was used to analyze behavior patterns. Regarding learning performance, the proposed two-tier test can improve students' programming skills. Regarding behavior patterns, the two-tier test approach facilitates students to develop a learning by reviewing strategy, which is useful to improve their programming skills.\",\n \"Title: The Development Of A Robot-Based Learning Companion: A User-Centered Design Approach\\nYear: 2015\\n\\nAbstract: A computer-vision-based method is widely employed to support the development of a variety of applications. In this vein, this study uses a computer-vision-based method to develop a playful learning system, which is a robot-based learning companion named RobotTell. Unlike existing playful learning systems, a user-centered design (UCD) approach is applied to assess the interface design of the RobotTell System so that the human-robot interaction can be enhanced during the design process. More specifically, the interface of the RobotTell system is assessed based on Nielsen's heuristics. Based on the results of Nielsen's heuristic evaluation, three versions of the RobotTell system were produced. The implications for the design of these three versions of the RobotTell system are discussed.\",\n \"Title: A Study Of Users' Reactions To A Mixed Online Discussion Model: A Lag Sequential Analysis Approach\\nYear: 2015\\n\\nAbstract: There is an increasing growth of Social Network Services (SNSs). A variety of SNSs are applied in online interpersonal platforms. Among them, asynchronous and synchronous discussions are widely examined. However, there is a lack of research into the effects of integrated discussion services that combines asynchronous and synchronous discussions. Thus, the study investigates users' performances and behavior patterns in a mixed discussion model that integrates asynchronous and synchronous discussions with a lag sequential analysis. The results showed that most groups chose to adopt the mixed discussion model (i.e., using both synchronous and asynchronous discussions), and only one group totally adopted the asynchronous discussion model. The study further analyzed the learners' learning effectiveness and behavioral patterns, and the results indicated that the groups using the mixed model had a positive performance to a certain extent in terms of learning effectiveness and knowledge construction. In addition, users with the mixed discussion model demonstrated diverse behaviors, which were more complex than that of those with a single-way discussion model. Furthermore, regarding the groups using \\\"balanced synchronous and asynchronous discussions\\\" and the groups \\\"mainly using synchronous discussion supplemented by less asynchronous discussion\\\" in the mixed model (simultaneously using synchronous and asynchronous discussions), those users who mainly used synchronous discussion supplemented by asynchronous discussion had more diverse behaviors of knowledge construction.\",\n \"Title: The Effects of Game-Based Learning on Mathematical Confidence and Performance: High Ability vs. Low Ability.\\nYear: 2014\\n\\nAbstract: Many students possess low confidence toward learning mathematics, which, in turn, may lead them to give up pursuing more mathematics knowledge. Recently, game-based learning (GBL) is regarded as a potential means in improving students' confidence. Thus, this study tried to promote students' confidence toward mathematics by using GBL. In addition, this study also investigated whether GBL is beneficial to all students with various abilities. The results demonstrated that this approach yielded better outcomes than the paper-based setting in both students' confidence and students' performance. The students with high and low levels of ability in the GBL group gained a significant improvement on the confidence toward mathematics. Additionally, low-ability students in the GBL group attained better mathematics performance than those in the paper-based setting.\",\n \"Title: Individual Differences and Joyful Assessment-Based Learning\\nYear: 2014\\n\\nAbstract: E-assessment learning, which combines the advantages of e-assessment and learning, was proposed in educational settings. However, it still makes most of learners feel anxious. Thus, this study develops a Joyful Assessment-based Learning System (JALS), which incorporates game-based learning into e-assessment learning. In order to get a complete understanding of how individual differences affect learners' reactions, an empirical study was conducted. Among various individual differences, cognitive styles and prior knowledge were considered as targets to investigate learners' reactions to the JALS. Our results suggest that FD learners preferred to use the game-based learning tool and demonstrated good performance. Conversely, the different level of prior knowledge may affect FI learners' learning preferences and learning performance.\",\n \"Title: Cognitive styles and the use of electronic journals in a mobile context.\\nYear: 2014\\n\\nAbstract: Purpose - The purpose of this paper is to examine how different cognitive style groups use electronic journals via mobile devices, with an emphasis on the use of search mechanisms, the approaches to judge the relevance and to refine the results, the use of horizontal screens, and on the influence of negative emotions. Design/methodology/approach - Participants' cognitive styles were identified as Holists and Serialists with Ford's Study Preference Questionnaire. The behaviour of the participants interacting with the ScienceDirect was observed and was analysed with both traditional statistics and data mining. Findings - Holists prefer to use multiple searches when undertaking the Expert Search only, which is a complex search mechanism. Conversely, Serialists have a tendency to use multiple searches, irrespective of the complexity of search mechanisms. Additionally, the Expert Search may be helpful for Serialists while the Basic Search seems beneficial to Holists. Regarding the judgement of relevance, Holists tend to use a variety of approaches while Serialists prefer to use a single approach. In general, the major results of this study confirm those of Vilar and Z. umer (2008). Originality/value - The novelty of this study covers two aspects. One is to examine the behaviour of Holists and Serialists in the mobile context of electronic journals and the other is to analyse data with an integrative approach that combines both traditional statistics and data mining. The former contributes to the understandings of how to design mobile electronic journals that can accommodate the needs of different cognitive style groups. The latter can not only strengthen the reliability of the results obtained from traditional statistics, but also discover findings neglected by traditional statistics.\",\n \"Title: The effects of gender differences on the use of annotatable multimedia e-readers\\nYear: 2015\\n\\nAbstract: AbstractPurpose - This paper aims to examine how gender differences influence students' reactions to the use of the annotatable multimedia e-reader AME. To reach this aim, we develop an AME where various annotation tools are provided to help students learn English in-class and after-class. Design/methodology/approach - An empirical study was conducted with 63 fifth-grade students from an elementary school. A pre-test and post-test were used to identify their prior knowledge and learning achievement, respectively. A questionnaire was applied to identify participants' perceptions towards the AME. Findings - The results show that students' post-test scores are significantly related to after-class behaviour, instead of in-class behaviour. Females prefer to use the text annotation and teachers' voice, but it is voice annotation that is beneficial to improve their learning achievement. Conversely, males prefer to use the text-to-speech only, but it is text annotation that is helpful to improve their learning achievement. Additionally, the ease of use affects males' intention to use the AME to learn English after-class while it has no effects on females. Originality/value - This study not only shows the importance of gender differences but also demonstrates the essence of after-class learning behaviour. More importantly, a framework is proposed to support designers to develop e-readers that can accommodate the preferences of females and males.\",\n \"Title: The influences of an interactive group-based videogame: Cognitive styles vs. prior ability\\nYear: 2015\\n\\nAbstract: This study aims to investigate how cognitive styles and the levels of prior abilities influence frustration tolerance and learning achievement in an interactive group-based videogame named “Multiple-Choice Practice Island.” In the aspect of cognitive styles, differences between field dependent and field independent students were explored. The results show that field independent students make better learning achievement but field dependent students demonstrate higher frustration tolerance. Besides, the low-ability students significantly made more improvement than the high-ability students. The low-ability students also demonstrate higher frustration tolerance in this study. The findings can guide designers how to develop adaptive group-based videogames that match the needs of each individual.\",\n \"Title: Investigation Of Multiple Human Factors In Personalized Learning\\nYear: 2016\\n\\nAbstract: In the past decade, a number of personalized learning systems have been developed and they mainly focus on learners' prior knowledge. On the other hand, previous research suggested that gender differences and cognitive styles have great effects on student learning. To this end, this study examines how human factors, especially gender differences and cognitive styles, affect learners' reactions to a personalized and non-personalized learning systems based on learners' prior knowledge. Forty-four university students participated in this study. The results show that females and Serialists show positive reactions to the personalized learning system, while males and Holists demonstrate similar reactions to the personalized learning system and the non-personalized learning system. The implications of these results for the design of personalized learning systems are discussed.\",\n \"Title: The impact of cognitive styles on perceptual distributed multimedia quality\\nYear: 2003\\n\\nAbstract: Multimedia technology has been widely used in web-based instruction, but previous studies have indicated that individual differences, especially cognitive styles, have significant effects on users' preferences with respect to presentation of multimedia content. However, such research has thus far neglected to examine the effect of cognitive styles on users' subjective perceptions of multimedia quality. This study aims to examine the relationships among users' cognitive styles, the multimedia Quality of Service (QoS) delivered by the underlying network, and Quality of Perception (QoP), which encompasses user levels of enjoyment and understanding of the informational content provided by multimedia material. Accordingly, 132 users took part in an experiment in which they were shown multimedia video clips presented with different values of two QoS parameters (frame rate and colour depth). Results show that, whilst the two QoS parameters do not impact user QoP multimedia content and dynamism levels significantly influence the user understanding and enjoyment component of QoP.\",\n \"Title: Matching/mismatching revisited: an empirical study of learning and teaching styles\\nYear: 2001\\n\\nAbstract: This paper presents results of a research project that explored the relationship between matching and mismatching instructional presentation style (breadth-first and depth-first) with students' cognitive style (field-dependence/-independence) in a computer-based learning environment, 73 postgraduate students were asked to create Web pages using HTML, using instructional materials that were either matched or mismatched with their cognitive styles. Significant differences in performance on a multiple choice test of conceptual knowledge were found for students learning in matched and mismatched conditions. Performance in matched conditions was significantly superior to that in mismatched conditions. However, significant effects were found for gender, matching mainly affecting male students. Performance on a practical test of Web page creation was not linked to matching or mismatching, but was linked to an interaction between gender and instructional presentation style, The findings provide support for the notion that matching and mismatching can have significant effects on learning outcomes. The paper concludes with suggestions for further research.\",\n \"Title: An Investigation of a Game-Based Anti-Drug System: Addictive Learners vs. Non-addictive Learners\\nYear: 2015\\n\\nAbstract: The spread of drugs causes serious problems recently. To enhance people awareness about drugs, several approaches were adopted to deliver anti-drug materials. However, limitations were observed in these approaches. Thus, this study develops a game-based anti-drug system (GADS), which incorporates game-based learning into anti-drug materials. In order to get a complete understanding of whether the GADS is suitable to both non-addictive users and addictive users, an empirical study was conducted. Drug experience was considered as a target to investigate learners' reactions to the GADS. The results indicated that non-addictive learners had better game performance and learning performance than addictive learners because addictive learners' working memory was negatively affected by drugs.\",\n \"Title: An Investigation of a Two-Tier Test Strategy in a University Calculus Course: Causes versus Consequences.\\nYear: 2016\\n\\nAbstract: Online tests have been identified as a core learning activity in higher education. Unlike conventional online tests, which cannot completely reflect students' learning status, two-tier tests not only consider students' answers, but also take into account reasons for their answers. Due to such significance, research into the two-tier tests had mushroomed but few studies examined why the two-tier te...\",\n \"Title: An Investigation Of A Joyful Peer Response System: High Ability Vs. Low Ability\\nYear: 2016\\n\\nAbstract: Existing peer response approaches majorly take place in traditional educational contexts, where students may not be highly motivated. On the other hand, game-based learning takes a joyful way to enhance students' motivation. Accordingly, we propose a joyful peer response (JPR), where game-based learning and peer response are integrated together. Furthermore, two empirical studies were conducted to investigate how high-and low-ability students perform in the JPR and how their perceptions are associated with their performance. Regarding the former, the findings suggest that the JPR is beneficial to high-and low-ability students. Regarding the latter, the findings imply that the writing performance of the high-ability students is majorly related to their perceptions of game elements. On the other hand, the writing performance of the low-ability students is related to their perceptions of both game elements and peer response.\",\n \"Title: Incorporating customization and personalization into game-based learning: A cognitive style perspective.\\nYear: 2016\\n\\nAbstract: In recent years, customization and personalization were widely applied to accommodate the needs of different cognitive style groups. Such two approaches have different advantages and disadvantages but there is a lack of studies to compare these two approaches from the perspective of game-based learning, which is currently popular in educational settings. To this end, we developed a customized game-based learning system and a personalized game-based learning system and conducted two empirical studies to examine how cognitive styles affected learner's reactions to these two game-based learning systems. The results from the customized game-based learning system showed that Holists might not always favor to listen to music because they frequently switched on/off music. On the other hand, Serialists did not prefer to use hints. In addition, learners with the customized game-based learning system had more positive perceptions while learners with the personalized game-based learning system had more negative perceptions though both systems were useful to enhance learners' learning performance, regardless of their cognitive styles.\",\n \"Title: The effects of task activities and gaming scales on eye reading and visual search performance.\\nYear: 2017\\n\\nAbstract: The effects of video gaming have been explored extensively in many learning-related fields. However, limited research has investigated whether the frequencies of exposing to gaming environments influence students' Chinese reading comprehension, arithmetic word problem test, spatial recognition, and graphical patterns recognition. To address this issue, this study adopted eye-tracking technology to explore learners' reading and visual search responses on various reading materials. Task effects and correlation analyses were conducted on a sample of 19 sophomores with varying gaming frequencies, eye reading fixations, and regressions. The results revealed that practice makes users have fewer fixation frequencies on a similar task. Meanwhile, the results demonstrated a positive gaming frequency correlation on eye fixations and regressions for Chinese reading comprehension and arithmetic word problem test, and a negative gaming frequency correlation for spatial and graphical patterns recognition tasks.\",\n \"Title: Exploring The Interactive Patterns Of Concept Map-Based Online Discussion: A Sequential Analysis Of Users' Operations, Cognitive Processing, And Knowledge Construction\\nYear: 2016\\n\\nAbstract: Concept maps can be used as a cognitive tool to assist learners' knowledge construction. However, in a concept map-based online discussion environment, studies that take into consideration learners' manipulative actions of composing concept maps, cognitive process among learners' discussion, and social knowledge construction at the same time are still lacking. Therefore, this study employed lag sequential analysis to empirically explore learners' behavioral patterns in a concept map-based online discussion environment. We explored learners' operations in the concept map tool, and their discussion behavioral patterns of cognitive processing phases and social knowledge construction. The results showed that it was easier for learners to define a node than to define a relationship when they used the concept map. Additionally, the concept map can enhance higher level cognitive processing phases and lead learners to return to the main discussions from off-topic discussions significantly.\",\n \"Title: The Influences of the 2D Image-Based Augmented Reality and Virtual Reality on Student Learning.\\nYear: 2017\\n\\nAbstract: Virtual reality (VR) learning environments can provide students with concepts of the simulated phenomena, but users are not allowed to interact with real elements. Conversely, augmented reality (AR) learning environments blend real-world environments so AR could enhance the effects of computer simulation and promote students' realistic experience. However, AR-based learning environments had a lot of dynamic real objects which may increase learners' mental effort. Moreover, paucity of research compared AR with VR and other mature technologies. Thus, the aim of this study is to compare the influence of the 2D imagebased VR and AR in an inquiry-based astronomy course. The findings of this study suggested that the real objects presented in the AR system could reduce the mental load because students could take the real objects of the AR system as the reference objects of the movement of the moon. Furthermore, the sense of the immediacy is increased due to the fact that peers appear on the AR system. Accordingly, the real objects and the sense of the immediacy not only enhance the learning motivations, but also encourage the students to keep conducting the tasks.\",\n \"Title: Developing a multimedia encyclopedia for little people and their families.\\nYear: 2017\\n\\nAbstract: Among a variety of disabilities, persons with achondroplasia, i.e., little people, have relatively minor symptoms though they have small bodies. Thus, they can learn how to lead an independent life. This paper presents a study on the development of a multimedia encyclopedia for little people and their families. Due to their different roles, this study examines how little people and their families reacted to the multimedia encyclopedia. In total, 18 little people and 15 family members participated in this study. Their interaction behavior was recorded in log files and their perceptions were identified by a questionnaire. The results indicated that little people preferred to use a search mechanism while their family members favored to use pop-up windows. Either little people or their families thought that the multimedia encyclopedia provided some new information that they did not know. However, family members’ attention, confidence and satisfaction with the multimedia encyclopedia were higher than those of the little people. This study can contribute to the knowledge of how to develop individualized multimedia encyclopedias that can accommodate the different preferences of little people and their family members.\",\n \"Title: Students' Reactions to Different Levels of Game Scenarios: A Cognitive Style Approach.\\nYear: 2017\\n\\nAbstract: Game-based learning comprises of a set of concrete scenarios, which can be presented with various presentation modalities (e.g., text, text with graphic, and context). Such presentation modalities disseminate information in different ways. On the other hand, cognitive styles affect how people process information. Accordingly, the study presented in this paper investigates the influences of cognitive styles on students' performance within different levels of game-based scenarios. More specifically, this study examines how students with different cognitive styles (i.e., Holist/Serialist) react to three presentation modalities (i.e., text, text with graphic, and context) in game-based scenarios. The results of an empirical study with 96 students revealed that all students made improvement, regardless of any presentation modalities. However, Holists significantly performed better than Serialists in the context version. According to the results, further suggestions for system design are discussed.\",\n \"Title: An investigation of the development of an animated E-book: A gender difference approach.\\nYear: 2018\\n\\nAbstract: Two empirical studies, i.e., preliminary study and main study, were included in this research. In the preliminary study, we developed a prototyped animated e-book system (PAES) that incorporated animation films into an e-book and examined how females and males reacted to design features provided by the PAES. The results from the preliminary study showed that gender differences significantly affected students' reactions to the PAES. Accordingly, the main study attempted to develop an animated e-book system (AES) that could accommodate gender differences based on the results from the preliminary study. The results from our main study showed that gender differences had no significant effects on students' learning performance and learning perceptions though females and males demonstrated different learning behavior. In other words, both females and males could equally benefit from the AES. Briefly, the major contribution of this research was to provide guidance on how to develop an animated e-book from a gender difference perspective. Such guidance can help designers to develop a versatile animated e-book that can support the needs and preferences of both females and males.\",\n \"Title: The influences of self-regulated learning support and prior knowledge on improving learning performance.\\nYear: 2018\\n\\nAbstract: Self-regulated learning (SRL) is helpful to students. On the other hand, prior knowledge has great effects on students' self-regulation and learning performance. To this end, this study aimed to examine how high prior knowledge students (HPKs) and low prior knowledge students (LPKs) behaved differently when interacting with a SRL environment. To achieve this aim, we proposed a self-regulated learning support system (SRLSS) for a mathematical course. The results showed that the gap of learning performance between the HPKs and LPKs was removed after a long-term learning process. Moreover, the LPKs and HPKs behaved similarly in the forethought and self-reflection phases but some behavior differences were found in the performance phase, where the LPKs relied on the notes and sought support the dashboard and quiz records while the HPKs did not demonstrate such a tendency. Our results' theoretical and methodological implications and possible applications for further research are also discussed.\",\n \"Title: Incorporating digital games into anti-drug material: non-drug-addicted learners vs. drug-addicted learners\\nYear: 2020\\n\\nAbstract: Drug abuse causes serious problems around the world. Thus, anti-drug material has been increasingly developed and used. However, conventional anti-drug material has until today shown insufficient effects on preventing drug abuse, especially for drug-addicted users. To this end, this study proposes a game-based anti-drug material and examines how drug-addicted people and non-drug-addicted people interact with the game-based anti-drug material. A lag sequential analysis approach was applied to analyze their different behavior patterns. The results revealed that the experience of drug addiction influenced the way individuals access drug information, complete tasks and reach goals during their gaming process. Based on the findings, some suggestions are proposed for the development of adaptive game-based anti-drug material that can accommodate the needs of drug-addicted and non-drug-addicted people to achieve the goal of universal access.\",\n \"Title: The Effects of Cognitive Styles on the Use of Hints in Academic English: A Learning Analytics Approach.\\nYear: 2017\\n\\nAbstract: Various hints have been proposed to scaffold students. Such hints can be broadly divided into direct hints and indirect hints. On the other hand, individual differences existed among learners. In particular, cognitive styles greatly affect student learning. However, there is a lack of studies to investigate how cognitive styles affect students' reactions to the use of hints in the context of academic English. To address this issue, this study aimed to examine the effects of cognitive styles on the use of hints in such a context. To obtain deep understandings, a learning-analytics approach was applied in this study, including quantitative measurement, qualitative evaluation and lag sequential analyses. The results from the quantitative measurement indicated that the frequencies of using Chinese hints and synonym hints were significantly associated with Serialists' task scores while such significant correlations were not found for Holists. The results from the qualitative evaluation showed that Holists favored to use synonym hints while Serialists preferred to use Chinese hints. The results from the lag sequential analyses suggested that Holists demonstrated iterative behavior while Serialists showed sequential behavior. In summary, cognitive styles have considerable influences on students' learning patterns in the context of Academic English.\",\n \"Title: An Investigation of the Effects of Scaffolding Task-Based English Logic Learning: A Cognitive Style Approach\\nYear: 2022\\n\\nAbstract: Task-based learning can be used to promote language learning. On the other hand, scaffolding instruction can help students without sufficient abilities to undertake tasks. Accordingly, we developed the Scaffolding Task-based English Logic Learning (S-TELL) by incorporating scaffolding instruction into Task-based Learning. The S-TELL was applied to facilitate research students to learn how to present clear logical relationships in academic writing. Additionally, an empirical study was conducted to investigate how different cognitive style groups (i.e., Serialists and Holists) reacted to the S-TELL, including learning performance, learning perception and learning behavior. Regarding learning performance, the findings showed that either Holists or Serialists made significant improvement via interacting with the S-TELL. Moreover, Holists significantly obtained higher post-scores than Serialists though they had similar task scores. Regarding learning perception, few perception differences existed between Serialists and Holists. Regarding learning behavior, Holists demonstrated more behavior sequences than Serialists. This was due to the fact that Holists often checked the current status and used the English hint while Serialists did not show such behavior. According to these findings, we proposed a framework, which can be employed to personalize the S-TELL in our future work so that the preferences of Holists and Serialists can be accommodated.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":81,"string":"81"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13571,"string":"13,571"}}},{"rowIdx":13572,"cells":{"question_id":{"kind":"string","value":"Rustam Stolkin"},"question":{"kind":"string","value":"Generate an abstract for the title \"Deformable Dictionary Learning for SAR Image Change Detection.\" using the following items: 1. Deformable dictionary learning \n2. Multitemporal image pairs \n3. Change detection \n4. Ground-truthed training images \n5. Shortage of training data"},"answer":{"kind":"string","value":"This paper proposes a novel method based on deformable dictionary learning for detecting the regions of change between multitemporal image pairs. We build on our previous work, which constructed a pair of dictionaries. The main shortcoming of this method was its dependence on a large amount of training data. In practice, there is often a shortage of ground-truthed training images, which limits the..."},"correct_answer":{"kind":"string","value":"This paper proposes a novel method based on deformable dictionary learning for detecting the regions of change between multitemporal image pairs. We build on our previous work, which constructed a pair of dictionaries. The main shortcoming of this method was its dependence on a large amount of training data. In practice, there is often a shortage of ground-truthed training images, which limits the..."},"documents":{"kind":"list like","value":["Title: Predicting the unobservable Visual 3D tracking with a probabilistic motion model.\nYear: 2011\n\nAbstract: Visual tracking of an object can provide a powerful source of feedback information during complex robotic manipulation operations, especially those in which there may be uncertainty about which new object pose may result from a planned manipulative action. At the same time, robotic manipulation can provide a challenging environment for visual tracking, with occlusions of the object by other objects or by the robot itself, and sudden changes in object pose that may be accompanied by motion blur. Recursive filtering techniques use motion models for predictor-corrector tracking, but the simple models typically used often fail to adequately predict the complex motions of manipulated objects. We show how statistical machine learning techniques can be used to train sophisticated motion predictors, which incorporate additional information by being conditioned on the planned manipulative action being executed. We then show how these learned predictors can be used to propagate the particles of a particle filter from one predictor-corrector step to the next, enabling a visual tracking algorithm to maintain plausible hypotheses about the location of an object, even during severe occlusion and other difficult conditions. We demonstrate the approach in the context of robotic push manipulation, where a 5-axis robot arm equipped with a rigid finger applies a series of pushes to an object, while it is tracked by a vision algorithm using a single camera.","Title: Learning to predict how rigid objects behave under simple manipulation.\nYear: 2011\n\nAbstract: An important problem in robotic manipulation is the ability to predict how objects behave under manipulative actions. This ability is necessary to allow planning of object manipulations. Physics simulators can be used to do this, but they model many kinds of object interaction poorly. An alternative is to learn a motion model for objects by interacting with them. In this paper we address the problem of learning to predict the interactions of rigid bodies in a probabilistic framework, and demonstrate the results in the domain of robotic push manipulation. A robot arm applies random pushes to various objects and observes the resulting motion with a vision system. The relationship between push actions and object motions is learned, and enables the robot to predict the motions that will result from new pushes. The learning does not make explicit use of physics knowledge, or any pre-coded physical constraints, nor is it even restricted to domains which obey any particular rules of physics. We use regression to learn efficiently how to predict the gross motion of a particular object. We further show how different density functions can encode different kinds of information about the behaviour of interacting objects. By combining these as a product of densities, we show how learned predictors can cope with a degree of generalisation to previously unencountered object shapes, subjected to previously unencountered push directions. Performance is evaluated through a combination of virtual experiments in a physics simulator, and real experiments with a 5-axis arm equipped with a simple, rigid finger.","Title: Optimal AUV path planning for extended missions in complex, fast-flowing estuarine environments\nYear: 2007\n\nAbstract: This paper addresses the problems of automatically planning Autonomous Underwater Vehicle (AUV) paths which best exploit complex current data, from computational estuarine model forecasts, while also avoiding obstacles. In particular we examine the possibilities for a novel type of AUV mission deployment in fast flowing tidal river regions which experience bi-directional current flow. These environments are interesting in that, by choosing an appropriate path in space and time, an AUV may both bypass adverse currents which are too fast to be overcome by the vehicle's motors and also exploit favorable currents to achieve far greater speeds than the motors could otherwise provide, while substantially saving energy. The AUV can \"ride\" currents both up and down the river, enabling extended monitoring of otherwise energy-exhausting, fast flow environments. The paper discusses suitable path parameterizations, cost functions and optimization techniques which enable optimal AUV paths to be efficiently generated. These paths take maximum advantage of the river currents in order to minimize energy expenditure, journey time and other cost parameters. The resulting path planner can automatically suggest useful alternative mission start and end times and locations to those specified by the user. Examples are presented for navigation in a simple simulation of the fast flowing Hudson River waters around Manhattan.","Title: Prediction learning in robotic pushing manipulation\nYear: 2009\n\nAbstract: This paper addresses the problem of learning about the interactions of rigid bodies. A probabilistic frame- work is presented for predicting the motion of one rigid body following contact with another. We describe an algorithm for learning these predictions from observations, which does not make use of physics and is not restricted to domains with particular physics. We demonstrate the method in a scenario where a robot arm applies pushes to objects. The probabilistic nature of the algorithm enables it to generalize from learned examples, to successfully predict the resulting object motion for previously unseen object poses, push directions and new objects with novel shape. We evaluate the method with empirical experiments in a physics simulator. I. INTRODUCTION In early childhood, humans and animals learn models for predicting the interactions with the environment (1). It seems unlikely that these models comprise an explicit encoding of Newtonian physics, and so must instead rely on a learned relationship between observed actions and their outcomes. This paper addresses the problem of learning to predict the motion of one body, which results from a forced interaction with another. We have chosen to investigate this problem in the context of robotic \"poking\" or \"pushing\" operations, be- cause this includes a large number of unstable manipulations and hence provides interesting situations. However the work is potentially more general. An algorithm is presented which learns to predict the motions of a rigid object that will result from an applied robotic pushing action. The algorithm does not rely on any understanding or encoding of Newtonian mechanics, but can be trained in simple online experiments in which a robot arm applies random pushes to objects of interest and extracts the resulting motions using a vision system. Properties of objects, and their interactions, are learned as distributions. Pushing operations are encountered frequently in robotics, but have received relatively little attention in the research community. They are important in that robotic grasping frequently involves a pushing phase, when one finger or jaw of a gripper contacts the workpiece before another. Furthermore, pushing may often be preferable to pick and place type operations if the robot lacks the size or strength necessary to lift an object. (2) was the first to identify pushing operations as funda- mental to manipulation, especially grasping. Mason develops a detailed analysis of the mechanics of pushed, sliding objects and determines conditions required for various 2D motions of a pushed object. (3) attempts to put quantitative bounds on the rate at which these predicted motions occur. (4) developed a method for finding the set of all possible","Title: Video with Ground-Truth for Validation of Visual Registration, Tracking and Navigation Algorithms\nYear: 2005\n\nAbstract: A fundamental task in computer vision is that of determining the position and orientation of a moving camera relative to an observed object or scene. Many such visual tracking algorithms have been proposed in the computer vision, artificial intelligence and robotics literature over the past 30 years. Predominantly, these remain unvalidated since the ground-truth camera positions and orientations at each frame in a video sequence are not available for comparison with the outputs of the proposed vision systems. A method is presented for generating real visual test data with complete underlying ground-truth. The method enables the production of long video sequences, filmed along complicated six degree of freedom trajectories, featuring a variety of objects, in a variety of different visibility conditions, for which complete ground-truth data is known including the camera position and orientation at every image frame, intrinsic camera calibration data, a lens distortion model and models of the viewed objects.","Title: An EM/E-MRF algorithm for adaptive model based tracking in extremely poor visibility\nYear: 2008\n\nAbstract: This paper addresses the problems of visual tracking in conditions of extremely poor visibility. The human visual system can often correctly interpret images that are of such poor quality that they contain insufficient explicit information to do so. We assert that such systems must therefore make use of prior knowledge in several forms. A tracking algorithm is presented which combines observed data (the current image) with predicted data derived from prior knowledge of the object being viewed and an estimate of the camera's motion. During image segmentation, a predicted image is used to estimate class conditional distribution models and an Extended-Markov Random Field technique is used to combine observed image data with expectations of that data within a probabilistic framework. Interpretations of scene content and camera position are then mutually improved using Expectation Maximisation. Models of background and tracked object are continually relearned and adapt iteratively with each new image frame. The algorithm is tested using real video sequences, filmed in poor visibility conditions with complete pre-measured ground-truth data.","Title: A novel selection evolutionary strategy for constrained optimization\nYear: 2013\n\nAbstract: The existence of infeasible solutions makes it very difficult to handle constrained optimization problems (COPs) in a way that ensures efficient, optimal and constraint-satisfying convergence. Although further optimization from feasible solutions will typically lead in a direction that generates further feasible solutions, certain infeasible solutions can also provide useful information about the optimal direction of improvement for the objective function. How well an algorithm makes use of these two solutions determines its performance on COPs. This paper proposes a novel selection evolutionary strategy (NSES) for constrained optimization. A self-adaptive selection method is introduced to exploit both informative infeasible and feasible solutions from a perspective of combining feasibility with multi-objective problem (MOP) techniques. Since the global optimal solution of a COP is a feasible non-dominated solution, both non-dominated solutions with low constraint violation and feasible ones with low objective values are beneficial to an evolution process. Thus, the exploration and exploitation of both of these two kinds of solutions are preferred during the selection procedure. Several theorems and properties are given to prove the above assertion. Furthermore, the performance of our method is evaluated using 22 well-known benchmark functions. Experimental results show that the proposed method outperforms state-of-the-art algorithms in terms of the speed of finding feasible solutions and the stability of converging to global optimal solutions. In particular, when dealing with problems that have zero feasibility ratios and more than one active constraint, our method provides feasible solutions within fewer fitness evaluations (FES) and converges to the optimal solutions more reliably than other popular methods from the literature.","Title: Change detection in SAR images by artificial immune multi-objective clustering.\nYear: 2014\n\nAbstract: This paper addresses the problem of unsupervised change detection in Synthetic Aperture Radar (SAR) images. Previous approaches have used evolutionary clustering optimization methods, which can suffer from reduced accuracy, because they often use only a single objective function and can easily become trapped at locally optimal values. To overcome these difficulties, we propose a new approach which combines the artificial immune system (AIS) theory with a multi-objective optimization algorithm. First, the self-adaptive artificial immune multi-objective algorithm is adopted to pre-sort the difference image. During this procedure, the difference image is categorized into three classes – changed class, unchanged class and uncertain samples. Second, based on wavelet decomposition to extract features from the difference image, the immune clonal multi-objective clustering algorithm is used to search for the optimal clustering centers of uncertain samples, labeling them as changed or unchanged. Experimental comparisons with four state-of-the-art approaches show that the proposed algorithm can obtain a higher accuracy, is more robust to noise, and finds solutions which are more globally optimal. Additionally, the proposed algorithm can improve the local search ability for the optimal solutions and produces better cluster centers.","Title: A Three-Component Fisher-Based Feature Weighting Method for Supervised PolSAR Image Classification\nYear: 2015\n\nAbstract: This letter presents a feature weighting method for polarimetric synthetic aperture radar (PolSAR) image classification. Appropriate feature weighting is essential for obtaining accurate classifications but so far has remained an open research problem. We propose in this letter a supervised three-component feature weighting method based on the Fisher linear discriminant. Fisher linear discriminant method is used to calculate a coefficient for each feature. Then, these coefficients are modified according to a three-component scattering power decomposition model, combining both physical and statistical scattering characteristics to adapt them for the particular scattering mechanisms inherent in PolSAR data and assigned to the coherency matrix to enhance the discriminating ability of the features. Freeman decomposition and Wishart classifier are used to classify the PolSAR image. The effectiveness of the proposed method is demonstrated by experiments NASA/JPL AIRSAR L-band and CSA Radarsat-2 C-band PolSAR images of the San Francisco area.","Title: Improved Memetic Algorithm Based on Route Distance Grouping for Multiobjective Large Scale Capacitated Arc Routing Problems.\nYear: 2016\n\nAbstract: The capacitated arc routing problem (CARP) has attracted considerable attention from researchers due to its broad potential for social applications. This paper builds on, and develops beyond, the cooperative coevolutionary algorithm based on route distance grouping (RDG-MAENS), recently proposed by Mei et al. Although Mei's method has proved superior to previous algorithms, we discuss several remaining drawbacks and propose solutions to overcome them. First, although RDG is used in searching for potential better solutions, the solution generated from the decomposed problem at each generation is not the best one, and the best solution found so far is not used for solving the current generation. Second, to determine which sub-population the individual belongs to simply according to the distance can lead to an imbalance in the number of the individuals among different sub-populations and the allocation of resources. Third, the method of Mei et al. was only used to solve single-objective CARP. To overcome the above issues, this paper proposes improving RDG-MAENS by updating the solutions immediately and applying them to solve the current solution through areas shared, and then according to the magnitude of the vector of the route direction, and a fast and simple allocation scheme is proposed to determine which decomposed problem the route belongs to. Finally, we combine the improved algorithm with an improved decomposition-based memetic algorithm to solve the multiobjective large scale CARP (LSCARP). Experimental results suggest that the proposed improved algorithm can achieve better results on both single-objective LSCARP and multiobjective LSCARP.","Title: Dynamic-context cooperative quantum-behaved particle swarm optimization based on multilevel thresholding applied to medical image segmentation.\nYear: 2015\n\nAbstract: This paper proposes a dynamic-context cooperative quantum-behaved particle swarm optimization algorithm. The proposed algorithm incorporates a new method for dynamically updating the context vector each time it completes a cooperation operation with other particles. We first explain how this leads to enhanced search ability and improved optimization over previous methods, and demonstrate this empirically with comparative experiments using benchmark test functions. We then demonstrate a practical application of the proposed method, by showing how it can be applied to optimize the parameters for Otsu image segmentation for processing medical images. Comparative experimental results show that the proposed method outperforms other state-of-the-art methods from the literature.","Title: A compressed sensing approach for efficient ensemble learning.\nYear: 2014\n\nAbstract: This paper presents a method for improved ensemble learning, by treating the optimization of an ensemble of classifiers as a compressed sensing problem. Ensemble learning methods improve the performance of a learned predictor by integrating a weighted combination of multiple predictive models. Ideally, the number of models needed in the ensemble should be minimized, while optimizing the weights associated with each included model. We solve this problem by treating it as an example of the compressed sensing problem, in which a sparse solution must be reconstructed from an under-determined linear system. Compressed sensing techniques are then employed to find an ensemble which is both small and effective. An additional contribution of this paper, is to present a new performance evaluation method (a new pairwise diversity measurement) called the roulette-wheel kappa-error. This method takes into account the different weightings of the classifiers, and also reduces the total number of pairs of classifiers needed in the kappa-error diagram, by selecting pairs through a roulette-wheel selection method according to the weightings of the classifiers. This approach can greatly improve the clarity and informativeness of the kappa-error diagram, especially when the number of classifiers in the ensemble is large. We use 25 different public data sets to evaluate and compare the performance of compressed sensing ensembles using four different sparse reconstruction algorithms, combined with two different classifier learning algorithms and two different training data manipulation techniques. We also give the comparison experiments of our method against another five state-of-the-art pruning methods. These experiments show that our method produces comparable or better accuracy, while being significantly faster than the compared methods.","Title: Two-level RRT planning for robotic push manipulation\nYear: 2012\n\nAbstract: This paper presents an algorithm for planning sequences of pushes, by which a robotic arm equipped with a single rigid finger can move a manipulated object (or manipulandum) towards a desired goal pose. Pushing is perhaps the most basic kind of manipulation, however it presents difficult challenges for planning, because of the complex relationship between manipulative pushing actions and resulting manipulandum motions. The motion planning literature has well developed paradigms for solving e.g. the piano-mover's problem, where the search occurs directly in the configuration space of the manipulandum object being moved. In contrast, in pushing manipulation, a plan must be built in the action space of the robot, which is only indirectly linked to the motion space of the manipulandum through a complex interaction for which inverse models may not be known. In this paper, we present a two stage approach to planning pushing operations. A global RRT path planner is used to explore the space of possible manipulandum configurations, while a local push planner makes use of predictive models of pushing interactions, to plan sequences of pushes to move the manipulandum from one RRT node to the next. The effectiveness of the algorithm is demonstrated in simulation experiments in which a robot must move a rigid body through complex 3D transformations by applying only a sequence of simple single finger pushes.","Title: Motion estimation using physical simulation\nYear: 2010\n\nAbstract: We consider the task of monocular visual motion estimation from video image sequences. We hypothesise that performance on the task can be improved by incorporating an understanding of physically likely and feasible object dynamics. We test this hypothesis by incorporating a physical simulator into a least-squares estimation procedure. We initialise a full trajectory estimate using RANSAC followed by gradient descent refinement. We present results for 2D image sequences consisting of single ambiguous, visible or occluded balls, as well as results for 3D computer-generated sequences of objects in free-flight with added noise. Results suggest that restricting the estimation to allow only motions that are feasible according to the physics simulator can produce marked improvement when the observed object motion is within the limits of the physics simulator and its world model. Conversely, merely penalising deviations from feasible physical dynamics produces a consistent but incremental improvement over more common dynamics models.","Title: Towards The Principled Study Of Variable Autonomy In Mobile Robots\nYear: 2015\n\nAbstract: Safety critical and demanding tasks (e.g search and rescue or hazardous environments inspection), can benefit from robotic systems that offer a spectrum of control modes. These can range from direct teleoperation to full autonomy. This paper describes a pilot-study experiment in which a variable autonomy robot completes a navigation task. It explores the comparative performances of the human-robot system at different autonomy levels under different sets of conditions. This is done from a Mixed-Initiative system investigation perspective. Sensor noise was added to degrade robot performance, while a secondary task induced varying degrees of additional workload on the human operator. Carrying out these experiments and analyzing the initial results, has highlighted the profound complexities of designing tasks, conditions, and performance metrics which are: principled; eliminate confounding factors; and yield scientifically rigorous insights into the intricacies of a collaborative system that combines both human and robot intelligences. A key contribution of this paper is to describe the lessons learned from attempting these experiments, and to suggest a variety of guidelines for other researchers to consider when designing experiments in this context.","Title: Single image super-resolution reconstruction based on genetic algorithm and regularization prior model.\nYear: 2016\n\nAbstract: Single image super-resolution (SR) reconstruction is an ill-posed inverse problem because the high-resolution (HR) image, obtained from the low-resolution (LR) image, is non-unique or unstable. In this paper, single image SR reconstruction is treated as an optimization problem, and a new single image SR method, based on a genetic algorithm and regularization prior model, is proposed. In the proposed method, the optimization problem is constructed with a regularization prior model which consists of the non-local means (NLMs) filter, total variation (TV) and adaptive sparse domain selection (ASDS) scheme for sparse representation. In order to avoid local optimization, we combine the genetic algorithm and the iterative shrinkage algorithm to deal with the regularization prior model. Compared with several other state-of-the-art algorithms, the proposed method demonstrates better performances in terms of both numerical analysis and visual effect.","Title: Subspace learning-based graph regularized feature selection.\nYear: 2016\n\nAbstract: In recent years, a variety of feature selection algorithms based on subspace learning have been proposed. However, such methods typically do not exploit information about the underlying geometry of the data. To overcome this shortcoming, we propose a novel algorithm called subspace learning-based graph regularized feature selection (SGFS). SGFS builds on the feature selection framework of subspace learning, but extends it by incorporating the idea of graph regularization, in which a feature map is constructed on the feature space in order to preserve geometric structure information on the feature manifold. Additionally, the L2,1-norm is used to constrain the feature selection matrix to ensure the sparsity of the feature array and avoid trivial solutions. The resulting method can provide more accurate discrimination information for feature selection. We evaluate SGFS by comparing it against five other state-of-the-art algorithms from the literature, on twelve publicly available benchmark data sets. Empirical results suggest that SGFS is more effective than the other five feature selection algorithms.","Title: Distractor-Supported Single Target Tracking In Extremely Cluttered Scenes\nYear: 2016\n\nAbstract: This paper presents a novel method for single target tracking in RGB images under conditions of extreme clutter and camouflage, including frequent occlusions by objects with similar appearance as the target. In contrast to conventional single target trackers, which onlymaintain the estimated target status, we propose a multi-level clustering-based robust estimation for online detection and learning of multiple targetlike regions, called distractors, when they appear near to the true target. To distinguish the target from these distractors, we exploit a global dynamic constraint (derived from the target and the distractors) in a feedback loop to improve single target tracking performance in situations where the target is camouflaged in highly cluttered scenes. Our proposed method successfully prevents the estimated target location from erroneously jumping to a distractor during occlusion or extreme camouflage interactions. To gain an insightful understanding of the evaluated trackers, we have augmented publicly available benchmark videos, by proposing a new set of clutter and camouflage sub-attributes, and annotating these sub-attributes for all frames in all sequences. Using this dataset, we first evaluate the effect of each key component of the tracker on the overall performance. Then, the proposed tracker is compared to other highly ranked single target tracking algorithms in the literature. The experimental results show that applying the proposed global dynamic constraint in a feedback loop can improve single target tracker performance, and demonstrate that the overall algorithm significantly outperforms other state-of-the-art single target trackers in highly cluttered scenes.","Title: Immune clonal algorithm based on directed evolution for multi-objective capacitated arc routing problem.\nYear: 2016\n\nAbstract: Display OmittedThe non-dominant solutions obtained by three algorithms on some small and large-scale instances.In the above figure, the former two lines of figures are small-scale instances, in which DE-ICA shows the best convergence on 2 instances compared with other two algorithms. Besides, on other 4 instances of small-scale instances, DE-ICA is not worse than D-MAENS and ID-MAENS. In general, DE-ICA shows a slightly advantage on small scale instances. The reason is that small scale instances have a smaller solution space and they are easier to solve. Consequently, other two compared algorithms have enough capacity to deal with it and get a similar performance with DE-ICA.In the latter two lines of figures, DE-ICA gets the best convergence on 5 instances (e4a, s1a, s4a, G1-A, G2-E). On these instances, DE-ICA can reach areas with both low total cost and the low makespan. As a result, the non-dominant solutions obtained by DE-ICA almost completely dominate the non-dominant solutions obtained by D-MAENS and ID-MAENS. On the remaining instances, DE-ICA is also not worse than the other two algorithms in convergence. In conclusion, DE-ICA demonstrates an obvious advantage in convergence on these large scale instances. The results above show that the DE-ICA is more suitable for the large-scale problems. This paper proposes and Immune Clonal Algorithm Based on Directed Evolution for Multi-Objective Capacitated Arc Routing Problem.The proposed algorithm expands the scale of the initial population to increase the diversity.It helps the antibody populations to share the neighborhood information in time.It applies a brand-new kind of comparison operator.It can evolve in the direction of the better population. The capacitated arc routing problem is playing an increasingly important role in our society, engendering increasing attention from the research community. Among the various models, multi-objective capacitated arc routing problem comes much closer to real-world problems. Therefore, this paper proposes an immune clonal algorithm based on directed evolution to solve this problem. Firstly, the proposed algorithm adopts the framework of the immune clonal algorithm and expands the scale of the initial antibody population in the initialization process to increase the diversity of the antibodies. Secondly, the proposed algorithm is combined with a decomposition strategy in the operations of the immune gene. Antibodies are classified to perform the immune genetic operations, which helps the antibody populations to share the neighborhood information in a timely manner. Thirdly, the proposed algorithm applies a novel kind of comparison operator to build the total population, which helps it to evolve in the direction of a better population and improves the quality of the antibodies. Experimental results suggest that the proposed algorithm can generate better non-dominant solutions than several compared state-of-the-art algorithms, especially for large-scale sets.","Title: A local-global coupled-layer puppet model for robust online human pose tracking.\nYear: 2016\n\nAbstract: We propose a new method for online tracking of articulated human body poses.Our method offers online sequential tracking from one frame to the next.Many other methods mutually optimize poses offline over all frames of a sequence.We propose a novel cross-coupled global-local model of articulated human body pose.We propose an adaptive penalty function for optimizing the pose estimates. This paper addresses the problem of online tracking of articulated human body poses in dynamic environments. Many previous approaches perform poorly in realistic applications: often future frames or entire sequences are used anticausally to mutually refine the poses in each individual frame, making online tracking impossible; tracking often relies on strong assumptions about e.g. clothing styles, body-part colours and constraints on body-part motion ranges, limiting such algorithms to a particular dataset; the use of holistic feature models limits the ability of optimisation-based matching to distinguish between pose errors of different body parts. We overcome these problems by proposing a coupled-layer framework, which uses the previous notions of deformable structure (DS) puppet models. The underlying idea is to decompose the global pose candidate in any particular frame into several local parts to obtain a refined pose. We introduce an adaptive penalty with our model to improve the searching scope for a local part pose, and also to overcome the problem of using fixed constraints. Since the pose is computed using only current and previous frames, our method is suitable for online sequential tracking. We have carried out empirical experiments using three different public benchmark datasets, comparing two variants of our algorithm against four recent state-of-the-art (SOA) methods from the literature. The results suggest comparatively strong performance of our method, regardless of weaker constraints and fewer assumptions about the scene, and despite the fact that our algorithm is performing online sequential tracking, whereas the comparison methods perform mutual optimisation backwards and forwards over all frames of the entire video sequence.","Title: Towards robotic decommissioning of legacy nuclear plant: Results of human-factors experiments with tele-robotic manipulation, and a discussion of challenges and approaches for decommissioning\nYear: 2016\n\nAbstract: This paper explores the problems of developing robotic systems for decommissioning legacy nuclear infrastructure, such as the many contaminated gloveboxes present in UK, USA and other countries. We begin with a discussion of these decomissioning challenges. We review the current manual methods for decommissioning alpha-contaminated plant, and review robotic approaches which might replace such direct human interventions. We then present our initial experiments with human test-subjects, exploring the ability of humans to control a remote robot to perform complex manipulation tasks. Our preliminary results reveal a number of interesting lessons: conventional tele-manipulation is very difficult and very slow without significant training; metrics for usability of such technology can be conflicting and hard to interpret; aptitude for tele-manipulation varies significantly between individuals; however such aptitude may be predicted by using spatial awareness tests to select prospective robot operators; additionally the abilities of people with different initial aptitudes appear to converge somewhat as learning progresses. An additional contribution of this paper is to show how rigorous scientific methodologies, drawn from the psychology and human-factors research fields, can be used to analyse the performance of humans using robots to perform practical tasks.","Title: Nonnegative Matrix Factorization with Rank Regularization and Hard Constraint.\nYear: 2017\n\nAbstract: Nonnegative matrix factorization (NMF) is well known to be an effective tool for dimensionality reduction in problems involving big data. For this reason, it frequently appears in many areas of scientific and engineering literature. This letter proposes a novel semisupervised NMF algorithm for overcoming a variety of problems associated with NMF algorithms, including poor use of prior information,...","Title: Guiding Trajectory Optimization by Demonstrated Distributions.\nYear: 2017\n\nAbstract: Trajectory optimization is an essential tool for motion planning under multiple constraints of robotic manipulators. Optimization-based methods can explicitly optimize a trajectory by leveraging prior knowledge of the system and have been used in various applications such as collision avoidance. However, these methods often require a hand-coded cost function in order to achieve the desired behavio...","Title: A fully end-to-end deep learning approach for real-time simultaneous 3D reconstruction and material recognition\nYear: 2017\n\nAbstract: This paper addresses the problem of simultaneous 3D reconstruction and material recognition and segmentation. Enabling robots to recognise different materials (concrete, metal etc.) in a scene is important for many tasks, e.g. robotic interventions in nuclear decommissioning. Previous work on 3D semantic reconstruction has predominantly focused on recognition of everyday domestic objects (tables, chairs etc.), whereas previous work on material recognition has largely been confined to single 2D images without any 3D reconstruction. Meanwhile, most 3D semantic reconstruction methods rely on computationally expensive post-processing, using Fully-Connected Conditional Random Fields (CRFs), to achieve consistent segmentations. In contrast, we propose a deep learning method which performs 3D reconstruction while simultaneously recognising different types of materials and labeling them at the pixel level. Unlike previous methods, we propose a fully end-to-end approach, which does not require hand-crafted features or CRF post-processing. Instead, we use only learned features, and the CRF segmentation constraints are incorporated inside the fully end-to-end learned system. We present the results of experiments, in which we trained our system to perform real-time 3D semantic reconstruction for 23 different materials in a real-world application. The run-time performance of the system can be boosted to around 10Hz, using a conventional GPU, which is enough to achieve realtime semantic reconstruction using a 30fps RGB-D camera. To the best of our knowledge, this work is the first real-time end-to-end system for simultaneous 3D reconstruction and material recognition.","Title: Weakly-supervised DCNN for RGB-D Object Recognition in Real-World Applications Which Lack Large-scale Annotated Training Data.\nYear: 2017\n\nAbstract: This paper addresses the problem of RGBD object recognition in real-world applications, where large amounts of annotated training data are typically unavailable. To overcome this problem, we propose a novel, weakly-supervised learning architecture (DCNN-GPC) which combines parametric models (a pair of Deep Convolutional Neural Networks (DCNN) for RGB and D modalities) with non-parametric models (Gaussian Process Classification). Our system is initially trained using a small amount of labeled data, and then automatically prop- agates labels to large-scale unlabeled data. We first run 3D- based objectness detection on RGBD videos to acquire many unlabeled object proposals, and then employ DCNN-GPC to label them. As a result, our multi-modal DCNN can be trained end-to-end using only a small amount of human annotation. Finally, our 3D-based objectness detection and multi-modal DCNN are integrated into a real-time detection and recognition pipeline. In our approach, bounding-box annotations are not required and boundary-aware detection is achieved. We also propose a novel way to pretrain a DCNN for the depth modality, by training on virtual depth images projected from CAD models. We pretrain our multi-modal DCNN on public 3D datasets, achieving performance comparable to state-of-the-art methods on Washington RGBS Dataset. We then finetune the network by further training on a small amount of annotated data from our novel dataset of industrial objects (nuclear waste simulants). Our weakly supervised approach has demonstrated to be highly effective in solving a novel RGBD object recognition application which lacks of human annotations.","Title: Learning modular and transferable forward models of the motions of push manipulated objects.\nYear: 2017\n\nAbstract: The ability to predict how objects behave during manipulation is an important problem. Models informed by mechanics are powerful, but are hard to tune. An alternative is to learn a model of the object's motion from data, to learn to predict. We study this for push manipulation. The paper starts by formulating a quasi-static prediction problem. We then pose the problem of learning to predict in two different frameworks: (i) regression and (ii) density estimation. Our architecture is modular: many simple, object specific, and context specific predictors are learned. We show empirically that such predictors outperform a rigid body dynamics engine tuned on the same data. We then extend the density estimation approach using a product of experts. This allows transfer of learned motion models to objects of novel shape, and to novel actions. With the right representation and learning method, these transferred models can match the prediction performance of a rigid body dynamics engine for novel objects or actions.","Title: Dynamic multi-level appearance models and adaptive clustered decision trees for single target tracking.\nYear: 2017\n\nAbstract: A robust tracking algorithm for tracking arbitrary objects in challenging video sequences.An adaptive clustered decision tree approach which dynamically selects the minimum combination of features to represent target parts.This adaptive clustered decision tree is utilized both to enable robust matching at the part level and to select the new parts for learning. This paper presents a tracking algorithm for arbitrary objects in challenging video sequences. Targets are modelled at three different levels of granularity (pixel, parts and bounding box levels), which are cross-constrained to enable robust model relearning. The main contribution is an adaptive clustered decision tree method which dynamically selects the minimum combination of features necessary to sufficiently represent each target part at each frame, thereby providing robustness with computational efficiency. The adaptive clustered decision tree is used in two separate ways: firstly for parts level matching between successive frames; secondly to select the best candidate image regions for learning new parts of the target. We test the tracker using two different tracking benchmarks (VOT2013-2014 and CVPR2013 tracking challenges), based on two different test methodologies, and show it to be more robust than the state-of-the-art methods from both of those tracking challenges, while also offering competitive tracking precision. Additionally, we evaluate the contribution of each key component of the tracker to overall performance; test the sensitivity of the tracker under different initialization conditions; investigate the effect of using features in different orders within the decision trees; illustrate the flexibility of the method for handling arbitrary kinds of features, by showing how it easily extends to handle RGB-D data.","Title: Autonomous vision-guided bi-manual grasping and manipulation\nYear: 2017\n\nAbstract: This paper describes the implementation, demonstration and evaluation of a variety of autonomous, vision-guided manipulation capabilities, using a dual-arm Baxter robot. Initially, symmetric coordinated bi-manual manipulation based on kinematic tracking algorithm was implemented on the robot to enable a master-slave manipulation system. We demonstrate the efficacy of this approach with a human-robot collaboration experiment, where a human operator moves the master arm along arbitrary trajectories and the slave arm automatically follows the master arm while maintaining a constant relative pose between the two end-effectors. Next, this concept was extended to perform dual-arm manipulation without human intervention. To this extent, an image-based visual servoing scheme has been developed to control the motion of arms for positioning them at a desired grasp locations. Next we combine this with a dynamic position controller to move the grasped object using both arms in a prescribed trajectory. The presented approach has been validated by performing numerous symmetric and asymmetric bi-manual manipulations at different conditions. Our experiments demonstrated 80% success rate in performing the symmetric dual-arm manipulation tasks; and 73% success rate in performing asymmetric dualarm manipulation tasks.","Title: Hybrid motion/force control: a review.\nYear: 2017\n\nAbstract: This paper reviews hybrid motion/force control, a control scheme which enables robots to perform tasks involving both motion, in the free space, and interactive force, at the contacts. Motivated by the large amount of literature on this topic, we facilitate comparison and elucidate the key differences among different approaches. An emphasis is placed on the study of the decoupling of motion control and force control. And we conclude that it is indeed possible to achieve a complete decoupling; however, this feature can be relaxed or sacrificed to reduce the robot's joint torques while still completing the task.","Title: Dense RGB-D semantic mapping with Pixel-Voxel neural network.\nYear: 2018\n\nAbstract: In this paper, a novel Pixel-Voxel network is proposed for dense 3D semantic mapping, which can perform dense 3D mapping while simultaneously recognizing and labelling the semantic category each point in the 3D map. In our approach, we fully leverage the advantages of different modalities. That is, the PixelNet can learn the high-level contextual information from 2D RGB images, and the VoxelNet can learn 3D geometrical shapes from the 3D point cloud. Unlike the existing architecture that fuses score maps from different modalities with equal weights, we propose a softmax weighted fusion stack that adaptively learns the varying contributions of PixelNet and VoxelNet and fuses the score maps according to their respective confidence levels. Our approach achieved competitive results on both the SUN RGB-D and NYU V2 benchmarks, while the runtime of the proposed system is boosted to around 13 Hz, enabling near-real-time performance using an i7 eight-cores PC with a single Titan X GPU.","Title: Mixed second order partial derivatives decomposition method for large scale optimization.\nYear: 2017\n\nAbstract: •A theoretical analysis of the interaction between variables is developed.•Three theorems and three lemma are presented.•A decomposition method based on mixed second order partial derivatives is proposed.","Title: Multi-objective artificial immune algorithm for fuzzy clustering based on multiple kernels.\nYear: 2017\n\nAbstract: This paper presents a multi-objective artificial immune algorithm for fuzzy clustering based on multiple kernels (MAFC). MAFC extends the classical Fuzzy C-Means (FCM) algorithm and improves some of its important limitations, such as vulnerability to local optima convergence, which can lead to poor clustering quality. MAFC unifies multi-kernel learning and multi-objective optimization in a joint clustering framework, which preserves the geometric information of the dataset. The multi-kernel method maps data from the feature space to kernel space by using kernel functions. Additionally, the introduction of multi-objective optimization helps to optimize between-cluster separation and within-cluster compactness simultaneously via two different clustering validity criteria. These properties help the proposed algorithm to avoid becoming stuck at local optima. Furthermore, this paper utilizes an artificial immune algorithm to address the multi-objective clustering problem and acquire a Pareto optimal solution set. The solution set is obtained through the process of antibody population initialization, clone proliferation, non-uniform mutation and uniformity maintaining strategy, which avoids the problems of degradation and prematurity which can occur with conventional genetic algorithms. Finally, we choose the best solution from the Pareto optimal solution set. We use a semi-supervised method to achieve the final clustering results. We compare our method against state-of-the-art methods from the literature by performing experiments with both UCI datasets and face datasets. The results suggest that MAFC is significantly more efficient for clustering and has a wider scope of application.","Title: Quantum-Inspired Immune Clonal Algorithm for solving large-scale capacitated arc routing problems.\nYear: 2018\n\nAbstract: In this paper, we present an approach to Large-Scale CARP called Quantum-Inspired Immune Clonal Algorithm (QICA-CARP). This algorithm combines the feature of artificial immune system and quantum computation ground on the qubit and the quantum superposition. We call an antibody of population quantum bit encoding, in QICA-CARP. For this encoding, to control the population with a high probability evolution towards a good schema we use the information on the current optimal antibody. The mutation strategy of quantum rotation gate accelerates the convergence of the original clone operator. Moreover, quantum crossover operator enhances the exchange of information and increases the diversity of the population. Furthermore, it avoids falling into local optimum. We also use the repair operator to amend the infeasible solutions to ensure the diversity of solutions. This makes QICA-CARP approximating the optimal solution. We demonstrate the effectiveness of our approach by a set of experiments and by Comparing the results of our approach with ones obtained with the RDG-MAENS and RAM using different test sets. Experimental results show that QICA-CARP outperforms other algorithms in terms of convergence rate and the quality of the obtained solutions. Especially, QICA-CARP converges to a better lower bound at a faster rate illustrating that it is suitable for solving large-scale CARP.","Title: Learning Monocular Visual Odometry With Dense 3d Mapping From Dense 3d Flow\nYear: 2018\n\nAbstract: This paper introduces a fully deep learning approach to monocular SLAM, which can perform simultaneous localization using a neural network for learning visual odometry (L-VO) and dense 3D mapping. Dense 2D flow and a depth image are generated from monocular images by sub-networks, which are then used by a 3D flow associated layer in the L-VO network to generate dense 3D flow. Given this 3D flow, the dual-stream L-VO network can then predict the 6DOF relative pose and furthermore reconstruct the vehicle trajectory. In order to learn the correlation between motion directions, the Bivariate Gaussian modeling is employed in the loss function. The L-VO network achieves an overall performance of 2:68% for average translational error and 0.0143 degrees/m for average rotational error on the KITTI odometry benchmark. Moreover, the learned depth is leveraged to generate a dense 3D map. As a result, an entire visual SLAM system, that is, learning monocular odometry combined with dense 3D mapping, is achieved.","Title: Robust Fusion of Color and Depth Data for RGB-D Target Tracking Using Adaptive Range-Invariant Depth Models and Spatio-Temporal Consistency Constraints.\nYear: 2018\n\nAbstract: This paper presents a novel robust method for single target tracking in RGB-D images, and also contributes a substantial new benchmark dataset for evaluating RGB-D trackers. While a target object's color distribution is reasonably motion-invariant, this is not true for the target's depth distribution, which continually varies as the target moves relative to the camera. It is therefore nontrivial t...","Title: Sensors, SLAM and Long-term Autonomy: A Review.\nYear: 2018\n\nAbstract: Simultaneous Localization and Mapping, commonly known as SLAM, has been an active research area in the field of Robotics over the past three decades. For solving the SLAM problem, every robot is equipped with either a single sensor or a combination of similar/different sensors. This paper attempts to review, discuss, evaluate and compare these sensors. Keeping an eye on future, this paper also assesses the characteristics of these sensors against factors critical to the long- term autonomy challenge.","Title: Weather Classification: A new multi-class dataset, data augmentation approach and comprehensive evaluations of Convolutional Neural Networks.\nYear: 2018\n\nAbstract: Weather conditions often disrupt the proper functioning of transportation systems. Present systems either deploy an array of sensors or use an in-vehicle camera to predict weather conditions. These solutions have resulted in incremental cost and limited scope. To ensure smooth operation of all transportation services in all-weather conditions, a reliable detection system is necessary to classify weather in wild. The challenges involved in solving this problem is that weather conditions are diverse in nature and there is an absence of discriminate features among various weather conditions. The existing works to solve this problem have been scene specific and have targeted classification of two categories of weather. In this paper, we have created a new open source dataset consisting of images depicting three classes of weather i. e rain, snow and fog called RFS Dataset. A novel algorithm has also been proposed which has used super pixel delimiting masks as a form of data augmentation, leading to reasonable results with respect to ten Convolutional Neural Network architectures.","Title: Real-Time Application Processing for FPGA-Based Resilient Embedded Systems in Harsh Environments.\nYear: 2018\n\nAbstract: Real-time embedded systems nowadays get employed in harsh environments such as space, nuclear sites to carry out critical operations. Along with the traditional software based (CPU) execution, FPGAs are now also emerging as a bright prospect to accomplish such routines. However, these platforms are often get plagued by faults generated due to the high radiations in such environments. As a result, the real-time applications running on the platform could also get jeopardized. Thus, efficient execution of a set of hard real-time applications on reconfigurable systems with anomaly detection and recovery mechanism is inevitable. This work aims at tackling such problem with a \"healing\" approach for extreme environments. Initially, the applications are intelligently partitioned for hardware and software execution, then attempts have been made to schedule hardware applications with intermittent preemption point. Upon detecting any abnormality on such distinct points, our approach orchestrates a healing mechanism to remediate the scenario without hampering the pre-determined schedule. Experimental validation of our proposed method reveals its effectiveness.","Title: Unsupervised feature selection based on kernel fisher discriminant analysis and regression learning\nYear: 2019\n\nAbstract: In this paper, we propose a new feature selection method called kernel fisher discriminant analysis and regression learning based algorithm for unsupervised feature selection. The existing feature selection methods are based on either manifold learning or discriminative techniques, each of which has some shortcomings. Although some studies show the advantages of two-steps method benefiting from both manifold learning and discriminative techniques, a joint formulation has been shown to be more efficient. To do so, we construct a global discriminant objective term of a clustering framework based on the kernel method. We add another term of regression learning into the objective function, which can impose the optimization to select a low-dimensional representation of the original dataset. We use L2,1-norm of the features to impose a sparse structure upon features, which can result in more discriminative features. We propose an algorithm to solve the optimization problem introduced in this paper. We further discuss convergence, parameter sensitivity, computational complexity, as well as the clustering and classification accuracy of the proposed algorithm. In order to demonstrate the effectiveness of the proposed algorithm, we perform a set of experiments with different available datasets. The results obtained by the proposed algorithm are compared against the state-of-the-art algorithms. These results show that our method outperforms the existing state-of-the-art methods in many cases on different datasets, but the improved performance comes with the cost of increased time complexity.","Title: Model-Free And Learning-Free Grasping By Local Contact Moment Matching\nYear: 2018\n\nAbstract: This paper addresses the problem of grasping arbitrarily shaped objects, observed as partial point-clouds, without requiring: models of the objects, physics parameters, training data, or other a-priori knowledge. A grasp metric is proposed based on Local Contact Moment (LoCoMo). LoCoMo combines zero-moment shift features, of both hand and object surface patches, to determine local similarity. This metric is then used to search for a set of feasible grasp poses with associated grasp likelihoods. LoCoMo overcomes some limitations of both classical grasp planners and learning-based approaches. Unlike force-closure analysis, LoCoMo does not require knowledge of physical parameters such as friction coefficients, and avoids assumptions about fingertip contacts, instead enabling robust contacts of large areas of hand and object surface. Unlike more recent learning-based approaches, LoCoMo does not require training data, and does not need any prototype grasp configurations to be taught by kinesthetic demonstration. We present results of real-robot experiments grasping 21 different objects, observed by a wrist-mounted depth camera. All objects are grasped successfully when presented to the robot individually. The robot also successfully clears cluttered heaps of objects by sequentially grasping and lifting objects until none remain.","Title: Choosing Grasps to Enable Collision-Free Post-Grasp Manipulations\nYear: 2018\n\nAbstract: Consider the task of grasping the handle of a door, and then pushing it until the door opens. These two fundamental robotics problems (selecting secure grasps of a hand on an object, e.g. the door handle, and planning collision-free trajectories of a robot arm that will move that object along a desired path) have predominantly been studied separately from one another. Thus, much of the grasping literature overlooks the fundamental purpose of grasping objects, which is typically to make them move in desirable ways. Given a desired post-grasp trajectory of the object, different choices of grasp will often determine whether or not collision-free post-grasp motions of the arm can be found, which will deliver that trajectory. We address this problem by examining a number of possible stable grasping configurations on an object. For each stable grasp, we explore the motion space of the manipulator which would be needed for post-grasp motions, to deliver the object along the desired trajectory. A criterion, based on potential fields in the post-grasp motion space, is used to assign a collision-cost to each grasp. A grasping configuration is then selected which enables the desired post-grasp object motion while minimising the proximity of all robot parts to obstacles during motion. We demonstrate our method with peg-in-hole and pick-and-place experiments in cluttered scenes, using a Franka Panda robot. Our approach is effective in selecting appropriate grasps, which enable both stable grasp and also desired post-grasp movements without collisions. We also show that, when grasps are selected based on grasp stability alone, without consideration for desired post-grasp manipulations, the corresponding post-grasp movements of the manipulator may result in collisions.","Title: A Fast Algorithm for SAR Image Segmentation Based on Key Pixels.\nYear: 2017\n\nAbstract: Recent high-performance clustering methods process all pixels when segmenting an image, which results in a very large time complexity of these algorithms. Additionally, the performance of such algorithms can be severely affected by noise when dealing with highly polluted images. To address these problems, we propose a new unsupervised algorithm for segmenting synthetic aperture radar images based ...","Title: A Spatial Fuzzy Clustering Algorithm With Kernel Metric Based on Immune Clone for SAR Image Segmentation.\nYear: 2016\n\nAbstract: The fuzzy c-means (FCM) clustering algorithm has been widely used in image segmentation. However, FCM exhibits poor robustness to noise, often leading to unsatisfactory segmentations on noisy images. Additionally, the FCM algorithm is sensitive to the choice of initial cluster centers. In order to solve these problems, this paper proposes clone kernel spatial FCM (CKS_FCM), which improves segmenta...","Title: An Enhanced Adaptive Coupled-Layer LGTracker++\nYear: 2013\n\nAbstract: This paper addresses the problems of tracking targets which undergo rapid and significant appearance changes. Our starting point is a successful, state-of-the-art tracker based on an adaptive coupled-layer visual model [10]. In this paper, we identify four important cases when the original tracker often fails: significant scale changes, environment clutter, and failures due to occlusion and rapid disordered movement. We suggest four new enhancements to solve these problems: we adapt the scale of the patches in addition to adapting the bounding box, marginal patch distributions are used to solve patch drifting in environment clutter, a memory is added and used to assist recovery from occlusion, situations where the tracker may lose the target are automatically detected, and a particle filter is substituted for the Kalman filter to help recover the target. We demonstrate the advantages of the enhanced tracker over the original tracker using a test toolkit [17]. We demonstrate the advantages of the enhanced tracker over the original tracker, as well as several other state-of-the art trackers from the literature.","Title: Non-Negative Spectral Learning and Sparse Regression-Based Dual-Graph Regularized Feature Selection.\nYear: 2018\n\nAbstract: Feature selection is an important approach for reducing the dimension of high-dimensional data. In recent years, many feature selection algorithms have been proposed, but most of them only exploit information from the data space. They often neglect useful information contained in the feature space, and do not make full use of the characteristics of the data. To overcome this problem, this paper pr...","Title: Region-sequence based six-stream CNN features for general and fine-grained human action recognition in videos.\nYear: 2018\n\nAbstract: •We proposed a new method for human fine-grained action recognition in videos.•Our method uses a coarse pose estimation method to cut video frames and get human body foreground patch sequence.•Our method focuses on human lower arm area to enhance effective pixels for fine-grained actions.•We propose an encoding method to process the last pooling layer features of CNN structure.","Title: SAR Targets Classification Based on Deep Memory Convolution Neural Networks and Transfer Parameters.\nYear: 2018\n\nAbstract: Deep learning has obtained state-of-the-art results in a variety of computer vision tasks and has also been used to solve SAR image classification problems. Deep learning algorithms typically require a large amount of training data to achieve high accuracy. In contrast, the size of SAR image datasets is often comparatively limited. Therefore, this paper proposes a novel method, deep memory convolu...","Title: A dynamic local cluster ratio-based band selection algorithm for hyperspectral images\nYear: 2019\n\nAbstract: Hyperspectral band selection algorithms can save the computational costs during image restoration and analysis. This paper proposes a novel unsupervised band selection method, based on dynamic local cluster ratio (DLCR). The contributions of this paper can be summarized as follows. First, the similarity matrix is calculated in a novel way. Conventional approaches compute the matrix from the Euclidean distances between bands and are vulnerable to noise. Our proposed method can improve the robustness to such noise. Second, we propose an enhanced clustering strategy which clusters each band individually. Third, a dynamic ranking strategy is used to select bands iteratively. Bands that are highly correlated with each other will be prevented from being added to avoid redundancy. DLCR demonstrates improved performance on the Indian Pines and Pavia University data sets, when compared against other methods from the literature.","Title: Dynamic grasp and trajectory planning for moving objects\nYear: 2019\n\nAbstract: This paper shows how a robot arm can follow and grasp moving objects tracked by a vision system, as is needed when a human hands over an object to the robot during collaborative working. While the object is being arbitrarily moved by the human co-worker, a set of likely grasps, generated by a learned grasp planner, are evaluated online to generate a feasible grasp with respect to both: the current configuration of the robot respecting the target grasp; and the constraints of finding a collision-free trajectory to reach that configuration. A task-based cost function enables relaxation of motion-planning constraints, enabling the robot to continue following the object by maintaining its end-effector near to a likely pre-grasp position throughout the object’s motion. We propose a method of dynamic switching between: a local planner, where the hand smoothly tracks the object, maintaining a steady relative pre-grasp pose; and a global planner, which rapidly moves the hand to a new grasp on a completely different part of the object, if the previous graspable part becomes unreachable. Various experiments are conducted using a real collaborative robot and the obtained results are discussed.","Title: Hypothesis-based Belief Planning for Dexterous Grasping.\nYear: 2019\n\nAbstract: Belief space planning is a viable alternative to formalise partially observable control problems and, in the recent years, its application to robot manipulation problems has grown. However, this planning approach was tried successfully only on simplified control problems. In this paper, we apply belief space planning to the problem of planning dexterous reach-to-grasp trajectories under object pose uncertainty. In our framework, the robot perceives the object to be grasped on-the-fly as a point cloud and compute a full 6D, non-Gaussian distribution over the objectu0027s pose (our belief space). The system has no limitations on the geometry of the object, i.e., non-convex objects can be represented, nor assumes that the point cloud is a complete representation of the object. A plan in the belief space is then created to reach and grasp the object, such that the information value of expected contacts along the trajectory is maximised to compensate for the pose uncertainty. If an unexpected contact occurs when performing the action, such information is used to refine the pose distribution and triggers a re-planning. Experimental results show that our planner (IR3ne) improves grasp reliability and compensates for the pose uncertainty such that it doubles the proportion of grasps that succeed on a first attempt.","Title: 2D Linear Time-Variant Controller for Human's Intention Detection for Reach-to-Grasp Trajectories in Novel Scenes.\nYear: 2019\n\nAbstract: Designing robotic assistance devices for manipulation tasks is challenging. This work is concerned with improving accuracy and usability of semi-autonomous robots, such as human operated manipulators or exoskeletons. The key insight is to develop a system that takes into account context- and user-awareness to take better decisions in how to assist the user. The context-awareness is implemented by enabling the system to automatically generate a set of candidate grasps and reach-to-grasp trajectories in novel, cluttered scenes. The user-awareness is implemented as a linear time-variant feedback controller to facilitate the motion towards the most promising grasp. Our approach is demonstrated in a simple 2D example in which participants are asked to grasp a specific object in a clutter scene. Our approach also reduce the number of controllable dimensions for the user by providing only control on x- and y-axis, while orientation of the end-effector and the pose of its fingers are inferred by the system. The experimental results show the benefits of our approach in terms of accuracy and execution time with respect to a pure manual control.","Title: Automatic Detection of Myocontrol Failures Based upon Situational Context Information\nYear: 2019\n\nAbstract: Myoelectric control systems for assistive devices are still unreliable. The user's input signals can become unstable over time due to e.g. fatigue, electrode displacement, or sweat. Hence, such controllers need to be constantly updated and heavily rely on user feedback. In this paper, we present an automatic failure detection method which learns when plausible predictions become unreliable and model updates are necessary. Our key insight is to enhance the control system with a set of generative models that learn sensible behaviour for a desired task from human demonstration. We illustrate our approach on a grasping scenario in Virtual Reality, in which the user is asked to grasp a bottle on a table. From demonstration our model learns the reach-to-grasp motion from a resting position to two grasps (power grasp and tridigital grasp) and how to predict the most adequate grasp from local context, e.g. tridigital grasp on the bottle cap or around the bottleneck. By measuring the error between new grasp attempts and the model prediction, the system can effectively detect which input commands do not reflect the user's intention. We evaluated our model in two cases: i) with both position and rotation information of the wrist pose, and ii) with only rotational information. Our results show that our approach detects statistically highly significant differences in error distributions with p<;0.001 between successful and failed grasp attempts in both cases.","Title: Metrics and Benchmarks for Remote Shared Controllers in Industrial Applications.\nYear: 2019\n\nAbstract: Remote manipulation is emerging as one of the key robotics tasks needed in extreme environments. Several researchers have investigated how to add AI components into shared controllers to improve their reliability. Nonetheless, the impact of novel research approaches in real-world applications can have a very slow in-take. We propose a set of benchmarks and metrics to evaluate how the AI components of remote shared control algorithms can improve the effectiveness of such frameworks for real industrial applications. We also present an empirical evaluation of a simple intelligent share controller against a manually operated manipulator in a tele-operated grasping scenario.","Title: Selective Ensemble Learning based Human Action Recognition Using Fusing Visual Features\nYear: 2018\n\nAbstract: The selection of motion feature directly affects the recognition effect of human action recognition method. Single feature is often affected by human appearance, environment, camera settings and other factors, and its recognition effect is limited. This paper propose a novel action recognition method by using selective ensemble learning, which is a special paradigm of ensemble learning. Moreover, this paper presents a fast and efficient action description feature and a novel recognition algorithm. Robust discriminant mixed features are learnt from behavioral video frames as behavioral descriptors, The recogniton algorithm using selective ensemble learning can achieve fast classification. Experimental results show that the proposed method achieves ideal recognition results on the self-built indoor behavior data set and public data set.","Title: Semi-Autonomous Behaviour Tree-Based Framework For Sorting Electric Vehicle Batteries Components\nYear: 2021\n\nAbstract: The process of recycling electric vehicle (EV) batteries currently represents a significant challenge to the waste management automation industry. One example of it is the necessity of removing and sorting dismantled components from EV battery pack. This paper proposes a novel framework to semi-automate the process of removing and sorting different objects from an EV battery pack using a mobile manipulator. The work exploits the Behaviour Trees model for cognitive task execution and monitoring, which links different robot capabilities such as navigation, object tracking and motion planning in a modular fashion. The framework was tested in simulation, in both static and dynamic environments, and it was evaluated based on task time and the number of objects that the robot successfully placed in the respective containers. Results suggested that the robot's success rate in accomplishing the task of sorting the battery components was 95% and 82% in static and dynamic environments, respectively.","Title: Trust, Shared Understanding and Locus of Control in Mixed-Initiative Robotic Systems\nYear: 2021\n\nAbstract: This paper investigates how trust, shared under-standing between a human operator and a robot, and the Locus of Control (LoC) personality trait, evolve and affect Human-Robot Interaction (HRI) in mixed-initiative robotic systems. As such systems become more advanced and able to instigate actions alongside human operators, there is a shift from robots being perceived as a tool to being a team-mate....","Title: Dual Quaternion-Based Visual Servoing for Grasping Moving Objects\nYear: 2021\n\nAbstract: This paper presents a new dual quaternion-based formulation for pose-based visual servoing. Extending our previous work on local contact moment (LoCoMo) based grasp planning, we demonstrate grasping of arbitrarily moving objects in 3D space. Instead of using the conventional axis-angle parameterization, dual quaternions allow designing the visual servoing task in a more compact manner and provide ...","Title: SpectGRASP: Robotic Grasping by Spectral Correlation\nYear: 2021\n\nAbstract: This paper presents a spectral correlation-based method (SpectGRASP) for robotic grasping of arbitrarily shaped, unknown objects. Given a point cloud of an object, SpectGRASP extracts contact points on the object's surface matching the hand configuration. It neither requires offline training nor a-priori object models. We propose a novel Binary Extended Gaussian Image (BEGI), which represents the point cloud surface normals of both object and robot fingers as signals on a 2-sphere. Spherical harmonics are then used to estimate the correlation between fingers and object BEGIs. The resulting spectral correlation density function provides a similarity measure of gripper and object surface normals. This is highly efficient in that it is simultaneously evaluated at all possible finger rotations in SO(3). A set of contact points are then extracted for each finger using rotations with high correlation values. We then use our previous work, Local Contact Moment (LoCoMo) similarity metric, to sequentially rank the generated grasps such that the one with maximum likelihood is executed. We evaluate the performance of SpectGRASP by conducting experiments with a 7-axis robot fitted with a parallel-jaw gripper, in a physics simulation environment. Obtained results indicate that the method not only can grasp individual objects, but also can successfully clear randomly organized groups of objects. The SpectGRASP method also outperforms the closest state-of-the-art method in terms of grasp generation time and grasp-efficiency.","Title: Towards Robotizing The Processes Of Testing Lithium-Ion Batteries\nYear: 2021\n\nAbstract: To boost the circular economy of the electric vehicle battery industry, an accurate assessment of the state of health of retired batteries is essential to assign them an appropriate value in the post automotive market and material degradation before recycling. In practice, the advanced battery testing techniques are usually limited to laboratory benches at the battery cell level and hardly used in the industrial environment at the battery module or pack level. This necessitates developing battery recycling facilities that can handle the assessment and testing undertakings for many batteries with different form factors. Towards this goal, for the first time, this article proposes proof of concept to automate the process of collecting the impedance data from a retired 24kWh Nissan LEAF battery module. The procedure entails the development of robot end-of-arm tooling that was connected to a Potentiostat. In this study, the robot was guided towards a fixed battery module using visual servoing technique, and then impedance control system was applied to create compliance between the end-of-arm tooling and the battery terminals. Moreover, an alarm system was designed and mounted on the robot's wrist to check the connectivity between a Potentiostat and the battery terminals. Subsequently, the electrochemical impedance spectroscopy test was run over a wide range of frequencies at a 5% state of charge. The electrochemical impedance spectroscopy data obtained from the automated test is validated by means of the three criteria (linearity, causality and stability) and compared with manually collected measurements under the same conditions. Results suggested the proposed automated configuration can accurately accomplish the electrochemical impedance spectroscopy test at the battery module level with no human intervention, which ensures safety and allows this advanced testing technique to be adopted in grading retired battery modules.","Title: A Bayesian-Based Approach to Human Operator Intent Recognition in Remote Mobile Robot Navigation\nYear: 2021\n\nAbstract: This paper addresses the problem of human operator intent recognition during teleoperated robot navigation. In this context, recognition of the operator’s intended navigational goal, could enable an artificial intelligence (AI) agent to assist the operator in an advanced human-robot interaction framework. We propose a Bayesian Operator Intent Recognition (BOIR) probabilistic method that utilizes: ...","Title: Fessonia - a Method for Real-Time Estimation of Human Operator Workload Using Behavioural Entropy.\nYear: 2021\n\nAbstract: This paper addresses the problem of the human operator cognitive workload estimation while controlling a robot. Being capable of assessing, in real-time, the operator's workload could help prevent calamitous events from occurring. This workload estimation could enable an AI to make informed decisions to assist or advise the operator, in an advanced human-robot interaction framework. We propose a method, named Fessonia, for real-time cognitive workload estimation from multiple parameters of an operator's driving behaviour via the use of behavioural entropy. Fessonia is comprised of: a method to calculate the entropy (i.e. unpredictability) of the operator driving behaviour profile; the Driver Profile Update algorithm which adapts the entropy calculations to the evolving driving profile of individual operators; and a Warning And Indication System that uses workload estimations to issue advice to the operator. Fessonia is evaluated in a robot teleoperation scenario that incorporated cognitively demanding secondary tasks to induce varying degrees of workload. The results demonstrate the ability of Fessonia to estimate different levels of imposed workload. Additionally, it is demonstrated that our approach is able to detect and adapt to the evolving driving profile of the different operators. Lastly, based on data obtained, a decrease in entropy is observed when a warning indication is issued, suggesting a more attentive approach focused on the primary navigation task.","Title: Nut Unfastening By Robotic Surface Exploration\nYear: 2021\n\nAbstract: In this paper, we present a novel concept and primary investigations regarding automated unfastening of hexagonal nuts by means of surface exploration with a compliant robot. In contrast to the conventional industrial approaches that rely on custom-designed motorised tools and mechanical tool changers, we propose to use robot fingers to position, grasp and unfasten unknown random-sized hexagonal nuts, which are arbitrarily positioned in the robot's task space. Inspired by how visually impaired people handle unknown objects, in this work, we use information observed from surface exploration to devise the unfastening strategy. It combines torque monitoring with active compliance for the robot fingers to smoothly explore the object's surface. We implement a shape estimation technique combining scaled iterative closest point and hypotrochoid approximation to estimate the location as well as contour profile of the hexagonal nut so as to accurately position the gripper fingers. We demonstrate this work in the context of dismantling an electrically driven vehicle battery pack. The experiments are conducted using a seven degrees of freedom (DoF)-compliant robot fitted with a two-finger gripper to unfasten four different sized randomly positioned hexagonal nuts. The obtained results suggest an overall exploration and unfastening success rate of 95% over an average of ten trials for each nut.","Title: Optimal grasp selection, and control for stabilising a grasped object, with respect to slippage and external forces\nYear: 2021\n\nAbstract: This paper explores the problem of how to grasp an object, and then control a robot arm so as to stabilise that object, under conditions where: i) there is significant slippage between the object and the robot’s fingers; and ii) the object is perturbed by external forces. For an n degrees of freedom (dof) robot, we treat the robot plus grasped object as an (n+1) dof system, where the grasped objec...","Title: Grasp Transfer for Deformable Objects by Functional Map Correspondence\nYear: 2022\n\nAbstract: Handling object deformations for robotic grasping is still a major problem to solve. In this paper, we propose an efficient learning-free solution for this problem where generated grasp hypotheses of a region of an object are adapted to its deformed configurations. To this end, we investigate the applicability of functional map (FM) correspondence, where the shape matching problem is treated as searching for correspondences between geometric functions in a reduced basis. For a user selected region of an object, a ranked list of grasp candidates is generated with local contact moment (LoCoMo) based grasp planner. The proposed FM-based methodology maps these candidates to an instance of the object that has suffered arbitrary level of deformation. The best grasp, by analysing its kinematic feasibility while respecting the original finger configuration as much as possible, is then executed on the object. We have compared the performance of our method with two different state-of-the-art correspondence mapping techniques in terms of grasp stability and region grasping accuracy for 4 different objects with 5 different deformations.","Title: Tracking linear deformable objects using slicing method\nYear: 2022\n\nAbstract: In this paper, an efficient novel method for tracking the linear deformable objects (LDOs) in real time is proposed. The method is developed based on recursively slicing a pointcloud into smaller pointclouds with sufficiently small variance. The performance of this method is investigated through a series of experiments with various camera resolutions in simulation when a robot end effector tracking an LDO using an RGBD camera, and in real word when the camera tracks a rope during a swing. The performance of the proposed method is compared with another state-of-the-art technique and the outcome is reported here.","Title: Predicting the Remaining Life of Lithium-ion Batteries Using a CNN-LSTM Model\nYear: 2022\n\nAbstract: Accurate predicting the remaining useful life of lithium-ion batteries is essential for the market of Electrical Vehicles (EVs) and the battery industry. However, diverse ageing processes, substantial battery variability, and dynamic operating circumstances are identified as main challenges for predicting the remaining useful life (RUL) of lithium-ion batteries (LIBs). This study proposes a machin...","Title: Robot-Assisted Nuclear Disaster Response: Report and Insights from a Field Exercise.\nYear: 2022\n\nAbstract: This paper reports on insights by robotics researchers that participated in a 5-day robot-assisted nuclear disaster response field exercise conducted by Kerntechnische Hilfdienst GmbH (KHG) in Karlsruhe, Germany. The German nuclear industry established KHG to provide a robot-assisted emergency response capability for nuclear accidents. We present a systematic description of the equipment used; the robot operators' training program; the field exercise and robot tasks; and the protocols followed during the exercise. Additionally, we provide insights and suggestions for advancing disaster response robotics based on these observations. Specifically, the main degradation in performance comes from the cognitive and attentional demands on the operator. Furthermore, robotic platforms and modules should aim to be robust and reliable in addition to their ease of use. Last, as emergency response stakeholders are often skeptical about using autonomous systems, we suggest adopting a variable autonomy paradigm to integrate autonomous robotic capabilities with the human-in-the-loop gradually. This middle ground between teleoperation and autonomy can increase end-user acceptance while directly alleviating some of the operator's robot control burden and maintaining the resilience of the human-in-the-loop."],"string":"[\n \"Title: Predicting the unobservable Visual 3D tracking with a probabilistic motion model.\\nYear: 2011\\n\\nAbstract: Visual tracking of an object can provide a powerful source of feedback information during complex robotic manipulation operations, especially those in which there may be uncertainty about which new object pose may result from a planned manipulative action. At the same time, robotic manipulation can provide a challenging environment for visual tracking, with occlusions of the object by other objects or by the robot itself, and sudden changes in object pose that may be accompanied by motion blur. Recursive filtering techniques use motion models for predictor-corrector tracking, but the simple models typically used often fail to adequately predict the complex motions of manipulated objects. We show how statistical machine learning techniques can be used to train sophisticated motion predictors, which incorporate additional information by being conditioned on the planned manipulative action being executed. We then show how these learned predictors can be used to propagate the particles of a particle filter from one predictor-corrector step to the next, enabling a visual tracking algorithm to maintain plausible hypotheses about the location of an object, even during severe occlusion and other difficult conditions. We demonstrate the approach in the context of robotic push manipulation, where a 5-axis robot arm equipped with a rigid finger applies a series of pushes to an object, while it is tracked by a vision algorithm using a single camera.\",\n \"Title: Learning to predict how rigid objects behave under simple manipulation.\\nYear: 2011\\n\\nAbstract: An important problem in robotic manipulation is the ability to predict how objects behave under manipulative actions. This ability is necessary to allow planning of object manipulations. Physics simulators can be used to do this, but they model many kinds of object interaction poorly. An alternative is to learn a motion model for objects by interacting with them. In this paper we address the problem of learning to predict the interactions of rigid bodies in a probabilistic framework, and demonstrate the results in the domain of robotic push manipulation. A robot arm applies random pushes to various objects and observes the resulting motion with a vision system. The relationship between push actions and object motions is learned, and enables the robot to predict the motions that will result from new pushes. The learning does not make explicit use of physics knowledge, or any pre-coded physical constraints, nor is it even restricted to domains which obey any particular rules of physics. We use regression to learn efficiently how to predict the gross motion of a particular object. We further show how different density functions can encode different kinds of information about the behaviour of interacting objects. By combining these as a product of densities, we show how learned predictors can cope with a degree of generalisation to previously unencountered object shapes, subjected to previously unencountered push directions. Performance is evaluated through a combination of virtual experiments in a physics simulator, and real experiments with a 5-axis arm equipped with a simple, rigid finger.\",\n \"Title: Optimal AUV path planning for extended missions in complex, fast-flowing estuarine environments\\nYear: 2007\\n\\nAbstract: This paper addresses the problems of automatically planning Autonomous Underwater Vehicle (AUV) paths which best exploit complex current data, from computational estuarine model forecasts, while also avoiding obstacles. In particular we examine the possibilities for a novel type of AUV mission deployment in fast flowing tidal river regions which experience bi-directional current flow. These environments are interesting in that, by choosing an appropriate path in space and time, an AUV may both bypass adverse currents which are too fast to be overcome by the vehicle's motors and also exploit favorable currents to achieve far greater speeds than the motors could otherwise provide, while substantially saving energy. The AUV can \\\"ride\\\" currents both up and down the river, enabling extended monitoring of otherwise energy-exhausting, fast flow environments. The paper discusses suitable path parameterizations, cost functions and optimization techniques which enable optimal AUV paths to be efficiently generated. These paths take maximum advantage of the river currents in order to minimize energy expenditure, journey time and other cost parameters. The resulting path planner can automatically suggest useful alternative mission start and end times and locations to those specified by the user. Examples are presented for navigation in a simple simulation of the fast flowing Hudson River waters around Manhattan.\",\n \"Title: Prediction learning in robotic pushing manipulation\\nYear: 2009\\n\\nAbstract: This paper addresses the problem of learning about the interactions of rigid bodies. A probabilistic frame- work is presented for predicting the motion of one rigid body following contact with another. We describe an algorithm for learning these predictions from observations, which does not make use of physics and is not restricted to domains with particular physics. We demonstrate the method in a scenario where a robot arm applies pushes to objects. The probabilistic nature of the algorithm enables it to generalize from learned examples, to successfully predict the resulting object motion for previously unseen object poses, push directions and new objects with novel shape. We evaluate the method with empirical experiments in a physics simulator. I. INTRODUCTION In early childhood, humans and animals learn models for predicting the interactions with the environment (1). It seems unlikely that these models comprise an explicit encoding of Newtonian physics, and so must instead rely on a learned relationship between observed actions and their outcomes. This paper addresses the problem of learning to predict the motion of one body, which results from a forced interaction with another. We have chosen to investigate this problem in the context of robotic \\\"poking\\\" or \\\"pushing\\\" operations, be- cause this includes a large number of unstable manipulations and hence provides interesting situations. However the work is potentially more general. An algorithm is presented which learns to predict the motions of a rigid object that will result from an applied robotic pushing action. The algorithm does not rely on any understanding or encoding of Newtonian mechanics, but can be trained in simple online experiments in which a robot arm applies random pushes to objects of interest and extracts the resulting motions using a vision system. Properties of objects, and their interactions, are learned as distributions. Pushing operations are encountered frequently in robotics, but have received relatively little attention in the research community. They are important in that robotic grasping frequently involves a pushing phase, when one finger or jaw of a gripper contacts the workpiece before another. Furthermore, pushing may often be preferable to pick and place type operations if the robot lacks the size or strength necessary to lift an object. (2) was the first to identify pushing operations as funda- mental to manipulation, especially grasping. Mason develops a detailed analysis of the mechanics of pushed, sliding objects and determines conditions required for various 2D motions of a pushed object. (3) attempts to put quantitative bounds on the rate at which these predicted motions occur. (4) developed a method for finding the set of all possible\",\n \"Title: Video with Ground-Truth for Validation of Visual Registration, Tracking and Navigation Algorithms\\nYear: 2005\\n\\nAbstract: A fundamental task in computer vision is that of determining the position and orientation of a moving camera relative to an observed object or scene. Many such visual tracking algorithms have been proposed in the computer vision, artificial intelligence and robotics literature over the past 30 years. Predominantly, these remain unvalidated since the ground-truth camera positions and orientations at each frame in a video sequence are not available for comparison with the outputs of the proposed vision systems. A method is presented for generating real visual test data with complete underlying ground-truth. The method enables the production of long video sequences, filmed along complicated six degree of freedom trajectories, featuring a variety of objects, in a variety of different visibility conditions, for which complete ground-truth data is known including the camera position and orientation at every image frame, intrinsic camera calibration data, a lens distortion model and models of the viewed objects.\",\n \"Title: An EM/E-MRF algorithm for adaptive model based tracking in extremely poor visibility\\nYear: 2008\\n\\nAbstract: This paper addresses the problems of visual tracking in conditions of extremely poor visibility. The human visual system can often correctly interpret images that are of such poor quality that they contain insufficient explicit information to do so. We assert that such systems must therefore make use of prior knowledge in several forms. A tracking algorithm is presented which combines observed data (the current image) with predicted data derived from prior knowledge of the object being viewed and an estimate of the camera's motion. During image segmentation, a predicted image is used to estimate class conditional distribution models and an Extended-Markov Random Field technique is used to combine observed image data with expectations of that data within a probabilistic framework. Interpretations of scene content and camera position are then mutually improved using Expectation Maximisation. Models of background and tracked object are continually relearned and adapt iteratively with each new image frame. The algorithm is tested using real video sequences, filmed in poor visibility conditions with complete pre-measured ground-truth data.\",\n \"Title: A novel selection evolutionary strategy for constrained optimization\\nYear: 2013\\n\\nAbstract: The existence of infeasible solutions makes it very difficult to handle constrained optimization problems (COPs) in a way that ensures efficient, optimal and constraint-satisfying convergence. Although further optimization from feasible solutions will typically lead in a direction that generates further feasible solutions, certain infeasible solutions can also provide useful information about the optimal direction of improvement for the objective function. How well an algorithm makes use of these two solutions determines its performance on COPs. This paper proposes a novel selection evolutionary strategy (NSES) for constrained optimization. A self-adaptive selection method is introduced to exploit both informative infeasible and feasible solutions from a perspective of combining feasibility with multi-objective problem (MOP) techniques. Since the global optimal solution of a COP is a feasible non-dominated solution, both non-dominated solutions with low constraint violation and feasible ones with low objective values are beneficial to an evolution process. Thus, the exploration and exploitation of both of these two kinds of solutions are preferred during the selection procedure. Several theorems and properties are given to prove the above assertion. Furthermore, the performance of our method is evaluated using 22 well-known benchmark functions. Experimental results show that the proposed method outperforms state-of-the-art algorithms in terms of the speed of finding feasible solutions and the stability of converging to global optimal solutions. In particular, when dealing with problems that have zero feasibility ratios and more than one active constraint, our method provides feasible solutions within fewer fitness evaluations (FES) and converges to the optimal solutions more reliably than other popular methods from the literature.\",\n \"Title: Change detection in SAR images by artificial immune multi-objective clustering.\\nYear: 2014\\n\\nAbstract: This paper addresses the problem of unsupervised change detection in Synthetic Aperture Radar (SAR) images. Previous approaches have used evolutionary clustering optimization methods, which can suffer from reduced accuracy, because they often use only a single objective function and can easily become trapped at locally optimal values. To overcome these difficulties, we propose a new approach which combines the artificial immune system (AIS) theory with a multi-objective optimization algorithm. First, the self-adaptive artificial immune multi-objective algorithm is adopted to pre-sort the difference image. During this procedure, the difference image is categorized into three classes – changed class, unchanged class and uncertain samples. Second, based on wavelet decomposition to extract features from the difference image, the immune clonal multi-objective clustering algorithm is used to search for the optimal clustering centers of uncertain samples, labeling them as changed or unchanged. Experimental comparisons with four state-of-the-art approaches show that the proposed algorithm can obtain a higher accuracy, is more robust to noise, and finds solutions which are more globally optimal. Additionally, the proposed algorithm can improve the local search ability for the optimal solutions and produces better cluster centers.\",\n \"Title: A Three-Component Fisher-Based Feature Weighting Method for Supervised PolSAR Image Classification\\nYear: 2015\\n\\nAbstract: This letter presents a feature weighting method for polarimetric synthetic aperture radar (PolSAR) image classification. Appropriate feature weighting is essential for obtaining accurate classifications but so far has remained an open research problem. We propose in this letter a supervised three-component feature weighting method based on the Fisher linear discriminant. Fisher linear discriminant method is used to calculate a coefficient for each feature. Then, these coefficients are modified according to a three-component scattering power decomposition model, combining both physical and statistical scattering characteristics to adapt them for the particular scattering mechanisms inherent in PolSAR data and assigned to the coherency matrix to enhance the discriminating ability of the features. Freeman decomposition and Wishart classifier are used to classify the PolSAR image. The effectiveness of the proposed method is demonstrated by experiments NASA/JPL AIRSAR L-band and CSA Radarsat-2 C-band PolSAR images of the San Francisco area.\",\n \"Title: Improved Memetic Algorithm Based on Route Distance Grouping for Multiobjective Large Scale Capacitated Arc Routing Problems.\\nYear: 2016\\n\\nAbstract: The capacitated arc routing problem (CARP) has attracted considerable attention from researchers due to its broad potential for social applications. This paper builds on, and develops beyond, the cooperative coevolutionary algorithm based on route distance grouping (RDG-MAENS), recently proposed by Mei et al. Although Mei's method has proved superior to previous algorithms, we discuss several remaining drawbacks and propose solutions to overcome them. First, although RDG is used in searching for potential better solutions, the solution generated from the decomposed problem at each generation is not the best one, and the best solution found so far is not used for solving the current generation. Second, to determine which sub-population the individual belongs to simply according to the distance can lead to an imbalance in the number of the individuals among different sub-populations and the allocation of resources. Third, the method of Mei et al. was only used to solve single-objective CARP. To overcome the above issues, this paper proposes improving RDG-MAENS by updating the solutions immediately and applying them to solve the current solution through areas shared, and then according to the magnitude of the vector of the route direction, and a fast and simple allocation scheme is proposed to determine which decomposed problem the route belongs to. Finally, we combine the improved algorithm with an improved decomposition-based memetic algorithm to solve the multiobjective large scale CARP (LSCARP). Experimental results suggest that the proposed improved algorithm can achieve better results on both single-objective LSCARP and multiobjective LSCARP.\",\n \"Title: Dynamic-context cooperative quantum-behaved particle swarm optimization based on multilevel thresholding applied to medical image segmentation.\\nYear: 2015\\n\\nAbstract: This paper proposes a dynamic-context cooperative quantum-behaved particle swarm optimization algorithm. The proposed algorithm incorporates a new method for dynamically updating the context vector each time it completes a cooperation operation with other particles. We first explain how this leads to enhanced search ability and improved optimization over previous methods, and demonstrate this empirically with comparative experiments using benchmark test functions. We then demonstrate a practical application of the proposed method, by showing how it can be applied to optimize the parameters for Otsu image segmentation for processing medical images. Comparative experimental results show that the proposed method outperforms other state-of-the-art methods from the literature.\",\n \"Title: A compressed sensing approach for efficient ensemble learning.\\nYear: 2014\\n\\nAbstract: This paper presents a method for improved ensemble learning, by treating the optimization of an ensemble of classifiers as a compressed sensing problem. Ensemble learning methods improve the performance of a learned predictor by integrating a weighted combination of multiple predictive models. Ideally, the number of models needed in the ensemble should be minimized, while optimizing the weights associated with each included model. We solve this problem by treating it as an example of the compressed sensing problem, in which a sparse solution must be reconstructed from an under-determined linear system. Compressed sensing techniques are then employed to find an ensemble which is both small and effective. An additional contribution of this paper, is to present a new performance evaluation method (a new pairwise diversity measurement) called the roulette-wheel kappa-error. This method takes into account the different weightings of the classifiers, and also reduces the total number of pairs of classifiers needed in the kappa-error diagram, by selecting pairs through a roulette-wheel selection method according to the weightings of the classifiers. This approach can greatly improve the clarity and informativeness of the kappa-error diagram, especially when the number of classifiers in the ensemble is large. We use 25 different public data sets to evaluate and compare the performance of compressed sensing ensembles using four different sparse reconstruction algorithms, combined with two different classifier learning algorithms and two different training data manipulation techniques. We also give the comparison experiments of our method against another five state-of-the-art pruning methods. These experiments show that our method produces comparable or better accuracy, while being significantly faster than the compared methods.\",\n \"Title: Two-level RRT planning for robotic push manipulation\\nYear: 2012\\n\\nAbstract: This paper presents an algorithm for planning sequences of pushes, by which a robotic arm equipped with a single rigid finger can move a manipulated object (or manipulandum) towards a desired goal pose. Pushing is perhaps the most basic kind of manipulation, however it presents difficult challenges for planning, because of the complex relationship between manipulative pushing actions and resulting manipulandum motions. The motion planning literature has well developed paradigms for solving e.g. the piano-mover's problem, where the search occurs directly in the configuration space of the manipulandum object being moved. In contrast, in pushing manipulation, a plan must be built in the action space of the robot, which is only indirectly linked to the motion space of the manipulandum through a complex interaction for which inverse models may not be known. In this paper, we present a two stage approach to planning pushing operations. A global RRT path planner is used to explore the space of possible manipulandum configurations, while a local push planner makes use of predictive models of pushing interactions, to plan sequences of pushes to move the manipulandum from one RRT node to the next. The effectiveness of the algorithm is demonstrated in simulation experiments in which a robot must move a rigid body through complex 3D transformations by applying only a sequence of simple single finger pushes.\",\n \"Title: Motion estimation using physical simulation\\nYear: 2010\\n\\nAbstract: We consider the task of monocular visual motion estimation from video image sequences. We hypothesise that performance on the task can be improved by incorporating an understanding of physically likely and feasible object dynamics. We test this hypothesis by incorporating a physical simulator into a least-squares estimation procedure. We initialise a full trajectory estimate using RANSAC followed by gradient descent refinement. We present results for 2D image sequences consisting of single ambiguous, visible or occluded balls, as well as results for 3D computer-generated sequences of objects in free-flight with added noise. Results suggest that restricting the estimation to allow only motions that are feasible according to the physics simulator can produce marked improvement when the observed object motion is within the limits of the physics simulator and its world model. Conversely, merely penalising deviations from feasible physical dynamics produces a consistent but incremental improvement over more common dynamics models.\",\n \"Title: Towards The Principled Study Of Variable Autonomy In Mobile Robots\\nYear: 2015\\n\\nAbstract: Safety critical and demanding tasks (e.g search and rescue or hazardous environments inspection), can benefit from robotic systems that offer a spectrum of control modes. These can range from direct teleoperation to full autonomy. This paper describes a pilot-study experiment in which a variable autonomy robot completes a navigation task. It explores the comparative performances of the human-robot system at different autonomy levels under different sets of conditions. This is done from a Mixed-Initiative system investigation perspective. Sensor noise was added to degrade robot performance, while a secondary task induced varying degrees of additional workload on the human operator. Carrying out these experiments and analyzing the initial results, has highlighted the profound complexities of designing tasks, conditions, and performance metrics which are: principled; eliminate confounding factors; and yield scientifically rigorous insights into the intricacies of a collaborative system that combines both human and robot intelligences. A key contribution of this paper is to describe the lessons learned from attempting these experiments, and to suggest a variety of guidelines for other researchers to consider when designing experiments in this context.\",\n \"Title: Single image super-resolution reconstruction based on genetic algorithm and regularization prior model.\\nYear: 2016\\n\\nAbstract: Single image super-resolution (SR) reconstruction is an ill-posed inverse problem because the high-resolution (HR) image, obtained from the low-resolution (LR) image, is non-unique or unstable. In this paper, single image SR reconstruction is treated as an optimization problem, and a new single image SR method, based on a genetic algorithm and regularization prior model, is proposed. In the proposed method, the optimization problem is constructed with a regularization prior model which consists of the non-local means (NLMs) filter, total variation (TV) and adaptive sparse domain selection (ASDS) scheme for sparse representation. In order to avoid local optimization, we combine the genetic algorithm and the iterative shrinkage algorithm to deal with the regularization prior model. Compared with several other state-of-the-art algorithms, the proposed method demonstrates better performances in terms of both numerical analysis and visual effect.\",\n \"Title: Subspace learning-based graph regularized feature selection.\\nYear: 2016\\n\\nAbstract: In recent years, a variety of feature selection algorithms based on subspace learning have been proposed. However, such methods typically do not exploit information about the underlying geometry of the data. To overcome this shortcoming, we propose a novel algorithm called subspace learning-based graph regularized feature selection (SGFS). SGFS builds on the feature selection framework of subspace learning, but extends it by incorporating the idea of graph regularization, in which a feature map is constructed on the feature space in order to preserve geometric structure information on the feature manifold. Additionally, the L2,1-norm is used to constrain the feature selection matrix to ensure the sparsity of the feature array and avoid trivial solutions. The resulting method can provide more accurate discrimination information for feature selection. We evaluate SGFS by comparing it against five other state-of-the-art algorithms from the literature, on twelve publicly available benchmark data sets. Empirical results suggest that SGFS is more effective than the other five feature selection algorithms.\",\n \"Title: Distractor-Supported Single Target Tracking In Extremely Cluttered Scenes\\nYear: 2016\\n\\nAbstract: This paper presents a novel method for single target tracking in RGB images under conditions of extreme clutter and camouflage, including frequent occlusions by objects with similar appearance as the target. In contrast to conventional single target trackers, which onlymaintain the estimated target status, we propose a multi-level clustering-based robust estimation for online detection and learning of multiple targetlike regions, called distractors, when they appear near to the true target. To distinguish the target from these distractors, we exploit a global dynamic constraint (derived from the target and the distractors) in a feedback loop to improve single target tracking performance in situations where the target is camouflaged in highly cluttered scenes. Our proposed method successfully prevents the estimated target location from erroneously jumping to a distractor during occlusion or extreme camouflage interactions. To gain an insightful understanding of the evaluated trackers, we have augmented publicly available benchmark videos, by proposing a new set of clutter and camouflage sub-attributes, and annotating these sub-attributes for all frames in all sequences. Using this dataset, we first evaluate the effect of each key component of the tracker on the overall performance. Then, the proposed tracker is compared to other highly ranked single target tracking algorithms in the literature. The experimental results show that applying the proposed global dynamic constraint in a feedback loop can improve single target tracker performance, and demonstrate that the overall algorithm significantly outperforms other state-of-the-art single target trackers in highly cluttered scenes.\",\n \"Title: Immune clonal algorithm based on directed evolution for multi-objective capacitated arc routing problem.\\nYear: 2016\\n\\nAbstract: Display OmittedThe non-dominant solutions obtained by three algorithms on some small and large-scale instances.In the above figure, the former two lines of figures are small-scale instances, in which DE-ICA shows the best convergence on 2 instances compared with other two algorithms. Besides, on other 4 instances of small-scale instances, DE-ICA is not worse than D-MAENS and ID-MAENS. In general, DE-ICA shows a slightly advantage on small scale instances. The reason is that small scale instances have a smaller solution space and they are easier to solve. Consequently, other two compared algorithms have enough capacity to deal with it and get a similar performance with DE-ICA.In the latter two lines of figures, DE-ICA gets the best convergence on 5 instances (e4a, s1a, s4a, G1-A, G2-E). On these instances, DE-ICA can reach areas with both low total cost and the low makespan. As a result, the non-dominant solutions obtained by DE-ICA almost completely dominate the non-dominant solutions obtained by D-MAENS and ID-MAENS. On the remaining instances, DE-ICA is also not worse than the other two algorithms in convergence. In conclusion, DE-ICA demonstrates an obvious advantage in convergence on these large scale instances. The results above show that the DE-ICA is more suitable for the large-scale problems. This paper proposes and Immune Clonal Algorithm Based on Directed Evolution for Multi-Objective Capacitated Arc Routing Problem.The proposed algorithm expands the scale of the initial population to increase the diversity.It helps the antibody populations to share the neighborhood information in time.It applies a brand-new kind of comparison operator.It can evolve in the direction of the better population. The capacitated arc routing problem is playing an increasingly important role in our society, engendering increasing attention from the research community. Among the various models, multi-objective capacitated arc routing problem comes much closer to real-world problems. Therefore, this paper proposes an immune clonal algorithm based on directed evolution to solve this problem. Firstly, the proposed algorithm adopts the framework of the immune clonal algorithm and expands the scale of the initial antibody population in the initialization process to increase the diversity of the antibodies. Secondly, the proposed algorithm is combined with a decomposition strategy in the operations of the immune gene. Antibodies are classified to perform the immune genetic operations, which helps the antibody populations to share the neighborhood information in a timely manner. Thirdly, the proposed algorithm applies a novel kind of comparison operator to build the total population, which helps it to evolve in the direction of a better population and improves the quality of the antibodies. Experimental results suggest that the proposed algorithm can generate better non-dominant solutions than several compared state-of-the-art algorithms, especially for large-scale sets.\",\n \"Title: A local-global coupled-layer puppet model for robust online human pose tracking.\\nYear: 2016\\n\\nAbstract: We propose a new method for online tracking of articulated human body poses.Our method offers online sequential tracking from one frame to the next.Many other methods mutually optimize poses offline over all frames of a sequence.We propose a novel cross-coupled global-local model of articulated human body pose.We propose an adaptive penalty function for optimizing the pose estimates. This paper addresses the problem of online tracking of articulated human body poses in dynamic environments. Many previous approaches perform poorly in realistic applications: often future frames or entire sequences are used anticausally to mutually refine the poses in each individual frame, making online tracking impossible; tracking often relies on strong assumptions about e.g. clothing styles, body-part colours and constraints on body-part motion ranges, limiting such algorithms to a particular dataset; the use of holistic feature models limits the ability of optimisation-based matching to distinguish between pose errors of different body parts. We overcome these problems by proposing a coupled-layer framework, which uses the previous notions of deformable structure (DS) puppet models. The underlying idea is to decompose the global pose candidate in any particular frame into several local parts to obtain a refined pose. We introduce an adaptive penalty with our model to improve the searching scope for a local part pose, and also to overcome the problem of using fixed constraints. Since the pose is computed using only current and previous frames, our method is suitable for online sequential tracking. We have carried out empirical experiments using three different public benchmark datasets, comparing two variants of our algorithm against four recent state-of-the-art (SOA) methods from the literature. The results suggest comparatively strong performance of our method, regardless of weaker constraints and fewer assumptions about the scene, and despite the fact that our algorithm is performing online sequential tracking, whereas the comparison methods perform mutual optimisation backwards and forwards over all frames of the entire video sequence.\",\n \"Title: Towards robotic decommissioning of legacy nuclear plant: Results of human-factors experiments with tele-robotic manipulation, and a discussion of challenges and approaches for decommissioning\\nYear: 2016\\n\\nAbstract: This paper explores the problems of developing robotic systems for decommissioning legacy nuclear infrastructure, such as the many contaminated gloveboxes present in UK, USA and other countries. We begin with a discussion of these decomissioning challenges. We review the current manual methods for decommissioning alpha-contaminated plant, and review robotic approaches which might replace such direct human interventions. We then present our initial experiments with human test-subjects, exploring the ability of humans to control a remote robot to perform complex manipulation tasks. Our preliminary results reveal a number of interesting lessons: conventional tele-manipulation is very difficult and very slow without significant training; metrics for usability of such technology can be conflicting and hard to interpret; aptitude for tele-manipulation varies significantly between individuals; however such aptitude may be predicted by using spatial awareness tests to select prospective robot operators; additionally the abilities of people with different initial aptitudes appear to converge somewhat as learning progresses. An additional contribution of this paper is to show how rigorous scientific methodologies, drawn from the psychology and human-factors research fields, can be used to analyse the performance of humans using robots to perform practical tasks.\",\n \"Title: Nonnegative Matrix Factorization with Rank Regularization and Hard Constraint.\\nYear: 2017\\n\\nAbstract: Nonnegative matrix factorization (NMF) is well known to be an effective tool for dimensionality reduction in problems involving big data. For this reason, it frequently appears in many areas of scientific and engineering literature. This letter proposes a novel semisupervised NMF algorithm for overcoming a variety of problems associated with NMF algorithms, including poor use of prior information,...\",\n \"Title: Guiding Trajectory Optimization by Demonstrated Distributions.\\nYear: 2017\\n\\nAbstract: Trajectory optimization is an essential tool for motion planning under multiple constraints of robotic manipulators. Optimization-based methods can explicitly optimize a trajectory by leveraging prior knowledge of the system and have been used in various applications such as collision avoidance. However, these methods often require a hand-coded cost function in order to achieve the desired behavio...\",\n \"Title: A fully end-to-end deep learning approach for real-time simultaneous 3D reconstruction and material recognition\\nYear: 2017\\n\\nAbstract: This paper addresses the problem of simultaneous 3D reconstruction and material recognition and segmentation. Enabling robots to recognise different materials (concrete, metal etc.) in a scene is important for many tasks, e.g. robotic interventions in nuclear decommissioning. Previous work on 3D semantic reconstruction has predominantly focused on recognition of everyday domestic objects (tables, chairs etc.), whereas previous work on material recognition has largely been confined to single 2D images without any 3D reconstruction. Meanwhile, most 3D semantic reconstruction methods rely on computationally expensive post-processing, using Fully-Connected Conditional Random Fields (CRFs), to achieve consistent segmentations. In contrast, we propose a deep learning method which performs 3D reconstruction while simultaneously recognising different types of materials and labeling them at the pixel level. Unlike previous methods, we propose a fully end-to-end approach, which does not require hand-crafted features or CRF post-processing. Instead, we use only learned features, and the CRF segmentation constraints are incorporated inside the fully end-to-end learned system. We present the results of experiments, in which we trained our system to perform real-time 3D semantic reconstruction for 23 different materials in a real-world application. The run-time performance of the system can be boosted to around 10Hz, using a conventional GPU, which is enough to achieve realtime semantic reconstruction using a 30fps RGB-D camera. To the best of our knowledge, this work is the first real-time end-to-end system for simultaneous 3D reconstruction and material recognition.\",\n \"Title: Weakly-supervised DCNN for RGB-D Object Recognition in Real-World Applications Which Lack Large-scale Annotated Training Data.\\nYear: 2017\\n\\nAbstract: This paper addresses the problem of RGBD object recognition in real-world applications, where large amounts of annotated training data are typically unavailable. To overcome this problem, we propose a novel, weakly-supervised learning architecture (DCNN-GPC) which combines parametric models (a pair of Deep Convolutional Neural Networks (DCNN) for RGB and D modalities) with non-parametric models (Gaussian Process Classification). Our system is initially trained using a small amount of labeled data, and then automatically prop- agates labels to large-scale unlabeled data. We first run 3D- based objectness detection on RGBD videos to acquire many unlabeled object proposals, and then employ DCNN-GPC to label them. As a result, our multi-modal DCNN can be trained end-to-end using only a small amount of human annotation. Finally, our 3D-based objectness detection and multi-modal DCNN are integrated into a real-time detection and recognition pipeline. In our approach, bounding-box annotations are not required and boundary-aware detection is achieved. We also propose a novel way to pretrain a DCNN for the depth modality, by training on virtual depth images projected from CAD models. We pretrain our multi-modal DCNN on public 3D datasets, achieving performance comparable to state-of-the-art methods on Washington RGBS Dataset. We then finetune the network by further training on a small amount of annotated data from our novel dataset of industrial objects (nuclear waste simulants). Our weakly supervised approach has demonstrated to be highly effective in solving a novel RGBD object recognition application which lacks of human annotations.\",\n \"Title: Learning modular and transferable forward models of the motions of push manipulated objects.\\nYear: 2017\\n\\nAbstract: The ability to predict how objects behave during manipulation is an important problem. Models informed by mechanics are powerful, but are hard to tune. An alternative is to learn a model of the object's motion from data, to learn to predict. We study this for push manipulation. The paper starts by formulating a quasi-static prediction problem. We then pose the problem of learning to predict in two different frameworks: (i) regression and (ii) density estimation. Our architecture is modular: many simple, object specific, and context specific predictors are learned. We show empirically that such predictors outperform a rigid body dynamics engine tuned on the same data. We then extend the density estimation approach using a product of experts. This allows transfer of learned motion models to objects of novel shape, and to novel actions. With the right representation and learning method, these transferred models can match the prediction performance of a rigid body dynamics engine for novel objects or actions.\",\n \"Title: Dynamic multi-level appearance models and adaptive clustered decision trees for single target tracking.\\nYear: 2017\\n\\nAbstract: A robust tracking algorithm for tracking arbitrary objects in challenging video sequences.An adaptive clustered decision tree approach which dynamically selects the minimum combination of features to represent target parts.This adaptive clustered decision tree is utilized both to enable robust matching at the part level and to select the new parts for learning. This paper presents a tracking algorithm for arbitrary objects in challenging video sequences. Targets are modelled at three different levels of granularity (pixel, parts and bounding box levels), which are cross-constrained to enable robust model relearning. The main contribution is an adaptive clustered decision tree method which dynamically selects the minimum combination of features necessary to sufficiently represent each target part at each frame, thereby providing robustness with computational efficiency. The adaptive clustered decision tree is used in two separate ways: firstly for parts level matching between successive frames; secondly to select the best candidate image regions for learning new parts of the target. We test the tracker using two different tracking benchmarks (VOT2013-2014 and CVPR2013 tracking challenges), based on two different test methodologies, and show it to be more robust than the state-of-the-art methods from both of those tracking challenges, while also offering competitive tracking precision. Additionally, we evaluate the contribution of each key component of the tracker to overall performance; test the sensitivity of the tracker under different initialization conditions; investigate the effect of using features in different orders within the decision trees; illustrate the flexibility of the method for handling arbitrary kinds of features, by showing how it easily extends to handle RGB-D data.\",\n \"Title: Autonomous vision-guided bi-manual grasping and manipulation\\nYear: 2017\\n\\nAbstract: This paper describes the implementation, demonstration and evaluation of a variety of autonomous, vision-guided manipulation capabilities, using a dual-arm Baxter robot. Initially, symmetric coordinated bi-manual manipulation based on kinematic tracking algorithm was implemented on the robot to enable a master-slave manipulation system. We demonstrate the efficacy of this approach with a human-robot collaboration experiment, where a human operator moves the master arm along arbitrary trajectories and the slave arm automatically follows the master arm while maintaining a constant relative pose between the two end-effectors. Next, this concept was extended to perform dual-arm manipulation without human intervention. To this extent, an image-based visual servoing scheme has been developed to control the motion of arms for positioning them at a desired grasp locations. Next we combine this with a dynamic position controller to move the grasped object using both arms in a prescribed trajectory. The presented approach has been validated by performing numerous symmetric and asymmetric bi-manual manipulations at different conditions. Our experiments demonstrated 80% success rate in performing the symmetric dual-arm manipulation tasks; and 73% success rate in performing asymmetric dualarm manipulation tasks.\",\n \"Title: Hybrid motion/force control: a review.\\nYear: 2017\\n\\nAbstract: This paper reviews hybrid motion/force control, a control scheme which enables robots to perform tasks involving both motion, in the free space, and interactive force, at the contacts. Motivated by the large amount of literature on this topic, we facilitate comparison and elucidate the key differences among different approaches. An emphasis is placed on the study of the decoupling of motion control and force control. And we conclude that it is indeed possible to achieve a complete decoupling; however, this feature can be relaxed or sacrificed to reduce the robot's joint torques while still completing the task.\",\n \"Title: Dense RGB-D semantic mapping with Pixel-Voxel neural network.\\nYear: 2018\\n\\nAbstract: In this paper, a novel Pixel-Voxel network is proposed for dense 3D semantic mapping, which can perform dense 3D mapping while simultaneously recognizing and labelling the semantic category each point in the 3D map. In our approach, we fully leverage the advantages of different modalities. That is, the PixelNet can learn the high-level contextual information from 2D RGB images, and the VoxelNet can learn 3D geometrical shapes from the 3D point cloud. Unlike the existing architecture that fuses score maps from different modalities with equal weights, we propose a softmax weighted fusion stack that adaptively learns the varying contributions of PixelNet and VoxelNet and fuses the score maps according to their respective confidence levels. Our approach achieved competitive results on both the SUN RGB-D and NYU V2 benchmarks, while the runtime of the proposed system is boosted to around 13 Hz, enabling near-real-time performance using an i7 eight-cores PC with a single Titan X GPU.\",\n \"Title: Mixed second order partial derivatives decomposition method for large scale optimization.\\nYear: 2017\\n\\nAbstract: •A theoretical analysis of the interaction between variables is developed.•Three theorems and three lemma are presented.•A decomposition method based on mixed second order partial derivatives is proposed.\",\n \"Title: Multi-objective artificial immune algorithm for fuzzy clustering based on multiple kernels.\\nYear: 2017\\n\\nAbstract: This paper presents a multi-objective artificial immune algorithm for fuzzy clustering based on multiple kernels (MAFC). MAFC extends the classical Fuzzy C-Means (FCM) algorithm and improves some of its important limitations, such as vulnerability to local optima convergence, which can lead to poor clustering quality. MAFC unifies multi-kernel learning and multi-objective optimization in a joint clustering framework, which preserves the geometric information of the dataset. The multi-kernel method maps data from the feature space to kernel space by using kernel functions. Additionally, the introduction of multi-objective optimization helps to optimize between-cluster separation and within-cluster compactness simultaneously via two different clustering validity criteria. These properties help the proposed algorithm to avoid becoming stuck at local optima. Furthermore, this paper utilizes an artificial immune algorithm to address the multi-objective clustering problem and acquire a Pareto optimal solution set. The solution set is obtained through the process of antibody population initialization, clone proliferation, non-uniform mutation and uniformity maintaining strategy, which avoids the problems of degradation and prematurity which can occur with conventional genetic algorithms. Finally, we choose the best solution from the Pareto optimal solution set. We use a semi-supervised method to achieve the final clustering results. We compare our method against state-of-the-art methods from the literature by performing experiments with both UCI datasets and face datasets. The results suggest that MAFC is significantly more efficient for clustering and has a wider scope of application.\",\n \"Title: Quantum-Inspired Immune Clonal Algorithm for solving large-scale capacitated arc routing problems.\\nYear: 2018\\n\\nAbstract: In this paper, we present an approach to Large-Scale CARP called Quantum-Inspired Immune Clonal Algorithm (QICA-CARP). This algorithm combines the feature of artificial immune system and quantum computation ground on the qubit and the quantum superposition. We call an antibody of population quantum bit encoding, in QICA-CARP. For this encoding, to control the population with a high probability evolution towards a good schema we use the information on the current optimal antibody. The mutation strategy of quantum rotation gate accelerates the convergence of the original clone operator. Moreover, quantum crossover operator enhances the exchange of information and increases the diversity of the population. Furthermore, it avoids falling into local optimum. We also use the repair operator to amend the infeasible solutions to ensure the diversity of solutions. This makes QICA-CARP approximating the optimal solution. We demonstrate the effectiveness of our approach by a set of experiments and by Comparing the results of our approach with ones obtained with the RDG-MAENS and RAM using different test sets. Experimental results show that QICA-CARP outperforms other algorithms in terms of convergence rate and the quality of the obtained solutions. Especially, QICA-CARP converges to a better lower bound at a faster rate illustrating that it is suitable for solving large-scale CARP.\",\n \"Title: Learning Monocular Visual Odometry With Dense 3d Mapping From Dense 3d Flow\\nYear: 2018\\n\\nAbstract: This paper introduces a fully deep learning approach to monocular SLAM, which can perform simultaneous localization using a neural network for learning visual odometry (L-VO) and dense 3D mapping. Dense 2D flow and a depth image are generated from monocular images by sub-networks, which are then used by a 3D flow associated layer in the L-VO network to generate dense 3D flow. Given this 3D flow, the dual-stream L-VO network can then predict the 6DOF relative pose and furthermore reconstruct the vehicle trajectory. In order to learn the correlation between motion directions, the Bivariate Gaussian modeling is employed in the loss function. The L-VO network achieves an overall performance of 2:68% for average translational error and 0.0143 degrees/m for average rotational error on the KITTI odometry benchmark. Moreover, the learned depth is leveraged to generate a dense 3D map. As a result, an entire visual SLAM system, that is, learning monocular odometry combined with dense 3D mapping, is achieved.\",\n \"Title: Robust Fusion of Color and Depth Data for RGB-D Target Tracking Using Adaptive Range-Invariant Depth Models and Spatio-Temporal Consistency Constraints.\\nYear: 2018\\n\\nAbstract: This paper presents a novel robust method for single target tracking in RGB-D images, and also contributes a substantial new benchmark dataset for evaluating RGB-D trackers. While a target object's color distribution is reasonably motion-invariant, this is not true for the target's depth distribution, which continually varies as the target moves relative to the camera. It is therefore nontrivial t...\",\n \"Title: Sensors, SLAM and Long-term Autonomy: A Review.\\nYear: 2018\\n\\nAbstract: Simultaneous Localization and Mapping, commonly known as SLAM, has been an active research area in the field of Robotics over the past three decades. For solving the SLAM problem, every robot is equipped with either a single sensor or a combination of similar/different sensors. This paper attempts to review, discuss, evaluate and compare these sensors. Keeping an eye on future, this paper also assesses the characteristics of these sensors against factors critical to the long- term autonomy challenge.\",\n \"Title: Weather Classification: A new multi-class dataset, data augmentation approach and comprehensive evaluations of Convolutional Neural Networks.\\nYear: 2018\\n\\nAbstract: Weather conditions often disrupt the proper functioning of transportation systems. Present systems either deploy an array of sensors or use an in-vehicle camera to predict weather conditions. These solutions have resulted in incremental cost and limited scope. To ensure smooth operation of all transportation services in all-weather conditions, a reliable detection system is necessary to classify weather in wild. The challenges involved in solving this problem is that weather conditions are diverse in nature and there is an absence of discriminate features among various weather conditions. The existing works to solve this problem have been scene specific and have targeted classification of two categories of weather. In this paper, we have created a new open source dataset consisting of images depicting three classes of weather i. e rain, snow and fog called RFS Dataset. A novel algorithm has also been proposed which has used super pixel delimiting masks as a form of data augmentation, leading to reasonable results with respect to ten Convolutional Neural Network architectures.\",\n \"Title: Real-Time Application Processing for FPGA-Based Resilient Embedded Systems in Harsh Environments.\\nYear: 2018\\n\\nAbstract: Real-time embedded systems nowadays get employed in harsh environments such as space, nuclear sites to carry out critical operations. Along with the traditional software based (CPU) execution, FPGAs are now also emerging as a bright prospect to accomplish such routines. However, these platforms are often get plagued by faults generated due to the high radiations in such environments. As a result, the real-time applications running on the platform could also get jeopardized. Thus, efficient execution of a set of hard real-time applications on reconfigurable systems with anomaly detection and recovery mechanism is inevitable. This work aims at tackling such problem with a \\\"healing\\\" approach for extreme environments. Initially, the applications are intelligently partitioned for hardware and software execution, then attempts have been made to schedule hardware applications with intermittent preemption point. Upon detecting any abnormality on such distinct points, our approach orchestrates a healing mechanism to remediate the scenario without hampering the pre-determined schedule. Experimental validation of our proposed method reveals its effectiveness.\",\n \"Title: Unsupervised feature selection based on kernel fisher discriminant analysis and regression learning\\nYear: 2019\\n\\nAbstract: In this paper, we propose a new feature selection method called kernel fisher discriminant analysis and regression learning based algorithm for unsupervised feature selection. The existing feature selection methods are based on either manifold learning or discriminative techniques, each of which has some shortcomings. Although some studies show the advantages of two-steps method benefiting from both manifold learning and discriminative techniques, a joint formulation has been shown to be more efficient. To do so, we construct a global discriminant objective term of a clustering framework based on the kernel method. We add another term of regression learning into the objective function, which can impose the optimization to select a low-dimensional representation of the original dataset. We use L2,1-norm of the features to impose a sparse structure upon features, which can result in more discriminative features. We propose an algorithm to solve the optimization problem introduced in this paper. We further discuss convergence, parameter sensitivity, computational complexity, as well as the clustering and classification accuracy of the proposed algorithm. In order to demonstrate the effectiveness of the proposed algorithm, we perform a set of experiments with different available datasets. The results obtained by the proposed algorithm are compared against the state-of-the-art algorithms. These results show that our method outperforms the existing state-of-the-art methods in many cases on different datasets, but the improved performance comes with the cost of increased time complexity.\",\n \"Title: Model-Free And Learning-Free Grasping By Local Contact Moment Matching\\nYear: 2018\\n\\nAbstract: This paper addresses the problem of grasping arbitrarily shaped objects, observed as partial point-clouds, without requiring: models of the objects, physics parameters, training data, or other a-priori knowledge. A grasp metric is proposed based on Local Contact Moment (LoCoMo). LoCoMo combines zero-moment shift features, of both hand and object surface patches, to determine local similarity. This metric is then used to search for a set of feasible grasp poses with associated grasp likelihoods. LoCoMo overcomes some limitations of both classical grasp planners and learning-based approaches. Unlike force-closure analysis, LoCoMo does not require knowledge of physical parameters such as friction coefficients, and avoids assumptions about fingertip contacts, instead enabling robust contacts of large areas of hand and object surface. Unlike more recent learning-based approaches, LoCoMo does not require training data, and does not need any prototype grasp configurations to be taught by kinesthetic demonstration. We present results of real-robot experiments grasping 21 different objects, observed by a wrist-mounted depth camera. All objects are grasped successfully when presented to the robot individually. The robot also successfully clears cluttered heaps of objects by sequentially grasping and lifting objects until none remain.\",\n \"Title: Choosing Grasps to Enable Collision-Free Post-Grasp Manipulations\\nYear: 2018\\n\\nAbstract: Consider the task of grasping the handle of a door, and then pushing it until the door opens. These two fundamental robotics problems (selecting secure grasps of a hand on an object, e.g. the door handle, and planning collision-free trajectories of a robot arm that will move that object along a desired path) have predominantly been studied separately from one another. Thus, much of the grasping literature overlooks the fundamental purpose of grasping objects, which is typically to make them move in desirable ways. Given a desired post-grasp trajectory of the object, different choices of grasp will often determine whether or not collision-free post-grasp motions of the arm can be found, which will deliver that trajectory. We address this problem by examining a number of possible stable grasping configurations on an object. For each stable grasp, we explore the motion space of the manipulator which would be needed for post-grasp motions, to deliver the object along the desired trajectory. A criterion, based on potential fields in the post-grasp motion space, is used to assign a collision-cost to each grasp. A grasping configuration is then selected which enables the desired post-grasp object motion while minimising the proximity of all robot parts to obstacles during motion. We demonstrate our method with peg-in-hole and pick-and-place experiments in cluttered scenes, using a Franka Panda robot. Our approach is effective in selecting appropriate grasps, which enable both stable grasp and also desired post-grasp movements without collisions. We also show that, when grasps are selected based on grasp stability alone, without consideration for desired post-grasp manipulations, the corresponding post-grasp movements of the manipulator may result in collisions.\",\n \"Title: A Fast Algorithm for SAR Image Segmentation Based on Key Pixels.\\nYear: 2017\\n\\nAbstract: Recent high-performance clustering methods process all pixels when segmenting an image, which results in a very large time complexity of these algorithms. Additionally, the performance of such algorithms can be severely affected by noise when dealing with highly polluted images. To address these problems, we propose a new unsupervised algorithm for segmenting synthetic aperture radar images based ...\",\n \"Title: A Spatial Fuzzy Clustering Algorithm With Kernel Metric Based on Immune Clone for SAR Image Segmentation.\\nYear: 2016\\n\\nAbstract: The fuzzy c-means (FCM) clustering algorithm has been widely used in image segmentation. However, FCM exhibits poor robustness to noise, often leading to unsatisfactory segmentations on noisy images. Additionally, the FCM algorithm is sensitive to the choice of initial cluster centers. In order to solve these problems, this paper proposes clone kernel spatial FCM (CKS_FCM), which improves segmenta...\",\n \"Title: An Enhanced Adaptive Coupled-Layer LGTracker++\\nYear: 2013\\n\\nAbstract: This paper addresses the problems of tracking targets which undergo rapid and significant appearance changes. Our starting point is a successful, state-of-the-art tracker based on an adaptive coupled-layer visual model [10]. In this paper, we identify four important cases when the original tracker often fails: significant scale changes, environment clutter, and failures due to occlusion and rapid disordered movement. We suggest four new enhancements to solve these problems: we adapt the scale of the patches in addition to adapting the bounding box, marginal patch distributions are used to solve patch drifting in environment clutter, a memory is added and used to assist recovery from occlusion, situations where the tracker may lose the target are automatically detected, and a particle filter is substituted for the Kalman filter to help recover the target. We demonstrate the advantages of the enhanced tracker over the original tracker using a test toolkit [17]. We demonstrate the advantages of the enhanced tracker over the original tracker, as well as several other state-of-the art trackers from the literature.\",\n \"Title: Non-Negative Spectral Learning and Sparse Regression-Based Dual-Graph Regularized Feature Selection.\\nYear: 2018\\n\\nAbstract: Feature selection is an important approach for reducing the dimension of high-dimensional data. In recent years, many feature selection algorithms have been proposed, but most of them only exploit information from the data space. They often neglect useful information contained in the feature space, and do not make full use of the characteristics of the data. To overcome this problem, this paper pr...\",\n \"Title: Region-sequence based six-stream CNN features for general and fine-grained human action recognition in videos.\\nYear: 2018\\n\\nAbstract: •We proposed a new method for human fine-grained action recognition in videos.•Our method uses a coarse pose estimation method to cut video frames and get human body foreground patch sequence.•Our method focuses on human lower arm area to enhance effective pixels for fine-grained actions.•We propose an encoding method to process the last pooling layer features of CNN structure.\",\n \"Title: SAR Targets Classification Based on Deep Memory Convolution Neural Networks and Transfer Parameters.\\nYear: 2018\\n\\nAbstract: Deep learning has obtained state-of-the-art results in a variety of computer vision tasks and has also been used to solve SAR image classification problems. Deep learning algorithms typically require a large amount of training data to achieve high accuracy. In contrast, the size of SAR image datasets is often comparatively limited. Therefore, this paper proposes a novel method, deep memory convolu...\",\n \"Title: A dynamic local cluster ratio-based band selection algorithm for hyperspectral images\\nYear: 2019\\n\\nAbstract: Hyperspectral band selection algorithms can save the computational costs during image restoration and analysis. This paper proposes a novel unsupervised band selection method, based on dynamic local cluster ratio (DLCR). The contributions of this paper can be summarized as follows. First, the similarity matrix is calculated in a novel way. Conventional approaches compute the matrix from the Euclidean distances between bands and are vulnerable to noise. Our proposed method can improve the robustness to such noise. Second, we propose an enhanced clustering strategy which clusters each band individually. Third, a dynamic ranking strategy is used to select bands iteratively. Bands that are highly correlated with each other will be prevented from being added to avoid redundancy. DLCR demonstrates improved performance on the Indian Pines and Pavia University data sets, when compared against other methods from the literature.\",\n \"Title: Dynamic grasp and trajectory planning for moving objects\\nYear: 2019\\n\\nAbstract: This paper shows how a robot arm can follow and grasp moving objects tracked by a vision system, as is needed when a human hands over an object to the robot during collaborative working. While the object is being arbitrarily moved by the human co-worker, a set of likely grasps, generated by a learned grasp planner, are evaluated online to generate a feasible grasp with respect to both: the current configuration of the robot respecting the target grasp; and the constraints of finding a collision-free trajectory to reach that configuration. A task-based cost function enables relaxation of motion-planning constraints, enabling the robot to continue following the object by maintaining its end-effector near to a likely pre-grasp position throughout the object’s motion. We propose a method of dynamic switching between: a local planner, where the hand smoothly tracks the object, maintaining a steady relative pre-grasp pose; and a global planner, which rapidly moves the hand to a new grasp on a completely different part of the object, if the previous graspable part becomes unreachable. Various experiments are conducted using a real collaborative robot and the obtained results are discussed.\",\n \"Title: Hypothesis-based Belief Planning for Dexterous Grasping.\\nYear: 2019\\n\\nAbstract: Belief space planning is a viable alternative to formalise partially observable control problems and, in the recent years, its application to robot manipulation problems has grown. However, this planning approach was tried successfully only on simplified control problems. In this paper, we apply belief space planning to the problem of planning dexterous reach-to-grasp trajectories under object pose uncertainty. In our framework, the robot perceives the object to be grasped on-the-fly as a point cloud and compute a full 6D, non-Gaussian distribution over the objectu0027s pose (our belief space). The system has no limitations on the geometry of the object, i.e., non-convex objects can be represented, nor assumes that the point cloud is a complete representation of the object. A plan in the belief space is then created to reach and grasp the object, such that the information value of expected contacts along the trajectory is maximised to compensate for the pose uncertainty. If an unexpected contact occurs when performing the action, such information is used to refine the pose distribution and triggers a re-planning. Experimental results show that our planner (IR3ne) improves grasp reliability and compensates for the pose uncertainty such that it doubles the proportion of grasps that succeed on a first attempt.\",\n \"Title: 2D Linear Time-Variant Controller for Human's Intention Detection for Reach-to-Grasp Trajectories in Novel Scenes.\\nYear: 2019\\n\\nAbstract: Designing robotic assistance devices for manipulation tasks is challenging. This work is concerned with improving accuracy and usability of semi-autonomous robots, such as human operated manipulators or exoskeletons. The key insight is to develop a system that takes into account context- and user-awareness to take better decisions in how to assist the user. The context-awareness is implemented by enabling the system to automatically generate a set of candidate grasps and reach-to-grasp trajectories in novel, cluttered scenes. The user-awareness is implemented as a linear time-variant feedback controller to facilitate the motion towards the most promising grasp. Our approach is demonstrated in a simple 2D example in which participants are asked to grasp a specific object in a clutter scene. Our approach also reduce the number of controllable dimensions for the user by providing only control on x- and y-axis, while orientation of the end-effector and the pose of its fingers are inferred by the system. The experimental results show the benefits of our approach in terms of accuracy and execution time with respect to a pure manual control.\",\n \"Title: Automatic Detection of Myocontrol Failures Based upon Situational Context Information\\nYear: 2019\\n\\nAbstract: Myoelectric control systems for assistive devices are still unreliable. The user's input signals can become unstable over time due to e.g. fatigue, electrode displacement, or sweat. Hence, such controllers need to be constantly updated and heavily rely on user feedback. In this paper, we present an automatic failure detection method which learns when plausible predictions become unreliable and model updates are necessary. Our key insight is to enhance the control system with a set of generative models that learn sensible behaviour for a desired task from human demonstration. We illustrate our approach on a grasping scenario in Virtual Reality, in which the user is asked to grasp a bottle on a table. From demonstration our model learns the reach-to-grasp motion from a resting position to two grasps (power grasp and tridigital grasp) and how to predict the most adequate grasp from local context, e.g. tridigital grasp on the bottle cap or around the bottleneck. By measuring the error between new grasp attempts and the model prediction, the system can effectively detect which input commands do not reflect the user's intention. We evaluated our model in two cases: i) with both position and rotation information of the wrist pose, and ii) with only rotational information. Our results show that our approach detects statistically highly significant differences in error distributions with p<;0.001 between successful and failed grasp attempts in both cases.\",\n \"Title: Metrics and Benchmarks for Remote Shared Controllers in Industrial Applications.\\nYear: 2019\\n\\nAbstract: Remote manipulation is emerging as one of the key robotics tasks needed in extreme environments. Several researchers have investigated how to add AI components into shared controllers to improve their reliability. Nonetheless, the impact of novel research approaches in real-world applications can have a very slow in-take. We propose a set of benchmarks and metrics to evaluate how the AI components of remote shared control algorithms can improve the effectiveness of such frameworks for real industrial applications. We also present an empirical evaluation of a simple intelligent share controller against a manually operated manipulator in a tele-operated grasping scenario.\",\n \"Title: Selective Ensemble Learning based Human Action Recognition Using Fusing Visual Features\\nYear: 2018\\n\\nAbstract: The selection of motion feature directly affects the recognition effect of human action recognition method. Single feature is often affected by human appearance, environment, camera settings and other factors, and its recognition effect is limited. This paper propose a novel action recognition method by using selective ensemble learning, which is a special paradigm of ensemble learning. Moreover, this paper presents a fast and efficient action description feature and a novel recognition algorithm. Robust discriminant mixed features are learnt from behavioral video frames as behavioral descriptors, The recogniton algorithm using selective ensemble learning can achieve fast classification. Experimental results show that the proposed method achieves ideal recognition results on the self-built indoor behavior data set and public data set.\",\n \"Title: Semi-Autonomous Behaviour Tree-Based Framework For Sorting Electric Vehicle Batteries Components\\nYear: 2021\\n\\nAbstract: The process of recycling electric vehicle (EV) batteries currently represents a significant challenge to the waste management automation industry. One example of it is the necessity of removing and sorting dismantled components from EV battery pack. This paper proposes a novel framework to semi-automate the process of removing and sorting different objects from an EV battery pack using a mobile manipulator. The work exploits the Behaviour Trees model for cognitive task execution and monitoring, which links different robot capabilities such as navigation, object tracking and motion planning in a modular fashion. The framework was tested in simulation, in both static and dynamic environments, and it was evaluated based on task time and the number of objects that the robot successfully placed in the respective containers. Results suggested that the robot's success rate in accomplishing the task of sorting the battery components was 95% and 82% in static and dynamic environments, respectively.\",\n \"Title: Trust, Shared Understanding and Locus of Control in Mixed-Initiative Robotic Systems\\nYear: 2021\\n\\nAbstract: This paper investigates how trust, shared under-standing between a human operator and a robot, and the Locus of Control (LoC) personality trait, evolve and affect Human-Robot Interaction (HRI) in mixed-initiative robotic systems. As such systems become more advanced and able to instigate actions alongside human operators, there is a shift from robots being perceived as a tool to being a team-mate....\",\n \"Title: Dual Quaternion-Based Visual Servoing for Grasping Moving Objects\\nYear: 2021\\n\\nAbstract: This paper presents a new dual quaternion-based formulation for pose-based visual servoing. Extending our previous work on local contact moment (LoCoMo) based grasp planning, we demonstrate grasping of arbitrarily moving objects in 3D space. Instead of using the conventional axis-angle parameterization, dual quaternions allow designing the visual servoing task in a more compact manner and provide ...\",\n \"Title: SpectGRASP: Robotic Grasping by Spectral Correlation\\nYear: 2021\\n\\nAbstract: This paper presents a spectral correlation-based method (SpectGRASP) for robotic grasping of arbitrarily shaped, unknown objects. Given a point cloud of an object, SpectGRASP extracts contact points on the object's surface matching the hand configuration. It neither requires offline training nor a-priori object models. We propose a novel Binary Extended Gaussian Image (BEGI), which represents the point cloud surface normals of both object and robot fingers as signals on a 2-sphere. Spherical harmonics are then used to estimate the correlation between fingers and object BEGIs. The resulting spectral correlation density function provides a similarity measure of gripper and object surface normals. This is highly efficient in that it is simultaneously evaluated at all possible finger rotations in SO(3). A set of contact points are then extracted for each finger using rotations with high correlation values. We then use our previous work, Local Contact Moment (LoCoMo) similarity metric, to sequentially rank the generated grasps such that the one with maximum likelihood is executed. We evaluate the performance of SpectGRASP by conducting experiments with a 7-axis robot fitted with a parallel-jaw gripper, in a physics simulation environment. Obtained results indicate that the method not only can grasp individual objects, but also can successfully clear randomly organized groups of objects. The SpectGRASP method also outperforms the closest state-of-the-art method in terms of grasp generation time and grasp-efficiency.\",\n \"Title: Towards Robotizing The Processes Of Testing Lithium-Ion Batteries\\nYear: 2021\\n\\nAbstract: To boost the circular economy of the electric vehicle battery industry, an accurate assessment of the state of health of retired batteries is essential to assign them an appropriate value in the post automotive market and material degradation before recycling. In practice, the advanced battery testing techniques are usually limited to laboratory benches at the battery cell level and hardly used in the industrial environment at the battery module or pack level. This necessitates developing battery recycling facilities that can handle the assessment and testing undertakings for many batteries with different form factors. Towards this goal, for the first time, this article proposes proof of concept to automate the process of collecting the impedance data from a retired 24kWh Nissan LEAF battery module. The procedure entails the development of robot end-of-arm tooling that was connected to a Potentiostat. In this study, the robot was guided towards a fixed battery module using visual servoing technique, and then impedance control system was applied to create compliance between the end-of-arm tooling and the battery terminals. Moreover, an alarm system was designed and mounted on the robot's wrist to check the connectivity between a Potentiostat and the battery terminals. Subsequently, the electrochemical impedance spectroscopy test was run over a wide range of frequencies at a 5% state of charge. The electrochemical impedance spectroscopy data obtained from the automated test is validated by means of the three criteria (linearity, causality and stability) and compared with manually collected measurements under the same conditions. Results suggested the proposed automated configuration can accurately accomplish the electrochemical impedance spectroscopy test at the battery module level with no human intervention, which ensures safety and allows this advanced testing technique to be adopted in grading retired battery modules.\",\n \"Title: A Bayesian-Based Approach to Human Operator Intent Recognition in Remote Mobile Robot Navigation\\nYear: 2021\\n\\nAbstract: This paper addresses the problem of human operator intent recognition during teleoperated robot navigation. In this context, recognition of the operator’s intended navigational goal, could enable an artificial intelligence (AI) agent to assist the operator in an advanced human-robot interaction framework. We propose a Bayesian Operator Intent Recognition (BOIR) probabilistic method that utilizes: ...\",\n \"Title: Fessonia - a Method for Real-Time Estimation of Human Operator Workload Using Behavioural Entropy.\\nYear: 2021\\n\\nAbstract: This paper addresses the problem of the human operator cognitive workload estimation while controlling a robot. Being capable of assessing, in real-time, the operator's workload could help prevent calamitous events from occurring. This workload estimation could enable an AI to make informed decisions to assist or advise the operator, in an advanced human-robot interaction framework. We propose a method, named Fessonia, for real-time cognitive workload estimation from multiple parameters of an operator's driving behaviour via the use of behavioural entropy. Fessonia is comprised of: a method to calculate the entropy (i.e. unpredictability) of the operator driving behaviour profile; the Driver Profile Update algorithm which adapts the entropy calculations to the evolving driving profile of individual operators; and a Warning And Indication System that uses workload estimations to issue advice to the operator. Fessonia is evaluated in a robot teleoperation scenario that incorporated cognitively demanding secondary tasks to induce varying degrees of workload. The results demonstrate the ability of Fessonia to estimate different levels of imposed workload. Additionally, it is demonstrated that our approach is able to detect and adapt to the evolving driving profile of the different operators. Lastly, based on data obtained, a decrease in entropy is observed when a warning indication is issued, suggesting a more attentive approach focused on the primary navigation task.\",\n \"Title: Nut Unfastening By Robotic Surface Exploration\\nYear: 2021\\n\\nAbstract: In this paper, we present a novel concept and primary investigations regarding automated unfastening of hexagonal nuts by means of surface exploration with a compliant robot. In contrast to the conventional industrial approaches that rely on custom-designed motorised tools and mechanical tool changers, we propose to use robot fingers to position, grasp and unfasten unknown random-sized hexagonal nuts, which are arbitrarily positioned in the robot's task space. Inspired by how visually impaired people handle unknown objects, in this work, we use information observed from surface exploration to devise the unfastening strategy. It combines torque monitoring with active compliance for the robot fingers to smoothly explore the object's surface. We implement a shape estimation technique combining scaled iterative closest point and hypotrochoid approximation to estimate the location as well as contour profile of the hexagonal nut so as to accurately position the gripper fingers. We demonstrate this work in the context of dismantling an electrically driven vehicle battery pack. The experiments are conducted using a seven degrees of freedom (DoF)-compliant robot fitted with a two-finger gripper to unfasten four different sized randomly positioned hexagonal nuts. The obtained results suggest an overall exploration and unfastening success rate of 95% over an average of ten trials for each nut.\",\n \"Title: Optimal grasp selection, and control for stabilising a grasped object, with respect to slippage and external forces\\nYear: 2021\\n\\nAbstract: This paper explores the problem of how to grasp an object, and then control a robot arm so as to stabilise that object, under conditions where: i) there is significant slippage between the object and the robot’s fingers; and ii) the object is perturbed by external forces. For an n degrees of freedom (dof) robot, we treat the robot plus grasped object as an (n+1) dof system, where the grasped objec...\",\n \"Title: Grasp Transfer for Deformable Objects by Functional Map Correspondence\\nYear: 2022\\n\\nAbstract: Handling object deformations for robotic grasping is still a major problem to solve. In this paper, we propose an efficient learning-free solution for this problem where generated grasp hypotheses of a region of an object are adapted to its deformed configurations. To this end, we investigate the applicability of functional map (FM) correspondence, where the shape matching problem is treated as searching for correspondences between geometric functions in a reduced basis. For a user selected region of an object, a ranked list of grasp candidates is generated with local contact moment (LoCoMo) based grasp planner. The proposed FM-based methodology maps these candidates to an instance of the object that has suffered arbitrary level of deformation. The best grasp, by analysing its kinematic feasibility while respecting the original finger configuration as much as possible, is then executed on the object. We have compared the performance of our method with two different state-of-the-art correspondence mapping techniques in terms of grasp stability and region grasping accuracy for 4 different objects with 5 different deformations.\",\n \"Title: Tracking linear deformable objects using slicing method\\nYear: 2022\\n\\nAbstract: In this paper, an efficient novel method for tracking the linear deformable objects (LDOs) in real time is proposed. The method is developed based on recursively slicing a pointcloud into smaller pointclouds with sufficiently small variance. The performance of this method is investigated through a series of experiments with various camera resolutions in simulation when a robot end effector tracking an LDO using an RGBD camera, and in real word when the camera tracks a rope during a swing. The performance of the proposed method is compared with another state-of-the-art technique and the outcome is reported here.\",\n \"Title: Predicting the Remaining Life of Lithium-ion Batteries Using a CNN-LSTM Model\\nYear: 2022\\n\\nAbstract: Accurate predicting the remaining useful life of lithium-ion batteries is essential for the market of Electrical Vehicles (EVs) and the battery industry. However, diverse ageing processes, substantial battery variability, and dynamic operating circumstances are identified as main challenges for predicting the remaining useful life (RUL) of lithium-ion batteries (LIBs). This study proposes a machin...\",\n \"Title: Robot-Assisted Nuclear Disaster Response: Report and Insights from a Field Exercise.\\nYear: 2022\\n\\nAbstract: This paper reports on insights by robotics researchers that participated in a 5-day robot-assisted nuclear disaster response field exercise conducted by Kerntechnische Hilfdienst GmbH (KHG) in Karlsruhe, Germany. The German nuclear industry established KHG to provide a robot-assisted emergency response capability for nuclear accidents. We present a systematic description of the equipment used; the robot operators' training program; the field exercise and robot tasks; and the protocols followed during the exercise. Additionally, we provide insights and suggestions for advancing disaster response robotics based on these observations. Specifically, the main degradation in performance comes from the cognitive and attentional demands on the operator. Furthermore, robotic platforms and modules should aim to be robust and reliable in addition to their ease of use. Last, as emergency response stakeholders are often skeptical about using autonomous systems, we suggest adopting a variable autonomy paradigm to integrate autonomous robotic capabilities with the human-in-the-loop gradually. This middle ground between teleoperation and autonomy can increase end-user acceptance while directly alleviating some of the operator's robot control burden and maintaining the resilience of the human-in-the-loop.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":67,"string":"67"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13572,"string":"13,572"}}},{"rowIdx":13573,"cells":{"question_id":{"kind":"string","value":"Fernando J Von Zuben"},"question":{"kind":"string","value":"Generate an abstract for the title \"Automatic feature selection for BCI: An analysis using the davies-bouldin index and extreme learning machines\" using the following items: 1. Feature selection\n2. Brain-Computer Interfaces (BCIs)\n3. Artificial immune network\n4. Extreme Learning Machine (ELM)\n5. Motor imagery paradigms"},"answer":{"kind":"string","value":"In this work, we present a novel framework for automatic feature selection in brain-computer interfaces (BCIs). The proposal, which manipulates features generated in the frequency domain by an estimate of the power spectral density of the EEG signals, is based on feature optimization (with both binary and real coding) using a state-of-the-art artificial immune network, the cob-aiNet. In order to analyze the performance of the proposed framework, two approaches are adopted: a direct use of the Davies-Bouldin index and the use of metrics associated with the operation of an extreme learning machine (ELM) in the role of a classifier. The results reveal that the proposal has the potential of improving the performance of a BCI system, and also provide elements for an analysis of the spectral content of EEG signals and of the performance of ELMs in motor imagery paradigms."},"correct_answer":{"kind":"string","value":"In this work, we present a novel framework for automatic feature selection in brain-computer interfaces (BCIs). The proposal, which manipulates features generated in the frequency domain by an estimate of the power spectral density of the EEG signals, is based on feature optimization (with both binary and real coding) using a state-of-the-art artificial immune network, the cob-aiNet. In order to analyze the performance of the proposed framework, two approaches are adopted: a direct use of the Davies-Bouldin index and the use of metrics associated with the operation of an extreme learning machine (ELM) in the role of a classifier. The results reveal that the proposal has the potential of improving the performance of a BCI system, and also provide elements for an analysis of the spectral content of EEG signals and of the performance of ELMs in motor imagery paradigms."},"documents":{"kind":"list like","value":["Title: Online learning in estimation of distribution algorithms for dynamic environments.\nYear: 2011\n\nAbstract: In this paper, we propose an estimation of distribution algorithm based on an inexpensive Gaussian mixture model with online learning, which will be employed in dynamic optimization. Here, the mixture model stores a vector of sufficient statistics of the best solutions, which is subsequently used to obtain the parameters of the Gaussian components. This approach is able to incorporate into the current mixture model potentially relevant information of the previous and current iterations. The online nature of the proposal is desirable in the context of dynamic optimization, where prompt reaction to new scenarios should be promoted. To analyze the performance of our proposal, a set of dynamic optimization problems in continuous domains was considered with distinct levels of complexity, and the obtained results were compared to the results produced by other existing algorithms in the dynamic optimization literature.","Title: An Immune-Evolutionary Algorithm for Multiple Rearrangements of Gene Expression Data\nYear: 2004\n\nAbstract: Microarray technologies are employed to simultaneously measure expression levels of thousands of genes. Data obtained from such experiments allow inference of individual gene functions, help to identify genes from specific tissues, to analyze the behavior of gene expression levels under various environmental conditions and under different cell cycle stages, and to identify inappropriately transcribed genes and several genetic diseases, among many other applications. As thousands of genes may be involved in a microarray experiment, computational tools for organizing and providing possible visualizations of the genes and their relationships are crucial to the understanding and analysis of the data. This work proposes an algorithm based on artificial immune systems for organizing gene expression data in order to simultaneously reveal multiple features in large amounts of data. A distinctive property of the proposed algorithm is the ability to provide a diversified set of high-quality rearrangements of the genes, opening up the possibility of identifying various co-regulated genes from representative graphical configurations of the expression levels. This is a very useful approach for biologists, because several co-regulated genes may exist under different conditions.","Title: Omni-aiNet: an immune-inspired approach for omni optimization\nYear: 2006\n\nAbstract: This work presents omni-aiNet, an immune-inspired algorithm developed to solve single and multi-objective optimization problems, either with single and multi-global solutions. The search engine is capable of automatically adapting the exploration of the search space according to the intrinsic demand of the optimization problem. This proposal unites the concepts of omni-optimization, already proposed in the literature, with distinctive procedures associated with immune-inspired concepts. Due to the immune inspiration, the omni-aiNet presents a population capable of adjusting its size during the execution of the algorithm, according to a predefined suppression threshold, and a new grid mechanism to control the spread of solutions in the objective space. The omni-aiNet was applied to several optimization problems and the obtained results are presented and analyzed.","Title: The Influence of the Pool of Candidates on the Performance of Selection and Combination Techniques in Ensembles\nYear: 2006\n\nAbstract: In this paper, we propose the use of an immuneinspired approach called opt-aiNet to generate a diverse set of high-performance candidates to compose an ensemble of neural network classifiers. Being a population-based search algorithm, the opt-aiNet is capable of maintaining diversity and finding many high-performance solutions simultaneously, which are known to be desired features when synthesizing an ensemble. Concerning the selection and combination phases, the most relevant selection and combination techniques already proposed in the literature have been considered. The main contribution of this paper is the indication that there is no pair of selection/combination technique that can be considered the best one, because the performance of the obtained ensemble varies significantly with the current composition of the pool of candidates already produced by the generation phase. Notwithstanding, this variability in performance is not restricted to the choice of opt-aiNet as the generative device. As a consequence, to overcome the performance of the best individual classifier, every possible pairs of selection and combination techniques should be tried. Only with such an exhaustive search (notice that the main computational burden is usually related to the generation phase), the performance of the ensemble was invariably superior to the performance of the best individual classifier on four benchmark classification problems.","Title: bicACO: An Ant Colony Inspired Biclustering Algorithm\nYear: 2008\n\nAbstract: A recent proposal developed to avoid some of the drawbacks presented by standard clustering algorithms is the so-called biclusteringtechnique [1], which performs clustering of rows and columns of the data matrix simultaneously, allowing the extraction of additional information from the dataset. Since the biclustering problem is combinatorial, and ant-based systems present several advantages when dealing with this kind of problems [2], in this work we propose an antinspired algorithm for biclustering, which was named bicACO.","Title: A multiobjective analysis of adaptive clustering algorithms for the definition of RBF neural network centers in regression problems\nYear: 2012\n\nAbstract: A variety of clustering algorithms have been applied to determine the internal structure of Radial Basis Function Neural Networks (RBFNNs). k-means algorithm is one of the most common choice for this task, although, like many other clustering algorithms, it needs to receive the number of prototypes a priori. This is a nontrivial procedure, mainly for real-world applications. An alternative is to use algorithms that automatically determine the number of prototypes. In this paper, we performed a multiobjective analysis involving three of these algorithms, which are: Adaptive Radius Immune Algorithm (ARIA), Affinity Propagation (AP), and Growing Neural Gas (GNG). For each one, the parameters that most influence the resulting number of prototypes composed the decision space, while the RBFNN RMSE and the number of prototypes formed the objective space. The experiments found that ARIA solutions achieved the best results for the multiobjective metrics adopted in this paper.","Title: An artificial immune network for multimodal function optimization on dynamic environments\nYear: 2005\n\nAbstract: Multimodal optimization algorithms inspired by the immune system are generally characterized by a dynamic control of the population size and by diversity maintenance along the search. One of the most popular proposals is denoted opt-aiNet (artificial immune network for optimization) and is extended here to deal with time-varying fitness functions. Additional procedures are designed to improve the overall performance and the robustness of the immune-inspired approach, giving rise to a version for dynamic optimization, denoted dopt-aiNet. Firstly, challenging benchmark problems in static multimodal optimization are considered to validate the new proposal. No parameter adjustment is necessary to adapt the algorithm according to the peculiarities of each problem. In the sequence, dynamic environments are considered, and usual evaluation indices are adopted to assess the performance of dopt-aiNet and compare with alternative solution procedures available in the literature.","Title: Multiple criteria optimization based on unsupervised learning and fuzzy inference applied to the vehicle routing problem\nYear: 2002\n\nAbstract: This paper presents a neuro-fuzzy system based on competitive learning to solve multiple criteria optimization problems. The proposed method promotes the simultaneous self-organization of several networks, employing unsupervised learning guided by a fuzzy rule base. The algorithm implements a policy of penalties and rewards, a strategy of neuron inhibition, insertion and pruning, and also takes into account certain statistical characteristics of the input space. A fuzzy inference system is designed to implement the decision making process under a multiobjective scenario, leading to an adaptive process of constraint relaxation. The effectiveness of the proposed method is attested by means of a series of computational simulations performed on standard data. In our simulations, we address two variants of the vehicle routing problem: the capacitated vehicle routing problem (CVRP) and the multiple traveling salesman problem (MTSP). There are a few works treating the vehicle routing problem by means of competitive learning. These approaches are briefly reviewed in this paper. We also present some improvements in the results of an implementation of tabu search by providing the solutions obtained by the neuro-fuzzy system as initial condition, showing that the proposed method can effectively produce satisfactory results when used in association with more dedicated approaches.","Title: Performance analysis of nonlinear echo state network readouts in signal processing tasks.\nYear: 2012\n\nAbstract: Echo state networks (ESNs) characterize an attractive alternative to conventional recurrent neural network (RNN) approaches as they offer the possibility of preserving, to a certain extent, the processing capability of a recurrent architecture and, at the same time, of simplifying the training process. However, the original ESN architecture cannot fully explore the potential of the RNN, given that only the second-order statistics of the signals are effectively used. In order to overcome this constraint, distinct proposals promote the use of a nonlinear readout aiming to explore higher-order available information though still maintaining a closed-form solution in the least-squares sense. In this work, we review two proposals of nonlinear readouts - a Volterra filter structure and an extreme learning machine - and analyze the performance of these architectures in the context of two relevant signal processing tasks: supervised channel equalization and chaotic time series prediction. The obtained results reveal that the nonlinear readout can be decisive in the process of aproximating the desired signal. Additionally, we discuss the possibility of combining both ideas of nonlinear readouts and preliminary results indicate that a performance improvement can be attained.","Title: Hybrid neural networks: An evolutionary approach with local search\nYear: 2002\n\nAbstract: Considering computational algorithms available in the literature, associated with supervised learning in feedforward neural networks, a wide range of distinct approaches can be identified. While the adjustment of the connection weights represents an omnipresent stage, the algorithms differ on three basic aspects: the technique chosen to determine the dimension of the multilayer neural network, the procedure adopted to specify the activation functions, and the kind of composition used to produce the output. Advanced learning algorithms should be developed to simultaneously treat all these aspects during learning, and an evolutionary learning algorithm with local search is proposed here. The essence of this approach is a synergy between genetic algorithms and conjugate gradient optimization, operating on a hybrid neural network architecture. As a consequence, the final neural network is automatically generated, and is characterized to be dedicated and computationally parsimonious.","Title: Evolving Phylogenetic Trees: an Alternative to Black-Box Approaches\nYear: 2002\n\nAbstract: This paper presents the main steps to reconstruct phylogenetic trees using an evolutionary algorithm with a wide range of alternative computational procedures for f itness evaluation and for tuning the search engine. Due to the fine equilibrium between exploration and exploitation of the search space, high- quality reconstructions are obtained among an explosion of candidates and with the additional requirement of adjusting the length of all branches in a given tree. A toolbox is available, and the matrix codification adopted, the a ssociated g e- netic operators, and the optimization procedures involved are clearly described to guarantee reproducibility.","Title: Recent Advances In Gene Expression Data Clustering: A Case Study With Comparative Results\nYear: 2004\n\nAbstract: Several advanced techniques have been proposed for data clustering and many of them have been applied to gene expression data, with partial success. The high dimensionality and the multitude of admissible perspectives for data analysis of gene expression require additional computational resources, such as hierarchical structures and dynamic allocation of resources. We present an immune-inspired hierarchical clustering device, called hierarchical artificial immune network (HaiNet), especially devoted to the analysis of gene expression data. This technique was applied to a newly generated data set, involving maize plants exposed to different aluminum concentrations. The performance of the algorithm was compared with that of a self-organizing map, which is commonly adopted to deal with gene expression data sets. More consistent and informative results were obtained with HaiNet.","Title: MOBAIS: A Bayesian Artificial Immune System for Multi-Objective Optimization\nYear: 2008\n\nAbstract: Significant progress has been made in theory and design of artificial immune systems (AISs) for solving multi-objective problems accurately. However, an aspect not yet widely addressed by the research reported in the literature is the lack of ability of the AIS to deal effectively with building blocks (high-quality partial solutions coded in the antibody). The available AISs present mechanisms for evolving the population that do not take into account the relationship among the variables of the problem, causing the disruption of these high-quality partial solutions. Recently, we proposed a novel immune-inspired approach for single-objective optimization as an attempt to avoid this drawback. Our proposal replaces the traditional mutation and cloning operators with a probabilistic model, more specifically a Bayesian network representing the joint distribution of promising solutions and, subsequently, uses this model for sampling new solutions. Now, in this paper we extend our methodology for solving multi-objective optimization problems. The proposal, called Multi-Objective Bayesian Artificial Immune System (MOBAIS), was evaluated in the well-known multi-objective Knapsack problem and its performance compares favorably with that produced by contenders such as NSGA-II, MISA, and mBOA.","Title: Immune-inspired Dynamic Optimization for Blind Spatial Equalization in Undermodeled Channels.\nYear: 2006\n\nAbstract: In this work, we propose an evolutionary-like approach to the problem of blind adaptive spatial filtering that is based on the decision-directed criterion and on the doptaiNet, an artificial immune network conceived to perform multimodal search in dynamic environments. The proposal was tested under static and time-varying undermodeled channel models, and, in all cases, its ability to find and track a solution close to the Wiener global optimum was attested. The obtained results reveal that the dopt-aiNet may decisively enhance the performance of adaptive arrays in scenarios built from elements that are representative of some aspects of real-world communication systems.","Title: Applying biclustering to text mining: an immune-inspired approach\nYear: 2007\n\nAbstract: With the rapid development of information technology, computers are proving to be a fundamental tool for the organization and classification of electronic texts, given the huge amount of available information. The existent methodologies for text mining apply standard clustering algorithms to group similar texts. However, these algorithms generally take into account only the global similarities between the texts and assign each one to only one cluster, limiting the amount of information that can be extracted from the texts. An alternative proposal capable of solving these drawbacks is the biclustering technique. The biclustering is able to perform clustering of rows and columns simultaneously, allowing a more comprehensive analysis of the texts. The main contribution of this paper is the development of an immune-inspired biclustering algorithm to carry out text mining, denoted BIC-aiNet. BIC-aiNet interprets the biclustering problem as several two-way bipartition problems, instead of considering a single two-way permutation framework. The experimental results indicate that our proposal is able to group similar texts efficiently and extract implicit useful information from groups of texts.","Title: Evolutionary Stigmergy In Multipurpose Navigation Systems\nYear: 2006\n\nAbstract: Autonomous robot navigation involves many challenges and difficulties which are augmented when multiple robots operate together. Sophisticated computational techniques are required to cope with autonomous navigation in collective robotics, being the biologically-inspired approaches the most frequently adopted. Stigmergy, i.e. the ants communication by means of pheromones, is the main biological metaphor used in this work to perform multi-robot communication. The robots will be able to mark regions of the environment with artificial pheromones, according to past experiences, assisting one another in a cooperative and indirect way to accomplish the navigation objectives. Each robot is controlled by an autonomous navigation system (ANS) based on Learning Classifier System, which evolves during navigation from no a priori knowledge. Besides learning to avoid obstacles and capture targets, the systems must also learn how and where to lay artificial pheromones. Some experiments and simulations are performed intending to particularly investigate the ANS from three main perspectives: capability of learning to achieve the navigation objectives in collective scenarios, adaptability in face of environmental changes and ability to obtain optimized navigation behaviors by means of stigmergy.","Title: Artificial homeostatic system: a novel approach\nYear: 2005\n\nAbstract: Many researchers are developing frameworks inspired by natural, especially biological, systems to solve complex real-world problems. This work extends previous work in the field of biologically inspired computing, proposing an artificial endocrine system for autonomous robot navigation. Having intrinsic self-organizing behaviour, the novel artificial endocrine system can be applied to a wide range of problems, particularly those that involve decision making under changing environmental conditions, such as autonomous robot navigation. This work draws on “embodied cognitive science”, including the study of intelligence, adaptivity, homeostasis, and the dynamic aspects of cognition, in order to help lay down fundamental principles and techniques for a novel approach to more biologically plausible artificial homeostatic systems. Results from using the artificial endocrine system to control a simulated robot are presented.","Title: A complex neighborhood based particle swarm optimization\nYear: 2009\n\nAbstract: This paper proposes a new variant of the PSO algorithm named Complex Neighborhood Particle Swarm Optimizer (CNPSO) for solving global optimization problems. In the CNPSO, the neighborhood of the particles is organized through a complex network which is modified during the search process. This evolution of the topology seeks to improve the influence of the most successful particles and it is fine tuned for maintaining the scale-free characteristics of the network while the optimization is being performed. The use of a scale-free topology instead of the usual regular or global neighborhoods is intended to bring to the search procedure a better capability of exploring promising regions without a premature convergence, which would result in the procedure being easily trapped in a local optimum. The performance of the CNPSO is compared with the standard PSO on some well-known and high-dimensional benchmark functions, ranging from multimodal to plateau-like problems. In all the cases the CNPSO outperformed the standard PSO.","Title: Data clustering with a neuro-immune network\nYear: 2005\n\nAbstract: This paper proposes a novel constructive learning algorithm for a competitive neural network. The proposed algorithm is developed by taking ideas from the immune system and demonstrates robustness for data clustering in the initial experiments reported here for three benchmark problems. Comparisons with results from the literature are also provided. To automatically segment the resultant neurons at the output, a tool from graph theory was used with promising results. A brief sensitivity analysis of the algorithm was performed in order to investigate the influence of the main user-defined parameters on the learning speed and accuracy of the results presented. General discussions and avenues for future works are also provided.","Title: Genetic programming for automating the development of data management algorithms in information technology systems\nYear: 2012\n\nAbstract: Information technology (IT) systems are present in almost all fields of human activity, with emphasis on processing, storage, and handling of datasets. Automated methods to provide access to data stored in databases have been proposed mainly for tasks related to knowledge discovery and datamining (KDD). However, for this purpose, the database is used only to query data in order to find relevant patterns associated with the records. Processes modelled on IT systems should manipulate the records to modify the state of the system. Linear genetic programming for databases (LGPDB) is a tool proposed here for automatic generation of programs that can query, delete, insert, and update records on databases. The obtained results indicate that the LGPDB approach is able to generate programs for effectively modelling processes of IT systems, opening the possibility of automating relevant stages of data manipulation, and thus allowing human programmers to focus on more complex tasks.","Title: Applying Biclustering to Perform Collaborative Filtering\nYear: 2007\n\nAbstract: Collaborative filtering (CF) is a method to perform automated suggestions for a user based on the opinion of other users with similar interest. Most of the CF algorithms do not take into account the existent duality between users and items, considering only the similarities between users or only the similarities between items. In this paper we propose a novel methodology for the CF capable of dealing with this situation. By proposing an immune-inspired biclustering technique to carry out clustering of rows and columns at the same time, our algorithm is able to group similarities between users and items. In order to evaluate the proposed methodology, we have applied it to MovieLens dataset which contains user's ratings to a large set of movies. The results indicate that our proposal is able to provide useful recommendations for the users, outperforming other methodologies for CF reported in the literature.","Title: Definition of Capacited p-Medians by a Modified Max Min Ant System with Local Search\nYear: 2004\n\nAbstract: This work introduces a modified MAX MIN Ant System (MMAS) designed to solve the Capacitated p-Medians Problem (CPMP). It presents the most relevant steps towards the implementation of an MMAS to solve the CPMP, including some improvements on the original MMAS algorithm, such as the use of a density model in the information heuristics and a local search adapted from the uncapacitated p-medians problem. Extensions to a recently proposed updating rule for the pheromone level, aiming at improving the MMAS ability to deal with large-scale instances, are also presented and discussed. Some simulations are performed using instances available from the literature, and well-known heuristics are employed for benchmarking.","Title: Evolving Arbitrarily Connected Feedforward Neural Networks via Genetic Algorithms\nYear: 2010\n\nAbstract: Though several approaches have already been proposed in the literature to evolve neural network topologies for solving a wide range of machine learning tasks, this paper presents an alternative one, capable of evolving arbitrarily connected feed forward neural networks (ACFNNs), including linear and nonlinear neurons. A genetic algorithm is conceived to adjust the topology and also to perform variable selection. The weights of the obtained neural networks, with arbitrary topologies, are adjusted using a simple descent gradient algorithm. The purpose is to obtain high-quality and parsimonious predictors for two real-world and one synthetic time series. The obtained results are compared with the ones produced by traditional MLP models and Mixtures of Heterogeneous Experts (MHEs).","Title: A constructive algorithm to synthesize arbitrarily connected feedforward neural networks\nYear: 2012\n\nAbstract: In this work we present a constructive algorithm capable of producing arbitrarily connected feedforward neural network architectures for classification problems. Architecture and synaptic weights of the neural network should be defined by the learning procedure. The main purpose is to obtain a parsimonious neural network, in the form of a hybrid and dedicate linear/nonlinear classification model, which can guide to high levels of performance in terms of generalization. Though not being a global optimization algorithm, nor a population-based metaheuristics, the constructive approach has mechanisms to avoid premature convergence, by mixing growing and pruning processes, and also by implementing a relaxation strategy for the learning error. The synaptic weights of the neural networks produced by the constructive mechanism are adjusted by a quasi-Newton method, and the decision to grow or prune the current network is based on a mutual information criterion. A set of benchmark experiments, including artificial and real datasets, indicates that the new proposal presents a favorable performance when compared with alternative approaches in the literature, such as traditional MLP, mixture of heterogeneous experts, cascade correlation networks and an evolutionary programming system, in terms of both classification accuracy and parsimony of the obtained classifier.","Title: Evolving phylogenetic trees: a multiobjective approach\nYear: 2007\n\nAbstract: This work presents the application of the omni-aiNet algorithm - an immune-inspired algorithm originally developed to solve single and multiobjective optimization problems - to the construction of phylogenetic trees. The main goal of this work is to automatically evolve a population of phylogenetic unrooted trees, possibly with distinct topologies, by minimizing at the same time the minimal evolution and the mean-squared error criteria. The obtained set of phylogenetic trees contains non-dominated individuals that form the Pareto front and that represent the trade-off of the two conflicting objectives. The proposal of multiple non-dominated solutions in a single run gives to the user the possibility of having distinct explanations for the difference observed in the terminal nodes of the tree, and also indicates the restrictive feedback provided by the individual application of well-known algorithms for phylogenetic reconstruction that takes into account both optimization criteria, like Neighbor Joining.","Title: A concentration-based artificial immune network for multi-objective optimization\nYear: 2011\n\nAbstract: Until recently, the main focus of researchers that develop algorithms for evolutionary multi-objective optimization has been the creation of mechanisms capable of obtaining sets of solutions that are as close as possible to the true Pareto front of the problem and also as diverse as possible in the objective space, to properly cover such front. However, an adequate maintenance of diversity in the decision space is also important, to efficiently solve several classes of problems and even to facilitate the post-optimization decision making process. This aspect has been widely studied in evolutionary single-objective optimization, what led to the development of several diversity maintenance techniques. Among them, the recently proposed concentration-based artificial immune network (cob-aiNet), which is capable of self-regulating the population size, presented promising results in multimodal problems. So, it is extended here to deal with multi-objective problems that require a proper maintenance of diversity in the decision space.","Title: Autonomous navigation system applied to collective robotics with ant-inspired communication\nYear: 2005\n\nAbstract: Research in collective robotics is motivated mainly by the possibility of achieving an efficient solution to multi-objective navigation tasks when multiple robots are employed, instead of a single robot. Several approaches have already been tried in multi-robot systems, but the bio-inspired ones are the most frequent. This paper proposes to augment an autonomous navigation system based on learning classifier systems for using in collective robotics, introducing an inter-robot communication mechanism inspired by ant stigmergy, with each robot acting independently and cooperatively. The navigation system has no innate basic behavior and all knowledge necessary to compose the decision-making artifact is evolved as a function of the environmental feedback only, during navigation. Repulsive and/or attractive pheromone trails are produced by the robots along navigation, following very simple rules. Basically, each robot has to perform obstacle avoidance and target search, and the status of the pheromone level at the position currently occupied by each robot will influence the coordination of the two fundamental behaviors. Experiments are performed in simulation, with comparative results indicating that the presence of the pheromone trails is responsible for significant improvements in the capture rate and in the length of the route adopted by each robot.","Title: Mixture of Experts Applied to Nonlinear Dynamic Systems Identification: A Comparative Study\nYear: 2002\n\nAbstract: A mixture of experts (ME) model provides a modularapproach wherein component neural networks are madespecialists on subparts of a problem. In this framework,that follows the \"divide-and-conquer\" philosophy, agating network learns how to softly partition the inputspace into regions to be each properly modeled by one ormore expert networks. In this paper, we investigate theapplication of different ME variants to some multivariatenonlinear dynamic systems identification problems whichare known to be difficult to be dealt with. The aim is toprovide a comparative performance analysis betweenvariable settings of the standard, gated, and localized MEmodels with more conventional NN models.","Title: ArTVox: Evolutionary Composition in Visual and Sound Domains\nYear: 2006\n\nAbstract: Computational creativity is certainly interesting and potentially important. ArTVox, a Java programmed evolutionary environment, arose from the attempt to emulate computational creativity applied to artistic production, in visual and sound domains, by using interactive genetic algorithms. Objects inspired in Kandinsky's artworks are being programmed in Shape, an auxiliary Java environment, to be inserted in ArTVox. Today, ArTVox creates and evolves visual compositions of geometric primitives that, by their turn, guide the sound production in another evolutionary environment, JaVox, integrated to ArTVox.","Title: A multi-Gaussian component EDA with restarting applied to direction of arrival tracking\nYear: 2013\n\nAbstract: This paper analyzes the application of a multi-population Gaussian-based estimation of distribution algorithm equipped with a restarting strategy and mutation, named MGcEDA, to the problem of estimating the Direction of Arrival (DOA) of time-varying plane waves impinging on a uniform linear array of sensors. This problem requires the minimization of a dynamic cost function which is non-linear, non-quadratic, multimodal and variant with respect to the signal-to-noise ratio. Experiments showed that MGcEDA was able to quickly respond to changes in the source features in scenarios with different levels of noise and number of signals. Moreover, MGcEDA outperforms a previously proposed approach in all considered experiments in terms of well known performance measures.","Title: Data Partition And Variable Selection For Time Series Prediction Using Wrappers\nYear: 2006\n\nAbstract: The purpose of this paper is a comparative study of a non-exhaustive, though representative, set of methodologies already available for the partition of the training dataset in time series prediction, and also for variable selection under the wrapper paradigm. The partition policy of the training dataset and the choice of a proper set of variables for the regression vector are known to have a significant influence in the accuracy of the predictor, no matter the choice of the prediction model. However, there has been no extensive search for a figure of merit supporting a comparative analysis. Here, two partition policies, denoted sequential and random, are compared, and among the Variable selection approaches using wrappers, forward selection is contrasted with sensitivity based pruning. Five real financial time series with trends and seasonality have been considered and multilayer perceptrons are adopted as the predictor. The obtained results indicate with high confidence that the rarely adopted random partition and the computationally intensive forward selection overcomes the contestants in the whole set of experiments.","Title: An Immune-Inspired Approach to Bayesian Networks\nYear: 2005\n\nAbstract: During the last few years, there has been an increasing interest in using heuristic search algorithms based on natural selection (the so-called \"evolutionary algorithms\") for solving a wide variety of problems. As in any other discipline, research on ...","Title: Handling Time-Varying Tsp Instances\nYear: 2006\n\nAbstract: Multimodal optimization algorithms are being adapted to deal with dynamic optimization, mainly due to their ability to provide a faster reaction to unexpected changes in the optimization surface. The faster reaction may be associated with the existence of two important attributes in population-based algorithms devoted to multimodal optimization: simultaneous maintenance of multiple local optima in the population; and self-regulation of the population size along the search. The optimization surface may be subject to variations motivated by one of two main reasons: modification of the objectives to be fulfilled and change in parameters of the problem. An immune-inspired algorithm specially designed to deal with combinatorial optimization is applied here to solve time-varying TSP instances, with the cost of going from one city to the other being a function of time. The proposal presents favorable results when compared to the results produced by a high-performance ant colony optimization algorithm of the literature.","Title: On-Line Approach for Loss Reduction in Electric Power Distribution Networks Using Learning Classifier Systems\nYear: 2001\n\nAbstract: The problem of minimization of energy losses in power distribution systems can be formulated as obtaining the \"best\" network configuration, through the manipulation of sectionalizing switches. Using graph terminology, we have a combinatorial optimization problem, whose solution corresponds to finding a minimum spanning tree for the network. As an on-line approach to loss reduction in electric power distribution networks, this paper relies on Learning Classifier Systems to continually proposed network configurations close to the one associated with minimum energy losses, in the case of time-varying profiles of energy requirement. In order to evolve the set of rules that composes the Classifier System, operators for selection, reproduction and mutation are applied. Case studies illustrate the possibilities of this approach.","Title: An Integrated System For Phylogenetic Inference Using Evolutionary Algorithms\nYear: 2002\n\nAbstract: This paper presents all the steps to reconstruct phylogenetic trees using an evolutionary algorithm. The fitness criterion is based on maximum likelihood and a toolbox is available. The codification, genetic operators and the optimization procedures involved are clearly described to guarantee reproducibility.","Title: Hybridizing mixtures of experts with support vector machines: Investigation into nonlinear dynamic systems identification\nYear: 2007\n\nAbstract: Mixture of experts (ME) models comprise a family of modular neural network architectures aiming at distilling complex problems into simple subtasks. This is done by deploying a separate gating module for softly dividing the input space into overlapping regions to be each assigned to one or more expert networks. Conversely, support vector machines (SVMs) refer to kernel-based methods, neural-network-alike models that constitute an approximate implementation of the structural risk minimization principle. Such learning machines follow the simple, but powerful idea of nonlinearly mapping input data into high-dimensional feature spaces wherein a linear decision surface discriminating different regions is properly designed. In this work, we formally characterize and empirically evaluate a novel approach, named as Mixture of Support Vector Machine Experts (MSVME), whose main purpose is to combine the complementary properties of both SVM and ME models. In the formal characterization, an algorithm based on a maximum likelihood criterion is considered for the MSVME training, and we demonstrate that it is possible to train each expert based on an SVM perspective. Regarding the empirical evaluation, simulation results involving nonlinear dynamic system identification problems are reported, contrasting the performance shown by the MSVME approach with that exhibited by conventional SVM and ME models.","Title: RABNET: a real-valued antibody network for data clustering\nYear: 2005\n\nAbstract: This paper proposes a novel constructive learning algorithm for a competitive neural network. The proposed algorithm is developed by taking ideas from the immune system and demonstrates robustness in the initial experiments reported here for a benchmark problem. Comparisons with results from the literature are also provided. To automatically segment the resultant neurons at the output, a tool from graph theory was used with promising results. General discussions and avenues for future works are also provided.","Title: Mapping Artificial Immune Systems into Learning Classifier Systems\nYear: 2002\n\nAbstract: This paper presents one form of mapping Artificial Immune Systems (AIS) into Learning Classifier Systems (LCS). Artificial Immune Systems can be defined as adaptive systems inspired by theoretical models and principles of the biological immune system and applied to solve problems in the most diverse domains, from biology to computing. Similar to Learning Classifier Systems, already used to model complex adaptive systems, a better understanding of Artificial Immune Systems can be obtained when they are analysed under the perspective of complex adaptive systems. One of the goals here is to determine complementary features of both systems (LCS and AIS), aiming at providing a novel mapping conception. The formal treatment proposed along the paper may then be used to integrate models for complex adaptive systems.","Title: A Hierarchical Immune Network Applied to Gene Expression Data\nYear: 2004\n\nAbstract: This paper describes a new proposal for gene expression data analysis. The method used is based on a hierarchical approach\n to a hybrid algorithm, which is composed of an artificial immune system, named aiNet, and a well known graph theoretic tool,\n the minimal spanning tree (MST). This algorithm has already proved to be efficient for clustering gene expression data, but\n its performance may decrease in some specific cases. However, through the use of a hierarchical approach of immune networks\n it is possible to improve the clustering capability of the hybrid algorithm, such that it becomes more efficient, even when\n the data set is complex. The proposed methodology is applied to the yeast data and gives important conclusions of the similarity\n relationships among genes within the data set.","Title: Pattern classification with mixtures of weighted least-squares support vector machine experts\nYear: 2009\n\nAbstract: Support Vector Machine (SVM) classifiers are high-performance classification models devised to comply with the structural\n risk minimization principle and to properly exploit the kernel artifice of nonlinearly mapping input data into high-dimensional\n feature spaces toward the automatic construction of better discriminating linear decision boundaries. Among several SVM variants,\n Least-Squares SVMs (LS-SVMs) have gained increased attention recently due mainly to their computationally attractive properties\n coming as the direct result of applying a modified formulation that makes use of a sum-squared-error cost function jointly\n with equality, instead of inequality, constraints. In this work, we present a flexible hybrid approach aimed at augmenting\n the proficiency of LS-SVM classifiers with regard to accuracy/generalization as well as to hyperparameter calibration issues.\n Such approach, named as Mixtures of Weighted Least-Squares Support Vector Machine Experts, centers around the fusion of the\n weighted variant of LS-SVMs with Mixtures of Experts models. After the formal characterization of the novel learning framework,\n simulation results obtained with respect to both binary and multiclass pattern classification problems are reported, ratifying\n the suitability of the novel hybrid approach in improving the performance issues considered.","Title: A proposal for direct-ordering gene expression data by self-organising maps\nYear: 2004\n\nAbstract: Microarray experiments are employed to simultaneously measure the expression level of thousands of genes. The possible applications of the available datasets include: inference of individual gene functions; identification of genes from specific tissues; analysis of the behaviour of gene expression levels under various environmental conditions and under different cell cycle stages; and characterisation of inappropriately transcribed genes and several genetic diseases. A fundamental step in the analysis of gene expression data is the detection of genes presenting similar expression patterns. Considering that microarray technology allows the inspection of a wide range of aspects related to the genome at once, and therefore thousands of genes may be involved in an experiment, new computational tools for data analysis and alternative visualisation strategies are crucial to understand and uncover information present in the data. In this work, the gene ordering problem is modelled as a shortest-path problem and we present an algorithm based on competitive learning for rearranging gene expression data in a linear order aiming to reveal trends in large amount of data. The effectiveness of the algorithm is attested by means of computational simulations performed on publicly accessible data sets.","Title: Support Vector Clustering Applied To Digital Communications\nYear: 2006\n\nAbstract: Support Vector Clustering (SVC) is a recently proposed clustering methodology with promising performance for high-dimensional and noisy datasets, and for clusters with arbitrary shape. This work addresses the application of SVC, a kernel-based method, in a context in which the channel equalization problem is conceived as a clustering task The main challenge, in this case, is to perform unsupervised clustering aiming at the design of an optimal Bayesian or a blind prediction-based receiver without resorting to a priori information about the transmission medium. The proposed technique employs a two-stage procedure - a combination between the use of SVC to obtain a first set of clusters and an auxiliary heuristic to help separating eventual multiple clouds contained in a single cluster and attribute centers to them via an iterated local search (ELS) algorithm. The obtained results indicate that kernel methods can be successfully applied to the field of signal processing.","Title: Improving a multi-objective multipopulation artificial immune network for biclustering\nYear: 2009\n\nAbstract: The biclustering technique was developed to avoid some of the drawbacks presented by standard clustering techniques. Given that biclustering requires the optimization of at least two conflicting objectives and that multiple independent solutions are desirable as the outcome, a few multi-objective evolutionary algorithms for biclustering were proposed in the literature. However, apart from the individual characteristics of the biclusters that should be optimized during their construction, several other global aspects should also be considered, such as the coverage of the dataset and the overlap among biclusters. These requirements will be addressed in this work with the MOM-aiNet+ algorithm, which is an improvement of the original multi-objective multipopulation artificial immune network denoted MOM-aiNet. Here, the MOM-aiNet+ algorithm will be described in detail, its main differences from the original MOM-aiNet will be highlighted, and both algorithms will be compared, together with three other proposals from the literature.","Title: Fault detection algorithm for telephone systems based on the danger theory\nYear: 2005\n\nAbstract: This work is aimed at presenting a fault detection algorithm composed of multiple interconnected modules, and operating according to the paradigm supported by the danger theory in immunology. This algorithm attempts to achieve significant features that a fault detection system is supposed to have when monitoring a telephone profile system. These features would basically be adaptability due to the strong variation that operational conditions may exhibit over time, and the decrease in the number of false positives, which can be generated when any abnormal behavior is erroneously classified as being a fault. Simulated scenarios have been conceived to validate the proposal, and the obtained results are then analyzed.","Title: Transductive Support Vector Machines for Cancer Diagnosis and Classification of Microarray Gene Expression Data\nYear: 2002\n\nAbstract: purpose of this paper is to present a powerful methodology for classification u sing g ene expression d ata. The following problems will be considered: determination of specific diagnosis categories using the small round blue cell tumors of childhood, and classification o f genes in functional groups. The classification task will be performed by means of Transductive Inference with SVM, that will be implemented based on training and testing d ata sets. In the case of training samples, experts have already previously classified the genes or diagnostic c ategories in their respective classes. So, given the testing samples, the purpose is to determine the corresponding class. Thereby, we will be able to classify in only one step each sample of the testing data set as a member or not of a c ertain class. As the transductive inference has not been p reviously applied to the classification of gene expression data, considering SVM as the classification tool, it will be compared with the traditional inductive method in a series of exhaustive experiments, with promising results. 1 - Introduction With the purpose of improving generalization and increasing the performance of the estimator, transductive inference is considered h ere to estimate function values in o nly one step. It represents an alternative approach to inductive inference, that requires two steps: the first step (inductive) consists of discovering the functional dependence among input-output variables, and the second step (deductive) uses this functional dependence to evaluate the output in the points of interest. We will apply this methodology to the following problems: (A) Gene classification of budding yeast Saccharomyces Cerevisiae in functional groups using data of gene expression from an experiment of DNA microarray hybridization (1). The data set is composed of the ratio of the e xpression levels of 2467 genes of budding yeast Saccharomyces cerevisiae, measured in 7 9 different experiments of DNA microarray hybridization. The genes were classified into five functional classes: tricarboxylic a cid cycle (TCA), respiration (Resp), cytoplasmic ribosomal genes (Ribo), proteasome (Prot). The genes belonging to each functional class will be classified as positive, and the others genes will be classified as negative. The gain function to be maximized in testing d ata will be: G = GN - FP - 2xFN, where FP and FN are the number of false positives and false negatives, and GN is the null gain (1). We will consider the complexity index as: Complexity = 1 - (G / GN ). Using this index,","Title: An immune-inspired multi-objective approach to the reconstruction of phylogenetic trees\nYear: 2010\n\nAbstract: This work presents the application of the omni-aiNet algorithm—an immune-inspired algorithm originally developed to solve single and multi-objective optimization problems—to the reconstruction of phylogenetic trees. The main goal here is to automatically evolve a population of phylogenetic unrooted trees, possibly with distinct topologies, by minimizing at the same time two optimization criteria: the minimum evolution and the mean-squared error. This proposal generates, in a single run, a set of non-dominated solutions that represent the trade-offs of the two conflicting objectives, and gives the user the possibility of having distinct explanations for the differences observed at the terminal nodes of the trees. A series of experimental results is also reported in this work, in order to illustrate the effectiveness of the proposal and its capability to overcome the restrictive feedback provided by the application of well-known algorithms for phylogenetic reconstruction, such as the Neighbor Joining. Besides, the methodology presented in this work is compared to the popular NSGA-II algorithm, also modified to solve phylogenetic reconstruction problems.","Title: A Multiagent-Based Constructive Approach for Feedforward Neural Networks\nYear: 2003\n\nAbstract: In this paper, a new constructive approach for the automatic definition of feedforward neural networks (FNNs) is introduced. Such approach (named MASCoNN) is multiagent-oriented and, thus, can be regarded as a kind of hybrid (synergetic) system. MASCoNN centers upon the employment of a two-level hierarchy of agent-based elements for the progressive allocation of neuronal building blocks. By this means, an FNN can be considered as an architectural organization of reactive neural agents, orchestrated by deliberative coordination entities via synaptic interactions. MASCoNN was successfully applied to implement nonlinear dynamic systems identification devices and some comparative results, involving alternative proposals, are analyzed here.","Title: Long-Term Time Series Prediction Using Wrappers For Variable Selection And Clustering For Data Partition\nYear: 2007\n\nAbstract: In an attempt to implement long-term time series prediction based on the recursive application of a one-step-ahead multilayer neural network predictor, we have considered the eleven short time series provided by the organizers of the Special Session NN3 Neural Network Forecasting Competition, and have proposed a joint application of a variable selection technique and a clustering procedure. The purpose was to define unbiased partition subsets and predictors with high generalization capability, based on a wrapper methodology. The proposed approach overcomes the performance of the predictor that considers all the lags in the regression vector. After obtaining the eleven long-term predictors, we conclude the paper presenting the eighteen multi-step predictions for each time series, as requested in the competition.","Title: Evaluating the Performance of a Biclustering Algorithm Applied to Collaborative Filtering - A Comparative Analysis\nYear: 2007\n\nAbstract: Collaborative filtering (CF) is a method to perform automated suggestions for a user based on the opinion of other users with similar interest. Most of the CF algorithms do not take into account the existent duality between users and items, considering only the similarities between users or only the similarities between items. The authors have proposed in a previous work a bio-inspired methodology for CF, namely BIC-aiNet, capable of clustering rows and columns of a data matrix simultaneously. The usefulness and performance of the methodology are reported in the literature. Now, the authors carry out more rigorous comparative experiments with BICaiNet and other techniques found in the literature, as well as evaluate the scalability of the algorithm in several datasets of different sizes. The results indicate that our proposal is able to provide useful recommendations for the users, outperforming other methodologies for CF.","Title: A Hybrid Ensemble Model Applied To The Short-Term Load Forecasting Problem\nYear: 2006\n\nAbstract: In this paper we present a methodology based on a combination of many distinct predictors in an ensemble, named hybrid ensemble model, to obtain a more accurate output using the results of single predictors. As basic components, we have used Artificial Neural Networks and Support Vector Machines models. In order to evaluate the performance, the hybrid model was required to predict a 24h daily series energy consumption of a Brazilian electrical operation unit located in the northeast of Brazil. The proposed ensemble model has reached an error 25% smaller than that achieved by the best single predictor. The model was initialized several times to confirm that ensembles of predictors also tend to produce low variance profiles.","Title: Application of an artificial immune system in a compositional timbre design technique\nYear: 2005\n\nAbstract: Computer generated sounds for music applications have many facets, of which timbre design is of groundbreaking significance. Timbre is a remarkable and rather complex phenomenon that has puzzled researchers for a long time. Actually, the nature of musical signals is not fully understood yet. In this paper, we present a sound synthesis method using an artificial immune network for data clustering, denoted aiNet. Sounds produced by the method are referred to as immunological sounds. Basically, antibody-sounds are generated to recognize a fixed and predefined set of antigen-sounds, thus producing timbral variants with the desired characteristics. The aiNet algorithm provides maintenance of diversity and an adaptive number of resultant antibody-sounds (memory cells), so that the intended aesthetical result is properly achieved by avoiding the formal definition of the timbral attributes. The initial set of antibody-sounds may be randomly generated vectors, sinusoidal waves with random frequency, or a set of loaded waveforms. To evaluate the obtained results we propose an affinity measure based on the average spectral distance from the memory cells to the antigen-sounds. With the validation of the affinity criterion, the experimental procedure is outlined, and the results are depicted and analyzed.","Title: A Michigan-like immune-inspired framework for performing independent component analysis over Galois fields of prime order\nYear: 2014\n\nAbstract: In this work, we present a novel bioinspired framework for performing ICA over finite (Galois) fields of prime order P. The proposal is based on a state-of-the-art immune-inspired algorithm, the cob-aiNet[C], which is employed to solve a combinatorial optimization problem - associated with a minimal entropy configuration - adopting a Michigan-like population structure. The simulation results reveal that the strategy is capable of reaching a performance similar to that of standard methods for lower-dimensional instances with the advantage of also handling scenarios with an elevated number of sources.","Title: Identification of a Multivariate Fermentation Process using Constructive Learning\nYear: 2002\n\nAbstract: In the present work, a constructive learning algorithmis employed to design an optimal one-hidden neuralnetwork structure that best approximates a givenmapping. The method determines not only the optimalnumber of hidden neurons but also the best activationfunction for each node. Here, the projection pursuittechnique is applied in association with the optimizationof the solvability condition, giving rise to a moreefficient and accurate computational learning algorithm.As each activation function of a hidden neuron isoptimally defined for every approximation problem,better rates of convergence are achieved. Since thetraining process operates the hidden neuronsindividually, a pertinent activation function employingHermite polynomials can be iteratively developed foreach neuron as a function of the learning set. Theproposed constructive learning algorithm wassuccessfully applied to identify a large-scale multivariateprocess, providing a multivariable model that was ableto describe the complex process dynamics, even in long-rangehorizon predictions.","Title: A neural architecture to address Reinforcement Learning problems\nYear: 2011\n\nAbstract: In this paper, the Reinforcement Learning problem is formulated equivalently to a Markov Decision Process. We address the solution of such problem using a novel Adaptive Dynamic Programming algorithm which is based on a Multilayer Perceptron Neural Network composed of a parameterized function approximator called Wire-Fitting. Extending such established model, this work makes use of concepts of eligibility to conceive faster learning algorithms. The advantage of the proposed approach is founded on the capability to handle continuous environments and to learn a better policy while following another. Simulation results involving the automatic control of an inverted pendulum are presented to indicate the effectiveness of the proposed algorithm.","Title: Feature Subset Selection by Means of a Bayesian Artificial Immune System\nYear: 2008\n\nAbstract: This paper proposes the application of a novel bio-inspired algorithm as a search engine to the feature subset selection problem. We may interpret our algorithm as an Estimation of Distribution Algorithm that adopts an Artificial Immune System to implement the search process in the space of all features and a Bayesian network to implement the probabilistic model of the promising solutions. The characteristics of the proposed algorithm are the capability of effectively identifying and manipulating building blocks, maintenance of diversity in the population, and automatic control of the population size. These properties allow the algorithm to perform a multimodal search, known to be of great relevance in feature selection problems. Experiments on five datasets were carried out in order to evaluate the proposed methodology in classification problems and its performance compares favorably to that produced by contenders.","Title: Query expansion using an immune-inspired biclustering algorithm\nYear: 2010\n\nAbstract: Query expansion is a technique utilized to improve the performance of information retrieval systems by automatically adding related terms to the initial query. These additional terms can be obtained from documents stored in a database. Usually, this task is performed by clustering the documents and then extracting representative terms from the clusters. Afterwards, a new search is performed in the whole database using the expanded set of terms. Recently, the authors have proposed an immune-inspired algorithm, namely BIC-aiNet, to perform biclustering of texts. Biclustering differs from standard clustering algorithms in the sense that the former can detect partial similarities in the attributes. The preliminary results indicated that our proposal is able to group similar texts effectively and the generated biclusters consistently presented relevant words to represent a category of texts. Motivated by this promising scenario, this paper better formalizes the proposal and investigates the usefulness of the whole methodology on larger datasets. The BIC-aiNet was applied to a set of documents aiming at identifying the set of relevant terms associated with each bicluster, giving rise to a query expansion tool. The obtained results were compared with those produced by two alternative proposals in the literature, and they indicate that these techniques tend to generate complementary results, as a consequence of the use of distinct similarity metrics.","Title: Interactive Control of Evolution Applied to Sound Synthesis\nYear: 2005\n\nAbstract: In this paper, we present a sound synthesis method that utilizes evolution as generative paradigm. Such sounds will be thereon referred to as evolutionary sounds. Upon defining a population of complex sounds, i.e. sound segments sampled from acoustical instruments and speech; we generated sounds that resulted from evolution applied to those populations. The methodology presented here is an extension to the Evolutionary Sound Synthesis Method (ESSynth) created recently. In ESSynth, a set of waveforms, the Population, is evolved towards another set, the Target, through the application of a Genetic Algorithm (GA). Fitness evaluation is a mathematical distance metric. We enhance features of the previous implementation herein and present the codification. The genetic operators and selection criterion applied are depicted together with the relevant genetic parameters involved in the process. To evaluate the results we present a sound taxonomy based on an objective and a subjective criterion. Those criteria are discussed, the experimental procedure is explained and the results are depicted and evaluated.","Title: BAIS: A Bayesian Artificial Immune System for the effective handling of building blocks\nYear: 2009\n\nAbstract: Significant progress has been made in theory and design of Artificial Immune Systems (AISs) for solving hard problems accurately. However, an aspect not yet widely addressed by the research reported in the literature is the lack of ability of the AISs to deal effectively with building blocks (partial high-quality solutions coded in the antibody). The available AISs present mechanisms for evolving the population that do not take into account the relationship among the variables of the problem, potentially causing the disruption of high-quality partial solutions. This paper proposes a novel AIS with abilities to identify and properly manipulate building blocks in optimization problems. Instead of using cloning and mutation to generate new individuals, our algorithm builds a probabilistic model representing the joint probability distribution of the promising solutions and, subsequently, uses this model for sampling new solutions. The probabilistic model used is a Bayesian network due to its capability of properly capturing the most relevant interactions among the variables. Therefore, our algorithm, called Bayesian Artificial Immune System (BAIS), represents a significant attempt to improve the performance of immune-inspired algorithms when dealing with building blocks, and hence to solve efficiently hard optimization problems with complex interactions among the variables. The performance of BAIS compares favorably with that produced by contenders such as state-of-the-art Estimation of Distribution Algorithms.","Title: Adaptive radius immune algorithm for data clustering\nYear: 2005\n\nAbstract: Many algorithms perform data clustering by compressing the original data into a more compact and interpretable representation, which can be more easily inspected for the presence of clusters. This, however, can be a risky alternative, because the simplified representation may contain distortions mainly related to the density information present in the data, which can considerably act on the clustering results. In order to treat this deficiency, this paper proposes an Adaptive Radius Immune Algorithm (ARIA), which is capable of maximally preserving the density information after compression by implementing an antibody adaptive suppression radius that varies inversely with the local density in the space. ARIA is tested with both artificial and real world problems obtaining a better performance than the aiNet algorithm and showing that preserving the density information leads to refined clustering results.","Title: Max Min Ant System And Capacitated P-Medians: Extensions And Improved Solutions\nYear: 2005\n\nAbstract: This work introduces a modified MAX MIN Ant System (MMAS) designed to solve the Capacitated p-Medians Problem (CPMP). It presents the most relevant steps towards the implementation of an MMAS to solve the CPMP, including some improvements on the original MMAS algorithm, such as the use of a density model in the information heuristics and a local search adapted from the uncapacitated pmedians problem. Extensions of a recently proposed updating rule for the pheromone level, aiming at improving the MMAS ability to deal with large-scale instances, are also presented and discussed. Some simulations are performed using instances available from the literature, and well-known heuristics are employed for benchmarking.","Title: Multi-objective Bayesian Artificial Immune System: Empirical Evaluation and Comparative Analyses\nYear: 2009\n\nAbstract: Recently, we have proposed a Multi-Objective Bayesian Artificial Immune System (MOBAIS) to deal effectively with building\n blocks (high-quality partial solutions coded in the solution vector) in combinatorial multi-objective problems. By replacing\n the mutation and cloning operators with a probabilistic model, more specifically a Bayesian network representing the joint\n distribution of promising solutions, MOBAIS takes into account the relationships among the variables of the problem, avoiding\n the disruption of already obtained high-quality partial solutions. The preliminary results have indicated that our proposal\n is able to properly build the Pareto front. Motivated by this scenario, this paper better formalizes the proposal and investigates\n its usefulness on more challenging problems. In addition, an important enhancement regarding the Bayesian network learning\n was incorporated into the algorithm in order to speed up its execution. To conclude, we compare MOBAIS with state-of-the-art\n algorithms taking into account quantitative aspects of the Pareto front found by the algorithms. MOBAIS outperforms the contenders\n in terms of the quality of the obtained solutions and requires an amount of computational resource inferior or compatible\n with the contenders.","Title: An Evolutionary Immune Network for Data Clustering\nYear: 2000\n\nAbstract: This paper explores basic aspects of the immune system and proposes a novel immune network model with the main goals of clustering and filtering unlabeled numerical data sets. It is not our concern to reproduce with confidence any immune phenomenon, but to show that immune concepts can be used to develop powerful computational tools for data processing. As important results of our model, the network evolved will be capable of reducing redundancy, describing data structure, including the shape of the clusters. The network will be implemented in association with a statistical inference technique, and its performance will be illustrated using two benchmark problems. The paper is concluded with a trade-off between the proposed network and artificial neural networks used to perform unsupervised learning.","Title: Self-organization and lateral interaction in echo state network reservoirs.\nYear: 2014\n\nAbstract: Echo state networks (ESNs) are recurrent structures that give rise to an interesting trade-off between achievable performance and tractability. This is a consequence of the fact that the key element of these networks – the recurrent intermediate layer known as dynamical reservoir – is not, as a rule, subject to supervised training, which is restricted to the linear output layer, also termed as readout. This trade-off, aside from being of theoretical significance, establishes ESNs as most attractive tools for both online and offline information processing. There are two key aspects to be taken into account in the ESN design: (i) the unsupervised definition of the synaptic weights of the reservoir and (ii) the definition of the structure and of the training strategy associated with the readout. This work is concerned with the first of these aspects: it proposes novel strategies for ESN reservoir design based on the theoretical framework built by Kohonen׳s classical works on self-organization – which includes the notions of short-range positive feedback and lateral inhibition – and also on the related and more recent notion of neural gas. It is shown, with the aid of a representative set of simulation results, that the proposed methodologies are capable of leading to significant performance improvements in the context of relevant information processing tasks – channel equalization and chaotic time series prediction – particularly when the input data suits well a cluster-based profile.","Title: A dynamic artificial immune algorithm applied to challenging benchmarking problems\nYear: 2009\n\nAbstract: In many real-world scenarios, in contrast to standard benchmark optimization problems, we may face some uncertainties regarding the objective function. One source of these uncertainties is a constantly changing environment in which the optima change their location over time. New heuristics or adaptations to already available algorithms must be conceived in order to deal with such problems. Among the desirable features that a search strategy should exhibit to deal with dynamic optimization are diversity maintenance, a memory of past solutions, and a multipopulation structure of candidate solutions. In this paper, an immune-inspired algorithm that presents these features, called dopt-aiNet, is properly adapted to deal with six newly proposed benchmark instances, and the obtained results are outlined according to the available specifications for the competition at the Congress on Evolutionary Computation 2009.","Title: Learning Ensembles of Neural Networks by Means of a Bayesian Artificial Immune System\nYear: 2011\n\nAbstract: In this paper, we apply an immune-inspired approach to design ensembles of heterogeneous neural networks for classification problems. Our proposal, called Bayesian artificial immune system, is an estimation of distribution algorithm that replaces the traditional mutation and cloning operators with a probabilistic model, more specifically a Bayesian network, representing the joint distribution of promising solutions. Among the additional attributes provided by the Bayesian framework inserted into an immune-inspired search algorithm are the automatic control of the population size along the search and the inherent ability to promote and preserve diversity among the candidate solutions. Both are attributes generally absent from alternative estimation of distribution algorithms, and both were shown to be useful attributes when implementing the generation and selection of components of the ensemble, thus leading to high-performance classifiers. Several aspects of the design are illustrated in practical applications, including a comparative analysis with other attempts to synthesize ensembles.","Title: A Heuristic Method Based on Unsupervised Learning and Fuzzy Inference for the Vehicle Routing Problem\nYear: 2002\n\nAbstract: This paper deals with a fuzzy-based system to solve thecapacitated vehicle routing problem. The proposedmethod makes use of a neural network employingunsupervised learning guided by a fuzzy rule base. Thealgorithm implements a policy of penalties and rewards, astrategy of neuron inhibition, insertion and pruning, andalso takes into account certain statistical characteristicsof the input space. Fuzzy theory is considered to minimizedrawbacks related to uncertainty and availability ofpartial information, guiding to an adaptive process ofconstraint relaxation. The effectiveness of the proposedmethod is attested by means of a series of computationalsimulations. As the proposed approach has no adaptationto any particular instance, it represents a good candidateto provide the initial condition for more dedicatedapproaches, like tabu search.","Title: Designing ensembles of fuzzy classification systems: an immune-inspired approach\nYear: 2005\n\nAbstract: In this work we propose an immune-based approach for designing of fuzzy systems. From numerical data and with membership function previously defined, the immune algorithm evolves a population of fuzzy classification rules based on the clonal selection, hypermutation and immune network principles. Once AIS are able to find multiple good solutions of the problem, accurate and diverse fuzzy systems are built in a single run. Hence, we construct an ensemble of these classifier in order to achieve better results. An ensemble of classifiers consists of a set of individual classifiers whose outputs are combined when classifying novel patterns. The good performance of an ensemble is strongly dependent of individual accuracy and diversity of its components. We evaluate the proposed methodology through computational experiments on some datasets. The results demonstrate that the performance of the obtained fuzzy systems in isolation is very good. However when we combine these systems, a significant improvement is obtained in the correct classification rate, outperforming the single best classifier.","Title: Evaluating the performance of a Bayesian Artificial Immune System for designing fuzzy rule bases\nYear: 2013\n\nAbstract: In this paper we perform a deep investigation about the usefulness of an immune-inspired algorithm to design accurate and compact fuzzy rule bases for classification problems. The algorithm, called Bayesian Artificial Immune System BAIS, incorporates a mechanism to learn a probability graphical model from the promising solutions found so far. Thus, BAIS utilizes this model to sample new candidate solutions. The probabilistic model utilized here is a Bayesian network due to its capability of expressing the relationships among the variables of the problem, avoiding the disruption of already obtained high-quality partial solutions building blocks. Besides the capability to identify and manipulate building blocks, the algorithm maintains diversity in the population, performs multimodal optimization and adjusts the size of the population automatically according to the problem. These attributes are generally absent from alternative algorithms, and can be considered useful attributes when generating fuzzy rule bases, thus guiding to high-performance classifiers. BAIS was evaluated in thirteen well-known classification problems and its performance compares favorably with that produced by contenders.","Title: Ordering Gene Expression Data Using One-dimensional Self-Organizing Maps\nYear: 2002\n\nAbstract: The microarray technology allows researchers to simultaneously measure gene expression levels of thousands of genes. Analysis of data pro- duced by such experiments provides knowledge about the gene function. An important step in the a nalysis of gene e xpression d ata is the detection o f genes with similar expression patterns. Real-time computational tools for or- ganization and visualization are crucial to understand and analyze the data. In this work, we make use of an algorithm based on self-organizing neural net- works for organizing g ene e xpression d ata in o rder to reveal t rends in g ene expression profiles under the biological viewpoint.","Title: Modular And Hierarchical Evolutionary Design Of Fuzzy Systems\nYear: 1999\n\nAbstract: The paper introduces a modular, hierarchical evolutionary method to design fuzzy systems. The method uses genetic algorithms to evolve a population of rule-based fuzzy models. For this purpose, three hierarchical modules are defined namely, partition, population of granules, and population of fuzzy graphs. The evolutionary process is designed to find the best set of fuzzy rules induced by the granularity required, the form of the membership functions and the inference procedure. This hierarchical configuration guides to the implementation of an effective process of cooperation and competition among rules, responsible for the small cardinality of the re-suiting set of rules. Simulation results show that the method does increase flexibility to design fuzzy models, preserves the computational tractability, and improves the descriptive nature of the final solution.","Title: On The Diversity Mechanisms Of Opt-Ainet: A Comparative Study With Fitness Sharing\nYear: 2010\n\nAbstract: Immune-inspired algorithms based on the Immune Network theory have been frequently claimed to be capable of maintaining diversity among the candidate solutions in their population. However, no specific study on this aspect to verify how the intrinsic diversity mechanisms of such immune algorithms behave, when compared to other approaches from the literature, has been made yet. Therefore, in this work we have addressed this issue, by taking the opt-aiNet algorithm (a popular immune-inspired algorithm developed for real-parameter optimization) and comparing its results with those of a modified version, in which the mechanisms associated with diversity maintenance were replaced by a traditional fitness sharing approach. Besides, two distance metrics were also considered for both algorithms: the traditional Euclidean distance, and the Line Distance, a metric proposed in the literature as capable of identifying whether two solutions belong to distinct local optima. The experiments were performed on six benchmark problems from the literature, each of them with distinct characteristics, and the results have shown that the original immune-inspired mechanisms of opt-aiNet are indeed more capable of stimulating the diversity of solutions, and also requiring a smaller amount of computational resources.","Title: New Perspectives For The Biclustering Problem\nYear: 2006\n\nAbstract: Multimodal optimization algorithms inspired by the immune system are generally characterized by a dynamic control of the population size and by diversity maintenance along the search. One of these proposals, denoted copt-aiNet (artificial immune network for combinatorial optimization), is used to deal with combinatorial problems like the Traveling Salesman Problem (TSP) and other permutation problems. In this paper, the copt-aiNet algorithm is extended and adapted to be applied to an important issue of modern data mining, the biclustering problem. The biclustering approach consists in simultaneously ordering the rows and columns of a given matrix, so that similar elements are grouped together. To illustrate the performance of the proposed method, two bitmap images are scrambled and used as input to the algorithm, and the biclustering procedure tries to restore the original image by grouping the pixels according to the similarity of colors in a neighborhood. Additionally, copt-aiNet is applied to gene expression data clustering, a classical problem of the bioinformatics literature, and its performance is compared with a hierarchical biclustering algorithm.","Title: Self-organizing Bio-inspired Sound Transformation\nYear: 2007\n\nAbstract: We present a time domain approach to explore a sound transformation paradigm for musical performance. Given a set of sounds containing a priori desired qualities and a population of agents interacting locally, the method generates both musical form and matter resulting from sonic trajectories. This proposal involves the use of bio-inspired algorithms, which possess intrinsic features of adaptive, self-organizing systems, as definers of generating and structuring processes of sound elements. Self-organization makes viable the temporal emergence of stable structures without an external organizing element. Regarding musical performance as a creative process that can be described using trajectories through the compositional space, and having the simultaneous emergence of musical matter and form resulting from the process itself as the final objective, the conception of a generative paradigm in computer music that does not contemplate a priori external organizing elements is the main focus of this proposal.","Title: A Wrapper For Projection Pursuit Learning\nYear: 2007\n\nAbstract: Constructive algorithms have shown to be reliable and effective methods for designing Artificial Neural Networks (ANN) with good accuracy and generalization capability, yet with parsimonious network structures. Projection Pursuit Learning (PPL) has demonstrated great flexibility and effectiveness in performing this task, though presenting some difficulties in the search for appropriate projection directions in input spaces with high dimensionality. Due to the existence of high-dimensional input spaces in the context of time series prediction, mainly under the existence of long-term dependencies in the time series, we propose here a method based on the wrapper methodology to perform variable selection, so that only a subset of highly-informative lags is going to be considered as the regression vector. The Yearly Sunspot Number time series is adopted as a case study and comparative analysis is performed considering alternative approaches in the literature, guiding to competitive results.","Title: An extended echo state network using Volterra filtering and principal component analysis.\nYear: 2012\n\nAbstract: Echo state networks (ESNs) can be interpreted as promoting an encouraging compromise between two seemingly conflicting objectives: (i) simplicity of the resulting mathematical model and (ii) capability to express a wide range of nonlinear dynamics. By imposing fixed weights to the recurrent connections, the echo state approach avoids the well-known difficulties faced by recurrent neural network training strategies, but still preserves, to a certain extent, the potential of the underlying structure due to the existence of feedback loops within the dynamical reservoir. Moreover, the overall training process is relatively simple, as it amounts essentially to adapting the readout, which usually corresponds to a linear combiner. However, the linear nature of the output layer may limit the capability of exploring the available information, since higher-order statistics of the signals are not taken into account. In this work, we present a novel architecture for an ESN in which the linear combiner is replaced by a Volterra filter structure. Additionally, the principal component analysis technique is used to reduce the number of effective signals transmitted to the output layer. This idea not only improves the processing capability of the network, but also preserves the simplicity of the training process. The proposed architecture is then analyzed in the context of a set of representative information extraction problems, more specifically supervised and unsupervised channel equalization, and blind separation of convolutive mixtures. The obtained results, when compared to those produced by already proposed ESN versions, highlight the benefits brought by the novel network proposal and characterize it as a promising tool to deal with challenging signal processing tasks.","Title: Multi-Objective Biclustering: When Non-dominated Solutions are not Enough\nYear: 2009\n\nAbstract: The biclustering technique was developed to avoid some of the drawbacks presented by standard clustering techniques, such\n as their impossibility of finding correlating data under a subset of features, and, consequently, to allow the extraction\n of more accurate information from datasets. Given that biclustering requires the optimization of at least two conflicting\n objectives (residue and volume) and that multiple independent solutions are desirable as the outcome, a few multi-objective\n evolutionary algorithms for biclustering were proposed in the literature. However, these algorithms only focus their search\n in the generation of a global set of non-dominated biclusters, which may be insufficient for most of the problems as the coverage\n of the dataset can be compromised. In order to overcome such problem, a multi-objective artificial immune system capable of\n performing a multipopulation search, named MOM-aiNet, was proposed. In this work, the MOM-aiNet algorithm will be described\n in detail, and an extensive set of experimental comparisons will be performed, with the obtained results of MOM-aiNet being\n confronted with those produced by the popular CC algorithm, by another immune-inspired approach for biclustering (BIC-aiNet),\n and by the multi-objective approach for biclustering proposed by Mitra & Banka.","Title: A Classifier System In Real Applications For Robot Navigation\nYear: 2003\n\nAbstract: This work presents an autonomous evolutionary system applied to control a mobile robot in unknown environments. The navigation system learns efficiently to deal with situations where the robot must capture targets avoiding collisions with obstacles. Toward this end, robot direction and speed must be properly defined. The evolutionary approach is based on a version of classifier systems, responsible for the proposition of a competitive process involving rules of elementary behaviour. A virtual environment is used to evolve the controller. With the evolved controller, a Khepera II robot is submitted to real navigation tasks, with no significant degradation in performance. As an additional experiment, the controller is also evolved in a real environment, and validated in a different and more complex environment, not previously experimented, attesting the generalisation capability of the proposal.","Title: Multivariate ant colony optimization in continuous search spaces\nYear: 2008\n\nAbstract: This work introduces an ant-inspired algorithm for optimization in continuous search spaces that is based on the generation of random vectors with multivariate Gaussian pdf. The proposed approach is called MACACO -- Multivariate Ant Colony Algorithm for Continuous Optimization -- and is able to simultaneously adapt all the dimensions of the random distribution employed to generate the new individuals at each iteration. In order to analyze MACACO's search efficiency, the approach was compared to a pair of counterparts: the Continuous Ant Colony System (CACS) and the approach known as Ant Colony Optimization in en (ACOR). The comparative analysis, which involves well-known benchmark problems from the literature, has indicated that MACACO outperforms CACS and ACOR in most cases as the quality of the final solution is concerned, and it is just about two times more costly than the least expensive contender.","Title: Topology of social networks and efficiency of collective intelligence methods\nYear: 2013\n\nAbstract: In this work we analyzed the role social networks play in the efficiency of collective problem-solving, evaluating whether the topological characteristics seen in real-world networks yield any performance improvement in such processes. To study this we used the Particle Swarm Optimization as a testbed for social groups performing a collective task, defining the structure of communication between individuals in the swarm through topologies generated by a model for the creation and evolution of social networks. The experimental results indicate that groups using these networks may, indeed, experience better performance in collective problem-solving, so that these groups were able to overcome the results achieved by swarms using classical neighborhoods for PSO and reached results very close to those found by swarms using the topology of DMS-PSO, usually considered to be part of the state-of-the-art of Particle Swarm Optimization.","Title: The construction of a Boolean competitive neural network using ideas from immunology\nYear: 2003\n\nAbstract: The immune system is capable of recognizing and responding to microorganisms and molecules that cannot be perceived by our sensory mechanisms, which send stimuli straight into the brain. It performs an accessory role for nervous cognition. This paper main goals are: (1) to show how some immune principles and theories can be used as sources of inspiration to develop novel neural network learning algorithms; (2) to survey the main works from the literature that employ the immune metaphor for the development of neural network architectures; and (3) to illustrate, with a new network model, how this source of inspiration can be actually used to develop a neural network learning algorithm. The novel learning algorithm proposed has the main features of competitive learning, automatic generation of the network structure and binary representation of the connection strengths (weights). The behavior of the algorithm is primarily described using a benchmark task, and some of its potential applications are illustrated using two simple real-world problems and a binary character recognition task. The results show that the network is a promising tool for solving problems that are inherently binary, and also that the immune system provides a new paradigm to search for neural network learning algorithms.","Title: Online Social Network Analysis: A Survey of Research Applications in Computer Science.\nYear: 2015\n\nAbstract: The emergence and popularization of online social networks suddenly made available a large amount of data from social organization, interaction and human behavior. All this information opens new perspectives and challenges to the study of social systems, being of interest to many fields. Although most online social networks are recent (less than fifteen years old), a vast amount of scientific papers was already published on this topic, dealing with a broad range of analytical methods and applications. This work describes how computational researches have approached this subject and the methods used to analyze such systems. Founded on a wide though non-exaustive review of the literature, a taxonomy is proposed to classify and describe different categories of research. Each research category is described and the main works, discoveries and perspectives are highlighted.","Title: Asymmetric Distributions from Constrained Mixtures.\nYear: 2015\n\nAbstract: This paper introduces constrained mixtures for continuous distributions, characterized by a mixture of distributions where each distribution has a shape similar to the base distribution and disjoint domains. This new concept is used to create generalized asymmetric versions of the Laplace and normal distributions, which are shown to define exponential families, with known conjugate priors, and to have maximum likelihood estimates for the original parameters, with known closed-form expressions. The asymmetric and symmetric normal distributions are compared in a linear regression example, showing that the asymmetric version performs at least as well as the symmetric one, and in a real world time-series problem, where a hidden Markov model is used to fit a stock index, indicating that the asymmetric version provides higher likelihood and may learn distribution models over states and transition distributions with considerably less entropy.","Title: Multi-task Sparse Structure Learning\nYear: 2014\n\nAbstract: Multi-task learning (MTL) aims to improve generalization performance by learning multiple related tasks simultaneously. While sometimes the underlying task relationship structure is known, often the structure needs to be estimated from data at hand. In this paper, we present a novel family of models for MTL, applicable to regression and classification problems, capable of learning the structure of task relationships. In particular, we consider a joint estimation problem of the task relationship structure and the individual task parameters, which is solved using alternating minimization. The task relationship structure learning component builds on recent advances in structure learning of Gaussian graphical models based on sparse estimators of the precision (inverse covariance) matrix. We illustrate the effectiveness of the proposed model on a variety of synthetic and benchmark datasets for regression and classification. We also consider the problem of combining climate model outputs for better projections of future climate, with focus on temperature in South America, and show that the proposed model outperforms several existing methods for the problem.","Title: Hybrid genetic training of gated mixtures of experts for nonlinear time series forecasting\nYear: 2003\n\nAbstract: In this paper, we introduce a genetic algorithm-based training mechanism (HGT-GAME) toward the automatic structural design and parameter configuration of gated mixtures of experts (ME). In HGT-GAME, a whole ME instance is codified into a given chromosome. By employing regulatory genes, our approach enables the automatic pruning and growing of experts in a way to properly match the complexity of the task at hand. Moreover, to leverage HGT-GAME's effectiveness a local search refinement upon each ME chromosome is performed in each generation via the gradient descent-learning algorithm. Forecasting experiments evaluate the performance of gated MEs trained with HGT-GAME.","Title: Implementation of an immuno-genetic network on a real Khepera II robot\nYear: 2003\n\nAbstract: The design of autonomous navigation systems for mobile robots, with simultaneous objectives to be satisfied such as garbage collection with integrity maintenance, requires refined coordination mechanisms to deal with modules of elementary behaviour. This paper shows the implementation on a real Khepera II robot of an immuno-genetic network for autonomous navigation that combines an evolutionary algorithm with a continuous immune network model. The proposed immuno-genetic system has the immune network implementing a dynamic process of decision-making, and the evolutionary algorithm defining the network structure. To be able to evaluate the controllers (immune networks) on the evolutionary process, a virtual environment was used for computer simulation, based on the characteristics of the navigation problem. The immune networks obtained by evolution were then analyzed and tested on new situations, presenting coordination capability in simple and more complex tasks. Some preliminary experiments on a real Khepera II robot demonstrate the feasibility of the evolved immune networks.","Title: Immune Learning Classifier Networks: Evolving Nodes and Connections\nYear: 2006\n\nAbstract: The design of an autonomous navigation system with multiple tasks to be accomplished in unknown environments represents a complex undertaking. With the simultaneous purposes of capturing targets and avoiding obstacles, the challenge may become still more intricate if the configuration of obstacles and targets creates local minima, like concave shapes and mazes between the robot and the target. Pure reactive navigation systems are not able to deal properly with such hampering scenarios, requiring additional cognitive apparatus. Concepts from immune network theory are then employed to convert an earlier reactive robot controller, based on learning classifier systems, into a connectionist device. Starting from no a priori knowledge, both the classifiers and their connections are evolved during the robot navigation. Some experiments with and without local minima are carried out and the proposed evolutionary network of classifiers was shown to produce connectionist navigation systems capable of successfully overcoming local minima.","Title: An Immunological Density-Preserving Approach to the Synthesis of RBF Neural Networks for classification\nYear: 2006\n\nAbstract: Radial basis function (RBF) neural networks are universal approximators and have been used for a wide range of applications. Aiming at reducing the number of neurons in the hidden layer, for regularization purposes, the center and dispersion of each RBF have to be properly defined by means of competitive learning. Only the output weights will be defined in a supervised manner. One of the drawbacks of such learning methodology, involving unsupervised and supervised learning, is that the centers will be defined so that regions in the input space with a high density of samples tend to be under-represented and those regions with a low density of samples tend to be over-represented. Additionally, few approaches provide a proper and individual indication of the dispersion of each RBF. This paper presents an immune density-preserving algorithm with adaptive radius, called ARIA, to determine the number of centers, their location and the dispersion of each RBF, based only on the available training data set. Considering classification problems, the algorithm to determine the hidden layer is compared to another immune-inspired algorithm called aiNet, K-means and the random choice of centers. The classification accuracy of the final network is compared to another density based approach and a decision tree classifier, C 5.0. The results are reported and analyzed.","Title: Bayesian Learning of Neural Networks by Means of Artificial Immune Systems\nYear: 2006\n\nAbstract: Once the design of artificial neural networks (ANN) may require the optimization of numerical and structural parameters, bio-inspired algorithms have been successfully applied to accomplish this task, since they are population-based search strategies capable of dealing successfully with complex and large search spaces, avoiding local minima. In this paper, we propose the use of an artificial immune system for learning feedforward ANN's topologies. Besides the number of neurons in the hidden layer, the algorithm also optimizes the type of activation function for each node. The use of a Bayesian framework to infer the weights and weight decay terms as well as to perform model selection allows us to find neural models with high generalization capability and low complexity, once the Occam's razor principle is incorporated into the framework. We demonstrate the applicability of the proposal on seven classification problems and promising results were obtained.","Title: Designing fuzzy rule bases with a Bayesian Artificial Immune System\nYear: 2011\n\nAbstract: In this paper we apply an immune-inspired approach to generate fuzzy rule bases for classification problems. Our proposal, called Bayesian Artificial Immune System (BAIS), is a hybrid algorithm that replaces the traditional mutation and cloning operators with a probabilistic model, more specifically a Bayesian network, representing the joint distribution of promising solutions. Thus, the algorithm takes into account the relationships among the variables of the problem, avoiding the disruption of already obtained high-quality partial solutions (building blocks). Besides the capability to identify and manipulate building blocks, the algorithm maintains diversity in the population, performs multimodal optimization and adjusts the size of the population automatically according to the problem. These attributes are generally absent from alternative algorithms, and can be considered useful attributes when generating fuzzy rule bases, thus guiding to high-performance classifiers. BAIS was evaluated in six well-known classification problems and its performance compares favorably with that produced by contenders.","Title: Learning Bayesian networks to perform feature selection\nYear: 2009\n\nAbstract: Bayesian networks have been widely applied to the feature selection problem. The existing approaches learn a Bayesian network from the available dataset and, afterward, utilize the Markov Blanket of the target feature as the criterion to select the relevant features. The Bayesian network learning can be viewed as a search and optimization procedure, where a search mechanism explores the space of all network structures while a scoring metric evaluates each candidate solution based on the likelihood. This paper investigates the application of an immune-inspired algorithm as the search procedure for obtaining high-quality Bayesian networks, motivated by the dynamical control of the population size and diversity along the search. Due to the resulting multimodal search capability, in a single run of the algorithm several subsets of features are obtained. Experiments on ten datasets were carried out in order to evaluate the proposed methodology in classification problems, and reduced-size subsets of features were produced.","Title: Regularized hypervolume selection for robust portfolio optimization in dynamic environments\nYear: 2013\n\nAbstract: This paper proposes a regularized hypervolume (SMetric) selection algorithm. The proposal is used for incorporating stability and diversification in financial portfolios obtained by solving a temporal sequence of multi-objective Mean Variance Problems (MVP) on real-world stock data, for short to longterm rebalancing periods. We also propose the usage of robust statistics for estimating the parameters of the assets returns distribution so that we are able to test two variants (with and without regularization) on dynamic environments under different levels of instability. The results suggest that the maximum attaining Sharpe Ratio portfolios obtained for the original MVP without regularization are unstable, yielding high turnover rates, whereas solving the robust MVP with regularization mitigated turnover, providing more stable solutions for unseen, dynamic environments. Finally, we report an apparent conflict between stability in the objective space and in the decision space.","Title: Anticipatory Stochastic Multi-Objective Optimization for uncertainty handling in portfolio selection\nYear: 2013\n\nAbstract: An anticipatory stochastic multi-objective model based on S-Metric maximization is proposed. The environment is assumed to be noisy and time-varying. This raises the question of how to incorporate anticipation in metaheuristics such that the Pareto optimal solutions can reflect the uncertainty about the subsequent environments. A principled anticipatory learning method for tracking the dynamics of the objective vectors is then proposed so that the estimated S-Metric contributions of each solution can integrate the underlying stochastic uncertainty. The proposal is assessed for minimum holding, cardinality constrained portfolio selection, using real-world stock data. Preliminary results suggest that, by taking into account the underlying uncertainty in the predictive knowledge provided by a Kalman filter, we were able to reduce the sum of squared errors prediction of the portfolios ex-post return and risk estimation in out-of-sample investment environments.","Title: A Gaussian Artificial Immune System for Multi-Objective optimization in continuous domains\nYear: 2010\n\nAbstract: This paper proposes a Multi-Objective Gaussian Artificial Immune System (MOGAIS) to deal effectively with building blocks (high-quality partial solutions coded in the solution vector) in multi-objective continuous optimization problems. By replacing the mutation and cloning operators with a probabilistic model, more specifically a Gaussian network representing the joint distribution of promising solutions, MOGAIS takes into account the relationships among the variables of the problem, avoiding the disruption of already obtained high-quality partial solutions. The algorithm was applied to three benchmarks and the results were compared with those produced by state-of-the-art algorithms.","Title: Heuristics to avoid redundant solutions on population-based multimodal continuous optimization\nYear: 2011\n\nAbstract: In population-based meta-heuristics, the generation and maintenance of diversity seem to be crucial to deal with multimodal continuous optimization. However, usually this crucial aspect is not an inherent feature of generally adopted meta-heuristics. In this paper, we propose to associate diversity maintenance with the detection and elimination of redundant candidate solutions in the search space, more specifically candidate solutions located at the same attraction basin of a local optimum. Two low computational cost heuristics are proposed to detect redundancy, in a pairwise comparison of candidate solutions and by extracting local features of the fitness landscape at runtime. Those heuristics are not tied to a specific class of algorithms, and are thus able to be incorporated into a broad range of population-based meta-heuristics, and even into multiple executions of non-population-based algorithms. In a set of experimental results, the two heuristics were implemented as an attached module of an already existing multipopulation meta heuristics, and the results indicate that they operate properly, no matter the number and conformation of the attraction basins in multimodal optimization problems.","Title: Evolutionary multi-objective optimization for the vendor-managed inventory routing problem\nYear: 2011\n\nAbstract: The class of inventory routing problems (IRPs) is present in several areas, including automotive industry and cash management for ATM networks. In the specific case of vendor-managed IRPs, in which the supplier is responsible for managing the product inventory in each client and for properly providing replenishments, the challenge is to determine which retailers should be served, the amount of product that should be delivered to each of these retailers, and which routes the distribution vehicles should follow, so that the associated costs are minimized. Although this is clearly a multi-objective optimization problem, in the literature it has been generally modeled as a single-objective problem, which limits the scope of the obtained results. Therefore, this work presents a multi-objective approach to solve one version of the IRP usually found in the scientific literature, by simultaneously minimizing both the inventory and transportation costs. The method proposed in this work is based on the well-known SPEA2 (Strength Pareto Evolutionary Algorithm) and includes innovative aspects mainly associated with the representation of candidate solutions, genetic operators and local search. The experiments were performed on a set of known benchmark IRPs from the literature, so that the obtained results could be properly compared to the best solution found for the single-objective version of each problem.","Title: Training multilayer perceptrons with a Gaussian Artificial Immune System\nYear: 2011\n\nAbstract: In this paper we apply an immune-inspired approach to train Multilayer Perceptrons (MLPs) for classification problems. Our proposal, called Gaussian Artificial Immune System (GAIS), is an estimation of distribution algorithm that replaces the traditional mutation and cloning operators with a probabilistic model, more specifically a Gaussian network, representing the joint distribution of promising solutions. Sub sequently, GAIS utilizes this probabilistic model for sampling new solutions. Thus, the algorithm takes into account the relationships among the variables of the problem, avoiding the disruption of already obtained high-quality partial solutions (building blocks). Besides the capability to identify and manipulate building blocks, the algorithm maintains diversity in the population, performs multimodal optimization and adjusts the size of the population automatically according to the problem. These attributes are generally absent from alternative algorithms, and all were shown to be useful attributes when optimizing the weights of MLPs, thus guiding to high-performance classifiers. GAIS was evaluated in six well-known classification problems and its performance compares favorably with that produced by contenders, such as opt-aiNet, IDEA and PSO.","Title: Multi-label structure learning with ising model selection\nYear: 2015\n\nAbstract: A common way of attacking multi-label classification problems is by splitting it into a set of binary classification problems, then solving each problem independently using traditional single-label methods. Nevertheless, by learning classifiers separately the information about the relationship between labels tends to be neglected. Built on recent advances in structure learning in Ising Markov Random Fields (I-MRF), we propose a multilabel classification algorithm that explicitly estimate and incorporate label dependence into the classifiers learning process by means of a sparse convex multitask learning formulation. Extensive experiments considering several existing multi-label algorithms indicate that the proposed method, while conceptually simple, outperforms the contenders in several datasets and performance metrics. Besides that, the conditional dependence graph encoded in the I-MRF provides a useful information that can be used in a posterior investigation regarding the reasons behind the relationship between labels.","Title: A Multitask Learning View on the Earth System Model Ensemble.\nYear: 2015\n\nAbstract: Earth system models (ESMs) are based on physical principles that are intended to emulate climate behavior. They're the primary mechanisms for obtaining projections of future conditions under different climate change scenarios. Because ESMs rely on the distinct modeling of certain physical processes and initial conditions, different ESMs can produce different responses for the same external forcing...","Title: On bicluster aggregation and its benefits for enumerative solutions\nYear: 2015\n\nAbstract: Biclustering involves the simultaneous clustering of objects and their attributes, thus defining local two-way clustering models. Recently, efficient algorithms were conceived to enumerate all biclusters in real-valued datasets. In this case, the solution composes a complete set of maximal and non-redundant biclusters. However, the ability to enumerate biclusters revealed a challenging scenario: in noisy datasets, each true bicluster may become highly fragmented and with a high degree of overlapping. It prevents a direct analysis of the obtained results. Aiming at reverting the fragmentation, we propose here two approaches for properly aggregating the whole set of enumerated biclusters: one based on single linkage and the other directly exploring the rate of overlapping. Both proposals were compared with each other and with the actual state-of-the-art in several experiments, and they not only significantly reduced the number of biclusters but also consistently increased the quality of the solution.","Title: Effects of Social Ties in Knowledge Diffusion: case study on PLOS ONE.\nYear: 2016\n\nAbstract: In order to capture the effects of social ties in knowledge diffusion, this paper examines the publication network that emerges from the collaboration of researchers, using citation information as means to estimate knowledge flow. For this purpose, we analyzed the papers published in the PLOS ONE journal finding strong evidence to support that the closer two authors are in the co-authorship network, the larger the probability that knowledge flow will occur between them. Moreover, we also found that when it comes to knowledge diffusion, strong co-authorship proximity is more determinant than geographic proximity.","Title: Spatial Projection of Multiple Climate Variables using Hierarchical Multitask Learning.\nYear: 2017\n\nAbstract: Future projection of climate is typically obtained by combining outputs from multiple Earth System Models (ESMs) for several climate variables such as temperature and precipitation. While IPCC has traditionally used a simple model output average, recent work has illustrated potential advantages of using a multitask learning (MTL) framework for projections of individual climate variables. In this paper we introduce a framework for hierarchical multitask learning (HMTL) with two levels of tasks such that each super-task, i.e., task at the top level, is itself a multitask learning problem over sub-tasks. For climate projections, each super-task focuses on projections of specific climate variables spatially using an MTL formulation. For the proposed HMTL approach, a group lasso regularization is added to couple parameters across the super-tasks, which in the climate context helps exploit relationships among the behavior of different climate variables at a given spatial location. We show that some recent works on MTL based on learning task dependency structures can be viewed as special cases of HMTL. Experiments on synthetic and real climate data show that HMTL produces better results than decoupled MTL methods applied separately on the super-tasks and HMTL significantly outperforms baselines for climate projection.","Title: An Extremal Optimization approach to parallel resonance constrained capacitor placement problem.\nYear: 2017\n\nAbstract: Installation of capacitors in distribution networks is one of the most used procedure to compensate reactive power generated by loads and, consequently, to reduce technical losses. So, the problem consists in identifying the optimal placement and sizing of capacitors. This problem is known in the literature as optimal capacitor placement problem. Neverthless, depending on the location and size of the capacitor, it may become a harmonic source, allowing capacitor to enter into resonance with the distribution network, causing several undesired side effects. In this work we propose a parsimonious method to deal with the capacitor placement problem that incorporates resonance constraints, ensuring that every allocated capacitor will not act as a harmonic source. This proposed algorithm is based upon a physical inspired metaheuristic known as Extremal Optimization. The results achieved showed that this proposal has reached significant gains when compared with other proposals that attempt repair, in a post-optimization stage, already obtained solutions which violate resonance constraints.","Title: Using Retweet Information as a Feature to Classify Messages Contents.\nYear: 2017\n\nAbstract: We investigate the use of machine learning algorithms to classify content published in Online Social Networks using as input solely user interaction data, instead of the actual message content. During a period of six months, we monitored and gathered data from users interacting with news messages on Twitter, creating thousands of information diffusion processes. The data set presented regular patterns on how messages were spread over the network by users, depending on its content, so we could build classifiers to predict the topic of a message using as input only information of how it was diffused. Thus, we demonstrate the explanatory power of user behavior data on identifying content present in Social Networks, proposing techniques for topic classification that can be used to assist traditional content identification strategies (such as natural language or image processing) in challenging contexts, or be applied in scenarios with limited information access.","Title: Efficient mining of maximal biclusters in mixed-attribute datasets.\nYear: 2017\n\nAbstract: This paper presents a novel enumerative biclustering algorithm to directly mine all maximal biclusters in mixed-attribute datasets (containing both numerical and categorical attributes), with or without missing values. The proposal is an extension of RIn-Close_CVC, which was originally conceived to mine perfect or perturbed biclusters with constant values on columns solely from numerical datasets, and without missing values. Even endowed with additional and more general features, the extended RIn-Close_CVC retains four key properties: (1) efficiency, (2) completeness, (3) correctness, and (4) non-redundancy. Our proposal is the first one to deal with mixed-attribute datasets without requiring any pre-processing step, such as discretization and itemization of real-valued attributes. This is a decisive aspect, because discretization and itemization implies a priori decisions, with information loss and no clear control over the consequences. On the other hand, even having to specify a priori an individual threshold for each numerical attribute, that will be used to indicate internal consistency per attribute, each threshold will be applied during the construction of the biclusters, shaping the peculiarities of the data distribution. We also explore the strong connection between biclustering and frequent pattern mining to (1) provide filters to select a compact bicluster set that exhibits high relevance and low redundancy, and (2) in the case of labeled datasets, automatically present the biclusters in a user-friendly and intuitive form, by means of quantitative class association rules. Our experimental results showed that the biclusters yield a parsimonious set of relevant rules, providing useful and interpretable models for five mixed-attribute labeled datasets.","Title: Many-Objective Ensemble-Based Multilabel Classification.\nYear: 2017\n\nAbstract: This paper proposes a many-objective ensemble-based algorithm to explore the relations among the labels on multilabel classification problems. This proposal consists in two phases. In the first one, a many-objective optimization method generates a set of candidate components exploring the relations among the labels, and the second one uses a stacking method to aggregate the components for each label. By balancing or not the relevance of each label, two versions were conceived for the proposal. The balanced one presented a good performance for recall and F1 metrics, and the unbalanced one for 1-Hamming loss and precision metrics.","Title: Ensembles of Multiobjective-Based Classifiers for Detection of Epileptic Seizures.\nYear: 2017\n\nAbstract: This paper proposes multiobjective-based classifiers to detect epileptic seizures using ensemble approaches, transfer-learning methods, and three alternative feature extraction techniques. Two aspects of the problem were investigated: (1) the relative merit of distinct proposals to synthesize an ensemble of classifiers, considering all the three feature extraction techniques; (2) the potential of an ensemble composed of transfer-learned classifiers. The blend approaches with the best performance detected all test seizures, with a high proportion of correctly detected samples inside the seizure interval and high proportion of time intervals correctly classified as non-seizures.","Title: RIn-Close_CVC2: an even more efficient enumerative algorithm for biclustering of numerical datasets.\nYear: 2018\n\nAbstract: RIn-Close_CVC is an efficient (take polynomial time per bicluster), complete (find all maximal biclusters), correct (all biclusters attend the user-defined level of consistency) and non-redundant (all the obtained biclusters are maximal and the same bicluster is not enumerated more than once) enumerative algorithm for mining maximal biclusters with constant values on columns in numerical datasets. Despite RIn-Close_CVC has all these outstanding properties, it has a high computational cost in terms of memory usage because it must keep a symbol table in memory to prevent a maximal bicluster to be found more than once. In this paper, we propose a new version of RIn-Close_CVC, named RIn-Close_CVC2, that does not use a symbol table to prevent redundant biclusters, and keeps all these four properties. We also prove that these algorithms actually possess these properties. Experiments are carried out with synthetic and real-world datasets to compare RIn-Close_CVC and RIn-Close_CVC2 in terms of memory usage and runtime. The experimental results show that RIn-Close_CVC2 brings a large reduction in memory usage and, in average, significant runtime gain when compared to its predecessor.","Title: Multi-task Sparse Structure Learning with Gaussian Copula Models.\nYear: 2016\n\nAbstract: Multi-task learning (MTL) aims to improve generalization performance by learning multiple related tasks simultaneously. While sometimes the underlying task relationship structure is known, often the structure needs to be estimated from data at hand. In this paper, we present a novel family of models for MTL, applicable to regression and classification problems, capable of learning the structure of tasks relationship. In particular, we consider a joint estimation problem of the tasks relationship structure and the individual task parameters, which is solved using alternating minimization. The task relationship revealed by structure learning is founded on recent advances in Gaussian graphical models endowed with sparse estimators of the precision (inverse covariance) matrix. An extension to include flexible Gaussian copula models that relaxes the Gaussian marginal assumption is also proposed. We illustrate the e ff ectiveness of the proposed model on a variety of synthetic and benchmark data sets for regression and classi fi cation. We also consider the problem of combining Earth System Model (ESM) outputs for better projections of future climate, with focus on projections of temperature by combining ESMs in South and North America, and show that the proposed model outperforms several existing methods for the problem.","Title: Simulated annealing for symbolic regression\nYear: 2021\n\nAbstract: ABSTRACTSymbolic regression aims to hypothesize a functional relationship involving explanatory variables and one or more dependent variables, based on examples of the desired input-output behavior. Genetic programming is a meta-heuristic commonly used in the literature to achieve this goal. Even though Symbolic Regression is sometimes associated with the potential of generating interpretable expressions, there is no guarantee that the returned function will not contain complicated constructs or even bloat. The Interaction-Transformation (IT) representation was recently proposed to alleviate this issue by constraining the search space to expressions following a simple and comprehensive pattern. In this paper, we resort to Simulated Annealing to search for a symbolic expression using the IT representation. Simulated Annealing exhibits an intrinsic ability to escape from poor local minima, which is demonstrated here to yield competitive results, particularly in terms of generalization, when compared with state-of-the-art Symbolic Regression techniques, that depend on population-based meta-heuristics, and committees of learning machines.","Title: Wind Speed Forecasting via Multi-task Learning\nYear: 2021\n\nAbstract: The viability of wind power massive use goes through an effective estimation of the power to be produced in wind farms, one of the fastest growing sources of renewable energy. Therefore, short-term wind speed forecasting involving several turbines and possibly multiple wind farms becomes a crucial practical demand. Once there are historical time series of wind speed intensity associated with subgroups of turbines, multiple tasks of time series prediction should be solved simultaneously. Aiming at performance improvement, we propose here the adoption of information sharing in multi-task learning frameworks, involving linear and nonlinear prediction models, and time series with and without differentiation. The nonlinear multi-task learning approach is based on extreme learning machines (ELMs) and guided to the best accuracies on a real-world case study of short-term wind speed forecasting. A suitable tuning of the variance in the distribution of the random hidden layer weights was demonstrated to be an essential preprocessing step for the ELMs, together with regularization routines. Additionally, it was found that properly incorporating tasks relationship into the learning process can systematically contribute to better generalization, particularly when compared to the single task learning counterparts.","Title: Online Convex Optimization of a Multi-task Fuzzy Rule-based Evolving System\nYear: 2022\n\nAbstract: This paper extends the recently conceived learning mechanism called EVeP (Extreme Value evolving Predictor), an evolving fuzzy-rule-based predictor characterized by innovative procedures to define the antecedent and consequent parts of the existing fuzzy rules. In EVeP, information granules are recursively updated and associated with Weibull distributions, a generalization of Gaussian distributions which incorporates more robust statistics to establish the region of influence of each fuzzy rule. Shared information from all the rules, in a multi-task formulation, is adopted to set the consequent parameters in EVeP. Given that the multi-task formulation is solved using batch learning and gradient descent, the computational cost per iteration tends to be high, being a concern in practical applications. Therefore, here the multi-task framework at the consequent part of the rules was revised to incorporate online convex optimization, given rise to EVeP_OCO. Now, antecedent and consequent parts of the rules are updated in a fully recursive way, with a clear reduction in the computational burden per iteration, particularly when the worst case scenarios are considered: the cost per iteration depends on the current number of rules to be updated. The case studies are composed of a variety of benchmark time series prediction problems. They demonstrate the significant gain in terms of computational cost per iteration, with an admissible reduction in performance by replacing a batch multi-task learning procedure by an online counterpart.","Title: Multi-Objective Bilevel Recommender System for Food Diets\nYear: 2022\n\nAbstract: This work aimed to develop a personalized multi-objective recommender system for food diets, which seeks to suggest to the user a diverse list of four meals a day (breakfast, lunch, snack, dinner). The efficient solutions are capable of simultaneously meeting a list of nutritional specifications and also minimizing both the concentration in a specific meal or food item and the total cost of acquisition and preparation. Efficient solutions are sought from the joint use of the NSGA-II and Gurobi optimization packages, after formulating the diet problem as a bilevel optimization: a combinatorial and multi-objective problem at the upper level - which food items from each category (the available food categories and the categories allocated to each of the four meals are defined in advance) should make up the diet - and a mathematical programming problem at the lower level - what is the optimal amount of the selected food items to compose the diet, given nutritional constraints. As Gurobi does not operate directly with a multi-objective optimization perspective, its lower-level objective function involves maximizing the total energy of the daily diet. Experimental results, considering fictitious food costs, show that NSGA-II and Gurobi operate in synergy, providing a diverse list of menus for the four daily meals, thus making a valuable approximation of the Pareto frontier. The distinctive aspect of this multi-objective bilevel solution to the diet problem, then, resides in the supply of diverse and, at the same time, efficient candidate solutions, in the sense of achieving the Pareto frontier and being scattered along its extension."],"string":"[\n \"Title: Online learning in estimation of distribution algorithms for dynamic environments.\\nYear: 2011\\n\\nAbstract: In this paper, we propose an estimation of distribution algorithm based on an inexpensive Gaussian mixture model with online learning, which will be employed in dynamic optimization. Here, the mixture model stores a vector of sufficient statistics of the best solutions, which is subsequently used to obtain the parameters of the Gaussian components. This approach is able to incorporate into the current mixture model potentially relevant information of the previous and current iterations. The online nature of the proposal is desirable in the context of dynamic optimization, where prompt reaction to new scenarios should be promoted. To analyze the performance of our proposal, a set of dynamic optimization problems in continuous domains was considered with distinct levels of complexity, and the obtained results were compared to the results produced by other existing algorithms in the dynamic optimization literature.\",\n \"Title: An Immune-Evolutionary Algorithm for Multiple Rearrangements of Gene Expression Data\\nYear: 2004\\n\\nAbstract: Microarray technologies are employed to simultaneously measure expression levels of thousands of genes. Data obtained from such experiments allow inference of individual gene functions, help to identify genes from specific tissues, to analyze the behavior of gene expression levels under various environmental conditions and under different cell cycle stages, and to identify inappropriately transcribed genes and several genetic diseases, among many other applications. As thousands of genes may be involved in a microarray experiment, computational tools for organizing and providing possible visualizations of the genes and their relationships are crucial to the understanding and analysis of the data. This work proposes an algorithm based on artificial immune systems for organizing gene expression data in order to simultaneously reveal multiple features in large amounts of data. A distinctive property of the proposed algorithm is the ability to provide a diversified set of high-quality rearrangements of the genes, opening up the possibility of identifying various co-regulated genes from representative graphical configurations of the expression levels. This is a very useful approach for biologists, because several co-regulated genes may exist under different conditions.\",\n \"Title: Omni-aiNet: an immune-inspired approach for omni optimization\\nYear: 2006\\n\\nAbstract: This work presents omni-aiNet, an immune-inspired algorithm developed to solve single and multi-objective optimization problems, either with single and multi-global solutions. The search engine is capable of automatically adapting the exploration of the search space according to the intrinsic demand of the optimization problem. This proposal unites the concepts of omni-optimization, already proposed in the literature, with distinctive procedures associated with immune-inspired concepts. Due to the immune inspiration, the omni-aiNet presents a population capable of adjusting its size during the execution of the algorithm, according to a predefined suppression threshold, and a new grid mechanism to control the spread of solutions in the objective space. The omni-aiNet was applied to several optimization problems and the obtained results are presented and analyzed.\",\n \"Title: The Influence of the Pool of Candidates on the Performance of Selection and Combination Techniques in Ensembles\\nYear: 2006\\n\\nAbstract: In this paper, we propose the use of an immuneinspired approach called opt-aiNet to generate a diverse set of high-performance candidates to compose an ensemble of neural network classifiers. Being a population-based search algorithm, the opt-aiNet is capable of maintaining diversity and finding many high-performance solutions simultaneously, which are known to be desired features when synthesizing an ensemble. Concerning the selection and combination phases, the most relevant selection and combination techniques already proposed in the literature have been considered. The main contribution of this paper is the indication that there is no pair of selection/combination technique that can be considered the best one, because the performance of the obtained ensemble varies significantly with the current composition of the pool of candidates already produced by the generation phase. Notwithstanding, this variability in performance is not restricted to the choice of opt-aiNet as the generative device. As a consequence, to overcome the performance of the best individual classifier, every possible pairs of selection and combination techniques should be tried. Only with such an exhaustive search (notice that the main computational burden is usually related to the generation phase), the performance of the ensemble was invariably superior to the performance of the best individual classifier on four benchmark classification problems.\",\n \"Title: bicACO: An Ant Colony Inspired Biclustering Algorithm\\nYear: 2008\\n\\nAbstract: A recent proposal developed to avoid some of the drawbacks presented by standard clustering algorithms is the so-called biclusteringtechnique [1], which performs clustering of rows and columns of the data matrix simultaneously, allowing the extraction of additional information from the dataset. Since the biclustering problem is combinatorial, and ant-based systems present several advantages when dealing with this kind of problems [2], in this work we propose an antinspired algorithm for biclustering, which was named bicACO.\",\n \"Title: A multiobjective analysis of adaptive clustering algorithms for the definition of RBF neural network centers in regression problems\\nYear: 2012\\n\\nAbstract: A variety of clustering algorithms have been applied to determine the internal structure of Radial Basis Function Neural Networks (RBFNNs). k-means algorithm is one of the most common choice for this task, although, like many other clustering algorithms, it needs to receive the number of prototypes a priori. This is a nontrivial procedure, mainly for real-world applications. An alternative is to use algorithms that automatically determine the number of prototypes. In this paper, we performed a multiobjective analysis involving three of these algorithms, which are: Adaptive Radius Immune Algorithm (ARIA), Affinity Propagation (AP), and Growing Neural Gas (GNG). For each one, the parameters that most influence the resulting number of prototypes composed the decision space, while the RBFNN RMSE and the number of prototypes formed the objective space. The experiments found that ARIA solutions achieved the best results for the multiobjective metrics adopted in this paper.\",\n \"Title: An artificial immune network for multimodal function optimization on dynamic environments\\nYear: 2005\\n\\nAbstract: Multimodal optimization algorithms inspired by the immune system are generally characterized by a dynamic control of the population size and by diversity maintenance along the search. One of the most popular proposals is denoted opt-aiNet (artificial immune network for optimization) and is extended here to deal with time-varying fitness functions. Additional procedures are designed to improve the overall performance and the robustness of the immune-inspired approach, giving rise to a version for dynamic optimization, denoted dopt-aiNet. Firstly, challenging benchmark problems in static multimodal optimization are considered to validate the new proposal. No parameter adjustment is necessary to adapt the algorithm according to the peculiarities of each problem. In the sequence, dynamic environments are considered, and usual evaluation indices are adopted to assess the performance of dopt-aiNet and compare with alternative solution procedures available in the literature.\",\n \"Title: Multiple criteria optimization based on unsupervised learning and fuzzy inference applied to the vehicle routing problem\\nYear: 2002\\n\\nAbstract: This paper presents a neuro-fuzzy system based on competitive learning to solve multiple criteria optimization problems. The proposed method promotes the simultaneous self-organization of several networks, employing unsupervised learning guided by a fuzzy rule base. The algorithm implements a policy of penalties and rewards, a strategy of neuron inhibition, insertion and pruning, and also takes into account certain statistical characteristics of the input space. A fuzzy inference system is designed to implement the decision making process under a multiobjective scenario, leading to an adaptive process of constraint relaxation. The effectiveness of the proposed method is attested by means of a series of computational simulations performed on standard data. In our simulations, we address two variants of the vehicle routing problem: the capacitated vehicle routing problem (CVRP) and the multiple traveling salesman problem (MTSP). There are a few works treating the vehicle routing problem by means of competitive learning. These approaches are briefly reviewed in this paper. We also present some improvements in the results of an implementation of tabu search by providing the solutions obtained by the neuro-fuzzy system as initial condition, showing that the proposed method can effectively produce satisfactory results when used in association with more dedicated approaches.\",\n \"Title: Performance analysis of nonlinear echo state network readouts in signal processing tasks.\\nYear: 2012\\n\\nAbstract: Echo state networks (ESNs) characterize an attractive alternative to conventional recurrent neural network (RNN) approaches as they offer the possibility of preserving, to a certain extent, the processing capability of a recurrent architecture and, at the same time, of simplifying the training process. However, the original ESN architecture cannot fully explore the potential of the RNN, given that only the second-order statistics of the signals are effectively used. In order to overcome this constraint, distinct proposals promote the use of a nonlinear readout aiming to explore higher-order available information though still maintaining a closed-form solution in the least-squares sense. In this work, we review two proposals of nonlinear readouts - a Volterra filter structure and an extreme learning machine - and analyze the performance of these architectures in the context of two relevant signal processing tasks: supervised channel equalization and chaotic time series prediction. The obtained results reveal that the nonlinear readout can be decisive in the process of aproximating the desired signal. Additionally, we discuss the possibility of combining both ideas of nonlinear readouts and preliminary results indicate that a performance improvement can be attained.\",\n \"Title: Hybrid neural networks: An evolutionary approach with local search\\nYear: 2002\\n\\nAbstract: Considering computational algorithms available in the literature, associated with supervised learning in feedforward neural networks, a wide range of distinct approaches can be identified. While the adjustment of the connection weights represents an omnipresent stage, the algorithms differ on three basic aspects: the technique chosen to determine the dimension of the multilayer neural network, the procedure adopted to specify the activation functions, and the kind of composition used to produce the output. Advanced learning algorithms should be developed to simultaneously treat all these aspects during learning, and an evolutionary learning algorithm with local search is proposed here. The essence of this approach is a synergy between genetic algorithms and conjugate gradient optimization, operating on a hybrid neural network architecture. As a consequence, the final neural network is automatically generated, and is characterized to be dedicated and computationally parsimonious.\",\n \"Title: Evolving Phylogenetic Trees: an Alternative to Black-Box Approaches\\nYear: 2002\\n\\nAbstract: This paper presents the main steps to reconstruct phylogenetic trees using an evolutionary algorithm with a wide range of alternative computational procedures for f itness evaluation and for tuning the search engine. Due to the fine equilibrium between exploration and exploitation of the search space, high- quality reconstructions are obtained among an explosion of candidates and with the additional requirement of adjusting the length of all branches in a given tree. A toolbox is available, and the matrix codification adopted, the a ssociated g e- netic operators, and the optimization procedures involved are clearly described to guarantee reproducibility.\",\n \"Title: Recent Advances In Gene Expression Data Clustering: A Case Study With Comparative Results\\nYear: 2004\\n\\nAbstract: Several advanced techniques have been proposed for data clustering and many of them have been applied to gene expression data, with partial success. The high dimensionality and the multitude of admissible perspectives for data analysis of gene expression require additional computational resources, such as hierarchical structures and dynamic allocation of resources. We present an immune-inspired hierarchical clustering device, called hierarchical artificial immune network (HaiNet), especially devoted to the analysis of gene expression data. This technique was applied to a newly generated data set, involving maize plants exposed to different aluminum concentrations. The performance of the algorithm was compared with that of a self-organizing map, which is commonly adopted to deal with gene expression data sets. More consistent and informative results were obtained with HaiNet.\",\n \"Title: MOBAIS: A Bayesian Artificial Immune System for Multi-Objective Optimization\\nYear: 2008\\n\\nAbstract: Significant progress has been made in theory and design of artificial immune systems (AISs) for solving multi-objective problems accurately. However, an aspect not yet widely addressed by the research reported in the literature is the lack of ability of the AIS to deal effectively with building blocks (high-quality partial solutions coded in the antibody). The available AISs present mechanisms for evolving the population that do not take into account the relationship among the variables of the problem, causing the disruption of these high-quality partial solutions. Recently, we proposed a novel immune-inspired approach for single-objective optimization as an attempt to avoid this drawback. Our proposal replaces the traditional mutation and cloning operators with a probabilistic model, more specifically a Bayesian network representing the joint distribution of promising solutions and, subsequently, uses this model for sampling new solutions. Now, in this paper we extend our methodology for solving multi-objective optimization problems. The proposal, called Multi-Objective Bayesian Artificial Immune System (MOBAIS), was evaluated in the well-known multi-objective Knapsack problem and its performance compares favorably with that produced by contenders such as NSGA-II, MISA, and mBOA.\",\n \"Title: Immune-inspired Dynamic Optimization for Blind Spatial Equalization in Undermodeled Channels.\\nYear: 2006\\n\\nAbstract: In this work, we propose an evolutionary-like approach to the problem of blind adaptive spatial filtering that is based on the decision-directed criterion and on the doptaiNet, an artificial immune network conceived to perform multimodal search in dynamic environments. The proposal was tested under static and time-varying undermodeled channel models, and, in all cases, its ability to find and track a solution close to the Wiener global optimum was attested. The obtained results reveal that the dopt-aiNet may decisively enhance the performance of adaptive arrays in scenarios built from elements that are representative of some aspects of real-world communication systems.\",\n \"Title: Applying biclustering to text mining: an immune-inspired approach\\nYear: 2007\\n\\nAbstract: With the rapid development of information technology, computers are proving to be a fundamental tool for the organization and classification of electronic texts, given the huge amount of available information. The existent methodologies for text mining apply standard clustering algorithms to group similar texts. However, these algorithms generally take into account only the global similarities between the texts and assign each one to only one cluster, limiting the amount of information that can be extracted from the texts. An alternative proposal capable of solving these drawbacks is the biclustering technique. The biclustering is able to perform clustering of rows and columns simultaneously, allowing a more comprehensive analysis of the texts. The main contribution of this paper is the development of an immune-inspired biclustering algorithm to carry out text mining, denoted BIC-aiNet. BIC-aiNet interprets the biclustering problem as several two-way bipartition problems, instead of considering a single two-way permutation framework. The experimental results indicate that our proposal is able to group similar texts efficiently and extract implicit useful information from groups of texts.\",\n \"Title: Evolutionary Stigmergy In Multipurpose Navigation Systems\\nYear: 2006\\n\\nAbstract: Autonomous robot navigation involves many challenges and difficulties which are augmented when multiple robots operate together. Sophisticated computational techniques are required to cope with autonomous navigation in collective robotics, being the biologically-inspired approaches the most frequently adopted. Stigmergy, i.e. the ants communication by means of pheromones, is the main biological metaphor used in this work to perform multi-robot communication. The robots will be able to mark regions of the environment with artificial pheromones, according to past experiences, assisting one another in a cooperative and indirect way to accomplish the navigation objectives. Each robot is controlled by an autonomous navigation system (ANS) based on Learning Classifier System, which evolves during navigation from no a priori knowledge. Besides learning to avoid obstacles and capture targets, the systems must also learn how and where to lay artificial pheromones. Some experiments and simulations are performed intending to particularly investigate the ANS from three main perspectives: capability of learning to achieve the navigation objectives in collective scenarios, adaptability in face of environmental changes and ability to obtain optimized navigation behaviors by means of stigmergy.\",\n \"Title: Artificial homeostatic system: a novel approach\\nYear: 2005\\n\\nAbstract: Many researchers are developing frameworks inspired by natural, especially biological, systems to solve complex real-world problems. This work extends previous work in the field of biologically inspired computing, proposing an artificial endocrine system for autonomous robot navigation. Having intrinsic self-organizing behaviour, the novel artificial endocrine system can be applied to a wide range of problems, particularly those that involve decision making under changing environmental conditions, such as autonomous robot navigation. This work draws on “embodied cognitive science”, including the study of intelligence, adaptivity, homeostasis, and the dynamic aspects of cognition, in order to help lay down fundamental principles and techniques for a novel approach to more biologically plausible artificial homeostatic systems. Results from using the artificial endocrine system to control a simulated robot are presented.\",\n \"Title: A complex neighborhood based particle swarm optimization\\nYear: 2009\\n\\nAbstract: This paper proposes a new variant of the PSO algorithm named Complex Neighborhood Particle Swarm Optimizer (CNPSO) for solving global optimization problems. In the CNPSO, the neighborhood of the particles is organized through a complex network which is modified during the search process. This evolution of the topology seeks to improve the influence of the most successful particles and it is fine tuned for maintaining the scale-free characteristics of the network while the optimization is being performed. The use of a scale-free topology instead of the usual regular or global neighborhoods is intended to bring to the search procedure a better capability of exploring promising regions without a premature convergence, which would result in the procedure being easily trapped in a local optimum. The performance of the CNPSO is compared with the standard PSO on some well-known and high-dimensional benchmark functions, ranging from multimodal to plateau-like problems. In all the cases the CNPSO outperformed the standard PSO.\",\n \"Title: Data clustering with a neuro-immune network\\nYear: 2005\\n\\nAbstract: This paper proposes a novel constructive learning algorithm for a competitive neural network. The proposed algorithm is developed by taking ideas from the immune system and demonstrates robustness for data clustering in the initial experiments reported here for three benchmark problems. Comparisons with results from the literature are also provided. To automatically segment the resultant neurons at the output, a tool from graph theory was used with promising results. A brief sensitivity analysis of the algorithm was performed in order to investigate the influence of the main user-defined parameters on the learning speed and accuracy of the results presented. General discussions and avenues for future works are also provided.\",\n \"Title: Genetic programming for automating the development of data management algorithms in information technology systems\\nYear: 2012\\n\\nAbstract: Information technology (IT) systems are present in almost all fields of human activity, with emphasis on processing, storage, and handling of datasets. Automated methods to provide access to data stored in databases have been proposed mainly for tasks related to knowledge discovery and datamining (KDD). However, for this purpose, the database is used only to query data in order to find relevant patterns associated with the records. Processes modelled on IT systems should manipulate the records to modify the state of the system. Linear genetic programming for databases (LGPDB) is a tool proposed here for automatic generation of programs that can query, delete, insert, and update records on databases. The obtained results indicate that the LGPDB approach is able to generate programs for effectively modelling processes of IT systems, opening the possibility of automating relevant stages of data manipulation, and thus allowing human programmers to focus on more complex tasks.\",\n \"Title: Applying Biclustering to Perform Collaborative Filtering\\nYear: 2007\\n\\nAbstract: Collaborative filtering (CF) is a method to perform automated suggestions for a user based on the opinion of other users with similar interest. Most of the CF algorithms do not take into account the existent duality between users and items, considering only the similarities between users or only the similarities between items. In this paper we propose a novel methodology for the CF capable of dealing with this situation. By proposing an immune-inspired biclustering technique to carry out clustering of rows and columns at the same time, our algorithm is able to group similarities between users and items. In order to evaluate the proposed methodology, we have applied it to MovieLens dataset which contains user's ratings to a large set of movies. The results indicate that our proposal is able to provide useful recommendations for the users, outperforming other methodologies for CF reported in the literature.\",\n \"Title: Definition of Capacited p-Medians by a Modified Max Min Ant System with Local Search\\nYear: 2004\\n\\nAbstract: This work introduces a modified MAX MIN Ant System (MMAS) designed to solve the Capacitated p-Medians Problem (CPMP). It presents the most relevant steps towards the implementation of an MMAS to solve the CPMP, including some improvements on the original MMAS algorithm, such as the use of a density model in the information heuristics and a local search adapted from the uncapacitated p-medians problem. Extensions to a recently proposed updating rule for the pheromone level, aiming at improving the MMAS ability to deal with large-scale instances, are also presented and discussed. Some simulations are performed using instances available from the literature, and well-known heuristics are employed for benchmarking.\",\n \"Title: Evolving Arbitrarily Connected Feedforward Neural Networks via Genetic Algorithms\\nYear: 2010\\n\\nAbstract: Though several approaches have already been proposed in the literature to evolve neural network topologies for solving a wide range of machine learning tasks, this paper presents an alternative one, capable of evolving arbitrarily connected feed forward neural networks (ACFNNs), including linear and nonlinear neurons. A genetic algorithm is conceived to adjust the topology and also to perform variable selection. The weights of the obtained neural networks, with arbitrary topologies, are adjusted using a simple descent gradient algorithm. The purpose is to obtain high-quality and parsimonious predictors for two real-world and one synthetic time series. The obtained results are compared with the ones produced by traditional MLP models and Mixtures of Heterogeneous Experts (MHEs).\",\n \"Title: A constructive algorithm to synthesize arbitrarily connected feedforward neural networks\\nYear: 2012\\n\\nAbstract: In this work we present a constructive algorithm capable of producing arbitrarily connected feedforward neural network architectures for classification problems. Architecture and synaptic weights of the neural network should be defined by the learning procedure. The main purpose is to obtain a parsimonious neural network, in the form of a hybrid and dedicate linear/nonlinear classification model, which can guide to high levels of performance in terms of generalization. Though not being a global optimization algorithm, nor a population-based metaheuristics, the constructive approach has mechanisms to avoid premature convergence, by mixing growing and pruning processes, and also by implementing a relaxation strategy for the learning error. The synaptic weights of the neural networks produced by the constructive mechanism are adjusted by a quasi-Newton method, and the decision to grow or prune the current network is based on a mutual information criterion. A set of benchmark experiments, including artificial and real datasets, indicates that the new proposal presents a favorable performance when compared with alternative approaches in the literature, such as traditional MLP, mixture of heterogeneous experts, cascade correlation networks and an evolutionary programming system, in terms of both classification accuracy and parsimony of the obtained classifier.\",\n \"Title: Evolving phylogenetic trees: a multiobjective approach\\nYear: 2007\\n\\nAbstract: This work presents the application of the omni-aiNet algorithm - an immune-inspired algorithm originally developed to solve single and multiobjective optimization problems - to the construction of phylogenetic trees. The main goal of this work is to automatically evolve a population of phylogenetic unrooted trees, possibly with distinct topologies, by minimizing at the same time the minimal evolution and the mean-squared error criteria. The obtained set of phylogenetic trees contains non-dominated individuals that form the Pareto front and that represent the trade-off of the two conflicting objectives. The proposal of multiple non-dominated solutions in a single run gives to the user the possibility of having distinct explanations for the difference observed in the terminal nodes of the tree, and also indicates the restrictive feedback provided by the individual application of well-known algorithms for phylogenetic reconstruction that takes into account both optimization criteria, like Neighbor Joining.\",\n \"Title: A concentration-based artificial immune network for multi-objective optimization\\nYear: 2011\\n\\nAbstract: Until recently, the main focus of researchers that develop algorithms for evolutionary multi-objective optimization has been the creation of mechanisms capable of obtaining sets of solutions that are as close as possible to the true Pareto front of the problem and also as diverse as possible in the objective space, to properly cover such front. However, an adequate maintenance of diversity in the decision space is also important, to efficiently solve several classes of problems and even to facilitate the post-optimization decision making process. This aspect has been widely studied in evolutionary single-objective optimization, what led to the development of several diversity maintenance techniques. Among them, the recently proposed concentration-based artificial immune network (cob-aiNet), which is capable of self-regulating the population size, presented promising results in multimodal problems. So, it is extended here to deal with multi-objective problems that require a proper maintenance of diversity in the decision space.\",\n \"Title: Autonomous navigation system applied to collective robotics with ant-inspired communication\\nYear: 2005\\n\\nAbstract: Research in collective robotics is motivated mainly by the possibility of achieving an efficient solution to multi-objective navigation tasks when multiple robots are employed, instead of a single robot. Several approaches have already been tried in multi-robot systems, but the bio-inspired ones are the most frequent. This paper proposes to augment an autonomous navigation system based on learning classifier systems for using in collective robotics, introducing an inter-robot communication mechanism inspired by ant stigmergy, with each robot acting independently and cooperatively. The navigation system has no innate basic behavior and all knowledge necessary to compose the decision-making artifact is evolved as a function of the environmental feedback only, during navigation. Repulsive and/or attractive pheromone trails are produced by the robots along navigation, following very simple rules. Basically, each robot has to perform obstacle avoidance and target search, and the status of the pheromone level at the position currently occupied by each robot will influence the coordination of the two fundamental behaviors. Experiments are performed in simulation, with comparative results indicating that the presence of the pheromone trails is responsible for significant improvements in the capture rate and in the length of the route adopted by each robot.\",\n \"Title: Mixture of Experts Applied to Nonlinear Dynamic Systems Identification: A Comparative Study\\nYear: 2002\\n\\nAbstract: A mixture of experts (ME) model provides a modularapproach wherein component neural networks are madespecialists on subparts of a problem. In this framework,that follows the \\\"divide-and-conquer\\\" philosophy, agating network learns how to softly partition the inputspace into regions to be each properly modeled by one ormore expert networks. In this paper, we investigate theapplication of different ME variants to some multivariatenonlinear dynamic systems identification problems whichare known to be difficult to be dealt with. The aim is toprovide a comparative performance analysis betweenvariable settings of the standard, gated, and localized MEmodels with more conventional NN models.\",\n \"Title: ArTVox: Evolutionary Composition in Visual and Sound Domains\\nYear: 2006\\n\\nAbstract: Computational creativity is certainly interesting and potentially important. ArTVox, a Java programmed evolutionary environment, arose from the attempt to emulate computational creativity applied to artistic production, in visual and sound domains, by using interactive genetic algorithms. Objects inspired in Kandinsky's artworks are being programmed in Shape, an auxiliary Java environment, to be inserted in ArTVox. Today, ArTVox creates and evolves visual compositions of geometric primitives that, by their turn, guide the sound production in another evolutionary environment, JaVox, integrated to ArTVox.\",\n \"Title: A multi-Gaussian component EDA with restarting applied to direction of arrival tracking\\nYear: 2013\\n\\nAbstract: This paper analyzes the application of a multi-population Gaussian-based estimation of distribution algorithm equipped with a restarting strategy and mutation, named MGcEDA, to the problem of estimating the Direction of Arrival (DOA) of time-varying plane waves impinging on a uniform linear array of sensors. This problem requires the minimization of a dynamic cost function which is non-linear, non-quadratic, multimodal and variant with respect to the signal-to-noise ratio. Experiments showed that MGcEDA was able to quickly respond to changes in the source features in scenarios with different levels of noise and number of signals. Moreover, MGcEDA outperforms a previously proposed approach in all considered experiments in terms of well known performance measures.\",\n \"Title: Data Partition And Variable Selection For Time Series Prediction Using Wrappers\\nYear: 2006\\n\\nAbstract: The purpose of this paper is a comparative study of a non-exhaustive, though representative, set of methodologies already available for the partition of the training dataset in time series prediction, and also for variable selection under the wrapper paradigm. The partition policy of the training dataset and the choice of a proper set of variables for the regression vector are known to have a significant influence in the accuracy of the predictor, no matter the choice of the prediction model. However, there has been no extensive search for a figure of merit supporting a comparative analysis. Here, two partition policies, denoted sequential and random, are compared, and among the Variable selection approaches using wrappers, forward selection is contrasted with sensitivity based pruning. Five real financial time series with trends and seasonality have been considered and multilayer perceptrons are adopted as the predictor. The obtained results indicate with high confidence that the rarely adopted random partition and the computationally intensive forward selection overcomes the contestants in the whole set of experiments.\",\n \"Title: An Immune-Inspired Approach to Bayesian Networks\\nYear: 2005\\n\\nAbstract: During the last few years, there has been an increasing interest in using heuristic search algorithms based on natural selection (the so-called \\\"evolutionary algorithms\\\") for solving a wide variety of problems. As in any other discipline, research on ...\",\n \"Title: Handling Time-Varying Tsp Instances\\nYear: 2006\\n\\nAbstract: Multimodal optimization algorithms are being adapted to deal with dynamic optimization, mainly due to their ability to provide a faster reaction to unexpected changes in the optimization surface. The faster reaction may be associated with the existence of two important attributes in population-based algorithms devoted to multimodal optimization: simultaneous maintenance of multiple local optima in the population; and self-regulation of the population size along the search. The optimization surface may be subject to variations motivated by one of two main reasons: modification of the objectives to be fulfilled and change in parameters of the problem. An immune-inspired algorithm specially designed to deal with combinatorial optimization is applied here to solve time-varying TSP instances, with the cost of going from one city to the other being a function of time. The proposal presents favorable results when compared to the results produced by a high-performance ant colony optimization algorithm of the literature.\",\n \"Title: On-Line Approach for Loss Reduction in Electric Power Distribution Networks Using Learning Classifier Systems\\nYear: 2001\\n\\nAbstract: The problem of minimization of energy losses in power distribution systems can be formulated as obtaining the \\\"best\\\" network configuration, through the manipulation of sectionalizing switches. Using graph terminology, we have a combinatorial optimization problem, whose solution corresponds to finding a minimum spanning tree for the network. As an on-line approach to loss reduction in electric power distribution networks, this paper relies on Learning Classifier Systems to continually proposed network configurations close to the one associated with minimum energy losses, in the case of time-varying profiles of energy requirement. In order to evolve the set of rules that composes the Classifier System, operators for selection, reproduction and mutation are applied. Case studies illustrate the possibilities of this approach.\",\n \"Title: An Integrated System For Phylogenetic Inference Using Evolutionary Algorithms\\nYear: 2002\\n\\nAbstract: This paper presents all the steps to reconstruct phylogenetic trees using an evolutionary algorithm. The fitness criterion is based on maximum likelihood and a toolbox is available. The codification, genetic operators and the optimization procedures involved are clearly described to guarantee reproducibility.\",\n \"Title: Hybridizing mixtures of experts with support vector machines: Investigation into nonlinear dynamic systems identification\\nYear: 2007\\n\\nAbstract: Mixture of experts (ME) models comprise a family of modular neural network architectures aiming at distilling complex problems into simple subtasks. This is done by deploying a separate gating module for softly dividing the input space into overlapping regions to be each assigned to one or more expert networks. Conversely, support vector machines (SVMs) refer to kernel-based methods, neural-network-alike models that constitute an approximate implementation of the structural risk minimization principle. Such learning machines follow the simple, but powerful idea of nonlinearly mapping input data into high-dimensional feature spaces wherein a linear decision surface discriminating different regions is properly designed. In this work, we formally characterize and empirically evaluate a novel approach, named as Mixture of Support Vector Machine Experts (MSVME), whose main purpose is to combine the complementary properties of both SVM and ME models. In the formal characterization, an algorithm based on a maximum likelihood criterion is considered for the MSVME training, and we demonstrate that it is possible to train each expert based on an SVM perspective. Regarding the empirical evaluation, simulation results involving nonlinear dynamic system identification problems are reported, contrasting the performance shown by the MSVME approach with that exhibited by conventional SVM and ME models.\",\n \"Title: RABNET: a real-valued antibody network for data clustering\\nYear: 2005\\n\\nAbstract: This paper proposes a novel constructive learning algorithm for a competitive neural network. The proposed algorithm is developed by taking ideas from the immune system and demonstrates robustness in the initial experiments reported here for a benchmark problem. Comparisons with results from the literature are also provided. To automatically segment the resultant neurons at the output, a tool from graph theory was used with promising results. General discussions and avenues for future works are also provided.\",\n \"Title: Mapping Artificial Immune Systems into Learning Classifier Systems\\nYear: 2002\\n\\nAbstract: This paper presents one form of mapping Artificial Immune Systems (AIS) into Learning Classifier Systems (LCS). Artificial Immune Systems can be defined as adaptive systems inspired by theoretical models and principles of the biological immune system and applied to solve problems in the most diverse domains, from biology to computing. Similar to Learning Classifier Systems, already used to model complex adaptive systems, a better understanding of Artificial Immune Systems can be obtained when they are analysed under the perspective of complex adaptive systems. One of the goals here is to determine complementary features of both systems (LCS and AIS), aiming at providing a novel mapping conception. The formal treatment proposed along the paper may then be used to integrate models for complex adaptive systems.\",\n \"Title: A Hierarchical Immune Network Applied to Gene Expression Data\\nYear: 2004\\n\\nAbstract: This paper describes a new proposal for gene expression data analysis. The method used is based on a hierarchical approach\\n to a hybrid algorithm, which is composed of an artificial immune system, named aiNet, and a well known graph theoretic tool,\\n the minimal spanning tree (MST). This algorithm has already proved to be efficient for clustering gene expression data, but\\n its performance may decrease in some specific cases. However, through the use of a hierarchical approach of immune networks\\n it is possible to improve the clustering capability of the hybrid algorithm, such that it becomes more efficient, even when\\n the data set is complex. The proposed methodology is applied to the yeast data and gives important conclusions of the similarity\\n relationships among genes within the data set.\",\n \"Title: Pattern classification with mixtures of weighted least-squares support vector machine experts\\nYear: 2009\\n\\nAbstract: Support Vector Machine (SVM) classifiers are high-performance classification models devised to comply with the structural\\n risk minimization principle and to properly exploit the kernel artifice of nonlinearly mapping input data into high-dimensional\\n feature spaces toward the automatic construction of better discriminating linear decision boundaries. Among several SVM variants,\\n Least-Squares SVMs (LS-SVMs) have gained increased attention recently due mainly to their computationally attractive properties\\n coming as the direct result of applying a modified formulation that makes use of a sum-squared-error cost function jointly\\n with equality, instead of inequality, constraints. In this work, we present a flexible hybrid approach aimed at augmenting\\n the proficiency of LS-SVM classifiers with regard to accuracy/generalization as well as to hyperparameter calibration issues.\\n Such approach, named as Mixtures of Weighted Least-Squares Support Vector Machine Experts, centers around the fusion of the\\n weighted variant of LS-SVMs with Mixtures of Experts models. After the formal characterization of the novel learning framework,\\n simulation results obtained with respect to both binary and multiclass pattern classification problems are reported, ratifying\\n the suitability of the novel hybrid approach in improving the performance issues considered.\",\n \"Title: A proposal for direct-ordering gene expression data by self-organising maps\\nYear: 2004\\n\\nAbstract: Microarray experiments are employed to simultaneously measure the expression level of thousands of genes. The possible applications of the available datasets include: inference of individual gene functions; identification of genes from specific tissues; analysis of the behaviour of gene expression levels under various environmental conditions and under different cell cycle stages; and characterisation of inappropriately transcribed genes and several genetic diseases. A fundamental step in the analysis of gene expression data is the detection of genes presenting similar expression patterns. Considering that microarray technology allows the inspection of a wide range of aspects related to the genome at once, and therefore thousands of genes may be involved in an experiment, new computational tools for data analysis and alternative visualisation strategies are crucial to understand and uncover information present in the data. In this work, the gene ordering problem is modelled as a shortest-path problem and we present an algorithm based on competitive learning for rearranging gene expression data in a linear order aiming to reveal trends in large amount of data. The effectiveness of the algorithm is attested by means of computational simulations performed on publicly accessible data sets.\",\n \"Title: Support Vector Clustering Applied To Digital Communications\\nYear: 2006\\n\\nAbstract: Support Vector Clustering (SVC) is a recently proposed clustering methodology with promising performance for high-dimensional and noisy datasets, and for clusters with arbitrary shape. This work addresses the application of SVC, a kernel-based method, in a context in which the channel equalization problem is conceived as a clustering task The main challenge, in this case, is to perform unsupervised clustering aiming at the design of an optimal Bayesian or a blind prediction-based receiver without resorting to a priori information about the transmission medium. The proposed technique employs a two-stage procedure - a combination between the use of SVC to obtain a first set of clusters and an auxiliary heuristic to help separating eventual multiple clouds contained in a single cluster and attribute centers to them via an iterated local search (ELS) algorithm. The obtained results indicate that kernel methods can be successfully applied to the field of signal processing.\",\n \"Title: Improving a multi-objective multipopulation artificial immune network for biclustering\\nYear: 2009\\n\\nAbstract: The biclustering technique was developed to avoid some of the drawbacks presented by standard clustering techniques. Given that biclustering requires the optimization of at least two conflicting objectives and that multiple independent solutions are desirable as the outcome, a few multi-objective evolutionary algorithms for biclustering were proposed in the literature. However, apart from the individual characteristics of the biclusters that should be optimized during their construction, several other global aspects should also be considered, such as the coverage of the dataset and the overlap among biclusters. These requirements will be addressed in this work with the MOM-aiNet+ algorithm, which is an improvement of the original multi-objective multipopulation artificial immune network denoted MOM-aiNet. Here, the MOM-aiNet+ algorithm will be described in detail, its main differences from the original MOM-aiNet will be highlighted, and both algorithms will be compared, together with three other proposals from the literature.\",\n \"Title: Fault detection algorithm for telephone systems based on the danger theory\\nYear: 2005\\n\\nAbstract: This work is aimed at presenting a fault detection algorithm composed of multiple interconnected modules, and operating according to the paradigm supported by the danger theory in immunology. This algorithm attempts to achieve significant features that a fault detection system is supposed to have when monitoring a telephone profile system. These features would basically be adaptability due to the strong variation that operational conditions may exhibit over time, and the decrease in the number of false positives, which can be generated when any abnormal behavior is erroneously classified as being a fault. Simulated scenarios have been conceived to validate the proposal, and the obtained results are then analyzed.\",\n \"Title: Transductive Support Vector Machines for Cancer Diagnosis and Classification of Microarray Gene Expression Data\\nYear: 2002\\n\\nAbstract: purpose of this paper is to present a powerful methodology for classification u sing g ene expression d ata. The following problems will be considered: determination of specific diagnosis categories using the small round blue cell tumors of childhood, and classification o f genes in functional groups. The classification task will be performed by means of Transductive Inference with SVM, that will be implemented based on training and testing d ata sets. In the case of training samples, experts have already previously classified the genes or diagnostic c ategories in their respective classes. So, given the testing samples, the purpose is to determine the corresponding class. Thereby, we will be able to classify in only one step each sample of the testing data set as a member or not of a c ertain class. As the transductive inference has not been p reviously applied to the classification of gene expression data, considering SVM as the classification tool, it will be compared with the traditional inductive method in a series of exhaustive experiments, with promising results. 1 - Introduction With the purpose of improving generalization and increasing the performance of the estimator, transductive inference is considered h ere to estimate function values in o nly one step. It represents an alternative approach to inductive inference, that requires two steps: the first step (inductive) consists of discovering the functional dependence among input-output variables, and the second step (deductive) uses this functional dependence to evaluate the output in the points of interest. We will apply this methodology to the following problems: (A) Gene classification of budding yeast Saccharomyces Cerevisiae in functional groups using data of gene expression from an experiment of DNA microarray hybridization (1). The data set is composed of the ratio of the e xpression levels of 2467 genes of budding yeast Saccharomyces cerevisiae, measured in 7 9 different experiments of DNA microarray hybridization. The genes were classified into five functional classes: tricarboxylic a cid cycle (TCA), respiration (Resp), cytoplasmic ribosomal genes (Ribo), proteasome (Prot). The genes belonging to each functional class will be classified as positive, and the others genes will be classified as negative. The gain function to be maximized in testing d ata will be: G = GN - FP - 2xFN, where FP and FN are the number of false positives and false negatives, and GN is the null gain (1). We will consider the complexity index as: Complexity = 1 - (G / GN ). Using this index,\",\n \"Title: An immune-inspired multi-objective approach to the reconstruction of phylogenetic trees\\nYear: 2010\\n\\nAbstract: This work presents the application of the omni-aiNet algorithm—an immune-inspired algorithm originally developed to solve single and multi-objective optimization problems—to the reconstruction of phylogenetic trees. The main goal here is to automatically evolve a population of phylogenetic unrooted trees, possibly with distinct topologies, by minimizing at the same time two optimization criteria: the minimum evolution and the mean-squared error. This proposal generates, in a single run, a set of non-dominated solutions that represent the trade-offs of the two conflicting objectives, and gives the user the possibility of having distinct explanations for the differences observed at the terminal nodes of the trees. A series of experimental results is also reported in this work, in order to illustrate the effectiveness of the proposal and its capability to overcome the restrictive feedback provided by the application of well-known algorithms for phylogenetic reconstruction, such as the Neighbor Joining. Besides, the methodology presented in this work is compared to the popular NSGA-II algorithm, also modified to solve phylogenetic reconstruction problems.\",\n \"Title: A Multiagent-Based Constructive Approach for Feedforward Neural Networks\\nYear: 2003\\n\\nAbstract: In this paper, a new constructive approach for the automatic definition of feedforward neural networks (FNNs) is introduced. Such approach (named MASCoNN) is multiagent-oriented and, thus, can be regarded as a kind of hybrid (synergetic) system. MASCoNN centers upon the employment of a two-level hierarchy of agent-based elements for the progressive allocation of neuronal building blocks. By this means, an FNN can be considered as an architectural organization of reactive neural agents, orchestrated by deliberative coordination entities via synaptic interactions. MASCoNN was successfully applied to implement nonlinear dynamic systems identification devices and some comparative results, involving alternative proposals, are analyzed here.\",\n \"Title: Long-Term Time Series Prediction Using Wrappers For Variable Selection And Clustering For Data Partition\\nYear: 2007\\n\\nAbstract: In an attempt to implement long-term time series prediction based on the recursive application of a one-step-ahead multilayer neural network predictor, we have considered the eleven short time series provided by the organizers of the Special Session NN3 Neural Network Forecasting Competition, and have proposed a joint application of a variable selection technique and a clustering procedure. The purpose was to define unbiased partition subsets and predictors with high generalization capability, based on a wrapper methodology. The proposed approach overcomes the performance of the predictor that considers all the lags in the regression vector. After obtaining the eleven long-term predictors, we conclude the paper presenting the eighteen multi-step predictions for each time series, as requested in the competition.\",\n \"Title: Evaluating the Performance of a Biclustering Algorithm Applied to Collaborative Filtering - A Comparative Analysis\\nYear: 2007\\n\\nAbstract: Collaborative filtering (CF) is a method to perform automated suggestions for a user based on the opinion of other users with similar interest. Most of the CF algorithms do not take into account the existent duality between users and items, considering only the similarities between users or only the similarities between items. The authors have proposed in a previous work a bio-inspired methodology for CF, namely BIC-aiNet, capable of clustering rows and columns of a data matrix simultaneously. The usefulness and performance of the methodology are reported in the literature. Now, the authors carry out more rigorous comparative experiments with BICaiNet and other techniques found in the literature, as well as evaluate the scalability of the algorithm in several datasets of different sizes. The results indicate that our proposal is able to provide useful recommendations for the users, outperforming other methodologies for CF.\",\n \"Title: A Hybrid Ensemble Model Applied To The Short-Term Load Forecasting Problem\\nYear: 2006\\n\\nAbstract: In this paper we present a methodology based on a combination of many distinct predictors in an ensemble, named hybrid ensemble model, to obtain a more accurate output using the results of single predictors. As basic components, we have used Artificial Neural Networks and Support Vector Machines models. In order to evaluate the performance, the hybrid model was required to predict a 24h daily series energy consumption of a Brazilian electrical operation unit located in the northeast of Brazil. The proposed ensemble model has reached an error 25% smaller than that achieved by the best single predictor. The model was initialized several times to confirm that ensembles of predictors also tend to produce low variance profiles.\",\n \"Title: Application of an artificial immune system in a compositional timbre design technique\\nYear: 2005\\n\\nAbstract: Computer generated sounds for music applications have many facets, of which timbre design is of groundbreaking significance. Timbre is a remarkable and rather complex phenomenon that has puzzled researchers for a long time. Actually, the nature of musical signals is not fully understood yet. In this paper, we present a sound synthesis method using an artificial immune network for data clustering, denoted aiNet. Sounds produced by the method are referred to as immunological sounds. Basically, antibody-sounds are generated to recognize a fixed and predefined set of antigen-sounds, thus producing timbral variants with the desired characteristics. The aiNet algorithm provides maintenance of diversity and an adaptive number of resultant antibody-sounds (memory cells), so that the intended aesthetical result is properly achieved by avoiding the formal definition of the timbral attributes. The initial set of antibody-sounds may be randomly generated vectors, sinusoidal waves with random frequency, or a set of loaded waveforms. To evaluate the obtained results we propose an affinity measure based on the average spectral distance from the memory cells to the antigen-sounds. With the validation of the affinity criterion, the experimental procedure is outlined, and the results are depicted and analyzed.\",\n \"Title: A Michigan-like immune-inspired framework for performing independent component analysis over Galois fields of prime order\\nYear: 2014\\n\\nAbstract: In this work, we present a novel bioinspired framework for performing ICA over finite (Galois) fields of prime order P. The proposal is based on a state-of-the-art immune-inspired algorithm, the cob-aiNet[C], which is employed to solve a combinatorial optimization problem - associated with a minimal entropy configuration - adopting a Michigan-like population structure. The simulation results reveal that the strategy is capable of reaching a performance similar to that of standard methods for lower-dimensional instances with the advantage of also handling scenarios with an elevated number of sources.\",\n \"Title: Identification of a Multivariate Fermentation Process using Constructive Learning\\nYear: 2002\\n\\nAbstract: In the present work, a constructive learning algorithmis employed to design an optimal one-hidden neuralnetwork structure that best approximates a givenmapping. The method determines not only the optimalnumber of hidden neurons but also the best activationfunction for each node. Here, the projection pursuittechnique is applied in association with the optimizationof the solvability condition, giving rise to a moreefficient and accurate computational learning algorithm.As each activation function of a hidden neuron isoptimally defined for every approximation problem,better rates of convergence are achieved. Since thetraining process operates the hidden neuronsindividually, a pertinent activation function employingHermite polynomials can be iteratively developed foreach neuron as a function of the learning set. Theproposed constructive learning algorithm wassuccessfully applied to identify a large-scale multivariateprocess, providing a multivariable model that was ableto describe the complex process dynamics, even in long-rangehorizon predictions.\",\n \"Title: A neural architecture to address Reinforcement Learning problems\\nYear: 2011\\n\\nAbstract: In this paper, the Reinforcement Learning problem is formulated equivalently to a Markov Decision Process. We address the solution of such problem using a novel Adaptive Dynamic Programming algorithm which is based on a Multilayer Perceptron Neural Network composed of a parameterized function approximator called Wire-Fitting. Extending such established model, this work makes use of concepts of eligibility to conceive faster learning algorithms. The advantage of the proposed approach is founded on the capability to handle continuous environments and to learn a better policy while following another. Simulation results involving the automatic control of an inverted pendulum are presented to indicate the effectiveness of the proposed algorithm.\",\n \"Title: Feature Subset Selection by Means of a Bayesian Artificial Immune System\\nYear: 2008\\n\\nAbstract: This paper proposes the application of a novel bio-inspired algorithm as a search engine to the feature subset selection problem. We may interpret our algorithm as an Estimation of Distribution Algorithm that adopts an Artificial Immune System to implement the search process in the space of all features and a Bayesian network to implement the probabilistic model of the promising solutions. The characteristics of the proposed algorithm are the capability of effectively identifying and manipulating building blocks, maintenance of diversity in the population, and automatic control of the population size. These properties allow the algorithm to perform a multimodal search, known to be of great relevance in feature selection problems. Experiments on five datasets were carried out in order to evaluate the proposed methodology in classification problems and its performance compares favorably to that produced by contenders.\",\n \"Title: Query expansion using an immune-inspired biclustering algorithm\\nYear: 2010\\n\\nAbstract: Query expansion is a technique utilized to improve the performance of information retrieval systems by automatically adding related terms to the initial query. These additional terms can be obtained from documents stored in a database. Usually, this task is performed by clustering the documents and then extracting representative terms from the clusters. Afterwards, a new search is performed in the whole database using the expanded set of terms. Recently, the authors have proposed an immune-inspired algorithm, namely BIC-aiNet, to perform biclustering of texts. Biclustering differs from standard clustering algorithms in the sense that the former can detect partial similarities in the attributes. The preliminary results indicated that our proposal is able to group similar texts effectively and the generated biclusters consistently presented relevant words to represent a category of texts. Motivated by this promising scenario, this paper better formalizes the proposal and investigates the usefulness of the whole methodology on larger datasets. The BIC-aiNet was applied to a set of documents aiming at identifying the set of relevant terms associated with each bicluster, giving rise to a query expansion tool. The obtained results were compared with those produced by two alternative proposals in the literature, and they indicate that these techniques tend to generate complementary results, as a consequence of the use of distinct similarity metrics.\",\n \"Title: Interactive Control of Evolution Applied to Sound Synthesis\\nYear: 2005\\n\\nAbstract: In this paper, we present a sound synthesis method that utilizes evolution as generative paradigm. Such sounds will be thereon referred to as evolutionary sounds. Upon defining a population of complex sounds, i.e. sound segments sampled from acoustical instruments and speech; we generated sounds that resulted from evolution applied to those populations. The methodology presented here is an extension to the Evolutionary Sound Synthesis Method (ESSynth) created recently. In ESSynth, a set of waveforms, the Population, is evolved towards another set, the Target, through the application of a Genetic Algorithm (GA). Fitness evaluation is a mathematical distance metric. We enhance features of the previous implementation herein and present the codification. The genetic operators and selection criterion applied are depicted together with the relevant genetic parameters involved in the process. To evaluate the results we present a sound taxonomy based on an objective and a subjective criterion. Those criteria are discussed, the experimental procedure is explained and the results are depicted and evaluated.\",\n \"Title: BAIS: A Bayesian Artificial Immune System for the effective handling of building blocks\\nYear: 2009\\n\\nAbstract: Significant progress has been made in theory and design of Artificial Immune Systems (AISs) for solving hard problems accurately. However, an aspect not yet widely addressed by the research reported in the literature is the lack of ability of the AISs to deal effectively with building blocks (partial high-quality solutions coded in the antibody). The available AISs present mechanisms for evolving the population that do not take into account the relationship among the variables of the problem, potentially causing the disruption of high-quality partial solutions. This paper proposes a novel AIS with abilities to identify and properly manipulate building blocks in optimization problems. Instead of using cloning and mutation to generate new individuals, our algorithm builds a probabilistic model representing the joint probability distribution of the promising solutions and, subsequently, uses this model for sampling new solutions. The probabilistic model used is a Bayesian network due to its capability of properly capturing the most relevant interactions among the variables. Therefore, our algorithm, called Bayesian Artificial Immune System (BAIS), represents a significant attempt to improve the performance of immune-inspired algorithms when dealing with building blocks, and hence to solve efficiently hard optimization problems with complex interactions among the variables. The performance of BAIS compares favorably with that produced by contenders such as state-of-the-art Estimation of Distribution Algorithms.\",\n \"Title: Adaptive radius immune algorithm for data clustering\\nYear: 2005\\n\\nAbstract: Many algorithms perform data clustering by compressing the original data into a more compact and interpretable representation, which can be more easily inspected for the presence of clusters. This, however, can be a risky alternative, because the simplified representation may contain distortions mainly related to the density information present in the data, which can considerably act on the clustering results. In order to treat this deficiency, this paper proposes an Adaptive Radius Immune Algorithm (ARIA), which is capable of maximally preserving the density information after compression by implementing an antibody adaptive suppression radius that varies inversely with the local density in the space. ARIA is tested with both artificial and real world problems obtaining a better performance than the aiNet algorithm and showing that preserving the density information leads to refined clustering results.\",\n \"Title: Max Min Ant System And Capacitated P-Medians: Extensions And Improved Solutions\\nYear: 2005\\n\\nAbstract: This work introduces a modified MAX MIN Ant System (MMAS) designed to solve the Capacitated p-Medians Problem (CPMP). It presents the most relevant steps towards the implementation of an MMAS to solve the CPMP, including some improvements on the original MMAS algorithm, such as the use of a density model in the information heuristics and a local search adapted from the uncapacitated pmedians problem. Extensions of a recently proposed updating rule for the pheromone level, aiming at improving the MMAS ability to deal with large-scale instances, are also presented and discussed. Some simulations are performed using instances available from the literature, and well-known heuristics are employed for benchmarking.\",\n \"Title: Multi-objective Bayesian Artificial Immune System: Empirical Evaluation and Comparative Analyses\\nYear: 2009\\n\\nAbstract: Recently, we have proposed a Multi-Objective Bayesian Artificial Immune System (MOBAIS) to deal effectively with building\\n blocks (high-quality partial solutions coded in the solution vector) in combinatorial multi-objective problems. By replacing\\n the mutation and cloning operators with a probabilistic model, more specifically a Bayesian network representing the joint\\n distribution of promising solutions, MOBAIS takes into account the relationships among the variables of the problem, avoiding\\n the disruption of already obtained high-quality partial solutions. The preliminary results have indicated that our proposal\\n is able to properly build the Pareto front. Motivated by this scenario, this paper better formalizes the proposal and investigates\\n its usefulness on more challenging problems. In addition, an important enhancement regarding the Bayesian network learning\\n was incorporated into the algorithm in order to speed up its execution. To conclude, we compare MOBAIS with state-of-the-art\\n algorithms taking into account quantitative aspects of the Pareto front found by the algorithms. MOBAIS outperforms the contenders\\n in terms of the quality of the obtained solutions and requires an amount of computational resource inferior or compatible\\n with the contenders.\",\n \"Title: An Evolutionary Immune Network for Data Clustering\\nYear: 2000\\n\\nAbstract: This paper explores basic aspects of the immune system and proposes a novel immune network model with the main goals of clustering and filtering unlabeled numerical data sets. It is not our concern to reproduce with confidence any immune phenomenon, but to show that immune concepts can be used to develop powerful computational tools for data processing. As important results of our model, the network evolved will be capable of reducing redundancy, describing data structure, including the shape of the clusters. The network will be implemented in association with a statistical inference technique, and its performance will be illustrated using two benchmark problems. The paper is concluded with a trade-off between the proposed network and artificial neural networks used to perform unsupervised learning.\",\n \"Title: Self-organization and lateral interaction in echo state network reservoirs.\\nYear: 2014\\n\\nAbstract: Echo state networks (ESNs) are recurrent structures that give rise to an interesting trade-off between achievable performance and tractability. This is a consequence of the fact that the key element of these networks – the recurrent intermediate layer known as dynamical reservoir – is not, as a rule, subject to supervised training, which is restricted to the linear output layer, also termed as readout. This trade-off, aside from being of theoretical significance, establishes ESNs as most attractive tools for both online and offline information processing. There are two key aspects to be taken into account in the ESN design: (i) the unsupervised definition of the synaptic weights of the reservoir and (ii) the definition of the structure and of the training strategy associated with the readout. This work is concerned with the first of these aspects: it proposes novel strategies for ESN reservoir design based on the theoretical framework built by Kohonen׳s classical works on self-organization – which includes the notions of short-range positive feedback and lateral inhibition – and also on the related and more recent notion of neural gas. It is shown, with the aid of a representative set of simulation results, that the proposed methodologies are capable of leading to significant performance improvements in the context of relevant information processing tasks – channel equalization and chaotic time series prediction – particularly when the input data suits well a cluster-based profile.\",\n \"Title: A dynamic artificial immune algorithm applied to challenging benchmarking problems\\nYear: 2009\\n\\nAbstract: In many real-world scenarios, in contrast to standard benchmark optimization problems, we may face some uncertainties regarding the objective function. One source of these uncertainties is a constantly changing environment in which the optima change their location over time. New heuristics or adaptations to already available algorithms must be conceived in order to deal with such problems. Among the desirable features that a search strategy should exhibit to deal with dynamic optimization are diversity maintenance, a memory of past solutions, and a multipopulation structure of candidate solutions. In this paper, an immune-inspired algorithm that presents these features, called dopt-aiNet, is properly adapted to deal with six newly proposed benchmark instances, and the obtained results are outlined according to the available specifications for the competition at the Congress on Evolutionary Computation 2009.\",\n \"Title: Learning Ensembles of Neural Networks by Means of a Bayesian Artificial Immune System\\nYear: 2011\\n\\nAbstract: In this paper, we apply an immune-inspired approach to design ensembles of heterogeneous neural networks for classification problems. Our proposal, called Bayesian artificial immune system, is an estimation of distribution algorithm that replaces the traditional mutation and cloning operators with a probabilistic model, more specifically a Bayesian network, representing the joint distribution of promising solutions. Among the additional attributes provided by the Bayesian framework inserted into an immune-inspired search algorithm are the automatic control of the population size along the search and the inherent ability to promote and preserve diversity among the candidate solutions. Both are attributes generally absent from alternative estimation of distribution algorithms, and both were shown to be useful attributes when implementing the generation and selection of components of the ensemble, thus leading to high-performance classifiers. Several aspects of the design are illustrated in practical applications, including a comparative analysis with other attempts to synthesize ensembles.\",\n \"Title: A Heuristic Method Based on Unsupervised Learning and Fuzzy Inference for the Vehicle Routing Problem\\nYear: 2002\\n\\nAbstract: This paper deals with a fuzzy-based system to solve thecapacitated vehicle routing problem. The proposedmethod makes use of a neural network employingunsupervised learning guided by a fuzzy rule base. Thealgorithm implements a policy of penalties and rewards, astrategy of neuron inhibition, insertion and pruning, andalso takes into account certain statistical characteristicsof the input space. Fuzzy theory is considered to minimizedrawbacks related to uncertainty and availability ofpartial information, guiding to an adaptive process ofconstraint relaxation. The effectiveness of the proposedmethod is attested by means of a series of computationalsimulations. As the proposed approach has no adaptationto any particular instance, it represents a good candidateto provide the initial condition for more dedicatedapproaches, like tabu search.\",\n \"Title: Designing ensembles of fuzzy classification systems: an immune-inspired approach\\nYear: 2005\\n\\nAbstract: In this work we propose an immune-based approach for designing of fuzzy systems. From numerical data and with membership function previously defined, the immune algorithm evolves a population of fuzzy classification rules based on the clonal selection, hypermutation and immune network principles. Once AIS are able to find multiple good solutions of the problem, accurate and diverse fuzzy systems are built in a single run. Hence, we construct an ensemble of these classifier in order to achieve better results. An ensemble of classifiers consists of a set of individual classifiers whose outputs are combined when classifying novel patterns. The good performance of an ensemble is strongly dependent of individual accuracy and diversity of its components. We evaluate the proposed methodology through computational experiments on some datasets. The results demonstrate that the performance of the obtained fuzzy systems in isolation is very good. However when we combine these systems, a significant improvement is obtained in the correct classification rate, outperforming the single best classifier.\",\n \"Title: Evaluating the performance of a Bayesian Artificial Immune System for designing fuzzy rule bases\\nYear: 2013\\n\\nAbstract: In this paper we perform a deep investigation about the usefulness of an immune-inspired algorithm to design accurate and compact fuzzy rule bases for classification problems. The algorithm, called Bayesian Artificial Immune System BAIS, incorporates a mechanism to learn a probability graphical model from the promising solutions found so far. Thus, BAIS utilizes this model to sample new candidate solutions. The probabilistic model utilized here is a Bayesian network due to its capability of expressing the relationships among the variables of the problem, avoiding the disruption of already obtained high-quality partial solutions building blocks. Besides the capability to identify and manipulate building blocks, the algorithm maintains diversity in the population, performs multimodal optimization and adjusts the size of the population automatically according to the problem. These attributes are generally absent from alternative algorithms, and can be considered useful attributes when generating fuzzy rule bases, thus guiding to high-performance classifiers. BAIS was evaluated in thirteen well-known classification problems and its performance compares favorably with that produced by contenders.\",\n \"Title: Ordering Gene Expression Data Using One-dimensional Self-Organizing Maps\\nYear: 2002\\n\\nAbstract: The microarray technology allows researchers to simultaneously measure gene expression levels of thousands of genes. Analysis of data pro- duced by such experiments provides knowledge about the gene function. An important step in the a nalysis of gene e xpression d ata is the detection o f genes with similar expression patterns. Real-time computational tools for or- ganization and visualization are crucial to understand and analyze the data. In this work, we make use of an algorithm based on self-organizing neural net- works for organizing g ene e xpression d ata in o rder to reveal t rends in g ene expression profiles under the biological viewpoint.\",\n \"Title: Modular And Hierarchical Evolutionary Design Of Fuzzy Systems\\nYear: 1999\\n\\nAbstract: The paper introduces a modular, hierarchical evolutionary method to design fuzzy systems. The method uses genetic algorithms to evolve a population of rule-based fuzzy models. For this purpose, three hierarchical modules are defined namely, partition, population of granules, and population of fuzzy graphs. The evolutionary process is designed to find the best set of fuzzy rules induced by the granularity required, the form of the membership functions and the inference procedure. This hierarchical configuration guides to the implementation of an effective process of cooperation and competition among rules, responsible for the small cardinality of the re-suiting set of rules. Simulation results show that the method does increase flexibility to design fuzzy models, preserves the computational tractability, and improves the descriptive nature of the final solution.\",\n \"Title: On The Diversity Mechanisms Of Opt-Ainet: A Comparative Study With Fitness Sharing\\nYear: 2010\\n\\nAbstract: Immune-inspired algorithms based on the Immune Network theory have been frequently claimed to be capable of maintaining diversity among the candidate solutions in their population. However, no specific study on this aspect to verify how the intrinsic diversity mechanisms of such immune algorithms behave, when compared to other approaches from the literature, has been made yet. Therefore, in this work we have addressed this issue, by taking the opt-aiNet algorithm (a popular immune-inspired algorithm developed for real-parameter optimization) and comparing its results with those of a modified version, in which the mechanisms associated with diversity maintenance were replaced by a traditional fitness sharing approach. Besides, two distance metrics were also considered for both algorithms: the traditional Euclidean distance, and the Line Distance, a metric proposed in the literature as capable of identifying whether two solutions belong to distinct local optima. The experiments were performed on six benchmark problems from the literature, each of them with distinct characteristics, and the results have shown that the original immune-inspired mechanisms of opt-aiNet are indeed more capable of stimulating the diversity of solutions, and also requiring a smaller amount of computational resources.\",\n \"Title: New Perspectives For The Biclustering Problem\\nYear: 2006\\n\\nAbstract: Multimodal optimization algorithms inspired by the immune system are generally characterized by a dynamic control of the population size and by diversity maintenance along the search. One of these proposals, denoted copt-aiNet (artificial immune network for combinatorial optimization), is used to deal with combinatorial problems like the Traveling Salesman Problem (TSP) and other permutation problems. In this paper, the copt-aiNet algorithm is extended and adapted to be applied to an important issue of modern data mining, the biclustering problem. The biclustering approach consists in simultaneously ordering the rows and columns of a given matrix, so that similar elements are grouped together. To illustrate the performance of the proposed method, two bitmap images are scrambled and used as input to the algorithm, and the biclustering procedure tries to restore the original image by grouping the pixels according to the similarity of colors in a neighborhood. Additionally, copt-aiNet is applied to gene expression data clustering, a classical problem of the bioinformatics literature, and its performance is compared with a hierarchical biclustering algorithm.\",\n \"Title: Self-organizing Bio-inspired Sound Transformation\\nYear: 2007\\n\\nAbstract: We present a time domain approach to explore a sound transformation paradigm for musical performance. Given a set of sounds containing a priori desired qualities and a population of agents interacting locally, the method generates both musical form and matter resulting from sonic trajectories. This proposal involves the use of bio-inspired algorithms, which possess intrinsic features of adaptive, self-organizing systems, as definers of generating and structuring processes of sound elements. Self-organization makes viable the temporal emergence of stable structures without an external organizing element. Regarding musical performance as a creative process that can be described using trajectories through the compositional space, and having the simultaneous emergence of musical matter and form resulting from the process itself as the final objective, the conception of a generative paradigm in computer music that does not contemplate a priori external organizing elements is the main focus of this proposal.\",\n \"Title: A Wrapper For Projection Pursuit Learning\\nYear: 2007\\n\\nAbstract: Constructive algorithms have shown to be reliable and effective methods for designing Artificial Neural Networks (ANN) with good accuracy and generalization capability, yet with parsimonious network structures. Projection Pursuit Learning (PPL) has demonstrated great flexibility and effectiveness in performing this task, though presenting some difficulties in the search for appropriate projection directions in input spaces with high dimensionality. Due to the existence of high-dimensional input spaces in the context of time series prediction, mainly under the existence of long-term dependencies in the time series, we propose here a method based on the wrapper methodology to perform variable selection, so that only a subset of highly-informative lags is going to be considered as the regression vector. The Yearly Sunspot Number time series is adopted as a case study and comparative analysis is performed considering alternative approaches in the literature, guiding to competitive results.\",\n \"Title: An extended echo state network using Volterra filtering and principal component analysis.\\nYear: 2012\\n\\nAbstract: Echo state networks (ESNs) can be interpreted as promoting an encouraging compromise between two seemingly conflicting objectives: (i) simplicity of the resulting mathematical model and (ii) capability to express a wide range of nonlinear dynamics. By imposing fixed weights to the recurrent connections, the echo state approach avoids the well-known difficulties faced by recurrent neural network training strategies, but still preserves, to a certain extent, the potential of the underlying structure due to the existence of feedback loops within the dynamical reservoir. Moreover, the overall training process is relatively simple, as it amounts essentially to adapting the readout, which usually corresponds to a linear combiner. However, the linear nature of the output layer may limit the capability of exploring the available information, since higher-order statistics of the signals are not taken into account. In this work, we present a novel architecture for an ESN in which the linear combiner is replaced by a Volterra filter structure. Additionally, the principal component analysis technique is used to reduce the number of effective signals transmitted to the output layer. This idea not only improves the processing capability of the network, but also preserves the simplicity of the training process. The proposed architecture is then analyzed in the context of a set of representative information extraction problems, more specifically supervised and unsupervised channel equalization, and blind separation of convolutive mixtures. The obtained results, when compared to those produced by already proposed ESN versions, highlight the benefits brought by the novel network proposal and characterize it as a promising tool to deal with challenging signal processing tasks.\",\n \"Title: Multi-Objective Biclustering: When Non-dominated Solutions are not Enough\\nYear: 2009\\n\\nAbstract: The biclustering technique was developed to avoid some of the drawbacks presented by standard clustering techniques, such\\n as their impossibility of finding correlating data under a subset of features, and, consequently, to allow the extraction\\n of more accurate information from datasets. Given that biclustering requires the optimization of at least two conflicting\\n objectives (residue and volume) and that multiple independent solutions are desirable as the outcome, a few multi-objective\\n evolutionary algorithms for biclustering were proposed in the literature. However, these algorithms only focus their search\\n in the generation of a global set of non-dominated biclusters, which may be insufficient for most of the problems as the coverage\\n of the dataset can be compromised. In order to overcome such problem, a multi-objective artificial immune system capable of\\n performing a multipopulation search, named MOM-aiNet, was proposed. In this work, the MOM-aiNet algorithm will be described\\n in detail, and an extensive set of experimental comparisons will be performed, with the obtained results of MOM-aiNet being\\n confronted with those produced by the popular CC algorithm, by another immune-inspired approach for biclustering (BIC-aiNet),\\n and by the multi-objective approach for biclustering proposed by Mitra & Banka.\",\n \"Title: A Classifier System In Real Applications For Robot Navigation\\nYear: 2003\\n\\nAbstract: This work presents an autonomous evolutionary system applied to control a mobile robot in unknown environments. The navigation system learns efficiently to deal with situations where the robot must capture targets avoiding collisions with obstacles. Toward this end, robot direction and speed must be properly defined. The evolutionary approach is based on a version of classifier systems, responsible for the proposition of a competitive process involving rules of elementary behaviour. A virtual environment is used to evolve the controller. With the evolved controller, a Khepera II robot is submitted to real navigation tasks, with no significant degradation in performance. As an additional experiment, the controller is also evolved in a real environment, and validated in a different and more complex environment, not previously experimented, attesting the generalisation capability of the proposal.\",\n \"Title: Multivariate ant colony optimization in continuous search spaces\\nYear: 2008\\n\\nAbstract: This work introduces an ant-inspired algorithm for optimization in continuous search spaces that is based on the generation of random vectors with multivariate Gaussian pdf. The proposed approach is called MACACO -- Multivariate Ant Colony Algorithm for Continuous Optimization -- and is able to simultaneously adapt all the dimensions of the random distribution employed to generate the new individuals at each iteration. In order to analyze MACACO's search efficiency, the approach was compared to a pair of counterparts: the Continuous Ant Colony System (CACS) and the approach known as Ant Colony Optimization in en (ACOR). The comparative analysis, which involves well-known benchmark problems from the literature, has indicated that MACACO outperforms CACS and ACOR in most cases as the quality of the final solution is concerned, and it is just about two times more costly than the least expensive contender.\",\n \"Title: Topology of social networks and efficiency of collective intelligence methods\\nYear: 2013\\n\\nAbstract: In this work we analyzed the role social networks play in the efficiency of collective problem-solving, evaluating whether the topological characteristics seen in real-world networks yield any performance improvement in such processes. To study this we used the Particle Swarm Optimization as a testbed for social groups performing a collective task, defining the structure of communication between individuals in the swarm through topologies generated by a model for the creation and evolution of social networks. The experimental results indicate that groups using these networks may, indeed, experience better performance in collective problem-solving, so that these groups were able to overcome the results achieved by swarms using classical neighborhoods for PSO and reached results very close to those found by swarms using the topology of DMS-PSO, usually considered to be part of the state-of-the-art of Particle Swarm Optimization.\",\n \"Title: The construction of a Boolean competitive neural network using ideas from immunology\\nYear: 2003\\n\\nAbstract: The immune system is capable of recognizing and responding to microorganisms and molecules that cannot be perceived by our sensory mechanisms, which send stimuli straight into the brain. It performs an accessory role for nervous cognition. This paper main goals are: (1) to show how some immune principles and theories can be used as sources of inspiration to develop novel neural network learning algorithms; (2) to survey the main works from the literature that employ the immune metaphor for the development of neural network architectures; and (3) to illustrate, with a new network model, how this source of inspiration can be actually used to develop a neural network learning algorithm. The novel learning algorithm proposed has the main features of competitive learning, automatic generation of the network structure and binary representation of the connection strengths (weights). The behavior of the algorithm is primarily described using a benchmark task, and some of its potential applications are illustrated using two simple real-world problems and a binary character recognition task. The results show that the network is a promising tool for solving problems that are inherently binary, and also that the immune system provides a new paradigm to search for neural network learning algorithms.\",\n \"Title: Online Social Network Analysis: A Survey of Research Applications in Computer Science.\\nYear: 2015\\n\\nAbstract: The emergence and popularization of online social networks suddenly made available a large amount of data from social organization, interaction and human behavior. All this information opens new perspectives and challenges to the study of social systems, being of interest to many fields. Although most online social networks are recent (less than fifteen years old), a vast amount of scientific papers was already published on this topic, dealing with a broad range of analytical methods and applications. This work describes how computational researches have approached this subject and the methods used to analyze such systems. Founded on a wide though non-exaustive review of the literature, a taxonomy is proposed to classify and describe different categories of research. Each research category is described and the main works, discoveries and perspectives are highlighted.\",\n \"Title: Asymmetric Distributions from Constrained Mixtures.\\nYear: 2015\\n\\nAbstract: This paper introduces constrained mixtures for continuous distributions, characterized by a mixture of distributions where each distribution has a shape similar to the base distribution and disjoint domains. This new concept is used to create generalized asymmetric versions of the Laplace and normal distributions, which are shown to define exponential families, with known conjugate priors, and to have maximum likelihood estimates for the original parameters, with known closed-form expressions. The asymmetric and symmetric normal distributions are compared in a linear regression example, showing that the asymmetric version performs at least as well as the symmetric one, and in a real world time-series problem, where a hidden Markov model is used to fit a stock index, indicating that the asymmetric version provides higher likelihood and may learn distribution models over states and transition distributions with considerably less entropy.\",\n \"Title: Multi-task Sparse Structure Learning\\nYear: 2014\\n\\nAbstract: Multi-task learning (MTL) aims to improve generalization performance by learning multiple related tasks simultaneously. While sometimes the underlying task relationship structure is known, often the structure needs to be estimated from data at hand. In this paper, we present a novel family of models for MTL, applicable to regression and classification problems, capable of learning the structure of task relationships. In particular, we consider a joint estimation problem of the task relationship structure and the individual task parameters, which is solved using alternating minimization. The task relationship structure learning component builds on recent advances in structure learning of Gaussian graphical models based on sparse estimators of the precision (inverse covariance) matrix. We illustrate the effectiveness of the proposed model on a variety of synthetic and benchmark datasets for regression and classification. We also consider the problem of combining climate model outputs for better projections of future climate, with focus on temperature in South America, and show that the proposed model outperforms several existing methods for the problem.\",\n \"Title: Hybrid genetic training of gated mixtures of experts for nonlinear time series forecasting\\nYear: 2003\\n\\nAbstract: In this paper, we introduce a genetic algorithm-based training mechanism (HGT-GAME) toward the automatic structural design and parameter configuration of gated mixtures of experts (ME). In HGT-GAME, a whole ME instance is codified into a given chromosome. By employing regulatory genes, our approach enables the automatic pruning and growing of experts in a way to properly match the complexity of the task at hand. Moreover, to leverage HGT-GAME's effectiveness a local search refinement upon each ME chromosome is performed in each generation via the gradient descent-learning algorithm. Forecasting experiments evaluate the performance of gated MEs trained with HGT-GAME.\",\n \"Title: Implementation of an immuno-genetic network on a real Khepera II robot\\nYear: 2003\\n\\nAbstract: The design of autonomous navigation systems for mobile robots, with simultaneous objectives to be satisfied such as garbage collection with integrity maintenance, requires refined coordination mechanisms to deal with modules of elementary behaviour. This paper shows the implementation on a real Khepera II robot of an immuno-genetic network for autonomous navigation that combines an evolutionary algorithm with a continuous immune network model. The proposed immuno-genetic system has the immune network implementing a dynamic process of decision-making, and the evolutionary algorithm defining the network structure. To be able to evaluate the controllers (immune networks) on the evolutionary process, a virtual environment was used for computer simulation, based on the characteristics of the navigation problem. The immune networks obtained by evolution were then analyzed and tested on new situations, presenting coordination capability in simple and more complex tasks. Some preliminary experiments on a real Khepera II robot demonstrate the feasibility of the evolved immune networks.\",\n \"Title: Immune Learning Classifier Networks: Evolving Nodes and Connections\\nYear: 2006\\n\\nAbstract: The design of an autonomous navigation system with multiple tasks to be accomplished in unknown environments represents a complex undertaking. With the simultaneous purposes of capturing targets and avoiding obstacles, the challenge may become still more intricate if the configuration of obstacles and targets creates local minima, like concave shapes and mazes between the robot and the target. Pure reactive navigation systems are not able to deal properly with such hampering scenarios, requiring additional cognitive apparatus. Concepts from immune network theory are then employed to convert an earlier reactive robot controller, based on learning classifier systems, into a connectionist device. Starting from no a priori knowledge, both the classifiers and their connections are evolved during the robot navigation. Some experiments with and without local minima are carried out and the proposed evolutionary network of classifiers was shown to produce connectionist navigation systems capable of successfully overcoming local minima.\",\n \"Title: An Immunological Density-Preserving Approach to the Synthesis of RBF Neural Networks for classification\\nYear: 2006\\n\\nAbstract: Radial basis function (RBF) neural networks are universal approximators and have been used for a wide range of applications. Aiming at reducing the number of neurons in the hidden layer, for regularization purposes, the center and dispersion of each RBF have to be properly defined by means of competitive learning. Only the output weights will be defined in a supervised manner. One of the drawbacks of such learning methodology, involving unsupervised and supervised learning, is that the centers will be defined so that regions in the input space with a high density of samples tend to be under-represented and those regions with a low density of samples tend to be over-represented. Additionally, few approaches provide a proper and individual indication of the dispersion of each RBF. This paper presents an immune density-preserving algorithm with adaptive radius, called ARIA, to determine the number of centers, their location and the dispersion of each RBF, based only on the available training data set. Considering classification problems, the algorithm to determine the hidden layer is compared to another immune-inspired algorithm called aiNet, K-means and the random choice of centers. The classification accuracy of the final network is compared to another density based approach and a decision tree classifier, C 5.0. The results are reported and analyzed.\",\n \"Title: Bayesian Learning of Neural Networks by Means of Artificial Immune Systems\\nYear: 2006\\n\\nAbstract: Once the design of artificial neural networks (ANN) may require the optimization of numerical and structural parameters, bio-inspired algorithms have been successfully applied to accomplish this task, since they are population-based search strategies capable of dealing successfully with complex and large search spaces, avoiding local minima. In this paper, we propose the use of an artificial immune system for learning feedforward ANN's topologies. Besides the number of neurons in the hidden layer, the algorithm also optimizes the type of activation function for each node. The use of a Bayesian framework to infer the weights and weight decay terms as well as to perform model selection allows us to find neural models with high generalization capability and low complexity, once the Occam's razor principle is incorporated into the framework. We demonstrate the applicability of the proposal on seven classification problems and promising results were obtained.\",\n \"Title: Designing fuzzy rule bases with a Bayesian Artificial Immune System\\nYear: 2011\\n\\nAbstract: In this paper we apply an immune-inspired approach to generate fuzzy rule bases for classification problems. Our proposal, called Bayesian Artificial Immune System (BAIS), is a hybrid algorithm that replaces the traditional mutation and cloning operators with a probabilistic model, more specifically a Bayesian network, representing the joint distribution of promising solutions. Thus, the algorithm takes into account the relationships among the variables of the problem, avoiding the disruption of already obtained high-quality partial solutions (building blocks). Besides the capability to identify and manipulate building blocks, the algorithm maintains diversity in the population, performs multimodal optimization and adjusts the size of the population automatically according to the problem. These attributes are generally absent from alternative algorithms, and can be considered useful attributes when generating fuzzy rule bases, thus guiding to high-performance classifiers. BAIS was evaluated in six well-known classification problems and its performance compares favorably with that produced by contenders.\",\n \"Title: Learning Bayesian networks to perform feature selection\\nYear: 2009\\n\\nAbstract: Bayesian networks have been widely applied to the feature selection problem. The existing approaches learn a Bayesian network from the available dataset and, afterward, utilize the Markov Blanket of the target feature as the criterion to select the relevant features. The Bayesian network learning can be viewed as a search and optimization procedure, where a search mechanism explores the space of all network structures while a scoring metric evaluates each candidate solution based on the likelihood. This paper investigates the application of an immune-inspired algorithm as the search procedure for obtaining high-quality Bayesian networks, motivated by the dynamical control of the population size and diversity along the search. Due to the resulting multimodal search capability, in a single run of the algorithm several subsets of features are obtained. Experiments on ten datasets were carried out in order to evaluate the proposed methodology in classification problems, and reduced-size subsets of features were produced.\",\n \"Title: Regularized hypervolume selection for robust portfolio optimization in dynamic environments\\nYear: 2013\\n\\nAbstract: This paper proposes a regularized hypervolume (SMetric) selection algorithm. The proposal is used for incorporating stability and diversification in financial portfolios obtained by solving a temporal sequence of multi-objective Mean Variance Problems (MVP) on real-world stock data, for short to longterm rebalancing periods. We also propose the usage of robust statistics for estimating the parameters of the assets returns distribution so that we are able to test two variants (with and without regularization) on dynamic environments under different levels of instability. The results suggest that the maximum attaining Sharpe Ratio portfolios obtained for the original MVP without regularization are unstable, yielding high turnover rates, whereas solving the robust MVP with regularization mitigated turnover, providing more stable solutions for unseen, dynamic environments. Finally, we report an apparent conflict between stability in the objective space and in the decision space.\",\n \"Title: Anticipatory Stochastic Multi-Objective Optimization for uncertainty handling in portfolio selection\\nYear: 2013\\n\\nAbstract: An anticipatory stochastic multi-objective model based on S-Metric maximization is proposed. The environment is assumed to be noisy and time-varying. This raises the question of how to incorporate anticipation in metaheuristics such that the Pareto optimal solutions can reflect the uncertainty about the subsequent environments. A principled anticipatory learning method for tracking the dynamics of the objective vectors is then proposed so that the estimated S-Metric contributions of each solution can integrate the underlying stochastic uncertainty. The proposal is assessed for minimum holding, cardinality constrained portfolio selection, using real-world stock data. Preliminary results suggest that, by taking into account the underlying uncertainty in the predictive knowledge provided by a Kalman filter, we were able to reduce the sum of squared errors prediction of the portfolios ex-post return and risk estimation in out-of-sample investment environments.\",\n \"Title: A Gaussian Artificial Immune System for Multi-Objective optimization in continuous domains\\nYear: 2010\\n\\nAbstract: This paper proposes a Multi-Objective Gaussian Artificial Immune System (MOGAIS) to deal effectively with building blocks (high-quality partial solutions coded in the solution vector) in multi-objective continuous optimization problems. By replacing the mutation and cloning operators with a probabilistic model, more specifically a Gaussian network representing the joint distribution of promising solutions, MOGAIS takes into account the relationships among the variables of the problem, avoiding the disruption of already obtained high-quality partial solutions. The algorithm was applied to three benchmarks and the results were compared with those produced by state-of-the-art algorithms.\",\n \"Title: Heuristics to avoid redundant solutions on population-based multimodal continuous optimization\\nYear: 2011\\n\\nAbstract: In population-based meta-heuristics, the generation and maintenance of diversity seem to be crucial to deal with multimodal continuous optimization. However, usually this crucial aspect is not an inherent feature of generally adopted meta-heuristics. In this paper, we propose to associate diversity maintenance with the detection and elimination of redundant candidate solutions in the search space, more specifically candidate solutions located at the same attraction basin of a local optimum. Two low computational cost heuristics are proposed to detect redundancy, in a pairwise comparison of candidate solutions and by extracting local features of the fitness landscape at runtime. Those heuristics are not tied to a specific class of algorithms, and are thus able to be incorporated into a broad range of population-based meta-heuristics, and even into multiple executions of non-population-based algorithms. In a set of experimental results, the two heuristics were implemented as an attached module of an already existing multipopulation meta heuristics, and the results indicate that they operate properly, no matter the number and conformation of the attraction basins in multimodal optimization problems.\",\n \"Title: Evolutionary multi-objective optimization for the vendor-managed inventory routing problem\\nYear: 2011\\n\\nAbstract: The class of inventory routing problems (IRPs) is present in several areas, including automotive industry and cash management for ATM networks. In the specific case of vendor-managed IRPs, in which the supplier is responsible for managing the product inventory in each client and for properly providing replenishments, the challenge is to determine which retailers should be served, the amount of product that should be delivered to each of these retailers, and which routes the distribution vehicles should follow, so that the associated costs are minimized. Although this is clearly a multi-objective optimization problem, in the literature it has been generally modeled as a single-objective problem, which limits the scope of the obtained results. Therefore, this work presents a multi-objective approach to solve one version of the IRP usually found in the scientific literature, by simultaneously minimizing both the inventory and transportation costs. The method proposed in this work is based on the well-known SPEA2 (Strength Pareto Evolutionary Algorithm) and includes innovative aspects mainly associated with the representation of candidate solutions, genetic operators and local search. The experiments were performed on a set of known benchmark IRPs from the literature, so that the obtained results could be properly compared to the best solution found for the single-objective version of each problem.\",\n \"Title: Training multilayer perceptrons with a Gaussian Artificial Immune System\\nYear: 2011\\n\\nAbstract: In this paper we apply an immune-inspired approach to train Multilayer Perceptrons (MLPs) for classification problems. Our proposal, called Gaussian Artificial Immune System (GAIS), is an estimation of distribution algorithm that replaces the traditional mutation and cloning operators with a probabilistic model, more specifically a Gaussian network, representing the joint distribution of promising solutions. Sub sequently, GAIS utilizes this probabilistic model for sampling new solutions. Thus, the algorithm takes into account the relationships among the variables of the problem, avoiding the disruption of already obtained high-quality partial solutions (building blocks). Besides the capability to identify and manipulate building blocks, the algorithm maintains diversity in the population, performs multimodal optimization and adjusts the size of the population automatically according to the problem. These attributes are generally absent from alternative algorithms, and all were shown to be useful attributes when optimizing the weights of MLPs, thus guiding to high-performance classifiers. GAIS was evaluated in six well-known classification problems and its performance compares favorably with that produced by contenders, such as opt-aiNet, IDEA and PSO.\",\n \"Title: Multi-label structure learning with ising model selection\\nYear: 2015\\n\\nAbstract: A common way of attacking multi-label classification problems is by splitting it into a set of binary classification problems, then solving each problem independently using traditional single-label methods. Nevertheless, by learning classifiers separately the information about the relationship between labels tends to be neglected. Built on recent advances in structure learning in Ising Markov Random Fields (I-MRF), we propose a multilabel classification algorithm that explicitly estimate and incorporate label dependence into the classifiers learning process by means of a sparse convex multitask learning formulation. Extensive experiments considering several existing multi-label algorithms indicate that the proposed method, while conceptually simple, outperforms the contenders in several datasets and performance metrics. Besides that, the conditional dependence graph encoded in the I-MRF provides a useful information that can be used in a posterior investigation regarding the reasons behind the relationship between labels.\",\n \"Title: A Multitask Learning View on the Earth System Model Ensemble.\\nYear: 2015\\n\\nAbstract: Earth system models (ESMs) are based on physical principles that are intended to emulate climate behavior. They're the primary mechanisms for obtaining projections of future conditions under different climate change scenarios. Because ESMs rely on the distinct modeling of certain physical processes and initial conditions, different ESMs can produce different responses for the same external forcing...\",\n \"Title: On bicluster aggregation and its benefits for enumerative solutions\\nYear: 2015\\n\\nAbstract: Biclustering involves the simultaneous clustering of objects and their attributes, thus defining local two-way clustering models. Recently, efficient algorithms were conceived to enumerate all biclusters in real-valued datasets. In this case, the solution composes a complete set of maximal and non-redundant biclusters. However, the ability to enumerate biclusters revealed a challenging scenario: in noisy datasets, each true bicluster may become highly fragmented and with a high degree of overlapping. It prevents a direct analysis of the obtained results. Aiming at reverting the fragmentation, we propose here two approaches for properly aggregating the whole set of enumerated biclusters: one based on single linkage and the other directly exploring the rate of overlapping. Both proposals were compared with each other and with the actual state-of-the-art in several experiments, and they not only significantly reduced the number of biclusters but also consistently increased the quality of the solution.\",\n \"Title: Effects of Social Ties in Knowledge Diffusion: case study on PLOS ONE.\\nYear: 2016\\n\\nAbstract: In order to capture the effects of social ties in knowledge diffusion, this paper examines the publication network that emerges from the collaboration of researchers, using citation information as means to estimate knowledge flow. For this purpose, we analyzed the papers published in the PLOS ONE journal finding strong evidence to support that the closer two authors are in the co-authorship network, the larger the probability that knowledge flow will occur between them. Moreover, we also found that when it comes to knowledge diffusion, strong co-authorship proximity is more determinant than geographic proximity.\",\n \"Title: Spatial Projection of Multiple Climate Variables using Hierarchical Multitask Learning.\\nYear: 2017\\n\\nAbstract: Future projection of climate is typically obtained by combining outputs from multiple Earth System Models (ESMs) for several climate variables such as temperature and precipitation. While IPCC has traditionally used a simple model output average, recent work has illustrated potential advantages of using a multitask learning (MTL) framework for projections of individual climate variables. In this paper we introduce a framework for hierarchical multitask learning (HMTL) with two levels of tasks such that each super-task, i.e., task at the top level, is itself a multitask learning problem over sub-tasks. For climate projections, each super-task focuses on projections of specific climate variables spatially using an MTL formulation. For the proposed HMTL approach, a group lasso regularization is added to couple parameters across the super-tasks, which in the climate context helps exploit relationships among the behavior of different climate variables at a given spatial location. We show that some recent works on MTL based on learning task dependency structures can be viewed as special cases of HMTL. Experiments on synthetic and real climate data show that HMTL produces better results than decoupled MTL methods applied separately on the super-tasks and HMTL significantly outperforms baselines for climate projection.\",\n \"Title: An Extremal Optimization approach to parallel resonance constrained capacitor placement problem.\\nYear: 2017\\n\\nAbstract: Installation of capacitors in distribution networks is one of the most used procedure to compensate reactive power generated by loads and, consequently, to reduce technical losses. So, the problem consists in identifying the optimal placement and sizing of capacitors. This problem is known in the literature as optimal capacitor placement problem. Neverthless, depending on the location and size of the capacitor, it may become a harmonic source, allowing capacitor to enter into resonance with the distribution network, causing several undesired side effects. In this work we propose a parsimonious method to deal with the capacitor placement problem that incorporates resonance constraints, ensuring that every allocated capacitor will not act as a harmonic source. This proposed algorithm is based upon a physical inspired metaheuristic known as Extremal Optimization. The results achieved showed that this proposal has reached significant gains when compared with other proposals that attempt repair, in a post-optimization stage, already obtained solutions which violate resonance constraints.\",\n \"Title: Using Retweet Information as a Feature to Classify Messages Contents.\\nYear: 2017\\n\\nAbstract: We investigate the use of machine learning algorithms to classify content published in Online Social Networks using as input solely user interaction data, instead of the actual message content. During a period of six months, we monitored and gathered data from users interacting with news messages on Twitter, creating thousands of information diffusion processes. The data set presented regular patterns on how messages were spread over the network by users, depending on its content, so we could build classifiers to predict the topic of a message using as input only information of how it was diffused. Thus, we demonstrate the explanatory power of user behavior data on identifying content present in Social Networks, proposing techniques for topic classification that can be used to assist traditional content identification strategies (such as natural language or image processing) in challenging contexts, or be applied in scenarios with limited information access.\",\n \"Title: Efficient mining of maximal biclusters in mixed-attribute datasets.\\nYear: 2017\\n\\nAbstract: This paper presents a novel enumerative biclustering algorithm to directly mine all maximal biclusters in mixed-attribute datasets (containing both numerical and categorical attributes), with or without missing values. The proposal is an extension of RIn-Close_CVC, which was originally conceived to mine perfect or perturbed biclusters with constant values on columns solely from numerical datasets, and without missing values. Even endowed with additional and more general features, the extended RIn-Close_CVC retains four key properties: (1) efficiency, (2) completeness, (3) correctness, and (4) non-redundancy. Our proposal is the first one to deal with mixed-attribute datasets without requiring any pre-processing step, such as discretization and itemization of real-valued attributes. This is a decisive aspect, because discretization and itemization implies a priori decisions, with information loss and no clear control over the consequences. On the other hand, even having to specify a priori an individual threshold for each numerical attribute, that will be used to indicate internal consistency per attribute, each threshold will be applied during the construction of the biclusters, shaping the peculiarities of the data distribution. We also explore the strong connection between biclustering and frequent pattern mining to (1) provide filters to select a compact bicluster set that exhibits high relevance and low redundancy, and (2) in the case of labeled datasets, automatically present the biclusters in a user-friendly and intuitive form, by means of quantitative class association rules. Our experimental results showed that the biclusters yield a parsimonious set of relevant rules, providing useful and interpretable models for five mixed-attribute labeled datasets.\",\n \"Title: Many-Objective Ensemble-Based Multilabel Classification.\\nYear: 2017\\n\\nAbstract: This paper proposes a many-objective ensemble-based algorithm to explore the relations among the labels on multilabel classification problems. This proposal consists in two phases. In the first one, a many-objective optimization method generates a set of candidate components exploring the relations among the labels, and the second one uses a stacking method to aggregate the components for each label. By balancing or not the relevance of each label, two versions were conceived for the proposal. The balanced one presented a good performance for recall and F1 metrics, and the unbalanced one for 1-Hamming loss and precision metrics.\",\n \"Title: Ensembles of Multiobjective-Based Classifiers for Detection of Epileptic Seizures.\\nYear: 2017\\n\\nAbstract: This paper proposes multiobjective-based classifiers to detect epileptic seizures using ensemble approaches, transfer-learning methods, and three alternative feature extraction techniques. Two aspects of the problem were investigated: (1) the relative merit of distinct proposals to synthesize an ensemble of classifiers, considering all the three feature extraction techniques; (2) the potential of an ensemble composed of transfer-learned classifiers. The blend approaches with the best performance detected all test seizures, with a high proportion of correctly detected samples inside the seizure interval and high proportion of time intervals correctly classified as non-seizures.\",\n \"Title: RIn-Close_CVC2: an even more efficient enumerative algorithm for biclustering of numerical datasets.\\nYear: 2018\\n\\nAbstract: RIn-Close_CVC is an efficient (take polynomial time per bicluster), complete (find all maximal biclusters), correct (all biclusters attend the user-defined level of consistency) and non-redundant (all the obtained biclusters are maximal and the same bicluster is not enumerated more than once) enumerative algorithm for mining maximal biclusters with constant values on columns in numerical datasets. Despite RIn-Close_CVC has all these outstanding properties, it has a high computational cost in terms of memory usage because it must keep a symbol table in memory to prevent a maximal bicluster to be found more than once. In this paper, we propose a new version of RIn-Close_CVC, named RIn-Close_CVC2, that does not use a symbol table to prevent redundant biclusters, and keeps all these four properties. We also prove that these algorithms actually possess these properties. Experiments are carried out with synthetic and real-world datasets to compare RIn-Close_CVC and RIn-Close_CVC2 in terms of memory usage and runtime. The experimental results show that RIn-Close_CVC2 brings a large reduction in memory usage and, in average, significant runtime gain when compared to its predecessor.\",\n \"Title: Multi-task Sparse Structure Learning with Gaussian Copula Models.\\nYear: 2016\\n\\nAbstract: Multi-task learning (MTL) aims to improve generalization performance by learning multiple related tasks simultaneously. While sometimes the underlying task relationship structure is known, often the structure needs to be estimated from data at hand. In this paper, we present a novel family of models for MTL, applicable to regression and classification problems, capable of learning the structure of tasks relationship. In particular, we consider a joint estimation problem of the tasks relationship structure and the individual task parameters, which is solved using alternating minimization. The task relationship revealed by structure learning is founded on recent advances in Gaussian graphical models endowed with sparse estimators of the precision (inverse covariance) matrix. An extension to include flexible Gaussian copula models that relaxes the Gaussian marginal assumption is also proposed. We illustrate the e ff ectiveness of the proposed model on a variety of synthetic and benchmark data sets for regression and classi fi cation. We also consider the problem of combining Earth System Model (ESM) outputs for better projections of future climate, with focus on projections of temperature by combining ESMs in South and North America, and show that the proposed model outperforms several existing methods for the problem.\",\n \"Title: Simulated annealing for symbolic regression\\nYear: 2021\\n\\nAbstract: ABSTRACTSymbolic regression aims to hypothesize a functional relationship involving explanatory variables and one or more dependent variables, based on examples of the desired input-output behavior. Genetic programming is a meta-heuristic commonly used in the literature to achieve this goal. Even though Symbolic Regression is sometimes associated with the potential of generating interpretable expressions, there is no guarantee that the returned function will not contain complicated constructs or even bloat. The Interaction-Transformation (IT) representation was recently proposed to alleviate this issue by constraining the search space to expressions following a simple and comprehensive pattern. In this paper, we resort to Simulated Annealing to search for a symbolic expression using the IT representation. Simulated Annealing exhibits an intrinsic ability to escape from poor local minima, which is demonstrated here to yield competitive results, particularly in terms of generalization, when compared with state-of-the-art Symbolic Regression techniques, that depend on population-based meta-heuristics, and committees of learning machines.\",\n \"Title: Wind Speed Forecasting via Multi-task Learning\\nYear: 2021\\n\\nAbstract: The viability of wind power massive use goes through an effective estimation of the power to be produced in wind farms, one of the fastest growing sources of renewable energy. Therefore, short-term wind speed forecasting involving several turbines and possibly multiple wind farms becomes a crucial practical demand. Once there are historical time series of wind speed intensity associated with subgroups of turbines, multiple tasks of time series prediction should be solved simultaneously. Aiming at performance improvement, we propose here the adoption of information sharing in multi-task learning frameworks, involving linear and nonlinear prediction models, and time series with and without differentiation. The nonlinear multi-task learning approach is based on extreme learning machines (ELMs) and guided to the best accuracies on a real-world case study of short-term wind speed forecasting. A suitable tuning of the variance in the distribution of the random hidden layer weights was demonstrated to be an essential preprocessing step for the ELMs, together with regularization routines. Additionally, it was found that properly incorporating tasks relationship into the learning process can systematically contribute to better generalization, particularly when compared to the single task learning counterparts.\",\n \"Title: Online Convex Optimization of a Multi-task Fuzzy Rule-based Evolving System\\nYear: 2022\\n\\nAbstract: This paper extends the recently conceived learning mechanism called EVeP (Extreme Value evolving Predictor), an evolving fuzzy-rule-based predictor characterized by innovative procedures to define the antecedent and consequent parts of the existing fuzzy rules. In EVeP, information granules are recursively updated and associated with Weibull distributions, a generalization of Gaussian distributions which incorporates more robust statistics to establish the region of influence of each fuzzy rule. Shared information from all the rules, in a multi-task formulation, is adopted to set the consequent parameters in EVeP. Given that the multi-task formulation is solved using batch learning and gradient descent, the computational cost per iteration tends to be high, being a concern in practical applications. Therefore, here the multi-task framework at the consequent part of the rules was revised to incorporate online convex optimization, given rise to EVeP_OCO. Now, antecedent and consequent parts of the rules are updated in a fully recursive way, with a clear reduction in the computational burden per iteration, particularly when the worst case scenarios are considered: the cost per iteration depends on the current number of rules to be updated. The case studies are composed of a variety of benchmark time series prediction problems. They demonstrate the significant gain in terms of computational cost per iteration, with an admissible reduction in performance by replacing a batch multi-task learning procedure by an online counterpart.\",\n \"Title: Multi-Objective Bilevel Recommender System for Food Diets\\nYear: 2022\\n\\nAbstract: This work aimed to develop a personalized multi-objective recommender system for food diets, which seeks to suggest to the user a diverse list of four meals a day (breakfast, lunch, snack, dinner). The efficient solutions are capable of simultaneously meeting a list of nutritional specifications and also minimizing both the concentration in a specific meal or food item and the total cost of acquisition and preparation. Efficient solutions are sought from the joint use of the NSGA-II and Gurobi optimization packages, after formulating the diet problem as a bilevel optimization: a combinatorial and multi-objective problem at the upper level - which food items from each category (the available food categories and the categories allocated to each of the four meals are defined in advance) should make up the diet - and a mathematical programming problem at the lower level - what is the optimal amount of the selected food items to compose the diet, given nutritional constraints. As Gurobi does not operate directly with a multi-objective optimization perspective, its lower-level objective function involves maximizing the total energy of the daily diet. Experimental results, considering fictitious food costs, show that NSGA-II and Gurobi operate in synergy, providing a diverse list of menus for the four daily meals, thus making a valuable approximation of the Pareto frontier. The distinctive aspect of this multi-objective bilevel solution to the diet problem, then, resides in the supply of diverse and, at the same time, efficient candidate solutions, in the sense of achieving the Pareto frontier and being scattered along its extension.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":112,"string":"112"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13573,"string":"13,573"}}},{"rowIdx":13574,"cells":{"question_id":{"kind":"string","value":"Gail E Kaiser"},"question":{"kind":"string","value":"Generate an abstract for the title \"A Uniform Programming Abstraction for Effecting Autonomic Adaptations onto Software Systems\" using the following items: 1. Autonomic systems\n2. Feedback control loop\n3. General-purpose programming model\n4. Low-level execution\n5. Reconfigurations"},"answer":{"kind":"string","value":"Most general-purpose work towards autonomic or self-managing systems has emphasized the front end of the feedback control loop, with some also concerned with controlling the back end enactment of runtime adaptations -- but usually employing an effector technology peculiar to one type of target system. While completely generic \"one size fits all\" effector technologies seem implausible, we propose a general purpose programming model and interaction layer that abstractsaway from the peculiarities of target specific effectors,enabling a uniform approach to controlling and coordinatingthe low-level execution of reconfigurations, repairs,micro-reboots, etc."},"correct_answer":{"kind":"string","value":"Most general-purpose work towards autonomic or self-managing systems has emphasized the front end of the feedback control loop, with some also concerned with controlling the back end enactment of runtime adaptations -- but usually employing an effector technology peculiar to one type of target system. While completely generic \"one size fits all\" effector technologies seem implausible, we propose a general purpose programming model and interaction layer that abstractsaway from the peculiarities of target specific effectors,enabling a uniform approach to controlling and coordinatingthe low-level execution of reconfigurations, repairs,micro-reboots, etc."},"documents":{"kind":"list like","value":["Title: Process Evolution in the Marvel Environment.\nYear: 1993\n\nAbstract: We present a schema and process evolution tool, called the Evolver, for the M ARVEL process-centered environment. The Evolver analyzes the differences between the new and installed process models of an existing environment, detecting each case where the notion of consistency defined by the process model has been strengthened or weakened. The Evolver then automatically updates the environment's objectbase to guarantee that the objects are consistent according to the new specifications. The Evolver can be applied while the installed process is in progress, temporarily halting normal operation while it updates the objectbase, after which development continues using the new process. We have had several months of experience using the Evolver to make repeated changes in the process that supports our own further development of MARVEL, and include in this paper one small but practical example of a recent change made to a real MARVEL process.","Title: Smarter recompilation\nYear: 1988\n\nAbstract: Tichy's Smart Recompilation method can be made smarter by permitting benign type inconsistencies between separately compiled modules. This enhanced method helps the programmer to make far-reaching changes in small, manageable steps.","Title: Experience with Marvel\nYear: 1989\n\nAbstract: MARVEL is a kernel for software development environments that enacts a processing modeling notation, called the MARVEL Strategy Language (MSL). The software process is defined in MSL as a collection of modular strategies, where different (sets of these) strategies are appropriate for different tool suites, user roles, phases of the lifecycle, management policies, etc. Each strategy specifies a partial object-oriented data model and a partial rule-based process model. The strategies are supplemented by a collection of envelopes, each of which describes how to apply one or more of the external tools used during software development activities; this includes the mapping from the inputs/outputs of these tools and the MARVEL objectbase and the command lines to invoke these tools. Multiple MSL strategies are merged to obtain the full data and process models for enaction by the MARVEL kernel.Each software development activity is described in MSL by a rule specifying the precondition and multiple postconditions of the activity. Both preconditions and postconditions are written in a subset of the first order predicate logic, stylized to incorporate directives for efficient queries and updates. A precondition is a logical statement that must be true before the activity is initiated; a postcondition is a logical statement asserted to be true when the activity completes. Multiple mutually exclusive postconditions reflect the multiple possible results of many activities. An activity is typically a particular combination of software development tool, options specific to that tool, and typed arguments. The notion of activity could be extended to include loading a new set of strategies to take over at this point in the process, to support both nesting of hierarchical activities and successive activity phases.Enaction of an MSL process model is implemented by what we call opportunistic processing, consisting primarily of forward and backward chaining among the rules. When a user initiates an activity, MARVEL applies backward chaining to attempt to satisfy its precondition. If this is not possible, MARVEL could use the rules in the style of an expert system to explain the problem. Once an activity has completed, MARVEL asserts one of its postconditions and uses forward chaining to automatically carry out other activities whose results are presumed to be needed in the near future.We have designed two facilities for controlling this processing, to prevent runaway automation. Hints are additional MSL preconditions that affect automatic but not user-initiated invocation of tools. Implicit queries represent a reflective capacity to keep tabs on estimated costs of automatic behavior and request user confirmation when certain user-selected thresholds are reached.MARVEL version 2.10 was recently completed, an upgrade over the version 2.01 that was demonstrated at the 3rd PSDE conference in November 1988. This latest version of MARVEL adds an X11 windows graphical user interface and new subsystems called Marvelizer and Organ. Marvelizer is a tool for immigrating existing software systems from the Unix file system or (a restricted class of) existing software development environment into a MARVEL objectbase. Organ is a set of commands for reorganizing MARVEL objectbases as the project and/or system structure evolves. Marvelizer has been used to immigrate Marvel itself, including about 35,000 lines of C code, into a C/Marvel objectbase. We would like to use the C/Marvel set of MSL strategies, which support activities using the user's favorite editor, lint, cc, ar, ranlib, less (and soon rcs, dbx and mail), for our own further development of MARVEL. Both MARVEL and C/Marvel were in use during the Spring 1989 semester by students in my E6123y Programming Environments and Software Tools course (admittedly a captive audience).We have discovered several difficulties with the MARVEL paradigm that seem likely to apply at least partially to enaction of other notations for software process modeling. First, constructing and debugging a process model is hard. We have run into many unanticipated interactions among rules, particularly among those defined in different strategies that happen to be loaded into the MARVEL kernel at the same time. We are working on debugging facilities that will uncover all direct and transitive dependencies by constructing a graphical dataflow graph of the currently loaded rules. It is feasible for a user to enable and disable individual rules, both during debugging to see how that affects the dataflow among rules and also during operation because certain behavior is not currently desired. This would be dangerous, however, if done by a naive user: if rules were turned off and on at arbitrary points, it would be possible for MARVEL to get into what is an inconsistent state with respect to the full set of rules in such a way that automatic recovery is impossible.Similar difficulties seem likely to arise for any enactable process modeling paradigm. The following questions need to be answered: How does an implementor go about writing, testing and debugging a process model? For non-trivial processes, the model is likely to be very large and hence should be modularized. What is the appropriate style of interface among process model modules and how is it enforced? Is the process model visible to the users and, if so, how is it presented? If the users can modify the process model during enaction, how does this affect continuing the process? Should (can) consistency be enforced or should (can) inconsistencies be detected and repaired?Second, constructing envelopes is even harder. We currently use the Shell languages (e.g., CShell) as notations for defining envelopes. Writing the Shell scripts requires an intimate knowledge of both the tool and the mapping of the MARVEL objectbase onto the file system. Although there is no way to avoid the former, we hope to minimize the latter by developing a higher level envelope language that understands the MARVEL objectbase implementation so the envelope writer doesn't have to. However, this is currently a relatively low priority due to lack of manpower.Again, similar difficulties seem inevitable for other enaction mechanisms intended to work with external tools. The questions: Is it possible (or desirable) to support existing tools, particularly tools where retrofitting is impractical? If only new tools are supported, can a relatively simple standard interface be developed and published so these tools can be developed separately by vendors? Would such an interface pre-empt too many design decisions, or would it be feasible for vendors to upgrade their current products?Third, supporting multiple users is hardest of all. The current MARVEL kernel does not provide any facilities for concurrency control. It would be easy to add locking of the entire objectbase, but a more flexible transaction model is needed. We are working on a number of alternative models that we think are promising for incorporation into the MARVEL kernel. Commit-serializability introduces two new transaction operations, split-transaction and join-transaction, that manipulate in-progress transactions to support coordination among multiple users. Participant transactions (AKA participation domains) distinguishes participants in group software development tasks from observers, and relaxes serializability as viewed by participants to permit cooperation.It is essential for enactable process models to support coordination and cooperation among the members of software development teams. There are two sets of questions: First, what are appropriate processes for programming-in-the-many? This is itself an active area of research, and there is certainly no consensus. One opinion, voiced at the previous workshop, is that the community should not be enacting process models until we understand the processes we are modeling. An opposing opinion is that enaction gives the community the capability to experiment with different processes without prematurely making a full scale commitment to a particular process. Assuming agreement with the second view, what process modeling notations and enaction mechanisms can support a reasonable range of cooperative work facilities?The database and operating system communities have agreed on atomic and serializable transactions, which can be implemented using a wide variety of technologies, as the basis for data consistency, failure recovery and concurrency control across their traditional application areas. Can the software engineering community develop an equally all-encompassing “transaction” model? Or is there some characteristic of the software development process such that coordination and cooperation policies will always require special support on a case-by-case basis? If so, what does this say about enaction of process models in general?In summary, MARVEL is implemented and in limited use, it works in some cases, it doesn't work in others, it raises a lot of questions, and both we the MARVEL group and we the software engineering community have a lot more work to do.","Title: Mechanisms (session summary)\nYear: 1989\n\nAbstract: Takuya Katayama lead the discussion on Mechanisms for Software Process Description. He classified software process views as reflected in the position papers into functional, behavioral and enactional, described some generally desirable properties of software process mechanisms, and listed those projects that seemed to have obtained non-toy experience.He made a plea for each of these projects to answer a questionnaire that, among other things, requested information regarding how the formalism handled ill-behaved processes. One generic example of an ill-behaved process was given: repeat function1, …, functionn until predicate1, …, predicatem are satisfied. It is not really clear to me that this process is ill-defined, since it seems to be computable using normal means; my opinion is that an ill-defined process is one that is made up and/or arbitrarily changed while it is in progress, in the sense of self-modifying code (e.g., Lee Osterweil's universal process program — an unfair characterization since Lee didn't show up to defend himself). In any case, Katayama's questionnaire will hopefully be addressed in the position papers for 6ISPW.The discussion focused on several interesting issues. The most controversial was instigated by Mark Dowson, when he asked what is the definition of “activity”. There were a number of different responses, but most fell into one of three camps.The first, which I'll call notation-oriented, defined an activity as the primitive (according to Watts Humphrey, the instruction set) of the process modeling notation. In general, the notation might lack facilities to express decomposition of an activity, or at least — as Colin Tully states — we choose to regard such units as decomposable. The second, people-oriented, defined an activity as a primitive in the real-world process being modeled. Bill Curtis said this might be a job that one person takes responsibility for, without interaction with other people; the process would be defined in terms of such interactions. The third, which I'll call process-oriented, was exemplified by Dewayne Perry's definition of an activity in terms of causes (preconditions?) and goals (postconditions?) with respect to the process. There was some argument over whether or not such an activity necessarily resulted in an observable change of state. Bob Balzer pointed out it is an open question as to whether syntax (notation) can realistically be separated from semantics (people and process).There was some argument in all three camps as to whether activities were really primitive, as assumed in the definitions above, or could be decomposable at a lower level of abstraction. One issue is whether or not a particular activity can be decomposed with respect to the question at hand (David Carr), for example, the encapsulation might be due to ignorance (Sam Redwine); alternatively, there may be no need for decomposition (Frank Belz).One response by Mark Dowson seems to cut across all three camps: tasks are defined as managed units (i.e., the units are scheduled and allocated resources), while activities are not managed separately from their enclosing tasks. But this leads to the distinction pointed out by Peter Feiler of machine-level scheduling and resource allocation versus process-level, since everything executed on a computer is scheduled and allocated resources at a fine granularity by the operating system. Taly Minsky expressed a concern that this entire discussion was at too high a level of abstraction, and we should return to more operational definitions; however, I think that discussion of scheduling and resource allocation is quite concrete, since we can define (computer or bureaucratic) algorithms to carry out these operations.At this juncture, Anthony Finkelstein pointed out the need for folk classification: what do the people involved in a process think are the meaningful units? This lead into a new controversy over the notion of “commitments”, which seemed to have at least as many definitions as “activities”. Here the division seemed to be over whether the most important concern was why things are being done versus what is being done; there was also an issue over what it meant for commitments to be retracted. (The conversation was rather random, so I can relay only a small subset of the interesting comments made by particular individuals.)Bill Curtis pointed out that computerized formalization and monitoring of commitments may appear to project personnel as Naziware; further, if there is a shared information base for rationale capture, who monitors and cleans it up? Sam Redwine gave the example of burying a project to track the fulfillment or lack thereof of presidential campaign promises. Walt Scacchi made a distinction between strong commitments, with real consequences if broken, and weak commitments, which are no big deal; Bill Curtis mentioned false commitments, based on what is acceptable to management or customers rather than how things are really done.According to Anthony Finkelstein, there are four kinds of commitment: commitment to a statement, no commitment to a statement, commitment to the negation of a statement, no commitment at all. Taly Minsky mentioned the commitment (or obligation) to restore inconsistent states to consistency; Ian Thomas pointed out such a commitment might be from an individual to himself, and not necessarily public.There was a digression and some brief confusion over the use of the term commitment as it relates to transactions, but that was determined to be irrelevant to the discussion at hand. Watts Humphrey defined a commitment as between people, freely entered into, based on intelligent knowledge, public, with a clear intention of performing, and prior notification and renegotiation if the commitment will not be met. In this sense, project personnel themselves act as two agents with respect to commitments, one that makes commitments while the other monitors whether the commitments are being fulfilled (Mark Dowson).After the break, two specific projects were briefly discussed.Karen Huff described her work on rule-based process modeling and planning. She made three main points: (1) components should be defined to permit dynamic construction of processes, in order to accommodate failure and iteration, particularly for ill-defined processes; (2) there is an important distinction between doing versus deciding what to do, i.e., it should be possible to reason about a course of action without committing to start its execution (Bob Balzer describes this as envisionment versus enaction); and (3) the process should be viewed in terms of its state of completion with respect to its goals rather than as an instruction pointer and a snapshot. She noted that the ability to decide was built-in, as opposed to a separate process.This led quite predictably to a discussion about planning (Bob Balzer) versus doing (Pamela Zave) versus planning in response to feedback from doing and measurement of progress (Sam Redwine). There were a number of questions, mostly from Dennis Heimbigner and Bob Balzer, regarding the power of Huff's notation (apparently Turing-complete in theory, but not fully implemented), its application to representing real processes (primarily Unix-based coding and testing), and the difficulties encountered (there is a problem with suspending and then resuming a plan after the world has changed). I mentioned the concern about side-effects from what might otherwise be considered an atomic activity, and how such side-effects could be detected (the best we can do is operating system support — which of course doesn't work for off-line activities) and modeled. Wilhelm Schaefer delineated the problem of combining process models for individuals, who can change their processes at will.Going on to the next brief project description, Walt Scacchi described his work on a hypertext environment, where an empirical evaluation was made with respect to student groups. The process was defined at the intergroup but not intragroup level, and there was no control over the work division among participants. Not surprisingly the unexpected circumstances arose at the intragroup level, which turned out to be decomposable after all and led to the greatest variation in productivity and quality. Individuals may introduce new tools spontaneously, and there's no general way to describe this. There is thus a need to produce new processes and apply them on the fly, as well a model interactions among multiple agents whose purpose may be to decide what to do next; since people meander, processes need to be able to meander. People operate in an open world situation, but the process is defined by a closed world formulation.Watts Humphrey pointed out that processes don't transplant well; outside processes are not accepted and a sense of ownership (achieved through improving the process) is needed. Marc Kellner brought up the issue of where do processes come from, e.g., DoD guidelines. This led to a discussion of description versus prescriptive processes (Marc Kellner) and abstraction versus instantiation of processes (Bob Balzer). Anthony Finkelstein questioned the idea of an emergent process. Walt Scacchi responded that modeling technology is too heavy to support modeling on the fly, and the effort of changing the model but be less than just doing the work correctly.At the end of the session, Colin Tully suggested holding a process description contest, with a benchmark process that all participants would define in their own process modeling notations. Anthony Finkelstein pointed out that this is problematical, however, since different notations address widely differing aspects of the process. It might be most reasonable to have at least two benchmarks, one involving upstream and the other downstream activities, with no expectation that every process modeling notation addresses both areas. Hopefully this will be addressed in the call for 6ISPW.","Title: Kinesthetics eXtreme: An External Infrastructure for Monitoring Distributed Legacy Systems\nYear: 2003\n\nAbstract: Autonomic computing - self-configuring, self-healing, self-optimizing applications, systems and networks - is widely believed to be a promising solution to ever- increasing system complexity and the spiraling costs of human system management as systems scale to global proportions. Most results to date, however, suggest ways to architect new software constructed from the ground up as autonomic systems, whereas in the real world organizations continue to use stovepipe legacy systems and/or build \"systems of systems\" that draw from a gamut of new and legacy components involving disparate technologies from numerous vendors. Our goal is to retrofit autonomic computing onto such systems, externally, without any need to understand or modify the code, and in many cases even when it is impossible to recompile. We present a meta-architecture implemented as active middleware infrastructure to explicitly add autonomic services via an attached feedback loop that provides continual monitoring and, as needed, reconfiguration and/or repair. Our lightweight design and separation of concerns enables easy adoption of individual components, as well as the full infrastructure, for use with a large variety of legacy, new systems, and systems of systems. We summarize several experiments spanning multiple domains.","Title: Database Support for Knowledge-Based Engineering Environments\nYear: 1988\n\nAbstract: The authors discuss Marvel, a knowledge-based engineering environment that can be instantiated with the entities and tools for a specific engineering project, together with rules regulating (technical) conduct of the project. The resulting environment maintains database representations of entities and carries out menial chores through controlled automation. The authors concentrate on the details of MSL (Marvel Strategy Language) and tool envelopes, including the algorithms used in the various parts of the Marvel system. These algorithms are different from those used for other expert systems because of Marvel's requirement for large-scale entities, the ability to load and unload strategies with different behavior and views of the database structure, and the overhead of tool processing effectively prohibiting 'undo' during backtracking.","Title: Marvel strategy language example\nYear: 1989\n\nAbstract: The DocPrep environment was developed by two students, Laura Johnson and Victor Kan, as their term project in the Spring 1989 E6123y Programming Environments and Software Tools course. Neither student had any prior knowledge of or experience with marvel, and were not familiar with its internal implementation. They successfully used DocPrep to produce their final project report. The twelve envelopes, bind, delete-sect-order, format, printdoc, printsec, review-format-err, review-spell-err, runeditor, specify-header, specify-printer, specify-sect-order and spell-check, are omitted.","Title: MELDing transactions and objects\nYear: 1991\n\nAbstract: MELD is an experimental object system under development at Columbia University. The object-oriented programming language supports classes, strong typing of instance variables, active values, multiple inheritance, and separate compilation of modular units called features that bundle together related classes and objects. These facilities were developed for an early version of MELD [3, 4], without persistence, concurrency or distribution.","Title: An object model for shared data\nYear: 1990\n\nAbstract: The classical object model supports private data within objects and clean interfaces among objects, and, by definition, does not permit sharing of data among arbitrary objects. This is a problem for certain real-world applications, where the same data logically belongs to multiple objects and may be distributed over multiple nodes on the network. Rather than give up the advantages of encapsulated objects in modeling real-world entities, a new object model that supports distribution of computation units from information-hiding concerns is proposed. A new object model is introduced, a motivating example from the financial services domain is described, and a new language based on the model is presented","Title: A Flexible Rule-Chaining Engine for Process-Based Software Engineering\nYear: 1994\n\nAbstract: We present the design of a new rule-based processengine that generalizes previous systems to supportprocess enforcement, automation, guidance, monitoring,delegation, planning, simulation, instrumentationand potentially other applications. Our approach isfully knowledge-based, tailored by knowledge regardingthe process assistance policies to be supported as wellas the process definition.1 IntroductionA process is a set of steps for developing a softwaresystem. Different processes are ...","Title: Federating Process-Centered Environments: The Oz Experience\nYear: 1998\n\nAbstract: We describe two models for federating process-centeredenvironments (PCEs): homogeneous federation among distinctinstances of the same environment framework enacting the same ordifferent process models, and heterogeneous federation amongdiverse process enactment systems. We identify the requirements andconsider possible architectures for each model, although weconcentrate primarily on the homogeneous case. The bulk of the paperpresents our choice of architecture, and correspondinginfrastructure, for homogeneous federation among MARVEL environment instances as realized in the OZ system. We brieflyconsider how a single MARVEL environment, or an OZfederation of MARVEL environments, might be integrated into aheterogeneous federation based on ProcessWall‘s facilities forinteroperating PCEs.","Title: A Marvelous Extended Transaction Processing Model\nYear: 1989\n\nAbstract: A key flaw in previous software development environment research is the lack of a transaction model supporting fault tolerance, concurrency control, consistent publication of changes and user control of commit and abort for software development activities. The atomicity properties of the classical transaction model make it unsuitable for industrial software development efforts. We sketch an extended transaction model with a commit-serializability semantics and describe the application of this model to the existing M ARVEL architecture for software development environments. MARVEL supports rule-based automation of software development activities and integrates commercial off-the-shelf tools.","Title: Experiments with rule based process modelling in an SDE\nYear: 1989\n\nAbstract: Without Abstract","Title: An architecture for WWW-based hypercode environments\nYear: 1997\n\nAbstract: A hypercode software engineering environment represents all plausible multimedia artifacts concerned with software development and evolution that can be placed or generated on-line, from source code to formal documentation to digital library resources to informal email and chat transcripts. A hypercode environment supports both internal (hypertext) and external (link server) links among these artifacts, which can be added incrementally as useful connections are discovered; project-specific hypermedia search and browsing; automated construction of artifacts and hyperlinks according the software process; application of tools to the artifacts according to the process workflow; and collaborative work for geographically dispersed teams. We present a general architecture for what we call hypermedia subwebs, and groupspace services operating on shared subwebs, based on World Wide Web technology - which could be applied over the Internet or within an intranet. We describe our realization in OzWeb.","Title: An architecture for integrating OODBs with WWW\nYear: 1996\n\nAbstract: The main topic of this paper is how to structure information so that the view of the web, both within and across web pages, is dynamically customizable. We present an architecture that integrates Object Oriented Databases with the World Wide Web to organize such dynamic structures. Different users, or the same user at different times, could have different views of the web. We discuss several architectural variants and and implementation issues. Our chosen architecture provides high flexibility for a wide variety of applications, ranging from managed healthcare to software development environments, and has been realized in the dkweb system.","Title: Report from the second international workshop on games and software engineering (GAS 2012)\nYear: 2013\n\nAbstract: The Second International Workshop on Games and Software Engineering (GAS 2012) was held in June 2012, collocated with ICSE 2012 in Zurich, Switzerland. GAS 2012 explores issues that crosscut the software engineering and the game engineering communities. Advances in game engineering techniques can be adopted by the software engineering community to develop more engaging applications across diverse domains: education; healthcare; fitness; sustainable activities (e.g., recycling awareness); and so on. Successful computer games feature a property that is not always found in traditional software: they are highly engaging. Engaging games enthrall players and result in users willing to spend increasing amounts of time and money playing them. Nine submissions were accepted for the workshop, which also featured a panel session and an informal poster session.","Title: VMVM: unit test virtualization for Java\nYear: 2014\n\nAbstract: As software evolves and grows, its regression test suites tend to grow as well. When these test suites become too large, they can eventually reach a point where they become too length to regularly execute. Previous work in Test Suite Minimization has reduced the number of tests in such suites by attempting to identify those that are redundant (e.g. by a coverage metric). Our approach to ameliorating the runtime of these large test suites is complementary, instead focusing on reducing the overhead of running each test, an approach that we call Unit Test Virtualization. This Tool Demonstration presents our implementation of Unit Test Virtualization, VMVM (pronounced \\\"vroom-vroom\\\") and summarizes an evaluation of our implementation on 20 real-world Java applications, showing that it reduces test suite execution time by up to 97% (on average, 62%). A companion video to this demonstration is available online, at https://www.youtube.com/watch?v=sRpqF3rJERI.","Title: Automating Content Extraction of HTML Documents\nYear: 2005\n\nAbstract: Web pages often contain clutter (such as unnecessary images and extraneous links) around the body of an article that distracts a user from actual content. Extraction of \"useful and relevant\" content from web pages has many applications, including cell phone and PDA browsing, speech rendering for the visually impaired, and text summarization. Most approaches to making content more readable involve changing font size or removing HTML and data components such as images, which takes away from a webpage's inherent look and feel. Unlike \"Content Reformatting,\" which aims to reproduce the entire webpage in a more convenient form, our solution directly addresses \"Content Extraction.\" We have developed a framework that employs an easily extensible set of techniques. It incorporates advantages of previous work on content extraction. Our key insight is to work with DOM trees, a W3C specified interface that allows programs to dynamically access document structure, rather than with raw HTML markup. We have implemented our approach in a publicly available Web proxy to extract content from HTML web pages. This proxy can be used both centrally, administered for groups of users, as well as by individuals for personal browsers. We have also, after receiving feedback from users about the proxy, created a revised version with improved performance and accessibility in mind.","Title: COMPASS: A Community-driven Parallelization Advisor for Sequential Software\nYear: 2009\n\nAbstract: The widespread adoption of multicores has renewed the emphasis on the use of parallelism to improve performance. The present and growing diversity in hardware architectures and software environments, however, continues to pose difficulties in the effective use of parallelism thus delaying a quick and smooth transition to the concurrency era. In this paper, we describe the research being conducted at Columbia University on a system called COMPASS that aims to simplify this transition by providing advice to programmers while they reengineer their code for parallelism. The advice proffered to the programmer is based on the wisdom collected from programmers who have already parallelized some similar code. The utility of COMPASS rests, not only on its ability to collect the wisdom unintrusively but also on its ability to automatically seek, find and synthesize this wisdom into advice that is tailored to the task at hand, i.e., the code the user is considering parallelizing and the environment in which the optimized program is planned to execute. COMPASS provides a platform and an extensible framework for sharing human expertise about code parallelization - widely, and on diverse hardware and software. By leveraging the “wisdom of crowds” model [30], which has been conjectured to scale exponentially and which has successfully worked for wikis, COMPASS aims to enable rapid propagation of knowledge about code parallelization in the context of the actual parallelization reengineering, and thus continue to extend the benefits of Moores law scaling to science and society.","Title: Parallel and distributed incremental attribute evaluation algorithms for multiuser software development environments\nYear: 1993\n\nAbstract: The problem of change propagation in multiuser software development environments distributed across a local-area network is addressed. The program is modeled as an attributed parse tree segmented among multiple user processes and changes are modeled as subtree replacements requested asynchronously by individual users. Change propagation is then implemented using decentralized incremental evaluation of an attribute grammar that defines the static semantic properties of the programming language. Building up to our primary result, we first present algorithms that support parallel evaluation on a centralized tree in response to single edits using a singe editing cursor and multiple edits with multiple editing cursors. Then we present our algorithm for parallel evaluation on a decentralized tree. We also present a protocol to guarantee reliability of the evaluation algorithm as components of the decentralized tree become unavailable due to failures and return to availability.","Title: A configuration process for a distributed software development environment\nYear: 1994\n\nAbstract: The authors describe work-in-progress on a configuration facility for a multi-site software development environment. The environment supports collaboration among geographically-dispersed teams of software developers. Addition and deletion of local subenvironment sites to a global environment is performed interactively inside any one of the existing local subenvironments, with the same user interface normally employed for invoking software development tools. This registration process is defined and executed using the same notation and mechanisms, respectively, as for the software development process. Everything described by the authors has been implemented and is working, but since they are in the midst of experimentation, they do not expect that the “final” system will be exactly as described","Title: Genspace: Exploring Social Networking Metaphors For Knowledge Sharing And Scientific Collaborative Work\nYear: 2008\n\nAbstract: Many collaborative applications, especially in scientific research, focus only on the sharing of tools or the sharing of data. We seek to introduce an approach to scientific collaboration that is based on the sharing of knowledge. We do this by automatically building organizational memory and enabling knowledge sharing by observing what users do with a particular tool or set of tools in the domain, through the addition of activity and usage monitoring facilities to standalone applications. Once this knowledge has been gathered, we apply social networking models to provide collaborative features to users, such as suggestions on tools to use, and automatically-generated sequences of actions based on past usage amongst the members of a social network or the entire community. In this work, we investigate social networking models as an approach to scientific knowledge sharing, and present an implementation called genSpace, which is built as an extension to the geWorkbench. platform for computational biologists. Last, we discuss the approach from the viewpoint of social software engineering.","Title: An Approach to Software Testing of Machine Learning Applications\nYear: 2007\n\nAbstract: Some machine learning applications are intended to learn properties of data sets where the correct answers are not already known to human users. It is challenging to test such ML software, because there is no reliable test oracle. We describe a software testing approach aimed at addressing this problem. We present our findings from testing implementations of two different ML ranking algorithms: Support Vector Machines and MartiRank.","Title: An Architectural Survey of Object Management Systems\nYear: 1992\n\nAbstract: Much work has been done in the last decade in the related areas of object-oriented programming languages and object-oriented databases. Researchers from both areas now seem to be working toward a common end, that of an object management system, or OMS. An OMS is constructed similarly to an OODB but provides a general purpose concurrent object-oriented programming language as well, complementing the OODB query facilities. In this paper, we will define several different types of object systems (object servers, persistent OOPL's, OODB's and OMS's) in terms of their interfaces and capabilities from the viewpoint of how these support the requirements of cooperative information systems. We will examine the distinguishing features and general architecture of systems of each type in the light of a general model of OMS architecture.","Title: A competitive-collaborative approach for introducing software engineering in a CS2 class\nYear: 2013\n\nAbstract: Introductory Computer Science (CS) classes are typically competitive in nature. The cutthroat nature of these classes comes from students attempting to get as high a grade as possible, which may or may not correlate with actual learning. Further, there is very little collaboration allowed in most introductory CS classes. Most assignments are completed individually since many educators feel that students learn the most, especially in introductory classes, by working alone. In addition to completing “normal” individual assignments, which have many benefits, we wanted to expose students to collaboration early (via, for example, team projects). In this paper, we describe how we leveraged competition and collaboration in a CS2 class to help students learn aspects of computer science better - in this case, good software design and software testing - and summarize student feedback.","Title: A Flexible Transaction Model for Software Engineering\nYear: 1990\n\nAbstract: It is generally recognized that the classical transaction model, providing atomicity and serializability, is too strong for certain application areas since it unnecessarily restricts concurrency. The author is concerned with supporting cooperative work in multiuser design environments, particularly teams of programmers cooperating to develop and maintain software systems. An extended transaction model that meets the special requirements of software engineering projects is presented, possible implementation techniques are described, and a number of issues regarding the incorporation of such a model into multiuser software development environments are discussed","Title: A Workgroup Model for Smart Pushing and Pulling\nYear: 1999\n\nAbstract: Our Workgroup Cache system operates as a virtual intranet, introducing a shared cache to members of the same workgroup. Users may be members of multiple workgroups at the same time. Criteria are associated with each workgroup to pull documents from an individual cache to the shared cache, or push from the shared cache to an individual cache. These criteria provide semantics of the workgroup' s tasks and interests to reduce latency for its members.","Title: MELDing data flow and object-oriented programming\nYear: 1987\n\nAbstract: MELD combines concepts from data flow and object-oriented programming languages in a unique approach to tool reusability. MELD provides three units of abstraction — equations, classes and features — that together allow sufficient options for granularity and encapsulation to support the implementation of reusable tools and the composition of existing tools in parallel (i.e., interleaved) as well as in series.","Title: A framework for immigrating existing software to new software development environments\nYear: 1991\n\nAbstract: A discussion is given on the problem of immigrating software artifacts from one software development environment (SDE) to another, for the purpose of upgrading to new SDEs as technology improves, while continuing development or maintenance of existing software systems. The authors taxonomise the larger problem of data migration, to establish the scope of immigration. They classify SDEs in terms of the ease of immigrating software artifacts from the data repository of the source SDE, without knowledge of its internal representation. A framework is presented for constructing automatic immigration tools as utilities provided by destination SDEs. They describe a specific immigration tool, called Marvelizer, implemented as part of the Marvel SDE and discuss experience using the tool","Title: Generation of run-time environments\nYear: 1986\n\nAbstract: Attribute grammars have been used for many years for automated compiler construction. Attribute grammars support the description of semantic analysis, code generation and some code optimization in a formal declarative style. Other tools support the automation of lexical analysis and parsing. However, there is one large part of compiler construction that is missing from our toolkit: run-time environments. This paper introduces an extension of attribute grammars that supports the generation of run-time environments. The extension also supports the generation of interpreters, symbolic debugging tools, and other execution-time facilities./","Title: Intelligent Assistance without Artificial Intelligence\nYear: 1987\n\nAbstract: SMILE is a multi-user software engineering environment that behaves as an intelligent assistant. SMILE presents a 'fileless environment', derives and transforms data to shelter users from entering redundant information, automatically invokes programming tools, and actively participates in the software development and maintenance process. Unlike other intelligent assistants, SMILE is not a rule-based environment: its knowledge of software objects and the programming process is hardcoded into the environment. We describe S MILE's functionality and explain how we achieved this functionality without reliance on artificial intelligence technology.","Title: Incremental parsing without a parser\nYear: 1985\n\nAbstract: This article describes an algorithm for incremental parsing of expressions in the context of syntax-directed editors for programming languages. Since a syntax-directed editor represents programs as trees and statements and expressions as nodes in trees, making minor modifications in an expression can be difficult. Consider, for example, changing a '' + '' operator to a ''*'' operator or adding a short subexpression at a syntactically but not structurally correct position, such as inserting '') * (d'' at the # mark in'' (a + b # + c)''. To make these changes in a typical syntax-directed editor, the user must understand the tree structure and type a number of tree-oriented construction and manipulation commands. This article describes an algorithm that allows the user to think in terms of the syntax of the expression as it is displayed on the screen (in infix notation) rather than in terms of its internal representation (which is effectively prefix), while maintaining the benefits of syntax-directed editing. This algorithm is significantly different from other incremental parsing algorithms in that it does not involve modifications to a traditional parsing algorithm or the overhead of maintaining a parser stack or any data structure other than the syntax tree. Instead, the algorithm applies tree transformations, in real-time as each token is inserted or deleted, to maintain a correct syntax tree.","Title: Multiple concurrency control policies in an object-oriented programming system\nYear: 1990\n\nAbstract: Object-oriented systems have been generally recognized as a good choice for approaching a range of parallel and distributed programming problems. But different parallel and distributed applications have different consistency models, so multiple concurrency control policies are needed. When objects are shared among applications with different policies, multiple policies must operate simultaneously and compatibly. We have investigated two consistency models, atomic blocks and classical serializable transactions, and the interactions both between these models and with ''uncontrolled'' concurrency, in the context of the MELD distributed object- oriented programming language. We present the relevant language constructs and discuss certain programming problems and implementation issues.","Title: Scaling Up Rule-Based Software Development Environments\nYear: 1991\n\nAbstract: Rule-based software development environments (RBDEs) model the software development process in terms of rules that encapsulate development activities, and assist in executing the process via forward and backward chaining over the rule base. We investigate the scaling up of RBDEs to support (1) multiple views of the rule base for multiple users and (2) evolution of the rule base over the lifetime of a project. Our approach is based on clarifying two distinct functions of rules and chaining: maintaining consistency and automation. By definition, consistency is mandatory whereas automation is not. Distinguishing the consistency and automation aspects of RBDE assistance mechanisms makes it possible to formalize the range of compatible views and the scope of mechanizable evolution steps. Throughout the paper, we use the MARVEL RBDE as an example application. Appeared in International Journal on Software Engineering & Knowledge Engineering , World Scientific, 2(1):59-78, March 1992.","Title: Making progress in cooperative transaction models\nYear: 1991\n\nAbstract: In the classical transaction model, transactions are consistency preserving units: a transaction is made up of a series of actions which, when executed in isolation in a reliable environment, transforms the database from one consistent state to another (2). A classical transaction management system guarantees the appearance of isolation and reliability, even though multiple transactions execute concurrently and hardware and software components fail. It does this by enforcing atomicity and serializability: atomicity means that either an entire transaction apparently executes to completion or not at all, while serializability means that the effects of the transactions are viewed as if the transactions had executed in some serial order, one completing before the next begins. This is accomplished by considering the objects read and written by concurrently executing transactions, and ensuring that either all updates are completed or none are, and that the read and write dependencies among the set of transactions correspond to some serial order of the transactions. There have been many proposals for extending the transaction model from its original data processing applications to software development, CAD/CAM and other forms of cooperative work. The notion of ''transaction'' is intuitively appealing in these domains, since forward progress often depends on making a related set of changes to a program, design or document in such a way that it is transformed from one consistent state to another. As with data processing applications, ''consistency'' depends on the requirements of the domain, such as a new system configuration passing regression and acceptance tests before the changed modules can be committed.","Title: A Mobile Agent Approach to Process-Based Dynamic Adaptation of Complex Software Systems\nYear: 2001\n\nAbstract: We describe an approach based upon software process technology to on-the-fly monitoring, redeployment, reconfiguration, and in general dynamic adaptation of distributed software applications. We choose the term dynamic adaptation to refer to modifications in structure and behavior that can be made to individual components, as well as sets thereof, or the overall target system configuration, such as adding, removing or substituting components, while the system is running and without bringing it down. The goal of dynamic adaptation is manifold: supporting run-time software composition, enforcing adherence to requirements, ensuring uptime and quality of service of mission-critical systems, recovering from and preventing faults, seamless system upgrading, etc. Our approach involves dispatching and coordinating software agents - named Worklets - via a process engine, since successful dynamic adaptation of a complex distributed software system often requires the concerted action of multiple agents on multiple components. The dynamic adaptation process must incorporate and decide upon knowledge about the specifications and architecture of the target software, as well as Worklets capabilities. Dynamic adaptation is correlated to a variety of other software processes - such as configuration management, deployment, validation and evolution - and allows addressing at run time a number of related concerns that are normally dealt with only at development time.","Title: Dynamic reconfiguration in an object-based programming language with distributed shared data.\nYear: 1991\n\nAbstract: significant to the financial analysts' strategies. Thus, On-line distributed applications generally allow conventional approaches such as interpretation and dynamic storage allocation. We present a scheme for In a previous paper (8), we introduced a distributed special cases of fine grained dynamic reconfiguration object-based programming model that addresses these sufficient for our application domain and show how it can problems. This programming model supports an be used for practical changes. We introduce new application architecture where price changes are language concepts to implement this scheme in the monitored by agents operating on behalf of individual context of an object-based programming language that portfolio managers. The sampling rate of each agent is supports shared data in a distributed environment. specific to the requirements of its portfolio manager, and it notifies its manager of changes at the granularity considered interesting by the manager's financial strategy. This model is suitable for other applications with similar","Title: Adaptive internet interactive team video\nYear: 2005\n\nAbstract: The increasing popularity of online courses has highlighted the lack of collaborative tools for student groups. In addition, the introduction of lecture videos into the online curriculum has drawn attention to the disparity in the network resources used by students. We present an e-Learning architecture and adaptation model called AI2TV (Adaptive Internet Interactive Team Video), which allows virtual students, possibly some or all disadvantaged in network resources, to collaboratively view a video in synchrony. AI2TV upholds the invariant that each student will view semantically equivalent content at all times. Video player actions, like play, pause and stop, can be initiated by any student and their results are seen by all the other students. These features allow group members to review a lecture video in tandem, facilitating the learning process. Experimental trials show that AI2TV can successfully synchronize video for distributed students while, at the same time, optimizing the video quality, given fluctuating bandwidth, by adaptively adjusting the quality level for each student.","Title: Infuse: a tool for automatically managing and coordinating source changes in large systems\nYear: 1987\n\nAbstract: In current change management tools, the actual changes occur outside the tool. In contrast, Infuse concentrates on the actual change process and provides facilities for both managing and coordinating source changes. Infuse provides facilities for automatically structuring the cooperation among programmers, propagating changes, and determining the consistency of changes, and provides a basis for negotiating the resolution of conflicting changes and for iterating over a set of changes.","Title: The CORD Appraoch to Extensible Concurrency Control\nYear: 1997\n\nAbstract: Database management systems (DBMSs) have been increasingly used for advanced application domains, such as software development environments, workflow management systems, computer-aided design and manufacturing, and managed healthcare. In these domains, the standard correctness model of serializability is often too restrictive. We introduce the notion of a Concurrency Control Language (CCL) that allows a database application designer to specify concurrency control policies to tailor the behavior of a transaction manager. A well-crafted set of policies defines an extended transaction model. The necessary semantic information required by the CCL run-time engine is extracted from a task manager, a (logical) module by definition included in all advanced applications. This module stores task models that encode the semantic information about the transactions submitted to the DBMS. We have designed a rule-based CCL, called CORD, and have implemented a run-time engine that can be hooked to a conventional transaction manager to implement the sophisticated concurrency control required by advanced database applications. We present an architecture for systems based on CORD and describe how we integrated the CORD engine with the Exodus Storage Manager to implement Altruistic Locking.","Title: Database research at Columbia University\nYear: 1998\n\nAbstract: this report,we describe the Columbia Fast Query Project (Section2), the JAM project (Section 3), the CARDGISproject (Section 4), the Columbia Internet InformationSearching Project (Section 5), the ColumbiaContent-Based Visual Query project (Section 6), andprojects associated with Columbia's ProgrammingSystems Laboratory (Section 7).","Title: Divergence control for distributed database systems\nYear: 1995\n\nAbstract: This paper presents distributed divergence control algorithms for epsilon serializability for both homogeneous and heterogeneous distributed databases. Epsilon serializability allows for more concurrency by permitting non-serializable interleavings of database operations among epsilon transactions. We first present a strict 2-phase locking divergence control algorithm and an optimistic divergence control algorithm for a homogeneous distributed database system, where the local orderings of all the sub-transactions of a distributed epsilon transaction are the same. In such an environment, the total inconsistency of a distributed epsilon transaction is simply the sum of those of all its sub-transactions. We then describe a divergence control algorithm for a heterogeneous distributed database system, where the local orderings of all the sub-transactions of a distributed epsilon transaction may not be the same and the total inconsistency of a distributed epsilon transaction may be greater than the sum of those of all its sub-transactions. As a result, in addition to executing a local divergence control algorithm in each site to maintain the local inconsistency, a global mechanism is needed to take into account the additional inconsistency.","Title: Distributed divergence control for epsilon serializability\nYear: 1993\n\nAbstract: Epsilon serializability (ESR) allows for more concurrency by permitting nonserializable interleavings of database operations among epsilon transactions (ETs). The authors present the design of distributed divergence control (DDC) algorithms for ESR in homogeneous and heterogeneous distributed databases. They first present a strict two-phase locking DDC algorithm (S2PLDDC) and an optimistic DDC algorithm (ODDC) for homogeneous distributed databases, where the local orderings of all the sub-ETs of a distributed ET are the same, and the total inconsistency of a distributed ET is simply the sum of that of all its sub-ETs. A superdatabase DDC algorithm is described for heterogeneous distributed databases, where the local orderings of all the sub-ETs of a distributed ET may not be the same, and the total inconsistency of a distributed ET may be greater than the sum of that of all its sub-ETs. As a result, in addition to local divergence control in each site, a global mechanism is needed to guarantee ESR","Title: An architecture for multi-user software development environments\nYear: 1993\n\nAbstract: We present an architecture for multi-user software development environments, covering general, process-centered and rule-based MUSDEs. Our architecture is founded on componentization, with particular concern for the capability to replace the synchronization component–to allow experimentation with novel concurrency control mechanisms–with minimal effects on other components while still suporting integration. The architecture has been implemented for the MARVEL SDE.","Title: The design and implementation of late binding in a distributed programming language\nYear: 1993\n\nAbstract: In distributed application domains where data change rapidly, it is often desirable for programs to obtain the latest available data values to achieve accurate computations. Example applications are financial services and network management. Such data are logically shared by a network of programs. Unlike data in traditional databases, rapidly changing data are usually not lockable by (client) programs and it is crucial to the computations to access their values in a timely manner. In these application domains, a typical program usually performs computations based on recently available data values obtained from the network. However, these data values may be inconsistent or obsolete, since the real data are external to the system and may change more rapidly than can be reflected by their copies within the system. Decision making based on such inaccurate computations can lead to substantial penalties. In this paper, we propose an approach to delaying data value retrieval until needed in distributed programming, considering data and configuration change rapidly. This approach offers the advantage of obtaining more recent data values, resulting in more accurate computations and decision making.","Title: Enveloping Sophisticated Tools into Process-Centered Environments\nYear: 1996\n\nAbstract: . We present a tool integration strategy based on enveloping pre-existing tools withoutsource code modifications or recompilation, and without assuming an extension language, applicationprogramming interface, or any other special capabilities on the part of the tool. This BlackBox enveloping (or wrapping) idea has existed for a long time, but was previously restricted torelatively simple tools. We describe the design and implementation of, and experimentation with,a new Black Box...","Title: GESDOR - a generic execution model for sharing of computer-interpretable clinical practice guidelines.\nYear: 2003\n\nAbstract: We developed the Guideline Execution by Semantic Decomposition of Representation (GESDOR) model to share guidelines encoded in different formats at the execution level. For this purpose, we extracted a set of generalized guideline execution tasks from the existing guideline representation models. We then created the mappings between specific guideline representation models and the set of the common guideline execution tasks. Finally, we developed a generic task-scheduling model to harmonize the existing approaches to guideline task scheduling. The evaluation has shown that the GESDOR model can be used for the effective execution of guidelines encoded in different formats, and thus realizes guideline sharing at the execution level.","Title: Adaptive Synchronization Of Semantically Compressed Instructional Videos For Collaborative Distance Learning\nYear: 2007\n\nAbstract: The increasing popularity of online courses has highlighted the need for collaborative learning tools for student groups. In this article, we present an e-Learning architecture and adaptation model called AI2TV (Adaptive Interactive Internet Team Video), which allows groups of students to collaboratively view instructional videos in synchrony. Video player actions, like play, pause and stop, can be initiated by any group member and and the results of those actions are synchronized with all the other students. These features allow students to review a lecture video in tandem, facilitating the learning process. AI2TV upholds the invariant that each student will receive semantically equivalent content at all times. Experimental trials show that AI2TV successfully synchronizes instructional videos for distributed students while concurrently optimizing the video quality, even under conditions of fluctuating bandwidth, by adaptively adjusting the quality level for each student while still maintaining the semantic content of the instructional videos.","Title: Extending attribute grammars to support programming-in-the-large\nYear: 1994\n\nAbstract: Attribute grammars add specification of static semantic properties to context-free grammars, which, in turn, describe the syntactic structure of program units. However, context-free grammars cannot express programming-in-the-large features common in modern programming languages, including unordered collections of units, included units, and sharing of included units. We present extensions to context-free grammars, and corresponding extensions to attribute grammars, suitable for defining such features. We explain how batch and incremental attribute-evaluation algorithms can be adapted to support these extensions, resulting in a uniform approach to intraunit and interunit static semantic analysis and translation of multiunit programs.","Title: A Virtual Environment for Collaborative Distance Learning with Video Synchronization\nYear: 2004\n\nAbstract: We present a 3D collaborative virtual environment, CHIME, in which geographically dispersed students can meet together in study groups or to work on team projects. Conventional educational materials from heterogeneous backend data sources are reflected in the virtual world through an automated metadata extraction and projection process that structurally organizes container materials into rooms and interconnecting doors, with atomic objects within containers depicted as furnishings and decorations. A novel in-world authoring tool makes it easy for instructors to design environments, with additional in-world modification afforded to the students themselves, in both cases without programming. Specialized educational services can also be added to virtual environments via programmed plugins. We present an example plugin that supports synchronized viewing of lecture videos by groups of students with widely varying bandwidths.","Title: RUST: A Retargetable Usability Testbed for Web Site Authentication Technologies\nYear: 2008\n\nAbstract: Website authentication technologies attempt to make the identity of a website clear to the user, by supplying in- formation about the identity of the website. In prac- tice however, usability issues can prevent users from cor- rectly identifying the websites they are interacting with. To help identify usability issues we present RUST, a Retargetable USability Testbed for website authentica- tion technologies. RUST is a testbed that consists of a test harness, which provides the ability to easily con- figure the environment for running usability study ses- sions, and a usability study design that evaluates usability based on spoofability, learnability, and acceptability. We present data collected by RUST and discuss preliminary results for two authentication technologies, Microsoft CardSpace and Verisign Secure Letterhead. Based on the data collected, we conclude that the testbed is useful for gathering data on a variety of technologies.","Title: Workgroup Middleware for Distributed Projects*\nYear: 1998\n\nAbstract: We have developed a middleware framework for workgroup environments that can support distributed software develop- ment and a variety of other application domains requiring document management and change management for distributed projects. The framework enables hypermedia-based integra- tion of arbitrary legacy and new information resources avail- able via a range of protocols, not necessarily known in ad- vance to us as the general framework developers nor even to the environment instance designers. The repositories in which such information resides may be dispersed across the Internet and/or an organizational intranet. The framework also permits a range of client models for user and tool inter- action, and applies an extensible suite of collaboration ser- vices, including but not limited to multi-participant workflow and coordination, to their information retrievals and updates. That is, the framework is interposed between clients, services and repositories — thus \"middleware\". We explain how our framework makes it easy to realize a comprehensive collec- tion of workgroup and workflow features we culled from a requirements survey conducted by NASA.","Title: Unit test virtualization with VMVM\nYear: 2014\n\nAbstract: Testing large software packages can become very time intensive. To address this problem, researchers have investigated techniques such as Test Suite Minimization. Test Suite Minimization reduces the number of tests in a suite by removing tests that appear redundant, at the risk of a reduction in fault-finding ability since it can be difficult to identify which tests are truly redundant. We take a completely different approach to solving the same problem of long running test suites by instead reducing the time needed to execute each test, an approach that we call Unit Test Virtualization. With Unit Test Virtualization, we reduce the overhead of isolating each unit test with a lightweight virtualization container. We describe the empirical analysis that grounds our approach and provide an implementation of Unit Test Virtualization targeting Java applications. We evaluated our implementation, VMVM, using 20 real-world Java applications and found that it reduces test suite execution time by up to 97% (on average, 62%) when compared to traditional unit test execution. We also compared VMVM to a well known Test Suite Minimization technique, finding the reduction provided by VMVM to be four times greater, while still executing every test with no loss of fault-finding ability.","Title: JPernLite: Extensible Transaction Services for the WWW\nYear: 1999\n\nAbstract: Concurrency control is one of the key problems in design and implementation of collaborative systems such as hypertext/hypermedia systems, CAD/CAM systems, and software development environments. Most existing systems store data in specialized databases with built-in concurrency control policies, usually implemented via locking. It is desirable to construct such collaborative systems on top of the World Wide Web, but most Web servers do not support even conventional transactions, let alone distributed (multi-Website) transactions or flexible concurrency control mechanisms oriented toward teamwork驴such as event notification, shared locks, and fine granularity locks. We present a transaction server that operates independently of Web servers or the collaborative systems, to fill the concurrency control gap. By default, the transaction server enforces the conventional atomic transaction model, where sets of operations are performed in an all-or-nothing fashion and isolated from concurrent users. The server can be tailored dynamically to apply more sophisticated concurrency control policies appropriate for collaboration. The transaction server also supports applications employing information resources other than Web servers, such as legacy databases, CORBA objects, and other hypermedia systems. Our implementation permits a wide range of system architecture styles.","Title: The Apache HTTP Server Project\nYear: 1997\n\nAbstract: Most reports of Internet collaboration refer to small scale operations among a few authors or designers. However, several projects have shown that the Internet can also be the locus for large scale collaboration. In these projects, contributors from around the world combine their individual forces and develop a product that rivals those of multibillion dollar corporations. The Apache HTTP Server Project is a case in point. This collaborative software development effort has created a robust, feature-rich HTTP server software package that currently dominates the public Internet market (46 percent compared with 16 percent for Microsoft and 12 percent for Netscape, according to a June 1997 survey published by Netcraft). The software and its source code are free, but Apache's popularity is more often attributed to performance than price. The project is managed by the Apache Group, a geographically distributed group of volunteers who use the Internet and Web to communicate, develop, and distribute the server and its related documentation. In addition, hundreds of users have contributed ideas, code, and documentation to the project","Title: Enveloping \"Persistent\" Tools for a Process-Centered Environment\nYear: 1995\n\nAbstract: this document are those of the authors and should not be interpreted asrepresenting the official policies, either expressed or implied, of ARPA, NSF, the US Government,AT&T, Bull or Xerox.","Title: Incremental Process Support for Code Reengineering\nYear: 1994\n\nAbstract: Reengineering a large code base can be a monumental task, and the situation becomes even worse if the code is concomitantly being modified. Since January 1992, we have been using the MARVEL process centered environment (PCE) for all of our software development, and are currently using it to develop the Oz PCE (MARVEL's successor). Towards this effort, we are reengineering Oz's code base to isolate the process engine, transaction manager, and object management system as separate components that can be used in arbitrary systems. In this paper, we show how a PCE can assist teams of users in carrying out code reengineering while allowing them to continue their normal code development. The key features to this approach are its incremental nature and the ability of the PCE to automate most of the tasks necessary to maintain the consistency of the code base","Title: Automatic extension of an ATN knowledge base\nYear: 1981\n\nAbstract: A computer program is described that acquires much of its knowledge from conversations among operators on Morse code radio networks. The system consists of a learning component and a language understander. The learning component extends a ‘core’ augmented transition network (ATN) knowledge base by generalizing from sentences taken from scripts of actual conversations. The extensions enable the understanding component to process a large number of sentences that are syntactically and semantically similar to the examples.","Title: WWW-based collaboration environments with distributed tool services\nYear: 1998\n\nAbstract: We have developed an architecture for a general-purpose framework for hypermedia collaboration environments that support purposeful work by orchestrated teams. The hypermedia represents all plausible multimedia artifacts concerned with the collaborative task(s) at hand that can be placed or generated on-line, from application-specific materials (e.g., source code, chip layouts, blueprints) to formal documentation to digital library resources to informal email and chat transcripts. The framework capabilities support both internal (WWW-style hypertext) and external (non-WWW open hypertext link server) links among these artifacts, which can be added incrementally as useful connections are discovered; project-specific intelligent hypermedia search and browsing; automated construction of artifacts and hyperlinks according to the semantics of the group and individual tasks and the overall workflow among the tasks; application of arbitrary tools to the artifacts; and collaborative work for geographically dispersed teams connected by the Internet and/or an intranet/extranet. We also present a general architecture for a WWW-based distributed tool launching service compatible with our collaboration environment framework. We describe our prototype realization of the framework in OzWeb. It reuses object-oriented data management for application-specific hyperbase organization, and workflow enactment and cooperative transactions as built-in services, which were originally developed for the Oz non-hypermedia environment. The tool service is implemented by the generic Rivendell component, which has been integrated into OzWeb as an example “foreign” (i.e., add-on) service. Rivendell could alternatively be employed in a stand-alone manner. We have several months experience using an OzWeb hypermedia collaboration environment for our own continuing software development work on the system.","Title: WWW access to legacy client/server applications\nYear: 1996\n\nAbstract: We describe a method for accessing Client/Server applications from standard World Wide Web browsers. An existing client for the system is modified to perform HTTP Proxy duties. Web browser users simply configure their browsers to use this HTTP Proxy, and can then access the system via specially encoded URLs that the HTTP Proxy intercepts and sends to the legacy server system. An example implementation using the Oz Process Centered Software Development Environment is presented.","Title: Transactions for concurrent object-oriented programming systems\nYear: 1989\n\nAbstract: Concurrent object-oriented programming systems (COOPS) require support for fault tolerance, concurrency control, consistent commitment of changes and program-initiated rollback. It is sometimes suggested that the classical transaction processing model successfully applied in databases and operating systems be integrated directly into COOPS facilities. This is clearly desirable, but by itself is too limiting. COOPS applications require several granularities of transaction-like facilities. A number of transaction-like mechanisms were addressed at the workshop, and no doubt there are many others suitable for COOPS. Here I briefly survey four levels of granularity that were discussed at the workshop.In their workshop paper Linearizable Concurrent Objects and in [1], Maurice Herlihy and Jeannette Wing describe a fine granularity correctness condition for COOPS, linearizability. Linearizability requires that each operation should appear to “take effect” instantaneously and that the order of non-concurrent operations should be preserved. My workshop paper Concurrent Meld discusses the incorporation of atomic blocks as a programming facility into the Meld COOPS. Atomic blocks in effect implement linearizability at the level of blocks (essentially compound statements, which may encompass an entire method).In his workshop paper Nested Objects and in [4], Bruce Martin describes small grain mechanisms for both externally serializable and semantically verifiable operations for COOPS. Externally serializable operations enforce serializability among top-level operations but permit non-serializable computations on subobjects. Semantically verifiable operations do not enforce serializability at all, but instead consider the semantics of apparently conflicting operations in preventing inconsistencies from being introduced.My paper also discusses our plans for integrating medium granularity transactions — that is, classical transactions with atomicity and serializability as the correctness conditions — into Meld. Serializable transactions have been implemented in a separate version of Meld. not yet integrated into the main line. This implementation uses an algorithm for distributed optimistic concurrency control with multiple versions [2], but two phase locking, timestamps, etc. seem equally feasible for COOPS depending on the anticipated applications.We are currently working on large grain transaction-like facilities for COOPS. Our particular concern is an important class of COOPS applications — open-ended activities [5] such as CAD/CAM. VLSI design, medical informatics, office automation and software development — characterized by uncertain duration, uncertain developments and interaction with other concurrent activities. Classical transactions are not suitable for open-ended activities for the following reasons:Fault tolerance implies atomicity, so if there is some failure, the transaction is rolled back and retried. This is entirely inappropriate when the 'transaction' consists of fixing a bug by browsing and editing a number of source files, compiling and linking, executing test cases and generating traces, etc., which may easily take several hours.Concurrency control requires that users are isolated, and the transactions appear to have been executed in some serial order. But it is not acceptable for a programmer to be locked out from editing a source file just because some other programmer had previously read the file but has not yet finished his changes to other files.Consistent commitment implies simultaneously making all of a transaction's updates publicly available only when the first transaction completes and commits. But programmers must release certain source files so they can be viewed and/or compiled and linked by other programmers cooperating on the same subsystem, while continuing in progress work on other files.An abort deletes all changes made during the transaction, so they are never available to other transactions. A programmer may realize that his original ideas about the cause of a program error were incorrect and decide to start over, but he may want to keep his incorrectly modified versions available for reference.Thus the traditional transaction model is not suitable for open-ended activities such as software development, at least where the 'transactions' are at the granularity of bug fixes, completion of a milestone, or release of a product. An extended transaction processing model, with supporting programming facilities, is necessary for COOPS.We have investigated an extended transaction model, commit-serializability, as a candidate for integration into Meld and other COOPS. The term commit-serializability denotes a model where all committed transactions are in fact serializable in the standard sense, but these transactions may not correspond in a simple way to those transactions that were initiated. In particular, the initiated transactions may be divided during operation and parts committed separately in such ways that these transactions are not serializable.Consider two in-progress transactions T1 and T2. T1 is divided under program or user control into T3 and T4, and shortly thereafter T3 commits while T4 continues. T2 may view the committed updates of T3, some of which were made by T1 before the division, and then itself commits. T4 may then view the committed updates of T2 before it commits. T2, T3 and T4 are serializable, but T1 and T2 are not. The originally initiated transaction T1 in effect disappears, and in particular is neither committed nor aborted.Commit-serializability is supported by two new transaction processing operations, split-transaction and join-transaction [5], in addition to the standard begin-transaction, commit-transaction and abort-transaction operations. The split-transaction operation supports the kind of division described above; the inverse join-transaction operation merges a completed transaction into an in-progress transaction to commit their results together.The two-way versions of the split and join operations take the following arguments. Either statement may be executed as the last statement of any method invoked as part of a transaction T. split-Transaction (A: (AReadSet, AWriteSet, AMessage), B: (BReadSet, BWriteSet, BMessage)) Join-Transaction (S: TID)When the split-transaction operation is invoked during a transaction T, there is a TReadSet consisting of all objects read by T but not updated and a TWriteSet consisting of all objects updated by T. In COOPS, TReadSet is the set of instance variables read by T, including the instance variables of objects (transitively) sent a message where the responding method accessed but did not modify these instance variables of that object. TWriteSet is the set of instance variables assigned by T, including the instance variables of objects (transitively) sent a message where the responding method modified these instance variables of that object. In both cases, the object whose method initiated T acts as coordinator of the cohort.TReadSet is divided, not necessarily disjointly, into AReadSet and BReadSet. TWriteSet is divided disjointly into AWriteSet and BWriteSet; AWriteSet normally must also be disjoint from BReadSet and BWriteSet from AReadSet. AMessage and BMessage are the selectors of messages sent to the same object ($self in Meld) to continue the two transactions. In the special case where A is immediately committed, AMessage is the operation commit-transaction, and objects in AWriteSet may also appear in either BReadSet or BWriteSet.When the join-transaction operation is invoked during a transaction T, target transaction S must be ongoing — and not in conflict with T. TReadSet and TWriteSet are added to SReadSet and SWriteSet, respectively, and S may continue or commit. One difficulty of join-transaction that did not arise for split-transaction is the necessity of naming another transaction S, other than the current one T. This can be done by treating each transaction as a special object, and references to such special objects can be stored in instance variables (TID in the figure), sent as arguments in messages and even sent messages — although these objects respond only to transaction operations. (This is orthogonal to treating different classes of transaction processing characteristics as mixin superclasses, as in Avalon [3].) More work on naming issues is needed.In the cases of both split-transaction and join-transaction, the originally initiated transaction T is divided or merged, respectively, so the net effect is as if it had never existed. The tables, logs, etc. used in the transaction manager implementation may be updated to expunge knowledge of T and replace it with knowledge of A and B or S, respectively, or this may be optimized to use T's identification for one of A or B, or for S.","Title: Garp: Graph Abstractions for Concurrent Programming\nYear: 1988\n\nAbstract: Several research projects are investigating parallel processing languages where dynamic process topologies can be constructed. Failure to impose abstractions on interprocess connection patterns can result in arbitrary interconnection topologies that are difficult to understand. We propose the use of a graph-grammar based formalism to control the complexities arising from trying to program such dynamic networks.","Title: Us and them: a study of privacy requirements across north america, asia, and europe\nYear: 2014\n\nAbstract: Data privacy when using online systems like Facebook and Amazon has become an increasingly popular topic in the last few years. However, only a little is known about how users and developers perceive privacy and which concrete measures would mitigate their privacy concerns. To investigate privacy requirements, we conducted an online survey with closed and open questions and collected 408 valid responses. Our results show that users often reduce privacy to security, with data sharing and data breaches being their biggest concerns. Users are more concerned about the content of their documents and their personal data such as location than about their interaction data. Unlike users, developers clearly prefer technical measures like data anonymization and think that privacy laws and policies are less effective. We also observed interesting differences between people from different geographies. For example, people from Europe are more concerned about data breaches than people from North America. People from Asia/Pacific and Europe believe that content and metadata are more critical for privacy than people from North America. Our results contribute to developing a user-driven privacy framework that is based on empirical evidence in addition to the legal, technical, and commercial perspectives.","Title: Manipulating managed execution runtimes to support self-healing systems\nYear: 2005\n\nAbstract: Self-healing systems require that repair mechanisms are available to resolve problems that arise while the system executes. Managed execution environments such as the Common Language Runtime (CLR) and Java Virtual Machine (JVM) provide a number of application services (application isolation, security sandboxing, garbage collection and structured exception handling) which are geared primarily at making managed applications more robust. However, none of these services directly enables applications to perform repairs or consistency checks of their components. From a design and implementation standpoint, the preferred way to enable repair in a self-healing system is to use an externalized repair/adaptation architecture rather than hardwiring adaptation logic inside the system where it is harder to analyze, reuse and extend. We present a framework that allows a repair engine to dynamically attach and detach to/from a managed application while it executes essentially adding repair mechanisms as another application service provided in the execution environment.","Title: Modeling Concurrency in Rule-Based Development Environments\nYear: 1990\n\nAbstract: The problem of cooperative work in the software development domain is explored, and a solution that combines object-oriented programming with rule-based modeling is proposed. The solution divides the problem into three components: how to detect potential conflicts between developers' concurrent activities, how to specify the consistency requirements of a project, and how to use the consistency specification to resolve potential conflicts. The focus is on the first component; the other two are merely sketched. The solution exploits recent advances in object-oriented databases, extended transaction models, and computer-supported cooperative work, all of which provide clues as to how to support cooperation while guaranteeing data consistencyed.","Title: An object-based programming model for shared data\nYear: 1992\n\nAbstract: The classical object model supports private data within objects and clean interfaces between objects, and by definition does not permit sharing of data among arbitrary objects. This is a problem for real-world applications, such as advanced financial services and integrated network management, where the same data logically belong to multiple objects and may be distributed over multiple nodes on the network. Rather than give up the advantages of encapsulated objects in modeling real-world entities, we propose a new object model that supports shared data in a distributed environment. The key is separating distribution of computation units from information-hiding concerns. Minimal units of data and control, called facets, may be shared among multiple objects and are grouped into processes. Thus, a single object, or information-hiding unit, may be distributed among multiple processes, or computation units. In other words, different facets of the same object may reside in different address spaces on different machines. We introduce our new object model, describe a motivating example from the financial domain, and then explain facets, objects, and processes, followed by timing and synchronization concerns.","Title: Properties of Machine Learning Applications for Use in Metamorphic Testing\nYear: 2008\n\nAbstract: It is challenging to test machine learning (ML) applica- tions, which are intended to learn properties of data sets where the correct answers are not already known. In the absence of a test oracle, one approach to testing these ap- plications is to use metamorphic testing, in which proper- ties of the application are exploited to define transforma- tion functions on the input, such that the new output will be unchanged or can easily be predicted based on the original output; if the output is not as expected, then a defect must exist in the application. Here, we seek to enumerate and classify the metamorphic properties of some machine learn- ing algorithms, and demonstrate how these can be applied to reveal defects in the applications of interest. In addition to the results of our testing, we present a set of properties that can be used to define these metamorphic relationships so that metamorphic testing can be used as a general ap- proach to testing machine learning applications.","Title: Vroom: Faster Build Processes for Java\nYear: 2015\n\nAbstract: Build processes are too slow. Because most of the build time for Java projects is spent executing tests, researchers have focused on speeding up testing. They've integrated two complementary approaches into a system that seamlessly supports Ant and Maven JUnit build processes. The first approach, unit test virtualization, isolates in-memory dependencies among test cases, which otherwise are isolated inefficiently by restarting the Java Virtual Machine (JVM) before every test. The system supports just-in-time reinitialization of only the small portion of memory needed by the next test, reusing a single JVM. The implementation of this approach is called VMVM (Virtual Machine in the Virtual Machine, pronounced \"vroom vroom\"). In addition, simple setup and tear-down resource management methods designed for sequential execution lead to conflicts when the resources are accessed concurrently. So, the second approach, virtualized unit test virtualization, isolates external dependencies such as files and network ports while long-running tests execute in parallel. For this, the system distributes testing jobs in round-robin manner among OS-level virtual machines. The result is, on average, a 51 percent speedup of application build times. The implementation of this approach is called VMVMVM (Virtual Machine in a Virtual Machine on a Virtual Machine \"vroom vroom vroom\").","Title: Phosphor: illuminating dynamic data flow in commodity jvms\nYear: 2014\n\nAbstract: Dynamic taint analysis is a well-known information flow analysis problem with many possible applications. Taint tracking allows for analysis of application data flow by assigning labels to data, and then propagating those labels through data flow. Taint tracking systems traditionally compromise among performance, precision, soundness, and portability. Performance can be critical, as these systems are often intended to be deployed to production environments, and hence must have low overhead. To be deployed in security-conscious settings, taint tracking must also be sound and precise. Dynamic taint tracking must be portable in order to be easily deployed and adopted for real world purposes, without requiring recompilation of the operating system or language interpreter, and without requiring access to application source code. We present Phosphor, a dynamic taint tracking system for the Java Virtual Machine (JVM) that simultaneously achieves our goals of performance, soundness, precision, and portability. Moreover, to our knowledge, it is the first portable general purpose taint tracking system for the JVM. We evaluated Phosphor's performance on two commonly used JVM languages (Java and Scala), on two successive revisions of two commonly used JVMs (Oracle's HotSpot and OpenJDK's IcedTea) and on Android's Dalvik Virtual Machine, finding its performance to be impressive: as low as 3% (53% on average; 220% at worst) using the DaCapo macro benchmark suite. This paper describes our approach toward achieving portable taint tracking in the JVM.","Title: Pebbles: fine-grained data management abstractions for modern operating systems\nYear: 2014\n\nAbstract: Support for fine-grained data management has all but disappeared from modern operating systems such as Android and iOS. Instead, we must rely on each individual application to manage our data properly - e.g., to delete our emails, documents, and photos in full upon request; to not collect more data than required for its function; and to back up our data to reliable backends. Yet, research studies and media articles constantly remind us of the poor data management practices applied by our applications. We have developed Pebbles, a fine-grained data management system that enables management at a powerful new level of abstraction: application-level data objects, such as emails, documents, notes, notebooks, bank accounts, etc. The key contribution is Pebbles's ability to discover such high-level objects in arbitrary applications without requiring any input from or modifications to these applications. Intuitively, it seems impossible for an OS-level service to understand object structures in unmodified applications, however we observe that the high-level storage abstractions embedded in modern OSes - relational databases and object-relational mappers - bear significant structural information that makes object recognition possible and accurate.","Title: Intelligent assistance for software development and maintenance\nYear: 1988\n\nAbstract: An environment is described, called Professor Marvel, that provides early error checking and answers questions about the program under development. The environment has a certain understanding of the systems being developed and how to use tools to produce software. It aids individual programmers and helps coordinate programming teams. The key components of Marvel are a database that stores data represented as objects, as in object-oriented languages, and a model of the development process that imposes a structure on programming activities. Marvel's support of insight and of opportunistic processing is discussed at length, as is the handling of side effects. A sample session is described.<>","Title: Rapid prototyping of concurrent programming languages\nYear: 1988\n\nAbstract: A novel approach is proposed for automatic generation of concurrent interpreters for formal specifications of the programming languages. It consists of a formal notation and supporting algorithms. The formal notation is called action equations, which are attribute grammars extended by the concepts of events and unification. The supporting algorithms include preprocessing algorithms to generate the interpreters and evaluation algorithms embedded in the generated interpreters. Attribute grammars are reviewed, the extension to the action equations paradigm is explained, and the synthesis of action equations with a unification strategy as a means for specifying and implementing synchronization primitives is presented. The approach is illustrated by a specification of CSP. A brief comparison to related work is included","Title: Infuse: fusing integration test management with change management\nYear: 1989\n\nAbstract: Infuse is an experimental software development environment focusing on change coordination during the maintenance/evolution phase of large scale software projects. Its core philosophy is to integrate strongly connected modules first and more weakly connected sets of modules later, moving up a hierarchy from singletons to clusters of interdependent modules and, finally, merging the change set into the baseline. We have previously described how Infuse enforces static consistency at each level of the hierarchy. We now extend our work to dynamic consistency — i.e., testing. Unit testing is done for the individual modules at the leaves of the hierarchy, integration testing for the intermediate clusters and acceptance testing at the root. Infuse supports this by partially automating the construction of test harnesses and regression test suites at each level of the hierarchy from components available from lower levels. Infuse is implemented for C, and is used to support its own evolution, but the implementation does not yet provide the test management described here.","Title: Testing reliable distributed applications through simulated events\nYear: 1989\n\nAbstract: There are many distributed applications that incorporate application-specific reliability algorithms that operate on top of general-purpose networking, operating system, and programming language facilities. The authors present a framework for application-level reliability testing suitable for a wide range of distributed applications using low-level events and the automatic generation of series of these events. They also describe how they used this testing approach for a particular application","Title: An information retrieval approach for automatically constructing software libraries\nYear: 1991\n\nAbstract: A technology for automatically assembling large software libraries which promote software reuse by helping the user locate the components closest to her/his needs is described. Software libraries are automatically assembled from a set of unorganized components by using information retrieval techniques. The construction of the library is done in two steps. First, attributes are automatically extracted from natural language documentation by using an indexing scheme based on the notions of lexical affinities and quantity of information. Then a hierarchy for browsing is automatically generated using a clustering technique which draws only on the information provided by the attributes. Due to the free-text indexing scheme, tools following this approach can accept free-style natural language queries.","Title: Debugging multithreaded programs with MPD\nYear: 1991\n\nAbstract: MPD, a multiprocessor debugger, is described. In MPD, users specify sequential or parallel event (breakpoint) expressions and associated actions; MPD automatically generates code to recognize these patterns of events and invokes the corresponding actions when the event expressions are satisfied during the program execution. The richness of the expression-specification language gives users a great deal of precision in formulating queries, obviating tedious searches through long traces of multithreaded software. Implementation issues are discussed.<>","Title: An object-based approach to implementing distributed concurrency control\nYear: 1991\n\nAbstract: Distributed concurrency control has been implemented by representing in-progress transactions as simulated objects, called transaction objects, that use normal message passing facilities. The implementation of an optimistic mechanism has been completed using transaction objects and a two-phase locking mechanism has been designed. The tradeoffs made and lessons learned, dealing with transactions both on objects and as objects, are discussed","Title: Using tool abstraction to compose systems\nYear: 1992\n\nAbstract: The tool abstraction paradigm, which supports the evolution of large-scale software systems by easing design changes in the system functions, is discussed. Systems that support tool abstraction are structured as a pool of abstract data structures shared by a collection of cooperating 'toolies', where each toolie provides a piece of the overall system function. When one toolie updates the shared data, other toolies must be notified: otherwise, cooperating-but-independent toolies may not execute, and the overall system function may be compromised. The KWIC (key word in context) index production system is used to illustrate the idea of tool abstraction. The relationship of tool abstraction to other concepts is examined.<>","Title: Rule chaining in Marvel: dynamic binding of parameters\nYear: 1992\n\nAbstract: Marvel is a software development environment that uses a rule-based model of the development process. When a user issues a command, Marvel invokes the corresponding rule and binds the rule's formal parameter to the object selected by the user. If firing the rule changes the object in a way that satisfies the conditions of other rules, Marvel automatically fires each of these rules and tries to bind their formal parameters to objects. Marvel must infer which objects to bind to the formal parameters of the rules in the chain. A problem arises when the classes of the objects manipulated by the rules are different, because Marvel has to determine which object to bind to the parameter of each rule in the chain. Recursive data definitions cause chaining between rules that act on different objects in the same class. Methods developed to address this chaining problem are discussed.<>","Title: A bi-level language for software process modeling\nYear: 1993\n\nAbstract: The authors present a multi-user implementation of a bi-level process modeling language (PML). Most process modeling formalisms are well-suited to one of two levels of specification, but not both. Some concentrate on global control flow and synchronization. These languages make it easy to define the broad outline of a process, but harder to refine the process by expressing constraints and policies on individual tools and data. Other process formalisms are inherently local. It is easy to define constraints, but far from straightforward to express control flow. Combining global and local formalisms is proposed to produce bi-level formalisms suitable for expressing the enacting large scale processes. The new PML is called the activity structures language. The activity structures language integrates global constrained expressions with local rules. Its implementation on top of the Marvel rule-based environment is described","Title: Dynamic inference of likely metamorphic properties to support differential testing\nYear: 2015\n\nAbstract: Metamorphic testing is an advanced technique to test programs without a true test oracle such as machine learning applications. Because these programs have no general oracle to identify their correctness, traditional testing techniques such as unit testing may not be helpful for developers to detect potential bugs. This paper presents a novel system, Kabu, which can dynamically infer properties of methods' states in programs that describe the characteristics of a method before and after transforming its input. These Metamorphic Properties (MPs) are pivotal to detecting potential bugs in programs without test oracles, but most previous work relies solely on human effort to identify them and only considers MPs between input parameters and output result (return value) of a program or method. This paper also proposes a testing concept, Metamorphic Differential Testing (MDT). By detecting different sets of MPs between different versions for the same method, Kabu reports potential bugs for human review. We have performed a preliminary evaluation of Kabu by comparing the MPs detected by humans with the MPs detected by Kabu. Our preliminary results are promising: Kabu can find more MPs than human developers, and MDT is effective at detecting function changes in methods.","Title: On effective testing of health care simulation software\nYear: 2011\n\nAbstract: Health care professionals rely on software to simulate anatomical and physiological elements of the human body for purposes of training, prototyping, and decision making. Software can also be used to simulate medical processes and protocols to measure cost effectiveness and resource utilization. Whereas much of the software engineering research into simulation software focuses on validation (determining that the simulation accurately models real-world activity), to date there has been little investigation into the testing of simulation software itself, that is, the ability to effectively search for errors in the implementation. This is particularly challenging because often there is no test oracle to indicate whether the results of the simulation are correct. In this paper, we present an approach to systematically testing simulation software in the absence of test oracles, and evaluate the effectiveness of the technique.","Title: Accelerating Maven by delaying test dependencies\nYear: 2015\n\nAbstract: Modular build systems (such as Maven) may simplify build maintenance, but significantly reduce opportunities for parallelism where they may be most helpful: when running tests. If tests are contained in each module, and modules contain dependencies on each other, their tests can not run in parallel between modules that have build dependencies. This poster will present a technique for achieving significantly greater parallelism in running the tests of Maven-built Java projects, cutting build times in half in our case study.","Title: The weHelp reference architecture for community-driven recommender systems\nYear: 2010\n\nAbstract: Recommender systems have become increasingly popular. Most research on recommender systems has focused on recommendation algorithms. There has been relatively little research, however, in the area of generalized system architectures for recommendation systems. In this paper, we introduce weHelp - a reference architecture for social recommender systems. Our architecture is designed to be application and domain agnostic, but we briefly discuss here how it applies to recommender systems for software engineering.","Title: Dynamic taint tracking for Java with phosphor (demo)\nYear: 2015\n\nAbstract: Dynamic taint tracking is an information flow analysis that can be applied to many areas of testing. Phosphor is the first portable, accurate and performant dynamic taint tracking system for Java. While previous systems for performing general-purpose taint tracking in the JVM required specialized research JVMs, Phosphor works with standard off-the-shelf JVMs (such as Oracle's HotSpot and OpenJDK's IcedTea). Phosphor also differs from previous portable JVM taint tracking systems that were not general purpose (e.g. tracked only tags on Strings and no other type), in that it tracks tags on all variables. We have also made several enhancements to Phosphor, to track taint tags through control flow (in addition to data flow), as well as to track an arbitrary number of relationships between taint tags (rather than be limited to only 32 tags). In this demonstration, we show how developers writing testing tools can benefit from Phosphor, and explain briefly how to interact with it.","Title: Efficient dependency detection for safe Java test acceleration\nYear: 2015\n\nAbstract: Slow builds remain a plague for software developers. The frequency with which code can be built (compiled, tested and packaged) directly impacts the productivity of developers: longer build times mean a longer wait before determining if a change to the application being built was successful. We have discovered that in the case of some languages, such as Java, the majority of build time is spent running tests, where dependencies between individual tests are complicated to discover, making many existing test acceleration techniques unsound to deploy in practice. Without knowledge of which tests are dependent on others, we cannot safely parallelize the execution of the tests, nor can we perform incremental testing (i.e., execute only a subset of an application's tests for each build). The previous techniques for detecting these dependencies did not scale to large test suites: given a test suite that normally ran in two hours, the best-case running scenario for the previous tool would have taken over 422 CPU days to find dependencies between all test methods (and would not soundly find all dependencies) — on the same project the exhaustive technique (to find all dependencies) would have taken over 1e300 years. We present a novel approach to detecting all dependencies between test cases in large projects that can enable safe exploitation of parallelism and test selection with a modest analysis cost.","Title: Identifying functionally similar code in complex codebases\nYear: 2016\n\nAbstract: Identifying similar code in software systems can assist many software engineering tasks such as program understanding and software refactoring. While most approaches focus on identifying code that looks alike, some techniques aim at detecting code that functions alike. Detecting these functional clones — code that functions alike — in object oriented languages remains an open question because of the difficulty in exposing and comparing programs' functionality effectively. We propose a novel technique, In-Vivo Clone Detection, that detects functional clones in arbitrary programs by identifying and mining their inputs and outputs. The key insight is to use existing workloads to execute programs and then measure functional similarities between programs based on their inputs and outputs, which mitigates the problems in object oriented languages reported by prior work. We implement such technique in our system, HitoshiIO, which is open source and freely available. Our experimental results show that HitoshiIO detects more than 800 functional clones across a corpus of 118 projects. In a random sample of the detected clones, HitoshiIO achieves 68+% true positive rate with only 15% false positive rate.","Title: Code relatives: detecting similarly behaving software.\nYear: 2016\n\nAbstract: Detecting “similar code” is useful for many software engineering tasks. Current tools can help detect code with statically similar syntactic and–or semantic features (code clones) and with dynamically similar functional input/output (simions). Unfortunately, some code fragments that behave similarly at the finer granularity of their execution traces may be ignored. In this paper, we propose the term “code relatives” to refer to code with similar execution behavior. We define code relatives and then present DyCLINK, our approach to detecting code relatives within and across codebases. DyCLINK records instruction-level traces from sample executions, organizes the traces into instruction-level dynamic dependence graphs, and employs our specialized subgraph matching algorithm to efficiently compare the executions of candidate code relatives. In our experiments, DyCLINK analyzed 422+ million prospective subgraph matches in only 43 minutes. We compared DyCLINK to one static code clone detector from the community and to our implementation of a dynamic simion detector. The results show that DyCLINK effectively detects code relatives with a reasonable analysis time.","Title: Challenges in Behavioral Code Clone Detection\nYear: 2016\n\nAbstract: When software engineering researchers discuss \"similar\" code, we often mean code determined by static analysis to be textually, syntactically or structurally similar, known as code clones (looks alike). Ideally, we would like to also include code that is behaviorally or functionally similar, even if it looks completely different. The state of the art in detecting these behavioral clones focuses on checking the functional equivalence of the inputs and outputs of code fragments, regardless of its internal behavior (focusing only on input and output states). We argue that with an advance in dynamic code clone detection towards detecting behavioral clones (i.e., those with similar execution behavior), we can greatly increase the applications of behavioral clones as a whole for general program understanding tasks.","Title: Obfuscation resilient search through executable classification.\nYear: 2018\n\nAbstract: Android applications are usually obfuscated before release, making it difficult to analyze them for malware presence or intellectual property violations. Obfuscators might hide the true intent of code by renaming variables and/or modifying program structures. It is challenging to search for executables relevant to an obfuscated application for developers to analyze efficiently. Prior approaches toward obfuscation resilient search have relied on certain structural parts of apps remaining as landmarks, un-touched by obfuscation. For instance, some prior approaches have assumed that the structural relationships between identifiers are not broken by obfuscators; others have assumed that control flow graphs maintain their structures. Both approaches can be easily defeated by a motivated obfuscator. We present a new approach, MACNETO, to search for programs relevant to obfuscated executables leveraging deep learning and principal components on instructions. MACNETO makes few assumptions about the kinds of modifications that an obfuscator might perform. We show that it has high search precision for executables obfuscated by a state-of-the-art obfuscator that changes control flow. Further, we also demonstrate the potential of MACNETO to help developers understand executables, where MACNETO infers keywords (which are from relevant un-obfuscated programs) for obfuscated executables.","Title: Replay without recording of production bugs for service oriented applications.\nYear: 2018\n\nAbstract: Short time-to-localize and time-to-fix for production bugs is extremely important for any 24x7 service-oriented application (SOA). Debugging buggy behavior in deployed applications is hard, as it requires careful reproduction of a similar environment and workload. Prior approaches for automatically reproducing production failures do not scale to large SOA systems. Our key insight is that for many failures in SOA systems (e.g., many semantic and performance bugs), a failure can automatically be reproduced solely by relaying network packets to replicas of suspect services, an insight that we validated through a manual study of 16 real bugs across five different systems. This paper presents Parikshan, an application monitoring framework that leverages user-space virtualization and network proxy technologies to provide a sandbox “debug” environment. In this “debug” environment, developers are free to attach debuggers and analysis tools without impacting performance or correctness of the production environment. In comparison to existing monitoring solutions that can slow down production applications, Parikshan allows application monitoring at significantly lower overhead.","Title: A Case Study on the Impact of Similarity Measure on Information Retrieval based Software Engineering Tasks.\nYear: 2018\n\nAbstract: Information Retrieval (IR) plays a pivotal role in diverse Software Engineering (SE) tasks, e.g., bug localization and triaging, code retrieval, requirements analysis, etc. The choice of similarity measure is the core component of an IR technique. The performance of any IR method critically depends on selecting an appropriate similarity measure for the given application domain. Since different SE tasks operate on different document types like bug reports, software descriptions, source code, etc. that often contain non-standard domain-specific vocabulary, it is essential to understand which similarity measures work best for different SE documents. This paper presents two case studies on the effect of different similarity measure on various SE documents w.r.t. two tasks: (i) project recommendation: finding similar GitHub projects and (ii) bug localization: retrieving buggy source file(s) correspond to a bug report. These tasks contain a diverse combination of textual (i.e. description, readme) and code (i.e. source code, API, import package) artifacts. observe that the performance of IR models varies when applied to different artifact types. find that, in general, the context-aware models achieve better performance on textual artifacts. In contrast, simple keyword-based bag-of-words models perform better on code artifacts. On the other hand, the probabilistic ranking model BM25 performs better on a mixture of text and code artifacts. We further investigate how such an informed choice of similarity measure impacts the performance of SE tools. In particular, we analyze two previously proposed tools for project recommendation and bug localization tasks, which leverage diverse software artifacts, and observe that an informed choice of similarity measure indeed leads to improved performance of the existing SE tools.","Title: Interfacing Oz with the PCTE OMS: A Case Study of Integrating a Legacy System with a Standard Object Management System\nYear: 1999\n\nAbstract: The integration of a legacy system and a standard Object Management System (OMS) is often a very challenging task. This paper details a case study, our experiment in inter facing Oz with the PCTE (Portable Common Tool Environment) Object Management System. Oz is a multi- user process-centered software development environment that has been under development in our lab since 1987, originally under the name Marvel. PCTE is a specification that de fines a language-independent interface providing support mechanisms for software engineering e nviron- ments (SEE). One of the premises of PCTE is that, in theory, an SEE such as Oz can be built (or extended) using the services provided by PCTE. The purpose of our experiment was to study how a legacy system such as Oz can be integrated into a new environment framework, e.g., PCTE. The architecture of the legacy system and the services of the framework are t he key factors in the integration approach. Because Oz historically has included a native OMS, our experiment focused on modifying Oz to use the PCTE OMS, which has an open and standard API. This paper de- scribes how several Oz components were changed to interface to the PCTE OMS. The resulting proof-of-concept hybrid system has process control and integration services provided by Oz, and data integration services provided in part by PCTE. We discuss in depth the solut ions to the con- currency control problems that arise in such an environment, where Oz and PCTE use different approaches to transaction management (i.e., each has its own transaction manager). The PCTE implementation used in our experiment was the Emeraude PCTE V 12.5.1, and the Oz version was V 1.0.1."],"string":"[\n \"Title: Process Evolution in the Marvel Environment.\\nYear: 1993\\n\\nAbstract: We present a schema and process evolution tool, called the Evolver, for the M ARVEL process-centered environment. The Evolver analyzes the differences between the new and installed process models of an existing environment, detecting each case where the notion of consistency defined by the process model has been strengthened or weakened. The Evolver then automatically updates the environment's objectbase to guarantee that the objects are consistent according to the new specifications. The Evolver can be applied while the installed process is in progress, temporarily halting normal operation while it updates the objectbase, after which development continues using the new process. We have had several months of experience using the Evolver to make repeated changes in the process that supports our own further development of MARVEL, and include in this paper one small but practical example of a recent change made to a real MARVEL process.\",\n \"Title: Smarter recompilation\\nYear: 1988\\n\\nAbstract: Tichy's Smart Recompilation method can be made smarter by permitting benign type inconsistencies between separately compiled modules. This enhanced method helps the programmer to make far-reaching changes in small, manageable steps.\",\n \"Title: Experience with Marvel\\nYear: 1989\\n\\nAbstract: MARVEL is a kernel for software development environments that enacts a processing modeling notation, called the MARVEL Strategy Language (MSL). The software process is defined in MSL as a collection of modular strategies, where different (sets of these) strategies are appropriate for different tool suites, user roles, phases of the lifecycle, management policies, etc. Each strategy specifies a partial object-oriented data model and a partial rule-based process model. The strategies are supplemented by a collection of envelopes, each of which describes how to apply one or more of the external tools used during software development activities; this includes the mapping from the inputs/outputs of these tools and the MARVEL objectbase and the command lines to invoke these tools. Multiple MSL strategies are merged to obtain the full data and process models for enaction by the MARVEL kernel.Each software development activity is described in MSL by a rule specifying the precondition and multiple postconditions of the activity. Both preconditions and postconditions are written in a subset of the first order predicate logic, stylized to incorporate directives for efficient queries and updates. A precondition is a logical statement that must be true before the activity is initiated; a postcondition is a logical statement asserted to be true when the activity completes. Multiple mutually exclusive postconditions reflect the multiple possible results of many activities. An activity is typically a particular combination of software development tool, options specific to that tool, and typed arguments. The notion of activity could be extended to include loading a new set of strategies to take over at this point in the process, to support both nesting of hierarchical activities and successive activity phases.Enaction of an MSL process model is implemented by what we call opportunistic processing, consisting primarily of forward and backward chaining among the rules. When a user initiates an activity, MARVEL applies backward chaining to attempt to satisfy its precondition. If this is not possible, MARVEL could use the rules in the style of an expert system to explain the problem. Once an activity has completed, MARVEL asserts one of its postconditions and uses forward chaining to automatically carry out other activities whose results are presumed to be needed in the near future.We have designed two facilities for controlling this processing, to prevent runaway automation. Hints are additional MSL preconditions that affect automatic but not user-initiated invocation of tools. Implicit queries represent a reflective capacity to keep tabs on estimated costs of automatic behavior and request user confirmation when certain user-selected thresholds are reached.MARVEL version 2.10 was recently completed, an upgrade over the version 2.01 that was demonstrated at the 3rd PSDE conference in November 1988. This latest version of MARVEL adds an X11 windows graphical user interface and new subsystems called Marvelizer and Organ. Marvelizer is a tool for immigrating existing software systems from the Unix file system or (a restricted class of) existing software development environment into a MARVEL objectbase. Organ is a set of commands for reorganizing MARVEL objectbases as the project and/or system structure evolves. Marvelizer has been used to immigrate Marvel itself, including about 35,000 lines of C code, into a C/Marvel objectbase. We would like to use the C/Marvel set of MSL strategies, which support activities using the user's favorite editor, lint, cc, ar, ranlib, less (and soon rcs, dbx and mail), for our own further development of MARVEL. Both MARVEL and C/Marvel were in use during the Spring 1989 semester by students in my E6123y Programming Environments and Software Tools course (admittedly a captive audience).We have discovered several difficulties with the MARVEL paradigm that seem likely to apply at least partially to enaction of other notations for software process modeling. First, constructing and debugging a process model is hard. We have run into many unanticipated interactions among rules, particularly among those defined in different strategies that happen to be loaded into the MARVEL kernel at the same time. We are working on debugging facilities that will uncover all direct and transitive dependencies by constructing a graphical dataflow graph of the currently loaded rules. It is feasible for a user to enable and disable individual rules, both during debugging to see how that affects the dataflow among rules and also during operation because certain behavior is not currently desired. This would be dangerous, however, if done by a naive user: if rules were turned off and on at arbitrary points, it would be possible for MARVEL to get into what is an inconsistent state with respect to the full set of rules in such a way that automatic recovery is impossible.Similar difficulties seem likely to arise for any enactable process modeling paradigm. The following questions need to be answered: How does an implementor go about writing, testing and debugging a process model? For non-trivial processes, the model is likely to be very large and hence should be modularized. What is the appropriate style of interface among process model modules and how is it enforced? Is the process model visible to the users and, if so, how is it presented? If the users can modify the process model during enaction, how does this affect continuing the process? Should (can) consistency be enforced or should (can) inconsistencies be detected and repaired?Second, constructing envelopes is even harder. We currently use the Shell languages (e.g., CShell) as notations for defining envelopes. Writing the Shell scripts requires an intimate knowledge of both the tool and the mapping of the MARVEL objectbase onto the file system. Although there is no way to avoid the former, we hope to minimize the latter by developing a higher level envelope language that understands the MARVEL objectbase implementation so the envelope writer doesn't have to. However, this is currently a relatively low priority due to lack of manpower.Again, similar difficulties seem inevitable for other enaction mechanisms intended to work with external tools. The questions: Is it possible (or desirable) to support existing tools, particularly tools where retrofitting is impractical? If only new tools are supported, can a relatively simple standard interface be developed and published so these tools can be developed separately by vendors? Would such an interface pre-empt too many design decisions, or would it be feasible for vendors to upgrade their current products?Third, supporting multiple users is hardest of all. The current MARVEL kernel does not provide any facilities for concurrency control. It would be easy to add locking of the entire objectbase, but a more flexible transaction model is needed. We are working on a number of alternative models that we think are promising for incorporation into the MARVEL kernel. Commit-serializability introduces two new transaction operations, split-transaction and join-transaction, that manipulate in-progress transactions to support coordination among multiple users. Participant transactions (AKA participation domains) distinguishes participants in group software development tasks from observers, and relaxes serializability as viewed by participants to permit cooperation.It is essential for enactable process models to support coordination and cooperation among the members of software development teams. There are two sets of questions: First, what are appropriate processes for programming-in-the-many? This is itself an active area of research, and there is certainly no consensus. One opinion, voiced at the previous workshop, is that the community should not be enacting process models until we understand the processes we are modeling. An opposing opinion is that enaction gives the community the capability to experiment with different processes without prematurely making a full scale commitment to a particular process. Assuming agreement with the second view, what process modeling notations and enaction mechanisms can support a reasonable range of cooperative work facilities?The database and operating system communities have agreed on atomic and serializable transactions, which can be implemented using a wide variety of technologies, as the basis for data consistency, failure recovery and concurrency control across their traditional application areas. Can the software engineering community develop an equally all-encompassing “transaction” model? Or is there some characteristic of the software development process such that coordination and cooperation policies will always require special support on a case-by-case basis? If so, what does this say about enaction of process models in general?In summary, MARVEL is implemented and in limited use, it works in some cases, it doesn't work in others, it raises a lot of questions, and both we the MARVEL group and we the software engineering community have a lot more work to do.\",\n \"Title: Mechanisms (session summary)\\nYear: 1989\\n\\nAbstract: Takuya Katayama lead the discussion on Mechanisms for Software Process Description. He classified software process views as reflected in the position papers into functional, behavioral and enactional, described some generally desirable properties of software process mechanisms, and listed those projects that seemed to have obtained non-toy experience.He made a plea for each of these projects to answer a questionnaire that, among other things, requested information regarding how the formalism handled ill-behaved processes. One generic example of an ill-behaved process was given: repeat function1, …, functionn until predicate1, …, predicatem are satisfied. It is not really clear to me that this process is ill-defined, since it seems to be computable using normal means; my opinion is that an ill-defined process is one that is made up and/or arbitrarily changed while it is in progress, in the sense of self-modifying code (e.g., Lee Osterweil's universal process program — an unfair characterization since Lee didn't show up to defend himself). In any case, Katayama's questionnaire will hopefully be addressed in the position papers for 6ISPW.The discussion focused on several interesting issues. The most controversial was instigated by Mark Dowson, when he asked what is the definition of “activity”. There were a number of different responses, but most fell into one of three camps.The first, which I'll call notation-oriented, defined an activity as the primitive (according to Watts Humphrey, the instruction set) of the process modeling notation. In general, the notation might lack facilities to express decomposition of an activity, or at least — as Colin Tully states — we choose to regard such units as decomposable. The second, people-oriented, defined an activity as a primitive in the real-world process being modeled. Bill Curtis said this might be a job that one person takes responsibility for, without interaction with other people; the process would be defined in terms of such interactions. The third, which I'll call process-oriented, was exemplified by Dewayne Perry's definition of an activity in terms of causes (preconditions?) and goals (postconditions?) with respect to the process. There was some argument over whether or not such an activity necessarily resulted in an observable change of state. Bob Balzer pointed out it is an open question as to whether syntax (notation) can realistically be separated from semantics (people and process).There was some argument in all three camps as to whether activities were really primitive, as assumed in the definitions above, or could be decomposable at a lower level of abstraction. One issue is whether or not a particular activity can be decomposed with respect to the question at hand (David Carr), for example, the encapsulation might be due to ignorance (Sam Redwine); alternatively, there may be no need for decomposition (Frank Belz).One response by Mark Dowson seems to cut across all three camps: tasks are defined as managed units (i.e., the units are scheduled and allocated resources), while activities are not managed separately from their enclosing tasks. But this leads to the distinction pointed out by Peter Feiler of machine-level scheduling and resource allocation versus process-level, since everything executed on a computer is scheduled and allocated resources at a fine granularity by the operating system. Taly Minsky expressed a concern that this entire discussion was at too high a level of abstraction, and we should return to more operational definitions; however, I think that discussion of scheduling and resource allocation is quite concrete, since we can define (computer or bureaucratic) algorithms to carry out these operations.At this juncture, Anthony Finkelstein pointed out the need for folk classification: what do the people involved in a process think are the meaningful units? This lead into a new controversy over the notion of “commitments”, which seemed to have at least as many definitions as “activities”. Here the division seemed to be over whether the most important concern was why things are being done versus what is being done; there was also an issue over what it meant for commitments to be retracted. (The conversation was rather random, so I can relay only a small subset of the interesting comments made by particular individuals.)Bill Curtis pointed out that computerized formalization and monitoring of commitments may appear to project personnel as Naziware; further, if there is a shared information base for rationale capture, who monitors and cleans it up? Sam Redwine gave the example of burying a project to track the fulfillment or lack thereof of presidential campaign promises. Walt Scacchi made a distinction between strong commitments, with real consequences if broken, and weak commitments, which are no big deal; Bill Curtis mentioned false commitments, based on what is acceptable to management or customers rather than how things are really done.According to Anthony Finkelstein, there are four kinds of commitment: commitment to a statement, no commitment to a statement, commitment to the negation of a statement, no commitment at all. Taly Minsky mentioned the commitment (or obligation) to restore inconsistent states to consistency; Ian Thomas pointed out such a commitment might be from an individual to himself, and not necessarily public.There was a digression and some brief confusion over the use of the term commitment as it relates to transactions, but that was determined to be irrelevant to the discussion at hand. Watts Humphrey defined a commitment as between people, freely entered into, based on intelligent knowledge, public, with a clear intention of performing, and prior notification and renegotiation if the commitment will not be met. In this sense, project personnel themselves act as two agents with respect to commitments, one that makes commitments while the other monitors whether the commitments are being fulfilled (Mark Dowson).After the break, two specific projects were briefly discussed.Karen Huff described her work on rule-based process modeling and planning. She made three main points: (1) components should be defined to permit dynamic construction of processes, in order to accommodate failure and iteration, particularly for ill-defined processes; (2) there is an important distinction between doing versus deciding what to do, i.e., it should be possible to reason about a course of action without committing to start its execution (Bob Balzer describes this as envisionment versus enaction); and (3) the process should be viewed in terms of its state of completion with respect to its goals rather than as an instruction pointer and a snapshot. She noted that the ability to decide was built-in, as opposed to a separate process.This led quite predictably to a discussion about planning (Bob Balzer) versus doing (Pamela Zave) versus planning in response to feedback from doing and measurement of progress (Sam Redwine). There were a number of questions, mostly from Dennis Heimbigner and Bob Balzer, regarding the power of Huff's notation (apparently Turing-complete in theory, but not fully implemented), its application to representing real processes (primarily Unix-based coding and testing), and the difficulties encountered (there is a problem with suspending and then resuming a plan after the world has changed). I mentioned the concern about side-effects from what might otherwise be considered an atomic activity, and how such side-effects could be detected (the best we can do is operating system support — which of course doesn't work for off-line activities) and modeled. Wilhelm Schaefer delineated the problem of combining process models for individuals, who can change their processes at will.Going on to the next brief project description, Walt Scacchi described his work on a hypertext environment, where an empirical evaluation was made with respect to student groups. The process was defined at the intergroup but not intragroup level, and there was no control over the work division among participants. Not surprisingly the unexpected circumstances arose at the intragroup level, which turned out to be decomposable after all and led to the greatest variation in productivity and quality. Individuals may introduce new tools spontaneously, and there's no general way to describe this. There is thus a need to produce new processes and apply them on the fly, as well a model interactions among multiple agents whose purpose may be to decide what to do next; since people meander, processes need to be able to meander. People operate in an open world situation, but the process is defined by a closed world formulation.Watts Humphrey pointed out that processes don't transplant well; outside processes are not accepted and a sense of ownership (achieved through improving the process) is needed. Marc Kellner brought up the issue of where do processes come from, e.g., DoD guidelines. This led to a discussion of description versus prescriptive processes (Marc Kellner) and abstraction versus instantiation of processes (Bob Balzer). Anthony Finkelstein questioned the idea of an emergent process. Walt Scacchi responded that modeling technology is too heavy to support modeling on the fly, and the effort of changing the model but be less than just doing the work correctly.At the end of the session, Colin Tully suggested holding a process description contest, with a benchmark process that all participants would define in their own process modeling notations. Anthony Finkelstein pointed out that this is problematical, however, since different notations address widely differing aspects of the process. It might be most reasonable to have at least two benchmarks, one involving upstream and the other downstream activities, with no expectation that every process modeling notation addresses both areas. Hopefully this will be addressed in the call for 6ISPW.\",\n \"Title: Kinesthetics eXtreme: An External Infrastructure for Monitoring Distributed Legacy Systems\\nYear: 2003\\n\\nAbstract: Autonomic computing - self-configuring, self-healing, self-optimizing applications, systems and networks - is widely believed to be a promising solution to ever- increasing system complexity and the spiraling costs of human system management as systems scale to global proportions. Most results to date, however, suggest ways to architect new software constructed from the ground up as autonomic systems, whereas in the real world organizations continue to use stovepipe legacy systems and/or build \\\"systems of systems\\\" that draw from a gamut of new and legacy components involving disparate technologies from numerous vendors. Our goal is to retrofit autonomic computing onto such systems, externally, without any need to understand or modify the code, and in many cases even when it is impossible to recompile. We present a meta-architecture implemented as active middleware infrastructure to explicitly add autonomic services via an attached feedback loop that provides continual monitoring and, as needed, reconfiguration and/or repair. Our lightweight design and separation of concerns enables easy adoption of individual components, as well as the full infrastructure, for use with a large variety of legacy, new systems, and systems of systems. We summarize several experiments spanning multiple domains.\",\n \"Title: Database Support for Knowledge-Based Engineering Environments\\nYear: 1988\\n\\nAbstract: The authors discuss Marvel, a knowledge-based engineering environment that can be instantiated with the entities and tools for a specific engineering project, together with rules regulating (technical) conduct of the project. The resulting environment maintains database representations of entities and carries out menial chores through controlled automation. The authors concentrate on the details of MSL (Marvel Strategy Language) and tool envelopes, including the algorithms used in the various parts of the Marvel system. These algorithms are different from those used for other expert systems because of Marvel's requirement for large-scale entities, the ability to load and unload strategies with different behavior and views of the database structure, and the overhead of tool processing effectively prohibiting 'undo' during backtracking.\",\n \"Title: Marvel strategy language example\\nYear: 1989\\n\\nAbstract: The DocPrep environment was developed by two students, Laura Johnson and Victor Kan, as their term project in the Spring 1989 E6123y Programming Environments and Software Tools course. Neither student had any prior knowledge of or experience with marvel, and were not familiar with its internal implementation. They successfully used DocPrep to produce their final project report. The twelve envelopes, bind, delete-sect-order, format, printdoc, printsec, review-format-err, review-spell-err, runeditor, specify-header, specify-printer, specify-sect-order and spell-check, are omitted.\",\n \"Title: MELDing transactions and objects\\nYear: 1991\\n\\nAbstract: MELD is an experimental object system under development at Columbia University. The object-oriented programming language supports classes, strong typing of instance variables, active values, multiple inheritance, and separate compilation of modular units called features that bundle together related classes and objects. These facilities were developed for an early version of MELD [3, 4], without persistence, concurrency or distribution.\",\n \"Title: An object model for shared data\\nYear: 1990\\n\\nAbstract: The classical object model supports private data within objects and clean interfaces among objects, and, by definition, does not permit sharing of data among arbitrary objects. This is a problem for certain real-world applications, where the same data logically belongs to multiple objects and may be distributed over multiple nodes on the network. Rather than give up the advantages of encapsulated objects in modeling real-world entities, a new object model that supports distribution of computation units from information-hiding concerns is proposed. A new object model is introduced, a motivating example from the financial services domain is described, and a new language based on the model is presented\",\n \"Title: A Flexible Rule-Chaining Engine for Process-Based Software Engineering\\nYear: 1994\\n\\nAbstract: We present the design of a new rule-based processengine that generalizes previous systems to supportprocess enforcement, automation, guidance, monitoring,delegation, planning, simulation, instrumentationand potentially other applications. Our approach isfully knowledge-based, tailored by knowledge regardingthe process assistance policies to be supported as wellas the process definition.1 IntroductionA process is a set of steps for developing a softwaresystem. Different processes are ...\",\n \"Title: Federating Process-Centered Environments: The Oz Experience\\nYear: 1998\\n\\nAbstract: We describe two models for federating process-centeredenvironments (PCEs): homogeneous federation among distinctinstances of the same environment framework enacting the same ordifferent process models, and heterogeneous federation amongdiverse process enactment systems. We identify the requirements andconsider possible architectures for each model, although weconcentrate primarily on the homogeneous case. The bulk of the paperpresents our choice of architecture, and correspondinginfrastructure, for homogeneous federation among MARVEL environment instances as realized in the OZ system. We brieflyconsider how a single MARVEL environment, or an OZfederation of MARVEL environments, might be integrated into aheterogeneous federation based on ProcessWall‘s facilities forinteroperating PCEs.\",\n \"Title: A Marvelous Extended Transaction Processing Model\\nYear: 1989\\n\\nAbstract: A key flaw in previous software development environment research is the lack of a transaction model supporting fault tolerance, concurrency control, consistent publication of changes and user control of commit and abort for software development activities. The atomicity properties of the classical transaction model make it unsuitable for industrial software development efforts. We sketch an extended transaction model with a commit-serializability semantics and describe the application of this model to the existing M ARVEL architecture for software development environments. MARVEL supports rule-based automation of software development activities and integrates commercial off-the-shelf tools.\",\n \"Title: Experiments with rule based process modelling in an SDE\\nYear: 1989\\n\\nAbstract: Without Abstract\",\n \"Title: An architecture for WWW-based hypercode environments\\nYear: 1997\\n\\nAbstract: A hypercode software engineering environment represents all plausible multimedia artifacts concerned with software development and evolution that can be placed or generated on-line, from source code to formal documentation to digital library resources to informal email and chat transcripts. A hypercode environment supports both internal (hypertext) and external (link server) links among these artifacts, which can be added incrementally as useful connections are discovered; project-specific hypermedia search and browsing; automated construction of artifacts and hyperlinks according the software process; application of tools to the artifacts according to the process workflow; and collaborative work for geographically dispersed teams. We present a general architecture for what we call hypermedia subwebs, and groupspace services operating on shared subwebs, based on World Wide Web technology - which could be applied over the Internet or within an intranet. We describe our realization in OzWeb.\",\n \"Title: An architecture for integrating OODBs with WWW\\nYear: 1996\\n\\nAbstract: The main topic of this paper is how to structure information so that the view of the web, both within and across web pages, is dynamically customizable. We present an architecture that integrates Object Oriented Databases with the World Wide Web to organize such dynamic structures. Different users, or the same user at different times, could have different views of the web. We discuss several architectural variants and and implementation issues. Our chosen architecture provides high flexibility for a wide variety of applications, ranging from managed healthcare to software development environments, and has been realized in the dkweb system.\",\n \"Title: Report from the second international workshop on games and software engineering (GAS 2012)\\nYear: 2013\\n\\nAbstract: The Second International Workshop on Games and Software Engineering (GAS 2012) was held in June 2012, collocated with ICSE 2012 in Zurich, Switzerland. GAS 2012 explores issues that crosscut the software engineering and the game engineering communities. Advances in game engineering techniques can be adopted by the software engineering community to develop more engaging applications across diverse domains: education; healthcare; fitness; sustainable activities (e.g., recycling awareness); and so on. Successful computer games feature a property that is not always found in traditional software: they are highly engaging. Engaging games enthrall players and result in users willing to spend increasing amounts of time and money playing them. Nine submissions were accepted for the workshop, which also featured a panel session and an informal poster session.\",\n \"Title: VMVM: unit test virtualization for Java\\nYear: 2014\\n\\nAbstract: As software evolves and grows, its regression test suites tend to grow as well. When these test suites become too large, they can eventually reach a point where they become too length to regularly execute. Previous work in Test Suite Minimization has reduced the number of tests in such suites by attempting to identify those that are redundant (e.g. by a coverage metric). Our approach to ameliorating the runtime of these large test suites is complementary, instead focusing on reducing the overhead of running each test, an approach that we call Unit Test Virtualization. This Tool Demonstration presents our implementation of Unit Test Virtualization, VMVM (pronounced \\\\\\\"vroom-vroom\\\\\\\") and summarizes an evaluation of our implementation on 20 real-world Java applications, showing that it reduces test suite execution time by up to 97% (on average, 62%). A companion video to this demonstration is available online, at https://www.youtube.com/watch?v=sRpqF3rJERI.\",\n \"Title: Automating Content Extraction of HTML Documents\\nYear: 2005\\n\\nAbstract: Web pages often contain clutter (such as unnecessary images and extraneous links) around the body of an article that distracts a user from actual content. Extraction of \\\"useful and relevant\\\" content from web pages has many applications, including cell phone and PDA browsing, speech rendering for the visually impaired, and text summarization. Most approaches to making content more readable involve changing font size or removing HTML and data components such as images, which takes away from a webpage's inherent look and feel. Unlike \\\"Content Reformatting,\\\" which aims to reproduce the entire webpage in a more convenient form, our solution directly addresses \\\"Content Extraction.\\\" We have developed a framework that employs an easily extensible set of techniques. It incorporates advantages of previous work on content extraction. Our key insight is to work with DOM trees, a W3C specified interface that allows programs to dynamically access document structure, rather than with raw HTML markup. We have implemented our approach in a publicly available Web proxy to extract content from HTML web pages. This proxy can be used both centrally, administered for groups of users, as well as by individuals for personal browsers. We have also, after receiving feedback from users about the proxy, created a revised version with improved performance and accessibility in mind.\",\n \"Title: COMPASS: A Community-driven Parallelization Advisor for Sequential Software\\nYear: 2009\\n\\nAbstract: The widespread adoption of multicores has renewed the emphasis on the use of parallelism to improve performance. The present and growing diversity in hardware architectures and software environments, however, continues to pose difficulties in the effective use of parallelism thus delaying a quick and smooth transition to the concurrency era. In this paper, we describe the research being conducted at Columbia University on a system called COMPASS that aims to simplify this transition by providing advice to programmers while they reengineer their code for parallelism. The advice proffered to the programmer is based on the wisdom collected from programmers who have already parallelized some similar code. The utility of COMPASS rests, not only on its ability to collect the wisdom unintrusively but also on its ability to automatically seek, find and synthesize this wisdom into advice that is tailored to the task at hand, i.e., the code the user is considering parallelizing and the environment in which the optimized program is planned to execute. COMPASS provides a platform and an extensible framework for sharing human expertise about code parallelization - widely, and on diverse hardware and software. By leveraging the “wisdom of crowds” model [30], which has been conjectured to scale exponentially and which has successfully worked for wikis, COMPASS aims to enable rapid propagation of knowledge about code parallelization in the context of the actual parallelization reengineering, and thus continue to extend the benefits of Moores law scaling to science and society.\",\n \"Title: Parallel and distributed incremental attribute evaluation algorithms for multiuser software development environments\\nYear: 1993\\n\\nAbstract: The problem of change propagation in multiuser software development environments distributed across a local-area network is addressed. The program is modeled as an attributed parse tree segmented among multiple user processes and changes are modeled as subtree replacements requested asynchronously by individual users. Change propagation is then implemented using decentralized incremental evaluation of an attribute grammar that defines the static semantic properties of the programming language. Building up to our primary result, we first present algorithms that support parallel evaluation on a centralized tree in response to single edits using a singe editing cursor and multiple edits with multiple editing cursors. Then we present our algorithm for parallel evaluation on a decentralized tree. We also present a protocol to guarantee reliability of the evaluation algorithm as components of the decentralized tree become unavailable due to failures and return to availability.\",\n \"Title: A configuration process for a distributed software development environment\\nYear: 1994\\n\\nAbstract: The authors describe work-in-progress on a configuration facility for a multi-site software development environment. The environment supports collaboration among geographically-dispersed teams of software developers. Addition and deletion of local subenvironment sites to a global environment is performed interactively inside any one of the existing local subenvironments, with the same user interface normally employed for invoking software development tools. This registration process is defined and executed using the same notation and mechanisms, respectively, as for the software development process. Everything described by the authors has been implemented and is working, but since they are in the midst of experimentation, they do not expect that the “final” system will be exactly as described\",\n \"Title: Genspace: Exploring Social Networking Metaphors For Knowledge Sharing And Scientific Collaborative Work\\nYear: 2008\\n\\nAbstract: Many collaborative applications, especially in scientific research, focus only on the sharing of tools or the sharing of data. We seek to introduce an approach to scientific collaboration that is based on the sharing of knowledge. We do this by automatically building organizational memory and enabling knowledge sharing by observing what users do with a particular tool or set of tools in the domain, through the addition of activity and usage monitoring facilities to standalone applications. Once this knowledge has been gathered, we apply social networking models to provide collaborative features to users, such as suggestions on tools to use, and automatically-generated sequences of actions based on past usage amongst the members of a social network or the entire community. In this work, we investigate social networking models as an approach to scientific knowledge sharing, and present an implementation called genSpace, which is built as an extension to the geWorkbench. platform for computational biologists. Last, we discuss the approach from the viewpoint of social software engineering.\",\n \"Title: An Approach to Software Testing of Machine Learning Applications\\nYear: 2007\\n\\nAbstract: Some machine learning applications are intended to learn properties of data sets where the correct answers are not already known to human users. It is challenging to test such ML software, because there is no reliable test oracle. We describe a software testing approach aimed at addressing this problem. We present our findings from testing implementations of two different ML ranking algorithms: Support Vector Machines and MartiRank.\",\n \"Title: An Architectural Survey of Object Management Systems\\nYear: 1992\\n\\nAbstract: Much work has been done in the last decade in the related areas of object-oriented programming languages and object-oriented databases. Researchers from both areas now seem to be working toward a common end, that of an object management system, or OMS. An OMS is constructed similarly to an OODB but provides a general purpose concurrent object-oriented programming language as well, complementing the OODB query facilities. In this paper, we will define several different types of object systems (object servers, persistent OOPL's, OODB's and OMS's) in terms of their interfaces and capabilities from the viewpoint of how these support the requirements of cooperative information systems. We will examine the distinguishing features and general architecture of systems of each type in the light of a general model of OMS architecture.\",\n \"Title: A competitive-collaborative approach for introducing software engineering in a CS2 class\\nYear: 2013\\n\\nAbstract: Introductory Computer Science (CS) classes are typically competitive in nature. The cutthroat nature of these classes comes from students attempting to get as high a grade as possible, which may or may not correlate with actual learning. Further, there is very little collaboration allowed in most introductory CS classes. Most assignments are completed individually since many educators feel that students learn the most, especially in introductory classes, by working alone. In addition to completing “normal” individual assignments, which have many benefits, we wanted to expose students to collaboration early (via, for example, team projects). In this paper, we describe how we leveraged competition and collaboration in a CS2 class to help students learn aspects of computer science better - in this case, good software design and software testing - and summarize student feedback.\",\n \"Title: A Flexible Transaction Model for Software Engineering\\nYear: 1990\\n\\nAbstract: It is generally recognized that the classical transaction model, providing atomicity and serializability, is too strong for certain application areas since it unnecessarily restricts concurrency. The author is concerned with supporting cooperative work in multiuser design environments, particularly teams of programmers cooperating to develop and maintain software systems. An extended transaction model that meets the special requirements of software engineering projects is presented, possible implementation techniques are described, and a number of issues regarding the incorporation of such a model into multiuser software development environments are discussed\",\n \"Title: A Workgroup Model for Smart Pushing and Pulling\\nYear: 1999\\n\\nAbstract: Our Workgroup Cache system operates as a virtual intranet, introducing a shared cache to members of the same workgroup. Users may be members of multiple workgroups at the same time. Criteria are associated with each workgroup to pull documents from an individual cache to the shared cache, or push from the shared cache to an individual cache. These criteria provide semantics of the workgroup' s tasks and interests to reduce latency for its members.\",\n \"Title: MELDing data flow and object-oriented programming\\nYear: 1987\\n\\nAbstract: MELD combines concepts from data flow and object-oriented programming languages in a unique approach to tool reusability. MELD provides three units of abstraction — equations, classes and features — that together allow sufficient options for granularity and encapsulation to support the implementation of reusable tools and the composition of existing tools in parallel (i.e., interleaved) as well as in series.\",\n \"Title: A framework for immigrating existing software to new software development environments\\nYear: 1991\\n\\nAbstract: A discussion is given on the problem of immigrating software artifacts from one software development environment (SDE) to another, for the purpose of upgrading to new SDEs as technology improves, while continuing development or maintenance of existing software systems. The authors taxonomise the larger problem of data migration, to establish the scope of immigration. They classify SDEs in terms of the ease of immigrating software artifacts from the data repository of the source SDE, without knowledge of its internal representation. A framework is presented for constructing automatic immigration tools as utilities provided by destination SDEs. They describe a specific immigration tool, called Marvelizer, implemented as part of the Marvel SDE and discuss experience using the tool\",\n \"Title: Generation of run-time environments\\nYear: 1986\\n\\nAbstract: Attribute grammars have been used for many years for automated compiler construction. Attribute grammars support the description of semantic analysis, code generation and some code optimization in a formal declarative style. Other tools support the automation of lexical analysis and parsing. However, there is one large part of compiler construction that is missing from our toolkit: run-time environments. This paper introduces an extension of attribute grammars that supports the generation of run-time environments. The extension also supports the generation of interpreters, symbolic debugging tools, and other execution-time facilities./\",\n \"Title: Intelligent Assistance without Artificial Intelligence\\nYear: 1987\\n\\nAbstract: SMILE is a multi-user software engineering environment that behaves as an intelligent assistant. SMILE presents a 'fileless environment', derives and transforms data to shelter users from entering redundant information, automatically invokes programming tools, and actively participates in the software development and maintenance process. Unlike other intelligent assistants, SMILE is not a rule-based environment: its knowledge of software objects and the programming process is hardcoded into the environment. We describe S MILE's functionality and explain how we achieved this functionality without reliance on artificial intelligence technology.\",\n \"Title: Incremental parsing without a parser\\nYear: 1985\\n\\nAbstract: This article describes an algorithm for incremental parsing of expressions in the context of syntax-directed editors for programming languages. Since a syntax-directed editor represents programs as trees and statements and expressions as nodes in trees, making minor modifications in an expression can be difficult. Consider, for example, changing a '' + '' operator to a ''*'' operator or adding a short subexpression at a syntactically but not structurally correct position, such as inserting '') * (d'' at the # mark in'' (a + b # + c)''. To make these changes in a typical syntax-directed editor, the user must understand the tree structure and type a number of tree-oriented construction and manipulation commands. This article describes an algorithm that allows the user to think in terms of the syntax of the expression as it is displayed on the screen (in infix notation) rather than in terms of its internal representation (which is effectively prefix), while maintaining the benefits of syntax-directed editing. This algorithm is significantly different from other incremental parsing algorithms in that it does not involve modifications to a traditional parsing algorithm or the overhead of maintaining a parser stack or any data structure other than the syntax tree. Instead, the algorithm applies tree transformations, in real-time as each token is inserted or deleted, to maintain a correct syntax tree.\",\n \"Title: Multiple concurrency control policies in an object-oriented programming system\\nYear: 1990\\n\\nAbstract: Object-oriented systems have been generally recognized as a good choice for approaching a range of parallel and distributed programming problems. But different parallel and distributed applications have different consistency models, so multiple concurrency control policies are needed. When objects are shared among applications with different policies, multiple policies must operate simultaneously and compatibly. We have investigated two consistency models, atomic blocks and classical serializable transactions, and the interactions both between these models and with ''uncontrolled'' concurrency, in the context of the MELD distributed object- oriented programming language. We present the relevant language constructs and discuss certain programming problems and implementation issues.\",\n \"Title: Scaling Up Rule-Based Software Development Environments\\nYear: 1991\\n\\nAbstract: Rule-based software development environments (RBDEs) model the software development process in terms of rules that encapsulate development activities, and assist in executing the process via forward and backward chaining over the rule base. We investigate the scaling up of RBDEs to support (1) multiple views of the rule base for multiple users and (2) evolution of the rule base over the lifetime of a project. Our approach is based on clarifying two distinct functions of rules and chaining: maintaining consistency and automation. By definition, consistency is mandatory whereas automation is not. Distinguishing the consistency and automation aspects of RBDE assistance mechanisms makes it possible to formalize the range of compatible views and the scope of mechanizable evolution steps. Throughout the paper, we use the MARVEL RBDE as an example application. Appeared in International Journal on Software Engineering & Knowledge Engineering , World Scientific, 2(1):59-78, March 1992.\",\n \"Title: Making progress in cooperative transaction models\\nYear: 1991\\n\\nAbstract: In the classical transaction model, transactions are consistency preserving units: a transaction is made up of a series of actions which, when executed in isolation in a reliable environment, transforms the database from one consistent state to another (2). A classical transaction management system guarantees the appearance of isolation and reliability, even though multiple transactions execute concurrently and hardware and software components fail. It does this by enforcing atomicity and serializability: atomicity means that either an entire transaction apparently executes to completion or not at all, while serializability means that the effects of the transactions are viewed as if the transactions had executed in some serial order, one completing before the next begins. This is accomplished by considering the objects read and written by concurrently executing transactions, and ensuring that either all updates are completed or none are, and that the read and write dependencies among the set of transactions correspond to some serial order of the transactions. There have been many proposals for extending the transaction model from its original data processing applications to software development, CAD/CAM and other forms of cooperative work. The notion of ''transaction'' is intuitively appealing in these domains, since forward progress often depends on making a related set of changes to a program, design or document in such a way that it is transformed from one consistent state to another. As with data processing applications, ''consistency'' depends on the requirements of the domain, such as a new system configuration passing regression and acceptance tests before the changed modules can be committed.\",\n \"Title: A Mobile Agent Approach to Process-Based Dynamic Adaptation of Complex Software Systems\\nYear: 2001\\n\\nAbstract: We describe an approach based upon software process technology to on-the-fly monitoring, redeployment, reconfiguration, and in general dynamic adaptation of distributed software applications. We choose the term dynamic adaptation to refer to modifications in structure and behavior that can be made to individual components, as well as sets thereof, or the overall target system configuration, such as adding, removing or substituting components, while the system is running and without bringing it down. The goal of dynamic adaptation is manifold: supporting run-time software composition, enforcing adherence to requirements, ensuring uptime and quality of service of mission-critical systems, recovering from and preventing faults, seamless system upgrading, etc. Our approach involves dispatching and coordinating software agents - named Worklets - via a process engine, since successful dynamic adaptation of a complex distributed software system often requires the concerted action of multiple agents on multiple components. The dynamic adaptation process must incorporate and decide upon knowledge about the specifications and architecture of the target software, as well as Worklets capabilities. Dynamic adaptation is correlated to a variety of other software processes - such as configuration management, deployment, validation and evolution - and allows addressing at run time a number of related concerns that are normally dealt with only at development time.\",\n \"Title: Dynamic reconfiguration in an object-based programming language with distributed shared data.\\nYear: 1991\\n\\nAbstract: significant to the financial analysts' strategies. Thus, On-line distributed applications generally allow conventional approaches such as interpretation and dynamic storage allocation. We present a scheme for In a previous paper (8), we introduced a distributed special cases of fine grained dynamic reconfiguration object-based programming model that addresses these sufficient for our application domain and show how it can problems. This programming model supports an be used for practical changes. We introduce new application architecture where price changes are language concepts to implement this scheme in the monitored by agents operating on behalf of individual context of an object-based programming language that portfolio managers. The sampling rate of each agent is supports shared data in a distributed environment. specific to the requirements of its portfolio manager, and it notifies its manager of changes at the granularity considered interesting by the manager's financial strategy. This model is suitable for other applications with similar\",\n \"Title: Adaptive internet interactive team video\\nYear: 2005\\n\\nAbstract: The increasing popularity of online courses has highlighted the lack of collaborative tools for student groups. In addition, the introduction of lecture videos into the online curriculum has drawn attention to the disparity in the network resources used by students. We present an e-Learning architecture and adaptation model called AI2TV (Adaptive Internet Interactive Team Video), which allows virtual students, possibly some or all disadvantaged in network resources, to collaboratively view a video in synchrony. AI2TV upholds the invariant that each student will view semantically equivalent content at all times. Video player actions, like play, pause and stop, can be initiated by any student and their results are seen by all the other students. These features allow group members to review a lecture video in tandem, facilitating the learning process. Experimental trials show that AI2TV can successfully synchronize video for distributed students while, at the same time, optimizing the video quality, given fluctuating bandwidth, by adaptively adjusting the quality level for each student.\",\n \"Title: Infuse: a tool for automatically managing and coordinating source changes in large systems\\nYear: 1987\\n\\nAbstract: In current change management tools, the actual changes occur outside the tool. In contrast, Infuse concentrates on the actual change process and provides facilities for both managing and coordinating source changes. Infuse provides facilities for automatically structuring the cooperation among programmers, propagating changes, and determining the consistency of changes, and provides a basis for negotiating the resolution of conflicting changes and for iterating over a set of changes.\",\n \"Title: The CORD Appraoch to Extensible Concurrency Control\\nYear: 1997\\n\\nAbstract: Database management systems (DBMSs) have been increasingly used for advanced application domains, such as software development environments, workflow management systems, computer-aided design and manufacturing, and managed healthcare. In these domains, the standard correctness model of serializability is often too restrictive. We introduce the notion of a Concurrency Control Language (CCL) that allows a database application designer to specify concurrency control policies to tailor the behavior of a transaction manager. A well-crafted set of policies defines an extended transaction model. The necessary semantic information required by the CCL run-time engine is extracted from a task manager, a (logical) module by definition included in all advanced applications. This module stores task models that encode the semantic information about the transactions submitted to the DBMS. We have designed a rule-based CCL, called CORD, and have implemented a run-time engine that can be hooked to a conventional transaction manager to implement the sophisticated concurrency control required by advanced database applications. We present an architecture for systems based on CORD and describe how we integrated the CORD engine with the Exodus Storage Manager to implement Altruistic Locking.\",\n \"Title: Database research at Columbia University\\nYear: 1998\\n\\nAbstract: this report,we describe the Columbia Fast Query Project (Section2), the JAM project (Section 3), the CARDGISproject (Section 4), the Columbia Internet InformationSearching Project (Section 5), the ColumbiaContent-Based Visual Query project (Section 6), andprojects associated with Columbia's ProgrammingSystems Laboratory (Section 7).\",\n \"Title: Divergence control for distributed database systems\\nYear: 1995\\n\\nAbstract: This paper presents distributed divergence control algorithms for epsilon serializability for both homogeneous and heterogeneous distributed databases. Epsilon serializability allows for more concurrency by permitting non-serializable interleavings of database operations among epsilon transactions. We first present a strict 2-phase locking divergence control algorithm and an optimistic divergence control algorithm for a homogeneous distributed database system, where the local orderings of all the sub-transactions of a distributed epsilon transaction are the same. In such an environment, the total inconsistency of a distributed epsilon transaction is simply the sum of those of all its sub-transactions. We then describe a divergence control algorithm for a heterogeneous distributed database system, where the local orderings of all the sub-transactions of a distributed epsilon transaction may not be the same and the total inconsistency of a distributed epsilon transaction may be greater than the sum of those of all its sub-transactions. As a result, in addition to executing a local divergence control algorithm in each site to maintain the local inconsistency, a global mechanism is needed to take into account the additional inconsistency.\",\n \"Title: Distributed divergence control for epsilon serializability\\nYear: 1993\\n\\nAbstract: Epsilon serializability (ESR) allows for more concurrency by permitting nonserializable interleavings of database operations among epsilon transactions (ETs). The authors present the design of distributed divergence control (DDC) algorithms for ESR in homogeneous and heterogeneous distributed databases. They first present a strict two-phase locking DDC algorithm (S2PLDDC) and an optimistic DDC algorithm (ODDC) for homogeneous distributed databases, where the local orderings of all the sub-ETs of a distributed ET are the same, and the total inconsistency of a distributed ET is simply the sum of that of all its sub-ETs. A superdatabase DDC algorithm is described for heterogeneous distributed databases, where the local orderings of all the sub-ETs of a distributed ET may not be the same, and the total inconsistency of a distributed ET may be greater than the sum of that of all its sub-ETs. As a result, in addition to local divergence control in each site, a global mechanism is needed to guarantee ESR\",\n \"Title: An architecture for multi-user software development environments\\nYear: 1993\\n\\nAbstract: We present an architecture for multi-user software development environments, covering general, process-centered and rule-based MUSDEs. Our architecture is founded on componentization, with particular concern for the capability to replace the synchronization component–to allow experimentation with novel concurrency control mechanisms–with minimal effects on other components while still suporting integration. The architecture has been implemented for the MARVEL SDE.\",\n \"Title: The design and implementation of late binding in a distributed programming language\\nYear: 1993\\n\\nAbstract: In distributed application domains where data change rapidly, it is often desirable for programs to obtain the latest available data values to achieve accurate computations. Example applications are financial services and network management. Such data are logically shared by a network of programs. Unlike data in traditional databases, rapidly changing data are usually not lockable by (client) programs and it is crucial to the computations to access their values in a timely manner. In these application domains, a typical program usually performs computations based on recently available data values obtained from the network. However, these data values may be inconsistent or obsolete, since the real data are external to the system and may change more rapidly than can be reflected by their copies within the system. Decision making based on such inaccurate computations can lead to substantial penalties. In this paper, we propose an approach to delaying data value retrieval until needed in distributed programming, considering data and configuration change rapidly. This approach offers the advantage of obtaining more recent data values, resulting in more accurate computations and decision making.\",\n \"Title: Enveloping Sophisticated Tools into Process-Centered Environments\\nYear: 1996\\n\\nAbstract: . We present a tool integration strategy based on enveloping pre-existing tools withoutsource code modifications or recompilation, and without assuming an extension language, applicationprogramming interface, or any other special capabilities on the part of the tool. This BlackBox enveloping (or wrapping) idea has existed for a long time, but was previously restricted torelatively simple tools. We describe the design and implementation of, and experimentation with,a new Black Box...\",\n \"Title: GESDOR - a generic execution model for sharing of computer-interpretable clinical practice guidelines.\\nYear: 2003\\n\\nAbstract: We developed the Guideline Execution by Semantic Decomposition of Representation (GESDOR) model to share guidelines encoded in different formats at the execution level. For this purpose, we extracted a set of generalized guideline execution tasks from the existing guideline representation models. We then created the mappings between specific guideline representation models and the set of the common guideline execution tasks. Finally, we developed a generic task-scheduling model to harmonize the existing approaches to guideline task scheduling. The evaluation has shown that the GESDOR model can be used for the effective execution of guidelines encoded in different formats, and thus realizes guideline sharing at the execution level.\",\n \"Title: Adaptive Synchronization Of Semantically Compressed Instructional Videos For Collaborative Distance Learning\\nYear: 2007\\n\\nAbstract: The increasing popularity of online courses has highlighted the need for collaborative learning tools for student groups. In this article, we present an e-Learning architecture and adaptation model called AI2TV (Adaptive Interactive Internet Team Video), which allows groups of students to collaboratively view instructional videos in synchrony. Video player actions, like play, pause and stop, can be initiated by any group member and and the results of those actions are synchronized with all the other students. These features allow students to review a lecture video in tandem, facilitating the learning process. AI2TV upholds the invariant that each student will receive semantically equivalent content at all times. Experimental trials show that AI2TV successfully synchronizes instructional videos for distributed students while concurrently optimizing the video quality, even under conditions of fluctuating bandwidth, by adaptively adjusting the quality level for each student while still maintaining the semantic content of the instructional videos.\",\n \"Title: Extending attribute grammars to support programming-in-the-large\\nYear: 1994\\n\\nAbstract: Attribute grammars add specification of static semantic properties to context-free grammars, which, in turn, describe the syntactic structure of program units. However, context-free grammars cannot express programming-in-the-large features common in modern programming languages, including unordered collections of units, included units, and sharing of included units. We present extensions to context-free grammars, and corresponding extensions to attribute grammars, suitable for defining such features. We explain how batch and incremental attribute-evaluation algorithms can be adapted to support these extensions, resulting in a uniform approach to intraunit and interunit static semantic analysis and translation of multiunit programs.\",\n \"Title: A Virtual Environment for Collaborative Distance Learning with Video Synchronization\\nYear: 2004\\n\\nAbstract: We present a 3D collaborative virtual environment, CHIME, in which geographically dispersed students can meet together in study groups or to work on team projects. Conventional educational materials from heterogeneous backend data sources are reflected in the virtual world through an automated metadata extraction and projection process that structurally organizes container materials into rooms and interconnecting doors, with atomic objects within containers depicted as furnishings and decorations. A novel in-world authoring tool makes it easy for instructors to design environments, with additional in-world modification afforded to the students themselves, in both cases without programming. Specialized educational services can also be added to virtual environments via programmed plugins. We present an example plugin that supports synchronized viewing of lecture videos by groups of students with widely varying bandwidths.\",\n \"Title: RUST: A Retargetable Usability Testbed for Web Site Authentication Technologies\\nYear: 2008\\n\\nAbstract: Website authentication technologies attempt to make the identity of a website clear to the user, by supplying in- formation about the identity of the website. In prac- tice however, usability issues can prevent users from cor- rectly identifying the websites they are interacting with. To help identify usability issues we present RUST, a Retargetable USability Testbed for website authentica- tion technologies. RUST is a testbed that consists of a test harness, which provides the ability to easily con- figure the environment for running usability study ses- sions, and a usability study design that evaluates usability based on spoofability, learnability, and acceptability. We present data collected by RUST and discuss preliminary results for two authentication technologies, Microsoft CardSpace and Verisign Secure Letterhead. Based on the data collected, we conclude that the testbed is useful for gathering data on a variety of technologies.\",\n \"Title: Workgroup Middleware for Distributed Projects*\\nYear: 1998\\n\\nAbstract: We have developed a middleware framework for workgroup environments that can support distributed software develop- ment and a variety of other application domains requiring document management and change management for distributed projects. The framework enables hypermedia-based integra- tion of arbitrary legacy and new information resources avail- able via a range of protocols, not necessarily known in ad- vance to us as the general framework developers nor even to the environment instance designers. The repositories in which such information resides may be dispersed across the Internet and/or an organizational intranet. The framework also permits a range of client models for user and tool inter- action, and applies an extensible suite of collaboration ser- vices, including but not limited to multi-participant workflow and coordination, to their information retrievals and updates. That is, the framework is interposed between clients, services and repositories — thus \\\"middleware\\\". We explain how our framework makes it easy to realize a comprehensive collec- tion of workgroup and workflow features we culled from a requirements survey conducted by NASA.\",\n \"Title: Unit test virtualization with VMVM\\nYear: 2014\\n\\nAbstract: Testing large software packages can become very time intensive. To address this problem, researchers have investigated techniques such as Test Suite Minimization. Test Suite Minimization reduces the number of tests in a suite by removing tests that appear redundant, at the risk of a reduction in fault-finding ability since it can be difficult to identify which tests are truly redundant. We take a completely different approach to solving the same problem of long running test suites by instead reducing the time needed to execute each test, an approach that we call Unit Test Virtualization. With Unit Test Virtualization, we reduce the overhead of isolating each unit test with a lightweight virtualization container. We describe the empirical analysis that grounds our approach and provide an implementation of Unit Test Virtualization targeting Java applications. We evaluated our implementation, VMVM, using 20 real-world Java applications and found that it reduces test suite execution time by up to 97% (on average, 62%) when compared to traditional unit test execution. We also compared VMVM to a well known Test Suite Minimization technique, finding the reduction provided by VMVM to be four times greater, while still executing every test with no loss of fault-finding ability.\",\n \"Title: JPernLite: Extensible Transaction Services for the WWW\\nYear: 1999\\n\\nAbstract: Concurrency control is one of the key problems in design and implementation of collaborative systems such as hypertext/hypermedia systems, CAD/CAM systems, and software development environments. Most existing systems store data in specialized databases with built-in concurrency control policies, usually implemented via locking. It is desirable to construct such collaborative systems on top of the World Wide Web, but most Web servers do not support even conventional transactions, let alone distributed (multi-Website) transactions or flexible concurrency control mechanisms oriented toward teamwork驴such as event notification, shared locks, and fine granularity locks. We present a transaction server that operates independently of Web servers or the collaborative systems, to fill the concurrency control gap. By default, the transaction server enforces the conventional atomic transaction model, where sets of operations are performed in an all-or-nothing fashion and isolated from concurrent users. The server can be tailored dynamically to apply more sophisticated concurrency control policies appropriate for collaboration. The transaction server also supports applications employing information resources other than Web servers, such as legacy databases, CORBA objects, and other hypermedia systems. Our implementation permits a wide range of system architecture styles.\",\n \"Title: The Apache HTTP Server Project\\nYear: 1997\\n\\nAbstract: Most reports of Internet collaboration refer to small scale operations among a few authors or designers. However, several projects have shown that the Internet can also be the locus for large scale collaboration. In these projects, contributors from around the world combine their individual forces and develop a product that rivals those of multibillion dollar corporations. The Apache HTTP Server Project is a case in point. This collaborative software development effort has created a robust, feature-rich HTTP server software package that currently dominates the public Internet market (46 percent compared with 16 percent for Microsoft and 12 percent for Netscape, according to a June 1997 survey published by Netcraft). The software and its source code are free, but Apache's popularity is more often attributed to performance than price. The project is managed by the Apache Group, a geographically distributed group of volunteers who use the Internet and Web to communicate, develop, and distribute the server and its related documentation. In addition, hundreds of users have contributed ideas, code, and documentation to the project\",\n \"Title: Enveloping \\\"Persistent\\\" Tools for a Process-Centered Environment\\nYear: 1995\\n\\nAbstract: this document are those of the authors and should not be interpreted asrepresenting the official policies, either expressed or implied, of ARPA, NSF, the US Government,AT&T, Bull or Xerox.\",\n \"Title: Incremental Process Support for Code Reengineering\\nYear: 1994\\n\\nAbstract: Reengineering a large code base can be a monumental task, and the situation becomes even worse if the code is concomitantly being modified. Since January 1992, we have been using the MARVEL process centered environment (PCE) for all of our software development, and are currently using it to develop the Oz PCE (MARVEL's successor). Towards this effort, we are reengineering Oz's code base to isolate the process engine, transaction manager, and object management system as separate components that can be used in arbitrary systems. In this paper, we show how a PCE can assist teams of users in carrying out code reengineering while allowing them to continue their normal code development. The key features to this approach are its incremental nature and the ability of the PCE to automate most of the tasks necessary to maintain the consistency of the code base\",\n \"Title: Automatic extension of an ATN knowledge base\\nYear: 1981\\n\\nAbstract: A computer program is described that acquires much of its knowledge from conversations among operators on Morse code radio networks. The system consists of a learning component and a language understander. The learning component extends a ‘core’ augmented transition network (ATN) knowledge base by generalizing from sentences taken from scripts of actual conversations. The extensions enable the understanding component to process a large number of sentences that are syntactically and semantically similar to the examples.\",\n \"Title: WWW-based collaboration environments with distributed tool services\\nYear: 1998\\n\\nAbstract: We have developed an architecture for a general-purpose framework for hypermedia collaboration environments that support purposeful work by orchestrated teams. The hypermedia represents all plausible multimedia artifacts concerned with the collaborative task(s) at hand that can be placed or generated on-line, from application-specific materials (e.g., source code, chip layouts, blueprints) to formal documentation to digital library resources to informal email and chat transcripts. The framework capabilities support both internal (WWW-style hypertext) and external (non-WWW open hypertext link server) links among these artifacts, which can be added incrementally as useful connections are discovered; project-specific intelligent hypermedia search and browsing; automated construction of artifacts and hyperlinks according to the semantics of the group and individual tasks and the overall workflow among the tasks; application of arbitrary tools to the artifacts; and collaborative work for geographically dispersed teams connected by the Internet and/or an intranet/extranet. We also present a general architecture for a WWW-based distributed tool launching service compatible with our collaboration environment framework. We describe our prototype realization of the framework in OzWeb. It reuses object-oriented data management for application-specific hyperbase organization, and workflow enactment and cooperative transactions as built-in services, which were originally developed for the Oz non-hypermedia environment. The tool service is implemented by the generic Rivendell component, which has been integrated into OzWeb as an example “foreign” (i.e., add-on) service. Rivendell could alternatively be employed in a stand-alone manner. We have several months experience using an OzWeb hypermedia collaboration environment for our own continuing software development work on the system.\",\n \"Title: WWW access to legacy client/server applications\\nYear: 1996\\n\\nAbstract: We describe a method for accessing Client/Server applications from standard World Wide Web browsers. An existing client for the system is modified to perform HTTP Proxy duties. Web browser users simply configure their browsers to use this HTTP Proxy, and can then access the system via specially encoded URLs that the HTTP Proxy intercepts and sends to the legacy server system. An example implementation using the Oz Process Centered Software Development Environment is presented.\",\n \"Title: Transactions for concurrent object-oriented programming systems\\nYear: 1989\\n\\nAbstract: Concurrent object-oriented programming systems (COOPS) require support for fault tolerance, concurrency control, consistent commitment of changes and program-initiated rollback. It is sometimes suggested that the classical transaction processing model successfully applied in databases and operating systems be integrated directly into COOPS facilities. This is clearly desirable, but by itself is too limiting. COOPS applications require several granularities of transaction-like facilities. A number of transaction-like mechanisms were addressed at the workshop, and no doubt there are many others suitable for COOPS. Here I briefly survey four levels of granularity that were discussed at the workshop.In their workshop paper Linearizable Concurrent Objects and in [1], Maurice Herlihy and Jeannette Wing describe a fine granularity correctness condition for COOPS, linearizability. Linearizability requires that each operation should appear to “take effect” instantaneously and that the order of non-concurrent operations should be preserved. My workshop paper Concurrent Meld discusses the incorporation of atomic blocks as a programming facility into the Meld COOPS. Atomic blocks in effect implement linearizability at the level of blocks (essentially compound statements, which may encompass an entire method).In his workshop paper Nested Objects and in [4], Bruce Martin describes small grain mechanisms for both externally serializable and semantically verifiable operations for COOPS. Externally serializable operations enforce serializability among top-level operations but permit non-serializable computations on subobjects. Semantically verifiable operations do not enforce serializability at all, but instead consider the semantics of apparently conflicting operations in preventing inconsistencies from being introduced.My paper also discusses our plans for integrating medium granularity transactions — that is, classical transactions with atomicity and serializability as the correctness conditions — into Meld. Serializable transactions have been implemented in a separate version of Meld. not yet integrated into the main line. This implementation uses an algorithm for distributed optimistic concurrency control with multiple versions [2], but two phase locking, timestamps, etc. seem equally feasible for COOPS depending on the anticipated applications.We are currently working on large grain transaction-like facilities for COOPS. Our particular concern is an important class of COOPS applications — open-ended activities [5] such as CAD/CAM. VLSI design, medical informatics, office automation and software development — characterized by uncertain duration, uncertain developments and interaction with other concurrent activities. Classical transactions are not suitable for open-ended activities for the following reasons:Fault tolerance implies atomicity, so if there is some failure, the transaction is rolled back and retried. This is entirely inappropriate when the 'transaction' consists of fixing a bug by browsing and editing a number of source files, compiling and linking, executing test cases and generating traces, etc., which may easily take several hours.Concurrency control requires that users are isolated, and the transactions appear to have been executed in some serial order. But it is not acceptable for a programmer to be locked out from editing a source file just because some other programmer had previously read the file but has not yet finished his changes to other files.Consistent commitment implies simultaneously making all of a transaction's updates publicly available only when the first transaction completes and commits. But programmers must release certain source files so they can be viewed and/or compiled and linked by other programmers cooperating on the same subsystem, while continuing in progress work on other files.An abort deletes all changes made during the transaction, so they are never available to other transactions. A programmer may realize that his original ideas about the cause of a program error were incorrect and decide to start over, but he may want to keep his incorrectly modified versions available for reference.Thus the traditional transaction model is not suitable for open-ended activities such as software development, at least where the 'transactions' are at the granularity of bug fixes, completion of a milestone, or release of a product. An extended transaction processing model, with supporting programming facilities, is necessary for COOPS.We have investigated an extended transaction model, commit-serializability, as a candidate for integration into Meld and other COOPS. The term commit-serializability denotes a model where all committed transactions are in fact serializable in the standard sense, but these transactions may not correspond in a simple way to those transactions that were initiated. In particular, the initiated transactions may be divided during operation and parts committed separately in such ways that these transactions are not serializable.Consider two in-progress transactions T1 and T2. T1 is divided under program or user control into T3 and T4, and shortly thereafter T3 commits while T4 continues. T2 may view the committed updates of T3, some of which were made by T1 before the division, and then itself commits. T4 may then view the committed updates of T2 before it commits. T2, T3 and T4 are serializable, but T1 and T2 are not. The originally initiated transaction T1 in effect disappears, and in particular is neither committed nor aborted.Commit-serializability is supported by two new transaction processing operations, split-transaction and join-transaction [5], in addition to the standard begin-transaction, commit-transaction and abort-transaction operations. The split-transaction operation supports the kind of division described above; the inverse join-transaction operation merges a completed transaction into an in-progress transaction to commit their results together.The two-way versions of the split and join operations take the following arguments. Either statement may be executed as the last statement of any method invoked as part of a transaction T. split-Transaction (A: (AReadSet, AWriteSet, AMessage), B: (BReadSet, BWriteSet, BMessage)) Join-Transaction (S: TID)When the split-transaction operation is invoked during a transaction T, there is a TReadSet consisting of all objects read by T but not updated and a TWriteSet consisting of all objects updated by T. In COOPS, TReadSet is the set of instance variables read by T, including the instance variables of objects (transitively) sent a message where the responding method accessed but did not modify these instance variables of that object. TWriteSet is the set of instance variables assigned by T, including the instance variables of objects (transitively) sent a message where the responding method modified these instance variables of that object. In both cases, the object whose method initiated T acts as coordinator of the cohort.TReadSet is divided, not necessarily disjointly, into AReadSet and BReadSet. TWriteSet is divided disjointly into AWriteSet and BWriteSet; AWriteSet normally must also be disjoint from BReadSet and BWriteSet from AReadSet. AMessage and BMessage are the selectors of messages sent to the same object ($self in Meld) to continue the two transactions. In the special case where A is immediately committed, AMessage is the operation commit-transaction, and objects in AWriteSet may also appear in either BReadSet or BWriteSet.When the join-transaction operation is invoked during a transaction T, target transaction S must be ongoing — and not in conflict with T. TReadSet and TWriteSet are added to SReadSet and SWriteSet, respectively, and S may continue or commit. One difficulty of join-transaction that did not arise for split-transaction is the necessity of naming another transaction S, other than the current one T. This can be done by treating each transaction as a special object, and references to such special objects can be stored in instance variables (TID in the figure), sent as arguments in messages and even sent messages — although these objects respond only to transaction operations. (This is orthogonal to treating different classes of transaction processing characteristics as mixin superclasses, as in Avalon [3].) More work on naming issues is needed.In the cases of both split-transaction and join-transaction, the originally initiated transaction T is divided or merged, respectively, so the net effect is as if it had never existed. The tables, logs, etc. used in the transaction manager implementation may be updated to expunge knowledge of T and replace it with knowledge of A and B or S, respectively, or this may be optimized to use T's identification for one of A or B, or for S.\",\n \"Title: Garp: Graph Abstractions for Concurrent Programming\\nYear: 1988\\n\\nAbstract: Several research projects are investigating parallel processing languages where dynamic process topologies can be constructed. Failure to impose abstractions on interprocess connection patterns can result in arbitrary interconnection topologies that are difficult to understand. We propose the use of a graph-grammar based formalism to control the complexities arising from trying to program such dynamic networks.\",\n \"Title: Us and them: a study of privacy requirements across north america, asia, and europe\\nYear: 2014\\n\\nAbstract: Data privacy when using online systems like Facebook and Amazon has become an increasingly popular topic in the last few years. However, only a little is known about how users and developers perceive privacy and which concrete measures would mitigate their privacy concerns. To investigate privacy requirements, we conducted an online survey with closed and open questions and collected 408 valid responses. Our results show that users often reduce privacy to security, with data sharing and data breaches being their biggest concerns. Users are more concerned about the content of their documents and their personal data such as location than about their interaction data. Unlike users, developers clearly prefer technical measures like data anonymization and think that privacy laws and policies are less effective. We also observed interesting differences between people from different geographies. For example, people from Europe are more concerned about data breaches than people from North America. People from Asia/Pacific and Europe believe that content and metadata are more critical for privacy than people from North America. Our results contribute to developing a user-driven privacy framework that is based on empirical evidence in addition to the legal, technical, and commercial perspectives.\",\n \"Title: Manipulating managed execution runtimes to support self-healing systems\\nYear: 2005\\n\\nAbstract: Self-healing systems require that repair mechanisms are available to resolve problems that arise while the system executes. Managed execution environments such as the Common Language Runtime (CLR) and Java Virtual Machine (JVM) provide a number of application services (application isolation, security sandboxing, garbage collection and structured exception handling) which are geared primarily at making managed applications more robust. However, none of these services directly enables applications to perform repairs or consistency checks of their components. From a design and implementation standpoint, the preferred way to enable repair in a self-healing system is to use an externalized repair/adaptation architecture rather than hardwiring adaptation logic inside the system where it is harder to analyze, reuse and extend. We present a framework that allows a repair engine to dynamically attach and detach to/from a managed application while it executes essentially adding repair mechanisms as another application service provided in the execution environment.\",\n \"Title: Modeling Concurrency in Rule-Based Development Environments\\nYear: 1990\\n\\nAbstract: The problem of cooperative work in the software development domain is explored, and a solution that combines object-oriented programming with rule-based modeling is proposed. The solution divides the problem into three components: how to detect potential conflicts between developers' concurrent activities, how to specify the consistency requirements of a project, and how to use the consistency specification to resolve potential conflicts. The focus is on the first component; the other two are merely sketched. The solution exploits recent advances in object-oriented databases, extended transaction models, and computer-supported cooperative work, all of which provide clues as to how to support cooperation while guaranteeing data consistencyed.\",\n \"Title: An object-based programming model for shared data\\nYear: 1992\\n\\nAbstract: The classical object model supports private data within objects and clean interfaces between objects, and by definition does not permit sharing of data among arbitrary objects. This is a problem for real-world applications, such as advanced financial services and integrated network management, where the same data logically belong to multiple objects and may be distributed over multiple nodes on the network. Rather than give up the advantages of encapsulated objects in modeling real-world entities, we propose a new object model that supports shared data in a distributed environment. The key is separating distribution of computation units from information-hiding concerns. Minimal units of data and control, called facets, may be shared among multiple objects and are grouped into processes. Thus, a single object, or information-hiding unit, may be distributed among multiple processes, or computation units. In other words, different facets of the same object may reside in different address spaces on different machines. We introduce our new object model, describe a motivating example from the financial domain, and then explain facets, objects, and processes, followed by timing and synchronization concerns.\",\n \"Title: Properties of Machine Learning Applications for Use in Metamorphic Testing\\nYear: 2008\\n\\nAbstract: It is challenging to test machine learning (ML) applica- tions, which are intended to learn properties of data sets where the correct answers are not already known. In the absence of a test oracle, one approach to testing these ap- plications is to use metamorphic testing, in which proper- ties of the application are exploited to define transforma- tion functions on the input, such that the new output will be unchanged or can easily be predicted based on the original output; if the output is not as expected, then a defect must exist in the application. Here, we seek to enumerate and classify the metamorphic properties of some machine learn- ing algorithms, and demonstrate how these can be applied to reveal defects in the applications of interest. In addition to the results of our testing, we present a set of properties that can be used to define these metamorphic relationships so that metamorphic testing can be used as a general ap- proach to testing machine learning applications.\",\n \"Title: Vroom: Faster Build Processes for Java\\nYear: 2015\\n\\nAbstract: Build processes are too slow. Because most of the build time for Java projects is spent executing tests, researchers have focused on speeding up testing. They've integrated two complementary approaches into a system that seamlessly supports Ant and Maven JUnit build processes. The first approach, unit test virtualization, isolates in-memory dependencies among test cases, which otherwise are isolated inefficiently by restarting the Java Virtual Machine (JVM) before every test. The system supports just-in-time reinitialization of only the small portion of memory needed by the next test, reusing a single JVM. The implementation of this approach is called VMVM (Virtual Machine in the Virtual Machine, pronounced \\\"vroom vroom\\\"). In addition, simple setup and tear-down resource management methods designed for sequential execution lead to conflicts when the resources are accessed concurrently. So, the second approach, virtualized unit test virtualization, isolates external dependencies such as files and network ports while long-running tests execute in parallel. For this, the system distributes testing jobs in round-robin manner among OS-level virtual machines. The result is, on average, a 51 percent speedup of application build times. The implementation of this approach is called VMVMVM (Virtual Machine in a Virtual Machine on a Virtual Machine \\\"vroom vroom vroom\\\").\",\n \"Title: Phosphor: illuminating dynamic data flow in commodity jvms\\nYear: 2014\\n\\nAbstract: Dynamic taint analysis is a well-known information flow analysis problem with many possible applications. Taint tracking allows for analysis of application data flow by assigning labels to data, and then propagating those labels through data flow. Taint tracking systems traditionally compromise among performance, precision, soundness, and portability. Performance can be critical, as these systems are often intended to be deployed to production environments, and hence must have low overhead. To be deployed in security-conscious settings, taint tracking must also be sound and precise. Dynamic taint tracking must be portable in order to be easily deployed and adopted for real world purposes, without requiring recompilation of the operating system or language interpreter, and without requiring access to application source code. We present Phosphor, a dynamic taint tracking system for the Java Virtual Machine (JVM) that simultaneously achieves our goals of performance, soundness, precision, and portability. Moreover, to our knowledge, it is the first portable general purpose taint tracking system for the JVM. We evaluated Phosphor's performance on two commonly used JVM languages (Java and Scala), on two successive revisions of two commonly used JVMs (Oracle's HotSpot and OpenJDK's IcedTea) and on Android's Dalvik Virtual Machine, finding its performance to be impressive: as low as 3% (53% on average; 220% at worst) using the DaCapo macro benchmark suite. This paper describes our approach toward achieving portable taint tracking in the JVM.\",\n \"Title: Pebbles: fine-grained data management abstractions for modern operating systems\\nYear: 2014\\n\\nAbstract: Support for fine-grained data management has all but disappeared from modern operating systems such as Android and iOS. Instead, we must rely on each individual application to manage our data properly - e.g., to delete our emails, documents, and photos in full upon request; to not collect more data than required for its function; and to back up our data to reliable backends. Yet, research studies and media articles constantly remind us of the poor data management practices applied by our applications. We have developed Pebbles, a fine-grained data management system that enables management at a powerful new level of abstraction: application-level data objects, such as emails, documents, notes, notebooks, bank accounts, etc. The key contribution is Pebbles's ability to discover such high-level objects in arbitrary applications without requiring any input from or modifications to these applications. Intuitively, it seems impossible for an OS-level service to understand object structures in unmodified applications, however we observe that the high-level storage abstractions embedded in modern OSes - relational databases and object-relational mappers - bear significant structural information that makes object recognition possible and accurate.\",\n \"Title: Intelligent assistance for software development and maintenance\\nYear: 1988\\n\\nAbstract: An environment is described, called Professor Marvel, that provides early error checking and answers questions about the program under development. The environment has a certain understanding of the systems being developed and how to use tools to produce software. It aids individual programmers and helps coordinate programming teams. The key components of Marvel are a database that stores data represented as objects, as in object-oriented languages, and a model of the development process that imposes a structure on programming activities. Marvel's support of insight and of opportunistic processing is discussed at length, as is the handling of side effects. A sample session is described.<>\",\n \"Title: Rapid prototyping of concurrent programming languages\\nYear: 1988\\n\\nAbstract: A novel approach is proposed for automatic generation of concurrent interpreters for formal specifications of the programming languages. It consists of a formal notation and supporting algorithms. The formal notation is called action equations, which are attribute grammars extended by the concepts of events and unification. The supporting algorithms include preprocessing algorithms to generate the interpreters and evaluation algorithms embedded in the generated interpreters. Attribute grammars are reviewed, the extension to the action equations paradigm is explained, and the synthesis of action equations with a unification strategy as a means for specifying and implementing synchronization primitives is presented. The approach is illustrated by a specification of CSP. A brief comparison to related work is included\",\n \"Title: Infuse: fusing integration test management with change management\\nYear: 1989\\n\\nAbstract: Infuse is an experimental software development environment focusing on change coordination during the maintenance/evolution phase of large scale software projects. Its core philosophy is to integrate strongly connected modules first and more weakly connected sets of modules later, moving up a hierarchy from singletons to clusters of interdependent modules and, finally, merging the change set into the baseline. We have previously described how Infuse enforces static consistency at each level of the hierarchy. We now extend our work to dynamic consistency — i.e., testing. Unit testing is done for the individual modules at the leaves of the hierarchy, integration testing for the intermediate clusters and acceptance testing at the root. Infuse supports this by partially automating the construction of test harnesses and regression test suites at each level of the hierarchy from components available from lower levels. Infuse is implemented for C, and is used to support its own evolution, but the implementation does not yet provide the test management described here.\",\n \"Title: Testing reliable distributed applications through simulated events\\nYear: 1989\\n\\nAbstract: There are many distributed applications that incorporate application-specific reliability algorithms that operate on top of general-purpose networking, operating system, and programming language facilities. The authors present a framework for application-level reliability testing suitable for a wide range of distributed applications using low-level events and the automatic generation of series of these events. They also describe how they used this testing approach for a particular application\",\n \"Title: An information retrieval approach for automatically constructing software libraries\\nYear: 1991\\n\\nAbstract: A technology for automatically assembling large software libraries which promote software reuse by helping the user locate the components closest to her/his needs is described. Software libraries are automatically assembled from a set of unorganized components by using information retrieval techniques. The construction of the library is done in two steps. First, attributes are automatically extracted from natural language documentation by using an indexing scheme based on the notions of lexical affinities and quantity of information. Then a hierarchy for browsing is automatically generated using a clustering technique which draws only on the information provided by the attributes. Due to the free-text indexing scheme, tools following this approach can accept free-style natural language queries.\",\n \"Title: Debugging multithreaded programs with MPD\\nYear: 1991\\n\\nAbstract: MPD, a multiprocessor debugger, is described. In MPD, users specify sequential or parallel event (breakpoint) expressions and associated actions; MPD automatically generates code to recognize these patterns of events and invokes the corresponding actions when the event expressions are satisfied during the program execution. The richness of the expression-specification language gives users a great deal of precision in formulating queries, obviating tedious searches through long traces of multithreaded software. Implementation issues are discussed.<>\",\n \"Title: An object-based approach to implementing distributed concurrency control\\nYear: 1991\\n\\nAbstract: Distributed concurrency control has been implemented by representing in-progress transactions as simulated objects, called transaction objects, that use normal message passing facilities. The implementation of an optimistic mechanism has been completed using transaction objects and a two-phase locking mechanism has been designed. The tradeoffs made and lessons learned, dealing with transactions both on objects and as objects, are discussed\",\n \"Title: Using tool abstraction to compose systems\\nYear: 1992\\n\\nAbstract: The tool abstraction paradigm, which supports the evolution of large-scale software systems by easing design changes in the system functions, is discussed. Systems that support tool abstraction are structured as a pool of abstract data structures shared by a collection of cooperating 'toolies', where each toolie provides a piece of the overall system function. When one toolie updates the shared data, other toolies must be notified: otherwise, cooperating-but-independent toolies may not execute, and the overall system function may be compromised. The KWIC (key word in context) index production system is used to illustrate the idea of tool abstraction. The relationship of tool abstraction to other concepts is examined.<>\",\n \"Title: Rule chaining in Marvel: dynamic binding of parameters\\nYear: 1992\\n\\nAbstract: Marvel is a software development environment that uses a rule-based model of the development process. When a user issues a command, Marvel invokes the corresponding rule and binds the rule's formal parameter to the object selected by the user. If firing the rule changes the object in a way that satisfies the conditions of other rules, Marvel automatically fires each of these rules and tries to bind their formal parameters to objects. Marvel must infer which objects to bind to the formal parameters of the rules in the chain. A problem arises when the classes of the objects manipulated by the rules are different, because Marvel has to determine which object to bind to the parameter of each rule in the chain. Recursive data definitions cause chaining between rules that act on different objects in the same class. Methods developed to address this chaining problem are discussed.<>\",\n \"Title: A bi-level language for software process modeling\\nYear: 1993\\n\\nAbstract: The authors present a multi-user implementation of a bi-level process modeling language (PML). Most process modeling formalisms are well-suited to one of two levels of specification, but not both. Some concentrate on global control flow and synchronization. These languages make it easy to define the broad outline of a process, but harder to refine the process by expressing constraints and policies on individual tools and data. Other process formalisms are inherently local. It is easy to define constraints, but far from straightforward to express control flow. Combining global and local formalisms is proposed to produce bi-level formalisms suitable for expressing the enacting large scale processes. The new PML is called the activity structures language. The activity structures language integrates global constrained expressions with local rules. Its implementation on top of the Marvel rule-based environment is described\",\n \"Title: Dynamic inference of likely metamorphic properties to support differential testing\\nYear: 2015\\n\\nAbstract: Metamorphic testing is an advanced technique to test programs without a true test oracle such as machine learning applications. Because these programs have no general oracle to identify their correctness, traditional testing techniques such as unit testing may not be helpful for developers to detect potential bugs. This paper presents a novel system, Kabu, which can dynamically infer properties of methods' states in programs that describe the characteristics of a method before and after transforming its input. These Metamorphic Properties (MPs) are pivotal to detecting potential bugs in programs without test oracles, but most previous work relies solely on human effort to identify them and only considers MPs between input parameters and output result (return value) of a program or method. This paper also proposes a testing concept, Metamorphic Differential Testing (MDT). By detecting different sets of MPs between different versions for the same method, Kabu reports potential bugs for human review. We have performed a preliminary evaluation of Kabu by comparing the MPs detected by humans with the MPs detected by Kabu. Our preliminary results are promising: Kabu can find more MPs than human developers, and MDT is effective at detecting function changes in methods.\",\n \"Title: On effective testing of health care simulation software\\nYear: 2011\\n\\nAbstract: Health care professionals rely on software to simulate anatomical and physiological elements of the human body for purposes of training, prototyping, and decision making. Software can also be used to simulate medical processes and protocols to measure cost effectiveness and resource utilization. Whereas much of the software engineering research into simulation software focuses on validation (determining that the simulation accurately models real-world activity), to date there has been little investigation into the testing of simulation software itself, that is, the ability to effectively search for errors in the implementation. This is particularly challenging because often there is no test oracle to indicate whether the results of the simulation are correct. In this paper, we present an approach to systematically testing simulation software in the absence of test oracles, and evaluate the effectiveness of the technique.\",\n \"Title: Accelerating Maven by delaying test dependencies\\nYear: 2015\\n\\nAbstract: Modular build systems (such as Maven) may simplify build maintenance, but significantly reduce opportunities for parallelism where they may be most helpful: when running tests. If tests are contained in each module, and modules contain dependencies on each other, their tests can not run in parallel between modules that have build dependencies. This poster will present a technique for achieving significantly greater parallelism in running the tests of Maven-built Java projects, cutting build times in half in our case study.\",\n \"Title: The weHelp reference architecture for community-driven recommender systems\\nYear: 2010\\n\\nAbstract: Recommender systems have become increasingly popular. Most research on recommender systems has focused on recommendation algorithms. There has been relatively little research, however, in the area of generalized system architectures for recommendation systems. In this paper, we introduce weHelp - a reference architecture for social recommender systems. Our architecture is designed to be application and domain agnostic, but we briefly discuss here how it applies to recommender systems for software engineering.\",\n \"Title: Dynamic taint tracking for Java with phosphor (demo)\\nYear: 2015\\n\\nAbstract: Dynamic taint tracking is an information flow analysis that can be applied to many areas of testing. Phosphor is the first portable, accurate and performant dynamic taint tracking system for Java. While previous systems for performing general-purpose taint tracking in the JVM required specialized research JVMs, Phosphor works with standard off-the-shelf JVMs (such as Oracle's HotSpot and OpenJDK's IcedTea). Phosphor also differs from previous portable JVM taint tracking systems that were not general purpose (e.g. tracked only tags on Strings and no other type), in that it tracks tags on all variables. We have also made several enhancements to Phosphor, to track taint tags through control flow (in addition to data flow), as well as to track an arbitrary number of relationships between taint tags (rather than be limited to only 32 tags). In this demonstration, we show how developers writing testing tools can benefit from Phosphor, and explain briefly how to interact with it.\",\n \"Title: Efficient dependency detection for safe Java test acceleration\\nYear: 2015\\n\\nAbstract: Slow builds remain a plague for software developers. The frequency with which code can be built (compiled, tested and packaged) directly impacts the productivity of developers: longer build times mean a longer wait before determining if a change to the application being built was successful. We have discovered that in the case of some languages, such as Java, the majority of build time is spent running tests, where dependencies between individual tests are complicated to discover, making many existing test acceleration techniques unsound to deploy in practice. Without knowledge of which tests are dependent on others, we cannot safely parallelize the execution of the tests, nor can we perform incremental testing (i.e., execute only a subset of an application's tests for each build). The previous techniques for detecting these dependencies did not scale to large test suites: given a test suite that normally ran in two hours, the best-case running scenario for the previous tool would have taken over 422 CPU days to find dependencies between all test methods (and would not soundly find all dependencies) — on the same project the exhaustive technique (to find all dependencies) would have taken over 1e300 years. We present a novel approach to detecting all dependencies between test cases in large projects that can enable safe exploitation of parallelism and test selection with a modest analysis cost.\",\n \"Title: Identifying functionally similar code in complex codebases\\nYear: 2016\\n\\nAbstract: Identifying similar code in software systems can assist many software engineering tasks such as program understanding and software refactoring. While most approaches focus on identifying code that looks alike, some techniques aim at detecting code that functions alike. Detecting these functional clones — code that functions alike — in object oriented languages remains an open question because of the difficulty in exposing and comparing programs' functionality effectively. We propose a novel technique, In-Vivo Clone Detection, that detects functional clones in arbitrary programs by identifying and mining their inputs and outputs. The key insight is to use existing workloads to execute programs and then measure functional similarities between programs based on their inputs and outputs, which mitigates the problems in object oriented languages reported by prior work. We implement such technique in our system, HitoshiIO, which is open source and freely available. Our experimental results show that HitoshiIO detects more than 800 functional clones across a corpus of 118 projects. In a random sample of the detected clones, HitoshiIO achieves 68+% true positive rate with only 15% false positive rate.\",\n \"Title: Code relatives: detecting similarly behaving software.\\nYear: 2016\\n\\nAbstract: Detecting “similar code” is useful for many software engineering tasks. Current tools can help detect code with statically similar syntactic and–or semantic features (code clones) and with dynamically similar functional input/output (simions). Unfortunately, some code fragments that behave similarly at the finer granularity of their execution traces may be ignored. In this paper, we propose the term “code relatives” to refer to code with similar execution behavior. We define code relatives and then present DyCLINK, our approach to detecting code relatives within and across codebases. DyCLINK records instruction-level traces from sample executions, organizes the traces into instruction-level dynamic dependence graphs, and employs our specialized subgraph matching algorithm to efficiently compare the executions of candidate code relatives. In our experiments, DyCLINK analyzed 422+ million prospective subgraph matches in only 43 minutes. We compared DyCLINK to one static code clone detector from the community and to our implementation of a dynamic simion detector. The results show that DyCLINK effectively detects code relatives with a reasonable analysis time.\",\n \"Title: Challenges in Behavioral Code Clone Detection\\nYear: 2016\\n\\nAbstract: When software engineering researchers discuss \\\"similar\\\" code, we often mean code determined by static analysis to be textually, syntactically or structurally similar, known as code clones (looks alike). Ideally, we would like to also include code that is behaviorally or functionally similar, even if it looks completely different. The state of the art in detecting these behavioral clones focuses on checking the functional equivalence of the inputs and outputs of code fragments, regardless of its internal behavior (focusing only on input and output states). We argue that with an advance in dynamic code clone detection towards detecting behavioral clones (i.e., those with similar execution behavior), we can greatly increase the applications of behavioral clones as a whole for general program understanding tasks.\",\n \"Title: Obfuscation resilient search through executable classification.\\nYear: 2018\\n\\nAbstract: Android applications are usually obfuscated before release, making it difficult to analyze them for malware presence or intellectual property violations. Obfuscators might hide the true intent of code by renaming variables and/or modifying program structures. It is challenging to search for executables relevant to an obfuscated application for developers to analyze efficiently. Prior approaches toward obfuscation resilient search have relied on certain structural parts of apps remaining as landmarks, un-touched by obfuscation. For instance, some prior approaches have assumed that the structural relationships between identifiers are not broken by obfuscators; others have assumed that control flow graphs maintain their structures. Both approaches can be easily defeated by a motivated obfuscator. We present a new approach, MACNETO, to search for programs relevant to obfuscated executables leveraging deep learning and principal components on instructions. MACNETO makes few assumptions about the kinds of modifications that an obfuscator might perform. We show that it has high search precision for executables obfuscated by a state-of-the-art obfuscator that changes control flow. Further, we also demonstrate the potential of MACNETO to help developers understand executables, where MACNETO infers keywords (which are from relevant un-obfuscated programs) for obfuscated executables.\",\n \"Title: Replay without recording of production bugs for service oriented applications.\\nYear: 2018\\n\\nAbstract: Short time-to-localize and time-to-fix for production bugs is extremely important for any 24x7 service-oriented application (SOA). Debugging buggy behavior in deployed applications is hard, as it requires careful reproduction of a similar environment and workload. Prior approaches for automatically reproducing production failures do not scale to large SOA systems. Our key insight is that for many failures in SOA systems (e.g., many semantic and performance bugs), a failure can automatically be reproduced solely by relaying network packets to replicas of suspect services, an insight that we validated through a manual study of 16 real bugs across five different systems. This paper presents Parikshan, an application monitoring framework that leverages user-space virtualization and network proxy technologies to provide a sandbox “debug” environment. In this “debug” environment, developers are free to attach debuggers and analysis tools without impacting performance or correctness of the production environment. In comparison to existing monitoring solutions that can slow down production applications, Parikshan allows application monitoring at significantly lower overhead.\",\n \"Title: A Case Study on the Impact of Similarity Measure on Information Retrieval based Software Engineering Tasks.\\nYear: 2018\\n\\nAbstract: Information Retrieval (IR) plays a pivotal role in diverse Software Engineering (SE) tasks, e.g., bug localization and triaging, code retrieval, requirements analysis, etc. The choice of similarity measure is the core component of an IR technique. The performance of any IR method critically depends on selecting an appropriate similarity measure for the given application domain. Since different SE tasks operate on different document types like bug reports, software descriptions, source code, etc. that often contain non-standard domain-specific vocabulary, it is essential to understand which similarity measures work best for different SE documents. This paper presents two case studies on the effect of different similarity measure on various SE documents w.r.t. two tasks: (i) project recommendation: finding similar GitHub projects and (ii) bug localization: retrieving buggy source file(s) correspond to a bug report. These tasks contain a diverse combination of textual (i.e. description, readme) and code (i.e. source code, API, import package) artifacts. observe that the performance of IR models varies when applied to different artifact types. find that, in general, the context-aware models achieve better performance on textual artifacts. In contrast, simple keyword-based bag-of-words models perform better on code artifacts. On the other hand, the probabilistic ranking model BM25 performs better on a mixture of text and code artifacts. We further investigate how such an informed choice of similarity measure impacts the performance of SE tools. In particular, we analyze two previously proposed tools for project recommendation and bug localization tasks, which leverage diverse software artifacts, and observe that an informed choice of similarity measure indeed leads to improved performance of the existing SE tools.\",\n \"Title: Interfacing Oz with the PCTE OMS: A Case Study of Integrating a Legacy System with a Standard Object Management System\\nYear: 1999\\n\\nAbstract: The integration of a legacy system and a standard Object Management System (OMS) is often a very challenging task. This paper details a case study, our experiment in inter facing Oz with the PCTE (Portable Common Tool Environment) Object Management System. Oz is a multi- user process-centered software development environment that has been under development in our lab since 1987, originally under the name Marvel. PCTE is a specification that de fines a language-independent interface providing support mechanisms for software engineering e nviron- ments (SEE). One of the premises of PCTE is that, in theory, an SEE such as Oz can be built (or extended) using the services provided by PCTE. The purpose of our experiment was to study how a legacy system such as Oz can be integrated into a new environment framework, e.g., PCTE. The architecture of the legacy system and the services of the framework are t he key factors in the integration approach. Because Oz historically has included a native OMS, our experiment focused on modifying Oz to use the PCTE OMS, which has an open and standard API. This paper de- scribes how several Oz components were changed to interface to the PCTE OMS. The resulting proof-of-concept hybrid system has process control and integration services provided by Oz, and data integration services provided in part by PCTE. We discuss in depth the solut ions to the con- currency control problems that arise in such an environment, where Oz and PCTE use different approaches to transaction management (i.e., each has its own transaction manager). The PCTE implementation used in our experiment was the Emeraude PCTE V 12.5.1, and the Oz version was V 1.0.1.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":93,"string":"93"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13574,"string":"13,574"}}},{"rowIdx":13575,"cells":{"question_id":{"kind":"string","value":"Timothy Roscoe"},"question":{"kind":"string","value":"Generate an abstract for the title \"Snailtrail: Generalizing Critical Paths For Online Analysis Of Distributed Dataflows\" using the following items: 1. Critical path analysis (CPA)\n2. SnailTrail system \n3. Critical participation metric\n4. Real-time performance analysis\n5. Distributed dataflow engines"},"answer":{"kind":"string","value":"We rigorously generalize critical path analysis (CPA) to long-running and streaming computations and present SnailTrail, a system built on Timely Dataflow, which applies our analysis to a range of popular distributed dataflow engines. Our technique uses the novel metric of critical participation, computed on time-based snapshots of execution traces, that provides immediate insights into specific parts of the computation. This allows SnailTrail to work online in real-time, rather than requiring complete offline traces as with traditional CPA. It is thus applicable to scenarios like model training in machine learning, and sensor stream processing.SnailTrail assumes only a highly general model of dataflow computation (which we define) and we show it can be applied to systems as diverse as Spark, Flink, TensorFlow, and Timely Dataflow itself. We further show with examples from all four of these systems that SnailTrail is fast and scalable, and that critical participation can deliver performance analysis and insights not available using prior techniques."},"correct_answer":{"kind":"string","value":"We rigorously generalize critical path analysis (CPA) to long-running and streaming computations and present SnailTrail, a system built on Timely Dataflow, which applies our analysis to a range of popular distributed dataflow engines. Our technique uses the novel metric of critical participation, computed on time-based snapshots of execution traces, that provides immediate insights into specific parts of the computation. This allows SnailTrail to work online in real-time, rather than requiring complete offline traces as with traditional CPA. It is thus applicable to scenarios like model training in machine learning, and sensor stream processing.SnailTrail assumes only a highly general model of dataflow computation (which we define) and we show it can be applied to systems as diverse as Spark, Flink, TensorFlow, and Timely Dataflow itself. We further show with examples from all four of these systems that SnailTrail is fast and scalable, and that critical participation can deliver performance analysis and insights not available using prior techniques."},"documents":{"kind":"list like","value":["Title: Hype and Virtue\nYear: 2007\n\nAbstract: In this paper, we question whether hypervisors are really acting as a disruptive force in OS research, instead argu- ing that they have so far changed very little at a techni- cal level. Essentially, we have retained the conventional Unix-like OS interface and added a new ABI based on PC hardware which is highly unsuitable for most purposes. Despite commercial excitement, focus on hypervisor de- sign may be leading OS research astray. However, adopt- ing a different approach to virtualization and recognizing its value to academic research holds the prospect of open- ing up kernel research to new directions.","Title: Implementing declarative overlays\nYear: 2005\n\nAbstract: Overlay networks are used today in a variety of distributed systems ranging from file-sharing and storage systems to communication infrastructures. However, designing, building and adapting these overlays to the intended application and the target environment is a difficult and time consuming process.To ease the development and the deployment of such overlay networks we have implemented P2, a system that uses a declarative logic language to express overlay networks in a highly compact and reusable form. P2 can express a Narada-style mesh network in 16 rules, and the Chord structured overlay in only 47 rules. P2 directly parses and executes such specifications using a dataflow architecture to construct and maintain overlay networks. We describe the P2 approach, how our implementation works, and show by experiment its promising trade-off point between specification complexity and performance.","Title: The Network Oracle\nYear: 2005\n\nAbstract: This paper sets out a high-level research agenda aimed at bui lding a collaborative, global end-system monitoring and information infrastructure for the Internet's core sta te. We argue that such a system is beneficial, feasible and timely, representing an important area for engagement a cross database and networking technologies. We start by hypothesizing the benefits to the Internet of a \"Netw ork Oracle\" that could answer real-time questions about the global state of the network. We then argue that data base and networking technology should make it possible to provide a useful approximation of such an oracle today, gathering information from a large number of end hosts and delivering useful views to each end system. We f urther argue that this can and should be done in a decentralized fashion. We provide an outline of such a syste m: it employs sensing agents, along with a distributed query/trigger/dissemination engine, and possible attrac tive end-user applications. A key point of our discussion is the timeliness and importance of a grassroots agenda for Int ernet monitoring and state-sharing. While significant social and economic barriers to deploying a centralized, pu blic Internet monitoring infrastructure exist, there are corresponding social and economic incentives for a collabo rative approach.","Title: The PlanetLab Platform\nYear: 2005\n\nAbstract: PlanetLab is an extensively used, global, community-maintained platform for researchers to develop, deploy, and evaluate widely-distributed\n applications such as Peer-to-Peer systems. Because of PlanetLab’s shared nature, and its unusual design goal of continuous\n replacement of components by the research community, it can also be viewed somewhat as a Peer-to-Peer system (and an ongoing\n research project) itself. This chapter describes PlanetLab’s goals and origins, and discusses in detail the design principles\n that have governed its development and growth so far. It also discusses some of the methodological issues in performing research\n using a platform like PlanetLab - what can be learned from experimentation on PlanetLab, and what research claims can be validated\n by the system.","Title: Dexferizer: a service for data transfer optimization\nYear: 2011\n\nAbstract: We present an approach to optimizing the transfer of data objects within a user's collection of computers and personal devices, subject to a variety of user-defined quality metrics such as cost, power consumption, and latency. By abstracting object transfer as a high-level service, and employing declarative networking techniques to cast object transfer as a constrained optimization problem, we show how to transparently exploit techniques as diverse as swarming and multi-hop transfer through virtual machines. Using a data replication system as a driving application, we demonstrate that our approach can easily accommodate flexible policies not easily implemented with existing solutions, and can thereby result in savings in time, power, bandwidth, or other costs.","Title: Application level ballooning for efficient server consolidation\nYear: 2013\n\nAbstract: Systems software like databases and language runtimes typically manage memory themselves to exploit application knowledge unavailable to the OS. Traditionally deployed on dedicated machines, they are designed to be statically configured with memory sufficient for peak load. In virtualization scenarios (cloud computing, server consolidation), however, static peak provisioning of RAM to applications dramatically reduces the efficiency and cost-saving benefits of virtualization. Unfortunately, existing memory \"ballooning\" techniques used to dynamically reallocate physical memory between VMs badly impact the performance of applications which manage their own memory. We address this problem by extending ballooning to applications (here, a database engine and Java runtime) so that memory can be efficiently and effectively moved between virtualized instances as the demands of each change over time. The results are significantly lower memory requirements to provide the same performance guarantees to a collocated set of VM running such applications, with minimal overhead or intrusive changes to application code.","Title: A better way to negotiate for testbed resources\nYear: 2011\n\nAbstract: Resource allocation is an increasing challenge for distributed network testbeds as computational and network resources are involved. Testbed designers have moved to a query-based model: clients provide a declarative description of their desired resources, and the provider allocate specific resources to meet the request. In this paper, we describe an new approach to negotiate testbed resources between clients and testbed providers: the clients specify their requests as constraints, and the providers reply with resource allocations expressed also as declarative set of constraints on resources. This gives providers more flexibility in late-binding of resources to requests, and opens up a wide design space to optimize resource allocation for efficiency, cost, utilization, or other metrics. Our simple first experiments suggest that the late-binding of resources enabled by representing resource reservation as constraints achieves better network resource utilization compared to the fixed assignment solution.","Title: R-OSGi: distributed applications through software modularization\nYear: 2007\n\nAbstract: In this paper we take advantage of the concepts developed for centralized module management, such as dynamic loading and unloading of modules, and show how they can be used to support the development and deployment of distributed applications. We do so through R-OSGi, a distributed middleware platform that extends the centralized, industry-standard OSGi specification to support distributed module management. To the developer, R-OSGi looks like a conventional module management tool. However, at deployment time, R-OSGi can be used to turn the application into a distributed application by simply indicating where the different modules should be deployed. At run time, R-OSGi represents distributed failures as module insertion and withdrawal operations so that the logic to deal with failures is the same as that employed to deal with dependencies among software modules. In doing so, R-OSGi greatly simplifies the development of distributed applications with no performance cost. In the paper we describe R-OSGi and several use cases. We also show with extensive experiments that R-OSGi has a performance comparable or better than that of RMI or UPnP, both commonly used distribution mechanisms with far less functionality than R-OSGi.","Title: BFT protocols under fire\nYear: 2008\n\nAbstract: Much recent work on Byzantine state machine replication focuses on protocols with improved performance under benign conditions (LANs, homogeneous replicas, limited crash faults), with relatively little evaluation under typical, practical conditions (WAN delays, packet loss, transient disconnection, shared resources). This makes it difficult for system designers to choose the appropriate protocol for a real target deployment. Moreover, most protocol implementations differ in their choice of runtime environment, crypto library, and transport, hindering direct protocol comparisons even under similar conditions. We present a simulation environment for such protocols that combines a declarative networking system with a robust network simulator. Protocols can be rapidly implemented from pseudocode in the high-level declarative language of the former, while network conditions and (measured) costs of communication packages and crypto primitives can be plugged into the latter. We show that the resulting simulator faithfully predicts the performance of native protocol implementations, both as published and as measured in our local network. We use the simulator to compare representative protocols under identical conditions and rapidly explore the effects of changes in the costs of crypto operations, workloads, network conditions and faults. For example, we show that Zyzzyva outperforms protocols like PBFT and Q/U undermost but not all conditions, indicating that one-size-fits-all protocols may be hard if not impossible to design in practice.","Title: Querying at Internet scale\nYear: 2004\n\nAbstract: We are developing a distributed query processor called PIER, which is designed to run on the scale of the entire Internet. PIER utilizes a Distributed Hash Table (DHT) as its communication substrate in order to achieve scalability, reliability, decentralized control, and load balancing. PIER enhances DHTs with declarative and algebraic query interfaces, and underneath those interfaces implements multihop, in-network versions of joins, aggregation, recursion, and query/result dissemination. PIER is currently being used for diverse applications, including network monitoring, keyword-based filesharing search, and network topology mapping. We will demonstrate PIER's functionality by showing system monitoring queries running on PlanetLab, a testbed of over 300 machines distributed across the globe.","Title: The Architecture of PIER: an Internet-Scale Query Processor\nYear: 2005\n\nAbstract: Abstract: This paper presents the architecture of PIER, an Internetscalequery engine we have been building over the last threeyears. PIER is the first general-purpose relational query processortargeted at a peer-to-peer (p2p) architecture of thousandsor millions of participating nodes on the Internet. Itsupports massively distributed, database-style dataflows forsnapshot and continuous queries. It is intended to serve as abuilding block for a diverse set of Internet-scale informationcentric...","Title: A blueprint for introducing disruptive technology into the Internet\nYear: 2003\n\nAbstract: This paper argues that a new class of geographically distributed network services is emerging, and that the most effective way to design, evaluate, and deploy these services is by using an overlay-based testbed. Unlike conventional network testbeds, however, we advocate an approach that supports both researchers that want to develop new services, and clients that want to use them. This dual use, in turn, suggests four design principles that are not widely supported in existing testbeds: services should be able to run continuously and access a slice of the overlay's resources, control over resources should be distributed, overlay management services should be unbundled and run in their own slices, and APIs should be designed to promote application development. We believe a testbed that supports these design principles will facilitate the emergence of a new service-oriented network architecture. Towards this end, the paper also briefly describes PlanetLab, an overlay network being designed with these four principles in mind.","Title: Operating system support for planetary-scale network services\nYear: 2004\n\nAbstract: PlanetLab is a geographically distributed overlay network designed to support the deployment and evaluation of planetary-scale network services. Two high-level goals shape its design. First, to enable a large research community to share the infrastructure, PlanetLab provides distributed virtualization, whereby each service runs in an isolated slice of PlanetLab's global resources. Second, to support competition among multiple network services, PlanetLab decouples the operating system running on each node from the network-wide services that define PlanetLab, a principle referred to as unbundled management. This paper describes how Planet-Lab realizes the goals of distributed virtualization and unbundled management, with a focus on the OS running on each node.","Title: Making the world (of communications) a different place\nYear: 2005\n\nAbstract: How might the computing and communications world be materially different in 10 to 15 years, and how might we define a research agenda that would get us to that world?","Title: Sophia: an Information Plane for networked systems\nYear: 2004\n\nAbstract: This paper motivates and describes an example network Information Plane, called Sophia, currently deployed on PlanetLab. Sophia is a distributed system that collects, stores, propagates, aggregates, and reacts to observations about the network's current conditions. Sophia's approach is novel: it can be viewed as a multi-user distributed expression evaluator in which sensors and actuators form the ground terms, and statements take on the complete expressiveness of a logic language like Prolog. This paper argues that this approach has several advantages in managing and controlling a complex, federated, and evolving network: (1) a declarative logic language provides a natural way to express the kinds of statements that are common to this application domain, through temporal and positional logic rules, facts and expressions; and (2) distributed evaluation of such logic expressions provides many opportunities for performance optimization yielding an efficient system.","Title: Transaction-Based Charging in Mnemosyne: A Peer-to-Peer Steganographic Storage System\nYear: 2002\n\nAbstract: Mnemosyne is a peer-to-peer steganographic storage system: one in which the existence ofa user's files cannot be verified without a key. This paper applies the techniques used in Mnemosyne - erasure codes and anonymous block writing - to move most ofthe administrative overhead of a commercial storage service over to the client, resulting in cost savings for the service provider.The contribution ofthis paper is to present a radically alternative way of charging for storage services. In place of renting some amount of space for some period of time, systems like Mnemosyne allow more flexible billing models closer to those proposed for network bandwidth, including versions of congestion pricing. We show how a reliable, commercial storage service using is feasible, and examine the details of the tradeoff it offers compared with conventional storage services.","Title: Multicore OS benchmarks: we can do better\nYear: 2011\n\nAbstract: Current multicore OS benchmarks do not provide workloads that sufficiently reflect real-world use: they typically run a single application, whereas real workloads consist of multiple concurrent programs. In this paper we show that this lack of mixed workloads leads to benchmarks that do not fully exercise the OS and are therefore inadequate at predicting real-world behavior. This implies that effective multicore OS benchmarks must includemixed workloads, but the main design challenge is choosing an appropriate mix. We present a principled approachwhich treats benchmark design as an optimization problem. Our solution leads to a workload mix that uses as much of a system's resources as possible, while also selecting applications whose performance is most sensitive to the availability of those resources.","Title: PlanetLab: an overlay testbed for broad-coverage services\nYear: 2003\n\nAbstract: PlanetLab is a global overlay network for developing and accessing broad-coverage network services. Our goal is to grow to 1000 geographically distributed nodes, connected by a disverse collection of links. PlanetLab allows multiple service to run concurrently and continuously, each in its own slice of PlanetLab. This paper discribes our initial implementation of PlanetLab, including the mechanisms used to impelment virtualization, and the collection of core services used to manage PlanetLab.","Title: Linkage in the Nemesis single address space operating system\nYear: 1994\n\nAbstract: The recent interest in single address space operating systems has resulted in a number of papers, most of which gloss over the issues of linking programs to run in multiple protection domains. Some of the confusion about 64-bit address spaces is due to the almost pervasive use of UNIX and UNIX-like operating systems (such as Mach, Chorus and Amoeba) and languages with poor enforcement of abstraction like C and C++.This paper describes some of the linkage structure of Nemesis, a multi-service operating system being developed as part of the Pegasus project. Nemesis provides a simple and efficient mechanism for program linkage which provides rich sharing of text at a level of individual object classes.","Title: Resource overbooking and application profiling in shared hosting platforms\nYear: 2002\n\nAbstract: In this paper, we present techniques for provisioning CPU and network resources in shared hosting platforms running potentially antagonistic third-party applications. The primary contribution of our work is to demonstrate the feasibility and benefits of overbooking resources in shared platforms, to maximize the platform yield: the revenue generated by the available resources. We do this by first deriving an accurate estimate of application resource needs by profiling applications on dedicated nodes, and then using these profiles to guide the placement of application components onto shared nodes. By overbooking cluster resources in a controlled fashion, our platform can provide performance guarantees to applications even when overbooked, and combine these techniques with commonly used QoS resource allocation mechanisms to provide application isolation and performance guarantees at run-time. When compared to provisioning based on the worst-case, the efficiency (and consequently revenue) benefits from controlled overbooking of resources can be dramatic. Specifically, experiments on our Linux cluster implementation indicate that overbooking resources by as little as 1% can increase the utilization of the cluster by a factor of two, and a 5% overbooking yields a 300--500% improvement, while still providing useful resource guarantees to applications.","Title: Plutarch: an argument for network pluralism\nYear: 2003\n\nAbstract: It is widely accepted that the current Internet architecture is insufficient for the future: problems such as address space scarcity, mobility and non-universal connectivity are already with us, and stand to be exacerbated by the explosion of wireless, ad-hoc and sensor networks. Furthermore, it is far from clear that the ubiquitous use of standard transport and name resolution protocols will remain practicable or even desirable.In this paper we propose Plutarch, a new inter-networking architecture. It subsumes existing architectures such as that determined by the Internet Protocol suite, but makes explicit the heterogeneity that contemporary inter-networking schemes attempt to mask. To handle this heterogeneity, we introduce the notions of context and interstitial function, and describe a supporting architecture. We discuss the benefits, present some potential scenarios, and consider the research challenges posed.","Title: Supercomputing operating systems: a naive view from over the fence\nYear: 2012\n\nAbstract: To exaggerate unfairly, from the perspective of mainstream OS research, the supercomputing community has a very different idea of the role (and appropriate design) of an OS. HPC people regard the OS as an annoying source of noise, whereas the former crowd see it as a thing of wondrous beauty and elegance, a sine qua non of usable everyday computing. This situation has existed without serious conflict erupting for years: OS researchers worried about PCs with one core (or, at most, a handful of cores) running a general-purpose OS and supporting a dynamic, bursty, diverse mix of hundreds of interactive, long-running, soft-realtime and/or background processes. Supercomputing people wanted one, highly parallel, program to finish as soon as possible so they could get on to the next one. With multicore, this all changed: highly parallel tasks will be the norm for future general-purpose computing. In 2007, my colleagues and I eagerly embarked on a new research OS for multicore computing, and looked forward to applying long-ignored (in our field) results from the HPC realm to our system. It didn't quite work out that way. In this talk I will look at what we found to be common to the two fields, and what we didn't, and speculate on where this might be going. I think there is a useful conversation to be had, and I'd like to help revive it.","Title: Friday: global comprehension for distributed replay\nYear: 2007\n\nAbstract: Debugging and profiling large-scale distributed applications is a daunting task. We present Friday, a system for debugging distributed applications that combines deterministic replay of components with the power of symbolic, low-level debugging and a simple language for expressing higher-level distributed conditions and actions. Friday allows the programmer to understand the collective state and dynamics of a distributed collection of coordinated application components. To evaluate Friday, we consider several distributed problems, including routing consistency in overlay networks, and temporal state abnormalities caused by route flaps. We show via micro-benchmarks and larger-scale application measurement that Friday can be used interactively to debug large distributed applications under replay on common hardware.","Title: CLANGER: an interpreted systems programming language\nYear: 1995\n\nAbstract: Clanger is a powerful, yet simple, command language for the Nemesis operating system. It uses runtime type information to interface directly with operating system components. Clanger is a combination of command-line interpreter, scripting language, debugger and prototyping tool. This paper describes why such a language is possible, how it is being implemented, and outlines the language as it currently stands.","Title: Towards realistic benchmarks for virtual infrastructure resource allocators\nYear: 2012\n\nAbstract: The rise of virtual infrastructures has renewed research interest in the design and implementation of resource allocators which allow users to reserve or use combinations of (virtual) nodes, switches, and network links on a variety of virtual infrastructures ranging from network testbeds to cloud computing facilities. However, since the area is still relatively new, work on resource allocation lacks realistic benchmarks. In this paper, we suggest how to generate realistic benchmarks for virtual infrastructure, and present our workload generator as an example, which is based on a 5-year trace of experiments submitted to the popular Emulab testbed. We demonstrate the importance as well as potential applications of such benchmarks by using the generated test workload in various evaluation scenarios.","Title: Building, deploying, and monitoring distributed applications with Eclipse and R-OSGI\nYear: 2007\n\nAbstract: Designing and testing distributed applications is still a difficult task that requires in-depth knowledge about networking issues. Eclipse is, among other things, a powerful and widely used IDE for the development of complex applications, in particular modular applications for the OSGi framework. Our R-OSGi middleware supports the seamless distribution of OSGi applications along the boundaries of services. By combining R-OSGi with Eclipse into the R-OSGi Deployment Tool, we give developers a tool that automatically handles distribution in a transparent way and integrates the capabilities of R-OSGi into the Eclipse workflow. With this tool, building, deploying, and monitoring distributed applications is as easy as writing OSGi applications in Eclipse and using a graphical editor to visually create distributed deployments of the modules. The tool can also be used to great effect by researchers to test and benchmark distributed applications and for education purposes.","Title: InfoSpect: using a logic language for system health monitoring in distributed systems\nYear: 2002\n\nAbstract: Dependable systems cannot be built without a monitoring and management component. In this paper we propose using a wide variety of information gathering tools coupled with custom scripts and a Prolog language engine to aggregate information from multiple sources. Complex queries, difficult to express in standard database languages, can then be used to answer questions about the system (e.g. the health of individual components) or to discover contradictions (e.g. inconsistent configurations). We describe our prototype implementation and present some early results.","Title: Mnemosyne: Peer-to-Peer Steganographic Storage\nYear: 2002\n\nAbstract: We present the design of Mnemosyne, a peer-to-peer steganographic storage service. Mnemosyne provides a high level of privacy and plausible deniability by using a large amount of shared distributed storage to hide data. Blocks are dispersed by secure hashing, and loss codes used for resiliency. We discuss the design of the system, and the challenges posed by traffic analysis.","Title: We need to talk about NICs\nYear: 2013\n\nAbstract: Operating systems fail both to efficiently exploit, and to effeectively manage, the considerable hardware resources of modern network interface controllers. We survey the kinds of hardware facilities available and their applicability, and then investigate (and critique) the reasons why OS designers eschew core support for such features. We then describe Dragonet, a new network stack design based on explicit descriptions of NIC capabilities, aimed at making the best use of today's and tomorrow's networking hardware. Dragonet represents both the physical capabilities of the network hardware and the current protocol state of the machine as dataflow graphs. We then embed the former into the latter, instantiating the remainder in software.","Title: Preventing Internet denial-of-service with capabilities\nYear: 2004\n\nAbstract: In this paper, we propose a new approach to preventing and constraining denial-of-service (DoS) attacks. Instead of being able to send anything to anyone at any time, in our architecture, nodes must first obtain \"permission to send\" from the destination; a receiver provides tokens, or capabilities, to those senders whose traffic it agrees to accept. The senders then include these tokens in packets. This enables verification points distributed around the network to check that traffic has been certified as legitimate by both endpoints and the path in between, and to cleanly discard unauthorized traffic. We show that our approach addresses many of the limitations of the currently popular approaches to DoS based on anomaly detection, traceback, and pushback. Further, we argue that our approach can be readily implemented in today's technology, is suitable for incremental deployment, and requires no more of a security infrastructure than that already needed to fix BGP's security weaknesses. Finally, our proposal facilitates innovation in application and networking protocols, something increasingly curtailed by existing DoS measures.","Title: Design Considerations for Information Planes\nYear: 2004\n\nAbstract: The concept of an information plane has emerged recently as an important part of large, decentral- ized systems that aspire to be self-managing, ranging from PlanetLab to the Internet itself. In this paper we describe what an information plane is, and report our experiences in developing and deploying an information plane for the Plan- etLab platform using the PIER distributed relational query processor. We recount the lessons we have learned from the experience, and the additional directions we intend to ex- plore in the PHI project, which aims at providing an infor- mation plane that can grow to serve a significant portion of the Internet.","Title: Mind the gap: reconnecting architecture and OS research\nYear: 2011\n\nAbstract: The goal of a computer system is to run an application workload securely, reliably, efficiently, and fast. A computer's hardware architecture and operating system exist to support this goal, and it would be nice if they cooperated as effectively as possible. Yet there is a growing gap between architectural research and OS research, which seems to be the result of poor communication about what actually matters. In this paper, we discuss this gap and what to do about it. We illustrate the opportunities for closing the gap using examples from some recent OS research.","Title: New Resource Control Issues in Shared Clusters\nYear: 2001\n\nAbstract: We claim that the renting of machine resources on clusters of servers introduces new systems challenges which are different from those hitherto encountered, either in multimedia systems or cluster-based computing. We characterize the requirements for such \"public computing platforms\" and discuss both how the scenario differs from more traditional multimedia resource control situations, and how some ideas from multimedia systems work can be reapplied in this new context. Finally, we discuss our ongoing work building a prototype public computing platform.","Title: Structured Peer-to-Peer Overlays Need Application-Driven Benchmarks\nYear: 2003\n\nAbstract: Considerable research effort has recently been devoted to the design of structured peer-to-peer overlays, a term we use to encompass Content-Addressable Networks (CANs), Distributed Hash Tables (DHTs), and Decentralized Object Location and Routing networks (DOLRs). These systems share the property that they consistently map a large set of identifiers to a set of nodes in a network, and while at first sight they provide very similar services, they nonetheless embody a wide variety of design alternatives. We present the case for developing application-driven benchmarks for such overlays, give a model of the services they provide applications, describe and present the results of two preliminary benchmarks, and discuss the implications of our tests for application writers. We are unaware of other empirical comparative work in this area.","Title: Predicate routing: enabling controlled networking\nYear: 2003\n\nAbstract: this paper, we present Predicate Routing(PR) as a solution to this problem. We briey describe ourcentralized implementation and then outline the extensionof Internet routing protocols to support Predicate Routing.In current IP networks, the state of the system is primarilyrepresented in an imperative fashion: routing tables andrewall rulesets local to each node strictly specify the actionto be performed on each arriving packet. In contrast,Predicate Routing represents the state of the...","Title: Resource overbooking and application profiling in a shared Internet hosting platform\nYear: 2009\n\nAbstract: In this article, we present techniques for provisioning CPU and network resources in shared Internet hosting platforms running potentially antagonistic third-party applications. The primary contribution of our work is to demonstrate the feasibility and benefits of overbooking resources in shared Internet platforms. Since an accurate estimate of an application's resource needs is necessary when overbooking resources, we present techniques to profile applications on dedicated nodes, possibly while in service, and use these profiles to guide the placement of application components onto shared nodes. We then propose techniques to overbook cluster resources in a controlled fashion. We outline an empirical appraoch to determine the degree of overbooking that allows a platform to achieve improvements in revenue while providing performance guarantees to Internet applications. We show how our techniques can be combined with commonly used QoS resource allocation mechanisms to provide application isolation and performance guarantees at run-time. We implement our techniques in a Linux cluster and evaluate them using common server applications. We find that the efficiency (and consequently revenue) benefits from controlled overbooking of resources can be dramatic. Specifically, we find that overbooking resources by as little as 1% we can increase the utilization of the cluster by a factor of two, and a 5% overbooking yields a 300--500% improvement, while still providing useful resource guarantees to applications.","Title: 30 seconds is not enough!: a study of operating system timer usage\nYear: 2008\n\nAbstract: The basic system timer facilities used by applications and OS kernels for scheduling timeouts and periodic activities have remained largely unchanged for decades, while hardware architectures and application loads have changed radically. This raises concerns with CPU overhead power management and application responsiveness. In this paper we study how kernel timers are used in the Linux and Vista kernels, and the instrumentation challenges and tradeoffs inherent in conducting such a study. We show how the same timer facilities serve at least five distinct purposes, and examine their performance characteristics under a selection of application workloads. We show that many timer parameters supplied by application and kernel programmers are somewhat arbitrary, and examine the potential benefit of adaptive timeouts. We also discuss the further implications of our results, both for enhancements to the system timer functionality in existing kernels, and for the clean-slate design of a system timer subsystem for new OS kernels, including the extent to which applications might require such an interface at all.","Title: Declarative networking: language, execution and optimization\nYear: 2006\n\nAbstract: The networking and distributed systems communities have recently explored a variety of new network architectures, both for application-level overlay networks, and as prototypes for a next-generation Internet architecture. In this context, we have investigated declarative networking: the use of a distributed recursive query engine as a powerful vehicle for accelerating innovation in network architectures [23, 24, 33]. Declarative networking represents a significant new application area for database research on recursive query processing. In this paper, we address fundamental database issues in this domain. First, we motivate and formally define the Network Datalog (NDlog) language for declarative network specifications. Second, we introduce and prove correct relaxed versions of the traditional semi-naïve query evaluation technique, to overcome fundamental problems of the traditional technique in an asynchronous distributed setting. Third, we consider the dynamics of network state, and formalize the iheventual consistencyl. of our programs even when bursts of updates can arrive in the midst of query execution. Fourth, we present a number of query optimization opportunities that arise in the declarative networking context, including applications of traditional techniques as well as new optimizations. Last, we present evaluation results of the above ideas implemented in our P2 declarative networking system, running on 100 machines over the Emulab network testbed.","Title: The SAHARA Model for Service Composition across Multiple Providers\nYear: 2002\n\nAbstract: Services are capabilities that enable applications and are of crucial importance to pervasive computing in next-generation networks. Service Composition is the construction of complex services from primitive ones; thus enabling rapid and flexible creation of new services. The presence of multiple independent service providers poses new and significant challenges. Managing trust across providers and verifying the performance of the components in composition become essential issues. Adapting the composed service to network and user dynamics by choosing service providers and instances is yet another challenge. In SAHARA, we are developing a comprehensive architecture for the creation, placement, and management of services for composition across independent providers. In this paper, we present a layered reference model for composition based on a classification of different kinds of composition. We then discuss the different overarching mechanisms necessary for the successful deployment of such an architecture through a variety of case-studies involving composition.","Title: Towards a practical, verified kernel\nYear: 2007\n\nAbstract: Optimistic synchronization allows concurrent execution of critical sections while performing dynamic conflict detection and recovery. Optimistic synchronization will increase performance only if critical regions are data independent--concurrent critical ...","Title: The design principles of PlanetLab\nYear: 2006\n\nAbstract: PlanetLab is a geographically distributed platform for deploying, evaluating, and accessing planetary-scale network services. PlanetLab is a shared community effort by a large international group of researchers, each of whom gets access to one or more isolated slices of PlanetLab's global resources. Because we deployed PlanetLab and started supporting users before we fully understood what its architecture would be, being able to evolve the system became a requirement. This paper examines the set of design principles that guided this evolution. Some of these principles were explicit at the project outset, and others have become crystallized as the platform has developed.","Title: Palimpsest: soft-capacity storage for planetary-scale services\nYear: 2003\n\nAbstract: Distributed writable storage systems typically provide NFS-like semantics and unbounded persistence for files. We claim that for planetary-scale distributed services such facilities are unnecessary and impose an unwanted overhead in complexity and ease of management. Furthermore, wide-area services have requirements not met by existing solutions, in particular capacity management and a realistic model for billing and charging. We argue for ephemeral storage systems which meet these requirements, and present Palimpsest, an early example being deployed on PlanetLab. Palimpsest is small and simple, yet provides soft capacity, congestion-based pricing, automatic reclamation of space, and a security model suitable for a shared storage facility for wide-area services. It does not \"fill up\" in the way that an ordinary filing system does. In fact, Palimpsest does not arbitrate resources--storage blocks--between clients at all.","Title: Rhizoma: a runtime for self-deploying, self-managing overlays\nYear: 2009\n\nAbstract: The trend towards cloud and utility computing infrastructures raises challenges not only for application development, but also for management: diverse resources, changing resource availability, and differing application requirements create a complex optimization problem. Most existing cloud applications are managed externally, and this separation can lead to increased response time to failures, and slower or less appropriate adaptation to resource availability and pricing changes. In this paper, we explore a different approach more akin to P2P systems: we closely couple a decentralized management runtime (\"Rhizoma\") with the application itself. The application expresses its resource requirements to the runtime as a constrained optimization problem. Rhizoma then fuses multiple real-time sources of resource availability data, from which it decides to acquire or release resources (such as virtual machines), redeploying the system to continually maximize its utility. Using PlanetLab as a challenging \"proving ground\" for cloud-based services, we present results showing Rhizoma's performance, overhead, and efficiency versus existing approaches, as well the system's ability to react to unexpected large-scale changes in resource availability.","Title: The multikernel: a new OS architecture for scalable multicore systems\nYear: 2009\n\nAbstract: Commodity computer systems contain more and more processor cores and exhibit increasingly diverse architectural tradeoffs, including memory hierarchies, interconnects, instruction sets and variants, and IO configurations. Previous high-performance computing systems have scaled in specific cases, but the dynamic nature of modern client and server workloads, coupled with the impossibility of statically optimizing an OS for all workloads and hardware variants pose serious challenges for operating system structures. We argue that the challenge of future multicore hardware is best met by embracing the networked nature of the machine, rethinking OS architecture using ideas from distributed systems. We investigate a new OS structure, the multikernel, that treats the machine as a network of independent cores, assumes no inter-core sharing at the lowest level, and moves traditional OS functionality to a distributed system of processes that communicate via message-passing. We have implemented a multikernel OS to show that the approach is promising, and we describe how traditional scalability problems for operating systems (such as memory management) can be effectively recast using messages and can exploit insights from distributed systems and networking. An evaluation of our prototype on multicore systems shows that, even on present-day machines, the performance of a multikernel is comparable with a conventional OS, and can scale better to support future hardware.","Title: Filet-o-fish: practical and dependable domain-specific languages for OS development\nYear: 2009\n\nAbstract: We address a persistent problem with using domain-specific languages to write operating systems: the effort of implementing, checking, and debugging the DSL usually outweighs any of its benefits. Because these DSLs generate C by templated string concatenation, they are tedious to write, fragile, and incompatible with automated verification tools. We present Filet-o-Fish (FoF), a semantic language to ease DSL construction. Building a DSL using FoF consists of safely composing semantically-rich building blocks. This has several advantages: input files for the DSL are formal specifications of the system's functionality, automated testing of the DSL is possible via existing tools, and we can prove that the C code generated by a given DSL respects the semantics expected by the developer. Early experience has been good: FoF is in daily use as part of the tool chain of the Barrelfish multicore OS, which makes extensive use of domain-specific languages to generate low-level OS code. We have found that the ability to rapidly generate DSLs we can rely on has changed how we have designed the OS.","Title: Ready for distribution?: turning modular into distributed applications with the R-OSGi deployment tool\nYear: 2007\n\nAbstract: In this demonstration we show drag-and-drop distribution of centralized, modular Java applications. Our system is based on OSGi, an industry standard for building Java applications out of modular units loosely connected through services. Since OSGi is a centralized system, we have elaborated a solution to seamlessly distribute OSGi applications along the boundaries of services and thereby turning arbitrary OSGi applications into distributed applications. In this demonstration, we present an Eclipse based tool that takes the source code of an OSGi application as input, produces a graph of its modules and module dependencies, and allows the user to deploy the application across a distributed system by dragging-and-dropping its constituent modules on different machines. By defining constraints on the distribution, the tool can also support advanced features like load-balancing or redundancy of modules.","Title: Dependable self-hosting distributed systems using constraints\nYear: 2008\n\nAbstract: We describe a technique for writing distributed applications which manage themselves over one or more utility computing infrastructures: by dynamically acquiring new computational resources, deploying themselves on these resources, and releasing others when no longer required. Unlike prior work, such management functionality is closely integrated with the application, allowing greater freedom in application-specific policies and faster response to failures and other changes in the environment without requiring any external management system. We address the programming complexity of this approach by applying constraint logic programming, and describe Rhizoma, an experimental runtime to explore these ideas. We present early experience of deploying self-hosting applications on PlanetLab using Rhizoma.","Title: Policy expressivity in the Anzere personal cloud\nYear: 2011\n\nAbstract: We present a technique for partially replicating data items at scale according to expressive policy specifications. Recent projects have addressed the challenge of policy-based replication of personal data (photos, music, etc.) within a network of devices, as an alternative to centralized online services. To date, the policies supported by such systems have been relatively simple, in order to facilitate scaling the policy calculation to large numbers of items. In this paper, we show how such replication systems can scale while supporting much more expressive policies than previous schemes: item replication expressed as constraints, devices referred to by predicates rather than explicitly named, and replication to storage nodes acquired on-demand from the cloud. These extensions introduce considerable complexity in policy evaluation, but we show a system can scale well by using equivalence classes to reduce the problem space. We validate our approach via deployment on an ensemble of devices (phones, PCs, cloud virtual machines, etc.), and show that it supports rich policies and high data volumes using simulations and real data based on personal usage in our group.","Title: Unifying synchronization and events in a multicore OS\nYear: 2012\n\nAbstract: In this paper, we argue that an operating system structured as a distributed system needs a coordination and a name service to make OS services work correctly. While a distributed structure allows applying algorithms from the distributed field, it also suffers from similar problems like synchronization, naming, distributed locking and coordination of service instances. Octopus, our solution to this problem in the context of a real OS, provides an easy-to-use, high-level, uniform coordination service with events at reasonable performance. Based on this service, we describe three real use cases: device management, OS service registry and boot-up coordination.","Title: Deployment of Query Plans on Multicores.\nYear: 2014\n\nAbstract: Efficient resource scheduling of multithreaded software on multicore hardware is difficult given the many parameters involved and the hardware heterogeneity of existing systems. In this paper we explore the efficient deployment of query plans over a multicore machine. We focus on shared query systems, and implement the proposed ideas using SharedDB. The goal of the paper is to explore how to deliver maximum performance and predictability, while minimizing resource utilization when deploying query plans on multicore machines. We propose to use resource activity vectors to characterize the behavior of individual database operators. We then present a novel deployment algorithm which uses these vectors together with dataflow information from the query plan to optimally assign relational operators to physical cores. Experiments demonstrate that this approach significantly reduces resource requirements while preserving performance and is robust across different server architectures.","Title: Cosh: Clear OS Data Sharing In An Incoherent World.\nYear: 2014\n\nAbstract: This paper tackles the problem of providing familiar OS abstractions for I/O (such as pipes, network sockets, and a shared file system) to applications on heterogeneous cores including accelerators, co-processors, and offload engines. We aim to isolate the implementation of these facilities from the details of a platform's memory architecture, which is likely to include a combination of cache-coherent shared memory, non-cache-coherent shared memory, and non-shared memory, all in the same system. We propose coherence-oblivious sharing (Cosh), a new OS abstraction that provides inter-process sharing with clear semantics on such diverse hardware. We have implemented a prototype of Cosh for the Barrelfish multikernel. We describe how to build common OS functionality using Cosh, and evaluate its performance on a heterogeneous system consisting of commodity cache-coherent CPUs and prototype Intel many-core co-processors.","Title: Grok the data center\nYear: 2014\n\nAbstract: In this paper, we claim that a vital part of managing a large enterprise data center is creating a model which encompasses all layers of the hardware and software stack, from network links up to application placement and configurations, and moreover that this model must be maintained in real time based on monitoring and trace data. Such a claim is not necessarily contentious, but we further argue that existing approaches to managing such facilities suffer from a lack of abstraction of the problem that limits their applicability: enterprise data centers are sufficiently diverse that it is hard to identify a reusable component that can express the complex range of queries that must be served in real time. The problem is less that of working out what data to gather, and how to process it, than it is one of building a flexible, reusable infrastructure above which to perform such processing. We therefore characterize the problem, and the systems challenges entailed by such a \\\"RDBMS for system management\\\". Finally, we report on our early exploration of the problem space, showing that even a limited version of such an engine can deliver useful value to a real, large-scale, production data center.","Title: Decoupling cores, kernels, and operating systems\nYear: 2014\n\nAbstract: We present Barrelfish/DC, an extension to the Barrelfish OS which decouples physical cores from a native OS kernel, and furthermore the kernel itself from the rest of the OS and application state. In Barrelfish/DC, native kernel code on any core can be quickly replaced, kernel state moved between cores, and cores added and removed from the system transparently to applications and OS processes, which continue to execute. Barrelfish/DC is a multikernel with two novel ideas: the use of boot drivers to abstract cores as regular devices, and a partitioned capability system for memory management which externalizes core-local kernel state. We show by performance measurements of real applications and device drivers that the approach is practical enough to be used for a number of purposes, such as online kernel upgrades, and temporarily delivering hard real-time performance by executing a process under a specialized, single-application kernel.","Title: Arrakis: the operating system is the control plane\nYear: 2014\n\nAbstract: Recent device hardware trends enable a new approach to the design of network server operating systems. In a traditional operating system, the kernel mediates access to device hardware by server applications to enforce process isolation as well as network and disk security. We have designed and implemented a new operating system, Arrakis, that splits the traditional role of the kernel in two. Applications have direct access to virtualized I/O devices, allowing most I/O operations to skip the kernel entirely, while the kernel is re-engineered to provide network and disk protection without kernel mediation of every operation. We describe the hardware and software changes needed to take advantage of this new abstraction, and we illustrate its power by showing improvements of 2 to 5 × in latency and 9 × throughput for a popular persistent NoSQL store relative to a well-tuned Linux implementation.","Title: Outlook on Operating Systems\nYear: 2016\n\nAbstract: Will OSs in 2025 still resemble the Unix-like consensus of today, or will a very different design achieve widespread adoption?","Title: Modeling NICs with Unicorn\nYear: 2013\n\nAbstract: NICs are increasingly complex and diverse, offering a wide range of hardware functionality to aid network protocol processing. Harnessing the power of NIC hardware requires the ability to control and reason about a variety of different feature sets in the network stack. Towards this goal, we propose Unicorn, a language for describing modern NICs. Unicorn offers a simple set of abstractions for modeling both NIC functionality and the state of a protocol stack. To evaluate its expressivity and potential, we present a non-trivial model for the Intel i82599 10GbE NIC, and an algorithm that uses graph embedding to optimize the use of NIC hardware in the network stack.","Title: SpaceJMP: Programming with Multiple Virtual Address Spaces.\nYear: 2016\n\nAbstract: Memory-centric computing demands careful organization of the virtual address space, but traditional methods for doing so are inflexible and inefficient. If an application wishes to address larger physical memory than virtual address bits allow, if it wishes to maintain pointer-based data structures beyond process lifetimes, or if it wishes to share large amounts of memory across simultaneously executing processes, legacy interfaces for managing the address space are cumbersome and often incur excessive overheads. We propose a new operating system design that promotes virtual address spaces to first-class citizens, enabling process threads to attach to, detach from, and switch between multiple virtual address spaces. Our work enables data-centric applications to utilize vast physical memory beyond the virtual range, represent persistent pointer-rich data structures without special pointer representations, and share large amounts of memory between processes efficiently.\n\nWe describe our prototype implementations in the DragonFly BSD and Barrelfish operating systems. We also present programming semantics and a compiler transformation to detect unsafe pointer usage. We demonstrate the benefits of our work on data-intensive applications such as the GUPS benchmark, the SAMTools genomics workflow, and the Redis key-value store.","Title: Shoal: Smart Allocation and Replication of Memory For Parallel Programs\nYear: 2015\n\nAbstract: Modern NUMA multi-core machines exhibit complex latency and throughput characteristics, making it hard to allocate memory optimally for a given program's access patterns. However, sub-optimal allocation can significantly impact performance of parallel programs. We present an array abstraction that allows data placement to be automatically inferred from program analysis, and implement the abstraction in Shoal, a runtime library for parallel programs on NUMA machines. In Shoal, arrays can be automatically replicated, distributed, or partitioned across NUMA domains based on annotating memory allocation statements to indicate access patterns. We further show how such annotations can be automatically provided by compilers for high-level domainspecific languages (for example, the Green-Marl graph language). Finally, we show how Shoal can exploit additional hardware such as programmable DMA copy engines to further improve parallel program performance. We demonstrate significant performance benefits from automatically selecting a good array implementation based on memory access patterns and machine characteristics. We present two case-studies: (i) Green-Marl, a graph analytics workload using automatically annotated code based on information extracted from the high-level program and (ii) a manually-annotated version of the PARSEC Streamcluster benchmark.","Title: Distributing the Data Plane for Remote Storage Access\nYear: 2015\n\nAbstract: Sub-microsecond network and memory latencies require fast user-level access to local and remote storage. While user-level access to local storage has been demonstrated recently, it does currently not extend to serverless parallel systems in datacenter environments. We propose direct user-level access to remote storage in a distributed setting, unifying fast data access and high-performance remote memory access programming. We discuss a minimal hardware extension of the IOMMU to enable direct remote storage access. In order to maintain optimal performance in the system, we use epoch-based accesses to allow fine-tuning of atomicity, consistency, isolation, and durability semantics. We also address the problem of user-managed coherent caching. Finally, we briefly discuss the design of DiDAFS, a Distributed Direct Access File System that enables efficient data analytics use-cases such as buffered producer-consumer synchronization and key-value stores as well as deeper integration of storage into high performance computing applications.","Title: Not Your Parents' Physical Address Space\nYear: 2015\n\nAbstract: A physical memory address is no longer the stable concept it was. We demonstrate how modern computer systems from rack-scale to SoCs have multiple physical address spaces, which overlap and intersect in complex, dynamic ways, and may be too small to even address available memory in the near future. We present a new model of representing and interpreting physical addresses in a machine for the purposes of memory management, and outline an implementation of the model in a memory system based on capabilities which can handle arbitrary translations between physical address spaces and still globally manage system memory. Finally, we point out future challenges in managing physical memory, of which our model and design are merely a foundation.","Title: Explaining Outputs in Modern Data Analytics.\nYear: 2016\n\nAbstract: We report on the design and implementation of a general framework for interactively explaining the outputs of modern data-parallel computations, including iterative data analytics. To produce explanations, existing works adopt a naive backward tracing approach which runs into known issues; naive backward tracing may identify: (i) too much information that is difficult to process, and (ii) not enough information to reproduce the output, which hinders the logical debugging of the program. The contribution of this work is twofold. First, we provide methods to effectively reduce the size of explanations based on the first occurrence of a record in an iterative computation. Second, we provide a general method for identifying explanations that are sufficient to reproduce the target output in arbitrary computations -- a problem for which no viable solution existed until now. We implement our approach on differential dataflow, a modern high-throughput, low-latency dataflow platform. We add a small (but extensible) set of rules to explain each of its data-parallel operators, and we implement these rules as differential dataflow operators themselves. This choice allows our implementation to inherit the performance characteristics of differential dataflow, and results in a system that efficiently computes and updates explanatory inputs even as the inputs of the reference computation change. We evaluate our system with various analytic tasks on real datasets, and we show that it produces concise explanations in tens of milliseconds, while remaining faster -- up to two orders of magnitude -- than even the best implementations that do not support explanations.","Title: So many performance events, so little time.\nYear: 2016\n\nAbstract: Many modern workloads are a heterogeneous mix of parallel applications running on machines with complex processors and memory systems. These workloads often interfere with each other by contending for the limited set of resources in a machine, leading to performance degradation. In principle, one can use hardware performance counters to characterize the root causes of interference inside a modern machine. However, we show that current facilities in today's operating systems mean that such an analysis requires careful consideration of the intricacies of specific hardware counters, domain knowledge about the applications, and a deep understanding of the underlying hardware architecture. In contrast, we present the design of Glatt, which is able to automatically identify the root causes of interference in a machine online. Glatt can expose this information to a runtime or the operating system for predicting better thread assignments or for dynamically increasing/decreasing parallelism within each runtime.","Title: Customized OS support for data-processing.\nYear: 2016\n\nAbstract: For decades, database engines have found the generic interfaces offered by the operating systems at odds with the need for efficient utilization of hardware resources. As a result, most engines circumvent the OS and manage hardware directly. With the growing complexity and heterogeneity of modern hardware, database engines are now facing a steep increase in the complexity they must absorb to achieve good performance. Taking advantage of recent proposals in operating system design, such as multi-kernels, in this paper we explore the development of a light weight OS kernel tailored for data processing and discuss its benefits for simplifying the design and improving the performance of data management systems.","Title: Machine-Aware Atomic Broadcast Trees for Multicores.\nYear: 2016\n\nAbstract: The performance of parallel programs on multicore machines often critically depends on group communication operations like barriers and reductions being highly tuned to hardware, a task requiring considerable developer skill. Smelt is a library that automatically builds efficient inter-core broadcast trees tuned to individual machines, using a machine model derived from hardware registers plus micro-benchmarks capturing the low-level machine characteristics missing from vendor specifications. Experiments on a wide variety of multicore machines show that near-optimal tree topologies and communication patterns are highly machine-dependent, but can nevertheless be derived by Smelt and often further improve performance over well-known static topologies. Furthermore, we show that the broadcast trees built by Smelt can be the basis for complex group operations like global barriers or state machine replication, and that the hardware-tuning provided by the underlying tree is sufficient to deliver as good or better performance than state-of-the-art approaches: the higher-level operations require no further hardware optimization.","Title: Formalizing Memory Accesses And Interrupts\nYear: 2017\n\nAbstract: The hardware/software boundary in modern heterogeneous multicore computers is increasingly complex, and diverse across different platforms. A single memory access by a core or DMA engine traverses multiple hardware translation and caching steps, and the destination memory cell or register often appears at different physical addresses for different cores. Interrupts pass through a complex topology of interrupt controllers and remappers before delivery to one or more cores, each with specific constraints on their configurations. System software must not only correctly understand the specific hardware at hand, but also configure it appropriately at runtime. We propose a formal model of address spaces and resources in a system that allows us to express and verify invariants of the system's runtime configuration, and illustrate (and motivate) it with several real platforms we have encountered in the process of OS implementation.","Title: Online Reconstruction of Structural Information from Datacenter Logs.\nYear: 2017\n\nAbstract: Well-run datacenter application architectures are heavily instrumented to provide detailed traces of messages and remote invocations. Reconstructing user sessions, call graphs, transaction trees, and other structural information from these messages, a process known as sessionization, is the foundation for a variety of diagnostic, profiling, and monitoring tasks essential to the operation of the datacenter. We present the design and implementation of a system which processes log streams at gigabits per second and reconstructs user sessions comprising millions of transactions per second in real time with modest compute resources, while dealing with clock skew, message loss, and other real-world phenomena that make such a task challenging. Our system is based on the Timely Dataflow framework for low latency, data-parallel computation, and we demonstrate its utility with a number of use-cases and traces from a large, operational, mission-critical enterprise data center.","Title: Separating Translation from Protection in Address Spaces with Dynamic Remapping\nYear: 2017\n\nAbstract: It is time to reconsider memory protection. The emergence of large non-volatile main memories, scalable interconnects, and rack-scale computers running large numbers of small \\\"micro services\\\" creates significant challenges for memory protection based solely on MMU mechanisms. Central to this is a tension between protection and translation: optimizing for translation performance often comes with a cost in protection flexibility. We argue that a key-based memory protection scheme, complementary to but separate from regular page-level translation, is a better match for this new world. We present MaKC, a new architecture which combines two levels of capability-based protection to scale fine-grained memory protection at both user and kernel level to large numbers of protection domains without compromising efficiency at scale or ease of revocation.","Title: Towards Correct-by-Construction Interrupt Routing on Real Hardware.\nYear: 2017\n\nAbstract: In this paper we address the problem of correctly configuring interrupts. The interrupt subsystem of a computer is increasingly complex: a zoo of different controllers with varying constraints and capabilities form a network with limited connectivity. An OS which aspires to provable correctness must manage a limited set of interrupt vectors, delegate interrupts to device drivers and configure the controllers correctly. No well-specified approach exists.\n\nAs a foundation for applying language-level techniques like program sketching and synthesis to this problem, we present a formal model for interrupt routing which can capture all the system topologies and interrupt controllers we have encountered in the wild, show applications of such a model not possible with informal, ad-hoc approaches like DeviceTrees, and finally discuss an implementation based on the model which forms the new interrupt subsystem of the Barrelfish OS.","Title: Megaphone: latency-conscious state migration for distributed streaming dataflows\nYear: 2018\n\nAbstract: We design and implement Megaphone, a data migration mechanism for stateful distributed dataflow engines with latency objectives. When compared to existing migration mechanisms, Megaphone has the following differentiating characteristics: (i) migrations can be subdivided to a configurable granularity to avoid latency spikes, and (ii) migrations can be prepared ahead of time to avoid runtime coordination. Megaphone is implemented as a library on an unmodified timely dataflow implementation, and provides an operator interface compatible with its existing APIs. We evaluate Megaphone on established benchmarks with varying amounts of state and observe that compared to naïve approaches Megaphone reduces service latencies during reconfiguration by orders of magnitude without significantly increasing steady-state overhead.","Title: DeltaPath: dataflow-based high-performance incremental routing.\nYear: 2018\n\nAbstract: Routing controllers must react quickly to failures, reconfigurations and workload or policy changes, to ensure service performance and cost-efficient network operation. We propose a general execution model which views routing as an incremental data-parallel computation on a graph-based network model plus a continuous stream of network changes. Our approach supports different routing objectives with only minor re-configuration of its core algorithm, and easily accomodates dynamic user-defined routing policies. Moreover, our prototype demonstrates excellent performance: on Google Jupiter topology it reacts with a median time of 350ms to link failures and serves more than two million path requests per second each with latency under 1ms. This is three orders-of-magnitude faster than the popular ONOS open-source SDN controller.","Title: Rack-Scale Capabilities: Fine-Grained Protection for Large-Scale Memories\nYear: 2019\n\nAbstract: Rack-scale systems with large, shared, disaggregated, and persistent memory need solid protection and authorization techniques. Our solution uses a memoryside capability enforcement processor that gates memory accesses through extended capabilities, enables finegrained access control beyond a single address space, and minimally disrupts the programming model.","Title: Memory-Side Protection With a Capability Enforcement Co-Processor.\nYear: 2019\n\nAbstract: Byte-addressable nonvolatile memory (NVM) blends the concepts of storage and memory and can radically improve data-centric applications, from in-memory databases to graph processing. By enabling large-capacity devices to be shared across multiple computing elements, fabric-attached NVM changes the nature of rack-scale systems and enables short-latency direct memory access while retaining data persistence properties and simplifying the software stack.\n\nAn adequate protection scheme is paramount when addressing shared and persistent memory, but mechanisms that rely on virtual memory paging suffer from the tension between performance (pushing toward large pages) and protection granularity (pushing toward small pages). To address this tension, capabilities are worth revisiting as a more powerful protection mechanism, but the long time needed to introduce new CPU features hampers the adoption of schemes that rely on instruction-set architecture support.\n\nThis article proposes the Capability Enforcement Co-Processor (CEP), a programmable memory controller that implements fine-grain protection through the capability model without requiring instruction-set support in the application CPU. CEP decouples capabilities from the application CPU instruction-set architecture, shortens time to adoption, and can rapidly evolve to embrace new persistent memory technologies, from NVDIMMs to native NVM devices, either locally connected or fabric attached in rack-scale configurations. CEP exposes an application interface based on memory handles that get internally converted to extended-pointer capabilities.\n\nThis article presents a proof of concept implementation of a distributed object store (Redis) with CEP. It also demonstrates a capability-enhanced file system (FUSE) implementation using CEP. Our proof of concept shows that CEP provides fine-grain protection while enabling direct memory access from application clients to the NVM, and that by doing so opens up important performance optimization opportunities (up to 4× reduction in latency in comparison to software-based security enforcement) without compromising security.\n\nFinally, we also sketch how a future hybrid model could improve the initial implementation by delegating some CEP functionality to a CHERI-enabled processor.","Title: Generating correct initial page tables from formal hardware descriptions\nYear: 2021\n\nAbstract: ABSTRACTModern hardware platforms are increasingly complex and heterogeneous. System software uses a hodgepodge of different mechanisms and representations to express the memory topology of the target platform. Considerable maintenance effort is required to keep them in sync while often sharing is impossible due to hard-coded values. Incorrect platform-specific values in the hardware initialization sequence can lead to security critical and hard-to-find bugs because of misconfigured translation hardware, inaccessible devices, or the use of bad pointers. We present a better way for system software to express and initialize memory hardware. We adopt an existing, powerful hardware description language, and efficiently compile it to generate correct initial page tables and memory maps for OS kernels and firmware from a single system description. We evaluate our system on multiple architectures and platforms, and demonstrate that we can use the generated data structures to successfully initialize translation hardware, devices, memory maps, and allocators enabling easy support of new hardware platforms.","Title: Declarative Power Sequencing\nYear: 2021\n\nAbstract: Modern computer server systems are increasingly managed at a low level by baseboard management controllers (BMCs). BMCs are processors with access to the most critical parts of the platform, below the level of OS or hypervisor, including control over power delivery to every system component. Buggy or poorly designed BMC software not only poses a security threat to a machine, it can permanently render the hardware inoperative. Despite this, there is little published work on how to rigorously engineer the power management functionality of BMCs so as to prevent this happening.This article takes a first step toward putting BMC software on a sound footing by specifying the hardware environment and the constraints necessary for safe and correct operation. This is best accomplished through automation: correct-by-construction power control sequences can be efficiently generated from a simple, trustworthy model of the platform's power tree that incorporates the sequencing requirements and safe voltage ranges of all components.We present both a modeling language for complex power-delivery networks and a tool to automatically generate safe, efficient power sequences for complex modern platforms. This not only increases the trustworthiness of a hitherto opaque yet critical element of platform firmware: regulator and chip power models are significantly simpler to produce than hand-written power sequences. This, combined with model reuse for common components, reduces both time and cost associated with platform bring-up for new hardware.We evaluate our tool using a new high-performance 2-socket server platform with >100W per socket TDP, tight voltage limits and 25 distinct power regulators needing configuration, showing both fast (<10s) tool runtime, and correct power sequencing of a live system."],"string":"[\n \"Title: Hype and Virtue\\nYear: 2007\\n\\nAbstract: In this paper, we question whether hypervisors are really acting as a disruptive force in OS research, instead argu- ing that they have so far changed very little at a techni- cal level. Essentially, we have retained the conventional Unix-like OS interface and added a new ABI based on PC hardware which is highly unsuitable for most purposes. Despite commercial excitement, focus on hypervisor de- sign may be leading OS research astray. However, adopt- ing a different approach to virtualization and recognizing its value to academic research holds the prospect of open- ing up kernel research to new directions.\",\n \"Title: Implementing declarative overlays\\nYear: 2005\\n\\nAbstract: Overlay networks are used today in a variety of distributed systems ranging from file-sharing and storage systems to communication infrastructures. However, designing, building and adapting these overlays to the intended application and the target environment is a difficult and time consuming process.To ease the development and the deployment of such overlay networks we have implemented P2, a system that uses a declarative logic language to express overlay networks in a highly compact and reusable form. P2 can express a Narada-style mesh network in 16 rules, and the Chord structured overlay in only 47 rules. P2 directly parses and executes such specifications using a dataflow architecture to construct and maintain overlay networks. We describe the P2 approach, how our implementation works, and show by experiment its promising trade-off point between specification complexity and performance.\",\n \"Title: The Network Oracle\\nYear: 2005\\n\\nAbstract: This paper sets out a high-level research agenda aimed at bui lding a collaborative, global end-system monitoring and information infrastructure for the Internet's core sta te. We argue that such a system is beneficial, feasible and timely, representing an important area for engagement a cross database and networking technologies. We start by hypothesizing the benefits to the Internet of a \\\"Netw ork Oracle\\\" that could answer real-time questions about the global state of the network. We then argue that data base and networking technology should make it possible to provide a useful approximation of such an oracle today, gathering information from a large number of end hosts and delivering useful views to each end system. We f urther argue that this can and should be done in a decentralized fashion. We provide an outline of such a syste m: it employs sensing agents, along with a distributed query/trigger/dissemination engine, and possible attrac tive end-user applications. A key point of our discussion is the timeliness and importance of a grassroots agenda for Int ernet monitoring and state-sharing. While significant social and economic barriers to deploying a centralized, pu blic Internet monitoring infrastructure exist, there are corresponding social and economic incentives for a collabo rative approach.\",\n \"Title: The PlanetLab Platform\\nYear: 2005\\n\\nAbstract: PlanetLab is an extensively used, global, community-maintained platform for researchers to develop, deploy, and evaluate widely-distributed\\n applications such as Peer-to-Peer systems. Because of PlanetLab’s shared nature, and its unusual design goal of continuous\\n replacement of components by the research community, it can also be viewed somewhat as a Peer-to-Peer system (and an ongoing\\n research project) itself. This chapter describes PlanetLab’s goals and origins, and discusses in detail the design principles\\n that have governed its development and growth so far. It also discusses some of the methodological issues in performing research\\n using a platform like PlanetLab - what can be learned from experimentation on PlanetLab, and what research claims can be validated\\n by the system.\",\n \"Title: Dexferizer: a service for data transfer optimization\\nYear: 2011\\n\\nAbstract: We present an approach to optimizing the transfer of data objects within a user's collection of computers and personal devices, subject to a variety of user-defined quality metrics such as cost, power consumption, and latency. By abstracting object transfer as a high-level service, and employing declarative networking techniques to cast object transfer as a constrained optimization problem, we show how to transparently exploit techniques as diverse as swarming and multi-hop transfer through virtual machines. Using a data replication system as a driving application, we demonstrate that our approach can easily accommodate flexible policies not easily implemented with existing solutions, and can thereby result in savings in time, power, bandwidth, or other costs.\",\n \"Title: Application level ballooning for efficient server consolidation\\nYear: 2013\\n\\nAbstract: Systems software like databases and language runtimes typically manage memory themselves to exploit application knowledge unavailable to the OS. Traditionally deployed on dedicated machines, they are designed to be statically configured with memory sufficient for peak load. In virtualization scenarios (cloud computing, server consolidation), however, static peak provisioning of RAM to applications dramatically reduces the efficiency and cost-saving benefits of virtualization. Unfortunately, existing memory \\\"ballooning\\\" techniques used to dynamically reallocate physical memory between VMs badly impact the performance of applications which manage their own memory. We address this problem by extending ballooning to applications (here, a database engine and Java runtime) so that memory can be efficiently and effectively moved between virtualized instances as the demands of each change over time. The results are significantly lower memory requirements to provide the same performance guarantees to a collocated set of VM running such applications, with minimal overhead or intrusive changes to application code.\",\n \"Title: A better way to negotiate for testbed resources\\nYear: 2011\\n\\nAbstract: Resource allocation is an increasing challenge for distributed network testbeds as computational and network resources are involved. Testbed designers have moved to a query-based model: clients provide a declarative description of their desired resources, and the provider allocate specific resources to meet the request. In this paper, we describe an new approach to negotiate testbed resources between clients and testbed providers: the clients specify their requests as constraints, and the providers reply with resource allocations expressed also as declarative set of constraints on resources. This gives providers more flexibility in late-binding of resources to requests, and opens up a wide design space to optimize resource allocation for efficiency, cost, utilization, or other metrics. Our simple first experiments suggest that the late-binding of resources enabled by representing resource reservation as constraints achieves better network resource utilization compared to the fixed assignment solution.\",\n \"Title: R-OSGi: distributed applications through software modularization\\nYear: 2007\\n\\nAbstract: In this paper we take advantage of the concepts developed for centralized module management, such as dynamic loading and unloading of modules, and show how they can be used to support the development and deployment of distributed applications. We do so through R-OSGi, a distributed middleware platform that extends the centralized, industry-standard OSGi specification to support distributed module management. To the developer, R-OSGi looks like a conventional module management tool. However, at deployment time, R-OSGi can be used to turn the application into a distributed application by simply indicating where the different modules should be deployed. At run time, R-OSGi represents distributed failures as module insertion and withdrawal operations so that the logic to deal with failures is the same as that employed to deal with dependencies among software modules. In doing so, R-OSGi greatly simplifies the development of distributed applications with no performance cost. In the paper we describe R-OSGi and several use cases. We also show with extensive experiments that R-OSGi has a performance comparable or better than that of RMI or UPnP, both commonly used distribution mechanisms with far less functionality than R-OSGi.\",\n \"Title: BFT protocols under fire\\nYear: 2008\\n\\nAbstract: Much recent work on Byzantine state machine replication focuses on protocols with improved performance under benign conditions (LANs, homogeneous replicas, limited crash faults), with relatively little evaluation under typical, practical conditions (WAN delays, packet loss, transient disconnection, shared resources). This makes it difficult for system designers to choose the appropriate protocol for a real target deployment. Moreover, most protocol implementations differ in their choice of runtime environment, crypto library, and transport, hindering direct protocol comparisons even under similar conditions. We present a simulation environment for such protocols that combines a declarative networking system with a robust network simulator. Protocols can be rapidly implemented from pseudocode in the high-level declarative language of the former, while network conditions and (measured) costs of communication packages and crypto primitives can be plugged into the latter. We show that the resulting simulator faithfully predicts the performance of native protocol implementations, both as published and as measured in our local network. We use the simulator to compare representative protocols under identical conditions and rapidly explore the effects of changes in the costs of crypto operations, workloads, network conditions and faults. For example, we show that Zyzzyva outperforms protocols like PBFT and Q/U undermost but not all conditions, indicating that one-size-fits-all protocols may be hard if not impossible to design in practice.\",\n \"Title: Querying at Internet scale\\nYear: 2004\\n\\nAbstract: We are developing a distributed query processor called PIER, which is designed to run on the scale of the entire Internet. PIER utilizes a Distributed Hash Table (DHT) as its communication substrate in order to achieve scalability, reliability, decentralized control, and load balancing. PIER enhances DHTs with declarative and algebraic query interfaces, and underneath those interfaces implements multihop, in-network versions of joins, aggregation, recursion, and query/result dissemination. PIER is currently being used for diverse applications, including network monitoring, keyword-based filesharing search, and network topology mapping. We will demonstrate PIER's functionality by showing system monitoring queries running on PlanetLab, a testbed of over 300 machines distributed across the globe.\",\n \"Title: The Architecture of PIER: an Internet-Scale Query Processor\\nYear: 2005\\n\\nAbstract: Abstract: This paper presents the architecture of PIER, an Internetscalequery engine we have been building over the last threeyears. PIER is the first general-purpose relational query processortargeted at a peer-to-peer (p2p) architecture of thousandsor millions of participating nodes on the Internet. Itsupports massively distributed, database-style dataflows forsnapshot and continuous queries. It is intended to serve as abuilding block for a diverse set of Internet-scale informationcentric...\",\n \"Title: A blueprint for introducing disruptive technology into the Internet\\nYear: 2003\\n\\nAbstract: This paper argues that a new class of geographically distributed network services is emerging, and that the most effective way to design, evaluate, and deploy these services is by using an overlay-based testbed. Unlike conventional network testbeds, however, we advocate an approach that supports both researchers that want to develop new services, and clients that want to use them. This dual use, in turn, suggests four design principles that are not widely supported in existing testbeds: services should be able to run continuously and access a slice of the overlay's resources, control over resources should be distributed, overlay management services should be unbundled and run in their own slices, and APIs should be designed to promote application development. We believe a testbed that supports these design principles will facilitate the emergence of a new service-oriented network architecture. Towards this end, the paper also briefly describes PlanetLab, an overlay network being designed with these four principles in mind.\",\n \"Title: Operating system support for planetary-scale network services\\nYear: 2004\\n\\nAbstract: PlanetLab is a geographically distributed overlay network designed to support the deployment and evaluation of planetary-scale network services. Two high-level goals shape its design. First, to enable a large research community to share the infrastructure, PlanetLab provides distributed virtualization, whereby each service runs in an isolated slice of PlanetLab's global resources. Second, to support competition among multiple network services, PlanetLab decouples the operating system running on each node from the network-wide services that define PlanetLab, a principle referred to as unbundled management. This paper describes how Planet-Lab realizes the goals of distributed virtualization and unbundled management, with a focus on the OS running on each node.\",\n \"Title: Making the world (of communications) a different place\\nYear: 2005\\n\\nAbstract: How might the computing and communications world be materially different in 10 to 15 years, and how might we define a research agenda that would get us to that world?\",\n \"Title: Sophia: an Information Plane for networked systems\\nYear: 2004\\n\\nAbstract: This paper motivates and describes an example network Information Plane, called Sophia, currently deployed on PlanetLab. Sophia is a distributed system that collects, stores, propagates, aggregates, and reacts to observations about the network's current conditions. Sophia's approach is novel: it can be viewed as a multi-user distributed expression evaluator in which sensors and actuators form the ground terms, and statements take on the complete expressiveness of a logic language like Prolog. This paper argues that this approach has several advantages in managing and controlling a complex, federated, and evolving network: (1) a declarative logic language provides a natural way to express the kinds of statements that are common to this application domain, through temporal and positional logic rules, facts and expressions; and (2) distributed evaluation of such logic expressions provides many opportunities for performance optimization yielding an efficient system.\",\n \"Title: Transaction-Based Charging in Mnemosyne: A Peer-to-Peer Steganographic Storage System\\nYear: 2002\\n\\nAbstract: Mnemosyne is a peer-to-peer steganographic storage system: one in which the existence ofa user's files cannot be verified without a key. This paper applies the techniques used in Mnemosyne - erasure codes and anonymous block writing - to move most ofthe administrative overhead of a commercial storage service over to the client, resulting in cost savings for the service provider.The contribution ofthis paper is to present a radically alternative way of charging for storage services. In place of renting some amount of space for some period of time, systems like Mnemosyne allow more flexible billing models closer to those proposed for network bandwidth, including versions of congestion pricing. We show how a reliable, commercial storage service using is feasible, and examine the details of the tradeoff it offers compared with conventional storage services.\",\n \"Title: Multicore OS benchmarks: we can do better\\nYear: 2011\\n\\nAbstract: Current multicore OS benchmarks do not provide workloads that sufficiently reflect real-world use: they typically run a single application, whereas real workloads consist of multiple concurrent programs. In this paper we show that this lack of mixed workloads leads to benchmarks that do not fully exercise the OS and are therefore inadequate at predicting real-world behavior. This implies that effective multicore OS benchmarks must includemixed workloads, but the main design challenge is choosing an appropriate mix. We present a principled approachwhich treats benchmark design as an optimization problem. Our solution leads to a workload mix that uses as much of a system's resources as possible, while also selecting applications whose performance is most sensitive to the availability of those resources.\",\n \"Title: PlanetLab: an overlay testbed for broad-coverage services\\nYear: 2003\\n\\nAbstract: PlanetLab is a global overlay network for developing and accessing broad-coverage network services. Our goal is to grow to 1000 geographically distributed nodes, connected by a disverse collection of links. PlanetLab allows multiple service to run concurrently and continuously, each in its own slice of PlanetLab. This paper discribes our initial implementation of PlanetLab, including the mechanisms used to impelment virtualization, and the collection of core services used to manage PlanetLab.\",\n \"Title: Linkage in the Nemesis single address space operating system\\nYear: 1994\\n\\nAbstract: The recent interest in single address space operating systems has resulted in a number of papers, most of which gloss over the issues of linking programs to run in multiple protection domains. Some of the confusion about 64-bit address spaces is due to the almost pervasive use of UNIX and UNIX-like operating systems (such as Mach, Chorus and Amoeba) and languages with poor enforcement of abstraction like C and C++.This paper describes some of the linkage structure of Nemesis, a multi-service operating system being developed as part of the Pegasus project. Nemesis provides a simple and efficient mechanism for program linkage which provides rich sharing of text at a level of individual object classes.\",\n \"Title: Resource overbooking and application profiling in shared hosting platforms\\nYear: 2002\\n\\nAbstract: In this paper, we present techniques for provisioning CPU and network resources in shared hosting platforms running potentially antagonistic third-party applications. The primary contribution of our work is to demonstrate the feasibility and benefits of overbooking resources in shared platforms, to maximize the platform yield: the revenue generated by the available resources. We do this by first deriving an accurate estimate of application resource needs by profiling applications on dedicated nodes, and then using these profiles to guide the placement of application components onto shared nodes. By overbooking cluster resources in a controlled fashion, our platform can provide performance guarantees to applications even when overbooked, and combine these techniques with commonly used QoS resource allocation mechanisms to provide application isolation and performance guarantees at run-time. When compared to provisioning based on the worst-case, the efficiency (and consequently revenue) benefits from controlled overbooking of resources can be dramatic. Specifically, experiments on our Linux cluster implementation indicate that overbooking resources by as little as 1% can increase the utilization of the cluster by a factor of two, and a 5% overbooking yields a 300--500% improvement, while still providing useful resource guarantees to applications.\",\n \"Title: Plutarch: an argument for network pluralism\\nYear: 2003\\n\\nAbstract: It is widely accepted that the current Internet architecture is insufficient for the future: problems such as address space scarcity, mobility and non-universal connectivity are already with us, and stand to be exacerbated by the explosion of wireless, ad-hoc and sensor networks. Furthermore, it is far from clear that the ubiquitous use of standard transport and name resolution protocols will remain practicable or even desirable.In this paper we propose Plutarch, a new inter-networking architecture. It subsumes existing architectures such as that determined by the Internet Protocol suite, but makes explicit the heterogeneity that contemporary inter-networking schemes attempt to mask. To handle this heterogeneity, we introduce the notions of context and interstitial function, and describe a supporting architecture. We discuss the benefits, present some potential scenarios, and consider the research challenges posed.\",\n \"Title: Supercomputing operating systems: a naive view from over the fence\\nYear: 2012\\n\\nAbstract: To exaggerate unfairly, from the perspective of mainstream OS research, the supercomputing community has a very different idea of the role (and appropriate design) of an OS. HPC people regard the OS as an annoying source of noise, whereas the former crowd see it as a thing of wondrous beauty and elegance, a sine qua non of usable everyday computing. This situation has existed without serious conflict erupting for years: OS researchers worried about PCs with one core (or, at most, a handful of cores) running a general-purpose OS and supporting a dynamic, bursty, diverse mix of hundreds of interactive, long-running, soft-realtime and/or background processes. Supercomputing people wanted one, highly parallel, program to finish as soon as possible so they could get on to the next one. With multicore, this all changed: highly parallel tasks will be the norm for future general-purpose computing. In 2007, my colleagues and I eagerly embarked on a new research OS for multicore computing, and looked forward to applying long-ignored (in our field) results from the HPC realm to our system. It didn't quite work out that way. In this talk I will look at what we found to be common to the two fields, and what we didn't, and speculate on where this might be going. I think there is a useful conversation to be had, and I'd like to help revive it.\",\n \"Title: Friday: global comprehension for distributed replay\\nYear: 2007\\n\\nAbstract: Debugging and profiling large-scale distributed applications is a daunting task. We present Friday, a system for debugging distributed applications that combines deterministic replay of components with the power of symbolic, low-level debugging and a simple language for expressing higher-level distributed conditions and actions. Friday allows the programmer to understand the collective state and dynamics of a distributed collection of coordinated application components. To evaluate Friday, we consider several distributed problems, including routing consistency in overlay networks, and temporal state abnormalities caused by route flaps. We show via micro-benchmarks and larger-scale application measurement that Friday can be used interactively to debug large distributed applications under replay on common hardware.\",\n \"Title: CLANGER: an interpreted systems programming language\\nYear: 1995\\n\\nAbstract: Clanger is a powerful, yet simple, command language for the Nemesis operating system. It uses runtime type information to interface directly with operating system components. Clanger is a combination of command-line interpreter, scripting language, debugger and prototyping tool. This paper describes why such a language is possible, how it is being implemented, and outlines the language as it currently stands.\",\n \"Title: Towards realistic benchmarks for virtual infrastructure resource allocators\\nYear: 2012\\n\\nAbstract: The rise of virtual infrastructures has renewed research interest in the design and implementation of resource allocators which allow users to reserve or use combinations of (virtual) nodes, switches, and network links on a variety of virtual infrastructures ranging from network testbeds to cloud computing facilities. However, since the area is still relatively new, work on resource allocation lacks realistic benchmarks. In this paper, we suggest how to generate realistic benchmarks for virtual infrastructure, and present our workload generator as an example, which is based on a 5-year trace of experiments submitted to the popular Emulab testbed. We demonstrate the importance as well as potential applications of such benchmarks by using the generated test workload in various evaluation scenarios.\",\n \"Title: Building, deploying, and monitoring distributed applications with Eclipse and R-OSGI\\nYear: 2007\\n\\nAbstract: Designing and testing distributed applications is still a difficult task that requires in-depth knowledge about networking issues. Eclipse is, among other things, a powerful and widely used IDE for the development of complex applications, in particular modular applications for the OSGi framework. Our R-OSGi middleware supports the seamless distribution of OSGi applications along the boundaries of services. By combining R-OSGi with Eclipse into the R-OSGi Deployment Tool, we give developers a tool that automatically handles distribution in a transparent way and integrates the capabilities of R-OSGi into the Eclipse workflow. With this tool, building, deploying, and monitoring distributed applications is as easy as writing OSGi applications in Eclipse and using a graphical editor to visually create distributed deployments of the modules. The tool can also be used to great effect by researchers to test and benchmark distributed applications and for education purposes.\",\n \"Title: InfoSpect: using a logic language for system health monitoring in distributed systems\\nYear: 2002\\n\\nAbstract: Dependable systems cannot be built without a monitoring and management component. In this paper we propose using a wide variety of information gathering tools coupled with custom scripts and a Prolog language engine to aggregate information from multiple sources. Complex queries, difficult to express in standard database languages, can then be used to answer questions about the system (e.g. the health of individual components) or to discover contradictions (e.g. inconsistent configurations). We describe our prototype implementation and present some early results.\",\n \"Title: Mnemosyne: Peer-to-Peer Steganographic Storage\\nYear: 2002\\n\\nAbstract: We present the design of Mnemosyne, a peer-to-peer steganographic storage service. Mnemosyne provides a high level of privacy and plausible deniability by using a large amount of shared distributed storage to hide data. Blocks are dispersed by secure hashing, and loss codes used for resiliency. We discuss the design of the system, and the challenges posed by traffic analysis.\",\n \"Title: We need to talk about NICs\\nYear: 2013\\n\\nAbstract: Operating systems fail both to efficiently exploit, and to effeectively manage, the considerable hardware resources of modern network interface controllers. We survey the kinds of hardware facilities available and their applicability, and then investigate (and critique) the reasons why OS designers eschew core support for such features. We then describe Dragonet, a new network stack design based on explicit descriptions of NIC capabilities, aimed at making the best use of today's and tomorrow's networking hardware. Dragonet represents both the physical capabilities of the network hardware and the current protocol state of the machine as dataflow graphs. We then embed the former into the latter, instantiating the remainder in software.\",\n \"Title: Preventing Internet denial-of-service with capabilities\\nYear: 2004\\n\\nAbstract: In this paper, we propose a new approach to preventing and constraining denial-of-service (DoS) attacks. Instead of being able to send anything to anyone at any time, in our architecture, nodes must first obtain \\\"permission to send\\\" from the destination; a receiver provides tokens, or capabilities, to those senders whose traffic it agrees to accept. The senders then include these tokens in packets. This enables verification points distributed around the network to check that traffic has been certified as legitimate by both endpoints and the path in between, and to cleanly discard unauthorized traffic. We show that our approach addresses many of the limitations of the currently popular approaches to DoS based on anomaly detection, traceback, and pushback. Further, we argue that our approach can be readily implemented in today's technology, is suitable for incremental deployment, and requires no more of a security infrastructure than that already needed to fix BGP's security weaknesses. Finally, our proposal facilitates innovation in application and networking protocols, something increasingly curtailed by existing DoS measures.\",\n \"Title: Design Considerations for Information Planes\\nYear: 2004\\n\\nAbstract: The concept of an information plane has emerged recently as an important part of large, decentral- ized systems that aspire to be self-managing, ranging from PlanetLab to the Internet itself. In this paper we describe what an information plane is, and report our experiences in developing and deploying an information plane for the Plan- etLab platform using the PIER distributed relational query processor. We recount the lessons we have learned from the experience, and the additional directions we intend to ex- plore in the PHI project, which aims at providing an infor- mation plane that can grow to serve a significant portion of the Internet.\",\n \"Title: Mind the gap: reconnecting architecture and OS research\\nYear: 2011\\n\\nAbstract: The goal of a computer system is to run an application workload securely, reliably, efficiently, and fast. A computer's hardware architecture and operating system exist to support this goal, and it would be nice if they cooperated as effectively as possible. Yet there is a growing gap between architectural research and OS research, which seems to be the result of poor communication about what actually matters. In this paper, we discuss this gap and what to do about it. We illustrate the opportunities for closing the gap using examples from some recent OS research.\",\n \"Title: New Resource Control Issues in Shared Clusters\\nYear: 2001\\n\\nAbstract: We claim that the renting of machine resources on clusters of servers introduces new systems challenges which are different from those hitherto encountered, either in multimedia systems or cluster-based computing. We characterize the requirements for such \\\"public computing platforms\\\" and discuss both how the scenario differs from more traditional multimedia resource control situations, and how some ideas from multimedia systems work can be reapplied in this new context. Finally, we discuss our ongoing work building a prototype public computing platform.\",\n \"Title: Structured Peer-to-Peer Overlays Need Application-Driven Benchmarks\\nYear: 2003\\n\\nAbstract: Considerable research effort has recently been devoted to the design of structured peer-to-peer overlays, a term we use to encompass Content-Addressable Networks (CANs), Distributed Hash Tables (DHTs), and Decentralized Object Location and Routing networks (DOLRs). These systems share the property that they consistently map a large set of identifiers to a set of nodes in a network, and while at first sight they provide very similar services, they nonetheless embody a wide variety of design alternatives. We present the case for developing application-driven benchmarks for such overlays, give a model of the services they provide applications, describe and present the results of two preliminary benchmarks, and discuss the implications of our tests for application writers. We are unaware of other empirical comparative work in this area.\",\n \"Title: Predicate routing: enabling controlled networking\\nYear: 2003\\n\\nAbstract: this paper, we present Predicate Routing(PR) as a solution to this problem. We briey describe ourcentralized implementation and then outline the extensionof Internet routing protocols to support Predicate Routing.In current IP networks, the state of the system is primarilyrepresented in an imperative fashion: routing tables andrewall rulesets local to each node strictly specify the actionto be performed on each arriving packet. In contrast,Predicate Routing represents the state of the...\",\n \"Title: Resource overbooking and application profiling in a shared Internet hosting platform\\nYear: 2009\\n\\nAbstract: In this article, we present techniques for provisioning CPU and network resources in shared Internet hosting platforms running potentially antagonistic third-party applications. The primary contribution of our work is to demonstrate the feasibility and benefits of overbooking resources in shared Internet platforms. Since an accurate estimate of an application's resource needs is necessary when overbooking resources, we present techniques to profile applications on dedicated nodes, possibly while in service, and use these profiles to guide the placement of application components onto shared nodes. We then propose techniques to overbook cluster resources in a controlled fashion. We outline an empirical appraoch to determine the degree of overbooking that allows a platform to achieve improvements in revenue while providing performance guarantees to Internet applications. We show how our techniques can be combined with commonly used QoS resource allocation mechanisms to provide application isolation and performance guarantees at run-time. We implement our techniques in a Linux cluster and evaluate them using common server applications. We find that the efficiency (and consequently revenue) benefits from controlled overbooking of resources can be dramatic. Specifically, we find that overbooking resources by as little as 1% we can increase the utilization of the cluster by a factor of two, and a 5% overbooking yields a 300--500% improvement, while still providing useful resource guarantees to applications.\",\n \"Title: 30 seconds is not enough!: a study of operating system timer usage\\nYear: 2008\\n\\nAbstract: The basic system timer facilities used by applications and OS kernels for scheduling timeouts and periodic activities have remained largely unchanged for decades, while hardware architectures and application loads have changed radically. This raises concerns with CPU overhead power management and application responsiveness. In this paper we study how kernel timers are used in the Linux and Vista kernels, and the instrumentation challenges and tradeoffs inherent in conducting such a study. We show how the same timer facilities serve at least five distinct purposes, and examine their performance characteristics under a selection of application workloads. We show that many timer parameters supplied by application and kernel programmers are somewhat arbitrary, and examine the potential benefit of adaptive timeouts. We also discuss the further implications of our results, both for enhancements to the system timer functionality in existing kernels, and for the clean-slate design of a system timer subsystem for new OS kernels, including the extent to which applications might require such an interface at all.\",\n \"Title: Declarative networking: language, execution and optimization\\nYear: 2006\\n\\nAbstract: The networking and distributed systems communities have recently explored a variety of new network architectures, both for application-level overlay networks, and as prototypes for a next-generation Internet architecture. In this context, we have investigated declarative networking: the use of a distributed recursive query engine as a powerful vehicle for accelerating innovation in network architectures [23, 24, 33]. Declarative networking represents a significant new application area for database research on recursive query processing. In this paper, we address fundamental database issues in this domain. First, we motivate and formally define the Network Datalog (NDlog) language for declarative network specifications. Second, we introduce and prove correct relaxed versions of the traditional semi-naïve query evaluation technique, to overcome fundamental problems of the traditional technique in an asynchronous distributed setting. Third, we consider the dynamics of network state, and formalize the iheventual consistencyl. of our programs even when bursts of updates can arrive in the midst of query execution. Fourth, we present a number of query optimization opportunities that arise in the declarative networking context, including applications of traditional techniques as well as new optimizations. Last, we present evaluation results of the above ideas implemented in our P2 declarative networking system, running on 100 machines over the Emulab network testbed.\",\n \"Title: The SAHARA Model for Service Composition across Multiple Providers\\nYear: 2002\\n\\nAbstract: Services are capabilities that enable applications and are of crucial importance to pervasive computing in next-generation networks. Service Composition is the construction of complex services from primitive ones; thus enabling rapid and flexible creation of new services. The presence of multiple independent service providers poses new and significant challenges. Managing trust across providers and verifying the performance of the components in composition become essential issues. Adapting the composed service to network and user dynamics by choosing service providers and instances is yet another challenge. In SAHARA, we are developing a comprehensive architecture for the creation, placement, and management of services for composition across independent providers. In this paper, we present a layered reference model for composition based on a classification of different kinds of composition. We then discuss the different overarching mechanisms necessary for the successful deployment of such an architecture through a variety of case-studies involving composition.\",\n \"Title: Towards a practical, verified kernel\\nYear: 2007\\n\\nAbstract: Optimistic synchronization allows concurrent execution of critical sections while performing dynamic conflict detection and recovery. Optimistic synchronization will increase performance only if critical regions are data independent--concurrent critical ...\",\n \"Title: The design principles of PlanetLab\\nYear: 2006\\n\\nAbstract: PlanetLab is a geographically distributed platform for deploying, evaluating, and accessing planetary-scale network services. PlanetLab is a shared community effort by a large international group of researchers, each of whom gets access to one or more isolated slices of PlanetLab's global resources. Because we deployed PlanetLab and started supporting users before we fully understood what its architecture would be, being able to evolve the system became a requirement. This paper examines the set of design principles that guided this evolution. Some of these principles were explicit at the project outset, and others have become crystallized as the platform has developed.\",\n \"Title: Palimpsest: soft-capacity storage for planetary-scale services\\nYear: 2003\\n\\nAbstract: Distributed writable storage systems typically provide NFS-like semantics and unbounded persistence for files. We claim that for planetary-scale distributed services such facilities are unnecessary and impose an unwanted overhead in complexity and ease of management. Furthermore, wide-area services have requirements not met by existing solutions, in particular capacity management and a realistic model for billing and charging. We argue for ephemeral storage systems which meet these requirements, and present Palimpsest, an early example being deployed on PlanetLab. Palimpsest is small and simple, yet provides soft capacity, congestion-based pricing, automatic reclamation of space, and a security model suitable for a shared storage facility for wide-area services. It does not \\\"fill up\\\" in the way that an ordinary filing system does. In fact, Palimpsest does not arbitrate resources--storage blocks--between clients at all.\",\n \"Title: Rhizoma: a runtime for self-deploying, self-managing overlays\\nYear: 2009\\n\\nAbstract: The trend towards cloud and utility computing infrastructures raises challenges not only for application development, but also for management: diverse resources, changing resource availability, and differing application requirements create a complex optimization problem. Most existing cloud applications are managed externally, and this separation can lead to increased response time to failures, and slower or less appropriate adaptation to resource availability and pricing changes. In this paper, we explore a different approach more akin to P2P systems: we closely couple a decentralized management runtime (\\\"Rhizoma\\\") with the application itself. The application expresses its resource requirements to the runtime as a constrained optimization problem. Rhizoma then fuses multiple real-time sources of resource availability data, from which it decides to acquire or release resources (such as virtual machines), redeploying the system to continually maximize its utility. Using PlanetLab as a challenging \\\"proving ground\\\" for cloud-based services, we present results showing Rhizoma's performance, overhead, and efficiency versus existing approaches, as well the system's ability to react to unexpected large-scale changes in resource availability.\",\n \"Title: The multikernel: a new OS architecture for scalable multicore systems\\nYear: 2009\\n\\nAbstract: Commodity computer systems contain more and more processor cores and exhibit increasingly diverse architectural tradeoffs, including memory hierarchies, interconnects, instruction sets and variants, and IO configurations. Previous high-performance computing systems have scaled in specific cases, but the dynamic nature of modern client and server workloads, coupled with the impossibility of statically optimizing an OS for all workloads and hardware variants pose serious challenges for operating system structures. We argue that the challenge of future multicore hardware is best met by embracing the networked nature of the machine, rethinking OS architecture using ideas from distributed systems. We investigate a new OS structure, the multikernel, that treats the machine as a network of independent cores, assumes no inter-core sharing at the lowest level, and moves traditional OS functionality to a distributed system of processes that communicate via message-passing. We have implemented a multikernel OS to show that the approach is promising, and we describe how traditional scalability problems for operating systems (such as memory management) can be effectively recast using messages and can exploit insights from distributed systems and networking. An evaluation of our prototype on multicore systems shows that, even on present-day machines, the performance of a multikernel is comparable with a conventional OS, and can scale better to support future hardware.\",\n \"Title: Filet-o-fish: practical and dependable domain-specific languages for OS development\\nYear: 2009\\n\\nAbstract: We address a persistent problem with using domain-specific languages to write operating systems: the effort of implementing, checking, and debugging the DSL usually outweighs any of its benefits. Because these DSLs generate C by templated string concatenation, they are tedious to write, fragile, and incompatible with automated verification tools. We present Filet-o-Fish (FoF), a semantic language to ease DSL construction. Building a DSL using FoF consists of safely composing semantically-rich building blocks. This has several advantages: input files for the DSL are formal specifications of the system's functionality, automated testing of the DSL is possible via existing tools, and we can prove that the C code generated by a given DSL respects the semantics expected by the developer. Early experience has been good: FoF is in daily use as part of the tool chain of the Barrelfish multicore OS, which makes extensive use of domain-specific languages to generate low-level OS code. We have found that the ability to rapidly generate DSLs we can rely on has changed how we have designed the OS.\",\n \"Title: Ready for distribution?: turning modular into distributed applications with the R-OSGi deployment tool\\nYear: 2007\\n\\nAbstract: In this demonstration we show drag-and-drop distribution of centralized, modular Java applications. Our system is based on OSGi, an industry standard for building Java applications out of modular units loosely connected through services. Since OSGi is a centralized system, we have elaborated a solution to seamlessly distribute OSGi applications along the boundaries of services and thereby turning arbitrary OSGi applications into distributed applications. In this demonstration, we present an Eclipse based tool that takes the source code of an OSGi application as input, produces a graph of its modules and module dependencies, and allows the user to deploy the application across a distributed system by dragging-and-dropping its constituent modules on different machines. By defining constraints on the distribution, the tool can also support advanced features like load-balancing or redundancy of modules.\",\n \"Title: Dependable self-hosting distributed systems using constraints\\nYear: 2008\\n\\nAbstract: We describe a technique for writing distributed applications which manage themselves over one or more utility computing infrastructures: by dynamically acquiring new computational resources, deploying themselves on these resources, and releasing others when no longer required. Unlike prior work, such management functionality is closely integrated with the application, allowing greater freedom in application-specific policies and faster response to failures and other changes in the environment without requiring any external management system. We address the programming complexity of this approach by applying constraint logic programming, and describe Rhizoma, an experimental runtime to explore these ideas. We present early experience of deploying self-hosting applications on PlanetLab using Rhizoma.\",\n \"Title: Policy expressivity in the Anzere personal cloud\\nYear: 2011\\n\\nAbstract: We present a technique for partially replicating data items at scale according to expressive policy specifications. Recent projects have addressed the challenge of policy-based replication of personal data (photos, music, etc.) within a network of devices, as an alternative to centralized online services. To date, the policies supported by such systems have been relatively simple, in order to facilitate scaling the policy calculation to large numbers of items. In this paper, we show how such replication systems can scale while supporting much more expressive policies than previous schemes: item replication expressed as constraints, devices referred to by predicates rather than explicitly named, and replication to storage nodes acquired on-demand from the cloud. These extensions introduce considerable complexity in policy evaluation, but we show a system can scale well by using equivalence classes to reduce the problem space. We validate our approach via deployment on an ensemble of devices (phones, PCs, cloud virtual machines, etc.), and show that it supports rich policies and high data volumes using simulations and real data based on personal usage in our group.\",\n \"Title: Unifying synchronization and events in a multicore OS\\nYear: 2012\\n\\nAbstract: In this paper, we argue that an operating system structured as a distributed system needs a coordination and a name service to make OS services work correctly. While a distributed structure allows applying algorithms from the distributed field, it also suffers from similar problems like synchronization, naming, distributed locking and coordination of service instances. Octopus, our solution to this problem in the context of a real OS, provides an easy-to-use, high-level, uniform coordination service with events at reasonable performance. Based on this service, we describe three real use cases: device management, OS service registry and boot-up coordination.\",\n \"Title: Deployment of Query Plans on Multicores.\\nYear: 2014\\n\\nAbstract: Efficient resource scheduling of multithreaded software on multicore hardware is difficult given the many parameters involved and the hardware heterogeneity of existing systems. In this paper we explore the efficient deployment of query plans over a multicore machine. We focus on shared query systems, and implement the proposed ideas using SharedDB. The goal of the paper is to explore how to deliver maximum performance and predictability, while minimizing resource utilization when deploying query plans on multicore machines. We propose to use resource activity vectors to characterize the behavior of individual database operators. We then present a novel deployment algorithm which uses these vectors together with dataflow information from the query plan to optimally assign relational operators to physical cores. Experiments demonstrate that this approach significantly reduces resource requirements while preserving performance and is robust across different server architectures.\",\n \"Title: Cosh: Clear OS Data Sharing In An Incoherent World.\\nYear: 2014\\n\\nAbstract: This paper tackles the problem of providing familiar OS abstractions for I/O (such as pipes, network sockets, and a shared file system) to applications on heterogeneous cores including accelerators, co-processors, and offload engines. We aim to isolate the implementation of these facilities from the details of a platform's memory architecture, which is likely to include a combination of cache-coherent shared memory, non-cache-coherent shared memory, and non-shared memory, all in the same system. We propose coherence-oblivious sharing (Cosh), a new OS abstraction that provides inter-process sharing with clear semantics on such diverse hardware. We have implemented a prototype of Cosh for the Barrelfish multikernel. We describe how to build common OS functionality using Cosh, and evaluate its performance on a heterogeneous system consisting of commodity cache-coherent CPUs and prototype Intel many-core co-processors.\",\n \"Title: Grok the data center\\nYear: 2014\\n\\nAbstract: In this paper, we claim that a vital part of managing a large enterprise data center is creating a model which encompasses all layers of the hardware and software stack, from network links up to application placement and configurations, and moreover that this model must be maintained in real time based on monitoring and trace data. Such a claim is not necessarily contentious, but we further argue that existing approaches to managing such facilities suffer from a lack of abstraction of the problem that limits their applicability: enterprise data centers are sufficiently diverse that it is hard to identify a reusable component that can express the complex range of queries that must be served in real time. The problem is less that of working out what data to gather, and how to process it, than it is one of building a flexible, reusable infrastructure above which to perform such processing. We therefore characterize the problem, and the systems challenges entailed by such a \\\\\\\"RDBMS for system management\\\\\\\". Finally, we report on our early exploration of the problem space, showing that even a limited version of such an engine can deliver useful value to a real, large-scale, production data center.\",\n \"Title: Decoupling cores, kernels, and operating systems\\nYear: 2014\\n\\nAbstract: We present Barrelfish/DC, an extension to the Barrelfish OS which decouples physical cores from a native OS kernel, and furthermore the kernel itself from the rest of the OS and application state. In Barrelfish/DC, native kernel code on any core can be quickly replaced, kernel state moved between cores, and cores added and removed from the system transparently to applications and OS processes, which continue to execute. Barrelfish/DC is a multikernel with two novel ideas: the use of boot drivers to abstract cores as regular devices, and a partitioned capability system for memory management which externalizes core-local kernel state. We show by performance measurements of real applications and device drivers that the approach is practical enough to be used for a number of purposes, such as online kernel upgrades, and temporarily delivering hard real-time performance by executing a process under a specialized, single-application kernel.\",\n \"Title: Arrakis: the operating system is the control plane\\nYear: 2014\\n\\nAbstract: Recent device hardware trends enable a new approach to the design of network server operating systems. In a traditional operating system, the kernel mediates access to device hardware by server applications to enforce process isolation as well as network and disk security. We have designed and implemented a new operating system, Arrakis, that splits the traditional role of the kernel in two. Applications have direct access to virtualized I/O devices, allowing most I/O operations to skip the kernel entirely, while the kernel is re-engineered to provide network and disk protection without kernel mediation of every operation. We describe the hardware and software changes needed to take advantage of this new abstraction, and we illustrate its power by showing improvements of 2 to 5 × in latency and 9 × throughput for a popular persistent NoSQL store relative to a well-tuned Linux implementation.\",\n \"Title: Outlook on Operating Systems\\nYear: 2016\\n\\nAbstract: Will OSs in 2025 still resemble the Unix-like consensus of today, or will a very different design achieve widespread adoption?\",\n \"Title: Modeling NICs with Unicorn\\nYear: 2013\\n\\nAbstract: NICs are increasingly complex and diverse, offering a wide range of hardware functionality to aid network protocol processing. Harnessing the power of NIC hardware requires the ability to control and reason about a variety of different feature sets in the network stack. Towards this goal, we propose Unicorn, a language for describing modern NICs. Unicorn offers a simple set of abstractions for modeling both NIC functionality and the state of a protocol stack. To evaluate its expressivity and potential, we present a non-trivial model for the Intel i82599 10GbE NIC, and an algorithm that uses graph embedding to optimize the use of NIC hardware in the network stack.\",\n \"Title: SpaceJMP: Programming with Multiple Virtual Address Spaces.\\nYear: 2016\\n\\nAbstract: Memory-centric computing demands careful organization of the virtual address space, but traditional methods for doing so are inflexible and inefficient. If an application wishes to address larger physical memory than virtual address bits allow, if it wishes to maintain pointer-based data structures beyond process lifetimes, or if it wishes to share large amounts of memory across simultaneously executing processes, legacy interfaces for managing the address space are cumbersome and often incur excessive overheads. We propose a new operating system design that promotes virtual address spaces to first-class citizens, enabling process threads to attach to, detach from, and switch between multiple virtual address spaces. Our work enables data-centric applications to utilize vast physical memory beyond the virtual range, represent persistent pointer-rich data structures without special pointer representations, and share large amounts of memory between processes efficiently.\\n\\nWe describe our prototype implementations in the DragonFly BSD and Barrelfish operating systems. We also present programming semantics and a compiler transformation to detect unsafe pointer usage. We demonstrate the benefits of our work on data-intensive applications such as the GUPS benchmark, the SAMTools genomics workflow, and the Redis key-value store.\",\n \"Title: Shoal: Smart Allocation and Replication of Memory For Parallel Programs\\nYear: 2015\\n\\nAbstract: Modern NUMA multi-core machines exhibit complex latency and throughput characteristics, making it hard to allocate memory optimally for a given program's access patterns. However, sub-optimal allocation can significantly impact performance of parallel programs. We present an array abstraction that allows data placement to be automatically inferred from program analysis, and implement the abstraction in Shoal, a runtime library for parallel programs on NUMA machines. In Shoal, arrays can be automatically replicated, distributed, or partitioned across NUMA domains based on annotating memory allocation statements to indicate access patterns. We further show how such annotations can be automatically provided by compilers for high-level domainspecific languages (for example, the Green-Marl graph language). Finally, we show how Shoal can exploit additional hardware such as programmable DMA copy engines to further improve parallel program performance. We demonstrate significant performance benefits from automatically selecting a good array implementation based on memory access patterns and machine characteristics. We present two case-studies: (i) Green-Marl, a graph analytics workload using automatically annotated code based on information extracted from the high-level program and (ii) a manually-annotated version of the PARSEC Streamcluster benchmark.\",\n \"Title: Distributing the Data Plane for Remote Storage Access\\nYear: 2015\\n\\nAbstract: Sub-microsecond network and memory latencies require fast user-level access to local and remote storage. While user-level access to local storage has been demonstrated recently, it does currently not extend to serverless parallel systems in datacenter environments. We propose direct user-level access to remote storage in a distributed setting, unifying fast data access and high-performance remote memory access programming. We discuss a minimal hardware extension of the IOMMU to enable direct remote storage access. In order to maintain optimal performance in the system, we use epoch-based accesses to allow fine-tuning of atomicity, consistency, isolation, and durability semantics. We also address the problem of user-managed coherent caching. Finally, we briefly discuss the design of DiDAFS, a Distributed Direct Access File System that enables efficient data analytics use-cases such as buffered producer-consumer synchronization and key-value stores as well as deeper integration of storage into high performance computing applications.\",\n \"Title: Not Your Parents' Physical Address Space\\nYear: 2015\\n\\nAbstract: A physical memory address is no longer the stable concept it was. We demonstrate how modern computer systems from rack-scale to SoCs have multiple physical address spaces, which overlap and intersect in complex, dynamic ways, and may be too small to even address available memory in the near future. We present a new model of representing and interpreting physical addresses in a machine for the purposes of memory management, and outline an implementation of the model in a memory system based on capabilities which can handle arbitrary translations between physical address spaces and still globally manage system memory. Finally, we point out future challenges in managing physical memory, of which our model and design are merely a foundation.\",\n \"Title: Explaining Outputs in Modern Data Analytics.\\nYear: 2016\\n\\nAbstract: We report on the design and implementation of a general framework for interactively explaining the outputs of modern data-parallel computations, including iterative data analytics. To produce explanations, existing works adopt a naive backward tracing approach which runs into known issues; naive backward tracing may identify: (i) too much information that is difficult to process, and (ii) not enough information to reproduce the output, which hinders the logical debugging of the program. The contribution of this work is twofold. First, we provide methods to effectively reduce the size of explanations based on the first occurrence of a record in an iterative computation. Second, we provide a general method for identifying explanations that are sufficient to reproduce the target output in arbitrary computations -- a problem for which no viable solution existed until now. We implement our approach on differential dataflow, a modern high-throughput, low-latency dataflow platform. We add a small (but extensible) set of rules to explain each of its data-parallel operators, and we implement these rules as differential dataflow operators themselves. This choice allows our implementation to inherit the performance characteristics of differential dataflow, and results in a system that efficiently computes and updates explanatory inputs even as the inputs of the reference computation change. We evaluate our system with various analytic tasks on real datasets, and we show that it produces concise explanations in tens of milliseconds, while remaining faster -- up to two orders of magnitude -- than even the best implementations that do not support explanations.\",\n \"Title: So many performance events, so little time.\\nYear: 2016\\n\\nAbstract: Many modern workloads are a heterogeneous mix of parallel applications running on machines with complex processors and memory systems. These workloads often interfere with each other by contending for the limited set of resources in a machine, leading to performance degradation. In principle, one can use hardware performance counters to characterize the root causes of interference inside a modern machine. However, we show that current facilities in today's operating systems mean that such an analysis requires careful consideration of the intricacies of specific hardware counters, domain knowledge about the applications, and a deep understanding of the underlying hardware architecture. In contrast, we present the design of Glatt, which is able to automatically identify the root causes of interference in a machine online. Glatt can expose this information to a runtime or the operating system for predicting better thread assignments or for dynamically increasing/decreasing parallelism within each runtime.\",\n \"Title: Customized OS support for data-processing.\\nYear: 2016\\n\\nAbstract: For decades, database engines have found the generic interfaces offered by the operating systems at odds with the need for efficient utilization of hardware resources. As a result, most engines circumvent the OS and manage hardware directly. With the growing complexity and heterogeneity of modern hardware, database engines are now facing a steep increase in the complexity they must absorb to achieve good performance. Taking advantage of recent proposals in operating system design, such as multi-kernels, in this paper we explore the development of a light weight OS kernel tailored for data processing and discuss its benefits for simplifying the design and improving the performance of data management systems.\",\n \"Title: Machine-Aware Atomic Broadcast Trees for Multicores.\\nYear: 2016\\n\\nAbstract: The performance of parallel programs on multicore machines often critically depends on group communication operations like barriers and reductions being highly tuned to hardware, a task requiring considerable developer skill. Smelt is a library that automatically builds efficient inter-core broadcast trees tuned to individual machines, using a machine model derived from hardware registers plus micro-benchmarks capturing the low-level machine characteristics missing from vendor specifications. Experiments on a wide variety of multicore machines show that near-optimal tree topologies and communication patterns are highly machine-dependent, but can nevertheless be derived by Smelt and often further improve performance over well-known static topologies. Furthermore, we show that the broadcast trees built by Smelt can be the basis for complex group operations like global barriers or state machine replication, and that the hardware-tuning provided by the underlying tree is sufficient to deliver as good or better performance than state-of-the-art approaches: the higher-level operations require no further hardware optimization.\",\n \"Title: Formalizing Memory Accesses And Interrupts\\nYear: 2017\\n\\nAbstract: The hardware/software boundary in modern heterogeneous multicore computers is increasingly complex, and diverse across different platforms. A single memory access by a core or DMA engine traverses multiple hardware translation and caching steps, and the destination memory cell or register often appears at different physical addresses for different cores. Interrupts pass through a complex topology of interrupt controllers and remappers before delivery to one or more cores, each with specific constraints on their configurations. System software must not only correctly understand the specific hardware at hand, but also configure it appropriately at runtime. We propose a formal model of address spaces and resources in a system that allows us to express and verify invariants of the system's runtime configuration, and illustrate (and motivate) it with several real platforms we have encountered in the process of OS implementation.\",\n \"Title: Online Reconstruction of Structural Information from Datacenter Logs.\\nYear: 2017\\n\\nAbstract: Well-run datacenter application architectures are heavily instrumented to provide detailed traces of messages and remote invocations. Reconstructing user sessions, call graphs, transaction trees, and other structural information from these messages, a process known as sessionization, is the foundation for a variety of diagnostic, profiling, and monitoring tasks essential to the operation of the datacenter. We present the design and implementation of a system which processes log streams at gigabits per second and reconstructs user sessions comprising millions of transactions per second in real time with modest compute resources, while dealing with clock skew, message loss, and other real-world phenomena that make such a task challenging. Our system is based on the Timely Dataflow framework for low latency, data-parallel computation, and we demonstrate its utility with a number of use-cases and traces from a large, operational, mission-critical enterprise data center.\",\n \"Title: Separating Translation from Protection in Address Spaces with Dynamic Remapping\\nYear: 2017\\n\\nAbstract: It is time to reconsider memory protection. The emergence of large non-volatile main memories, scalable interconnects, and rack-scale computers running large numbers of small \\\\\\\"micro services\\\\\\\" creates significant challenges for memory protection based solely on MMU mechanisms. Central to this is a tension between protection and translation: optimizing for translation performance often comes with a cost in protection flexibility. We argue that a key-based memory protection scheme, complementary to but separate from regular page-level translation, is a better match for this new world. We present MaKC, a new architecture which combines two levels of capability-based protection to scale fine-grained memory protection at both user and kernel level to large numbers of protection domains without compromising efficiency at scale or ease of revocation.\",\n \"Title: Towards Correct-by-Construction Interrupt Routing on Real Hardware.\\nYear: 2017\\n\\nAbstract: In this paper we address the problem of correctly configuring interrupts. The interrupt subsystem of a computer is increasingly complex: a zoo of different controllers with varying constraints and capabilities form a network with limited connectivity. An OS which aspires to provable correctness must manage a limited set of interrupt vectors, delegate interrupts to device drivers and configure the controllers correctly. No well-specified approach exists.\\n\\nAs a foundation for applying language-level techniques like program sketching and synthesis to this problem, we present a formal model for interrupt routing which can capture all the system topologies and interrupt controllers we have encountered in the wild, show applications of such a model not possible with informal, ad-hoc approaches like DeviceTrees, and finally discuss an implementation based on the model which forms the new interrupt subsystem of the Barrelfish OS.\",\n \"Title: Megaphone: latency-conscious state migration for distributed streaming dataflows\\nYear: 2018\\n\\nAbstract: We design and implement Megaphone, a data migration mechanism for stateful distributed dataflow engines with latency objectives. When compared to existing migration mechanisms, Megaphone has the following differentiating characteristics: (i) migrations can be subdivided to a configurable granularity to avoid latency spikes, and (ii) migrations can be prepared ahead of time to avoid runtime coordination. Megaphone is implemented as a library on an unmodified timely dataflow implementation, and provides an operator interface compatible with its existing APIs. We evaluate Megaphone on established benchmarks with varying amounts of state and observe that compared to naïve approaches Megaphone reduces service latencies during reconfiguration by orders of magnitude without significantly increasing steady-state overhead.\",\n \"Title: DeltaPath: dataflow-based high-performance incremental routing.\\nYear: 2018\\n\\nAbstract: Routing controllers must react quickly to failures, reconfigurations and workload or policy changes, to ensure service performance and cost-efficient network operation. We propose a general execution model which views routing as an incremental data-parallel computation on a graph-based network model plus a continuous stream of network changes. Our approach supports different routing objectives with only minor re-configuration of its core algorithm, and easily accomodates dynamic user-defined routing policies. Moreover, our prototype demonstrates excellent performance: on Google Jupiter topology it reacts with a median time of 350ms to link failures and serves more than two million path requests per second each with latency under 1ms. This is three orders-of-magnitude faster than the popular ONOS open-source SDN controller.\",\n \"Title: Rack-Scale Capabilities: Fine-Grained Protection for Large-Scale Memories\\nYear: 2019\\n\\nAbstract: Rack-scale systems with large, shared, disaggregated, and persistent memory need solid protection and authorization techniques. Our solution uses a memoryside capability enforcement processor that gates memory accesses through extended capabilities, enables finegrained access control beyond a single address space, and minimally disrupts the programming model.\",\n \"Title: Memory-Side Protection With a Capability Enforcement Co-Processor.\\nYear: 2019\\n\\nAbstract: Byte-addressable nonvolatile memory (NVM) blends the concepts of storage and memory and can radically improve data-centric applications, from in-memory databases to graph processing. By enabling large-capacity devices to be shared across multiple computing elements, fabric-attached NVM changes the nature of rack-scale systems and enables short-latency direct memory access while retaining data persistence properties and simplifying the software stack.\\n\\nAn adequate protection scheme is paramount when addressing shared and persistent memory, but mechanisms that rely on virtual memory paging suffer from the tension between performance (pushing toward large pages) and protection granularity (pushing toward small pages). To address this tension, capabilities are worth revisiting as a more powerful protection mechanism, but the long time needed to introduce new CPU features hampers the adoption of schemes that rely on instruction-set architecture support.\\n\\nThis article proposes the Capability Enforcement Co-Processor (CEP), a programmable memory controller that implements fine-grain protection through the capability model without requiring instruction-set support in the application CPU. CEP decouples capabilities from the application CPU instruction-set architecture, shortens time to adoption, and can rapidly evolve to embrace new persistent memory technologies, from NVDIMMs to native NVM devices, either locally connected or fabric attached in rack-scale configurations. CEP exposes an application interface based on memory handles that get internally converted to extended-pointer capabilities.\\n\\nThis article presents a proof of concept implementation of a distributed object store (Redis) with CEP. It also demonstrates a capability-enhanced file system (FUSE) implementation using CEP. Our proof of concept shows that CEP provides fine-grain protection while enabling direct memory access from application clients to the NVM, and that by doing so opens up important performance optimization opportunities (up to 4× reduction in latency in comparison to software-based security enforcement) without compromising security.\\n\\nFinally, we also sketch how a future hybrid model could improve the initial implementation by delegating some CEP functionality to a CHERI-enabled processor.\",\n \"Title: Generating correct initial page tables from formal hardware descriptions\\nYear: 2021\\n\\nAbstract: ABSTRACTModern hardware platforms are increasingly complex and heterogeneous. System software uses a hodgepodge of different mechanisms and representations to express the memory topology of the target platform. Considerable maintenance effort is required to keep them in sync while often sharing is impossible due to hard-coded values. Incorrect platform-specific values in the hardware initialization sequence can lead to security critical and hard-to-find bugs because of misconfigured translation hardware, inaccessible devices, or the use of bad pointers. We present a better way for system software to express and initialize memory hardware. We adopt an existing, powerful hardware description language, and efficiently compile it to generate correct initial page tables and memory maps for OS kernels and firmware from a single system description. We evaluate our system on multiple architectures and platforms, and demonstrate that we can use the generated data structures to successfully initialize translation hardware, devices, memory maps, and allocators enabling easy support of new hardware platforms.\",\n \"Title: Declarative Power Sequencing\\nYear: 2021\\n\\nAbstract: Modern computer server systems are increasingly managed at a low level by baseboard management controllers (BMCs). BMCs are processors with access to the most critical parts of the platform, below the level of OS or hypervisor, including control over power delivery to every system component. Buggy or poorly designed BMC software not only poses a security threat to a machine, it can permanently render the hardware inoperative. Despite this, there is little published work on how to rigorously engineer the power management functionality of BMCs so as to prevent this happening.This article takes a first step toward putting BMC software on a sound footing by specifying the hardware environment and the constraints necessary for safe and correct operation. This is best accomplished through automation: correct-by-construction power control sequences can be efficiently generated from a simple, trustworthy model of the platform's power tree that incorporates the sequencing requirements and safe voltage ranges of all components.We present both a modeling language for complex power-delivery networks and a tool to automatically generate safe, efficient power sequences for complex modern platforms. This not only increases the trustworthiness of a hitherto opaque yet critical element of platform firmware: regulator and chip power models are significantly simpler to produce than hand-written power sequences. This, combined with model reuse for common components, reduces both time and cost associated with platform bring-up for new hardware.We evaluate our tool using a new high-performance 2-socket server platform with >100W per socket TDP, tight voltage limits and 25 distinct power regulators needing configuration, showing both fast (<10s) tool runtime, and correct power sequencing of a live system.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":74,"string":"74"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13575,"string":"13,575"}}},{"rowIdx":13576,"cells":{"question_id":{"kind":"string","value":"Christophe Rosenberger"},"question":{"kind":"string","value":"Generate an abstract for the title \"Hybrid template update system for unimodal biometric systems\" using the following items: Semi-supervised template update systems, intra-class variability, hybrid system, biometric sub-references, keystroke-dynamics authentication system."},"answer":{"kind":"string","value":"Semi-supervised template update systems allow to automatically take into account the intra-class variability of the biometric data over time. Such systems can be inefficient by including too many impostor's samples or skipping too many genuine's samples. In the first case, the biometric reference drifts from the real biometric data and attracts more often impostors. In the second case, the biometric reference does not evolve quickly enough and also progressively drifts from the real biometric data. We propose a hybrid system using several biometric sub-references in order to increase performance of self-update systems by reducing the previously cited errors. The proposition is validated for a keystroke-dynamics authentication system (this modality suffers of high variability over time) on two consequent datasets from the state of the art."},"correct_answer":{"kind":"string","value":"Semi-supervised template update systems allow to automatically take into account the intra-class variability of the biometric data over time. Such systems can be inefficient by including too many impostor's samples or skipping too many genuine's samples. In the first case, the biometric reference drifts from the real biometric data and attracts more often impostors. In the second case, the biometric reference does not evolve quickly enough and also progressively drifts from the real biometric data. We propose a hybrid system using several biometric sub-references in order to increase performance of self-update systems by reducing the previously cited errors. The proposition is validated for a keystroke-dynamics authentication system (this modality suffers of high variability over time) on two consequent datasets from the state of the art."},"documents":{"kind":"list like","value":["Title: An e-payment Architecture Ensuring a High Level of Privacy Protection.\nYear: 2013\n\nAbstract: Online shopping is becoming more and more interesting for clients because of the ease of use and the large choice of products. As a consequence, 2.3 billion online clients have been identified in 2011. This rapid increase was accompagnied by various frauds, including stolen smart cards or fraudulent repudiation. Several e-payment systems have been proposed to reduce these security threats and the 3D-Secure protocol is becoming a standard for the payment on the Internet. Nevertheless, this protocol has not been studied in-depth, particularly in terms of privacy. This paper proposes a detailed description and an analysis of the 3D-Secure protocol, through a new privacy-orienting model for e-payment architectures. Some improvements of 3D-Secure protocol, concerning the protection of banking information, are also presented. Then, this article presents and analyses a new online payment architecture centered on the privacy of individuals.","Title: Low Cost and Usable Multimodal Biometric System Based on Keystroke Dynamics and 2D Face Recognition\nYear: 2010\n\nAbstract: We propose in this paper a low cost multimodal biometric system combining keystroke dynamics and 2D face recognition. The objective of the proposed system is to be used while keeping in mind: good performances, acceptability, and espect of privacy. Different fusion methods have been used (min, max, mul, svm, weighted sum configured with genetic algorithms, and, genetic programming) on the scores of three keystroke dynamics algorithms and two 2D face recognition ones. This multimodal biometric system improves the recognition rate in comparison with each individual method. On a chimeric database composed of 100 individuals, the best keystroke dynamics method obtains an EER of 8.77%, the best face recognition one has an EER of 6.38%, while the best proposed fusion system provides an EER of 2.22%.","Title: A tele-operated mobile ultrasound scanner using a light-weight robot.\nYear: 2005\n\nAbstract: This paper presents a new tele-operated robotic chain for real-time ultrasound image acquisition and medical diagnosis. This system has been developed in the frame of the Mobile Tele-Echography Using an Ultralight Robot European Project. A light-weight six degrees-of-freedom serial robot, with a remote center of motion, has been specially designed for this application. It holds and moves a real probe on a distant patient according to the expert gesture and permits an image acquisition using a standard ultrasound device. The combination of mechanical structure choice for the robot and dedicated control law, particularly nearby the singular configuration allows a good path following and a robotized gesture accuracy. The choice of compression techniques for image transmission enables a compromise between flow and quality. These combined approaches, for robotics and image processing, enable the medical specialist to better control the remote ultrasound probe holder system and to receive stable and good quality ultrasound images to make a diagnosis via any type of communication link from terrestrial to satellite. Clinical tests have been performed since April 2003. They used both satellite or Integrated Services Digital Network lines with a theoretical bandwidth of 384 Kb/s. They showed the tele-echography system helped to identify 66% of lesions and 83% of symptomatic pathologies.","Title: A new soft biometric approach for keystroke dynamics based on gender recognition\nYear: 2012\n\nAbstract: Keystroke dynamics allows to authenticate individuals through their way of typing on a computer keyboard. In this study, we are interested in static shared secret keystroke dynamics (all the users type the same password). We present new soft biometrics information which can be extracted from keystroke typing patterns: the gender of the user. This is the first study, to our knowledge, experimenting such kind of information in the field of keystroke dynamics. We present a method for gender recognition through keystroke dynamics with more than 91% of accuracy, on the tested dataset, and we show the improvement on keystroke dynamics authentication method using such kind of information through pattern and score fusion. We obtain a gain of 20% when using gender information against a classical keystroke dynamics method.","Title: Genetic programming for multibiometrics\nYear: 2012\n\nAbstract: Biometric systems suffer from some drawbacks: a biometric system can provide in general good performances except with some individuals as its performance depends highly on the quality of the capture... One solution to solve some of these problems is to use multibiometrics where different biometric systems are combined together (multiple captures of the same biometric modality, multiple feature extraction algorithms, multiple biometric modalities...). In this paper, we are interested in score level fusion functions application (i.e., we use a multibiometric authentication scheme which accept or deny the claimant for using an application). In the state of the art, the weighted sum of scores (which is a linear classifier) and the use of an SVM (which is a non linear classifier) provided by different biometric systems provide one of the best performances. We present a new method based on the use of genetic programming giving similar or better performances (depending on the complexity of the database). We derive a score fusion function by assembling some classical primitives functions (+,*,-,... ). We have validated the proposed method on three significant biometric benchmark datasets from the state of the art.","Title: Service provider authentication assurance\nYear: 2012\n\nAbstract: The concept of authentication assurance traditionally refers to the robustness of methods and mechanisms for user authentication, including the robustness of initial registration and provisioning of user credentials, as well as the robustness of mechanisms that enforce user authentication during operation. However, the user is not the only party that needs to be authenticated to ensure security of online transactions. In fact, online service provision always involves two parties, typically the user on the client side and the service provider on the server side, so that mutual authentication between the two sides is required. In contrast to the unilateral focus on user authentication by industry and academia, it is in fact equally important for the user to correctly authenticate the service provider. Unfortunately, little attention is paid to the problem of correctly authentication the service provider. This paper proposes a framework for server and service provider authentication assurance, similarly to frameworks for user authentication assurance that have already been specified, or are currently under development by many national governments.","Title: 2D color shape recognition using Zernike moments\nYear: 2005\n\nAbstract: 2D Zernike moments belong to the useful object invariant descriptors which have been successfully applied in pattern recognition tasks. The main problem of using Zernike moments invariants is that they are not able to discriminate two objects having the same shape. In this paper, an approach based on Zernike moments applied on color images is proposed. A Support Vector Machine is used for object classification. For object segmentation, the connected component labeling algorithm is used. Compared with the classical method, this approach shows higher accuracy in object recognition. Some experimental results on the COIL-100 database are presented.","Title: Abnormal events detection based on spatio-temporal co-occurences\nYear: 2009\n\nAbstract: We explore a location-basedapproachforbehaviormod- eling and abnormality detection. In contrast to the con- ventional object-based approach where an object may first be tagged, identified, classified, and tracked, we proceed directly with event characterization and behavior model- ing at the pixel(s) level based on motion labels obtained from background subtraction. Since events are temporally and spatially dependent, this calls for techniques that ac- count for statistics of spatio-temporal events. Based on mo- tion labels, we learn co-occurrence statistics for normal events across space-time. For one (or many) key pixel(s), we estimate a co-occurrence matrix that accounts for any two active labels which co-occur simultaneously within the same spatio-temporal volume. This co-occurrence matrix is then used as a potential function in a Markov Random Field (MRF) model to describe the probability of observa- tions within the same spatio-temporal volume. The MRF distribution implicitly accounts for speed, direction, as well as the average size of the objects passing in front of each key pixel. Furthermore, when the spatio-temporal volume is large enough, the co-occurrence distribution contains the average normal path followed by moving objects. The learned normal co-occurrence distribution can be used for abnormaldetection. Our method hasbeen tested on various outdoor videos representing various challenges.","Title: Segmentation framework based on label field fusion.\nYear: 2007\n\nAbstract: In this paper, we put forward a novel fusion framework that mixes together label fields instead of observation data as is usually the case. Our framework takes as input two label fields: a quickly estimated and to-be-refined segmentation map and a spatial region map that exhibits the shape of the main objects of the scene. These two label fields are fused together with a global energy function that is minimized with a deterministic iterative conditional mode algorithm. As explained in the paper, the energy function may implement a pure fusion strategy or a fusion-reaction function. In the latter case, a data-related term is used to make the optimization problem well posed. We believe that the conceptual simplicity, the small number of parameters, the use of a simple and fast deterministic optimizer that admits a natural implementation on a parallel architecture are among the main advantages of our approach. Our fusion framework is adapted to various computer vision applications among which are motion segmentation, motion estimation and occlusion detection.","Title: Fast computation of the performance evaluation of biometric systems: Application to multibiometrics\nYear: 2013\n\nAbstract: The performance evaluation of biometric systems is a crucial step when designing and evaluating such systems. The evaluation process uses the Equal Error Rate (EER) metric proposed by the International Organization for Standardization (ISO/IEC). The EER metric is a powerful metric which allows easily comparing and evaluating biometric systems. However, the computation time of the EER is, most of the time, very intensive. In this paper, we propose a fast method which computes an approximated value of the EER. We illustrate the benefit of the proposed method on two applications: the computing of non parametric confidence intervals and the use of genetic algorithms to compute the parameters of fusion functions. Experimental results show the superiority of the proposed EER approximation method in term of computing time, and the interest of its use to reduce the learning of parameters with genetic algorithms. The proposed method opens new perspectives for the development of secure multibiometrics systems by speeding up their computation time.","Title: Enhancing the privacy of electronic passports\nYear: 2012\n\nAbstract: We address in this paper the problem of privacy in the current architecture in electronic passports for the storage and transmission of biometric data such as fingerprints. The current architecture provides a good protection of biometric personal data but brute force attack could be used in the near future using cloud computing. We propose a new solution combining cryptographic protocols and cancellable biometrics to solve this problem. The biocode of an individual is protected by cryptographic keys exchanged by the PACE protocol. We put into obviousness the benefit of the proposed solution in terms of security and privacy.","Title: Comparative study of global invariant descriptors for object recognition\nYear: 2008\n\nAbstract: Although many object invariant descriptors have been proposed in the literature, putting them into practice to obtain a robust recognition system that is able to face several perturbations is still a studied problem. After presenting the most commonly used global invariant descriptors, a comparative study permits us to show their ability to discriminate between objects with little training. The Columbia Object Image Library database (COIL-100), which presents a same object translated, rotated, and scaled, is used to test the invariant features of geometrical transforms. Partial object occultation or presence of complex background are examples of used images to test the robustness of the studied descriptors. We compare them in both a global and a local context (computed on the neighborhood of a pixel). The scale invariant feature transform descriptor is used as a reference for local invariant descriptors. This study shows the relative performance of invariant descriptors used in both a global and a local context and identifies the different situations for which they are best suited. (C) 2008 SPIE and IS&T.","Title: A Real Time Human Detection System Based on Far Infrared Vision\nYear: 2008\n\nAbstract: We present in this article a human detection and tracking algorithm using infrared vision in order to have reliable information on a room occupation. We intend to use this information to limit energetic consumption (light, heating). We perform first, a foreground segmentation with a Gaussian background model. A tracking step based on connected components intersections allows to collect information on 2D displacements of moving objects in the image plane. A classification based on a cascade of boosted classifiers is used for the recognition. Experimental results show the efficiency of the proposed algorithm.","Title: Performance Evaluation of Biometric Template Update\nYear: 2012\n\nAbstract: Template update allows to modify the biometric reference of a user while he uses the biometric system. With such kind of mechanism we expect the biometric system uses always an up to date representation of the user, by capturing his intra-class (temporary or permanent) variability. Although several studies exist in the literature, there is no commonly adopted evaluation scheme. This does not ease the comparison of the different systems of the literature. In this paper, we show that using different evaluation procedures can lead in different, and contradictory, interpretations of the results. We use a keystroke dynamics (which is a modality suffering of template ageing quickly) template update system on a dataset consisting of height different sessions to illustrate this point. Even if we do not answer to this problematic, it shows that it is necessary to normalize the template update evaluation procedures.","Title: A New Protocol to Evaluate the Resistance of Template Update Systems against Zero-Effort Attacks\nYear: 2013\n\nAbstract: Since several years, there is a high interest in the creation of template update mechanisms in biometrics. By automatically updating over time the biometric reference of an individual, these update systems aim at avoiding a performance decreases over time. However, update errors occur when impostor samples are included in the biometric reference. In this paper, we are interested in analysing the behavior of template update against zero effort attacks depending on the difficulty of these attacks which is parametrized by the ratio of impostors samples included. We propose a protocol, which allows to evaluate an update mechanism under such kind of attacks by preserving an important amount of genuine samples, even when the ratio of impostor samples is important. Then, we apply this protocol on an existing template update system from the state of the art at various attack rates in order to assess its benefits.","Title: Comparative study of localization metrics for the evaluation of image interpretation systems\nYear: 2010\n\nAbstract: Image interpretation is particularly important in many real applications (video monitoring, biometrics, etc.). Due to the proliferation of image interpretation systems in the literature, their evaluation still remains a crucial stake. Among all the tasks in this field, the quality of object localization is often evaluated through an evaluation metric. We propose to review these techniques and study their reliability. We first propose a generic definition of a localization algorithm. Then, different state of the art techniques to evaluate image interpretation results are detailed. Secondly, we focus on metrics that enable us to evaluate localization results. We propose a general methodology to analyze the behavior of an evaluation metric, considered here as a black box (its definition is not even supposed to be known). We define the properties that these metrics should fulfill. We then perform a comparative study of 33 localization metrics from the state of the art. Experimental results conducted on a large and significant image database permit us to determine metrics that should be used in the future for the evaluation of object localization results. (C) 2010 SPIE and IS&T. [DOI: 10.1117/1.3446803]","Title: Calibration and quality control of cherries by artificial vision\nYear: 2004\n\nAbstract: We describe an artificial vision system for the quality control of cherries. We develop image-processing software determining three types of information describing the quality of the fruit: the color as an indicator of ripeness, the presence of defects such as cracking, and the size. The sorting of cherries conditioned by all these criteria is carried out at a high cadence (20 cherries/s). Real-time image processing is then necessary. A simple sensor is used to synchronize the processing and the sorting of cherries. Air actuators are used to eject them after quality control by vision. We present the architecture of the developed system. We show its efficiency through experimental results and we give some perspectives of improvement. (C) 2004 SPIE and IST.","Title: Comparative study of contour detection evaluation criteria based on dissimilarity measures\nYear: 2008\n\nAbstract: We present in this article a comparative study of well-known supervised evaluation criteria that enable the quantification of the quality of contour detection algorithms. The tested criteria are often used or combined in the literature to create new ones. Though these criteria are classical ones, none comparison has been made, on a large amount of data, to understand their relative behaviors. The objective of this article is to overcome this lack using large test databases both in a synthetic and a real context allowing a comparison in various situations and application fields and consequently to start a general comparison which could be extended by any person interested in this topic. After a review of the most common criteria used for the quantification of the quality of contour detection algorithms, their respective performances are presented using synthetic segmentation results in order to show their performance relevance face to undersegmentation, oversegmentation, or situations combining these two perturbations. These criteria are then tested on natural images in order to process the diversity of the possible encountered situations. The used databases and the following study can constitute the ground works for any researcher who wants to confront a new criterion face to well-known ones.","Title: Evaluation Protocol for Localization Metrics\nYear: 2008\n\nAbstract: Localization metrics permit to quantify the correctness of object detection in an image interpretation result. This paper deals with the definition of a protocol in order to evaluate the behavior of localization metrics. We first define some properties that metrics should verify and create a synthetic database that enables to verify those properties on different metrics. After presenting the tested localization metrics, the results obtained following the proposed protocol are exposed. Finally, some conclusions and perspectives are given.","Title: Evaluation metric for image understanding\nYear: 2009\n\nAbstract: We propose in this paper a new evaluation metric that enables to quantify the quality of an image interpretation result. This metric takes into account the a priori knowledge used by the interpretation algorithm and the ground truth associated with the original image. We combine two metrics that evaluate the localization and recognition results of each detected object. We show that the proposed metric fulfills some theoretical properties and has a correct behavior face to empirical experiments on an image benchmark database. We think that this metric could be a reliable reference for image and video understanding competitions.","Title: Evaluation of Human Detection Algorithms in Image Sequences\nYear: 2010\n\nAbstract: This paper deals with the general evaluation of human detection algorithms. We first present the algorithms implemented within the CAPTHOM project dedicated to the development of a vision-based system for human detection and tracking in an indoor environment using a static camera. We then show how a global evaluation metric we developped for the evaluation of understanding algorithms taking into account both localization and recognition precision of each single interpretation result, can be a useful tool for industrials to guide them in the elaboration of suitable and optimized algorithms.","Title: Evolutionary computation approaches for shape modelling and fitting\nYear: 2005\n\nAbstract: This paper proposes and analyzes different evolutionary computation techniques for conjointly determining a model and its associated parameters. The context of 3D reconstruction of objects by a functional representation illustrates the ability of the proposed approaches to perform this task using real data, a set of 3D points on or near the surface of the real object. The final recovered model can then be used efficiently in further modelling, animation or analysis applications. The first approach is based on multiple genetic algorithms that find the correct model and parameters by successive approximations. The second approach is based on a standard strongly-typed implementation of genetic programming. This study shows radical differences between the results produced by each technique on a simple problem, and points toward future improvements to join the best features of both approaches.","Title: Comparative Study of Local Descriptors for Measuring Object Taxonomy\nYear: 2009\n\nAbstract: Many object descriptors have been proposed in the state of the art. For many reasons (occlusion, point of view, acquisition conditions...), local descriptors have a better robustness for image understanding applications. The goal of this paper is to make a comparative study of eight recent local descriptors. The objective is here to quantify their ability to generate automatically an object taxonomy. In order to answer this question, we use the Caltech256 benchmark which provides a large object taxonomy used as reference. This study shows that SIFT, differential invariants and shape context descriptors are the best ones to achieve this goal.","Title: Segmentation evaluation using a support vector machine\nYear: 2005\n\nAbstract: Segmentation evaluation is a very difficult task even for an expert. We propose in this article a new unsupervised evaluation criterion of an image segmentation result. The quality of a segmentation result is derived without any a priori knowledge by taking into account different evaluation criteria from the literature. We first compare six unsupervised evaluation criteria on a database composed of synthetic gray level images. Vinet's measure is used as an objective function to compare the behavior of the different criteria. We propose in this paper to fuse the best ones by a support vector machine. We illustrate the efficiency of the proposed approach through some experimental results.","Title: Can Chronological Information be Used as a Soft Biometric in Keystroke Dynamics?\nYear: 2012\n\nAbstract: Keystroke dynamics is a behavioral biometric modality which uses typing patterns on a keyboard to recognize individuals. The way of typing the password slightly changes with time, because of various factors (including user's training). This modification in the way of typing results in a decrease of performance recognition over time. In this paper, we analyse the correlation between the comparison score between a query, and a reference and the number of times the user has typed the password. After having quantified his correlation, we analyse the possibility of using stacked classification to ake this aspect in consideration during authentication. Then, we verify if it is possible to classify users on their way of evolving their typing pattern. Results show that even if comparison scores are correlated with the number of time the user has typed the password, the use of a stacked classifier does not improve the results much.","Title: Unconstrained keystroke dynamics authentication with shared secret\nYear: 2011\n\nAbstract: mong all the existing biometric modalities, authentication systems based on keystroke dynamics present interesting advantages. These solutions are well accepted by users and cheap as no additional sensor is required for authenticating the user before accessing to an application. In the last thirty years, many researchers have proposed, different algorithms aimed at increasing the performance of this approach. Their main drawback lies on the large number of data required for the enrollment step. As a consequence, the verification system is barely usable, because the enrollment is too restrictive. In this work, we propose a new method based on the Support Vector Machine (SVM) learning satisfying industrial conditions (i.e., few samples per user are needed during the enrollment phase to create its template). In this method, users are authenticated through the keystroke dynamics of a shared secret (chosen by the system administrator). We use the GREYC keystroke database that is composed of a large number of users (100) for validation purposes. We compared the proposed method with six methods from the literature (selected based on their ability to work with few enrollment samples). Experimental results show that, even though the computation time to build the template can be longer with our method (54 s against 3 s for most of the others), its performance outperforms the other methods in an industrial context (Equal Error Rate of 15.28% against 16.79% and 17.02% for the two best methods of the state-of-the-art, on our dataset and five samples to create the template, with a better computation time than the second best method).","Title: Multi-Agent Based Prototyping of Agriculture Robots\nYear: 2006\n\nAbstract: In this paper, we propose a multi-agents system for the simulation of prototypes for different agriculture robots that can be employed in the harvesting of a vineyard. We show that this prototyping method works well and is appropriate for a system involving several robots. We present the behaviors of each robot, their interactions and some scenarios for the testing of the system. Of specific interest is the cooperative behavior of the robots in such a task. Preliminary results indicate that some optimal choices toward the development of the harvesting system can be correctly identified based on these simulations. The choices can then be used when designing and building the system prototype.","Title: A Support Vector Machine As An Estimator Of Mountain Papaya Ripeness Using Resonant Frequency Or Frequency Centroid\nYear: 2006\n\nAbstract: Mountain papaya fruits (Vasconcella pubescens) were tested for firmness with a nondestructive acoustic method for 14 days after harvest. The response of each fruit was analyzed with the Fourier transform to obtain a firmness index (FI) based on the second resonant frequency and with the Short Time Fourier Transform (STFT) to obtain a spectrogram frequency centroid (FC) index. The indexes were processed with a support vector machine (SVM) learning procedure in which days since harvest was taken as the basic truth of ripeness which the measured indexes attempt to estimate. The analysis of the results demonstrate that different groupings of the days into classes to be estimated give widely varying recognition rates and that the best rates are obtained when the classes are delimited using prior knowledge.","Title: Toward a complete adaptive analysis of an image\nYear: 2003\n\nAbstract: We present an adaptive method that allows the classification of the regions of an image in two classes: textured and uniform regions (weak gray-level variance). Information on the type of texture (stochastic, deterministic) and its granularity (macroscopic, microscopic) are extracted. The developed method is achieved in two steps. It first enables the determination of the global context of an, image (image mainly composed of uniform areas or textured ones) and the localization of the textured and uniform areas.. The second step characterizes each detected area by considering some appropriate features: mean and variance in the case of uniform regions and classical relevant texture attributes (derived from a statistical analysis) associated with some new attributes that we define in the case of textured regions. These complementary features are determined from a texture model derived from the Wold decomposition of the autocovariance function. They enable the acquisition of some information on the type of texture and its granularity. We show the efficiency of the method through two examples of image segmentation. (C) 2003 SPIE and IST.","Title: Secure payment with NFC mobile phone in the SmartTouch project.\nYear: 2008\n\nAbstract: This paper deals with the ITEA (Information Technology for European Advancement) SmartTouch project whose one of the objectives is to use NFC mobile phones for a secure payment. An experience in real conditions has been carried out at Strasbourg in France, in 2007. This trial is the first experiment of a NFC-based payment application that fully supports the international EMV standard and the PayPass program. We focus in this paper on the security analysis which can occur in a NFC mobile phone payment. In the first section, we give a simplified description of the prototype. In the second section, we present the security concepts as a major issue in this scenario. Roughly, the main requirement is to have the same level of security as standard EMV payment transactions. The methodology described in the third section concerns the payment application and its communication with the payment terminal through a NFC link. The last part of this article presents different research perspectives to overstep the encountered problems.","Title: Soft Biometrics For Keystroke Dynamics\nYear: 2013\n\nAbstract: Keystroke dynamics is a viable and practical way as an addition to security for identity verification. It can be combined with passphrases authentication resulting in a more secure verification system. This paper presents a new soft biometric approach for keystroke dynamics. Soft biometrics traits are physical, behavioral or adhered human characteristics, which have been derived from the way human beings normally distinguish their peers (e.g. height, gender, hair color etc.). Those attributes have a low discriminating power, thus not capable of identification performance. Additionally, they are fully available to everyone which makes them privacy-safe. Thus, in this study, it consists of extracting information from the keystroke dynamics templates with the ability to recognise the hand(s) used (i.e. one/two hand(s)); the gender; the age category; and the handedness of a user when he/she types a given password or passphrase on a keyboard. Experiments were conducted on a keystroke dynamics database of 110 users and our experimental results show that the proposed methods are promising.","Title: Keystroke dynamics authentication for collaborative systems\nYear: 2009\n\nAbstract: We present in this paper a study on the ability and the benefits of using a keystroke dynamics authentication method for collaborative systems. Authentication is a challenging issue in order to guarantee the security of use of collaborative systems during the access control step. Many solutions exist in the state of the art such as the use of one time passwords or smart-cards We focus in this paper on biometric based solutions that do not necessitate any additional sensor. Keystroke dynamics is an interesting solution as it uses only the keyboard and is invisible for users. Many methods have been published in this field. We make a comparative study of many of them considering the operational constraints of use for collaborative systems.","Title: Texture detection for image analysis\nYear: 2005\n\nAbstract: Many applications such as image compression, pre-processing or segmentation require some information from the regions composing an image. The main objective of this paper is to define a methodology to extract some local information from an image. Each region is characterized in terms of homogeneity (region composed with the same grey-level or a single texture) and its type (textured or uniform). The decision criterion is based on the use of classical texture attributes (cooccurrence matrix and grey-levels moments) and a support vector machine in order to realize the fusion of the different attributes. We then characterize each region considering its type by appropriate features.","Title: Web-Based Benchmark for Keystroke Dynamics Biometric Systems: A Statistical Analysis\nYear: 2012\n\nAbstract: Most keystroke dynamics studies have been evaluated using a specific kind of dataset in which users type an imposed login and password. Moreover, these studies are optimistics since most of them use different acquisition protocols, private datasets, controlled environment, etc. In order to enhance the accuracy of keystroke dynamics' performance, the main contribution of this paper is twofold. First, we provide a new kind of dataset in which users have typed both an imposed and a chosen pairs of logins and passwords. In addition, the keystroke dynamics samples are collected in a web-based uncontrolled environment (OS, keyboards, browser, etc.). Such kind of dataset is important since it provides us more realistic results of keystroke dynamics' performance in comparison to the literature (controlled environment, etc.). Second, we present a statistical analysis of well known assertions such as the relationship between performance and password size, impact of fusion schemes on system overall performance, and others such as the relationship between performance and entropy. We put into obviousness in this paper some new results on keystroke dynamics in realistic conditions.","Title: An integrated framework combining Bio-Hashed minutiae template and PKCS15 compliant card for a better secure management of fingerprint cancelable templates\nYear: 2013\n\nAbstract: We address in this paper the problem of privacy in fingerprint biometric systems. Today, cancelable techniques have been proposed to deal with this issue. Ideally, such transforms are one-way. However, even if they are with provable security, they remain vulnerable when the user-specific key that achieves cancelability property is stolen. The prominence of the cancelable template confidentiality to maintain the irreversibility property was also demonstrated for many proposed constructions. To prevent possible coming attacks, it becomes important to securely manage this key as well as the transformed template in order to avoid them being leaked simultaneously and thus compromised. To better manage the user credentials of cancelable constructs, we propose a new solution combining a trusted architecture and a cancelable fingerprint template. Therefore, a Bio-Hashed minutiae template based on a chip matching algorithm is proposed. A pkcs15 compliant cancelable biometric system for fingerprint privacy preserving is implemented on a smartcard. This closed system satisfies the safe management of the sensitive templates. The proposed solution is proved to be well resistant to different attacks.","Title: Unsupervised performance evaluation of image segmentation\nYear: 2006\n\nAbstract: We present in this paper a study of unsupervised evaluation criteria that enable the quantification of the quality of an image segmentation result. These evaluation criteria compute some statistics for each region or class in a segmentation result. Such an evaluation criterion can be useful for different applications: the comparison of segmentation results, the automatic choice of the best fitted parameters of a segmentation method for a given image, or the definition of new segmentation methods by optimization. We first present the state of art of unsupervised evaluation, and then, we compare six unsupervised evaluation criteria. For this comparative study, we use a database composed of 8400 synthetic gray-level images segmented in four different ways. Vinet's measure (correct classification rate) is used as an objective criterion to compare the behavior of the different criteria. Finally, we present the experimental results on the segmentation evaluation of a few gray-level natural images.","Title: Palm Vein Verification System Based on SIFT Matching\nYear: 2009\n\nAbstract: We present in this communication a new biometric system based on the use of hand veins acquired by an infrared imager. After the preprocessing stage and binarization, the vein image is characterized by specific patterns. One originality of the proposed system is to use SIFT descriptors for the verification process. The developed method only necessitates a single image for the enrollment step allowing a very fast verification. The experimental results on a database containing images of 24 individuals acquired after two sessions show the efficiency of the proposed method.","Title: A new supervised evaluation criterion for region based segmentation methods\nYear: 2007\n\nAbstract: We present in this article a new supervised evaluation criterion that enables the quantification of the quality of region segmentation algorithms. This criterion is compared with seven well-known criteria available in this context. To that end, we test the different methods on natural images by using a subjective evaluation involving different experts from the French community in image processing. Experimental results show the benefit of this new criterion.","Title: Detecting Half-Occlusion with a Fast Region-Based Fusion Procedure\nYear: 2006\n\nAbstract: This paper presents a novel region-based approach for detecting occlusion between two consecutive frames. Based on a generalization of Marr and Poggio's uniqueness assumption, the explicit goal of our method is to reduce the number of false positives while optimizing the hit rate. To do so, our method relies on a fusion procedure that blends together two segmentation maps: one pre-estimated occlusion binary map and one color segmentation map. While the occlusion map is obtained after a simple thresholding proce- dure, the color segmentation map is obtained with an unsupervised Marko- vian approach. Assuming that the color segmentation regions exhibit more precise edges, the occlusion areas are iteratively modified to fit the color- region shapes. Since our method has been entirely implemented on a parallel architecture (a Graphics Processor Unit), its processing times are remarkably low. Our method is compared with other occlusion approaches both quanti- tatively and qualitatively on scenes that represent different challenges.","Title: Quality assessment of image-based biometric information.\nYear: 2015\n\nAbstract: The quality of biometric raw data is one of the main factors affecting the overall performance of biometric systems. Poor biometric samples increase the enrollment failure and decrease the system performance. Hence, controlling the quality of the acquired biometric raw data is essential in order to have useful biometric authentication systems. Towards this goal, we present a generic methodology for the quality assessment of image-based biometric modality combining two types of information: 1) image quality and 2) pattern-based quality using the scale-invariant feature transformation (SIFT) descriptor. The associated metric has the advantages of being multimodal (face, fingerprint, and hand veins) and independent from the used authentication system. Six benchmark databases and one biometric verification system are used to illustrate the benefits of the proposed metric. A comparison study with the National Institute of Standards and Technology (NIST) fingerprint image quality (NFIQ) metric proposed by the NIST shows the benefits of the presented metric.","Title: Image Watermarking with Biometric Data for Copyright Protection\nYear: 2015\n\nAbstract: In this paper, we deal with the proof of ownership or legitimate usage of a digital content, such as an image, in order to tackle the illegitimate copy. The proposed scheme based on the combination of the watermarking and cancelable biometrics does not require a trusted third party, all the exchanges are between the provider and the customer. The use of cancelable biometrics allows us to provide a privacy compliant proof of identity. We illustrate the robustness of this method against intentional and unintentional attacks of the watermarked content.","Title: Soft biometrics for keystroke dynamics: Profiling individuals while typing passwords\nYear: 2014\n\nAbstract: This paper presents a new profiling approach of individuals based on soft biometrics for keystroke dynamics. Soft biometric traits are unique representation of a person, which can be in a form of physical, behavioural or biological human characteristics that differentiate between him/her into a group people (e.g. gender, age, height, colour, race etc.). Keystroke dynamics is a behavioural biometric modality to recognise how a person types on a keyboard. In this paper, we consider the following soft traits: the hand category (i.e. if the user types with one or two hands), the gender category, the age category and the handedness category. For this purpose, we collected a new database. Two cases are studied: static passwords and free text. By combining machine learning and fusion process, the results are promising.","Title: One-Time Biometrics for Online Banking and Electronic Payment Authentication.\nYear: 2014\n\nAbstract: Online banking and electronic payment systems on the Internet are becoming increasingly advanced. On the machine level, transactions take place between client and server hosts through a secure channel protected with SSL/TLS. User authentication is typically based on two or more factors. Nevertheless, the development of various malwares and social engineering attacks transform the user's PC in an untrusted device and thereby making user authentication vulnerable. This paper investigates how user authentication with biometrics can be made more robust in the online banking context by using a specific device called OffPAD. This context requires that authentication is realized by the bank and not only by the user (or by the personal device) contrary to standard banking systems. More precisely, a new protocol for the generation of one- time passwords from biometric data is presented, ensuring the security and privacy of the entire transaction. Experimental results show an excellent performance considering with regard to false positives. The security analysis of our protocol also illustrates the benefits in terms of strengthened security.","Title: Analysis of embedded applications by evolutionary fuzzing\nYear: 2014\n\nAbstract: In this paper, we propose to use fuzzing techniques to discover vulnerabilities in programs hosted into smart cards used for telecommunications or banking purposes (SIM cards, credit cards, secure element into NFC mobile devices...). Those programs - called applets - usually host sensitive applications and manipulate sensitive data. A flaw by design or by implementation in one of those applet could have disastrous consequences. The proposed approach uses a genetic algorithm to optimize the vulnerabilities search. We illustrate the benefit of the proposed method on a MasterCard M/Chip applet through experimental results.","Title: Towards a New Tool for the Evaluation of the Quality of Ultrasound Compressed Images\nYear: 2006\n\nAbstract: This paper presents a new tool for the evaluation of ultrasound image compression. The goal is to measure the image quality as easily as with a statistical criterion, and with the same reliability as the one provided by the medical assessment. An initial experiment is proposed to medical experts and represents our reference value for the comparison of evaluation criteria. Twenty-one statistical criteria are selected from the literature. A cumulative absolute similarity measure is defined as a distance between the criterion to evaluate and the reference value. A first fusion method based on a linear combination of criteria is proposed to improve the results obtained by each of them separately. The second proposed approach combines different statistical criteria and uses the medical assessment in a training phase with a support vector machine. Some experimental results are given and show the benefit of fusion.","Title: Similarity-based matching for face authentication\nYear: 2008\n\nAbstract: We propose in this paper a face authentication method based on a similarity measure. The SIFT de- scriptor is used to define some interest keypoints char- acterized by an invariant parameter. A graph is then built where nodes correspond to these keypoints. We model the authentication problem as a graph matching process. Experimental results on the AR database show an EER equals to 12% with only one image used for the enrollment and with images simulating real conditions.","Title: Emerging trends in biometric authentication\nYear: 2009\n\nAbstract: The main objective of biometrics is to identify or authenticate an individual, given his/her different physical or behavioral characteristics. This approach has already been employed in our daily life in various applications and forms (USB biometric key, access control, ...). Nevertheless, this technology could be used for multiple other applications. In this presentation, we survey the trends that have been observed for the proliferation of this technology such as: definitions of contemporary unconstrained biometric systems that necessitate or call for low cost sensors and few steps for enrollment (some examples based on face, veins and keystroke dynamics will illustrate this point); evaluation of biometric systems that take into account performance, acceptability, and security; and privacy related applications, such as those in biohashing techniques.","Title: Security EvaBio: An analysis tool for the security evaluation of biometric authentication systems\nYear: 2012\n\nAbstract: Biometric systems present several drawbacks that may significantly decrease their utility. Nowadays, several platforms (such as the FVC-onGoing) exist to assess the performance of such systems. Despite this, none platform exists for the security evaluation of biometric systems. Hence, the aim of this paper is to present an on-line platform for the security evaluation of biometric systems. The key benefits of the presented platform are twofold. First, it provides the research community in biometrics an evaluation tool to assess biometric systems in term of security. Second, the platform provides a database of common threats and vulnerabilities of biometric systems that can be updated by researchers feedbacks. The presented tool is modality-independent. A keystroke dynamics system is used to illustrate the benefits of the presented platform.","Title: Texture based fingerprint BioHashing: Attacks and robustness\nYear: 2012\n\nAbstract: BioHashing is a popular biometric template protection scheme defined in the last decade. Most of previous studies on this algorithm focus on the performance optimization or the use of this privacy protection scheme on many types of biometric modalities (face, fingerprint, palmprint...). The objective of this paper is to study the robustness of this algorithm on a texture based representation by testing and simulating operational attacks. We consider in this study some quantitative measures of the robustness of the BioHashing algorithm and we show some new results on its security on fingerprints represented by texture features.","Title: Analysis of the acquisition process for keystroke dynamics\nYear: 2012\n\nAbstract: In order to evaluate authentication methods for keystroke dynamics, it is necessary to create new datasets. We present in this paper an analysis study of the factors, involved during the acquisition process, which affect the performance of keystroke-based authentication systems. More generally speaking, we are looking for the optimal keystroke data acquisition scenario which may deeply affect the performance of such systems. Results show that (1) it is better to choose passwords based on short, simple and known words; and (2) during the acquisition process of a new keystroke dynamics dataset, there is no significant preference for displaying or telling to a user the text to type.","Title: Challenge-based speaker recognition for mobile authentication\nYear: 2012\n\nAbstract: User authentication is a major trend to guarantee the security of electronic transactions when using mobile devices such as tablets or mobile phones. Biometrics is for us the only real user authentication method. In this article, we propose to realize a speaker recognition approach to achieve this goal. We use a challenge-based method to avoid the replay attack (especially if the impostor has recorded the user's voice). In this case, free text recognition is realized. Experimental results on the CMU database show very good results, while providing low computation times.","Title: Biohashing for Securing Minutiae Template\nYear: 2010\n\nAbstract: The storage of fingerprints is an important issue as this biometric modality is more and more deployed for real applications. The a prori impossibility to revoke a biometric template (like a password) in case of theft, is a major concern for privacy reasons. We propose in this paper a new method to secure fingerprint minutiae templates by storing a bio code while keeping good recognition results. We show the efficiency of the method in comparison to some published methods for different scenarios.","Title: Study on Color Spaces for Single Image Enrolment Face Authentication\nYear: 2010\n\nAbstract: We propose in this paper to study different color spaces for representing an image for the face authentication application. We used a generic algorithm based on a matching of keypoints using sift descriptors computed on one color component. Ten color spaces have been studied on four large and significant benchmark databases (ENSIB, FACES94, AR and FERET). We show that all color spaces do not provide the same efficiency and the use of the color information allows an interesting improvement of verification results.","Title: A contactless e-health information system with privacy\nYear: 2013\n\nAbstract: E-health information systems give rise to many security and privacy concerns. Security of these systems involves a large amount of sensitive data shared by several actors, such as doctors or nurses in various institutions. However, the privacy preserving issue, including data minimization and data sovereignty, is not necessary treated. This paper presents an e-health infrastructure intended to minimize personal data disclosure, with an access to right system based on a secret sharing. We analyze the security of the proposed infrastructure in accordance with medical constraints through contactless transactions.","Title: Soft biometrics database: A benchmark for keystroke dynamics biometric systems\nYear: 2013\n\nAbstract: Among all the existing biometric modalities, authentication systems based on keystroke dynamics are particularly interesting for usability reasons. Many researchers proposed in the last decades some algorithms to increase the efficiency of this biometric modality. Propose in this paper: a benchmark testing suite composed of a database containing multiple data (keystroke dynamics templates, soft biometric traits ...), which will be made available for the research community and a software that is already available for the scientific community for the evaluation of keystroke dynamics based systems. We also built the proposed biometric database on soft biometric traits for keystroke dynamics to suit the experiment. 110 people had voluntarily participated and gave their soft biometrics data i.e. the way of typing gender age and handedness.","Title: Fingerprint quality assessment using a no-reference image quality metric\nYear: 2013\n\nAbstract: The quality assessment of the acquired biometric raw data is very important as it deeply affects the performance of biometric systems and consequently their usability. Poor quality samples increase the enrolment failures, and decrease the system performance. In this paper, we present a new quality assessment metric of fingerprints. Its main originality lies in the use of a no-reference image quality metric. The proposed quality metric combines two types of parameters through a weighted sum optimized by a genetic algorithm: 1) image quality criterion and 2) pattern-based quality criteria (salient and patch-based features). BOZORTH3 matching system and the FVC2002 DB3 fingerprint database are used to clarify the benefits of the presented quality metric.","Title: Analysis of template update strategies for keystroke dynamics\nYear: 2011\n\nAbstract: Keystroke dynamics is a behavioral biometrics showing a degradation of performance when used over time. This is due to the fact that the user improves his/her way of typing while using the system, therefore the test samples may be different from the initial template computed at an earlier stage. One way to bypass this problem is to use template update mechanisms. We propose in this work, new semi-supervised update mechanisms, inspired from known supervised ones. These schemes rely on the choice of two thresholds (an acceptance threshold and an update threshold) which are fixed manually depending on the performance of the system and the level of tolerance in possible inclusion of impostor data in the update template. We also propose a new evaluation scheme for update mechanisms, taking into account performance evolution over several time-sessions. Our results show an improvement of 50% in the supervised scheme and of 45% in the semi-supervised one with a configuration of the parameters chosen so that we do not accept many erroneous data.","Title: Toward a distributed benchmarking tool for biometrics\nYear: 2011\n\nAbstract: Research in computer science evolves very quickly. In order to prove the efficiency of a new algorithm, it is generally necessary to show some results on a large and significant benchmark used in the state of the art. This fact has for consequence the need of a large computation capability in a research laboratory. We address in this paper the performance evaluation of biometric systems through distributed computing. In this domain, we need to realize with the same algorithm many computations on different data (generally corresponding to images). We propose a solution based on a software we developed for facilitating this task. The proposed architecture is composed of a server distributing computation tasks on all the available clients. Experimental results show the benefit of the proposed software.","Title: Comparative study of texture feature for rotation invariant RECOGNITION\nYear: 2007\n\nAbstract: Human visual system easily and rapidly recognizes a scene or image under different affine transformations, which is not the true for the machine. Rotation is more complex than translation and engenders more difficulties in analysis. This paper address evaluation and comparison of texture descriptors, particularly Local Relational String, under rotation effects. Many methods are invariant for geometric transformation, but this is not sufficient to handle the classification problem. We show in this study, when training samples represent a large range of rotated textures, methods with high discriminative properties leads to a very good classification rate despite their no invariance for rotation.","Title: A minutiae level fusion for AFIS systems\nYear: 2009\n\nAbstract: Minutiae correspond essentially to the termination and bifurcation of fingerprint patterns. Since the quality of fingerprint images is often low, extraction algorithms produce a large number of false alarms. A complete minutiae extraction scheme for automatic fingerprint recognition systems is presented. A great deal of work has been dedicated to fingerprint enhancement. We propose a new fusion approach to combine different extraction algorithms. The proposed method, as confirmed by simulations, allows a good overall performance to be achieved suitable for identification forensic applications.","Title: EVALUATING THE SEGMENTATION RESULT OF A GRAY-LEVEL IMAGE\nYear: 2004\n\nAbstract: We propose in this communication an unsupervised crite- rion that enables to quantify the quality of a segmentation result of a gray-level image. The originality of this method lies in the possibility to evaluate the segmentation results for all kinds of images including textured ones. This method is based upon a new criterion that takes into account the intra- region and inter-regions disparities by considering the type of region (textured or uniform). These two disparity mea- sures are computed from adaptive attributes calculated from the segmentation result. Experimental results show the effi- ciency of this technique for synthetic and natural images with textured regions.","Title: Subjective evaluation of image understanding results\nYear: 2010\n\nAbstract: Image understanding has many applications. Given an image and a ground truth, it is possible to measure the quality of understanding results provided by different algorithms or parameters. In this paper, we ask some users to make a subjective evaluation of image understanding results by sorting them from the best to the worst. We compared the results with some provided by a metric we defined recently. Experimental results show the good behavior of this metric compared to the human judgment.","Title: Blind minutiae selection for standard minutiae templates\nYear: 2015\n\nAbstract: The embedded applications of fingerprint proposed so far are chiefly based on the minutiae template. This kind of system is not resource-free and minutiae template is generally sacrificed to cover the shortage. This paper presents several simple yet efficient no-image minutiae selection approaches (NIMS) for the standard minutiae templates (ISO/IEC 19794-2). With the reduced-templates obtained by using the proposed methods, the overall performance can be guaranteed in comparing with the results generated by the original templates. The interoperability tests are performed with several FVC databases. An additional analysis with the quality of the enrollment samples is also carried out. The experimental results demonstrate the validity and efficiency of the proposed approaches.","Title: Operational bio-hash to preserve privacy of fingerprint minutiae templates.\nYear: 2013\n\nAbstract: The storage of fingerprints is an important issue as this biometric modality is more and more deployed for real applications. Considering minutiae templates as sensitive information, a key question concerns the secure and privacy management of this digital identity. Indeed, if an attacker obtains the minutiae template of a subject, he/she will be able to generate a fingerprint having the same char...","Title: Evaluation of biometric systems: a study of users' acceptance and satisfaction\nYear: 2012\n\nAbstract: This paper presents a modality-independent evaluation methodology to study users' acceptance and satisfaction of biometric systems. It uses a survey questionnaire for data collection, and some data-mining tools for their analysis. We have applied it on two biometric systems developed in our research laboratory. The results from this survey show the necessity of taking users point of view when designing and evaluating biometric systems. A panel of 100 volunteers was more satised from the keystroke system than the face one. Users surprisingly considered that its perceived performance was also better, even if the used face system has a better performance with an EER of 8.76% than the keystroke one with an EER of 17.51%. The robustness of a system against attacks, computation time required during the verication phase and its easiness to use have been identied as important factors inuencing their opinions regarding the tested systems.","Title: Evaluation metric of an image understanding result\nYear: 2015\n\nAbstract: Image processing algorithms include methods that process images from their acquisition to the extraction of useful information for a given application. Among interpretation algorithms, some are designed to detect, localize, and identify one or several objects in an image. The problem addressed is the evaluation of the interpretation results of an image or a video given an associated ground truth. Challenges are multiple, such as the comparison of algorithms, evaluation of an algorithm during its development, or the definition of its optimal settings. We propose a new metric for evaluating the interpretation result of an image. The advantage of the proposed metric is to evaluate a result by taking into account the quality of the localization, recognition, and detection of objects of interest in the image. Several parameters allow us to change the behavior of this metric for a given application. Its behavior has been tested on a large database and showed interesting results. (C) 2015 SPIE and IS&T","Title: Comparative study of background subtraction algorithms\nYear: 2010\n\nAbstract: We present a comparative study of several state-of-the-art background subtraction methods. Approaches ranging from simple background subtraction with global thresholding to more sophisticated statistical methods have been implemented and tested on different videos with ground truth. The goal is to provide a solid analytic ground to underscore the strengths and weaknesses of the most widely implemented motion detection methods. The methods are compared based on their robustness to different types of video, their memory requirements, and the computational effort they require. The impact of a Markovian prior as well as some postprocessing operators are also evaluated. Most of the videos used come from state-of-the-art benchmark databases and represent different challenges such as poor SNR, multimodal background motion, and camera jitter. Overall, we not only help to better understand for which type of videos each method best suits but also estimate how, sophisticated methods are better compared to basic background subtraction methods. (C) 2010 SPIE and IS&T. [DOI: 10.1117/1.3456695]","Title: Keystroke dynamics performance enhancement with soft biometrics\nYear: 2015\n\nAbstract: It is accepted that the way a person types on a keyboard contains timing patterns, which can be used to classify him/her, is known as keystroke dynamics. Keystroke dynamics is a behavioural biometric modality, whose performances, however, are worse than morphological modalities such as fingerprint, iris recognition or face recognition. To cope with this, we propose to combine keystroke dynamics with soft biometrics. Soft biometrics refers to biometric characteristics that are not sufficient to authenticate a user (e.g. height, gender, skin/eye/hair colour). Concerning keystroke dynamics, three soft categories are considered: gender, age and handedness. We present different methods to combine the results of a classical keystroke dynamics system with such soft criteria. By applying simple sum and multiply rules, our experiments suggest that the combination approach performs better than the classification approach with best result of 5.41% of equal error rate. The efficiency of our approaches is illustrated on a public database.","Title: Evaluation of the quality of ultrasound image compression by fusion of criteria with a support vector machine\nYear: 2005\n\nAbstract: In the framework of a robotized tele-echography, ultrasound images are compressed and sent from a patient station to an expert one. An important task concerns the evaluation of the quality of the compressed images. Indeed, transmitted images are the only feedback information available to the medical expert to remotely control the distant robotized system and to propose a diagnosis. Our objective is to measure the image quality with a statistical criterion and with the same reliability as the medical assessment. We propose in this work a new method for the comparison of compression results. The proposed approach combines different statistical criteria and uses the medical assessment in a training phase with a support vector machine. We show the benefit of this methodology through some experimental results.","Title: Generation of Local and Expected Behaviors of a Smart Card Application to Detect Software Anomaly\nYear: 2015\n\nAbstract: The electronic payment transaction involves the use of a smart card. A card application is a software, corresponding to standards and non-proprietary and proprietary specifications, and is stored in the smart card. Despite increased security with Euro pay Mastercard Visa (EMV) specifications, attacks still exist due to anomalies in the card application. The validation of the card application enables the detection of any anomaly, improving the overall security of electronic payment transactions. Among the different ways of validating a card application, we can use the verification of required behaviors. These behavior can be materialized as properties of commands sent by the terminal and responses from the smart card, using the Application Protocol Data Unit (APDU) from the ISO/IEC 7816 standard [1]. However, the creation of these behaviors is complicated. We propose in this article a way to automatically create such behaviors by using a genetic algorithm technique.","Title: Fingerprint Quality Assessment with Multiple Segmentation.\nYear: 2015\n\nAbstract: Image quality is an important factor for automated fingerprint identification systems (AFIS) because the matching performance could be significantly affected by poor quality samples. Most of the existing studies mainly focus on calculating a quality index via either a single feature or a combination of multiple features, and some others achieve this purpose with learning approaches which may depend on a prior-knowledge of matching performance. In this paper, a general framework for estimating fingerprint image quality is proposed by fusing features in segmentation phase. The quality index is indicated by a ratio of the pixel number of the integrated foreground area to the size (pixel number) of the fingerprint image. The potential advantage of this framework is that it could be improved by integrating other segmentation approaches or quality features rather than fusing them in a more complicated manner. The experiment is performed with several fingerprint datasets created via different sensors. Experimental results obtained from a dual evaluation approach demonstrate the validity of the proposed method in improving the overall performance.","Title: Quality assessment of fingerprints with minutiae delaunay triangulation\nYear: 2015\n\nAbstract: This article proposes a new quality assessment method of fingerprint, represented by only a set of minutiae points. The proposed quality metric is modeled with the convex-hull and Delaunay triangulation of the minutiae points. The validity of this quality metric is verified on several Fingerprint Verification Competition (FVC) databases by referring to an image-based metric from the state of the art (considered as the reference). The experiments of the utility-based evaluation approach demonstrate that the proposed quality metric is able to generate a desired result. We reveal the possibility of assessing fingerprint quality when only the minutiae template is available.","Title: Fingerprint Quality Assessment Combining Blind Image Quality, Texture and Minutiae Features.\nYear: 2015\n\nAbstract: Biometric sample quality assessment approaches are generally designed in terms of utility property due to thepotential difference between human perception of quality and the biometric quality requirements for a recognitionsystem. This study proposes a utility based quality assessment method of fingerprints by consideringseveral complementary aspects: 1) Image quality assessment without any reference which is consistent withhuman conception of inspecting quality, 2) Textural features related to the fingerprint image and 3) minutiaefeatures which correspond to the most used information for matching. The proposed quality metric is obtainedby a linear combination of these features and is validated with a reference metric using different approaches.Experiments performed on several trial databases show the benefit of the proposed fingerprint quality metric.","Title: EvaBio platform for the evaluation biometric system: Application to the optimization of the enrollment process for fingerprints devices\nYear: 2015\n\nAbstract: Nowadays, when someone wants to make a payment with a smartcard, the user has to enter a pin code to be identified. Only biometrics is able to authenticate a user; yet biometric information is sensitive. To ensure the security and privacy of biometric data, OCC (On-Card-Comparison) has been proposed. This approach consists in storing biometric data in a secure zone on a smartcard and computing the verification decision in a Secure Element (SE). The purpose of this paper is to propose an evaluation platform for testing biometric systems such as the analysis of performance and security on biometric OCC. Based on two examples, we illustrate its different uses in an operational context. The first example focus on the “Quality module” which allows to choose the enrollment by considering the fingerprint quality with one proposed metric. The second one addresses the minutiae reduction of the fingerprint template when the number of minutiae is higher than expected by the OCC.","Title: Comparative study on texture features for fingerprint recognition: application to the BioHashing template protection scheme.\nYear: 2016\n\nAbstract: In order to ensure privacy constraints in biometrics, new algorithms called template protection schemes have been proposed in the last 10 years in the literature. Most of these algorithms require as input a feature having a fixed size. Texture features can be used within this context for fingerprints as the number of minutiae varies in general for different captures. BioHashing is a two authentication factor algorithm that can be used to ensure privacy while using biometrics. In this work, we compare different recent texture features from the literature within the BioHashing scheme while considering many constraints: efficiency, maximal representation size, and constant size description. Experiments are conducted on three fingerprint databases from the FVC competition. Results permit us to conclude on the texture features to be used within this context. (C) 2016 SPIE and IS&T","Title: Literature review of fingerprint quality assessment and its evaluation.\nYear: 2016\n\nAbstract: Fingerprint quality assessment (FQA) has been a challenging issue due to a variety of noisy information contained in the samples, such as physical defect and distortions caused by sensing devices. Existing studies have made efforts to find out more suitable techniques for assessing fingerprint quality but it is difficult to achieve a common solution because of, for example, different image setting...","Title: Memory Carving in Embedded Devices: Separate the Wheat from the Chaff.\nYear: 2016\n\nAbstract: This paper investigates memory carving techniques for embedded devices. Given that cryptographic material in memory dumps makes carving techniques inefficient, we introduce a methodology to distinguish meaningful information from cryptographic material in small-sized memory dumps. The proposed methodology uses an adaptive boosting technique with statistical tests. Experimented on EMV cards, the methodology recognized 92% of meaningful information and 98% of cryptographic material.","Title: An Observe-and-Detect Methodology for the Security and Functional Testing of Smart Card Applications.\nYear: 2016\n\nAbstract: Smart cards are tamper resistant devices but vulnerabilities are sometimes discovered. We address in this paper the security and the functional testing of embedded applications in smart cards. We propose an original methodology for the evaluation of applications and we show its benefit by comparing it to a classical certification process. The proposed method is based on the observation of the APDU (Application Protocol Data unit) communication with the smart card. Some specific properties are verified as a complementary method in the evaluation process and allows the on-the-fly detection of an anomaly and the reasons that triggered this anomaly during the test. Here are presented two uses of this method: a simple use to illustrate the use of properties to verify an implementation of an application and a more complex illustration by applying the fuzzing method to show what we can obtain with the proposed approach, i.e. an analysis of an anomaly.","Title: A Continuous LoA Compliant Trust Evaluation Method.\nYear: 2016\n\nAbstract: The trust provided by authentication systems is commonly expressed with a Level of Assurance (LoA see 3). If it can be considered as a first process to simplify the expression of trust during the authentication step, it does not handle all the aspects of the authentication mechanism and especially it fails to integrate continuous authentication systems. In this paper, we propose a model based on the Dempster Shafer theory to merge continuous authentication system with more traditional static authentication scheme and to assign a continuous trust level to the current LoA. In addition, this method is proved to be compliant with the LoA frameworks.","Title: In What Way Is It Possible to Impersonate You Bypassing Fingerprint Sensors?\nYear: 2016\n\nAbstract: In this study, we want to determine if an attacker with different a priori information on the sensor parameters during the enrollment process can impersonate an individual. We have tested different apriorism such as the fingerprint class, sensor type, image resolution and the number of minutiae. Two different matching algorithms have been used on specific designed databases generated with SFinge. Two attack scenarii have been tested. Obtained results show that the knowledge of fingerprint class and resolution influence the attack success.","Title: Synchronous One Time Biometrics with Pattern Based Authentication\nYear: 2016\n\nAbstract: One time passwords are commonly used for authentication purposes in electronic transactions. Nevertheless, providing such a one time password is not really a strong authentication proof %as it can be given by an impostor. because the token generating the passwords can be given by an impostor. In order to cope with this problem, biometric recognition is more and more employed. Even if biometric data are strongly linked with the user, their revocability nor diversity is possible, without an adapted post-processing. Biometric template protection schemes, including the BioHashing algorithm, are used to manage the underlying privacy and security issues. These schemes are used for the protection of several biometric modalities, but are not necessary adapted for all of them. In this paper, we propose a new protocol combining protected biometric data and a classical synchronous one time password to enhance the security of user authentication while preserving usability and privacy. Behavioral biometrics is used to provide a fast and a usable solution for users. We show through experimental results the efficiency of the proposed method.","Title: Privacy Preserving Transparent Mobile Authentication\nYear: 2017\n\nAbstract: Transparent authentication on mobile phones suffers from privacy issues especially when biometric information is involved. In this paper, we propose a solution to address those two issues using the Biohashing algorithm on behavioral information extracted from a mobile phone. The authentication scenario is tested on a dataset composed of 100 users and shows promising results with a 10% EER in the worst case scenario (i.e when protection key is compromised) and a 1% EER in the best case one. In addition, privacy concerns are discussed and experimentally evaluated both in a quantitative and qualitative ways. This opens new perspectives concerning online authentication using smartphone sensing abilities.","Title: Fingerprint Class Recognition For Securing Emv Transaction\nYear: 2017\n\nAbstract: Fingerprint analysis is a very important issue in biometry. The minutiae representation of a fingerprint is the most used modality to identify people or authorize access when using a biometric system. In this paper, we propose some features based on triangle parameters from the Delaunay triangulation of minutiae. We show the benefit of these features to recognize the type of a fingerprint without any access to the associated fingerprint image.","Title: Memory carving can finally unveil your embedded personal data\nYear: 2017\n\nAbstract: Smart cards are involved in most of activities, and they gather and record plenty of personal data. A manual interpretation of these raw data is difficult without specifications. This task becomes really tedious applied to plenty of devices. The paper introduces the first method to automatically retrieve textual information from memory dumps of smart cards. Given the data structure and encoding are assumed to be unknown, the method is based on text statistics and characteristics of smart cards to discard false positives. The experiments performed on more than 350 memory dumps revealed that the method can automatically retrieve more than 99% of textual information available in a dump, while keeping the false positive rate as low as 5.5%.","Title: Contrasting False Identities in Social Networks by Trust Chains and Biometric Reinforcement\nYear: 2017\n\nAbstract: Fake identities and identity theft are issues whose relevance is increasing in the social network domain. This paper deals with this problem by proposing an innovative approach which combines a collaborative mechanism implementing a trust graph with keystroke-dynamic-recognition techniques to trust identities. The trust of each node is computed on the basis of neighborhood recognition and behavioral biometric support. The model leverages the word of mouth propagation and a settable degree of redundancy to obtain robustness. Experimental results show the benefit of the proposed solution even if attack nodes are present in the social network.","Title: Retrieving dates in smart card dumps is as hard as finding a needle in a haystack\nYear: 2017\n\nAbstract: This paper introduces a method to automatically retrieve dates from smart card memory dumps when the card specifications are unknown. It exploits specificities of smart cards, using a multi-dump analysis augmented with contextual information. The experiments performed on more than 180 real smart cards show that our method is highly successful in removing false positives.","Title: Vision-Based System for Human Detection and Tracking in Indoor Environment\nYear: 2010\n\nAbstract: In this paper, we propose a vision-based system for human detection and tracking in indoor environment us- ing a static camera. The proposed method is based on ob- ject recognition in still images combined with methods us- ing temporal information from the video. Doing that, we improve the performance of the overall system and reduce the task complexity. We first use background subtraction to limit the search space of the classifier. The segmentation is realized by modeling each background pixel by a single Gaussian model. As each connected component detected by the background subtraction potentially corresponds to one person, each blob is independently tracked. The tracking process is based on the analysis of connected components position and interest points tracking. In order to know the nature of various objects that could be present in the scene, we use multiple cascades of boosted classifiers based on Haar-like filters. We also present in this article a wide eval- uation of this system based on a large set of videos.","Title: Comparative Study of Digital Fingerprint Quality Assessment Metrics\nYear: 2018\n\nAbstract: The quality assessment of biometric data proceeds as a toll to decide whether a biometric sample may be used to generate the user's reference template. Many studies showed its significant impact on the subsequent performance of the biometric system. Since many metrics are proposed for this purpose by researchers or standardization institutions, their relevance should be studied in particular to evaluate their relative usefulness. This paper provides a comparative study of fingerprint quality assessment (FQA) metrics. We consider the enrollment selection validation approach to perform an objective comparison of them. We show the efficiency of 7 well known FQA metrics on 9 datasets. Results show a dependency of these metrics to the dataset (i.e. the fingerprint sensor) and for matchers a similar behavior.","Title: Rhu Keystroke Touchscreen Benchmark\nYear: 2018\n\nAbstract: Biometric systems are currently widely used in many applications to control and verify individual's identity. Keystroke dynamics modality has been shown as a promising solution that would be used in many applications such as e-payment and banking applications. However, such systems suffer from several performance limitations (such as cross-devices problem) that prevent their widespread of use in real applications. The objective of this paper is to provide researchers and developers with a public touchscreen-based benchmark collected using a mobile phone and a tablet (both portrait and landscape orientation each). Such a benchmark can be used to assess keystroke-based matching algorithms. Furthermore, It is mainly developed to measure the robustness of keystroke matching algorithms vis-'a-vis cross-devices and orientation variations. An online visualizer for the database is also given to researchers allowing them to visualize the acquired keystroke signals.","Title: Enhancing the Security of Transformation Based Biometric Template Protection Schemes\nYear: 2018\n\nAbstract: Template protection is a crucial issue in biometrics. Many algorithms have been proposed in the literature among secure computing approaches, crypto-biometric algorithm and feature transformation schemes. The BioHashing algorithm belongs to this last category and has very interesting properties. Among them, we can cite its genericity since it could be applied on any biometric modality, the possible cancelability of the generated BioCode and its efficiency when the secret is not stolen by an impostor. Its main drawback is its weakness face to a combined attack (zero effort with the stolen secret scenario). In this paper, we propose a transformation-based biometric template protection scheme as an improvement of the BioHashing algorithm where the projection matrix is generated by combining the secret and the biometric data. Experimental results on two biometric modalities, namely digital fingerprint and finger knuckle print images, show the benefits of the proposed method face to attacks while keeping a good efficiency.","Title: User Dependent Template Update for Keystroke Dynamics Recognition\nYear: 2018\n\nAbstract: Regarding the fact that individuals have different interactions with biometric authentication systems, several techniques have been developed in the literature to model different users categories. Doddington Zoo is a concept of categorizing users behaviors into animal groups to reflect their characteristics with respect to biometric systems. This concept was developed for different biometric modalities including keystroke dynamics. The present study extends this biometric classification, by proposing a novel adaptive strategy based on the Doddinghton Zoo, for the recognition of the user's keystroke dynamics. The obtained results demonstrate competitive performances on significant keystroke dynamics datasets.","Title: Analysis of Keystroke Dynamics for the Generation of Synthetic Datasets\nYear: 2018\n\nAbstract: Biometrics is an emerging technology more and more present in our daily life. However, building biometric systems requires a large amount of data that may be difficult to collect. Collecting such sensitive data is also very time consuming and constrained, s.a. GDPR legislation. In the case of keystroke dynamics, existing databases have less than 200 users. For these reasons, we aim at generating a keystroke dynamics synthetic dataset. This paper presents the generation of keystroke data from known users as a first step towards the generation of synthetic datasets, and could also be used to impersonate users' identity.","Title: A COMPARATIVE STUDY OF SUPERVISED EVALUATION CRITERIA FOR IMAGE SEGMENTATION\nYear: 2004\n\nAbstract: This paper presents a comparative study of five supervised evaluation criteria for image segmentation. The different cri- teria have been tested on a selection of hundred images ex- tracted from the c Corel database for which manual segmen- tation results provided by experts are available. Nine seg- mentation algorithms have been considered, most of which are based on threshold selection. In order to compare the behavior of the different criteria towards over- and underseg- mentation, three thresholds have been taken into account, for each selected image, to simulate the various situations. Ex- perimental results permit to reveal the advantages and limi- tations of the studied criteria.","Title: EvaBio a New Modular Platform to Evaluate Biometric System\nYear: 2015\n\nAbstract: when someone wants to make a payment with a smartcard, the user has to enter a pin code to be identified. Only biometrics is able to authenticate a user; yet biometric information is sensitive. To ensure the security and privacy of biometric data, OCC (On-Card-Comparison) has been proposed. This approach consists in storing biometric data in a secure zone on a smartcard and computing the verification decision in a Secure Element (SE). The purpose of this paper is to propose an evaluation platform for testing biometric systems such as the analysis of performance and security on biometric OCC. Based on four examples, we illustrate its different uses in an operationnal context. The first example focus on the \"Quality module\" which allows to choose the enrollment by considering the fingerprint quality with one proposed metric. The second one addresses the minutiae reduction of the fingerprint template when the number of minutiae is higher than expected by the OCC. The third is based on sensors acquisition module to create databases and made attacks on sensors. The last one is the evaluation module, it permit to visualize results after an evaluation.","Title: Double serial adaptation mechanism for keystroke dynamics authentication based on a single password.\nYear: 2019\n\nAbstract: Cyber-attacks have spread all over the world to steal information such as trade secrets, intellectual property and banking data. Facing the danger of the insecurity of saved data (personal, professional, official, etc.), keystroke dynamics was proposed as an interesting, non-intrusive, inexpensive, permanent and weakly constrained solution for users. Based on the typing rhythm of users, it improves logical access security. Nevertheless, it was demonstrated that such an authentication mechanism would need a larger number of samples to enroll the typing characteristics of users. Moreover, these registered characteristics generally undergo aging effects after a time span. Different solutions have been suggested to remedy these variability problems, including template adaptation. In this paper, we propose a double serial adaptation strategy that considers a single-capture-based enrollment process. When using the authentication system, the template of users and the decision/adaptation thresholds are updated. Experimental results on three public keystroke dynamics datasets show the benefits of the proposed method.","Title: Analysis of Doddington zoo classification for user dependent template update: Application to keystroke dynamics recognition\nYear: 2019\n\nAbstract: Biometric authentication systems are increasingly considered in different access control applications. Regarding that users have completely different interactions with these authentication systems, several techniques have been developed in the literature to model distinctive users categories. Doddington zoo is a biometric menagerie that defines and labels user groups with animal species to reflect their behavior with the biometric systems. This menagerie was developed for different biometric modalities including keystroke dynamics. The present study proposes a user dependent adaptive strategy based on the Doddington zoo, for the recognition of the user’s keystroke dynamics. The novelty of the proposed approach lies in applying an adaptive strategy specific to the characteristics of each user of the Doddington zoo menagerie aiming to solve the intra-class variation problems. The obtained results demonstrate competitive performances on significant keystroke dynamics datasets WEBGREYC and CMU.","Title: Statistical modeling of keystroke dynamics samples for the generation of synthetic datasets\nYear: 2019\n\nAbstract: Biometrics is an emerging technology more and more present in our daily life. However, building biometric systems requires a large amount of data that may be difficult to collect. Collecting such sensitive data is also very time consuming and constrained, s.a. GDPR legislation in Europe. In the case of keystroke dynamics, most existing databases have less than 200 users. For these reasons, it is crucial for this biometric modality to be able to generate a significant and realistic synthetic dataset of keystroke dynamics samples. We propose in this paper an original approach for the generation of synthetic keystroke data given samples from known users as a first step towards the generation of synthetic datasets. Experimental results show the capability of the proposed statistical model to generate realistic samples from existing datasets in the literature.","Title: Privacy Aura for Transparent Authentication on Multiple Smart Devices\nYear: 2021\n\nAbstract: Nowadays, users carry multiple connected devices such as a smartphone, laptop, connected watch.... Security constraints limit user's usability especially when using all of them intensively during the day (social media, work). In this paper, we propose the privacy Aura concept corresponding to the circle of trust in the neighborhood of each smart device to facilitate user authentication. Many data (phone calls habits, biometrics, localization) can be collected to realize a transparent and privacy compliant authentication on each device. The confidence on user authentication on each device can be transferred to another one if it is located in the same Aura. This is the main contribution of the paper. We show through illustrations the benefit of the proposed solution.","Title: Classifying Biometric Systems Users among the Doddington Zoo: Application to Keystroke Dynamics\nYear: 2021\n\nAbstract: Doddington zoo defines four categories of users when using a biometric system related to their difficulty to be recognized or attacked. In this paper, we propose an original work consisting in predicting for any biometric modality the associated animal in the Doddington menagerie related to a user given few captured biometric samples. Such a prediction could be useful for many applications, as for example, to adapt the behavior of biometric systems to each user. In this work, we apply this methodology to keystroke dynamics as it is an interesting behavioral biometric modality for user authentication. It consists in analyzing the way of typing of a user in order to recognize him/her. We use a significant keystroke dynamics dataset and we demonstrate through experimental results the benefit of the proposed approach.","Title: Face Transparent User Authentication Respecting Privacy\nYear: 2021\n\nAbstract: User authentication is an important issue to guarantee the security of information systems. For users, this task is in general inconvenient and boring. We propose in this paper a new transparent user authentication based on face recognition. During the use of a laptop or a smartphone, face features are computed, protected and sent to a trust party server that measures at any time the confidence of...","Title: Keystroke Dynamics Classification Based On LSTM and BLSTM Models\nYear: 2021\n\nAbstract: By adopting keystroke dynamics, authentication applications can integrate advanced identity proofing technology for detecting fraud and prevent unauthorized access. However, understanding a user's keystroke dynamics behavior in real applications is a challenging task regarding that this behavior is notably changing over time. To mitigate this problem, we apply, in this paper, the long short-term m...","Title: How Artificial Intelligence can be used for Behavioral Identification?\nYear: 2021\n\nAbstract: Nowadays, users interact with computer systems. Behavioral biometrics consists of analyzing user's interactions for identification and verification applications. This approach could be very useful for enhancing security and improving user experience and many privacy concerns are also related. In this paper, we address the problem of user identification considering their behaviors. How efficient ar...","Title: Impact Of Environmental Conditions On Fingerprint Systems Performance\nYear: 2021\n\nAbstract: Biometrics testing has for objective to determine the performance of a biometric system in order to guarantee security and user experience requirements. Providing trust in biometric systems is a key for many manufacturers. The performance is usually measured through the computation of matching scores between legitimate and impostor samples from a given database. Different bias in particular those ..."],"string":"[\n \"Title: An e-payment Architecture Ensuring a High Level of Privacy Protection.\\nYear: 2013\\n\\nAbstract: Online shopping is becoming more and more interesting for clients because of the ease of use and the large choice of products. As a consequence, 2.3 billion online clients have been identified in 2011. This rapid increase was accompagnied by various frauds, including stolen smart cards or fraudulent repudiation. Several e-payment systems have been proposed to reduce these security threats and the 3D-Secure protocol is becoming a standard for the payment on the Internet. Nevertheless, this protocol has not been studied in-depth, particularly in terms of privacy. This paper proposes a detailed description and an analysis of the 3D-Secure protocol, through a new privacy-orienting model for e-payment architectures. Some improvements of 3D-Secure protocol, concerning the protection of banking information, are also presented. Then, this article presents and analyses a new online payment architecture centered on the privacy of individuals.\",\n \"Title: Low Cost and Usable Multimodal Biometric System Based on Keystroke Dynamics and 2D Face Recognition\\nYear: 2010\\n\\nAbstract: We propose in this paper a low cost multimodal biometric system combining keystroke dynamics and 2D face recognition. The objective of the proposed system is to be used while keeping in mind: good performances, acceptability, and espect of privacy. Different fusion methods have been used (min, max, mul, svm, weighted sum configured with genetic algorithms, and, genetic programming) on the scores of three keystroke dynamics algorithms and two 2D face recognition ones. This multimodal biometric system improves the recognition rate in comparison with each individual method. On a chimeric database composed of 100 individuals, the best keystroke dynamics method obtains an EER of 8.77%, the best face recognition one has an EER of 6.38%, while the best proposed fusion system provides an EER of 2.22%.\",\n \"Title: A tele-operated mobile ultrasound scanner using a light-weight robot.\\nYear: 2005\\n\\nAbstract: This paper presents a new tele-operated robotic chain for real-time ultrasound image acquisition and medical diagnosis. This system has been developed in the frame of the Mobile Tele-Echography Using an Ultralight Robot European Project. A light-weight six degrees-of-freedom serial robot, with a remote center of motion, has been specially designed for this application. It holds and moves a real probe on a distant patient according to the expert gesture and permits an image acquisition using a standard ultrasound device. The combination of mechanical structure choice for the robot and dedicated control law, particularly nearby the singular configuration allows a good path following and a robotized gesture accuracy. The choice of compression techniques for image transmission enables a compromise between flow and quality. These combined approaches, for robotics and image processing, enable the medical specialist to better control the remote ultrasound probe holder system and to receive stable and good quality ultrasound images to make a diagnosis via any type of communication link from terrestrial to satellite. Clinical tests have been performed since April 2003. They used both satellite or Integrated Services Digital Network lines with a theoretical bandwidth of 384 Kb/s. They showed the tele-echography system helped to identify 66% of lesions and 83% of symptomatic pathologies.\",\n \"Title: A new soft biometric approach for keystroke dynamics based on gender recognition\\nYear: 2012\\n\\nAbstract: Keystroke dynamics allows to authenticate individuals through their way of typing on a computer keyboard. In this study, we are interested in static shared secret keystroke dynamics (all the users type the same password). We present new soft biometrics information which can be extracted from keystroke typing patterns: the gender of the user. This is the first study, to our knowledge, experimenting such kind of information in the field of keystroke dynamics. We present a method for gender recognition through keystroke dynamics with more than 91% of accuracy, on the tested dataset, and we show the improvement on keystroke dynamics authentication method using such kind of information through pattern and score fusion. We obtain a gain of 20% when using gender information against a classical keystroke dynamics method.\",\n \"Title: Genetic programming for multibiometrics\\nYear: 2012\\n\\nAbstract: Biometric systems suffer from some drawbacks: a biometric system can provide in general good performances except with some individuals as its performance depends highly on the quality of the capture... One solution to solve some of these problems is to use multibiometrics where different biometric systems are combined together (multiple captures of the same biometric modality, multiple feature extraction algorithms, multiple biometric modalities...). In this paper, we are interested in score level fusion functions application (i.e., we use a multibiometric authentication scheme which accept or deny the claimant for using an application). In the state of the art, the weighted sum of scores (which is a linear classifier) and the use of an SVM (which is a non linear classifier) provided by different biometric systems provide one of the best performances. We present a new method based on the use of genetic programming giving similar or better performances (depending on the complexity of the database). We derive a score fusion function by assembling some classical primitives functions (+,*,-,... ). We have validated the proposed method on three significant biometric benchmark datasets from the state of the art.\",\n \"Title: Service provider authentication assurance\\nYear: 2012\\n\\nAbstract: The concept of authentication assurance traditionally refers to the robustness of methods and mechanisms for user authentication, including the robustness of initial registration and provisioning of user credentials, as well as the robustness of mechanisms that enforce user authentication during operation. However, the user is not the only party that needs to be authenticated to ensure security of online transactions. In fact, online service provision always involves two parties, typically the user on the client side and the service provider on the server side, so that mutual authentication between the two sides is required. In contrast to the unilateral focus on user authentication by industry and academia, it is in fact equally important for the user to correctly authenticate the service provider. Unfortunately, little attention is paid to the problem of correctly authentication the service provider. This paper proposes a framework for server and service provider authentication assurance, similarly to frameworks for user authentication assurance that have already been specified, or are currently under development by many national governments.\",\n \"Title: 2D color shape recognition using Zernike moments\\nYear: 2005\\n\\nAbstract: 2D Zernike moments belong to the useful object invariant descriptors which have been successfully applied in pattern recognition tasks. The main problem of using Zernike moments invariants is that they are not able to discriminate two objects having the same shape. In this paper, an approach based on Zernike moments applied on color images is proposed. A Support Vector Machine is used for object classification. For object segmentation, the connected component labeling algorithm is used. Compared with the classical method, this approach shows higher accuracy in object recognition. Some experimental results on the COIL-100 database are presented.\",\n \"Title: Abnormal events detection based on spatio-temporal co-occurences\\nYear: 2009\\n\\nAbstract: We explore a location-basedapproachforbehaviormod- eling and abnormality detection. In contrast to the con- ventional object-based approach where an object may first be tagged, identified, classified, and tracked, we proceed directly with event characterization and behavior model- ing at the pixel(s) level based on motion labels obtained from background subtraction. Since events are temporally and spatially dependent, this calls for techniques that ac- count for statistics of spatio-temporal events. Based on mo- tion labels, we learn co-occurrence statistics for normal events across space-time. For one (or many) key pixel(s), we estimate a co-occurrence matrix that accounts for any two active labels which co-occur simultaneously within the same spatio-temporal volume. This co-occurrence matrix is then used as a potential function in a Markov Random Field (MRF) model to describe the probability of observa- tions within the same spatio-temporal volume. The MRF distribution implicitly accounts for speed, direction, as well as the average size of the objects passing in front of each key pixel. Furthermore, when the spatio-temporal volume is large enough, the co-occurrence distribution contains the average normal path followed by moving objects. The learned normal co-occurrence distribution can be used for abnormaldetection. Our method hasbeen tested on various outdoor videos representing various challenges.\",\n \"Title: Segmentation framework based on label field fusion.\\nYear: 2007\\n\\nAbstract: In this paper, we put forward a novel fusion framework that mixes together label fields instead of observation data as is usually the case. Our framework takes as input two label fields: a quickly estimated and to-be-refined segmentation map and a spatial region map that exhibits the shape of the main objects of the scene. These two label fields are fused together with a global energy function that is minimized with a deterministic iterative conditional mode algorithm. As explained in the paper, the energy function may implement a pure fusion strategy or a fusion-reaction function. In the latter case, a data-related term is used to make the optimization problem well posed. We believe that the conceptual simplicity, the small number of parameters, the use of a simple and fast deterministic optimizer that admits a natural implementation on a parallel architecture are among the main advantages of our approach. Our fusion framework is adapted to various computer vision applications among which are motion segmentation, motion estimation and occlusion detection.\",\n \"Title: Fast computation of the performance evaluation of biometric systems: Application to multibiometrics\\nYear: 2013\\n\\nAbstract: The performance evaluation of biometric systems is a crucial step when designing and evaluating such systems. The evaluation process uses the Equal Error Rate (EER) metric proposed by the International Organization for Standardization (ISO/IEC). The EER metric is a powerful metric which allows easily comparing and evaluating biometric systems. However, the computation time of the EER is, most of the time, very intensive. In this paper, we propose a fast method which computes an approximated value of the EER. We illustrate the benefit of the proposed method on two applications: the computing of non parametric confidence intervals and the use of genetic algorithms to compute the parameters of fusion functions. Experimental results show the superiority of the proposed EER approximation method in term of computing time, and the interest of its use to reduce the learning of parameters with genetic algorithms. The proposed method opens new perspectives for the development of secure multibiometrics systems by speeding up their computation time.\",\n \"Title: Enhancing the privacy of electronic passports\\nYear: 2012\\n\\nAbstract: We address in this paper the problem of privacy in the current architecture in electronic passports for the storage and transmission of biometric data such as fingerprints. The current architecture provides a good protection of biometric personal data but brute force attack could be used in the near future using cloud computing. We propose a new solution combining cryptographic protocols and cancellable biometrics to solve this problem. The biocode of an individual is protected by cryptographic keys exchanged by the PACE protocol. We put into obviousness the benefit of the proposed solution in terms of security and privacy.\",\n \"Title: Comparative study of global invariant descriptors for object recognition\\nYear: 2008\\n\\nAbstract: Although many object invariant descriptors have been proposed in the literature, putting them into practice to obtain a robust recognition system that is able to face several perturbations is still a studied problem. After presenting the most commonly used global invariant descriptors, a comparative study permits us to show their ability to discriminate between objects with little training. The Columbia Object Image Library database (COIL-100), which presents a same object translated, rotated, and scaled, is used to test the invariant features of geometrical transforms. Partial object occultation or presence of complex background are examples of used images to test the robustness of the studied descriptors. We compare them in both a global and a local context (computed on the neighborhood of a pixel). The scale invariant feature transform descriptor is used as a reference for local invariant descriptors. This study shows the relative performance of invariant descriptors used in both a global and a local context and identifies the different situations for which they are best suited. (C) 2008 SPIE and IS&T.\",\n \"Title: A Real Time Human Detection System Based on Far Infrared Vision\\nYear: 2008\\n\\nAbstract: We present in this article a human detection and tracking algorithm using infrared vision in order to have reliable information on a room occupation. We intend to use this information to limit energetic consumption (light, heating). We perform first, a foreground segmentation with a Gaussian background model. A tracking step based on connected components intersections allows to collect information on 2D displacements of moving objects in the image plane. A classification based on a cascade of boosted classifiers is used for the recognition. Experimental results show the efficiency of the proposed algorithm.\",\n \"Title: Performance Evaluation of Biometric Template Update\\nYear: 2012\\n\\nAbstract: Template update allows to modify the biometric reference of a user while he uses the biometric system. With such kind of mechanism we expect the biometric system uses always an up to date representation of the user, by capturing his intra-class (temporary or permanent) variability. Although several studies exist in the literature, there is no commonly adopted evaluation scheme. This does not ease the comparison of the different systems of the literature. In this paper, we show that using different evaluation procedures can lead in different, and contradictory, interpretations of the results. We use a keystroke dynamics (which is a modality suffering of template ageing quickly) template update system on a dataset consisting of height different sessions to illustrate this point. Even if we do not answer to this problematic, it shows that it is necessary to normalize the template update evaluation procedures.\",\n \"Title: A New Protocol to Evaluate the Resistance of Template Update Systems against Zero-Effort Attacks\\nYear: 2013\\n\\nAbstract: Since several years, there is a high interest in the creation of template update mechanisms in biometrics. By automatically updating over time the biometric reference of an individual, these update systems aim at avoiding a performance decreases over time. However, update errors occur when impostor samples are included in the biometric reference. In this paper, we are interested in analysing the behavior of template update against zero effort attacks depending on the difficulty of these attacks which is parametrized by the ratio of impostors samples included. We propose a protocol, which allows to evaluate an update mechanism under such kind of attacks by preserving an important amount of genuine samples, even when the ratio of impostor samples is important. Then, we apply this protocol on an existing template update system from the state of the art at various attack rates in order to assess its benefits.\",\n \"Title: Comparative study of localization metrics for the evaluation of image interpretation systems\\nYear: 2010\\n\\nAbstract: Image interpretation is particularly important in many real applications (video monitoring, biometrics, etc.). Due to the proliferation of image interpretation systems in the literature, their evaluation still remains a crucial stake. Among all the tasks in this field, the quality of object localization is often evaluated through an evaluation metric. We propose to review these techniques and study their reliability. We first propose a generic definition of a localization algorithm. Then, different state of the art techniques to evaluate image interpretation results are detailed. Secondly, we focus on metrics that enable us to evaluate localization results. We propose a general methodology to analyze the behavior of an evaluation metric, considered here as a black box (its definition is not even supposed to be known). We define the properties that these metrics should fulfill. We then perform a comparative study of 33 localization metrics from the state of the art. Experimental results conducted on a large and significant image database permit us to determine metrics that should be used in the future for the evaluation of object localization results. (C) 2010 SPIE and IS&T. [DOI: 10.1117/1.3446803]\",\n \"Title: Calibration and quality control of cherries by artificial vision\\nYear: 2004\\n\\nAbstract: We describe an artificial vision system for the quality control of cherries. We develop image-processing software determining three types of information describing the quality of the fruit: the color as an indicator of ripeness, the presence of defects such as cracking, and the size. The sorting of cherries conditioned by all these criteria is carried out at a high cadence (20 cherries/s). Real-time image processing is then necessary. A simple sensor is used to synchronize the processing and the sorting of cherries. Air actuators are used to eject them after quality control by vision. We present the architecture of the developed system. We show its efficiency through experimental results and we give some perspectives of improvement. (C) 2004 SPIE and IST.\",\n \"Title: Comparative study of contour detection evaluation criteria based on dissimilarity measures\\nYear: 2008\\n\\nAbstract: We present in this article a comparative study of well-known supervised evaluation criteria that enable the quantification of the quality of contour detection algorithms. The tested criteria are often used or combined in the literature to create new ones. Though these criteria are classical ones, none comparison has been made, on a large amount of data, to understand their relative behaviors. The objective of this article is to overcome this lack using large test databases both in a synthetic and a real context allowing a comparison in various situations and application fields and consequently to start a general comparison which could be extended by any person interested in this topic. After a review of the most common criteria used for the quantification of the quality of contour detection algorithms, their respective performances are presented using synthetic segmentation results in order to show their performance relevance face to undersegmentation, oversegmentation, or situations combining these two perturbations. These criteria are then tested on natural images in order to process the diversity of the possible encountered situations. The used databases and the following study can constitute the ground works for any researcher who wants to confront a new criterion face to well-known ones.\",\n \"Title: Evaluation Protocol for Localization Metrics\\nYear: 2008\\n\\nAbstract: Localization metrics permit to quantify the correctness of object detection in an image interpretation result. This paper deals with the definition of a protocol in order to evaluate the behavior of localization metrics. We first define some properties that metrics should verify and create a synthetic database that enables to verify those properties on different metrics. After presenting the tested localization metrics, the results obtained following the proposed protocol are exposed. Finally, some conclusions and perspectives are given.\",\n \"Title: Evaluation metric for image understanding\\nYear: 2009\\n\\nAbstract: We propose in this paper a new evaluation metric that enables to quantify the quality of an image interpretation result. This metric takes into account the a priori knowledge used by the interpretation algorithm and the ground truth associated with the original image. We combine two metrics that evaluate the localization and recognition results of each detected object. We show that the proposed metric fulfills some theoretical properties and has a correct behavior face to empirical experiments on an image benchmark database. We think that this metric could be a reliable reference for image and video understanding competitions.\",\n \"Title: Evaluation of Human Detection Algorithms in Image Sequences\\nYear: 2010\\n\\nAbstract: This paper deals with the general evaluation of human detection algorithms. We first present the algorithms implemented within the CAPTHOM project dedicated to the development of a vision-based system for human detection and tracking in an indoor environment using a static camera. We then show how a global evaluation metric we developped for the evaluation of understanding algorithms taking into account both localization and recognition precision of each single interpretation result, can be a useful tool for industrials to guide them in the elaboration of suitable and optimized algorithms.\",\n \"Title: Evolutionary computation approaches for shape modelling and fitting\\nYear: 2005\\n\\nAbstract: This paper proposes and analyzes different evolutionary computation techniques for conjointly determining a model and its associated parameters. The context of 3D reconstruction of objects by a functional representation illustrates the ability of the proposed approaches to perform this task using real data, a set of 3D points on or near the surface of the real object. The final recovered model can then be used efficiently in further modelling, animation or analysis applications. The first approach is based on multiple genetic algorithms that find the correct model and parameters by successive approximations. The second approach is based on a standard strongly-typed implementation of genetic programming. This study shows radical differences between the results produced by each technique on a simple problem, and points toward future improvements to join the best features of both approaches.\",\n \"Title: Comparative Study of Local Descriptors for Measuring Object Taxonomy\\nYear: 2009\\n\\nAbstract: Many object descriptors have been proposed in the state of the art. For many reasons (occlusion, point of view, acquisition conditions...), local descriptors have a better robustness for image understanding applications. The goal of this paper is to make a comparative study of eight recent local descriptors. The objective is here to quantify their ability to generate automatically an object taxonomy. In order to answer this question, we use the Caltech256 benchmark which provides a large object taxonomy used as reference. This study shows that SIFT, differential invariants and shape context descriptors are the best ones to achieve this goal.\",\n \"Title: Segmentation evaluation using a support vector machine\\nYear: 2005\\n\\nAbstract: Segmentation evaluation is a very difficult task even for an expert. We propose in this article a new unsupervised evaluation criterion of an image segmentation result. The quality of a segmentation result is derived without any a priori knowledge by taking into account different evaluation criteria from the literature. We first compare six unsupervised evaluation criteria on a database composed of synthetic gray level images. Vinet's measure is used as an objective function to compare the behavior of the different criteria. We propose in this paper to fuse the best ones by a support vector machine. We illustrate the efficiency of the proposed approach through some experimental results.\",\n \"Title: Can Chronological Information be Used as a Soft Biometric in Keystroke Dynamics?\\nYear: 2012\\n\\nAbstract: Keystroke dynamics is a behavioral biometric modality which uses typing patterns on a keyboard to recognize individuals. The way of typing the password slightly changes with time, because of various factors (including user's training). This modification in the way of typing results in a decrease of performance recognition over time. In this paper, we analyse the correlation between the comparison score between a query, and a reference and the number of times the user has typed the password. After having quantified his correlation, we analyse the possibility of using stacked classification to ake this aspect in consideration during authentication. Then, we verify if it is possible to classify users on their way of evolving their typing pattern. Results show that even if comparison scores are correlated with the number of time the user has typed the password, the use of a stacked classifier does not improve the results much.\",\n \"Title: Unconstrained keystroke dynamics authentication with shared secret\\nYear: 2011\\n\\nAbstract: mong all the existing biometric modalities, authentication systems based on keystroke dynamics present interesting advantages. These solutions are well accepted by users and cheap as no additional sensor is required for authenticating the user before accessing to an application. In the last thirty years, many researchers have proposed, different algorithms aimed at increasing the performance of this approach. Their main drawback lies on the large number of data required for the enrollment step. As a consequence, the verification system is barely usable, because the enrollment is too restrictive. In this work, we propose a new method based on the Support Vector Machine (SVM) learning satisfying industrial conditions (i.e., few samples per user are needed during the enrollment phase to create its template). In this method, users are authenticated through the keystroke dynamics of a shared secret (chosen by the system administrator). We use the GREYC keystroke database that is composed of a large number of users (100) for validation purposes. We compared the proposed method with six methods from the literature (selected based on their ability to work with few enrollment samples). Experimental results show that, even though the computation time to build the template can be longer with our method (54 s against 3 s for most of the others), its performance outperforms the other methods in an industrial context (Equal Error Rate of 15.28% against 16.79% and 17.02% for the two best methods of the state-of-the-art, on our dataset and five samples to create the template, with a better computation time than the second best method).\",\n \"Title: Multi-Agent Based Prototyping of Agriculture Robots\\nYear: 2006\\n\\nAbstract: In this paper, we propose a multi-agents system for the simulation of prototypes for different agriculture robots that can be employed in the harvesting of a vineyard. We show that this prototyping method works well and is appropriate for a system involving several robots. We present the behaviors of each robot, their interactions and some scenarios for the testing of the system. Of specific interest is the cooperative behavior of the robots in such a task. Preliminary results indicate that some optimal choices toward the development of the harvesting system can be correctly identified based on these simulations. The choices can then be used when designing and building the system prototype.\",\n \"Title: A Support Vector Machine As An Estimator Of Mountain Papaya Ripeness Using Resonant Frequency Or Frequency Centroid\\nYear: 2006\\n\\nAbstract: Mountain papaya fruits (Vasconcella pubescens) were tested for firmness with a nondestructive acoustic method for 14 days after harvest. The response of each fruit was analyzed with the Fourier transform to obtain a firmness index (FI) based on the second resonant frequency and with the Short Time Fourier Transform (STFT) to obtain a spectrogram frequency centroid (FC) index. The indexes were processed with a support vector machine (SVM) learning procedure in which days since harvest was taken as the basic truth of ripeness which the measured indexes attempt to estimate. The analysis of the results demonstrate that different groupings of the days into classes to be estimated give widely varying recognition rates and that the best rates are obtained when the classes are delimited using prior knowledge.\",\n \"Title: Toward a complete adaptive analysis of an image\\nYear: 2003\\n\\nAbstract: We present an adaptive method that allows the classification of the regions of an image in two classes: textured and uniform regions (weak gray-level variance). Information on the type of texture (stochastic, deterministic) and its granularity (macroscopic, microscopic) are extracted. The developed method is achieved in two steps. It first enables the determination of the global context of an, image (image mainly composed of uniform areas or textured ones) and the localization of the textured and uniform areas.. The second step characterizes each detected area by considering some appropriate features: mean and variance in the case of uniform regions and classical relevant texture attributes (derived from a statistical analysis) associated with some new attributes that we define in the case of textured regions. These complementary features are determined from a texture model derived from the Wold decomposition of the autocovariance function. They enable the acquisition of some information on the type of texture and its granularity. We show the efficiency of the method through two examples of image segmentation. (C) 2003 SPIE and IST.\",\n \"Title: Secure payment with NFC mobile phone in the SmartTouch project.\\nYear: 2008\\n\\nAbstract: This paper deals with the ITEA (Information Technology for European Advancement) SmartTouch project whose one of the objectives is to use NFC mobile phones for a secure payment. An experience in real conditions has been carried out at Strasbourg in France, in 2007. This trial is the first experiment of a NFC-based payment application that fully supports the international EMV standard and the PayPass program. We focus in this paper on the security analysis which can occur in a NFC mobile phone payment. In the first section, we give a simplified description of the prototype. In the second section, we present the security concepts as a major issue in this scenario. Roughly, the main requirement is to have the same level of security as standard EMV payment transactions. The methodology described in the third section concerns the payment application and its communication with the payment terminal through a NFC link. The last part of this article presents different research perspectives to overstep the encountered problems.\",\n \"Title: Soft Biometrics For Keystroke Dynamics\\nYear: 2013\\n\\nAbstract: Keystroke dynamics is a viable and practical way as an addition to security for identity verification. It can be combined with passphrases authentication resulting in a more secure verification system. This paper presents a new soft biometric approach for keystroke dynamics. Soft biometrics traits are physical, behavioral or adhered human characteristics, which have been derived from the way human beings normally distinguish their peers (e.g. height, gender, hair color etc.). Those attributes have a low discriminating power, thus not capable of identification performance. Additionally, they are fully available to everyone which makes them privacy-safe. Thus, in this study, it consists of extracting information from the keystroke dynamics templates with the ability to recognise the hand(s) used (i.e. one/two hand(s)); the gender; the age category; and the handedness of a user when he/she types a given password or passphrase on a keyboard. Experiments were conducted on a keystroke dynamics database of 110 users and our experimental results show that the proposed methods are promising.\",\n \"Title: Keystroke dynamics authentication for collaborative systems\\nYear: 2009\\n\\nAbstract: We present in this paper a study on the ability and the benefits of using a keystroke dynamics authentication method for collaborative systems. Authentication is a challenging issue in order to guarantee the security of use of collaborative systems during the access control step. Many solutions exist in the state of the art such as the use of one time passwords or smart-cards We focus in this paper on biometric based solutions that do not necessitate any additional sensor. Keystroke dynamics is an interesting solution as it uses only the keyboard and is invisible for users. Many methods have been published in this field. We make a comparative study of many of them considering the operational constraints of use for collaborative systems.\",\n \"Title: Texture detection for image analysis\\nYear: 2005\\n\\nAbstract: Many applications such as image compression, pre-processing or segmentation require some information from the regions composing an image. The main objective of this paper is to define a methodology to extract some local information from an image. Each region is characterized in terms of homogeneity (region composed with the same grey-level or a single texture) and its type (textured or uniform). The decision criterion is based on the use of classical texture attributes (cooccurrence matrix and grey-levels moments) and a support vector machine in order to realize the fusion of the different attributes. We then characterize each region considering its type by appropriate features.\",\n \"Title: Web-Based Benchmark for Keystroke Dynamics Biometric Systems: A Statistical Analysis\\nYear: 2012\\n\\nAbstract: Most keystroke dynamics studies have been evaluated using a specific kind of dataset in which users type an imposed login and password. Moreover, these studies are optimistics since most of them use different acquisition protocols, private datasets, controlled environment, etc. In order to enhance the accuracy of keystroke dynamics' performance, the main contribution of this paper is twofold. First, we provide a new kind of dataset in which users have typed both an imposed and a chosen pairs of logins and passwords. In addition, the keystroke dynamics samples are collected in a web-based uncontrolled environment (OS, keyboards, browser, etc.). Such kind of dataset is important since it provides us more realistic results of keystroke dynamics' performance in comparison to the literature (controlled environment, etc.). Second, we present a statistical analysis of well known assertions such as the relationship between performance and password size, impact of fusion schemes on system overall performance, and others such as the relationship between performance and entropy. We put into obviousness in this paper some new results on keystroke dynamics in realistic conditions.\",\n \"Title: An integrated framework combining Bio-Hashed minutiae template and PKCS15 compliant card for a better secure management of fingerprint cancelable templates\\nYear: 2013\\n\\nAbstract: We address in this paper the problem of privacy in fingerprint biometric systems. Today, cancelable techniques have been proposed to deal with this issue. Ideally, such transforms are one-way. However, even if they are with provable security, they remain vulnerable when the user-specific key that achieves cancelability property is stolen. The prominence of the cancelable template confidentiality to maintain the irreversibility property was also demonstrated for many proposed constructions. To prevent possible coming attacks, it becomes important to securely manage this key as well as the transformed template in order to avoid them being leaked simultaneously and thus compromised. To better manage the user credentials of cancelable constructs, we propose a new solution combining a trusted architecture and a cancelable fingerprint template. Therefore, a Bio-Hashed minutiae template based on a chip matching algorithm is proposed. A pkcs15 compliant cancelable biometric system for fingerprint privacy preserving is implemented on a smartcard. This closed system satisfies the safe management of the sensitive templates. The proposed solution is proved to be well resistant to different attacks.\",\n \"Title: Unsupervised performance evaluation of image segmentation\\nYear: 2006\\n\\nAbstract: We present in this paper a study of unsupervised evaluation criteria that enable the quantification of the quality of an image segmentation result. These evaluation criteria compute some statistics for each region or class in a segmentation result. Such an evaluation criterion can be useful for different applications: the comparison of segmentation results, the automatic choice of the best fitted parameters of a segmentation method for a given image, or the definition of new segmentation methods by optimization. We first present the state of art of unsupervised evaluation, and then, we compare six unsupervised evaluation criteria. For this comparative study, we use a database composed of 8400 synthetic gray-level images segmented in four different ways. Vinet's measure (correct classification rate) is used as an objective criterion to compare the behavior of the different criteria. Finally, we present the experimental results on the segmentation evaluation of a few gray-level natural images.\",\n \"Title: Palm Vein Verification System Based on SIFT Matching\\nYear: 2009\\n\\nAbstract: We present in this communication a new biometric system based on the use of hand veins acquired by an infrared imager. After the preprocessing stage and binarization, the vein image is characterized by specific patterns. One originality of the proposed system is to use SIFT descriptors for the verification process. The developed method only necessitates a single image for the enrollment step allowing a very fast verification. The experimental results on a database containing images of 24 individuals acquired after two sessions show the efficiency of the proposed method.\",\n \"Title: A new supervised evaluation criterion for region based segmentation methods\\nYear: 2007\\n\\nAbstract: We present in this article a new supervised evaluation criterion that enables the quantification of the quality of region segmentation algorithms. This criterion is compared with seven well-known criteria available in this context. To that end, we test the different methods on natural images by using a subjective evaluation involving different experts from the French community in image processing. Experimental results show the benefit of this new criterion.\",\n \"Title: Detecting Half-Occlusion with a Fast Region-Based Fusion Procedure\\nYear: 2006\\n\\nAbstract: This paper presents a novel region-based approach for detecting occlusion between two consecutive frames. Based on a generalization of Marr and Poggio's uniqueness assumption, the explicit goal of our method is to reduce the number of false positives while optimizing the hit rate. To do so, our method relies on a fusion procedure that blends together two segmentation maps: one pre-estimated occlusion binary map and one color segmentation map. While the occlusion map is obtained after a simple thresholding proce- dure, the color segmentation map is obtained with an unsupervised Marko- vian approach. Assuming that the color segmentation regions exhibit more precise edges, the occlusion areas are iteratively modified to fit the color- region shapes. Since our method has been entirely implemented on a parallel architecture (a Graphics Processor Unit), its processing times are remarkably low. Our method is compared with other occlusion approaches both quanti- tatively and qualitatively on scenes that represent different challenges.\",\n \"Title: Quality assessment of image-based biometric information.\\nYear: 2015\\n\\nAbstract: The quality of biometric raw data is one of the main factors affecting the overall performance of biometric systems. Poor biometric samples increase the enrollment failure and decrease the system performance. Hence, controlling the quality of the acquired biometric raw data is essential in order to have useful biometric authentication systems. Towards this goal, we present a generic methodology for the quality assessment of image-based biometric modality combining two types of information: 1) image quality and 2) pattern-based quality using the scale-invariant feature transformation (SIFT) descriptor. The associated metric has the advantages of being multimodal (face, fingerprint, and hand veins) and independent from the used authentication system. Six benchmark databases and one biometric verification system are used to illustrate the benefits of the proposed metric. A comparison study with the National Institute of Standards and Technology (NIST) fingerprint image quality (NFIQ) metric proposed by the NIST shows the benefits of the presented metric.\",\n \"Title: Image Watermarking with Biometric Data for Copyright Protection\\nYear: 2015\\n\\nAbstract: In this paper, we deal with the proof of ownership or legitimate usage of a digital content, such as an image, in order to tackle the illegitimate copy. The proposed scheme based on the combination of the watermarking and cancelable biometrics does not require a trusted third party, all the exchanges are between the provider and the customer. The use of cancelable biometrics allows us to provide a privacy compliant proof of identity. We illustrate the robustness of this method against intentional and unintentional attacks of the watermarked content.\",\n \"Title: Soft biometrics for keystroke dynamics: Profiling individuals while typing passwords\\nYear: 2014\\n\\nAbstract: This paper presents a new profiling approach of individuals based on soft biometrics for keystroke dynamics. Soft biometric traits are unique representation of a person, which can be in a form of physical, behavioural or biological human characteristics that differentiate between him/her into a group people (e.g. gender, age, height, colour, race etc.). Keystroke dynamics is a behavioural biometric modality to recognise how a person types on a keyboard. In this paper, we consider the following soft traits: the hand category (i.e. if the user types with one or two hands), the gender category, the age category and the handedness category. For this purpose, we collected a new database. Two cases are studied: static passwords and free text. By combining machine learning and fusion process, the results are promising.\",\n \"Title: One-Time Biometrics for Online Banking and Electronic Payment Authentication.\\nYear: 2014\\n\\nAbstract: Online banking and electronic payment systems on the Internet are becoming increasingly advanced. On the machine level, transactions take place between client and server hosts through a secure channel protected with SSL/TLS. User authentication is typically based on two or more factors. Nevertheless, the development of various malwares and social engineering attacks transform the user's PC in an untrusted device and thereby making user authentication vulnerable. This paper investigates how user authentication with biometrics can be made more robust in the online banking context by using a specific device called OffPAD. This context requires that authentication is realized by the bank and not only by the user (or by the personal device) contrary to standard banking systems. More precisely, a new protocol for the generation of one- time passwords from biometric data is presented, ensuring the security and privacy of the entire transaction. Experimental results show an excellent performance considering with regard to false positives. The security analysis of our protocol also illustrates the benefits in terms of strengthened security.\",\n \"Title: Analysis of embedded applications by evolutionary fuzzing\\nYear: 2014\\n\\nAbstract: In this paper, we propose to use fuzzing techniques to discover vulnerabilities in programs hosted into smart cards used for telecommunications or banking purposes (SIM cards, credit cards, secure element into NFC mobile devices...). Those programs - called applets - usually host sensitive applications and manipulate sensitive data. A flaw by design or by implementation in one of those applet could have disastrous consequences. The proposed approach uses a genetic algorithm to optimize the vulnerabilities search. We illustrate the benefit of the proposed method on a MasterCard M/Chip applet through experimental results.\",\n \"Title: Towards a New Tool for the Evaluation of the Quality of Ultrasound Compressed Images\\nYear: 2006\\n\\nAbstract: This paper presents a new tool for the evaluation of ultrasound image compression. The goal is to measure the image quality as easily as with a statistical criterion, and with the same reliability as the one provided by the medical assessment. An initial experiment is proposed to medical experts and represents our reference value for the comparison of evaluation criteria. Twenty-one statistical criteria are selected from the literature. A cumulative absolute similarity measure is defined as a distance between the criterion to evaluate and the reference value. A first fusion method based on a linear combination of criteria is proposed to improve the results obtained by each of them separately. The second proposed approach combines different statistical criteria and uses the medical assessment in a training phase with a support vector machine. Some experimental results are given and show the benefit of fusion.\",\n \"Title: Similarity-based matching for face authentication\\nYear: 2008\\n\\nAbstract: We propose in this paper a face authentication method based on a similarity measure. The SIFT de- scriptor is used to define some interest keypoints char- acterized by an invariant parameter. A graph is then built where nodes correspond to these keypoints. We model the authentication problem as a graph matching process. Experimental results on the AR database show an EER equals to 12% with only one image used for the enrollment and with images simulating real conditions.\",\n \"Title: Emerging trends in biometric authentication\\nYear: 2009\\n\\nAbstract: The main objective of biometrics is to identify or authenticate an individual, given his/her different physical or behavioral characteristics. This approach has already been employed in our daily life in various applications and forms (USB biometric key, access control, ...). Nevertheless, this technology could be used for multiple other applications. In this presentation, we survey the trends that have been observed for the proliferation of this technology such as: definitions of contemporary unconstrained biometric systems that necessitate or call for low cost sensors and few steps for enrollment (some examples based on face, veins and keystroke dynamics will illustrate this point); evaluation of biometric systems that take into account performance, acceptability, and security; and privacy related applications, such as those in biohashing techniques.\",\n \"Title: Security EvaBio: An analysis tool for the security evaluation of biometric authentication systems\\nYear: 2012\\n\\nAbstract: Biometric systems present several drawbacks that may significantly decrease their utility. Nowadays, several platforms (such as the FVC-onGoing) exist to assess the performance of such systems. Despite this, none platform exists for the security evaluation of biometric systems. Hence, the aim of this paper is to present an on-line platform for the security evaluation of biometric systems. The key benefits of the presented platform are twofold. First, it provides the research community in biometrics an evaluation tool to assess biometric systems in term of security. Second, the platform provides a database of common threats and vulnerabilities of biometric systems that can be updated by researchers feedbacks. The presented tool is modality-independent. A keystroke dynamics system is used to illustrate the benefits of the presented platform.\",\n \"Title: Texture based fingerprint BioHashing: Attacks and robustness\\nYear: 2012\\n\\nAbstract: BioHashing is a popular biometric template protection scheme defined in the last decade. Most of previous studies on this algorithm focus on the performance optimization or the use of this privacy protection scheme on many types of biometric modalities (face, fingerprint, palmprint...). The objective of this paper is to study the robustness of this algorithm on a texture based representation by testing and simulating operational attacks. We consider in this study some quantitative measures of the robustness of the BioHashing algorithm and we show some new results on its security on fingerprints represented by texture features.\",\n \"Title: Analysis of the acquisition process for keystroke dynamics\\nYear: 2012\\n\\nAbstract: In order to evaluate authentication methods for keystroke dynamics, it is necessary to create new datasets. We present in this paper an analysis study of the factors, involved during the acquisition process, which affect the performance of keystroke-based authentication systems. More generally speaking, we are looking for the optimal keystroke data acquisition scenario which may deeply affect the performance of such systems. Results show that (1) it is better to choose passwords based on short, simple and known words; and (2) during the acquisition process of a new keystroke dynamics dataset, there is no significant preference for displaying or telling to a user the text to type.\",\n \"Title: Challenge-based speaker recognition for mobile authentication\\nYear: 2012\\n\\nAbstract: User authentication is a major trend to guarantee the security of electronic transactions when using mobile devices such as tablets or mobile phones. Biometrics is for us the only real user authentication method. In this article, we propose to realize a speaker recognition approach to achieve this goal. We use a challenge-based method to avoid the replay attack (especially if the impostor has recorded the user's voice). In this case, free text recognition is realized. Experimental results on the CMU database show very good results, while providing low computation times.\",\n \"Title: Biohashing for Securing Minutiae Template\\nYear: 2010\\n\\nAbstract: The storage of fingerprints is an important issue as this biometric modality is more and more deployed for real applications. The a prori impossibility to revoke a biometric template (like a password) in case of theft, is a major concern for privacy reasons. We propose in this paper a new method to secure fingerprint minutiae templates by storing a bio code while keeping good recognition results. We show the efficiency of the method in comparison to some published methods for different scenarios.\",\n \"Title: Study on Color Spaces for Single Image Enrolment Face Authentication\\nYear: 2010\\n\\nAbstract: We propose in this paper to study different color spaces for representing an image for the face authentication application. We used a generic algorithm based on a matching of keypoints using sift descriptors computed on one color component. Ten color spaces have been studied on four large and significant benchmark databases (ENSIB, FACES94, AR and FERET). We show that all color spaces do not provide the same efficiency and the use of the color information allows an interesting improvement of verification results.\",\n \"Title: A contactless e-health information system with privacy\\nYear: 2013\\n\\nAbstract: E-health information systems give rise to many security and privacy concerns. Security of these systems involves a large amount of sensitive data shared by several actors, such as doctors or nurses in various institutions. However, the privacy preserving issue, including data minimization and data sovereignty, is not necessary treated. This paper presents an e-health infrastructure intended to minimize personal data disclosure, with an access to right system based on a secret sharing. We analyze the security of the proposed infrastructure in accordance with medical constraints through contactless transactions.\",\n \"Title: Soft biometrics database: A benchmark for keystroke dynamics biometric systems\\nYear: 2013\\n\\nAbstract: Among all the existing biometric modalities, authentication systems based on keystroke dynamics are particularly interesting for usability reasons. Many researchers proposed in the last decades some algorithms to increase the efficiency of this biometric modality. Propose in this paper: a benchmark testing suite composed of a database containing multiple data (keystroke dynamics templates, soft biometric traits ...), which will be made available for the research community and a software that is already available for the scientific community for the evaluation of keystroke dynamics based systems. We also built the proposed biometric database on soft biometric traits for keystroke dynamics to suit the experiment. 110 people had voluntarily participated and gave their soft biometrics data i.e. the way of typing gender age and handedness.\",\n \"Title: Fingerprint quality assessment using a no-reference image quality metric\\nYear: 2013\\n\\nAbstract: The quality assessment of the acquired biometric raw data is very important as it deeply affects the performance of biometric systems and consequently their usability. Poor quality samples increase the enrolment failures, and decrease the system performance. In this paper, we present a new quality assessment metric of fingerprints. Its main originality lies in the use of a no-reference image quality metric. The proposed quality metric combines two types of parameters through a weighted sum optimized by a genetic algorithm: 1) image quality criterion and 2) pattern-based quality criteria (salient and patch-based features). BOZORTH3 matching system and the FVC2002 DB3 fingerprint database are used to clarify the benefits of the presented quality metric.\",\n \"Title: Analysis of template update strategies for keystroke dynamics\\nYear: 2011\\n\\nAbstract: Keystroke dynamics is a behavioral biometrics showing a degradation of performance when used over time. This is due to the fact that the user improves his/her way of typing while using the system, therefore the test samples may be different from the initial template computed at an earlier stage. One way to bypass this problem is to use template update mechanisms. We propose in this work, new semi-supervised update mechanisms, inspired from known supervised ones. These schemes rely on the choice of two thresholds (an acceptance threshold and an update threshold) which are fixed manually depending on the performance of the system and the level of tolerance in possible inclusion of impostor data in the update template. We also propose a new evaluation scheme for update mechanisms, taking into account performance evolution over several time-sessions. Our results show an improvement of 50% in the supervised scheme and of 45% in the semi-supervised one with a configuration of the parameters chosen so that we do not accept many erroneous data.\",\n \"Title: Toward a distributed benchmarking tool for biometrics\\nYear: 2011\\n\\nAbstract: Research in computer science evolves very quickly. In order to prove the efficiency of a new algorithm, it is generally necessary to show some results on a large and significant benchmark used in the state of the art. This fact has for consequence the need of a large computation capability in a research laboratory. We address in this paper the performance evaluation of biometric systems through distributed computing. In this domain, we need to realize with the same algorithm many computations on different data (generally corresponding to images). We propose a solution based on a software we developed for facilitating this task. The proposed architecture is composed of a server distributing computation tasks on all the available clients. Experimental results show the benefit of the proposed software.\",\n \"Title: Comparative study of texture feature for rotation invariant RECOGNITION\\nYear: 2007\\n\\nAbstract: Human visual system easily and rapidly recognizes a scene or image under different affine transformations, which is not the true for the machine. Rotation is more complex than translation and engenders more difficulties in analysis. This paper address evaluation and comparison of texture descriptors, particularly Local Relational String, under rotation effects. Many methods are invariant for geometric transformation, but this is not sufficient to handle the classification problem. We show in this study, when training samples represent a large range of rotated textures, methods with high discriminative properties leads to a very good classification rate despite their no invariance for rotation.\",\n \"Title: A minutiae level fusion for AFIS systems\\nYear: 2009\\n\\nAbstract: Minutiae correspond essentially to the termination and bifurcation of fingerprint patterns. Since the quality of fingerprint images is often low, extraction algorithms produce a large number of false alarms. A complete minutiae extraction scheme for automatic fingerprint recognition systems is presented. A great deal of work has been dedicated to fingerprint enhancement. We propose a new fusion approach to combine different extraction algorithms. The proposed method, as confirmed by simulations, allows a good overall performance to be achieved suitable for identification forensic applications.\",\n \"Title: EVALUATING THE SEGMENTATION RESULT OF A GRAY-LEVEL IMAGE\\nYear: 2004\\n\\nAbstract: We propose in this communication an unsupervised crite- rion that enables to quantify the quality of a segmentation result of a gray-level image. The originality of this method lies in the possibility to evaluate the segmentation results for all kinds of images including textured ones. This method is based upon a new criterion that takes into account the intra- region and inter-regions disparities by considering the type of region (textured or uniform). These two disparity mea- sures are computed from adaptive attributes calculated from the segmentation result. Experimental results show the effi- ciency of this technique for synthetic and natural images with textured regions.\",\n \"Title: Subjective evaluation of image understanding results\\nYear: 2010\\n\\nAbstract: Image understanding has many applications. Given an image and a ground truth, it is possible to measure the quality of understanding results provided by different algorithms or parameters. In this paper, we ask some users to make a subjective evaluation of image understanding results by sorting them from the best to the worst. We compared the results with some provided by a metric we defined recently. Experimental results show the good behavior of this metric compared to the human judgment.\",\n \"Title: Blind minutiae selection for standard minutiae templates\\nYear: 2015\\n\\nAbstract: The embedded applications of fingerprint proposed so far are chiefly based on the minutiae template. This kind of system is not resource-free and minutiae template is generally sacrificed to cover the shortage. This paper presents several simple yet efficient no-image minutiae selection approaches (NIMS) for the standard minutiae templates (ISO/IEC 19794-2). With the reduced-templates obtained by using the proposed methods, the overall performance can be guaranteed in comparing with the results generated by the original templates. The interoperability tests are performed with several FVC databases. An additional analysis with the quality of the enrollment samples is also carried out. The experimental results demonstrate the validity and efficiency of the proposed approaches.\",\n \"Title: Operational bio-hash to preserve privacy of fingerprint minutiae templates.\\nYear: 2013\\n\\nAbstract: The storage of fingerprints is an important issue as this biometric modality is more and more deployed for real applications. Considering minutiae templates as sensitive information, a key question concerns the secure and privacy management of this digital identity. Indeed, if an attacker obtains the minutiae template of a subject, he/she will be able to generate a fingerprint having the same char...\",\n \"Title: Evaluation of biometric systems: a study of users' acceptance and satisfaction\\nYear: 2012\\n\\nAbstract: This paper presents a modality-independent evaluation methodology to study users' acceptance and satisfaction of biometric systems. It uses a survey questionnaire for data collection, and some data-mining tools for their analysis. We have applied it on two biometric systems developed in our research laboratory. The results from this survey show the necessity of taking users point of view when designing and evaluating biometric systems. A panel of 100 volunteers was more satised from the keystroke system than the face one. Users surprisingly considered that its perceived performance was also better, even if the used face system has a better performance with an EER of 8.76% than the keystroke one with an EER of 17.51%. The robustness of a system against attacks, computation time required during the verication phase and its easiness to use have been identied as important factors inuencing their opinions regarding the tested systems.\",\n \"Title: Evaluation metric of an image understanding result\\nYear: 2015\\n\\nAbstract: Image processing algorithms include methods that process images from their acquisition to the extraction of useful information for a given application. Among interpretation algorithms, some are designed to detect, localize, and identify one or several objects in an image. The problem addressed is the evaluation of the interpretation results of an image or a video given an associated ground truth. Challenges are multiple, such as the comparison of algorithms, evaluation of an algorithm during its development, or the definition of its optimal settings. We propose a new metric for evaluating the interpretation result of an image. The advantage of the proposed metric is to evaluate a result by taking into account the quality of the localization, recognition, and detection of objects of interest in the image. Several parameters allow us to change the behavior of this metric for a given application. Its behavior has been tested on a large database and showed interesting results. (C) 2015 SPIE and IS&T\",\n \"Title: Comparative study of background subtraction algorithms\\nYear: 2010\\n\\nAbstract: We present a comparative study of several state-of-the-art background subtraction methods. Approaches ranging from simple background subtraction with global thresholding to more sophisticated statistical methods have been implemented and tested on different videos with ground truth. The goal is to provide a solid analytic ground to underscore the strengths and weaknesses of the most widely implemented motion detection methods. The methods are compared based on their robustness to different types of video, their memory requirements, and the computational effort they require. The impact of a Markovian prior as well as some postprocessing operators are also evaluated. Most of the videos used come from state-of-the-art benchmark databases and represent different challenges such as poor SNR, multimodal background motion, and camera jitter. Overall, we not only help to better understand for which type of videos each method best suits but also estimate how, sophisticated methods are better compared to basic background subtraction methods. (C) 2010 SPIE and IS&T. [DOI: 10.1117/1.3456695]\",\n \"Title: Keystroke dynamics performance enhancement with soft biometrics\\nYear: 2015\\n\\nAbstract: It is accepted that the way a person types on a keyboard contains timing patterns, which can be used to classify him/her, is known as keystroke dynamics. Keystroke dynamics is a behavioural biometric modality, whose performances, however, are worse than morphological modalities such as fingerprint, iris recognition or face recognition. To cope with this, we propose to combine keystroke dynamics with soft biometrics. Soft biometrics refers to biometric characteristics that are not sufficient to authenticate a user (e.g. height, gender, skin/eye/hair colour). Concerning keystroke dynamics, three soft categories are considered: gender, age and handedness. We present different methods to combine the results of a classical keystroke dynamics system with such soft criteria. By applying simple sum and multiply rules, our experiments suggest that the combination approach performs better than the classification approach with best result of 5.41% of equal error rate. The efficiency of our approaches is illustrated on a public database.\",\n \"Title: Evaluation of the quality of ultrasound image compression by fusion of criteria with a support vector machine\\nYear: 2005\\n\\nAbstract: In the framework of a robotized tele-echography, ultrasound images are compressed and sent from a patient station to an expert one. An important task concerns the evaluation of the quality of the compressed images. Indeed, transmitted images are the only feedback information available to the medical expert to remotely control the distant robotized system and to propose a diagnosis. Our objective is to measure the image quality with a statistical criterion and with the same reliability as the medical assessment. We propose in this work a new method for the comparison of compression results. The proposed approach combines different statistical criteria and uses the medical assessment in a training phase with a support vector machine. We show the benefit of this methodology through some experimental results.\",\n \"Title: Generation of Local and Expected Behaviors of a Smart Card Application to Detect Software Anomaly\\nYear: 2015\\n\\nAbstract: The electronic payment transaction involves the use of a smart card. A card application is a software, corresponding to standards and non-proprietary and proprietary specifications, and is stored in the smart card. Despite increased security with Euro pay Mastercard Visa (EMV) specifications, attacks still exist due to anomalies in the card application. The validation of the card application enables the detection of any anomaly, improving the overall security of electronic payment transactions. Among the different ways of validating a card application, we can use the verification of required behaviors. These behavior can be materialized as properties of commands sent by the terminal and responses from the smart card, using the Application Protocol Data Unit (APDU) from the ISO/IEC 7816 standard [1]. However, the creation of these behaviors is complicated. We propose in this article a way to automatically create such behaviors by using a genetic algorithm technique.\",\n \"Title: Fingerprint Quality Assessment with Multiple Segmentation.\\nYear: 2015\\n\\nAbstract: Image quality is an important factor for automated fingerprint identification systems (AFIS) because the matching performance could be significantly affected by poor quality samples. Most of the existing studies mainly focus on calculating a quality index via either a single feature or a combination of multiple features, and some others achieve this purpose with learning approaches which may depend on a prior-knowledge of matching performance. In this paper, a general framework for estimating fingerprint image quality is proposed by fusing features in segmentation phase. The quality index is indicated by a ratio of the pixel number of the integrated foreground area to the size (pixel number) of the fingerprint image. The potential advantage of this framework is that it could be improved by integrating other segmentation approaches or quality features rather than fusing them in a more complicated manner. The experiment is performed with several fingerprint datasets created via different sensors. Experimental results obtained from a dual evaluation approach demonstrate the validity of the proposed method in improving the overall performance.\",\n \"Title: Quality assessment of fingerprints with minutiae delaunay triangulation\\nYear: 2015\\n\\nAbstract: This article proposes a new quality assessment method of fingerprint, represented by only a set of minutiae points. The proposed quality metric is modeled with the convex-hull and Delaunay triangulation of the minutiae points. The validity of this quality metric is verified on several Fingerprint Verification Competition (FVC) databases by referring to an image-based metric from the state of the art (considered as the reference). The experiments of the utility-based evaluation approach demonstrate that the proposed quality metric is able to generate a desired result. We reveal the possibility of assessing fingerprint quality when only the minutiae template is available.\",\n \"Title: Fingerprint Quality Assessment Combining Blind Image Quality, Texture and Minutiae Features.\\nYear: 2015\\n\\nAbstract: Biometric sample quality assessment approaches are generally designed in terms of utility property due to thepotential difference between human perception of quality and the biometric quality requirements for a recognitionsystem. This study proposes a utility based quality assessment method of fingerprints by consideringseveral complementary aspects: 1) Image quality assessment without any reference which is consistent withhuman conception of inspecting quality, 2) Textural features related to the fingerprint image and 3) minutiaefeatures which correspond to the most used information for matching. The proposed quality metric is obtainedby a linear combination of these features and is validated with a reference metric using different approaches.Experiments performed on several trial databases show the benefit of the proposed fingerprint quality metric.\",\n \"Title: EvaBio platform for the evaluation biometric system: Application to the optimization of the enrollment process for fingerprints devices\\nYear: 2015\\n\\nAbstract: Nowadays, when someone wants to make a payment with a smartcard, the user has to enter a pin code to be identified. Only biometrics is able to authenticate a user; yet biometric information is sensitive. To ensure the security and privacy of biometric data, OCC (On-Card-Comparison) has been proposed. This approach consists in storing biometric data in a secure zone on a smartcard and computing the verification decision in a Secure Element (SE). The purpose of this paper is to propose an evaluation platform for testing biometric systems such as the analysis of performance and security on biometric OCC. Based on two examples, we illustrate its different uses in an operational context. The first example focus on the “Quality module” which allows to choose the enrollment by considering the fingerprint quality with one proposed metric. The second one addresses the minutiae reduction of the fingerprint template when the number of minutiae is higher than expected by the OCC.\",\n \"Title: Comparative study on texture features for fingerprint recognition: application to the BioHashing template protection scheme.\\nYear: 2016\\n\\nAbstract: In order to ensure privacy constraints in biometrics, new algorithms called template protection schemes have been proposed in the last 10 years in the literature. Most of these algorithms require as input a feature having a fixed size. Texture features can be used within this context for fingerprints as the number of minutiae varies in general for different captures. BioHashing is a two authentication factor algorithm that can be used to ensure privacy while using biometrics. In this work, we compare different recent texture features from the literature within the BioHashing scheme while considering many constraints: efficiency, maximal representation size, and constant size description. Experiments are conducted on three fingerprint databases from the FVC competition. Results permit us to conclude on the texture features to be used within this context. (C) 2016 SPIE and IS&T\",\n \"Title: Literature review of fingerprint quality assessment and its evaluation.\\nYear: 2016\\n\\nAbstract: Fingerprint quality assessment (FQA) has been a challenging issue due to a variety of noisy information contained in the samples, such as physical defect and distortions caused by sensing devices. Existing studies have made efforts to find out more suitable techniques for assessing fingerprint quality but it is difficult to achieve a common solution because of, for example, different image setting...\",\n \"Title: Memory Carving in Embedded Devices: Separate the Wheat from the Chaff.\\nYear: 2016\\n\\nAbstract: This paper investigates memory carving techniques for embedded devices. Given that cryptographic material in memory dumps makes carving techniques inefficient, we introduce a methodology to distinguish meaningful information from cryptographic material in small-sized memory dumps. The proposed methodology uses an adaptive boosting technique with statistical tests. Experimented on EMV cards, the methodology recognized 92% of meaningful information and 98% of cryptographic material.\",\n \"Title: An Observe-and-Detect Methodology for the Security and Functional Testing of Smart Card Applications.\\nYear: 2016\\n\\nAbstract: Smart cards are tamper resistant devices but vulnerabilities are sometimes discovered. We address in this paper the security and the functional testing of embedded applications in smart cards. We propose an original methodology for the evaluation of applications and we show its benefit by comparing it to a classical certification process. The proposed method is based on the observation of the APDU (Application Protocol Data unit) communication with the smart card. Some specific properties are verified as a complementary method in the evaluation process and allows the on-the-fly detection of an anomaly and the reasons that triggered this anomaly during the test. Here are presented two uses of this method: a simple use to illustrate the use of properties to verify an implementation of an application and a more complex illustration by applying the fuzzing method to show what we can obtain with the proposed approach, i.e. an analysis of an anomaly.\",\n \"Title: A Continuous LoA Compliant Trust Evaluation Method.\\nYear: 2016\\n\\nAbstract: The trust provided by authentication systems is commonly expressed with a Level of Assurance (LoA see 3). If it can be considered as a first process to simplify the expression of trust during the authentication step, it does not handle all the aspects of the authentication mechanism and especially it fails to integrate continuous authentication systems. In this paper, we propose a model based on the Dempster Shafer theory to merge continuous authentication system with more traditional static authentication scheme and to assign a continuous trust level to the current LoA. In addition, this method is proved to be compliant with the LoA frameworks.\",\n \"Title: In What Way Is It Possible to Impersonate You Bypassing Fingerprint Sensors?\\nYear: 2016\\n\\nAbstract: In this study, we want to determine if an attacker with different a priori information on the sensor parameters during the enrollment process can impersonate an individual. We have tested different apriorism such as the fingerprint class, sensor type, image resolution and the number of minutiae. Two different matching algorithms have been used on specific designed databases generated with SFinge. Two attack scenarii have been tested. Obtained results show that the knowledge of fingerprint class and resolution influence the attack success.\",\n \"Title: Synchronous One Time Biometrics with Pattern Based Authentication\\nYear: 2016\\n\\nAbstract: One time passwords are commonly used for authentication purposes in electronic transactions. Nevertheless, providing such a one time password is not really a strong authentication proof %as it can be given by an impostor. because the token generating the passwords can be given by an impostor. In order to cope with this problem, biometric recognition is more and more employed. Even if biometric data are strongly linked with the user, their revocability nor diversity is possible, without an adapted post-processing. Biometric template protection schemes, including the BioHashing algorithm, are used to manage the underlying privacy and security issues. These schemes are used for the protection of several biometric modalities, but are not necessary adapted for all of them. In this paper, we propose a new protocol combining protected biometric data and a classical synchronous one time password to enhance the security of user authentication while preserving usability and privacy. Behavioral biometrics is used to provide a fast and a usable solution for users. We show through experimental results the efficiency of the proposed method.\",\n \"Title: Privacy Preserving Transparent Mobile Authentication\\nYear: 2017\\n\\nAbstract: Transparent authentication on mobile phones suffers from privacy issues especially when biometric information is involved. In this paper, we propose a solution to address those two issues using the Biohashing algorithm on behavioral information extracted from a mobile phone. The authentication scenario is tested on a dataset composed of 100 users and shows promising results with a 10% EER in the worst case scenario (i.e when protection key is compromised) and a 1% EER in the best case one. In addition, privacy concerns are discussed and experimentally evaluated both in a quantitative and qualitative ways. This opens new perspectives concerning online authentication using smartphone sensing abilities.\",\n \"Title: Fingerprint Class Recognition For Securing Emv Transaction\\nYear: 2017\\n\\nAbstract: Fingerprint analysis is a very important issue in biometry. The minutiae representation of a fingerprint is the most used modality to identify people or authorize access when using a biometric system. In this paper, we propose some features based on triangle parameters from the Delaunay triangulation of minutiae. We show the benefit of these features to recognize the type of a fingerprint without any access to the associated fingerprint image.\",\n \"Title: Memory carving can finally unveil your embedded personal data\\nYear: 2017\\n\\nAbstract: Smart cards are involved in most of activities, and they gather and record plenty of personal data. A manual interpretation of these raw data is difficult without specifications. This task becomes really tedious applied to plenty of devices. The paper introduces the first method to automatically retrieve textual information from memory dumps of smart cards. Given the data structure and encoding are assumed to be unknown, the method is based on text statistics and characteristics of smart cards to discard false positives. The experiments performed on more than 350 memory dumps revealed that the method can automatically retrieve more than 99% of textual information available in a dump, while keeping the false positive rate as low as 5.5%.\",\n \"Title: Contrasting False Identities in Social Networks by Trust Chains and Biometric Reinforcement\\nYear: 2017\\n\\nAbstract: Fake identities and identity theft are issues whose relevance is increasing in the social network domain. This paper deals with this problem by proposing an innovative approach which combines a collaborative mechanism implementing a trust graph with keystroke-dynamic-recognition techniques to trust identities. The trust of each node is computed on the basis of neighborhood recognition and behavioral biometric support. The model leverages the word of mouth propagation and a settable degree of redundancy to obtain robustness. Experimental results show the benefit of the proposed solution even if attack nodes are present in the social network.\",\n \"Title: Retrieving dates in smart card dumps is as hard as finding a needle in a haystack\\nYear: 2017\\n\\nAbstract: This paper introduces a method to automatically retrieve dates from smart card memory dumps when the card specifications are unknown. It exploits specificities of smart cards, using a multi-dump analysis augmented with contextual information. The experiments performed on more than 180 real smart cards show that our method is highly successful in removing false positives.\",\n \"Title: Vision-Based System for Human Detection and Tracking in Indoor Environment\\nYear: 2010\\n\\nAbstract: In this paper, we propose a vision-based system for human detection and tracking in indoor environment us- ing a static camera. The proposed method is based on ob- ject recognition in still images combined with methods us- ing temporal information from the video. Doing that, we improve the performance of the overall system and reduce the task complexity. We first use background subtraction to limit the search space of the classifier. The segmentation is realized by modeling each background pixel by a single Gaussian model. As each connected component detected by the background subtraction potentially corresponds to one person, each blob is independently tracked. The tracking process is based on the analysis of connected components position and interest points tracking. In order to know the nature of various objects that could be present in the scene, we use multiple cascades of boosted classifiers based on Haar-like filters. We also present in this article a wide eval- uation of this system based on a large set of videos.\",\n \"Title: Comparative Study of Digital Fingerprint Quality Assessment Metrics\\nYear: 2018\\n\\nAbstract: The quality assessment of biometric data proceeds as a toll to decide whether a biometric sample may be used to generate the user's reference template. Many studies showed its significant impact on the subsequent performance of the biometric system. Since many metrics are proposed for this purpose by researchers or standardization institutions, their relevance should be studied in particular to evaluate their relative usefulness. This paper provides a comparative study of fingerprint quality assessment (FQA) metrics. We consider the enrollment selection validation approach to perform an objective comparison of them. We show the efficiency of 7 well known FQA metrics on 9 datasets. Results show a dependency of these metrics to the dataset (i.e. the fingerprint sensor) and for matchers a similar behavior.\",\n \"Title: Rhu Keystroke Touchscreen Benchmark\\nYear: 2018\\n\\nAbstract: Biometric systems are currently widely used in many applications to control and verify individual's identity. Keystroke dynamics modality has been shown as a promising solution that would be used in many applications such as e-payment and banking applications. However, such systems suffer from several performance limitations (such as cross-devices problem) that prevent their widespread of use in real applications. The objective of this paper is to provide researchers and developers with a public touchscreen-based benchmark collected using a mobile phone and a tablet (both portrait and landscape orientation each). Such a benchmark can be used to assess keystroke-based matching algorithms. Furthermore, It is mainly developed to measure the robustness of keystroke matching algorithms vis-'a-vis cross-devices and orientation variations. An online visualizer for the database is also given to researchers allowing them to visualize the acquired keystroke signals.\",\n \"Title: Enhancing the Security of Transformation Based Biometric Template Protection Schemes\\nYear: 2018\\n\\nAbstract: Template protection is a crucial issue in biometrics. Many algorithms have been proposed in the literature among secure computing approaches, crypto-biometric algorithm and feature transformation schemes. The BioHashing algorithm belongs to this last category and has very interesting properties. Among them, we can cite its genericity since it could be applied on any biometric modality, the possible cancelability of the generated BioCode and its efficiency when the secret is not stolen by an impostor. Its main drawback is its weakness face to a combined attack (zero effort with the stolen secret scenario). In this paper, we propose a transformation-based biometric template protection scheme as an improvement of the BioHashing algorithm where the projection matrix is generated by combining the secret and the biometric data. Experimental results on two biometric modalities, namely digital fingerprint and finger knuckle print images, show the benefits of the proposed method face to attacks while keeping a good efficiency.\",\n \"Title: User Dependent Template Update for Keystroke Dynamics Recognition\\nYear: 2018\\n\\nAbstract: Regarding the fact that individuals have different interactions with biometric authentication systems, several techniques have been developed in the literature to model different users categories. Doddington Zoo is a concept of categorizing users behaviors into animal groups to reflect their characteristics with respect to biometric systems. This concept was developed for different biometric modalities including keystroke dynamics. The present study extends this biometric classification, by proposing a novel adaptive strategy based on the Doddinghton Zoo, for the recognition of the user's keystroke dynamics. The obtained results demonstrate competitive performances on significant keystroke dynamics datasets.\",\n \"Title: Analysis of Keystroke Dynamics for the Generation of Synthetic Datasets\\nYear: 2018\\n\\nAbstract: Biometrics is an emerging technology more and more present in our daily life. However, building biometric systems requires a large amount of data that may be difficult to collect. Collecting such sensitive data is also very time consuming and constrained, s.a. GDPR legislation. In the case of keystroke dynamics, existing databases have less than 200 users. For these reasons, we aim at generating a keystroke dynamics synthetic dataset. This paper presents the generation of keystroke data from known users as a first step towards the generation of synthetic datasets, and could also be used to impersonate users' identity.\",\n \"Title: A COMPARATIVE STUDY OF SUPERVISED EVALUATION CRITERIA FOR IMAGE SEGMENTATION\\nYear: 2004\\n\\nAbstract: This paper presents a comparative study of five supervised evaluation criteria for image segmentation. The different cri- teria have been tested on a selection of hundred images ex- tracted from the c Corel database for which manual segmen- tation results provided by experts are available. Nine seg- mentation algorithms have been considered, most of which are based on threshold selection. In order to compare the behavior of the different criteria towards over- and underseg- mentation, three thresholds have been taken into account, for each selected image, to simulate the various situations. Ex- perimental results permit to reveal the advantages and limi- tations of the studied criteria.\",\n \"Title: EvaBio a New Modular Platform to Evaluate Biometric System\\nYear: 2015\\n\\nAbstract: when someone wants to make a payment with a smartcard, the user has to enter a pin code to be identified. Only biometrics is able to authenticate a user; yet biometric information is sensitive. To ensure the security and privacy of biometric data, OCC (On-Card-Comparison) has been proposed. This approach consists in storing biometric data in a secure zone on a smartcard and computing the verification decision in a Secure Element (SE). The purpose of this paper is to propose an evaluation platform for testing biometric systems such as the analysis of performance and security on biometric OCC. Based on four examples, we illustrate its different uses in an operationnal context. The first example focus on the \\\"Quality module\\\" which allows to choose the enrollment by considering the fingerprint quality with one proposed metric. The second one addresses the minutiae reduction of the fingerprint template when the number of minutiae is higher than expected by the OCC. The third is based on sensors acquisition module to create databases and made attacks on sensors. The last one is the evaluation module, it permit to visualize results after an evaluation.\",\n \"Title: Double serial adaptation mechanism for keystroke dynamics authentication based on a single password.\\nYear: 2019\\n\\nAbstract: Cyber-attacks have spread all over the world to steal information such as trade secrets, intellectual property and banking data. Facing the danger of the insecurity of saved data (personal, professional, official, etc.), keystroke dynamics was proposed as an interesting, non-intrusive, inexpensive, permanent and weakly constrained solution for users. Based on the typing rhythm of users, it improves logical access security. Nevertheless, it was demonstrated that such an authentication mechanism would need a larger number of samples to enroll the typing characteristics of users. Moreover, these registered characteristics generally undergo aging effects after a time span. Different solutions have been suggested to remedy these variability problems, including template adaptation. In this paper, we propose a double serial adaptation strategy that considers a single-capture-based enrollment process. When using the authentication system, the template of users and the decision/adaptation thresholds are updated. Experimental results on three public keystroke dynamics datasets show the benefits of the proposed method.\",\n \"Title: Analysis of Doddington zoo classification for user dependent template update: Application to keystroke dynamics recognition\\nYear: 2019\\n\\nAbstract: Biometric authentication systems are increasingly considered in different access control applications. Regarding that users have completely different interactions with these authentication systems, several techniques have been developed in the literature to model distinctive users categories. Doddington zoo is a biometric menagerie that defines and labels user groups with animal species to reflect their behavior with the biometric systems. This menagerie was developed for different biometric modalities including keystroke dynamics. The present study proposes a user dependent adaptive strategy based on the Doddington zoo, for the recognition of the user’s keystroke dynamics. The novelty of the proposed approach lies in applying an adaptive strategy specific to the characteristics of each user of the Doddington zoo menagerie aiming to solve the intra-class variation problems. The obtained results demonstrate competitive performances on significant keystroke dynamics datasets WEBGREYC and CMU.\",\n \"Title: Statistical modeling of keystroke dynamics samples for the generation of synthetic datasets\\nYear: 2019\\n\\nAbstract: Biometrics is an emerging technology more and more present in our daily life. However, building biometric systems requires a large amount of data that may be difficult to collect. Collecting such sensitive data is also very time consuming and constrained, s.a. GDPR legislation in Europe. In the case of keystroke dynamics, most existing databases have less than 200 users. For these reasons, it is crucial for this biometric modality to be able to generate a significant and realistic synthetic dataset of keystroke dynamics samples. We propose in this paper an original approach for the generation of synthetic keystroke data given samples from known users as a first step towards the generation of synthetic datasets. Experimental results show the capability of the proposed statistical model to generate realistic samples from existing datasets in the literature.\",\n \"Title: Privacy Aura for Transparent Authentication on Multiple Smart Devices\\nYear: 2021\\n\\nAbstract: Nowadays, users carry multiple connected devices such as a smartphone, laptop, connected watch.... Security constraints limit user's usability especially when using all of them intensively during the day (social media, work). In this paper, we propose the privacy Aura concept corresponding to the circle of trust in the neighborhood of each smart device to facilitate user authentication. Many data (phone calls habits, biometrics, localization) can be collected to realize a transparent and privacy compliant authentication on each device. The confidence on user authentication on each device can be transferred to another one if it is located in the same Aura. This is the main contribution of the paper. We show through illustrations the benefit of the proposed solution.\",\n \"Title: Classifying Biometric Systems Users among the Doddington Zoo: Application to Keystroke Dynamics\\nYear: 2021\\n\\nAbstract: Doddington zoo defines four categories of users when using a biometric system related to their difficulty to be recognized or attacked. In this paper, we propose an original work consisting in predicting for any biometric modality the associated animal in the Doddington menagerie related to a user given few captured biometric samples. Such a prediction could be useful for many applications, as for example, to adapt the behavior of biometric systems to each user. In this work, we apply this methodology to keystroke dynamics as it is an interesting behavioral biometric modality for user authentication. It consists in analyzing the way of typing of a user in order to recognize him/her. We use a significant keystroke dynamics dataset and we demonstrate through experimental results the benefit of the proposed approach.\",\n \"Title: Face Transparent User Authentication Respecting Privacy\\nYear: 2021\\n\\nAbstract: User authentication is an important issue to guarantee the security of information systems. For users, this task is in general inconvenient and boring. We propose in this paper a new transparent user authentication based on face recognition. During the use of a laptop or a smartphone, face features are computed, protected and sent to a trust party server that measures at any time the confidence of...\",\n \"Title: Keystroke Dynamics Classification Based On LSTM and BLSTM Models\\nYear: 2021\\n\\nAbstract: By adopting keystroke dynamics, authentication applications can integrate advanced identity proofing technology for detecting fraud and prevent unauthorized access. However, understanding a user's keystroke dynamics behavior in real applications is a challenging task regarding that this behavior is notably changing over time. To mitigate this problem, we apply, in this paper, the long short-term m...\",\n \"Title: How Artificial Intelligence can be used for Behavioral Identification?\\nYear: 2021\\n\\nAbstract: Nowadays, users interact with computer systems. Behavioral biometrics consists of analyzing user's interactions for identification and verification applications. This approach could be very useful for enhancing security and improving user experience and many privacy concerns are also related. In this paper, we address the problem of user identification considering their behaviors. How efficient ar...\",\n \"Title: Impact Of Environmental Conditions On Fingerprint Systems Performance\\nYear: 2021\\n\\nAbstract: Biometrics testing has for objective to determine the performance of a biometric system in order to guarantee security and user experience requirements. Providing trust in biometric systems is a key for many manufacturers. The performance is usually measured through the computation of matching scores between legitimate and impostor samples from a given database. Different bias in particular those ...\"\n]","isConversational":false},"profile_size":{"kind":"number","value":103,"string":"103"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13576,"string":"13,576"}}},{"rowIdx":13577,"cells":{"question_id":{"kind":"string","value":"Rong Jin"},"question":{"kind":"string","value":"Generate an abstract for the title \"Wearable Autonomous Microsystem With Electrochemical Gas Sensor Array For Real-Time Health And Safety Monitoring\" using the following items: Airborne pollution, explosive gases, wearable sensor, electrochemical sensors, real-time monitoring."},"answer":{"kind":"string","value":"Airborne pollution and explosive gases threaten human health and occupational safety, therefore generating high demand for a wearable autonomous multi-analyte gas sensor system for real-time environmental monitoring. This paper presents a system level solution through synergistic integration of sensors, electronics, and data analysis algorithms. Electrochemical sensors featuring ionic liquids were chosen to provide low-power room-temperature operation, rapid response, high sensitivity, good selectivity, and a long operating life with low maintenance. The system utilizes a multi-mode electrochemical instrumentation circuit that combines all signal condition functions within a single microelectronics chip to minimize system cost, size and power consumption. Embedded sensor array signal processing algorithms enable gas classification and concentration estimation within a real-world mixture of analytes. System design and integration methodologies are described, and preliminary results are shown for a first generation SO2 sensor and a thumb-drive sized prototype system."},"correct_answer":{"kind":"string","value":"Airborne pollution and explosive gases threaten human health and occupational safety, therefore generating high demand for a wearable autonomous multi-analyte gas sensor system for real-time environmental monitoring. This paper presents a system level solution through synergistic integration of sensors, electronics, and data analysis algorithms. Electrochemical sensors featuring ionic liquids were chosen to provide low-power room-temperature operation, rapid response, high sensitivity, good selectivity, and a long operating life with low maintenance. The system utilizes a multi-mode electrochemical instrumentation circuit that combines all signal condition functions within a single microelectronics chip to minimize system cost, size and power consumption. Embedded sensor array signal processing algorithms enable gas classification and concentration estimation within a real-world mixture of analytes. System design and integration methodologies are described, and preliminary results are shown for a first generation SO2 sensor and a thumb-drive sized prototype system."},"documents":{"kind":"list like","value":["Title: Integrating User Feedback Log into Relevance Feedback by Coupled SVM for Content-Based Image Retrieval\nYear: 2005\n\nAbstract: Relevance feedback has been shown as an important tool to boost the retrieval performance in content-based image retrieval. In the past decade, various algorithms have been proposed to formulate relevance feedback in contentbased image retrieval. Traditional relevance feedback techniques mainly carry out the learning tasks by focusing lowlevel visual features of image content with little consideration on log information of user feedback. However, from a long-term learning perspective, the user feedback log is one of the most important resources to bridge the semantic gap problem in image retrieval. In this paper we propose a novel technique to integrate the log information of user feedback into relevance feedback for image retrieval. Our algorithms construction is based on a coupled support vector machine which learns consistently with the two types of information: the low-level image content and the user feedback log. We present a mathematical formulation of the problem and develop a practical algorithm to solve the problem effectively. Experimental results show that our proposed scheme is effective and promising.","Title: An Information Geometry Approach for Distance Metric Learning\nYear: 2009\n\nAbstract: In this paper, we propose a framework for metric learning based on information geome- try. The key idea is to construct two kernel matrices for the given training data: one is based on the distance metric and the other is based on the assigned class labels. Inspired by the idea of information geometry, we re- late these two kernel matrices to two Gaus- sian distributions, and the difierence between the two kernel matrices is then computed by the Kullback-Leibler (KL) divergence be- tween the two Gaussian distributions. The optimal distance metric is then found by min- imizing the divergence between the two dis- tributions. We present two metric learning algorithms, one for linear distance metric and one for nonlinear distance using a kernel func- tion. Unlike many existing algorithms for metric learning that require solving a non- trivial optimization problem and are compu- tationally expensive when the data dimen- sion is high, the proposed algorithms have a closed-form solution and are computationally more e-cient. Extensive experiments with data classiflcation and face recognition show that the proposed algorithms are compara- ble to or better than the state-of-the-art al- gorithms for metric learning.","Title: CUR Algorithm with Incomplete Matrix Observation.\nYear: 2014\n\nAbstract: CUR matrix decomposition is a randomized algorithm that can efficiently compute the low rank approximation for a given rectangle matrix. One limitation with the existing CUR algorithms is that they require an access to the full matrix A for computing U. In this work, we aim to alleviate this limitation. In particular, we assume that besides having an access to randomly sampled d rows and d columns from A, we only observe a subset of randomly sampled entries from A. Our goal is to develop a low rank approximation algorithm, similar to CUR, based on (i) randomly sampled rows and columns from A, and (ii) randomly sampled entries from A. The proposed algorithm is able to perfectly recover the target matrix A with only O(rn log n) number of observed entries. In addition, instead of having to solve an optimization problem involved trace norm regularization, the proposed algorithm only needs to solve a standard regression problem. Finally, unlike most matrix completion theories that hold only when the target matrix is of low rank, we show a strong guarantee for the proposed algorithm even when the target matrix is not low rank.","Title: Calibration of staggered Y-shaped array by 120 degree rotation\nYear: 2012\n\nAbstract: The calibration of aperture synthesis radiometers (ASRs) by the standard method of noise injection would seriously increase hardware requirements and system complexity. To reduce the network of noise injection, redundant space calibration (RSC) method has been considered. However the number of error phases is larger than the number of independent equations. In this paper, an external calibration method with 120-degree antenna array rotation for geostationary ASR with staggered Y-shaped array is proposed. More redundant baselines are produced by rotation. The equations by rotation and by RSC are combined for solving the phase errors. Basic procedure on this method and preliminary simulation are introduced. As another advantage, if twice 120-degree rotation is taken, one of the antenna arms can be removed from staggered Y-shaped arrays with the missing baselines supplied by the rotations. In this way, the number of antenna elements can be reduced by 1/3.","Title: Learning to trade off between exploration and exploitation in multiclass bandit prediction\nYear: 2011\n\nAbstract: We study multi-class bandit prediction, an online learning problem where the learner only receives a partial feedback in each trial indicating whether the predicted class label is correct. The exploration vs. exploitation tradeoff strategy is a well-known technique for online learning with incomplete feedback (i.e., bandit setup). Banditron [8], a multi-class online learning algorithm for bandit setting, maximizes the run-time gain by balancing between exploration and exploitation with a fixed tradeoff parameter. The performance of Banditron can be quite sensitive to the choice of the tradeoff parameter and therefore effective algorithms to automatically tune this parameter is desirable. In this paper, we propose three learning strategies to automatically adjust the tradeoff parameter for Banditron. Our extensive empirical study with multiple real-world data sets verifies the efficacy of the proposed approach in learning the exploration vs. exploitation tradeoff parameter.","Title: An efficient algorithm for local distance metric learning\nYear: 2006\n\nAbstract: Learning application-specific distance metrics from labeled data is critical for both statistical classification and information retrieval. Most of the earlier work in this area has focused on finding metrics that simultaneously optimize compactness and separability in a global sense. Specifically, such distance metrics attempt to keep all of the data points in each class close together while ensuring that data points from different classes are separated. However, particularly when classes exhibit multimodal data distributions, these goals conflict and thus cannot be simultaneously satisfied. This paper proposes a Local Distance Metric (LDM) that aims to optimize local compactness and local separability. We present an efficient algorithm that employs eigenvector analysis, and bound optimization to learn the LDM from training data in a probabilistic framework. We demonstrate that LDM achieves significant improvements in both classification and retrieval accuracy compared to global distance learning and kernel-based KNN.","Title: A smoothed boosting algorithm using probabilistic output codes\nYear: 2005\n\nAbstract: AdaBoost.OC has shown to be an effective method in boosting \"weak\" binary classifiers for multi-class learning. It employs the Error Correcting Output Code (ECOC) method to convert a multi-class learning problem into a set of binary classification problems, and applies the AdaBoost algorithm to solve them efficiently. In this paper, we propose a new boosting algorithm that improves the AdaBoost.OC algorithm in two aspects: 1) It introduces a smoothing mechanism into the boosting algorithm to alleviate the potential overfitting problem with the AdaBoost algorithm, and 2) It introduces a probabilistic coding scheme to generate binary codes for multiple classes such that training errors can be efficiently reduced. Empirical studies with seven UCI datasets have indicated that the proposed boosting algorithm is more robust and effective than the AdaBoost.OC algorithm for multi-class learning.","Title: Excess Risk Bounds for Exponentially Concave Losses.\nYear: 2014\n\nAbstract: The overarching goal of this paper is to derive excess risk bounds for learning from exp-concave loss functions in passive and sequential learning settings. Exp-concave loss functions encompass several fundamental problems in machine learning such as squared loss in linear regression, logistic loss in classification, and negative logarithm loss in portfolio management. In batch setting, we obtain sharp bounds on the performance of empirical risk minimization performed in a linear hypothesis space and with respect to the exp-concave loss functions. We also extend the results to the online setting where the learner receives the training examples in a sequential manner. We propose an online learning algorithm that is a properly modified version of online Newton method to obtain sharp risk bounds. Under an additional mild assumption on the loss function, we show that in both settings we are able to achieve an excess risk bound of $O(d\\log n/n)$ that holds with a high probability.","Title: Improved Bounds for the Nyström Method With Application to Kernel Classification\nYear: 2013\n\nAbstract: We develop two approaches for analyzing the approximation error bound for the Nyström method that approximates a positive semidefinite (PSD) matrix by sampling a small set of columns, one based on a concentration inequality for integral operators, and one based on random matrix theory. We show that the approximation error, measured in the spectral norm, can be improved from $O(N/\\\\sqrt {m})$ to $O(N/m^{1 - \\\\rho })$ in the case of large eigengap, where $N$ is the total number of data points, $m$ is the number of sampled data points, and $\\\\rho \\\\in (0, 1/2)$ is a positive constant that characterizes the eigengap. When the eigenvalues of the kernel matrix follow a $p$-power law, our analysis based on random matrix theory further improves the bound to $O(N/m^{p - 1})$ under an incoherence assumption. We present a kernel classification approach based on the Nyström method and derive its generalization performance using the improved bound. We show that when the eigenvalues of the kernel matrix follow a $p$-power law, we can reduce the number of support vectors to $N^{2p/(p^{2} - 1)}$, which is sublinear in $N$ when $p 1+\\\\sqrt {2}$, without seriously sacrificing its generalization performance.","Title: Language model for IR using collection information\nYear: 2002\n\nAbstract: Information retrieval using meta data can be traced back to the early age of IR where documents are represented by the controlled vocabulary. In this paper, we explore the usage of meta-data information under the framework of language model. We present a new language model that is able to take advantage of the category information for documents to improve the retrieval accuracy. We compare the new language model with the traditional language model over the TREC4 dataset where the collection information for documents is obtained using the k-means clustering method. The new language model outperforms the traditional language model, which verifies our statement.","Title: Semi-supervised multi-label learning by constrained non-negative matrix factorization\nYear: 2006\n\nAbstract: We present a novel framework for multi-label learning that explicitly addresses the challenge arising from the large number of classes and a small size of training data. The key assumption behind this work is that two examples tend to have large overlap in their assigned class memberships if they share high similarity in their input patterns. We capitalize this assumption by first computing two sets of similarities, one based on the input patterns of examples, and the other based on the class memberships of the examples. We then search for the optimal assignment of class memberships to the unlabeled data that minimizes the difference between these two sets of similarities. The optimization problem is formulated as a constrained Non-negative Matrix Factorization (NMF) problem, and an algorithm is presented to efficiently find the solution. Compared to the existing approaches for multi-label learning, the proposed approach is advantageous in that it is able to explore both the unlabeled data and the correlation among different classes simultaneously. Experiments with text categorization show that our approach performs significantly better than several state-of-the-art classification techniques when the number of classes is large and the size of training data is small.","Title: A graph-based framework for relation propagation and its application to multi-label learning\nYear: 2006\n\nAbstract: Label propagation exploits the structure of the unlabeled documents by propagating the label information of the training documents to the unlabeled documents. The limitation with the existing label propagation approaches is that they can only deal with a single type of objects. We propose a framework, named \"relation propagation\", that allows for information propagated among multiple types of objects. Empirical studies with multi-label text categorization showed that the proposed algorithm is more effective than several semi-supervised learning algorithms in that it is capable of exploring the correlation among different categories and the structure of unlabeled documents simultaneously.","Title: Fast Bounded Online Gradient Descent Algorithms for Scalable Kernel-Based Online Learning\nYear: 2012\n\nAbstract: Kernel-based online learning has often shown state-of-the-art performance for many online learning tasks. It, however, suffers from a major shortcoming, that is, the unbounded number of support vectors, making it non-scalable and unsuitable for applications with large-scale datasets. In this work, we study the problem of bounded kernel-based online learning that aims to constrain the number of support vectors by a predefined budget. Although several algorithms have been proposed in literature, they are neither computationally efficient due to their intensive budget maintenance strategy nor effective due to the use of simple Perceptron algorithm. To overcome these limitations, we propose a framework for bounded kernel-based online learning based on an online gradient descent approach. We propose two efficient algorithms of bounded online gradient descent (BOGD) for scalable kernel-based online learning: (i) BOGD by maintaining support vectors using uniform sampling, and (ii) BOGD++ by maintaining support vectors using non-uniform sampling. We present theoretical analysis of regret bound for both algorithms, and found promising empirical performance in terms of both efficacy and efficiency by comparing them to several well-known algorithms for bounded kernel-based online learning on large-scale datasets.","Title: A Bayesian framework for knowledge driven regression model in micro-array data analysis.\nYear: 2008\n\nAbstract: This paper addresses the sparse data problem in the linear regression model, namely the number of variables is significantly larger than the number of the data points for regression. We assume that in addition to the measured data points, the prior knowledge about the input variables may be provided in the form of pair wise similarity. We presented a full Bayesian framework to effectively exploit the similarity information of the input variables for linear regression. Empirical studies with gene expression data show that the regression errors can be reduced significantly by incorporating the similarity information derived from gene ontology.","Title: Distance metric learning from uncertain side information for automated photo tagging\nYear: 2011\n\nAbstract: Automated photo tagging is an important technique for many intelligent multimedia information systems, for example, smart photo management system and intelligent digital media library. To attack the challenge, several machine learning techniques have been developed and applied for automated photo tagging. For example, supervised learning techniques have been applied to automated photo tagging by training statistical classifiers from a collection of manually labeled examples. Although the existing approaches work well for small testbeds with relatively small number of annotation words, due to the long-standing challenge of object recognition, they often perform poorly in large-scale problems. Another limitation of the existing approaches is that they require a set of high-quality labeled data, which is not only expensive to collect but also time consuming. In this article, we investigate a social image based annotation scheme by exploiting implicit side information that is available for a large number of social photos from the social web sites. The key challenge of our intelligent annotation scheme is how to learn an effective distance metric based on implicit side information (visual or textual) of social photos. To this end, we present a novel “Probabilistic Distance Metric Learning” (PDML) framework, which can learn optimized metrics by effectively exploiting the implicit side information vastly available on the social web. We apply the proposed technique to photo annotation tasks based on a large social image testbed with over 1 million tagged photos crawled from a social photo sharing portal. Encouraging results show that the proposed technique is effective and promising for social photo based annotation tasks.","Title: Semi-supervised Clustering by Input Pattern Assisted Pairwise Similarity Matrix Completion.\nYear: 2013\n\nAbstract: Many semi-supervised clustering algorithms have been proposed to improve the clustering accuracy by effectively exploring the available side information that is usually in the form of pairwise constraints. Despite the progress, there are two main shortcomings of the existing semi-supervised clustering algorithms. First, they have to deal with non-convex optimization problems, leading to clustering results that are sensitive to the initialization. Second, none of these algorithms is equipped with theoretical guarantee regarding the clustering performance. We address these limitations by developing a framework for semi-supervised clustering based on \\it input pattern assisted matrix completion. The key idea is to cast clustering into a matrix completion problem, and solve it efficiently by exploiting the correlation between input patterns and cluster assignments. Our analysis shows that under appropriate conditions, only O(\\log n) pairwise constraints are needed to accurately recover the true cluster partition. We verify the effectiveness of the proposed algorithm by comparing it to the state-of-the-art semi-supervised clustering algorithms on several benchmark datasets.","Title: Active Query Selection For Semi-Supervised Clustering\nYear: 2008\n\nAbstract: Semi-supervised clustering allows a user to specify available prior knowledge about the data to improve the clustering performance. A common way to express this information is in the form of pair-wise constraints. A number of studies have shown that, in general, these constraints improve the resulting data partition. However the choice of constraints is critical since improperly chosen constraints might actually degrade the clustering performance. We focus on constraint (also known as query) selection for improving the performance of semi-supervised clustering algorithms. We present an active query selection mechanism, where the queries are selected using a min-max criterion. Experimental results on a variety of datasets, using MPCK-means as the underlying semi-clustering algorithm, demonstrate the superior performance of the proposed query selection procedure.","Title: A language modeling framework for resource selection and results merging\nYear: 2002\n\nAbstract: Statistical language models have been proposed recently for several information retrieval tasks, including the resource selection task in distributed information retrieval. This paper extends the language modeling approach to integrate resource selection, ad-hoc searching, and merging of results from different text databases into a single probabilistic retrieval model. This new approach is designed primarily for Intranet environments, where it is reasonable to assume that resource providers are relatively homogeneous and can adopt the same kind of search engine. Experiments demonstrate that this new, integrated approach is at least as effective as the prior state-of-the-art in distributed IR.","Title: SWITCH: A Novel Approach to Ensemble Learning for Heterogeneous Data\nYear: 2004\n\nAbstract: The standard framework of machine learning problems assumes that the available data is independent and identically distributed (i.i.d.). However, in some applications such as image classification, the training data are often collected from multiple sources and heterogeneous. Ensemble learning is a proven effective approach to heterogeneous data, which uses multiple classification models to capture the diverse aspects of heterogeneous data. If an ensemble can learn the relationship between different portions of data and their corresponding models, the ensemble can selectively apply models to unseen data according to the learned relationship. We propose a novel approach to enable the learning of the relationships between data and models by creating a set of I switches' that can route a testing instance to appropriate classification models in an ensemble. Our empirical study on both real-world data and benchmark data shows that the proposed approach to ensemble learning can achieve significant performance improvement for heterogeneous data.","Title: Learning Bregman Distance Functions for Semi-Supervised Clustering\nYear: 2012\n\nAbstract: Learning distance functions with side information plays a key role in many data mining applications. Conventional distance metric learning approaches often assume that the target distance function is represented in some form of Mahalanobis distance. These approaches usually work well when data are in low dimensionality, but often become computationally expensive or even infeasible when handling high-dimensional data. In this paper, we propose a novel scheme of learning nonlinear distance functions with side information. It aims to learn a Bregman distance function using a nonparametric approach that is similar to Support Vector Machines. We emphasize that the proposed scheme is more general than the conventional approach for distance metric learning, and is able to handle high-dimensional data efficiently. We verify the efficacy of the proposed distance learning method with extensive experiments on semi-supervised clustering. The comparison with state-of-the-art approaches for learning distance functions with side information reveals clear advantages of the proposed technique.","Title: Error model and calibration of synthetic aperture interferometric radiometer based on visibility function\nYear: 2012\n\nAbstract: Traditionally, the system errors of synthetic aperture interferometric radiometers (SAIRs) are analyzed and classified by the causes they are generated, which is easy for understanding but complicated for calibration. This paper presents a new error model based on the visibility function to analyze the system errors of SAIRs. In this model the system errors are classified by their effects on the ideal visibility function, which is simple for calibration. Firstly, the error model based on the visibility function is proposed, in which the errors according to their effects on the ideal visibility function are grouped in: 1) the orientation-independent multiplicative error, 2) the orientation-dependent multiplicative error, and 3) the additive error. Then, the two overall calibration methods respectively using external reference source and external reference scene are proposed. Finally, the preliminary calibration experiment results on the 8mm-band HUST-SAIR prototype show the validity of the error model and the overall calibration methods. The study in this paper provides a means to reduce the complexity of calibration routine of SAIRs and to improve the imaging performance of SAIRs.","Title: Robust feature induction for support vector machines\nYear: 2004\n\nAbstract: The goal of feature induction is to automatically create nonlinear combinations of existing features as additional input features to improve classification accuracy. Typically, nonlinear features are introduced into a support vector machine (SVM) through a nonlinear kernel function. One disadvantage of such an approach is that the feature space induced by a kernel function is usually of high dimension and therefore will substantially increase the chance of over-fitting the training data. Another disadvantage is that nonlinear features are induced implicitly and therefore are difficult for people to understand which induced features are critical to the classification performance. In this paper, we propose a boosting-style algorithm that can explicitly induces important nonlinear features for SVMs. We present empirical studies with discussion to show that this approach is effective in improving classification accuracy for SVMs. The comparison with an SVM model using nonlinear kernels also indicates that this approach is effective and robust, particularly when the number of training data is small.","Title: Online learning by ellipsoid method\nYear: 2009\n\nAbstract: In this work, we extend the ellipsoid method, which was originally designed for convex optimization, for online learning. The key idea is to approximate by an ellipsoid the classification hypotheses that are consistent with all the training examples received so far. This is in contrast to most online learning algorithms where only a single classifier is maintained at each iteration. Efficient algorithms are presented for updating both the centroid and the positive definite matrix of ellipsoid given a misclassified example. In addition to the classical ellipsoid method, an improved version for online learning is also presented. Mistake bounds for both ellipsoid methods are derived. Evaluation with the USPS dataset and three UCI data-sets shows encouraging results when comparing the proposed online learning algorithm to two state-of-the-art online learners.","Title: A Kernel Approach to Multi-Task Learning with Task-Specific Kernels\nYear: 2012\n\nAbstract: Several kernel-based methods for multi-task learning have been proposed,which leverage relations among tasks as regularization to enhance the overall learning accuracies.These methods assume that the tasks share the same kernel,which could limit their applications because in practice different tasks may need different kernels.The main challenge of introducing multiple kernels into multiple tasks is that models from different reproducing kernel Hilbert spaces(RKHSs) are not comparable,making it difficult to exploit relations among tasks.This paper addresses the challenge by formalizing the problem in the square integrable space(SIS).Specially,it proposes a kernel-based method which makes use of a regularization term defined in SIS to represent task relations.We prove a new representer theorem for the proposed approach in SIS.We further derive a practical method for solving the learning problem and conduct consistency analysis of the method.We discuss the relationship between our method and an existing method.We also give an SVM(support vector machine)based implementation of our method for multi-label classification.Experiments on an artificial example and two real-world datasets show that the proposed method performs better than the existing method.","Title: Collaborative filtering with decoupled models for preferences and ratings\nYear: 2003\n\nAbstract: In this paper, we describe a new model for collaborative filtering. The motivation of this work comes from the fact that two users with very similar preferences on items may have very different rating schemes. For example, one user may tend to assign a higher rating to all items than another user. Unlike previous models of collaborative filtering, which determine the similarity between two users only based on their rating performance, our model treats the user's preferences on items separately from the user's rating scheme. More specifically, for each user, we build two separate models: a preference model capturing which items are favored by the user and a rating model capturing how the user would rate an item given the preference information. The similarity of two users is computed based on the underlying preference model, instead of the surface ratings. We compare the new model with several representative previous approaches on two data sets. Experiment results show that the new model outperforms all the previous approaches that are tested consistently on both data sets.","Title: Graffiti-ID: matching and retrieval of graffiti images\nYear: 2009\n\nAbstract: Graffiti are abundant in most urban neighborhoods and are considered a nuisance and an eyesore. Yet, law enforcement agencies have found them to be useful for understanding gang activities, and uncovering the extent of a gang's territory in large metropolitan areas. The current method for matching and retrieving graffiti is based on a manual database search that is not only inaccurate but also time consuming. We present a content-based image retrieval (CBIR) system for automatic matching and retrieval of graffiti images. Our system represents each graffiti image by a bag of SIFT (Scale Invariant Feature Transform) features. The similarity between a query image and a graffiti image in the database is computed based on the number of matched SIFT features between the two images under certain geometric constraints. Experimental results on two graffiti databases with thousands of graffiti images show encouraging results.","Title: Passive Learning with Target Risk\nYear: 2013\n\nAbstract: In this paper we consider learning in passive setting but with a slight modification. We assume that the target expected loss, also referred to as target risk, is provided in advance for learner as prior knowledge. Unlike most studies in the learning theory that only incorporate the prior knowledge into the generalization bounds, we are able to explicitly utilize the target risk in the learning process. Our analysis reveals a surprising result on the sample complexity of learning: by exploiting the target risk in the learning algorithm, we show that when the loss function is both strongly convex and smooth, the sample complexity reduces to $\\O(\\log (\\frac{1}{\\epsilon}))$, an exponential improvement compared to the sample complexity $\\O(\\frac{1}{\\epsilon})$ for learning with strongly convex loss functions. Furthermore, our proof is constructive and is based on a computationally efficient stochastic optimization algorithm for such settings which demonstrate that the proposed algorithm is practically useful.","Title: Meta-Classification of Multimedia Classifiers\nYear: 2002\n\nAbstract: Combining multiple classifiers is of particular interest in the multimedia systems, since there is usually data of very different types/modalities that should be mined or analyzed. Our wearable 'experience collection' system unobtrusively records the wearer's conversation, recognizes the face of the dialog partner and remembers his/her voice. When the system sees the same person's face or hears the same voice it can then use a summary of the last conversation with this person to remind the wearer. To correctly identify a person from a mixture of video and audio stream, classification judgments from individual modality classifiers must be combined effectively to yield a more accurate decision. A meta-classification strategy of combining multimodal classifiers using Support Vector Machine is proposed. Preliminary, empirical results show that combining different face recognition and speaker identification technology by meta-classification is dramatically more effective than weighted interpolation. Meta-classification is general enough to be applied to any application that needs to combine multiple classifiers without much modification.","Title: Data analysis for a multimedia library\nYear: 2000\n\nAbstract: This section describes the indexing, search, and retrieval of various combinations of audio, video, text, and image media\n and the automated content processing that enables it. The intent is to provide a framework for data analysis in multimedia\n digital libraries. The organization of this article is as follows: The introduction briefly distinguishes digital from traditional\n libraries and touches on the specific issues important to searching the content of multimedia libraries. The second section\n introduces the Informedia Digital Video Library as an example of a multimedia library, including a quick tour of the functionality.\n The next section discusses the processing of audio and image information, as it relates to a multimedia library. Section four\n illustrates the interplay between audio and video information using a video information retrieval experiment as an example.\n Section five discusses the exporting and sharing of metadata in a digital library using MPEG–7. Finally, section 6 presents\n one vision of a future digital library, where all personal memory can be recorded and accessed.","Title: Online Stochastic Optimization with Multiple Objectives\nYear: 2012\n\nAbstract: In this paper we propose a general framework to characterize and solve the stochastic optimization problems with multiple objectives underlying many real world learning applications. We first propose a projection based algorithm which attains an $O(T^{-1/3})$ convergence rate. Then, by leveraging on the theory of Lagrangian in constrained optimization, we devise a novel primal-dual stochastic approximation algorithm which attains the optimal convergence rate of $O(T^{-1/2})$ for general Lipschitz continuous objectives.","Title: Exploitation and exploration in a performance based contextual advertising system\nYear: 2010\n\nAbstract: The dynamic marketplace in online advertising calls for ranking systems that are optimized to consistently promote and capitalize better performing ads. The streaming nature of online data inevitably makes an advertising system choose between maximizing its expected revenue according to its current knowledge in short term (exploitation) and trying to learn more about the unknown to improve its knowledge (exploration), since the latter might increase its revenue in the future. The exploitation and exploration (EE) tradeoff has been extensively studied in the reinforcement learning community, however, not been paid much attention in online advertising until recently. In this paper, we develop two novel EE strategies for online advertising. Specifically, our methods can adaptively balance the two aspects of EE by automatically learning the optimal tradeoff and incorporating confidence metrics of historical performance. Within a deliberately designed offline simulation framework we apply our algorithms to an industry leading performance based contextual advertising system and conduct extensive evaluations with real online event log data. The experimental results and detailed analysis reveal several important findings of EE behaviors in online advertising and demonstrate that our algorithms perform superiorly in terms of ad reach and click-through-rate (CTR).","Title: Batch mode active learning and its application to medical image classification\nYear: 2006\n\nAbstract: The goal of active learning is to select the most informative examples for manual labeling. Most of the previous studies in active learning have focused on selecting a single unlabeled example in each iteration. This could be inefficient since the classification model has to be retrained for every labeled example. In this paper, we present a framework for \"batch mode active learning\" that applies the Fisher information matrix to select a number of informative examples simultaneously. The key computational challenge is how to efficiently identify the subset of unlabeled examples that can result in the largest reduction in the Fisher information. To resolve this challenge, we propose an efficient greedy algorithm that is based on the property of submodular functions. Our empirical studies with five UCI datasets and one real-world medical image classification show that the proposed batch mode active learning algorithm is more effective than the state-of-the-art algorithms for active learning.","Title: Online Kernel Learning with a Near Optimal Sparsity Bound.\nYear: 2013\n\nAbstract: In this work, we focus on Online Sparse Kernel Learning that aims to online learn a kernel classifier with a bounded number of support vectors. Although many online learning algorithms have been proposed to learn a sparse kernel classifier, most of them fail to bound the number of support vectors used by the final solution which is the average of the intermediate kernel classifiers generated by online algorithms. The key idea of the proposed algorithm is to measure the difficulty in correctly classifying a training example by the derivative of a smooth loss function, and give a more chance to a difficult example to be a support vector than an easy one via a sampling scheme. Our analysis shows that when the loss function is smooth, the proposed algorithm yields similar performance guarantee as the standard online learning algorithm but with a near optimal number of support vectors (up to a poly(lnT) factor). Our empirical study shows promising performance of the proposed algorithm compared to the state-of-the-art algorithms for online sparse kernel learning.","Title: Identifying neuronal assemblies with local and global connectivity with scale space spectral clustering\nYear: 2007\n\nAbstract: A nonparametric approach is proposed to identify clusters of functionally interdependent neurons, independent of the time scale at which they are maximally correlated. The neural point processes are represented in a N-dimensional scale space using the Haar wavelet transform. A similarity measure between any given pair of neurons is defined in the scale space. Clusters of ''similar'' neurons are identified by first reducing the N-dimensional scale space representation using principal components to obtain a Q-dimensional space. The weighted principal components are subsequently used to connect each neuron to the others in a graph representation. A probabilistic spectral clustering algorithm is used to perform graph partitioning by maximizing cluster compactness. Performance is compared to that of the k-means and the expectation-maximization algorithms for 120 neurons with time-varying intensity functions consisting of spontaneous background activity and phased response elicited at distinct time scales.","Title: Machine learning for information retrieval\nYear: 2011\n\nAbstract: In recent years, we have witnessed successful application of machine learning techniques to a wide range of information retrieval problems, including Web search engines, recommendation systems, online advertising, etc. It is thus critical for researchers in the information retrieval community to understand the core machine learning techniques. In order to accommodate audiences with different levels of understanding of machine learning, we divide this tutorial into two sessions: the first session will focus on basic machine learning concepts and tools; in the second session, we will introduce more advanced topics in machine learning, and will present recent developments in machine learning and its application to information retrieval. Each season is self-contained. Session 1: Core Learning Technologies for Information Retrieval. This session of the tutorial will cover the core machine learning methods, basic optimization techniques and key information retrieval applications. In particular, it includes: 1). Core concepts in machine learning, such as supervised learning/unsupervised learning, bias and variance trade off, and probabilistic models; 2). Useful concepts and algorithms in optimization including the first and second order gradient methods, and Expectation and Maximization; 3). The application of machine learning methods to key information retrieval problems including text classification, collaborative filtering, clustering and learning to rank; Session 2: Emerging Learning Technologies for Information Retrieval. This session will cover more advanced machine learning techniques that have started to be utilized in information retrieval applications. In particular, it will cover: 1). Advanced Optimization Techniques including stochastic optimization and smooth minimization; 2). Emerging Learning Techniques such as Multiple-Instance Learning, Active Learning and Semi-supervised Learning. The tutorial will benefit a large body of audience in the information retrieval community, ranging from students who are new to machine learning to the seasoned researchers who would like to understand the recent advance in machine learning for information retrieval research. This tutorial will also benefit the practitioners who apply learning techniques to real-world information retrieval systems.","Title: Multi-label Multiple Kernel Learning\nYear: 2008\n\nAbstract: We present a multi-label multiple kernel learning (MKL) formulation in which the data are embedded into a low-dimensional space directed by the instance- label correlations encoded into a hypergraph. We formulate the problem in the kernel-induced feature space and propose to learn the kernel matrix as a linear combination of a given collection of kernel matrices in the MKL framework. The proposed learning formulation leads to a non-smooth min-max problem, which can be cast into a semi-infinite linear program (SILP). We further propose an ap- proximate formulation with a guaranteed error bound which involves an uncon- strained convex optimization problem. In addition, we show that the objective function of the approximate formulation is differentiable with Lipschitz continu- ous gradient, and hence existing methods can be employed to compute the optimal solution efficiently. We apply the proposed formulation to the automated annota- tion of Drosophila gene expression pattern images, and promising results have been reported in comparison with representative algorithms.","Title: Symmetric Statistical Translation Models for Automatic Image Annotation\nYear: 2005\n\nAbstract: Automatic image annotation provides means for users to search image collections on the semantic level using natural language queries. In the past, statistical machine translation models have been successfully applied to automatic image annotation. A problem with this approach is that, due to the skewed distribution of term frequency for annotation words, common words have been overly favored, which leaves little room for uncommon words to be used in auto-annotations. In contrast, studies on information retrieval have revealed that uncommon words are at least as important as common words since they are also often used in users' queries. Unlike the previous studies where a single type of statistical translation model is considered for automatic image annotation, in this paper, we studied two types of statistical translation models: a forward translation model, which translates visual information into textual words, and a backward model, which translates textual words into visual images. In particular, we propose a new statistical translation model, named regularization-based symmetric statistical translation model, which combines strength of forward and backward models to alleviate the problem of overly favoring common words. Our empirical studies with the Corel dataset have shown that the proposed model performs considerably better than the existing translation model and a state-of-the-art approach for automatic image annotation.","Title: Large-scale near-duplicate image retrieval by kernel density estimation.\nYear: 2012\n\nAbstract: Bag-of-words model is one of the most widely used methods in the recent studies of multimedia data retrieval. The key idea of the bag-of-words model is to quantize the bag of local features, for example SIFT, to a histogram of visual words and then standard information retrieval technologies developed from text retrieval can be applied directly. Despite its success, one problem of the bag-of-words model is that the two key steps, i.e., feature quantization and retrieval, are separated. In other words, the step of generating bag-of-words representation is not optimized for the step of retrieval which often leads to a sub-optimal performance. In this paper we propose a statistical framework for large-scale near-duplication image retrieval which unifies the two steps by introducing kernel density function. The central idea of the proposed method is to represent each image by a kernel density function and the similarity between the query image and a database image is then estimated as the query likelihood. In order to make the proposed method applicable to large-scale data sets, we have developed efficient algorithms for both estimating the density function of each image and computing the query likelihood. Our empirical studies confirm that the proposed method is not only more effective but also more efficient than the bag-of-words model.","Title: Flexible Mixture Model for Collaborative Filtering\nYear: 2003\n\nAbstract: This paper presents a flexible mixture model (FMM) for collaborative filtering. FMM extends existing partitioning/clustering algorithms for collaborative filtering by clustering both users and items together simultaneously without assuming that each user and item should only belong to a single cluster. Furthermore, with the introduction of 'preference' nodes, the proposed framework is able to explicitly model how users rate items, which can vary dramatically, even among the users with similar tastes on items. Empirical study over two datasets of movie ratings has shown that our new algorithm outperforms five other collaborative filtering algorithms substantially.","Title: Using knowledge driven matrix factorization to reconstruct modular gene regulatory network\nYear: 2008\n\nAbstract: Reconstructing gene networks from micro-array data can provide information on the mechanisms that govern cellular processes. Numerous studies have been devoted to addressing this problem. A popular method is to view the gene network as a Bayesian inference network, and to apply structure learning methods to determine the topology of the gene network. There are, however, several shortcomings with the Bayesian structure learning approach for reconstructing gene networks. They include high computational cost associated with analyzing a large number of genes and inefficiency in exploiting prior knowledge of co-regulation that could be derived from Gene Ontology (GO) information. In this paper, we present a knowledge driven matrix factorization (KMF) framework for reconstructing modular gene networks that addresses these shortcomings. In KMF, gene expression data is initially used to estimate the correlation matrix. The gene modules and the interactions among the modules are derived by factorizing the correlation matrix. The prior knowledge in GO is integrated into matrix factorization to help identify the gene modules. An alternating optimization algorithm is presented to efficiently find the solution. Experiments show that our algorithm performs significantly better in identifying gene modules than several state-of-the-art algorithms, and the interactions among the modules uncovered by our algorithm are proved to be biologically meaningful.","Title: An efficient primal dual prox method for non-smooth optimization\nYear: 2012\n\nAbstract: We study the non-smooth optimization problems in machine learning, where both the loss function and the regularizer are non-smooth functions. Previous studies on efficient empirical loss minimization assume either a smooth loss function or a strongly convex regularizer, making them unsuitable for non-smooth optimization. We develop a simple yet efficient method for a family of non-smooth optimization problems where the dual form of the loss function is bilinear in primal and dual variables. We cast a non-smooth optimization problem into a minimax optimization problem, and develop a primal dual prox method that solves the minimax optimization problem at a rate of $$O(1/T)$$O(1/T) assuming that the proximal step can be efficiently solved, significantly faster than a standard subgradient descent method that has an $$O(1/\\sqrt{T})$$O(1/T) convergence rate. Our empirical studies verify the efficiency of the proposed method for various non-smooth optimization problems that arise ubiquitously in machine learning by comparing it to the state-of-the-art first order methods.","Title: Video Retrieval with the Informedia Digital Video Library System\nYear: 2001\n\nAbstract: Background: The Informedia Digital Video Library System. The Informedia Digital Video Library [1] was the only NSF DLI project focusing specifically on information extraction from video and audio content. Over a terabyte of online data was collected, with automatically generated metadata and indices for retrieving videos from this library. The architecture for the project was based on the premise that real-time constraints on library and associated metadata creation could be relaxed in order to realize increased automation and deeper parsing and indexing for identifying the library contents and breaking it into segments. Library creation was an offline activity, with library exploration by users occurring online and making use of the generated metadata and segmentation. The goal of the Informedia interface was to enable quick access to relevant information in a digital video library, leveraging from derived metadata and the partitioning of the video into small segments. Figure 1 shows the IDVLS interface following a query. In this figure, a set of results is displayed at the bottom. The display includes a window containing a headline, and a pictorial menu of video segments each represented with a thumbnail image at approximately ? resolution of the video in the horizontal and vertical dimensions. The headline window automatically pops up whenever the mouse is positioned over a result item; the headline window for the first result is shown. IDVLS also supports other ways of navigating and browsing the digital video library. These interface features were essential to deal with the ambiguity of the derived data generated by speech recognition, image processing, and natural language processing. Consider the filmstrip and video playback IDVLS window shown in Figure 2.For this actual video in the IDVLS library, the segmentation process failed, resulting in a thirty-minute segment. This long segment was one of the returned results for the query \"Mir collision. \" The filmstrip in Figure 2 shows that the segment is more than just a story on the Russian space station, but rather begins with a commercial, then the weather, and then coverage of Hong Kong before addressing Mir. By overlaying the filmstrip and video playback windows with match location information, the user can quickly see that matches don't occur until later in the segment, after these other stories that were irrelevant to the query. The match bars are optionally color-coded to specific query words; in Figure 2 \"Mir\" matches are in red and \"collision\" matches in purple. When the user moved the mouse over the match bars in the filmstrip, a text window displayed the actual matching word from the transcript or Video OCR metadata for that particular match; \"Mir\" is shown in one such text window in Figure 2.By investigating the distribution of match locations on the filmstrip, the user can determine the relevance of the returned result and the location of interest within the segment. The user can click on a match bar to jump directly to that point in the video segment. Hence, clicking the mouse as shown in Figure 2 would start playing the video at this mention of \"Mir\" with the overhead shot of people at desks. Similarly, IDVLS provided \"seek to next match\" and \"seek to previous match\" buttons in the video player allowing the user to quickly jump from one match to the next. In the example of Figure 2, these interface features allowed the user to bypass problems in segmentation and jump directly to the \"Mir\" story without having to first watch the opening video on other topics.","Title: A maximum coherence model for dictionary-based cross-language information retrieval\nYear: 2005\n\nAbstract: One key to cross-language information retrieval is how to efficiently resolve the translation ambiguity of queries given their short length. This problem is even more challenging when only bilingual dictionaries are available, which is the focus of this paper. In the previous research of cross-language information retrieval using bilingual dictionaries, the word co-occurrence statistics is used to determine the most likely translations of queries. In this paper, we propose a novel statistical model, named ``maximum coherence model'', which estimates the translation probabilities of query words that are consistent with the word co-occurrence statistics. Unlike the previous work, where a binary decision is made for the selection of translations, the new model maintains the uncertainty in translating query words when their sense ambiguity is difficult to resolve. Furthermore, this new model is able to estimate translations of multiple query words simultaneously. This is in contrast to many previous approaches where translations of individual query words are determined independently. Empirical studies with TREC datasets have shown that the maximum coherence model achieves a relative 10% - 40% improvement in cross-language information retrieval, comparing to other approaches that also use word co-occurrence statistics for sense disambiguation.","Title: A Probabilistic Approach for Optimizing Spectral Clustering\nYear: 2005\n\nAbstract: Spectral clustering enjoys its success in both data cluster ing and semi- supervised learning. But, most spectral clustering algori thms cannot handle multi-class clustering problems directly. Additio nal strategies are needed to extend spectral clustering algorithms to multi-class cluster- ing problems. Furthermore, most spectral clustering algorithms employ hard cluster membership, which is likely to be trapped by the local op- timum. In this paper, we present a new spectral clustering algorithm, named \"Soft Cut\". It improves the normalized cut algorithm by intro- ducing soft membership, and can be efficiently computed usin g a bound optimization algorithm. Our experiments with a variety of datasets have shown the promising performance of the proposed clustering algorithm.","Title: Query translation disambiguation as graph partitioning\nYear: 2005\n\nAbstract: Resolving ambiguity in the process of query translation is crucial to cross-language information retrieval when only a bilingual dictionary is available. In this paper we propose a novel approach for query translation disambiguation, named \"spectral query translation model\". The proposed approach views the problem of query translation disambiguation as a graph partitioning problem. For a given query, a weighted graph is first created for all possible translations of query words based on the co-occurrence statistics of the translation words. The best translation of the query is then determined by the most strongly connected component within the graph. The proposed approach distinguishes from previous approaches in that the translations of all query words are estimated simultaneously. Furthermore, translation probabilities are introduced in the proposed approach to capture the uncertainty in translating queries. Empirical studies with TREC datasets have shown that the spectral query translation model achieves a relative 20% -50% improvement in cross-language information retrieval, compared to other approaches that also exploit word co-occurrence statistics for query translation disambiguation.","Title: A Bayesian reconstruction algorithm for synthesis aperture imaging radiometer\nYear: 2012\n\nAbstract: Regularization methods are very efficient in reconstructing the radiometric brightness temperature maps form interferometric measurements. The performance of these methods depends on selecting an optimal regularization parameter, which is typically defined by using the cross-validation or the L-curve. In this paper, we present a novel brightness temperature reconstruction method, which can automatically learn the optimal parameter by Bayesian learning without reducing the performance of the brightness temperature image. To support the theory, numerical simulations are presented and analyzed with emphasis on stability and error analysis.","Title: The Research on Teaching Method of Basics Course of Computer based on Cluster Analysis\nYear: 2010\n\nAbstract: Due to the difference between rural and urban and regional disparity of computer basic education produced at their middle school period, computer application level of the freshmen make a great difference. According to this situation, corresponding skill survey is designed. Data obtained from the survey is analyzed by K-means cluster algorithm. And the students are classified into three types according to their master degree of computer application. The study group is composed of the students selected from each type. This can promote them to study from each other and draw the strong points of others to offset their own weakness. The teachers can adjust the emphasis of their teaching to different classes according to the master degree of computer application of the classes. Good teaching effect is achieved by using this teaching method in Basics Course of Computer.","Title: Unsupervised transfer classification: application to text categorization\nYear: 2010\n\nAbstract: We study the problem of building the classification model for a target class in the absence of any labeled training example for that class. To address this difficult learning problem, we extend the idea of transfer learning by assuming that the following side information is available: (i) a collection of labeled examples belonging to other classes in the problem domain, called the auxiliary classes; (ii) the class information including the prior of the target class and the correlation between the target class and the auxiliary classes. Our goal is to construct the classification model for the target class by leveraging the above data and information. We refer to this learning problem as unsupervised transfer classification. Our framework is based on the generalized maximum entropy model that is effective in transferring the label information of the auxiliary classes to the target class. A theoretical analysis shows that under certain assumption, the classification model obtained by the proposed approach converges to the optimal model when it is learned from the labeled examples for the target class. Empirical study on text categorization over four different data sets verifies the effectiveness of the proposed approach.","Title: Learn to weight terms in information retrieval using category information\nYear: 2005\n\nAbstract: How to assign appropriate weights to terms is one of the critical issues in information retrieval. Many term weighting schemes are unsupervised. They are either based on the empirical observation in information retrieval, or based on generative approaches for language modeling. As a result, the existing term weighting schemes are usually insufficient in distinguishing informative words from the uninformative ones, which is crucial to the performance of information retrieval. In this paper, we present supervised term weighting schemes that automatically learn term weights based on the correlation between word frequency and category information of documents. Empirical studies with the ImageCLEF dataset have indicated that the proposed methods perform substantially better than the state-of-the-art approaches for term weighting and other alternatives that exploit category information for information retrieval.","Title: Efficient Convex Relaxation for Transductive Support Vector Machine\nYear: 2007\n\nAbstract: We consider the problem of Support Vector Machine transduction, which involves a combinatorial problem with exponential computational complexity in the num- ber of unlabeled examples. Although several studies are devoted to Transductive SVM, they suffer either from the high computation complexity or from the so- lutions of local optimum. To address this problem, we propose solving Trans- ductive SVM via a convex relaxation, which converts the NP-hard problem to a semi-definite programming. Compared with the other SDP relaxation for Trans- ductive SVM, the proposed algorithm is computationally more efficient with the number of free parameters reduced from O(n2) to O(n) where n is the number of examples. Empirical study with several benchmark data sets shows the promising performance of the proposed algorithm in comparison with other state-of-the-art implementations of Transductive SVM.","Title: A Robust Regularization Kernel Regression Algorithm for Passive Millimeter Wave Imaging Target Detection.\nYear: 2012\n\nAbstract: This letter deals with small target detection in passive millimeter wave (PMMW) imaging. Specifically, it focuses on a general detection scheme, where, first, the background is suppressed through a background prediction algorithm, and then the detection is accomplished. A precise prediction of the background is essential to a successful outcome. In practical applications, background estimation pro...","Title: Title generation for machine-translated documents\nYear: 2001\n\nAbstract: In this paper, we present and compare automatically generated titles for machine-translated documents using several different statistics-based methods. A Naïve Bayesian, a K-Nearest Neighbour, a TF-IDF and an iterative Expectation-Maximization method for title generation were applied to 1000 original English news documents and again to the same documents translated from English into Portuguese, French or German and back to English using SYSTRAN. The AutoSummarization function of Microsoft Word was used as a base line. Results on several metrics show that the statistics-based methods of title generation for machine-translated documents are fairly language independent and title generation is possible at a level approaching the accuracy of titles generated for the original English documents.","Title: Stochastic Optimization of Smooth Loss.\nYear: 2013\n\nAbstract: In this paper, we first prove a high probability bound rather than an expectation bound for stochastic optimization with smooth loss. Furthermore, the existing analysis requires the knowledge of optimal classifier for tuning the step size in order to achieve the desired bound. However, this information is usually not accessible in advanced. We also propose a strategy to address the limitation.","Title: Semi-supervised Collaborative Text Classification\nYear: 2007\n\nAbstract: Most text categorization methods require text content of documents that is often difficult to obtain. We consider \"Collaborative Text Categorization\", where each document is represented by the feedback from a large number of users. Our study focuses on the semi-supervised case in which one key challenge is that a significant number of users have not rated any labeled document. To address this problem, we examine several semi-supervised learning methods and our empirical study shows that collaborative text categorization is more effective than content-based text categorization and the manifold regularization is more effective than other state-of-the-art semi-supervised learning methods.","Title: A Bayesian Approach Toward Finding Communities and Their Evolutions in Dynamic Social Networks\nYear: 2009\n\nAbstract: Although a large body of work are devoted to finding communities in static social networks, only a few studies examined the dynamics of communities in evolving social networks. In this paper, we propose a dynamic stochastic block model for finding communities and their evolutions in a dynamic social network. The proposed model captures the evolution of communities by explicitly modeling the transition of community memberships for individual nodes in the network. Unlike many existing approaches for modeling social networks that estimate parameters by their most likely values (i.e., point estimation), in this study, we employ a Bayesian treatment for parameter estimation that computes the posterior distributions for all the unknown parameters. This Bayesian treatment allows us to capture the uncertainty in parameter values and therefore is more robust to data noise than point estimation. In addition, an efficient algorithm is developed for Bayesian inference to handle large sparse social networks. Extensive experimental studies based on both synthetic data and real-life data demonstrate that our model achieves higher accuracy and reveals more insights in the data than several state-of-the- art algorithms.","Title: Using a probabilistic source model for comparing images\nYear: 2002\n\nAbstract: We propose a probabilistic model for image retrieval. To obtain the similarity between the query image IQ and any image I' in the collection, the model computes the probability of generating the image I' given the observation of the query image IQ. We compare our probabilistic model for image retrieval with a color histogram based image retrieval method and the IBM QBIC image search engine. The evaluation used the 11-hour video retrieval collection (80,000 extracted images) and associated queries from the 2001 TREC-10 information retrieval evaluations. The experimental results show that the probabilistic model dramatically outperforms the color histogram based image retrieval method and the IBM QBIC image search engine by 40%.","Title: Title Generation Using a Training Corpus\nYear: 2001\n\nAbstract: This paper discusses fundamental issues involved in word selection for title generation. We review several methods for title generation, namely extractive summarization and two versions of a Naïve Bayesian, and compare the performance of those methods using an F1 metric. In addition, we introduce a novel approach to title generation using the k-nearest neighbor (KNN) algorithm. Both the KNN method and a limited-vocabulary Naïve Bayesian method outperform the other evaluated methods with an F1 score of around 20%. Since KNN produces complete and legible titles, we conclude that KNN is a very promising method for title generation, provided good content overlap exists between the training corpus and the test documents.","Title: Tag Completion for Image Retrieval.\nYear: 2013\n\nAbstract: Many social image search engines are based on keyword/tag matching. This is because tag based image retrieval (TBIR) is not only efficient but also effective. The performance of TBIR is highly dependent on the availability and quality of manual tags. Recent studies have shown that manual tags are often unreliable and inconsistent. In addition, since many users tend to choose general and ambiguous tags in order to minimize their efforts in choosing appropriate words, tags that are specific to the visual content of images tend to be missing or noisy, leading to a limited performance of TBIR. To address this challenge, we study the problem of \\textit{tag completion} where the goal is to automatically fill in the missing tags as well as correct noisy tags for given images. We represent the image-tag relation by a \\textit{tag matrix}, and search for the optimal tag matrix consistent with both the observed tags and the visual similarity. We propose an efficient algorithm for solving the related optimization problem. Extensive empirical studies show that the proposed algorithm is significantly more effective than the state-of-the-art algorithms. Our studies also verify that the proposed algorithm is computationally efficient and scales well to large databases.","Title: Correlated Label Propagation with Application to Multi-label Learning\nYear: 2006\n\nAbstract: Many computer vision applications, such as scene analysis and medical image interpretation, are ill-suited for traditional classification where each image can only be associated with a single class. This has stimulated recent work in multi-label learning where a given image can be tagged with multiple class labels. A serious problem with existing approaches is that they are unable to exploit correlations between class labels. This paper presents a novel framework for multi-label learning termed Correlated Label Propagation (CLP) that explicitly models interactions between labels in an efficient manner. As in standard label propagation, labels attached to training data points are propagated to test data points; however, unlike standard algorithms that treat each label independently, CLP simultaneously co-propagates multiple labels. Existing work eschews such an approach since naive algorithms for label co-propagation are intractable. We present an algorithm based on properties of submodular functions that efficiently finds an optimal solution. Our experiments demonstrate that CLP leads to significant gains in precision/recall against standard techniques on two real-world computer vision tasks involving several hundred labels.","Title: Inferring functional cortical networks from spike train ensembles using Dynamic Bayesian Networks\nYear: 2009\n\nAbstract: A fundamental goal in systems neuroscience is to infer the functional connectivity among neuronal elements coordinating information processing in the brain. In this work, we investigate the applicability of Dynamic Bayesian Networks (DBN) in inferring the structure of cortical networks from the observed spike trains. DBNs have unique features that make them capable of detecting causal relationships between spike trains such as modeling time-dependent relationships, detecting non-linear interactions and inferring connectivity between neurons from the observed ensemble activity. A probabilistic point process model was used to assess the performance under systematic variations of the model parameters. Results demonstrate the utility of DBN in inferring functional connectivity in cortical network models.","Title: On predicting rare classes with SVM ensembles in scene classification\nYear: 2003\n\nAbstract: Scene classification is an important technique to infer high-level semantic scene categories from low-level visual features. However, in the real world the positive data for many scenes may be rare, which degrades the performance of many classifiers. In this paper, we propose SVM ensembles to address the rare class problem. Various classifier combination strategies are investigated, including majority voting, sum rule, neural network gater and hierarchical SVMs. We also compare our method with two other common approaches for dealing with the rare class problem. Our experimental results show that hierarchical SVMs can achieve significantly better and more stable performance than other strategies, as well as high computational efficiency.","Title: BoostCluster: boosting clustering by pairwise constraints\nYear: 2007\n\nAbstract: Data clustering is an important task in many disciplines. A large number of studies have attempted to improve clustering by using the side information that is often encoded as pairwise constraints. However, these studies focus on designing special clustering algorithms that can effectively exploit the pairwise constraints. We present a boosting framework for data clustering,termed as BoostCluster, that is able to iteratively improve the accuracy of any given clustering algorithm by exploiting the pairwise constraints. The key challenge in designing a boosting framework for data clustering is how to influence an arbitrary clustering algorithm with the side information since clustering algorithms by definition are unsupervised. The proposed framework addresses this problem by dynamically generating new data representations at each iteration that are, on the one hand, adapted to the clustering results at previous iterations by the given algorithm, and on the other hand consistent with the given side information. Our empirical study shows that the proposed boosting framework is effective in improving the performance of a number of popular clustering algorithms (K-means, partitional SingleLink, spectral clustering), and its performance is comparable to the state-of-the-art algorithms for data clustering with side information.","Title: Sensor localization in an obstructed environment\nYear: 2005\n\nAbstract: Sensor localization can be divided into two categories: range-based approaches, and rang-free approaches. Although range-based approaches tend to be more accurate than range-free approaches, they are more sensitive to errors in distance measurement. Despite of the efforts on recovering sensors’ Euclidean coordinates from erroneous distance measurements, as will be illustrated in this paper, they are still prone to distance errors, particularly in an obstruction abundant environment. In this paper, we propose a new algorithm for sensor localization based on Multiscale Radio Transmission Power(MRTP). It gradually increases the scale level of transmission power, and the distance is determined by the minimal scale of received signals. Unlike the range-based approaches, which treat each measured distance as an approximation of the true one, in our new approach, the measured distance serves as a constraint that limits the feasible location of sensors. Our simulations have shown that the MRTP-based approach is able to provide accurate and robust estimation of location, especially in an area abundant in obstructions, where most current approaches fail to perform well.","Title: A study of methods for normalizing user ratings in collaborative filtering\nYear: 2004\n\nAbstract: The goal of collaborative filtering is to make recommendations for a test user by utilizing the rating information of users who share interests similar to the test user. Because ratings are determined not only by user interests but also the rating habits of users, it is important to normalize ratings of different users to the same scale. In this paper, we compare two different normalization strategies for user ratings, namely the Gaussian normalization method and the decoupling normalization method. Particularly, we incorporated these two rating normalization methods into two collaborative filtering algorithms, and evaluated their effectiveness on the EachMovie dataset. The experiment results have shown that the decoupling method for rating normalization is more effective than the Gaussian normalization method in improving the performance of collaborative filtering algorithms.","Title: Online feature selection for mining big data\nYear: 2012\n\nAbstract: Most studies of online learning require accessing all the attributes/features of training instances. Such a classical setting is not always appropriate for real-world applications when data instances are of high dimensionality or the access to it is expensive to acquire the full set of attributes/features. To address this limitation, we investigate the problem of Online Feature Selection (OFS) in which the online learner is only allowed to maintain a classifier involved a small and fixed number of features. The key challenge of Online Feature Selection is how to make accurate prediction using a small and fixed number of active features. This is in contrast to the classical setup of online learning where all the features are active and can be used for prediction. We address this challenge by studying sparsity regularization and truncation techniques. Specifically, we present an effective algorithm to solve the problem, give the theoretical analysis, and evaluate the empirical performance of the proposed algorithms for online feature selection on several public datasets. We also demonstrate the application of our online feature selection technique to tackle real-world problems of big data mining, which is significantly more scalable than some well-known batch feature selection algorithms. The encouraging results of our experiments validate the efficacy and efficiency of the proposed techniques for large-scale applications.","Title: Design and Implementation of an Open Programmable Router Compliant to IETF ForCES Specifications\nYear: 2007\n\nAbstract: IETF ForCES (Forwarding and Control Element Separation) is defining specifications for interfaces and modular resources abstractions of network equipments. ForCES brings flexible, programmable, and cost-effective advantages over tradi-tional architectures for network equipments. In this paper, the design and implementation of a ForCES-based router (ForTER) was illustrated. The framework of ForCES was introduced firstly. Then a layered model for the implementation of ForTER was presented. Next, the implementation details of Control Ele-ment (CE) and Forwarding Element (FE) of ForTER were intro-duced, which complied with the ForCES framework and related IETF ForCES specifications. Finally, results of experiments for routing and for SNMP support showed the feasibility of the sys-tem architecture of ForTER, which also fundamentally testified the feasibility of ForCES and related specifications that are being standardized by IETF ForCES working group.","Title: A Bayesian framework for community detection integrating content and link\nYear: 2012\n\nAbstract: This paper addresses the problem of community detection in networked data that combines link and content analysis. Most existing work combines link and content information by a generative model. There are two major shortcomings with the existing approaches. First, they assume that the probability of creating a link between two nodes is determined only by the community memberships of the nodes; however other factors(e.g. popularity) could also affect the link pattern. Second, they use generative models to model the content of individual nodes, whereas these generative models are vulnerable to the content attributes that are irrelevant to communities. We propose a Bayesian framework for combining link and content information for community detection that explicitly addresses these shortcomings. A new link model is presented that introduces a random variable to capture the node popularity when deciding the link between two nodes; a discriminative model is used to determine the community membership of a node by its content. An approximate inference algorithm is presented for efficient Bayesian inference. Our empirical study shows that the proposed framework outperforms several state-of-the-art approaches in combining link and content information for community detection.","Title: A framework for incorporating class priors into discriminative classification\nYear: 2005\n\nAbstract: Discriminative and generative methods provide two distinct approaches to machine learning classification. One advantage of generative approaches is that they naturally model the prior class distributions. In contrast, discriminative approaches directly model the conditional distribution of class given inputs, so the class priors are only implicitly obtained if the input density is known. In this paper, we propose a framework for incorporating class prior proportions into discriminative methods in order to improve their classification accuracy. The basic idea is to enforce that the distribution of class labels predicted on the test data by the discriminative model is consistent with the class priors. Therefore, the discriminative model has to not only fit the training data well but also predict class labels for the test data that are consistent with the class priors. Experiments on five different UCI datasets and one image database show that this framework is effective in improving the classification accuracy when the training data and the test data come from the same class proportions, even if the test data does not have exactly the same feature distribution as the training data.","Title: A Boosting Framework for Visuality-Preserving Distance Metric Learning and Its Application to Medical Image Retrieval\nYear: 2010\n\nAbstract: Similarity measurement is a critical component in content-based image retrieval systems, and learning a good distance metric can significantly improve retrieval performance. However, despite extensive study, there are several major shortcomings with the existing approaches for distance metric learning that can significantly affect their application to medical image retrieval. In particular, “similarity” can mean very different things in image retrieval: resemblance in visual appearance (e.g., two images that look like one another) or similarity in semantic annotation (e.g., two images of tumors that look quite different yet are both malignant). Current approaches for distance metric learning typically address only one goal without consideration of the other. This is problematic for medical image retrieval where the goal is to assist doctors in decision making. In these applications, given a query image, the goal is to retrieve similar images from a reference library whose semantic annotations could provide the medical professional with greater insight into the possible interpretations of the query image. If the system were to retrieve images that did not look like the query, then users would be less likely to trust the system; on the other hand, retrieving images that appear superficially similar to the query but are semantically unrelated is undesirable because that could lead users toward an incorrect diagnosis. Hence, learning a distance metric that preserves both visual resemblance and semantic similarity is important. We emphasize that, although our study is focused on medical image retrieval, the problem addressed in this work is critical to many image retrieval systems. We present a boosting framework for distance metric learning that aims to preserve both visual and semantic similarities. The boosting framework first learns a binary representation using side information, in the form of labeled pairs, and then computes the distance as a weighted Hamming distance using the learned binary representation. A boosting algorithm is presented to efficiently learn the distance function. We evaluate the proposed algorithm on a mammographic image reference library with an Interactive Search-Assisted Decision Support (ISADS) system and on the medical image data set from ImageCLEF. Our results show that the boosting framework compares favorably to state-of-the-art approaches for distance metric learning in retrieval accuracy, with much lower computational cost. Additional evaluation with the COREL collection shows that our algorithm works well for regular image data sets.","Title: Active multiple kernel learning for interactive 3D object retrieval systems\nYear: 2011\n\nAbstract: An effective relevance feedback solution plays a key role in interactive intelligent 3D object retrieval systems. In this work, we investigate the relevance feedback problem for interactive intelligent 3D object retrieval, with the focus on studying effective machine learning algorithms for improving the user's interaction in the retrieval task. One of the key challenges is to learn appropriate kernel similarity measure between 3D objects through the relevance feedback interaction with users. We address this challenge by presenting a novel framework of Active multiple kernel learning (AMKL), which exploits multiple kernel learning techniques for relevance feedback in interactive 3D object retrieval. The proposed framework aims to efficiently identify an optimal combination of multiple kernels by asking the users to label the most informative 3D images. We evaluate the proposed techniques on a dataset of over 10,000 3D models collected from the World Wide Web. Our experimental results show that the proposed AMKL technique is significantly more effective for 3D object retrieval than the regular relevance feedback techniques widely used in interactive content-based image retrieval, and thus is promising for enhancing user's interaction in such interactive intelligent retrieval systems.","Title: Ranking refinement and its application to information retrieval\nYear: 2008\n\nAbstract: We consider the problem of ranking refinement, i.e., to improve the accuracy of an existing ranking function with a small set of labeled instances. We are, particularly, interested in learning a better ranking function using two complementary sources of information, ranking information given by the existing ranking function (i.e., a base ranker) and that obtained from users' feedbacks. This problem is very important in information retrieval where the feedback is gradually collected. The key challenge in combining the two sources of information arises from the fact that the ranking information presented by the base ranker tends to be imperfect and the ranking information obtained from users' feedbacks tends to be noisy. We present a novel boosting framework for ranking refinement that can effectively leverage the uses of the two sources of information. Our empirical study shows that the proposed algorithm is effective for ranking refinement, and furthermore significantly outperforms the baseline algorithms that incorporate the outputs from the base ranker as an additional feature.","Title: Learning a distance metric from multi-instance multi-label data\nYear: 2009\n\nAbstract: Multi-instance multi-label learning (MIML) refers to the learning problems where each example is represented by a bag/collection of instances and is labeled by multiple labels. An example application of MIML is visual object recognition in which each image is represented by multiple key points (i.e., instances) and is assigned to multiple object categories. In this paper, we study the problem of learning a distance metric from multi-instance multi-label data. It is significantly more challenging than the conventional setup of distance metric learning because it is difficult to associate instances in a bag with its assigned class labels. We propose an iterative algorithm for MIML distance metric learning: it first estimates the association between instances in a bag and its assigned class labels, and learns a distance metric from the estimated association by a discriminative analysis; the learned metric will be used to update the association between instances and class labels, which is further used to improve the learning of distance metric. We evaluate the proposed algorithm by the task of automated image annotation, a well known MIML problem. Our empirical study shows an encouraging result when combining the proposed algorithm with citation-kNN, a state-of-the-art algorithm for multi-instance learning.","Title: Automated Vocabulary Acquisition and Interpretation in Multimodal Conversational Systems\nYear: 2007\n\nAbstract: Motivated by psycholinguistic findings that eye gaze is tightly linked to human lan- guage production, we developed an unsuper- vised approach based on translation models to automatically learn the mappings between words and objects on a graphic display dur- ing human machine conversation. The ex- perimental results indicate that user eye gaze can provide useful information to establish such mappings, which have important impli- cations in automatically acquiring and inter- preting user vocabularies for conversational systems.","Title: Efficient Algorithm for Localized Support Vector Machine\nYear: 2010\n\nAbstract: This paper presents a framework called localized support vector machine (LSVM) for classifying data with nonlinear decision surfaces. Instead of building a sophisticated global model from the training data, LSVM constructs multiple linear SVMs, each of which is designed to accurately classify a given test example. A major limitation of this framework is its high computational cost since a unique model must be constructed for each test example. To overcome this limitation, we propose an efficient implementation of LSVM, termed profile SVM (PSVM). PSVM partitions the training examples into clusters and builds a separate linear SVM model for each cluster. Our empirical results show that (1) LSVM and PSVM outperform nonlinear SVM for all 20 of the evaluated data sets and (2) PSVM achieves comparable performance as LSVM in terms of model accuracy but with significant computational savings. We also demonstrate the efficacy of the proposed approaches in terms of classifying data with spatial and temporal dependencies.","Title: Efficient Kernel Clustering Using Random Fourier Features\nYear: 2012\n\nAbstract: Kernel clustering algorithms have the ability to capture the non-linear structure inherent in many real world data sets and thereby, achieve better clustering performance than Euclidean distance based clustering algorithms. However, their quadratic computational complexity renders them non-scalable to large data sets. In this paper, we employ random Fourier maps, originally proposed for large scale classification, to accelerate kernel clustering. The key idea behind the use of random Fourier maps for clustering is to project the data into a low-dimensional space where the inner product of the transformed data points approximates the kernel similarity between them. An efficient linear clustering algorithm can then be applied to the points in the transformed space. We also propose an improved scheme which uses the top singular vectors of the transformed data matrix to perform clustering, and yields a better approximation of kernel clustering under appropriate conditions. Our empirical studies demonstrate that the proposed schemes can be efficiently applied to large data sets containing millions of data points, while achieving accuracy similar to that achieved by state-of-the-art kernel clustering algorithms.","Title: Multi-Task Learning in Square Integrable Space.\nYear: 2011\n\nAbstract: Several kernel based methods for multi-task learning have been proposed, which leverage relations among tasks as regularization to enhance the overall learning accuracies. These methods assume that the tasks share the same kernel, which could limit their applications because in practice different tasks may need different kernels. The main challenge of introducing multiple kernels into multiple tasks is that models from different Reproducing Kernel Hilbert Spaces (RKHSs) are not comparable, making it difficult to exploit relations among tasks. This paper addresses the challenge by formalizing the problem in the Square Integrable Space (SIS). Specially, it proposes a kernel based method which makes use of a regularization term defined in the SIS to represent task relations. We prove a new representer theorem for the proposed approach in SIS. We further derive a practical method for solving the learning problem and conduct consistency analysis of the method. We discuss the relations between our method and an existing method. We also give an SVM based implementation of our method for multi-label classification. Experiments on two real-world data sets show that the proposed method performs better than the existing method. Copyright © 2011, Association for the Advancement of Artificial Intelligence. All rights reserved.","Title: A kernel density based approach for large scale image retrieval\nYear: 2011\n\nAbstract: Local image features, such as SIFT descriptors, have been shown to be effective for content-based image retrieval (CBIR). In order to achieve efficient image retrieval using local features, most existing approaches represent an image by a bag-of-words model in which every local feature is quantized into a visual word. Given the bag-of-words representation for images, a text search engine is then used to efficiently find the matched images for a given query. The main drawback with these approaches is that the two key steps, i.e., key point quantization and image matching, are separated, leading to sub-optimal performance in image retrieval. In this work, we present a statistical framework for large-scale image retrieval that unifies key point quantization and image matching by introducing kernel density function. The key ideas of the proposed framework are (a) each image is represented by a kernel density function from which the observed key points are sampled, and (b) the similarity of a gallery image to a query image is estimated as the likelihood of generating the key points in the query image by the kernel density function of the gallery image. We present efficient algorithms for kernel density estimation as well as for effective image matching. Experiments with large-scale image retrieval confirm that the proposed method is not only more effective but also more efficient than the state-of-the-art approaches in identifying visually similar images for given queries from large image databases.","Title: Array Configuration Design of One-Dimensional Mirrored Interferometric Aperture Synthesis\nYear: 2012\n\nAbstract: Mirrored interferometric aperture synthesis (IAS) (MIAS) is a novel interferometry which has the potential to reduce system complexity compared with that of conventional IAS. Array configuration design plays a major role in imaging. As the concept of MIAS was proposed just a few years ago, the problem of array configuration design has not been solved yet. In this letter, the principles of array configuration design of 1-D MIAS are proposed, and the corresponding optimization model is established and refined. The optimal array configurations are presented through simulated annealing method and show that 1-D MIAS can achieve almost twice as much of spatial resolution as 1-D IAS with the same array size.","Title: Non-parametric mixture models for clustering\nYear: 2010\n\nAbstract: Mixture models have been widely used for data clustering. However, commonly used mixture models are generally of a parametric form (e.g., mixture of Gaussian distributions or GMM), which significantly limits their capacity in fitting diverse multidimensional data distributions encountered in practice.We propose a non-parametric mixture model (NMM) for data clustering in order to detect clusters generated from arbitrary unknown distributions, using non-parametric kernel density estimates. The proposed model is non-parametric since the generative distribution of each data point depends only on the rest of the data points and the chosen kernel. A leave-one-out likelihood maximization is performed to estimate the parameters of the model. The NMM approach, when applied to cluster high dimensional text datasets significantly outperforms the state-of-the-art and classical approaches such as K-means, Gaussian Mixture Models, spectral clustering and linkage methods.","Title: Semi-supervised text categorization by active search\nYear: 2008\n\nAbstract: In automated text categorization, given a small number of labeled documents, it is very challenging, if not impossible, to build a reliable classifier that is able to achieve high classification accuracy. To address this problem, a novel web-assisted text categorization framework is proposed in this paper. Important keywords are first automatically identified from the available labeled documents to form the queries. Search engines are then utilized to retrieve from the Web a multitude of relevant documents, which are then exploited by a semi-supervised framework. To our best knowledge, this work is the first study of this kind. Extensive experimental study shows the encouraging results of the proposed text categorization framework: using Google as the web search engine, the proposed framework is able to reduce the classification error by 30% when compared with the state-of-the-art supervised text categorization method.","Title: Linguistic theories in efficient multimodal reference resolution: an empirical investigation\nYear: 2005\n\nAbstract: Multimodal conversational interfaces provide a natural means for users to communicate with computer systems through multiple modalities such as speech, gesture, and gaze. To build effective multimodal interfaces, understanding user multimodal inputs is important. Previous linguistic and cognitive studies indicate that user language behavior does not occur randomly, but rather follows certain linguistic and cognitive principles. Therefore, this paper investigates the use of linguistic theories in multimodal interpretation. In particular, we present a greedy algorithm that incorporates Conversation Implicature and Givenness Hierarchy for efficient multimodal reference resolution. Empirical studies indicate that this algorithm significantly reduces the complexity in multimodal reference resolution compared to a previous graph-matching approach. One major advantage of this greedy algorithm is that the prior linguistic and cognitive knowledge can be used to guide the search and significantly prune the search space. Because of its simplicity and generality, this approach has the potential to improve the robustness of interpretation and provide a more practical solution to multimodal input interpretation.","Title: Unsupervised Ensemble Ranking: Application to Large-Scale Image Retrieval\nYear: 2010\n\nAbstract: The continued explosion in the growth of image and video databases makes automatic image search and retrieval an extremely important problem. Among the various approaches to Content-based Image Retrieval (CBIR), image similarity based on local point descriptors has shown promising performance. However, this approach suffers from the scalability problem. Although bag-of-words model resolves the scalability problem, it suffers from loss in retrieval accuracy. We circumvent this performance loss by an ensemble ranking approach in which rankings from multiple bag-of-words models are combined to obtain more accurate retrieval results. An unsupervised algorithm is developed to learn the weights for fusing the rankings from multiple bag-of-words models. Experimental results on a database of 100,000 images show that this approach is both efficient and effective in finding visually similar images.","Title: A Random Matrix Approach to Differential Privacy and Structure Preserved Social Network Graph Publishing.\nYear: 2013\n\nAbstract: Online social networks are being increasingly used for analyzing various societal phenomena such as epidemiology, information dissemination, marketing and sentiment flow. Popular analysis techniques such as clustering and influential node analysis, require the computation of eigenvectors of the real graph's adjacency matrix. Recent de-anonymization attacks on Netflix and AOL datasets show that an open access to such graphs pose privacy threats. Among the various privacy preserving models, Differential privacy provides the strongest privacy guarantees. In this paper we propose a privacy preserving mechanism for publishing social network graph data, which satisfies differential privacy guarantees by utilizing a combination of theory of random matrix and that of differential privacy. The key idea is to project each row of an adjacency matrix to a low dimensional space using the random projection approach and then perturb the projected matrix with random noise. We show that as compared to existing approaches for differential private approximation of eigenvectors, our approach is computationally efficient, preserves the utility and satisfies differential privacy. We evaluate our approach on social network graphs of Facebook, Live Journal and Pokec. The results show that even for high values of noise variance sigma=1 the clustering quality given by normalized mutual information gain is as low as 0.74. For influential node discovery, the propose approach is able to correctly recover 80 of the most influential nodes. We also compare our results with an approach presented in [43], which directly perturbs the eigenvector of the original data by a Laplacian noise. The results show that this approach requires a large random perturbation in order to preserve the differential privacy, which leads to a poor estimation of eigenvectors for large social networks.","Title: Automatic title generation for spoken broadcast news\nYear: 2001\n\nAbstract: In this paper, we implemented a set of title generation methods using training set of 21190 news stories and evaluated them on an independent test corpus of 1006 broadcast news documents, comparing the results over manual transcription to the results over automatically recognized speech. We use both F1 and the average number of correct title words in the correct order as metric. Overall, the results show that title generation for speech recognized news documents is possible at a level approaching the accuracy of titles generated for perfect text transcriptions.","Title: Automated annotation of Drosophila gene expression patterns using a controlled vocabulary.\nYear: 2008\n\nAbstract: Regulation of gene expression in space and time directs its localization to a specific subset of cells during development. Systematic determination of the spatiotemporal dynamics of gene expression plays an important role in understanding the regulatory networks driving development. An atlas for the gene expression patterns of fruit fly Drosophila melanogaster has been created by whole-mount in situ hybridization, and it documents the dynamic changes of gene expression pattern during Drosophila embryogenesis. The spatial and temporal patterns of gene expression are integrated by anatomical terms from a controlled vocabulary linking together intermediate tissues developed from one another. Currently, the terms are assigned to patterns manually. However, the number of patterns generated by high-throughput in situ hybridization is rapidly increasing. It is, therefore, tempting to approach this problem by employing computational methods.In this article, we present a novel computational framework for annotating gene expression patterns using a controlled vocabulary. In the currently available high-throughput data, annotation terms are assigned to groups of patterns rather than to individual images. We propose to extract invariant features from images, and construct pyramid match kernels to measure the similarity between sets of patterns. To exploit the complementary information conveyed by different features and incorporate the correlation among patterns sharing common structures, we propose efficient convex formulations to integrate the kernels derived from various features. The proposed framework is evaluated by comparing its annotation with that of human curators, and promising performance in terms of F1 score has been reported.","Title: Learning nonparametric kernel matrices from pairwise constraints\nYear: 2007\n\nAbstract: Many kernel learning methods have to assume parametric forms for the target kernel functions, which significantly limits the capability of kernels in fitting diverse patterns. Some kernel learning methods assume the target kernel matrix to be a linear combination of parametric kernel matrices. This assumption again importantly limits the flexibility of the target kernel matrices. The key challenge with nonparametric kernel learning arises from the difficulty in linking the nonparametric kernels to the input patterns. In this paper, we resolve this problem by introducing the graph Laplacian of the observed data as a regularizer when optimizing the kernel matrix with respect to the pairwise constraints. We formulate the problem into Semi-Definite Programs (SDP), and propose an efficient algorithm to solve the SDP problem. The extensive evaluation on clustering with pairwise constraints shows that the proposed nonparametric kernel learning method is more effective than other state-of-the-art kernel learning techniques.","Title: First ACM SIGMM international workshop onsocial media (WSM'09)\nYear: 2009\n\nAbstract: The ACM SIGMM International Workshop on Social Media (WSM'09) is the first workshop held in conjunction with the ACM International Multimedia Conference (MM'09) at Bejing, P.R. China, 2009. This workshop provides a forum for researchers and practitioners from all over the world to share information on their latest investigations on social media analysis, exploration, search, mining, and emerging new social media applications.","Title: Combining Link and Content for Community Detection: A Discriminative Approach\nYear: 2009\n\nAbstract: In this paper, we consider the problem of combining link and content analysis for community detection from networked data, such as paper citation networks and Word Wide Web. Most existing approaches combine link and content infor- mation by a generative model that generates both links and contents via a shared set of community memberships. These generative models have some shortcomings in that they failed to consider additional factors that could affect the community memberships and isolate the contents that are irrelevant to community memberships. To explicitly ad- dress these shortcomings, we propose a discriminative model for combining the link and content analysis for community detection. First, we propose a conditional model for link analysis and in the model, we introduce hidden variables to explicitly model the popularity of nodes. Second, to allevi- ate the impact of irrelevant content attributes, we develop a discriminative model for content analysis. These two models are unified seamlessly via the community memberships. We present efficient algorithms to solve the related optimization problems based on bound optimization and alternating pro- jection. Extensive experiments with benchmark data sets show that the proposed framework significantly outperforms the state-of-the-art approaches for combining link and con- tent analysis for community detection.","Title: Video Classification and Retrieval with the Informedia Digital Video Library System\nYear: 2002\n\nAbstract: and Title) 0.0348 0.1445 0.0724Text + Movie + Image retrieval (Color + Texture) 0.0892 0.220 0.1046Text + Movie + Color + Texture + PRF Classification 0.0924 0.216 0.1124Table 1 Video Retrieval Results on the 25 queries of the 2003 TREC video track evaluation.","Title: Sparse Multiple Kernel Learning with Geometric Convergence Rate\nYear: 2013\n\nAbstract: In this paper, we study the problem of sparse multiple kernel learning (MKL), where the goal is to efficiently learn a combination of a fixed small number of kernels from a large pool that could lead to a kernel classifier with a small prediction error. We develop an efficient algorithm based on the greedy coordinate descent algorithm, that is able to achieve a geometric convergence rate under appropriate conditions. The convergence rate is achieved by measuring the size of functional gradients by an empirical $\\ell_2$ norm that depends on the empirical data distribution. This is in contrast to previous algorithms that use a functional norm to measure the size of gradients, which is independent from the data samples. We also establish a generalization error bound of the learned sparse kernel classifier using the technique of local Rademacher complexity.","Title: Title generation for spoken broadcast news using a training corpus\nYear: 2000\n\nAbstract: The problem of title generation involves finding the essence of a document and expressing it in only a few words. The results of a query to the Informedia Digital Video Library are summarized through an automatically generated title for each retrieved news story. When the document is errorful, as with speech-recognized broadcast news stories, the title creation challenge becomes even greater. We implemented a set of title word selection strategies and evaluated them on an independent test corpus of 579 broadcast news documents, comparing manual transcription results to automatically recognized speech using the CMU Sphinx speech recognition system with a 64000-word broadcast news language model. Using a training collection of 21190 transcribed broadcast news stories, we trained several systems to produce appropriate title words, i.e. Naïve Bayesian approach with full vocabulary, Naïve Bayesian approach with limited vocabulary, nearest neighbor approach and extractive approach. The F1 results shows that the nearest neighbor approach is a quick and easy way of generating good titles for speech recognized documents (F1 = 15.2%), while a Nave Bayesian approach with limited vocabulary also does well on our F1 measure (F1 = 21.6%), which ignores word order in the titles. Overall, the results show that title generation for speech recognized news documents is possible at a level approaching the accuracy of titles generated for perfect text transcriptions. One surprising phenomenon is that extractive approach performances slightly better for speech recognized documents than for manual transcripts.","Title: On the use of dynamic Bayesian networks in reconstructing functional neuronal networks from spike train ensembles.\nYear: 2010\n\nAbstract: Coordination among cortical neurons is believed to be a key element in mediating many high-level cortical processes such as perception, attention, learning, and memory formation. Inferring the structure of the neural circuitry underlying this coordination is important to characterize the highly nonlinear, time-varying interactions between cortical neurons in the presence of complex stimuli. In this work, we investigate the applicability of dynamic Bayesian networks (DBNs) in inferring the effective connectivity between spiking cortical neurons from their observed spike trains. We demonstrate that DBNs can infer the underlying nonlinear and time-varying causal interactions between these neurons and can discriminate between mono- and polysynaptic links between them under certain constraints governing their putative connectivity. We analyzed conditionally Poisson spike train data mimicking spiking activity of cortical networks of small and moderately large size. The performance was assessed and compared to other methods under systematic variations of the network structure to mimic a wide range of responses typically observed in the cortex. Results demonstrate the utility of DBN in inferring the effective connectivity in cortical networks.","Title: A compensation method to improve the performance of IPI-based entity recognition system in body sensor networks.\nYear: 2013\n\nAbstract: Security of wireless body sensor networks (BSNs) with telemedicine applications remains a crucial issue. A family of novel biometrics schemes has been recently proposed for node recognition and cryptographic key distribution without any pre-deployment in BSNs, where dynamic entity identifiers (EIs) generated from physiological signals captured by individual sensor nodes are used for nodes to recognize each other. As the recognition performance of EIs determines the maximal performance that can be achieved in such biometric systems, several kinds of EI generation schemes have been proposed. The inter-pulse intervals based EI generation scheme is more promising for such applications in actual scenarios because of its acceptable recognition performance. However, it was found that such generated EIs by true pairs of nodes, i.e. two nodes from the same BSN, have some error pattern which could be considered while doing node recognition or key distribution for an improved success rate. To address the problem, this work proposes an error-correcting code based compensation method which can be easily combined together with the key distribution process to achieve an improved recognition performance. Results of statistical analysis with experimental data collected from 14 subjects show that the bit difference between EIs from true pairs of nodes can be effectively reduced with the proposed method.","Title: Learning to rank image tags with limited training examples.\nYear: 2015\n\nAbstract: With an increasing number of images that are available in social media, image annotation has emerged as an important research topic due to its application in image matching and retrieval. Most studies cast image annotation into a multilabel classification problem. The main shortcoming of this approach is that it requires a large number of training images with clean and complete annotations in order to learn a reliable model for tag prediction. We address this limitation by developing a novel approach that combines the strength of tag ranking with the power of matrix recovery. Instead of having to make a binary decision for each tag, our approach ranks tags in the descending order of their relevance to the given image, significantly simplifying the problem. In addition, the proposed method aggregates the prediction models for different tags into a matrix, and casts tag ranking into a matrix recovery problem. It introduces the matrix trace norm to explicitly control the model complexity, so that a reliable prediction model can be learned for tag ranking even when the tag space is large and the number of training images is limited. Experiments on multiple well-known image data sets demonstrate the effectiveness of the proposed framework for tag ranking compared with the state-of-the-art approaches for image annotation and tag ranking.","Title: CUR Algorithm for Partially Observed Matrices.\nYear: 2014\n\nAbstract: CUR matrix decomposition computes the low rank approximation of a given matrix by using the actual rows and columns of the matrix. It has been a very useful tool for handling large matrices. One limitation with the existing algorithms for CUR matrix decomposition is that they cannot deal with entries in a partially observed matrix, while incomplete matrices are found in many real world applications. In this work, we alleviate this limitation by developing a CUR decomposition algorithm for partially observed matrices. In particular, the proposed algorithm computes the low rank approximation of the target matrix based on (i) the randomly sampled rows and columns, and (ii) a subset of observed entries that are randomly sampled from the matrix. Our analysis shows the error bound, measured by spectral norm, for the proposed algorithm when the target matrix is of full rank. We also show that only O(nr ln r) observed entries are needed by the proposed algorithm to perfectly recover a rank r matrix of size n × n, which improves the sample complexity of the existing algorithms for matrix completion. Empirical studies on both synthetic and real-world datasets verify our theoretical claims and demonstrate the effectiveness of the proposed algorithm.","Title: A Simple Homotopy Proximal Mapping for Compressive Sensing.\nYear: 2014\n\nAbstract: In this paper, we present a novel yet simple homotopy proximal mapping algorithm for compressive sensing. The algorithm adopts a simple proximal mapping of the $\\ell_1$ norm at each iteration and gradually reduces the regularization parameter for the $\\ell_1$ norm. We prove a global linear convergence of the proposed homotopy proximal mapping (HPM) algorithm for solving compressive sensing under three different settings (i) sparse signal recovery under noiseless measurements, (ii) sparse signal recovery under noisy measurements, and (iii) nearly-sparse signal recovery under sub-gaussian noisy measurements. In particular, we show that when the measurement matrix satisfies Restricted Isometric Properties (RIP), our theoretical results in settings (i) and (ii) almost recover the best condition on the RIP constants for compressive sensing. In addition, in setting (iii), our results for sparse signal recovery are better than the previous results, and furthermore our analysis explicitly exhibits that more observations lead to not only more accurate recovery but also faster convergence. Compared with previous studies on linear convergence for sparse signal recovery, our algorithm is simple and efficient, and our results are better and provide more insights. Finally our empirical studies provide further support for the proposed homotopy proximal mapping algorithm and verify the theoretical results.","Title: Analysis of Nuclear Norm Regularization for Full-rank Matrix Completion.\nYear: 2015\n\nAbstract: In this paper, we provide a theoretical analysis of the nuclear-norm regularized least squares for full-rank matrix completion. Although similar formulations have been examined by previous studies, their results are unsatisfactory because only additive upper bounds are provided. Under the assumption that the top eigenspaces of the target matrix are incoherent, we derive a relative upper bound for recovering the best low-rank approximation of the unknown matrix. Our relative upper bound is tighter than previous additive bounds of other methods if the mass of the target matrix is concentrated on its top eigenspaces, and also implies perfect recovery if it is low-rank. The analysis is built upon the optimality condition of the regularized formulation and existing guarantees for low-rank matrix completion. To the best of our knowledge, this is first time such a relative bound is proved for the regularized formulation of matrix completion.","Title: Random Projections for Classification: A Recovery Approach.\nYear: 2014\n\nAbstract: Random projection has been widely used in data classification. It maps high-dimensional data into a low-dimensional subspace in order to reduce the computational cost in solving the related optimization problem. While previous studies are focused on analyzing the classification performance in the low-dimensional space, in this paper, we consider the recovery problem, i.e., how to accurately recover the optimal solution to the original high-dimensional optimization problem based on the solution learned after random projection. We present a simple algorithm, termed dual random projection, which uses the dual solution of the low-dimensional optimization problem to recover the optimal solution to the original problem. Our theoretical analysis shows that with a high probability, the proposed algorithm is able to accurately recover the optimal solution to the original problem, provided that the data matrix is (approximately) low-rank and/or optimal solution is (approximately) sparse. We further show that the proposed algorithm can be applied iteratively to reducing the recovery error exponentially.","Title: A Novel Auxiliary-Free Zero Inductor Current Detection Scheme For Step Down Non-Isolated Led Driver\nYear: 2014\n\nAbstract: A novel auxiliary-free zero crossing detection (ZCD) circuit is presented in this paper, which is especially suitable for step down non-isolated LED drivers operating in boundary conduction mode (BCM). Without extra auxiliary winding, the proposed method can fast and correctly detect zero inductor current information, which only requires an integrated NMOS power switch. The step down non-isolated LED driver with the proposed ZCD scheme has been implemented in CZ6H 0.35 um standard CMOS process, results show that the proposed ZCD scheme can always work properly regardless of the variation of line voltage and load condition.","Title: Optimization Of Periphery Circuits In A 1k-Bit Pcram Chip For Highly Reliable Write And Read Operations\nYear: 2014\n\nAbstract: A 1K-bit phase change random access memory (PCRAM) with improved periphery circuits for better reliable operations has been successfully developed in 130 nm CMOS technology. A flexible write driver is proposed to provide a novel continuous step-down pulses by studying programming strategies while a reliable read circuit is designed by investigating the special transition characteristics of PCRAM, leading to an effective write operation and a non-destructive read operation without any additional changes of the storage states. In addition, a large sense margin has been achieved and the read results corresponding well with the write operations, which demonstrate the influences of technology variations have been considerably decreased with the proposed periphery circuits.","Title: A Smart Method Of Optimizing The Read/Write Current On Pcm Array\nYear: 2014\n\nAbstract: A method for optimizing the read/write current on phase change memory array is proposed. A smart current adjustment circuit is designed to bi-directionally alter the internal read/write current. The best read/write condition based on arrays can be found through this adjustment approach. Example of a 2-dimensional shmoo test on a 16 k-bit phase change array implemented in 0.13 mu m CMOS technology is given. The resistance distribution can also be roughly obtained. This method, taking advantage of the peripheral circuit, provides statistical yield data on a variety of read/write current, thus offering reliable and helpful indicators for chip parameter setup and process optimization.","Title: Efficient Algorithms for Robust One-bit Compressive Sensing.\nYear: 2014\n\nAbstract: While the conventional compressive sensing assumes measurements of infinite precision, one-bit compressive sensing considers an extreme setting where each measurement is quantized to just a single bit. In this paper, we study the vector recovery problem from noisy one-bit measurements, and develop two novel algorithms with formal theoretical guarantees. First, we propose a passive algorithm, which is very efficient in the sense it only needs to solve a convex optimization problem that has a closed-form solution. Despite the apparent simplicity, our theoretical analysis reveals that the proposed algorithm can recover both the exactly sparse and the approximately sparse vectors. In particular, for a sparse vector with s nonzero elements, the sample complexity is O(s \\log n/ε^2), where n is the dimensionality and εis the recovery error. This result improves significantly over the previously best known sample complexity in the noisy setting, which is O(s \\log n/ε^4). Second, in the case that the noise model is known, we develop an adaptive algorithm based on the principle of active learning. The key idea is to solicit the sign information only when it cannot be inferred from the current estimator. Compared with the passive algorithm, the adaptive one has a lower sample complexity if a high-precision solution is desired.","Title: A Single-Pass Algorithm for Efficiently Recovering Sparse Cluster Centers of High-dimensional Data.\nYear: 2014\n\nAbstract: Learning a statistical model for high-dimensional data is an important topic in machine learning. Although this problem has been well studied in the supervised setting, little is known about its unsupervised counterpart. In this work, we focus on the problem of clustering high-dimensional data with sparse centers. In particular, we address the following open question in unsupervised learning: “is it possible to reliably cluster high-dimensional data when the number of samples is smaller than the data dimensionality?\" We develop an efficient clustering algorithm that is able to estimate sparse cluster centers with a single pass over the data. Our theoretical analysis shows that the proposed algorithm is able to accurately recover cluster centers with only O(s\\log d) number of samples (data points), provided all the cluster centers are s-sparse vectors in a d dimensional space. Experimental results verify both the effectiveness and efficiency of the proposed clustering algorithm compared to the state-of-the-art algorithms on several benchmark datasets.","Title: Distance metric learning using dropout: a structured regularization approach\nYear: 2014\n\nAbstract: Distance metric learning (DML) aims to learn a distance metric better than Euclidean distance. It has been successfully applied to various tasks, e.g., classification, clustering and information retrieval. Many DML algorithms suffer from the over-fitting problem because of a large number of parameters to be determined in DML. In this paper, we exploit the dropout technique, which has been successfully applied in deep learning to alleviate the over-fitting problem, for DML. Different from the previous studies that only apply dropout to training data, we apply dropout to both the learned metrics and the training data. We illustrate that application of dropout to DML is essentially equivalent to matrix norm based regularization. Compared with the standard regularization scheme in DML, dropout is advantageous in simulating the structured regularizers which have shown consistently better performance than non structured regularizers. We verify, both empirically and theoretically, that dropout is effective in regulating the learned metric to avoid the over-fitting problem. Last, we examine the idea of wrapping the dropout technique in the state-of-art DML methods and observe that the dropout technique can significantly improve the performance of the original DML methods.","Title: MagPairing: Exploiting magnetometers for pairing smartphones in close proximity\nYear: 2014\n\nAbstract: With the prevalence of mobile computing, lots of wireless devices need to establish secure communication on the fly without pre-shared secrets. Device pairing is critical for bootstrapping secure communication between two previously unassociated devices over the wireless channel. Using auxiliary out-of-band channels involving visual, acoustic, tactile or vibrational sensors has been proposed as a feasible option to facilitate device pairing. However, these methods usually require users to perform additional tasks such as copying, comparing, and shaking. It is preferable to have a natural and intuitive pairing method with minimal user tasks. In this paper, we introduce a new method, called MagPairing, for pairing smartphones in close proximity by exploiting correlated magnetometer readings. In MagPairing, users only need to naturally tap the smartphones together for a few seconds without performing any additional operations in authentication and key establishment. Our method exploits the fact that smartphones are equipped with tiny magnets. Highly correlated magnetic field patterns are produced when two smartphones are close to each other. We design MagPairing protocol and implement it on Android smartphones. We conduct extensive simulation and experiments to evaluate MagPairing. Experimental results show that MagPairing can successfully pair two smartphones with 4.5 seconds on average. It is immune to man-in-the-middle attack even when the attacker is a few centimeters away from the pairing devices.","Title: A Simple Homotopy Algorithm for Compressive Sensing.\nYear: 2015\n\nAbstract: In this paper, we consider the problem of recovering the s largest elements of an arbitrary vector from noisy measurements. Inspired by previous work, we develop an homotopy algorithm which solves the l(1)-regularized least square problem for a sequence of decreasing values of the regularization parameter. Compared to the previous method, our algorithm is more efficient in the sense it only updates the solution once for each intermediate problem, and more practical in the sense it has a simple stopping criterion by checking the sparsity of the intermediate solution. Theoretical analysis reveals that our method enjoys a linear convergence rate in reducing the recovery error. Furthermore, our guarantee for recovering the top s elements of the target vector is tighter than previous results, and that for recovering the target vector itself matches the state of the art in compressive sensing.","Title: Practical secret key agreement for full-duplex near field communications\nYear: 2014\n\nAbstract: Near Field Communication (NFC) is a promising short distance radio communication technology for many useful applications. Although its communication range is short, NFC alone does not guarantee secure communication and is subject to security attacks, such as eavesdropping attack. Generating a shared key and using symmetric key cryptography to secure the communication between NFC devices is a feasible solution to prevent various attacks. However, conventional Diffie-Hellman key agreement protocol is not preferable for resource constrained NFC devices due to its extensive computational overhead and energy consumption. In this paper, we propose a practical, fast and energy-efficient key agreement scheme, which uses random bits transmission with waveform shaking, for NFC devices by exploiting its off-the-shelf full-duplex capability. In the proposed method, two devices send random bits to each other simultaneously without strict synchronization or perfect match of amplitude and phase. On the contrary, the method randomly introduces synchronization offset and mismatch of amplitude and phase for each bit transmission in order to prevent a passive attacker from determining the generated key. A shared bit can be established when two devices send different bits. We conduct theoretical analysis on the correctness and security strength of the method, and extensive simulations to evaluate its effectiveness. We build a testbed based on USRP software defined radio and conduct proof-of-concept experiments to evaluate the method in a real-world environment. It shows that the proposed method achieves a high key generation rate of about 26kbps and is immune to eavesdropping attack even when the attacker is within several centimeters from the legitimate devices. The proposed method is a practical, fast, energy-efficient, and secure key agreement scheme for resource-constrained NFC devices.","Title: Provide IPv4 Service Using Pure IPv6 Servers with Stateless NAT64 Translator\nYear: 2014\n\nAbstract: The current Internet is suffering from IPv4 address depletion. Stateless NAT64 is proposed to facilitate the IPv4/IPv6 co-existence and transition. In the past, the stateless NAT64 is mainly for IPv6 clients to access both IPv6 and IPv4 servers. As the fast growth of IP address consuming Internet services, the support for pure IPv6 servers is becoming urgent. In this paper, we propose a design of stateless NAT64 based translator with server port mapping and application layer proxy support to ensure both the IPv4 and IPv6 accessibility to pure IPv6 servers. By deploying such translators, pure IPv6 server infrastructure can be built and the transition process can be promoted.","Title: Image Tag Completion By Noisy Matrix Recovery\nYear: 2014\n\nAbstract: It is now generally recognized that user-provided image tags are incomplete and noisy. In this study, we focus on the problem of tag completion that aims to simultaneously enrich the missing tags and remove noisy tags. The novel component of the proposed framework is a noisy matrix recovery algorithm. It assumes that the observed tags are independently sampled from an unknown tag matrix and our goal is to recover the tag matrix based on the sampled tags. We show theoretically that the proposed noisy tag matrix recovery algorithm is able to simultaneously recover the missing tags and de-emphasize the noisy tags even with a limited number of observations. In addition, a graph Laplacian based component is introduced to combine the noisy matrix recovery component with visual features. Our empirical study with multiple benchmark datasets for image tagging shows that the proposed algorithm outperforms state-of-the-art approaches in terms of both effectiveness and efficiency when handling missing and noisy tags.","Title: Nystrom Approximation For Sparse Kernel Methods: Theoretical Analysis And Empirical Evaluation\nYear: 2015\n\nAbstract: Nystrom approximation is an effective approach to accelerate the computation of kernel matrices in many kernel methods. In this paper, we consider the Nystrom approximation for sparse kernel methods. Instead of relying on the low-rank assumption of the original kernels, which sometimes does not hold in some applications, we take advantage of the restricted eigenvalue condition, which has been proved to be robust for sparse kernel methods. Based on the restricted eigenvalue condition, we have provided not only the approximation bound for the original kernel matrix but also the recovery bound for the sparse solutions of sparse kernel regression. In addition to the theoretical analysis, we also demonstrate the good performance of the Nystrom approximation for sparse kernel regression on real world data sets.","Title: Online Bandit Learning For A Special Class Of Non-Convex Losses\nYear: 2015\n\nAbstract: In online bandit learning, the learner aims to minimize a sequence of losses, while only observing the value of each loss at a single point. Although various algorithms and theories have been developed for online bandit learning, most of them are limited to convex losses. In this paper, we investigate the problem of online bandit learning with non-convex losses, and develop an efficient algorithm with formal theoretical guarantees. To be specific, we consider a class of losses which is a composition of a non-increasing scalar function and a linear function. This setting models a wide range of supervised learning applications such as online classification with a non-convex loss. Theoretical analysis shows that our algorithm achieves an (O) over tilde (poly(d)T-2/3) regret bound when the variation of the loss function is small. To the best of our knowledge, this is the first work in online bandit learning that does not rely on convexity.","Title: Delay analysis of physical layer key generation in multi-user dynamic wireless networks\nYear: 2014\n\nAbstract: Secret key generation by extracting the shared randomness in wireless fading channel is a promising way to ensure wireless communication security. Previous works only consider key generation in static networks, but real-world key establishments are usually dynamic. In this work, for the first time we investigate the pairwise key generation in dynamic wireless networks with a center node (eg. access point (AP)) and random arrival users. We establish the key generation model for this kind of networks. We propose a method based on discrete Markov chain to calculate the average time a user will spend on waiting and completing the key generation (average key generation delay, AKGD). Our method can tackle both serial and parallel key generation scheduling under various conditions. We conduct extensive simulations to show the effectiveness of our model and method. The analytical and simulation results match to each other.","Title: Multiple Kernel Learning for Visual Object Recognition: A Review\nYear: 2014\n\nAbstract: Multiple kernel learning (MKL) is a principled approach for selecting and combining kernels for a given recognition task. A number of studies have shown that MKL is a useful tool for object recognition, where each image is represented by multiple sets of features and MKL is applied to combine different feature sets. We review the state-of-the-art for MKL, including different formulations and algorithms for solving the related optimization problems, with the focus on their applications to object recognition. One dilemma faced by practitioners interested in using MKL for object recognition is that different studies often provide conflicting results about the effectiveness and efficiency of MKL. To resolve this, we conduct extensive experiments on standard datasets to evaluate various approaches to MKL for object recognition. We argue that the seemingly contradictory conclusions offered by studies are due to different experimental setups. The conclusions of our study are: (i) given a sufficient number of training examples and feature/kernel types, MKL is more effective for object recognition than simple kernel combination (e.g., choosing the best performing kernel or average of kernels); and (ii) among the various approaches proposed for MKL, the sequential minimal optimization, semi-infinite programming, and level method based ones are computationally most efficient.","Title: Assessing temporal and spatial evolution of clusters of functionally interdependent neurons using graph partitioning techniques.\nYear: 2006\n\nAbstract: This paper suggests a new approach for identifying clusters of neurons with correlated spiking activity in large-size neuronal ensembles recorded with high-density microelectrode arrays. The nonparametric approach relies on mapping the neuronal spike trains to a 'scale space' using a nested multiresolution projection. Similarity measures can be arbitrarily defined in the scale space independent of the fixed bin width classically used to assess neuronal correlation. This representation allows efficient graph partitioning techniques to be used to identify clusters of correlated firing within distinct behavioral contexts. We use a new probabilistic spectral clustering algorithm that simultaneously maximizes cluster aggregation based on similarity measures. The technique is able to efficiently identify functionally interdependent neurons regardless of the temporal scale from which rate functions are typically estimated. We report the clustering performance of the algorithm applied to a synthesized neurophysiological data set and compare it to known clustering techniques to illustrate the substantial gain in the performance.","Title: A knowledge driven regression model for gene expression and microarray analysis.\nYear: 2006\n\nAbstract: The linear regression model has been widely used in the analysis of gene expression and microarray data to identify a subset of genes that are important to a given metabolic function. One of the key challenges in applying the linear regression model to gene expression data analysis arises from the sparse data problem, in which the number of genes is significantly larger than the number of conditions. To resolve this problem, we present a knowledge driven regression model that incorporates the knowledge of genes from the Gene Ontology (GO) database into the linear regression model. It is based on the assumption that two genes are likely to be assigned similar weights when they share similar sets of GO codes. Empirical studies show that the proposed knowledge driven regression model is effective in reducing the regression errors, and furthermore effective in identifying genes that are relevant to a given metabolite.","Title: Semi-supervised SVM batch mode active learning for image retrieval\nYear: 2008\n\nAbstract: Active learning has been shown as a key technique for improving content-based image retrieval (CBIR) perfor- mance. Among various methods, support vector machine (SVM) active learning is popular for its application to rel- evance feedback in CBIR. However, the regular SVM ac- tive learning has two main drawbacks when used for rele- vance feedback. First, SVM often suffers from learning with a small number of labeled examples, which is the case in relevance feedback. Second, SVM active learning usually does not take into account the redundancy among exam- ples, and therefore could select multiple examples in rele- vance feedback that are similar (or even identical) to each other. In this paper, we propose a novel scheme that ex- ploits both semi-supervised kernel learning and batch mode active learning for relevance feedback in CBIR. In partic- ular, a kernel function is first learned from a mixture of la- beled and unlabeled examples. The kernel will then be used to effectively identify the informative and diverse examples for active learning via a min-max framework. An empiri- cal study with relevance feedback of CBIR showed that the proposed scheme is significantly more effective than other state-of-the-art approaches.","Title: Speedup of fuzzy and possibilistic kernel c-means for large-scale clustering\nYear: 2011\n\nAbstract: The ubiquity of personal computing technology has produced an abundance of staggeringly large data sets-the Library of Congress has stored over 160 terabytes of web data and it is estimated that Facebook alone logs over 25 terabytes of data per day. There is a great need for systems by which one can elucidate the similarity and dissimilarity among and between groups in these data sets. Clustering is one way to find these groups. In this paper, we propose an approximation method for the fuzzy and possibilistic kernel c-means clustering algorithms. Our approximation constrains the cluster centers to be linear combinations of a size m randomly selected subset of the n input objects, where m ≪ n. The proposed algorithm only requires an m × n rectangular portion of the full n × n kernel matrix and the n diagonal values, resulting in significant memory savings. Furthermore, the computational complexity of the c-means algorithm is substantially reduced. We demonstrate that up to 3 orders of magnitude of speedup are possible while achieving almost the same performance as the original kernel c-means algorithm.","Title: Tattoo Image Matching and Retrieval\nYear: 2012\n\nAbstract: An automated tattoo image matching system achieves significantly better results for forensic and law enforcement applications than traditional keyword-based matching.","Title: Online visual vocabulary pruning using pairwise constraints\nYear: 2010\n\nAbstract: Given a pair of images represented using bag-of-visual-words and a label corresponding to whether the images are “related”(must-link constraint) or “unrelated” (cannot-link constraint), we address the problem of selecting a subset of visual words that are salient in explaining the relation between the image pair. In particular, a subset of features is selected such that the distance computed using these features satisfies the given pairwise constraints. An efficient online feature selection algorithm is presented based on the dual-gradient descent approach. Side information in the form of pair-wise constraints is incorporated into the feature selection stage, providing the user with flexibility to use an unsupervised or semi-supervised algorithm at a later stage. Correlated subsets of visual words, usually resulting from hierarchical quantization process (called groups), are exploited to select a significantly smaller vocabulary. A group-LASSO regularizer is used to drive as many feature weights to zero as possible. We evaluate the quality of the pruned vocabulary by clustering the data using the resulting feature subset. Experiments on PASCAL VOC 2007 dataset using 5000 visual keywords, resulted in around 80% reduction in the number of keywords, with little or no loss in performance.","Title: Content-based image retrieval: An application to tattoo images\nYear: 2009\n\nAbstract: Tattoo images on human body have been routinely collected and used in law enforcement to assist in suspect and victim identification. However, the current practice of matching tattoos is based on keywords. Assigning keywords to individual tattoo images is both tedious and subjective. We have developed a content-based image retrieval system for a tattoo image database. The system automatically extracts image features based on the Scale Invariant Feature Transform (SIFT). Side information, i.e., body location of tattoos and tattoo classes, is utilized to improve the retrieval time and retrieval accuracy. Geometrical constraints are also introduced in SIFT keypoint matching to reduce false retrievals. Experimental results on 1,000 queries against an operational database of 63,593 tattoo images show a rank-20 accuracy of 94.2%; the average matching time per query is 2.9 sec. on Intel Core 2, 2.66 GHz, 3 GB RAM processor.","Title: Efficient multi-label ranking for multi-class learning: Application to object recognition\nYear: 2009\n\nAbstract: Multi-label learning is useful in visual object recogni- tion when several objects are present in an image. Conven- tional approaches implement multi-label learning as a set of binary classification problems, but they suffer from im- balanced data distributions when the number of classes is large. In this paper, we address multi-label learning with many classes via a ranking approach, termed multi-label ranking. Given a test image, the proposed scheme aims to order all the object classes such that the relevant classes are ranked higher than the irrelevant ones. We present an efficient algorithm for multi-label ranking based on the idea of block coordinate descent. The proposed algorithm is ap- plied to visual object recognition. Empirical results on the PASCAL VOC 2006 and 2007 data sets show promising re- sults in comparison to the state-of-the-art algorithms for multi-label learning.","Title: A Hybrid Calibration Method for Aperture Synthesis Radiometers.\nYear: 2016\n\nAbstract: The calibration of aperture synthesis radiometers (ASRs) with large arrays by internal noise injection suffers from high increased mass and volume and is not capable to deal with the out-of-plane distortion of antenna arrays. In this letter, a hybrid calibration method is proposed, which combines the redundant baselines and reduced noise injection. The hybrid method needs only a small noise inject...","Title: Inter-functional analysis of high-throughput phenotype data by non-parametric clustering and its application to photosynthesis\nYear: 2016\n\nAbstract: Motivation: Phenomics is the study of the properties and behaviors of organisms (i.e. their phenotypes) on a high-throughput scale. New computational tools are needed to analyze complex phenomics data, which consists of multiple traits/behaviors that interact with each other and are dependent on external factors, such as genotype and environmental conditions, in a way that has not been well studied. Results: We deployed an efficient framework for partitioning complex and high dimensional phenotype data into distinct functional groups. To achieve this, we represented measured phenotype data from each genotype as a cloud-of-points, and developed a novel non-parametric clustering algorithm to cluster all the genotypes. When compared with conventional clustering approaches, the new method is advantageous in that it makes no assumption about the parametric form of the underlying data distribution and is thus particularly suitable for phenotype data analysis. We demonstrated the utility of the new clustering technique by distinguishing novel phenotypic patterns in both synthetic data and a high-throughput plant photosynthetic phenotype dataset. We biologically verified the clustering results using four Arabidopsis chloroplast mutant lines.","Title: MagPairing: Pairing Smartphones in Close Proximity Using Magnetometers.\nYear: 2016\n\nAbstract: With the prevalence of mobile computing, lots of wireless devices need to establish secure communication on the fly without pre-shared secrets. Device pairing is critical for bootstrapping secure communication between two previously unassociated devices over the wireless channel. Using auxiliary out-of-band channels involving visual, acoustic, tactile, or vibrational sensors has been proposed as a...","Title: Theory of Dual-sparse Regularized Randomized Reduction.\nYear: 2015\n\nAbstract: In this paper, we study randomized reduction methods, which reduce high-dimensional features into low-dimensional space by randomized methods (e.g., random projection, random hashing), for large-scale high-dimensional classification. Previous theoretical results on randomized reduction methods hinge on strong assumptions about the data, e.g., low rank of the data matrix or a large separable margin of classification, which hinder their applications in broad domains. To address these limitations, we propose dual-sparse regularized randomized reduction methods that introduce a sparse regularizer into the reduced dual problem. Under a mild condition that the original dual solution is a (nearly) sparse vector, we show that the resulting dual solution is close to the original dual solution and concentrates on its support set. In numerical experiments, we present an empirical study to support the analysis and we also present a novel application of the dual-sparse regularized randomized reduction methods to reducing the communication cost of distributed learning from large-scale high-dimensional data.","Title: An Explicit Sampling Dependent Spectral Error Bound for Column Subset Selection\nYear: 2015\n\nAbstract: In this paper, we consider the problem of column subset selection. We present a novel analysis of the spectral norm reconstruction for a simple randomized algorithm and establish a new bound that depends explicitly on the sampling probabilities. The sampling dependent error bound (i) allows us to better understand the tradeoff in the reconstruction error due to sampling probabilities, (ii) exhibits more insights than existing error bounds that exploit specific probability distributions, and (iii) implies better sampling distributions. In particular, we show that a sampling distribution with probabilities proportional to the square root of the statistical leverage scores is better than uniform sampling, and is better than leverage-based sampling when the statistical leverage scores are very nonuniform. And by solving a constrained optimization problem related to the error bound with an efficient bisection search we are able to achieve better performance than using either the leverage-based distribution or that proportional to the square root of the statistical leverage scores. Numerical simulations demonstrate the benefits of the new sampling distributions for low-rank matrix approximation and least square approximation compared to state-of-the art algorithms.","Title: Finding Multiple Stable Clusterings\nYear: 2015\n\nAbstract: Multi-clustering, which tries to find multiple independent ways to partition a data set into groups, has enjoyed many applications, such as customer relationship management, bioinformatics and healthcare informatics. This paper addresses two fundamental questions in multi-clustering: How to model quality of clusterings and how to find multiple stable clusterings (MSC). We introduce to multi-clustering the notion of clustering stability based on Laplacian eigengap, which was originally used by the regularized spectral learning method for similarity matrix learning. We mathematically prove that the larger the eigengap, the more stable the clustering. Furthermore, we propose a novel multi-clustering method MSC. An advantage of our method comparing to the state-of-the-art multi-clustering methods is that our method can provide users a feature subspace to understand each clustering solution. Another advantage is that MSC does not need users to specify the number of clusters and the number of alternative clusterings, which is usually difficult for users without any guidance. Our method can heuristically estimate the number of stable clusterings in a data set. We also discuss a practical way to make MSC applicable to large-scale data. We report an extensive empirical study that clearly demonstrates the effectiveness of our method.","Title: Stream Clustering: Efficient Kernel-Based Approximation Using Importance Sampling.\nYear: 2015\n\nAbstract: Stream clustering methods, which group continuous, temporally ordered dynamic data instances, have been used in a number of applications such as stock market analysis, network analysis, and cosmological analysis. Most of the popular stream clustering algorithms are linear in nature, i.e. they assume that the data is linearly separable in the input space and use measures such as the Euclidean distance to define the inter-point similarity. Though these linear clustering algorithms are efficient, they do no achieve acceptable cluster quality on real-world data. Kernel-based clustering algorithms, which use non-linear similarity measures, yield better cluster quality, but are unsuitable for clustering data streams due to their high running time and memory complexity. We propose an efficient kernel-based clustering algorithm, called the Approximate Stream Kernel k-means, which uses importance sampling to sample a subset of the data stream, and clusters the entire stream based on each data point's similarity to the sampled data points in real-time. Every time a data point is sampled, the kernel matrix representing the similarity between the sampled points is updated, and projected to a low dimensional space spanned by its top eigenvectors. The data points are clustered in this low-dimensional space using the k-means algorithm. Thus, the Approximate Stream Kernel k-means algorithm performs clustering in linear time using kernel-based similarity. We show that only a small number of points need to be sampled from the data stream, and the resulting approximation error is well-bounded. Using several large benchmark data sets to simulate data streams, we demonstrate that the proposed algorithm achieves a significant speedup over other kernel-based clustering algorithms with minimal loss in cluster quality. We also demonstrate the practical applicability of the proposed algorithm by using it to efficiently find trending topics in the Twitter stream data.","Title: Big Data Analytics: Optimization and Randomization\nYear: 2015\n\nAbstract: As the scale and dimensionality of data continue to grow in many applications of data analytics (e.g., bioinformatics, finance, computer vision, medical informatics), it becomes critical to develop efficient and effective algorithms to solve numerous machine learning and data mining problems. This tutorial will focus on simple yet practically effective techniques and algorithms for big data analytics. In the first part, we plan to present the state-of-the-art large-scale optimization algorithms, including various stochastic gradient descent methods, stochastic coordinate descent methods and distributed optimization algorithms, for solving various machine learning problems. In the second part, we will focus on randomized approximation algorithms for learning from large-scale data. We will discuss i) randomized algorithms for low-rank matrix approximation; ii) approximation techniques for solving kernel learning problems; iii) randomized reduction methods for addressing the high-dimensional challenge. Along with the description of algorithms, we will also present some empirical results to facilitate understanding of different algorithms and comparison between them.","Title: Stock Constrained Recommendation in Tmall\nYear: 2015\n\nAbstract: A large number of recommender systems have been developed to serve users with interesting news, ads, products or other contents. One main limitation with the existing work is that they do not take into account the inventory size of of items to be recommended. As a result, popular items are likely to be out of stock soon as they have been recommended and sold to many users, significantly affecting the impact of recommendation and user experience. This observation motivates us to develop a novel aware recommender system. It jointly optimizes the recommended items for all users based on both user preference and inventory sizes of different items. It requires solving a non-smooth optimization involved estimating a matrix of nxn, where n is the number of items. With the proliferation of items, this approach can quickly become computationally infeasible. We address this challenge by developing a dual method that reduces the number of variables from n^2 to n, significantly improving the computational efficiency. We also extend this approach to the online setting, which is particularly important for big promotion events. Our empirical studies based on a real benchmark data with 100 millions of user visits from Tmall verify the effectiveness of the proposed approach.","Title: SecNFC: Securing inductively-coupled Near Field Communication at physical layer\nYear: 2015\n\nAbstract: Near Field Communication (NFC) is widely used today in many useful applicatious, such as contactless payment, identification, and file exchange. Due to the limitations on computation, power, and cost of NFC devices, NFC systems are often lack of encryption or weakly encrypted, leaving them exposed to security attacks. One solution for this problem is to iustall strong cryptographic protocols on NFC devices. However, it involves upgrading and revoking deployed NFC devices, which is costly and impractical. Moreover, encryption algorithms are usually cousidered expeusive for resource constrained NFC devices in terms of computation overhead and energy cousumption. In this paper, aiming at securing NFC from eavesdropping at physical layer, we propose SecNFC, a solution that changes the signals on the initiators to hide the communicatious, but does not require any changes to the targets. In this work, for the first time, we address a practical problem of synchronization offset between two NFC terminals, which can be exploited by an eavesdropper to compromise the transmitted bits. SecNFC takes into consideration the synchronization offset and solves the challenge by introducing blocking signals around the transition point when the target switches load between on and off. With this solution, even with the existence of synchronization offset, eavesdroppers cannot extract any bit information from eavesdropping. We conduct extensive simulations to evaluate the performance of SecNFC. We also build a testbed based on USRP software defined radio and off-theshelf NFC tags to evaluate SecNFC in a real-world environment. Both simulation and experimental results show that SecNFC can efficiently prevent NFC from eavesdropping with a slight and tolerable decoding performance degradation at the initiator.","Title: Physical layer key agreement under signal injection attacks\nYear: 2015\n\nAbstract: Physical layer key agreement techniques derive a symmetric cryptographic key from the wireless fading channel between two wireless devices by exploiting channel randomness and reciprocity. Existing works mainly focus on the security analysis and protocol design of the techniques under passive attacks. The study on physical layer key agreement techniques under active attacks is largely open. In this paper, we present a new form of high threatening active attack, named signal injection attack. By injecting the similar signals to both keying devices, the attacker aims at manipulating the channel measurements and compromising a portion of the key. We further propose a countermeasure to the signal injection attack, pHY-UIR (PHYsical layer key agreement with User Introduced Randomness). In PHY-UIR, both keying devices independently introduce randomness into the channel probing frames, and compose common random series by combining the randomness in the fading channel and the ones introduced by users together. With this solution, the composed series and injected signals become uncorrelated. Thus, the final key will automatically exclude the contaminated portion related to injected signal while persisting the other portion related to random fading channel. Moreover, the contaminated composed series at two keying devices become decorrelated, which help detect the attack. We analyze the security strength of PHY-UIR and conduct extensive simulations to evaluate it. Both theoretical analysis and simulations demonstrate the effectiveness of PHy-UIR. We also perform proof-of-concept experiments by using software defined radios in a real-world environment. We show that signal injection attack is feasible in practice and leads to a strong correlation (0.75) between the injected signal and channel measurements at legitimate users for existing key generation methods. PHY-UIR is immune to the signal injection attack and results in low correlation (0.15) between the injected signal and the composed random signals at legitimate users.","Title: A reconfigurable system of routing and switching node based on ATC\nYear: 2015\n\nAbstract: Flexible Architecture of Reconfigurable Infrastructure (FARI) is a new information and communication infrastructure network. It is a kind of Software Defined Network. The provided resources by the underlying network are abstracted into Atomic Capabilities. This paper focuses on the bottom reconstruction technology in the reconfigurable routing and switching nodes of the FARI based on ATC theory. An ATC dynamically reconfigurable prototype system is designed, and a user management platform for testing is also implemented. Test results show that the system can dynamically reconfigure the underlying nodes.","Title: An Atomic Capability Chain Composition Method\nYear: 2015\n\nAbstract: FARI (Flexible Architecture of Reconfigurable Infrastructure) is proposed for a future information and communication infrastructure network. This architecture is dynamic and flexible, and can meet various requirements of future networks. Based on the atomic capability theory of FARI and the traditional I/O matching algorithm, an atomic capability chain composition method is proposed. This method can combine a series of atomic capabilities to an atomic capability chain, according to special business requirement. And it can automatically compose some ACs with independent function characteristics into a chain which satisfies the business requirement. Traditional I/O matching algorithm just specifies the matching between input and output parameters. Our improved I/O matching algorithm specifies not only the matching of I/O parameters, but also the matching of metadata which is specific to ACs. At last, an example is provided to illustrate how this method works, and a simulation is given to show the efficiency of this method.","Title: Lower and Upper Bounds on the Generalization of Stochastic Exponentially Concave Optimization\nYear: 2015\n\nAbstract: In this paper we derive high probability lower and upper bounds on the excess risk of stochastic optimization of exponentially concave loss functions. Exponentially concave loss functions encompass several fundamental problems in machine learning such as squared loss in linear regression, logistic loss in classification, and negative logarithm loss in portfolio management. We demonstrate an O(d log T/T ) upper bound on the excess risk of stochastic online Newton step algorithm, and an O(d/T ) lower bound on the excess risk of any stochastic optimization method for squared loss, indicating that the obtained upper bound is optimal up to a logarithmic factor. The analysis of upper bound is based on recent advances in concentration inequalities for bounding self-normalized martingales, which is interesting by its own right, and the proof technique used to achieve the lower bound is a probabilistic method and relies on an information-theoretic minimax analysis.","Title: Efficient Sparse Recovery Via Adaptive Non-Convex Regularizers With Oracle Property\nYear: 2014\n\nAbstract: The main shortcoming of sparse recovery with a convex regularizer is that it is a biased estimator and therefore will result in a suboptimal performance in many cases. Recent studies have shown, both theoretically and empirically, that non-convex regularizer is able to overcome the biased estimation problem. Although multiple algorithms have been developed for sparse recovery with non-convex regularization, they are either computationally demanding or not equipped with the desired properties (i.e. optimal recovery error, selection consistency and oracle property). In this work, we develop an algorithm for efficient sparse recovery based on proximal gradient descent. The key feature of the proposed algorithm is introducing adaptive non-convex regularizers whose shrinking threshold vary over iterations. The algorithm is compatible with most popular non-convex regularizers, achieves a geometric convergence rate for the recovery error, is selection consistent, and most importantly has the oracle property. Based on the proposed framework, we suggest to use a so-called ACCQ regularizer, which is equivalent to zero proximal projection gap adaptive hard-thresholding. Experiments with both synthetic data sets and real images verify both the efficiency and effectiveness of the proposed method compared to the state-of-the-art methods for sparse recovery.","Title: Topic Modeling in Semantic Space with Keywords\nYear: 2015\n\nAbstract: A common and convenient approach for user to describe his information need is to provide a set of keywords. Therefore, the technique to understand the need becomes crucial. In this paper, for the information need about a topic or category, we propose a novel method called TDCS(Topic Distilling with Compressive Sensing) for explicit and accurate modeling the topic implied by several keywords. The task is transformed as a topic reconstruction problem in the semantic space with a reasonable intuition that the topic is sparse in the semantic space. The latent semantic space could be mined from documents via unsupervised methods, e.g. LSI. Compressive sensing is leveraged to obtain a sparse representation from only a few keywords. In order to make the distilled topic more robust, an iterative learning approach is adopted. The experiment results show the effectiveness of our method. Moreover, with only a few semantic concepts remained for the topic, our method is efficient for subsequent text mining tasks.","Title: Sparse Kernel Clustering of Massive High-Dimensional Data sets with Large Number of Clusters\nYear: 2015\n\nAbstract: In clustering applications involving documents and images, in addition to the large number of data points (N) and their high dimensionality (d), the number of clusters (C) into which the data need to be partitioned is also large. Kernel-based clustering algorithms, which have been shown to perform better than linear clustering algorithms, have high running time complexity in terms of N, d and C. We propose an efficient sparse kernel k-means clustering algorithm, which incrementally samples the most informative points from the data set using importance sampling, and constructs a sparse kernel matrix using these sampled points. Each row in this matrix corresponds to a data point's similarity with its p-nearest neighbors among the sampled points (p -- N). This sparse kernel matrix is used to perform clustering and obtain the cluster labels. This combination of sampling and sparsity reduces both the running time and memory complexity of kernel clustering. In order to further enhance its efficiency, the proposed algorithm projects the data on to the top C eigenvectors of the sparse kernel matrix and clusters these eigenvectors using a modified k-means algorithm. The running time of the proposed sparse kernel k-means algorithm is linear in N and d, and logarithmic in C. We show analytically that only a small number of points need to be sampled from the data set, and the resulting approximation error is well-bounded. We demonstrate, using several large high-dimensional text and image data sets, that the proposed algorithm is significantly faster than classical kernel-based clustering algorithms, while maintaining clustering quality.","Title: Accelerated Sparse Linear Regression via Random Projection.\nYear: 2016\n\nAbstract: In this paper, we present an accelerated numerical method based on random projection for sparse linear regression. Previous studies have shown that under appropriate conditions, gradient-based methods enjoy a geometric convergence rate when applied to this problem. However, the time complexity of evaluating the gradient is as large as O(nd), where n is the number of data points and d is the dimensionality, making those methods inefficient for large-scale and high-dimensional dataset. To address this limitation, we first utilize random projection to find a rank-k approximator for the data matrix, and reduce the cost of gradient evaluation to O(nk + dk), a significant improvement when k is much smaller than d and n. Then, we solve the sparse linear regression problem via a proximal gradient method with a homotopy strategy to generate sparse intermediate solutions. Theoretical analysis shows that our method also achieves a global geometric convergence rate, and moreover the sparsity of all the intermediate solutions are well-bounded over the iterations. Finally, we conduct experiments to demonstrate the efficiency of the proposed method.","Title: Stochastic Optimization for Kernel PCA.\nYear: 2016\n\nAbstract: Kernel Principal Component Analysis (PCA) is a popular extension of PCA which is able to find nonlinear patterns from data. However, the application of kernel PCA to large-scale problems remains a big challenge, due to its quadratic space complexity and cubic time complexity in the number of examples. To address this limitation, we utilize techniques from stochastic optimization to solve kernel PCA with linear space and time complexities per iteration. Specifically, we formulate it as a stochastic composite optimization problem, where a nuclear norm regularizer is introduced to promote low-rankness, and then develop a simple algorithm based on stochastic proximal gradient descent. During the optimization process, the proposed algorithm always maintains a low-rank factorization of iterates that can be conveniently held in memory. Compared to previous iterative approaches, a remarkable property of our algorithm is that it is equipped with an explicit rate of convergence. Theoretical analysis shows that the solution of our algorithm converges to the optimal one at an O(1/T) rate, where T is the number of iterations.","Title: Tracking Slowly Moving Clairvoyant: Optimal Dynamic Regret of Online Learning with True and Noisy Gradient.\nYear: 2016\n\nAbstract: This work focuses on dynamic regret of online convex optimization that compares the performance of online learning to a clairvoyant who knows the sequence of loss functions in advance and hence selects the minimizer of the loss function at each step. By assuming that the clairvoyant moves slowly (i.e., the minimizers change slowly), we present several improved variation-based upper bounds of the dynamic regret under the true and noisy gradient feedback, which are optimal in light of the presented lower bounds. The key to our analysis is to explore a regularity metric that measures the temporal changes in the clairvoyantu0027s minimizers, to which we refer as path variation. Firstly, we present a general lower bound in terms of the path variation, and then show that under full information or gradient feedback we are able to achieve an optimal dynamic regret. Secondly, we present a lower bound with noisy gradient feedback and then show that we can achieve optimal dynamic regrets under a stochastic gradient feedback and two-point bandit feedback. Moreover, for a sequence of smooth loss functions that admit a small variation in the gradients, our dynamic regret under the two-point bandit feedback matches what is achieved with full information.","Title: Social Graph Publishing with Privacy Guarantees\nYear: 2016\n\nAbstract: Online social network graphs provide useful insights on various social phenomena such as information dissemination and epidemiology. Unfortunately, social network providers often refuse to publish their social network graphs due to privacy concerns. Recently, differential privacy has become the widely accepted criteria for privacy preserving data publishing because it provides strongest privacy guarantees for publishing sensitive datasets. Although some work has been done on publishing matrices with differential privacy, they are computationally unpractical as they are not designed to handle large matrices such as the adjacency matrices of OSN graphs. In this paper, we propose a random matrix approach to OSN graph publishing, which achieves storage and computational efficiency by reducing the dimensions of adjacency matrices and achieves differential privacy by adding a small amount of noise. Our key idea is to first project each row of an adjacency matrix into a low dimensional space using random projection, and then perturb the projected matrix with random noise, and finally publish the perturbed and projected matrix. In this paper, we first prove that random projection plus random perturbation preserve differential privacy, and also that the random noise required to achieve differential privacy is small. We then validate the proposed approach and evaluate the utility of the published data for two different applications, namely node clustering and node ranking, using publicly available OSN graphs of Facebook, Live Journal, and Pokec.","Title: Large-scale Robust Online Matching and Its Application in E-commerce\nYear: 2016\n\nAbstract: This talk will be focused on large-scale matching problem that aims to find the optimal assignment of tasks to different agents under linear constraints. Large-scale matching has found numerous applications in e-commerce. An well known example is budget aware online advertisement. A common practice in online advertisement is to find, for each opportunity or user, the advertisements that fit best with his/her interests. The main shortcoming with this greedy approach is that it did not take into account the budget limits set by advertisers. Our studies, as well as others, have shown that by carefully taking into budget limits of individual advertisers, we could significantly improve the performance of the advertisement system. Despite of rich literature, two important issues are often overlooked in the previous studies of matching/assignment problem. The first issues arises from the fact that most quantities used by optimization are estimated based on historical data and therefore are likely to be inaccurate and unreliable. The second challenge is how to perform online matching as in many e-commerce problems, tasks are created in an online fashion and algorithm has to make assignment decision immediately when every task emerges. We refer to these two issues as challenges of “robust matching” and “online matching”. To address the first challenge, I will introduce two different techniques for robust matching. The first approach is based on the theory of robust optimization that takes into account the uncertainties of estimated quantities when performing optimization. The second approach is based on the theory of two-sided matching whose result only depends on the partial preference of estimated quantities. To deal with the challenge of online matching, I will discuss two online optimization techniques, one based on theory of primal-dual online optimization and one based on minimizing dynamic regret under long term constraints. We verify the effectiveness of all these approaches by applying them to real-world projects developed in Alibaba.","Title: Array Configuration Optimization For Rotating Mirrored Aperture Synthesis (Rmas)\nYear: 2016\n\nAbstract: Aperture synthesis with high spatial resolution would seriously increase the hardware requirements and system complexity. To reduce the system complexity, Rotating mirrored aperture synthesis (RMAS) that combines the antenna array rotation with mirrored aperture synthesis was proposed to reduce the number of antennas. In this paper, the sieving method is presented to optimize the array configuration for RMAS. The initial simulation results demonstrate the validity of the proposed method and the optimized array of RMAS can reconstruct the brightness temperature of the observed scene with fewer antennas.","Title: Correction Of Smos Data In Coastal Area Of South China Sea Based On Land Contamination Analysis\nYear: 2016\n\nAbstract: For SMOS data, significant errors exist in coastal areas because of the contamination by the nearby land. In this paper, the analysis of SMOS brightness temperature (TB) data in coastal area of South China Sea (SCS) reveals a decreasing trend with the increase of the distance to coast. A SMOS TB measurement model according to geophysical parameters and SMOS/MIRAS antenna array is established. The simulation shows that TB is large within 40 km since land is much warmer than ocean and is observed by the mainlobe of the antenna array. Then the simulation result of land contamination is used for correction of SMOS measured TB data in coastal area of SCS. Then the corrected TB is applied for sea surface salinity (SSS) retrieval, which reduces SSS outliers and improves the accuracy of SSS in coastal area of SCS.","Title: Strongly Adaptive Regret Implies Optimally Dynamic Regret.\nYear: 2017\n\nAbstract: To cope with changing environments, recent developments in online learning have introduced the concepts of adaptive regret and dynamic regret independently. In this paper, we illustrate an intrinsic connection between these two concepts by showing that the dynamic regret can be expressed in terms of the adaptive regret and the functional variation. This observation implies that strongly adaptive algorithms can be directly leveraged to minimize the dynamic regret. As a result, we present a series of strongly adaptive algorithms whose dynamic regrets are minimax optimal for convex functions, exponentially concave functions, and strongly convex functions, respectively. To the best of our knowledge, this is the first time that such kind of dynamic regret bound is established for exponentially concave functions. Moreover, all of those adaptive algorithms do not need any prior knowledge of the functional variation, which is a significant advantage over previous specialized methods for minimizing dynamic regret.","Title: Sparse Learning with Stochastic Composite Optimization.\nYear: 2017\n\nAbstract: In this paper, we study Stochastic Composite Optimization (SCO) for sparse learning that aims to learn a sparse solution from a composite function. Most of the recent SCO algorithms have already reached the optimal expected convergence rate $\\\\mathcal {O}(1/\\\\lambda T)$ , but they often fail to deliver sparse solutions at the end either due to the limited sparsity regularization during stochastic optimization (SO) or due to the limitation in online-to-batch conversion. Even when the objective function is strongly convex, their high probability bounds can only attain $\\\\mathcal {O}(\\\\sqrt{\\\\log (1/\\\\delta)/T})$ with $\\\\delta$ is the failure probability, which is much worse than the expected convergence rate. To address these limitations, we propose a simple yet effective two-phase Stochastic Composite Optimization scheme by adding a novel powerful sparse online-to-batch conversion to the general Stochastic Optimization algorithms. We further develop three concrete algorithms, OptimalSL, LastSL and AverageSL, directly under our scheme to prove the effectiveness of the proposed scheme. Both the theoretical analysis and the experiment results show that our methods can really outperform the existing methods at the ability of sparse learning and at the meantime we can improve the high probability bound to approximately $\\\\mathcal {O}(\\\\log (\\\\log (T)/\\\\delta)/\\\\lambda T)$ .","Title: Experiences with Multi-modal Collaborative Virtual Laboratory (MMCVL).\nYear: 2017\n\nAbstract: Training through Multi-modal Collaborative Virtual Laboratory (MMCVL) aims at developing or improving skills of students without the need of them being physically present. MMCVL addresses the problems faced by the current education system such as lack of resources and safe use of expensive laboratory equipment. It also provides a means for easy collaboration between students and teachers from different universities and schools. Multimedia data in terms of 3D mesh, skeleton, audio and virtual object physics interaction are incorporated in MMCVL to provide a 3D Tele-Immersion environment. The system uses low-cost RGB-D cameras such as Microsoft Kinect V2 to capture and generate 3D model of the person by extracting him/her from the entire captured data and immersing it in different interactive virtual environments. Different lab environments can be virtually created to provide for and replace the real laboratories. Object interactions allow for users to interact with the virtual objects present in the scene similar to real lab setup. Recording module of the toolkit allows future reference for training and testing of the student. The system also facilitates virtual feedback by giving the correct measurements, video demonstrations as well as audio communication with the collaborating teacher. Two different Chemistry experiments were performed to evaluate the user experience and usability aspects of MMCVL. Results obtained from the user study analysis prove the system to be fun and realistic, and at the same time engaging and motivating for learning laboratory experiments.","Title: Empirical Risk Minimization for Stochastic Convex Optimization: $O(1/n)$- and $O(1/n^2)$-type of Risk Bounds.\nYear: 2017\n\nAbstract: Although there exist plentiful theories of empirical risk minimization (ERM) for supervised learning, current theoretical understandings of ERM for a related problem---stochastic convex optimization (SCO), are limited. In this work, we strengthen the realm of ERM for SCO by exploiting smoothness and strong convexity conditions to improve the risk bounds. First, we establish an $widetilde{O}(d/n + sqrt{F_*/n})$ risk bound when the random function is nonnegative, convex and smooth, and the expected function is Lipschitz continuous, where $d$ is the dimensionality of the problem, $n$ is the number of samples, and $F_*$ is the minimal risk. Thus, when $F_*$ is small we obtain an $widetilde{O}(d/n)$ risk bound, which is analogous to the $widetilde{O}(1/n)$ optimistic rate of ERM for supervised learning. Second, if the objective function is also $lambda$-strongly convex, we prove an $widetilde{O}(d/n + kappa F_*/n )$ risk bound where $kappa$ is the condition number, and improve it to $O(1/[lambda n^2] + kappa F_*/n)$ when $n=widetilde{Omega}(kappa d)$. As a result, we obtain an $O(kappa/n^2)$ risk bound under the condition that $n$ is large and $F_*$ is small, which to the best of our knowledge, is the first $O(1/n^2)$-type of risk bound of ERM. Third, we stress that the above results are established in a unified framework, which allows us to derive new risk bounds under weaker conditions, e.g., without convexity of the random function and Lipschitz continuity of the expected function. Finally, we demonstrate that to achieve an $O(1/[lambda n^2] + kappa F_*/n)$ risk bound for supervised learning, the $widetilde{Omega}(kappa d)$ requirement on $n$ can be replaced with $Omega(kappa^2)$, which is dimensionality-independent.","Title: A Two-Stage Approach for Learning a Sparse Model with Sharp Excess Risk Analysis.\nYear: 2017\n\nAbstract: This paper aims to provide a sharp excess risk guarantee for learning a sparse linear model without any assumptions about the strong convexity of the expected loss and the sparsity of the optimal solution in hindsight. Given a target level for the excess risk, an interesting question to ask is how many examples and how large the support set of the solution are enough for learning a good model with the target excess risk. To answer these questions, we present a two-stage algorithm that (i) in the first stage an epoch based stochastic optimization algorithm is exploited with an established O(1/epsilon) bound on the sample complexity; and (ii) in the second stage a distribution dependent randomized sparsification is presented with an O(1/epsilon) bound on the sparsity (referred to as support complexity) of the resulting model. Compared to previous works, our contributions lie at (i) we reduce the order of the sample complexity from O(1/epsilon(2)) to O(1/epsilon) without the strong convexity assumption; and (ii) we reduce the constant in O(1/epsilon) for the sparsity by exploring the distribution dependent sampling.","Title: Extremely Low Bit Neural Network: Squeeze the Last Bit Out with ADMM.\nYear: 2018\n\nAbstract: Although deep learning models are highly effective for various learning tasks, their high computational costs prohibit the deployment to scenarios where either memory or computational resources are limited. In this paper, we focus on compressing and accelerating deep models with network weights represented by very small numbers of bits, referred to as extremely low bit neural network. We model this problem as a discretely constrained optimization problem. Borrowing the idea from Alternating Direction Method of Multipliers (ADMM), we decouple the continuous parameters from the discrete constraints of network, and cast the original hard problem into several subproblems. We propose to solve these subproblems using extragradient and iterative quantization algorithms that lead to considerably faster convergency compared to conventional optimization methods. Extensive experiments on image recognition and object detection verify that the proposed algorithm is more effective than state-of-the-art approaches when coming to extremely low bit neural network.","Title: QoE Studies on Interactive 3D Tele-Immersion\nYear: 2017\n\nAbstract: Users' Quality of Experience (QoE) in Interactive 3D Tele-Immersion (i3DTI) systems is influenced by several factors such as the quality of the \"live\" 3D avatars of the users, network latency, rendering methodology (head mounted display or regular TV type of display), etc. Hence, it becomes important to answer the question: \"Is Visual Quality (VQ) the only factor to be considered or do better immersion and faster interactions matter, in having good QoE?\" To answer this question, in this paper, a highly optimized state-of-the-art i3DTI framework implementation is introduced along with a soccer-inspired penalty shootout game. This game allows users to experience various situations, view different angles, perceive delays, track virtual ball motion, and play naturally using their entire bodies. A head mounted display device - Oculus Rift allows the users to get completely immersed and perform better in the penalty shootout game compared to watching themselves play on a 3D TV. This scenario is obtained in a controlled lab setting with ultra-high-speed network that has ultra-low latency, high VQ, fast and realistic interactions. Such ideal conditions are not typically available in a wide area Internet. Hence, for faster interaction scenario, we used lesser RGB-D cameras for 3D reconstruction, thereby reducing the model quality significantly. The high responsiveness of the game masked the user's perception of quality, and resulted in them not noticing the lower VQ of the reconstructed 3D models. Based on the results from the user study that focused on immersion, interaction and VQ aspects; users felt the game to be visually appealing, intuitive, engaging, and highly entertaining in all of the scenarios.","Title: The Opensesame Nist 2016 Speaker Recognition Evaluation System\nYear: 2017\n\nAbstract: Last two decades have witnessed a significant progress in speaker recognition, as evidenced by the improving performance in the speaker recognition evaluations (SRE) hosted by NIST. Despite the progress, only a few research is focused on speaker recognition with short duration and language mismatch condition, which often leads to poor recognition performance. In NIST SRE2016, these concerns were first systematically investigated by the speaker recognition community. In this study, we address these challenges from the viewpoint of feature extraction and modeling. In particular, we improve the robustness of features by combining GMM and DNN based iVector extraction approaches, and improve the reliability of the back-end model by exploiting symmetric SVM that can effectively leverage the unlabeled data. Finally, we introduce distance metric learning to improve the generalization capacity of the development data that is usually in limited size. Then a fusion strategy is adopted to collectively boost the performance. The effectiveness of the proposed scheme for speaker recognition is demonstrated on SRE2016 evaluation data: compared with DNN-iVector PLDA baseline system, our method yields 25.6% relative improvement in terms of min_Cprimary.","Title: Learndna: An Interactive Vr Application For Learning Dna Structure\nYear: 2018\n\nAbstract: Understanding complex concepts is known to be difficult and tedious. This complexity is very often due to lack of visual representation of abstract concepts, that cannot be easily observed. Representing these ideas in a visual Virtual Environment (VE) is a very important application of Virtual Reality (VR). This paper presents an immersive VR application for teaching students about one such complex concept of DNA structure. Building virtual environments that show the double helix structure of DNA and allowing students to interact with it, provides an illustrative and engaging experience. A crucial factor that determines usability of such systems is the correlation between actual concept and its virtual representation. Our virtual DNA structure is very similar to the actual structure, thus highlighting its complexity. This VE is built for Oculus Rift with simple clicker based manipulation and travel, to allow easy portability to other VR devices. The system also provides test based assessment to gauge students' understanding. The evaluation of this VE was carried out with the help from a group of Subject Matter Experts (SMEs). Through their feedback, we refined system design and subject content, as well as survey questionnaires. A subsequent user study was conducted in a local community college. Students were recruited from two biology classes and offered extra credit based on results recorded by this system. The study gave pre-test prior to virtual experiment, post-test following virtual experiment and survey session after post-test.All SMEs felt that the application was useful for learning about DNA. Comparing the pre-test and post-test performance of the students, there was an average of 40% improvement in the scores, indicating that LearnDNA was able to facilitate learning the concepts. Most students also felt that the application was engaging and fun to use.","Title: Physical layer multi-user key generation in wireless networks.\nYear: 2018\n\nAbstract: Secret key generation by extracting the shared randomness in the wireless fading channel from physical layer is an interesting topic of practical value. Previous works have focused on the study of physical layer key generation with two nodes from the view point of key generation rate (KGR). Information theoretic limits and the KGRs in implementation have been derived. However, in real-world applications, the physical layer key generation problem involving multiple nodes is the common case, which lacks sufficient study so far. Multi-node case differs from two-node case in that there are two more important considerations: (1) the trade-off between KGR and probing efficiency at individual node pair; (2) channel probing schedule among multiple node pairs. This paper aims at minimizing the (shorten as OWT) through the optimization of probing rates at individual node pair and channel probing schedule. The theoretical lower bound of OWT is derived first, then a practical method (MUKEM) is proposed to compute reasonable probing rates and channel probing schedule for multiple node pairs to obtain a short OWT. Simulations are conducted to evaluate the effectiveness of our method. The results show that 70 % of OWT can be reduced by using our method comparing with one-by-one key generations; while it is only about 8 % longer than the lower bound of OWT.","Title: Learning with Non-Convex Truncated Losses by SGD.\nYear: 2018\n\nAbstract: Learning with a {it convex loss} function has been a dominating paradigm for many years. It remains an interesting question how non-convex loss functions help improve the generalization of learning with broad applicability. In this paper, we study a family of objective functions formed by truncating traditional loss functions, which is applicable to both shallow learning and deep learning. Truncating loss functions has potential to be less vulnerable and more robust to large noise in observations that could be adversarial. More importantly, it is a generic technique without assuming the knowledge of noise distribution. To justify non-convex learning with truncated losses, we establish excess risk bounds of empirical risk minimization based on truncated losses for heavy-tailed output, and statistical error of an approximate stationary point found by stochastic gradient descent (SGD) method. Our experiments for shallow and deep learning for regression with outliers, corrupted data and heavy-tailed noise further justify the proposed method.","Title: Robust Optimization over Multiple Domains\nYear: 2018\n\nAbstract: In this work, we study the problem of learning a single model for multiple domains. Unlike the conventional machine learning scenario where each domain can have the corresponding model, multiple domains (i.e., applications/users) may share the same machine learning model due to maintenance loads in cloud computing services. For example, a digit-recognition model should be applicable to hand-written digits, house numbers, car plates, etc. Therefore, an ideal model for cloud computing has to perform well at each applicable domain. To address this new challenge from cloud computing, we develop a framework of robust optimization over multiple domains. In lieu of minimizing the empirical risk, we aim to learn a model optimized to the adversarial distribution over multiple domains. Hence, we propose to learn the model and the adversarial distribution simultaneously with the stochastic algorithm for efficiency. Theoretically, we analyze the convergence rate for convex and non-convex models. To our best knowledge, we first study the convergence rate of learning a robust non-convex model with a practical algorithm. Furthermore, we demonstrate that the robustness of the framework and the convergence rate can be further enhanced by appropriate regularizers over the adversarial distribution. The empirical study on real-world fine-grained visual categorization and digits recognition tasks verifies the effectiveness and efficiency of the proposed framework.","Title: Multinomial Logit Bandit with Linear Utility Functions.\nYear: 2018\n\nAbstract: Multinomial logit bandit is a sequential subset selection problem which arises in many applications. In each round, the player selects a $K$-cardinality subset from $N$ candidate items, and receives a reward which is governed by a {it multinomial logit} (MNL) choice model considering both item utility and substitution property among items. The playeru0027s objective is to dynamically learn the parameters of MNL model and maximize cumulative reward over a finite horizon $T$. This problem faces the exploration-exploitation dilemma, and the involved combinatorial nature makes it non-trivial. In recent years, there have developed some algorithms by exploiting specific characteristics of the MNL model, but all of them estimate the parameters of MNL model separately and incur a regret no better than $tilde{O}big(sqrt{NT}big)$ which is not preferred for large candidate set size $N$. In this paper, we consider the {it linear utility} MNL choice model whose item utilities are represented as linear functions of $d$-dimension item features, and propose an algorithm, titled {bf LUMB}, to exploit the underlying structure. It is proven that the proposed algorithm achieves $tilde{O}big(dKsqrt{T}big)$ regret which is free of candidate set size. Experiments show the superiority of the proposed algorithm.","Title: Detecting communities and their evolutions in dynamic social networks—a Bayesian approach\nYear: 2011\n\nAbstract: Although a large body of work is devoted to finding communities in static social networks, only a few studies examined the dynamics of communities in evolving social networks. In this paper, we propose a dynamic stochastic block model for finding communities and their evolution in a dynamic social network. The proposed model captures the evolution of communities by explicitly modeling the transition of community memberships for individual nodes in the network. Unlike many existing approaches for modeling social networks that estimate parameters by their most likely values (i.e., point estimation), in this study, we employ a Bayesian treatment for parameter estimation that computes the posterior distributions for all the unknown parameters. This Bayesian treatment allows us to capture the uncertainty in parameter values and therefore is more robust to data noise than point estimation. In addition, an efficient algorithm is developed for Bayesian inference to handle large sparse social networks. Extensive experimental studies based on both synthetic data and real-life data demonstrate that our model achieves higher accuracy and reveals more insights in the data than several state-of-the-art algorithms.","Title: Visual Search at Alibaba.\nYear: 2018\n\nAbstract: This paper introduces the large scale visual search algorithm and system infrastructure at Alibaba. The following challenges are discussed under the E-commercial circumstance at Alibaba (a) how to handle heterogeneous image data and bridge the gap between real-shot images from user query and the online images. (b) how to deal with large scale indexing for massive updating data. (c) how to train deep models for effective feature representation without huge human annotations. (d) how to improve the user engagement by considering the quality of the content. We take advantage of large image collection of Alibaba and state-of-the-art deep learning techniques to perform visual search at scale. We present solutions and implementation details to overcome those problems and also share our learnings from building such a large scale commercial visual search engine. Specifically, model and search-based fusion approach is introduced to effectively predict categories. Also, we propose a deep CNN model for joint detection and feature learning by mining user click behavior. The binary index engine is designed to scale up indexing without compromising recall and precision. Finally, we apply all the stages into an end-to-end system architecture, which can simultaneously achieve highly efficient and scalable performance adapting to real-shot images. Extensive experiments demonstrate the advancement of each module in our system. We hope visual search at Alibaba becomes more widely incorporated into today's commercial applications.","Title: Dynamic Regret of Strongly Adaptive Methods.\nYear: 2018\n\nAbstract: To cope with changing environments, recent developments in online learning have introduced the concepts of adaptive regret and dynamic regret independently. In this paper, we illustrate an intrinsic connection between these two concepts by showing that the dynamic regret can be expressed in terms of the adaptive regret and the functional variation. This observation implies that strongly adaptive algorithms can be directly leveraged to minimize the dynamic regret. As a result, we present a series of strongly adaptive algorithms that have small dynamic regrets for convex functions, exponentially concave functions, and strongly convex functions, respectively. To the best of our knowledge, this is the first time that exponential concavity is utilized to upper bound the dynamic regret. Moreover, all of those adaptive algorithms do not need any prior knowledge of the functional variation, which is a significant advantage over previous specialized methods for minimizing dynamic regret.","Title: Secure Inductive-Coupled Near Field Communication at Physical Layer.\nYear: 2018\n\nAbstract: Near field communication (NFC) is widely used today in many useful applications, such as contactless payment, identification, and file exchange. Due to the limitations on computation, power, and cost of NFC devices, NFC systems often lack encryption or are weakly encrypted, leaving them exposed to security attacks. One solution for this problem is to install strong cryptographic protocols on NFC d...","Title: A Congestion Control Method Of Sdn Data Center Based On Reinforcement Learning\nYear: 2018\n\nAbstract: With the development of cloud computing and big data, the internal communication business in data center has increased dramatically, and then the traffic in data center has also significantly increased. The bandwidth of data center is difficult to meet the bandwidth requirements of those intensive applications, and data center is facing a risk of network congestion. Under the background of the development of network intelligence, software-defined network (SDN) should demonstrate its intelligence as a future network architecture. In this paper, we introduce reinforcement learning into the SDN data center to implement congestion control based on flow. We improve the Q-learning and Sarsa algorithms and propose two methods of congestion control based on the algorithms. Test results show that these two congestion control methods can control congestion effectively. And Sarsa method has a better performance of link utilization. The average link utilization of the Sarsa method is 2.4% higher than the Q-learning method and is 4.48% higher than the on-demand method.","Title: Balancing the QOS and Security in Dijkstra Algorithm by SDN Technology.\nYear: 2018\n\nAbstract: Dijkstra algorithm is widely used in a lot of common network routing protocols. We consider the problem of quality of service (QoS) and the Security features of the network routing area using software defined networks (SDN). The SDN framework enables an efficient decoupled implementation of dynamic routing protocols which could aware the communication network status. In this work we consider the varying delay status of the communication network along with other network security parameters. The routing problem is formulated as a multi-constrained shortest path problem. A new improved Dijkstra algorithm is presented named as QS-Dijkstra. The implement and experiment show that QS-Dijkstra algorithm is able to minimize traffic routing through vulnerable links while satisfying the QoS constraints of the network.","Title: A simple homotopy proximal mapping algorithm for compressive sensing\nYear: 2019\n\nAbstract: In this paper, we present novel yet simple homotopy proximal mapping algorithms for reconstructing a sparse signal from (noisy) linear measurements of the signal or for learning a sparse linear model from observed data, where the former task is well-known in the field of compressive sensing and the latter task is known as model selection in statistics and machine learning. The algorithms adopt a simple proximal mapping of the \\(\\ell _1\\) norm at each iteration and gradually reduces the regularization parameter for the \\(\\ell _1\\) norm. We prove a global linear convergence of the proposed homotopy proximal mapping (HPM) algorithms for recovering the sparse signal under three different settings (i) sparse signal recovery under noiseless measurements, (ii) sparse signal recovery under noisy measurements, and (iii) nearly-sparse signal recovery under sub-Gaussian noisy measurements. In particular, we show that when the measurement matrix satisfies restricted isometric properties (RIP), one of the proposed algorithms with an appropriate setting of a parameter based on the RIP constants converges linearly to the optimal solution up to the noise level. In addition, in setting (iii), a practical variant of the proposed algorithms does not rely on the RIP constants and our results for sparse signal recovery are better than the previous results in the sense that our recovery error bound is smaller. Furthermore, our analysis explicitly exhibits that more observations lead to not only more accurate recovery but also faster convergence. Finally our empirical studies provide further support for the proposed homotopy proximal mapping algorithm and verify the theoretical results.","Title: Semi-Supervised Boosting for Multi-Class Classification\nYear: 2008\n\nAbstract: Most semi-supervised learning algorithms have been designed for binary classification, and are extended to multi-class classification by approaches such as one-against-the-rest. The main shortcoming of these approaches is that they are unable to exploit the fact that each example is only assigned to one class. Additional problems with extending semi-supervised binary classifiers to multi-class problems include imbalanced classification and different output scales of different binary classifiers. We propose a semi-supervised boosting framework, termed Multi-Class Semi-Supervised Boosting (MCSSB), that directly solves the semi-supervised multi-class learning problem. Compared to the existing semi-supervised boosting methods, the proposed framework is advantageous in that it exploits both classification confidence and similarities among examples when deciding the pseudo-labels for unlabeled examples. Empirical study with a number of UCI datasets shows that the proposed MCSSB algorithm performs better than the state-of-the-art boosting algorithms for semi-supervised learning.","Title: Tattoo-ID: automatic tattoo image retrieval for suspect and victim identification\nYear: 2007\n\nAbstract: Tattoos are used by law enforcement agencies for identification of a victim or a suspect using a false identity. Current method for matching tattoos is based on human-assigned class labels that is time consuming, subjective and has limited performance. It is desirable to build a content-based image retrieval (CBIR) system for automatic matching and retrieval of tattoos. We examine several key design issues related to building a prototype CBIR system for tattoo image database. Our system computes the similarity between the query and stored tattoos based on image content to retrieve the most similar tattoos. The performance of the system is evaluated on a database of 2,157 tattoos representing 20 different classes. Effects of segmentation errors, image transformations (e.g., blurring, illumination), influence of semantic labels and relevance feedback are also studied.","Title: Discriminative semi-supervised feature selection via manifold regularization.\nYear: 2009\n\nAbstract: Feature selection has attracted a huge amount of interest in both research and application communities of data mining. We consider the problem of semi-supervised feature selection, where we are given a small amount of labeled examples and a large amount of unlabeled examples. Since a small number of labeled samples are usually insufficient for identifying the relevant features, the critical problem arising from semi-supervised feature selection is how to take advantage of the information underneath the unlabeled data. To address this problem, we propose a novel discriminative semi-supervised feature selection method based on the idea of manifold regularization. The proposed approach selects features through maximizing the classification margin between different classes and simultaneously exploiting the geometry of the probability distribution that generates both labeled and unlabeled data. In comparison with previous semi-supervised feature selection algorithms, our proposed semi-supervised feature selection method is an embedded feature selection method and is able to find more discriminative features. We formulate the proposed feature selection method into a convex-concave optimization problem, where the saddle point corresponds to the optimal solution. To find the optimal solution, the level method, a fairly recent optimization method, is employed. We also present a theoretic proof of the convergence rate for the application of the level method to our problem. Empirical evaluation on several benchmark data sets demonstrates the effectiveness of the proposed semi-supervised feature selection method.","Title: Delay Analysis of Physical-Layer Key Generation in Dynamic Roadside-to-Vehicle Networks.\nYear: 2017\n\nAbstract: Secret key generation by extracting the shared randomness in a wireless fading channel is a promising way to ensure wireless communication security. Previous studies only consider key generation in static networks, but real-world key establishments are usually dynamic. In this paper, for the first time, we investigate the pairwise key generation in dynamic wireless networks with a center node and ...","Title: Which Factorization Machine Modeling is Better: A Theoretical Answer with Optimal Guarantee.\nYear: 2019\n\nAbstract: Factorization machine (FM) is a popular machine learning model to capture the second order feature interactions. The optimal learning guarantee of FM and its generalized version is not yet developed. For a rank k generalized FM of d dimensional input, the previous best known sampling complexity is O[k(3)d . polylog(kd)] under Gaussian distribution. This bound is sub-optimal comparing to the information theoretical lower bound O(kd). In this work, we aim to tighten this bound towards optimal and generalize the analysis to sub-gaussian distribution. We prove that when the input data satisfies the so-called tau-Moment Invertible Property, the sampling complexity of generalized FM can be improved to O[k(2)d . polylog(kd)/tau(2)]. When the second order self-interaction terms are excluded in the generalized FM, the bound can be improved to the optimal O[kd . polylog(kd)] up to the logarithmic factors. Our analysis also suggests that the positive semi-definite constraint in the conventional FM is redundant as it does not improve the sampling complexity while making the model difficult to optimize We evaluate our improved FM model in real-time high precision GPS signal calibration task to validate its superiority.","Title: Conservative Exploration for Semi-Bandits with Linear Generalization: A Product Selection Problem for Urban Warehouses.\nYear: 2019\n\nAbstract: The recent rising popularity of ultra-fast delivery services on retail platforms fuels increasing use of urban warehouses, whose proximity to customers makes fast deliveries viable. The space limit in urban warehouses poses a problem for online retailers: number of products (SKUs) they carry is no longer the more, better, yet it can still be significantly large, reaching hundreds or thousands in a product category. In this paper, we study algorithms for dynamically identifying a large number of products (i.e., SKUs) with top customer purchase probabilities on fly, from an ocean of potential products to offer on retailersu0027 ultra-fast delivery platforms. distill product selection problem into a semi-bandit model with linear generalization. There are in total $N$ different arms, each with a feature vector of dimension $d$. The player pulls $K$ arms in each period and observes bandit feedback from each of pulled arms. We focus on setting where $K$ is much greater than number of total time periods $T$ or dimension of product features $d$. We first analyze a standard UCB algorithm and show its regret bound can be expressed as sum of a $T$-independent part $tilde O(K d^{3/2})$ and a $T$-dependent part $tilde O(dsqrt{KT})$, which we refer to as and variable respectively. To reduce fixed cost for large $K$ values, we propose a novel online learning algorithm, with more conservative exploration steps, and show its fixed cost is reduced by a factor of $d$ to $tilde O(K sqrt{d})$. Moreover, we test algorithms on an industrial dataset from Alibaba Group. Experimental results show that our new algorithm reduces total regret of standard UCB algorithm by at least 10%.","Title: Stochastic Optimization for DC Functions and Non-smooth Non-convex Regularizers with Non-asymptotic Convergence\nYear: 2019\n\nAbstract: Difference of convex (DC) functions cover a broad family of non-convex and possibly non-smooth and non-differentiable functions, and have wide applications in machine learning and statistics. Although deterministic algorithms for DC functions have been extensively studied, stochastic optimization that is more suitable for learning with big data remains under-explored. In this paper, we propose new stochastic optimization algorithms and study their first-order convergence theories for solving a broad family of DC functions. We improve the existing algorithms and theories of stochastic optimization for DC functions from both practical and theoretical perspectives. On the practical side, our algorithm is more user-friendly without requiring a large mini-batch size and more efficient by saving unnecessary computations. On the theoretical side, our convergence analysis does not necessarily require the involved functions to be smooth with Lipschitz continuous gradient. Instead, the convergence rate of the proposed stochastic algorithm is automatically adaptive to the H\\\"{o}lder continuity of the gradient of one component function. Moreover, we extend the proposed stochastic algorithms for DC functions to solve problems with a general non-convex non-differentiable regularizer, which does not necessarily have a DC decomposition but enjoys an efficient proximal mapping. To the best of our knowledge, this is the first work that gives the first non-asymptotic convergence for solving non-convex optimization whose objective has a general non-convex non-differentiable regularizer.","Title: Publishing Social Network Graph Eigen-Spectrum with Privacy Guarantees\nYear: 2020\n\nAbstract: Online social networks (OSNs) often refuse to publish their social network graphs due to privacy concerns. Recently, differential privacy has become the widely accepted criteria for privacy preserving data publishing. Although some work has been done on publishing matrices with differential privacy, they are computationally unpractical as they are not designed to handle large matrices such as adja...","Title: Distributed Structured Compressive Sensing-Based Time-Frequency Joint Channel Estimation for Massive MIMO-OFDM Systems\nYear: 2019\n\nAbstract: In massive multi-input multi-output orthogonal frequency division multiplexing (MIMO-OFDM) systems, accurate channel state information (CSI) is essential to realize system performance gains such as high spectrum and energy efficiency. However, high-dimensional CSI acquisition requires prohibitively high pilot overhead, which leads to a significant reduction in spectrum efficiency and energy efficiency. In this paper, we propose a more efficient time-frequency joint channel estimation scheme for massive MIMO-OFDM systems to resolve those problems. First, partial channel common support (PCCS) is obtained by using time-domain training. Second, utilizing the spatiotemporal common sparse property of the MIMO channels and the obtained PCCS information, we propose the priori-information aided distributed structured sparsity adaptive matching pursuit (PA-DS-SAMP) algorithm to achieve accurate channel estimation in frequency domain. Third, through performance analysis of the proposed algorithm, two signal power reference thresholds are given, which can ensure that the signal can be recovered accurately under power-limited noise and accurately recovered according to probability under Gaussian noise. Finally, pilot design, computational complexity, spectrum efficiency, and energy efficiency are discussed as well. Simulation results show that the proposed method achieves higher channel estimation accuracy while requiring lower pilot sequence overhead compared with other methods.","Title: On the Computation and Communication Complexity of Parallel SGD with Dynamic Batch Sizes for Stochastic Non-Convex Optimization.\nYear: 2019\n\nAbstract: For SGD based distributed stochastic optimization, computation complexity, measured by the convergence rate in terms of the number of stochastic gradient calls, and communication complexity, measured by the number of inter-node communication rounds, are two most important performance metrics. The classical data-parallel implementation of SGD over $N$ workers can achieve linear speedup of its convergence rate but incurs an inter-node communication round at each batch. We study the benefit of using dynamically increasing batch sizes in parallel SGD for stochastic non-convex optimization by charactering the attained convergence rate and the required number of communication rounds. We show that for stochastic non-convex optimization under the P-L condition, the classical data-parallel SGD with exponentially increasing batch sizes can achieve the fastest known $O(1/(NT))$ convergence with linear speedup using only $\\log(T)$ communication rounds. For general stochastic non-convex optimization, we propose a Catalyst-like algorithm to achieve the fastest known $O(1/\\sqrt{NT})$ convergence with only $O(\\sqrt{NT}\\log(\\frac{T}{N}))$ communication rounds.","Title: Robust Gaussian Process Regression for Real-Time High Precision GPS Signal Enhancement.\nYear: 2019\n\nAbstract: Satellite-based positioning system such as GPS often suffers from large amount of noise that degrades the positioning accuracy dramatically especially in real-time applications. In this work, we consider a data-mining approach to enhance the GPS signal. We build a large-scale high precision GPS receiver grid system to collect real-time GPS signals for training. The Gaussian Process (GP) regression is chosen to model the vertical Total Electron Content (vTEC) distribution of the ionosphere of the Earth. Our experiments show that the noise in the real-time GPS signals often exceeds the breakdown point of the conventional robust regression methods resulting in sub-optimal system performance. We propose a three-step approach to address this challenge. In the first step we perform a set of signal validity tests to separate the signals into clean and dirty groups. In the second step, we train an initial model on the clean signals and then reweigting the dirty signals based on the residual error. A final model is retrained on both the clean signals and the reweighted dirty signals. In the theoretical analysis, we prove that the proposed three-step approach is able to tolerate much higher noise level than the vanilla robust regression methods if two reweighting rules are followed. We validate the superiority of the proposed method in our real-time high precision positioning system against several popular state-of-the-art robust regression methods. Our method achieves centimeter positioning accuracy in the benchmark region with probability $78.4%$ , outperforming the second best baseline method by a margin of $8.3%$. The benchmark takes 6 hours on 20,000 CPU cores or 14 years on a single CPU.","Title: XNAS: Neural Architecture Search with Expert Advice.\nYear: 2019\n\nAbstract: This paper introduces a novel optimization method for differential neural architecture search, based on the theory of prediction with expert advice. Its optimization criterion is well fitted for an architecture-selection, i.e., it minimizes the regret incurred by a sub-optimal selection of operations. Unlike previous search relaxations, that require hard pruning of architectures, our method is designed to dynamically wipe out inferior architectures and enhance superior ones. It achieves an optimal worst-case regret bound and suggests the use of multiple learning-rates, based on the amount of information carried by the backward gradients. Experiments show that our algorithm achieves a strong performance over several image classification datasets. Specifically, it obtains an error rate of 1.6% for CIFAR-10, 23.9% for ImageNet under mobile settings, and achieves state-of-the-art results on three additional datasets.","Title: Train A One-Million-Way Instance Classifier For Unsupervised Visual Representation Learning\nYear: 2021\n\nAbstract: This paper presents a simple unsupervised visual representation learning method with a pretext task of discriminating all images in a dataset using a parametric, instance-level classifier. The overall framework is a replica of a supervised classification model, where semantic classes (e.g., dog, bird, and ship) are replaced by instance IDs. However, scaling up the classification task from thousands of semantic labels to millions of instance labels brings specific challenges including 1) the large-scale softmax computation; 2) the slow convergence due to the infrequent visiting of instance samples; and 3) the massive number of negative classes that can be noisy. This work presents several novel techniques to handle these difficulties. First, we introduce a hybrid parallel training framework to make large-scale training feasible. Second, we present a raw-feature initialization mechanism for classification weights, which we assume offers a contrastive prior for instance discrimination and can clearly speed up converge in our experiments. Finally, we propose to smooth the labels of a few hardest classes to avoid optimizing over very similar negative pairs. While being conceptually simple, our framework achieves competitive or superior performance compared to state-of-the-art unsupervised approaches, i.e., SimCLR, MoCoV2, and PIC under ImageNet linear evaluation protocol and on several downstream visual tasks, verifying that full instance classification is a strong pretraining technique for many semantic visual tasks.","Title: Communication Efficient SGD via Gradient Sampling with Bayes Prior\nYear: 2021\n\nAbstract: Gradient compression has been widely adopted in data-parallel distributed training of deep neural networks to reduce communication overhead. Some literatures have demonstrated that large gradients are more important than small ones because they contain more information, such as Top-k compressor. Other mainstream methods, like random-k compressor and gradient quantization, usually treat all gradients equally. Different from all of them, we regard large and small gradients selection as the exploitation and exploration of gradient information, respectively. And we find taking both of them into consideration is the key to boost the final accuracy. So, we propose a novel gradient compressor: Gradient Sampling with Bayes Prior in this paper. Specifically, we sample important/large gradients based on the global gradient distribution, which is periodically updated across multiple workers. Then we introduce Bayes Prior into distribution model to further explore the gradients. We prove the convergence of our method for smooth non-convex problems in the distributed system. Compared with methods that running after high compression ratio at the expense of accuracy, we pursue no loss of accuracy and the actual acceleration benefit in practice. Experimental comparisons on a variety of computer vision tasks (e.g. image classification and object detection) and backbones (ResNet, MobileNetV2, InceptionV3 and AlexNet) show that our approach outperforms the state-of-the-art techniques in terms of both speed and accuracy, with the limitation of 100x compression ratio.","Title: Effective Model Sparsification by Scheduled Grow-and-Prune Methods\nYear: 2022\n\nAbstract: Deep neural networks (DNNs) are effective in solving many real-world problems. Larger DNN models usually exhibit better quality (e.g., accuracy) but their excessive computation results in long training and inference time. Model sparsification can reduce the computation and memory cost while maintaining model quality. Most existing sparsification algorithms unidirectionally remove weights, while others randomly or greedily explore a small subset of weights in each layer. The inefficiency of the algorithms reduces the achievable sparsity level. In addition, many algorithms still require pre-trained dense models and thus suffer from large memory footprint and long training time. In this paper, we propose a novel scheduled grow-and-prune (GaP) methodology without pre-training the dense models. It addresses the shortcomings of the previous works by repeatedly growing a subset of layers to dense and then pruning back to sparse after some training. Experiments have shown that such models can match or beat the quality of highly optimized dense models at 80% sparsity on a variety of tasks, such as image classification, objective detection, 3D object part segmentation, and translation. They also outperform other state-of-the-art (SOTA) pruning methods, including pruning from pre-trained dense models. As an example, a 90% sparse ResNet-50 obtained via GaP achieves 77.9% top-1 accuracy on ImageNet, improving the SOTA results by 1.5%.","Title: Graph Convolution for Re-Ranking in Person Re-Identification.\nYear: 2022\n\nAbstract: Nowadays, deep learning is widely applied to extract features for similarity computation in person re-identification (re-ID) and have achieved great success. However, due to the non-overlapping between training and testing IDs, the difference between the data used for model training and the testing data makes the performance of learned feature degraded during testing. Hence, re-ranking is proposed to mitigate this issue and various algorithms have been developed. However, most of existing re-ranking methods focus on replacing the Euclidean distance with sophisticated distance metrics, which are not friendly to downstream tasks and hard to be used for fast retrieval of massive data in real applications. In this work, we propose a graph-based re-ranking method to improve learned features while still keeping Euclidean distance as the similarity metric. Inspired by graph convolution networks, we develop an operator to propagate features over an appropriate graph. Since graph is the essential key for the propagation, two important criteria are considered for designing the graph, and three different graphs are explored accordingly. Furthermore, a simple yet effective method is proposed to generate a profile vector for each tracklet in videos, which helps extend our method to video re-ID. Extensive experiments on three benchmark data sets, e.g., Market-1501, Duke, and MARS, demonstrate the effectiveness of our proposed approach.","Title: Scaled ReLU Matters for Training Vision Transformers.\nYear: 2022\n\nAbstract: Vision transformers (ViTs) have been an alternative design paradigm to convolutional neural networks (CNNs). However, the training of ViTs is much harder than CNNs, as it is sensitive to the training parameters, such as learning rate, optimizer and warmup epoch. The reasons for training difficulty are empirically analysed in the paper Early Convolutions Help Transformers See Better, and the authors conjecture that the issue lies with the patchify-stem of ViT models. In this paper, we further investigate this problem and extend the above conclusion: only early convolutions do not help for stable training, but the scaled ReLU operation in the convolutional stem (conv-stem) matters. We verify, both theoretically and empirically, that scaled ReLU in conv-stem not only improves training stabilization, but also increases the diversity of patch tokens, thus boosting peak performance with a large margin via adding few parameters and flops. In addition, extensive experiments are conducted to demonstrate that previous ViTs are far from being well trained, further showing that ViTs have great potential to be a better substitute of CNNs.","Title: CDTrans: Cross-domain Transformer for Unsupervised Domain Adaptation\nYear: 2022\n\nAbstract: Unsupervised domain adaptation (UDA) aims to transfer knowledge learned from a labeled source domain to a different unlabeled target domain. Most existing UDA methods focus on learning domain-invariant feature representation, either from the domain level or category level, using convolution neural networks (CNNs)-based frameworks. One fundamental problem for the category level based UDA is the production of pseudo labels for samples in target domain, which are usually too noisy for accurate domain alignment, inevitably compromising the UDA performance. With the success of Transformer in various tasks, we find that the cross-attention in Transformer is robust to the noisy input pairs for better feature alignment, thus in this paper Transformer is adopted for the challenging UDA task. Specifically, to generate accurate input pairs, we design a two-way center-aware labeling algorithm to produce pseudo labels for target samples. Along with the pseudo labels, a weight-sharing triple-branch transformer framework is proposed to apply self-attention and cross-attention for source/target feature learning and source-target domain alignment, respectively. Such design explicitly enforces the framework to learn discriminative domain-specific and domain-invariant representations simultaneously. The proposed method is dubbed CDTrans (cross-domain transformer), and it provides one of the first attempts to solve UDA tasks with a pure transformer solution. Extensive experiments show that our proposed method achieves the best performance on Office-Home, VisDA-2017, and DomainNet datasets.","Title: Rethinking Supervised Pre-Training for Better Downstream Transferring\nYear: 2022\n\nAbstract: The pretrain-finetune paradigm has shown outstanding performance on many applications of deep learning, where a model is pre-trained on a upstream large dataset (e.g. ImageNet), and is then fine-tuned to different downstream tasks. Though for most cases, the pre-training stage is conducted based on supervised methods, recent works on self-supervised pre-training have shown powerful transferability and even outperform supervised pre-training on multiple downstream tasks. It thus remains an open question how to better generalize supervised pre-training model to downstream tasks. In this paper, we argue that the worse transferability of existing supervised pre-training methods arise from the negligence of valuable intra-class semantic difference. This is because these methods tend to push images from the same class close to each other despite of the large diversity in their visual contents, a problem to which referred as \"overfit of upstream tasks\". To alleviate this problem, we propose a new supervised pre-training method based on Leave-One-Out K-Nearest-Neighbor, or LOOK for short. It relieves the problem of overfitting upstream tasks by only requiring each image to share its class label with most of its k nearest neighbors, thus allowing each class to exhibit a multi-mode distribution and consequentially preserving part of intra-class difference for better transferring to downstream tasks. We developed efficient implementation of the proposed method that scales well to large datasets. Experimental studies on multiple downstream tasks show that LOOK outperforms other state-of-the-art methods for supervised and self-supervised pre-training.","Title: Weakly Supervised Representation Learning with Coarse Labels.\nYear: 2021\n\nAbstract: With the development of computational power and techniques for data collection, deep learning demonstrates a superior performance over most existing algorithms on visual benchmark data sets. Many efforts have been devoted to studying the mechanism of deep learning. One important observation is that deep learning can learn the discriminative patterns from raw materials directly in a task-dependent manner. Therefore, the representations obtained by deep learning outperform hand-crafted features significantly. However, for some real-world applications, it is too expensive to collect the task-specific labels, such as visual search in online shopping. Compared to the limited availability of these task-specific labels, their coarse-class labels are much more affordable, but representations learned from them can be suboptimal for the target task. To mitigate this challenge, we propose an algorithm to learn the fine-grained patterns for the target task, when only its coarse-class labels are available. More importantly, we provide a theoretical guarantee for this. Extensive experiments on real-world data sets demonstrate that the proposed method can significantly improve the performance of learned representations on the target task, when only coarse-class information is available for training.","Title: KVT: k-NN Attention for Boosting Vision Transformers.\nYear: 2022\n\nAbstract: Convolutional Neural Networks (CNNs) have dominated computer vision for years, due to its ability in capturing locality and translation invariance. Recently, many vision transformer architectures have been proposed and they show promising performance. A key component in vision transformers is the fully-connected self-attention which is more powerful than CNNs in modelling long range dependencies. However, since the current dense self-attention uses all image patches (tokens) to compute attention matrix, it may neglect locality of images patches and involve noisy tokens (e.g., clutter background and occlusion), leading to a slow training process and potential degradation of performance. To address these problems, we propose the k-NN attention for boosting vision transformers. Specifically, instead of involving all the tokens for attention matrix calculation, we only select the top-k similar tokens from the keys for each query to compute the attention map. The proposed k-NN attention naturally inherits the local bias of CNNs without introducing convolutional operations, as nearby tokens tend to be more similar than others. In addition, the k-NN attention allows for the exploration of long range correlation and at the same time filters out irrelevant tokens by choosing the most similar tokens from the entire image. Despite its simplicity, we verify, both theoretically and empirically, that k-NN attention is powerful in speeding up training and distilling noise from input tokens. Extensive experiments are conducted by using 11 different vision transformer architectures to verify that the proposed k-NN attention can work with any existing transformer architectures to improve its prediction performance. The codes are available at https://github.com/damo-cv/KVT.","Title: An Online Method for A Class of Distributionally Robust Optimization with Non-convex Objectives.\nYear: 2021\n\nAbstract: In this paper, we propose a practical online method for solving a class of distributional robust optimization (DRO) with non-convex objectives, which has important applications in machine learning for improving the robustness of neural networks. In the literature, most methods for solving DRO are based on stochastic primal-dual methods. However, primal-dual methods for DRO suffer from several drawbacks: (1) manipulating a high-dimensional dual variable corresponding to the size of data is time expensive; (2) they are not friendly to online learning where data is coming sequentially. To address these issues, we consider a class of DRO with an KL divergence regularization on the dual variables, transform the min-max problem into a compositional minimization problem, and propose practical duality-free online stochastic methods without requiring a large mini-batch size. We establish the state-of-the-art complexities of the proposed methods with and without a Polyak-Łojasiewicz (PL) condition of the objective. Empirical studies on large-scale deep learning tasks (i) demonstrate that our method can speed up the training by more than 2 times than baseline methods and save days of training time on a large-scale dataset with ∼ 265K images, and (ii) verify the supreme performance of DRO over Empirical Risk Minimization (ERM) on imbalanced datasets. Of independent interest, the proposed method can be also used for solving a family of stochastic compositional problems with state-of-the-art complexities.","Title: TransFGU: A Top-Down Approach to Fine-Grained Unsupervised Semantic Segmentation.\nYear: 2022\n\nAbstract: Unsupervised semantic segmentation aims to obtain high-level semantic representation on low-level visual features without manual annotations. Most existing methods are bottom-up approaches that try to group pixels into regions based on their visual cues or certain predefined rules. As a result, it is difficult for these bottom-up approaches to generate fine-grained semantic segmentation when coming to complicated scenes with multiple objects and some objects sharing similar visual appearance. In contrast, we propose the first top-down unsupervised semantic segmentation framework for fine-grained segmentation in extremely complicated scenarios. Specifically, we first obtain rich high-level structured semantic concept information from large-scale vision data in a self-supervised learning manner, and use such information as a prior to discover potential semantic categories presented in target datasets. Secondly, the discovered high-level semantic categories are mapped to low-level pixel features by calculating the class activate map (CAM) with respect to certain discovered semantic representation. Lastly, the obtained CAMs serve as pseudo labels to train the segmentation module and produce the final semantic segmentation. Experimental results on multiple semantic segmentation benchmarks show that our top-down unsupervised segmentation is robust to both object-centric and scene-centric datasets under different semantic granularity levels, and outperforms all the current state-of-the-art bottom-up methods. Our code is available at https://github.com/damo-cv/TransFGU.","Title: Decoupling and Recoupling Spatiotemporal Representation for RGB-D-based Motion Recognition\nYear: 2022\n\nAbstract: Decoupling spatiotemporal representation refers to decomposing the spatial and temporal features into dimension-independent factors. Although previous RGB-D-based motion recognition methods have achieved promising performance through the tightly coupled multi-modal spatiotemporal representation, they still suffer from (i) optimization difficulty under small data setting due to the tightly spatiotemporal-entangled modeling; (ii) information redundancy as it usually contains lots of marginal information that is weakly relevant to classification; and (iii) low interaction between multi-modal spatiotemporal information caused by insufficient late fusion. To alleviate these drawbacks, we propose to decouple and recouple spatiotemporal representation for RGB-D-based motion recognition. Specifically, we disentangle the task of learning spatiotemporal representation into 3 sub-tasks: (1) Learning high-quality and dimension independent features through a decoupled spatial and temporal modeling network. (2) Recoupling the decoupled representation to establish stronger space-time dependency. (3) Introducing a Cross-modal Adaptive Posterior Fusion (CAPF) mechanism to capture cross-modal spatiotemporal information from RGB-D data. Seamless combination of these novel designs forms a robust spatiotemporal representation and achieves better performance than state-of-the-art methods on four public motion datasets. Our code is available at https://github.com/damo-cv/MotionRGBD.","Title: FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecasting.\nYear: 2022\n\nAbstract: Long-term time series forecasting is challenging since prediction accuracy tends to decrease dramatically with the increasing horizon. Although Transformer-based methods have significantly improved state-of-the-art results for long-term forecasting, they are not only computationally expensive but more importantly, are unable to capture the global view of time series (e.g. overall trend). To address these problems, we propose to combine Transformer with the seasonal-trend decomposition method, in which the decomposition method captures the global profile of time series while Transformers capture more detailed structures. To further enhance the performance of Transformer for long-term prediction, we exploit the fact that most time series tend to have a sparse representation in a well-known basis such as Fourier transform, and develop a frequency enhanced Transformer. Besides being more effective, the proposed method, termed as Frequency Enhanced Decomposed Transformer (FEDformer), is more efficient than standard Transformer with a linear complexity to the sequence length. Our empirical studies with six benchmark datasets show that compared with state-of-the-art methods, Fedformer can reduce prediction error by 14.8% and 22.6% for multivariate and univariate time series, respectively. Code is publicly available at https://github.com/MAZiqing/FEDformer.","Title: A Trend-Driven Fashion Design System for Rapid Response Marketing in E-commerce.\nYear: 2022\n\nAbstract: Fashion is the form we express ourselves to the world and has grown into one of the largest industries in the world. Despite the significant evolvement of the fashion industry over the past decades, it is still a great challenge to respond to the diverse preferences of a large number of different consumers in time and accurately. To deal with the problem, we present an innovative demonstration of a trend-driven fashion design system using deep generative modeling, which enables automatic fashion design and editing based on trend reports. Our system consists of three components, including trend-driven fashion design, interactive fashion editing, and popularity estimation. The system offers a unified framework for the mass production of fashion designs that conform to the trend, which helps businesses better respond to market demands.","Title: CHEX: CHannel EXploration for CNN Model Compression\nYear: 2022\n\nAbstract: Channel pruning has been broadly recognized as an effective technique to reduce the computation and memory cost of deep convolutional neural networks. However, conventional pruning methods have limitations in that: they are restricted to pruning process only, and they require a fully pre-trained large model. Such limitations may lead to sub-optimal model quality as well as excessive memory and training cost. In this paper, we propose a novel Channel Exploration methodology, dubbed as CHEX, to rectify these problems. As opposed to pruning-only strategy, we propose to repeatedly prune and regrow the channels throughout the training process, which reduces the risk of pruning important channels prematurely. More exactly: From intra-Layer's aspect, we tackle the channel pruning problem via a well-known column subset selection (CSS) formulation. From inter-Layer's aspect, our regrowing stages open a path for dynamically re-allocating the number of channels across all the layers under a global channel sparsity constraint. In addition, all the exploration process is done in a single training from scratch without the need of a pre-trained large model. Experimental results demonstrate that CHEX can effectively reduce the FLOPs of diverse CNN architectures on a variety of computer vision tasks, including image classification, object detection, instance segmentation, and 3D vision. For example, our compressed ResNet-50 model on ImageNet dataset achieves 76% top-l accuracy with only 25% FLOPs of the original ResNet-50 model, outperforming previous state-of-the-art channel pruning methods. The checkpoints and code are available at here.","Title: Learning from Untrimmed Videos: Self-Supervised Video Representation Learning with Hierarchical Consistency\nYear: 2022\n\nAbstract: Natural videos provide rich visual contents for selfsupervised learning. Yet most existing approaches for learning spatio-temporal representations rely on manually trimmed videos, leading to limited diversity in visual patterns and limited performance gain. In this work, we aim to learn representations by leveraging more abundant information in untrimmed videos. To this end, we propose to learn a hierarchy of consistencies in videos, i.e., visual consistency and topical consistency, corresponding respectively to clip pairs that tend to be visually similar when separated by a short time span and share similar topics when separated by a long time span. Specifically, a hierarchical consistency learning framework HiCo is presented, where the visually consistent pairs are encouraged to have the same representation through contrastive learning, while the topically consistent pairs are coupled through a topical classifier that distinguishes whether they are topicrelated. Further, we impose a gradual sampling algorithm for proposed hierarchical consistency learning, and demonstrate its theoretical superiority. Empirically, we show that not only HiCo can generate stronger representations on untrimmed videos, it also improves the representation quality when applied to trimmed videos. This is in contrast to standard contrastive learning that fails to learn appropriate representations from untrimmed videos.","Title: Entroformer: A Transformer-based Entropy Model for Learned Image Compression\nYear: 2022\n\nAbstract: One critical component in lossy deep image compression is the entropy model, which predicts the probability distribution of the quantized latent representation in the encoding and decoding modules. Previous works build entropy models upon convolutional neural networks which are inefficient in capturing global dependencies. In this work, we propose a novel transformer-based entropy model, termed Entroformer, to capture long-range dependencies in probability distribution estimation effectively and efficiently. Different from vision transformers in image classification, the Entroformer is highly optimized for image compression, including a top-k self-attention and a diamond relative position encoding. Meanwhile, we further expand this architecture with a parallel bidirectional context model to speed up the decoding process. The experiments show that the Entroformer achieves state-of-the-art performance on image compression while being time-efficient.","Title: Learning to Generalize to More: Continuous Semantic Augmentation for Neural Machine Translation\nYear: 2022\n\nAbstract: The principal task in supervised neural machine translation (NMT) is to learn to generate target sentences conditioned on the source inputs from a set of parallel sentence pairs, and thus produce a model capable of generalizing to unseen instances. However, it is commonly observed that the generalization performance of the model is highly influenced by the amount of parallel data used in training. Although data augmentation is widely used to enrich the training data, conventional methods with discrete manipulations fail to generate diverse and faithful training samples. In this paper, we present a novel data augmentation paradigm termed Continuous Semantic Augmentation (CSANMT), which augments each training instance with an adjacency semantic region that could cover adequate variants of literal expression under the same meaning. We conduct extensive experiments on both rich-resource and low-resource settings involving various language pairs, including WMT14 English ->{German,French}, NIST Chinese -> English and multiple low-resource IWSLT translation tasks. The provided empirical evidences show that CSANMT sets a new level of performance among existing augmentation techniques, improving on the state-of-the-art by a large margin.(1)","Title: Hybrid Relation Guided Set Matching for Few-shot Action Recognition\nYear: 2022\n\nAbstract: Current few-shot action recognition methods reach impressive performance by learning discriminative features for each video via episodic training and designing various temporal alignment strategies. Nevertheless, they are limited in that (a) learning individual features without considering the entire task may lose the most relevant information in the current episode, and (b) these alignment strategies may fail in misaligned instances. To overcome the two limitations, we propose a novel Hybrid Relation guided Set Matching (HyRSM) approach that incorporates two key components: hybrid relation module and set matching metric. The purpose of the hybrid relation module is to learn task-specific embeddings by fully exploiting associated relations within and cross videos in an episode. Built upon the task-specific features, we reformulate distance measure between query and support videos as a set matching problem and further design a bidirectional Mean Hausdorff Metric to improve the resilience to misaligned instances. By this means, the proposed HyRSM can be highly informative and flexible to predict query categories under the few-shot settings. We evaluate HyRSM on six challenging benchmarks, and the experimental results show its superiority over the state-of-the-art methods by a convincing margin. Project page: https://hyrsm-cvpr2022.github.io/.","Title: Low- and Moderate-Level RFI Detection Using Point-Source Ripple for Synthetic Aperture Interferometric Radiometer\nYear: 2022\n\nAbstract: Finding out and turning off the radio-frequency interference (RFI) sources as many as possible is an important work for synthetic aperture interferometric radiometer (SAIR). While the impact of strong RFIs on reconstructed brightness temperature (BT) image is clearly observable, the presence of low- or moderate-level RFIs is often hard to be aware of, leaving the RFI detection and localization a difficult problem. In this article, an improved RFI detection method is proposed by making use of the point-source ripple to give prominence to the RFIs presented in the SAIR imagery. The method tests based on Soil Moisture and Ocean Salinity satellite (SMOS) visibility samples and the performance evaluations by numerical simulations validate the correctness and effectiveness of the developed method.","Title: ANN Softmax: Acceleration of Extreme Classification Training\nYear: 2021\n\nAbstract: Thanks to the popularity of GPU and the growth of its computational power, more and more deep learning tasks, such as face recognition, image retrieval and word embedding, can take advantage of extreme classification to improve accuracy. However, it remains a big challenge to train a deep model with millions of classes efficiently due to the huge memory and computation consumption in the last layer. By sampling a small set of classes to avoid the total classes calculation, sampling-based approaches have been proved to be an effective solution. But most of them suffer from the following two issues: i) the important classes are ignored or only partly sampled, such as the methods using random sampling scheme or retrieval techniques of low recall (e.g., locality-sensitive hashing), resulting in the degradation of accuracy; ii) inefficient implementation owing to incompatibility with GPU, like selective softmax. It uses hashing forest to help select classes, but the search process is implemented in CPU. To address the above issues, we propose a new sampling-based softmax called ANN Softmax in this paper. Specifically, we employ binary quantization with inverted file system to improve the recall of important classes. With the help of dedicated kernel design, it can be totally parallelized in mainstream training framework. Then, we find the size of important classes that are recalled by each training sample has a great impact on the final accuracy, so we introduce sample grouping optimization to well approximate the full classes training. Experimental evaluations on two tasks (Embedding Learning and Classification) and ten datasets (e.g., MegaFace, ImageNet, SKU datasets) demonstrate our proposed method maintains the same precision as Full Softmax for different loss objectives, including cross entropy loss, ArcFace, CosFace and D-Softmax loss, with only 1/10 sampled classes, which outperforms the state-of-the-art techniques. Moreover, we implement ANN Softmax in a complete GPU pipeline that can accelerate the training more than 4.3x. Equipped our method with a 256 GPUs cluster, the time of training a classifier of 300 million classes on our SKU-300M dataset can be reduced to ten days.","Title: Joint Inversion Algorithm of Sea Surface Temperature From Microwave and Infrared Brightness Temperature\nYear: 2022\n\nAbstract: The demand for high-precision sea surface temperature (SST) has been growing rapidly in recent years because SST is one of the key parameters to describe the thermal state of the sea surface. This article analyzes the differences between microwave remote sensing and infrared remote sensing for SST, including the spatial resolution difference and the penetration depth difference. In order to improve the accuracy of retrieved SST, this article proposes a joint inversion algorithm for SST from combining microwave brightness temperature (BT) and infrared BT, which has also taken the influence of wind speed and atmosphere into consideration. Experiments confirm that SST data obtained from the joint inversion algorithm are more accurate than those obtained from the existing inversion algorithms.","Title: Estimation of Dry Matter and N Nutrient Status of Choy Sum by Analyzing Canopy Images and Plant Height Information\nYear: 2022\n\nAbstract: The estimation accuracy of plant dry matter by spectra- or remote sensing-based methods tends to decline when canopy coverage approaches closure; this is known as the saturation problem. This study aimed to enhance the estimation accuracy of plant dry matter and subsequently use the critical nitrogen dilution curve (CNDC) to diagnose N in Choy Sum by analyzing the combined information of canopy imaging and plant height. A three-year experiment with different N levels (0, 25, 50, 100, 150, and 200 kg center dot ha(-1)) was conducted on Choy Sum. Variables of canopy coverage (CC) and plant height were used to build the dry matter and N estimation model. The results showed that the yields of N-0 and N-25 were significantly lower than those of high-N treatments (N-50, N-100, N-150, and N-200) for all three years. The variables of CC x Height had a significant linear relationship with dry matter, with R-2 values above 0.87. The good performance of the CC x Height-based model implied that the saturation problem of dry matter prediction was well-addressed. By contrast, the relationship between dry matter and CC was best fitted by an exponential function. CNDC models built based on CC x Height information could satisfactorily differentiate groups of N deficiency and N abundance treatments, implying their feasibility in diagnosing N status. N application rates of 50-100 kgN/ha are recommended as optimal for a good yield of Choy Sum production in the study region."],"string":"[\n \"Title: Integrating User Feedback Log into Relevance Feedback by Coupled SVM for Content-Based Image Retrieval\\nYear: 2005\\n\\nAbstract: Relevance feedback has been shown as an important tool to boost the retrieval performance in content-based image retrieval. In the past decade, various algorithms have been proposed to formulate relevance feedback in contentbased image retrieval. Traditional relevance feedback techniques mainly carry out the learning tasks by focusing lowlevel visual features of image content with little consideration on log information of user feedback. However, from a long-term learning perspective, the user feedback log is one of the most important resources to bridge the semantic gap problem in image retrieval. In this paper we propose a novel technique to integrate the log information of user feedback into relevance feedback for image retrieval. Our algorithms construction is based on a coupled support vector machine which learns consistently with the two types of information: the low-level image content and the user feedback log. We present a mathematical formulation of the problem and develop a practical algorithm to solve the problem effectively. Experimental results show that our proposed scheme is effective and promising.\",\n \"Title: An Information Geometry Approach for Distance Metric Learning\\nYear: 2009\\n\\nAbstract: In this paper, we propose a framework for metric learning based on information geome- try. The key idea is to construct two kernel matrices for the given training data: one is based on the distance metric and the other is based on the assigned class labels. Inspired by the idea of information geometry, we re- late these two kernel matrices to two Gaus- sian distributions, and the difierence between the two kernel matrices is then computed by the Kullback-Leibler (KL) divergence be- tween the two Gaussian distributions. The optimal distance metric is then found by min- imizing the divergence between the two dis- tributions. We present two metric learning algorithms, one for linear distance metric and one for nonlinear distance using a kernel func- tion. Unlike many existing algorithms for metric learning that require solving a non- trivial optimization problem and are compu- tationally expensive when the data dimen- sion is high, the proposed algorithms have a closed-form solution and are computationally more e-cient. Extensive experiments with data classiflcation and face recognition show that the proposed algorithms are compara- ble to or better than the state-of-the-art al- gorithms for metric learning.\",\n \"Title: CUR Algorithm with Incomplete Matrix Observation.\\nYear: 2014\\n\\nAbstract: CUR matrix decomposition is a randomized algorithm that can efficiently compute the low rank approximation for a given rectangle matrix. One limitation with the existing CUR algorithms is that they require an access to the full matrix A for computing U. In this work, we aim to alleviate this limitation. In particular, we assume that besides having an access to randomly sampled d rows and d columns from A, we only observe a subset of randomly sampled entries from A. Our goal is to develop a low rank approximation algorithm, similar to CUR, based on (i) randomly sampled rows and columns from A, and (ii) randomly sampled entries from A. The proposed algorithm is able to perfectly recover the target matrix A with only O(rn log n) number of observed entries. In addition, instead of having to solve an optimization problem involved trace norm regularization, the proposed algorithm only needs to solve a standard regression problem. Finally, unlike most matrix completion theories that hold only when the target matrix is of low rank, we show a strong guarantee for the proposed algorithm even when the target matrix is not low rank.\",\n \"Title: Calibration of staggered Y-shaped array by 120 degree rotation\\nYear: 2012\\n\\nAbstract: The calibration of aperture synthesis radiometers (ASRs) by the standard method of noise injection would seriously increase hardware requirements and system complexity. To reduce the network of noise injection, redundant space calibration (RSC) method has been considered. However the number of error phases is larger than the number of independent equations. In this paper, an external calibration method with 120-degree antenna array rotation for geostationary ASR with staggered Y-shaped array is proposed. More redundant baselines are produced by rotation. The equations by rotation and by RSC are combined for solving the phase errors. Basic procedure on this method and preliminary simulation are introduced. As another advantage, if twice 120-degree rotation is taken, one of the antenna arms can be removed from staggered Y-shaped arrays with the missing baselines supplied by the rotations. In this way, the number of antenna elements can be reduced by 1/3.\",\n \"Title: Learning to trade off between exploration and exploitation in multiclass bandit prediction\\nYear: 2011\\n\\nAbstract: We study multi-class bandit prediction, an online learning problem where the learner only receives a partial feedback in each trial indicating whether the predicted class label is correct. The exploration vs. exploitation tradeoff strategy is a well-known technique for online learning with incomplete feedback (i.e., bandit setup). Banditron [8], a multi-class online learning algorithm for bandit setting, maximizes the run-time gain by balancing between exploration and exploitation with a fixed tradeoff parameter. The performance of Banditron can be quite sensitive to the choice of the tradeoff parameter and therefore effective algorithms to automatically tune this parameter is desirable. In this paper, we propose three learning strategies to automatically adjust the tradeoff parameter for Banditron. Our extensive empirical study with multiple real-world data sets verifies the efficacy of the proposed approach in learning the exploration vs. exploitation tradeoff parameter.\",\n \"Title: An efficient algorithm for local distance metric learning\\nYear: 2006\\n\\nAbstract: Learning application-specific distance metrics from labeled data is critical for both statistical classification and information retrieval. Most of the earlier work in this area has focused on finding metrics that simultaneously optimize compactness and separability in a global sense. Specifically, such distance metrics attempt to keep all of the data points in each class close together while ensuring that data points from different classes are separated. However, particularly when classes exhibit multimodal data distributions, these goals conflict and thus cannot be simultaneously satisfied. This paper proposes a Local Distance Metric (LDM) that aims to optimize local compactness and local separability. We present an efficient algorithm that employs eigenvector analysis, and bound optimization to learn the LDM from training data in a probabilistic framework. We demonstrate that LDM achieves significant improvements in both classification and retrieval accuracy compared to global distance learning and kernel-based KNN.\",\n \"Title: A smoothed boosting algorithm using probabilistic output codes\\nYear: 2005\\n\\nAbstract: AdaBoost.OC has shown to be an effective method in boosting \\\"weak\\\" binary classifiers for multi-class learning. It employs the Error Correcting Output Code (ECOC) method to convert a multi-class learning problem into a set of binary classification problems, and applies the AdaBoost algorithm to solve them efficiently. In this paper, we propose a new boosting algorithm that improves the AdaBoost.OC algorithm in two aspects: 1) It introduces a smoothing mechanism into the boosting algorithm to alleviate the potential overfitting problem with the AdaBoost algorithm, and 2) It introduces a probabilistic coding scheme to generate binary codes for multiple classes such that training errors can be efficiently reduced. Empirical studies with seven UCI datasets have indicated that the proposed boosting algorithm is more robust and effective than the AdaBoost.OC algorithm for multi-class learning.\",\n \"Title: Excess Risk Bounds for Exponentially Concave Losses.\\nYear: 2014\\n\\nAbstract: The overarching goal of this paper is to derive excess risk bounds for learning from exp-concave loss functions in passive and sequential learning settings. Exp-concave loss functions encompass several fundamental problems in machine learning such as squared loss in linear regression, logistic loss in classification, and negative logarithm loss in portfolio management. In batch setting, we obtain sharp bounds on the performance of empirical risk minimization performed in a linear hypothesis space and with respect to the exp-concave loss functions. We also extend the results to the online setting where the learner receives the training examples in a sequential manner. We propose an online learning algorithm that is a properly modified version of online Newton method to obtain sharp risk bounds. Under an additional mild assumption on the loss function, we show that in both settings we are able to achieve an excess risk bound of $O(d\\\\log n/n)$ that holds with a high probability.\",\n \"Title: Improved Bounds for the Nyström Method With Application to Kernel Classification\\nYear: 2013\\n\\nAbstract: We develop two approaches for analyzing the approximation error bound for the Nyström method that approximates a positive semidefinite (PSD) matrix by sampling a small set of columns, one based on a concentration inequality for integral operators, and one based on random matrix theory. We show that the approximation error, measured in the spectral norm, can be improved from $O(N/\\\\\\\\sqrt {m})$ to $O(N/m^{1 - \\\\\\\\rho })$ in the case of large eigengap, where $N$ is the total number of data points, $m$ is the number of sampled data points, and $\\\\\\\\rho \\\\\\\\in (0, 1/2)$ is a positive constant that characterizes the eigengap. When the eigenvalues of the kernel matrix follow a $p$-power law, our analysis based on random matrix theory further improves the bound to $O(N/m^{p - 1})$ under an incoherence assumption. We present a kernel classification approach based on the Nyström method and derive its generalization performance using the improved bound. We show that when the eigenvalues of the kernel matrix follow a $p$-power law, we can reduce the number of support vectors to $N^{2p/(p^{2} - 1)}$, which is sublinear in $N$ when $p 1+\\\\\\\\sqrt {2}$, without seriously sacrificing its generalization performance.\",\n \"Title: Language model for IR using collection information\\nYear: 2002\\n\\nAbstract: Information retrieval using meta data can be traced back to the early age of IR where documents are represented by the controlled vocabulary. In this paper, we explore the usage of meta-data information under the framework of language model. We present a new language model that is able to take advantage of the category information for documents to improve the retrieval accuracy. We compare the new language model with the traditional language model over the TREC4 dataset where the collection information for documents is obtained using the k-means clustering method. The new language model outperforms the traditional language model, which verifies our statement.\",\n \"Title: Semi-supervised multi-label learning by constrained non-negative matrix factorization\\nYear: 2006\\n\\nAbstract: We present a novel framework for multi-label learning that explicitly addresses the challenge arising from the large number of classes and a small size of training data. The key assumption behind this work is that two examples tend to have large overlap in their assigned class memberships if they share high similarity in their input patterns. We capitalize this assumption by first computing two sets of similarities, one based on the input patterns of examples, and the other based on the class memberships of the examples. We then search for the optimal assignment of class memberships to the unlabeled data that minimizes the difference between these two sets of similarities. The optimization problem is formulated as a constrained Non-negative Matrix Factorization (NMF) problem, and an algorithm is presented to efficiently find the solution. Compared to the existing approaches for multi-label learning, the proposed approach is advantageous in that it is able to explore both the unlabeled data and the correlation among different classes simultaneously. Experiments with text categorization show that our approach performs significantly better than several state-of-the-art classification techniques when the number of classes is large and the size of training data is small.\",\n \"Title: A graph-based framework for relation propagation and its application to multi-label learning\\nYear: 2006\\n\\nAbstract: Label propagation exploits the structure of the unlabeled documents by propagating the label information of the training documents to the unlabeled documents. The limitation with the existing label propagation approaches is that they can only deal with a single type of objects. We propose a framework, named \\\"relation propagation\\\", that allows for information propagated among multiple types of objects. Empirical studies with multi-label text categorization showed that the proposed algorithm is more effective than several semi-supervised learning algorithms in that it is capable of exploring the correlation among different categories and the structure of unlabeled documents simultaneously.\",\n \"Title: Fast Bounded Online Gradient Descent Algorithms for Scalable Kernel-Based Online Learning\\nYear: 2012\\n\\nAbstract: Kernel-based online learning has often shown state-of-the-art performance for many online learning tasks. It, however, suffers from a major shortcoming, that is, the unbounded number of support vectors, making it non-scalable and unsuitable for applications with large-scale datasets. In this work, we study the problem of bounded kernel-based online learning that aims to constrain the number of support vectors by a predefined budget. Although several algorithms have been proposed in literature, they are neither computationally efficient due to their intensive budget maintenance strategy nor effective due to the use of simple Perceptron algorithm. To overcome these limitations, we propose a framework for bounded kernel-based online learning based on an online gradient descent approach. We propose two efficient algorithms of bounded online gradient descent (BOGD) for scalable kernel-based online learning: (i) BOGD by maintaining support vectors using uniform sampling, and (ii) BOGD++ by maintaining support vectors using non-uniform sampling. We present theoretical analysis of regret bound for both algorithms, and found promising empirical performance in terms of both efficacy and efficiency by comparing them to several well-known algorithms for bounded kernel-based online learning on large-scale datasets.\",\n \"Title: A Bayesian framework for knowledge driven regression model in micro-array data analysis.\\nYear: 2008\\n\\nAbstract: This paper addresses the sparse data problem in the linear regression model, namely the number of variables is significantly larger than the number of the data points for regression. We assume that in addition to the measured data points, the prior knowledge about the input variables may be provided in the form of pair wise similarity. We presented a full Bayesian framework to effectively exploit the similarity information of the input variables for linear regression. Empirical studies with gene expression data show that the regression errors can be reduced significantly by incorporating the similarity information derived from gene ontology.\",\n \"Title: Distance metric learning from uncertain side information for automated photo tagging\\nYear: 2011\\n\\nAbstract: Automated photo tagging is an important technique for many intelligent multimedia information systems, for example, smart photo management system and intelligent digital media library. To attack the challenge, several machine learning techniques have been developed and applied for automated photo tagging. For example, supervised learning techniques have been applied to automated photo tagging by training statistical classifiers from a collection of manually labeled examples. Although the existing approaches work well for small testbeds with relatively small number of annotation words, due to the long-standing challenge of object recognition, they often perform poorly in large-scale problems. Another limitation of the existing approaches is that they require a set of high-quality labeled data, which is not only expensive to collect but also time consuming. In this article, we investigate a social image based annotation scheme by exploiting implicit side information that is available for a large number of social photos from the social web sites. The key challenge of our intelligent annotation scheme is how to learn an effective distance metric based on implicit side information (visual or textual) of social photos. To this end, we present a novel “Probabilistic Distance Metric Learning” (PDML) framework, which can learn optimized metrics by effectively exploiting the implicit side information vastly available on the social web. We apply the proposed technique to photo annotation tasks based on a large social image testbed with over 1 million tagged photos crawled from a social photo sharing portal. Encouraging results show that the proposed technique is effective and promising for social photo based annotation tasks.\",\n \"Title: Semi-supervised Clustering by Input Pattern Assisted Pairwise Similarity Matrix Completion.\\nYear: 2013\\n\\nAbstract: Many semi-supervised clustering algorithms have been proposed to improve the clustering accuracy by effectively exploring the available side information that is usually in the form of pairwise constraints. Despite the progress, there are two main shortcomings of the existing semi-supervised clustering algorithms. First, they have to deal with non-convex optimization problems, leading to clustering results that are sensitive to the initialization. Second, none of these algorithms is equipped with theoretical guarantee regarding the clustering performance. We address these limitations by developing a framework for semi-supervised clustering based on \\\\it input pattern assisted matrix completion. The key idea is to cast clustering into a matrix completion problem, and solve it efficiently by exploiting the correlation between input patterns and cluster assignments. Our analysis shows that under appropriate conditions, only O(\\\\log n) pairwise constraints are needed to accurately recover the true cluster partition. We verify the effectiveness of the proposed algorithm by comparing it to the state-of-the-art semi-supervised clustering algorithms on several benchmark datasets.\",\n \"Title: Active Query Selection For Semi-Supervised Clustering\\nYear: 2008\\n\\nAbstract: Semi-supervised clustering allows a user to specify available prior knowledge about the data to improve the clustering performance. A common way to express this information is in the form of pair-wise constraints. A number of studies have shown that, in general, these constraints improve the resulting data partition. However the choice of constraints is critical since improperly chosen constraints might actually degrade the clustering performance. We focus on constraint (also known as query) selection for improving the performance of semi-supervised clustering algorithms. We present an active query selection mechanism, where the queries are selected using a min-max criterion. Experimental results on a variety of datasets, using MPCK-means as the underlying semi-clustering algorithm, demonstrate the superior performance of the proposed query selection procedure.\",\n \"Title: A language modeling framework for resource selection and results merging\\nYear: 2002\\n\\nAbstract: Statistical language models have been proposed recently for several information retrieval tasks, including the resource selection task in distributed information retrieval. This paper extends the language modeling approach to integrate resource selection, ad-hoc searching, and merging of results from different text databases into a single probabilistic retrieval model. This new approach is designed primarily for Intranet environments, where it is reasonable to assume that resource providers are relatively homogeneous and can adopt the same kind of search engine. Experiments demonstrate that this new, integrated approach is at least as effective as the prior state-of-the-art in distributed IR.\",\n \"Title: SWITCH: A Novel Approach to Ensemble Learning for Heterogeneous Data\\nYear: 2004\\n\\nAbstract: The standard framework of machine learning problems assumes that the available data is independent and identically distributed (i.i.d.). However, in some applications such as image classification, the training data are often collected from multiple sources and heterogeneous. Ensemble learning is a proven effective approach to heterogeneous data, which uses multiple classification models to capture the diverse aspects of heterogeneous data. If an ensemble can learn the relationship between different portions of data and their corresponding models, the ensemble can selectively apply models to unseen data according to the learned relationship. We propose a novel approach to enable the learning of the relationships between data and models by creating a set of I switches' that can route a testing instance to appropriate classification models in an ensemble. Our empirical study on both real-world data and benchmark data shows that the proposed approach to ensemble learning can achieve significant performance improvement for heterogeneous data.\",\n \"Title: Learning Bregman Distance Functions for Semi-Supervised Clustering\\nYear: 2012\\n\\nAbstract: Learning distance functions with side information plays a key role in many data mining applications. Conventional distance metric learning approaches often assume that the target distance function is represented in some form of Mahalanobis distance. These approaches usually work well when data are in low dimensionality, but often become computationally expensive or even infeasible when handling high-dimensional data. In this paper, we propose a novel scheme of learning nonlinear distance functions with side information. It aims to learn a Bregman distance function using a nonparametric approach that is similar to Support Vector Machines. We emphasize that the proposed scheme is more general than the conventional approach for distance metric learning, and is able to handle high-dimensional data efficiently. We verify the efficacy of the proposed distance learning method with extensive experiments on semi-supervised clustering. The comparison with state-of-the-art approaches for learning distance functions with side information reveals clear advantages of the proposed technique.\",\n \"Title: Error model and calibration of synthetic aperture interferometric radiometer based on visibility function\\nYear: 2012\\n\\nAbstract: Traditionally, the system errors of synthetic aperture interferometric radiometers (SAIRs) are analyzed and classified by the causes they are generated, which is easy for understanding but complicated for calibration. This paper presents a new error model based on the visibility function to analyze the system errors of SAIRs. In this model the system errors are classified by their effects on the ideal visibility function, which is simple for calibration. Firstly, the error model based on the visibility function is proposed, in which the errors according to their effects on the ideal visibility function are grouped in: 1) the orientation-independent multiplicative error, 2) the orientation-dependent multiplicative error, and 3) the additive error. Then, the two overall calibration methods respectively using external reference source and external reference scene are proposed. Finally, the preliminary calibration experiment results on the 8mm-band HUST-SAIR prototype show the validity of the error model and the overall calibration methods. The study in this paper provides a means to reduce the complexity of calibration routine of SAIRs and to improve the imaging performance of SAIRs.\",\n \"Title: Robust feature induction for support vector machines\\nYear: 2004\\n\\nAbstract: The goal of feature induction is to automatically create nonlinear combinations of existing features as additional input features to improve classification accuracy. Typically, nonlinear features are introduced into a support vector machine (SVM) through a nonlinear kernel function. One disadvantage of such an approach is that the feature space induced by a kernel function is usually of high dimension and therefore will substantially increase the chance of over-fitting the training data. Another disadvantage is that nonlinear features are induced implicitly and therefore are difficult for people to understand which induced features are critical to the classification performance. In this paper, we propose a boosting-style algorithm that can explicitly induces important nonlinear features for SVMs. We present empirical studies with discussion to show that this approach is effective in improving classification accuracy for SVMs. The comparison with an SVM model using nonlinear kernels also indicates that this approach is effective and robust, particularly when the number of training data is small.\",\n \"Title: Online learning by ellipsoid method\\nYear: 2009\\n\\nAbstract: In this work, we extend the ellipsoid method, which was originally designed for convex optimization, for online learning. The key idea is to approximate by an ellipsoid the classification hypotheses that are consistent with all the training examples received so far. This is in contrast to most online learning algorithms where only a single classifier is maintained at each iteration. Efficient algorithms are presented for updating both the centroid and the positive definite matrix of ellipsoid given a misclassified example. In addition to the classical ellipsoid method, an improved version for online learning is also presented. Mistake bounds for both ellipsoid methods are derived. Evaluation with the USPS dataset and three UCI data-sets shows encouraging results when comparing the proposed online learning algorithm to two state-of-the-art online learners.\",\n \"Title: A Kernel Approach to Multi-Task Learning with Task-Specific Kernels\\nYear: 2012\\n\\nAbstract: Several kernel-based methods for multi-task learning have been proposed,which leverage relations among tasks as regularization to enhance the overall learning accuracies.These methods assume that the tasks share the same kernel,which could limit their applications because in practice different tasks may need different kernels.The main challenge of introducing multiple kernels into multiple tasks is that models from different reproducing kernel Hilbert spaces(RKHSs) are not comparable,making it difficult to exploit relations among tasks.This paper addresses the challenge by formalizing the problem in the square integrable space(SIS).Specially,it proposes a kernel-based method which makes use of a regularization term defined in SIS to represent task relations.We prove a new representer theorem for the proposed approach in SIS.We further derive a practical method for solving the learning problem and conduct consistency analysis of the method.We discuss the relationship between our method and an existing method.We also give an SVM(support vector machine)based implementation of our method for multi-label classification.Experiments on an artificial example and two real-world datasets show that the proposed method performs better than the existing method.\",\n \"Title: Collaborative filtering with decoupled models for preferences and ratings\\nYear: 2003\\n\\nAbstract: In this paper, we describe a new model for collaborative filtering. The motivation of this work comes from the fact that two users with very similar preferences on items may have very different rating schemes. For example, one user may tend to assign a higher rating to all items than another user. Unlike previous models of collaborative filtering, which determine the similarity between two users only based on their rating performance, our model treats the user's preferences on items separately from the user's rating scheme. More specifically, for each user, we build two separate models: a preference model capturing which items are favored by the user and a rating model capturing how the user would rate an item given the preference information. The similarity of two users is computed based on the underlying preference model, instead of the surface ratings. We compare the new model with several representative previous approaches on two data sets. Experiment results show that the new model outperforms all the previous approaches that are tested consistently on both data sets.\",\n \"Title: Graffiti-ID: matching and retrieval of graffiti images\\nYear: 2009\\n\\nAbstract: Graffiti are abundant in most urban neighborhoods and are considered a nuisance and an eyesore. Yet, law enforcement agencies have found them to be useful for understanding gang activities, and uncovering the extent of a gang's territory in large metropolitan areas. The current method for matching and retrieving graffiti is based on a manual database search that is not only inaccurate but also time consuming. We present a content-based image retrieval (CBIR) system for automatic matching and retrieval of graffiti images. Our system represents each graffiti image by a bag of SIFT (Scale Invariant Feature Transform) features. The similarity between a query image and a graffiti image in the database is computed based on the number of matched SIFT features between the two images under certain geometric constraints. Experimental results on two graffiti databases with thousands of graffiti images show encouraging results.\",\n \"Title: Passive Learning with Target Risk\\nYear: 2013\\n\\nAbstract: In this paper we consider learning in passive setting but with a slight modification. We assume that the target expected loss, also referred to as target risk, is provided in advance for learner as prior knowledge. Unlike most studies in the learning theory that only incorporate the prior knowledge into the generalization bounds, we are able to explicitly utilize the target risk in the learning process. Our analysis reveals a surprising result on the sample complexity of learning: by exploiting the target risk in the learning algorithm, we show that when the loss function is both strongly convex and smooth, the sample complexity reduces to $\\\\O(\\\\log (\\\\frac{1}{\\\\epsilon}))$, an exponential improvement compared to the sample complexity $\\\\O(\\\\frac{1}{\\\\epsilon})$ for learning with strongly convex loss functions. Furthermore, our proof is constructive and is based on a computationally efficient stochastic optimization algorithm for such settings which demonstrate that the proposed algorithm is practically useful.\",\n \"Title: Meta-Classification of Multimedia Classifiers\\nYear: 2002\\n\\nAbstract: Combining multiple classifiers is of particular interest in the multimedia systems, since there is usually data of very different types/modalities that should be mined or analyzed. Our wearable 'experience collection' system unobtrusively records the wearer's conversation, recognizes the face of the dialog partner and remembers his/her voice. When the system sees the same person's face or hears the same voice it can then use a summary of the last conversation with this person to remind the wearer. To correctly identify a person from a mixture of video and audio stream, classification judgments from individual modality classifiers must be combined effectively to yield a more accurate decision. A meta-classification strategy of combining multimodal classifiers using Support Vector Machine is proposed. Preliminary, empirical results show that combining different face recognition and speaker identification technology by meta-classification is dramatically more effective than weighted interpolation. Meta-classification is general enough to be applied to any application that needs to combine multiple classifiers without much modification.\",\n \"Title: Data analysis for a multimedia library\\nYear: 2000\\n\\nAbstract: This section describes the indexing, search, and retrieval of various combinations of audio, video, text, and image media\\n and the automated content processing that enables it. The intent is to provide a framework for data analysis in multimedia\\n digital libraries. The organization of this article is as follows: The introduction briefly distinguishes digital from traditional\\n libraries and touches on the specific issues important to searching the content of multimedia libraries. The second section\\n introduces the Informedia Digital Video Library as an example of a multimedia library, including a quick tour of the functionality.\\n The next section discusses the processing of audio and image information, as it relates to a multimedia library. Section four\\n illustrates the interplay between audio and video information using a video information retrieval experiment as an example.\\n Section five discusses the exporting and sharing of metadata in a digital library using MPEG–7. Finally, section 6 presents\\n one vision of a future digital library, where all personal memory can be recorded and accessed.\",\n \"Title: Online Stochastic Optimization with Multiple Objectives\\nYear: 2012\\n\\nAbstract: In this paper we propose a general framework to characterize and solve the stochastic optimization problems with multiple objectives underlying many real world learning applications. We first propose a projection based algorithm which attains an $O(T^{-1/3})$ convergence rate. Then, by leveraging on the theory of Lagrangian in constrained optimization, we devise a novel primal-dual stochastic approximation algorithm which attains the optimal convergence rate of $O(T^{-1/2})$ for general Lipschitz continuous objectives.\",\n \"Title: Exploitation and exploration in a performance based contextual advertising system\\nYear: 2010\\n\\nAbstract: The dynamic marketplace in online advertising calls for ranking systems that are optimized to consistently promote and capitalize better performing ads. The streaming nature of online data inevitably makes an advertising system choose between maximizing its expected revenue according to its current knowledge in short term (exploitation) and trying to learn more about the unknown to improve its knowledge (exploration), since the latter might increase its revenue in the future. The exploitation and exploration (EE) tradeoff has been extensively studied in the reinforcement learning community, however, not been paid much attention in online advertising until recently. In this paper, we develop two novel EE strategies for online advertising. Specifically, our methods can adaptively balance the two aspects of EE by automatically learning the optimal tradeoff and incorporating confidence metrics of historical performance. Within a deliberately designed offline simulation framework we apply our algorithms to an industry leading performance based contextual advertising system and conduct extensive evaluations with real online event log data. The experimental results and detailed analysis reveal several important findings of EE behaviors in online advertising and demonstrate that our algorithms perform superiorly in terms of ad reach and click-through-rate (CTR).\",\n \"Title: Batch mode active learning and its application to medical image classification\\nYear: 2006\\n\\nAbstract: The goal of active learning is to select the most informative examples for manual labeling. Most of the previous studies in active learning have focused on selecting a single unlabeled example in each iteration. This could be inefficient since the classification model has to be retrained for every labeled example. In this paper, we present a framework for \\\"batch mode active learning\\\" that applies the Fisher information matrix to select a number of informative examples simultaneously. The key computational challenge is how to efficiently identify the subset of unlabeled examples that can result in the largest reduction in the Fisher information. To resolve this challenge, we propose an efficient greedy algorithm that is based on the property of submodular functions. Our empirical studies with five UCI datasets and one real-world medical image classification show that the proposed batch mode active learning algorithm is more effective than the state-of-the-art algorithms for active learning.\",\n \"Title: Online Kernel Learning with a Near Optimal Sparsity Bound.\\nYear: 2013\\n\\nAbstract: In this work, we focus on Online Sparse Kernel Learning that aims to online learn a kernel classifier with a bounded number of support vectors. Although many online learning algorithms have been proposed to learn a sparse kernel classifier, most of them fail to bound the number of support vectors used by the final solution which is the average of the intermediate kernel classifiers generated by online algorithms. The key idea of the proposed algorithm is to measure the difficulty in correctly classifying a training example by the derivative of a smooth loss function, and give a more chance to a difficult example to be a support vector than an easy one via a sampling scheme. Our analysis shows that when the loss function is smooth, the proposed algorithm yields similar performance guarantee as the standard online learning algorithm but with a near optimal number of support vectors (up to a poly(lnT) factor). Our empirical study shows promising performance of the proposed algorithm compared to the state-of-the-art algorithms for online sparse kernel learning.\",\n \"Title: Identifying neuronal assemblies with local and global connectivity with scale space spectral clustering\\nYear: 2007\\n\\nAbstract: A nonparametric approach is proposed to identify clusters of functionally interdependent neurons, independent of the time scale at which they are maximally correlated. The neural point processes are represented in a N-dimensional scale space using the Haar wavelet transform. A similarity measure between any given pair of neurons is defined in the scale space. Clusters of ''similar'' neurons are identified by first reducing the N-dimensional scale space representation using principal components to obtain a Q-dimensional space. The weighted principal components are subsequently used to connect each neuron to the others in a graph representation. A probabilistic spectral clustering algorithm is used to perform graph partitioning by maximizing cluster compactness. Performance is compared to that of the k-means and the expectation-maximization algorithms for 120 neurons with time-varying intensity functions consisting of spontaneous background activity and phased response elicited at distinct time scales.\",\n \"Title: Machine learning for information retrieval\\nYear: 2011\\n\\nAbstract: In recent years, we have witnessed successful application of machine learning techniques to a wide range of information retrieval problems, including Web search engines, recommendation systems, online advertising, etc. It is thus critical for researchers in the information retrieval community to understand the core machine learning techniques. In order to accommodate audiences with different levels of understanding of machine learning, we divide this tutorial into two sessions: the first session will focus on basic machine learning concepts and tools; in the second session, we will introduce more advanced topics in machine learning, and will present recent developments in machine learning and its application to information retrieval. Each season is self-contained. Session 1: Core Learning Technologies for Information Retrieval. This session of the tutorial will cover the core machine learning methods, basic optimization techniques and key information retrieval applications. In particular, it includes: 1). Core concepts in machine learning, such as supervised learning/unsupervised learning, bias and variance trade off, and probabilistic models; 2). Useful concepts and algorithms in optimization including the first and second order gradient methods, and Expectation and Maximization; 3). The application of machine learning methods to key information retrieval problems including text classification, collaborative filtering, clustering and learning to rank; Session 2: Emerging Learning Technologies for Information Retrieval. This session will cover more advanced machine learning techniques that have started to be utilized in information retrieval applications. In particular, it will cover: 1). Advanced Optimization Techniques including stochastic optimization and smooth minimization; 2). Emerging Learning Techniques such as Multiple-Instance Learning, Active Learning and Semi-supervised Learning. The tutorial will benefit a large body of audience in the information retrieval community, ranging from students who are new to machine learning to the seasoned researchers who would like to understand the recent advance in machine learning for information retrieval research. This tutorial will also benefit the practitioners who apply learning techniques to real-world information retrieval systems.\",\n \"Title: Multi-label Multiple Kernel Learning\\nYear: 2008\\n\\nAbstract: We present a multi-label multiple kernel learning (MKL) formulation in which the data are embedded into a low-dimensional space directed by the instance- label correlations encoded into a hypergraph. We formulate the problem in the kernel-induced feature space and propose to learn the kernel matrix as a linear combination of a given collection of kernel matrices in the MKL framework. The proposed learning formulation leads to a non-smooth min-max problem, which can be cast into a semi-infinite linear program (SILP). We further propose an ap- proximate formulation with a guaranteed error bound which involves an uncon- strained convex optimization problem. In addition, we show that the objective function of the approximate formulation is differentiable with Lipschitz continu- ous gradient, and hence existing methods can be employed to compute the optimal solution efficiently. We apply the proposed formulation to the automated annota- tion of Drosophila gene expression pattern images, and promising results have been reported in comparison with representative algorithms.\",\n \"Title: Symmetric Statistical Translation Models for Automatic Image Annotation\\nYear: 2005\\n\\nAbstract: Automatic image annotation provides means for users to search image collections on the semantic level using natural language queries. In the past, statistical machine translation models have been successfully applied to automatic image annotation. A problem with this approach is that, due to the skewed distribution of term frequency for annotation words, common words have been overly favored, which leaves little room for uncommon words to be used in auto-annotations. In contrast, studies on information retrieval have revealed that uncommon words are at least as important as common words since they are also often used in users' queries. Unlike the previous studies where a single type of statistical translation model is considered for automatic image annotation, in this paper, we studied two types of statistical translation models: a forward translation model, which translates visual information into textual words, and a backward model, which translates textual words into visual images. In particular, we propose a new statistical translation model, named regularization-based symmetric statistical translation model, which combines strength of forward and backward models to alleviate the problem of overly favoring common words. Our empirical studies with the Corel dataset have shown that the proposed model performs considerably better than the existing translation model and a state-of-the-art approach for automatic image annotation.\",\n \"Title: Large-scale near-duplicate image retrieval by kernel density estimation.\\nYear: 2012\\n\\nAbstract: Bag-of-words model is one of the most widely used methods in the recent studies of multimedia data retrieval. The key idea of the bag-of-words model is to quantize the bag of local features, for example SIFT, to a histogram of visual words and then standard information retrieval technologies developed from text retrieval can be applied directly. Despite its success, one problem of the bag-of-words model is that the two key steps, i.e., feature quantization and retrieval, are separated. In other words, the step of generating bag-of-words representation is not optimized for the step of retrieval which often leads to a sub-optimal performance. In this paper we propose a statistical framework for large-scale near-duplication image retrieval which unifies the two steps by introducing kernel density function. The central idea of the proposed method is to represent each image by a kernel density function and the similarity between the query image and a database image is then estimated as the query likelihood. In order to make the proposed method applicable to large-scale data sets, we have developed efficient algorithms for both estimating the density function of each image and computing the query likelihood. Our empirical studies confirm that the proposed method is not only more effective but also more efficient than the bag-of-words model.\",\n \"Title: Flexible Mixture Model for Collaborative Filtering\\nYear: 2003\\n\\nAbstract: This paper presents a flexible mixture model (FMM) for collaborative filtering. FMM extends existing partitioning/clustering algorithms for collaborative filtering by clustering both users and items together simultaneously without assuming that each user and item should only belong to a single cluster. Furthermore, with the introduction of 'preference' nodes, the proposed framework is able to explicitly model how users rate items, which can vary dramatically, even among the users with similar tastes on items. Empirical study over two datasets of movie ratings has shown that our new algorithm outperforms five other collaborative filtering algorithms substantially.\",\n \"Title: Using knowledge driven matrix factorization to reconstruct modular gene regulatory network\\nYear: 2008\\n\\nAbstract: Reconstructing gene networks from micro-array data can provide information on the mechanisms that govern cellular processes. Numerous studies have been devoted to addressing this problem. A popular method is to view the gene network as a Bayesian inference network, and to apply structure learning methods to determine the topology of the gene network. There are, however, several shortcomings with the Bayesian structure learning approach for reconstructing gene networks. They include high computational cost associated with analyzing a large number of genes and inefficiency in exploiting prior knowledge of co-regulation that could be derived from Gene Ontology (GO) information. In this paper, we present a knowledge driven matrix factorization (KMF) framework for reconstructing modular gene networks that addresses these shortcomings. In KMF, gene expression data is initially used to estimate the correlation matrix. The gene modules and the interactions among the modules are derived by factorizing the correlation matrix. The prior knowledge in GO is integrated into matrix factorization to help identify the gene modules. An alternating optimization algorithm is presented to efficiently find the solution. Experiments show that our algorithm performs significantly better in identifying gene modules than several state-of-the-art algorithms, and the interactions among the modules uncovered by our algorithm are proved to be biologically meaningful.\",\n \"Title: An efficient primal dual prox method for non-smooth optimization\\nYear: 2012\\n\\nAbstract: We study the non-smooth optimization problems in machine learning, where both the loss function and the regularizer are non-smooth functions. Previous studies on efficient empirical loss minimization assume either a smooth loss function or a strongly convex regularizer, making them unsuitable for non-smooth optimization. We develop a simple yet efficient method for a family of non-smooth optimization problems where the dual form of the loss function is bilinear in primal and dual variables. We cast a non-smooth optimization problem into a minimax optimization problem, and develop a primal dual prox method that solves the minimax optimization problem at a rate of $$O(1/T)$$O(1/T) assuming that the proximal step can be efficiently solved, significantly faster than a standard subgradient descent method that has an $$O(1/\\\\sqrt{T})$$O(1/T) convergence rate. Our empirical studies verify the efficiency of the proposed method for various non-smooth optimization problems that arise ubiquitously in machine learning by comparing it to the state-of-the-art first order methods.\",\n \"Title: Video Retrieval with the Informedia Digital Video Library System\\nYear: 2001\\n\\nAbstract: Background: The Informedia Digital Video Library System. The Informedia Digital Video Library [1] was the only NSF DLI project focusing specifically on information extraction from video and audio content. Over a terabyte of online data was collected, with automatically generated metadata and indices for retrieving videos from this library. The architecture for the project was based on the premise that real-time constraints on library and associated metadata creation could be relaxed in order to realize increased automation and deeper parsing and indexing for identifying the library contents and breaking it into segments. Library creation was an offline activity, with library exploration by users occurring online and making use of the generated metadata and segmentation. The goal of the Informedia interface was to enable quick access to relevant information in a digital video library, leveraging from derived metadata and the partitioning of the video into small segments. Figure 1 shows the IDVLS interface following a query. In this figure, a set of results is displayed at the bottom. The display includes a window containing a headline, and a pictorial menu of video segments each represented with a thumbnail image at approximately ? resolution of the video in the horizontal and vertical dimensions. The headline window automatically pops up whenever the mouse is positioned over a result item; the headline window for the first result is shown. IDVLS also supports other ways of navigating and browsing the digital video library. These interface features were essential to deal with the ambiguity of the derived data generated by speech recognition, image processing, and natural language processing. Consider the filmstrip and video playback IDVLS window shown in Figure 2.For this actual video in the IDVLS library, the segmentation process failed, resulting in a thirty-minute segment. This long segment was one of the returned results for the query \\\"Mir collision. \\\" The filmstrip in Figure 2 shows that the segment is more than just a story on the Russian space station, but rather begins with a commercial, then the weather, and then coverage of Hong Kong before addressing Mir. By overlaying the filmstrip and video playback windows with match location information, the user can quickly see that matches don't occur until later in the segment, after these other stories that were irrelevant to the query. The match bars are optionally color-coded to specific query words; in Figure 2 \\\"Mir\\\" matches are in red and \\\"collision\\\" matches in purple. When the user moved the mouse over the match bars in the filmstrip, a text window displayed the actual matching word from the transcript or Video OCR metadata for that particular match; \\\"Mir\\\" is shown in one such text window in Figure 2.By investigating the distribution of match locations on the filmstrip, the user can determine the relevance of the returned result and the location of interest within the segment. The user can click on a match bar to jump directly to that point in the video segment. Hence, clicking the mouse as shown in Figure 2 would start playing the video at this mention of \\\"Mir\\\" with the overhead shot of people at desks. Similarly, IDVLS provided \\\"seek to next match\\\" and \\\"seek to previous match\\\" buttons in the video player allowing the user to quickly jump from one match to the next. In the example of Figure 2, these interface features allowed the user to bypass problems in segmentation and jump directly to the \\\"Mir\\\" story without having to first watch the opening video on other topics.\",\n \"Title: A maximum coherence model for dictionary-based cross-language information retrieval\\nYear: 2005\\n\\nAbstract: One key to cross-language information retrieval is how to efficiently resolve the translation ambiguity of queries given their short length. This problem is even more challenging when only bilingual dictionaries are available, which is the focus of this paper. In the previous research of cross-language information retrieval using bilingual dictionaries, the word co-occurrence statistics is used to determine the most likely translations of queries. In this paper, we propose a novel statistical model, named ``maximum coherence model'', which estimates the translation probabilities of query words that are consistent with the word co-occurrence statistics. Unlike the previous work, where a binary decision is made for the selection of translations, the new model maintains the uncertainty in translating query words when their sense ambiguity is difficult to resolve. Furthermore, this new model is able to estimate translations of multiple query words simultaneously. This is in contrast to many previous approaches where translations of individual query words are determined independently. Empirical studies with TREC datasets have shown that the maximum coherence model achieves a relative 10% - 40% improvement in cross-language information retrieval, comparing to other approaches that also use word co-occurrence statistics for sense disambiguation.\",\n \"Title: A Probabilistic Approach for Optimizing Spectral Clustering\\nYear: 2005\\n\\nAbstract: Spectral clustering enjoys its success in both data cluster ing and semi- supervised learning. But, most spectral clustering algori thms cannot handle multi-class clustering problems directly. Additio nal strategies are needed to extend spectral clustering algorithms to multi-class cluster- ing problems. Furthermore, most spectral clustering algorithms employ hard cluster membership, which is likely to be trapped by the local op- timum. In this paper, we present a new spectral clustering algorithm, named \\\"Soft Cut\\\". It improves the normalized cut algorithm by intro- ducing soft membership, and can be efficiently computed usin g a bound optimization algorithm. Our experiments with a variety of datasets have shown the promising performance of the proposed clustering algorithm.\",\n \"Title: Query translation disambiguation as graph partitioning\\nYear: 2005\\n\\nAbstract: Resolving ambiguity in the process of query translation is crucial to cross-language information retrieval when only a bilingual dictionary is available. In this paper we propose a novel approach for query translation disambiguation, named \\\"spectral query translation model\\\". The proposed approach views the problem of query translation disambiguation as a graph partitioning problem. For a given query, a weighted graph is first created for all possible translations of query words based on the co-occurrence statistics of the translation words. The best translation of the query is then determined by the most strongly connected component within the graph. The proposed approach distinguishes from previous approaches in that the translations of all query words are estimated simultaneously. Furthermore, translation probabilities are introduced in the proposed approach to capture the uncertainty in translating queries. Empirical studies with TREC datasets have shown that the spectral query translation model achieves a relative 20% -50% improvement in cross-language information retrieval, compared to other approaches that also exploit word co-occurrence statistics for query translation disambiguation.\",\n \"Title: A Bayesian reconstruction algorithm for synthesis aperture imaging radiometer\\nYear: 2012\\n\\nAbstract: Regularization methods are very efficient in reconstructing the radiometric brightness temperature maps form interferometric measurements. The performance of these methods depends on selecting an optimal regularization parameter, which is typically defined by using the cross-validation or the L-curve. In this paper, we present a novel brightness temperature reconstruction method, which can automatically learn the optimal parameter by Bayesian learning without reducing the performance of the brightness temperature image. To support the theory, numerical simulations are presented and analyzed with emphasis on stability and error analysis.\",\n \"Title: The Research on Teaching Method of Basics Course of Computer based on Cluster Analysis\\nYear: 2010\\n\\nAbstract: Due to the difference between rural and urban and regional disparity of computer basic education produced at their middle school period, computer application level of the freshmen make a great difference. According to this situation, corresponding skill survey is designed. Data obtained from the survey is analyzed by K-means cluster algorithm. And the students are classified into three types according to their master degree of computer application. The study group is composed of the students selected from each type. This can promote them to study from each other and draw the strong points of others to offset their own weakness. The teachers can adjust the emphasis of their teaching to different classes according to the master degree of computer application of the classes. Good teaching effect is achieved by using this teaching method in Basics Course of Computer.\",\n \"Title: Unsupervised transfer classification: application to text categorization\\nYear: 2010\\n\\nAbstract: We study the problem of building the classification model for a target class in the absence of any labeled training example for that class. To address this difficult learning problem, we extend the idea of transfer learning by assuming that the following side information is available: (i) a collection of labeled examples belonging to other classes in the problem domain, called the auxiliary classes; (ii) the class information including the prior of the target class and the correlation between the target class and the auxiliary classes. Our goal is to construct the classification model for the target class by leveraging the above data and information. We refer to this learning problem as unsupervised transfer classification. Our framework is based on the generalized maximum entropy model that is effective in transferring the label information of the auxiliary classes to the target class. A theoretical analysis shows that under certain assumption, the classification model obtained by the proposed approach converges to the optimal model when it is learned from the labeled examples for the target class. Empirical study on text categorization over four different data sets verifies the effectiveness of the proposed approach.\",\n \"Title: Learn to weight terms in information retrieval using category information\\nYear: 2005\\n\\nAbstract: How to assign appropriate weights to terms is one of the critical issues in information retrieval. Many term weighting schemes are unsupervised. They are either based on the empirical observation in information retrieval, or based on generative approaches for language modeling. As a result, the existing term weighting schemes are usually insufficient in distinguishing informative words from the uninformative ones, which is crucial to the performance of information retrieval. In this paper, we present supervised term weighting schemes that automatically learn term weights based on the correlation between word frequency and category information of documents. Empirical studies with the ImageCLEF dataset have indicated that the proposed methods perform substantially better than the state-of-the-art approaches for term weighting and other alternatives that exploit category information for information retrieval.\",\n \"Title: Efficient Convex Relaxation for Transductive Support Vector Machine\\nYear: 2007\\n\\nAbstract: We consider the problem of Support Vector Machine transduction, which involves a combinatorial problem with exponential computational complexity in the num- ber of unlabeled examples. Although several studies are devoted to Transductive SVM, they suffer either from the high computation complexity or from the so- lutions of local optimum. To address this problem, we propose solving Trans- ductive SVM via a convex relaxation, which converts the NP-hard problem to a semi-definite programming. Compared with the other SDP relaxation for Trans- ductive SVM, the proposed algorithm is computationally more efficient with the number of free parameters reduced from O(n2) to O(n) where n is the number of examples. Empirical study with several benchmark data sets shows the promising performance of the proposed algorithm in comparison with other state-of-the-art implementations of Transductive SVM.\",\n \"Title: A Robust Regularization Kernel Regression Algorithm for Passive Millimeter Wave Imaging Target Detection.\\nYear: 2012\\n\\nAbstract: This letter deals with small target detection in passive millimeter wave (PMMW) imaging. Specifically, it focuses on a general detection scheme, where, first, the background is suppressed through a background prediction algorithm, and then the detection is accomplished. A precise prediction of the background is essential to a successful outcome. In practical applications, background estimation pro...\",\n \"Title: Title generation for machine-translated documents\\nYear: 2001\\n\\nAbstract: In this paper, we present and compare automatically generated titles for machine-translated documents using several different statistics-based methods. A Naïve Bayesian, a K-Nearest Neighbour, a TF-IDF and an iterative Expectation-Maximization method for title generation were applied to 1000 original English news documents and again to the same documents translated from English into Portuguese, French or German and back to English using SYSTRAN. The AutoSummarization function of Microsoft Word was used as a base line. Results on several metrics show that the statistics-based methods of title generation for machine-translated documents are fairly language independent and title generation is possible at a level approaching the accuracy of titles generated for the original English documents.\",\n \"Title: Stochastic Optimization of Smooth Loss.\\nYear: 2013\\n\\nAbstract: In this paper, we first prove a high probability bound rather than an expectation bound for stochastic optimization with smooth loss. Furthermore, the existing analysis requires the knowledge of optimal classifier for tuning the step size in order to achieve the desired bound. However, this information is usually not accessible in advanced. We also propose a strategy to address the limitation.\",\n \"Title: Semi-supervised Collaborative Text Classification\\nYear: 2007\\n\\nAbstract: Most text categorization methods require text content of documents that is often difficult to obtain. We consider \\\"Collaborative Text Categorization\\\", where each document is represented by the feedback from a large number of users. Our study focuses on the semi-supervised case in which one key challenge is that a significant number of users have not rated any labeled document. To address this problem, we examine several semi-supervised learning methods and our empirical study shows that collaborative text categorization is more effective than content-based text categorization and the manifold regularization is more effective than other state-of-the-art semi-supervised learning methods.\",\n \"Title: A Bayesian Approach Toward Finding Communities and Their Evolutions in Dynamic Social Networks\\nYear: 2009\\n\\nAbstract: Although a large body of work are devoted to finding communities in static social networks, only a few studies examined the dynamics of communities in evolving social networks. In this paper, we propose a dynamic stochastic block model for finding communities and their evolutions in a dynamic social network. The proposed model captures the evolution of communities by explicitly modeling the transition of community memberships for individual nodes in the network. Unlike many existing approaches for modeling social networks that estimate parameters by their most likely values (i.e., point estimation), in this study, we employ a Bayesian treatment for parameter estimation that computes the posterior distributions for all the unknown parameters. This Bayesian treatment allows us to capture the uncertainty in parameter values and therefore is more robust to data noise than point estimation. In addition, an efficient algorithm is developed for Bayesian inference to handle large sparse social networks. Extensive experimental studies based on both synthetic data and real-life data demonstrate that our model achieves higher accuracy and reveals more insights in the data than several state-of-the- art algorithms.\",\n \"Title: Using a probabilistic source model for comparing images\\nYear: 2002\\n\\nAbstract: We propose a probabilistic model for image retrieval. To obtain the similarity between the query image IQ and any image I' in the collection, the model computes the probability of generating the image I' given the observation of the query image IQ. We compare our probabilistic model for image retrieval with a color histogram based image retrieval method and the IBM QBIC image search engine. The evaluation used the 11-hour video retrieval collection (80,000 extracted images) and associated queries from the 2001 TREC-10 information retrieval evaluations. The experimental results show that the probabilistic model dramatically outperforms the color histogram based image retrieval method and the IBM QBIC image search engine by 40%.\",\n \"Title: Title Generation Using a Training Corpus\\nYear: 2001\\n\\nAbstract: This paper discusses fundamental issues involved in word selection for title generation. We review several methods for title generation, namely extractive summarization and two versions of a Naïve Bayesian, and compare the performance of those methods using an F1 metric. In addition, we introduce a novel approach to title generation using the k-nearest neighbor (KNN) algorithm. Both the KNN method and a limited-vocabulary Naïve Bayesian method outperform the other evaluated methods with an F1 score of around 20%. Since KNN produces complete and legible titles, we conclude that KNN is a very promising method for title generation, provided good content overlap exists between the training corpus and the test documents.\",\n \"Title: Tag Completion for Image Retrieval.\\nYear: 2013\\n\\nAbstract: Many social image search engines are based on keyword/tag matching. This is because tag based image retrieval (TBIR) is not only efficient but also effective. The performance of TBIR is highly dependent on the availability and quality of manual tags. Recent studies have shown that manual tags are often unreliable and inconsistent. In addition, since many users tend to choose general and ambiguous tags in order to minimize their efforts in choosing appropriate words, tags that are specific to the visual content of images tend to be missing or noisy, leading to a limited performance of TBIR. To address this challenge, we study the problem of \\\\textit{tag completion} where the goal is to automatically fill in the missing tags as well as correct noisy tags for given images. We represent the image-tag relation by a \\\\textit{tag matrix}, and search for the optimal tag matrix consistent with both the observed tags and the visual similarity. We propose an efficient algorithm for solving the related optimization problem. Extensive empirical studies show that the proposed algorithm is significantly more effective than the state-of-the-art algorithms. Our studies also verify that the proposed algorithm is computationally efficient and scales well to large databases.\",\n \"Title: Correlated Label Propagation with Application to Multi-label Learning\\nYear: 2006\\n\\nAbstract: Many computer vision applications, such as scene analysis and medical image interpretation, are ill-suited for traditional classification where each image can only be associated with a single class. This has stimulated recent work in multi-label learning where a given image can be tagged with multiple class labels. A serious problem with existing approaches is that they are unable to exploit correlations between class labels. This paper presents a novel framework for multi-label learning termed Correlated Label Propagation (CLP) that explicitly models interactions between labels in an efficient manner. As in standard label propagation, labels attached to training data points are propagated to test data points; however, unlike standard algorithms that treat each label independently, CLP simultaneously co-propagates multiple labels. Existing work eschews such an approach since naive algorithms for label co-propagation are intractable. We present an algorithm based on properties of submodular functions that efficiently finds an optimal solution. Our experiments demonstrate that CLP leads to significant gains in precision/recall against standard techniques on two real-world computer vision tasks involving several hundred labels.\",\n \"Title: Inferring functional cortical networks from spike train ensembles using Dynamic Bayesian Networks\\nYear: 2009\\n\\nAbstract: A fundamental goal in systems neuroscience is to infer the functional connectivity among neuronal elements coordinating information processing in the brain. In this work, we investigate the applicability of Dynamic Bayesian Networks (DBN) in inferring the structure of cortical networks from the observed spike trains. DBNs have unique features that make them capable of detecting causal relationships between spike trains such as modeling time-dependent relationships, detecting non-linear interactions and inferring connectivity between neurons from the observed ensemble activity. A probabilistic point process model was used to assess the performance under systematic variations of the model parameters. Results demonstrate the utility of DBN in inferring functional connectivity in cortical network models.\",\n \"Title: On predicting rare classes with SVM ensembles in scene classification\\nYear: 2003\\n\\nAbstract: Scene classification is an important technique to infer high-level semantic scene categories from low-level visual features. However, in the real world the positive data for many scenes may be rare, which degrades the performance of many classifiers. In this paper, we propose SVM ensembles to address the rare class problem. Various classifier combination strategies are investigated, including majority voting, sum rule, neural network gater and hierarchical SVMs. We also compare our method with two other common approaches for dealing with the rare class problem. Our experimental results show that hierarchical SVMs can achieve significantly better and more stable performance than other strategies, as well as high computational efficiency.\",\n \"Title: BoostCluster: boosting clustering by pairwise constraints\\nYear: 2007\\n\\nAbstract: Data clustering is an important task in many disciplines. A large number of studies have attempted to improve clustering by using the side information that is often encoded as pairwise constraints. However, these studies focus on designing special clustering algorithms that can effectively exploit the pairwise constraints. We present a boosting framework for data clustering,termed as BoostCluster, that is able to iteratively improve the accuracy of any given clustering algorithm by exploiting the pairwise constraints. The key challenge in designing a boosting framework for data clustering is how to influence an arbitrary clustering algorithm with the side information since clustering algorithms by definition are unsupervised. The proposed framework addresses this problem by dynamically generating new data representations at each iteration that are, on the one hand, adapted to the clustering results at previous iterations by the given algorithm, and on the other hand consistent with the given side information. Our empirical study shows that the proposed boosting framework is effective in improving the performance of a number of popular clustering algorithms (K-means, partitional SingleLink, spectral clustering), and its performance is comparable to the state-of-the-art algorithms for data clustering with side information.\",\n \"Title: Sensor localization in an obstructed environment\\nYear: 2005\\n\\nAbstract: Sensor localization can be divided into two categories: range-based approaches, and rang-free approaches. Although range-based approaches tend to be more accurate than range-free approaches, they are more sensitive to errors in distance measurement. Despite of the efforts on recovering sensors’ Euclidean coordinates from erroneous distance measurements, as will be illustrated in this paper, they are still prone to distance errors, particularly in an obstruction abundant environment. In this paper, we propose a new algorithm for sensor localization based on Multiscale Radio Transmission Power(MRTP). It gradually increases the scale level of transmission power, and the distance is determined by the minimal scale of received signals. Unlike the range-based approaches, which treat each measured distance as an approximation of the true one, in our new approach, the measured distance serves as a constraint that limits the feasible location of sensors. Our simulations have shown that the MRTP-based approach is able to provide accurate and robust estimation of location, especially in an area abundant in obstructions, where most current approaches fail to perform well.\",\n \"Title: A study of methods for normalizing user ratings in collaborative filtering\\nYear: 2004\\n\\nAbstract: The goal of collaborative filtering is to make recommendations for a test user by utilizing the rating information of users who share interests similar to the test user. Because ratings are determined not only by user interests but also the rating habits of users, it is important to normalize ratings of different users to the same scale. In this paper, we compare two different normalization strategies for user ratings, namely the Gaussian normalization method and the decoupling normalization method. Particularly, we incorporated these two rating normalization methods into two collaborative filtering algorithms, and evaluated their effectiveness on the EachMovie dataset. The experiment results have shown that the decoupling method for rating normalization is more effective than the Gaussian normalization method in improving the performance of collaborative filtering algorithms.\",\n \"Title: Online feature selection for mining big data\\nYear: 2012\\n\\nAbstract: Most studies of online learning require accessing all the attributes/features of training instances. Such a classical setting is not always appropriate for real-world applications when data instances are of high dimensionality or the access to it is expensive to acquire the full set of attributes/features. To address this limitation, we investigate the problem of Online Feature Selection (OFS) in which the online learner is only allowed to maintain a classifier involved a small and fixed number of features. The key challenge of Online Feature Selection is how to make accurate prediction using a small and fixed number of active features. This is in contrast to the classical setup of online learning where all the features are active and can be used for prediction. We address this challenge by studying sparsity regularization and truncation techniques. Specifically, we present an effective algorithm to solve the problem, give the theoretical analysis, and evaluate the empirical performance of the proposed algorithms for online feature selection on several public datasets. We also demonstrate the application of our online feature selection technique to tackle real-world problems of big data mining, which is significantly more scalable than some well-known batch feature selection algorithms. The encouraging results of our experiments validate the efficacy and efficiency of the proposed techniques for large-scale applications.\",\n \"Title: Design and Implementation of an Open Programmable Router Compliant to IETF ForCES Specifications\\nYear: 2007\\n\\nAbstract: IETF ForCES (Forwarding and Control Element Separation) is defining specifications for interfaces and modular resources abstractions of network equipments. ForCES brings flexible, programmable, and cost-effective advantages over tradi-tional architectures for network equipments. In this paper, the design and implementation of a ForCES-based router (ForTER) was illustrated. The framework of ForCES was introduced firstly. Then a layered model for the implementation of ForTER was presented. Next, the implementation details of Control Ele-ment (CE) and Forwarding Element (FE) of ForTER were intro-duced, which complied with the ForCES framework and related IETF ForCES specifications. Finally, results of experiments for routing and for SNMP support showed the feasibility of the sys-tem architecture of ForTER, which also fundamentally testified the feasibility of ForCES and related specifications that are being standardized by IETF ForCES working group.\",\n \"Title: A Bayesian framework for community detection integrating content and link\\nYear: 2012\\n\\nAbstract: This paper addresses the problem of community detection in networked data that combines link and content analysis. Most existing work combines link and content information by a generative model. There are two major shortcomings with the existing approaches. First, they assume that the probability of creating a link between two nodes is determined only by the community memberships of the nodes; however other factors(e.g. popularity) could also affect the link pattern. Second, they use generative models to model the content of individual nodes, whereas these generative models are vulnerable to the content attributes that are irrelevant to communities. We propose a Bayesian framework for combining link and content information for community detection that explicitly addresses these shortcomings. A new link model is presented that introduces a random variable to capture the node popularity when deciding the link between two nodes; a discriminative model is used to determine the community membership of a node by its content. An approximate inference algorithm is presented for efficient Bayesian inference. Our empirical study shows that the proposed framework outperforms several state-of-the-art approaches in combining link and content information for community detection.\",\n \"Title: A framework for incorporating class priors into discriminative classification\\nYear: 2005\\n\\nAbstract: Discriminative and generative methods provide two distinct approaches to machine learning classification. One advantage of generative approaches is that they naturally model the prior class distributions. In contrast, discriminative approaches directly model the conditional distribution of class given inputs, so the class priors are only implicitly obtained if the input density is known. In this paper, we propose a framework for incorporating class prior proportions into discriminative methods in order to improve their classification accuracy. The basic idea is to enforce that the distribution of class labels predicted on the test data by the discriminative model is consistent with the class priors. Therefore, the discriminative model has to not only fit the training data well but also predict class labels for the test data that are consistent with the class priors. Experiments on five different UCI datasets and one image database show that this framework is effective in improving the classification accuracy when the training data and the test data come from the same class proportions, even if the test data does not have exactly the same feature distribution as the training data.\",\n \"Title: A Boosting Framework for Visuality-Preserving Distance Metric Learning and Its Application to Medical Image Retrieval\\nYear: 2010\\n\\nAbstract: Similarity measurement is a critical component in content-based image retrieval systems, and learning a good distance metric can significantly improve retrieval performance. However, despite extensive study, there are several major shortcomings with the existing approaches for distance metric learning that can significantly affect their application to medical image retrieval. In particular, “similarity” can mean very different things in image retrieval: resemblance in visual appearance (e.g., two images that look like one another) or similarity in semantic annotation (e.g., two images of tumors that look quite different yet are both malignant). Current approaches for distance metric learning typically address only one goal without consideration of the other. This is problematic for medical image retrieval where the goal is to assist doctors in decision making. In these applications, given a query image, the goal is to retrieve similar images from a reference library whose semantic annotations could provide the medical professional with greater insight into the possible interpretations of the query image. If the system were to retrieve images that did not look like the query, then users would be less likely to trust the system; on the other hand, retrieving images that appear superficially similar to the query but are semantically unrelated is undesirable because that could lead users toward an incorrect diagnosis. Hence, learning a distance metric that preserves both visual resemblance and semantic similarity is important. We emphasize that, although our study is focused on medical image retrieval, the problem addressed in this work is critical to many image retrieval systems. We present a boosting framework for distance metric learning that aims to preserve both visual and semantic similarities. The boosting framework first learns a binary representation using side information, in the form of labeled pairs, and then computes the distance as a weighted Hamming distance using the learned binary representation. A boosting algorithm is presented to efficiently learn the distance function. We evaluate the proposed algorithm on a mammographic image reference library with an Interactive Search-Assisted Decision Support (ISADS) system and on the medical image data set from ImageCLEF. Our results show that the boosting framework compares favorably to state-of-the-art approaches for distance metric learning in retrieval accuracy, with much lower computational cost. Additional evaluation with the COREL collection shows that our algorithm works well for regular image data sets.\",\n \"Title: Active multiple kernel learning for interactive 3D object retrieval systems\\nYear: 2011\\n\\nAbstract: An effective relevance feedback solution plays a key role in interactive intelligent 3D object retrieval systems. In this work, we investigate the relevance feedback problem for interactive intelligent 3D object retrieval, with the focus on studying effective machine learning algorithms for improving the user's interaction in the retrieval task. One of the key challenges is to learn appropriate kernel similarity measure between 3D objects through the relevance feedback interaction with users. We address this challenge by presenting a novel framework of Active multiple kernel learning (AMKL), which exploits multiple kernel learning techniques for relevance feedback in interactive 3D object retrieval. The proposed framework aims to efficiently identify an optimal combination of multiple kernels by asking the users to label the most informative 3D images. We evaluate the proposed techniques on a dataset of over 10,000 3D models collected from the World Wide Web. Our experimental results show that the proposed AMKL technique is significantly more effective for 3D object retrieval than the regular relevance feedback techniques widely used in interactive content-based image retrieval, and thus is promising for enhancing user's interaction in such interactive intelligent retrieval systems.\",\n \"Title: Ranking refinement and its application to information retrieval\\nYear: 2008\\n\\nAbstract: We consider the problem of ranking refinement, i.e., to improve the accuracy of an existing ranking function with a small set of labeled instances. We are, particularly, interested in learning a better ranking function using two complementary sources of information, ranking information given by the existing ranking function (i.e., a base ranker) and that obtained from users' feedbacks. This problem is very important in information retrieval where the feedback is gradually collected. The key challenge in combining the two sources of information arises from the fact that the ranking information presented by the base ranker tends to be imperfect and the ranking information obtained from users' feedbacks tends to be noisy. We present a novel boosting framework for ranking refinement that can effectively leverage the uses of the two sources of information. Our empirical study shows that the proposed algorithm is effective for ranking refinement, and furthermore significantly outperforms the baseline algorithms that incorporate the outputs from the base ranker as an additional feature.\",\n \"Title: Learning a distance metric from multi-instance multi-label data\\nYear: 2009\\n\\nAbstract: Multi-instance multi-label learning (MIML) refers to the learning problems where each example is represented by a bag/collection of instances and is labeled by multiple labels. An example application of MIML is visual object recognition in which each image is represented by multiple key points (i.e., instances) and is assigned to multiple object categories. In this paper, we study the problem of learning a distance metric from multi-instance multi-label data. It is significantly more challenging than the conventional setup of distance metric learning because it is difficult to associate instances in a bag with its assigned class labels. We propose an iterative algorithm for MIML distance metric learning: it first estimates the association between instances in a bag and its assigned class labels, and learns a distance metric from the estimated association by a discriminative analysis; the learned metric will be used to update the association between instances and class labels, which is further used to improve the learning of distance metric. We evaluate the proposed algorithm by the task of automated image annotation, a well known MIML problem. Our empirical study shows an encouraging result when combining the proposed algorithm with citation-kNN, a state-of-the-art algorithm for multi-instance learning.\",\n \"Title: Automated Vocabulary Acquisition and Interpretation in Multimodal Conversational Systems\\nYear: 2007\\n\\nAbstract: Motivated by psycholinguistic findings that eye gaze is tightly linked to human lan- guage production, we developed an unsuper- vised approach based on translation models to automatically learn the mappings between words and objects on a graphic display dur- ing human machine conversation. The ex- perimental results indicate that user eye gaze can provide useful information to establish such mappings, which have important impli- cations in automatically acquiring and inter- preting user vocabularies for conversational systems.\",\n \"Title: Efficient Algorithm for Localized Support Vector Machine\\nYear: 2010\\n\\nAbstract: This paper presents a framework called localized support vector machine (LSVM) for classifying data with nonlinear decision surfaces. Instead of building a sophisticated global model from the training data, LSVM constructs multiple linear SVMs, each of which is designed to accurately classify a given test example. A major limitation of this framework is its high computational cost since a unique model must be constructed for each test example. To overcome this limitation, we propose an efficient implementation of LSVM, termed profile SVM (PSVM). PSVM partitions the training examples into clusters and builds a separate linear SVM model for each cluster. Our empirical results show that (1) LSVM and PSVM outperform nonlinear SVM for all 20 of the evaluated data sets and (2) PSVM achieves comparable performance as LSVM in terms of model accuracy but with significant computational savings. We also demonstrate the efficacy of the proposed approaches in terms of classifying data with spatial and temporal dependencies.\",\n \"Title: Efficient Kernel Clustering Using Random Fourier Features\\nYear: 2012\\n\\nAbstract: Kernel clustering algorithms have the ability to capture the non-linear structure inherent in many real world data sets and thereby, achieve better clustering performance than Euclidean distance based clustering algorithms. However, their quadratic computational complexity renders them non-scalable to large data sets. In this paper, we employ random Fourier maps, originally proposed for large scale classification, to accelerate kernel clustering. The key idea behind the use of random Fourier maps for clustering is to project the data into a low-dimensional space where the inner product of the transformed data points approximates the kernel similarity between them. An efficient linear clustering algorithm can then be applied to the points in the transformed space. We also propose an improved scheme which uses the top singular vectors of the transformed data matrix to perform clustering, and yields a better approximation of kernel clustering under appropriate conditions. Our empirical studies demonstrate that the proposed schemes can be efficiently applied to large data sets containing millions of data points, while achieving accuracy similar to that achieved by state-of-the-art kernel clustering algorithms.\",\n \"Title: Multi-Task Learning in Square Integrable Space.\\nYear: 2011\\n\\nAbstract: Several kernel based methods for multi-task learning have been proposed, which leverage relations among tasks as regularization to enhance the overall learning accuracies. These methods assume that the tasks share the same kernel, which could limit their applications because in practice different tasks may need different kernels. The main challenge of introducing multiple kernels into multiple tasks is that models from different Reproducing Kernel Hilbert Spaces (RKHSs) are not comparable, making it difficult to exploit relations among tasks. This paper addresses the challenge by formalizing the problem in the Square Integrable Space (SIS). Specially, it proposes a kernel based method which makes use of a regularization term defined in the SIS to represent task relations. We prove a new representer theorem for the proposed approach in SIS. We further derive a practical method for solving the learning problem and conduct consistency analysis of the method. We discuss the relations between our method and an existing method. We also give an SVM based implementation of our method for multi-label classification. Experiments on two real-world data sets show that the proposed method performs better than the existing method. Copyright © 2011, Association for the Advancement of Artificial Intelligence. All rights reserved.\",\n \"Title: A kernel density based approach for large scale image retrieval\\nYear: 2011\\n\\nAbstract: Local image features, such as SIFT descriptors, have been shown to be effective for content-based image retrieval (CBIR). In order to achieve efficient image retrieval using local features, most existing approaches represent an image by a bag-of-words model in which every local feature is quantized into a visual word. Given the bag-of-words representation for images, a text search engine is then used to efficiently find the matched images for a given query. The main drawback with these approaches is that the two key steps, i.e., key point quantization and image matching, are separated, leading to sub-optimal performance in image retrieval. In this work, we present a statistical framework for large-scale image retrieval that unifies key point quantization and image matching by introducing kernel density function. The key ideas of the proposed framework are (a) each image is represented by a kernel density function from which the observed key points are sampled, and (b) the similarity of a gallery image to a query image is estimated as the likelihood of generating the key points in the query image by the kernel density function of the gallery image. We present efficient algorithms for kernel density estimation as well as for effective image matching. Experiments with large-scale image retrieval confirm that the proposed method is not only more effective but also more efficient than the state-of-the-art approaches in identifying visually similar images for given queries from large image databases.\",\n \"Title: Array Configuration Design of One-Dimensional Mirrored Interferometric Aperture Synthesis\\nYear: 2012\\n\\nAbstract: Mirrored interferometric aperture synthesis (IAS) (MIAS) is a novel interferometry which has the potential to reduce system complexity compared with that of conventional IAS. Array configuration design plays a major role in imaging. As the concept of MIAS was proposed just a few years ago, the problem of array configuration design has not been solved yet. In this letter, the principles of array configuration design of 1-D MIAS are proposed, and the corresponding optimization model is established and refined. The optimal array configurations are presented through simulated annealing method and show that 1-D MIAS can achieve almost twice as much of spatial resolution as 1-D IAS with the same array size.\",\n \"Title: Non-parametric mixture models for clustering\\nYear: 2010\\n\\nAbstract: Mixture models have been widely used for data clustering. However, commonly used mixture models are generally of a parametric form (e.g., mixture of Gaussian distributions or GMM), which significantly limits their capacity in fitting diverse multidimensional data distributions encountered in practice.We propose a non-parametric mixture model (NMM) for data clustering in order to detect clusters generated from arbitrary unknown distributions, using non-parametric kernel density estimates. The proposed model is non-parametric since the generative distribution of each data point depends only on the rest of the data points and the chosen kernel. A leave-one-out likelihood maximization is performed to estimate the parameters of the model. The NMM approach, when applied to cluster high dimensional text datasets significantly outperforms the state-of-the-art and classical approaches such as K-means, Gaussian Mixture Models, spectral clustering and linkage methods.\",\n \"Title: Semi-supervised text categorization by active search\\nYear: 2008\\n\\nAbstract: In automated text categorization, given a small number of labeled documents, it is very challenging, if not impossible, to build a reliable classifier that is able to achieve high classification accuracy. To address this problem, a novel web-assisted text categorization framework is proposed in this paper. Important keywords are first automatically identified from the available labeled documents to form the queries. Search engines are then utilized to retrieve from the Web a multitude of relevant documents, which are then exploited by a semi-supervised framework. To our best knowledge, this work is the first study of this kind. Extensive experimental study shows the encouraging results of the proposed text categorization framework: using Google as the web search engine, the proposed framework is able to reduce the classification error by 30% when compared with the state-of-the-art supervised text categorization method.\",\n \"Title: Linguistic theories in efficient multimodal reference resolution: an empirical investigation\\nYear: 2005\\n\\nAbstract: Multimodal conversational interfaces provide a natural means for users to communicate with computer systems through multiple modalities such as speech, gesture, and gaze. To build effective multimodal interfaces, understanding user multimodal inputs is important. Previous linguistic and cognitive studies indicate that user language behavior does not occur randomly, but rather follows certain linguistic and cognitive principles. Therefore, this paper investigates the use of linguistic theories in multimodal interpretation. In particular, we present a greedy algorithm that incorporates Conversation Implicature and Givenness Hierarchy for efficient multimodal reference resolution. Empirical studies indicate that this algorithm significantly reduces the complexity in multimodal reference resolution compared to a previous graph-matching approach. One major advantage of this greedy algorithm is that the prior linguistic and cognitive knowledge can be used to guide the search and significantly prune the search space. Because of its simplicity and generality, this approach has the potential to improve the robustness of interpretation and provide a more practical solution to multimodal input interpretation.\",\n \"Title: Unsupervised Ensemble Ranking: Application to Large-Scale Image Retrieval\\nYear: 2010\\n\\nAbstract: The continued explosion in the growth of image and video databases makes automatic image search and retrieval an extremely important problem. Among the various approaches to Content-based Image Retrieval (CBIR), image similarity based on local point descriptors has shown promising performance. However, this approach suffers from the scalability problem. Although bag-of-words model resolves the scalability problem, it suffers from loss in retrieval accuracy. We circumvent this performance loss by an ensemble ranking approach in which rankings from multiple bag-of-words models are combined to obtain more accurate retrieval results. An unsupervised algorithm is developed to learn the weights for fusing the rankings from multiple bag-of-words models. Experimental results on a database of 100,000 images show that this approach is both efficient and effective in finding visually similar images.\",\n \"Title: A Random Matrix Approach to Differential Privacy and Structure Preserved Social Network Graph Publishing.\\nYear: 2013\\n\\nAbstract: Online social networks are being increasingly used for analyzing various societal phenomena such as epidemiology, information dissemination, marketing and sentiment flow. Popular analysis techniques such as clustering and influential node analysis, require the computation of eigenvectors of the real graph's adjacency matrix. Recent de-anonymization attacks on Netflix and AOL datasets show that an open access to such graphs pose privacy threats. Among the various privacy preserving models, Differential privacy provides the strongest privacy guarantees. In this paper we propose a privacy preserving mechanism for publishing social network graph data, which satisfies differential privacy guarantees by utilizing a combination of theory of random matrix and that of differential privacy. The key idea is to project each row of an adjacency matrix to a low dimensional space using the random projection approach and then perturb the projected matrix with random noise. We show that as compared to existing approaches for differential private approximation of eigenvectors, our approach is computationally efficient, preserves the utility and satisfies differential privacy. We evaluate our approach on social network graphs of Facebook, Live Journal and Pokec. The results show that even for high values of noise variance sigma=1 the clustering quality given by normalized mutual information gain is as low as 0.74. For influential node discovery, the propose approach is able to correctly recover 80 of the most influential nodes. We also compare our results with an approach presented in [43], which directly perturbs the eigenvector of the original data by a Laplacian noise. The results show that this approach requires a large random perturbation in order to preserve the differential privacy, which leads to a poor estimation of eigenvectors for large social networks.\",\n \"Title: Automatic title generation for spoken broadcast news\\nYear: 2001\\n\\nAbstract: In this paper, we implemented a set of title generation methods using training set of 21190 news stories and evaluated them on an independent test corpus of 1006 broadcast news documents, comparing the results over manual transcription to the results over automatically recognized speech. We use both F1 and the average number of correct title words in the correct order as metric. Overall, the results show that title generation for speech recognized news documents is possible at a level approaching the accuracy of titles generated for perfect text transcriptions.\",\n \"Title: Automated annotation of Drosophila gene expression patterns using a controlled vocabulary.\\nYear: 2008\\n\\nAbstract: Regulation of gene expression in space and time directs its localization to a specific subset of cells during development. Systematic determination of the spatiotemporal dynamics of gene expression plays an important role in understanding the regulatory networks driving development. An atlas for the gene expression patterns of fruit fly Drosophila melanogaster has been created by whole-mount in situ hybridization, and it documents the dynamic changes of gene expression pattern during Drosophila embryogenesis. The spatial and temporal patterns of gene expression are integrated by anatomical terms from a controlled vocabulary linking together intermediate tissues developed from one another. Currently, the terms are assigned to patterns manually. However, the number of patterns generated by high-throughput in situ hybridization is rapidly increasing. It is, therefore, tempting to approach this problem by employing computational methods.In this article, we present a novel computational framework for annotating gene expression patterns using a controlled vocabulary. In the currently available high-throughput data, annotation terms are assigned to groups of patterns rather than to individual images. We propose to extract invariant features from images, and construct pyramid match kernels to measure the similarity between sets of patterns. To exploit the complementary information conveyed by different features and incorporate the correlation among patterns sharing common structures, we propose efficient convex formulations to integrate the kernels derived from various features. The proposed framework is evaluated by comparing its annotation with that of human curators, and promising performance in terms of F1 score has been reported.\",\n \"Title: Learning nonparametric kernel matrices from pairwise constraints\\nYear: 2007\\n\\nAbstract: Many kernel learning methods have to assume parametric forms for the target kernel functions, which significantly limits the capability of kernels in fitting diverse patterns. Some kernel learning methods assume the target kernel matrix to be a linear combination of parametric kernel matrices. This assumption again importantly limits the flexibility of the target kernel matrices. The key challenge with nonparametric kernel learning arises from the difficulty in linking the nonparametric kernels to the input patterns. In this paper, we resolve this problem by introducing the graph Laplacian of the observed data as a regularizer when optimizing the kernel matrix with respect to the pairwise constraints. We formulate the problem into Semi-Definite Programs (SDP), and propose an efficient algorithm to solve the SDP problem. The extensive evaluation on clustering with pairwise constraints shows that the proposed nonparametric kernel learning method is more effective than other state-of-the-art kernel learning techniques.\",\n \"Title: First ACM SIGMM international workshop onsocial media (WSM'09)\\nYear: 2009\\n\\nAbstract: The ACM SIGMM International Workshop on Social Media (WSM'09) is the first workshop held in conjunction with the ACM International Multimedia Conference (MM'09) at Bejing, P.R. China, 2009. This workshop provides a forum for researchers and practitioners from all over the world to share information on their latest investigations on social media analysis, exploration, search, mining, and emerging new social media applications.\",\n \"Title: Combining Link and Content for Community Detection: A Discriminative Approach\\nYear: 2009\\n\\nAbstract: In this paper, we consider the problem of combining link and content analysis for community detection from networked data, such as paper citation networks and Word Wide Web. Most existing approaches combine link and content infor- mation by a generative model that generates both links and contents via a shared set of community memberships. These generative models have some shortcomings in that they failed to consider additional factors that could affect the community memberships and isolate the contents that are irrelevant to community memberships. To explicitly ad- dress these shortcomings, we propose a discriminative model for combining the link and content analysis for community detection. First, we propose a conditional model for link analysis and in the model, we introduce hidden variables to explicitly model the popularity of nodes. Second, to allevi- ate the impact of irrelevant content attributes, we develop a discriminative model for content analysis. These two models are unified seamlessly via the community memberships. We present efficient algorithms to solve the related optimization problems based on bound optimization and alternating pro- jection. Extensive experiments with benchmark data sets show that the proposed framework significantly outperforms the state-of-the-art approaches for combining link and con- tent analysis for community detection.\",\n \"Title: Video Classification and Retrieval with the Informedia Digital Video Library System\\nYear: 2002\\n\\nAbstract: and Title) 0.0348 0.1445 0.0724Text + Movie + Image retrieval (Color + Texture) 0.0892 0.220 0.1046Text + Movie + Color + Texture + PRF Classification 0.0924 0.216 0.1124Table 1 Video Retrieval Results on the 25 queries of the 2003 TREC video track evaluation.\",\n \"Title: Sparse Multiple Kernel Learning with Geometric Convergence Rate\\nYear: 2013\\n\\nAbstract: In this paper, we study the problem of sparse multiple kernel learning (MKL), where the goal is to efficiently learn a combination of a fixed small number of kernels from a large pool that could lead to a kernel classifier with a small prediction error. We develop an efficient algorithm based on the greedy coordinate descent algorithm, that is able to achieve a geometric convergence rate under appropriate conditions. The convergence rate is achieved by measuring the size of functional gradients by an empirical $\\\\ell_2$ norm that depends on the empirical data distribution. This is in contrast to previous algorithms that use a functional norm to measure the size of gradients, which is independent from the data samples. We also establish a generalization error bound of the learned sparse kernel classifier using the technique of local Rademacher complexity.\",\n \"Title: Title generation for spoken broadcast news using a training corpus\\nYear: 2000\\n\\nAbstract: The problem of title generation involves finding the essence of a document and expressing it in only a few words. The results of a query to the Informedia Digital Video Library are summarized through an automatically generated title for each retrieved news story. When the document is errorful, as with speech-recognized broadcast news stories, the title creation challenge becomes even greater. We implemented a set of title word selection strategies and evaluated them on an independent test corpus of 579 broadcast news documents, comparing manual transcription results to automatically recognized speech using the CMU Sphinx speech recognition system with a 64000-word broadcast news language model. Using a training collection of 21190 transcribed broadcast news stories, we trained several systems to produce appropriate title words, i.e. Naïve Bayesian approach with full vocabulary, Naïve Bayesian approach with limited vocabulary, nearest neighbor approach and extractive approach. The F1 results shows that the nearest neighbor approach is a quick and easy way of generating good titles for speech recognized documents (F1 = 15.2%), while a Nave Bayesian approach with limited vocabulary also does well on our F1 measure (F1 = 21.6%), which ignores word order in the titles. Overall, the results show that title generation for speech recognized news documents is possible at a level approaching the accuracy of titles generated for perfect text transcriptions. One surprising phenomenon is that extractive approach performances slightly better for speech recognized documents than for manual transcripts.\",\n \"Title: On the use of dynamic Bayesian networks in reconstructing functional neuronal networks from spike train ensembles.\\nYear: 2010\\n\\nAbstract: Coordination among cortical neurons is believed to be a key element in mediating many high-level cortical processes such as perception, attention, learning, and memory formation. Inferring the structure of the neural circuitry underlying this coordination is important to characterize the highly nonlinear, time-varying interactions between cortical neurons in the presence of complex stimuli. In this work, we investigate the applicability of dynamic Bayesian networks (DBNs) in inferring the effective connectivity between spiking cortical neurons from their observed spike trains. We demonstrate that DBNs can infer the underlying nonlinear and time-varying causal interactions between these neurons and can discriminate between mono- and polysynaptic links between them under certain constraints governing their putative connectivity. We analyzed conditionally Poisson spike train data mimicking spiking activity of cortical networks of small and moderately large size. The performance was assessed and compared to other methods under systematic variations of the network structure to mimic a wide range of responses typically observed in the cortex. Results demonstrate the utility of DBN in inferring the effective connectivity in cortical networks.\",\n \"Title: A compensation method to improve the performance of IPI-based entity recognition system in body sensor networks.\\nYear: 2013\\n\\nAbstract: Security of wireless body sensor networks (BSNs) with telemedicine applications remains a crucial issue. A family of novel biometrics schemes has been recently proposed for node recognition and cryptographic key distribution without any pre-deployment in BSNs, where dynamic entity identifiers (EIs) generated from physiological signals captured by individual sensor nodes are used for nodes to recognize each other. As the recognition performance of EIs determines the maximal performance that can be achieved in such biometric systems, several kinds of EI generation schemes have been proposed. The inter-pulse intervals based EI generation scheme is more promising for such applications in actual scenarios because of its acceptable recognition performance. However, it was found that such generated EIs by true pairs of nodes, i.e. two nodes from the same BSN, have some error pattern which could be considered while doing node recognition or key distribution for an improved success rate. To address the problem, this work proposes an error-correcting code based compensation method which can be easily combined together with the key distribution process to achieve an improved recognition performance. Results of statistical analysis with experimental data collected from 14 subjects show that the bit difference between EIs from true pairs of nodes can be effectively reduced with the proposed method.\",\n \"Title: Learning to rank image tags with limited training examples.\\nYear: 2015\\n\\nAbstract: With an increasing number of images that are available in social media, image annotation has emerged as an important research topic due to its application in image matching and retrieval. Most studies cast image annotation into a multilabel classification problem. The main shortcoming of this approach is that it requires a large number of training images with clean and complete annotations in order to learn a reliable model for tag prediction. We address this limitation by developing a novel approach that combines the strength of tag ranking with the power of matrix recovery. Instead of having to make a binary decision for each tag, our approach ranks tags in the descending order of their relevance to the given image, significantly simplifying the problem. In addition, the proposed method aggregates the prediction models for different tags into a matrix, and casts tag ranking into a matrix recovery problem. It introduces the matrix trace norm to explicitly control the model complexity, so that a reliable prediction model can be learned for tag ranking even when the tag space is large and the number of training images is limited. Experiments on multiple well-known image data sets demonstrate the effectiveness of the proposed framework for tag ranking compared with the state-of-the-art approaches for image annotation and tag ranking.\",\n \"Title: CUR Algorithm for Partially Observed Matrices.\\nYear: 2014\\n\\nAbstract: CUR matrix decomposition computes the low rank approximation of a given matrix by using the actual rows and columns of the matrix. It has been a very useful tool for handling large matrices. One limitation with the existing algorithms for CUR matrix decomposition is that they cannot deal with entries in a partially observed matrix, while incomplete matrices are found in many real world applications. In this work, we alleviate this limitation by developing a CUR decomposition algorithm for partially observed matrices. In particular, the proposed algorithm computes the low rank approximation of the target matrix based on (i) the randomly sampled rows and columns, and (ii) a subset of observed entries that are randomly sampled from the matrix. Our analysis shows the error bound, measured by spectral norm, for the proposed algorithm when the target matrix is of full rank. We also show that only O(nr ln r) observed entries are needed by the proposed algorithm to perfectly recover a rank r matrix of size n × n, which improves the sample complexity of the existing algorithms for matrix completion. Empirical studies on both synthetic and real-world datasets verify our theoretical claims and demonstrate the effectiveness of the proposed algorithm.\",\n \"Title: A Simple Homotopy Proximal Mapping for Compressive Sensing.\\nYear: 2014\\n\\nAbstract: In this paper, we present a novel yet simple homotopy proximal mapping algorithm for compressive sensing. The algorithm adopts a simple proximal mapping of the $\\\\ell_1$ norm at each iteration and gradually reduces the regularization parameter for the $\\\\ell_1$ norm. We prove a global linear convergence of the proposed homotopy proximal mapping (HPM) algorithm for solving compressive sensing under three different settings (i) sparse signal recovery under noiseless measurements, (ii) sparse signal recovery under noisy measurements, and (iii) nearly-sparse signal recovery under sub-gaussian noisy measurements. In particular, we show that when the measurement matrix satisfies Restricted Isometric Properties (RIP), our theoretical results in settings (i) and (ii) almost recover the best condition on the RIP constants for compressive sensing. In addition, in setting (iii), our results for sparse signal recovery are better than the previous results, and furthermore our analysis explicitly exhibits that more observations lead to not only more accurate recovery but also faster convergence. Compared with previous studies on linear convergence for sparse signal recovery, our algorithm is simple and efficient, and our results are better and provide more insights. Finally our empirical studies provide further support for the proposed homotopy proximal mapping algorithm and verify the theoretical results.\",\n \"Title: Analysis of Nuclear Norm Regularization for Full-rank Matrix Completion.\\nYear: 2015\\n\\nAbstract: In this paper, we provide a theoretical analysis of the nuclear-norm regularized least squares for full-rank matrix completion. Although similar formulations have been examined by previous studies, their results are unsatisfactory because only additive upper bounds are provided. Under the assumption that the top eigenspaces of the target matrix are incoherent, we derive a relative upper bound for recovering the best low-rank approximation of the unknown matrix. Our relative upper bound is tighter than previous additive bounds of other methods if the mass of the target matrix is concentrated on its top eigenspaces, and also implies perfect recovery if it is low-rank. The analysis is built upon the optimality condition of the regularized formulation and existing guarantees for low-rank matrix completion. To the best of our knowledge, this is first time such a relative bound is proved for the regularized formulation of matrix completion.\",\n \"Title: Random Projections for Classification: A Recovery Approach.\\nYear: 2014\\n\\nAbstract: Random projection has been widely used in data classification. It maps high-dimensional data into a low-dimensional subspace in order to reduce the computational cost in solving the related optimization problem. While previous studies are focused on analyzing the classification performance in the low-dimensional space, in this paper, we consider the recovery problem, i.e., how to accurately recover the optimal solution to the original high-dimensional optimization problem based on the solution learned after random projection. We present a simple algorithm, termed dual random projection, which uses the dual solution of the low-dimensional optimization problem to recover the optimal solution to the original problem. Our theoretical analysis shows that with a high probability, the proposed algorithm is able to accurately recover the optimal solution to the original problem, provided that the data matrix is (approximately) low-rank and/or optimal solution is (approximately) sparse. We further show that the proposed algorithm can be applied iteratively to reducing the recovery error exponentially.\",\n \"Title: A Novel Auxiliary-Free Zero Inductor Current Detection Scheme For Step Down Non-Isolated Led Driver\\nYear: 2014\\n\\nAbstract: A novel auxiliary-free zero crossing detection (ZCD) circuit is presented in this paper, which is especially suitable for step down non-isolated LED drivers operating in boundary conduction mode (BCM). Without extra auxiliary winding, the proposed method can fast and correctly detect zero inductor current information, which only requires an integrated NMOS power switch. The step down non-isolated LED driver with the proposed ZCD scheme has been implemented in CZ6H 0.35 um standard CMOS process, results show that the proposed ZCD scheme can always work properly regardless of the variation of line voltage and load condition.\",\n \"Title: Optimization Of Periphery Circuits In A 1k-Bit Pcram Chip For Highly Reliable Write And Read Operations\\nYear: 2014\\n\\nAbstract: A 1K-bit phase change random access memory (PCRAM) with improved periphery circuits for better reliable operations has been successfully developed in 130 nm CMOS technology. A flexible write driver is proposed to provide a novel continuous step-down pulses by studying programming strategies while a reliable read circuit is designed by investigating the special transition characteristics of PCRAM, leading to an effective write operation and a non-destructive read operation without any additional changes of the storage states. In addition, a large sense margin has been achieved and the read results corresponding well with the write operations, which demonstrate the influences of technology variations have been considerably decreased with the proposed periphery circuits.\",\n \"Title: A Smart Method Of Optimizing The Read/Write Current On Pcm Array\\nYear: 2014\\n\\nAbstract: A method for optimizing the read/write current on phase change memory array is proposed. A smart current adjustment circuit is designed to bi-directionally alter the internal read/write current. The best read/write condition based on arrays can be found through this adjustment approach. Example of a 2-dimensional shmoo test on a 16 k-bit phase change array implemented in 0.13 mu m CMOS technology is given. The resistance distribution can also be roughly obtained. This method, taking advantage of the peripheral circuit, provides statistical yield data on a variety of read/write current, thus offering reliable and helpful indicators for chip parameter setup and process optimization.\",\n \"Title: Efficient Algorithms for Robust One-bit Compressive Sensing.\\nYear: 2014\\n\\nAbstract: While the conventional compressive sensing assumes measurements of infinite precision, one-bit compressive sensing considers an extreme setting where each measurement is quantized to just a single bit. In this paper, we study the vector recovery problem from noisy one-bit measurements, and develop two novel algorithms with formal theoretical guarantees. First, we propose a passive algorithm, which is very efficient in the sense it only needs to solve a convex optimization problem that has a closed-form solution. Despite the apparent simplicity, our theoretical analysis reveals that the proposed algorithm can recover both the exactly sparse and the approximately sparse vectors. In particular, for a sparse vector with s nonzero elements, the sample complexity is O(s \\\\log n/ε^2), where n is the dimensionality and εis the recovery error. This result improves significantly over the previously best known sample complexity in the noisy setting, which is O(s \\\\log n/ε^4). Second, in the case that the noise model is known, we develop an adaptive algorithm based on the principle of active learning. The key idea is to solicit the sign information only when it cannot be inferred from the current estimator. Compared with the passive algorithm, the adaptive one has a lower sample complexity if a high-precision solution is desired.\",\n \"Title: A Single-Pass Algorithm for Efficiently Recovering Sparse Cluster Centers of High-dimensional Data.\\nYear: 2014\\n\\nAbstract: Learning a statistical model for high-dimensional data is an important topic in machine learning. Although this problem has been well studied in the supervised setting, little is known about its unsupervised counterpart. In this work, we focus on the problem of clustering high-dimensional data with sparse centers. In particular, we address the following open question in unsupervised learning: “is it possible to reliably cluster high-dimensional data when the number of samples is smaller than the data dimensionality?\\\" We develop an efficient clustering algorithm that is able to estimate sparse cluster centers with a single pass over the data. Our theoretical analysis shows that the proposed algorithm is able to accurately recover cluster centers with only O(s\\\\log d) number of samples (data points), provided all the cluster centers are s-sparse vectors in a d dimensional space. Experimental results verify both the effectiveness and efficiency of the proposed clustering algorithm compared to the state-of-the-art algorithms on several benchmark datasets.\",\n \"Title: Distance metric learning using dropout: a structured regularization approach\\nYear: 2014\\n\\nAbstract: Distance metric learning (DML) aims to learn a distance metric better than Euclidean distance. It has been successfully applied to various tasks, e.g., classification, clustering and information retrieval. Many DML algorithms suffer from the over-fitting problem because of a large number of parameters to be determined in DML. In this paper, we exploit the dropout technique, which has been successfully applied in deep learning to alleviate the over-fitting problem, for DML. Different from the previous studies that only apply dropout to training data, we apply dropout to both the learned metrics and the training data. We illustrate that application of dropout to DML is essentially equivalent to matrix norm based regularization. Compared with the standard regularization scheme in DML, dropout is advantageous in simulating the structured regularizers which have shown consistently better performance than non structured regularizers. We verify, both empirically and theoretically, that dropout is effective in regulating the learned metric to avoid the over-fitting problem. Last, we examine the idea of wrapping the dropout technique in the state-of-art DML methods and observe that the dropout technique can significantly improve the performance of the original DML methods.\",\n \"Title: MagPairing: Exploiting magnetometers for pairing smartphones in close proximity\\nYear: 2014\\n\\nAbstract: With the prevalence of mobile computing, lots of wireless devices need to establish secure communication on the fly without pre-shared secrets. Device pairing is critical for bootstrapping secure communication between two previously unassociated devices over the wireless channel. Using auxiliary out-of-band channels involving visual, acoustic, tactile or vibrational sensors has been proposed as a feasible option to facilitate device pairing. However, these methods usually require users to perform additional tasks such as copying, comparing, and shaking. It is preferable to have a natural and intuitive pairing method with minimal user tasks. In this paper, we introduce a new method, called MagPairing, for pairing smartphones in close proximity by exploiting correlated magnetometer readings. In MagPairing, users only need to naturally tap the smartphones together for a few seconds without performing any additional operations in authentication and key establishment. Our method exploits the fact that smartphones are equipped with tiny magnets. Highly correlated magnetic field patterns are produced when two smartphones are close to each other. We design MagPairing protocol and implement it on Android smartphones. We conduct extensive simulation and experiments to evaluate MagPairing. Experimental results show that MagPairing can successfully pair two smartphones with 4.5 seconds on average. It is immune to man-in-the-middle attack even when the attacker is a few centimeters away from the pairing devices.\",\n \"Title: A Simple Homotopy Algorithm for Compressive Sensing.\\nYear: 2015\\n\\nAbstract: In this paper, we consider the problem of recovering the s largest elements of an arbitrary vector from noisy measurements. Inspired by previous work, we develop an homotopy algorithm which solves the l(1)-regularized least square problem for a sequence of decreasing values of the regularization parameter. Compared to the previous method, our algorithm is more efficient in the sense it only updates the solution once for each intermediate problem, and more practical in the sense it has a simple stopping criterion by checking the sparsity of the intermediate solution. Theoretical analysis reveals that our method enjoys a linear convergence rate in reducing the recovery error. Furthermore, our guarantee for recovering the top s elements of the target vector is tighter than previous results, and that for recovering the target vector itself matches the state of the art in compressive sensing.\",\n \"Title: Practical secret key agreement for full-duplex near field communications\\nYear: 2014\\n\\nAbstract: Near Field Communication (NFC) is a promising short distance radio communication technology for many useful applications. Although its communication range is short, NFC alone does not guarantee secure communication and is subject to security attacks, such as eavesdropping attack. Generating a shared key and using symmetric key cryptography to secure the communication between NFC devices is a feasible solution to prevent various attacks. However, conventional Diffie-Hellman key agreement protocol is not preferable for resource constrained NFC devices due to its extensive computational overhead and energy consumption. In this paper, we propose a practical, fast and energy-efficient key agreement scheme, which uses random bits transmission with waveform shaking, for NFC devices by exploiting its off-the-shelf full-duplex capability. In the proposed method, two devices send random bits to each other simultaneously without strict synchronization or perfect match of amplitude and phase. On the contrary, the method randomly introduces synchronization offset and mismatch of amplitude and phase for each bit transmission in order to prevent a passive attacker from determining the generated key. A shared bit can be established when two devices send different bits. We conduct theoretical analysis on the correctness and security strength of the method, and extensive simulations to evaluate its effectiveness. We build a testbed based on USRP software defined radio and conduct proof-of-concept experiments to evaluate the method in a real-world environment. It shows that the proposed method achieves a high key generation rate of about 26kbps and is immune to eavesdropping attack even when the attacker is within several centimeters from the legitimate devices. The proposed method is a practical, fast, energy-efficient, and secure key agreement scheme for resource-constrained NFC devices.\",\n \"Title: Provide IPv4 Service Using Pure IPv6 Servers with Stateless NAT64 Translator\\nYear: 2014\\n\\nAbstract: The current Internet is suffering from IPv4 address depletion. Stateless NAT64 is proposed to facilitate the IPv4/IPv6 co-existence and transition. In the past, the stateless NAT64 is mainly for IPv6 clients to access both IPv6 and IPv4 servers. As the fast growth of IP address consuming Internet services, the support for pure IPv6 servers is becoming urgent. In this paper, we propose a design of stateless NAT64 based translator with server port mapping and application layer proxy support to ensure both the IPv4 and IPv6 accessibility to pure IPv6 servers. By deploying such translators, pure IPv6 server infrastructure can be built and the transition process can be promoted.\",\n \"Title: Image Tag Completion By Noisy Matrix Recovery\\nYear: 2014\\n\\nAbstract: It is now generally recognized that user-provided image tags are incomplete and noisy. In this study, we focus on the problem of tag completion that aims to simultaneously enrich the missing tags and remove noisy tags. The novel component of the proposed framework is a noisy matrix recovery algorithm. It assumes that the observed tags are independently sampled from an unknown tag matrix and our goal is to recover the tag matrix based on the sampled tags. We show theoretically that the proposed noisy tag matrix recovery algorithm is able to simultaneously recover the missing tags and de-emphasize the noisy tags even with a limited number of observations. In addition, a graph Laplacian based component is introduced to combine the noisy matrix recovery component with visual features. Our empirical study with multiple benchmark datasets for image tagging shows that the proposed algorithm outperforms state-of-the-art approaches in terms of both effectiveness and efficiency when handling missing and noisy tags.\",\n \"Title: Nystrom Approximation For Sparse Kernel Methods: Theoretical Analysis And Empirical Evaluation\\nYear: 2015\\n\\nAbstract: Nystrom approximation is an effective approach to accelerate the computation of kernel matrices in many kernel methods. In this paper, we consider the Nystrom approximation for sparse kernel methods. Instead of relying on the low-rank assumption of the original kernels, which sometimes does not hold in some applications, we take advantage of the restricted eigenvalue condition, which has been proved to be robust for sparse kernel methods. Based on the restricted eigenvalue condition, we have provided not only the approximation bound for the original kernel matrix but also the recovery bound for the sparse solutions of sparse kernel regression. In addition to the theoretical analysis, we also demonstrate the good performance of the Nystrom approximation for sparse kernel regression on real world data sets.\",\n \"Title: Online Bandit Learning For A Special Class Of Non-Convex Losses\\nYear: 2015\\n\\nAbstract: In online bandit learning, the learner aims to minimize a sequence of losses, while only observing the value of each loss at a single point. Although various algorithms and theories have been developed for online bandit learning, most of them are limited to convex losses. In this paper, we investigate the problem of online bandit learning with non-convex losses, and develop an efficient algorithm with formal theoretical guarantees. To be specific, we consider a class of losses which is a composition of a non-increasing scalar function and a linear function. This setting models a wide range of supervised learning applications such as online classification with a non-convex loss. Theoretical analysis shows that our algorithm achieves an (O) over tilde (poly(d)T-2/3) regret bound when the variation of the loss function is small. To the best of our knowledge, this is the first work in online bandit learning that does not rely on convexity.\",\n \"Title: Delay analysis of physical layer key generation in multi-user dynamic wireless networks\\nYear: 2014\\n\\nAbstract: Secret key generation by extracting the shared randomness in wireless fading channel is a promising way to ensure wireless communication security. Previous works only consider key generation in static networks, but real-world key establishments are usually dynamic. In this work, for the first time we investigate the pairwise key generation in dynamic wireless networks with a center node (eg. access point (AP)) and random arrival users. We establish the key generation model for this kind of networks. We propose a method based on discrete Markov chain to calculate the average time a user will spend on waiting and completing the key generation (average key generation delay, AKGD). Our method can tackle both serial and parallel key generation scheduling under various conditions. We conduct extensive simulations to show the effectiveness of our model and method. The analytical and simulation results match to each other.\",\n \"Title: Multiple Kernel Learning for Visual Object Recognition: A Review\\nYear: 2014\\n\\nAbstract: Multiple kernel learning (MKL) is a principled approach for selecting and combining kernels for a given recognition task. A number of studies have shown that MKL is a useful tool for object recognition, where each image is represented by multiple sets of features and MKL is applied to combine different feature sets. We review the state-of-the-art for MKL, including different formulations and algorithms for solving the related optimization problems, with the focus on their applications to object recognition. One dilemma faced by practitioners interested in using MKL for object recognition is that different studies often provide conflicting results about the effectiveness and efficiency of MKL. To resolve this, we conduct extensive experiments on standard datasets to evaluate various approaches to MKL for object recognition. We argue that the seemingly contradictory conclusions offered by studies are due to different experimental setups. The conclusions of our study are: (i) given a sufficient number of training examples and feature/kernel types, MKL is more effective for object recognition than simple kernel combination (e.g., choosing the best performing kernel or average of kernels); and (ii) among the various approaches proposed for MKL, the sequential minimal optimization, semi-infinite programming, and level method based ones are computationally most efficient.\",\n \"Title: Assessing temporal and spatial evolution of clusters of functionally interdependent neurons using graph partitioning techniques.\\nYear: 2006\\n\\nAbstract: This paper suggests a new approach for identifying clusters of neurons with correlated spiking activity in large-size neuronal ensembles recorded with high-density microelectrode arrays. The nonparametric approach relies on mapping the neuronal spike trains to a 'scale space' using a nested multiresolution projection. Similarity measures can be arbitrarily defined in the scale space independent of the fixed bin width classically used to assess neuronal correlation. This representation allows efficient graph partitioning techniques to be used to identify clusters of correlated firing within distinct behavioral contexts. We use a new probabilistic spectral clustering algorithm that simultaneously maximizes cluster aggregation based on similarity measures. The technique is able to efficiently identify functionally interdependent neurons regardless of the temporal scale from which rate functions are typically estimated. We report the clustering performance of the algorithm applied to a synthesized neurophysiological data set and compare it to known clustering techniques to illustrate the substantial gain in the performance.\",\n \"Title: A knowledge driven regression model for gene expression and microarray analysis.\\nYear: 2006\\n\\nAbstract: The linear regression model has been widely used in the analysis of gene expression and microarray data to identify a subset of genes that are important to a given metabolic function. One of the key challenges in applying the linear regression model to gene expression data analysis arises from the sparse data problem, in which the number of genes is significantly larger than the number of conditions. To resolve this problem, we present a knowledge driven regression model that incorporates the knowledge of genes from the Gene Ontology (GO) database into the linear regression model. It is based on the assumption that two genes are likely to be assigned similar weights when they share similar sets of GO codes. Empirical studies show that the proposed knowledge driven regression model is effective in reducing the regression errors, and furthermore effective in identifying genes that are relevant to a given metabolite.\",\n \"Title: Semi-supervised SVM batch mode active learning for image retrieval\\nYear: 2008\\n\\nAbstract: Active learning has been shown as a key technique for improving content-based image retrieval (CBIR) perfor- mance. Among various methods, support vector machine (SVM) active learning is popular for its application to rel- evance feedback in CBIR. However, the regular SVM ac- tive learning has two main drawbacks when used for rele- vance feedback. First, SVM often suffers from learning with a small number of labeled examples, which is the case in relevance feedback. Second, SVM active learning usually does not take into account the redundancy among exam- ples, and therefore could select multiple examples in rele- vance feedback that are similar (or even identical) to each other. In this paper, we propose a novel scheme that ex- ploits both semi-supervised kernel learning and batch mode active learning for relevance feedback in CBIR. In partic- ular, a kernel function is first learned from a mixture of la- beled and unlabeled examples. The kernel will then be used to effectively identify the informative and diverse examples for active learning via a min-max framework. An empiri- cal study with relevance feedback of CBIR showed that the proposed scheme is significantly more effective than other state-of-the-art approaches.\",\n \"Title: Speedup of fuzzy and possibilistic kernel c-means for large-scale clustering\\nYear: 2011\\n\\nAbstract: The ubiquity of personal computing technology has produced an abundance of staggeringly large data sets-the Library of Congress has stored over 160 terabytes of web data and it is estimated that Facebook alone logs over 25 terabytes of data per day. There is a great need for systems by which one can elucidate the similarity and dissimilarity among and between groups in these data sets. Clustering is one way to find these groups. In this paper, we propose an approximation method for the fuzzy and possibilistic kernel c-means clustering algorithms. Our approximation constrains the cluster centers to be linear combinations of a size m randomly selected subset of the n input objects, where m ≪ n. The proposed algorithm only requires an m × n rectangular portion of the full n × n kernel matrix and the n diagonal values, resulting in significant memory savings. Furthermore, the computational complexity of the c-means algorithm is substantially reduced. We demonstrate that up to 3 orders of magnitude of speedup are possible while achieving almost the same performance as the original kernel c-means algorithm.\",\n \"Title: Tattoo Image Matching and Retrieval\\nYear: 2012\\n\\nAbstract: An automated tattoo image matching system achieves significantly better results for forensic and law enforcement applications than traditional keyword-based matching.\",\n \"Title: Online visual vocabulary pruning using pairwise constraints\\nYear: 2010\\n\\nAbstract: Given a pair of images represented using bag-of-visual-words and a label corresponding to whether the images are “related”(must-link constraint) or “unrelated” (cannot-link constraint), we address the problem of selecting a subset of visual words that are salient in explaining the relation between the image pair. In particular, a subset of features is selected such that the distance computed using these features satisfies the given pairwise constraints. An efficient online feature selection algorithm is presented based on the dual-gradient descent approach. Side information in the form of pair-wise constraints is incorporated into the feature selection stage, providing the user with flexibility to use an unsupervised or semi-supervised algorithm at a later stage. Correlated subsets of visual words, usually resulting from hierarchical quantization process (called groups), are exploited to select a significantly smaller vocabulary. A group-LASSO regularizer is used to drive as many feature weights to zero as possible. We evaluate the quality of the pruned vocabulary by clustering the data using the resulting feature subset. Experiments on PASCAL VOC 2007 dataset using 5000 visual keywords, resulted in around 80% reduction in the number of keywords, with little or no loss in performance.\",\n \"Title: Content-based image retrieval: An application to tattoo images\\nYear: 2009\\n\\nAbstract: Tattoo images on human body have been routinely collected and used in law enforcement to assist in suspect and victim identification. However, the current practice of matching tattoos is based on keywords. Assigning keywords to individual tattoo images is both tedious and subjective. We have developed a content-based image retrieval system for a tattoo image database. The system automatically extracts image features based on the Scale Invariant Feature Transform (SIFT). Side information, i.e., body location of tattoos and tattoo classes, is utilized to improve the retrieval time and retrieval accuracy. Geometrical constraints are also introduced in SIFT keypoint matching to reduce false retrievals. Experimental results on 1,000 queries against an operational database of 63,593 tattoo images show a rank-20 accuracy of 94.2%; the average matching time per query is 2.9 sec. on Intel Core 2, 2.66 GHz, 3 GB RAM processor.\",\n \"Title: Efficient multi-label ranking for multi-class learning: Application to object recognition\\nYear: 2009\\n\\nAbstract: Multi-label learning is useful in visual object recogni- tion when several objects are present in an image. Conven- tional approaches implement multi-label learning as a set of binary classification problems, but they suffer from im- balanced data distributions when the number of classes is large. In this paper, we address multi-label learning with many classes via a ranking approach, termed multi-label ranking. Given a test image, the proposed scheme aims to order all the object classes such that the relevant classes are ranked higher than the irrelevant ones. We present an efficient algorithm for multi-label ranking based on the idea of block coordinate descent. The proposed algorithm is ap- plied to visual object recognition. Empirical results on the PASCAL VOC 2006 and 2007 data sets show promising re- sults in comparison to the state-of-the-art algorithms for multi-label learning.\",\n \"Title: A Hybrid Calibration Method for Aperture Synthesis Radiometers.\\nYear: 2016\\n\\nAbstract: The calibration of aperture synthesis radiometers (ASRs) with large arrays by internal noise injection suffers from high increased mass and volume and is not capable to deal with the out-of-plane distortion of antenna arrays. In this letter, a hybrid calibration method is proposed, which combines the redundant baselines and reduced noise injection. The hybrid method needs only a small noise inject...\",\n \"Title: Inter-functional analysis of high-throughput phenotype data by non-parametric clustering and its application to photosynthesis\\nYear: 2016\\n\\nAbstract: Motivation: Phenomics is the study of the properties and behaviors of organisms (i.e. their phenotypes) on a high-throughput scale. New computational tools are needed to analyze complex phenomics data, which consists of multiple traits/behaviors that interact with each other and are dependent on external factors, such as genotype and environmental conditions, in a way that has not been well studied. Results: We deployed an efficient framework for partitioning complex and high dimensional phenotype data into distinct functional groups. To achieve this, we represented measured phenotype data from each genotype as a cloud-of-points, and developed a novel non-parametric clustering algorithm to cluster all the genotypes. When compared with conventional clustering approaches, the new method is advantageous in that it makes no assumption about the parametric form of the underlying data distribution and is thus particularly suitable for phenotype data analysis. We demonstrated the utility of the new clustering technique by distinguishing novel phenotypic patterns in both synthetic data and a high-throughput plant photosynthetic phenotype dataset. We biologically verified the clustering results using four Arabidopsis chloroplast mutant lines.\",\n \"Title: MagPairing: Pairing Smartphones in Close Proximity Using Magnetometers.\\nYear: 2016\\n\\nAbstract: With the prevalence of mobile computing, lots of wireless devices need to establish secure communication on the fly without pre-shared secrets. Device pairing is critical for bootstrapping secure communication between two previously unassociated devices over the wireless channel. Using auxiliary out-of-band channels involving visual, acoustic, tactile, or vibrational sensors has been proposed as a...\",\n \"Title: Theory of Dual-sparse Regularized Randomized Reduction.\\nYear: 2015\\n\\nAbstract: In this paper, we study randomized reduction methods, which reduce high-dimensional features into low-dimensional space by randomized methods (e.g., random projection, random hashing), for large-scale high-dimensional classification. Previous theoretical results on randomized reduction methods hinge on strong assumptions about the data, e.g., low rank of the data matrix or a large separable margin of classification, which hinder their applications in broad domains. To address these limitations, we propose dual-sparse regularized randomized reduction methods that introduce a sparse regularizer into the reduced dual problem. Under a mild condition that the original dual solution is a (nearly) sparse vector, we show that the resulting dual solution is close to the original dual solution and concentrates on its support set. In numerical experiments, we present an empirical study to support the analysis and we also present a novel application of the dual-sparse regularized randomized reduction methods to reducing the communication cost of distributed learning from large-scale high-dimensional data.\",\n \"Title: An Explicit Sampling Dependent Spectral Error Bound for Column Subset Selection\\nYear: 2015\\n\\nAbstract: In this paper, we consider the problem of column subset selection. We present a novel analysis of the spectral norm reconstruction for a simple randomized algorithm and establish a new bound that depends explicitly on the sampling probabilities. The sampling dependent error bound (i) allows us to better understand the tradeoff in the reconstruction error due to sampling probabilities, (ii) exhibits more insights than existing error bounds that exploit specific probability distributions, and (iii) implies better sampling distributions. In particular, we show that a sampling distribution with probabilities proportional to the square root of the statistical leverage scores is better than uniform sampling, and is better than leverage-based sampling when the statistical leverage scores are very nonuniform. And by solving a constrained optimization problem related to the error bound with an efficient bisection search we are able to achieve better performance than using either the leverage-based distribution or that proportional to the square root of the statistical leverage scores. Numerical simulations demonstrate the benefits of the new sampling distributions for low-rank matrix approximation and least square approximation compared to state-of-the art algorithms.\",\n \"Title: Finding Multiple Stable Clusterings\\nYear: 2015\\n\\nAbstract: Multi-clustering, which tries to find multiple independent ways to partition a data set into groups, has enjoyed many applications, such as customer relationship management, bioinformatics and healthcare informatics. This paper addresses two fundamental questions in multi-clustering: How to model quality of clusterings and how to find multiple stable clusterings (MSC). We introduce to multi-clustering the notion of clustering stability based on Laplacian eigengap, which was originally used by the regularized spectral learning method for similarity matrix learning. We mathematically prove that the larger the eigengap, the more stable the clustering. Furthermore, we propose a novel multi-clustering method MSC. An advantage of our method comparing to the state-of-the-art multi-clustering methods is that our method can provide users a feature subspace to understand each clustering solution. Another advantage is that MSC does not need users to specify the number of clusters and the number of alternative clusterings, which is usually difficult for users without any guidance. Our method can heuristically estimate the number of stable clusterings in a data set. We also discuss a practical way to make MSC applicable to large-scale data. We report an extensive empirical study that clearly demonstrates the effectiveness of our method.\",\n \"Title: Stream Clustering: Efficient Kernel-Based Approximation Using Importance Sampling.\\nYear: 2015\\n\\nAbstract: Stream clustering methods, which group continuous, temporally ordered dynamic data instances, have been used in a number of applications such as stock market analysis, network analysis, and cosmological analysis. Most of the popular stream clustering algorithms are linear in nature, i.e. they assume that the data is linearly separable in the input space and use measures such as the Euclidean distance to define the inter-point similarity. Though these linear clustering algorithms are efficient, they do no achieve acceptable cluster quality on real-world data. Kernel-based clustering algorithms, which use non-linear similarity measures, yield better cluster quality, but are unsuitable for clustering data streams due to their high running time and memory complexity. We propose an efficient kernel-based clustering algorithm, called the Approximate Stream Kernel k-means, which uses importance sampling to sample a subset of the data stream, and clusters the entire stream based on each data point's similarity to the sampled data points in real-time. Every time a data point is sampled, the kernel matrix representing the similarity between the sampled points is updated, and projected to a low dimensional space spanned by its top eigenvectors. The data points are clustered in this low-dimensional space using the k-means algorithm. Thus, the Approximate Stream Kernel k-means algorithm performs clustering in linear time using kernel-based similarity. We show that only a small number of points need to be sampled from the data stream, and the resulting approximation error is well-bounded. Using several large benchmark data sets to simulate data streams, we demonstrate that the proposed algorithm achieves a significant speedup over other kernel-based clustering algorithms with minimal loss in cluster quality. We also demonstrate the practical applicability of the proposed algorithm by using it to efficiently find trending topics in the Twitter stream data.\",\n \"Title: Big Data Analytics: Optimization and Randomization\\nYear: 2015\\n\\nAbstract: As the scale and dimensionality of data continue to grow in many applications of data analytics (e.g., bioinformatics, finance, computer vision, medical informatics), it becomes critical to develop efficient and effective algorithms to solve numerous machine learning and data mining problems. This tutorial will focus on simple yet practically effective techniques and algorithms for big data analytics. In the first part, we plan to present the state-of-the-art large-scale optimization algorithms, including various stochastic gradient descent methods, stochastic coordinate descent methods and distributed optimization algorithms, for solving various machine learning problems. In the second part, we will focus on randomized approximation algorithms for learning from large-scale data. We will discuss i) randomized algorithms for low-rank matrix approximation; ii) approximation techniques for solving kernel learning problems; iii) randomized reduction methods for addressing the high-dimensional challenge. Along with the description of algorithms, we will also present some empirical results to facilitate understanding of different algorithms and comparison between them.\",\n \"Title: Stock Constrained Recommendation in Tmall\\nYear: 2015\\n\\nAbstract: A large number of recommender systems have been developed to serve users with interesting news, ads, products or other contents. One main limitation with the existing work is that they do not take into account the inventory size of of items to be recommended. As a result, popular items are likely to be out of stock soon as they have been recommended and sold to many users, significantly affecting the impact of recommendation and user experience. This observation motivates us to develop a novel aware recommender system. It jointly optimizes the recommended items for all users based on both user preference and inventory sizes of different items. It requires solving a non-smooth optimization involved estimating a matrix of nxn, where n is the number of items. With the proliferation of items, this approach can quickly become computationally infeasible. We address this challenge by developing a dual method that reduces the number of variables from n^2 to n, significantly improving the computational efficiency. We also extend this approach to the online setting, which is particularly important for big promotion events. Our empirical studies based on a real benchmark data with 100 millions of user visits from Tmall verify the effectiveness of the proposed approach.\",\n \"Title: SecNFC: Securing inductively-coupled Near Field Communication at physical layer\\nYear: 2015\\n\\nAbstract: Near Field Communication (NFC) is widely used today in many useful applicatious, such as contactless payment, identification, and file exchange. Due to the limitations on computation, power, and cost of NFC devices, NFC systems are often lack of encryption or weakly encrypted, leaving them exposed to security attacks. One solution for this problem is to iustall strong cryptographic protocols on NFC devices. However, it involves upgrading and revoking deployed NFC devices, which is costly and impractical. Moreover, encryption algorithms are usually cousidered expeusive for resource constrained NFC devices in terms of computation overhead and energy cousumption. In this paper, aiming at securing NFC from eavesdropping at physical layer, we propose SecNFC, a solution that changes the signals on the initiators to hide the communicatious, but does not require any changes to the targets. In this work, for the first time, we address a practical problem of synchronization offset between two NFC terminals, which can be exploited by an eavesdropper to compromise the transmitted bits. SecNFC takes into consideration the synchronization offset and solves the challenge by introducing blocking signals around the transition point when the target switches load between on and off. With this solution, even with the existence of synchronization offset, eavesdroppers cannot extract any bit information from eavesdropping. We conduct extensive simulations to evaluate the performance of SecNFC. We also build a testbed based on USRP software defined radio and off-theshelf NFC tags to evaluate SecNFC in a real-world environment. Both simulation and experimental results show that SecNFC can efficiently prevent NFC from eavesdropping with a slight and tolerable decoding performance degradation at the initiator.\",\n \"Title: Physical layer key agreement under signal injection attacks\\nYear: 2015\\n\\nAbstract: Physical layer key agreement techniques derive a symmetric cryptographic key from the wireless fading channel between two wireless devices by exploiting channel randomness and reciprocity. Existing works mainly focus on the security analysis and protocol design of the techniques under passive attacks. The study on physical layer key agreement techniques under active attacks is largely open. In this paper, we present a new form of high threatening active attack, named signal injection attack. By injecting the similar signals to both keying devices, the attacker aims at manipulating the channel measurements and compromising a portion of the key. We further propose a countermeasure to the signal injection attack, pHY-UIR (PHYsical layer key agreement with User Introduced Randomness). In PHY-UIR, both keying devices independently introduce randomness into the channel probing frames, and compose common random series by combining the randomness in the fading channel and the ones introduced by users together. With this solution, the composed series and injected signals become uncorrelated. Thus, the final key will automatically exclude the contaminated portion related to injected signal while persisting the other portion related to random fading channel. Moreover, the contaminated composed series at two keying devices become decorrelated, which help detect the attack. We analyze the security strength of PHY-UIR and conduct extensive simulations to evaluate it. Both theoretical analysis and simulations demonstrate the effectiveness of PHy-UIR. We also perform proof-of-concept experiments by using software defined radios in a real-world environment. We show that signal injection attack is feasible in practice and leads to a strong correlation (0.75) between the injected signal and channel measurements at legitimate users for existing key generation methods. PHY-UIR is immune to the signal injection attack and results in low correlation (0.15) between the injected signal and the composed random signals at legitimate users.\",\n \"Title: A reconfigurable system of routing and switching node based on ATC\\nYear: 2015\\n\\nAbstract: Flexible Architecture of Reconfigurable Infrastructure (FARI) is a new information and communication infrastructure network. It is a kind of Software Defined Network. The provided resources by the underlying network are abstracted into Atomic Capabilities. This paper focuses on the bottom reconstruction technology in the reconfigurable routing and switching nodes of the FARI based on ATC theory. An ATC dynamically reconfigurable prototype system is designed, and a user management platform for testing is also implemented. Test results show that the system can dynamically reconfigure the underlying nodes.\",\n \"Title: An Atomic Capability Chain Composition Method\\nYear: 2015\\n\\nAbstract: FARI (Flexible Architecture of Reconfigurable Infrastructure) is proposed for a future information and communication infrastructure network. This architecture is dynamic and flexible, and can meet various requirements of future networks. Based on the atomic capability theory of FARI and the traditional I/O matching algorithm, an atomic capability chain composition method is proposed. This method can combine a series of atomic capabilities to an atomic capability chain, according to special business requirement. And it can automatically compose some ACs with independent function characteristics into a chain which satisfies the business requirement. Traditional I/O matching algorithm just specifies the matching between input and output parameters. Our improved I/O matching algorithm specifies not only the matching of I/O parameters, but also the matching of metadata which is specific to ACs. At last, an example is provided to illustrate how this method works, and a simulation is given to show the efficiency of this method.\",\n \"Title: Lower and Upper Bounds on the Generalization of Stochastic Exponentially Concave Optimization\\nYear: 2015\\n\\nAbstract: In this paper we derive high probability lower and upper bounds on the excess risk of stochastic optimization of exponentially concave loss functions. Exponentially concave loss functions encompass several fundamental problems in machine learning such as squared loss in linear regression, logistic loss in classification, and negative logarithm loss in portfolio management. We demonstrate an O(d log T/T ) upper bound on the excess risk of stochastic online Newton step algorithm, and an O(d/T ) lower bound on the excess risk of any stochastic optimization method for squared loss, indicating that the obtained upper bound is optimal up to a logarithmic factor. The analysis of upper bound is based on recent advances in concentration inequalities for bounding self-normalized martingales, which is interesting by its own right, and the proof technique used to achieve the lower bound is a probabilistic method and relies on an information-theoretic minimax analysis.\",\n \"Title: Efficient Sparse Recovery Via Adaptive Non-Convex Regularizers With Oracle Property\\nYear: 2014\\n\\nAbstract: The main shortcoming of sparse recovery with a convex regularizer is that it is a biased estimator and therefore will result in a suboptimal performance in many cases. Recent studies have shown, both theoretically and empirically, that non-convex regularizer is able to overcome the biased estimation problem. Although multiple algorithms have been developed for sparse recovery with non-convex regularization, they are either computationally demanding or not equipped with the desired properties (i.e. optimal recovery error, selection consistency and oracle property). In this work, we develop an algorithm for efficient sparse recovery based on proximal gradient descent. The key feature of the proposed algorithm is introducing adaptive non-convex regularizers whose shrinking threshold vary over iterations. The algorithm is compatible with most popular non-convex regularizers, achieves a geometric convergence rate for the recovery error, is selection consistent, and most importantly has the oracle property. Based on the proposed framework, we suggest to use a so-called ACCQ regularizer, which is equivalent to zero proximal projection gap adaptive hard-thresholding. Experiments with both synthetic data sets and real images verify both the efficiency and effectiveness of the proposed method compared to the state-of-the-art methods for sparse recovery.\",\n \"Title: Topic Modeling in Semantic Space with Keywords\\nYear: 2015\\n\\nAbstract: A common and convenient approach for user to describe his information need is to provide a set of keywords. Therefore, the technique to understand the need becomes crucial. In this paper, for the information need about a topic or category, we propose a novel method called TDCS(Topic Distilling with Compressive Sensing) for explicit and accurate modeling the topic implied by several keywords. The task is transformed as a topic reconstruction problem in the semantic space with a reasonable intuition that the topic is sparse in the semantic space. The latent semantic space could be mined from documents via unsupervised methods, e.g. LSI. Compressive sensing is leveraged to obtain a sparse representation from only a few keywords. In order to make the distilled topic more robust, an iterative learning approach is adopted. The experiment results show the effectiveness of our method. Moreover, with only a few semantic concepts remained for the topic, our method is efficient for subsequent text mining tasks.\",\n \"Title: Sparse Kernel Clustering of Massive High-Dimensional Data sets with Large Number of Clusters\\nYear: 2015\\n\\nAbstract: In clustering applications involving documents and images, in addition to the large number of data points (N) and their high dimensionality (d), the number of clusters (C) into which the data need to be partitioned is also large. Kernel-based clustering algorithms, which have been shown to perform better than linear clustering algorithms, have high running time complexity in terms of N, d and C. We propose an efficient sparse kernel k-means clustering algorithm, which incrementally samples the most informative points from the data set using importance sampling, and constructs a sparse kernel matrix using these sampled points. Each row in this matrix corresponds to a data point's similarity with its p-nearest neighbors among the sampled points (p -- N). This sparse kernel matrix is used to perform clustering and obtain the cluster labels. This combination of sampling and sparsity reduces both the running time and memory complexity of kernel clustering. In order to further enhance its efficiency, the proposed algorithm projects the data on to the top C eigenvectors of the sparse kernel matrix and clusters these eigenvectors using a modified k-means algorithm. The running time of the proposed sparse kernel k-means algorithm is linear in N and d, and logarithmic in C. We show analytically that only a small number of points need to be sampled from the data set, and the resulting approximation error is well-bounded. We demonstrate, using several large high-dimensional text and image data sets, that the proposed algorithm is significantly faster than classical kernel-based clustering algorithms, while maintaining clustering quality.\",\n \"Title: Accelerated Sparse Linear Regression via Random Projection.\\nYear: 2016\\n\\nAbstract: In this paper, we present an accelerated numerical method based on random projection for sparse linear regression. Previous studies have shown that under appropriate conditions, gradient-based methods enjoy a geometric convergence rate when applied to this problem. However, the time complexity of evaluating the gradient is as large as O(nd), where n is the number of data points and d is the dimensionality, making those methods inefficient for large-scale and high-dimensional dataset. To address this limitation, we first utilize random projection to find a rank-k approximator for the data matrix, and reduce the cost of gradient evaluation to O(nk + dk), a significant improvement when k is much smaller than d and n. Then, we solve the sparse linear regression problem via a proximal gradient method with a homotopy strategy to generate sparse intermediate solutions. Theoretical analysis shows that our method also achieves a global geometric convergence rate, and moreover the sparsity of all the intermediate solutions are well-bounded over the iterations. Finally, we conduct experiments to demonstrate the efficiency of the proposed method.\",\n \"Title: Stochastic Optimization for Kernel PCA.\\nYear: 2016\\n\\nAbstract: Kernel Principal Component Analysis (PCA) is a popular extension of PCA which is able to find nonlinear patterns from data. However, the application of kernel PCA to large-scale problems remains a big challenge, due to its quadratic space complexity and cubic time complexity in the number of examples. To address this limitation, we utilize techniques from stochastic optimization to solve kernel PCA with linear space and time complexities per iteration. Specifically, we formulate it as a stochastic composite optimization problem, where a nuclear norm regularizer is introduced to promote low-rankness, and then develop a simple algorithm based on stochastic proximal gradient descent. During the optimization process, the proposed algorithm always maintains a low-rank factorization of iterates that can be conveniently held in memory. Compared to previous iterative approaches, a remarkable property of our algorithm is that it is equipped with an explicit rate of convergence. Theoretical analysis shows that the solution of our algorithm converges to the optimal one at an O(1/T) rate, where T is the number of iterations.\",\n \"Title: Tracking Slowly Moving Clairvoyant: Optimal Dynamic Regret of Online Learning with True and Noisy Gradient.\\nYear: 2016\\n\\nAbstract: This work focuses on dynamic regret of online convex optimization that compares the performance of online learning to a clairvoyant who knows the sequence of loss functions in advance and hence selects the minimizer of the loss function at each step. By assuming that the clairvoyant moves slowly (i.e., the minimizers change slowly), we present several improved variation-based upper bounds of the dynamic regret under the true and noisy gradient feedback, which are optimal in light of the presented lower bounds. The key to our analysis is to explore a regularity metric that measures the temporal changes in the clairvoyantu0027s minimizers, to which we refer as path variation. Firstly, we present a general lower bound in terms of the path variation, and then show that under full information or gradient feedback we are able to achieve an optimal dynamic regret. Secondly, we present a lower bound with noisy gradient feedback and then show that we can achieve optimal dynamic regrets under a stochastic gradient feedback and two-point bandit feedback. Moreover, for a sequence of smooth loss functions that admit a small variation in the gradients, our dynamic regret under the two-point bandit feedback matches what is achieved with full information.\",\n \"Title: Social Graph Publishing with Privacy Guarantees\\nYear: 2016\\n\\nAbstract: Online social network graphs provide useful insights on various social phenomena such as information dissemination and epidemiology. Unfortunately, social network providers often refuse to publish their social network graphs due to privacy concerns. Recently, differential privacy has become the widely accepted criteria for privacy preserving data publishing because it provides strongest privacy guarantees for publishing sensitive datasets. Although some work has been done on publishing matrices with differential privacy, they are computationally unpractical as they are not designed to handle large matrices such as the adjacency matrices of OSN graphs. In this paper, we propose a random matrix approach to OSN graph publishing, which achieves storage and computational efficiency by reducing the dimensions of adjacency matrices and achieves differential privacy by adding a small amount of noise. Our key idea is to first project each row of an adjacency matrix into a low dimensional space using random projection, and then perturb the projected matrix with random noise, and finally publish the perturbed and projected matrix. In this paper, we first prove that random projection plus random perturbation preserve differential privacy, and also that the random noise required to achieve differential privacy is small. We then validate the proposed approach and evaluate the utility of the published data for two different applications, namely node clustering and node ranking, using publicly available OSN graphs of Facebook, Live Journal, and Pokec.\",\n \"Title: Large-scale Robust Online Matching and Its Application in E-commerce\\nYear: 2016\\n\\nAbstract: This talk will be focused on large-scale matching problem that aims to find the optimal assignment of tasks to different agents under linear constraints. Large-scale matching has found numerous applications in e-commerce. An well known example is budget aware online advertisement. A common practice in online advertisement is to find, for each opportunity or user, the advertisements that fit best with his/her interests. The main shortcoming with this greedy approach is that it did not take into account the budget limits set by advertisers. Our studies, as well as others, have shown that by carefully taking into budget limits of individual advertisers, we could significantly improve the performance of the advertisement system. Despite of rich literature, two important issues are often overlooked in the previous studies of matching/assignment problem. The first issues arises from the fact that most quantities used by optimization are estimated based on historical data and therefore are likely to be inaccurate and unreliable. The second challenge is how to perform online matching as in many e-commerce problems, tasks are created in an online fashion and algorithm has to make assignment decision immediately when every task emerges. We refer to these two issues as challenges of “robust matching” and “online matching”. To address the first challenge, I will introduce two different techniques for robust matching. The first approach is based on the theory of robust optimization that takes into account the uncertainties of estimated quantities when performing optimization. The second approach is based on the theory of two-sided matching whose result only depends on the partial preference of estimated quantities. To deal with the challenge of online matching, I will discuss two online optimization techniques, one based on theory of primal-dual online optimization and one based on minimizing dynamic regret under long term constraints. We verify the effectiveness of all these approaches by applying them to real-world projects developed in Alibaba.\",\n \"Title: Array Configuration Optimization For Rotating Mirrored Aperture Synthesis (Rmas)\\nYear: 2016\\n\\nAbstract: Aperture synthesis with high spatial resolution would seriously increase the hardware requirements and system complexity. To reduce the system complexity, Rotating mirrored aperture synthesis (RMAS) that combines the antenna array rotation with mirrored aperture synthesis was proposed to reduce the number of antennas. In this paper, the sieving method is presented to optimize the array configuration for RMAS. The initial simulation results demonstrate the validity of the proposed method and the optimized array of RMAS can reconstruct the brightness temperature of the observed scene with fewer antennas.\",\n \"Title: Correction Of Smos Data In Coastal Area Of South China Sea Based On Land Contamination Analysis\\nYear: 2016\\n\\nAbstract: For SMOS data, significant errors exist in coastal areas because of the contamination by the nearby land. In this paper, the analysis of SMOS brightness temperature (TB) data in coastal area of South China Sea (SCS) reveals a decreasing trend with the increase of the distance to coast. A SMOS TB measurement model according to geophysical parameters and SMOS/MIRAS antenna array is established. The simulation shows that TB is large within 40 km since land is much warmer than ocean and is observed by the mainlobe of the antenna array. Then the simulation result of land contamination is used for correction of SMOS measured TB data in coastal area of SCS. Then the corrected TB is applied for sea surface salinity (SSS) retrieval, which reduces SSS outliers and improves the accuracy of SSS in coastal area of SCS.\",\n \"Title: Strongly Adaptive Regret Implies Optimally Dynamic Regret.\\nYear: 2017\\n\\nAbstract: To cope with changing environments, recent developments in online learning have introduced the concepts of adaptive regret and dynamic regret independently. In this paper, we illustrate an intrinsic connection between these two concepts by showing that the dynamic regret can be expressed in terms of the adaptive regret and the functional variation. This observation implies that strongly adaptive algorithms can be directly leveraged to minimize the dynamic regret. As a result, we present a series of strongly adaptive algorithms whose dynamic regrets are minimax optimal for convex functions, exponentially concave functions, and strongly convex functions, respectively. To the best of our knowledge, this is the first time that such kind of dynamic regret bound is established for exponentially concave functions. Moreover, all of those adaptive algorithms do not need any prior knowledge of the functional variation, which is a significant advantage over previous specialized methods for minimizing dynamic regret.\",\n \"Title: Sparse Learning with Stochastic Composite Optimization.\\nYear: 2017\\n\\nAbstract: In this paper, we study Stochastic Composite Optimization (SCO) for sparse learning that aims to learn a sparse solution from a composite function. Most of the recent SCO algorithms have already reached the optimal expected convergence rate $\\\\\\\\mathcal {O}(1/\\\\\\\\lambda T)$ , but they often fail to deliver sparse solutions at the end either due to the limited sparsity regularization during stochastic optimization (SO) or due to the limitation in online-to-batch conversion. Even when the objective function is strongly convex, their high probability bounds can only attain $\\\\\\\\mathcal {O}(\\\\\\\\sqrt{\\\\\\\\log (1/\\\\\\\\delta)/T})$ with $\\\\\\\\delta$ is the failure probability, which is much worse than the expected convergence rate. To address these limitations, we propose a simple yet effective two-phase Stochastic Composite Optimization scheme by adding a novel powerful sparse online-to-batch conversion to the general Stochastic Optimization algorithms. We further develop three concrete algorithms, OptimalSL, LastSL and AverageSL, directly under our scheme to prove the effectiveness of the proposed scheme. Both the theoretical analysis and the experiment results show that our methods can really outperform the existing methods at the ability of sparse learning and at the meantime we can improve the high probability bound to approximately $\\\\\\\\mathcal {O}(\\\\\\\\log (\\\\\\\\log (T)/\\\\\\\\delta)/\\\\\\\\lambda T)$ .\",\n \"Title: Experiences with Multi-modal Collaborative Virtual Laboratory (MMCVL).\\nYear: 2017\\n\\nAbstract: Training through Multi-modal Collaborative Virtual Laboratory (MMCVL) aims at developing or improving skills of students without the need of them being physically present. MMCVL addresses the problems faced by the current education system such as lack of resources and safe use of expensive laboratory equipment. It also provides a means for easy collaboration between students and teachers from different universities and schools. Multimedia data in terms of 3D mesh, skeleton, audio and virtual object physics interaction are incorporated in MMCVL to provide a 3D Tele-Immersion environment. The system uses low-cost RGB-D cameras such as Microsoft Kinect V2 to capture and generate 3D model of the person by extracting him/her from the entire captured data and immersing it in different interactive virtual environments. Different lab environments can be virtually created to provide for and replace the real laboratories. Object interactions allow for users to interact with the virtual objects present in the scene similar to real lab setup. Recording module of the toolkit allows future reference for training and testing of the student. The system also facilitates virtual feedback by giving the correct measurements, video demonstrations as well as audio communication with the collaborating teacher. Two different Chemistry experiments were performed to evaluate the user experience and usability aspects of MMCVL. Results obtained from the user study analysis prove the system to be fun and realistic, and at the same time engaging and motivating for learning laboratory experiments.\",\n \"Title: Empirical Risk Minimization for Stochastic Convex Optimization: $O(1/n)$- and $O(1/n^2)$-type of Risk Bounds.\\nYear: 2017\\n\\nAbstract: Although there exist plentiful theories of empirical risk minimization (ERM) for supervised learning, current theoretical understandings of ERM for a related problem---stochastic convex optimization (SCO), are limited. In this work, we strengthen the realm of ERM for SCO by exploiting smoothness and strong convexity conditions to improve the risk bounds. First, we establish an $widetilde{O}(d/n + sqrt{F_*/n})$ risk bound when the random function is nonnegative, convex and smooth, and the expected function is Lipschitz continuous, where $d$ is the dimensionality of the problem, $n$ is the number of samples, and $F_*$ is the minimal risk. Thus, when $F_*$ is small we obtain an $widetilde{O}(d/n)$ risk bound, which is analogous to the $widetilde{O}(1/n)$ optimistic rate of ERM for supervised learning. Second, if the objective function is also $lambda$-strongly convex, we prove an $widetilde{O}(d/n + kappa F_*/n )$ risk bound where $kappa$ is the condition number, and improve it to $O(1/[lambda n^2] + kappa F_*/n)$ when $n=widetilde{Omega}(kappa d)$. As a result, we obtain an $O(kappa/n^2)$ risk bound under the condition that $n$ is large and $F_*$ is small, which to the best of our knowledge, is the first $O(1/n^2)$-type of risk bound of ERM. Third, we stress that the above results are established in a unified framework, which allows us to derive new risk bounds under weaker conditions, e.g., without convexity of the random function and Lipschitz continuity of the expected function. Finally, we demonstrate that to achieve an $O(1/[lambda n^2] + kappa F_*/n)$ risk bound for supervised learning, the $widetilde{Omega}(kappa d)$ requirement on $n$ can be replaced with $Omega(kappa^2)$, which is dimensionality-independent.\",\n \"Title: A Two-Stage Approach for Learning a Sparse Model with Sharp Excess Risk Analysis.\\nYear: 2017\\n\\nAbstract: This paper aims to provide a sharp excess risk guarantee for learning a sparse linear model without any assumptions about the strong convexity of the expected loss and the sparsity of the optimal solution in hindsight. Given a target level for the excess risk, an interesting question to ask is how many examples and how large the support set of the solution are enough for learning a good model with the target excess risk. To answer these questions, we present a two-stage algorithm that (i) in the first stage an epoch based stochastic optimization algorithm is exploited with an established O(1/epsilon) bound on the sample complexity; and (ii) in the second stage a distribution dependent randomized sparsification is presented with an O(1/epsilon) bound on the sparsity (referred to as support complexity) of the resulting model. Compared to previous works, our contributions lie at (i) we reduce the order of the sample complexity from O(1/epsilon(2)) to O(1/epsilon) without the strong convexity assumption; and (ii) we reduce the constant in O(1/epsilon) for the sparsity by exploring the distribution dependent sampling.\",\n \"Title: Extremely Low Bit Neural Network: Squeeze the Last Bit Out with ADMM.\\nYear: 2018\\n\\nAbstract: Although deep learning models are highly effective for various learning tasks, their high computational costs prohibit the deployment to scenarios where either memory or computational resources are limited. In this paper, we focus on compressing and accelerating deep models with network weights represented by very small numbers of bits, referred to as extremely low bit neural network. We model this problem as a discretely constrained optimization problem. Borrowing the idea from Alternating Direction Method of Multipliers (ADMM), we decouple the continuous parameters from the discrete constraints of network, and cast the original hard problem into several subproblems. We propose to solve these subproblems using extragradient and iterative quantization algorithms that lead to considerably faster convergency compared to conventional optimization methods. Extensive experiments on image recognition and object detection verify that the proposed algorithm is more effective than state-of-the-art approaches when coming to extremely low bit neural network.\",\n \"Title: QoE Studies on Interactive 3D Tele-Immersion\\nYear: 2017\\n\\nAbstract: Users' Quality of Experience (QoE) in Interactive 3D Tele-Immersion (i3DTI) systems is influenced by several factors such as the quality of the \\\"live\\\" 3D avatars of the users, network latency, rendering methodology (head mounted display or regular TV type of display), etc. Hence, it becomes important to answer the question: \\\"Is Visual Quality (VQ) the only factor to be considered or do better immersion and faster interactions matter, in having good QoE?\\\" To answer this question, in this paper, a highly optimized state-of-the-art i3DTI framework implementation is introduced along with a soccer-inspired penalty shootout game. This game allows users to experience various situations, view different angles, perceive delays, track virtual ball motion, and play naturally using their entire bodies. A head mounted display device - Oculus Rift allows the users to get completely immersed and perform better in the penalty shootout game compared to watching themselves play on a 3D TV. This scenario is obtained in a controlled lab setting with ultra-high-speed network that has ultra-low latency, high VQ, fast and realistic interactions. Such ideal conditions are not typically available in a wide area Internet. Hence, for faster interaction scenario, we used lesser RGB-D cameras for 3D reconstruction, thereby reducing the model quality significantly. The high responsiveness of the game masked the user's perception of quality, and resulted in them not noticing the lower VQ of the reconstructed 3D models. Based on the results from the user study that focused on immersion, interaction and VQ aspects; users felt the game to be visually appealing, intuitive, engaging, and highly entertaining in all of the scenarios.\",\n \"Title: The Opensesame Nist 2016 Speaker Recognition Evaluation System\\nYear: 2017\\n\\nAbstract: Last two decades have witnessed a significant progress in speaker recognition, as evidenced by the improving performance in the speaker recognition evaluations (SRE) hosted by NIST. Despite the progress, only a few research is focused on speaker recognition with short duration and language mismatch condition, which often leads to poor recognition performance. In NIST SRE2016, these concerns were first systematically investigated by the speaker recognition community. In this study, we address these challenges from the viewpoint of feature extraction and modeling. In particular, we improve the robustness of features by combining GMM and DNN based iVector extraction approaches, and improve the reliability of the back-end model by exploiting symmetric SVM that can effectively leverage the unlabeled data. Finally, we introduce distance metric learning to improve the generalization capacity of the development data that is usually in limited size. Then a fusion strategy is adopted to collectively boost the performance. The effectiveness of the proposed scheme for speaker recognition is demonstrated on SRE2016 evaluation data: compared with DNN-iVector PLDA baseline system, our method yields 25.6% relative improvement in terms of min_Cprimary.\",\n \"Title: Learndna: An Interactive Vr Application For Learning Dna Structure\\nYear: 2018\\n\\nAbstract: Understanding complex concepts is known to be difficult and tedious. This complexity is very often due to lack of visual representation of abstract concepts, that cannot be easily observed. Representing these ideas in a visual Virtual Environment (VE) is a very important application of Virtual Reality (VR). This paper presents an immersive VR application for teaching students about one such complex concept of DNA structure. Building virtual environments that show the double helix structure of DNA and allowing students to interact with it, provides an illustrative and engaging experience. A crucial factor that determines usability of such systems is the correlation between actual concept and its virtual representation. Our virtual DNA structure is very similar to the actual structure, thus highlighting its complexity. This VE is built for Oculus Rift with simple clicker based manipulation and travel, to allow easy portability to other VR devices. The system also provides test based assessment to gauge students' understanding. The evaluation of this VE was carried out with the help from a group of Subject Matter Experts (SMEs). Through their feedback, we refined system design and subject content, as well as survey questionnaires. A subsequent user study was conducted in a local community college. Students were recruited from two biology classes and offered extra credit based on results recorded by this system. The study gave pre-test prior to virtual experiment, post-test following virtual experiment and survey session after post-test.All SMEs felt that the application was useful for learning about DNA. Comparing the pre-test and post-test performance of the students, there was an average of 40% improvement in the scores, indicating that LearnDNA was able to facilitate learning the concepts. Most students also felt that the application was engaging and fun to use.\",\n \"Title: Physical layer multi-user key generation in wireless networks.\\nYear: 2018\\n\\nAbstract: Secret key generation by extracting the shared randomness in the wireless fading channel from physical layer is an interesting topic of practical value. Previous works have focused on the study of physical layer key generation with two nodes from the view point of key generation rate (KGR). Information theoretic limits and the KGRs in implementation have been derived. However, in real-world applications, the physical layer key generation problem involving multiple nodes is the common case, which lacks sufficient study so far. Multi-node case differs from two-node case in that there are two more important considerations: (1) the trade-off between KGR and probing efficiency at individual node pair; (2) channel probing schedule among multiple node pairs. This paper aims at minimizing the (shorten as OWT) through the optimization of probing rates at individual node pair and channel probing schedule. The theoretical lower bound of OWT is derived first, then a practical method (MUKEM) is proposed to compute reasonable probing rates and channel probing schedule for multiple node pairs to obtain a short OWT. Simulations are conducted to evaluate the effectiveness of our method. The results show that 70 % of OWT can be reduced by using our method comparing with one-by-one key generations; while it is only about 8 % longer than the lower bound of OWT.\",\n \"Title: Learning with Non-Convex Truncated Losses by SGD.\\nYear: 2018\\n\\nAbstract: Learning with a {it convex loss} function has been a dominating paradigm for many years. It remains an interesting question how non-convex loss functions help improve the generalization of learning with broad applicability. In this paper, we study a family of objective functions formed by truncating traditional loss functions, which is applicable to both shallow learning and deep learning. Truncating loss functions has potential to be less vulnerable and more robust to large noise in observations that could be adversarial. More importantly, it is a generic technique without assuming the knowledge of noise distribution. To justify non-convex learning with truncated losses, we establish excess risk bounds of empirical risk minimization based on truncated losses for heavy-tailed output, and statistical error of an approximate stationary point found by stochastic gradient descent (SGD) method. Our experiments for shallow and deep learning for regression with outliers, corrupted data and heavy-tailed noise further justify the proposed method.\",\n \"Title: Robust Optimization over Multiple Domains\\nYear: 2018\\n\\nAbstract: In this work, we study the problem of learning a single model for multiple domains. Unlike the conventional machine learning scenario where each domain can have the corresponding model, multiple domains (i.e., applications/users) may share the same machine learning model due to maintenance loads in cloud computing services. For example, a digit-recognition model should be applicable to hand-written digits, house numbers, car plates, etc. Therefore, an ideal model for cloud computing has to perform well at each applicable domain. To address this new challenge from cloud computing, we develop a framework of robust optimization over multiple domains. In lieu of minimizing the empirical risk, we aim to learn a model optimized to the adversarial distribution over multiple domains. Hence, we propose to learn the model and the adversarial distribution simultaneously with the stochastic algorithm for efficiency. Theoretically, we analyze the convergence rate for convex and non-convex models. To our best knowledge, we first study the convergence rate of learning a robust non-convex model with a practical algorithm. Furthermore, we demonstrate that the robustness of the framework and the convergence rate can be further enhanced by appropriate regularizers over the adversarial distribution. The empirical study on real-world fine-grained visual categorization and digits recognition tasks verifies the effectiveness and efficiency of the proposed framework.\",\n \"Title: Multinomial Logit Bandit with Linear Utility Functions.\\nYear: 2018\\n\\nAbstract: Multinomial logit bandit is a sequential subset selection problem which arises in many applications. In each round, the player selects a $K$-cardinality subset from $N$ candidate items, and receives a reward which is governed by a {it multinomial logit} (MNL) choice model considering both item utility and substitution property among items. The playeru0027s objective is to dynamically learn the parameters of MNL model and maximize cumulative reward over a finite horizon $T$. This problem faces the exploration-exploitation dilemma, and the involved combinatorial nature makes it non-trivial. In recent years, there have developed some algorithms by exploiting specific characteristics of the MNL model, but all of them estimate the parameters of MNL model separately and incur a regret no better than $tilde{O}big(sqrt{NT}big)$ which is not preferred for large candidate set size $N$. In this paper, we consider the {it linear utility} MNL choice model whose item utilities are represented as linear functions of $d$-dimension item features, and propose an algorithm, titled {bf LUMB}, to exploit the underlying structure. It is proven that the proposed algorithm achieves $tilde{O}big(dKsqrt{T}big)$ regret which is free of candidate set size. Experiments show the superiority of the proposed algorithm.\",\n \"Title: Detecting communities and their evolutions in dynamic social networks—a Bayesian approach\\nYear: 2011\\n\\nAbstract: Although a large body of work is devoted to finding communities in static social networks, only a few studies examined the dynamics of communities in evolving social networks. In this paper, we propose a dynamic stochastic block model for finding communities and their evolution in a dynamic social network. The proposed model captures the evolution of communities by explicitly modeling the transition of community memberships for individual nodes in the network. Unlike many existing approaches for modeling social networks that estimate parameters by their most likely values (i.e., point estimation), in this study, we employ a Bayesian treatment for parameter estimation that computes the posterior distributions for all the unknown parameters. This Bayesian treatment allows us to capture the uncertainty in parameter values and therefore is more robust to data noise than point estimation. In addition, an efficient algorithm is developed for Bayesian inference to handle large sparse social networks. Extensive experimental studies based on both synthetic data and real-life data demonstrate that our model achieves higher accuracy and reveals more insights in the data than several state-of-the-art algorithms.\",\n \"Title: Visual Search at Alibaba.\\nYear: 2018\\n\\nAbstract: This paper introduces the large scale visual search algorithm and system infrastructure at Alibaba. The following challenges are discussed under the E-commercial circumstance at Alibaba (a) how to handle heterogeneous image data and bridge the gap between real-shot images from user query and the online images. (b) how to deal with large scale indexing for massive updating data. (c) how to train deep models for effective feature representation without huge human annotations. (d) how to improve the user engagement by considering the quality of the content. We take advantage of large image collection of Alibaba and state-of-the-art deep learning techniques to perform visual search at scale. We present solutions and implementation details to overcome those problems and also share our learnings from building such a large scale commercial visual search engine. Specifically, model and search-based fusion approach is introduced to effectively predict categories. Also, we propose a deep CNN model for joint detection and feature learning by mining user click behavior. The binary index engine is designed to scale up indexing without compromising recall and precision. Finally, we apply all the stages into an end-to-end system architecture, which can simultaneously achieve highly efficient and scalable performance adapting to real-shot images. Extensive experiments demonstrate the advancement of each module in our system. We hope visual search at Alibaba becomes more widely incorporated into today's commercial applications.\",\n \"Title: Dynamic Regret of Strongly Adaptive Methods.\\nYear: 2018\\n\\nAbstract: To cope with changing environments, recent developments in online learning have introduced the concepts of adaptive regret and dynamic regret independently. In this paper, we illustrate an intrinsic connection between these two concepts by showing that the dynamic regret can be expressed in terms of the adaptive regret and the functional variation. This observation implies that strongly adaptive algorithms can be directly leveraged to minimize the dynamic regret. As a result, we present a series of strongly adaptive algorithms that have small dynamic regrets for convex functions, exponentially concave functions, and strongly convex functions, respectively. To the best of our knowledge, this is the first time that exponential concavity is utilized to upper bound the dynamic regret. Moreover, all of those adaptive algorithms do not need any prior knowledge of the functional variation, which is a significant advantage over previous specialized methods for minimizing dynamic regret.\",\n \"Title: Secure Inductive-Coupled Near Field Communication at Physical Layer.\\nYear: 2018\\n\\nAbstract: Near field communication (NFC) is widely used today in many useful applications, such as contactless payment, identification, and file exchange. Due to the limitations on computation, power, and cost of NFC devices, NFC systems often lack encryption or are weakly encrypted, leaving them exposed to security attacks. One solution for this problem is to install strong cryptographic protocols on NFC d...\",\n \"Title: A Congestion Control Method Of Sdn Data Center Based On Reinforcement Learning\\nYear: 2018\\n\\nAbstract: With the development of cloud computing and big data, the internal communication business in data center has increased dramatically, and then the traffic in data center has also significantly increased. The bandwidth of data center is difficult to meet the bandwidth requirements of those intensive applications, and data center is facing a risk of network congestion. Under the background of the development of network intelligence, software-defined network (SDN) should demonstrate its intelligence as a future network architecture. In this paper, we introduce reinforcement learning into the SDN data center to implement congestion control based on flow. We improve the Q-learning and Sarsa algorithms and propose two methods of congestion control based on the algorithms. Test results show that these two congestion control methods can control congestion effectively. And Sarsa method has a better performance of link utilization. The average link utilization of the Sarsa method is 2.4% higher than the Q-learning method and is 4.48% higher than the on-demand method.\",\n \"Title: Balancing the QOS and Security in Dijkstra Algorithm by SDN Technology.\\nYear: 2018\\n\\nAbstract: Dijkstra algorithm is widely used in a lot of common network routing protocols. We consider the problem of quality of service (QoS) and the Security features of the network routing area using software defined networks (SDN). The SDN framework enables an efficient decoupled implementation of dynamic routing protocols which could aware the communication network status. In this work we consider the varying delay status of the communication network along with other network security parameters. The routing problem is formulated as a multi-constrained shortest path problem. A new improved Dijkstra algorithm is presented named as QS-Dijkstra. The implement and experiment show that QS-Dijkstra algorithm is able to minimize traffic routing through vulnerable links while satisfying the QoS constraints of the network.\",\n \"Title: A simple homotopy proximal mapping algorithm for compressive sensing\\nYear: 2019\\n\\nAbstract: In this paper, we present novel yet simple homotopy proximal mapping algorithms for reconstructing a sparse signal from (noisy) linear measurements of the signal or for learning a sparse linear model from observed data, where the former task is well-known in the field of compressive sensing and the latter task is known as model selection in statistics and machine learning. The algorithms adopt a simple proximal mapping of the \\\\(\\\\ell _1\\\\) norm at each iteration and gradually reduces the regularization parameter for the \\\\(\\\\ell _1\\\\) norm. We prove a global linear convergence of the proposed homotopy proximal mapping (HPM) algorithms for recovering the sparse signal under three different settings (i) sparse signal recovery under noiseless measurements, (ii) sparse signal recovery under noisy measurements, and (iii) nearly-sparse signal recovery under sub-Gaussian noisy measurements. In particular, we show that when the measurement matrix satisfies restricted isometric properties (RIP), one of the proposed algorithms with an appropriate setting of a parameter based on the RIP constants converges linearly to the optimal solution up to the noise level. In addition, in setting (iii), a practical variant of the proposed algorithms does not rely on the RIP constants and our results for sparse signal recovery are better than the previous results in the sense that our recovery error bound is smaller. Furthermore, our analysis explicitly exhibits that more observations lead to not only more accurate recovery but also faster convergence. Finally our empirical studies provide further support for the proposed homotopy proximal mapping algorithm and verify the theoretical results.\",\n \"Title: Semi-Supervised Boosting for Multi-Class Classification\\nYear: 2008\\n\\nAbstract: Most semi-supervised learning algorithms have been designed for binary classification, and are extended to multi-class classification by approaches such as one-against-the-rest. The main shortcoming of these approaches is that they are unable to exploit the fact that each example is only assigned to one class. Additional problems with extending semi-supervised binary classifiers to multi-class problems include imbalanced classification and different output scales of different binary classifiers. We propose a semi-supervised boosting framework, termed Multi-Class Semi-Supervised Boosting (MCSSB), that directly solves the semi-supervised multi-class learning problem. Compared to the existing semi-supervised boosting methods, the proposed framework is advantageous in that it exploits both classification confidence and similarities among examples when deciding the pseudo-labels for unlabeled examples. Empirical study with a number of UCI datasets shows that the proposed MCSSB algorithm performs better than the state-of-the-art boosting algorithms for semi-supervised learning.\",\n \"Title: Tattoo-ID: automatic tattoo image retrieval for suspect and victim identification\\nYear: 2007\\n\\nAbstract: Tattoos are used by law enforcement agencies for identification of a victim or a suspect using a false identity. Current method for matching tattoos is based on human-assigned class labels that is time consuming, subjective and has limited performance. It is desirable to build a content-based image retrieval (CBIR) system for automatic matching and retrieval of tattoos. We examine several key design issues related to building a prototype CBIR system for tattoo image database. Our system computes the similarity between the query and stored tattoos based on image content to retrieve the most similar tattoos. The performance of the system is evaluated on a database of 2,157 tattoos representing 20 different classes. Effects of segmentation errors, image transformations (e.g., blurring, illumination), influence of semantic labels and relevance feedback are also studied.\",\n \"Title: Discriminative semi-supervised feature selection via manifold regularization.\\nYear: 2009\\n\\nAbstract: Feature selection has attracted a huge amount of interest in both research and application communities of data mining. We consider the problem of semi-supervised feature selection, where we are given a small amount of labeled examples and a large amount of unlabeled examples. Since a small number of labeled samples are usually insufficient for identifying the relevant features, the critical problem arising from semi-supervised feature selection is how to take advantage of the information underneath the unlabeled data. To address this problem, we propose a novel discriminative semi-supervised feature selection method based on the idea of manifold regularization. The proposed approach selects features through maximizing the classification margin between different classes and simultaneously exploiting the geometry of the probability distribution that generates both labeled and unlabeled data. In comparison with previous semi-supervised feature selection algorithms, our proposed semi-supervised feature selection method is an embedded feature selection method and is able to find more discriminative features. We formulate the proposed feature selection method into a convex-concave optimization problem, where the saddle point corresponds to the optimal solution. To find the optimal solution, the level method, a fairly recent optimization method, is employed. We also present a theoretic proof of the convergence rate for the application of the level method to our problem. Empirical evaluation on several benchmark data sets demonstrates the effectiveness of the proposed semi-supervised feature selection method.\",\n \"Title: Delay Analysis of Physical-Layer Key Generation in Dynamic Roadside-to-Vehicle Networks.\\nYear: 2017\\n\\nAbstract: Secret key generation by extracting the shared randomness in a wireless fading channel is a promising way to ensure wireless communication security. Previous studies only consider key generation in static networks, but real-world key establishments are usually dynamic. In this paper, for the first time, we investigate the pairwise key generation in dynamic wireless networks with a center node and ...\",\n \"Title: Which Factorization Machine Modeling is Better: A Theoretical Answer with Optimal Guarantee.\\nYear: 2019\\n\\nAbstract: Factorization machine (FM) is a popular machine learning model to capture the second order feature interactions. The optimal learning guarantee of FM and its generalized version is not yet developed. For a rank k generalized FM of d dimensional input, the previous best known sampling complexity is O[k(3)d . polylog(kd)] under Gaussian distribution. This bound is sub-optimal comparing to the information theoretical lower bound O(kd). In this work, we aim to tighten this bound towards optimal and generalize the analysis to sub-gaussian distribution. We prove that when the input data satisfies the so-called tau-Moment Invertible Property, the sampling complexity of generalized FM can be improved to O[k(2)d . polylog(kd)/tau(2)]. When the second order self-interaction terms are excluded in the generalized FM, the bound can be improved to the optimal O[kd . polylog(kd)] up to the logarithmic factors. Our analysis also suggests that the positive semi-definite constraint in the conventional FM is redundant as it does not improve the sampling complexity while making the model difficult to optimize We evaluate our improved FM model in real-time high precision GPS signal calibration task to validate its superiority.\",\n \"Title: Conservative Exploration for Semi-Bandits with Linear Generalization: A Product Selection Problem for Urban Warehouses.\\nYear: 2019\\n\\nAbstract: The recent rising popularity of ultra-fast delivery services on retail platforms fuels increasing use of urban warehouses, whose proximity to customers makes fast deliveries viable. The space limit in urban warehouses poses a problem for online retailers: number of products (SKUs) they carry is no longer the more, better, yet it can still be significantly large, reaching hundreds or thousands in a product category. In this paper, we study algorithms for dynamically identifying a large number of products (i.e., SKUs) with top customer purchase probabilities on fly, from an ocean of potential products to offer on retailersu0027 ultra-fast delivery platforms. distill product selection problem into a semi-bandit model with linear generalization. There are in total $N$ different arms, each with a feature vector of dimension $d$. The player pulls $K$ arms in each period and observes bandit feedback from each of pulled arms. We focus on setting where $K$ is much greater than number of total time periods $T$ or dimension of product features $d$. We first analyze a standard UCB algorithm and show its regret bound can be expressed as sum of a $T$-independent part $tilde O(K d^{3/2})$ and a $T$-dependent part $tilde O(dsqrt{KT})$, which we refer to as and variable respectively. To reduce fixed cost for large $K$ values, we propose a novel online learning algorithm, with more conservative exploration steps, and show its fixed cost is reduced by a factor of $d$ to $tilde O(K sqrt{d})$. Moreover, we test algorithms on an industrial dataset from Alibaba Group. Experimental results show that our new algorithm reduces total regret of standard UCB algorithm by at least 10%.\",\n \"Title: Stochastic Optimization for DC Functions and Non-smooth Non-convex Regularizers with Non-asymptotic Convergence\\nYear: 2019\\n\\nAbstract: Difference of convex (DC) functions cover a broad family of non-convex and possibly non-smooth and non-differentiable functions, and have wide applications in machine learning and statistics. Although deterministic algorithms for DC functions have been extensively studied, stochastic optimization that is more suitable for learning with big data remains under-explored. In this paper, we propose new stochastic optimization algorithms and study their first-order convergence theories for solving a broad family of DC functions. We improve the existing algorithms and theories of stochastic optimization for DC functions from both practical and theoretical perspectives. On the practical side, our algorithm is more user-friendly without requiring a large mini-batch size and more efficient by saving unnecessary computations. On the theoretical side, our convergence analysis does not necessarily require the involved functions to be smooth with Lipschitz continuous gradient. Instead, the convergence rate of the proposed stochastic algorithm is automatically adaptive to the H\\\\\\\"{o}lder continuity of the gradient of one component function. Moreover, we extend the proposed stochastic algorithms for DC functions to solve problems with a general non-convex non-differentiable regularizer, which does not necessarily have a DC decomposition but enjoys an efficient proximal mapping. To the best of our knowledge, this is the first work that gives the first non-asymptotic convergence for solving non-convex optimization whose objective has a general non-convex non-differentiable regularizer.\",\n \"Title: Publishing Social Network Graph Eigen-Spectrum with Privacy Guarantees\\nYear: 2020\\n\\nAbstract: Online social networks (OSNs) often refuse to publish their social network graphs due to privacy concerns. Recently, differential privacy has become the widely accepted criteria for privacy preserving data publishing. Although some work has been done on publishing matrices with differential privacy, they are computationally unpractical as they are not designed to handle large matrices such as adja...\",\n \"Title: Distributed Structured Compressive Sensing-Based Time-Frequency Joint Channel Estimation for Massive MIMO-OFDM Systems\\nYear: 2019\\n\\nAbstract: In massive multi-input multi-output orthogonal frequency division multiplexing (MIMO-OFDM) systems, accurate channel state information (CSI) is essential to realize system performance gains such as high spectrum and energy efficiency. However, high-dimensional CSI acquisition requires prohibitively high pilot overhead, which leads to a significant reduction in spectrum efficiency and energy efficiency. In this paper, we propose a more efficient time-frequency joint channel estimation scheme for massive MIMO-OFDM systems to resolve those problems. First, partial channel common support (PCCS) is obtained by using time-domain training. Second, utilizing the spatiotemporal common sparse property of the MIMO channels and the obtained PCCS information, we propose the priori-information aided distributed structured sparsity adaptive matching pursuit (PA-DS-SAMP) algorithm to achieve accurate channel estimation in frequency domain. Third, through performance analysis of the proposed algorithm, two signal power reference thresholds are given, which can ensure that the signal can be recovered accurately under power-limited noise and accurately recovered according to probability under Gaussian noise. Finally, pilot design, computational complexity, spectrum efficiency, and energy efficiency are discussed as well. Simulation results show that the proposed method achieves higher channel estimation accuracy while requiring lower pilot sequence overhead compared with other methods.\",\n \"Title: On the Computation and Communication Complexity of Parallel SGD with Dynamic Batch Sizes for Stochastic Non-Convex Optimization.\\nYear: 2019\\n\\nAbstract: For SGD based distributed stochastic optimization, computation complexity, measured by the convergence rate in terms of the number of stochastic gradient calls, and communication complexity, measured by the number of inter-node communication rounds, are two most important performance metrics. The classical data-parallel implementation of SGD over $N$ workers can achieve linear speedup of its convergence rate but incurs an inter-node communication round at each batch. We study the benefit of using dynamically increasing batch sizes in parallel SGD for stochastic non-convex optimization by charactering the attained convergence rate and the required number of communication rounds. We show that for stochastic non-convex optimization under the P-L condition, the classical data-parallel SGD with exponentially increasing batch sizes can achieve the fastest known $O(1/(NT))$ convergence with linear speedup using only $\\\\log(T)$ communication rounds. For general stochastic non-convex optimization, we propose a Catalyst-like algorithm to achieve the fastest known $O(1/\\\\sqrt{NT})$ convergence with only $O(\\\\sqrt{NT}\\\\log(\\\\frac{T}{N}))$ communication rounds.\",\n \"Title: Robust Gaussian Process Regression for Real-Time High Precision GPS Signal Enhancement.\\nYear: 2019\\n\\nAbstract: Satellite-based positioning system such as GPS often suffers from large amount of noise that degrades the positioning accuracy dramatically especially in real-time applications. In this work, we consider a data-mining approach to enhance the GPS signal. We build a large-scale high precision GPS receiver grid system to collect real-time GPS signals for training. The Gaussian Process (GP) regression is chosen to model the vertical Total Electron Content (vTEC) distribution of the ionosphere of the Earth. Our experiments show that the noise in the real-time GPS signals often exceeds the breakdown point of the conventional robust regression methods resulting in sub-optimal system performance. We propose a three-step approach to address this challenge. In the first step we perform a set of signal validity tests to separate the signals into clean and dirty groups. In the second step, we train an initial model on the clean signals and then reweigting the dirty signals based on the residual error. A final model is retrained on both the clean signals and the reweighted dirty signals. In the theoretical analysis, we prove that the proposed three-step approach is able to tolerate much higher noise level than the vanilla robust regression methods if two reweighting rules are followed. We validate the superiority of the proposed method in our real-time high precision positioning system against several popular state-of-the-art robust regression methods. Our method achieves centimeter positioning accuracy in the benchmark region with probability $78.4%$ , outperforming the second best baseline method by a margin of $8.3%$. The benchmark takes 6 hours on 20,000 CPU cores or 14 years on a single CPU.\",\n \"Title: XNAS: Neural Architecture Search with Expert Advice.\\nYear: 2019\\n\\nAbstract: This paper introduces a novel optimization method for differential neural architecture search, based on the theory of prediction with expert advice. Its optimization criterion is well fitted for an architecture-selection, i.e., it minimizes the regret incurred by a sub-optimal selection of operations. Unlike previous search relaxations, that require hard pruning of architectures, our method is designed to dynamically wipe out inferior architectures and enhance superior ones. It achieves an optimal worst-case regret bound and suggests the use of multiple learning-rates, based on the amount of information carried by the backward gradients. Experiments show that our algorithm achieves a strong performance over several image classification datasets. Specifically, it obtains an error rate of 1.6% for CIFAR-10, 23.9% for ImageNet under mobile settings, and achieves state-of-the-art results on three additional datasets.\",\n \"Title: Train A One-Million-Way Instance Classifier For Unsupervised Visual Representation Learning\\nYear: 2021\\n\\nAbstract: This paper presents a simple unsupervised visual representation learning method with a pretext task of discriminating all images in a dataset using a parametric, instance-level classifier. The overall framework is a replica of a supervised classification model, where semantic classes (e.g., dog, bird, and ship) are replaced by instance IDs. However, scaling up the classification task from thousands of semantic labels to millions of instance labels brings specific challenges including 1) the large-scale softmax computation; 2) the slow convergence due to the infrequent visiting of instance samples; and 3) the massive number of negative classes that can be noisy. This work presents several novel techniques to handle these difficulties. First, we introduce a hybrid parallel training framework to make large-scale training feasible. Second, we present a raw-feature initialization mechanism for classification weights, which we assume offers a contrastive prior for instance discrimination and can clearly speed up converge in our experiments. Finally, we propose to smooth the labels of a few hardest classes to avoid optimizing over very similar negative pairs. While being conceptually simple, our framework achieves competitive or superior performance compared to state-of-the-art unsupervised approaches, i.e., SimCLR, MoCoV2, and PIC under ImageNet linear evaluation protocol and on several downstream visual tasks, verifying that full instance classification is a strong pretraining technique for many semantic visual tasks.\",\n \"Title: Communication Efficient SGD via Gradient Sampling with Bayes Prior\\nYear: 2021\\n\\nAbstract: Gradient compression has been widely adopted in data-parallel distributed training of deep neural networks to reduce communication overhead. Some literatures have demonstrated that large gradients are more important than small ones because they contain more information, such as Top-k compressor. Other mainstream methods, like random-k compressor and gradient quantization, usually treat all gradients equally. Different from all of them, we regard large and small gradients selection as the exploitation and exploration of gradient information, respectively. And we find taking both of them into consideration is the key to boost the final accuracy. So, we propose a novel gradient compressor: Gradient Sampling with Bayes Prior in this paper. Specifically, we sample important/large gradients based on the global gradient distribution, which is periodically updated across multiple workers. Then we introduce Bayes Prior into distribution model to further explore the gradients. We prove the convergence of our method for smooth non-convex problems in the distributed system. Compared with methods that running after high compression ratio at the expense of accuracy, we pursue no loss of accuracy and the actual acceleration benefit in practice. Experimental comparisons on a variety of computer vision tasks (e.g. image classification and object detection) and backbones (ResNet, MobileNetV2, InceptionV3 and AlexNet) show that our approach outperforms the state-of-the-art techniques in terms of both speed and accuracy, with the limitation of 100x compression ratio.\",\n \"Title: Effective Model Sparsification by Scheduled Grow-and-Prune Methods\\nYear: 2022\\n\\nAbstract: Deep neural networks (DNNs) are effective in solving many real-world problems. Larger DNN models usually exhibit better quality (e.g., accuracy) but their excessive computation results in long training and inference time. Model sparsification can reduce the computation and memory cost while maintaining model quality. Most existing sparsification algorithms unidirectionally remove weights, while others randomly or greedily explore a small subset of weights in each layer. The inefficiency of the algorithms reduces the achievable sparsity level. In addition, many algorithms still require pre-trained dense models and thus suffer from large memory footprint and long training time. In this paper, we propose a novel scheduled grow-and-prune (GaP) methodology without pre-training the dense models. It addresses the shortcomings of the previous works by repeatedly growing a subset of layers to dense and then pruning back to sparse after some training. Experiments have shown that such models can match or beat the quality of highly optimized dense models at 80% sparsity on a variety of tasks, such as image classification, objective detection, 3D object part segmentation, and translation. They also outperform other state-of-the-art (SOTA) pruning methods, including pruning from pre-trained dense models. As an example, a 90% sparse ResNet-50 obtained via GaP achieves 77.9% top-1 accuracy on ImageNet, improving the SOTA results by 1.5%.\",\n \"Title: Graph Convolution for Re-Ranking in Person Re-Identification.\\nYear: 2022\\n\\nAbstract: Nowadays, deep learning is widely applied to extract features for similarity computation in person re-identification (re-ID) and have achieved great success. However, due to the non-overlapping between training and testing IDs, the difference between the data used for model training and the testing data makes the performance of learned feature degraded during testing. Hence, re-ranking is proposed to mitigate this issue and various algorithms have been developed. However, most of existing re-ranking methods focus on replacing the Euclidean distance with sophisticated distance metrics, which are not friendly to downstream tasks and hard to be used for fast retrieval of massive data in real applications. In this work, we propose a graph-based re-ranking method to improve learned features while still keeping Euclidean distance as the similarity metric. Inspired by graph convolution networks, we develop an operator to propagate features over an appropriate graph. Since graph is the essential key for the propagation, two important criteria are considered for designing the graph, and three different graphs are explored accordingly. Furthermore, a simple yet effective method is proposed to generate a profile vector for each tracklet in videos, which helps extend our method to video re-ID. Extensive experiments on three benchmark data sets, e.g., Market-1501, Duke, and MARS, demonstrate the effectiveness of our proposed approach.\",\n \"Title: Scaled ReLU Matters for Training Vision Transformers.\\nYear: 2022\\n\\nAbstract: Vision transformers (ViTs) have been an alternative design paradigm to convolutional neural networks (CNNs). However, the training of ViTs is much harder than CNNs, as it is sensitive to the training parameters, such as learning rate, optimizer and warmup epoch. The reasons for training difficulty are empirically analysed in the paper Early Convolutions Help Transformers See Better, and the authors conjecture that the issue lies with the patchify-stem of ViT models. In this paper, we further investigate this problem and extend the above conclusion: only early convolutions do not help for stable training, but the scaled ReLU operation in the convolutional stem (conv-stem) matters. We verify, both theoretically and empirically, that scaled ReLU in conv-stem not only improves training stabilization, but also increases the diversity of patch tokens, thus boosting peak performance with a large margin via adding few parameters and flops. In addition, extensive experiments are conducted to demonstrate that previous ViTs are far from being well trained, further showing that ViTs have great potential to be a better substitute of CNNs.\",\n \"Title: CDTrans: Cross-domain Transformer for Unsupervised Domain Adaptation\\nYear: 2022\\n\\nAbstract: Unsupervised domain adaptation (UDA) aims to transfer knowledge learned from a labeled source domain to a different unlabeled target domain. Most existing UDA methods focus on learning domain-invariant feature representation, either from the domain level or category level, using convolution neural networks (CNNs)-based frameworks. One fundamental problem for the category level based UDA is the production of pseudo labels for samples in target domain, which are usually too noisy for accurate domain alignment, inevitably compromising the UDA performance. With the success of Transformer in various tasks, we find that the cross-attention in Transformer is robust to the noisy input pairs for better feature alignment, thus in this paper Transformer is adopted for the challenging UDA task. Specifically, to generate accurate input pairs, we design a two-way center-aware labeling algorithm to produce pseudo labels for target samples. Along with the pseudo labels, a weight-sharing triple-branch transformer framework is proposed to apply self-attention and cross-attention for source/target feature learning and source-target domain alignment, respectively. Such design explicitly enforces the framework to learn discriminative domain-specific and domain-invariant representations simultaneously. The proposed method is dubbed CDTrans (cross-domain transformer), and it provides one of the first attempts to solve UDA tasks with a pure transformer solution. Extensive experiments show that our proposed method achieves the best performance on Office-Home, VisDA-2017, and DomainNet datasets.\",\n \"Title: Rethinking Supervised Pre-Training for Better Downstream Transferring\\nYear: 2022\\n\\nAbstract: The pretrain-finetune paradigm has shown outstanding performance on many applications of deep learning, where a model is pre-trained on a upstream large dataset (e.g. ImageNet), and is then fine-tuned to different downstream tasks. Though for most cases, the pre-training stage is conducted based on supervised methods, recent works on self-supervised pre-training have shown powerful transferability and even outperform supervised pre-training on multiple downstream tasks. It thus remains an open question how to better generalize supervised pre-training model to downstream tasks. In this paper, we argue that the worse transferability of existing supervised pre-training methods arise from the negligence of valuable intra-class semantic difference. This is because these methods tend to push images from the same class close to each other despite of the large diversity in their visual contents, a problem to which referred as \\\"overfit of upstream tasks\\\". To alleviate this problem, we propose a new supervised pre-training method based on Leave-One-Out K-Nearest-Neighbor, or LOOK for short. It relieves the problem of overfitting upstream tasks by only requiring each image to share its class label with most of its k nearest neighbors, thus allowing each class to exhibit a multi-mode distribution and consequentially preserving part of intra-class difference for better transferring to downstream tasks. We developed efficient implementation of the proposed method that scales well to large datasets. Experimental studies on multiple downstream tasks show that LOOK outperforms other state-of-the-art methods for supervised and self-supervised pre-training.\",\n \"Title: Weakly Supervised Representation Learning with Coarse Labels.\\nYear: 2021\\n\\nAbstract: With the development of computational power and techniques for data collection, deep learning demonstrates a superior performance over most existing algorithms on visual benchmark data sets. Many efforts have been devoted to studying the mechanism of deep learning. One important observation is that deep learning can learn the discriminative patterns from raw materials directly in a task-dependent manner. Therefore, the representations obtained by deep learning outperform hand-crafted features significantly. However, for some real-world applications, it is too expensive to collect the task-specific labels, such as visual search in online shopping. Compared to the limited availability of these task-specific labels, their coarse-class labels are much more affordable, but representations learned from them can be suboptimal for the target task. To mitigate this challenge, we propose an algorithm to learn the fine-grained patterns for the target task, when only its coarse-class labels are available. More importantly, we provide a theoretical guarantee for this. Extensive experiments on real-world data sets demonstrate that the proposed method can significantly improve the performance of learned representations on the target task, when only coarse-class information is available for training.\",\n \"Title: KVT: k-NN Attention for Boosting Vision Transformers.\\nYear: 2022\\n\\nAbstract: Convolutional Neural Networks (CNNs) have dominated computer vision for years, due to its ability in capturing locality and translation invariance. Recently, many vision transformer architectures have been proposed and they show promising performance. A key component in vision transformers is the fully-connected self-attention which is more powerful than CNNs in modelling long range dependencies. However, since the current dense self-attention uses all image patches (tokens) to compute attention matrix, it may neglect locality of images patches and involve noisy tokens (e.g., clutter background and occlusion), leading to a slow training process and potential degradation of performance. To address these problems, we propose the k-NN attention for boosting vision transformers. Specifically, instead of involving all the tokens for attention matrix calculation, we only select the top-k similar tokens from the keys for each query to compute the attention map. The proposed k-NN attention naturally inherits the local bias of CNNs without introducing convolutional operations, as nearby tokens tend to be more similar than others. In addition, the k-NN attention allows for the exploration of long range correlation and at the same time filters out irrelevant tokens by choosing the most similar tokens from the entire image. Despite its simplicity, we verify, both theoretically and empirically, that k-NN attention is powerful in speeding up training and distilling noise from input tokens. Extensive experiments are conducted by using 11 different vision transformer architectures to verify that the proposed k-NN attention can work with any existing transformer architectures to improve its prediction performance. The codes are available at https://github.com/damo-cv/KVT.\",\n \"Title: An Online Method for A Class of Distributionally Robust Optimization with Non-convex Objectives.\\nYear: 2021\\n\\nAbstract: In this paper, we propose a practical online method for solving a class of distributional robust optimization (DRO) with non-convex objectives, which has important applications in machine learning for improving the robustness of neural networks. In the literature, most methods for solving DRO are based on stochastic primal-dual methods. However, primal-dual methods for DRO suffer from several drawbacks: (1) manipulating a high-dimensional dual variable corresponding to the size of data is time expensive; (2) they are not friendly to online learning where data is coming sequentially. To address these issues, we consider a class of DRO with an KL divergence regularization on the dual variables, transform the min-max problem into a compositional minimization problem, and propose practical duality-free online stochastic methods without requiring a large mini-batch size. We establish the state-of-the-art complexities of the proposed methods with and without a Polyak-Łojasiewicz (PL) condition of the objective. Empirical studies on large-scale deep learning tasks (i) demonstrate that our method can speed up the training by more than 2 times than baseline methods and save days of training time on a large-scale dataset with ∼ 265K images, and (ii) verify the supreme performance of DRO over Empirical Risk Minimization (ERM) on imbalanced datasets. Of independent interest, the proposed method can be also used for solving a family of stochastic compositional problems with state-of-the-art complexities.\",\n \"Title: TransFGU: A Top-Down Approach to Fine-Grained Unsupervised Semantic Segmentation.\\nYear: 2022\\n\\nAbstract: Unsupervised semantic segmentation aims to obtain high-level semantic representation on low-level visual features without manual annotations. Most existing methods are bottom-up approaches that try to group pixels into regions based on their visual cues or certain predefined rules. As a result, it is difficult for these bottom-up approaches to generate fine-grained semantic segmentation when coming to complicated scenes with multiple objects and some objects sharing similar visual appearance. In contrast, we propose the first top-down unsupervised semantic segmentation framework for fine-grained segmentation in extremely complicated scenarios. Specifically, we first obtain rich high-level structured semantic concept information from large-scale vision data in a self-supervised learning manner, and use such information as a prior to discover potential semantic categories presented in target datasets. Secondly, the discovered high-level semantic categories are mapped to low-level pixel features by calculating the class activate map (CAM) with respect to certain discovered semantic representation. Lastly, the obtained CAMs serve as pseudo labels to train the segmentation module and produce the final semantic segmentation. Experimental results on multiple semantic segmentation benchmarks show that our top-down unsupervised segmentation is robust to both object-centric and scene-centric datasets under different semantic granularity levels, and outperforms all the current state-of-the-art bottom-up methods. Our code is available at https://github.com/damo-cv/TransFGU.\",\n \"Title: Decoupling and Recoupling Spatiotemporal Representation for RGB-D-based Motion Recognition\\nYear: 2022\\n\\nAbstract: Decoupling spatiotemporal representation refers to decomposing the spatial and temporal features into dimension-independent factors. Although previous RGB-D-based motion recognition methods have achieved promising performance through the tightly coupled multi-modal spatiotemporal representation, they still suffer from (i) optimization difficulty under small data setting due to the tightly spatiotemporal-entangled modeling; (ii) information redundancy as it usually contains lots of marginal information that is weakly relevant to classification; and (iii) low interaction between multi-modal spatiotemporal information caused by insufficient late fusion. To alleviate these drawbacks, we propose to decouple and recouple spatiotemporal representation for RGB-D-based motion recognition. Specifically, we disentangle the task of learning spatiotemporal representation into 3 sub-tasks: (1) Learning high-quality and dimension independent features through a decoupled spatial and temporal modeling network. (2) Recoupling the decoupled representation to establish stronger space-time dependency. (3) Introducing a Cross-modal Adaptive Posterior Fusion (CAPF) mechanism to capture cross-modal spatiotemporal information from RGB-D data. Seamless combination of these novel designs forms a robust spatiotemporal representation and achieves better performance than state-of-the-art methods on four public motion datasets. Our code is available at https://github.com/damo-cv/MotionRGBD.\",\n \"Title: FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecasting.\\nYear: 2022\\n\\nAbstract: Long-term time series forecasting is challenging since prediction accuracy tends to decrease dramatically with the increasing horizon. Although Transformer-based methods have significantly improved state-of-the-art results for long-term forecasting, they are not only computationally expensive but more importantly, are unable to capture the global view of time series (e.g. overall trend). To address these problems, we propose to combine Transformer with the seasonal-trend decomposition method, in which the decomposition method captures the global profile of time series while Transformers capture more detailed structures. To further enhance the performance of Transformer for long-term prediction, we exploit the fact that most time series tend to have a sparse representation in a well-known basis such as Fourier transform, and develop a frequency enhanced Transformer. Besides being more effective, the proposed method, termed as Frequency Enhanced Decomposed Transformer (FEDformer), is more efficient than standard Transformer with a linear complexity to the sequence length. Our empirical studies with six benchmark datasets show that compared with state-of-the-art methods, Fedformer can reduce prediction error by 14.8% and 22.6% for multivariate and univariate time series, respectively. Code is publicly available at https://github.com/MAZiqing/FEDformer.\",\n \"Title: A Trend-Driven Fashion Design System for Rapid Response Marketing in E-commerce.\\nYear: 2022\\n\\nAbstract: Fashion is the form we express ourselves to the world and has grown into one of the largest industries in the world. Despite the significant evolvement of the fashion industry over the past decades, it is still a great challenge to respond to the diverse preferences of a large number of different consumers in time and accurately. To deal with the problem, we present an innovative demonstration of a trend-driven fashion design system using deep generative modeling, which enables automatic fashion design and editing based on trend reports. Our system consists of three components, including trend-driven fashion design, interactive fashion editing, and popularity estimation. The system offers a unified framework for the mass production of fashion designs that conform to the trend, which helps businesses better respond to market demands.\",\n \"Title: CHEX: CHannel EXploration for CNN Model Compression\\nYear: 2022\\n\\nAbstract: Channel pruning has been broadly recognized as an effective technique to reduce the computation and memory cost of deep convolutional neural networks. However, conventional pruning methods have limitations in that: they are restricted to pruning process only, and they require a fully pre-trained large model. Such limitations may lead to sub-optimal model quality as well as excessive memory and training cost. In this paper, we propose a novel Channel Exploration methodology, dubbed as CHEX, to rectify these problems. As opposed to pruning-only strategy, we propose to repeatedly prune and regrow the channels throughout the training process, which reduces the risk of pruning important channels prematurely. More exactly: From intra-Layer's aspect, we tackle the channel pruning problem via a well-known column subset selection (CSS) formulation. From inter-Layer's aspect, our regrowing stages open a path for dynamically re-allocating the number of channels across all the layers under a global channel sparsity constraint. In addition, all the exploration process is done in a single training from scratch without the need of a pre-trained large model. Experimental results demonstrate that CHEX can effectively reduce the FLOPs of diverse CNN architectures on a variety of computer vision tasks, including image classification, object detection, instance segmentation, and 3D vision. For example, our compressed ResNet-50 model on ImageNet dataset achieves 76% top-l accuracy with only 25% FLOPs of the original ResNet-50 model, outperforming previous state-of-the-art channel pruning methods. The checkpoints and code are available at here.\",\n \"Title: Learning from Untrimmed Videos: Self-Supervised Video Representation Learning with Hierarchical Consistency\\nYear: 2022\\n\\nAbstract: Natural videos provide rich visual contents for selfsupervised learning. Yet most existing approaches for learning spatio-temporal representations rely on manually trimmed videos, leading to limited diversity in visual patterns and limited performance gain. In this work, we aim to learn representations by leveraging more abundant information in untrimmed videos. To this end, we propose to learn a hierarchy of consistencies in videos, i.e., visual consistency and topical consistency, corresponding respectively to clip pairs that tend to be visually similar when separated by a short time span and share similar topics when separated by a long time span. Specifically, a hierarchical consistency learning framework HiCo is presented, where the visually consistent pairs are encouraged to have the same representation through contrastive learning, while the topically consistent pairs are coupled through a topical classifier that distinguishes whether they are topicrelated. Further, we impose a gradual sampling algorithm for proposed hierarchical consistency learning, and demonstrate its theoretical superiority. Empirically, we show that not only HiCo can generate stronger representations on untrimmed videos, it also improves the representation quality when applied to trimmed videos. This is in contrast to standard contrastive learning that fails to learn appropriate representations from untrimmed videos.\",\n \"Title: Entroformer: A Transformer-based Entropy Model for Learned Image Compression\\nYear: 2022\\n\\nAbstract: One critical component in lossy deep image compression is the entropy model, which predicts the probability distribution of the quantized latent representation in the encoding and decoding modules. Previous works build entropy models upon convolutional neural networks which are inefficient in capturing global dependencies. In this work, we propose a novel transformer-based entropy model, termed Entroformer, to capture long-range dependencies in probability distribution estimation effectively and efficiently. Different from vision transformers in image classification, the Entroformer is highly optimized for image compression, including a top-k self-attention and a diamond relative position encoding. Meanwhile, we further expand this architecture with a parallel bidirectional context model to speed up the decoding process. The experiments show that the Entroformer achieves state-of-the-art performance on image compression while being time-efficient.\",\n \"Title: Learning to Generalize to More: Continuous Semantic Augmentation for Neural Machine Translation\\nYear: 2022\\n\\nAbstract: The principal task in supervised neural machine translation (NMT) is to learn to generate target sentences conditioned on the source inputs from a set of parallel sentence pairs, and thus produce a model capable of generalizing to unseen instances. However, it is commonly observed that the generalization performance of the model is highly influenced by the amount of parallel data used in training. Although data augmentation is widely used to enrich the training data, conventional methods with discrete manipulations fail to generate diverse and faithful training samples. In this paper, we present a novel data augmentation paradigm termed Continuous Semantic Augmentation (CSANMT), which augments each training instance with an adjacency semantic region that could cover adequate variants of literal expression under the same meaning. We conduct extensive experiments on both rich-resource and low-resource settings involving various language pairs, including WMT14 English ->{German,French}, NIST Chinese -> English and multiple low-resource IWSLT translation tasks. The provided empirical evidences show that CSANMT sets a new level of performance among existing augmentation techniques, improving on the state-of-the-art by a large margin.(1)\",\n \"Title: Hybrid Relation Guided Set Matching for Few-shot Action Recognition\\nYear: 2022\\n\\nAbstract: Current few-shot action recognition methods reach impressive performance by learning discriminative features for each video via episodic training and designing various temporal alignment strategies. Nevertheless, they are limited in that (a) learning individual features without considering the entire task may lose the most relevant information in the current episode, and (b) these alignment strategies may fail in misaligned instances. To overcome the two limitations, we propose a novel Hybrid Relation guided Set Matching (HyRSM) approach that incorporates two key components: hybrid relation module and set matching metric. The purpose of the hybrid relation module is to learn task-specific embeddings by fully exploiting associated relations within and cross videos in an episode. Built upon the task-specific features, we reformulate distance measure between query and support videos as a set matching problem and further design a bidirectional Mean Hausdorff Metric to improve the resilience to misaligned instances. By this means, the proposed HyRSM can be highly informative and flexible to predict query categories under the few-shot settings. We evaluate HyRSM on six challenging benchmarks, and the experimental results show its superiority over the state-of-the-art methods by a convincing margin. Project page: https://hyrsm-cvpr2022.github.io/.\",\n \"Title: Low- and Moderate-Level RFI Detection Using Point-Source Ripple for Synthetic Aperture Interferometric Radiometer\\nYear: 2022\\n\\nAbstract: Finding out and turning off the radio-frequency interference (RFI) sources as many as possible is an important work for synthetic aperture interferometric radiometer (SAIR). While the impact of strong RFIs on reconstructed brightness temperature (BT) image is clearly observable, the presence of low- or moderate-level RFIs is often hard to be aware of, leaving the RFI detection and localization a difficult problem. In this article, an improved RFI detection method is proposed by making use of the point-source ripple to give prominence to the RFIs presented in the SAIR imagery. The method tests based on Soil Moisture and Ocean Salinity satellite (SMOS) visibility samples and the performance evaluations by numerical simulations validate the correctness and effectiveness of the developed method.\",\n \"Title: ANN Softmax: Acceleration of Extreme Classification Training\\nYear: 2021\\n\\nAbstract: Thanks to the popularity of GPU and the growth of its computational power, more and more deep learning tasks, such as face recognition, image retrieval and word embedding, can take advantage of extreme classification to improve accuracy. However, it remains a big challenge to train a deep model with millions of classes efficiently due to the huge memory and computation consumption in the last layer. By sampling a small set of classes to avoid the total classes calculation, sampling-based approaches have been proved to be an effective solution. But most of them suffer from the following two issues: i) the important classes are ignored or only partly sampled, such as the methods using random sampling scheme or retrieval techniques of low recall (e.g., locality-sensitive hashing), resulting in the degradation of accuracy; ii) inefficient implementation owing to incompatibility with GPU, like selective softmax. It uses hashing forest to help select classes, but the search process is implemented in CPU. To address the above issues, we propose a new sampling-based softmax called ANN Softmax in this paper. Specifically, we employ binary quantization with inverted file system to improve the recall of important classes. With the help of dedicated kernel design, it can be totally parallelized in mainstream training framework. Then, we find the size of important classes that are recalled by each training sample has a great impact on the final accuracy, so we introduce sample grouping optimization to well approximate the full classes training. Experimental evaluations on two tasks (Embedding Learning and Classification) and ten datasets (e.g., MegaFace, ImageNet, SKU datasets) demonstrate our proposed method maintains the same precision as Full Softmax for different loss objectives, including cross entropy loss, ArcFace, CosFace and D-Softmax loss, with only 1/10 sampled classes, which outperforms the state-of-the-art techniques. Moreover, we implement ANN Softmax in a complete GPU pipeline that can accelerate the training more than 4.3x. Equipped our method with a 256 GPUs cluster, the time of training a classifier of 300 million classes on our SKU-300M dataset can be reduced to ten days.\",\n \"Title: Joint Inversion Algorithm of Sea Surface Temperature From Microwave and Infrared Brightness Temperature\\nYear: 2022\\n\\nAbstract: The demand for high-precision sea surface temperature (SST) has been growing rapidly in recent years because SST is one of the key parameters to describe the thermal state of the sea surface. This article analyzes the differences between microwave remote sensing and infrared remote sensing for SST, including the spatial resolution difference and the penetration depth difference. In order to improve the accuracy of retrieved SST, this article proposes a joint inversion algorithm for SST from combining microwave brightness temperature (BT) and infrared BT, which has also taken the influence of wind speed and atmosphere into consideration. Experiments confirm that SST data obtained from the joint inversion algorithm are more accurate than those obtained from the existing inversion algorithms.\",\n \"Title: Estimation of Dry Matter and N Nutrient Status of Choy Sum by Analyzing Canopy Images and Plant Height Information\\nYear: 2022\\n\\nAbstract: The estimation accuracy of plant dry matter by spectra- or remote sensing-based methods tends to decline when canopy coverage approaches closure; this is known as the saturation problem. This study aimed to enhance the estimation accuracy of plant dry matter and subsequently use the critical nitrogen dilution curve (CNDC) to diagnose N in Choy Sum by analyzing the combined information of canopy imaging and plant height. A three-year experiment with different N levels (0, 25, 50, 100, 150, and 200 kg center dot ha(-1)) was conducted on Choy Sum. Variables of canopy coverage (CC) and plant height were used to build the dry matter and N estimation model. The results showed that the yields of N-0 and N-25 were significantly lower than those of high-N treatments (N-50, N-100, N-150, and N-200) for all three years. The variables of CC x Height had a significant linear relationship with dry matter, with R-2 values above 0.87. The good performance of the CC x Height-based model implied that the saturation problem of dry matter prediction was well-addressed. By contrast, the relationship between dry matter and CC was best fitted by an exponential function. CNDC models built based on CC x Height information could satisfactorily differentiate groups of N deficiency and N abundance treatments, implying their feasibility in diagnosing N status. N application rates of 50-100 kgN/ha are recommended as optimal for a good yield of Choy Sum production in the study region.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":200,"string":"200"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13577,"string":"13,577"}}},{"rowIdx":13578,"cells":{"question_id":{"kind":"string","value":"Jie Cao"},"question":{"kind":"string","value":"Generate an abstract for the title \"A novel noise filter based on interesting pattern mining for bag-of-features images.\" using the following items: 1. Cosine interesting pattern\n2. Noise filter\n3. Shortest patterns\n4. Mining process\n5. Pruning technique."},"answer":{"kind":"string","value":"•We first propose to utilize the cosine interesting pattern to construct the noise filter.•We then prove that to filter noise only requires to mine the shortest cosine interesting patterns, which dramatically simplifies the mining process.•We present an in-breadth pruning technique to further speed up the mining process."},"correct_answer":{"kind":"string","value":"•We first propose to utilize the cosine interesting pattern to construct the noise filter.•We then prove that to filter noise only requires to mine the shortest cosine interesting patterns, which dramatically simplifies the mining process.•We present an in-breadth pruning technique to further speed up the mining process."},"documents":{"kind":"list like","value":["Title: Exploiting space-time status for service recommendation\nYear: 2012\n\nAbstract: The prevalence of smart devices allows people to record their space-time status. This paper focuses on exploiting user space-time status and the related semantic information for service recommendation. Firstly, event DAG is employed to organize the space-time information generated based on the service invocation history. Generation algorithm of the event DAG is then proposed. Secondly, a novel collaborative filtering based recommendation algorithm is designed. Potentially interesting services in the target node and its subsequent nodes can be recommended. In our implementation, the user space-time status is generated from the 4D city models (3D location + time) with semantic information. A prototype system is implemented to generate service invocation logs of different users. These simulative logs are utilized to evaluate the effectiveness and efficiency of our proposed method.","Title: A theoretic framework of K-means-based consensus clustering\nYear: 2013\n\nAbstract: Consensus clustering emerges as a promising solution to find cluster structures from data. As an efficient approach for consensus clustering, the K-means based method has garnered attention in the literature, but the existing research is still preliminary and fragmented. In this paper, we provide a systematic study on the framework of K-means-based Consensus Clustering (KCC). We first formulate the general definition of KCC, and then reveal a necessary and sufficient condition for utility functions that work for KCC, on both complete and incomplete basic partitionings. Experimental results on various real-world data sets demonstrate that KCC is highly efficient and is comparable to the state-of-the-art methods in terms of clustering quality. In addition, KCC shows high robustness to incomplete basic partitionings with substantial missing values.","Title: A Cloud System For Community Extraction From Super-Large Scale Social Networks\nYear: 2013\n\nAbstract: This demo showcase the Community Extraction Cloud (CEC) system. The key idea is to drop weak-tie nodes by efficiently extracting core nodes based on the novel concept of asymptotically equivalent structures (AES) and parallel AES mining algorithm. Meanwhile, to facilitate storing and processing of massive networks, several cloud computing technologies including HDFS, Katta, and Hama are seamlessly integrated into CEC system.","Title: A novel 3d correspondence-less method for MRI and paxinos-watson atlas of rat brain registration\nYear: 2006\n\nAbstract: In this paper, a novel three-dimensional registration method for Magnetic Resonance Image (MRI) and Paxinos-Watson Atlas of rat brain involving 3D affine transformations is proposed. For the purpose of adapting to a large range of transform parameters in a high registration accuracy, the registration procedure includes two steps: a principle components analysis (PCA) based coarse registration and a Hausdorff distance based fine registration. Both steps are free from the correspondence of the feature points of MRI and Atlas. We implemented this registration method in a rat brain 3D reconstruction and analysis system. Experiments have demonstrated that this two-step method can be successfully applied to registering the low resolution and noise affection MRI with Paxinos-Watson Atlas of rat brain.","Title: The Design and Research of Video Grid of Scalable Multilevel Architecture\nYear: 2009\n\nAbstract: Based on the video grid of two level architecture, this paper proposes to use some client computers with high performance as temporary video servers so as to relieve the payload of the local video server, and to expand the original video grid of two level architecture into that of scalable multilevel architecture as the data stream of a local video server reaches the threshold. A detailed discussion is made about the design and process pattern of this expanded video grid and the strategies of service transfer from one server to another when a temporary video server fails. The simulation results produced by VGSim suggest that the service capacity of the video grid of scalable multilevel architecture based on data stream control has been efficiently improved than that of the video grid of two-level architecture.","Title: Personalized Recommendation On Multi-Layer Context Graph\nYear: 2013\n\nAbstract: Recommender systems have been successfully dealing with the problem of information overload. A considerable amount of research has been conducted on recommender systems, but most existing approaches only focus on user and item dimensions and neglect any additional contextual information, such as time and location. In this paper, we propose a Multi-Layer Context Graph (MLCG) model which incorporates a variety of contextual information into a recommendation process and models the interactions between users and items for better recommendation. Moreover, we provide a new ranking algorithm based on Personalized PageRank for recommendation in MLCG, which captures users' preferences and current situations. The experiments on two real-world datasets demonstrate the effectiveness of our approach.","Title: Detecting global outliers from large distributed databases.\nYear: 2012\n\nAbstract: In this paper, we present an innovative system, coined as DISTROD (a.k.a DISTRibuted Outlier Detector), for detecting outliers from distributed databases. DISTROD is able to effectively detect the so-called global outliers from distributed databases that are consistent with those produced by the centralized detection paradigm. Experimental evaluation demonstrates the good performance of DISTROD in terms of effectiveness and speed. © 2012 IEEE.","Title: Research System and Flow of Dynamic Alliance Based on Grid\nYear: 2007\n\nAbstract: Along with enterprise alliance that strides the district, strides the trade and strides nation recombining, there needs a public technical service platform urgently in order to ensure the alliance resources shared effectively. It also should response the changes of market fast. The emergence of grid technology provides a good way to solve this kind of problems. This paper set forth procreant reason, characteristic, system structure and work flow of dynamic alliance on grid in order to realize share of computing resource, storage resource, dada resource, information resource, software resource, network resource, communication resource, knowledge resource, expert resource, and to eliminate information independence and resource independence, then realize high capability structure sharing and cooperating in network virtual environment.","Title: Multi-Feature Fusion Face Recognition Based On Kernel Discriminate Local Preserve Projection Algorithm Under Smart Environment\nYear: 2012\n\nAbstract: In this paper, a new face recognition method based on kernel discriminate local preserve projection(KDLPP) and Multi-feature fusion under smart environment is proposed. In order to solve the small sample size problem, combined with kernel theory and QR decomposition, a new face recognition algorithm named kernel discriminate local preserve projection is proposed based on discriminate local preserve projection algorithm. considered the external features are useful in face recognition, because hair is a highly variable feature of human face, so we combined hair features and DCT features on the feature layer. The experiments on the AMI database indicate the proposed method can enhance the accuracy of the recognition system effectively.","Title: Data cloud for distributed data mining via pipelined mapreduce\nYear: 2011\n\nAbstract: Distributed data mining (DDM) which often utilizes autonomous agents is a process to extract globally interesting associations, classifiers, clusters, and other patterns from distributed data. As datasets double in size every year, moving the data repeatedly to distant CPUs brings about high communication cost. In this paper, data cloud is utilized to implement DDM in order to move the data rather than moving computation. MapReduce is a popular programming model for implementing data-centric distributed computing. Firstly, a kind of cloud system architecture for DDM is proposed. Secondly, a modified MapReduce framework called pipelined MapReduce is presented. We select Apriori as a case study and discuss its implementation within MapReduce framework. Several experiments are conducted at last. Experimental results show that with moderate number of map tasks, the execution time of DDM algorithms (i.e., Apriori) can be reduced remarkably. Performance comparison between traditional and our pipelined MapReduce has shown that the map task and reduce task in our pipelined MapReduce can run in a parallel manner, and our pipelined MapReduce greatly decreases the execution time of DDM algorithm. Data cloud is suitable for a multitude of DDM algorithms and can provide significant speedups.","Title: The Application of P2P in Video Grid\nYear: 2009\n\nAbstract: The POP servers usually have overloaded burden in the current Video Grid. This paper puts forward a technique to import P2P mechanism into Video Grid to alleviate the burden of POP servers. PGVOD is proposed under this technique. It can send video streams buffered in client computers to other clients so as to alleviate the burden of POP servers, shorten the transmission path of video streams in network, to improve the scalability of the Video Grid. The design schema of the framework and the operation mode of nodes in PGVOD system, the design and the process of management of dynamic update of the LCN are described in detail. Finally, PGVOD is compared with a VOD system--Video Grid of C/S structure. The result shows that PGVOD can save bandwidth of network; improve the throughput and performance of the system efficiently.","Title: Discovering Semantics from Multiple Correlated Time Series Stream.\nYear: 2013\n\nAbstract: In this paper, we study a challenging problem of mining data generating rules and state transforming rules (i.e., semantics) underneath multiple correlated time series streams. A novel Correlation field-based Semantics Learning Framework (CfSLF) is proposed to learn the semantic. In the framework, we use Hidden Markov Random Field (HMRF) method to model relationship between latent states and observations in multiple correlated time series to learn data generating rules. The transforming rules are learned from corresponding latent state sequence of multiple time series based on Markov chain character. The reusable semantics learned by CfSLF can be fed into various analysis tools, such as prediction or anomaly detection. Moreover, we present two algorithms based on the semantics, which can later be applied to next-n step prediction and anomaly detection. Experiments on real world data sets demonstrate the efficiency and effectiveness of the proposed method. © Springer-Verlag 2013.","Title: A hybrid model using analytic network process and gray relational analysis for bank's IT outsourcing vendor selection.\nYear: 2012\n\nAbstract: Purpose - Considering the limitation of the single stage vendor selection model, this paper proposes a two-stage vendor selection framework for IT outsourcing in microfinance banks. Design/methodology/approach - The paper attempts to realize a complete analysis at company level using grey systems theory for shaping the relations among variables. With the social choice function Dodgson function, the first stage is a trial phase that helps the decision-maker find the potential vendors, then, the decision-maker employs those chosen vendors for the final selection with modified grey relational analysis (GRA) integrated analytic network process (ANP), which emphasizes the interrelation among those selection criteria, and avoids the subjective estimation of experts and practitioners. The case of a microfinance bank IT outsourcing vendor selection is used to verify the proposed approach. Findings - The results of the empirical study show that the proposed method is practical for ranking competing vendors in terms of their overall performance with respect to multiple interdependence criteria for the bank's IT outsourcing. Practical implications - The method exposed in the paper can be used for other supplier selection by modifying some criteria and weight. Originality/value - The paper provides a method for ranking competing vendors in terms of their overall performance with respect to multiple interdependence criteria for IT outsourcing.","Title: Keyword extraction based on sequential pattern mining\nYear: 2011\n\nAbstract: Keyword extraction is to automatically extract keywords that capture the main topic discussed in a given document. In this paper, a new keyword extraction algorithm based on sequential patterns is proposed. By preprocessing, a document is represented as sequences of words where a sequential pattern mining algorithm is applied on, and important sequential patterns are mined that reflect the semantic relatedness between words. Both statistical features and pattern features within words are used to build the keyword extraction model. The algorithm is independent of languages and does not need the help of a semantic dictionary to get the semantic features. Experimental results on Chinese journal articles show that the proposed algorithm always outperforms the baseline method KEA.","Title: A novel neural network approach to cDNA microarray image segmentation.\nYear: 2013\n\nAbstract: Microarray technology has become a great source of information for biologists to understand the workings of DNA which is one of the most complex codes in nature. Microarray images typically contain several thousands of small spots, each of which represents a different gene in the experiment. One of the key steps in extracting information from a microarray image is the segmentation whose aim is to identify which pixels within an image represent which gene. This task is greatly complicated by noise within the image and a wide degree of variation in the values of the pixels belonging to a typical spot. In the past there have been many methods proposed for the segmentation of microarray image. In this paper, a new method utilizing a series of artificial neural networks, which are based on multi-layer perceptron (MLP) and Kohonen networks, is proposed. The proposed method is applied to a set of real-world cDNA images. Quantitative comparisons between the proposed method and commercial software GenePix(®) are carried out in terms of the peak signal-to-noise ratio (PSNR). This method is shown to not only deliver results comparable and even superior to existing techniques but also have a faster run time.","Title: Online mining abnormal period patterns from multiple medical sensor data streams\nYear: 2014\n\nAbstract: With the advanced technology of medical devices and sensors, an abundance of medical data streams are available. However, data analysis techniques are very limited, especially for processing massive multiple physiological streams that may only be understood by medical experts. The state-of-the-art techniques only allow multiple medical devices to independently monitor different physiological parameters for the patient's status, thus they signal too many false alarms, creating unnecessary noise, especially in the Intensive Care Unit (ICU). An effective solution which has been recently studied is to integrate information from multiple physiologic parameters to reduce alarms. But it is a challenge to detect abnormalities from high frequently changed physiological streams data, since abnormalities occur gradually due to the complex situation of patients. An analysis of ICU physiological data streams shows that many vital physiological parameters are changed periodically (such as heart rate, arterial pressure, and respiratory impedance) and thus abnormalities are generally abnormal period patterns. In this paper, we develop a Mining Abnormal Period Patterns from Multiple Physiological Streams (MAPPMPS) method to detect and rank abnormalities in medical sensor streams. The efficiency and effectiveness of the MAPPMPS method is demonstrated by a real-world massive database of multiple physiological streams sampled in ICU, comprising 250 patients' streams (each stream involving over 1.3 million data points) with a total size of 28 GB data.","Title: Towards a Tricksy Group Shilling Attack Model against Recommender Systems.\nYear: 2012\n\nAbstract: The robustness of recommender systems has drawn recently more and more attention of both industry and academia. Although a multitude of studies have been devoted to shilling attack modeling and detection, few of them focus on group shilling attack. The attackers in a shilling group work together to manipulate the output of the recommender system. Meanwhile, since the rating profiles in a shilling group are carefully designed, it is hard to detect them by traditional methods. This paper presents a generative model to create shilling group in which every pair of attackers has high diversity. In particular, both strict and loose versions of group shilling attack generation algorithm are proposed. Experimental results on MovieLens data set demonstrate that the shilling group generated by the our model can not only exert large negative effect to recommender systems, but also avoid the detection by the traditional methods. © Springer-Verlag 2012.","Title: An Improved Protocol for Deadlock and Livelock Avoidance Resource Co-allocation in Network Computing\nYear: 2010\n\nAbstract: A multitude of applications require simultaneous access to multiple kinds of resources scatted in distributed sites. This problem is known as resource co-allocation which has evolved as a hot topic in network computing. How to design a kind of high-performance protocol for deadlock and livelock avoidance resource co-allocation becomes a challenging problem. In this paper, we propose a new protocol OODP3 (Optimal ODP3) which is based on the currently popular protocol ODP3 (Order-based Deadlock Prevention Protocol with Parallel requests).OODP3 not only inherits the advantage of ODP3 but also guarantees the fulfillment of resource co-allocation within polynomial time. Theoretical proof is conducted to verify the correctness of OODP3. Experimental results also show that OODP3 achieves the better performance improvements than the existing deadlock and livelock avoidance protocol.","Title: Community Detection In Multi-Relational Social Networks\nYear: 2013\n\nAbstract: Multi-relational networks are ubiquitous in many fields such as bibliography, twitter, and healthcare. There have been many studies in the literature targeting at discovering communities from social networks. However, most of them have focused on single-relational networks. A hint of methods detected communities from multi-relational networks by converting them to single-relational networks first. Nevertheless, they commonly assumed different relations were independent from each other, which is obviously unreal to real-life cases. In this paper, we attempt to address this challenge by introducing a novel co-ranking framework, named MutuRank. It makes full use of the mutual influence between relations and actors to transform the multi-relational network to the single-relational network. We then present GMM-NK (Gaussian Mixture Model with Neighbor Knowledge) based on local consistency principle to enhance the performance of spectral clustering process in discovering overlapping communities. Experimental results on both synthetic and real-world data demonstrate the effectiveness of the proposed method.","Title: A loan default discrimination model using cost-sensitive support vector machine improved by PSO\nYear: 2013\n\nAbstract: This study proposes a novel PSO---CS-SVM model that hybridizes the particle swarm optimization (PSO) and cost sensitive support vector machine (CS-SVM) to deal with the problem of unbalanced data classification and asymmetry misclassification cost in loan default discrimination problem. Cost sensitive learning is applied to the standard SVM by integrating misclassification cost of each sample into standard SVM and PSO is employed for parameter determination of the CS-SVM. Meantime, the financial data are discretized by using the self-organizing mapping neural network. And the evaluation indices are reduced without information loss by genetic algorithm for decreasing the complexity of the model. The effectiveness of integrated model of CS-SVM and PSO is verified by three experiments comparing with traditional CS-SVM, PSO---SVM, SVM and BP neural network through real loan default data of companies in China. The corresponding results indicate that the accuracy rate, hit rate, covering rate and lift coefficient are improved dramatically by the developed approach. The proposed method can control the different types of errors distribution with various cost of misclassification accurately, reduce the total misclassification cost largely, and distinguish the loan default problems effectively.","Title: Pick-Up Tree Based Route Recommendation from Taxi Trajectories.\nYear: 2012\n\nAbstract: Recommending suitable routes to taxi drivers for picking up passengers is helpful to raise their incomes and reduce the gasoline consumption. In this paper, a pick-up tree based route recommender system is proposed to minimize the traveling distance without carrying passengers for a given taxis set. Firstly, we apply clustering approach to the GPS trajectory data of a large number of taxis that indicates state variance from \"free\" to \"occupied\", and take the centroids as potential pick-up points. Secondly, we propose a heuristic based on skyline computation to construct a pick-up tree in which current position is its root node that connects all centroids. Then, we present a probability model to estimate gasoline consumption of every route. By adopting the estimated gasoline consumption as the weight of every route, the weighted Round-Robin recommendation method for the set of taxis is proposed. Our experimental results on real-world taxi trajectories data set have shown that the proposed recommendation method effectively reduce the driving distance before carrying passengers, especially when the number of cabs becomes large. Meanwhile, the time-cost of our method is also lower than the existing methods. © 2012 Springer-Verlag.","Title: Localization and Labeling of Rat Brain in MR Image Based on Paxinos-Watson Atlas\nYear: 2006\n\nAbstract: Localization and labeling of function regions in brain is an important topic in experimental brain sciences because the huge amount of data collected by neuroscientists will become meaningless if we cannot give them a precise description of their locations. In this paper, we proposed a localization and labelling method of 3D MR image of rat brain based on Paxinos-Watson atlas. Our objective is to use the specific atlas to accomplish localization and labeling of specified tissue of interest (TOI) to mimic a veteran expert such that invisible or unclear anatomic function regions in the MR images of rat brain can be automatically identified and marked. We proposed a multi-step method to locate and label the TOIs from the MR image of rat brain. Firstly, preprocessing. It alms at the digitization and 3D reconstruction of the atlas and MRI of rat brain. Secondly, two-step registration. The global registration is to eliminate the big misalign and section angle offset as well as the scale between the MRI and atlas. We can choose some unambiguous and characteristic points manually, and based on these correspondences a coarse registration is obtained using affine model. The local registration is to address individual variability of rat brain that can be performed by using Snake model. Thirdly, post-processing. The goal is to locate and label the TOIs in the selected MR image of rat brain slice guided by well-registered atlas. The experiments demonstrated the feasibility of our method.","Title: A novel mutation differential evolution for global optimization.\nYear: 2015\n\nAbstract: Differential evolution DE is a simple and powerful population-based evolutionary algorithm. The success of DE in solving a specific problem crucially depends on appropriately choosing generation strategies and control parameter values. A novel mutation DE MDE is proposed to improve generation strategy DE/best/2. Adaptive mutation is conducted to current population when the population clusters around local optima. Control parameters are adapted based on constants. The performance of MDE is extensively evaluated on eighteen benchmark functions and compares favorably with the several DE variants.","Title: K-Means-Based Consensus Clustering: A Unified View\nYear: 2015\n\nAbstract: The objective of consensus clustering is to find a single partitioning which agrees as much as possible with existing basic partitionings. Consensus clustering emerges as a promising solution to find cluster structures from heterogeneous data. As an efficient approach for consensus clustering, the K-means based method has garnered attention in the literature, however the existing research efforts are still preliminary and fragmented. To that end, in this paper, we provide a systematic study of K-means-based consensus clustering (KCC). Specifically, we first reveal a necessary and sufficient condition for utility functions which work for KCC. This helps to establish a unified framework for KCC on both complete and incomplete data sets. Also, we investigate some important factors, such as the quality and diversity of basic partitionings, which may affect the performances of KCC. Experimental results on various realworld data sets demonstrate that KCC is highly efficient and is comparable to the state-of-the-art methods in terms of clustering quality. In addition, KCC shows high robustness to incomplete basic partitionings with many missing values.","Title: cDNA microarray adaptive segmentation\nYear: 2014\n\nAbstract: DNA microarray technology has enabled biologists to study all the genes within an entire organism to obtain a global view of genes@? interaction and regulation. This technology has a great potential in obtaining a deep understanding of the functional organization of cells. This paper is concerned with improving the processes involved in the analysis of microarray image data. The main focus is to clarify an image@?s feature space in an unsupervised manner. Rather than using the raw microarray image, it proposes to produce filtered versions of the image data by applying nonlinear anisotropic diffusion so that the dynamic range of the image could be increased and, therefore, a better ability of signal extraction could be achieved. In this paper, a novel segmentation algorithm is developed. This algorithm is based on the Cellular Neural Network computational paradigm integrated with median and anisotropic diffusion filters. The AnaLogic CNN Simulation Toolbox for MATLAB (InstantVision Toolboxes for MATLAB) is used during the segmentation process. Quantitative comparisons among the proposed methods and GenePix^(R) are carried out from both the objective and subjective points of view. It is shown that analogic algorithm integrated with Complex Diffusion filter is the best one to be applied to accomplish the segmentation.","Title: Polygonal approximation using integer particle swarm optimization.\nYear: 2014\n\nAbstract: Polygonal approximation is an effective yet challenging digital curve representation for image analysis, pattern recognition and computer vision. This paper proposes a novel approach, integer particle swarm optimization (iPSO), for polygonal approximation. When compared to the traditional binary version of particle swarm optimization (bPSO), the new iPSO directly uses an integer vector to represent the candidate solution and provides a more efficient and convenient means for solution processing. The velocity and position updating mechanisms in iPSO not only have clear physical meaning, but also guarantee the optimality of the solutions. The method is suitable for polygonal approximation which could otherwise be an intractable optimization problem. The proposed method has been tested on commonly used synthesized shapes and lake contours extracted from the maps of four famous lakes in the world. The experimental results show that the proposed iPSO has better solution quality and computational efficiency than the bPSO-based methods and better solution quality than the other state-of-the-art methods.","Title: State Estimation For Complex Systems With Randomly Occurring Nonlinearities And Randomly Missing Measurements\nYear: 2014\n\nAbstract: This paper is concerned with the state estimation problem for the complex networked systems with randomly occurring nonlinearities and randomly missing measurements. The nonlinearities are included to describe the phenomena of nonlinear disturbances which exist in the network and may occur in a probabilistic way. Considering the fact that probabilistic data missing may occur in the process of information transmission, we introduce the randomly data missing into the sensor measurements. The aim of this paper is to design a state estimator to estimate the true states of the considered complex network through the available output measurements. By using a Lyapunov functional and some stochastic analysis techniques, sufficient criteria are obtained in the form of linear matrix inequalities under which the estimation error dynamics is globally asymptotically stable in the mean square. Furthermore, the state estimator gain is also obtained. Finally, a numerical example is employed to illustrate the effectiveness of the proposed state estimation conditions.","Title: Detecting Genuine Communities from Large-Scale Social Networks: A Pattern-Based Method.\nYear: 2014\n\nAbstract: Community detection is a long-standing yet very difficult task in social network analysis. It becomes more challenging as many online social networking sites are evolving into super-large scales. Numerous methods have been proposed for community detection from massive networks, but how to reconcile the partitioning efficiency and the community quality remains an open problem. In this paper, we att...","Title: A framework for component selection in collaborative sensing application development\nYear: 2014\n\nAbstract: Wireless sensor network-based technologies and applications have attracted a lot of attention in the past two decades because of their huge potential to change people's way of life. These applications usually need close collaboration among multiple sensors, gateways, services and end users. When developing these applications, system designers and practitioners usually face several performance requirements such as the accuracy, battery life and system reliability. Given the hard requirements in system performance, how to choose an optimal combination from various sensors, algorithms and collaborative systems to form the application is the most important problem that practitioners need to address. Ad hoc solutions were proposed in specific applications in the past; however, a general methodology that can be easily applied to future applications is lacking. In this paper, we take the challenge and propose a general framework aiming to address the component selection problem, illustrate how this framework can be applied to real life applications through a case study, and discuss challenging issues and two interesting finds from our implementation.","Title: Local Community Extraction for Non-overlapping and Overlapping Community Detection.\nYear: 2014\n\nAbstract: The scale of current networked system is becoming increasingly large, which exerts significant challenges to acquire the knowledge of the entire graph structure, and most global community detection methods often suffer from the computational inefficiency. Local community detection aims at finding a community structure starting from a seed vertex without global information. In this article, we propose a Local Community Extraction algorithm (LCE) to find the local community from a seed vertex. First, a local search model is carefully designed to determine candidate vertices to be preserved or discarded, which only relies on the local/incomplete knowledge rather than the global view of the network. Second, we expand LCE for the global non-overlapping community detection, in which the labels of detected local communities are seen as vertices' attributive tags. Finally, we adopt the results of LCE to calculate a membership matrix, which can been used to detect the global overlapping community of a graph. Experimental results on four real-life networks demonstrate the advantage of LCE over the existing degree-based and similarity-based local community detection methods by either effectiveness or efficiency validity.","Title: Research on the Forecasting Model for Hypertension Drug Efficacy Based on Medical Data\nYear: 2014\n\nAbstract: Hypertension is a common disease endangering human health, and every year the number of cases increases at a speed of ten millions. So far there are about 230 million hypertension patients in China, and the long-term drug treatment for hypertension patients is an effective measure to control blood pressure. In this paper, based on the evaluation period of patients with Medication Possession Ratio (MPR) and blood pressure analysis, the beta distribution model which is used to find the relationship between them is established. Then genetic algorithms (GA) and cross validation are carried out on the model and we make comparative analysis with linear distribution model. The experimental results show that the beta distribution model can well forecast the drug efficacy of patients. Only through long-term drug therapy, can the blood pressure be effectively controlled.","Title: Overlapping Community Extraction: A Link Hypergraph Partitioning Based Method\nYear: 2014\n\nAbstract: Real-world networks often contain communities with pervasive overlaps such that nodes simultaneously belong to several groups. Community extraction, emerging in recent years, is considered to be a promising solution for finding meaningful communities from social networks. In this paper, we explore overlapping community extraction from a link partitioning perspective. First, we define the local link structure composed of a set of closely interrelated links, by extending the similarity of link-pairs to that of a group of links. Second, based upon our prior work, we transform the problem of mining local link structures into a pattern mining problem, and thus present an efficient mining algorithm. Third, we propose to use the hypergraph to assemble all local link structures, and employ hMETIS for hypergraph partitioning. Finally, based on extracted link communities, we restore the membership of nodes in the original graph owing to its links. Experimental results on various real-life social networks validate the effectiveness of the proposed method.","Title: A backbone extraction method with Local Search for complex weighted networks\nYear: 2014\n\nAbstract: The backbone is the natural abstraction of a complex network, which can help people to understand it in a more simplified form. Backbone extraction becomes more challenging as many networks are evolving into large scale and the weight distributions are spanning several orders of magnitude. Traditional filter-based methods tend to include many outliers into the backbone. What is more, they often suffer from the computational inefficiency-the exhaustive search of all nodes or edges is often prohibitively expensive. In this work, we propose a Local Search based Backbone Extraction Heuristic (LS-BEH) to find the backbone in a complex weighted network. First, a strict filtering rule is carefully designed to determine edges to be preserved or discarded. Second, we present a local search model to examine part of edges in an iterative way. Experimental results on two real-life networks demonstrate the advantage of LS-BEH over the classic disparity filter method by either effectiveness or efficiency validity.","Title: Collaborative Topic Ranking: Leveraging Item Meta-Data For Sparsity Reduction\nYear: 2015\n\nAbstract: Pair-wise ranking methods have been widely used in recommender systems to deal with implicit feedback. They attempt to discriminate between a handful of observed items and the large set of unobserved items. In these approaches, however, user preferences and item characteristics cannot be estimated reliably due to overfitting given highly sparse data. To alleviate this problem, in this paper, we propose a novel hierarchical Bayesian framework which incorporates \"bag-of-words\" type meta-data on items into pair-wise ranking models for one-class collaborative filtering. The main idea of our method lies in extending the pair-wise ranking with a probabilistic topic modeling. Instead of regularizing item factors through a zero-mean Gaussian prior, our method introduces item-specific topic proportions as priors for item factors. As a by-product, interpretable latent factors for users and items may help explain recommendations in some applications. We conduct an experimental study on a real and publicly available dataset, and the results show that our algorithm is effective in providing accurate recommendation and interpreting user factors and item factors.","Title: Simulation of the recharging method of implantable biosensors based on a wearable incoherent light source.\nYear: 2014\n\nAbstract: Recharging implantable electronics from the outside of the human body is very important for applications such as implantable biosensors and other implantable electronics. In this paper, a recharging method for implantable biosensors based on a wearable incoherent light source has been proposed and simulated. Firstly, we develop a model of the incoherent light source and a multi-layer model of skin tissue. Secondly, the recharging processes of the proposed method have been simulated and tested experimentally, whereby some important conclusions have been reached. Our results indicate that the proposed method will offer a convenient, safe and low-cost recharging method for implantable biosensors, which should promote the application of implantable electronics.","Title: An efficient and scalable density-based clustering algorithm for datasets with complex structures.\nYear: 2016\n\nAbstract: As a research branch of data mining, clustering, as an unsupervised learning scheme, focuses on assigning objects in the dataset into several groups, called clusters, without any prior knowledge. Density-Based Spatial Clustering of Applications with Noise (DBSCAN) is one of the most widely used clustering algorithms for spatial datasets, which can detect any shapes of clusters and can automatically identify noise points. However, there are several troublesome limitations of DBSCAN: (1) the performance of the algorithm depends on two specified parameters, ε and MinPts in which ε represents the maximum radius of a neighborhood from the observing point and MinPts means the minimum number of data points contained in such a neighborhood. (2) The time consumption for searching the nearest neighbors of each object is intolerable in the cluster expansion. (3) Selecting different starting points results in quite different consequences. (4) DBSCAN is unable to identify adjacent clusters of various densities. In addition to these restrictions about DBSCAN mentioned above, the identification of border points is often ignored. In our paper, we successfully solve the above problems. Firstly, we improve the traditional locality sensitive hashing method to implement fast query of nearest neighbors. Secondly, several definitions are redefined on the basis of the influence space of each object, which takes the nearest neighbors and the reverse nearest neighbors into account. The influence space is proved to be sensitive to local density changes to successfully reduce the amount of parameters and identify adjacent clusters of different densities. Moreover, this new relationship based on the influence space makes the insensitivity to the ordering of inputting points possible. Finally, a new concept—core density reachable based on the influence space is put forward which aims to distinguish between border objects and noisy objects. Several experiments are performed which demonstrate that the performance of our proposed algorithm is better than the traditional DBSCAN algorithm and the improved algorithm IS-DBSCAN.","Title: Adaptive event-triggered H∞ filtering for T-S fuzzy system with time delay.\nYear: 2016\n\nAbstract: This paper is concerned with adaptive event-triggered H∞ filter design for a class of T-S fuzzy systems with time delay. Firstly, an adaptive event-triggered scheme is introduced, which can adaptively adjust the communication threshold to save the limited communication resource. Secondly, a T-S fuzzy model is applied to approximate the nonlinear dynamics of the plant. By using Lyapunov function, sufficient conditions for the existence of the desired filter are established in terms of linear matrix inequalities such that the filtering error dynamics is locally mean square asymptotically stable. Then, the explicit expression is provided for the designed filter parameters. Finally, a simulation example is employed to illustrate the design method.","Title: Discovering shilling groups in a real e-commerce platform.\nYear: 2016\n\nAbstract: Purpose - With the popularity of e-commerce, shilling attack is becoming more rampant in online shopping websites. Shilling attackers publish mendacious ratings as well as reviews for promoting or suppressing target products. The purpose of this paper is to investigate group shilling, a new typed shilling attack, behavior in a real e-commerce platform (e.g. Amazon. cn). Design/methodology/approach - Several behavioral features are proposed for modeling the shilling group, and thus an unsupervised ranking method based on principal component analysis (PCA) is presented for identifying shilling groups from real users on Amazon. cn. Findings - As indicated by the behavior analysis, the proposed method has successfully identified a number of shilling groups on Amazon. Meanwhile, the effectiveness of the proposed features and accuracy of the proposed unsupervised method are carefully validated. Originality/value - This paper presents a set of solutions for discovering shilling groups when the ground truth labels are hard to be obtained in real environment, including candidate groups generation, behavioral features definition and unsupervised detection.","Title: The consistency measures of multi-granularity linguistic group decision making.\nYear: 2015\n\nAbstract: When different experts have different uncertainty degrees on the phenomenon, several linguistic term sets with different granularity of uncertainty are necessary. In this paper, we deal with the consistency problem with multi-granularity linguistic term sets applied to group decision making. Firstly, we develop a transformation model to maintain the uncertainty degrees of different granularity associated with each expert. We use the uncertain linguistic variables as the unified form. Some computational rules about uncertain linguistic variables are given. Then consistency index of linguistic preference relations based on the distance of a multi-granularity linguistic preference relation is defined. Chi-square statistic is used to establish the consistency thresholds. Finally two numerical examples are used to show the process and effects of our new proposed method.","Title: Detecting overlapping communities in poly-relational networks\nYear: 2015\n\nAbstract: Discovering communities can promote the understanding of the structure, function and evolution in various systems. Overlapping community detection in poly-relational networks has gained much more interests in recent years, due to the fact that poly-relational networks and communities with pervasive overlap are prevalent in the real world. A plethora of methods detect communities from the poly-relational network by converting it to mono-relational networks first. Nevertheless, they commonly assume different relations are independent from each other, which is obviously unreal to real-life cases. In this paper, we attempt to relax this strong assumption by introducing a novel co-ranking framework, named MutuRank. It makes full use of the mutual influence between relations and actors to transform the poly-relational network to the mono-relational network. We then present a novel GMM-NK (Gaussian Mixture Model with Neighbor Knowledge) algorithm incorporating the impact from neighbors into the traditional GMM. Experimental results both on synthetic networks and the real-world network have verified the effectiveness of MutuRank and GMM-NK.","Title: An Ensemble Differential Evolution For Numerical Optimization\nYear: 2015\n\nAbstract: The success of differential evolution (DE) in solving a specific problem crucially depends on appropriately choosing generation strategies and control parameter values. The mutation strategies of DE are classified into two groups: DE/rand/k without best solution and DE/best/k with best solution. The proposed algorithm utilizes two mutation strategies. The first one is from DE/rand/k and the second one is from DE/best/k. The proposed algorithm uses two control parameter settings. It randomly combines them to generate trial vectors. The novel mechanism improves the convergence rate of DE and maintains diversity of the population. The performance of the proposed algorithm is extensively evaluated on all the CEC2005 test functions and compares favorably with the several DE variants.","Title: AAC-OT: Accountable Oblivious Transfer With Access Control\nYear: 2015\n\nAbstract: To prevent illegal users accessing the database and protect users' privacy, oblivious transfer with access control (AC-OT) was proposed. In an AC-OT scheme, the database provider can encrypt the records and publish corresponding access control lists (ACLs). Prior to accessing the records, a user needs to obtain anonymous credentials from the issuer. Subsequently, an authorized user can obtain the intended records without the database provider knowing its choices. Although AC-OT schemes have shown a lot of merits, there are some practical issues: 1) one of the inherited problems in anonymous credentials is timely revocation and 2) how to prevent malicious users overusing the records. In this paper, we propose an accountable AC-OT scheme to address these issues. In our scheme, an authorized user can access the protected records without the database provider knowing his personal information and choices if: 1) he has obtained the required credentials listed in the ACLs and 2) the number of the access times for each record is no more than the specified bound. Notably, the database provider can trace and revoke the user who overused the records even in the lifetime of his credentials. To the best of our knowledge, it is the first AC-OT scheme where timely revocation and overuse detection are considered.","Title: A Hybrid Evolutionary Algorithm For Numerical Optimization Problem\nYear: 2015\n\nAbstract: The hybrid artificial bee colony (ABC) algorithm with differential evolution (DE) techniques (HABCwDE) is proposed for numerical optimization in this paper. The HABCwDE adopts multiple candidate solution generation strategies (CSGSes) from DE techniques to generate new solutions in the framework of the ABC algorithm. In the HABCwDE algorithm, three CSGSes and three groups of parameter settings are employed. The performance of HABCwDE and some other evolutionary algorithms are tested on 26 state-of-the-art benchmark functions. Experimental results demonstrate that HABCwDE is very competitive, and that it is an effective way to improve the performance of ABC algorithm by employing CSGSes from DE techniques.","Title: Multimodel Biometrics Fusion Based On Far And Frr Using Triangular Norm\nYear: 2015\n\nAbstract: Multibiomitric systems are expected to be more accurate due to the presence of multiple evidences, score level fusion is the most commonly used approach in multibiometrics. In this paper, A novel approach is proposed for the fusion at score level fusion based on False Reject Rate(FRR) and False Accept Rate(FRR) using triangular norms(t-norms). This study aims at tapping the potential of t-norms for information fusion at first, at the second, it transfers scores into Transfer function based on corresponding FRRs and FARs, thus avoiding calculating posterior probability of a certain score. Experiment result shown that the proposed method renders very good performance as it is quite computationally and outperforms the traditional score level fusion schemes, the experimental result also confirms the effectiveness of the proposed method to improve the performance of multibiometric system.","Title: 3D Model-Based Food Traceability Information Extraction Framework.\nYear: 2015\n\nAbstract: In this paper, we propose a 3D model-based food traceability information extraction method for processing video surveillance data. The proposed method first builds a 3D model of the surveillance area. Then, the video cameras are mapped in the 3D model and the coordinate transform functions from the 2D camera coordinates to the 3D model coordinates are calculated. Next, the object detection method is applied to identify the target which is then mapped into the 3D coordinates so that its 3D trajectory can be generated. Finally, we merge multiple trajectories from different cameras to create the complete traceability information for the target object. According to the experimental results, the proposed method can efficiently extract useful traceability information for a video surveillance system.","Title: Spammers Detection from Product Reviews: A Hybrid Model\nYear: 2015\n\nAbstract: Driven by profits, spam reviews for product promotion or suppression become increasingly rampant in online shopping platforms. This paper focuses on detecting hidden spam users based on product reviews. In the literature, there have been tremendous studies suggesting diversified methods for spammer detection, but whether these methods can be combined effectively for higher performance remains unclear. Along this line, a hybrid PU-learning-based Spammer Detection (hPSD) model is proposed in this paper. On one hand, hPSD can detect multi-type spammers by injecting or recognizing only a small portion of positive samples, which meets particularly real-world application scenarios. More importantly, hPSD can leverage both user features and user relations to build a spammer classifier via a semi-supervised hybrid learning framework. Experimental results on movie data sets with shilling injection show that hPSD outperforms several state-of-the-art baseline methods. In particular, hPSD shows great potential in detecting hidden spammers as well as their underlying employers from a real-life Amazon data set. These demonstrate the effectiveness and practical value of hPSD for real-life applications.","Title: An improved neural network algorithm based on artificial bee colony algorithm and its application in sewage treatment\nYear: 2015\n\nAbstract: As an advanced artificial intelligence technology, error back-propagation (BP) neural network algorithm has been widely applied to electronics, communications, automation and other fields. However, traditional BP neural network algorithm has the disadvantages, such as inclination to stick into local optima, and slow convergence, which exert a great impact on the processing performance, and also limit its further application. Artificial bee colony (ABC) algorithm has been developed as a new heuristic algorithm in recent years based on foraging behavior, which is also proposed to be a global optimization algorithm. As a simple and reliable algorithm, ABC algorithm is easy to be implemented. In this paper, an optimized BP neural network algorithm based on ABC algorithm is proposed so as to integrate the advantages of both algorithms. Simulation result shows that compared with the conventional BP neural network algorithm, the new algorithm has gained a significant development not only in prediction accuracy but also in convergence speed.","Title: A survival analysis method for stock market prediction\nYear: 2015\n\nAbstract: Stock market prediction focus on developing approaches to determine the future price of a stock or other financial product. The key task of stock market prediction is to determine the timing for the buying or selling of stock, undoubtedly, it is very difficult due to the high volatility and nonlinear relationships driven by short-term fluctuations in investment demand. In this work, we address this problem by adopting the Cox's hazard model to predict a stock's future rising or dropping probabilities. Specifically, we define the problem of Buy-and-Sell-Point Prediction from the survival analysis perspective. The Cox's hazard model is proposed as the model of choice for this prediction problem due to several reasons including the ability to model the dynamics in the stock movement, and to easily incorporate different types of technical indexes as covariates. In the experiment, we apply the trained model for the stock market forecasting on six stocks in Shanghai Stock Exchange. The results show that the proposed model is superior to several baseline models in terms of accuracy, and the stock return evaluations have revealed that the profits produced by the proposed model are higher.","Title: Scaling Information-Theoretic Text Clustering: A Sampling-based Approximate Method\nYear: 2014\n\nAbstract: Info-Kmeans, a K-means clustering method employingKL-divergence as the proximity function, is one ofthe representative methods in information-theoretic clustering.With the explosive growth of online texts such as online reviewsand user-generated content, the text is becoming more sparseand much bigger, which poses significant challenges on botheffectiveness and efficiency issues of text clustering. In our priorwork, we presented a Summation-bAsed Incremental Learning(SAIL) algorithm, which can avoid the zero-feature dilemma ofhighly sparse texts. In this paper, we propose a sampling-basedapproximate approach for scaling SAIL algorithm to deal withthe large-scale of texts. Particularly, an instance-level randomsampling is invoked to reduce the number of instances to beexamined during each iteration, which substantially speedsup the clustering on big text data. Furthermore, we provethat the margin of errors introduced by random samplingcan be controlled in a small range. Extensive experiments oneight real-life text datasets demonstrate the advantage of theproposed sampling-based approximate clustering method. Inparticular, our method shows merits in both effectiveness andefficiency on clustering performance.","Title: Efficient Parallel Community Detection in Large Edge-Intensive Networks\nYear: 2014\n\nAbstract: Community detection is a classic and very difficulttask in social network analysis. A large number of methodshave been developed for both efficient and effective communitydetection. However, much of the existing methods are heavilydependent on the number of links in the network, and thusthey often suffer from the computational inefficiency whenmeeting large edge-intensive networks. In this paper, we presenta novel SIMPLifying and Ensembling (SIMPLE) frameworkfor parallel community detection. It employs the random linksampling to simplify the network and obtain basic partitioningson every sampled graphs. Then, the K-means-based ConsensusClustering is used to ensemble a number of basic partitioningsto get high-quality community structures. Meanwhile, stepsof random sampling and sampled graph partitioning areencapsulated into MapReduce to further improve the efficiency.Experiments on four real-world social networks analyze keyparameters and factors inside SIMPLE, and demonstrate theeffectiveness of the SIMPLE.","Title: A distributed SVM method based on the iterative MapReduce\nYear: 2015\n\nAbstract: Linear classification is useful in many applications, but training large-scale data remains an important research issue. Recent advances in linear classification have shown that distributed methods can be efficient in improving the training time. However, for most of the existing training methods,based on MPI or Hadoop, the communication between nodes is the bottleneck. To shorten the communication between nodes, we propose and analyze a method for distributed support vector machine and implement it on an iterative MapReduce framework. Through our distributed method, the local SVMs are generic and can make use of the state-of-the-art SVM solvers. Unlike previous attempts to parallelize SVMs the algorithm does not make assumptions on the density of the support vectors, i.e., the efficiency of the algorithm holds also for the “difficult” cases where the number of support vectors is very high. The performance of the our method is evaluated in an experimental environment. By partitioning the training dataset into smaller subsets and optimizing the partitioned subsets across a cluster of computers, we reduce the training time significantly while maintaining a high level of accuracy in both binary and multiclass classifications.","Title: Building the Computational Virtual Reality Environment for Anaesthetists' Training and Practice\nYear: 2015\n\nAbstract: Understanding the real world based on visualisation and prediction is essential for the decision-maker. We build a computational virtual reality environment to improve visualisation, understanding and prediction of the physical world and to guide action. It develops a five-dimensional, computer-generated, computational Virtual Reality Environment for Anaesthesia (VREA). Our online prediction will be calculated based on the correlation and composition computing with respect to the three dimensions: horizontal, vertical and individual. The novel musical notes based anesthetic simulator is proposed to identify the abnormality and visualize the online medical time series. The experiments with the online ECG data will present a real-time case to show the effectiveness and efficiency of our proposed system and algorithms.","Title: Statistical characterization of TVWS noise: An experimental study\nYear: 2012\n\nAbstract: This paper presents results of a study about the statistical properties of noise (including unknown interference) in blank channels of TV bands, i.e., TV white spaces. Measurements were conducted to collect data in three different scenarios: sunny-day laboratory, sunny-day curbside and rainy-day laboratory. Impulsive noise has been modeled as a pulse train and the pulse amplitude, pulse duration time and inter-arrival time in both time and frequency domains were defined and considered as typical statistical variables. Several comparisons were made to study the statistical properties of TV white space noise, in two different channels (672~680MHz and 768~776MHz), under three different scenarios, and with several typical distribution models (Rayleigh, Gamma, lognormal and Middleton A). It has been found that noises in different TV white spaces have highly consistent statistical properties, whereas under different scenarios may exhibit diverse statistical properties. A Kolmogorov-Smirnov test was applied to find the closest distributions within the considered candidate distributions. Two new models for the amplitude probability distribution of noise in the frequency domain were proposed and validated, both consisting of a low-amplitude thermal noise component and a high-amplitude impulsive noise component.","Title: Accountable mobile E-commerce scheme via identity-based plaintext-checkable encryption.\nYear: 2016\n\nAbstract: Privacy and accountability are two important issues in mobile e-commerce. In this paper, we formalized the definition and security model of identity-based plaintext-checkable encryption (IBPCE) schemes, and proposed an IBPCE scheme. Furthermore, we designed an accountable mobile e-commerce (AMEC) scheme. This scheme can provide the following features: (1) users can register to the e-commerce system by using their mobile identities, such as mobile phone numbers; (2) the transactions between a buyer and a vendor are encrypted; (3) if there is a dispute, an offline adjudicator can identify who is dishonest by checking the encrypted transactions. Hence, our scheme is an ideal primitive to balance the need for privacy and accountability. In mobile e-commerce systems, users conduct transactions using wireless or Internet-based devices, such as mobile phones and tablets. It is different from traditional e-commerce systems relying on workstations or desktops, which is usually used in a fixed location. Recently, privacy and accountability have become users' primary concerns in mobile e-commerce applications. In this paper, a novel mobile e-commerce scheme is developed to address the fundamental requirements. We first propose an identity-based plaintext-checkable encryption (IBPCE) scheme where anyone can check whether a ciphertext is the encryption of a plaintext under a specific identity without knowing the secret key. Furthermore, the proposed IBPCE scheme is incorporated into the mobile e-commerce scenario, which results in an accountable mobile e-commerce (AMEC) scheme. Our proposed AMEC scheme has several superior features: (1) Users can register to the e-commerce system by using their mobile identities, such as mobile phone numbers; (2) The transactions between a buyer and a vendor are encrypted; (3) If there is a dispute, an offline adjudicator can identify who is dishonest by checking the encrypted transactions. We evaluate the proposed scheme and confirm that the new scheme can effectively balance the need for privacy and accountability.","Title: K-Anonymisation Of Social Network By Vertex And Edge Modification\nYear: 2016\n\nAbstract: With the rapid growth of social networks, a large quantity of social network data is being collected, and published for analysis. Privacy is one of the major issues when publishing social network data. To address the issue, several social network anonymisation approaches have been proposed. However, these methods usually introduce a large amount of distortion to the original social network graph, thus, the utility of the social network is not preserved very well. In this paper, we present a two-phase approach: in the first phase, we get the target degree of each vertex; the community structure of the social network and path length between vertices are considered when finding candidates to increase the degree of vertex to its target degree in the second phase. Experimental results on real world datasets show that the utility of the social network is well preserved with our approach.","Title: Game theory based emotional evolution analysis for chinese online reviews.\nYear: 2016\n\nAbstract: Sentiment analysis has become one of the mainstream researches in social network analysis. Its impact can be seen in many practical applications, ranging from public opinion analysis to marketing of public praise and information prediction. However, most of the existing research has been performed in the sentiment classification for subjective text, the emotional evolution analysis for complex interactive text (e.g., online reviews) has not yet been thoroughly targeted by the research community. This paper is concerned on short-text Chinese online reviews collected from Tianya forum. First, an efficient affective computing framework is proposed to capture the underlying emotions of Chinese online reviews. It can accurately calculate the semantic orientation of the entire review, without requiring expensive manual labeling of seed words. As users' attitudes might influence with each other, predicting their future emotional behaviors that only relying on the emotional values of historical reviews is very one-sided. Therefore, we propose a game theory based emotional evolution prediction algorithm combining the affective computing, in which the mixed nash equilibrium strategies are calculated as the future emotional behavior of interactive users. Then, experimental results on the large-scaled review dataset are provided to demonstrate the effectiveness and accurateness of our approaches. Finally, by applying the research results on the pairwise happiness-popularity coordination evaluation, we have revealed some interesting phenomenon on the \\\"World View\\\" board in Tianya forum.","Title: LED: A fast overlapping communities detection algorithm based on structural clustering.\nYear: 2016\n\nAbstract: Abstract Community detection in social networks is a fundamental task of complex network analysis. Community is usually regarded as a functional unit. Networks in real world more or less have overlapping community structure while traditional community detection algorithms assume that one vertex can only belong to one community. This paper proposes an efficient overlapping community detection algorithm named LED ( Loop Edges Delete ). LED algorithm is based on Structural Clustering, which converts structural similarity between vertices to weights of network. The evaluations of the LED algorithm are conducted both from classical networks from literature and C-DBLP, which is a huge and real-life co-author social network in China. The results show that LED is superior to other methods in accuracy, efficiency, comparing with FastModurity and GN algorithm.","Title: Edge Computing: Vision and Challenges.\nYear: 2016\n\nAbstract: The proliferation of Internet of Things (IoT) and the success of rich cloud services have pushed the horizon of a new computing paradigm, edge computing, which calls for processing the data at the edge of the network. Edge computing has the potential to address the concerns of response time requirement, battery life constraint, bandwidth cost saving, as well as data safety and privacy. In this pap...","Title: Stability analysis of switched cellular neural networks: A mode-dependent average dwell time approach.\nYear: 2016\n\nAbstract: This paper addresses the exponential stability of switched cellular neural networks by using the mode-dependent average dwell time (MDADT) approach. This method is quite different from the traditional average dwell time (ADT) method in permitting each subsystem to have its own average dwell time. Detailed investigations have been carried out for two cases. One is that all subsystems are stable and the other is that stable subsystems coexist with unstable subsystems. By employing Lyapunov functionals, linear matrix inequalities (LMIs), Jessen-type inequality, Wirtinger-based inequality, reciprocally convex approach, we derived some novel and less conservative conditions on exponential stability of the networks. Comparing to ADT, the proposed MDADT show that the minimal dwell time of each subsystem is smaller and the switched system stabilizes faster. The obtained results extend and improve some existing ones. Moreover, the validness and effectiveness of these results are demonstrated through numerical simulations.","Title: Real-time target detection and recognition with deep convolutional networks for intelligent visual surveillance.\nYear: 2016\n\nAbstract: Moving target detection and tracking, recognition, behaviors analysis are the key issues in the intelligent visual surveillance system (IVSS). The challenge is how to process the real-time video stream in an effective way in case that we could find the interested objects for analysis. However, the traditional video surveillance technology often does not meet the needs of real-time key frame recognition for the on-line intelligent video monitoring system. In our paper, we adopt the state-of-the-art Faster R-CNN [7] that takes advantages of convolutional neural networks into our real-time target recognition system - Deep Intelligent Visual Surveillance (DIVS). The key aspects of our DIVS are consisted of four parts: (i) Getting the real-time video images from remote cameras; (ii) Processing the data with the deep learning framework caffe [23] built for Faster R-CNN; (iii) Storing the valuable data with MySQL; (iv) Data presentation on the website. Experiments based on our system validated the effectiveness, stability and accuracy of our proposed solutions.","Title: Similarity based leaf image retrieval using multiscale R-angle description.\nYear: 2016\n\nAbstract: Leaf image identification is a significant and challenging application of computer vision and image processing. A central issue associated with this task is how to effectively and efficiently describe the leaf images and measure their similarities. In this paper, a novel shape descriptor termed R-angle is proposed. R-angle describes the curvature of the contour by measuring the angle between the intersections of the shape contour with a circle of radius R centered at points sampled around the contour. It is intrinsically invariant to group transforms including scaling, rotation and translation. Varying the parameter R of the proposed R-angle naturally introduces the notation of scale, which we leverage to provide a coarse-to-fine description of the local curvature. A local scale arrangement is proposed by taking the distance between each contour point and the center of the shape to be the maximum scale for a given contour point. Two matching schemes, including L1-norm matching and dynamic programming based matching, are applied to measure the similarities of the leaf shapes. The retrieval experiments conducted on two challenging leaf image datasets indicate that the proposed method significantly outperforms the state-of-the-art methods for leaf identification. An additional experiment on an animal dataset also indicates its potential for general shape recognition.","Title: DGWC: Distributed and generic web crawler for online information extraction\nYear: 2016\n\nAbstract: Online information has become important data source to analyze the public opinion and behavior, which is significant for social management and business decision. Web crawler systems target at automatically download and parse web pages to extract expected online information. However, as the rapid increasing of web pages and the heterogeneous page structures, the performance and the rules of parsing have become two serious challenges to web crawler systems. In this paper, we propose a distributed and generic web crawler system (DGWC), in which spiders are scheduled to parallel access and parse web pages to improve performance, utilized a shared and memory based database. Furthermore, we package the spider program and the dependencies in a container called Docker to make the system easily horizontal scaling. Last but not the least, a statistics-based approach is proposed to extract the main text using supervised-learning classifier instead of parsing the page structures. Experimental results on real-world data validate the efficiency and effectiveness of DGWC.","Title: Parallelizing K-Means-Based Clustering on Spark\nYear: 2016\n\nAbstract: K-means is in fact a family of clustering algorithms with different distance functions and a variety of extension, e.g., fuzzy clustering and consensus clustering. Nevertheless, K-means-based clustering algorithms employ the similar two-phase iterative procedure including distance computation and centroids updating. Therefore, to explore the parallel implementations of this two-phase iterative procedure on Spark is not only universal to a wealth of clustering algorithms but also meets the practical needs addressed by big data. This paper contributes to reveal implementation details for parallelizing K-means-based clustering on Spark. In particular, we first introduce the boundary of so-called K-means-based clustering, and then present the overall parallelizable framework on Spark. We discuss the technical barrier and their alternative strategies for each step. Experimental results on both large-scale UCI datasets and text datasets demonstrate the effectiveness and efficiency of our implementations.","Title: Online Public Opinion System: Design and Applications\nYear: 2016\n\nAbstract: Online Public Opinion Systems (OPOS) target at collecting, analyzing, summarizing and monitoring massive public opinions on the Internet in real time. Meanwhile, OPOS often have the ability to identify the key or sudden events, and thus notify related people immediately for rapid responses to these events. As part of this endeavor, this paper introduces the architecture and techniques of an OPOS that has been used by several large enterprises. This self-designed OPOS generally contains data layer, computation layer and application layer from bottom to up. We first introduce the basic function and key techniques of each layer, and then present several typical yet important algorithms on the computation layer. Experimental results on real-world data validate the effectiveness of algorithms fixed in our system. Last but not the least, a system demonstration in a ship-building company is provided to justify the value of our OPOS for real enterprises.","Title: Self-representation dimensionality reduction for multi-model classification.\nYear: 2017\n\nAbstract: Feature selection remove the noisy/irrelevant samples and select the subset of representative features, in general, from the high-dimensional space of data has been a fatal significant technique in computer vision and machine learning. Afterwards, motivated by the interpretable ability of feature selection patterns, beside, and the successful use of low-rank constraint in static and sparse learning in the field of machine learning. We present a novel feature selection model with unsupervised learning by using low-rank regression on loss function, and a sparsity term plus K-means clustering method on regularization term during this article. In order to distinguish from those existing state-of-the-art attribute selection measures, the propose method have described as follows: (1) represent the every feature by other features (including itself) via utilize the corresponding loss function with a feature-level self-express way; (2) embed K-means to generate pseudo class label information for the attribute selection as an pseudo supervised method, because of the supervised learning usually have the better recognition results than unsupervised learning; (3) also use the low-rank constraint to feature selection which considers two aspects of information inherent in data. The low-rank constraint takes the correlation of response variables into account, while an 2, p-norm regularizer considers the correlation between feature vectors and their corresponding response variables. The extensive relevant results of experiment on three multi-model comparison data demonstrated that our new unsupervised feature selection pattern outperforms the related approaches.","Title: Htcrl: A Range-Free Location Algorithm Based On Homothetic Triangle Cyclic Refinement In Wireless Sensor Networks\nYear: 2017\n\nAbstract: Wireless sensor networks (WSN) have become a significant technology in recent years. They can be widely used in many applications. WSNs consist of a large number of sensor nodes and each of them is energy-constrained and low-power dissipation. Most of the sensor nodes are tiny sensors with small memories and do not acquire their own locations. This means determining the locations of the unknown sensor nodes is one of the key issues in WSN. In this paper, an improved APIT algorithm HTCRL (Homothetic Triangle Cyclic Refinement Location) is proposed, which is based on the principle of the homothetic triangle. It adopts perpendicular median surface cutting to narrow down target area in order to decrease the average localization error rate. It reduces the probability of misjudgment by adding the conditions of judgment. It can get a relatively high accuracy compared with the typical APIT algorithm without any additional hardware equipment or increasing the communication overhead.","Title: A recommendation engine for travel products based on topic sequential patterns.\nYear: 2017\n\nAbstract: Travel products recommendation has become one of emerging issues in the realm of recommendation systems. The widely-used collaborative filtering algorithms are usually difficult to be used for recommending travel products due to a number of reasons, including (1) the content of travel products is very complex, (2) the user-item matrix is extremely sparse, and (3) the cold-start users are widely existing. To tackle these issues, we try to exploit Web server logs for generating recommendation, and present a novel recommendation engine (SECT for short) for travel products based on topic sequential patterns. In detail, we first extract topics from semantic description of every Web page. Then, we mine topic frequent sequential patterns and their target products to form click patterns library. At last, we propose a Markov n-gram model for matching the real-time click-stream of users with the click patterns library and thus computing recommendation scores. Experimental results on a real-world travel dataset demonstrate that the SECT prevails over the state-of-art baseline algorithms. In particular, SECT shows merits in improving the both coverage and accuracy for recommending products to cold-start users. Also, SECT is effective to recommend long tail items and outperform baseline algorithms.","Title: Recent Progress of Face Image Synthesis\nYear: 2017\n\nAbstract: Face synthesis has been a fascinating yet challenging problem in computer vision and machine learning. Its main research effort is to design algorithms to generate photo-realistic face images via given semantic domain. It has been a crucial prepossessing step of main-stream face recognition approaches and an excellent test of AI ability to use complicated probability distributions. In this paper, we provide a comprehensive review of typical face synthesis works that involve traditional methods as well as advanced deep learning approaches. Particularly, Generative Adversarial Net (GAN) is highlighted to generate photo-realistic and identity preserving results. Furthermore, the public available databases and evaluation metrics are introduced in details. We end the review with discussing unsolved difficulties and promising directions for future research.","Title: An accurate traffic classification model based on support vector machines.\nYear: 2017\n\nAbstract: Network traffic classification is a fundamental research topic on high-performance network protocol design and network operation management. Compared with other state-of-the-art studies done on the network traffic classification, machine learning ML methods are more flexible and intelligent, which can automatically search for and describe useful structural patterns in a supplied traffic dataset. As a typical ML method, support vector machines SVMs based on statistical theory has high classification accuracy and stability. However, the performance of SVM classifier can be severely affected by the data scale, feature dimension, and parameters of the classifier. In this paper, a real-time accurate SVM training model named SPP-SVM is proposed. An SPP-SVM is deducted from the scaling dataset and employs principal component analysis PCA to extract data features and verify its relevant traffic features obtained from PCA. By employing PCA algorithm to do the dimension extraction, SPP-SVM confirms the critical component features, reduces the redundancy among them, and lowers the original feature dimension so as to reduce the over fitting and increase its generalization effectively. The optimal working parameters of kernel function used in SPP-SVM are derived automatically from improved particle swarm optimization algorithm, which will optimize the global solution and make its inertia weight coefficient adaptive without searching for the parameters in a wide range, traversing all the parameter points in the grid and adjusting steps gradually. The performance of its two- and multi-class classifiers is proved over 2 sets of traffic traces, coming from different topological points on the Internet. Experiments show that the SPP-SVM's two- and multi-class classifiers are superior to the typical supervised ML algorithms and performs significantly better than traditional SVM in classification accuracy, dimension, and elapsed time.","Title: Edgeos(H): A Home Operating System For Internet Of Everything\nYear: 2017\n\nAbstract: The proliferation of Internet of Everything (IoE) and the success of rich Cloud services have pushed the horizon of a new computing paradigm, Edge Computing, which calls for processing the data at the edge of the network. Smart home as a typical IoE application is being widely adapted into people's lives. Edge Computing has the potential to empower the smart home, but it needs more contribution from the community before it truly benefits our lives. In this paper, we present the vision of EdgeOS(H), a home operating system for Internet of Everything. We further discuss functional challenges, namely programming interface, self-management, data management, security & privacy, and naming, as well as non-functional challenges, such as user experience, system cost, delay, and the lack of availability of open testbed. Within each challenge we also discuss the potential directions that are worth further investigation.","Title: Semi-SGD: Semi-Supervised Learning Based Spammer Group Detection in Product Reviews\nYear: 2017\n\nAbstract: The purchase decision of customers in e-commerce platforms is strongly influenced by product ratings and reviews. Driven by the profits, review spammers post fake reviews to promote their products or demote their competitors' products. Differ from individual spammers, the spammer groups manipulate reviews together and can be more damaging. Existing work for spammer group detection extract candidate groups from review data and identify the spammer groups using unsupervised spamicity ranking methods. However, the labeled and unlabeled data are existing simultaneously in practice and no method makes good use of both these data in spammer group detection. In this paper, we propose a semi-supervised learning based spammer group detection method (Semi-SGD), which trains a Naive Bayes classifier on a small set of labeled data as an initial classifier, and then incorporates unlabeled data with Expectation Maximization (EM) algorithm to improve the initial classifier iteratively. Experiments on Amazon.cn datasets show that our proposed Semi-SGD is efficient and effective.","Title: A Novel De-Noising Method for Improving the Performance of Full-Waveform LiDAR Using Differential Optical Path.\nYear: 2017\n\nAbstract: A novel de-noising method for improving the performance of full-waveform light detection and ranging (LiDAR) based on differential optical path is proposed, and the mathematical models of this method are developed and verified. Backscattered full-waveform signal (BFWS) is detected by two avalanche photodiodes placed before and after the focus of the focusing lens. On the basis of the proposed method, some simulations are carried out and conclusions are achieved. (1) Background noise can be suppressed effectively and peak points of the BFWS are transformed into negative-going zero-crossing points as stop timing moments. (2) The relative increment percentage of the signal-to-noise ratio based on the proposed method first dramatically increases with the increase of the distance, and then the improvement gets smaller by increasing the distance. (3) The differential Gaussian fitting with the Levenberg-Marquardt algorithm is applied, and the results show that it can decompose the BFWS with high accuracy. (4) The differential distance should not be larger than c/2 x (rmin), and two variable gain amplifiers can eliminate the inconsistency of two differential beams. The results are beneficial for designing a better performance full-waveform LiDAR.","Title: An Effective Training Method For Deep Convolutional Neural Network.\nYear: 2017\n\nAbstract: In this paper, we propose the nonlinearity generation method to speed up and stabilize the training of deep convolutional neural networks. The proposed method modifies a family of activation functions as nonlinearity generators (NGs). NGs make the activation functions linear symmetric for their inputs to lower model capacity, and automatically introduce nonlinearity to enhance the capacity of the model during training. The proposed method can be considered an unusual form of regularization: the model parameters are obtained by training a relatively low-capacity model, that is relatively easy to optimize at the beginning, with only a few iterations, and these parameters are reused for the initialization of a higher-capacity model. We derive the upper and lower bounds of variance of the weight variation, and show that the initial symmetric structure of NGs helps stabilize training. We evaluate the proposed method on different frameworks of convolutional neural networks over two object recognition benchmark tasks (CIFAR-10 and CIFAR-100). Experimental results showed that the proposed method allows us to (1) speed up the convergence of training, (2) allow for less careful weight initialization, (3) improve or at least maintain the performance of the model at negligible extra computational cost, and (4) easily train a very deep model.","Title: Fuzzy Consensus Clustering With Applications on Big Data.\nYear: 2017\n\nAbstract: Consensus clustering aims to find a single partition of data that agrees as much as possible with existing basic partitions. Given its robustness and generalizability, consensus clustering has emerged as a promising solution to find cluster structures inside heterogeneous big data rising from various application domains. In the area of fuzzy systems, however, research along this line is still in i...","Title: Analysis of Grain Storage Loss Based on Decision Tree Algorithm.\nYear: 2017\n\nAbstract: Different grain storage factors will cause different degrees of grain loss. In this paper, the data mining method is used to study the loss of grain storage, and the grain loss analysis and forecasting model based on decision tree algorithm is proposed. The paper analyzes and predicts the grain loss caused by different grain storage factors. And the influence of model parameters on model fitting and accuracy is verified by the verification curve. Then the decision tree model is optimized by the method of grid search and cross validation, which improves the prediction accuracy of the decision tree model to analyze the grain loss.","Title: Analysis of the grain loss in harvest based on logistic regression.\nYear: 2017\n\nAbstract: In this paper, the grain loss assessment was studied based on logistic regression, and 5400 samples of 31 provinces in our country in the year 2012-2014 were selected, and the 7 typical provinces among them were respectively tested. Using the logistic regression model to predict the loss rate of grain harvesting step, the prediction result is 86.25%. The stochastic gradient descent algorithm is used to optimize the parameters of the model, when the learning_rate is 0.1, the prediction results of grain harvest losses of up to 92.53%, which further improves the prediction accuracy of grain harvest step loss.","Title: Study on prediction model of grain post-harvest loss.\nYear: 2017\n\nAbstract: With the arrival of the information age, a great deal of data has been produced in a series of process of grain post-harvest. The rational use of these data allows us to obtain more intelligent, in-depth and valuable information. In this paper, we set up a variety of prediction models for the consumption of grain post-harvest loss, and select the appropriate classifier through comparison. On this basis, the dimension reduction is processed and the confusion matrix is used as the evaluation index to evaluate the prediction effect. Make correlation analysis of the data, obtained the main influencing factors of post-harvest consumption loss link. Finally, visualize the results of the process.","Title: TIDM: Topic-Specific Information Detection Model.\nYear: 2017\n\nAbstract: Nowadays information control and detection on the social network have become a problem that we should solve as soon as possible. Unfortunately, due to the informal expressions, detecting the massive data on the internet is a big challenge based on the traditional text mining methods such as Topic Model. In our paper, we propose a simple 4-Tuple Structure instead of the raw text event which usually contains many meaningless words. Using the word embedding technique, we propose the Topic-Specific Information Detection Model (TIDM) for detecting the specific information. For training the words and idiomatic phrases, we adopt the supervise learning technique: manually constructing a specific Semantic Dataset for training our model. Our experiments based on the Amazon Reviews demonstrate that the TIDM can effectively detect and recognize the information.","Title: A generalized game theoretic framework for mining communities in complex networks.\nYear: 2018\n\nAbstract: •We introduce a game theoretic framework to model community formation in networks.•We deduce many commonly used objective functions as potential functions in the model.•The proposed framework has been proved to exist fixed (equilibrium) point.•A synchronous learning mechanism has be proposed to reach the local equilibrium.","Title: A Simplified FRI Sampling System for Pulse Streams Based on Constraint Random Modulation.\nYear: 2018\n\nAbstract: The recent finite rate of innovation (FRI) framework provides effective sub-Nyquist sampling of pulse streams, allowing recovery of such signals from a set of Fourier coefficients. In this brief, a multi-channel FRI sampling system is presented to sample distinct bands of Fourier coefficients. This is achieved through modulating the desired spectrum band to baseband and then filtering with a low-p...","Title: Detecting Spammer Groups From Product Reviews: A Partially Supervised Learning Model.\nYear: 2018\n\nAbstract: Nowadays, online product reviews play a crucial role in the purchase decision of consumers. A high proportion of positive reviews will bring substantial sales growth, while negative reviews will cause sales loss. Driven by the immense financial profits, many spammers try to promote their products or demote their competitors' products by posting fake and biased online reviews. By registering a number of accounts or releasing tasks in crowdsourcing platforms, many individual spammers could be organized as spammer groups to manipulate the product reviews together and can be more damaging. Existing works on spammer group detection extract spammer group candidates from review data and identify the real spammer groups using unsupervised spamicity ranking methods. Actually, according to the previous research, labeling a small number of spammer groups is easier than one assumes, however, few methods try to make good use of these important labeled data. In this paper, we propose a partially supervised learning model (PSGD) to detect spammer groups. By labeling some spammer groups as positive instances, PSGD applies positive unlabeled learning (PU-Learning) to study a classifier as spammer group detector from positive instances (labeled spammer groups) and unlabeled instances (unlabeled groups). Specifically, we extract reliable negative set in terms of the positive instances and the distinctive features. By combining the positive instances, extracted negative instances and unlabeled instances, we convert the PU-Learning problem into the well-known semi supervised learning problem, and then use a Naive Bayesian model and an EM algorithm to train a classifier for spammer group detection. Experiments on real-life Amazon.cn data set show that the proposed PSGD is effective and outperforms the state-of-the-art spammer group detection methods.","Title: Hybrid Attention Networks For Chinese Short Text Classification\nYear: 2017\n\nAbstract: To improve the classification performance for Chinese short text with automatic semantic feature selection, in this paper we propose the Hybrid Attention Networks (HANs) which combines the word-and character-level selective attentions. The model firstly applies RNN and CNN to extract the semantic features of texts. Then it captures class-related attentive representation from word-and character-level features. Finally, all of the features are concatenated and fed into the output layer for classification. Experimental results on 32-class and 5-class datasets show that, our model outperforms multiple baselines by combining not only the word-and character-level features of the texts, but also class-related semantic features by attentive mechanism.","Title: A novel subgraph K+ -isomorphism method in social network based on graph similarity detection.\nYear: 2018\n\nAbstract: In this paper, we propose a novel (K^{+})-isomorphism method to achieve K-anonymization state among subgraphs or detected communities in a given social network. Our proposed (K^{+})-isomorphism method firstly partitions the subgraphs we have detected into some similar-subgraph clusters followed by graph modification conducted in every cluster. In this way, it is feasible to publish preserved structures of communities or subgraphs and every preserved structure actually represents a cluster of at least K subgraphs or communities which are isomorphic to each other. The contributions of this paper are listed as follows: On the one hand, we improve a maximum common subgraph detection algorithm, MPD(_{-})V, which is a core technique for graph similarity detection involved in partition phase of our proposed (K^{+})-isomorphism method; on the other hand, with minor adjustment, we utilize some current techniques as an innovative combination to finish the partition and modification of similar-community cluster in (K^{+})-isomorphism method. The experiments have shown that the improved MPD(_{-})V method has much better efficiency to search larger common subgraphs with acceptable performance compared with its prototype and other techniques. Moreover, our proposed (K^{+})-isomorphism method can achieve the K-isomorphism state with less modification of original network structure, or lower anonymization cost compared to the current K-isomorphism method.","Title: Load Balanced GANs for Multi-view Face Image Synthesis.\nYear: 2018\n\nAbstract: Multi-view face synthesis from a single image is an ill-posed problem and often suffers from serious appearance distortion. Producing photo-realistic and identity preserving multi-view results is still a not well defined synthesis problem. This paper proposes Load Balanced Generative Adversarial Networks (LB-GAN) to precisely rotate the yaw angle of an input face image to any specified angle. LB-GAN decomposes the challenging synthesis problem into two well constrained subtasks that correspond to a face normalizer and a face editor respectively. The normalizer first frontalizes an input image, and then the editor rotates the frontalized image to a desired pose guided by a remote code. In order to generate photo-realistic local details, the normalizer and the editor are trained in a two-stage manner and regulated by a conditional self-cycle loss and an attention based L2 loss. Exhaustive experiments on controlled and uncontrolled environments demonstrate that the proposed method not only improves the visual realism of multi-view synthetic images, but also preserves identity information well.","Title: A Weighted Collaboration Network Generalization Method For Privacy Protection In C-Dblp\nYear: 2018\n\nAbstract: The increasing population of online communication and telecommunication has interested scholars and researchers considering their social networks. These social networks datasets play an exceptionally important role in the research of data mining. However, large amounts of social network data are produced by using social networking applications. And these data inevitably contain a large amount of personal privacy information. Therefore, in order to avoid disclosure of privacy, the data holders need adopt privacy protection before these data are released. Furthermore, most current methods of privacy protection are based on the simple graph only. The weight values on the edges represent the tightness between the nodes. The algorithm based on weights in privacy protection field is still relatively rare. In real social networks, the weight can indicate tightness between two individuals of social relations. The weight may be as attackersar background knowledge to re-identify the target individual and lead to loss of privacy. In this paper, we consider protecting the weighted social networks from weight-based attacks and propose a method based on the weighted social networks, named k-weighted generalization anonymity (KWGA). And This method combines k-anonymous with generalization method to ensure the security of the social network data when it is published. In order to ensure the higher validity of privacy protection, this paper introduces a concept of the weight difference to reduce the modification for weight graph. Finally, we firstly use real dataset C-DBLP to verify the validity of our method perform much better than the Fast k-degree anonymity (FKDA) in average clustering coefficient (ACC), global clustering coefficient (GCC), average path length (APL) and rate of edges change four aspects. Furthermore, we also use two common datasets in the research of weighted graphs to verify our algorithm.","Title: GLEAM: a graph clustering framework based on potential game optimization for large-scale social networks.\nYear: 2018\n\nAbstract: With the growing explosion of online social networks, the study of large-scale graph clustering has attracted considerable interest. Most of traditional methods view the graph clustering problem as an optimization problem based on a given objective function; however, there are few methodical theories for the emergence of clusters over real-life networks. In this paper, each actor in online social networks is viewed as a selfish player in a non-cooperative game. The strategy associated with each node is defined as the cluster membership vector, and each one’s incentive is to maximize its own social identity by adopting the most suitable strategy. The definition of utility function in our game model is inspired by the conformity psychology, which is defined as the weighted average of one’s social identity by participating different clusters. With this setting, the proposed game can well match a potential game. So that the cluster could be shaped by the actions of those closely interactive users who adopt the same strategy in a Nash equilibrium. To this end, we propose a novel Graph cLustering framework based on potEntial gAme optiMization (GLEAM) for parallel graph clustering. It first utilize the cosine similarity to weight each edge in the original network. Then, an initial partition, including a number of clusters dominated by those potential “leader nodes”, is created by a fast heuristic process. Third, a potential game-based weighted Modularity optimization is used to improve the initial partition. Finally, we introduce the notion of potentially attractive cluster, and then discover the overlapping partition of the graph using a simple double-threshold procedure. Three phases in GLEAM are carefully designed for parallel execution. Experiments on real-world networks analyze the convergence inside GLEAM, and demonstrate the high performance of GLEAM by comparing it with the state-of-the-art community detection approaches in the literature.","Title: H∞ filtering for networked systems with hybrid-triggered communication mechanism and stochastic cyber attacks.\nYear: 2017\n\nAbstract: This paper concentrates on investigating H∞ filtering for networked systems with hybrid-triggered communication mechanism and stochastic cyber attacks. Random variables satisfying Bernoulli distribution are introduced to describe the hybrid-triggered scheme and stochastic cyber attacks, respectively. Firstly, a mathematical H∞ filtering error model with hybrid-triggered communication mechanism is constructed under the stochastic cyber attacks. Secondly, by using Lyapunov stability theory and linear matrix inequality (LMI) techniques, the sufficient conditions which can guarantee the stability of augmented filtering error system are obtained and the parameters of the designed filter can be presented in an explicit form. Finally, numerical examples are given to demonstrate the feasibility of the designed filter.","Title: A novel POMDP-based server RAM caching algorithm for VoD systems.\nYear: 2018\n\nAbstract: In contrast to other services on the Internet, streaming media service needs to fetch data from local disks more frequently, since it always lasts long and the bit rate is quite high. In addition, because of the much slower reading/writing speed of disk than random access memory (RAM), adopting advisable RAM caching policy can efficiently reduce disk I/O. In this paper, we study the problem of reducing disk I/O by using a novel approach. We first provide a new popularity estimate algorithm. Then a formal optimization problem about average disk I/O is presented, and a suboptimal caching algorithm for a special case of the problem is given. Furthermore, a partially observable Markov decision process (POMDP) model is constructed for the caching problem. Based on the model, popularity is taken advantage of to predict clients’ randomized behaviors, data replacing decisions are made when the defined observations occur, and the impact of caching actions on disk performance for future infinite steps is assessed. The method of event-based optimization is applied in search of the optimal stochastic policy. Disk I/O, as the long-run average performance measure, is optimized by applying the policy-gradient algorithm. The simulation results illustrate that data required by clients could be better predicted and lower disk I/O could be achieved by using the model proposed in this paper.","Title: Quantized state estimation for neural networks with cyber attacks and hybrid triggered communication scheme.\nYear: 2018\n\nAbstract: This paper is concerned with the issue of quantized state estimation for neural networks with cyber attacks and hybrid triggered communication scheme. In order to reduce the pressure of the network transmission and save the network resources, the hybrid triggered scheme and quantization are introduced. The hybrid triggered scheme consists of time triggered scheme and event triggered scheme, in which the stochastic switch is described by a variable satisfying Bernoulli distribution. First, by taking the effect of hybrid triggered scheme and quantization into consideration, a mathematical model for estimating the state of neural networks is constructed. Second, by using linear matrix inequality (LMI) techniques and Lyapunov stability theory, the sufficient conditions are given which can ensure the stability of estimating error system under hybrid triggered scheme, and the designing algorithm of desired state estimator is also presented in terms of LMIs. Finally, a numerical example is given to show the usefulness of the proposed approach.","Title: Graph classification based on graph set reconstruction and graph kernel feature reduction.\nYear: 2018\n\nAbstract: Graph, a kind of structured data, is widely used to model complex relationships among objects, and has been used in various of scientific and engineering fields, such as bioinformatics, network intrusion detection, social network, etc. Building an automatic and highly accurate classification method for graphs becomes quite necessary for predicting unknown graphs or understanding complex structures among different categories. The kernel method is regarded as a powerful solution to graph classification, which consists of two steps, namely, graph kernel mapping and classification. However, the feature selection process is ignored, and those sub-structures with low discriminative power result in classification accuracy decrease. In order to solve this problem, we propose an efficient graph classification algorithm based on graph set reconstruction and graph kernel feature reduction. First of all, the least discriminative frequent subgraphs and part of the infrequent subgraphs are removed to reconstruct the original graph set. Then we adopt the graph-kernel-based discriminant analysis method to perform feature reduction on the well-reconstructed graph set. At last, the whole framework of the graph classification method is introduced and any commonly used classifiers can be utilized. Extensive experimental results on a series of bioinformatics benchmarks show that our graph classification algorithm demonstrates a significant improvement of prediction comparing with other graph-kernel-based classification approaches.","Title: Enhance the Performance of Network Computation by a Tunable Weighting Strategy.\nYear: 2018\n\nAbstract: Networked systems with high computational efficiency are desired in many applications ranging from sociology to engineering. Generally, the performance of the network computation can be enhanced by two ways: rewiring and weighting. In this paper, we proposed a new two-modes weighting strategy based on the concept of communication neighbor graph, which takes use of both the local and global topolog...","Title: A modified artificial bee colony approach for the 0-1 knapsack problem.\nYear: 2018\n\nAbstract: The 0-1 knapsack problem (KP01) is one of the classical NP-hard problems in operation research and has a number of engineering applications. In this paper, the BABC-DE (binary artificial bee colony algorithm with differential evolution), a modified artificial bee colony algorithm, is proposed to solve KP01. In BABC-DE, a new binary searching operator which comprehensively considers the memory and neighbour information is designed in the employed bee phase, and the mutation and crossover operations of differential evolution are adopted in the onlooker bee phase. In order to make the searching solution feasible, a repair operator based on greedy strategy is employed. Experimental results on different dimensional KP01s verify the efficiency of the proposed method, and it gets superior performance compared with other five metaheuristic algorithms.","Title: A variable resolution feedback improving the performances of object detection and recognition.\nYear: 2018\n\nAbstract: In order to improve the performances of object detection and recognition, a two-stage framework combined with variable resolution control strategy is proposed. The images in low resolution and high resolution are employed in object detection and object recognition, respectively. Meanwhile, a feedback mechanism used in two-stage framework is proved to effectively improve the performances of object detection. The results show that under low resolution, the accuracy of object detection based on fixed resolution 320 x 240 without feedback mechanism is 36.7%. However, the accuracy of object detection of the proposed method based on variable resolution increases to 95.3%. Under high resolution, compared with the method based on fixed resolution 1280 x 960 using the two-stage framework, the time consumption of the proposed method with variable resolution decreases by 51.4%, while keeping almost identical recognition accuracy.","Title: A Hybrid Bionic Image Sensor Achieving FOV Extension and Foveated Imaging.\nYear: 2018\n\nAbstract: Based on bionic compound eye and human foveated imaging mechanisms, a hybrid bionic image sensor (HBIS) is proposed in this paper to extend the field of view (FOV) with high resolution. First, the hybrid bionic imaging model was developed and the structure parameters of the HBIS were deduced. Second, the properties of the HBIS were simulated, including FOV extension, super-resolution imaging, foveal ratio and so on. Third, a prototype of the HBIS was developed to validate the theory. Imaging experiments were carried out, and the results are in accordance with the simulations, proving the potential of the HBIS for large FOV and high-resolution imaging with low cost.","Title: A comparative study of clustering ensemble algorithms.\nYear: 2018\n\nAbstract: Since clustering ensemble was proposed, it has rapidly attracted much attention. This paper makes an overview of recent research on clustering ensemble about generative mechanism, selective clustering ensemble, consensus function and application. Twelve clustering ensemble algorithms are described and compared to choose a basic one. The experiment shows that using k-means with different initializations as generative mechanism and average-linkage agglomerative clustering as consensus function is the best one. As ensemble size increases, the performance of clustering ensemble improves. The basic clustering ensemble algorithm with suitable ensemble size is compared with clustering algorithms and the experiment shows that clustering ensemble is better than clustering. The influence of diversity on clustering ensemble is instructive to selecting members. The experiment shows that selecting members in high quality and big diversity for low-dimensional data sets, and selecting members in high quality and median diversity for high-dimensional data sets are better than traditional clustering ensemble.","Title: A hybrid short-term traffic flow forecasting model based on time series multifractal characteristics.\nYear: 2018\n\nAbstract: Short-term traffic flow forecasting is a key problem in the area of intelligent transportation systems (ITS). Timely and accurate traffic state prediction is also the prerequisite of realizing proactive traffic control and dynamic traffic assignment effectively. In this paper, a new hybrid model for short-term traffic flow forecasting, which is built based on multifractal characteristics of traffic flow time series, is proposed. The hybrid model decomposes traffic flow series into four different components, namely a periodic part, a trend part, a stationary part and a volatility part, to unearth the traffic features hidden behind the data. Four parts are treated and modeled separately by using different methods, such as spectral analysis, time series and statistical volatility analysis, to further explore the underlying traffic patterns and improve forecasting accuracy. Performance of the proposed hybrid model is investigated with traffic flow data from freeway I-694 EB in the Twin Cities. The experimental results indicate that the proposed model outperforms in capturing nonlinear volatility and improving forecasting accuracy than traditional forecasting methods, especially for the multi-step ahead forecasting. Compared with the ARIMA-GARCH model, it gets an improvement of 8.23% in RMSE for one-step ahead forecasting and 10.69% for ten-step ahead forecasting. It is better than the hybrid model newly proposed in literature (Zhang et al. Transp Res Part C: Emerg Technol 43(1):65–78 2014) and gets an improvement of 1.27% in forecasting accuracy.","Title: Microwave Metamaterial Absorber for Non-Destructive Sensing Applications of Grain.\nYear: 2018\n\nAbstract: In this work, we propose a metamaterial absorber at microwave frequencies with significant sensitivity and non-destructive sensing capability for grain samples. This absorber is composed of cross-resonators periodically arranged on an ultrathin substrate, a sensing layer filled with grain samples, and a metal ground. The cross-resonator array is fabricated using the printed circuit board process on an FR-4 board. The performance of the proposed metamaterial is demonstrated with both full-wave simulation and measurement results, and the working mechanism is revealed through multi-reflection interference theory. It can serve as a non-contact sensor for food quality control such as adulteration, variety, etc. by detecting shifts in the resonant frequencies. As a direct application, it is shown that the resonant frequency displays a significant blue shift from 7.11 GHz to 7.52 GHz when the mass fraction of stale rice in the mixture of fresh and stale rice is changed from 0% to 100%. In addition, the absorber shows a distinct difference in the resonant absorption frequency for different varieties of grain, which also makes it a candidate for a grain classification sensor. The presented scheme could open up opportunities for microwave metamaterial absorbers to be applied as efficient sensors in the non-destructive evaluation of agricultural and food product quality.","Title: A Sequential Recommendation for Mobile Apps: What Will User Click Next App?\nYear: 2018\n\nAbstract: With the advances in smartphones users install abundant apps to facilitate their daily lives. Both users and related developers have increasing requirements to understand the mobile App usage pattern, for individual and commercial use. Respectively, personalized App recommendation methods and systems have emerged as a novel attractive topic that can demonstrate the human App usage behavior. The mobile Apps recommendation can serve as a cornerstone for a variety of intelligent services, such as fast-launching UIs, intelligent user-phone interactions, and battery management of cellphones. In this paper, we develop a novel App recommendation framework combining the historical App usage data with the sequence of recently-used Apps. Specifically, our framework is an extension of the user-based collaborative filtering technique, where the set of nearest neighbors is employed for training the prediction model. However, our prediction scheme is constructed on the temporal sequential data and is modeled by using the chain-augmented Naive Bayes model. Experiments with a real mobile Internet record dataset demonstrate that the accuracy of our framework outperforms several baseline App recommendation approaches.","Title: Understanding Customer Behavior In Shopping Mall From Indoor Tracking Data\nYear: 2018\n\nAbstract: The prosperity of various indoor positioning technologies makes possible the large collection of tracking data in indoor spaces. Much of the focus has been on several fundamental problems such as indoor localization, indoor space modeling, indoor data cleansing, indexing and querying. However, this paper attempts to analyze customer behavior from a unique indoor tracking data, which will promote the convergence between various applications and the underlying data. In particular, we introduce a real-life indoor tracking data collected at an electrical mall in China. Then, we cluster users into several groups and summarize the most characteristic behaviors of each cluster. Last but not least, we analyze customer's individual behaviors through two aspects: 1) the regression model is used to reveal hot regions where customers are likely to stay in long time; and 2) a transition matrix combined with the connectivity is presented to demonstrate hot paths.","Title: Skip-connection convolutional neural network for still image crowd counting.\nYear: 2018\n\nAbstract: In recent years, crowd counting in still images has attracted many research interests due to its applications in public safety. However, it remains a challenging task for reasons of perspective and scale variations. In this paper, we propose an effective Skip-connection Convolutional Neural Network (SCNN) for crowd counting to overcome the issue of scale variations. The proposed SCNN architecture consists of several multi-scale units to extract multi-scale features. Each multi-scale unit including three convolutional layers builds connections between the input and each convolutional layer. In addition, we propose a scale-related training method to improve the accuracy and robustness of crowd counting. We evaluate our method on three crowd counting benchmarks. Experimental results verify the efficiency of the proposed method, and it achieves superior performance compared with other methods.","Title: A multivariate short-term traffic flow forecasting method based on wavelet analysis and seasonal time series.\nYear: 2018\n\nAbstract: Short-term traffic flow forecasting is a key step to achieve the performance of intelligent transportation system (ITS). Timely and accurate traffic information prediction is also the prerequisite of realizing proactive traffic control and dynamic traffic assignment effectively. Based on the fact that univariate forecasting methods have limited forecasting abilities when the data is missing or erroneous and that single models make no full use of information underline data, a new hybrid method with multivariate for short-term traffic flow forecasting is proposed. This method combines statistical analysis method with computational intelligence techniques to mine the characteristic of traffic flow as well as forecast short-term traffic state. First, the wavelet de-noising is employed to remove the noise information. Then, time series analysis is used to analyze time-varying and periodic characteristic of traffic flow. Furthermore, the seasonal auto-regressive moving average with external input (SARIMAX) is established to fit traffic flow with occupancy as exogenous variables. Finally, wavelet forecast is adopted to forecast the values of occupancy which are used as exogenous input, and a WSARIMAX is constructed to forecast traffic flow. Using the relationship of flow and occupancy at the same road section and taking traffic flow and occupancy data from freeway I-694 EB in the Twin Cities as endogenous variables and exogenous variables respectively, this paper studies the forecasting performance of the proposed method. The study results are encouraging. Compared with SARIMA newly proposed in literature, WSARIMA and SARIMAX improved method with wavelet analysis and multivariate modeling method, the proposed method gets improvements of 12.95%, 12.62% and 10.41% in forecasting accuracy of one-step ahead respectively. For ten-steps ahead forecasting, it gets improvements of 18.87%, 17.05% and 2.57% in forecasting accuracy respectively.","Title: Few-shot learning for short text classification.\nYear: 2018\n\nAbstract: Due to the limited length and freely constructed sentence structures, it is a difficult classification task for short text classification. In this paper, a short text classification framework based on Siamese CNNs and few-shot learning is proposed. The Siamese CNNs will learn the discriminative text encoding so as to help classifiers distinguish those obscure or informal sentence. The different sentence structures and different descriptions of a topic are viewed as ‘prototypes’, which will be learned by few-shot learning strategy to improve the classifier’s generalization. Our experimental results show that the proposed framework leads to better results in accuracies on twitter classifications and outperforms some popular traditional text classification methods and a few deep network approaches.","Title: Recombined multinomial tree based on saddle-point approximation and its application to Levy models options pricing.\nYear: 2019\n\nAbstract: This paper studies the constructing methods of a recombined multinomial tree based on saddle-point approximation and its application to Levy models options pricing. Firstly, the Levy process and the European option pricing are introduced. Then, we used the characteristic function of Levy process to generate density function value at discrete point in a certain range, based on the saddle-point approximation method. Then, we provide the method to construct recombined multinomial tree and give the pricing formula of European option and path-dependent option pricing based on the backward iteration. Finally, we used the CGMY process to demonstrate its application to European option, American option and American barrier option pricing. It proves that saddle-point approximation turns the inverse Fourier integral transform problem to several function value calculation. Comparing to IFFT, the speed of calculation using this method is faster, and it avoid the negative probability density function value based on IFFT. Because of the linear growth of the number of nodes, we can extend the saddle-point approximation method to the calculation of path-dependent option.","Title: Parallel Attribute Reduction Algorithm for Complex Heterogeneous Data Using MapReduce.\nYear: 2018\n\nAbstract: Parallel attribute reduction is one of the most important topics in current research on rough set theory. Although some parallel algorithms were well documented, most of them are still faced with some challenges for effectively dealing with the complex heterogeneous data including categorical and numerical attributes. Aiming at this problem, a novel attribute reduction algorithm based on neighborhood multigranulation rough sets was developed to process the massive heterogeneous data in the parallel way. The MapReduce-based parallelization method for attribute reduction was proposed in the framework of neighborhood multigranulation rough sets. To improve the reduction efficiency, the hashing Map/Reduce functions were designed to speed up the positive region calculation. Thereafter, a quick parallel attribute reduction algorithm using MapReduce was developed. The effectiveness and superiority of this parallel algorithm were demonstrated by theoretical analysis and comparison experiments.","Title: A pattern-based topic detection and analysis system on Chinese tweets.\nYear: 2018\n\nAbstract: •A FP-growth-like algorithm to extract cosine interesting patterns as topics.•A distributed topic detecting algorithm on Spark.•Evolutionary and sentimental analysis on the detected topics.","Title: Dynamic Cluster Formation Game for Attributed Graph Clustering.\nYear: 2019\n\nAbstract: Besides the topological structure, there are additional information, i.e., node attributes, on top of the plain graphs. Usually, these systems can be well modeled by attributed graphs, where nodes represent component actors, a set of attributes describe users' portraits and edges indicate their connections. An elusive question associated with attributed graphs is to study how clusters with common ...","Title: hPSD: A Hybrid PU-Learning-Based Spammer Detection Model for Product Reviews.\nYear: 2020\n\nAbstract: Spammers, who manipulate online reviews to promote or suppress products, are flooding in online commerce. To combat this trend, there has been a great deal of research focused on detecting review spammers, most of which design diversified features and thus develop various classifiers. The widespread growth of crowdsourcing platforms has created large-scale deceptive review writers who behave more like normal users, that the way they can more easily evade detection by the classifiers that are purely based on fixed characteristics. In this paper, we propose a hybrid semisupervised learning model titled hybrid PU-learning-based spammer detection (hPSD) for spammer detection to leverage both the users’ characteristics and the user–product relations. Specifically, the hPSD model can iteratively detect multitype spammers by injecting different positive samples, and allows the construction of classifiers in a semisupervised hybrid learning framework. Comprehensive experiments on movie dataset with shilling injection confirm the superior performance of hPSD over existing baseline methods. The hPSD is then utilized to detect the hidden spammers from real-life Amazon data. A set of spammers and their underlying employers (e.g., book publishers) are successfully discovered and validated. These demonstrate that hPSD meets the real-world application scenarios and can thus effectively detect the potentially deceptive review writers.","Title: Enabling Semantics in oneM2M Service Delivery Platform\nYear: 2018\n\nAbstract: With the burgeoning of Edge computing, data from the Internet of Things (IoT) and Machine-to-Machine (M2M) systems are shared and consumed at the edge of the network. How to understand and leverage the data from other systems is a crucial challenge for information exchange and communication among IoT/M2M. Semantic paved the road for the knowledge-driven IoT/M2M systems. Context awareness becomes possible with the help of semantic information. In this paper, we proposed the methods of support semantic service in a sophisticated IoT/M2M system with reasonable overhead. Moreover, we also discussed the approaches to support a secure mechanism such as access control directly within the semantic information to boost the performance of the system. Our experiment shows that the proposed method could effectively and securely provide semantic service to the IoT/M2M system.","Title: A Comparative Study of Subgraph Matching Isomorphic Methods in Social Networks.\nYear: 2018\n\nAbstract: With the fast development of social networks, more and more data has been generated. Finding useful information among these data is important. Subgraph matching is the method that can be used in social networks for social search, recommender systems, and so on. In this paper, we choose five typical graph matching methods which are VF2, SPath, Turbo(ISO), BoostIso, and RI to value their performance and scalability in social networks. These methods are verified by three social network data sets whose node's number vary from thousands to millions. According to the experiments, we can find that VF2 and RI is applicable for rather small graphs while SPath performs better in large graph when average degree of graph is small. When graph with high average degree, Turbo(ISO) and BoostIso performs better. What is more, we also compare the efficiency of different matching order when searching. The order choosing strategy proposed by RI is better.","Title: Improving lazy decision tree for imbalanced classification by using skew-insensitive criteria\nYear: 2019\n\nAbstract: Lazy decision tree (LazyDT) constructs a customized decision tree for each test instance, which consists of only a single path from the root to a leaf node. LazyDT has two strengths in comparison with eager decision trees. One is that LazyDT can build shorter decision paths than eager decision trees, and the other is that LazyDT can avoid unnecessary data fragmentation. However, the split criterion used for constructing a customized tree in LazyDT is information gain, which is skew-sensitive. When learning from imbalanced data sets, class imbalance impedes their ability to learn the minority class concept. In this paper, we use Hellinger distance and K-L divergence as split criteria to build two types of lazy decision trees. An experimental framework is performed across a wide range of imbalanced data sets to investigate the effectiveness of our methods when comparing with the other methods including lazy decision tree, C4.5, Hellinger distance based decision tree and support vector machine. In addition, we also use SMOTE to preprocess the highly imbalance data sets in the experiment and evaluate its effectiveness. The experimental results, which contrasted through nonparametric statistical tests, demonstrate that using Hellinger distance and K-L divergence as the split criterion can improve the performances of LazyDT for imbalanced classification effectively.","Title: Feature Selection Using Forest Optimization Algorithm Based On Contribution Degree\nYear: 2018\n\nAbstract: As a combinatorial optimization problem, feature selection has been widely used in machine learning and data mining In this paper, a feature selection method using forest optimization algorithm based on contribution degree is proposed. The proposed method uses a contribution degree strategy which is embedded in forest optimization algorithm. The goal of the contribution degree is to guide the search process of the forest optimization algorithm to select features according to high class correlation and low redundancy between features. The proposed algorithm is verified on some data sets from the UCI repository and the experiments show that the proposed method improves the classification accuracy compared with some other methods.","Title: Combining Non-Uniform Time Slice and Finite Difference to Improve 3D Ghost Imaging.\nYear: 2019\n\nAbstract: Three-dimensional ghost imaging (3DGI) using a detector is widely used in many applications. The performance of 3DGI based on a uniform time slice is difficult to improve because obtaining an accurate time-slice position remains a challenge. This paper reports a novel structure based on non-uniform time slice combined with finite difference. In this approach, finite difference is beneficial to improving sensitivity of zero crossing to accurately obtain the position of the target in the field of view. Simultaneously, non-uniform time slice is used to quickly obtain 3DGI on an interesting target. Results show that better performances of 3DGI are obtained by our proposed method compared to the traditional method. Moreover, the relation between time slice and the signal-noise-ratio of 3DGI is discussed, and the optimal differential distance is obtained, thus motivating the development of a high-performance 3DGI.","Title: Buoyancy Force Acting on Underground Structures considering Seepage of Confined Water.\nYear: 2019\n\nAbstract: The antifloating property of underground structures in areas with high underground water levels is a key design aspect. Evaluating the buoyancy forces acting on underground structures is complicated, particularly in the presence of confined water beneath the structures. Herein, the effects of the permeability coefficient of layered soil, hydraulic gradient, and embedment depth of the aquiclude on the buoyancy force acting on underground structures are investigated through three model tests: (1) calibration of the test system, (2) buoyancy force acting on a structure located in homogeneous soil considering vertical direction seepage, and (3) buoyancy force acting on a structure located in layered soil considering vertical seepage of confined water. The results show that the pore pressure along the structure and the buoyancy force acting on the underground structure considering seepage are greater than those obtained under hydrostatic conditions. The raising ratios of the pore pressure and buoyancy force are equal to the vertical hydraulic gradient when seepage occurs in homogeneous soil. In the presence of confined water, the raising ratio is significantly greater than the hydraulic gradient. In the cases studied herein, the raising ratio is approximately twice the hydraulic gradient. Simplified equations are proposed to calculate the buoyancy force acting on underground structures considering the vertical seepage of confined water. Finally, a finite element analysis is carried out to verify the conclusions obtained from the model test and the rationality of the proposed equations.","Title: Location-Centered House Price Prediction: A Multi-Task Learning Approach\nYear: 2019\n\nAbstract: AbstractAccurate house prediction is of great significance to various real estate stakeholders such as house owners, buyers, and investors. We propose a location-centered prediction framework that differs from existing work in terms of data profiling and prediction model. Regarding data profiling, we make an important observation as follows – besides the in-house features such as floor area, the location plays a critical role in house price prediction. Unfortunately, existing work either overlooked it or had a coarse grained measurement of locations. Thereby, we define and capture a fine-grained location profile powered by a diverse range of location data sources, including transportation profile, education profile, suburb profile based on census data, and facility profile. Regarding the choice of prediction model, we observe that a variety of approaches either consider the entire data for modeling, or split the entire house data and model each partition independently. However, such modeling ignores the relatedness among partitions, and for all prediction scenarios, there may not be sufficient training samples per partition for the latter approach. We address this problem by conducting a careful study of exploiting the Multi-Task Learning (MTL) model. Specifically, we map the strategies for splitting the entire house data to the ways the tasks are defined in MTL, and select specific MTL-based methods with different regularization terms to capture and exploit the relatedness among tasks. Based on real-world house transaction data collected in Melbourne, Australia, we design extensive experimental evaluations, and the results indicate a significant superiority of MTL-based methods over state-of-the-art approaches. Meanwhile, we conduct an in-depth analysis on the impact of task definitions and method selections in MTL on the prediction performance, and demonstrate that the impact of task definitions on prediction performance far exceeds that of method selections.","Title: Improving task scheduling with parallelism awareness in heterogeneous computational environments.\nYear: 2019\n\nAbstract: Task scheduling is a key function for executing tasks in heterogeneous computational environments, efficiently. While the available computing resources are not fully used when applying existing scheduling methods as they consider that a task is executed on one single core or on a server without parallel tasks by assuming that the task exhausts the server. Therefore, in this paper, we focus on the problem of executing tasks with deadline constraints with parallelism awareness where the parallel degree of each task can be tuned between one and its maximum according to the available cores of the server it assigned to during its execution. We first model the problem as an optimization problem maximizing the overall utilization of servers, and propose a set of scheduling methods with parallelism awareness (SPA), each of which iteratively allocates as much resources and as soon as possible to the assigned task with the earliest deadline on a server, based on existing scheduling algorithms, and present two SPA instances to illustrate the implement of SPA. Experiment results show a great performance improvement in various aspects, e.g., resource utilization, task violations, finish time, and energy efficiency, when executing tasks heterogeneous computational systems using SPA.","Title: CAMAS: A cluster-aware multiagent system for attributed graph clustering.\nYear: 2017\n\nAbstract: Abstract Attributed graphs describe nodes via attribute vectors and also relationships between different nodes via edges. To partition nodes into clusters with tighter correlations, an effective way is applying clustering techniques on attributed graphs based on various criteria such as node connectivity and/or attribute similarity. Even though clusters typically form around nodes with tight edges and similar attributes, existing methods have only focused on one of these two data modalities. In this paper, we comprehend each node as an autonomous agent and develop an accurate and scalable multiagent system for extracting overlapping clusters in attributed graphs. First, a kernel function with a tunable bandwidth factor δ is introduced to measure the influence of each agent, and those agents with highest local influence can be viewed as the “leader” agents. Then, a novel local expansion strategy is proposed, which can be applied by each leader agent to absorb the most relevant followers in the graph. Finally, we design the cluster-aware multiagent system (CAMAS), in which agents communicate with each other freely under an efficient communication mechanism. Using the proposed multiagent system, we are able to uncover the optimal overlapping cluster configuration, i.e. nodes within one cluster are not only connected closely with each other but also with similar attributes. Our method is highly efficient, and the computational time is shown that nearly linearly dependent on the number of edges when δ ∈ [0.5, 1). Finally, applications of the proposed method on a variety of synthetic benchmark graphs and real-life attributed graphs are demonstrated to verify the systematic performance.","Title: Efficiently Translating Complex SQL Query to MapReduce Jobflow on Cloud\nYear: 2020\n\nAbstract: MapReduce is a widely-used programming model in cloud environment for parallel processing large-scale data sets. The combination of the high-level language with a SQL-to-MapReduce translator allows programmers to code using SQL-like declarative language, so that each program can afterwards be complied into a MapReduce jobflow automatically. This way is helpful to narrow the gap between non-profess...","Title: Emergency Rescue Vehicle Dispatch Planning Using A Hybrid Algorithm\nYear: 2018\n\nAbstract: This paper describes the emergency rescue vehicle transportation network within the entire rescue period, and imitates rescue vehicle to select rescue route and to allocate emergency resource. The presented emergency rescue vehicle dispatch model seeks to minimize rescue time as the first objective function, minimize delay cost as the second objective function and maximize lifesaving utility as the last objective function in disaster response operations. To solve the proposed multiple objective model, a hybrid algorithm named nondominated sorting genetic algorithm (NSGA-II) with ant colony algorithm and a NSGA-II with random crossover and mutation, which can find better initial solution, are presented. In order to further prove the validity of the model and algorithm, a more complicated case is cited. Computational results are reported to illustrate the performance of the proposed model and algorithm. Statistical analysis confirms that the proposed random crossover and mutation operator outperforms the original crossover and mutation operator. The sensitivity analysis proves which parameter is more important for objective function values.","Title: Understanding Crowdsourcing Systems from a Multiagent Perspective and Approach.\nYear: 2018\n\nAbstract: Crowdsourcing has recently been significantly explored. Although related surveys have been conducted regarding this subject, each has mainly consisted of a review of a single aspect of crowdsourcing systems or on the application of crowdsourcing in a specific application domain. A crowdsourcing system is a comprehensive set of multiple entities, including various elements and processes. Multiagent computing has already been widely envisioned as a powerful paradigm for modeling autonomous multi-entity systems with adaptation to dynamic environments. Therefore, this article presents a novel multiagent perspective and approach to understanding crowdsourcing systems, which can be used to correlate the research on crowdsourcing and multiagent systems and inspire possible interdisciplinary research between the two areas. This article mainly discusses the following two aspects: (1) The multiagent perspective can be used for conducting a comprehensive survey on the state of the art of crowdsourcing, and (2) the multiagent approach can bring about concrete enhancements for crowdsourcing technology and inspire future research directions that enable crowdsourcing research to overcome the typical challenges in crowdsourcing technology. Finally, this article discusses the advantages and disadvantages of the multiagent perspective by comparing it with two other popular perspectives on crowdsourcing: the business perspective and the technical perspective.","Title: An Efficient Distributed-Computing Framework for Association-Rule-Based Recommendation\nYear: 2018\n\nAbstract: The association-rule-based recommendation model is one of the most widely used commercial recommendation engines in e-commerce websites. Existing studies mostly focus on how to select eligible rules to enhance the recommendation performance, but the efficiency of recommendation has been paid few attentions. To remedy this, this paper develops a distributed-computing framework for improving the computational efficiency of rule-based recommendation. Specifically, a tree-typed structure called Ordered-Patterns Forest (OPF) is designed to compress and store frequent patterns. Then, we transform eligible rules mining to a path-searching problem on OPF, and present a path-searching algorithm running on single machine. Finally, a load-balanced strategy for data partitioning is clarified. Experimental results demonstrate that the efficiency improved remarkably by the proposed OPF, compared with the traditional Brute-Force method.","Title: Geometry-Aware Face Completion and Editing.\nYear: 2018\n\nAbstract: Face completion is a challenging generation task because it requires generating visually pleasing new pixels that are semantically consistent with the unmasked face region. This paper proposes a geometry-aware Face Completion and Editing NETwork (FCENet) by systematically studying facial geometry from the unmasked region. Firstly, a facial geometry estimator is learned to estimate facial landmark heatmaps and parsing maps from the unmasked face image. Then, an encoder-decoder structure generator serves to complete a face image and disentangle its mask areas conditioned on both the masked face image and the estimated facial geometry images. Besides, since low-rank property exists in manually labeled masks, a low-rank regularization term is imposed on the disentangled masks, enforcing our completion network to manage occlusion area with various shape and size. Furthermore, our network can generate diverse results from the same masked input by modifying estimated facial geometry, which provides a flexible mean to edit the completed face appearance. Extensive experimental results qualitatively and quantitatively demonstrate that our network is able to generate visually pleasing face completion results and edit face attributes as well.","Title: the Algorithm of Warehouse Vehicle Trajectory Intelligent Identification\nYear: 2018\n\nAbstract: Vehicle location is the key in the warehouse vehicle trajectory identification and warning, RFID signal transmission model in a closed warehouse environment in the pseudo range includes not only the RFID signal generated by the NLOS transmission distance of unknown offset, also includes the measurement of actual measurement error exists. Therefore, this paper uses the pseudo range positioning method to realize the vehicle location in the warehouse. In the study of pseudo range positioning method, position information by solving hyperbolic equations to obtain the warehouse of the vehicle, the vehicle positioning contains many constraint optimization problems, and using the distance difference method to reduce the impact of the unknown displacement and the measurement error of vehicle positioning in the process of the warehouse, to improve the positioning accuracy of warehouse vehicles.","Title: Research on Agricultural Environment Prediction Based on Deep Learning\nYear: 2018\n\nAbstract: The environmental security of agriculture is closely related to human beings. Analytical training of agricultural environmental data, forecasting its development trend, has positive significance for the protection of the safety of agricultural products. This paper proposes an agricultural environment prediction model based on deep learning LSTM (Long Short-Term Memory). By analyzing the agricultural environment parameters of the current period, the environmental parameters of the next moment can be predicted to achieve the purpose of early warning. The experimental results show that the model’s prediction results have little deviation from the actual values; on this basis, the LSTM model is optimized to replace LSTM with GRU(Gated Recurrent Unit), and the model is more effective.","Title: Maintenance and Management of Marine Communication and Navigation Equipment Based on Virtual Reality\nYear: 2018\n\nAbstract: According to the present situation and existing problems of marine communication and navigation equipment maintenance and management, this paper expounds the virtual reality technology in equipment maintenance and management. We design and implement a VR based maintenance and management system for marine communication and navigation equipment. The system consists of four parts: VR editing module, VR training module, user behavior analysis module, VR intelligent deduction module and user management module function module. The design and hardware structure of the virtual system is given, and the management task modules on this platform are designed, finally realized the function simulation. Functional verification proved the system can effectively improve the equipment management level of electro-technical officer (ETO).","Title: Deep rolling: A novel emotion prediction model for a multi-participant communication context.\nYear: 2019\n\nAbstract: •Propose a novel emotion prediction model, Deep Rolling, for target participant in a multi-participant communication context. The proposed model can capture the emotional stimulation caused by other participants in the current communication context.•Introduce data factorization into the proposed model, Deep Rolling, to enhance the prediction efficiency.•By adopting movie subtitle as dataset, Deep Rolling is utilized to find roles who have the most important emotional impact on the prediction target.","Title: SSE: A New Selective Initialization Strategy for Snapshot Ensembling\nYear: 2018\n\nAbstract: Deep neural network may suffer large variance in the task of small-sample classification. To overcome this problem, and motivated by that ensemble method is an effective way of reducing variance, this paper proposes a new ensemble method called Selective Snapshot Ensembling, which introduces a new selective initialization strategy into Snapshot Ensembling. Snapshot Ensembling, an ensemble method of neural networks, has been demonstrated its effectiveness on many datasets. The proposed method is able to discard some poor initializations of network parameters, and taking an example of cross-entropy loss, poor initializations here refer to those initializations which cannot make the network converge to a very small value of loss. In particular, we set a threshold of epochs in Snapshot Ensembling to monitor and select good initialization. Experimental results on two small-sample datasets, the LabelMe dataset and the 80 AI-Challeger dataset, demonstrate that the proposed ensemble method outperforms Snapshot Ensembling and other baseline methods.","Title: Knowledge Granularity Based Incremental Attribute Reduction for Hybrid Attribute Data\nYear: 2018\n\nAbstract: Traditional attribute reduction algorithms cannot dynamically update the results of the reduction effectively. The existing incremental attribute reduction algorithm compensates, to a certain extent, the defects of the traditional attribute algorithm, but most of them focused on either categorical attributes data or numerical attributes data simply. In order to process the dynamic data with hybrid attributes, and by combining the advantages of the neighborhood rough sets model that it processes hybrid attributes effectively, a knowledge granularity calculating method based on the neighborhood rough set model is proposed. Through dynamically updating the neighborhood particles to incrementally get the knowledge granularity. According to the knowledge granularity, an incremental reduction algorithm is further developed. The validity of the proposed algorithm is demonstrated by experimental analysis.","Title: Simulation of Continuous Enterprise Energy Consumption Process with Uncertain Factors\nYear: 2018\n\nAbstract: Continuous enterprise energy consumption process is a hybrid system with diverse structure such as parallel, serial and circuitous connection etc., and influenced by various uncertain factors. To analyze the energy consuming situation more accurately, based on the model of continuous enterprise energy consumption process with uncertain factors, the simulation method for continuous enterprise energy consumption process with uncertain factors was presented, and the simulation algorithm, in which the steady running range and the instantaneous firing speed can change dynamically, was proposed in detail. The effectiveness and superiority of this method were illustrated by simulation results.","Title: Compressed binary discernibility matrix based incremental attribute reduction algorithm for group dynamic data.\nYear: 2019\n\nAbstract: The datasets in real-world applications often vary dynamically over time. Moreover, datasets often expand by introducing a group of data in many cases rather than a single object one by one. The traditional incremental attribute reduction approaches for a single dynamic object may not be applied to such cases. Focusing on this issue, a compressed binary discernibility matrix is introduced and an incremental attribute reduction algorithm for group dynamic data is developed. The single dynamic object and the group dynamic objects are both considered in this algorithm. According to the dynamic data is a single object or a group of objects, different branches can be chosen to update the compressed binary discernibility matrix. Thereafter, the incremental reduction result can be obtained based on the updated compressed binary discernibility matrix. The validity of this algorithm is demonstrated by simulation and experimental analysis.","Title: Large-Margin Regularized Softmax Cross-Entropy Loss.\nYear: 2019\n\nAbstract: Softmax cross-entropy loss with L2 regularization is commonly adopted in the machine learning and neural network community. Considering that the traditional softmax cross-entropy loss simply focuses on fitting or classifying the training data accurately but does not explicitly encourage a large decision margin for classification, some loss functions are proposed to improve the generalization performance by solving the problem. However, these loss functions enhance the difficulty of model optimization. In addition, inspired by regularized logistic regression, where the regularized term is responsible for adjusting the width of decision margin, which can be seen as an approximation of support vector machine, we proposed a large-margin regularization method for softmax cross-entropy loss. The advantages of the proposed loss are twofold as follows: the first is the generalization performance improvement, and the second is easy optimization. The experimental results on three small-sample datasets show that our regularization method achieves good performance and outperforms the existing popular regularization methods of neural networks.","Title: Cross-spectral Face Completion for NIR-VIS Heterogeneous Face Recognition.\nYear: 2019\n\nAbstract: Near infrared-visible (NIR-VIS) heterogeneous face recognition refers to the process of matching NIR to VIS face images. Current heterogeneous methods try to extend VIS face recognition methods to the NIR spectrum by synthesizing VIS images from NIR images. However, due to self-occlusion and sensing gap, NIR face images lose some visible lighting contents so that they are always incomplete compared to VIS face images. This paper models high resolution heterogeneous face synthesis as a complementary combination of two components, a texture inpainting component and pose correction component. The inpainting component synthesizes and inpaints VIS image textures from NIR image textures. The correction component maps any pose in NIR images to a frontal pose in VIS images, resulting in paired NIR and VIS textures. A warping procedure is developed to integrate the two components into an end-to-end deep network. A fine-grained discriminator and a wavelet-based discriminator are designed to supervise intra-class variance and visual quality respectively. One UV loss, two adversarial losses and one pixel loss are imposed to ensure synthesis results. We demonstrate that by attaching the correction component, we can simplify heterogeneous face synthesis from one-to-many unpaired image translation to one-to-one paired image translation, and minimize spectral and pose discrepancy during heterogeneous recognition. Extensive experimental results show that our network not only generates high-resolution VIS face images and but also facilitates the accuracy improvement of heterogeneous face recognition.","Title: Biphasic Learning of GANs for High-Resolution Image-to-Image Translation.\nYear: 2019\n\nAbstract: Despite that the performance of image-to-image translation has been significantly improved by recent progress in generative models, current methods still suffer from severe degradation in training stability and sample quality when applied to the high-resolution situation. In this work, we present a novel training framework for GANs, namely biphasic learning, to achieve image-to-image translation in multiple visual domains at $1024^2$ resolution. Our core idea is to design an adjustable objective function that varies across training phases. Within the biphasic learning framework, we propose a novel inherited adversarial loss to achieve the enhancement of model capacity and stabilize the training phase transition. Furthermore, we introduce a perceptual-level consistency loss through mutual information estimation and maximization. To verify the superiority of the proposed method, we apply it to a wide range of face-related synthesis tasks and conduct experiments on multiple large-scale datasets. Through comprehensive quantitative analyses, we demonstrate that our method significantly outperforms existing methods.","Title: Dual Cross-Entropy Loss for Small-Sample Fine-Grained Vehicle Classification\nYear: 2019\n\nAbstract: Fine-grained vehicle classification is a challenging topic in computer vision due to the high intraclass variance and low interclass variance. Recently, considerable progress has been made in fine-grained vehicle classification due to the huge success of deep neural networks. Most studies of fine-grained vehicle classification based on neural networks, focus on the neural network structure to improve the classification performance. In contrast to existing works on fine-grained vehicle classification, we focus on the loss function of the neural network. We add a regularization term to the cross-entropy loss and propose a new loss function, \n<italic xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">Dual Cross-Entropy Loss</italic>\n. The regularization term places a constraint on the probability that a data point is assigned to a class other than its ground-truth class, which can alleviate the vanishing of the gradient when the value of the cross-entropy loss is close to zero. To demonstrate the effectiveness of our loss function, we perform two sets of experiments. The first set is conducted on a small-sample fine-grained vehicle classification dataset, the Stanford Cars-196 dataset. The second set is conducted on two small-sample datasets, the LabelMe dataset and the UIUC-Sports dataset, as well as on one large-sample dataset, the CIFAR-10 dataset. The experimental results show that the proposed loss function improves the fine-grained vehicle classification performance and has good performance on three other general image classification tasks.","Title: 3D Aided Duet GANs for Multi-view Face Image Synthesis\nYear: 2019\n\nAbstract: Multi-view face synthesis from a single image is an ill-posed computer vision problem. It often suffers from appearance distortions if it is not well-defined. Producing photo-realistic and identity preserving multi-view results is still a not well-defined synthesis problem. This paper proposes 3D aided duet generative adversarial networks (AD-GAN) to precisely rotate the yaw angle of an input face...","Title: Batch Allocation for Tasks with Overlapping Skill Requirements in Crowdsourcing\nYear: 2019\n\nAbstract: Existing studies on crowdsourcing often adopt the retail-style allocation approach, in which tasks are allocated individually and independently. However, such retail-style task allocation has the following problems: 1) each task is executed independently from scratch, thus the execution of one task seldom utilize the results of other tasks and the requester must pay in full for the task; 2) many workers only undertake a very small number of tasks contemporaneously, thus the workers’ skills and time may not be fully utilized. We observe that many complex tasks in real-world crowdsourcing platforms have similar skill requirements and long deadlines. Based on these real-world observations, this paper presents a novel batch allocation approach for tasks with overlapping skill requirements. Requesters’ real payment can be discounted because the real execution cost of tasks can be reduced due to batch allocation and execution, and each worker's real earnings may increase because he/she can undertake more tasks contemporaneously. This batch allocation optimization problem is proved to be NP-hard. Then, two types of heuristic approaches are designed: layered batch allocation and core-based batch allocation. The former approach mainly utilizes the hierarchy pattern to form all possible batches, which can achieve better performance but may require higher computational cost since all possible batches are formed and observed; the latter approach selects core tasks to form batches, which can achieve suboptimal performance with lower complexity and significantly reduce computational cost. With the theoretical analyses and experiments on a real-world Upwork dataset in which the proposed approaches are compared with the previous benchmark retail-style allocation approach, we find that our approaches have better performances in terms of total payment by requesters and average income of workers, as well as maintaining close successful task completion probability and consuming less task allocation time.","Title: Effects of pore structure and salinity on the imbibition of shale samples using physical simulation and NMR technique: A case from Chang 7 shale, Ordos basin\nYear: 2021\n\nAbstract: AbstractA novel experimental study is conducted to reveal the effect of pore structure and salinity on spontaneous imbibition of shale samples using field emission scanning electron microscopy and low-field nuclear magnetic resonance (NMR). Core samples selected from Chang 7 shale, Ordos basin, contain different proportions of mineral components and exhibit different pore structures, as demonstrated by scanning electron microscopy. During the imbibition process, T2 spectra are obtained for each core sample under different salinities and imbibition time periods using NMR. The results are then used to characterize and compare the water imbibition process in different shale samples, with a focus on investigating the effects of pore structure and salinity. The results show that the amount of absorption water is positively correlated with the development of microcracks and with large pores in shale samples. The shale samples are shown to develop more microcracks as the imbibition time increases and the salinity decreases. These microcracks can improve the samples’ capacity of imbibition effectively, therefore increasing water absorption. Moreover, it is indicated that the imbibition capacity increases with an increasing content of illite clay minerals in the shale samples. Furthermore, the greatest amount of adsorbed water is seen when the water salinity is as high as the synthetic formation water. This is important in choosing hydraulic fracturing fluids, knowing that high-salinity fracturing fluids can effectively reduce water absorption in shale, therefore contributing to low flow-back and better fracturing performance.","Title: Effects of pore structure and salinity on the imbibition of shale samples using physical simulation and NMR technique: A case from Chang 7 shale, Ordos basin\nYear: 2021\n\nAbstract: AbstractA novel experimental study is conducted to reveal the effect of pore structure and salinity on spontaneous imbibition of shale samples using field emission scanning electron microscopy and low-field nuclear magnetic resonance (NMR). Core samples selected from Chang 7 shale, Ordos basin, contain different proportions of mineral components and exhibit different pore structures, as demonstrated by scanning electron microscopy. During the imbibition process, T2 spectra are obtained for each core sample under different salinities and imbibition time periods using NMR. The results are then used to characterize and compare the water imbibition process in different shale samples, with a focus on investigating the effects of pore structure and salinity. The results show that the amount of absorption water is positively correlated with the development of microcracks and with large pores in shale samples. The shale samples are shown to develop more microcracks as the imbibition time increases and the salinity decreases. These microcracks can improve the samples’ capacity of imbibition effectively, therefore increasing water absorption. Moreover, it is indicated that the imbibition capacity increases with an increasing content of illite clay minerals in the shale samples. Furthermore, the greatest amount of adsorbed water is seen when the water salinity is as high as the synthetic formation water. This is important in choosing hydraulic fracturing fluids, knowing that high-salinity fracturing fluids can effectively reduce water absorption in shale, therefore contributing to low flow-back and better fracturing performance.","Title: Detecting Prosumer-Community Groups in Smart Grids From the Multiagent Perspective\nYear: 2019\n\nAbstract: One of the greatest advancements of the modern era is the evolution of smart grid (SG), which integrates information communication technologies with advanced power electronic technologies to cope with the global energy shortage. The users in SGs are often called the “prosumers,” who not only consume energy but also generate the energy and share it with the utility grid or with other energy consumers. In order to promote sustainable prosumer management in SGs, one of the feasible strategies is to aggregate the prosumers from different locations, but with similar energy behaviors and cohesive interconnections, such groups of prosumers are also called the prosumer-community groups (PCGs). The contribution of this paper is threefold. First, we provide a generalized definition of individual prosumer’s energy density, which can be used to detect the underlying leader prosumers in SGs. Second, we formulate the PCG detection (PCG-D) as a multiobjective optimization problem, and present a novel dynamic game model to find the locally Pareto-optimal PCG structure. Third, we propose a partially visible multiagent system (PVMAS), where the viewing angles of both prosumers and PCGs are mutually restricted. The significance of our PVMAS is that it can nicely lead itself to parallelization for PCG-D, due to the fact that the feature updating of each agent is independent of each other. We conduct a series of comprehensive experiments on the simulated SG datasets to validate the performance of PVMAS through comparing it with existing community detection approaches in the literature.","Title: Graph K-means based on Leader Identification, Dynamic Game and Opinion Dynamics\nYear: 2020\n\nAbstract: With the explosion of social media networks, many modern applications are concerning about people's connections, which leads to the so-called social computing. An elusive question is to study how opinion communities form and evolve in real-world networks with great individual diversity and complex human connections. In this scenario, the classic K-means technique and its extended versions could no...","Title: Link prediction in temporal networks: Integrating survival analysis and game theory.\nYear: 2019\n\nAbstract: Link prediction is an important task in complex network analysis and can be found in many real-world applications such as recommendation systems, information retrieval, and marketing analysis of social networks. This paper focuses on studying the evolution mechanism of real-world temporal networks. Specifically, given a set of temporal links during a fixed time window, how to predict the existence of links at any point in the future. To address this problem, we propose a novel semi-supervised learning framework, which integrates both survival analysis and game theory. First, we carefully define the ϵ-adjacent network sequence, and make use of time stamp on each link to generate the baseline network evolution sequence. Next, to capture the law of network evolution, we employ the Cox Proportional Hazard Model (Cox PHM) to study the relative hazard associated with each temporal link, so as to estimate the coefficients of covariates, which are defined as a set of neighborhood based proximity features. To narrow the area of inquiry, we further propose a game theory based two-way selection mechanism to predict the future network topology. We finally encapsulate these two new technologies in a robust Autonomy-Oriented-Computing (AOC) multi-agent system, and propose a paralleled algorithm to conduct the temporal link prediction task. Extensive experiments were applied to real-world temporal networks to demonstrate both effectiveness and scalability of the proposed approach.","Title: Attribute-Based Information Flow Control\nYear: 2019\n\nAbstract: Information flow control (IFC) regulates where information is permitted to travel within information systems. To enforce IFC, access control encryption (ACE) was proposed to support both the no read-up rule and the no write-down rule. There are some problems in existing schemes. First, the communication cost is linear with the number of receivers. Second, senders are not authenticated, namely an unauthorized sender can send a message to a receiver. To reduce communication cost and implement sender authentication, we propose an attribute-based IFC (ABIFC) scheme by introducing attribute-based systems into IFC. Our ABIFC scheme captures the following features: (i) flexible IFC policies are defined over a universal set of descriptive attributes; (ii) both the no read-up rule and the no write-down rule are supported; (iii) the communication cost is linear with the number of required attributes, instead of receivers; (iv) receivers can outsource heavy computation to a server without compromising data confidentiality; (v) authorized senders can control release their attributes when sending messages to receivers. To the best of our knowledge, it is the first IFC scheme where flexible policies are defined over descriptive attributes and outsourced computation is supported.","Title: Integrated planning for public health emergencies: A modified model for controlling H1N1 pandemic\nYear: 2020\n\nAbstract: Infectious disease outbreaks have occurred many times in the past decades and are more likely to occur in the future. Recently, Buyuktahtakin et al. (2018) proposed a new epidemics-logistics model to control the 2014 Ebola outbreak in West Africa. Considering that different diseases have dissimilar diffusion dynamics and can cause different public health emergencies, we modify the proposed model by changing capacity constraint, and then apply it to control the 2009 H1N1 outbreak in China. We formulate the problem to be a mixed-integer non-linear programming model (MINLP) and simultaneously determine when to open the new isolated wards and when to close the unused isolated wards. The test results reveal that our model could provide effective suggestions for controlling the H1N1 outbreak, including the appropriate capacity setting and the minimum budget required with different intervention start times.","Title: Accurate Human Gesture Sensing With Coarse-Grained RF Signatures.\nYear: 2019\n\nAbstract: RF-based gesture sensing and recognition has increasingly attracted intense academic and industrial interest due to its various device-free applications in daily life, such as elder monitoring, mobile games. State-of-the-art approaches achieved accurate gesture sensing by using fine-grained RF signatures (such as CSI, Doppler effect) while could not achieve the same accuracy with coarse-grained RF signatures such as received signal strength (RSS). This paper presents rRuler, a novel feature extraction method which aims to get fine-grained human gesture features with coarse-grained RSS readings, which means rought ruler could measure fine things. In order to further verify the performance of rRuler, we further propose rRuler-HMM, a hidden Markov model (HMM) based human gesture sensing and prediction algorithm which utilizes the features extracted by rRuler as input. We implemented rRuler and rRuler-HMM using TI Sensortag platforms and off-the-shelf (CTOS) laptops in an indoor environment, extensively performance evaluations show that rRuler and rRuler-HMM stand out for their low cost and high practicability, and the average gesture sensing accuracy of rRuler-HMM can achieve 95.71% in NLoS scenario and 97.14% in LoS scenario, respectively, which is similar to the performance that fine-grained RF signatures based approaches could achieve.","Title: Robust Task Grouping with Representative Tasks for Clustered Multi-Task Learning\nYear: 2019\n\nAbstract: Multi-task learning aims to learn multiple tasks jointly by sharing information among related tasks such that the generalization performance over different tasks could be improved. Although multi-task learning has been demonstrated to obtain performance gain in comparison with the single task learning, the main challenge that learning what to share with whom is still not fully resolved. In this paper, we propose a robust clustered multi-task learning approach that clusters tasks into several groups by learning the representative tasks. The main assumption behind our approach is that each task can be represented by a linear combination of some representative tasks that can characterize all tasks. The correlation between tasks can be indicated by the corresponding combination coefficient. By imposing a row-sparse constraint on the correlation matrix, our approach could select the representative tasks and encourage information sharing among the related tasks. In addition, the $l_1,2 $-norm is applied to the representation loss to enhance the robustness of our approach. To solve the resulting bi-convex optimization problem, we design an efficient optimization method based on the alternating direction method of multipliers and accelerated proximal gradient method. Finally, experimental results on synthetic and real-world data sets validate the effectiveness of the proposed approach.","Title: Integrating Bi-Dynamic Routing Capsule Network with Label-Constraint for Text classification\nYear: 2020\n\nAbstract: Neural-based text classification methods have attracted increasing attention in recent years. Unlike the standard text classification methods, neural-based text classification methods perform the representation operation and end-to-end learning on the text data. Many useful insights can be derived from neural based text classifiers as demonstrated by an ever-growing body of work focused on text mining. However, in the real-world, text can be both complex and noisy which can pose a problem for effective text classification. An effective way to deal with this issue is to incorporate self-attention and capsule networks into text mining solutions. In this paper, we propose a Bi-dynamic routing Capsule Network with Label-constraint (BCNL) model for text classification, which moves beyond the limitations of previous methods by automatically learning the task-relevant and label-relevant words of text. Specifically, we use a Bi-LSTM and self-attention with position encoder network to learn text embeddings. Meanwhile, we propose a bi-dynamic routing capsule network with label-constraint to adjust the category distribute of text capsules. Through extensive experiments on four datasets, we observe that our method outperforms state-of-the-art baseline methods.","Title: Random compact Gaussian kernel: Application to ELM classification and regression\nYear: 2021\n\nAbstract: Extreme learning machine (ELM) kernels using random feature mapping have recently gained a lot of popularity because they require low human supervision. However, the superiority in the mapping mechanism of ELM kernels is accompanied by a higher computation cost, rendering the kernel learning algorithms hard to tackle large scale learning tasks. On the other hand, the implicit mapping used in the conventional Gaussian kernel is computationally cheaper than the explicit computation of the ELM kernel, but requires trivial human intervention. This paper proposes to merge both properties by defining a new kernel, the random compact Gaussian (RCG) kernel. The random feature mapping property enables RCG kernel to save parameters selection time, while the implicitly mapping property enables RCG kernel to save kernel calculation time. The proposed kernel works by scaling one kernel parameter in the conventional Gaussian kernel to multiple kernel parameters, and generating all parameters randomly based on a continuous probability distribution. We prove that the RCG kernel is a Mercer kernel. The kernel is calculated implicitly before seeing the training samples and used to train ELMs. The experiments on 25 binary classification and regression benchmark problems show that the RCG kernel typically outperforms other competitive kernels. Compared to ELM kernel, RCG kernel not only achieves the better generalization performance on most datasets, but also needs much less kernel calculation cost. In addition, the sensitivity analysis of the kernel parameters of k-fold cross validation is conducted and the results show that the RCG kernel is robust and stable for repeated trials.","Title: A Comparative Study on Schema-Guided Dialogue State Tracking\nYear: 2021\n\nAbstract: Frame-based state representation is widely used in modern task-oriented dialog systems to model user intentions and slot values. However, a fixed design of domain ontology makes it difficult to extend to new services and APIs. Recent work proposed to use natural language descriptions to define the domain ontology instead of tag names for each intent or slot, thus offering a dynamic set of schema. In this paper, we conduct in-depth comparative studies to understand the use of natural language description for schema in dialog state tracking. Our discussion mainly covers three aspects: encoder architectures, impact of supplementary training, and effective schema description styles. We introduce a set of newly designed bench-marking descriptions and reveal the model robustness on both homogeneous and heterogeneous description styles in training and evaluation.","Title: Complementary Fourier Single-Pixel Imaging\nYear: 2021\n\nAbstract: Single-pixel imaging, with the advantages of a wide spectrum, beyond-visual-field imaging, and robustness to light scattering, has attracted increasing attention in recent years. Fourier single-pixel imaging (FSI) can reconstruct sharp images under sub-Nyquist sampling. However, the conventional FSI has difficulty balancing imaging quality and efficiency. To overcome this issue, we proposed a novel approach called complementary Fourier single-pixel imaging (CFSI) to reduce the number of measurements while retaining its robustness. The complementary nature of Fourier patterns based on a four-step phase-shift algorithm is combined with the complementary nature of a digital micromirror device. CFSI only requires two phase-shifted patterns to obtain one Fourier spectral value. Four light intensity values are obtained by loading the two patterns, and the spectral value is calculated through differential measurement, which has good robustness to noise. The proposed method is verified by simulations and experiments compared with FSI based on two-, three-, and four-step phase shift algorithms. CFSI performed better than the other methods under the condition that the best imaging quality of CFSI is not reached. The reported technique provides an alternative approach to realize real-time and high-quality imaging.","Title: An Efficient Alpha Seeding Method For Optimized Extreme Learning Machine-Based Feature Selection Algorithm\nYear: 2021\n\nAbstract: Embedded feature selection algorithms, such as support vector machine based recursive feature elimination (SVM-RFE), have proven to be effective for many real applications. However, due to the model selection problem, SVM-RFE naturally suffers from a heavy computational burden as well as high computational complexity. To solve these issues, this paper proposes using an optimized extreme learning machine (OELM) model instead of SVM. This model, referred to as OELM-RFE provides an efficient active set solver for training the OELM algorithm. We also present an effective alpha seeding algorithm to efficiently solve successive quadratic programming (QP) problems inherent in OELM. One of the salient characteristics of OELM-RFE is that it has only one tuning parameter: the penalty constant C. Experimental results from work on benchmark datasets show that OELM-RFE tends to have higher prediction accuracy than SVM-RFE, and requires fewer model selection efforts. In addition, the alpha seeding method works better on more datasets.","Title: Group-Oriented Task Allocation for Crowdsourcing in Social Networks\nYear: 2021\n\nAbstract: Previous crowdsourcing studies often adopted the individual-oriented approach that outsources a task to an individual worker or team formation-based approach that outsources a task to an artificially formed team of workers. Nowadays, workers are often naturally organized into groups through social networks. To address such common issue of grouped workers in real crowdsourcing systems, this article...","Title: FaceInpainter: High Fidelity Face Adaptation to Heterogeneous Domains\nYear: 2021\n\nAbstract: In this work, we propose a novel two-stage framework named Facelnpainter to implement controllable Identity-Guided Face Inpainting (IGFI) under heterogeneous domains. Concretely, by explicitly disentangling foreground and background of the target face, the first stage focuses on adaptive face fitting to the fixed background via a Styled Face Inpainting Network (SFI-Net), with 3D priors and texture code of the target, as well as identity factor of the source face. It is challenging to deal with the inconsistency between the new identity of the source and the original background of the target, concerning the face shape and appearance on the fused boundary. The second stage consists of a Joint Refinement Network (JR-Net) to refine the swapped face. It leverages AdaIN considering identity and multi-scale texture codes, for feature transformation of the decoded face from SFI-Net with facial occlusions. We adopt the contextual loss to implicitly preserve the attributes, encouraging face deformation and fewer texture distortions. Experimental results demonstrate that our approach handles high-quality identity adaptation to heterogeneous domains, exhibiting the competitive performance compared with state-of-the-art methods concerning both attribute and identity fidelity.","Title: A hierarchical knowledge guided backtracking search algorithm with self-learning strategy\nYear: 2021\n\nAbstract: To improve the performance of the backtracking search optimization algorithm (BSA), a multi-population cooperative evolution strategy guided BSA with hierarchical knowledge (HKBSA) is proposed in this paper. According to the domain knowledge of the candidates in objective space, the population is divided into the dominant population, the ordinary population and the inferior population. The information between the sub-populations has interacted with the evolution processes of the three sub-populations. The individuals in the dominant population are maintained as the optimal solutions and are utilized to guide the evolution of the other two sub-populations. A multi-strategy mutation mechanism is applied to solve non-separable problems. The distribution vector of inferior individuals is constructed by sampling, and a mechanism of the individual generation with feedback is proposed by combining self-learning strategy and elite learning strategy. The convergence of HKBSA is analyzed with the Markov model. Compared with the state-of-the-art BSA variants, HKBSA outperforms other algorithms in terms of the speed of convergence, solution accuracy and stability.","Title: Reinforcing Pretrained Models for Generating Attractive Text Advertisements\nYear: 2021\n\nAbstract: ABSTRACTWe study how pretrained language models can be enhanced by using deep reinforcement learning to generate attractive text advertisements that reach the high quality standard of real-world advertiser mediums. To improve ad attractiveness without hampering user experience, we propose a model-based reinforcement learning framework for text ad generation, which constructs a model for the environment dynamics and avoids large sample complexity. Based on the framework, we develop Masked-Sequence Policy Gradient, a reinforcement learning algorithm that integrates efficiently with pretrained models and explores the action space effectively. Our method has been deployed to production in Microsoft Bing. Automatic offline experiments, human evaluation, and online experiments demonstrate the superior performance of our method.","Title: Energy-Aware Offloading Based On Priority In Mobile Cloud Computing\nYear: 2021\n\nAbstract: Smartphones and portable devices have been widely used in our daily life. However, these portable devices cannot be used in a lot of environments due to limitations in battery capacity and computing speed. Therefore, how to save energy consumption and improve processing ability has become a hot issue. The previous works were mainly to transfer some local tasks to remote compute nodes by intelligently selecting the execution route of tasks based on Directed Acyclic Graph (DAG). Different from previous work, this work mainly attempts to consider task urgency and system load based on Dynamic Voltage and Frequency Scaling (DVFS) model. Firstly, tasks are given different priorities according to the energy consumption of different locations and task urgency. Secondly, a reference working state is given according to system load to ensure that each task (regardless of job deadline) is completed with minimum energy consumption under the average system load. Finally, a heuristic algorithm is used to schedule tasks on mobile devices and remote cloud resources based on the priority of tasks and reference status of resources. Our method is called AODVFS-Adaptive Offloading based on DVFS model. Comparison with other methods shows that our method saves average energy consumption and increases the number of completed tasks.","Title: Can AoI and Delay be Minimized Simultaneously with Short-Packet Transmission?\nYear: 2021\n\nAbstract: Age of information (AoI) and delay are both critical performance metrics to enable emerging applications such as remote estimation and real-time control in factory automation. In particular, it remains an open question whether AoI and delay can be minimized simultaneously in short-packet transmission for factory automation etc., as they are affected by block length and update rate in a complex man...","Title: Global Relation-Aware Attention Network for Image-Text Retrieval\nYear: 2021\n\nAbstract: ABSTRACTThe cross-modal image-text retrieval has attracted extensive attention in recent years, which contributes to the development of search engine. Fine-grained features and cross-attention have been widely used in past researches to reach the goal of cross-modal image-text matching. Although cross-related methods have achieved remarkable results, the features must be encoded again in evaluation phase due to the interaction of the two modalities, which is unsuitable for actual scenarios of search engine development. In addition, the aggregated feature does not contain sufficient semantics since it is merely obtained by simple mean pooling. Furthermore, connecting weights of self-attention blocks are target position invariant, which lacks the expected adaptability. To tackle these limitations, in this paper, we propose a novel Global Relation-aware Attention Network (GRAN) for image-text retrieval by designing Global Attention Module (GAM) and Relation-aware Attention Module (RAM) which play an important role in modeling the global feature and the relationships of local fragments. Firstly, we propose Global Attention Module (GAM) followed the fine-grained features to obtain meaningful global feature. Secondly, we use several stacked transformer encoders to further encode features separately. Finally, we propose Relation-aware Attention Module (RAM) to generate a vector which represents the relation information to infer the attention intensity of pairwise fragments. The local features, the global feature, and their relations are considered jointly to conduct an efficient image-text retrieval. Extensive experiments are conducted on the benchmark datasets of Flickr30K and MSCOCO, demonstrating the superiority of our method. On the Flickr30K, compared to the state-of-the-art method TERAN, we improve [email protected](K=1) metric by 5.8% and 4.0 on the image and text retrieval tasks, respectively.","Title: Measuring the Network Vulnerability Based on Markov Criticality\nYear: 2022\n\nAbstract: AbstractVulnerability assessment—a critical issue for networks—attempts to foresee unexpected destructive events or hostile attacks in the whole system. In this article, we consider a new Markov global connectivity metric—Kemeny constant, and take its derivative called Markov criticality to identify critical links. Markov criticality allows us to find links that are most influential on the derivative of Kemeny constant. Thus, we can utilize it to identity a critical link (i, j) from node i to node j, such that removing it leads to a minimization of networks’ global connectivity, i.e., the Kemeny constant. Furthermore, we also define a novel vulnerability index to measure the average speed by which we can disconnect a specified ratio of links with network decomposition. Our method is of high efficiency, which can be easily employed to calculate the Markov criticality in real-life networks. Comprehensive experiments on several synthetic and real-life networks have demonstrated our method’s better performance by comparing it with state-of-the-art baseline approaches.","Title: A hierarchical guidance strategy assisted fruit fly optimization algorithm with cooperative learning mechanism\nYear: 2021\n\nAbstract: •A novel cooperative learning fruit fly optimization algorithm (HGCLFOA) is proposed.•The hierarchical guidance strategy for local search is implemented in the olfactory search.•The inferior solution repairing (ISR) strategy is employed to modify the search direction.•The hybrid GEDA with previous information is applied to guide the evolution in vision search.•A probability selection strategy based on feedback of objective space is introduced.","Title: A two-stage evolutionary strategy based MOEA/D to multi-objective problems\nYear: 2021\n\nAbstract: •A two-stage evolution strategy is proposed for solving multi-objective problem.•The convergence and diversity should be balanced in multi-objective optimization.•A local searching method can improve the diversity of solutions in the space.•Experimental results have been presented by using statistical method.","Title: Learning to traverse over graphs with a Monte Carlo tree search-based self-play framework\nYear: 2021\n\nAbstract: The combinatorial optimization (CO) problems on the graph are the core and classic problems in artificial intelligence (AI) and operations research (OR). For example, the Vehicle Routing Problem (VRP) and Traveling Salesman Problem (TSP) are fascinating NP-hard problems and have important significance for the existing transportation system. Traditional methods such as heuristics methods, exact algorithms, and solution solvers can already find approximate solutions on small-scale graphs. However, they are helpless for large-scale graphs and other problems with similar structures. Moreover, traditional methods often require artificially designed heuristic functions to aid decision-making. In recent years, more and more work has focused on applying deep learning and reinforcement learning (RL) to learn heuristics, which allows us to learn the internal structure of the graph end-to-end and find the optimal path under the guidance of heuristic rules. However, most of these still need manual assistance, and the RL method used has the problems of low sampling efficiency and small searchable space. This paper proposes a novel framework (called OmegaZero) based on Alphago Zero, which does not prescribe expert experience or label data but is trained through self-play. We divide the learning into two stages: in the first stage, we employ graph attention network (GAT) and GRU to learn node representations and memory history trajectories. In the second stage, we employ Monte Carlo tree search (MCTS) and deep RL to search for the solution space and train the model.","Title: A Comparison Of The Different Stages Of Dust Events Over Beijing In March 2021: The Effects Of The Vertical Structure On Near-Surface Particle Concentration\nYear: 2021\n\nAbstract: Mineral dust is of great importance to climate change, air quality, and human health. In this study, multisource data, including the reanalysis data and remote sensing data, were used to compare the three dust events that occurred in the March of 2021 over Beijing and reveal the effects of atmospheric vertical structure on near-surface dust concentration. The combined effect of the Mongolian cyclone and a wide persistent cold-front induced two events (E1: from March 15 to 16 and E3: from March 28 to 29). E1 was more intense, more extensive, and longer-lasting than E3 due to the combination of the stronger Mongolian cyclone, slower high/cold surface pressure, and the low-level jet. However, under the appropriate configurations of temperature and pressure fields between high and low altitudes, weak updrafts were still induced and could elevate dust up to 850 hPa, as occurred during E2 on March 22 and 23. The dust emission was inferior to E1 and E3, which contributes to the low dust concentration near the surface in E2. On the other hand, the downdraft strength directly affected both the vertical distribution of dust and the concentration of surface particles. There was a strong temporal consistency between the occurrence of the downdraft and the dust touchdown. In E1, the continuous strong downdraft caused the maximum dust concentration to be above 4000 mu g/m(3) at around 200 m. In contrast, the maximum height of the dust mass concentration in E3 occurred at about 800 m due to the transient downdraft, which weakened its effect on surface visibility. Besides, the weak vertical motion in E2 caused most of the dust to become suspended in the air. Overall, the large dust emission resulted from active updrafts in the source region, and the lengthy strong downdrafts led to the ultrahigh particle concentration near the surface.","Title: A multi-task learning approach for improving travel recommendation with keywords generation\nYear: 2021\n\nAbstract: Travel recommendation is very critical to helping users quickly find products or services that they are interested in. The key to travel recommender systems is learning user shopping intentions, which are expressed through various supervision signals, such as the clicked products and their titles. Existing travel recommendation methods commonly infer user intentions from click behaviors on travel products. However, remarkable keywords in the product title, such as departure, destination, travel time, hotel, and transportation are paid less attention. To this end, we hypothesize that modeling click sequences and product keywords in title jointly would result in a more holistic representation of a product and towards more accurate recommendations. Thus, we propose a TRKG (short for Travel Recommendation with Keywords Generation) model, which fulfills the travel recommendation and keywords generation tasks simultaneously. To generate explainable outputs, unlike most previous approaches that regard the product title as a hidden feature vector, TRKG regards keywords in the product title as an additional supervision signal. Meanwhile, TRKG integrates the long-term and short-term user preferences in the travel recommendation component and the keywords generation component. To evaluate the proposed model, we constructed datasets from a large tourism e-commerce website in China. Extensive experiments demonstrate that the proposed method yields significant improvements over state-of-the-art methods.","Title: Information Age-Delay Correlation and Optimization With Finite Block Length\nYear: 2021\n\nAbstract: Both information age and delay are critical performance metrics of emerging time-sensitive applications. However, their inherent correlation in the finite block length (FBL) regime has remained uninvestigated as it is affected by block length and update rate in a complex manner. In this paper, closed-form expressions for average age of information (AoI), peak AoI (PAoI) and delay are derived for a...","Title: Toward a Green Secure Relay System for Mission-Critical IoT: Hybrid Duplex Relay Selection and Resource Allocation in the Finite Block Length Regime\nYear: 2021\n\nAbstract: In this paper, we investigate the physical layer security of a relay system toward secure and green mission-critical Internet of Things, in the finite block length regime, which was hardly considered in the literature. Closed-form expressions for statistical secrecy throughput of full-duplex (FD) and half-duplex (HD) relay systems are respectively derived and verified by simulation results. To max...","Title: Candidate Box Fusion Based Approach To Adjust Position Of The Candidate Box For Object Detection\nYear: 2021\n\nAbstract: The method of object detection has been applied to all aspects in our lives. Although object detection methods based on deep learning have been widely used in various fields, there are still some overlooked problems in the candidate box selection stage. The detection results of traditional candidate box selection methods can only select a relatively optimal maximum candidate box. If the maximum candidate box is still not accurate enough, this type of methods will not be able to do adjust it. To solve this problem, an object detection method based on the multiple candidate box fusion is proposed. The method can not only retain the maximum candidate box and delete the non-maximum candidate box, but also adjust the position of the maximum candidate box again. Thereby a more accurate maximum candidate box can be obtained. In order to verify the generalization ability of the method, the candidate box fusion method is combined with the two object detection frameworks: faster R-CNN model and YOLOv3 model. The results of these experiments prove that the proposed method can achieve higher detection accuracy and complete the object detection task more effectively.","Title: Single Haze Image Restoration Under Non-Uniform Dense Scattering Media\nYear: 2021\n\nAbstract: We demonstrate an efficient restoration method based on sky segmentation and block optimized transmission (SSBOT) to restore the visibility of single haze image under non-uniform scattering medium. The SSBOT method uses sky region segmentation to determine the haze-opaque region, and estimates the global atmospheric light using quadtree-decomposition. The proposed SSBOT estimates the transmission in every sub-images of the degraded image with dark channel method with coefficient correction. The experimental results show that SSBOT can efficiently restore images degraded by non-uniform dense scattering medium. The proposed method can be used for various security and surveillance applications.","Title: The Eulerian Distribution on the Involutions of the Hyperoctahedral Group is Indeed gamma-Positive\nYear: 2021\n\nAbstract: Let I-n(B) denote the set of the involutions of the hyperoctahedral group B-n, and let des(B)(pi) denote the number of descents of the permutation pi is an element of B-n. We settle a problem of Moustakas which states that I-n(B)(t) := Sigma(pi is an element of InB) t(desB(pi)) is gamma-positive for n >= 1.","Title: Improving Subgraph Recognition with Variational Graph Information Bottleneck\nYear: 2022\n\nAbstract: Subgraph recognition aims at discovering a compressed substructure of a graph that is most informative to the graph property. It can be formulated by optimizing Graph Information Bottleneck (GIB) with a mutual information estimator. However, GIB suffers from training instability and degenerated results due to its intrinsic optimization process. To tackle these issues, we reformulate the subgraph recognition problem into two steps: graph perturbation and subgraph selection, leading to a novel Variational Graph Information Bottleneck (VGIB) framework. VGIB first employs the noise injection to modulate the information flow from the input graph to the perturbed graph. Then, the perturbed graph is encouraged to be informative to the graph property. VGIB further obtains the desired subgraph by filtering out the noise in the perturbed graph. With the customized noise prior for each input, the VGIB objective is endowed with a tractable variational upper bound, leading to a superior empirical performance as well as theoretical properties. Extensive experiments on graph interpretation, explainability of Graph Neural Networks, and graph classification show that VGIB finds better subgraphs than existing methods 1 1 Code is avaliable on https://github.com/Samyu0304/VGIB.","Title: Random radial basis function kernel-based support vector machine\nYear: 2021\n\nAbstract: The main computational cost of building a support vector machine (SVM) training model lies in tuning the hyperparameters, including the kernel parameters and penalty constant C. This paper introduces a new kernel, the random radial basis function (RRBF) kernel, which all kernel parameters can be assigned to randomly. The key idea of the RRBF is to extend a one-dimensional parameter to d-dimensional parameters by simple modification of the traditional RBF kernel. We prove the universal approximation capability of a SVM with the RRBF kernel, proposing a simple model selection algorithm. The experiments on benchmark datasets show that SVM with the RRBF kernel clearly outperforms the traditional RBF kernel and other popular kernels, and the results are quite insensitive to C. Compared with other kernels, a SVM with the RRBF kernel is not sensitive to the penalty constant C, and it can be tuned in a wide range. Additionally, this property also enables a SVM to find the hyperparameters quickly. A simulation of parameter sensitivity study also show that a SVM is more stable w.r.t. to the different choices of kernel parameters.","Title: Multi-view Graph Attention Network for Travel Recommendation\nYear: 2022\n\nAbstract: As an e-commerce feature, the recommender system can enhance the consumer shopping experience and create huge benefits for businesses. The e-tourism has become one of the largest service industries with the application and popularity of recommender systems. Many studies have confirmed that the travel product recommendation is widely different from traditional recommendations. Due to the financial and time costs, travel products are usually browsed and purchased relatively infrequently compared with other traditional products (e.g., books, movies and food). In addition, choosing the appropriate travel product will be influenced by many factors, such as departure, destination and price. To tackle this challenging problem, we propose a MV-GAN (short for Multi-View Graph Attention Network for travel recommendation) model. It enriches user and product semantics through both metapath-guided neighbors aggregation and multi-view fusion in heterogeneous travel product recommendation graph. In particular, we design node-level and path-level attention networks for learning user and product representations from every single view. To collaboratively integrate multiple types of relationships in different views, a view-level attention mechanism is proposed to aggregate the node representations and obtain global user and product representations. We evaluate the proposed method on a public dataset and a dataset constructed from a large tourism e-commerce website in China. Extensive experiments not only validate the effectiveness of MV-GAN, but also show its potentially good interpretability.","Title: Engaging voluntary contributions in online review platforms: The effects of a hierarchical badges system\nYear: 2022\n\nAbstract: Employing a hierarchical badges system to sustain users' participation and content contribution has been a common practice in online review platforms. However, it is still unclear whether this incentive mechanism is effective in inducing users to provide more substantial and useful reviews. In this paper, we draw on the goalsetting theory and status hierarchies theory to study how the hierarchical badges system impacts users' behavior such as their effort levels and expressed opinions, using the data of 234,220 reviews from a large online review platform. Our results show that users significantly increase their efforts as their status are promoted, manifesting as review frequency, length, and readability. In terms of expressed opinions, users with higher badges are less likely to post extremely negative ratings and use negative emotional words. Our results have important implications for business modes that rely on user-generated content, such as online review platforms, social question and answer sites, social media, and crowdsourcing systems.","Title: Pilot Overhead vs. Pilot Power: Short Packet Structure Optimization for URLLC over Continuous Fading\nYear: 2021\n\nAbstract: In this paper, we investigate how to strike the balance between pilot overhead and pilot power for short packet ultra reliable and low latency communication (URLLC) systems under continuous fading, as the previous work has not considered the joint impact of pilot overhead and pilot power on the system throughput and has assumed block fading only. It is revealed that, at low to moderate signal to noise ratio (SNR) or in continuous fading, pilot power boosting can reduce pilot overhead and improve throughput significantly, and that at high SNR, pilot overhead plays a dominant role in enhancing the throughput while maintaining a low peak to average power ratio. For throughput formulation, a closed-form expression for the asymptotic block error probability under continuous fading is derived with respect to pilot power, pilot length and block length. And for throughput maximization, the near-optimal pilot power and the near-optimal block length are given in closed form by solving complicated transcendental equations. Based on the analysis, a low-complexity joint pilot power, pilot length and block length optimization (JPLLO) algorithm is proposed, which achieves a near-optimal performance and a dramatic complexity reduction compared to exhaustive search, converging within only 1-3 iterations. The JPLLO algorithm also significantly outperforms the previous joint optimization algorithm under continuous fading.","Title: A Novel Sentiment Polarity Detection Framework for Chinese\nYear: 2022\n\nAbstract: Nowadays, mining opinions or sentiment from online user-generated text has become a research hot spot. Although a large amount of lexicon-based Chinese polarity detection works have been done, the existing methods have one common flaw: that even the same word can have opposite polarities among different seed lexicons. This is known as polarity fuzziness. To enhance the performance of Chinese senti...","Title: DAN-SuperPoint: Self-Supervised Feature Point Detection Algorithm with Dual Attention Network\nYear: 2022\n\nAbstract: In view of the poor performance of traditional feature point detection methods in low-texture situations, we design a new self-supervised feature extraction network that can be applied to the visual odometer (VO) front-end feature extraction module based on the deep learning method. First, the network uses the feature pyramid structure to perform multi-scale feature fusion to obtain a feature map containing multi-scale information. Then, the feature map is passed through the position attention module and the channel attention module to obtain the feature dependency relationship of the spatial dimension and the channel dimension, respectively, and the weighted spatial feature map and the channel feature map are added element by element to enhance the feature representation. Finally, the weighted feature maps are trained for detectors and descriptors respectively. In addition, in order to improve the prediction accuracy of feature point locations and speed up the network convergence, we add a confidence loss term and a tolerance loss term to the loss functions of the detector and descriptor, respectively. The experiments show that our network achieves satisfactory performance under the Hpatches dataset and KITTI dataset, indicating the reliability of the network.","Title: An intermediary utility-based service search and structure organization approach in service-oriented MAS\nYear: 2022\n\nAbstract: In distributed service-oriented multiagent systems (SOMASs), agents cooperate autonomously to accomplish service discovery. In a SOMAS application, each agent plays two roles: (i) satisfying the service demands as a service provider, and (ii) forwarding the received queries as an intermediary agent. This study defines that the intermediary utility of an agent is positively related to the success rate of the service discovery in which it participates. If the intermediary utility of an agent is high (e.g., when the neighbors of the agent have abundant service resources), then the agent can better guide service searches. Based on this characteristic, this study proposes an intermediary utilities-based service search and structure organization approach to improve the efficiency of service discovery. In this approach, a service query has a higher probability of being forwarded to agents with higher intermediary utilities, which accelerates the service search. Additionally, the connectivity of the system structure can be preserved by the proposed approach, which ensures that the services can be accessed globally in the system. Extensive experiments demonstrate that the proposed approach can achieve better performance regarding the success rate of service discovery and average forwarding time of service queries comparing to previous approaches.","Title: Method to enhance deep learning fault diagnosis by generating adversarial samples\nYear: 2022\n\nAbstract: Modern industrial fields utilize complex mechanical equipment and machinery, which are closely linked, and equipment faults are difficult to express. Therefore, fault diagnosis is important to ensure the safety of complex mechanical equipment in modern industries. Deep learning has achieved excellent results with recent fault diagnosis methods. At present, three common deep learning models (MLP, CNN, and RNN models) can achieve diagnosis rates close to 100% with original fault diagnosis data and a signal-to-noise ratio above 10 dB. However, we found that the diagnostic rate of these three models was completely incorrect when an adversarial sample with a signal-to-noise ratio noise greater than 10 dB was added to the original sample. We propose a GAN-based adversarial signal generative adversarial network (AdvSGAN) in this paper. We conduct experiments on the CWRU dataset and conclude that we can easily obtain adversarial noise and generate training samples through AdvSGAN. With the addition of adversarial data training, the diagnostic rate of the model on these adversarial samples increased from less than 5% to 98.69%, 97.38% and 96.94%. Hence, this method increases the reliability of our deep learning model. (C) 2021 Elsevier B.V. All rights reserved.","Title: MVE-FLK: A multi-task legal judgment prediction via multi-view encoder fusing legal keywords\nYear: 2022\n\nAbstract: Legal Judgment Prediction (LJP) aims to predict the judgment result based on the fact description of a criminal case, and is gradually becoming a hot research topic in the legal realm. Generally, a classic LJP contains three subtasks, i.e., applicable law article prediction, charge prediction, and term of penalty prediction. In real-world scenarios, both charge prediction and applicable law article prediction are actually the tasks of multi-class classification with multi-label learning. However, most existing studies model them as the problems of multi-class classification with single-label learning. Besides, they only consider the context of the fact description, and ignore the exploitation of effective keywords that are widely existed in abundant law articles. To fill the above gaps, we propose a novel multi-task legal judgment prediction framework via multi-view encoder fusing legal keywords, named MVE-FLK, to jointly model multiple subtasks in LJP. Specifically, the multi-view encoder is the core module of MVE-FLK, in this module, we devise a word and sentence encoder (WSE) with an attention mechanism to fuse legal keywords. And then, we develop a multi-view attention network to combine WSE with classic Transformer and DAN (Deep Averaging Network) for encoding the case from multiple views. After that, we propose a multi-task prediction module by developing a novel keywords fusing approach to enhance the performance of multi-task prediction. In addition, we devise a unique prediction principle for each subtask at a fine-grained level, which effectively improves the performance of subtasks. The experimental results on two real-life legal datasets show that our model yields significant prediction performance advantages over six competitive methods.","Title: Average AoI Minimization in UAV-Assisted Data Collection With RF Wireless Power Transfer: A Deep Reinforcement Learning Scheme\nYear: 2022\n\nAbstract: This article studies the unmanned aerial vehicle (UAV)-assisted wireless powered network, where a UAV is dispatched to wirelessly charge multiple ground nodes (GNs) by using radio frequency (RF) energy transfer and then the GNs use their harvested energy to upload the sensed information to the UAV. At each moment, the UAV is scheduled to charge the GNs or only one GN is scheduled to upload its dat...","Title: Integer particle swarm optimization based task scheduling for device-edge- cloud cooperative computing to improve SLA satisfaction\nYear: 2022\n\nAbstract: Task scheduling helps to improve the resource efficiency and the user satisfaction for Device-Edge-Cloud Cooperative Computing (DE3C), by properly mapping requested tasks to hybrid device-edge-cloud resources. In this paper, we focused on the task scheduling problem for optimizing the Service-Level Agreement (SLA) satisfaction and the resource efficiency in DE3C environments. Existing works only focused on one or two of three sub-problems (offloading decision, task assignment and task ordering), leading to a sub-optimal solution. To address this issue, we first formulated the problem as a binary nonlinear programming, and proposed an integer particle swarm optimization method (IPSO) to solve the problem in a reasonable time. With integer coding of task assignment to computing cores, our proposed method exploited IPSO to jointly solve the problems of offloading decision and task assignment, and integrated earliest deadline first scheme into the IPSO to solve the task ordering problem for each core. Extensive experimental results showed that our method achieved upto 953% and 964% better performance than that of several classical and state-of-the-art task scheduling methods in SLA satisfaction and resource efficiency, respectively.","Title: A multipopulation cooperative coevolutionary whale optimization algorithm with a two-stage orthogonal learning mechanism\nYear: 2022\n\nAbstract: This paper designed a multipopulation cooperative coevolutionary framework with a two-stage orthogonal learning (OL) mechanism for the whale optimization algorithm (MCCWOA) to improve the performance of the whale optimization algorithm (WOA). In the framework, a prediction model of the neighborhood structure is established by discovering the guidance information of the following iteration process in the objective space at the first-stage OL. In the second-stage OL, an auxiliary vector pool with various features in the decision space is introduced to guide the candidates falling in the stagnant status to conduct more valuable exploration. According to the domain knowledge of the candidates, the population is divided into the elite population, the intermediate population, and the inferior population. The information of the subpopulations has interacted with the corresponding historical populations in the evolution processes to enhance the ability of cooperative coevolution among individuals. A standard set of comprehensive benchmark cases and three engineering cases are utilized to verify the advantages of the proposed algorithm. The results of the statistical analysis, diversity analysis, and convergence analysis testified that the MCCWOA outperforms the 15 state-of-the-art algorithms regarding efficiency and significance.","Title: A Novel Transformer-Based Attention Network for Image Dehazing\nYear: 2022\n\nAbstract: Image dehazing is challenging due to the problem of ill-posed parameter estimation. Numerous prior-based and learning-based methods have achieved great success. However, most learning-based methods use the changes and connections between scale and depth in convolutional neural networks for feature extraction. Although the performance is greatly improved compared with the prior-based methods, the performance in extracting detailed information is inferior. In this paper, we proposed an image dehazing model built with a convolutional neural network and Transformer, called Transformer for image dehazing (TID). First, we propose a Transformer-based channel attention module (TCAM), using a spatial attention module as its supplement. These two modules form an attention module that enhances channel and spatial features. Second, we use a multiscale parallel residual network as the backbone, which can extract feature information of different scales to achieve feature fusion. We experimented on the RESIDE dataset, and then conducted extensive comparisons and ablation studies with state-of-the-art methods. Experimental results show that our proposed method effectively improves the quality of the restored image, and it is also better than the existing attention modules in performance.","Title: Partial NIR-VIS Heterogeneous Face Recognition With Automatic Saliency Search\nYear: 2021\n\nAbstract: Near-infrared-visual (NIR-VIS) heterogeneous face recognition (HFR) aims to match NIR face images with the corresponding VIS ones. It is a challenging task due to the sensing gaps among different modalities. Occlusions in the input face images make the task extremely complex. To tackle these problems, we present a Saliency Search Network (SSN) to extract domain-invariant identity features. We propose to automatically search the efficient parts of face images in a modality-aware manner, and remove redundant information. Moreover, the searching process is guided by an information bottleneck network, which mitigates the overfitting problems caused by small datasets. Extensive experiments on both complete and partial NIR-VIS HFR on multiple datasets demonstrate the effectiveness and robustness of the proposed method to modality discrepancy and occlusions.","Title: A constrained multi-objective optimization algorithm with two cooperative populations\nYear: 2022\n\nAbstract: Constrained multi-objective problems (CMOPs) require balancing convergence, diversity, and feasibility of solutions. Unfortunately, the existing constrained multi-objective optimization algorithms (CMOEAs) exhibit poor performance when solving the CMOPs with complex feasible regions. To solve this shortcoming, this work proposes an improved algorithm named the CMOEA-TCP, which maintains two populations cooperating to push the solutions to approximate the constrained Pareto front. Specifically, one population is obtained by the Pareto-based method and aims to strengthen the algorithm’s convergence ability. Meanwhile, another population is maintained by decomposition-based method and devoted to improving its diversity. The two populations work cooperatively during the entire evolution process with the constraint-handling technique. The performance of the CMOEA- TCP is verified on three benchmark suites with 34 problems. The experimental results demonstrate that the CMOEA-TCP can achieve performance comparable to or better than the other six state-of-the-art CMOEAs on the majority of considered problems.","Title: Retina-like Computational Ghost Imaging for an Axially Moving Target\nYear: 2022\n\nAbstract: Unlike traditional optical imaging schemes, computational ghost imaging (CGI) provides a way to reconstruct images with the spatial distribution information of illumination patterns and the light intensity collected by a single-pixel detector or bucket detector. Compared with stationary scenes, the relative motion between the target and the imaging system in a dynamic scene causes the degradation of reconstructed images. Therefore, we propose a time-variant retina-like computational ghost imaging method for axially moving targets. The illuminated patterns are specially designed with retina-like structures, and the radii of foveal region can be modified according to the axial movement of target. By using the time-variant retina-like patterns and compressive sensing algorithms, high-quality imaging results are obtained. Experimental verification has shown its effectiveness in improving the reconstruction quality of axially moving targets. The proposed method retains the inherent merits of CGI and provides a useful reference for high-quality GI reconstruction of a moving target.","Title: Android Unit Test Case Generation Based on the Strategy of Multi-Dimensional Coverage\nYear: 2021\n\nAbstract: Although Android apps are becoming more and more popular, apps tend to contain defects which can ultimately manifest as failures (or crashes) to end-users. In order to improve software quality, people have proposed different automated tools for testing Android applications. The automatic generation of test cases is an important means to improve the efficiency of software testing, but many tools re...","Title: Average Descent Rate Singular Value Decomposition and Two-Dimensional Residual Neural Network for Fault Diagnosis of Rotating Machinery\nYear: 2022\n\nAbstract: Fault diagnosis of rotating machinery is difficult under the strong noisy environment. Although singular value decomposition (SVD) can remove noise from vibration signals, the singular value threshold is normally determined by expert experience. To solve this problem, a fault diagnosis method based on average descent rate (ADR)-SVD and two-dimensional residual neural network (Resnet) is proposed. First, ADR-SVD uses the ADR index to construct the singular value descent rate difference spectrum and uses the maximum value of the spectrum as the singular value threshold. The noise reduction process of ADR-SVD requires little expert experience. Then, in order to adaptively identify the fault features of the signals, we introduce Gramian angular difference field (GADF), which can transform the one-dimensional signals into two-dimensional images and preserve the temporal correlation of the one-dimensional signals. Finally, we construct a two-dimensional Resnet to learn image features and identify fault types. The proposed method is tested on Case Western Reserve University (CWRU) bearing dataset, Driveline Dynamic Simulator (DDS) gearbox dataset, and University of Connecticut (UoC) gearbox dataset under the strong noisy environment, which achieves the accuracies of 98.00%, 99.00%, and 98.88%, respectively. The accuracies of other deep learning methods and singular value difference spectrum method are below 95%. The comparisons show that the proposed method has better noise reduction effect and can diagnose the fault type more accurately.","Title: A novel hyperchaotic image encryption algorithm with simultaneous shuffling and diffusion\nYear: 2022\n\nAbstract: The performance of an image encryption algorithm based on chaos is largely determined by the nonlinear characteristics of the underlying chaotic system. This paper proposes a mixed one- and two-dimensional chaotic map (MOTDCM) that has a wider hyperchaotic interval, a larger maximum Lyapunov exponent, and more complex nonlinear dynamics than most existing chaotic systems. Using the hyperchaotic sequences generated by the MOTDCM, a novel image encryption algorithm with different structures is proposed, in which shuffling and diffusion are carried out simultaneously from the perspective of the whole input image. Simulation results and a comparative analysis show that the proposed encryption algorithm has a large key space, high sensitivity to the secret key, and good statistical ciphertext properties. It has a better diffusion effect than existing algorithms and meets the imposed security requirements within only one round of operation, with a reduction in algorithm complexity and an improvement in encryption efficiency. Experimental results demonstrate that this encryption algorithm has good performance and can resist chosen-plaintext attacks and known-plaintext attacks effectively.","Title: Experimental Study on the Coupling Mechanism of Sensors under a Strong Electromagnetic Pulse\nYear: 2022\n\nAbstract: This paper presents a study on the electromagnetic effects of a main fuel control system under the action of a strong electromagnetic pulse. It assesses the electromagnetic sensitivity of a speed sensor and a linear variable differential transformer (LVDT) sensor. This assessment focuses on the control system's electromagnetic effects and the sensors' coupling signals. The effects and signals were determined using radiation and pulse current injection tests. Analysis of the signal at the system power port shows that it is the same as those for the two test methods. Based on analysis of the working mechanism and terminal signals of the sensors, the electromagnetic sensitivity of the system is different under the different electromagnetic pulse conditions. The electromagnetic sensitivity characteristics of the sensors were verified, and the electromagnetic effects of the main fuel regulation control system were analyzed. Meanwhile, the degree of sensor coupling mechanism caused by the electromagnetic coupling effects of the main fuel regulation control system under strong electromagnetic pulses were studied. These findings have clear practical implications for electromagnetic pulse protection of aero-engine control systems.","Title: Scalable semantic-enhanced supervised hashing for cross-modal retrieval\nYear: 2022\n\nAbstract: Cross-modal retrieval is a classic scenario that can describe the same semantics from multiple angles and provide us with more abundant and diversified information ; thus, it has abundant applications in information retrieval, data mining, and machine learning. Generally, supervised information can enhance the ability to learn hash features, and obtain better retrieval accuracy than unsupervised methods. However, how to effectively solve the discrete binary constraint problem, and how to retain more semantic information in the process of hash learning, which constitute important parts of cross-modal retrieval, still have some challenges. To mitigate these problems, in this work, we propose a novel cross-modal hashing approach, called, scalable pairwise embedding constraint hashing (SPECH). The SPECH approach employs the loss of likelihood similarity technique using pairwise sample data to measure the semantic similarity of heterogeneous modal samples. In light of this, the SPECH approach can maximize the utilization of available data from heterogeneous modal samples for training models. In addition, we further enhance the discriminative capability of hash codes and reduce the intraclass differences in heterogeneous modal samples, by embedding the more discriminative semantic label attribute space regression into the to-be-learned process of hash codes. Comprehensive experiments on three benchmark datasets show that the proposed SPECH approach achieves significantly better retrieval accuracy and outperforms several state-of-the-art approaches on cross-modal retrieval applications.","Title: Status Prediction for Age of Information Oriented Short-Packet Transmission in Industrial IoT\nYear: 2022\n\nAbstract: Age of information (AoI), which measures the freshness of information, is a critical performance metric of time-sensitive applications of industrial Internet of things (IIoT) with short-packet transmission (SPT). In this paper, we investigate the suitability of predicting the status updates at source and sending them to destination in advance for AoI oriented SPT systems, in the presence of prediction error as well as transmission error. A predictive transmission scheme is proposed, where proactive transmission termination is adopted as soon as a prediction error is detected, and also multiple correlated features of the status is considered. A closed-form expression for the average AoI with respect to prediction horizon (related to prediction error probability) and blocklength (related to transmission error probability) is derived for the multi-feature source scenario. Also, the prediction error probability with respect to prediction horizon is derived in closed form. It is proved that the average AoI performance can benefit from status prediction, even under high prediction error probability. Simulation results demonstrate the correctness of the analytical results, and show that the proposed prediction scheme outperforms the prediction approach with no transmission termination, and there exists an optimal prediction horizon in terms of average AoI. A tight approximation of the optimal prediction horizon is derived for the special case of single-feature status, which achieves a near-optimal performance, with a much lower complexity than exhaustive search.","Title: A Bayesian optimal social law synthesizing mechanism for strategical agents\nYear: 2022\n\nAbstract: One of the effective and well studied approaches for coordinating multiagent systems is to synthesize social laws which restrict the behavior of individual agents. We show that when rational behavior of the agents and private information are considered, the optimal social law synthesizing problem naturally evolves into a setting which can be handled by the framework of algorithmic mechanism design. We focus on the Bayesian case in this paper, that is, the probability distribution of each agent’s private cost is known by the public. In this case, our problem closely relates to path/spanning-tree auctions and Myerson’s optimal auction mechanism, but the optimization objective is new, that is, we focus on profit maximization instead of payment maximization in a reverse auction, and as far as we know in this setting no existing mechanism can be directly applied. By studying this problem: We further extend the logic-based framework of social law optimization problem to the strategic case, and show that it becomes a new problem of algorithmic mechanism design; We find out a mechanism that is incentive compatible, individually rational and maximize the expected profit for all input cost profiles; However, we can show that this mechanism is computational intractable (FP\n \n \n \n $$^{NP}$$\n \n -complete); So, we try to specify Computation Tree Logic semantics as a set of linear-integer constraints, design a Integer-Linear Programming based algorithm for computing the proposed mechanism, enabling it to be handled by current ILP solvers, and finally find out a tractable 2-approximation mechanism.","Title: Independent Pilots Versus Shared Pilots: Short Frame Structure Optimization for Heterogeneous-Traffic URLLC Networks\nYear: 2022\n\nAbstract: We investigate a multi-device ultra-reliable low-latency communication system with heterogeneous traffic and finite block length over temporally-correlated fading channels. In light of the challenging demand for accurate channel estimation with limited pilot in a short frame, two frame structures, which respectively adopt independent pilots and shared pilot, are investigated. Block lengths and pilot lengths are jointly optimized for the two frame structures, through instantaneous channel state information (CSI) based dynamic optimization and statistical CSI based static optimization, to strike the tradeoffs among performance, complexity and signaling overhead. The proposed joint optimization algorithms significantly outperform the existing approaches that solely optimize block lengths or pilot lengths. The dynamic optimization algorithms achieve near-optimal performance at dramatic complexity reduction over exhaustive search, and maintain robustness against traffic heterogeneity. Also, the static optimization algorithms are conducted offline, while still outperforming the previous instantaneous CSI based dynamic optimization approaches. It is demonstrated that the independent-pilot frame structure with dynamic optimization is preferable in the scenario with high traffic heterogeneity or high mobility, and that the shared-pilot frame structure with static optimization presents a comparable performance to the former in the case of low mobility, incurring negligible complexity and signaling overhead.","Title: An explainable multi-sparsity multi-kernel nonconvex optimization least-squares classifier method via ADMM\nYear: 2022\n\nAbstract: Convex optimization techniques are extensively applied to various models, algorithms, and applications of machine learning and data mining. For optimization-based classification methods, the sparsity principle can greatly help to select simple classifier models, while the single- and multi-kernel methods can effectively address nonlinearly separable problems. However, the limited sparsity and kernel methods hinder the improvement of the predictive accuracy, efficiency, iterative update, and interpretable classification model. In this paper, we propose a new Explainable Multi-sparsity Multi-kernel Nonconvex Optimization Least-squares Classifier (EM2NOLC) model, which is an optimization problem with a least-squares objective function and multi-sparsity multi-kernel nonconvex constraints, aiming to address the aforementioned issues. Based on reconstructed multiple kernel learning (MKL), the proposed model can extract important instances and features by finding the sparse coefficient and kernel weight vectors, which are used to compute importance or contribution to classification and obtain the explainable prediction. The corresponding EM2NOLC algorithm is implemented with the Alternating Direction Method of Multipliers (ADMM) method. On the real classification datasets, compared with the three ADMM classifiers of Linear Support Vector Machine Classifier, SVMC, Least Absolute Shrinkage and Selection Operator Classifier, the two MKL classifiers of SimpleMKL and EasyMKL, and the gradient descent classifier of Feature Selection for SVMC, our proposed EM2NOLC generally obtains the best predictive performance and explainable results with the least number of important instances and features having different contribution percentages.","Title: Characterizing the fuzzy community structure in link graph via the likelihood optimization\nYear: 2022\n\nAbstract: Detecting fuzzy communities in networks is a critical but challenging task in many fields including biology, technology, social system and so on. Current technology is largely reliant on network topology, which can’t provide much useful information such module correlations and hierarchies. This paper introduces a new fuzzy community detection method by maximizing the likelihood function, which can dynamically map the link community structure to the optimal state. To specify the direction and weight of the link graph, the directional node graph is first converted into a new type of link graph. Next, we define a communityunit consisting of membership and correlation information in a link graph. Specially, we define this likelihood function with a posteriori probability that aims to find the optimal partition under the prior conditions of given link graph and granularity. Furthermore, based on the spectral analysis of Markov transition matrices, we give a strict mathematical analysis for identifying the optimal number of network communities by analyzing the stability of community structures. Finally, extensive experiments on both artificial benchmarks and real-world networks show that our algorithm consistently achieving higher accuracy than classical methods.","Title: Adaptive Transformer-Based Conditioned Variational Autoencoder for Incomplete Social Event Classification\nYear: 2022\n\nAbstract: ABSTRACTWith the rapid development of the Internet and the expanding scale of social media, incomplete social event classification has increasingly become a challenging task. The key for incomplete social event classification is to accurately leverage the image-level and text-level information. However, most of the existing approaches may suffer from the following limitations: (1) Most Generative Models use the available features to generate the incomplete modality features for social events classification while ignoring the rich semantic label information. (2) The majority of existing multi-modal methods just simply concatenate the coarse-grained image features and text features of the event to get the multi-modal features to classify social events, which ignores the irrelevant multi-modal features and limits their modeling capabilities. To tackle these challenges, in this paper, we propose an Adaptive Transformer-Based Conditioned Variational Autoencoder Network (AT-CVAE) for incomplete social event classification. In the AT-CVAE, we propose a novel Transformer-based Conditioned Variational Autoencoder to jointly model the textual information, visual information and label information into a unified deep model, which can generate more discriminative latent features and enhance the performance of incomplete social event classification. Furthermore, the Mixture-of-Experts Mechanism is utilized to dynamically acquire the weights of each multi-modal information, which can better filter out the irrelevant multi-modal information and capture the vitally important information. Extensive experiments are conducted on two public event datasets, demonstrating the superior performance of our AT-CVAE method.","Title: Order-aware Human Interaction Manipulation\nYear: 2022\n\nAbstract: ABSTRACTThe majority of current techniques for pose transfer disregard the interactions between the transferred person and the surrounding instances, resulting in context inconsistency when applied to complicated situations. To tackle this issue, we propose InterOrderNet, a novel framework to perform order-aware interaction learning. The proposed InterOrderNet learns the relative order on the direction of the z-axis among instances to describe instance-level occlusions. Not only does learning this order guarantee the context consistency of human pose transfer, but it also enhances its generalization to natural scenes. Additionally, we present a novel unsupervised method, named Imitative Contrastive Learning, which sidesteps the requirements of order annotations. Existing pose transfer methods are easy to be integrated into the proposed InterOrderNet. Extensive experiments demonstrate that InterOrderNet enables these methods to perform interaction manipulation.","Title: An Augmented Benchmark Dataset for Geometric Question Answering through Dual Parallel Text Encoding.\nYear: 2022\n\nAbstract: Automatic math problem solving has attracted much attention of NLP researchers recently. However, most of the works focus on the solving of Math Word Problems (MWPs). In this paper, we study on the Geometric Problem Solving based on neural networks. Solving geometric problems requires the integration of text and diagram information as well as the knowledge of the relevant theorems. The lack of high-quality datasets and efficient neural geometric solvers impedes the development of automatic geometric problems solving. Based on GeoQA, we newly annotate 2,518 geometric problems with richer types and greater difficulty to form an augmented benchmark dataset GeoQA+, containing 6,027 problems in training set and 7,528 totally. We further perform data augmentation method to expand the training set to 12,054. Besides, we design a Dual Parallel text Encoder DPE to efficiently encode long and medium-length problem text. The experimental results validate the effectiveness of GeoQA+ and DPE module, and the accuracy of automatic geometric problem solving is improved to 66.09%.","Title: Age of Loop Oriented Wireless Networked Control System: Communication and Control Co-Design in the FBL Regime\nYear: 2022\n\nAbstract: In this paper, age of loop (AoL) is investigated for communication and control co-design in wireless networked control systems (WNCSs), to ensure information freshness in the sensor-controller-actuator loop with two-way delay. Also, the impact of fading channel and finite block length due to limited transmission bits are considered. A closed-form expression for Peak AoL(PAoL) is derived with respect to block length and the number of allowable transmissions in the presence of block error probability. It is shown that the average PAoL is less than the sum of the average peak Age of information (PAoI) in uplink (PAoU) and the PAoI in downlink (PAoD), due to the coupling of uplink and downlink. In addition, the average PAoL is proved to be convex with respect to block length and mono-decreasing with respect to the allowable transmission time. We then propose a low-complexity PAoL-aware block length adaptation scheme with retransmission policy and optimal control. Simulation results verify the correctness of our analytical results and demonstrate the effectiveness of the proposed PAoL-aware block length adaptation scheme in terms of information freshness and control performance.","Title: 3D Printing of PEDOT:PSS-PU-PAA Hydrogels with Excellent Mechanical and Electrical Performance for EMG Electrodes\nYear: 2022\n\nAbstract: Bioelectronics has been developed for recording the electrophysiological activity of diagnostic and therapeutic devices. However, current bioelectrodes still imperfectly comply with tissues, which results in high interfacial impedance and even mechanical detachment. Herein, we report a simple yet effective approach to overcome such hurdles by designing a highly conductive, adhesive hydrogel composite based on freeze-dried poly(3,4-ethylenedioxythiophene):poly (styrene sulfonate) (PEDOT:PSS), polyurethane (PU), and poly(acrylic acid) (PAA). With the continuous phase-separation of PEDOT:PSS, PU, and PAA, the resultant composite hydrogels can simultaneously achieve high adhesion (lap-shear strength > 8 kPa), stretchability (fracture strain > 1100%), and electrical conductivity (conductivity > 2 S/m) by overcoming the traditional trade-off between mechanical and electrical properties in conducting polymer hydrogels. Moreover, such hydrogels are readily applicable to advanced manufacturing techniques such as 3D printing. We further fabricated skin electrodes and achieved high quality and high signal-to-noise ratio EMG signal recording of the forearm.","Title: Impact of Quasi-Biweekly Oscillation on Southeast Asian Cold Surge Rainfall Monitored by TRMM Satellite Observation\nYear: 2022\n\nAbstract: Based on the Tropical Rainfall Measuring Mission (TRMM) satellite observation and ERA5 re-analysis dataset, this paper studies the influence of the northwestward-propagating quasi-biweekly oscillation (QBWO) over the western North Pacific on cold surge rainfall (CSR) over Southeast Asia. Cold surges are the most important driver affecting Southeast Asian rainfall on a synoptic scale. The presence of the QBWO during phases 6-8, in which the associated active convection coupling with a cyclonic circulation reaches Southeast Asia, provides a favorable environment for the increase of CSR. The increase in CSR primarily occurs east of the Philippines, leading to a high likelihood of triggering extreme rainfall. The effects from the QBWO are independent of those from the active MJO phases over Southeast Asia. Additionally, cold surge activity could also be influenced by the QBWO. An examination of the QBWO and MJO indicates that the most preferred phases for the occurrence of cold surges are the time when phase 1 of the QBWO co-exists with phase 7 of the MJO or the time when phase 7 of the QBWO couples with phase 5 of the MJO. Accordingly, about 40% of the total cold surge days would fall in either combination."],"string":"[\n \"Title: Exploiting space-time status for service recommendation\\nYear: 2012\\n\\nAbstract: The prevalence of smart devices allows people to record their space-time status. This paper focuses on exploiting user space-time status and the related semantic information for service recommendation. Firstly, event DAG is employed to organize the space-time information generated based on the service invocation history. Generation algorithm of the event DAG is then proposed. Secondly, a novel collaborative filtering based recommendation algorithm is designed. Potentially interesting services in the target node and its subsequent nodes can be recommended. In our implementation, the user space-time status is generated from the 4D city models (3D location + time) with semantic information. A prototype system is implemented to generate service invocation logs of different users. These simulative logs are utilized to evaluate the effectiveness and efficiency of our proposed method.\",\n \"Title: A theoretic framework of K-means-based consensus clustering\\nYear: 2013\\n\\nAbstract: Consensus clustering emerges as a promising solution to find cluster structures from data. As an efficient approach for consensus clustering, the K-means based method has garnered attention in the literature, but the existing research is still preliminary and fragmented. In this paper, we provide a systematic study on the framework of K-means-based Consensus Clustering (KCC). We first formulate the general definition of KCC, and then reveal a necessary and sufficient condition for utility functions that work for KCC, on both complete and incomplete basic partitionings. Experimental results on various real-world data sets demonstrate that KCC is highly efficient and is comparable to the state-of-the-art methods in terms of clustering quality. In addition, KCC shows high robustness to incomplete basic partitionings with substantial missing values.\",\n \"Title: A Cloud System For Community Extraction From Super-Large Scale Social Networks\\nYear: 2013\\n\\nAbstract: This demo showcase the Community Extraction Cloud (CEC) system. The key idea is to drop weak-tie nodes by efficiently extracting core nodes based on the novel concept of asymptotically equivalent structures (AES) and parallel AES mining algorithm. Meanwhile, to facilitate storing and processing of massive networks, several cloud computing technologies including HDFS, Katta, and Hama are seamlessly integrated into CEC system.\",\n \"Title: A novel 3d correspondence-less method for MRI and paxinos-watson atlas of rat brain registration\\nYear: 2006\\n\\nAbstract: In this paper, a novel three-dimensional registration method for Magnetic Resonance Image (MRI) and Paxinos-Watson Atlas of rat brain involving 3D affine transformations is proposed. For the purpose of adapting to a large range of transform parameters in a high registration accuracy, the registration procedure includes two steps: a principle components analysis (PCA) based coarse registration and a Hausdorff distance based fine registration. Both steps are free from the correspondence of the feature points of MRI and Atlas. We implemented this registration method in a rat brain 3D reconstruction and analysis system. Experiments have demonstrated that this two-step method can be successfully applied to registering the low resolution and noise affection MRI with Paxinos-Watson Atlas of rat brain.\",\n \"Title: The Design and Research of Video Grid of Scalable Multilevel Architecture\\nYear: 2009\\n\\nAbstract: Based on the video grid of two level architecture, this paper proposes to use some client computers with high performance as temporary video servers so as to relieve the payload of the local video server, and to expand the original video grid of two level architecture into that of scalable multilevel architecture as the data stream of a local video server reaches the threshold. A detailed discussion is made about the design and process pattern of this expanded video grid and the strategies of service transfer from one server to another when a temporary video server fails. The simulation results produced by VGSim suggest that the service capacity of the video grid of scalable multilevel architecture based on data stream control has been efficiently improved than that of the video grid of two-level architecture.\",\n \"Title: Personalized Recommendation On Multi-Layer Context Graph\\nYear: 2013\\n\\nAbstract: Recommender systems have been successfully dealing with the problem of information overload. A considerable amount of research has been conducted on recommender systems, but most existing approaches only focus on user and item dimensions and neglect any additional contextual information, such as time and location. In this paper, we propose a Multi-Layer Context Graph (MLCG) model which incorporates a variety of contextual information into a recommendation process and models the interactions between users and items for better recommendation. Moreover, we provide a new ranking algorithm based on Personalized PageRank for recommendation in MLCG, which captures users' preferences and current situations. The experiments on two real-world datasets demonstrate the effectiveness of our approach.\",\n \"Title: Detecting global outliers from large distributed databases.\\nYear: 2012\\n\\nAbstract: In this paper, we present an innovative system, coined as DISTROD (a.k.a DISTRibuted Outlier Detector), for detecting outliers from distributed databases. DISTROD is able to effectively detect the so-called global outliers from distributed databases that are consistent with those produced by the centralized detection paradigm. Experimental evaluation demonstrates the good performance of DISTROD in terms of effectiveness and speed. © 2012 IEEE.\",\n \"Title: Research System and Flow of Dynamic Alliance Based on Grid\\nYear: 2007\\n\\nAbstract: Along with enterprise alliance that strides the district, strides the trade and strides nation recombining, there needs a public technical service platform urgently in order to ensure the alliance resources shared effectively. It also should response the changes of market fast. The emergence of grid technology provides a good way to solve this kind of problems. This paper set forth procreant reason, characteristic, system structure and work flow of dynamic alliance on grid in order to realize share of computing resource, storage resource, dada resource, information resource, software resource, network resource, communication resource, knowledge resource, expert resource, and to eliminate information independence and resource independence, then realize high capability structure sharing and cooperating in network virtual environment.\",\n \"Title: Multi-Feature Fusion Face Recognition Based On Kernel Discriminate Local Preserve Projection Algorithm Under Smart Environment\\nYear: 2012\\n\\nAbstract: In this paper, a new face recognition method based on kernel discriminate local preserve projection(KDLPP) and Multi-feature fusion under smart environment is proposed. In order to solve the small sample size problem, combined with kernel theory and QR decomposition, a new face recognition algorithm named kernel discriminate local preserve projection is proposed based on discriminate local preserve projection algorithm. considered the external features are useful in face recognition, because hair is a highly variable feature of human face, so we combined hair features and DCT features on the feature layer. The experiments on the AMI database indicate the proposed method can enhance the accuracy of the recognition system effectively.\",\n \"Title: Data cloud for distributed data mining via pipelined mapreduce\\nYear: 2011\\n\\nAbstract: Distributed data mining (DDM) which often utilizes autonomous agents is a process to extract globally interesting associations, classifiers, clusters, and other patterns from distributed data. As datasets double in size every year, moving the data repeatedly to distant CPUs brings about high communication cost. In this paper, data cloud is utilized to implement DDM in order to move the data rather than moving computation. MapReduce is a popular programming model for implementing data-centric distributed computing. Firstly, a kind of cloud system architecture for DDM is proposed. Secondly, a modified MapReduce framework called pipelined MapReduce is presented. We select Apriori as a case study and discuss its implementation within MapReduce framework. Several experiments are conducted at last. Experimental results show that with moderate number of map tasks, the execution time of DDM algorithms (i.e., Apriori) can be reduced remarkably. Performance comparison between traditional and our pipelined MapReduce has shown that the map task and reduce task in our pipelined MapReduce can run in a parallel manner, and our pipelined MapReduce greatly decreases the execution time of DDM algorithm. Data cloud is suitable for a multitude of DDM algorithms and can provide significant speedups.\",\n \"Title: The Application of P2P in Video Grid\\nYear: 2009\\n\\nAbstract: The POP servers usually have overloaded burden in the current Video Grid. This paper puts forward a technique to import P2P mechanism into Video Grid to alleviate the burden of POP servers. PGVOD is proposed under this technique. It can send video streams buffered in client computers to other clients so as to alleviate the burden of POP servers, shorten the transmission path of video streams in network, to improve the scalability of the Video Grid. The design schema of the framework and the operation mode of nodes in PGVOD system, the design and the process of management of dynamic update of the LCN are described in detail. Finally, PGVOD is compared with a VOD system--Video Grid of C/S structure. The result shows that PGVOD can save bandwidth of network; improve the throughput and performance of the system efficiently.\",\n \"Title: Discovering Semantics from Multiple Correlated Time Series Stream.\\nYear: 2013\\n\\nAbstract: In this paper, we study a challenging problem of mining data generating rules and state transforming rules (i.e., semantics) underneath multiple correlated time series streams. A novel Correlation field-based Semantics Learning Framework (CfSLF) is proposed to learn the semantic. In the framework, we use Hidden Markov Random Field (HMRF) method to model relationship between latent states and observations in multiple correlated time series to learn data generating rules. The transforming rules are learned from corresponding latent state sequence of multiple time series based on Markov chain character. The reusable semantics learned by CfSLF can be fed into various analysis tools, such as prediction or anomaly detection. Moreover, we present two algorithms based on the semantics, which can later be applied to next-n step prediction and anomaly detection. Experiments on real world data sets demonstrate the efficiency and effectiveness of the proposed method. © Springer-Verlag 2013.\",\n \"Title: A hybrid model using analytic network process and gray relational analysis for bank's IT outsourcing vendor selection.\\nYear: 2012\\n\\nAbstract: Purpose - Considering the limitation of the single stage vendor selection model, this paper proposes a two-stage vendor selection framework for IT outsourcing in microfinance banks. Design/methodology/approach - The paper attempts to realize a complete analysis at company level using grey systems theory for shaping the relations among variables. With the social choice function Dodgson function, the first stage is a trial phase that helps the decision-maker find the potential vendors, then, the decision-maker employs those chosen vendors for the final selection with modified grey relational analysis (GRA) integrated analytic network process (ANP), which emphasizes the interrelation among those selection criteria, and avoids the subjective estimation of experts and practitioners. The case of a microfinance bank IT outsourcing vendor selection is used to verify the proposed approach. Findings - The results of the empirical study show that the proposed method is practical for ranking competing vendors in terms of their overall performance with respect to multiple interdependence criteria for the bank's IT outsourcing. Practical implications - The method exposed in the paper can be used for other supplier selection by modifying some criteria and weight. Originality/value - The paper provides a method for ranking competing vendors in terms of their overall performance with respect to multiple interdependence criteria for IT outsourcing.\",\n \"Title: Keyword extraction based on sequential pattern mining\\nYear: 2011\\n\\nAbstract: Keyword extraction is to automatically extract keywords that capture the main topic discussed in a given document. In this paper, a new keyword extraction algorithm based on sequential patterns is proposed. By preprocessing, a document is represented as sequences of words where a sequential pattern mining algorithm is applied on, and important sequential patterns are mined that reflect the semantic relatedness between words. Both statistical features and pattern features within words are used to build the keyword extraction model. The algorithm is independent of languages and does not need the help of a semantic dictionary to get the semantic features. Experimental results on Chinese journal articles show that the proposed algorithm always outperforms the baseline method KEA.\",\n \"Title: A novel neural network approach to cDNA microarray image segmentation.\\nYear: 2013\\n\\nAbstract: Microarray technology has become a great source of information for biologists to understand the workings of DNA which is one of the most complex codes in nature. Microarray images typically contain several thousands of small spots, each of which represents a different gene in the experiment. One of the key steps in extracting information from a microarray image is the segmentation whose aim is to identify which pixels within an image represent which gene. This task is greatly complicated by noise within the image and a wide degree of variation in the values of the pixels belonging to a typical spot. In the past there have been many methods proposed for the segmentation of microarray image. In this paper, a new method utilizing a series of artificial neural networks, which are based on multi-layer perceptron (MLP) and Kohonen networks, is proposed. The proposed method is applied to a set of real-world cDNA images. Quantitative comparisons between the proposed method and commercial software GenePix(®) are carried out in terms of the peak signal-to-noise ratio (PSNR). This method is shown to not only deliver results comparable and even superior to existing techniques but also have a faster run time.\",\n \"Title: Online mining abnormal period patterns from multiple medical sensor data streams\\nYear: 2014\\n\\nAbstract: With the advanced technology of medical devices and sensors, an abundance of medical data streams are available. However, data analysis techniques are very limited, especially for processing massive multiple physiological streams that may only be understood by medical experts. The state-of-the-art techniques only allow multiple medical devices to independently monitor different physiological parameters for the patient's status, thus they signal too many false alarms, creating unnecessary noise, especially in the Intensive Care Unit (ICU). An effective solution which has been recently studied is to integrate information from multiple physiologic parameters to reduce alarms. But it is a challenge to detect abnormalities from high frequently changed physiological streams data, since abnormalities occur gradually due to the complex situation of patients. An analysis of ICU physiological data streams shows that many vital physiological parameters are changed periodically (such as heart rate, arterial pressure, and respiratory impedance) and thus abnormalities are generally abnormal period patterns. In this paper, we develop a Mining Abnormal Period Patterns from Multiple Physiological Streams (MAPPMPS) method to detect and rank abnormalities in medical sensor streams. The efficiency and effectiveness of the MAPPMPS method is demonstrated by a real-world massive database of multiple physiological streams sampled in ICU, comprising 250 patients' streams (each stream involving over 1.3 million data points) with a total size of 28 GB data.\",\n \"Title: Towards a Tricksy Group Shilling Attack Model against Recommender Systems.\\nYear: 2012\\n\\nAbstract: The robustness of recommender systems has drawn recently more and more attention of both industry and academia. Although a multitude of studies have been devoted to shilling attack modeling and detection, few of them focus on group shilling attack. The attackers in a shilling group work together to manipulate the output of the recommender system. Meanwhile, since the rating profiles in a shilling group are carefully designed, it is hard to detect them by traditional methods. This paper presents a generative model to create shilling group in which every pair of attackers has high diversity. In particular, both strict and loose versions of group shilling attack generation algorithm are proposed. Experimental results on MovieLens data set demonstrate that the shilling group generated by the our model can not only exert large negative effect to recommender systems, but also avoid the detection by the traditional methods. © Springer-Verlag 2012.\",\n \"Title: An Improved Protocol for Deadlock and Livelock Avoidance Resource Co-allocation in Network Computing\\nYear: 2010\\n\\nAbstract: A multitude of applications require simultaneous access to multiple kinds of resources scatted in distributed sites. This problem is known as resource co-allocation which has evolved as a hot topic in network computing. How to design a kind of high-performance protocol for deadlock and livelock avoidance resource co-allocation becomes a challenging problem. In this paper, we propose a new protocol OODP3 (Optimal ODP3) which is based on the currently popular protocol ODP3 (Order-based Deadlock Prevention Protocol with Parallel requests).OODP3 not only inherits the advantage of ODP3 but also guarantees the fulfillment of resource co-allocation within polynomial time. Theoretical proof is conducted to verify the correctness of OODP3. Experimental results also show that OODP3 achieves the better performance improvements than the existing deadlock and livelock avoidance protocol.\",\n \"Title: Community Detection In Multi-Relational Social Networks\\nYear: 2013\\n\\nAbstract: Multi-relational networks are ubiquitous in many fields such as bibliography, twitter, and healthcare. There have been many studies in the literature targeting at discovering communities from social networks. However, most of them have focused on single-relational networks. A hint of methods detected communities from multi-relational networks by converting them to single-relational networks first. Nevertheless, they commonly assumed different relations were independent from each other, which is obviously unreal to real-life cases. In this paper, we attempt to address this challenge by introducing a novel co-ranking framework, named MutuRank. It makes full use of the mutual influence between relations and actors to transform the multi-relational network to the single-relational network. We then present GMM-NK (Gaussian Mixture Model with Neighbor Knowledge) based on local consistency principle to enhance the performance of spectral clustering process in discovering overlapping communities. Experimental results on both synthetic and real-world data demonstrate the effectiveness of the proposed method.\",\n \"Title: A loan default discrimination model using cost-sensitive support vector machine improved by PSO\\nYear: 2013\\n\\nAbstract: This study proposes a novel PSO---CS-SVM model that hybridizes the particle swarm optimization (PSO) and cost sensitive support vector machine (CS-SVM) to deal with the problem of unbalanced data classification and asymmetry misclassification cost in loan default discrimination problem. Cost sensitive learning is applied to the standard SVM by integrating misclassification cost of each sample into standard SVM and PSO is employed for parameter determination of the CS-SVM. Meantime, the financial data are discretized by using the self-organizing mapping neural network. And the evaluation indices are reduced without information loss by genetic algorithm for decreasing the complexity of the model. The effectiveness of integrated model of CS-SVM and PSO is verified by three experiments comparing with traditional CS-SVM, PSO---SVM, SVM and BP neural network through real loan default data of companies in China. The corresponding results indicate that the accuracy rate, hit rate, covering rate and lift coefficient are improved dramatically by the developed approach. The proposed method can control the different types of errors distribution with various cost of misclassification accurately, reduce the total misclassification cost largely, and distinguish the loan default problems effectively.\",\n \"Title: Pick-Up Tree Based Route Recommendation from Taxi Trajectories.\\nYear: 2012\\n\\nAbstract: Recommending suitable routes to taxi drivers for picking up passengers is helpful to raise their incomes and reduce the gasoline consumption. In this paper, a pick-up tree based route recommender system is proposed to minimize the traveling distance without carrying passengers for a given taxis set. Firstly, we apply clustering approach to the GPS trajectory data of a large number of taxis that indicates state variance from \\\"free\\\" to \\\"occupied\\\", and take the centroids as potential pick-up points. Secondly, we propose a heuristic based on skyline computation to construct a pick-up tree in which current position is its root node that connects all centroids. Then, we present a probability model to estimate gasoline consumption of every route. By adopting the estimated gasoline consumption as the weight of every route, the weighted Round-Robin recommendation method for the set of taxis is proposed. Our experimental results on real-world taxi trajectories data set have shown that the proposed recommendation method effectively reduce the driving distance before carrying passengers, especially when the number of cabs becomes large. Meanwhile, the time-cost of our method is also lower than the existing methods. © 2012 Springer-Verlag.\",\n \"Title: Localization and Labeling of Rat Brain in MR Image Based on Paxinos-Watson Atlas\\nYear: 2006\\n\\nAbstract: Localization and labeling of function regions in brain is an important topic in experimental brain sciences because the huge amount of data collected by neuroscientists will become meaningless if we cannot give them a precise description of their locations. In this paper, we proposed a localization and labelling method of 3D MR image of rat brain based on Paxinos-Watson atlas. Our objective is to use the specific atlas to accomplish localization and labeling of specified tissue of interest (TOI) to mimic a veteran expert such that invisible or unclear anatomic function regions in the MR images of rat brain can be automatically identified and marked. We proposed a multi-step method to locate and label the TOIs from the MR image of rat brain. Firstly, preprocessing. It alms at the digitization and 3D reconstruction of the atlas and MRI of rat brain. Secondly, two-step registration. The global registration is to eliminate the big misalign and section angle offset as well as the scale between the MRI and atlas. We can choose some unambiguous and characteristic points manually, and based on these correspondences a coarse registration is obtained using affine model. The local registration is to address individual variability of rat brain that can be performed by using Snake model. Thirdly, post-processing. The goal is to locate and label the TOIs in the selected MR image of rat brain slice guided by well-registered atlas. The experiments demonstrated the feasibility of our method.\",\n \"Title: A novel mutation differential evolution for global optimization.\\nYear: 2015\\n\\nAbstract: Differential evolution DE is a simple and powerful population-based evolutionary algorithm. The success of DE in solving a specific problem crucially depends on appropriately choosing generation strategies and control parameter values. A novel mutation DE MDE is proposed to improve generation strategy DE/best/2. Adaptive mutation is conducted to current population when the population clusters around local optima. Control parameters are adapted based on constants. The performance of MDE is extensively evaluated on eighteen benchmark functions and compares favorably with the several DE variants.\",\n \"Title: K-Means-Based Consensus Clustering: A Unified View\\nYear: 2015\\n\\nAbstract: The objective of consensus clustering is to find a single partitioning which agrees as much as possible with existing basic partitionings. Consensus clustering emerges as a promising solution to find cluster structures from heterogeneous data. As an efficient approach for consensus clustering, the K-means based method has garnered attention in the literature, however the existing research efforts are still preliminary and fragmented. To that end, in this paper, we provide a systematic study of K-means-based consensus clustering (KCC). Specifically, we first reveal a necessary and sufficient condition for utility functions which work for KCC. This helps to establish a unified framework for KCC on both complete and incomplete data sets. Also, we investigate some important factors, such as the quality and diversity of basic partitionings, which may affect the performances of KCC. Experimental results on various realworld data sets demonstrate that KCC is highly efficient and is comparable to the state-of-the-art methods in terms of clustering quality. In addition, KCC shows high robustness to incomplete basic partitionings with many missing values.\",\n \"Title: cDNA microarray adaptive segmentation\\nYear: 2014\\n\\nAbstract: DNA microarray technology has enabled biologists to study all the genes within an entire organism to obtain a global view of genes@? interaction and regulation. This technology has a great potential in obtaining a deep understanding of the functional organization of cells. This paper is concerned with improving the processes involved in the analysis of microarray image data. The main focus is to clarify an image@?s feature space in an unsupervised manner. Rather than using the raw microarray image, it proposes to produce filtered versions of the image data by applying nonlinear anisotropic diffusion so that the dynamic range of the image could be increased and, therefore, a better ability of signal extraction could be achieved. In this paper, a novel segmentation algorithm is developed. This algorithm is based on the Cellular Neural Network computational paradigm integrated with median and anisotropic diffusion filters. The AnaLogic CNN Simulation Toolbox for MATLAB (InstantVision Toolboxes for MATLAB) is used during the segmentation process. Quantitative comparisons among the proposed methods and GenePix^(R) are carried out from both the objective and subjective points of view. It is shown that analogic algorithm integrated with Complex Diffusion filter is the best one to be applied to accomplish the segmentation.\",\n \"Title: Polygonal approximation using integer particle swarm optimization.\\nYear: 2014\\n\\nAbstract: Polygonal approximation is an effective yet challenging digital curve representation for image analysis, pattern recognition and computer vision. This paper proposes a novel approach, integer particle swarm optimization (iPSO), for polygonal approximation. When compared to the traditional binary version of particle swarm optimization (bPSO), the new iPSO directly uses an integer vector to represent the candidate solution and provides a more efficient and convenient means for solution processing. The velocity and position updating mechanisms in iPSO not only have clear physical meaning, but also guarantee the optimality of the solutions. The method is suitable for polygonal approximation which could otherwise be an intractable optimization problem. The proposed method has been tested on commonly used synthesized shapes and lake contours extracted from the maps of four famous lakes in the world. The experimental results show that the proposed iPSO has better solution quality and computational efficiency than the bPSO-based methods and better solution quality than the other state-of-the-art methods.\",\n \"Title: State Estimation For Complex Systems With Randomly Occurring Nonlinearities And Randomly Missing Measurements\\nYear: 2014\\n\\nAbstract: This paper is concerned with the state estimation problem for the complex networked systems with randomly occurring nonlinearities and randomly missing measurements. The nonlinearities are included to describe the phenomena of nonlinear disturbances which exist in the network and may occur in a probabilistic way. Considering the fact that probabilistic data missing may occur in the process of information transmission, we introduce the randomly data missing into the sensor measurements. The aim of this paper is to design a state estimator to estimate the true states of the considered complex network through the available output measurements. By using a Lyapunov functional and some stochastic analysis techniques, sufficient criteria are obtained in the form of linear matrix inequalities under which the estimation error dynamics is globally asymptotically stable in the mean square. Furthermore, the state estimator gain is also obtained. Finally, a numerical example is employed to illustrate the effectiveness of the proposed state estimation conditions.\",\n \"Title: Detecting Genuine Communities from Large-Scale Social Networks: A Pattern-Based Method.\\nYear: 2014\\n\\nAbstract: Community detection is a long-standing yet very difficult task in social network analysis. It becomes more challenging as many online social networking sites are evolving into super-large scales. Numerous methods have been proposed for community detection from massive networks, but how to reconcile the partitioning efficiency and the community quality remains an open problem. In this paper, we att...\",\n \"Title: A framework for component selection in collaborative sensing application development\\nYear: 2014\\n\\nAbstract: Wireless sensor network-based technologies and applications have attracted a lot of attention in the past two decades because of their huge potential to change people's way of life. These applications usually need close collaboration among multiple sensors, gateways, services and end users. When developing these applications, system designers and practitioners usually face several performance requirements such as the accuracy, battery life and system reliability. Given the hard requirements in system performance, how to choose an optimal combination from various sensors, algorithms and collaborative systems to form the application is the most important problem that practitioners need to address. Ad hoc solutions were proposed in specific applications in the past; however, a general methodology that can be easily applied to future applications is lacking. In this paper, we take the challenge and propose a general framework aiming to address the component selection problem, illustrate how this framework can be applied to real life applications through a case study, and discuss challenging issues and two interesting finds from our implementation.\",\n \"Title: Local Community Extraction for Non-overlapping and Overlapping Community Detection.\\nYear: 2014\\n\\nAbstract: The scale of current networked system is becoming increasingly large, which exerts significant challenges to acquire the knowledge of the entire graph structure, and most global community detection methods often suffer from the computational inefficiency. Local community detection aims at finding a community structure starting from a seed vertex without global information. In this article, we propose a Local Community Extraction algorithm (LCE) to find the local community from a seed vertex. First, a local search model is carefully designed to determine candidate vertices to be preserved or discarded, which only relies on the local/incomplete knowledge rather than the global view of the network. Second, we expand LCE for the global non-overlapping community detection, in which the labels of detected local communities are seen as vertices' attributive tags. Finally, we adopt the results of LCE to calculate a membership matrix, which can been used to detect the global overlapping community of a graph. Experimental results on four real-life networks demonstrate the advantage of LCE over the existing degree-based and similarity-based local community detection methods by either effectiveness or efficiency validity.\",\n \"Title: Research on the Forecasting Model for Hypertension Drug Efficacy Based on Medical Data\\nYear: 2014\\n\\nAbstract: Hypertension is a common disease endangering human health, and every year the number of cases increases at a speed of ten millions. So far there are about 230 million hypertension patients in China, and the long-term drug treatment for hypertension patients is an effective measure to control blood pressure. In this paper, based on the evaluation period of patients with Medication Possession Ratio (MPR) and blood pressure analysis, the beta distribution model which is used to find the relationship between them is established. Then genetic algorithms (GA) and cross validation are carried out on the model and we make comparative analysis with linear distribution model. The experimental results show that the beta distribution model can well forecast the drug efficacy of patients. Only through long-term drug therapy, can the blood pressure be effectively controlled.\",\n \"Title: Overlapping Community Extraction: A Link Hypergraph Partitioning Based Method\\nYear: 2014\\n\\nAbstract: Real-world networks often contain communities with pervasive overlaps such that nodes simultaneously belong to several groups. Community extraction, emerging in recent years, is considered to be a promising solution for finding meaningful communities from social networks. In this paper, we explore overlapping community extraction from a link partitioning perspective. First, we define the local link structure composed of a set of closely interrelated links, by extending the similarity of link-pairs to that of a group of links. Second, based upon our prior work, we transform the problem of mining local link structures into a pattern mining problem, and thus present an efficient mining algorithm. Third, we propose to use the hypergraph to assemble all local link structures, and employ hMETIS for hypergraph partitioning. Finally, based on extracted link communities, we restore the membership of nodes in the original graph owing to its links. Experimental results on various real-life social networks validate the effectiveness of the proposed method.\",\n \"Title: A backbone extraction method with Local Search for complex weighted networks\\nYear: 2014\\n\\nAbstract: The backbone is the natural abstraction of a complex network, which can help people to understand it in a more simplified form. Backbone extraction becomes more challenging as many networks are evolving into large scale and the weight distributions are spanning several orders of magnitude. Traditional filter-based methods tend to include many outliers into the backbone. What is more, they often suffer from the computational inefficiency-the exhaustive search of all nodes or edges is often prohibitively expensive. In this work, we propose a Local Search based Backbone Extraction Heuristic (LS-BEH) to find the backbone in a complex weighted network. First, a strict filtering rule is carefully designed to determine edges to be preserved or discarded. Second, we present a local search model to examine part of edges in an iterative way. Experimental results on two real-life networks demonstrate the advantage of LS-BEH over the classic disparity filter method by either effectiveness or efficiency validity.\",\n \"Title: Collaborative Topic Ranking: Leveraging Item Meta-Data For Sparsity Reduction\\nYear: 2015\\n\\nAbstract: Pair-wise ranking methods have been widely used in recommender systems to deal with implicit feedback. They attempt to discriminate between a handful of observed items and the large set of unobserved items. In these approaches, however, user preferences and item characteristics cannot be estimated reliably due to overfitting given highly sparse data. To alleviate this problem, in this paper, we propose a novel hierarchical Bayesian framework which incorporates \\\"bag-of-words\\\" type meta-data on items into pair-wise ranking models for one-class collaborative filtering. The main idea of our method lies in extending the pair-wise ranking with a probabilistic topic modeling. Instead of regularizing item factors through a zero-mean Gaussian prior, our method introduces item-specific topic proportions as priors for item factors. As a by-product, interpretable latent factors for users and items may help explain recommendations in some applications. We conduct an experimental study on a real and publicly available dataset, and the results show that our algorithm is effective in providing accurate recommendation and interpreting user factors and item factors.\",\n \"Title: Simulation of the recharging method of implantable biosensors based on a wearable incoherent light source.\\nYear: 2014\\n\\nAbstract: Recharging implantable electronics from the outside of the human body is very important for applications such as implantable biosensors and other implantable electronics. In this paper, a recharging method for implantable biosensors based on a wearable incoherent light source has been proposed and simulated. Firstly, we develop a model of the incoherent light source and a multi-layer model of skin tissue. Secondly, the recharging processes of the proposed method have been simulated and tested experimentally, whereby some important conclusions have been reached. Our results indicate that the proposed method will offer a convenient, safe and low-cost recharging method for implantable biosensors, which should promote the application of implantable electronics.\",\n \"Title: An efficient and scalable density-based clustering algorithm for datasets with complex structures.\\nYear: 2016\\n\\nAbstract: As a research branch of data mining, clustering, as an unsupervised learning scheme, focuses on assigning objects in the dataset into several groups, called clusters, without any prior knowledge. Density-Based Spatial Clustering of Applications with Noise (DBSCAN) is one of the most widely used clustering algorithms for spatial datasets, which can detect any shapes of clusters and can automatically identify noise points. However, there are several troublesome limitations of DBSCAN: (1) the performance of the algorithm depends on two specified parameters, ε and MinPts in which ε represents the maximum radius of a neighborhood from the observing point and MinPts means the minimum number of data points contained in such a neighborhood. (2) The time consumption for searching the nearest neighbors of each object is intolerable in the cluster expansion. (3) Selecting different starting points results in quite different consequences. (4) DBSCAN is unable to identify adjacent clusters of various densities. In addition to these restrictions about DBSCAN mentioned above, the identification of border points is often ignored. In our paper, we successfully solve the above problems. Firstly, we improve the traditional locality sensitive hashing method to implement fast query of nearest neighbors. Secondly, several definitions are redefined on the basis of the influence space of each object, which takes the nearest neighbors and the reverse nearest neighbors into account. The influence space is proved to be sensitive to local density changes to successfully reduce the amount of parameters and identify adjacent clusters of different densities. Moreover, this new relationship based on the influence space makes the insensitivity to the ordering of inputting points possible. Finally, a new concept—core density reachable based on the influence space is put forward which aims to distinguish between border objects and noisy objects. Several experiments are performed which demonstrate that the performance of our proposed algorithm is better than the traditional DBSCAN algorithm and the improved algorithm IS-DBSCAN.\",\n \"Title: Adaptive event-triggered H∞ filtering for T-S fuzzy system with time delay.\\nYear: 2016\\n\\nAbstract: This paper is concerned with adaptive event-triggered H∞ filter design for a class of T-S fuzzy systems with time delay. Firstly, an adaptive event-triggered scheme is introduced, which can adaptively adjust the communication threshold to save the limited communication resource. Secondly, a T-S fuzzy model is applied to approximate the nonlinear dynamics of the plant. By using Lyapunov function, sufficient conditions for the existence of the desired filter are established in terms of linear matrix inequalities such that the filtering error dynamics is locally mean square asymptotically stable. Then, the explicit expression is provided for the designed filter parameters. Finally, a simulation example is employed to illustrate the design method.\",\n \"Title: Discovering shilling groups in a real e-commerce platform.\\nYear: 2016\\n\\nAbstract: Purpose - With the popularity of e-commerce, shilling attack is becoming more rampant in online shopping websites. Shilling attackers publish mendacious ratings as well as reviews for promoting or suppressing target products. The purpose of this paper is to investigate group shilling, a new typed shilling attack, behavior in a real e-commerce platform (e.g. Amazon. cn). Design/methodology/approach - Several behavioral features are proposed for modeling the shilling group, and thus an unsupervised ranking method based on principal component analysis (PCA) is presented for identifying shilling groups from real users on Amazon. cn. Findings - As indicated by the behavior analysis, the proposed method has successfully identified a number of shilling groups on Amazon. Meanwhile, the effectiveness of the proposed features and accuracy of the proposed unsupervised method are carefully validated. Originality/value - This paper presents a set of solutions for discovering shilling groups when the ground truth labels are hard to be obtained in real environment, including candidate groups generation, behavioral features definition and unsupervised detection.\",\n \"Title: The consistency measures of multi-granularity linguistic group decision making.\\nYear: 2015\\n\\nAbstract: When different experts have different uncertainty degrees on the phenomenon, several linguistic term sets with different granularity of uncertainty are necessary. In this paper, we deal with the consistency problem with multi-granularity linguistic term sets applied to group decision making. Firstly, we develop a transformation model to maintain the uncertainty degrees of different granularity associated with each expert. We use the uncertain linguistic variables as the unified form. Some computational rules about uncertain linguistic variables are given. Then consistency index of linguistic preference relations based on the distance of a multi-granularity linguistic preference relation is defined. Chi-square statistic is used to establish the consistency thresholds. Finally two numerical examples are used to show the process and effects of our new proposed method.\",\n \"Title: Detecting overlapping communities in poly-relational networks\\nYear: 2015\\n\\nAbstract: Discovering communities can promote the understanding of the structure, function and evolution in various systems. Overlapping community detection in poly-relational networks has gained much more interests in recent years, due to the fact that poly-relational networks and communities with pervasive overlap are prevalent in the real world. A plethora of methods detect communities from the poly-relational network by converting it to mono-relational networks first. Nevertheless, they commonly assume different relations are independent from each other, which is obviously unreal to real-life cases. In this paper, we attempt to relax this strong assumption by introducing a novel co-ranking framework, named MutuRank. It makes full use of the mutual influence between relations and actors to transform the poly-relational network to the mono-relational network. We then present a novel GMM-NK (Gaussian Mixture Model with Neighbor Knowledge) algorithm incorporating the impact from neighbors into the traditional GMM. Experimental results both on synthetic networks and the real-world network have verified the effectiveness of MutuRank and GMM-NK.\",\n \"Title: An Ensemble Differential Evolution For Numerical Optimization\\nYear: 2015\\n\\nAbstract: The success of differential evolution (DE) in solving a specific problem crucially depends on appropriately choosing generation strategies and control parameter values. The mutation strategies of DE are classified into two groups: DE/rand/k without best solution and DE/best/k with best solution. The proposed algorithm utilizes two mutation strategies. The first one is from DE/rand/k and the second one is from DE/best/k. The proposed algorithm uses two control parameter settings. It randomly combines them to generate trial vectors. The novel mechanism improves the convergence rate of DE and maintains diversity of the population. The performance of the proposed algorithm is extensively evaluated on all the CEC2005 test functions and compares favorably with the several DE variants.\",\n \"Title: AAC-OT: Accountable Oblivious Transfer With Access Control\\nYear: 2015\\n\\nAbstract: To prevent illegal users accessing the database and protect users' privacy, oblivious transfer with access control (AC-OT) was proposed. In an AC-OT scheme, the database provider can encrypt the records and publish corresponding access control lists (ACLs). Prior to accessing the records, a user needs to obtain anonymous credentials from the issuer. Subsequently, an authorized user can obtain the intended records without the database provider knowing its choices. Although AC-OT schemes have shown a lot of merits, there are some practical issues: 1) one of the inherited problems in anonymous credentials is timely revocation and 2) how to prevent malicious users overusing the records. In this paper, we propose an accountable AC-OT scheme to address these issues. In our scheme, an authorized user can access the protected records without the database provider knowing his personal information and choices if: 1) he has obtained the required credentials listed in the ACLs and 2) the number of the access times for each record is no more than the specified bound. Notably, the database provider can trace and revoke the user who overused the records even in the lifetime of his credentials. To the best of our knowledge, it is the first AC-OT scheme where timely revocation and overuse detection are considered.\",\n \"Title: A Hybrid Evolutionary Algorithm For Numerical Optimization Problem\\nYear: 2015\\n\\nAbstract: The hybrid artificial bee colony (ABC) algorithm with differential evolution (DE) techniques (HABCwDE) is proposed for numerical optimization in this paper. The HABCwDE adopts multiple candidate solution generation strategies (CSGSes) from DE techniques to generate new solutions in the framework of the ABC algorithm. In the HABCwDE algorithm, three CSGSes and three groups of parameter settings are employed. The performance of HABCwDE and some other evolutionary algorithms are tested on 26 state-of-the-art benchmark functions. Experimental results demonstrate that HABCwDE is very competitive, and that it is an effective way to improve the performance of ABC algorithm by employing CSGSes from DE techniques.\",\n \"Title: Multimodel Biometrics Fusion Based On Far And Frr Using Triangular Norm\\nYear: 2015\\n\\nAbstract: Multibiomitric systems are expected to be more accurate due to the presence of multiple evidences, score level fusion is the most commonly used approach in multibiometrics. In this paper, A novel approach is proposed for the fusion at score level fusion based on False Reject Rate(FRR) and False Accept Rate(FRR) using triangular norms(t-norms). This study aims at tapping the potential of t-norms for information fusion at first, at the second, it transfers scores into Transfer function based on corresponding FRRs and FARs, thus avoiding calculating posterior probability of a certain score. Experiment result shown that the proposed method renders very good performance as it is quite computationally and outperforms the traditional score level fusion schemes, the experimental result also confirms the effectiveness of the proposed method to improve the performance of multibiometric system.\",\n \"Title: 3D Model-Based Food Traceability Information Extraction Framework.\\nYear: 2015\\n\\nAbstract: In this paper, we propose a 3D model-based food traceability information extraction method for processing video surveillance data. The proposed method first builds a 3D model of the surveillance area. Then, the video cameras are mapped in the 3D model and the coordinate transform functions from the 2D camera coordinates to the 3D model coordinates are calculated. Next, the object detection method is applied to identify the target which is then mapped into the 3D coordinates so that its 3D trajectory can be generated. Finally, we merge multiple trajectories from different cameras to create the complete traceability information for the target object. According to the experimental results, the proposed method can efficiently extract useful traceability information for a video surveillance system.\",\n \"Title: Spammers Detection from Product Reviews: A Hybrid Model\\nYear: 2015\\n\\nAbstract: Driven by profits, spam reviews for product promotion or suppression become increasingly rampant in online shopping platforms. This paper focuses on detecting hidden spam users based on product reviews. In the literature, there have been tremendous studies suggesting diversified methods for spammer detection, but whether these methods can be combined effectively for higher performance remains unclear. Along this line, a hybrid PU-learning-based Spammer Detection (hPSD) model is proposed in this paper. On one hand, hPSD can detect multi-type spammers by injecting or recognizing only a small portion of positive samples, which meets particularly real-world application scenarios. More importantly, hPSD can leverage both user features and user relations to build a spammer classifier via a semi-supervised hybrid learning framework. Experimental results on movie data sets with shilling injection show that hPSD outperforms several state-of-the-art baseline methods. In particular, hPSD shows great potential in detecting hidden spammers as well as their underlying employers from a real-life Amazon data set. These demonstrate the effectiveness and practical value of hPSD for real-life applications.\",\n \"Title: An improved neural network algorithm based on artificial bee colony algorithm and its application in sewage treatment\\nYear: 2015\\n\\nAbstract: As an advanced artificial intelligence technology, error back-propagation (BP) neural network algorithm has been widely applied to electronics, communications, automation and other fields. However, traditional BP neural network algorithm has the disadvantages, such as inclination to stick into local optima, and slow convergence, which exert a great impact on the processing performance, and also limit its further application. Artificial bee colony (ABC) algorithm has been developed as a new heuristic algorithm in recent years based on foraging behavior, which is also proposed to be a global optimization algorithm. As a simple and reliable algorithm, ABC algorithm is easy to be implemented. In this paper, an optimized BP neural network algorithm based on ABC algorithm is proposed so as to integrate the advantages of both algorithms. Simulation result shows that compared with the conventional BP neural network algorithm, the new algorithm has gained a significant development not only in prediction accuracy but also in convergence speed.\",\n \"Title: A survival analysis method for stock market prediction\\nYear: 2015\\n\\nAbstract: Stock market prediction focus on developing approaches to determine the future price of a stock or other financial product. The key task of stock market prediction is to determine the timing for the buying or selling of stock, undoubtedly, it is very difficult due to the high volatility and nonlinear relationships driven by short-term fluctuations in investment demand. In this work, we address this problem by adopting the Cox's hazard model to predict a stock's future rising or dropping probabilities. Specifically, we define the problem of Buy-and-Sell-Point Prediction from the survival analysis perspective. The Cox's hazard model is proposed as the model of choice for this prediction problem due to several reasons including the ability to model the dynamics in the stock movement, and to easily incorporate different types of technical indexes as covariates. In the experiment, we apply the trained model for the stock market forecasting on six stocks in Shanghai Stock Exchange. The results show that the proposed model is superior to several baseline models in terms of accuracy, and the stock return evaluations have revealed that the profits produced by the proposed model are higher.\",\n \"Title: Scaling Information-Theoretic Text Clustering: A Sampling-based Approximate Method\\nYear: 2014\\n\\nAbstract: Info-Kmeans, a K-means clustering method employingKL-divergence as the proximity function, is one ofthe representative methods in information-theoretic clustering.With the explosive growth of online texts such as online reviewsand user-generated content, the text is becoming more sparseand much bigger, which poses significant challenges on botheffectiveness and efficiency issues of text clustering. In our priorwork, we presented a Summation-bAsed Incremental Learning(SAIL) algorithm, which can avoid the zero-feature dilemma ofhighly sparse texts. In this paper, we propose a sampling-basedapproximate approach for scaling SAIL algorithm to deal withthe large-scale of texts. Particularly, an instance-level randomsampling is invoked to reduce the number of instances to beexamined during each iteration, which substantially speedsup the clustering on big text data. Furthermore, we provethat the margin of errors introduced by random samplingcan be controlled in a small range. Extensive experiments oneight real-life text datasets demonstrate the advantage of theproposed sampling-based approximate clustering method. Inparticular, our method shows merits in both effectiveness andefficiency on clustering performance.\",\n \"Title: Efficient Parallel Community Detection in Large Edge-Intensive Networks\\nYear: 2014\\n\\nAbstract: Community detection is a classic and very difficulttask in social network analysis. A large number of methodshave been developed for both efficient and effective communitydetection. However, much of the existing methods are heavilydependent on the number of links in the network, and thusthey often suffer from the computational inefficiency whenmeeting large edge-intensive networks. In this paper, we presenta novel SIMPLifying and Ensembling (SIMPLE) frameworkfor parallel community detection. It employs the random linksampling to simplify the network and obtain basic partitioningson every sampled graphs. Then, the K-means-based ConsensusClustering is used to ensemble a number of basic partitioningsto get high-quality community structures. Meanwhile, stepsof random sampling and sampled graph partitioning areencapsulated into MapReduce to further improve the efficiency.Experiments on four real-world social networks analyze keyparameters and factors inside SIMPLE, and demonstrate theeffectiveness of the SIMPLE.\",\n \"Title: A distributed SVM method based on the iterative MapReduce\\nYear: 2015\\n\\nAbstract: Linear classification is useful in many applications, but training large-scale data remains an important research issue. Recent advances in linear classification have shown that distributed methods can be efficient in improving the training time. However, for most of the existing training methods,based on MPI or Hadoop, the communication between nodes is the bottleneck. To shorten the communication between nodes, we propose and analyze a method for distributed support vector machine and implement it on an iterative MapReduce framework. Through our distributed method, the local SVMs are generic and can make use of the state-of-the-art SVM solvers. Unlike previous attempts to parallelize SVMs the algorithm does not make assumptions on the density of the support vectors, i.e., the efficiency of the algorithm holds also for the “difficult” cases where the number of support vectors is very high. The performance of the our method is evaluated in an experimental environment. By partitioning the training dataset into smaller subsets and optimizing the partitioned subsets across a cluster of computers, we reduce the training time significantly while maintaining a high level of accuracy in both binary and multiclass classifications.\",\n \"Title: Building the Computational Virtual Reality Environment for Anaesthetists' Training and Practice\\nYear: 2015\\n\\nAbstract: Understanding the real world based on visualisation and prediction is essential for the decision-maker. We build a computational virtual reality environment to improve visualisation, understanding and prediction of the physical world and to guide action. It develops a five-dimensional, computer-generated, computational Virtual Reality Environment for Anaesthesia (VREA). Our online prediction will be calculated based on the correlation and composition computing with respect to the three dimensions: horizontal, vertical and individual. The novel musical notes based anesthetic simulator is proposed to identify the abnormality and visualize the online medical time series. The experiments with the online ECG data will present a real-time case to show the effectiveness and efficiency of our proposed system and algorithms.\",\n \"Title: Statistical characterization of TVWS noise: An experimental study\\nYear: 2012\\n\\nAbstract: This paper presents results of a study about the statistical properties of noise (including unknown interference) in blank channels of TV bands, i.e., TV white spaces. Measurements were conducted to collect data in three different scenarios: sunny-day laboratory, sunny-day curbside and rainy-day laboratory. Impulsive noise has been modeled as a pulse train and the pulse amplitude, pulse duration time and inter-arrival time in both time and frequency domains were defined and considered as typical statistical variables. Several comparisons were made to study the statistical properties of TV white space noise, in two different channels (672~680MHz and 768~776MHz), under three different scenarios, and with several typical distribution models (Rayleigh, Gamma, lognormal and Middleton A). It has been found that noises in different TV white spaces have highly consistent statistical properties, whereas under different scenarios may exhibit diverse statistical properties. A Kolmogorov-Smirnov test was applied to find the closest distributions within the considered candidate distributions. Two new models for the amplitude probability distribution of noise in the frequency domain were proposed and validated, both consisting of a low-amplitude thermal noise component and a high-amplitude impulsive noise component.\",\n \"Title: Accountable mobile E-commerce scheme via identity-based plaintext-checkable encryption.\\nYear: 2016\\n\\nAbstract: Privacy and accountability are two important issues in mobile e-commerce. In this paper, we formalized the definition and security model of identity-based plaintext-checkable encryption (IBPCE) schemes, and proposed an IBPCE scheme. Furthermore, we designed an accountable mobile e-commerce (AMEC) scheme. This scheme can provide the following features: (1) users can register to the e-commerce system by using their mobile identities, such as mobile phone numbers; (2) the transactions between a buyer and a vendor are encrypted; (3) if there is a dispute, an offline adjudicator can identify who is dishonest by checking the encrypted transactions. Hence, our scheme is an ideal primitive to balance the need for privacy and accountability. In mobile e-commerce systems, users conduct transactions using wireless or Internet-based devices, such as mobile phones and tablets. It is different from traditional e-commerce systems relying on workstations or desktops, which is usually used in a fixed location. Recently, privacy and accountability have become users' primary concerns in mobile e-commerce applications. In this paper, a novel mobile e-commerce scheme is developed to address the fundamental requirements. We first propose an identity-based plaintext-checkable encryption (IBPCE) scheme where anyone can check whether a ciphertext is the encryption of a plaintext under a specific identity without knowing the secret key. Furthermore, the proposed IBPCE scheme is incorporated into the mobile e-commerce scenario, which results in an accountable mobile e-commerce (AMEC) scheme. Our proposed AMEC scheme has several superior features: (1) Users can register to the e-commerce system by using their mobile identities, such as mobile phone numbers; (2) The transactions between a buyer and a vendor are encrypted; (3) If there is a dispute, an offline adjudicator can identify who is dishonest by checking the encrypted transactions. We evaluate the proposed scheme and confirm that the new scheme can effectively balance the need for privacy and accountability.\",\n \"Title: K-Anonymisation Of Social Network By Vertex And Edge Modification\\nYear: 2016\\n\\nAbstract: With the rapid growth of social networks, a large quantity of social network data is being collected, and published for analysis. Privacy is one of the major issues when publishing social network data. To address the issue, several social network anonymisation approaches have been proposed. However, these methods usually introduce a large amount of distortion to the original social network graph, thus, the utility of the social network is not preserved very well. In this paper, we present a two-phase approach: in the first phase, we get the target degree of each vertex; the community structure of the social network and path length between vertices are considered when finding candidates to increase the degree of vertex to its target degree in the second phase. Experimental results on real world datasets show that the utility of the social network is well preserved with our approach.\",\n \"Title: Game theory based emotional evolution analysis for chinese online reviews.\\nYear: 2016\\n\\nAbstract: Sentiment analysis has become one of the mainstream researches in social network analysis. Its impact can be seen in many practical applications, ranging from public opinion analysis to marketing of public praise and information prediction. However, most of the existing research has been performed in the sentiment classification for subjective text, the emotional evolution analysis for complex interactive text (e.g., online reviews) has not yet been thoroughly targeted by the research community. This paper is concerned on short-text Chinese online reviews collected from Tianya forum. First, an efficient affective computing framework is proposed to capture the underlying emotions of Chinese online reviews. It can accurately calculate the semantic orientation of the entire review, without requiring expensive manual labeling of seed words. As users' attitudes might influence with each other, predicting their future emotional behaviors that only relying on the emotional values of historical reviews is very one-sided. Therefore, we propose a game theory based emotional evolution prediction algorithm combining the affective computing, in which the mixed nash equilibrium strategies are calculated as the future emotional behavior of interactive users. Then, experimental results on the large-scaled review dataset are provided to demonstrate the effectiveness and accurateness of our approaches. Finally, by applying the research results on the pairwise happiness-popularity coordination evaluation, we have revealed some interesting phenomenon on the \\\\\\\"World View\\\\\\\" board in Tianya forum.\",\n \"Title: LED: A fast overlapping communities detection algorithm based on structural clustering.\\nYear: 2016\\n\\nAbstract: Abstract Community detection in social networks is a fundamental task of complex network analysis. Community is usually regarded as a functional unit. Networks in real world more or less have overlapping community structure while traditional community detection algorithms assume that one vertex can only belong to one community. This paper proposes an efficient overlapping community detection algorithm named LED ( Loop Edges Delete ). LED algorithm is based on Structural Clustering, which converts structural similarity between vertices to weights of network. The evaluations of the LED algorithm are conducted both from classical networks from literature and C-DBLP, which is a huge and real-life co-author social network in China. The results show that LED is superior to other methods in accuracy, efficiency, comparing with FastModurity and GN algorithm.\",\n \"Title: Edge Computing: Vision and Challenges.\\nYear: 2016\\n\\nAbstract: The proliferation of Internet of Things (IoT) and the success of rich cloud services have pushed the horizon of a new computing paradigm, edge computing, which calls for processing the data at the edge of the network. Edge computing has the potential to address the concerns of response time requirement, battery life constraint, bandwidth cost saving, as well as data safety and privacy. In this pap...\",\n \"Title: Stability analysis of switched cellular neural networks: A mode-dependent average dwell time approach.\\nYear: 2016\\n\\nAbstract: This paper addresses the exponential stability of switched cellular neural networks by using the mode-dependent average dwell time (MDADT) approach. This method is quite different from the traditional average dwell time (ADT) method in permitting each subsystem to have its own average dwell time. Detailed investigations have been carried out for two cases. One is that all subsystems are stable and the other is that stable subsystems coexist with unstable subsystems. By employing Lyapunov functionals, linear matrix inequalities (LMIs), Jessen-type inequality, Wirtinger-based inequality, reciprocally convex approach, we derived some novel and less conservative conditions on exponential stability of the networks. Comparing to ADT, the proposed MDADT show that the minimal dwell time of each subsystem is smaller and the switched system stabilizes faster. The obtained results extend and improve some existing ones. Moreover, the validness and effectiveness of these results are demonstrated through numerical simulations.\",\n \"Title: Real-time target detection and recognition with deep convolutional networks for intelligent visual surveillance.\\nYear: 2016\\n\\nAbstract: Moving target detection and tracking, recognition, behaviors analysis are the key issues in the intelligent visual surveillance system (IVSS). The challenge is how to process the real-time video stream in an effective way in case that we could find the interested objects for analysis. However, the traditional video surveillance technology often does not meet the needs of real-time key frame recognition for the on-line intelligent video monitoring system. In our paper, we adopt the state-of-the-art Faster R-CNN [7] that takes advantages of convolutional neural networks into our real-time target recognition system - Deep Intelligent Visual Surveillance (DIVS). The key aspects of our DIVS are consisted of four parts: (i) Getting the real-time video images from remote cameras; (ii) Processing the data with the deep learning framework caffe [23] built for Faster R-CNN; (iii) Storing the valuable data with MySQL; (iv) Data presentation on the website. Experiments based on our system validated the effectiveness, stability and accuracy of our proposed solutions.\",\n \"Title: Similarity based leaf image retrieval using multiscale R-angle description.\\nYear: 2016\\n\\nAbstract: Leaf image identification is a significant and challenging application of computer vision and image processing. A central issue associated with this task is how to effectively and efficiently describe the leaf images and measure their similarities. In this paper, a novel shape descriptor termed R-angle is proposed. R-angle describes the curvature of the contour by measuring the angle between the intersections of the shape contour with a circle of radius R centered at points sampled around the contour. It is intrinsically invariant to group transforms including scaling, rotation and translation. Varying the parameter R of the proposed R-angle naturally introduces the notation of scale, which we leverage to provide a coarse-to-fine description of the local curvature. A local scale arrangement is proposed by taking the distance between each contour point and the center of the shape to be the maximum scale for a given contour point. Two matching schemes, including L1-norm matching and dynamic programming based matching, are applied to measure the similarities of the leaf shapes. The retrieval experiments conducted on two challenging leaf image datasets indicate that the proposed method significantly outperforms the state-of-the-art methods for leaf identification. An additional experiment on an animal dataset also indicates its potential for general shape recognition.\",\n \"Title: DGWC: Distributed and generic web crawler for online information extraction\\nYear: 2016\\n\\nAbstract: Online information has become important data source to analyze the public opinion and behavior, which is significant for social management and business decision. Web crawler systems target at automatically download and parse web pages to extract expected online information. However, as the rapid increasing of web pages and the heterogeneous page structures, the performance and the rules of parsing have become two serious challenges to web crawler systems. In this paper, we propose a distributed and generic web crawler system (DGWC), in which spiders are scheduled to parallel access and parse web pages to improve performance, utilized a shared and memory based database. Furthermore, we package the spider program and the dependencies in a container called Docker to make the system easily horizontal scaling. Last but not the least, a statistics-based approach is proposed to extract the main text using supervised-learning classifier instead of parsing the page structures. Experimental results on real-world data validate the efficiency and effectiveness of DGWC.\",\n \"Title: Parallelizing K-Means-Based Clustering on Spark\\nYear: 2016\\n\\nAbstract: K-means is in fact a family of clustering algorithms with different distance functions and a variety of extension, e.g., fuzzy clustering and consensus clustering. Nevertheless, K-means-based clustering algorithms employ the similar two-phase iterative procedure including distance computation and centroids updating. Therefore, to explore the parallel implementations of this two-phase iterative procedure on Spark is not only universal to a wealth of clustering algorithms but also meets the practical needs addressed by big data. This paper contributes to reveal implementation details for parallelizing K-means-based clustering on Spark. In particular, we first introduce the boundary of so-called K-means-based clustering, and then present the overall parallelizable framework on Spark. We discuss the technical barrier and their alternative strategies for each step. Experimental results on both large-scale UCI datasets and text datasets demonstrate the effectiveness and efficiency of our implementations.\",\n \"Title: Online Public Opinion System: Design and Applications\\nYear: 2016\\n\\nAbstract: Online Public Opinion Systems (OPOS) target at collecting, analyzing, summarizing and monitoring massive public opinions on the Internet in real time. Meanwhile, OPOS often have the ability to identify the key or sudden events, and thus notify related people immediately for rapid responses to these events. As part of this endeavor, this paper introduces the architecture and techniques of an OPOS that has been used by several large enterprises. This self-designed OPOS generally contains data layer, computation layer and application layer from bottom to up. We first introduce the basic function and key techniques of each layer, and then present several typical yet important algorithms on the computation layer. Experimental results on real-world data validate the effectiveness of algorithms fixed in our system. Last but not the least, a system demonstration in a ship-building company is provided to justify the value of our OPOS for real enterprises.\",\n \"Title: Self-representation dimensionality reduction for multi-model classification.\\nYear: 2017\\n\\nAbstract: Feature selection remove the noisy/irrelevant samples and select the subset of representative features, in general, from the high-dimensional space of data has been a fatal significant technique in computer vision and machine learning. Afterwards, motivated by the interpretable ability of feature selection patterns, beside, and the successful use of low-rank constraint in static and sparse learning in the field of machine learning. We present a novel feature selection model with unsupervised learning by using low-rank regression on loss function, and a sparsity term plus K-means clustering method on regularization term during this article. In order to distinguish from those existing state-of-the-art attribute selection measures, the propose method have described as follows: (1) represent the every feature by other features (including itself) via utilize the corresponding loss function with a feature-level self-express way; (2) embed K-means to generate pseudo class label information for the attribute selection as an pseudo supervised method, because of the supervised learning usually have the better recognition results than unsupervised learning; (3) also use the low-rank constraint to feature selection which considers two aspects of information inherent in data. The low-rank constraint takes the correlation of response variables into account, while an 2, p-norm regularizer considers the correlation between feature vectors and their corresponding response variables. The extensive relevant results of experiment on three multi-model comparison data demonstrated that our new unsupervised feature selection pattern outperforms the related approaches.\",\n \"Title: Htcrl: A Range-Free Location Algorithm Based On Homothetic Triangle Cyclic Refinement In Wireless Sensor Networks\\nYear: 2017\\n\\nAbstract: Wireless sensor networks (WSN) have become a significant technology in recent years. They can be widely used in many applications. WSNs consist of a large number of sensor nodes and each of them is energy-constrained and low-power dissipation. Most of the sensor nodes are tiny sensors with small memories and do not acquire their own locations. This means determining the locations of the unknown sensor nodes is one of the key issues in WSN. In this paper, an improved APIT algorithm HTCRL (Homothetic Triangle Cyclic Refinement Location) is proposed, which is based on the principle of the homothetic triangle. It adopts perpendicular median surface cutting to narrow down target area in order to decrease the average localization error rate. It reduces the probability of misjudgment by adding the conditions of judgment. It can get a relatively high accuracy compared with the typical APIT algorithm without any additional hardware equipment or increasing the communication overhead.\",\n \"Title: A recommendation engine for travel products based on topic sequential patterns.\\nYear: 2017\\n\\nAbstract: Travel products recommendation has become one of emerging issues in the realm of recommendation systems. The widely-used collaborative filtering algorithms are usually difficult to be used for recommending travel products due to a number of reasons, including (1) the content of travel products is very complex, (2) the user-item matrix is extremely sparse, and (3) the cold-start users are widely existing. To tackle these issues, we try to exploit Web server logs for generating recommendation, and present a novel recommendation engine (SECT for short) for travel products based on topic sequential patterns. In detail, we first extract topics from semantic description of every Web page. Then, we mine topic frequent sequential patterns and their target products to form click patterns library. At last, we propose a Markov n-gram model for matching the real-time click-stream of users with the click patterns library and thus computing recommendation scores. Experimental results on a real-world travel dataset demonstrate that the SECT prevails over the state-of-art baseline algorithms. In particular, SECT shows merits in improving the both coverage and accuracy for recommending products to cold-start users. Also, SECT is effective to recommend long tail items and outperform baseline algorithms.\",\n \"Title: Recent Progress of Face Image Synthesis\\nYear: 2017\\n\\nAbstract: Face synthesis has been a fascinating yet challenging problem in computer vision and machine learning. Its main research effort is to design algorithms to generate photo-realistic face images via given semantic domain. It has been a crucial prepossessing step of main-stream face recognition approaches and an excellent test of AI ability to use complicated probability distributions. In this paper, we provide a comprehensive review of typical face synthesis works that involve traditional methods as well as advanced deep learning approaches. Particularly, Generative Adversarial Net (GAN) is highlighted to generate photo-realistic and identity preserving results. Furthermore, the public available databases and evaluation metrics are introduced in details. We end the review with discussing unsolved difficulties and promising directions for future research.\",\n \"Title: An accurate traffic classification model based on support vector machines.\\nYear: 2017\\n\\nAbstract: Network traffic classification is a fundamental research topic on high-performance network protocol design and network operation management. Compared with other state-of-the-art studies done on the network traffic classification, machine learning ML methods are more flexible and intelligent, which can automatically search for and describe useful structural patterns in a supplied traffic dataset. As a typical ML method, support vector machines SVMs based on statistical theory has high classification accuracy and stability. However, the performance of SVM classifier can be severely affected by the data scale, feature dimension, and parameters of the classifier. In this paper, a real-time accurate SVM training model named SPP-SVM is proposed. An SPP-SVM is deducted from the scaling dataset and employs principal component analysis PCA to extract data features and verify its relevant traffic features obtained from PCA. By employing PCA algorithm to do the dimension extraction, SPP-SVM confirms the critical component features, reduces the redundancy among them, and lowers the original feature dimension so as to reduce the over fitting and increase its generalization effectively. The optimal working parameters of kernel function used in SPP-SVM are derived automatically from improved particle swarm optimization algorithm, which will optimize the global solution and make its inertia weight coefficient adaptive without searching for the parameters in a wide range, traversing all the parameter points in the grid and adjusting steps gradually. The performance of its two- and multi-class classifiers is proved over 2 sets of traffic traces, coming from different topological points on the Internet. Experiments show that the SPP-SVM's two- and multi-class classifiers are superior to the typical supervised ML algorithms and performs significantly better than traditional SVM in classification accuracy, dimension, and elapsed time.\",\n \"Title: Edgeos(H): A Home Operating System For Internet Of Everything\\nYear: 2017\\n\\nAbstract: The proliferation of Internet of Everything (IoE) and the success of rich Cloud services have pushed the horizon of a new computing paradigm, Edge Computing, which calls for processing the data at the edge of the network. Smart home as a typical IoE application is being widely adapted into people's lives. Edge Computing has the potential to empower the smart home, but it needs more contribution from the community before it truly benefits our lives. In this paper, we present the vision of EdgeOS(H), a home operating system for Internet of Everything. We further discuss functional challenges, namely programming interface, self-management, data management, security & privacy, and naming, as well as non-functional challenges, such as user experience, system cost, delay, and the lack of availability of open testbed. Within each challenge we also discuss the potential directions that are worth further investigation.\",\n \"Title: Semi-SGD: Semi-Supervised Learning Based Spammer Group Detection in Product Reviews\\nYear: 2017\\n\\nAbstract: The purchase decision of customers in e-commerce platforms is strongly influenced by product ratings and reviews. Driven by the profits, review spammers post fake reviews to promote their products or demote their competitors' products. Differ from individual spammers, the spammer groups manipulate reviews together and can be more damaging. Existing work for spammer group detection extract candidate groups from review data and identify the spammer groups using unsupervised spamicity ranking methods. However, the labeled and unlabeled data are existing simultaneously in practice and no method makes good use of both these data in spammer group detection. In this paper, we propose a semi-supervised learning based spammer group detection method (Semi-SGD), which trains a Naive Bayes classifier on a small set of labeled data as an initial classifier, and then incorporates unlabeled data with Expectation Maximization (EM) algorithm to improve the initial classifier iteratively. Experiments on Amazon.cn datasets show that our proposed Semi-SGD is efficient and effective.\",\n \"Title: A Novel De-Noising Method for Improving the Performance of Full-Waveform LiDAR Using Differential Optical Path.\\nYear: 2017\\n\\nAbstract: A novel de-noising method for improving the performance of full-waveform light detection and ranging (LiDAR) based on differential optical path is proposed, and the mathematical models of this method are developed and verified. Backscattered full-waveform signal (BFWS) is detected by two avalanche photodiodes placed before and after the focus of the focusing lens. On the basis of the proposed method, some simulations are carried out and conclusions are achieved. (1) Background noise can be suppressed effectively and peak points of the BFWS are transformed into negative-going zero-crossing points as stop timing moments. (2) The relative increment percentage of the signal-to-noise ratio based on the proposed method first dramatically increases with the increase of the distance, and then the improvement gets smaller by increasing the distance. (3) The differential Gaussian fitting with the Levenberg-Marquardt algorithm is applied, and the results show that it can decompose the BFWS with high accuracy. (4) The differential distance should not be larger than c/2 x (rmin), and two variable gain amplifiers can eliminate the inconsistency of two differential beams. The results are beneficial for designing a better performance full-waveform LiDAR.\",\n \"Title: An Effective Training Method For Deep Convolutional Neural Network.\\nYear: 2017\\n\\nAbstract: In this paper, we propose the nonlinearity generation method to speed up and stabilize the training of deep convolutional neural networks. The proposed method modifies a family of activation functions as nonlinearity generators (NGs). NGs make the activation functions linear symmetric for their inputs to lower model capacity, and automatically introduce nonlinearity to enhance the capacity of the model during training. The proposed method can be considered an unusual form of regularization: the model parameters are obtained by training a relatively low-capacity model, that is relatively easy to optimize at the beginning, with only a few iterations, and these parameters are reused for the initialization of a higher-capacity model. We derive the upper and lower bounds of variance of the weight variation, and show that the initial symmetric structure of NGs helps stabilize training. We evaluate the proposed method on different frameworks of convolutional neural networks over two object recognition benchmark tasks (CIFAR-10 and CIFAR-100). Experimental results showed that the proposed method allows us to (1) speed up the convergence of training, (2) allow for less careful weight initialization, (3) improve or at least maintain the performance of the model at negligible extra computational cost, and (4) easily train a very deep model.\",\n \"Title: Fuzzy Consensus Clustering With Applications on Big Data.\\nYear: 2017\\n\\nAbstract: Consensus clustering aims to find a single partition of data that agrees as much as possible with existing basic partitions. Given its robustness and generalizability, consensus clustering has emerged as a promising solution to find cluster structures inside heterogeneous big data rising from various application domains. In the area of fuzzy systems, however, research along this line is still in i...\",\n \"Title: Analysis of Grain Storage Loss Based on Decision Tree Algorithm.\\nYear: 2017\\n\\nAbstract: Different grain storage factors will cause different degrees of grain loss. In this paper, the data mining method is used to study the loss of grain storage, and the grain loss analysis and forecasting model based on decision tree algorithm is proposed. The paper analyzes and predicts the grain loss caused by different grain storage factors. And the influence of model parameters on model fitting and accuracy is verified by the verification curve. Then the decision tree model is optimized by the method of grid search and cross validation, which improves the prediction accuracy of the decision tree model to analyze the grain loss.\",\n \"Title: Analysis of the grain loss in harvest based on logistic regression.\\nYear: 2017\\n\\nAbstract: In this paper, the grain loss assessment was studied based on logistic regression, and 5400 samples of 31 provinces in our country in the year 2012-2014 were selected, and the 7 typical provinces among them were respectively tested. Using the logistic regression model to predict the loss rate of grain harvesting step, the prediction result is 86.25%. The stochastic gradient descent algorithm is used to optimize the parameters of the model, when the learning_rate is 0.1, the prediction results of grain harvest losses of up to 92.53%, which further improves the prediction accuracy of grain harvest step loss.\",\n \"Title: Study on prediction model of grain post-harvest loss.\\nYear: 2017\\n\\nAbstract: With the arrival of the information age, a great deal of data has been produced in a series of process of grain post-harvest. The rational use of these data allows us to obtain more intelligent, in-depth and valuable information. In this paper, we set up a variety of prediction models for the consumption of grain post-harvest loss, and select the appropriate classifier through comparison. On this basis, the dimension reduction is processed and the confusion matrix is used as the evaluation index to evaluate the prediction effect. Make correlation analysis of the data, obtained the main influencing factors of post-harvest consumption loss link. Finally, visualize the results of the process.\",\n \"Title: TIDM: Topic-Specific Information Detection Model.\\nYear: 2017\\n\\nAbstract: Nowadays information control and detection on the social network have become a problem that we should solve as soon as possible. Unfortunately, due to the informal expressions, detecting the massive data on the internet is a big challenge based on the traditional text mining methods such as Topic Model. In our paper, we propose a simple 4-Tuple Structure instead of the raw text event which usually contains many meaningless words. Using the word embedding technique, we propose the Topic-Specific Information Detection Model (TIDM) for detecting the specific information. For training the words and idiomatic phrases, we adopt the supervise learning technique: manually constructing a specific Semantic Dataset for training our model. Our experiments based on the Amazon Reviews demonstrate that the TIDM can effectively detect and recognize the information.\",\n \"Title: A generalized game theoretic framework for mining communities in complex networks.\\nYear: 2018\\n\\nAbstract: •We introduce a game theoretic framework to model community formation in networks.•We deduce many commonly used objective functions as potential functions in the model.•The proposed framework has been proved to exist fixed (equilibrium) point.•A synchronous learning mechanism has be proposed to reach the local equilibrium.\",\n \"Title: A Simplified FRI Sampling System for Pulse Streams Based on Constraint Random Modulation.\\nYear: 2018\\n\\nAbstract: The recent finite rate of innovation (FRI) framework provides effective sub-Nyquist sampling of pulse streams, allowing recovery of such signals from a set of Fourier coefficients. In this brief, a multi-channel FRI sampling system is presented to sample distinct bands of Fourier coefficients. This is achieved through modulating the desired spectrum band to baseband and then filtering with a low-p...\",\n \"Title: Detecting Spammer Groups From Product Reviews: A Partially Supervised Learning Model.\\nYear: 2018\\n\\nAbstract: Nowadays, online product reviews play a crucial role in the purchase decision of consumers. A high proportion of positive reviews will bring substantial sales growth, while negative reviews will cause sales loss. Driven by the immense financial profits, many spammers try to promote their products or demote their competitors' products by posting fake and biased online reviews. By registering a number of accounts or releasing tasks in crowdsourcing platforms, many individual spammers could be organized as spammer groups to manipulate the product reviews together and can be more damaging. Existing works on spammer group detection extract spammer group candidates from review data and identify the real spammer groups using unsupervised spamicity ranking methods. Actually, according to the previous research, labeling a small number of spammer groups is easier than one assumes, however, few methods try to make good use of these important labeled data. In this paper, we propose a partially supervised learning model (PSGD) to detect spammer groups. By labeling some spammer groups as positive instances, PSGD applies positive unlabeled learning (PU-Learning) to study a classifier as spammer group detector from positive instances (labeled spammer groups) and unlabeled instances (unlabeled groups). Specifically, we extract reliable negative set in terms of the positive instances and the distinctive features. By combining the positive instances, extracted negative instances and unlabeled instances, we convert the PU-Learning problem into the well-known semi supervised learning problem, and then use a Naive Bayesian model and an EM algorithm to train a classifier for spammer group detection. Experiments on real-life Amazon.cn data set show that the proposed PSGD is effective and outperforms the state-of-the-art spammer group detection methods.\",\n \"Title: Hybrid Attention Networks For Chinese Short Text Classification\\nYear: 2017\\n\\nAbstract: To improve the classification performance for Chinese short text with automatic semantic feature selection, in this paper we propose the Hybrid Attention Networks (HANs) which combines the word-and character-level selective attentions. The model firstly applies RNN and CNN to extract the semantic features of texts. Then it captures class-related attentive representation from word-and character-level features. Finally, all of the features are concatenated and fed into the output layer for classification. Experimental results on 32-class and 5-class datasets show that, our model outperforms multiple baselines by combining not only the word-and character-level features of the texts, but also class-related semantic features by attentive mechanism.\",\n \"Title: A novel subgraph K+ -isomorphism method in social network based on graph similarity detection.\\nYear: 2018\\n\\nAbstract: In this paper, we propose a novel (K^{+})-isomorphism method to achieve K-anonymization state among subgraphs or detected communities in a given social network. Our proposed (K^{+})-isomorphism method firstly partitions the subgraphs we have detected into some similar-subgraph clusters followed by graph modification conducted in every cluster. In this way, it is feasible to publish preserved structures of communities or subgraphs and every preserved structure actually represents a cluster of at least K subgraphs or communities which are isomorphic to each other. The contributions of this paper are listed as follows: On the one hand, we improve a maximum common subgraph detection algorithm, MPD(_{-})V, which is a core technique for graph similarity detection involved in partition phase of our proposed (K^{+})-isomorphism method; on the other hand, with minor adjustment, we utilize some current techniques as an innovative combination to finish the partition and modification of similar-community cluster in (K^{+})-isomorphism method. The experiments have shown that the improved MPD(_{-})V method has much better efficiency to search larger common subgraphs with acceptable performance compared with its prototype and other techniques. Moreover, our proposed (K^{+})-isomorphism method can achieve the K-isomorphism state with less modification of original network structure, or lower anonymization cost compared to the current K-isomorphism method.\",\n \"Title: Load Balanced GANs for Multi-view Face Image Synthesis.\\nYear: 2018\\n\\nAbstract: Multi-view face synthesis from a single image is an ill-posed problem and often suffers from serious appearance distortion. Producing photo-realistic and identity preserving multi-view results is still a not well defined synthesis problem. This paper proposes Load Balanced Generative Adversarial Networks (LB-GAN) to precisely rotate the yaw angle of an input face image to any specified angle. LB-GAN decomposes the challenging synthesis problem into two well constrained subtasks that correspond to a face normalizer and a face editor respectively. The normalizer first frontalizes an input image, and then the editor rotates the frontalized image to a desired pose guided by a remote code. In order to generate photo-realistic local details, the normalizer and the editor are trained in a two-stage manner and regulated by a conditional self-cycle loss and an attention based L2 loss. Exhaustive experiments on controlled and uncontrolled environments demonstrate that the proposed method not only improves the visual realism of multi-view synthetic images, but also preserves identity information well.\",\n \"Title: A Weighted Collaboration Network Generalization Method For Privacy Protection In C-Dblp\\nYear: 2018\\n\\nAbstract: The increasing population of online communication and telecommunication has interested scholars and researchers considering their social networks. These social networks datasets play an exceptionally important role in the research of data mining. However, large amounts of social network data are produced by using social networking applications. And these data inevitably contain a large amount of personal privacy information. Therefore, in order to avoid disclosure of privacy, the data holders need adopt privacy protection before these data are released. Furthermore, most current methods of privacy protection are based on the simple graph only. The weight values on the edges represent the tightness between the nodes. The algorithm based on weights in privacy protection field is still relatively rare. In real social networks, the weight can indicate tightness between two individuals of social relations. The weight may be as attackersar background knowledge to re-identify the target individual and lead to loss of privacy. In this paper, we consider protecting the weighted social networks from weight-based attacks and propose a method based on the weighted social networks, named k-weighted generalization anonymity (KWGA). And This method combines k-anonymous with generalization method to ensure the security of the social network data when it is published. In order to ensure the higher validity of privacy protection, this paper introduces a concept of the weight difference to reduce the modification for weight graph. Finally, we firstly use real dataset C-DBLP to verify the validity of our method perform much better than the Fast k-degree anonymity (FKDA) in average clustering coefficient (ACC), global clustering coefficient (GCC), average path length (APL) and rate of edges change four aspects. Furthermore, we also use two common datasets in the research of weighted graphs to verify our algorithm.\",\n \"Title: GLEAM: a graph clustering framework based on potential game optimization for large-scale social networks.\\nYear: 2018\\n\\nAbstract: With the growing explosion of online social networks, the study of large-scale graph clustering has attracted considerable interest. Most of traditional methods view the graph clustering problem as an optimization problem based on a given objective function; however, there are few methodical theories for the emergence of clusters over real-life networks. In this paper, each actor in online social networks is viewed as a selfish player in a non-cooperative game. The strategy associated with each node is defined as the cluster membership vector, and each one’s incentive is to maximize its own social identity by adopting the most suitable strategy. The definition of utility function in our game model is inspired by the conformity psychology, which is defined as the weighted average of one’s social identity by participating different clusters. With this setting, the proposed game can well match a potential game. So that the cluster could be shaped by the actions of those closely interactive users who adopt the same strategy in a Nash equilibrium. To this end, we propose a novel Graph cLustering framework based on potEntial gAme optiMization (GLEAM) for parallel graph clustering. It first utilize the cosine similarity to weight each edge in the original network. Then, an initial partition, including a number of clusters dominated by those potential “leader nodes”, is created by a fast heuristic process. Third, a potential game-based weighted Modularity optimization is used to improve the initial partition. Finally, we introduce the notion of potentially attractive cluster, and then discover the overlapping partition of the graph using a simple double-threshold procedure. Three phases in GLEAM are carefully designed for parallel execution. Experiments on real-world networks analyze the convergence inside GLEAM, and demonstrate the high performance of GLEAM by comparing it with the state-of-the-art community detection approaches in the literature.\",\n \"Title: H∞ filtering for networked systems with hybrid-triggered communication mechanism and stochastic cyber attacks.\\nYear: 2017\\n\\nAbstract: This paper concentrates on investigating H∞ filtering for networked systems with hybrid-triggered communication mechanism and stochastic cyber attacks. Random variables satisfying Bernoulli distribution are introduced to describe the hybrid-triggered scheme and stochastic cyber attacks, respectively. Firstly, a mathematical H∞ filtering error model with hybrid-triggered communication mechanism is constructed under the stochastic cyber attacks. Secondly, by using Lyapunov stability theory and linear matrix inequality (LMI) techniques, the sufficient conditions which can guarantee the stability of augmented filtering error system are obtained and the parameters of the designed filter can be presented in an explicit form. Finally, numerical examples are given to demonstrate the feasibility of the designed filter.\",\n \"Title: A novel POMDP-based server RAM caching algorithm for VoD systems.\\nYear: 2018\\n\\nAbstract: In contrast to other services on the Internet, streaming media service needs to fetch data from local disks more frequently, since it always lasts long and the bit rate is quite high. In addition, because of the much slower reading/writing speed of disk than random access memory (RAM), adopting advisable RAM caching policy can efficiently reduce disk I/O. In this paper, we study the problem of reducing disk I/O by using a novel approach. We first provide a new popularity estimate algorithm. Then a formal optimization problem about average disk I/O is presented, and a suboptimal caching algorithm for a special case of the problem is given. Furthermore, a partially observable Markov decision process (POMDP) model is constructed for the caching problem. Based on the model, popularity is taken advantage of to predict clients’ randomized behaviors, data replacing decisions are made when the defined observations occur, and the impact of caching actions on disk performance for future infinite steps is assessed. The method of event-based optimization is applied in search of the optimal stochastic policy. Disk I/O, as the long-run average performance measure, is optimized by applying the policy-gradient algorithm. The simulation results illustrate that data required by clients could be better predicted and lower disk I/O could be achieved by using the model proposed in this paper.\",\n \"Title: Quantized state estimation for neural networks with cyber attacks and hybrid triggered communication scheme.\\nYear: 2018\\n\\nAbstract: This paper is concerned with the issue of quantized state estimation for neural networks with cyber attacks and hybrid triggered communication scheme. In order to reduce the pressure of the network transmission and save the network resources, the hybrid triggered scheme and quantization are introduced. The hybrid triggered scheme consists of time triggered scheme and event triggered scheme, in which the stochastic switch is described by a variable satisfying Bernoulli distribution. First, by taking the effect of hybrid triggered scheme and quantization into consideration, a mathematical model for estimating the state of neural networks is constructed. Second, by using linear matrix inequality (LMI) techniques and Lyapunov stability theory, the sufficient conditions are given which can ensure the stability of estimating error system under hybrid triggered scheme, and the designing algorithm of desired state estimator is also presented in terms of LMIs. Finally, a numerical example is given to show the usefulness of the proposed approach.\",\n \"Title: Graph classification based on graph set reconstruction and graph kernel feature reduction.\\nYear: 2018\\n\\nAbstract: Graph, a kind of structured data, is widely used to model complex relationships among objects, and has been used in various of scientific and engineering fields, such as bioinformatics, network intrusion detection, social network, etc. Building an automatic and highly accurate classification method for graphs becomes quite necessary for predicting unknown graphs or understanding complex structures among different categories. The kernel method is regarded as a powerful solution to graph classification, which consists of two steps, namely, graph kernel mapping and classification. However, the feature selection process is ignored, and those sub-structures with low discriminative power result in classification accuracy decrease. In order to solve this problem, we propose an efficient graph classification algorithm based on graph set reconstruction and graph kernel feature reduction. First of all, the least discriminative frequent subgraphs and part of the infrequent subgraphs are removed to reconstruct the original graph set. Then we adopt the graph-kernel-based discriminant analysis method to perform feature reduction on the well-reconstructed graph set. At last, the whole framework of the graph classification method is introduced and any commonly used classifiers can be utilized. Extensive experimental results on a series of bioinformatics benchmarks show that our graph classification algorithm demonstrates a significant improvement of prediction comparing with other graph-kernel-based classification approaches.\",\n \"Title: Enhance the Performance of Network Computation by a Tunable Weighting Strategy.\\nYear: 2018\\n\\nAbstract: Networked systems with high computational efficiency are desired in many applications ranging from sociology to engineering. Generally, the performance of the network computation can be enhanced by two ways: rewiring and weighting. In this paper, we proposed a new two-modes weighting strategy based on the concept of communication neighbor graph, which takes use of both the local and global topolog...\",\n \"Title: A modified artificial bee colony approach for the 0-1 knapsack problem.\\nYear: 2018\\n\\nAbstract: The 0-1 knapsack problem (KP01) is one of the classical NP-hard problems in operation research and has a number of engineering applications. In this paper, the BABC-DE (binary artificial bee colony algorithm with differential evolution), a modified artificial bee colony algorithm, is proposed to solve KP01. In BABC-DE, a new binary searching operator which comprehensively considers the memory and neighbour information is designed in the employed bee phase, and the mutation and crossover operations of differential evolution are adopted in the onlooker bee phase. In order to make the searching solution feasible, a repair operator based on greedy strategy is employed. Experimental results on different dimensional KP01s verify the efficiency of the proposed method, and it gets superior performance compared with other five metaheuristic algorithms.\",\n \"Title: A variable resolution feedback improving the performances of object detection and recognition.\\nYear: 2018\\n\\nAbstract: In order to improve the performances of object detection and recognition, a two-stage framework combined with variable resolution control strategy is proposed. The images in low resolution and high resolution are employed in object detection and object recognition, respectively. Meanwhile, a feedback mechanism used in two-stage framework is proved to effectively improve the performances of object detection. The results show that under low resolution, the accuracy of object detection based on fixed resolution 320 x 240 without feedback mechanism is 36.7%. However, the accuracy of object detection of the proposed method based on variable resolution increases to 95.3%. Under high resolution, compared with the method based on fixed resolution 1280 x 960 using the two-stage framework, the time consumption of the proposed method with variable resolution decreases by 51.4%, while keeping almost identical recognition accuracy.\",\n \"Title: A Hybrid Bionic Image Sensor Achieving FOV Extension and Foveated Imaging.\\nYear: 2018\\n\\nAbstract: Based on bionic compound eye and human foveated imaging mechanisms, a hybrid bionic image sensor (HBIS) is proposed in this paper to extend the field of view (FOV) with high resolution. First, the hybrid bionic imaging model was developed and the structure parameters of the HBIS were deduced. Second, the properties of the HBIS were simulated, including FOV extension, super-resolution imaging, foveal ratio and so on. Third, a prototype of the HBIS was developed to validate the theory. Imaging experiments were carried out, and the results are in accordance with the simulations, proving the potential of the HBIS for large FOV and high-resolution imaging with low cost.\",\n \"Title: A comparative study of clustering ensemble algorithms.\\nYear: 2018\\n\\nAbstract: Since clustering ensemble was proposed, it has rapidly attracted much attention. This paper makes an overview of recent research on clustering ensemble about generative mechanism, selective clustering ensemble, consensus function and application. Twelve clustering ensemble algorithms are described and compared to choose a basic one. The experiment shows that using k-means with different initializations as generative mechanism and average-linkage agglomerative clustering as consensus function is the best one. As ensemble size increases, the performance of clustering ensemble improves. The basic clustering ensemble algorithm with suitable ensemble size is compared with clustering algorithms and the experiment shows that clustering ensemble is better than clustering. The influence of diversity on clustering ensemble is instructive to selecting members. The experiment shows that selecting members in high quality and big diversity for low-dimensional data sets, and selecting members in high quality and median diversity for high-dimensional data sets are better than traditional clustering ensemble.\",\n \"Title: A hybrid short-term traffic flow forecasting model based on time series multifractal characteristics.\\nYear: 2018\\n\\nAbstract: Short-term traffic flow forecasting is a key problem in the area of intelligent transportation systems (ITS). Timely and accurate traffic state prediction is also the prerequisite of realizing proactive traffic control and dynamic traffic assignment effectively. In this paper, a new hybrid model for short-term traffic flow forecasting, which is built based on multifractal characteristics of traffic flow time series, is proposed. The hybrid model decomposes traffic flow series into four different components, namely a periodic part, a trend part, a stationary part and a volatility part, to unearth the traffic features hidden behind the data. Four parts are treated and modeled separately by using different methods, such as spectral analysis, time series and statistical volatility analysis, to further explore the underlying traffic patterns and improve forecasting accuracy. Performance of the proposed hybrid model is investigated with traffic flow data from freeway I-694 EB in the Twin Cities. The experimental results indicate that the proposed model outperforms in capturing nonlinear volatility and improving forecasting accuracy than traditional forecasting methods, especially for the multi-step ahead forecasting. Compared with the ARIMA-GARCH model, it gets an improvement of 8.23% in RMSE for one-step ahead forecasting and 10.69% for ten-step ahead forecasting. It is better than the hybrid model newly proposed in literature (Zhang et al. Transp Res Part C: Emerg Technol 43(1):65–78 2014) and gets an improvement of 1.27% in forecasting accuracy.\",\n \"Title: Microwave Metamaterial Absorber for Non-Destructive Sensing Applications of Grain.\\nYear: 2018\\n\\nAbstract: In this work, we propose a metamaterial absorber at microwave frequencies with significant sensitivity and non-destructive sensing capability for grain samples. This absorber is composed of cross-resonators periodically arranged on an ultrathin substrate, a sensing layer filled with grain samples, and a metal ground. The cross-resonator array is fabricated using the printed circuit board process on an FR-4 board. The performance of the proposed metamaterial is demonstrated with both full-wave simulation and measurement results, and the working mechanism is revealed through multi-reflection interference theory. It can serve as a non-contact sensor for food quality control such as adulteration, variety, etc. by detecting shifts in the resonant frequencies. As a direct application, it is shown that the resonant frequency displays a significant blue shift from 7.11 GHz to 7.52 GHz when the mass fraction of stale rice in the mixture of fresh and stale rice is changed from 0% to 100%. In addition, the absorber shows a distinct difference in the resonant absorption frequency for different varieties of grain, which also makes it a candidate for a grain classification sensor. The presented scheme could open up opportunities for microwave metamaterial absorbers to be applied as efficient sensors in the non-destructive evaluation of agricultural and food product quality.\",\n \"Title: A Sequential Recommendation for Mobile Apps: What Will User Click Next App?\\nYear: 2018\\n\\nAbstract: With the advances in smartphones users install abundant apps to facilitate their daily lives. Both users and related developers have increasing requirements to understand the mobile App usage pattern, for individual and commercial use. Respectively, personalized App recommendation methods and systems have emerged as a novel attractive topic that can demonstrate the human App usage behavior. The mobile Apps recommendation can serve as a cornerstone for a variety of intelligent services, such as fast-launching UIs, intelligent user-phone interactions, and battery management of cellphones. In this paper, we develop a novel App recommendation framework combining the historical App usage data with the sequence of recently-used Apps. Specifically, our framework is an extension of the user-based collaborative filtering technique, where the set of nearest neighbors is employed for training the prediction model. However, our prediction scheme is constructed on the temporal sequential data and is modeled by using the chain-augmented Naive Bayes model. Experiments with a real mobile Internet record dataset demonstrate that the accuracy of our framework outperforms several baseline App recommendation approaches.\",\n \"Title: Understanding Customer Behavior In Shopping Mall From Indoor Tracking Data\\nYear: 2018\\n\\nAbstract: The prosperity of various indoor positioning technologies makes possible the large collection of tracking data in indoor spaces. Much of the focus has been on several fundamental problems such as indoor localization, indoor space modeling, indoor data cleansing, indexing and querying. However, this paper attempts to analyze customer behavior from a unique indoor tracking data, which will promote the convergence between various applications and the underlying data. In particular, we introduce a real-life indoor tracking data collected at an electrical mall in China. Then, we cluster users into several groups and summarize the most characteristic behaviors of each cluster. Last but not least, we analyze customer's individual behaviors through two aspects: 1) the regression model is used to reveal hot regions where customers are likely to stay in long time; and 2) a transition matrix combined with the connectivity is presented to demonstrate hot paths.\",\n \"Title: Skip-connection convolutional neural network for still image crowd counting.\\nYear: 2018\\n\\nAbstract: In recent years, crowd counting in still images has attracted many research interests due to its applications in public safety. However, it remains a challenging task for reasons of perspective and scale variations. In this paper, we propose an effective Skip-connection Convolutional Neural Network (SCNN) for crowd counting to overcome the issue of scale variations. The proposed SCNN architecture consists of several multi-scale units to extract multi-scale features. Each multi-scale unit including three convolutional layers builds connections between the input and each convolutional layer. In addition, we propose a scale-related training method to improve the accuracy and robustness of crowd counting. We evaluate our method on three crowd counting benchmarks. Experimental results verify the efficiency of the proposed method, and it achieves superior performance compared with other methods.\",\n \"Title: A multivariate short-term traffic flow forecasting method based on wavelet analysis and seasonal time series.\\nYear: 2018\\n\\nAbstract: Short-term traffic flow forecasting is a key step to achieve the performance of intelligent transportation system (ITS). Timely and accurate traffic information prediction is also the prerequisite of realizing proactive traffic control and dynamic traffic assignment effectively. Based on the fact that univariate forecasting methods have limited forecasting abilities when the data is missing or erroneous and that single models make no full use of information underline data, a new hybrid method with multivariate for short-term traffic flow forecasting is proposed. This method combines statistical analysis method with computational intelligence techniques to mine the characteristic of traffic flow as well as forecast short-term traffic state. First, the wavelet de-noising is employed to remove the noise information. Then, time series analysis is used to analyze time-varying and periodic characteristic of traffic flow. Furthermore, the seasonal auto-regressive moving average with external input (SARIMAX) is established to fit traffic flow with occupancy as exogenous variables. Finally, wavelet forecast is adopted to forecast the values of occupancy which are used as exogenous input, and a WSARIMAX is constructed to forecast traffic flow. Using the relationship of flow and occupancy at the same road section and taking traffic flow and occupancy data from freeway I-694 EB in the Twin Cities as endogenous variables and exogenous variables respectively, this paper studies the forecasting performance of the proposed method. The study results are encouraging. Compared with SARIMA newly proposed in literature, WSARIMA and SARIMAX improved method with wavelet analysis and multivariate modeling method, the proposed method gets improvements of 12.95%, 12.62% and 10.41% in forecasting accuracy of one-step ahead respectively. For ten-steps ahead forecasting, it gets improvements of 18.87%, 17.05% and 2.57% in forecasting accuracy respectively.\",\n \"Title: Few-shot learning for short text classification.\\nYear: 2018\\n\\nAbstract: Due to the limited length and freely constructed sentence structures, it is a difficult classification task for short text classification. In this paper, a short text classification framework based on Siamese CNNs and few-shot learning is proposed. The Siamese CNNs will learn the discriminative text encoding so as to help classifiers distinguish those obscure or informal sentence. The different sentence structures and different descriptions of a topic are viewed as ‘prototypes’, which will be learned by few-shot learning strategy to improve the classifier’s generalization. Our experimental results show that the proposed framework leads to better results in accuracies on twitter classifications and outperforms some popular traditional text classification methods and a few deep network approaches.\",\n \"Title: Recombined multinomial tree based on saddle-point approximation and its application to Levy models options pricing.\\nYear: 2019\\n\\nAbstract: This paper studies the constructing methods of a recombined multinomial tree based on saddle-point approximation and its application to Levy models options pricing. Firstly, the Levy process and the European option pricing are introduced. Then, we used the characteristic function of Levy process to generate density function value at discrete point in a certain range, based on the saddle-point approximation method. Then, we provide the method to construct recombined multinomial tree and give the pricing formula of European option and path-dependent option pricing based on the backward iteration. Finally, we used the CGMY process to demonstrate its application to European option, American option and American barrier option pricing. It proves that saddle-point approximation turns the inverse Fourier integral transform problem to several function value calculation. Comparing to IFFT, the speed of calculation using this method is faster, and it avoid the negative probability density function value based on IFFT. Because of the linear growth of the number of nodes, we can extend the saddle-point approximation method to the calculation of path-dependent option.\",\n \"Title: Parallel Attribute Reduction Algorithm for Complex Heterogeneous Data Using MapReduce.\\nYear: 2018\\n\\nAbstract: Parallel attribute reduction is one of the most important topics in current research on rough set theory. Although some parallel algorithms were well documented, most of them are still faced with some challenges for effectively dealing with the complex heterogeneous data including categorical and numerical attributes. Aiming at this problem, a novel attribute reduction algorithm based on neighborhood multigranulation rough sets was developed to process the massive heterogeneous data in the parallel way. The MapReduce-based parallelization method for attribute reduction was proposed in the framework of neighborhood multigranulation rough sets. To improve the reduction efficiency, the hashing Map/Reduce functions were designed to speed up the positive region calculation. Thereafter, a quick parallel attribute reduction algorithm using MapReduce was developed. The effectiveness and superiority of this parallel algorithm were demonstrated by theoretical analysis and comparison experiments.\",\n \"Title: A pattern-based topic detection and analysis system on Chinese tweets.\\nYear: 2018\\n\\nAbstract: •A FP-growth-like algorithm to extract cosine interesting patterns as topics.•A distributed topic detecting algorithm on Spark.•Evolutionary and sentimental analysis on the detected topics.\",\n \"Title: Dynamic Cluster Formation Game for Attributed Graph Clustering.\\nYear: 2019\\n\\nAbstract: Besides the topological structure, there are additional information, i.e., node attributes, on top of the plain graphs. Usually, these systems can be well modeled by attributed graphs, where nodes represent component actors, a set of attributes describe users' portraits and edges indicate their connections. An elusive question associated with attributed graphs is to study how clusters with common ...\",\n \"Title: hPSD: A Hybrid PU-Learning-Based Spammer Detection Model for Product Reviews.\\nYear: 2020\\n\\nAbstract: Spammers, who manipulate online reviews to promote or suppress products, are flooding in online commerce. To combat this trend, there has been a great deal of research focused on detecting review spammers, most of which design diversified features and thus develop various classifiers. The widespread growth of crowdsourcing platforms has created large-scale deceptive review writers who behave more like normal users, that the way they can more easily evade detection by the classifiers that are purely based on fixed characteristics. In this paper, we propose a hybrid semisupervised learning model titled hybrid PU-learning-based spammer detection (hPSD) for spammer detection to leverage both the users’ characteristics and the user–product relations. Specifically, the hPSD model can iteratively detect multitype spammers by injecting different positive samples, and allows the construction of classifiers in a semisupervised hybrid learning framework. Comprehensive experiments on movie dataset with shilling injection confirm the superior performance of hPSD over existing baseline methods. The hPSD is then utilized to detect the hidden spammers from real-life Amazon data. A set of spammers and their underlying employers (e.g., book publishers) are successfully discovered and validated. These demonstrate that hPSD meets the real-world application scenarios and can thus effectively detect the potentially deceptive review writers.\",\n \"Title: Enabling Semantics in oneM2M Service Delivery Platform\\nYear: 2018\\n\\nAbstract: With the burgeoning of Edge computing, data from the Internet of Things (IoT) and Machine-to-Machine (M2M) systems are shared and consumed at the edge of the network. How to understand and leverage the data from other systems is a crucial challenge for information exchange and communication among IoT/M2M. Semantic paved the road for the knowledge-driven IoT/M2M systems. Context awareness becomes possible with the help of semantic information. In this paper, we proposed the methods of support semantic service in a sophisticated IoT/M2M system with reasonable overhead. Moreover, we also discussed the approaches to support a secure mechanism such as access control directly within the semantic information to boost the performance of the system. Our experiment shows that the proposed method could effectively and securely provide semantic service to the IoT/M2M system.\",\n \"Title: A Comparative Study of Subgraph Matching Isomorphic Methods in Social Networks.\\nYear: 2018\\n\\nAbstract: With the fast development of social networks, more and more data has been generated. Finding useful information among these data is important. Subgraph matching is the method that can be used in social networks for social search, recommender systems, and so on. In this paper, we choose five typical graph matching methods which are VF2, SPath, Turbo(ISO), BoostIso, and RI to value their performance and scalability in social networks. These methods are verified by three social network data sets whose node's number vary from thousands to millions. According to the experiments, we can find that VF2 and RI is applicable for rather small graphs while SPath performs better in large graph when average degree of graph is small. When graph with high average degree, Turbo(ISO) and BoostIso performs better. What is more, we also compare the efficiency of different matching order when searching. The order choosing strategy proposed by RI is better.\",\n \"Title: Improving lazy decision tree for imbalanced classification by using skew-insensitive criteria\\nYear: 2019\\n\\nAbstract: Lazy decision tree (LazyDT) constructs a customized decision tree for each test instance, which consists of only a single path from the root to a leaf node. LazyDT has two strengths in comparison with eager decision trees. One is that LazyDT can build shorter decision paths than eager decision trees, and the other is that LazyDT can avoid unnecessary data fragmentation. However, the split criterion used for constructing a customized tree in LazyDT is information gain, which is skew-sensitive. When learning from imbalanced data sets, class imbalance impedes their ability to learn the minority class concept. In this paper, we use Hellinger distance and K-L divergence as split criteria to build two types of lazy decision trees. An experimental framework is performed across a wide range of imbalanced data sets to investigate the effectiveness of our methods when comparing with the other methods including lazy decision tree, C4.5, Hellinger distance based decision tree and support vector machine. In addition, we also use SMOTE to preprocess the highly imbalance data sets in the experiment and evaluate its effectiveness. The experimental results, which contrasted through nonparametric statistical tests, demonstrate that using Hellinger distance and K-L divergence as the split criterion can improve the performances of LazyDT for imbalanced classification effectively.\",\n \"Title: Feature Selection Using Forest Optimization Algorithm Based On Contribution Degree\\nYear: 2018\\n\\nAbstract: As a combinatorial optimization problem, feature selection has been widely used in machine learning and data mining In this paper, a feature selection method using forest optimization algorithm based on contribution degree is proposed. The proposed method uses a contribution degree strategy which is embedded in forest optimization algorithm. The goal of the contribution degree is to guide the search process of the forest optimization algorithm to select features according to high class correlation and low redundancy between features. The proposed algorithm is verified on some data sets from the UCI repository and the experiments show that the proposed method improves the classification accuracy compared with some other methods.\",\n \"Title: Combining Non-Uniform Time Slice and Finite Difference to Improve 3D Ghost Imaging.\\nYear: 2019\\n\\nAbstract: Three-dimensional ghost imaging (3DGI) using a detector is widely used in many applications. The performance of 3DGI based on a uniform time slice is difficult to improve because obtaining an accurate time-slice position remains a challenge. This paper reports a novel structure based on non-uniform time slice combined with finite difference. In this approach, finite difference is beneficial to improving sensitivity of zero crossing to accurately obtain the position of the target in the field of view. Simultaneously, non-uniform time slice is used to quickly obtain 3DGI on an interesting target. Results show that better performances of 3DGI are obtained by our proposed method compared to the traditional method. Moreover, the relation between time slice and the signal-noise-ratio of 3DGI is discussed, and the optimal differential distance is obtained, thus motivating the development of a high-performance 3DGI.\",\n \"Title: Buoyancy Force Acting on Underground Structures considering Seepage of Confined Water.\\nYear: 2019\\n\\nAbstract: The antifloating property of underground structures in areas with high underground water levels is a key design aspect. Evaluating the buoyancy forces acting on underground structures is complicated, particularly in the presence of confined water beneath the structures. Herein, the effects of the permeability coefficient of layered soil, hydraulic gradient, and embedment depth of the aquiclude on the buoyancy force acting on underground structures are investigated through three model tests: (1) calibration of the test system, (2) buoyancy force acting on a structure located in homogeneous soil considering vertical direction seepage, and (3) buoyancy force acting on a structure located in layered soil considering vertical seepage of confined water. The results show that the pore pressure along the structure and the buoyancy force acting on the underground structure considering seepage are greater than those obtained under hydrostatic conditions. The raising ratios of the pore pressure and buoyancy force are equal to the vertical hydraulic gradient when seepage occurs in homogeneous soil. In the presence of confined water, the raising ratio is significantly greater than the hydraulic gradient. In the cases studied herein, the raising ratio is approximately twice the hydraulic gradient. Simplified equations are proposed to calculate the buoyancy force acting on underground structures considering the vertical seepage of confined water. Finally, a finite element analysis is carried out to verify the conclusions obtained from the model test and the rationality of the proposed equations.\",\n \"Title: Location-Centered House Price Prediction: A Multi-Task Learning Approach\\nYear: 2019\\n\\nAbstract: AbstractAccurate house prediction is of great significance to various real estate stakeholders such as house owners, buyers, and investors. We propose a location-centered prediction framework that differs from existing work in terms of data profiling and prediction model. Regarding data profiling, we make an important observation as follows – besides the in-house features such as floor area, the location plays a critical role in house price prediction. Unfortunately, existing work either overlooked it or had a coarse grained measurement of locations. Thereby, we define and capture a fine-grained location profile powered by a diverse range of location data sources, including transportation profile, education profile, suburb profile based on census data, and facility profile. Regarding the choice of prediction model, we observe that a variety of approaches either consider the entire data for modeling, or split the entire house data and model each partition independently. However, such modeling ignores the relatedness among partitions, and for all prediction scenarios, there may not be sufficient training samples per partition for the latter approach. We address this problem by conducting a careful study of exploiting the Multi-Task Learning (MTL) model. Specifically, we map the strategies for splitting the entire house data to the ways the tasks are defined in MTL, and select specific MTL-based methods with different regularization terms to capture and exploit the relatedness among tasks. Based on real-world house transaction data collected in Melbourne, Australia, we design extensive experimental evaluations, and the results indicate a significant superiority of MTL-based methods over state-of-the-art approaches. Meanwhile, we conduct an in-depth analysis on the impact of task definitions and method selections in MTL on the prediction performance, and demonstrate that the impact of task definitions on prediction performance far exceeds that of method selections.\",\n \"Title: Improving task scheduling with parallelism awareness in heterogeneous computational environments.\\nYear: 2019\\n\\nAbstract: Task scheduling is a key function for executing tasks in heterogeneous computational environments, efficiently. While the available computing resources are not fully used when applying existing scheduling methods as they consider that a task is executed on one single core or on a server without parallel tasks by assuming that the task exhausts the server. Therefore, in this paper, we focus on the problem of executing tasks with deadline constraints with parallelism awareness where the parallel degree of each task can be tuned between one and its maximum according to the available cores of the server it assigned to during its execution. We first model the problem as an optimization problem maximizing the overall utilization of servers, and propose a set of scheduling methods with parallelism awareness (SPA), each of which iteratively allocates as much resources and as soon as possible to the assigned task with the earliest deadline on a server, based on existing scheduling algorithms, and present two SPA instances to illustrate the implement of SPA. Experiment results show a great performance improvement in various aspects, e.g., resource utilization, task violations, finish time, and energy efficiency, when executing tasks heterogeneous computational systems using SPA.\",\n \"Title: CAMAS: A cluster-aware multiagent system for attributed graph clustering.\\nYear: 2017\\n\\nAbstract: Abstract Attributed graphs describe nodes via attribute vectors and also relationships between different nodes via edges. To partition nodes into clusters with tighter correlations, an effective way is applying clustering techniques on attributed graphs based on various criteria such as node connectivity and/or attribute similarity. Even though clusters typically form around nodes with tight edges and similar attributes, existing methods have only focused on one of these two data modalities. In this paper, we comprehend each node as an autonomous agent and develop an accurate and scalable multiagent system for extracting overlapping clusters in attributed graphs. First, a kernel function with a tunable bandwidth factor δ is introduced to measure the influence of each agent, and those agents with highest local influence can be viewed as the “leader” agents. Then, a novel local expansion strategy is proposed, which can be applied by each leader agent to absorb the most relevant followers in the graph. Finally, we design the cluster-aware multiagent system (CAMAS), in which agents communicate with each other freely under an efficient communication mechanism. Using the proposed multiagent system, we are able to uncover the optimal overlapping cluster configuration, i.e. nodes within one cluster are not only connected closely with each other but also with similar attributes. Our method is highly efficient, and the computational time is shown that nearly linearly dependent on the number of edges when δ ∈ [0.5, 1). Finally, applications of the proposed method on a variety of synthetic benchmark graphs and real-life attributed graphs are demonstrated to verify the systematic performance.\",\n \"Title: Efficiently Translating Complex SQL Query to MapReduce Jobflow on Cloud\\nYear: 2020\\n\\nAbstract: MapReduce is a widely-used programming model in cloud environment for parallel processing large-scale data sets. The combination of the high-level language with a SQL-to-MapReduce translator allows programmers to code using SQL-like declarative language, so that each program can afterwards be complied into a MapReduce jobflow automatically. This way is helpful to narrow the gap between non-profess...\",\n \"Title: Emergency Rescue Vehicle Dispatch Planning Using A Hybrid Algorithm\\nYear: 2018\\n\\nAbstract: This paper describes the emergency rescue vehicle transportation network within the entire rescue period, and imitates rescue vehicle to select rescue route and to allocate emergency resource. The presented emergency rescue vehicle dispatch model seeks to minimize rescue time as the first objective function, minimize delay cost as the second objective function and maximize lifesaving utility as the last objective function in disaster response operations. To solve the proposed multiple objective model, a hybrid algorithm named nondominated sorting genetic algorithm (NSGA-II) with ant colony algorithm and a NSGA-II with random crossover and mutation, which can find better initial solution, are presented. In order to further prove the validity of the model and algorithm, a more complicated case is cited. Computational results are reported to illustrate the performance of the proposed model and algorithm. Statistical analysis confirms that the proposed random crossover and mutation operator outperforms the original crossover and mutation operator. The sensitivity analysis proves which parameter is more important for objective function values.\",\n \"Title: Understanding Crowdsourcing Systems from a Multiagent Perspective and Approach.\\nYear: 2018\\n\\nAbstract: Crowdsourcing has recently been significantly explored. Although related surveys have been conducted regarding this subject, each has mainly consisted of a review of a single aspect of crowdsourcing systems or on the application of crowdsourcing in a specific application domain. A crowdsourcing system is a comprehensive set of multiple entities, including various elements and processes. Multiagent computing has already been widely envisioned as a powerful paradigm for modeling autonomous multi-entity systems with adaptation to dynamic environments. Therefore, this article presents a novel multiagent perspective and approach to understanding crowdsourcing systems, which can be used to correlate the research on crowdsourcing and multiagent systems and inspire possible interdisciplinary research between the two areas. This article mainly discusses the following two aspects: (1) The multiagent perspective can be used for conducting a comprehensive survey on the state of the art of crowdsourcing, and (2) the multiagent approach can bring about concrete enhancements for crowdsourcing technology and inspire future research directions that enable crowdsourcing research to overcome the typical challenges in crowdsourcing technology. Finally, this article discusses the advantages and disadvantages of the multiagent perspective by comparing it with two other popular perspectives on crowdsourcing: the business perspective and the technical perspective.\",\n \"Title: An Efficient Distributed-Computing Framework for Association-Rule-Based Recommendation\\nYear: 2018\\n\\nAbstract: The association-rule-based recommendation model is one of the most widely used commercial recommendation engines in e-commerce websites. Existing studies mostly focus on how to select eligible rules to enhance the recommendation performance, but the efficiency of recommendation has been paid few attentions. To remedy this, this paper develops a distributed-computing framework for improving the computational efficiency of rule-based recommendation. Specifically, a tree-typed structure called Ordered-Patterns Forest (OPF) is designed to compress and store frequent patterns. Then, we transform eligible rules mining to a path-searching problem on OPF, and present a path-searching algorithm running on single machine. Finally, a load-balanced strategy for data partitioning is clarified. Experimental results demonstrate that the efficiency improved remarkably by the proposed OPF, compared with the traditional Brute-Force method.\",\n \"Title: Geometry-Aware Face Completion and Editing.\\nYear: 2018\\n\\nAbstract: Face completion is a challenging generation task because it requires generating visually pleasing new pixels that are semantically consistent with the unmasked face region. This paper proposes a geometry-aware Face Completion and Editing NETwork (FCENet) by systematically studying facial geometry from the unmasked region. Firstly, a facial geometry estimator is learned to estimate facial landmark heatmaps and parsing maps from the unmasked face image. Then, an encoder-decoder structure generator serves to complete a face image and disentangle its mask areas conditioned on both the masked face image and the estimated facial geometry images. Besides, since low-rank property exists in manually labeled masks, a low-rank regularization term is imposed on the disentangled masks, enforcing our completion network to manage occlusion area with various shape and size. Furthermore, our network can generate diverse results from the same masked input by modifying estimated facial geometry, which provides a flexible mean to edit the completed face appearance. Extensive experimental results qualitatively and quantitatively demonstrate that our network is able to generate visually pleasing face completion results and edit face attributes as well.\",\n \"Title: the Algorithm of Warehouse Vehicle Trajectory Intelligent Identification\\nYear: 2018\\n\\nAbstract: Vehicle location is the key in the warehouse vehicle trajectory identification and warning, RFID signal transmission model in a closed warehouse environment in the pseudo range includes not only the RFID signal generated by the NLOS transmission distance of unknown offset, also includes the measurement of actual measurement error exists. Therefore, this paper uses the pseudo range positioning method to realize the vehicle location in the warehouse. In the study of pseudo range positioning method, position information by solving hyperbolic equations to obtain the warehouse of the vehicle, the vehicle positioning contains many constraint optimization problems, and using the distance difference method to reduce the impact of the unknown displacement and the measurement error of vehicle positioning in the process of the warehouse, to improve the positioning accuracy of warehouse vehicles.\",\n \"Title: Research on Agricultural Environment Prediction Based on Deep Learning\\nYear: 2018\\n\\nAbstract: The environmental security of agriculture is closely related to human beings. Analytical training of agricultural environmental data, forecasting its development trend, has positive significance for the protection of the safety of agricultural products. This paper proposes an agricultural environment prediction model based on deep learning LSTM (Long Short-Term Memory). By analyzing the agricultural environment parameters of the current period, the environmental parameters of the next moment can be predicted to achieve the purpose of early warning. The experimental results show that the model’s prediction results have little deviation from the actual values; on this basis, the LSTM model is optimized to replace LSTM with GRU(Gated Recurrent Unit), and the model is more effective.\",\n \"Title: Maintenance and Management of Marine Communication and Navigation Equipment Based on Virtual Reality\\nYear: 2018\\n\\nAbstract: According to the present situation and existing problems of marine communication and navigation equipment maintenance and management, this paper expounds the virtual reality technology in equipment maintenance and management. We design and implement a VR based maintenance and management system for marine communication and navigation equipment. The system consists of four parts: VR editing module, VR training module, user behavior analysis module, VR intelligent deduction module and user management module function module. The design and hardware structure of the virtual system is given, and the management task modules on this platform are designed, finally realized the function simulation. Functional verification proved the system can effectively improve the equipment management level of electro-technical officer (ETO).\",\n \"Title: Deep rolling: A novel emotion prediction model for a multi-participant communication context.\\nYear: 2019\\n\\nAbstract: •Propose a novel emotion prediction model, Deep Rolling, for target participant in a multi-participant communication context. The proposed model can capture the emotional stimulation caused by other participants in the current communication context.•Introduce data factorization into the proposed model, Deep Rolling, to enhance the prediction efficiency.•By adopting movie subtitle as dataset, Deep Rolling is utilized to find roles who have the most important emotional impact on the prediction target.\",\n \"Title: SSE: A New Selective Initialization Strategy for Snapshot Ensembling\\nYear: 2018\\n\\nAbstract: Deep neural network may suffer large variance in the task of small-sample classification. To overcome this problem, and motivated by that ensemble method is an effective way of reducing variance, this paper proposes a new ensemble method called Selective Snapshot Ensembling, which introduces a new selective initialization strategy into Snapshot Ensembling. Snapshot Ensembling, an ensemble method of neural networks, has been demonstrated its effectiveness on many datasets. The proposed method is able to discard some poor initializations of network parameters, and taking an example of cross-entropy loss, poor initializations here refer to those initializations which cannot make the network converge to a very small value of loss. In particular, we set a threshold of epochs in Snapshot Ensembling to monitor and select good initialization. Experimental results on two small-sample datasets, the LabelMe dataset and the 80 AI-Challeger dataset, demonstrate that the proposed ensemble method outperforms Snapshot Ensembling and other baseline methods.\",\n \"Title: Knowledge Granularity Based Incremental Attribute Reduction for Hybrid Attribute Data\\nYear: 2018\\n\\nAbstract: Traditional attribute reduction algorithms cannot dynamically update the results of the reduction effectively. The existing incremental attribute reduction algorithm compensates, to a certain extent, the defects of the traditional attribute algorithm, but most of them focused on either categorical attributes data or numerical attributes data simply. In order to process the dynamic data with hybrid attributes, and by combining the advantages of the neighborhood rough sets model that it processes hybrid attributes effectively, a knowledge granularity calculating method based on the neighborhood rough set model is proposed. Through dynamically updating the neighborhood particles to incrementally get the knowledge granularity. According to the knowledge granularity, an incremental reduction algorithm is further developed. The validity of the proposed algorithm is demonstrated by experimental analysis.\",\n \"Title: Simulation of Continuous Enterprise Energy Consumption Process with Uncertain Factors\\nYear: 2018\\n\\nAbstract: Continuous enterprise energy consumption process is a hybrid system with diverse structure such as parallel, serial and circuitous connection etc., and influenced by various uncertain factors. To analyze the energy consuming situation more accurately, based on the model of continuous enterprise energy consumption process with uncertain factors, the simulation method for continuous enterprise energy consumption process with uncertain factors was presented, and the simulation algorithm, in which the steady running range and the instantaneous firing speed can change dynamically, was proposed in detail. The effectiveness and superiority of this method were illustrated by simulation results.\",\n \"Title: Compressed binary discernibility matrix based incremental attribute reduction algorithm for group dynamic data.\\nYear: 2019\\n\\nAbstract: The datasets in real-world applications often vary dynamically over time. Moreover, datasets often expand by introducing a group of data in many cases rather than a single object one by one. The traditional incremental attribute reduction approaches for a single dynamic object may not be applied to such cases. Focusing on this issue, a compressed binary discernibility matrix is introduced and an incremental attribute reduction algorithm for group dynamic data is developed. The single dynamic object and the group dynamic objects are both considered in this algorithm. According to the dynamic data is a single object or a group of objects, different branches can be chosen to update the compressed binary discernibility matrix. Thereafter, the incremental reduction result can be obtained based on the updated compressed binary discernibility matrix. The validity of this algorithm is demonstrated by simulation and experimental analysis.\",\n \"Title: Large-Margin Regularized Softmax Cross-Entropy Loss.\\nYear: 2019\\n\\nAbstract: Softmax cross-entropy loss with L2 regularization is commonly adopted in the machine learning and neural network community. Considering that the traditional softmax cross-entropy loss simply focuses on fitting or classifying the training data accurately but does not explicitly encourage a large decision margin for classification, some loss functions are proposed to improve the generalization performance by solving the problem. However, these loss functions enhance the difficulty of model optimization. In addition, inspired by regularized logistic regression, where the regularized term is responsible for adjusting the width of decision margin, which can be seen as an approximation of support vector machine, we proposed a large-margin regularization method for softmax cross-entropy loss. The advantages of the proposed loss are twofold as follows: the first is the generalization performance improvement, and the second is easy optimization. The experimental results on three small-sample datasets show that our regularization method achieves good performance and outperforms the existing popular regularization methods of neural networks.\",\n \"Title: Cross-spectral Face Completion for NIR-VIS Heterogeneous Face Recognition.\\nYear: 2019\\n\\nAbstract: Near infrared-visible (NIR-VIS) heterogeneous face recognition refers to the process of matching NIR to VIS face images. Current heterogeneous methods try to extend VIS face recognition methods to the NIR spectrum by synthesizing VIS images from NIR images. However, due to self-occlusion and sensing gap, NIR face images lose some visible lighting contents so that they are always incomplete compared to VIS face images. This paper models high resolution heterogeneous face synthesis as a complementary combination of two components, a texture inpainting component and pose correction component. The inpainting component synthesizes and inpaints VIS image textures from NIR image textures. The correction component maps any pose in NIR images to a frontal pose in VIS images, resulting in paired NIR and VIS textures. A warping procedure is developed to integrate the two components into an end-to-end deep network. A fine-grained discriminator and a wavelet-based discriminator are designed to supervise intra-class variance and visual quality respectively. One UV loss, two adversarial losses and one pixel loss are imposed to ensure synthesis results. We demonstrate that by attaching the correction component, we can simplify heterogeneous face synthesis from one-to-many unpaired image translation to one-to-one paired image translation, and minimize spectral and pose discrepancy during heterogeneous recognition. Extensive experimental results show that our network not only generates high-resolution VIS face images and but also facilitates the accuracy improvement of heterogeneous face recognition.\",\n \"Title: Biphasic Learning of GANs for High-Resolution Image-to-Image Translation.\\nYear: 2019\\n\\nAbstract: Despite that the performance of image-to-image translation has been significantly improved by recent progress in generative models, current methods still suffer from severe degradation in training stability and sample quality when applied to the high-resolution situation. In this work, we present a novel training framework for GANs, namely biphasic learning, to achieve image-to-image translation in multiple visual domains at $1024^2$ resolution. Our core idea is to design an adjustable objective function that varies across training phases. Within the biphasic learning framework, we propose a novel inherited adversarial loss to achieve the enhancement of model capacity and stabilize the training phase transition. Furthermore, we introduce a perceptual-level consistency loss through mutual information estimation and maximization. To verify the superiority of the proposed method, we apply it to a wide range of face-related synthesis tasks and conduct experiments on multiple large-scale datasets. Through comprehensive quantitative analyses, we demonstrate that our method significantly outperforms existing methods.\",\n \"Title: Dual Cross-Entropy Loss for Small-Sample Fine-Grained Vehicle Classification\\nYear: 2019\\n\\nAbstract: Fine-grained vehicle classification is a challenging topic in computer vision due to the high intraclass variance and low interclass variance. Recently, considerable progress has been made in fine-grained vehicle classification due to the huge success of deep neural networks. Most studies of fine-grained vehicle classification based on neural networks, focus on the neural network structure to improve the classification performance. In contrast to existing works on fine-grained vehicle classification, we focus on the loss function of the neural network. We add a regularization term to the cross-entropy loss and propose a new loss function, \\n<italic xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">Dual Cross-Entropy Loss</italic>\\n. The regularization term places a constraint on the probability that a data point is assigned to a class other than its ground-truth class, which can alleviate the vanishing of the gradient when the value of the cross-entropy loss is close to zero. To demonstrate the effectiveness of our loss function, we perform two sets of experiments. The first set is conducted on a small-sample fine-grained vehicle classification dataset, the Stanford Cars-196 dataset. The second set is conducted on two small-sample datasets, the LabelMe dataset and the UIUC-Sports dataset, as well as on one large-sample dataset, the CIFAR-10 dataset. The experimental results show that the proposed loss function improves the fine-grained vehicle classification performance and has good performance on three other general image classification tasks.\",\n \"Title: 3D Aided Duet GANs for Multi-view Face Image Synthesis\\nYear: 2019\\n\\nAbstract: Multi-view face synthesis from a single image is an ill-posed computer vision problem. It often suffers from appearance distortions if it is not well-defined. Producing photo-realistic and identity preserving multi-view results is still a not well-defined synthesis problem. This paper proposes 3D aided duet generative adversarial networks (AD-GAN) to precisely rotate the yaw angle of an input face...\",\n \"Title: Batch Allocation for Tasks with Overlapping Skill Requirements in Crowdsourcing\\nYear: 2019\\n\\nAbstract: Existing studies on crowdsourcing often adopt the retail-style allocation approach, in which tasks are allocated individually and independently. However, such retail-style task allocation has the following problems: 1) each task is executed independently from scratch, thus the execution of one task seldom utilize the results of other tasks and the requester must pay in full for the task; 2) many workers only undertake a very small number of tasks contemporaneously, thus the workers’ skills and time may not be fully utilized. We observe that many complex tasks in real-world crowdsourcing platforms have similar skill requirements and long deadlines. Based on these real-world observations, this paper presents a novel batch allocation approach for tasks with overlapping skill requirements. Requesters’ real payment can be discounted because the real execution cost of tasks can be reduced due to batch allocation and execution, and each worker's real earnings may increase because he/she can undertake more tasks contemporaneously. This batch allocation optimization problem is proved to be NP-hard. Then, two types of heuristic approaches are designed: layered batch allocation and core-based batch allocation. The former approach mainly utilizes the hierarchy pattern to form all possible batches, which can achieve better performance but may require higher computational cost since all possible batches are formed and observed; the latter approach selects core tasks to form batches, which can achieve suboptimal performance with lower complexity and significantly reduce computational cost. With the theoretical analyses and experiments on a real-world Upwork dataset in which the proposed approaches are compared with the previous benchmark retail-style allocation approach, we find that our approaches have better performances in terms of total payment by requesters and average income of workers, as well as maintaining close successful task completion probability and consuming less task allocation time.\",\n \"Title: Effects of pore structure and salinity on the imbibition of shale samples using physical simulation and NMR technique: A case from Chang 7 shale, Ordos basin\\nYear: 2021\\n\\nAbstract: AbstractA novel experimental study is conducted to reveal the effect of pore structure and salinity on spontaneous imbibition of shale samples using field emission scanning electron microscopy and low-field nuclear magnetic resonance (NMR). Core samples selected from Chang 7 shale, Ordos basin, contain different proportions of mineral components and exhibit different pore structures, as demonstrated by scanning electron microscopy. During the imbibition process, T2 spectra are obtained for each core sample under different salinities and imbibition time periods using NMR. The results are then used to characterize and compare the water imbibition process in different shale samples, with a focus on investigating the effects of pore structure and salinity. The results show that the amount of absorption water is positively correlated with the development of microcracks and with large pores in shale samples. The shale samples are shown to develop more microcracks as the imbibition time increases and the salinity decreases. These microcracks can improve the samples’ capacity of imbibition effectively, therefore increasing water absorption. Moreover, it is indicated that the imbibition capacity increases with an increasing content of illite clay minerals in the shale samples. Furthermore, the greatest amount of adsorbed water is seen when the water salinity is as high as the synthetic formation water. This is important in choosing hydraulic fracturing fluids, knowing that high-salinity fracturing fluids can effectively reduce water absorption in shale, therefore contributing to low flow-back and better fracturing performance.\",\n \"Title: Effects of pore structure and salinity on the imbibition of shale samples using physical simulation and NMR technique: A case from Chang 7 shale, Ordos basin\\nYear: 2021\\n\\nAbstract: AbstractA novel experimental study is conducted to reveal the effect of pore structure and salinity on spontaneous imbibition of shale samples using field emission scanning electron microscopy and low-field nuclear magnetic resonance (NMR). Core samples selected from Chang 7 shale, Ordos basin, contain different proportions of mineral components and exhibit different pore structures, as demonstrated by scanning electron microscopy. During the imbibition process, T2 spectra are obtained for each core sample under different salinities and imbibition time periods using NMR. The results are then used to characterize and compare the water imbibition process in different shale samples, with a focus on investigating the effects of pore structure and salinity. The results show that the amount of absorption water is positively correlated with the development of microcracks and with large pores in shale samples. The shale samples are shown to develop more microcracks as the imbibition time increases and the salinity decreases. These microcracks can improve the samples’ capacity of imbibition effectively, therefore increasing water absorption. Moreover, it is indicated that the imbibition capacity increases with an increasing content of illite clay minerals in the shale samples. Furthermore, the greatest amount of adsorbed water is seen when the water salinity is as high as the synthetic formation water. This is important in choosing hydraulic fracturing fluids, knowing that high-salinity fracturing fluids can effectively reduce water absorption in shale, therefore contributing to low flow-back and better fracturing performance.\",\n \"Title: Detecting Prosumer-Community Groups in Smart Grids From the Multiagent Perspective\\nYear: 2019\\n\\nAbstract: One of the greatest advancements of the modern era is the evolution of smart grid (SG), which integrates information communication technologies with advanced power electronic technologies to cope with the global energy shortage. The users in SGs are often called the “prosumers,” who not only consume energy but also generate the energy and share it with the utility grid or with other energy consumers. In order to promote sustainable prosumer management in SGs, one of the feasible strategies is to aggregate the prosumers from different locations, but with similar energy behaviors and cohesive interconnections, such groups of prosumers are also called the prosumer-community groups (PCGs). The contribution of this paper is threefold. First, we provide a generalized definition of individual prosumer’s energy density, which can be used to detect the underlying leader prosumers in SGs. Second, we formulate the PCG detection (PCG-D) as a multiobjective optimization problem, and present a novel dynamic game model to find the locally Pareto-optimal PCG structure. Third, we propose a partially visible multiagent system (PVMAS), where the viewing angles of both prosumers and PCGs are mutually restricted. The significance of our PVMAS is that it can nicely lead itself to parallelization for PCG-D, due to the fact that the feature updating of each agent is independent of each other. We conduct a series of comprehensive experiments on the simulated SG datasets to validate the performance of PVMAS through comparing it with existing community detection approaches in the literature.\",\n \"Title: Graph K-means based on Leader Identification, Dynamic Game and Opinion Dynamics\\nYear: 2020\\n\\nAbstract: With the explosion of social media networks, many modern applications are concerning about people's connections, which leads to the so-called social computing. An elusive question is to study how opinion communities form and evolve in real-world networks with great individual diversity and complex human connections. In this scenario, the classic K-means technique and its extended versions could no...\",\n \"Title: Link prediction in temporal networks: Integrating survival analysis and game theory.\\nYear: 2019\\n\\nAbstract: Link prediction is an important task in complex network analysis and can be found in many real-world applications such as recommendation systems, information retrieval, and marketing analysis of social networks. This paper focuses on studying the evolution mechanism of real-world temporal networks. Specifically, given a set of temporal links during a fixed time window, how to predict the existence of links at any point in the future. To address this problem, we propose a novel semi-supervised learning framework, which integrates both survival analysis and game theory. First, we carefully define the ϵ-adjacent network sequence, and make use of time stamp on each link to generate the baseline network evolution sequence. Next, to capture the law of network evolution, we employ the Cox Proportional Hazard Model (Cox PHM) to study the relative hazard associated with each temporal link, so as to estimate the coefficients of covariates, which are defined as a set of neighborhood based proximity features. To narrow the area of inquiry, we further propose a game theory based two-way selection mechanism to predict the future network topology. We finally encapsulate these two new technologies in a robust Autonomy-Oriented-Computing (AOC) multi-agent system, and propose a paralleled algorithm to conduct the temporal link prediction task. Extensive experiments were applied to real-world temporal networks to demonstrate both effectiveness and scalability of the proposed approach.\",\n \"Title: Attribute-Based Information Flow Control\\nYear: 2019\\n\\nAbstract: Information flow control (IFC) regulates where information is permitted to travel within information systems. To enforce IFC, access control encryption (ACE) was proposed to support both the no read-up rule and the no write-down rule. There are some problems in existing schemes. First, the communication cost is linear with the number of receivers. Second, senders are not authenticated, namely an unauthorized sender can send a message to a receiver. To reduce communication cost and implement sender authentication, we propose an attribute-based IFC (ABIFC) scheme by introducing attribute-based systems into IFC. Our ABIFC scheme captures the following features: (i) flexible IFC policies are defined over a universal set of descriptive attributes; (ii) both the no read-up rule and the no write-down rule are supported; (iii) the communication cost is linear with the number of required attributes, instead of receivers; (iv) receivers can outsource heavy computation to a server without compromising data confidentiality; (v) authorized senders can control release their attributes when sending messages to receivers. To the best of our knowledge, it is the first IFC scheme where flexible policies are defined over descriptive attributes and outsourced computation is supported.\",\n \"Title: Integrated planning for public health emergencies: A modified model for controlling H1N1 pandemic\\nYear: 2020\\n\\nAbstract: Infectious disease outbreaks have occurred many times in the past decades and are more likely to occur in the future. Recently, Buyuktahtakin et al. (2018) proposed a new epidemics-logistics model to control the 2014 Ebola outbreak in West Africa. Considering that different diseases have dissimilar diffusion dynamics and can cause different public health emergencies, we modify the proposed model by changing capacity constraint, and then apply it to control the 2009 H1N1 outbreak in China. We formulate the problem to be a mixed-integer non-linear programming model (MINLP) and simultaneously determine when to open the new isolated wards and when to close the unused isolated wards. The test results reveal that our model could provide effective suggestions for controlling the H1N1 outbreak, including the appropriate capacity setting and the minimum budget required with different intervention start times.\",\n \"Title: Accurate Human Gesture Sensing With Coarse-Grained RF Signatures.\\nYear: 2019\\n\\nAbstract: RF-based gesture sensing and recognition has increasingly attracted intense academic and industrial interest due to its various device-free applications in daily life, such as elder monitoring, mobile games. State-of-the-art approaches achieved accurate gesture sensing by using fine-grained RF signatures (such as CSI, Doppler effect) while could not achieve the same accuracy with coarse-grained RF signatures such as received signal strength (RSS). This paper presents rRuler, a novel feature extraction method which aims to get fine-grained human gesture features with coarse-grained RSS readings, which means rought ruler could measure fine things. In order to further verify the performance of rRuler, we further propose rRuler-HMM, a hidden Markov model (HMM) based human gesture sensing and prediction algorithm which utilizes the features extracted by rRuler as input. We implemented rRuler and rRuler-HMM using TI Sensortag platforms and off-the-shelf (CTOS) laptops in an indoor environment, extensively performance evaluations show that rRuler and rRuler-HMM stand out for their low cost and high practicability, and the average gesture sensing accuracy of rRuler-HMM can achieve 95.71% in NLoS scenario and 97.14% in LoS scenario, respectively, which is similar to the performance that fine-grained RF signatures based approaches could achieve.\",\n \"Title: Robust Task Grouping with Representative Tasks for Clustered Multi-Task Learning\\nYear: 2019\\n\\nAbstract: Multi-task learning aims to learn multiple tasks jointly by sharing information among related tasks such that the generalization performance over different tasks could be improved. Although multi-task learning has been demonstrated to obtain performance gain in comparison with the single task learning, the main challenge that learning what to share with whom is still not fully resolved. In this paper, we propose a robust clustered multi-task learning approach that clusters tasks into several groups by learning the representative tasks. The main assumption behind our approach is that each task can be represented by a linear combination of some representative tasks that can characterize all tasks. The correlation between tasks can be indicated by the corresponding combination coefficient. By imposing a row-sparse constraint on the correlation matrix, our approach could select the representative tasks and encourage information sharing among the related tasks. In addition, the $l_1,2 $-norm is applied to the representation loss to enhance the robustness of our approach. To solve the resulting bi-convex optimization problem, we design an efficient optimization method based on the alternating direction method of multipliers and accelerated proximal gradient method. Finally, experimental results on synthetic and real-world data sets validate the effectiveness of the proposed approach.\",\n \"Title: Integrating Bi-Dynamic Routing Capsule Network with Label-Constraint for Text classification\\nYear: 2020\\n\\nAbstract: Neural-based text classification methods have attracted increasing attention in recent years. Unlike the standard text classification methods, neural-based text classification methods perform the representation operation and end-to-end learning on the text data. Many useful insights can be derived from neural based text classifiers as demonstrated by an ever-growing body of work focused on text mining. However, in the real-world, text can be both complex and noisy which can pose a problem for effective text classification. An effective way to deal with this issue is to incorporate self-attention and capsule networks into text mining solutions. In this paper, we propose a Bi-dynamic routing Capsule Network with Label-constraint (BCNL) model for text classification, which moves beyond the limitations of previous methods by automatically learning the task-relevant and label-relevant words of text. Specifically, we use a Bi-LSTM and self-attention with position encoder network to learn text embeddings. Meanwhile, we propose a bi-dynamic routing capsule network with label-constraint to adjust the category distribute of text capsules. Through extensive experiments on four datasets, we observe that our method outperforms state-of-the-art baseline methods.\",\n \"Title: Random compact Gaussian kernel: Application to ELM classification and regression\\nYear: 2021\\n\\nAbstract: Extreme learning machine (ELM) kernels using random feature mapping have recently gained a lot of popularity because they require low human supervision. However, the superiority in the mapping mechanism of ELM kernels is accompanied by a higher computation cost, rendering the kernel learning algorithms hard to tackle large scale learning tasks. On the other hand, the implicit mapping used in the conventional Gaussian kernel is computationally cheaper than the explicit computation of the ELM kernel, but requires trivial human intervention. This paper proposes to merge both properties by defining a new kernel, the random compact Gaussian (RCG) kernel. The random feature mapping property enables RCG kernel to save parameters selection time, while the implicitly mapping property enables RCG kernel to save kernel calculation time. The proposed kernel works by scaling one kernel parameter in the conventional Gaussian kernel to multiple kernel parameters, and generating all parameters randomly based on a continuous probability distribution. We prove that the RCG kernel is a Mercer kernel. The kernel is calculated implicitly before seeing the training samples and used to train ELMs. The experiments on 25 binary classification and regression benchmark problems show that the RCG kernel typically outperforms other competitive kernels. Compared to ELM kernel, RCG kernel not only achieves the better generalization performance on most datasets, but also needs much less kernel calculation cost. In addition, the sensitivity analysis of the kernel parameters of k-fold cross validation is conducted and the results show that the RCG kernel is robust and stable for repeated trials.\",\n \"Title: A Comparative Study on Schema-Guided Dialogue State Tracking\\nYear: 2021\\n\\nAbstract: Frame-based state representation is widely used in modern task-oriented dialog systems to model user intentions and slot values. However, a fixed design of domain ontology makes it difficult to extend to new services and APIs. Recent work proposed to use natural language descriptions to define the domain ontology instead of tag names for each intent or slot, thus offering a dynamic set of schema. In this paper, we conduct in-depth comparative studies to understand the use of natural language description for schema in dialog state tracking. Our discussion mainly covers three aspects: encoder architectures, impact of supplementary training, and effective schema description styles. We introduce a set of newly designed bench-marking descriptions and reveal the model robustness on both homogeneous and heterogeneous description styles in training and evaluation.\",\n \"Title: Complementary Fourier Single-Pixel Imaging\\nYear: 2021\\n\\nAbstract: Single-pixel imaging, with the advantages of a wide spectrum, beyond-visual-field imaging, and robustness to light scattering, has attracted increasing attention in recent years. Fourier single-pixel imaging (FSI) can reconstruct sharp images under sub-Nyquist sampling. However, the conventional FSI has difficulty balancing imaging quality and efficiency. To overcome this issue, we proposed a novel approach called complementary Fourier single-pixel imaging (CFSI) to reduce the number of measurements while retaining its robustness. The complementary nature of Fourier patterns based on a four-step phase-shift algorithm is combined with the complementary nature of a digital micromirror device. CFSI only requires two phase-shifted patterns to obtain one Fourier spectral value. Four light intensity values are obtained by loading the two patterns, and the spectral value is calculated through differential measurement, which has good robustness to noise. The proposed method is verified by simulations and experiments compared with FSI based on two-, three-, and four-step phase shift algorithms. CFSI performed better than the other methods under the condition that the best imaging quality of CFSI is not reached. The reported technique provides an alternative approach to realize real-time and high-quality imaging.\",\n \"Title: An Efficient Alpha Seeding Method For Optimized Extreme Learning Machine-Based Feature Selection Algorithm\\nYear: 2021\\n\\nAbstract: Embedded feature selection algorithms, such as support vector machine based recursive feature elimination (SVM-RFE), have proven to be effective for many real applications. However, due to the model selection problem, SVM-RFE naturally suffers from a heavy computational burden as well as high computational complexity. To solve these issues, this paper proposes using an optimized extreme learning machine (OELM) model instead of SVM. This model, referred to as OELM-RFE provides an efficient active set solver for training the OELM algorithm. We also present an effective alpha seeding algorithm to efficiently solve successive quadratic programming (QP) problems inherent in OELM. One of the salient characteristics of OELM-RFE is that it has only one tuning parameter: the penalty constant C. Experimental results from work on benchmark datasets show that OELM-RFE tends to have higher prediction accuracy than SVM-RFE, and requires fewer model selection efforts. In addition, the alpha seeding method works better on more datasets.\",\n \"Title: Group-Oriented Task Allocation for Crowdsourcing in Social Networks\\nYear: 2021\\n\\nAbstract: Previous crowdsourcing studies often adopted the individual-oriented approach that outsources a task to an individual worker or team formation-based approach that outsources a task to an artificially formed team of workers. Nowadays, workers are often naturally organized into groups through social networks. To address such common issue of grouped workers in real crowdsourcing systems, this article...\",\n \"Title: FaceInpainter: High Fidelity Face Adaptation to Heterogeneous Domains\\nYear: 2021\\n\\nAbstract: In this work, we propose a novel two-stage framework named Facelnpainter to implement controllable Identity-Guided Face Inpainting (IGFI) under heterogeneous domains. Concretely, by explicitly disentangling foreground and background of the target face, the first stage focuses on adaptive face fitting to the fixed background via a Styled Face Inpainting Network (SFI-Net), with 3D priors and texture code of the target, as well as identity factor of the source face. It is challenging to deal with the inconsistency between the new identity of the source and the original background of the target, concerning the face shape and appearance on the fused boundary. The second stage consists of a Joint Refinement Network (JR-Net) to refine the swapped face. It leverages AdaIN considering identity and multi-scale texture codes, for feature transformation of the decoded face from SFI-Net with facial occlusions. We adopt the contextual loss to implicitly preserve the attributes, encouraging face deformation and fewer texture distortions. Experimental results demonstrate that our approach handles high-quality identity adaptation to heterogeneous domains, exhibiting the competitive performance compared with state-of-the-art methods concerning both attribute and identity fidelity.\",\n \"Title: A hierarchical knowledge guided backtracking search algorithm with self-learning strategy\\nYear: 2021\\n\\nAbstract: To improve the performance of the backtracking search optimization algorithm (BSA), a multi-population cooperative evolution strategy guided BSA with hierarchical knowledge (HKBSA) is proposed in this paper. According to the domain knowledge of the candidates in objective space, the population is divided into the dominant population, the ordinary population and the inferior population. The information between the sub-populations has interacted with the evolution processes of the three sub-populations. The individuals in the dominant population are maintained as the optimal solutions and are utilized to guide the evolution of the other two sub-populations. A multi-strategy mutation mechanism is applied to solve non-separable problems. The distribution vector of inferior individuals is constructed by sampling, and a mechanism of the individual generation with feedback is proposed by combining self-learning strategy and elite learning strategy. The convergence of HKBSA is analyzed with the Markov model. Compared with the state-of-the-art BSA variants, HKBSA outperforms other algorithms in terms of the speed of convergence, solution accuracy and stability.\",\n \"Title: Reinforcing Pretrained Models for Generating Attractive Text Advertisements\\nYear: 2021\\n\\nAbstract: ABSTRACTWe study how pretrained language models can be enhanced by using deep reinforcement learning to generate attractive text advertisements that reach the high quality standard of real-world advertiser mediums. To improve ad attractiveness without hampering user experience, we propose a model-based reinforcement learning framework for text ad generation, which constructs a model for the environment dynamics and avoids large sample complexity. Based on the framework, we develop Masked-Sequence Policy Gradient, a reinforcement learning algorithm that integrates efficiently with pretrained models and explores the action space effectively. Our method has been deployed to production in Microsoft Bing. Automatic offline experiments, human evaluation, and online experiments demonstrate the superior performance of our method.\",\n \"Title: Energy-Aware Offloading Based On Priority In Mobile Cloud Computing\\nYear: 2021\\n\\nAbstract: Smartphones and portable devices have been widely used in our daily life. However, these portable devices cannot be used in a lot of environments due to limitations in battery capacity and computing speed. Therefore, how to save energy consumption and improve processing ability has become a hot issue. The previous works were mainly to transfer some local tasks to remote compute nodes by intelligently selecting the execution route of tasks based on Directed Acyclic Graph (DAG). Different from previous work, this work mainly attempts to consider task urgency and system load based on Dynamic Voltage and Frequency Scaling (DVFS) model. Firstly, tasks are given different priorities according to the energy consumption of different locations and task urgency. Secondly, a reference working state is given according to system load to ensure that each task (regardless of job deadline) is completed with minimum energy consumption under the average system load. Finally, a heuristic algorithm is used to schedule tasks on mobile devices and remote cloud resources based on the priority of tasks and reference status of resources. Our method is called AODVFS-Adaptive Offloading based on DVFS model. Comparison with other methods shows that our method saves average energy consumption and increases the number of completed tasks.\",\n \"Title: Can AoI and Delay be Minimized Simultaneously with Short-Packet Transmission?\\nYear: 2021\\n\\nAbstract: Age of information (AoI) and delay are both critical performance metrics to enable emerging applications such as remote estimation and real-time control in factory automation. In particular, it remains an open question whether AoI and delay can be minimized simultaneously in short-packet transmission for factory automation etc., as they are affected by block length and update rate in a complex man...\",\n \"Title: Global Relation-Aware Attention Network for Image-Text Retrieval\\nYear: 2021\\n\\nAbstract: ABSTRACTThe cross-modal image-text retrieval has attracted extensive attention in recent years, which contributes to the development of search engine. Fine-grained features and cross-attention have been widely used in past researches to reach the goal of cross-modal image-text matching. Although cross-related methods have achieved remarkable results, the features must be encoded again in evaluation phase due to the interaction of the two modalities, which is unsuitable for actual scenarios of search engine development. In addition, the aggregated feature does not contain sufficient semantics since it is merely obtained by simple mean pooling. Furthermore, connecting weights of self-attention blocks are target position invariant, which lacks the expected adaptability. To tackle these limitations, in this paper, we propose a novel Global Relation-aware Attention Network (GRAN) for image-text retrieval by designing Global Attention Module (GAM) and Relation-aware Attention Module (RAM) which play an important role in modeling the global feature and the relationships of local fragments. Firstly, we propose Global Attention Module (GAM) followed the fine-grained features to obtain meaningful global feature. Secondly, we use several stacked transformer encoders to further encode features separately. Finally, we propose Relation-aware Attention Module (RAM) to generate a vector which represents the relation information to infer the attention intensity of pairwise fragments. The local features, the global feature, and their relations are considered jointly to conduct an efficient image-text retrieval. Extensive experiments are conducted on the benchmark datasets of Flickr30K and MSCOCO, demonstrating the superiority of our method. On the Flickr30K, compared to the state-of-the-art method TERAN, we improve [email protected](K=1) metric by 5.8% and 4.0 on the image and text retrieval tasks, respectively.\",\n \"Title: Measuring the Network Vulnerability Based on Markov Criticality\\nYear: 2022\\n\\nAbstract: AbstractVulnerability assessment—a critical issue for networks—attempts to foresee unexpected destructive events or hostile attacks in the whole system. In this article, we consider a new Markov global connectivity metric—Kemeny constant, and take its derivative called Markov criticality to identify critical links. Markov criticality allows us to find links that are most influential on the derivative of Kemeny constant. Thus, we can utilize it to identity a critical link (i, j) from node i to node j, such that removing it leads to a minimization of networks’ global connectivity, i.e., the Kemeny constant. Furthermore, we also define a novel vulnerability index to measure the average speed by which we can disconnect a specified ratio of links with network decomposition. Our method is of high efficiency, which can be easily employed to calculate the Markov criticality in real-life networks. Comprehensive experiments on several synthetic and real-life networks have demonstrated our method’s better performance by comparing it with state-of-the-art baseline approaches.\",\n \"Title: A hierarchical guidance strategy assisted fruit fly optimization algorithm with cooperative learning mechanism\\nYear: 2021\\n\\nAbstract: •A novel cooperative learning fruit fly optimization algorithm (HGCLFOA) is proposed.•The hierarchical guidance strategy for local search is implemented in the olfactory search.•The inferior solution repairing (ISR) strategy is employed to modify the search direction.•The hybrid GEDA with previous information is applied to guide the evolution in vision search.•A probability selection strategy based on feedback of objective space is introduced.\",\n \"Title: A two-stage evolutionary strategy based MOEA/D to multi-objective problems\\nYear: 2021\\n\\nAbstract: •A two-stage evolution strategy is proposed for solving multi-objective problem.•The convergence and diversity should be balanced in multi-objective optimization.•A local searching method can improve the diversity of solutions in the space.•Experimental results have been presented by using statistical method.\",\n \"Title: Learning to traverse over graphs with a Monte Carlo tree search-based self-play framework\\nYear: 2021\\n\\nAbstract: The combinatorial optimization (CO) problems on the graph are the core and classic problems in artificial intelligence (AI) and operations research (OR). For example, the Vehicle Routing Problem (VRP) and Traveling Salesman Problem (TSP) are fascinating NP-hard problems and have important significance for the existing transportation system. Traditional methods such as heuristics methods, exact algorithms, and solution solvers can already find approximate solutions on small-scale graphs. However, they are helpless for large-scale graphs and other problems with similar structures. Moreover, traditional methods often require artificially designed heuristic functions to aid decision-making. In recent years, more and more work has focused on applying deep learning and reinforcement learning (RL) to learn heuristics, which allows us to learn the internal structure of the graph end-to-end and find the optimal path under the guidance of heuristic rules. However, most of these still need manual assistance, and the RL method used has the problems of low sampling efficiency and small searchable space. This paper proposes a novel framework (called OmegaZero) based on Alphago Zero, which does not prescribe expert experience or label data but is trained through self-play. We divide the learning into two stages: in the first stage, we employ graph attention network (GAT) and GRU to learn node representations and memory history trajectories. In the second stage, we employ Monte Carlo tree search (MCTS) and deep RL to search for the solution space and train the model.\",\n \"Title: A Comparison Of The Different Stages Of Dust Events Over Beijing In March 2021: The Effects Of The Vertical Structure On Near-Surface Particle Concentration\\nYear: 2021\\n\\nAbstract: Mineral dust is of great importance to climate change, air quality, and human health. In this study, multisource data, including the reanalysis data and remote sensing data, were used to compare the three dust events that occurred in the March of 2021 over Beijing and reveal the effects of atmospheric vertical structure on near-surface dust concentration. The combined effect of the Mongolian cyclone and a wide persistent cold-front induced two events (E1: from March 15 to 16 and E3: from March 28 to 29). E1 was more intense, more extensive, and longer-lasting than E3 due to the combination of the stronger Mongolian cyclone, slower high/cold surface pressure, and the low-level jet. However, under the appropriate configurations of temperature and pressure fields between high and low altitudes, weak updrafts were still induced and could elevate dust up to 850 hPa, as occurred during E2 on March 22 and 23. The dust emission was inferior to E1 and E3, which contributes to the low dust concentration near the surface in E2. On the other hand, the downdraft strength directly affected both the vertical distribution of dust and the concentration of surface particles. There was a strong temporal consistency between the occurrence of the downdraft and the dust touchdown. In E1, the continuous strong downdraft caused the maximum dust concentration to be above 4000 mu g/m(3) at around 200 m. In contrast, the maximum height of the dust mass concentration in E3 occurred at about 800 m due to the transient downdraft, which weakened its effect on surface visibility. Besides, the weak vertical motion in E2 caused most of the dust to become suspended in the air. Overall, the large dust emission resulted from active updrafts in the source region, and the lengthy strong downdrafts led to the ultrahigh particle concentration near the surface.\",\n \"Title: A multi-task learning approach for improving travel recommendation with keywords generation\\nYear: 2021\\n\\nAbstract: Travel recommendation is very critical to helping users quickly find products or services that they are interested in. The key to travel recommender systems is learning user shopping intentions, which are expressed through various supervision signals, such as the clicked products and their titles. Existing travel recommendation methods commonly infer user intentions from click behaviors on travel products. However, remarkable keywords in the product title, such as departure, destination, travel time, hotel, and transportation are paid less attention. To this end, we hypothesize that modeling click sequences and product keywords in title jointly would result in a more holistic representation of a product and towards more accurate recommendations. Thus, we propose a TRKG (short for Travel Recommendation with Keywords Generation) model, which fulfills the travel recommendation and keywords generation tasks simultaneously. To generate explainable outputs, unlike most previous approaches that regard the product title as a hidden feature vector, TRKG regards keywords in the product title as an additional supervision signal. Meanwhile, TRKG integrates the long-term and short-term user preferences in the travel recommendation component and the keywords generation component. To evaluate the proposed model, we constructed datasets from a large tourism e-commerce website in China. Extensive experiments demonstrate that the proposed method yields significant improvements over state-of-the-art methods.\",\n \"Title: Information Age-Delay Correlation and Optimization With Finite Block Length\\nYear: 2021\\n\\nAbstract: Both information age and delay are critical performance metrics of emerging time-sensitive applications. However, their inherent correlation in the finite block length (FBL) regime has remained uninvestigated as it is affected by block length and update rate in a complex manner. In this paper, closed-form expressions for average age of information (AoI), peak AoI (PAoI) and delay are derived for a...\",\n \"Title: Toward a Green Secure Relay System for Mission-Critical IoT: Hybrid Duplex Relay Selection and Resource Allocation in the Finite Block Length Regime\\nYear: 2021\\n\\nAbstract: In this paper, we investigate the physical layer security of a relay system toward secure and green mission-critical Internet of Things, in the finite block length regime, which was hardly considered in the literature. Closed-form expressions for statistical secrecy throughput of full-duplex (FD) and half-duplex (HD) relay systems are respectively derived and verified by simulation results. To max...\",\n \"Title: Candidate Box Fusion Based Approach To Adjust Position Of The Candidate Box For Object Detection\\nYear: 2021\\n\\nAbstract: The method of object detection has been applied to all aspects in our lives. Although object detection methods based on deep learning have been widely used in various fields, there are still some overlooked problems in the candidate box selection stage. The detection results of traditional candidate box selection methods can only select a relatively optimal maximum candidate box. If the maximum candidate box is still not accurate enough, this type of methods will not be able to do adjust it. To solve this problem, an object detection method based on the multiple candidate box fusion is proposed. The method can not only retain the maximum candidate box and delete the non-maximum candidate box, but also adjust the position of the maximum candidate box again. Thereby a more accurate maximum candidate box can be obtained. In order to verify the generalization ability of the method, the candidate box fusion method is combined with the two object detection frameworks: faster R-CNN model and YOLOv3 model. The results of these experiments prove that the proposed method can achieve higher detection accuracy and complete the object detection task more effectively.\",\n \"Title: Single Haze Image Restoration Under Non-Uniform Dense Scattering Media\\nYear: 2021\\n\\nAbstract: We demonstrate an efficient restoration method based on sky segmentation and block optimized transmission (SSBOT) to restore the visibility of single haze image under non-uniform scattering medium. The SSBOT method uses sky region segmentation to determine the haze-opaque region, and estimates the global atmospheric light using quadtree-decomposition. The proposed SSBOT estimates the transmission in every sub-images of the degraded image with dark channel method with coefficient correction. The experimental results show that SSBOT can efficiently restore images degraded by non-uniform dense scattering medium. The proposed method can be used for various security and surveillance applications.\",\n \"Title: The Eulerian Distribution on the Involutions of the Hyperoctahedral Group is Indeed gamma-Positive\\nYear: 2021\\n\\nAbstract: Let I-n(B) denote the set of the involutions of the hyperoctahedral group B-n, and let des(B)(pi) denote the number of descents of the permutation pi is an element of B-n. We settle a problem of Moustakas which states that I-n(B)(t) := Sigma(pi is an element of InB) t(desB(pi)) is gamma-positive for n >= 1.\",\n \"Title: Improving Subgraph Recognition with Variational Graph Information Bottleneck\\nYear: 2022\\n\\nAbstract: Subgraph recognition aims at discovering a compressed substructure of a graph that is most informative to the graph property. It can be formulated by optimizing Graph Information Bottleneck (GIB) with a mutual information estimator. However, GIB suffers from training instability and degenerated results due to its intrinsic optimization process. To tackle these issues, we reformulate the subgraph recognition problem into two steps: graph perturbation and subgraph selection, leading to a novel Variational Graph Information Bottleneck (VGIB) framework. VGIB first employs the noise injection to modulate the information flow from the input graph to the perturbed graph. Then, the perturbed graph is encouraged to be informative to the graph property. VGIB further obtains the desired subgraph by filtering out the noise in the perturbed graph. With the customized noise prior for each input, the VGIB objective is endowed with a tractable variational upper bound, leading to a superior empirical performance as well as theoretical properties. Extensive experiments on graph interpretation, explainability of Graph Neural Networks, and graph classification show that VGIB finds better subgraphs than existing methods 1 1 Code is avaliable on https://github.com/Samyu0304/VGIB.\",\n \"Title: Random radial basis function kernel-based support vector machine\\nYear: 2021\\n\\nAbstract: The main computational cost of building a support vector machine (SVM) training model lies in tuning the hyperparameters, including the kernel parameters and penalty constant C. This paper introduces a new kernel, the random radial basis function (RRBF) kernel, which all kernel parameters can be assigned to randomly. The key idea of the RRBF is to extend a one-dimensional parameter to d-dimensional parameters by simple modification of the traditional RBF kernel. We prove the universal approximation capability of a SVM with the RRBF kernel, proposing a simple model selection algorithm. The experiments on benchmark datasets show that SVM with the RRBF kernel clearly outperforms the traditional RBF kernel and other popular kernels, and the results are quite insensitive to C. Compared with other kernels, a SVM with the RRBF kernel is not sensitive to the penalty constant C, and it can be tuned in a wide range. Additionally, this property also enables a SVM to find the hyperparameters quickly. A simulation of parameter sensitivity study also show that a SVM is more stable w.r.t. to the different choices of kernel parameters.\",\n \"Title: Multi-view Graph Attention Network for Travel Recommendation\\nYear: 2022\\n\\nAbstract: As an e-commerce feature, the recommender system can enhance the consumer shopping experience and create huge benefits for businesses. The e-tourism has become one of the largest service industries with the application and popularity of recommender systems. Many studies have confirmed that the travel product recommendation is widely different from traditional recommendations. Due to the financial and time costs, travel products are usually browsed and purchased relatively infrequently compared with other traditional products (e.g., books, movies and food). In addition, choosing the appropriate travel product will be influenced by many factors, such as departure, destination and price. To tackle this challenging problem, we propose a MV-GAN (short for Multi-View Graph Attention Network for travel recommendation) model. It enriches user and product semantics through both metapath-guided neighbors aggregation and multi-view fusion in heterogeneous travel product recommendation graph. In particular, we design node-level and path-level attention networks for learning user and product representations from every single view. To collaboratively integrate multiple types of relationships in different views, a view-level attention mechanism is proposed to aggregate the node representations and obtain global user and product representations. We evaluate the proposed method on a public dataset and a dataset constructed from a large tourism e-commerce website in China. Extensive experiments not only validate the effectiveness of MV-GAN, but also show its potentially good interpretability.\",\n \"Title: Engaging voluntary contributions in online review platforms: The effects of a hierarchical badges system\\nYear: 2022\\n\\nAbstract: Employing a hierarchical badges system to sustain users' participation and content contribution has been a common practice in online review platforms. However, it is still unclear whether this incentive mechanism is effective in inducing users to provide more substantial and useful reviews. In this paper, we draw on the goalsetting theory and status hierarchies theory to study how the hierarchical badges system impacts users' behavior such as their effort levels and expressed opinions, using the data of 234,220 reviews from a large online review platform. Our results show that users significantly increase their efforts as their status are promoted, manifesting as review frequency, length, and readability. In terms of expressed opinions, users with higher badges are less likely to post extremely negative ratings and use negative emotional words. Our results have important implications for business modes that rely on user-generated content, such as online review platforms, social question and answer sites, social media, and crowdsourcing systems.\",\n \"Title: Pilot Overhead vs. Pilot Power: Short Packet Structure Optimization for URLLC over Continuous Fading\\nYear: 2021\\n\\nAbstract: In this paper, we investigate how to strike the balance between pilot overhead and pilot power for short packet ultra reliable and low latency communication (URLLC) systems under continuous fading, as the previous work has not considered the joint impact of pilot overhead and pilot power on the system throughput and has assumed block fading only. It is revealed that, at low to moderate signal to noise ratio (SNR) or in continuous fading, pilot power boosting can reduce pilot overhead and improve throughput significantly, and that at high SNR, pilot overhead plays a dominant role in enhancing the throughput while maintaining a low peak to average power ratio. For throughput formulation, a closed-form expression for the asymptotic block error probability under continuous fading is derived with respect to pilot power, pilot length and block length. And for throughput maximization, the near-optimal pilot power and the near-optimal block length are given in closed form by solving complicated transcendental equations. Based on the analysis, a low-complexity joint pilot power, pilot length and block length optimization (JPLLO) algorithm is proposed, which achieves a near-optimal performance and a dramatic complexity reduction compared to exhaustive search, converging within only 1-3 iterations. The JPLLO algorithm also significantly outperforms the previous joint optimization algorithm under continuous fading.\",\n \"Title: A Novel Sentiment Polarity Detection Framework for Chinese\\nYear: 2022\\n\\nAbstract: Nowadays, mining opinions or sentiment from online user-generated text has become a research hot spot. Although a large amount of lexicon-based Chinese polarity detection works have been done, the existing methods have one common flaw: that even the same word can have opposite polarities among different seed lexicons. This is known as polarity fuzziness. To enhance the performance of Chinese senti...\",\n \"Title: DAN-SuperPoint: Self-Supervised Feature Point Detection Algorithm with Dual Attention Network\\nYear: 2022\\n\\nAbstract: In view of the poor performance of traditional feature point detection methods in low-texture situations, we design a new self-supervised feature extraction network that can be applied to the visual odometer (VO) front-end feature extraction module based on the deep learning method. First, the network uses the feature pyramid structure to perform multi-scale feature fusion to obtain a feature map containing multi-scale information. Then, the feature map is passed through the position attention module and the channel attention module to obtain the feature dependency relationship of the spatial dimension and the channel dimension, respectively, and the weighted spatial feature map and the channel feature map are added element by element to enhance the feature representation. Finally, the weighted feature maps are trained for detectors and descriptors respectively. In addition, in order to improve the prediction accuracy of feature point locations and speed up the network convergence, we add a confidence loss term and a tolerance loss term to the loss functions of the detector and descriptor, respectively. The experiments show that our network achieves satisfactory performance under the Hpatches dataset and KITTI dataset, indicating the reliability of the network.\",\n \"Title: An intermediary utility-based service search and structure organization approach in service-oriented MAS\\nYear: 2022\\n\\nAbstract: In distributed service-oriented multiagent systems (SOMASs), agents cooperate autonomously to accomplish service discovery. In a SOMAS application, each agent plays two roles: (i) satisfying the service demands as a service provider, and (ii) forwarding the received queries as an intermediary agent. This study defines that the intermediary utility of an agent is positively related to the success rate of the service discovery in which it participates. If the intermediary utility of an agent is high (e.g., when the neighbors of the agent have abundant service resources), then the agent can better guide service searches. Based on this characteristic, this study proposes an intermediary utilities-based service search and structure organization approach to improve the efficiency of service discovery. In this approach, a service query has a higher probability of being forwarded to agents with higher intermediary utilities, which accelerates the service search. Additionally, the connectivity of the system structure can be preserved by the proposed approach, which ensures that the services can be accessed globally in the system. Extensive experiments demonstrate that the proposed approach can achieve better performance regarding the success rate of service discovery and average forwarding time of service queries comparing to previous approaches.\",\n \"Title: Method to enhance deep learning fault diagnosis by generating adversarial samples\\nYear: 2022\\n\\nAbstract: Modern industrial fields utilize complex mechanical equipment and machinery, which are closely linked, and equipment faults are difficult to express. Therefore, fault diagnosis is important to ensure the safety of complex mechanical equipment in modern industries. Deep learning has achieved excellent results with recent fault diagnosis methods. At present, three common deep learning models (MLP, CNN, and RNN models) can achieve diagnosis rates close to 100% with original fault diagnosis data and a signal-to-noise ratio above 10 dB. However, we found that the diagnostic rate of these three models was completely incorrect when an adversarial sample with a signal-to-noise ratio noise greater than 10 dB was added to the original sample. We propose a GAN-based adversarial signal generative adversarial network (AdvSGAN) in this paper. We conduct experiments on the CWRU dataset and conclude that we can easily obtain adversarial noise and generate training samples through AdvSGAN. With the addition of adversarial data training, the diagnostic rate of the model on these adversarial samples increased from less than 5% to 98.69%, 97.38% and 96.94%. Hence, this method increases the reliability of our deep learning model. (C) 2021 Elsevier B.V. All rights reserved.\",\n \"Title: MVE-FLK: A multi-task legal judgment prediction via multi-view encoder fusing legal keywords\\nYear: 2022\\n\\nAbstract: Legal Judgment Prediction (LJP) aims to predict the judgment result based on the fact description of a criminal case, and is gradually becoming a hot research topic in the legal realm. Generally, a classic LJP contains three subtasks, i.e., applicable law article prediction, charge prediction, and term of penalty prediction. In real-world scenarios, both charge prediction and applicable law article prediction are actually the tasks of multi-class classification with multi-label learning. However, most existing studies model them as the problems of multi-class classification with single-label learning. Besides, they only consider the context of the fact description, and ignore the exploitation of effective keywords that are widely existed in abundant law articles. To fill the above gaps, we propose a novel multi-task legal judgment prediction framework via multi-view encoder fusing legal keywords, named MVE-FLK, to jointly model multiple subtasks in LJP. Specifically, the multi-view encoder is the core module of MVE-FLK, in this module, we devise a word and sentence encoder (WSE) with an attention mechanism to fuse legal keywords. And then, we develop a multi-view attention network to combine WSE with classic Transformer and DAN (Deep Averaging Network) for encoding the case from multiple views. After that, we propose a multi-task prediction module by developing a novel keywords fusing approach to enhance the performance of multi-task prediction. In addition, we devise a unique prediction principle for each subtask at a fine-grained level, which effectively improves the performance of subtasks. The experimental results on two real-life legal datasets show that our model yields significant prediction performance advantages over six competitive methods.\",\n \"Title: Average AoI Minimization in UAV-Assisted Data Collection With RF Wireless Power Transfer: A Deep Reinforcement Learning Scheme\\nYear: 2022\\n\\nAbstract: This article studies the unmanned aerial vehicle (UAV)-assisted wireless powered network, where a UAV is dispatched to wirelessly charge multiple ground nodes (GNs) by using radio frequency (RF) energy transfer and then the GNs use their harvested energy to upload the sensed information to the UAV. At each moment, the UAV is scheduled to charge the GNs or only one GN is scheduled to upload its dat...\",\n \"Title: Integer particle swarm optimization based task scheduling for device-edge- cloud cooperative computing to improve SLA satisfaction\\nYear: 2022\\n\\nAbstract: Task scheduling helps to improve the resource efficiency and the user satisfaction for Device-Edge-Cloud Cooperative Computing (DE3C), by properly mapping requested tasks to hybrid device-edge-cloud resources. In this paper, we focused on the task scheduling problem for optimizing the Service-Level Agreement (SLA) satisfaction and the resource efficiency in DE3C environments. Existing works only focused on one or two of three sub-problems (offloading decision, task assignment and task ordering), leading to a sub-optimal solution. To address this issue, we first formulated the problem as a binary nonlinear programming, and proposed an integer particle swarm optimization method (IPSO) to solve the problem in a reasonable time. With integer coding of task assignment to computing cores, our proposed method exploited IPSO to jointly solve the problems of offloading decision and task assignment, and integrated earliest deadline first scheme into the IPSO to solve the task ordering problem for each core. Extensive experimental results showed that our method achieved upto 953% and 964% better performance than that of several classical and state-of-the-art task scheduling methods in SLA satisfaction and resource efficiency, respectively.\",\n \"Title: A multipopulation cooperative coevolutionary whale optimization algorithm with a two-stage orthogonal learning mechanism\\nYear: 2022\\n\\nAbstract: This paper designed a multipopulation cooperative coevolutionary framework with a two-stage orthogonal learning (OL) mechanism for the whale optimization algorithm (MCCWOA) to improve the performance of the whale optimization algorithm (WOA). In the framework, a prediction model of the neighborhood structure is established by discovering the guidance information of the following iteration process in the objective space at the first-stage OL. In the second-stage OL, an auxiliary vector pool with various features in the decision space is introduced to guide the candidates falling in the stagnant status to conduct more valuable exploration. According to the domain knowledge of the candidates, the population is divided into the elite population, the intermediate population, and the inferior population. The information of the subpopulations has interacted with the corresponding historical populations in the evolution processes to enhance the ability of cooperative coevolution among individuals. A standard set of comprehensive benchmark cases and three engineering cases are utilized to verify the advantages of the proposed algorithm. The results of the statistical analysis, diversity analysis, and convergence analysis testified that the MCCWOA outperforms the 15 state-of-the-art algorithms regarding efficiency and significance.\",\n \"Title: A Novel Transformer-Based Attention Network for Image Dehazing\\nYear: 2022\\n\\nAbstract: Image dehazing is challenging due to the problem of ill-posed parameter estimation. Numerous prior-based and learning-based methods have achieved great success. However, most learning-based methods use the changes and connections between scale and depth in convolutional neural networks for feature extraction. Although the performance is greatly improved compared with the prior-based methods, the performance in extracting detailed information is inferior. In this paper, we proposed an image dehazing model built with a convolutional neural network and Transformer, called Transformer for image dehazing (TID). First, we propose a Transformer-based channel attention module (TCAM), using a spatial attention module as its supplement. These two modules form an attention module that enhances channel and spatial features. Second, we use a multiscale parallel residual network as the backbone, which can extract feature information of different scales to achieve feature fusion. We experimented on the RESIDE dataset, and then conducted extensive comparisons and ablation studies with state-of-the-art methods. Experimental results show that our proposed method effectively improves the quality of the restored image, and it is also better than the existing attention modules in performance.\",\n \"Title: Partial NIR-VIS Heterogeneous Face Recognition With Automatic Saliency Search\\nYear: 2021\\n\\nAbstract: Near-infrared-visual (NIR-VIS) heterogeneous face recognition (HFR) aims to match NIR face images with the corresponding VIS ones. It is a challenging task due to the sensing gaps among different modalities. Occlusions in the input face images make the task extremely complex. To tackle these problems, we present a Saliency Search Network (SSN) to extract domain-invariant identity features. We propose to automatically search the efficient parts of face images in a modality-aware manner, and remove redundant information. Moreover, the searching process is guided by an information bottleneck network, which mitigates the overfitting problems caused by small datasets. Extensive experiments on both complete and partial NIR-VIS HFR on multiple datasets demonstrate the effectiveness and robustness of the proposed method to modality discrepancy and occlusions.\",\n \"Title: A constrained multi-objective optimization algorithm with two cooperative populations\\nYear: 2022\\n\\nAbstract: Constrained multi-objective problems (CMOPs) require balancing convergence, diversity, and feasibility of solutions. Unfortunately, the existing constrained multi-objective optimization algorithms (CMOEAs) exhibit poor performance when solving the CMOPs with complex feasible regions. To solve this shortcoming, this work proposes an improved algorithm named the CMOEA-TCP, which maintains two populations cooperating to push the solutions to approximate the constrained Pareto front. Specifically, one population is obtained by the Pareto-based method and aims to strengthen the algorithm’s convergence ability. Meanwhile, another population is maintained by decomposition-based method and devoted to improving its diversity. The two populations work cooperatively during the entire evolution process with the constraint-handling technique. The performance of the CMOEA- TCP is verified on three benchmark suites with 34 problems. The experimental results demonstrate that the CMOEA-TCP can achieve performance comparable to or better than the other six state-of-the-art CMOEAs on the majority of considered problems.\",\n \"Title: Retina-like Computational Ghost Imaging for an Axially Moving Target\\nYear: 2022\\n\\nAbstract: Unlike traditional optical imaging schemes, computational ghost imaging (CGI) provides a way to reconstruct images with the spatial distribution information of illumination patterns and the light intensity collected by a single-pixel detector or bucket detector. Compared with stationary scenes, the relative motion between the target and the imaging system in a dynamic scene causes the degradation of reconstructed images. Therefore, we propose a time-variant retina-like computational ghost imaging method for axially moving targets. The illuminated patterns are specially designed with retina-like structures, and the radii of foveal region can be modified according to the axial movement of target. By using the time-variant retina-like patterns and compressive sensing algorithms, high-quality imaging results are obtained. Experimental verification has shown its effectiveness in improving the reconstruction quality of axially moving targets. The proposed method retains the inherent merits of CGI and provides a useful reference for high-quality GI reconstruction of a moving target.\",\n \"Title: Android Unit Test Case Generation Based on the Strategy of Multi-Dimensional Coverage\\nYear: 2021\\n\\nAbstract: Although Android apps are becoming more and more popular, apps tend to contain defects which can ultimately manifest as failures (or crashes) to end-users. In order to improve software quality, people have proposed different automated tools for testing Android applications. The automatic generation of test cases is an important means to improve the efficiency of software testing, but many tools re...\",\n \"Title: Average Descent Rate Singular Value Decomposition and Two-Dimensional Residual Neural Network for Fault Diagnosis of Rotating Machinery\\nYear: 2022\\n\\nAbstract: Fault diagnosis of rotating machinery is difficult under the strong noisy environment. Although singular value decomposition (SVD) can remove noise from vibration signals, the singular value threshold is normally determined by expert experience. To solve this problem, a fault diagnosis method based on average descent rate (ADR)-SVD and two-dimensional residual neural network (Resnet) is proposed. First, ADR-SVD uses the ADR index to construct the singular value descent rate difference spectrum and uses the maximum value of the spectrum as the singular value threshold. The noise reduction process of ADR-SVD requires little expert experience. Then, in order to adaptively identify the fault features of the signals, we introduce Gramian angular difference field (GADF), which can transform the one-dimensional signals into two-dimensional images and preserve the temporal correlation of the one-dimensional signals. Finally, we construct a two-dimensional Resnet to learn image features and identify fault types. The proposed method is tested on Case Western Reserve University (CWRU) bearing dataset, Driveline Dynamic Simulator (DDS) gearbox dataset, and University of Connecticut (UoC) gearbox dataset under the strong noisy environment, which achieves the accuracies of 98.00%, 99.00%, and 98.88%, respectively. The accuracies of other deep learning methods and singular value difference spectrum method are below 95%. The comparisons show that the proposed method has better noise reduction effect and can diagnose the fault type more accurately.\",\n \"Title: A novel hyperchaotic image encryption algorithm with simultaneous shuffling and diffusion\\nYear: 2022\\n\\nAbstract: The performance of an image encryption algorithm based on chaos is largely determined by the nonlinear characteristics of the underlying chaotic system. This paper proposes a mixed one- and two-dimensional chaotic map (MOTDCM) that has a wider hyperchaotic interval, a larger maximum Lyapunov exponent, and more complex nonlinear dynamics than most existing chaotic systems. Using the hyperchaotic sequences generated by the MOTDCM, a novel image encryption algorithm with different structures is proposed, in which shuffling and diffusion are carried out simultaneously from the perspective of the whole input image. Simulation results and a comparative analysis show that the proposed encryption algorithm has a large key space, high sensitivity to the secret key, and good statistical ciphertext properties. It has a better diffusion effect than existing algorithms and meets the imposed security requirements within only one round of operation, with a reduction in algorithm complexity and an improvement in encryption efficiency. Experimental results demonstrate that this encryption algorithm has good performance and can resist chosen-plaintext attacks and known-plaintext attacks effectively.\",\n \"Title: Experimental Study on the Coupling Mechanism of Sensors under a Strong Electromagnetic Pulse\\nYear: 2022\\n\\nAbstract: This paper presents a study on the electromagnetic effects of a main fuel control system under the action of a strong electromagnetic pulse. It assesses the electromagnetic sensitivity of a speed sensor and a linear variable differential transformer (LVDT) sensor. This assessment focuses on the control system's electromagnetic effects and the sensors' coupling signals. The effects and signals were determined using radiation and pulse current injection tests. Analysis of the signal at the system power port shows that it is the same as those for the two test methods. Based on analysis of the working mechanism and terminal signals of the sensors, the electromagnetic sensitivity of the system is different under the different electromagnetic pulse conditions. The electromagnetic sensitivity characteristics of the sensors were verified, and the electromagnetic effects of the main fuel regulation control system were analyzed. Meanwhile, the degree of sensor coupling mechanism caused by the electromagnetic coupling effects of the main fuel regulation control system under strong electromagnetic pulses were studied. These findings have clear practical implications for electromagnetic pulse protection of aero-engine control systems.\",\n \"Title: Scalable semantic-enhanced supervised hashing for cross-modal retrieval\\nYear: 2022\\n\\nAbstract: Cross-modal retrieval is a classic scenario that can describe the same semantics from multiple angles and provide us with more abundant and diversified information ; thus, it has abundant applications in information retrieval, data mining, and machine learning. Generally, supervised information can enhance the ability to learn hash features, and obtain better retrieval accuracy than unsupervised methods. However, how to effectively solve the discrete binary constraint problem, and how to retain more semantic information in the process of hash learning, which constitute important parts of cross-modal retrieval, still have some challenges. To mitigate these problems, in this work, we propose a novel cross-modal hashing approach, called, scalable pairwise embedding constraint hashing (SPECH). The SPECH approach employs the loss of likelihood similarity technique using pairwise sample data to measure the semantic similarity of heterogeneous modal samples. In light of this, the SPECH approach can maximize the utilization of available data from heterogeneous modal samples for training models. In addition, we further enhance the discriminative capability of hash codes and reduce the intraclass differences in heterogeneous modal samples, by embedding the more discriminative semantic label attribute space regression into the to-be-learned process of hash codes. Comprehensive experiments on three benchmark datasets show that the proposed SPECH approach achieves significantly better retrieval accuracy and outperforms several state-of-the-art approaches on cross-modal retrieval applications.\",\n \"Title: Status Prediction for Age of Information Oriented Short-Packet Transmission in Industrial IoT\\nYear: 2022\\n\\nAbstract: Age of information (AoI), which measures the freshness of information, is a critical performance metric of time-sensitive applications of industrial Internet of things (IIoT) with short-packet transmission (SPT). In this paper, we investigate the suitability of predicting the status updates at source and sending them to destination in advance for AoI oriented SPT systems, in the presence of prediction error as well as transmission error. A predictive transmission scheme is proposed, where proactive transmission termination is adopted as soon as a prediction error is detected, and also multiple correlated features of the status is considered. A closed-form expression for the average AoI with respect to prediction horizon (related to prediction error probability) and blocklength (related to transmission error probability) is derived for the multi-feature source scenario. Also, the prediction error probability with respect to prediction horizon is derived in closed form. It is proved that the average AoI performance can benefit from status prediction, even under high prediction error probability. Simulation results demonstrate the correctness of the analytical results, and show that the proposed prediction scheme outperforms the prediction approach with no transmission termination, and there exists an optimal prediction horizon in terms of average AoI. A tight approximation of the optimal prediction horizon is derived for the special case of single-feature status, which achieves a near-optimal performance, with a much lower complexity than exhaustive search.\",\n \"Title: A Bayesian optimal social law synthesizing mechanism for strategical agents\\nYear: 2022\\n\\nAbstract: One of the effective and well studied approaches for coordinating multiagent systems is to synthesize social laws which restrict the behavior of individual agents. We show that when rational behavior of the agents and private information are considered, the optimal social law synthesizing problem naturally evolves into a setting which can be handled by the framework of algorithmic mechanism design. We focus on the Bayesian case in this paper, that is, the probability distribution of each agent’s private cost is known by the public. In this case, our problem closely relates to path/spanning-tree auctions and Myerson’s optimal auction mechanism, but the optimization objective is new, that is, we focus on profit maximization instead of payment maximization in a reverse auction, and as far as we know in this setting no existing mechanism can be directly applied. By studying this problem: We further extend the logic-based framework of social law optimization problem to the strategic case, and show that it becomes a new problem of algorithmic mechanism design; We find out a mechanism that is incentive compatible, individually rational and maximize the expected profit for all input cost profiles; However, we can show that this mechanism is computational intractable (FP\\n \\n \\n \\n $$^{NP}$$\\n \\n -complete); So, we try to specify Computation Tree Logic semantics as a set of linear-integer constraints, design a Integer-Linear Programming based algorithm for computing the proposed mechanism, enabling it to be handled by current ILP solvers, and finally find out a tractable 2-approximation mechanism.\",\n \"Title: Independent Pilots Versus Shared Pilots: Short Frame Structure Optimization for Heterogeneous-Traffic URLLC Networks\\nYear: 2022\\n\\nAbstract: We investigate a multi-device ultra-reliable low-latency communication system with heterogeneous traffic and finite block length over temporally-correlated fading channels. In light of the challenging demand for accurate channel estimation with limited pilot in a short frame, two frame structures, which respectively adopt independent pilots and shared pilot, are investigated. Block lengths and pilot lengths are jointly optimized for the two frame structures, through instantaneous channel state information (CSI) based dynamic optimization and statistical CSI based static optimization, to strike the tradeoffs among performance, complexity and signaling overhead. The proposed joint optimization algorithms significantly outperform the existing approaches that solely optimize block lengths or pilot lengths. The dynamic optimization algorithms achieve near-optimal performance at dramatic complexity reduction over exhaustive search, and maintain robustness against traffic heterogeneity. Also, the static optimization algorithms are conducted offline, while still outperforming the previous instantaneous CSI based dynamic optimization approaches. It is demonstrated that the independent-pilot frame structure with dynamic optimization is preferable in the scenario with high traffic heterogeneity or high mobility, and that the shared-pilot frame structure with static optimization presents a comparable performance to the former in the case of low mobility, incurring negligible complexity and signaling overhead.\",\n \"Title: An explainable multi-sparsity multi-kernel nonconvex optimization least-squares classifier method via ADMM\\nYear: 2022\\n\\nAbstract: Convex optimization techniques are extensively applied to various models, algorithms, and applications of machine learning and data mining. For optimization-based classification methods, the sparsity principle can greatly help to select simple classifier models, while the single- and multi-kernel methods can effectively address nonlinearly separable problems. However, the limited sparsity and kernel methods hinder the improvement of the predictive accuracy, efficiency, iterative update, and interpretable classification model. In this paper, we propose a new Explainable Multi-sparsity Multi-kernel Nonconvex Optimization Least-squares Classifier (EM2NOLC) model, which is an optimization problem with a least-squares objective function and multi-sparsity multi-kernel nonconvex constraints, aiming to address the aforementioned issues. Based on reconstructed multiple kernel learning (MKL), the proposed model can extract important instances and features by finding the sparse coefficient and kernel weight vectors, which are used to compute importance or contribution to classification and obtain the explainable prediction. The corresponding EM2NOLC algorithm is implemented with the Alternating Direction Method of Multipliers (ADMM) method. On the real classification datasets, compared with the three ADMM classifiers of Linear Support Vector Machine Classifier, SVMC, Least Absolute Shrinkage and Selection Operator Classifier, the two MKL classifiers of SimpleMKL and EasyMKL, and the gradient descent classifier of Feature Selection for SVMC, our proposed EM2NOLC generally obtains the best predictive performance and explainable results with the least number of important instances and features having different contribution percentages.\",\n \"Title: Characterizing the fuzzy community structure in link graph via the likelihood optimization\\nYear: 2022\\n\\nAbstract: Detecting fuzzy communities in networks is a critical but challenging task in many fields including biology, technology, social system and so on. Current technology is largely reliant on network topology, which can’t provide much useful information such module correlations and hierarchies. This paper introduces a new fuzzy community detection method by maximizing the likelihood function, which can dynamically map the link community structure to the optimal state. To specify the direction and weight of the link graph, the directional node graph is first converted into a new type of link graph. Next, we define a communityunit consisting of membership and correlation information in a link graph. Specially, we define this likelihood function with a posteriori probability that aims to find the optimal partition under the prior conditions of given link graph and granularity. Furthermore, based on the spectral analysis of Markov transition matrices, we give a strict mathematical analysis for identifying the optimal number of network communities by analyzing the stability of community structures. Finally, extensive experiments on both artificial benchmarks and real-world networks show that our algorithm consistently achieving higher accuracy than classical methods.\",\n \"Title: Adaptive Transformer-Based Conditioned Variational Autoencoder for Incomplete Social Event Classification\\nYear: 2022\\n\\nAbstract: ABSTRACTWith the rapid development of the Internet and the expanding scale of social media, incomplete social event classification has increasingly become a challenging task. The key for incomplete social event classification is to accurately leverage the image-level and text-level information. However, most of the existing approaches may suffer from the following limitations: (1) Most Generative Models use the available features to generate the incomplete modality features for social events classification while ignoring the rich semantic label information. (2) The majority of existing multi-modal methods just simply concatenate the coarse-grained image features and text features of the event to get the multi-modal features to classify social events, which ignores the irrelevant multi-modal features and limits their modeling capabilities. To tackle these challenges, in this paper, we propose an Adaptive Transformer-Based Conditioned Variational Autoencoder Network (AT-CVAE) for incomplete social event classification. In the AT-CVAE, we propose a novel Transformer-based Conditioned Variational Autoencoder to jointly model the textual information, visual information and label information into a unified deep model, which can generate more discriminative latent features and enhance the performance of incomplete social event classification. Furthermore, the Mixture-of-Experts Mechanism is utilized to dynamically acquire the weights of each multi-modal information, which can better filter out the irrelevant multi-modal information and capture the vitally important information. Extensive experiments are conducted on two public event datasets, demonstrating the superior performance of our AT-CVAE method.\",\n \"Title: Order-aware Human Interaction Manipulation\\nYear: 2022\\n\\nAbstract: ABSTRACTThe majority of current techniques for pose transfer disregard the interactions between the transferred person and the surrounding instances, resulting in context inconsistency when applied to complicated situations. To tackle this issue, we propose InterOrderNet, a novel framework to perform order-aware interaction learning. The proposed InterOrderNet learns the relative order on the direction of the z-axis among instances to describe instance-level occlusions. Not only does learning this order guarantee the context consistency of human pose transfer, but it also enhances its generalization to natural scenes. Additionally, we present a novel unsupervised method, named Imitative Contrastive Learning, which sidesteps the requirements of order annotations. Existing pose transfer methods are easy to be integrated into the proposed InterOrderNet. Extensive experiments demonstrate that InterOrderNet enables these methods to perform interaction manipulation.\",\n \"Title: An Augmented Benchmark Dataset for Geometric Question Answering through Dual Parallel Text Encoding.\\nYear: 2022\\n\\nAbstract: Automatic math problem solving has attracted much attention of NLP researchers recently. However, most of the works focus on the solving of Math Word Problems (MWPs). In this paper, we study on the Geometric Problem Solving based on neural networks. Solving geometric problems requires the integration of text and diagram information as well as the knowledge of the relevant theorems. The lack of high-quality datasets and efficient neural geometric solvers impedes the development of automatic geometric problems solving. Based on GeoQA, we newly annotate 2,518 geometric problems with richer types and greater difficulty to form an augmented benchmark dataset GeoQA+, containing 6,027 problems in training set and 7,528 totally. We further perform data augmentation method to expand the training set to 12,054. Besides, we design a Dual Parallel text Encoder DPE to efficiently encode long and medium-length problem text. The experimental results validate the effectiveness of GeoQA+ and DPE module, and the accuracy of automatic geometric problem solving is improved to 66.09%.\",\n \"Title: Age of Loop Oriented Wireless Networked Control System: Communication and Control Co-Design in the FBL Regime\\nYear: 2022\\n\\nAbstract: In this paper, age of loop (AoL) is investigated for communication and control co-design in wireless networked control systems (WNCSs), to ensure information freshness in the sensor-controller-actuator loop with two-way delay. Also, the impact of fading channel and finite block length due to limited transmission bits are considered. A closed-form expression for Peak AoL(PAoL) is derived with respect to block length and the number of allowable transmissions in the presence of block error probability. It is shown that the average PAoL is less than the sum of the average peak Age of information (PAoI) in uplink (PAoU) and the PAoI in downlink (PAoD), due to the coupling of uplink and downlink. In addition, the average PAoL is proved to be convex with respect to block length and mono-decreasing with respect to the allowable transmission time. We then propose a low-complexity PAoL-aware block length adaptation scheme with retransmission policy and optimal control. Simulation results verify the correctness of our analytical results and demonstrate the effectiveness of the proposed PAoL-aware block length adaptation scheme in terms of information freshness and control performance.\",\n \"Title: 3D Printing of PEDOT:PSS-PU-PAA Hydrogels with Excellent Mechanical and Electrical Performance for EMG Electrodes\\nYear: 2022\\n\\nAbstract: Bioelectronics has been developed for recording the electrophysiological activity of diagnostic and therapeutic devices. However, current bioelectrodes still imperfectly comply with tissues, which results in high interfacial impedance and even mechanical detachment. Herein, we report a simple yet effective approach to overcome such hurdles by designing a highly conductive, adhesive hydrogel composite based on freeze-dried poly(3,4-ethylenedioxythiophene):poly (styrene sulfonate) (PEDOT:PSS), polyurethane (PU), and poly(acrylic acid) (PAA). With the continuous phase-separation of PEDOT:PSS, PU, and PAA, the resultant composite hydrogels can simultaneously achieve high adhesion (lap-shear strength > 8 kPa), stretchability (fracture strain > 1100%), and electrical conductivity (conductivity > 2 S/m) by overcoming the traditional trade-off between mechanical and electrical properties in conducting polymer hydrogels. Moreover, such hydrogels are readily applicable to advanced manufacturing techniques such as 3D printing. We further fabricated skin electrodes and achieved high quality and high signal-to-noise ratio EMG signal recording of the forearm.\",\n \"Title: Impact of Quasi-Biweekly Oscillation on Southeast Asian Cold Surge Rainfall Monitored by TRMM Satellite Observation\\nYear: 2022\\n\\nAbstract: Based on the Tropical Rainfall Measuring Mission (TRMM) satellite observation and ERA5 re-analysis dataset, this paper studies the influence of the northwestward-propagating quasi-biweekly oscillation (QBWO) over the western North Pacific on cold surge rainfall (CSR) over Southeast Asia. Cold surges are the most important driver affecting Southeast Asian rainfall on a synoptic scale. The presence of the QBWO during phases 6-8, in which the associated active convection coupling with a cyclonic circulation reaches Southeast Asia, provides a favorable environment for the increase of CSR. The increase in CSR primarily occurs east of the Philippines, leading to a high likelihood of triggering extreme rainfall. The effects from the QBWO are independent of those from the active MJO phases over Southeast Asia. Additionally, cold surge activity could also be influenced by the QBWO. An examination of the QBWO and MJO indicates that the most preferred phases for the occurrence of cold surges are the time when phase 1 of the QBWO co-exists with phase 7 of the MJO or the time when phase 7 of the QBWO couples with phase 5 of the MJO. Accordingly, about 40% of the total cold surge days would fall in either combination.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":200,"string":"200"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13578,"string":"13,578"}}},{"rowIdx":13579,"cells":{"question_id":{"kind":"string","value":"Bing Xiao"},"question":{"kind":"string","value":"Generate an abstract for the title \"An Analysis Method of the Supply-side Influencing Factors of \"Vocational Master of Education\" Based on ISM\" using the following items: 1. Postgraduate programs\n2. Vocational master of education\n3. Interpretative structural model\n4. Influencing factors\n5. Synergy effect"},"answer":{"kind":"string","value":"In order to explore new approaches of training postgraduates of taught postgraduate programs, this paper proposed an analysis method to discuss the supply-side influencing factors of \"vocational master of education\" using the interpretative structural model (ISM). The research results showed that the quality of postgraduates, recruitment plan and the number of graduated students are the direct influencing factors of the supply side of \"vocational master of education\". Teaching with practices, dissertation, postgraduate management, employment policy and the actual recruitment are the necessary influencing factors. Training program and courses, faculty, basic environment, entrance examination, program promotion, professional direction and planned enrolment are the indirect influencing factors. The strategic environment, the profession of secondary vocational schools and the demand of teachers in these schools are the fundamental influencing factors. Therefore, when studying the supply side of \"vocational master of education\", we analysed top-down with specific emphasis on its hierarchy to strengthen the synergy effect of the subject."},"correct_answer":{"kind":"string","value":"In order to explore new approaches of training postgraduates of taught postgraduate programs, this paper proposed an analysis method to discuss the supply-side influencing factors of \"vocational master of education\" using the interpretative structural model (ISM). The research results showed that the quality of postgraduates, recruitment plan and the number of graduated students are the direct influencing factors of the supply side of \"vocational master of education\". Teaching with practices, dissertation, postgraduate management, employment policy and the actual recruitment are the necessary influencing factors. Training program and courses, faculty, basic environment, entrance examination, program promotion, professional direction and planned enrolment are the indirect influencing factors. The strategic environment, the profession of secondary vocational schools and the demand of teachers in these schools are the fundamental influencing factors. Therefore, when studying the supply side of \"vocational master of education\", we analysed top-down with specific emphasis on its hierarchy to strengthen the synergy effect of the subject."},"documents":{"kind":"list like","value":["Title: Attitude Stabilization of Spacecrafts Under Actuator Saturation and Partial Loss of Control Effectiveness\nYear: 2013\n\nAbstract: A practical solution is presented to the problem of fault tolerant attitude stabilization for a rigid spacecraft by using feedback from attitude orientation only. The attitude system, represented by modified Rodriguez parameters, is considered in the presence of external disturbances, uncertain inertia parameters, and actuator saturation. A low-cost control scheme is developed to compensate for the partial loss of actuator effectiveness fault. The derived controller not only has the capability to protect the control effort from actuator saturation but also guarantees all the signals in the closed-loop system to be uniformly ultimately bounded. Another feature of the approach is that the implementation of the controller does not require any rate sensor to measure angular velocity. An example is included to verify those highly desirable features in comparison with the conventional velocity-free control strategy.","Title: Taxihailer: A Situation-Specific Taxi Pick-Up Points Recommendation System\nYear: 2014\n\nAbstract: This demonstration presents TaxiHailer, a situation-specific recommendation system for passengers who are eager to find a taxi. Given a query with departure point, destination and time, it recommends pick-up points within a specified distance and ranked by potential waiting time. Unlike existing works, we consider three sets of features to build regression models, as well as Poisson process models for road segment clusters. We evaluate and choose the most proper models for each cluster under different situations. Also, TaxiHailer gives destination-aware recommendations for pick-up points with driving directions. We evaluate our recommendation results based on real GPS datasets.","Title: Exponential convergence of an epidemic model with continuously distributed delays\nYear: 2008\n\nAbstract: In this paper the convergence behavior of an epidemic model with continuously distributed delays is considered. Some new sufficient conditions are established to ensure that all the solutions of the epidemic model with permitted initial conditions converge exponentially to zero.","Title: Competitive analysis of network load balancing\nYear: 1997\n\nAbstract: This paper presents a theoretical analysis of the Load Balancing Problem (LBP) in a network of processing units. The performance objective is to minimize the makespan, i.e., the time spent to finish all jobs in a network of processing units. Because of the communication delay that results from the network topology, it is impossible to have a strategy which obtains the exact optimum under all load distributions. Instead, we measure the information efficiency of a load balancing policy by the worst case ratio of the solution (for each load distribution) of a load balancing policy to the optimal solution (for the same load distribution) assuming that processors have complete information about the load distribution over the network. This ratio is called the competitive ratio of this strategy [17, 24, 34]. In particular, a policy is called competitive if this ratio is bounded by a constant. As a first step, we discuss the centralized LBP, where all the processors have complete information of the load distribution over a network. Its solution serves as a benchmark to compare with realistic strategies, both in theoretical analysis, and experimental and simulational studies of distributed algorithms. We show that when jobs have different sizes, even with preemptive scheduling, LBP is NP–complete. When the jobs are of the same size, we give a polynomial algorithm, using network–flow techniques, which extends to approximate solutions for jobs of different sizes. We apply this benchmark solution in order to analyze the competitiveness for three network topologies: completely connected graphs, rings, and hierarchical complete k -ary trees. The constant competitive ratio results for complete network and hierarchical complete k -ary trees are applied to a study on the issues of network designs suitable for the LBP. We further discuss the problem for general networks with jobs of different sizes for slightly weaker results than those for the constant competitive ratio requirement. Finally, we comment on the related issues of job partitioning over parallel/distributed systems.","Title: Photo-sketch synthesis and recognition based on subspace learning\nYear: 2010\n\nAbstract: This paper aims to reducing difference between sketches and photos by synthesizing sketches from photos, and vice versa, and then performing sketch-sketch/photo-photo recognition with subspace learning based methods. Pseudo-sketch/pseudo-photo patches are synthesized with embedded hidden Markov model. Because these patches are assembled by averaging their overlapping area in most of the local strategy based methods, which leads to blurring effect to the resulted pseudo-sketch/pseudo-photo, we integrate the patches with image quilting. Experiments are carried out to demonstrate that the proposed method is effective to produce pseudo-sketch/pseudo-photo with high quality and achieve promising recognition results.","Title: Existence and uniqueness of almost periodic solutions for a class of Hopfield neural networks with neutral delays\nYear: 2009\n\nAbstract: In this work Hopfield neural networks with neutral delays are considered. Some sufficient conditions for the existence and uniqueness of almost periodic solutions are established by using the fixed point theorem and differential inequality techniques. The results of this work are new and complement previously known results.","Title: A knot or not a knot? SETting the record 'straight' on proteins\nYear: 2003\n\nAbstract: A novel knot found in the SET domain is examined in the light of five recent crystal structures and their descriptions in the literature. Using the algorithm of Taylor it was established that the backbone chain does not form a true knot. However, only two crosslinks corresponding to hydrogen-bonds were needed to form a knotted structure. Such loosely knotted structures formed by hydrogen-bonded crosslinks were assessed as lying between covalent crosslinks (such as disulphide bonds) and threaded-loops which are formed by close (unbonded) contacts between different parts of the chain. The term pseudo-knot was introduced (from the RNA field) to distinguish hydrogen-bonded 'knots'.","Title: Attitude Tracking Control of Rigid Spacecraft With Actuator Misalignment and Fault\nYear: 2013\n\nAbstract: A control scheme is developed to address the attitude tracking problem of a rigid spacecraft. A sufficient condition to accommodate actuator misalignment is presented. The controller asymptotically stabilizes the closed-loop attitude tracking system. The desired attitude trajectories can be tracked in finite time, even in the presence of uncertain inertia tensor, external disturbances, actuator fault, and actuator misalignment. The attitude tracking performance of the controller is evaluated through an illustrative example.","Title: Adaptive frame rate up-conversion based on motion classification\nYear: 2008\n\nAbstract: In this paper, a new technique on video frame rate up-conversion (FRUC) is presented by combining the adaptive motion classification (AMC) for image sequences and the mixed motion estimation (ME). In the proposed FRUC scheme, the AMC classifies sequences into global and local motion sequences (G/LMS) whose advantages are fully utilized. The mixture of bidirectional ME (BDME) and unidirectional ME (UDME) is proposed to avoid hole and ''background replace object'' phenomenon in interpolated frames. Moreover, the motion vectors (MVs) refinement technique is employed to smooth MVs field. For LMS, the motion detection is used to deal with static background and occlusions are located by using forward prediction of motion regions. Correctional sum of absolute difference (CSAD) which gives more precise physical MVs field than conventional SAD is applied during all of the ME process. Simulation results demonstrate the effectiveness of the proposed FRUC algorithm.","Title: The general result of Gol'dberg's theorem concerning the growth of meromorphic solutions of algebraic differential equations\nYear: 2009\n\nAbstract: In this paper, we give a general estimate result of Gol'dberg's theorem concerning finite growth order of meromorphic solutions of first-order algebraic differential equations by using the normal family theory. It is an extending result of corresponding theorem for Bergweiler and Barsegian and so on. We also give some examples to show that our result is sharp in special case.","Title: Deterministic load balancing in computer networks\nYear: 1990\n\nAbstract: The paper presents theoretical analysis of the deterministic complexity of the load balancing problem (LBP). Because of difficulty of the general problem, research in the area mostly restricts itself to probabilistic or approximation algorithms, or to the average behavior of a network. The paper provides deterministic analysis of the problem for general networks. It focuses on the worst-case complexity analysis of the problem. It shows certain cases of the LBP to be NP-complete. The paper also discusses situations closely related to computer networks, where there is a global view of load distribution in the network; it provides a polynomial algorithm for solving the load balancing problem in this network.","Title: Visual Quality Assessment Of The Synthesized Sketch\nYear: 2013\n\nAbstract: The existing image and video quality assessment algorithms are devoted to the degradation of quality caused by some artefacts or noise. We propose a novel image quality assessment for the synthesized sketch. When the sketch is synthesized from a photo, the quality of the resulted sketch is evaluated from two aspects. First, the synthesis of sketch may be affected by artefacts and therefore its visual quality should be measured with no-reference quality evaluation algorithm; secondly, the synthesized sketch should preserve discrimination information as the corresponding photo, and their similarity is also examined in the proposed method, with the photo as reference image. Because sketch synthesis aims at enhancing the performance of sketch-photo recognition and the purpose of sketch quality assessment is to determine whether the quality of the resulted sketch is sufficient for sketch-photo recognition, the proposed sketch quality assessment is associated with the recognition rate of sketch-photo recognition, that is to say, the rationality of the proposed method is judged according to the performance of sketch-photo recognition. A set of experiments are performed to prove that our method is more valid than the traditional methods.","Title: A Face Recognition Scheme Based On Embedded Hidden Markov Model And Selective Ensemble Strategy\nYear: 2009\n\nAbstract: As an effective method, the embedded hidden Markov model (E-HMM) has been widely used in pattern recognition. On applying the E-HMM to face recognition, the performance heavily depends on the selection of model parameters. Aiming at the problem of model selection, a selective ensemble of multi E-HMMs based face recognition algorithm is proposed. Experimental results illustrate that compared with the traditional E-HMM based face recognition algorithm the proposed method cannot only obtain better and more stable recognition performance, but also achieve higher generalization ability.","Title: Adaptive fault tolerant control using integral sliding mode strategy with application to flexible spacecraft\nYear: 2013\n\nAbstract: Adaptive-based integral sliding mode control scheme is developed to solve the actuator fault-tolerant compensation problem for linear time-invariant system in the presence of unknown actuator faults and external disturbances. A nonlinear integral-type sliding manifold is first presented that incorporates a virtual nominal control to achieve prescribed specifications of the perturbed system, and an adaptive sliding mode controller is constructed to automatically compensate for external disturbances and unknown time-invariant faults. It is shown that the proposed controller has the capability to guarantee that the resulting closed-loop system is asymptotically stable. Control design methodology is then extended to tackle with the unknown time-varying actuator faults. It is proved that any given level of gain attenuation from external disturbance/parametric estimation error to system output is achieved with the developed control law. The closed-loop performance of the new control solution derived here is evaluated extensively through numerical simulations in which the flexible spacecraft attitude control under both the external disturbances and actuator faults are considered.","Title: Biview face recognition in the shape-texture domain\nYear: 2013\n\nAbstract: Face recognition is one of the biometric identification methods with the highest potential. The existing face recognition algorithms relying on the texture information of face images are affected greatly by the variation of expression, scale and illumination. Whereas the algorithms based on the shape topology weaken the influence of illumination to some extent, but the impact of expression, scale and illumination on face recognition is still unsolved. To this end, we propose a new method for face recognition by integrating texture information with shape information, called biview face recognition algorithm. The texture models are constructed by using subspace learning methods and shape topologies are formed by building graphs for face images. The proposed biview face recognition method is compared with recognition algorithms merely based on texture or shape information. Experimental results of recognizing faces under the variation of illumination, expression and scale demonstrate that the performance of the proposed biview face recognition outperforms texture-based and shape-based algorithms.","Title: A new approach for face recognition by sketches in photos\nYear: 2009\n\nAbstract: Face recognition by sketches in photos remains a challenging task. Unlike the existing sketch-photo recognition methods, which convert a photo into sketch and then perform the sketch-photo recognition through sketch-sketch recognition, this paper devotes to synthesizing a photo from the sketch and transforming the sketch-photo recognition to photo-photo recognition to achieve better performance in mixture pattern recognition. The contribution of this paper mainly focuses on two aspects: (1) in view of that there are no many research findings of sketch-photo recognition based on the pseudo-photo synthesis and the existing methods require a large set of training samples, which is nearly impossible to achieve for the high cost of sketch acquisition, we make use of embedded hidden Markov model (EHMM), which can learn the nonlinearity of sketch-photo pair with less training samples, to produce pseudo-photos in terms of sketches; and (2) photos and sketches are divided into patches and pseudo-photo is generated by combining pseudo-photo patches, which makes pseudo-photo more recognizable. Experimental results demonstrate that the newly proposed method is effective to identify face sketches in photo set.","Title: Predicting Users' Age Range In Micro-Blog Network\nYear: 2013\n\nAbstract: In this report, we present our work on WISE 2013 Challenge Track II to predict the age range of Weibo users. In this challenge, a dataset consisting of Sina Weibo user information was presented. The goal of the challenge is to predict users age range. With personal information and original tweets for over one million users as training data, we analyze and process the dataset, and experiment a series of prediction methods including SVM, decision tree etc. The result shows that ensemble classifiers based on AdaBoost achieves the best prediction results in this challenge.","Title: A survey of graph edit distance\nYear: 2010\n\nAbstract: Inexact graph matching has been one of the significant research foci in the area of pattern analysis. As an important way to measure the similarity between pairwise graphs error-tolerantly, graph edit distance (GED) is the base of inexact graph matching. The research advance of GED is surveyed in order to provide a review of the existing literatures and offer some insights into the studies of GED. Since graphs may be attributed or non-attributed and the definition of costs for edit operations is various, the existing GED algorithms are categorized according to these two factors and described in detail. After these algorithms are analyzed and their limitations are identified, several promising directions for further research are proposed.","Title: Adaptive Sliding Mode Fault Tolerant Attitude Tracking Control for Flexible Spacecraft Under Actuator Saturation.\nYear: 2012\n\nAbstract: A novel fault tolerant attitude tracking control scheme is developed for flexible spacecraft with partial loss of actuator effectiveness fault. Neural networks are first introduced to account for system uncertainties, and an adaptive sliding mode controller is derived by using on-line updating law to estimate the bound of actuator fault such that any information of the fault is not required. To further address actuator saturation problem, a modified fault tolerant control law is then presented to ensure that the resulting control signal will never incur saturation. It is shown that the roll, pitch and yaw angle trajectories can globally asymptotically track the desired attitude in the face of faulty actuator, system uncertainties, external disturbances and even actuator saturation. A simulation example of a flexible spacecraft is given to illustrate the effectiveness of the proposed controller. © 2012 IEEE.","Title: HMM-based graph edit distance for image indexing\nYear: 2008\n\nAbstract: Most of the existing graph edit distance (GED) algorithms require cost functions which are difficult to be defined exactly. In this article, we propose a cost function free algorithm for computing GED. It only depends on the distribution of nodes rather than node or edge attributes in graphs. Hidden Markov model (HMM) is employed to model the distribution of feature points and thus dissimilarity measure of graphs can be posed as distance of HMMs. A fast algorithm of Kullback-Leibler Distance, suitable for computing the distance between two probability models, is adopted to compute the distance of HMMs. Experimental results demonstrate that the proposed GED algorithm can characterize the structure variety of graphs effectively and is available for clustering and indexing images of both rigid and nonrigid bodies. © 2008 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 18, 209–218, 2008","Title: Image categorization: Graph edit distance + edge direction histogram\nYear: 2008\n\nAbstract: This paper presents a novel algorithm for computing graph edit distance (GED) in image categorization. This algorithm is purely structural, i.e., it needs only connectivity structure of the graph and does not draw on node or edge attributes. There are two major contributions: (1) Introducing edge direction histogram (EDH) to characterize shape features of images. It is shown that GED can be employed as distance of EDHs. This algorithm is completely independent on cost function which is difficult to be defined exactly. (2) Computing distance of EDHs with earth mover distance (EMD) which takes neighborhood bins into account so as to compute distance of EDHs correctly. A set of experiments demonstrate that the newly presented algorithm is available for classifying and clustering images and is immune to the planar rotation of images. Compared with GED from spectral seriation, our algorithm can capture the structure change of graphs better and consume 12.79% time used by the former one. The average classification rate is 5% and average clustering rate is 25% higher than the spectral seriation method. © 2008 Elsevier Ltd. All rights reserved.","Title: Finite-Time Attitude Tracking of Spacecraft With Fault-Tolerant Capability\nYear: 2015\n\nAbstract: A finite-time attitude tracking control scheme is presented for rigid spacecraft subject to constant but unknown inertia and external disturbances. The controller, developed using sliding mode control technique, has great fault-tolerant capability to accommodate four types of actuator faults. Different from most of the existing works on attitude fault-tolerant control (FTC), the developed controller guarantees the desired attitude to be followed in finite time, which is critical for FTC systems. Moreover, the convergence time is an explicit parameter for designer’s choice. Thus, the controller design simply meets the finite-time requirement. The attitude-tracking performance is evaluated through a numerical example.","Title: A Review on Recent Development of Spacecraft Attitude Fault Tolerant Control System.\nYear: 2016\n\nAbstract: Motivated by several accidents, attitude control of a spacecraft subject to faults/failures has gained considerable attention in a wider range of aerospace engineering and academic communities. This paper is concerned with industrial practices and theoretical approaches for fault tolerant control (FTC) and fault detection and diagnosis (FDD) in spacecraft attitude control system. An overview on re...","Title: Tracking Control of Robotic Manipulators With Uncertain Kinematics and Dynamics.\nYear: 2016\n\nAbstract: This paper investigates a difficult problem of tracking control for robotic manipulations with guaranteed high accuracy. Uncertain kinematics, unknown torques including unknown gravitational torque, unknown friction torque, and uncertain dynamics induced by uncertain moment of inertia and disturbance, are addressed. The approach is developed in the framework of observer-based control design. Two s...","Title: A Novel Disturbance Estimation Scheme for Formation Control of Ocean Surface Vessels.\nYear: 2017\n\nAbstract: To achieve high-accuracy formation control of multiple ocean surface vessels, the external disturbance acting on each vessel should be accommodated. In this paper, a novel disturbance estimation scheme is presented to solve this problem, and to provide the future formation controller design with precise estimation information. This approach is developed in the theoretical framework of a terminal s...","Title: NIOSIT: efficient data access for log-structured merge-tree style storage systems.\nYear: 2017\n\nAbstract: Recent years, the log-structured merge-tree(LSM-tree) style storage has been widely adopted in distributed data storage systems(e.g. Bigtable and HBase) and commercial database systems(e.g. Ocean-Base, Cassandra, SQLite, etc.) to provide both large-volume storage capacity and high-performance data updates. Write operations become easier as the LSM-tree style storage avoids writing in place by updating a data copy in memory. However, read operations are affected as it requires an additional step during a data compaction to check if there exists the newest update of data record in memory, which brings many of costly empty reads in real usage since the volume of immutable data is pervasively far more massive than incremental delta data. To address this issue, we design a new network request processing mechanism to allow data access being processed in an auxiliary lightweight network communication IO thread. And a Bloom filter is incorporated with the network IO thread to effectively filter out the empty reads. We also analyze the efficiency advantage of the mechanism and introduce its detailed implementation based on the well-known OceanBase system. Experimental study using the YCSB benchmark demonstrates the proposed mechanism can significantly achieve 20 percent to 30 percent better performance than existing method.","Title: A Fault-Tolerant Cooperative Positioning Approach for Multiple UAVs.\nYear: 2017\n\nAbstract: Due to the increasing potentials and benefits of unmanned aerial vehicles (UAVs) in civil and surveillance applications, the cooperative fight using multiple UAVs has attracted more and more attention. However, fault-tolerant cooperative positioning for malfunctions in global positioning system (GPS) is one of the challenges that need to be addressed in various practical applications of UAVs. Motivated by solving this issue, this paper presents a reliable cooperative positioning approach for multiple UAVs cooperative fiight by using the geometric azimuth angle and the inclination angle relative to the reference UAVs. In this proposed cooperative positioning system (COPS), these relative angles are measured through radio direction finding equipped on UAVs. With the horizontal dilution of precision of the various reference UAVs carefully analyzed, the optimal reference UAVs set can be first selected. Based on a constant acceleration kinematic model, an extended Kalman filter is then employed to improve the positioning accuracy. A COPS/GPS faulttolerant navigation algorithm is finally developed to accommodate GPS fault. Simulation results are further presented to verify that the proposed algorithm can guarantee satisfactory navigation of the UAVs even when their GPS cannot work.","Title: Efficient Multi-version Storage Engine for Main Memory Data Store.\nYear: 2017\n\nAbstract: Multi-version storage engine is the fundamental component of modern main memory data store using the popular multiple version concurrency control (MVCC). The straightforward implementation of storage engine is to use a linked list to store multiple versions of an object. A read operation has to traverse the list for the specified version, which incurs pointer chasing. An optimization method implemented in HyPer is to store the current version in the object header, which is friendly to read the latest snapshot of data. However, a read operation still needs one extra pointer chasing in memory when accessing an object being updated. In this paper, we propose an efficient multi-version storage (EMS), a new storage engine for main memory data store. EMS embeds two latest versions in each object header, so that it can avoid the overhead of traversal of version list, especially in the update-intensive scenario. We present an implementation mechanism of widely used snapshot isolation level over EMS. The experimental results demonstrate that EMS outperforms the exiting multi-version storage engine of well-known main memory data stores in terms of throughput without excessive memory consumption.","Title: A New Disturbance Attenuation Control Scheme for Quadrotor Unmanned Aerial Vehicles.\nYear: 2017\n\nAbstract: This paper addresses a difficult problem of high-accuracy control for quadrotor unmanned aerial vehicles (UAVs) subject to external disturbance force and unknown disturbance torque. An observer-based full control scheme is presented. In the strategy, two observer-based estimators are first designed to estimate external disturbance force and torque, respectively. With the application of the precise...","Title: A Data-Driven Clustering Approach for Fault Diagnosis.\nYear: 2017\n\nAbstract: Clustering is an important approach in fault diagnosis. The dominant sets algorithm is a graph-based clustering algorithm, which defines the dominant set as a concept of a cluster. In this paper, we make an in-depth investigation of the dominant sets algorithm. As a result, we find that this algorithm is dependent on the similarity parameter in constructing the pairwise similarity matrix, and has the tendency to generate spherical clusters only. Based on the merits and drawbacks of this algorithm, we apply the histogram equalization transformation to the similarity matrices for the purpose of removing the influence of similarity parameters, and then use a density-based cluster expansion process to improve the clustering results. In experimental validation of the proposed algorithm, we use two criterions to evaluate the clustering results in order to arrive at convincing conclusions. Data clustering experiments on ten data sets and fault detection experiments on the Tennessee Eastman process demonstrate the effectiveness of the proposed algorithm.","Title: An Intelligent Actuator Fault Reconstruction Scheme for Robotic Manipulators.\nYear: 2018\n\nAbstract: This paper investigates a difficult problem of reconstructing actuator faults for robotic manipulators. An intelligent approach with fast reconstruction property is developed. This is achieved by using observer technique. This scheme is capable of precisely reconstructing the actual actuator fault. It is shown by Lyapunov stability analysis that the reconstruction error can converge to zero after ...","Title: The development of an electronic control system for diesel-natural gas dual-fuel engine.\nYear: 2018\n\nAbstract: To meet the actual application requirements for the diesel-natural gas dual-fuel engine refit, a new electronic control system for the dual-fuel engine is developed in the study. Specifically, an active mode switching board is developed to achieve flexible switching between the pure diesel mode and dual-fuel mode. A diesel nozzle physical simulator is developed to ensure that the original diesel electronic control unit does not trigger fault alarm when engine works in the dual-fuel mode. Moreover, a dual-fuel electronic control unit, which uses the high-speed and multicore TMS320F28M35 as its microcontroller unit, is additionally developed on the basis of the original diesel electronic control unit. The peak and hold current shape for the diesel nozzle is intelligently controlled by the software program in C2000 core. Experiments reveal that the developed electronic control system can select a proper working mode according to the engine operation condition and smoothly switch the working mode without any fault alarms.","Title: Semi-Supervised Image Segmentation With Globalized Probability Of Boundary And Simple Linear Iterative Clustering\nYear: 2017\n\nAbstract: Semi-supervised image segmentation in order to get a divided image from the tag portion. In this paper, a new semi-supervised image segmentation framework with global boundaries of probability (gPb) and simple linear iterative clustering (SLIC) is proposed. We use SLIC to perform image segmentation of super pixels, and we have constructed a graph, in which each vertex is a superpixel. Then we utilize gPb to define the edge weights for the graph. Semi-supervised learning based image (GSSL) method, and the use of labeled samples to calculate anchor anchor category, the category may be further determined by the anchor point of the sample. The results of the experiments certify the robustness and effectiveness of the method proposed in this paper.","Title: A Graying Algorithm Based On Maximum Preserving Perception Features Of Images Whole Structure\nYear: 2017\n\nAbstract: Although color images have gradually become the mainstream in image representation, gray images are still drawing more attention because of their small data amount and simple operation. Existing graying algorithms are difficult to ensure that the whole structure of images is not damaged and contrast difference information is consistent after dimensionality reduction. They cannot convey high quality gray images with human visual information. An optimization algorithm is proposed in this paper based on improving the objective energy function to retain these visual characteristics utmostly. The gray variance constraint is constructed to utmostly retain overall perceptual features of different gradients of the original color image regions, so as to ensure the stability of the global structure of the image. Let the difference between gray and color contrast be approximated to construct chromatic aberration of pixel pairs constraint term, and the contrast information is retained maximally. The final energy function is obtained by the linear combination of these two constraint terms. The iterative method is used to obtain the optimal channel weight coefficients which satisfy the minimum energy function and then get the result image. Finally, the experimental results show that the algorithm in this paper is basically consistent with the human visual perception of image contrast changes, and it can maintain the details and the global structure better.","Title: Adaptive Quasi-Optimal Higher Order Sliding-Mode Control Without Gain Overestimation.\nYear: 2018\n\nAbstract: This paper presents an adaptive quasi-optimal higher order sliding-mode control (HOSMC) scheme, which is able to avoid gain overestimation. The overall scheme consists of two elements: 1) a quasi-optimal control law that provides fast finite-time stabilization for a chain of integrators; and 2) an adaptive HOSMC with integration of the quasi-optimal control and the integral sliding-mode concept. T...","Title: Adaptive attitude tracking control of a 3-degrees-of-freedom experimental helicopter with actuator dead-zone.\nYear: 2019\n\nAbstract: In this article, the attitude tracking problem of a three rotational degrees-of-freedom experimental helicopter is studied based on adaptive fuzzy control method. The system unmodeled dynamics, external disturbances, and actuator dead-zone fault are taken into consideration simultaneously. The uncertainties and the faults are compensated by resorting to the fuzzy logic system. Based on adaptive backstepping technique and Lyapunov stability theory, state feedback controllers are designed such that the reference signals can be tracked by the elevation and pitch angles, respectively, within small errors. Finally, experimental results are given to demonstrate the reliability of the proposed method.","Title: Observer-based control for robotic manipulations with uncertain kinematics and dynamics\nYear: 2016\n\nAbstract: This paper investigates the problem of tracking control for robotic manipulations with high-accuracy guaranteed. Uncertain kinematics, unknown torques including unknown gravitational torque, unknown friction torque, and uncertain dynamics induced by uncertain moment of inertia and disturbance, are addressed. The approach is developed in the framework of observer-based control design. Two sliding mode observers are proposed to handle uncertain kinematics and to estimate unknown torques, respectively. Using the estimated information, a control law is then synthesized to guarantee that the desired trajectory can be followed after finite-time with zero tracking error. Experimental results are presented to show the performance of the proposed control approach.","Title: An fast reconstruction approach for actuator fault in robot manipulators\nYear: 2016\n\nAbstract: A theoretical framework of fast actuator fault reconstruction approach design is presented for robot manipulators in this paper. Taking advantages of terminal sliding mode observer (TSMO) technique, a TSMO-based reconstruction law is developed. Using this approach, it is shown that the actual actuator fault can be reconstructed within a finite time. Moreover, zero reconstruction error is guaranteed. Hence, a fast and precise reconstruction performance for actuator fault is provided. Another key advantage of this approach is that, it is independent on robot controller, mathematical model of actuator fault and even its time-profile. Simulation results are presented to verify the reconstruction performance of the proposed approach.","Title: A Fusion Method For Node Importance Measurement In Complex Networks\nYear: 2018\n\nAbstract: For the problem that closeness is difficult to effectively distinguish the importance of some nodes in complex networks, a new method of node importance measurement is proposed, which fuse the degree and closeness based on node re-ranking in segmentation. According to the network propagation dynamics model and Kendalls Tau coefficient, accuracy indicator and ranking stability indicator for evaluating measurement methods are given. Using the proposed method, simulations are carried out on Barabasi-Albert(BA) scale-free networks and ER random networks with different structures. The results show that compared with degree and closeness, fusion method not only has better measurement accuracy, but also has higher ranking stability.","Title: Research of the Enhanced Anti-Xprobe2\nYear: 2016\n\nAbstract: Anti-Xprobe2 defenses OS fingerprinting detection of Xprobe2 by camouflaging response packets. In this paper, aiming at that the separation of the detection data is achieved using a static data set in the original event separation module, which results in higher false alarm rate, the enhanced An-ti-Xprobe2 is proposed. Event separation module is improved, which is described by a finite state machine. The effectiveness of the proposed method is proved with experimental results.","Title: A Structure Simple Controller for Satellite Attitude Tracking Maneuver.\nYear: 2017\n\nAbstract: This paper addresses a difficult problem of designing a control approach with simple structure to perform attitude tracking maneuver for rigid satellites. The satellite is subject to external disturbance torque and uncertain inertia parameters. An observer-based estimation law is first proposed to reconstruct the uncertain dynamics. It is shown that such estimation can be achieved with zero estima...","Title: Exponential Tracking Control of Robotic Manipulators With Uncertain Dynamics and Kinematics.\nYear: 2019\n\nAbstract: This paper addresses a long-standing yet well documented open problem on task-space trajectory tracking control of robotic manipulators subject to both uncertain dynamics and uncertain kinematics. The main contribution is to establish a theoretical framework for designing an observer-based controller to achieve exponential tracking control. Two observers are designed for precisely estimating the u...","Title: Further results on output-feedback stabilization of stochastic upper-triangular systems with state and input time-delays\nYear: 2018\n\nAbstract: Based on stochastic time-delay system stability criterion and a homogeneous domination approach, the output-feedback stabilization problem for a class of more general stochastic upper-triangular systems with state and input time-delays has been solved in this paper. Firstly, the initial system is changed into an equivalent one with a designed scalar by introducing a set of coordinate transformations. After that, by designing an implementable homogeneous reduced-order observer, and tactfully selecting a suitable Lyapunov-Krasoviskii functional and a low gain scale, a delay-independent output-feedback controller is explicitly constructed. Finally, the globally asymptotically stability in probability of the closed-loop system is ensured by rigorous proof. The simulation results demonstrate the efficiency of the proposed design scheme.","Title: Further results on output-feedback stabilization of stochastic upper-triangular systems with state and input time-delays\nYear: 2018\n\nAbstract: Based on stochastic time-delay system stability criterion and a homogeneous domination approach, the output-feedback stabilization problem for a class of more general stochastic upper-triangular systems with state and input time-delays has been solved in this paper. Firstly, the initial system is changed into an equivalent one with a designed scalar by introducing a set of coordinate transformations. After that, by designing an implementable homogeneous reduced-order observer, and tactfully selecting a suitable Lyapunov-Krasoviskii functional and a low gain scale, a delay-independent output-feedback controller is explicitly constructed. Finally, the globally asymptotically stability in probability of the closed-loop system is ensured by rigorous proof. The simulation results demonstrate the efficiency of the proposed design scheme.","Title: Reconfigurable Tolerant Control of Uncertain Mechanical Systems With Actuator Faults: A Sliding Mode Observer-Based Approach.\nYear: 2018\n\nAbstract: This paper studies a key issue of developing reconfigurable fault-tolerant control to retain a nominal feedback controller and simultaneously handles actuator faults and system uncertainty, while the closed-loop system is stabilized with all control objectives achieved. A theoretical architecture of a reconfigurable control design is presented for a class of uncertain mechanical systems by using a...","Title: Discrete Fireworks Algorithm for Clustering in Wireless Sensor Networks.\nYear: 2018\n\nAbstract: Grouping the sensor nodes into clusters is an approach to save energy in wireless sensor networks (WSNs). We proposed a new solution to improve the performance of clustering based on a novel swarm intelligence algorithm. Firstly, the objective function for clustering optimization is defined. Secondly, discrete fireworks algorithm for clustering (DFWA-C) in WSNs is designed to calculate the optimal number of clusters and to find the cluster-heads. At last, simulation is conducted using the DFWA-C and relevant algorithms respectively. Results show that the proposed algorithm could obtain the number of clusters which is close to the theoretical optimal value, and can effectively reduce energy consumption to prolong the lifetime of WSNs.","Title: Cooperative geometric localization for a ground target based on the relative distances by multiple UAVs\nYear: 2019\n\nAbstract: Based on the locations of several unmanned aerial vehicles (UAVs) and their relative distances from a target, a ground target cooperative geometric localization method that is more effective than a traditional approach is proposed in this paper. First, an algorithm for determining the location of the target is described. The effectiveness and suitability of the proposed algorithm are then shown. Next, to investigate the location accuracy of the proposed method, the influence of three critical factors, namely, the flight altitude, UAV position errors, and measurement errors, is analyzed. Furthermore, for the required location accuracy, the feasible regions of these factors are determined based on their influence, and the location accuracy will satisfy the requirements if all factors are within the feasible regions. Finally, simulation results from the MATLAB/Simulink toolbox are presented to show the effectiveness of the proposed method and the availability of the feasible regions.","Title: Accurate State-of-Charge Estimation Approach for Lithium-Ion Batteries by Gated Recurrent Unit With Ensemble Optimizer\nYear: 2019\n\nAbstract: State-of-charge (SoC) estimation is indispensable for battery management systems (BMSs). Accurate SoC estimation can improve the efficiency of battery utilization, especially for electric vehicles (EVs). Several kinds of battery SoC estimation approaches have been developed, but a simple and efficient method for battery SoC estimation that can adapt to a variety of lithium-ion batteries is worth exploring. To this end, a recurrent neural network (RNN) model based on a gated recurrent unit (GRU) is presented for battery SoC estimation. The GRU-RNN model can rapidly learn its own parameters by means of an ensemble optimization method based on the Nadam and AdaMax optimizers. The Nadam optimizer is used in the model pre-training phase to find the minimum optimized value as soon as possible, and then the AdaMax optimizer is used in the model fine-tuning phase to further determine the model parameters. To validate the effectiveness and robustness of the proposed method, the GRU-RNN model was trained and tested with three kinds of dynamic loading profiles and compared with existing SoC estimation methods. The experimental results show that the proposed method dramatically reduces the model training time and increases estimation accuracy.","Title: Semi-supervised Classification with Graph Structure Similarity and Extended Label Propagation\nYear: 2019\n\nAbstract: Graph-based semi-supervised learning (GSSL) has attracted great attention over the past decade. However, there are still several open problems: 1) how to construct a graph that effectively represents the underlying structure of data and 2) how to incorporate label information of the labeled samples into a procedure of label propagation. Our solution mainly focuses on two aspects: 1) we propose a new graph construction technique by fusing local and global structural similarity (FLGSS). Based on an initial graph structure such as K-nearest neighbors (KNN), we utilize different types of link prediction algorithms to extract local and global graph structure information. These two types of structure information are fused into a graph structure that enhances the ability to represent the data correlation. 2) By incorporating the label correlation with feature similarity of samples, we propose an extended label propagation algorithm (ELP). Through experiments on three different types of datasets, it is shown that our method outperforms other widely used graph construction methods. The extended label inference algorithm achieves better classification results than some state-of-the-art methods. The proposed FLGSS method starts from KNN graph and two link prediction algorithms are performed to construct the graph. With the time complexity analysis, we theoretically deduce that the time complexity of FLGSS is not beyond that of KNN. Meanwhile, the time complexity of ELP remains the same as that of the traditional LP algorithm.","Title: Neural Network Based Finite-Time Attitude Tracking Control of Spacecraft With Angular Velocity Sensor Failures and Actuator Saturation\nYear: 2022\n\nAbstract: This article investigates the finite-time attitude tracking control problem of spacecraft with angular velocity sensor failures and actuator saturation. A model-free angular velocity observer is proposed first to reconstruct the unmeasurable angular velocity in finite time. Invoking the reconstructed information, a saturated finite-time attitude tracking controller is then designed via the fast te...","Title: Design of Small Multiband Full-Screen Smartwatch Antenna for IoT Applications\nYear: 2021\n\nAbstract: Smartwatch is a potential candidate for the Internet-of-Things (IoT) hub. However, the performance of smartwatch antennas is severely restricted by the smartwatch structure, especially when the antennas are designed by traditional methods. For adapting smartwatches to the role of IoT hub, a novel method of designing the multiband smartwatch antenna is presented in this article, aiming at increasin...","Title: Centered Error Entropy Kalman Filter With Application To Satellite Attitude Determination\nYear: 2021\n\nAbstract: Due to unavoidable factors, heavy-tailed noise appears in satellite attitude estimation. Traditional Kalman filter is prone to performance degradation and even filtering divergence when facing non-Gaussian noise. The existing robust algorithms have limited accuracy. To improve the attitude determination accuracy under non-Gaussian noise, we use the centered error entropy (CEE) criterion to derive a new filter named centered error entropy Kalman filter (CEEKF). CEEKF is formed by maximizing the CEE cost function. In the CEEKF algorithm, the prior state values are transmitted the same as the classical Kalman filter, and the posterior states are calculated by the fixed-point iteration method. The CEE EKF (CEE-EKF) algorithm is also derived to improve filtering accuracy in the case of the nonlinear system. We also give the convergence conditions of the iteration algorithm and the computational complexity analysis of CEEKF. The results of the two simulation examples validate the robustness of the algorithm we presented.","Title: Fixed-time Trajectory Tracking Control of a Wheeled Mobile Robot\nYear: 2021\n\nAbstract: This work investigates the fixed-time trajectory tracking control problem of a wheeled mobile robot (WMR) under external disturbances. Firstly, the dynamic error system of the WMR is transformed into a second-order attitude error subsystem and a third-order position error subsystem, respectively. Subsequently, the super-twisting-like algorithms are proposed for the trajectory tracking of the WMR, ...","Title: A neural network learning-based global optimization approach for aero-engine transient control schedule\nYear: 2022\n\nAbstract: Transient performance of aero-engine determines the maneuverability of aircraft. The optimal control schedule can tap the potential transient performance with subject to each constraint. Thus the transient time can be minimized with the optimal transient control schedule. This transient schedule is described by a strong constrained and nonlinear problem. It is therefore challenging to present an optimal method to achieve the best transient schedule. Motivated by solving this problem, a surrogate-assisted optimization framework is presented by using the learning capability of neural networks. It is achieved by presenting a sequential ensemble radial basis function (RBF) neural network-based optimization (SERO) algorithm. The advantage of the RBF neural network such as excellent prediction accuracy is taken and integrated into the surrogate-assisted optimization algorithm to improve the algorithm performance. With the application of the proposed scheme, the global optimization schedule is guaranteed for aero-engine transient control. Numerical validation is finally carried out by applying the presented optimization framework to a mixed-flow aero-engine transient control schedule. It is demonstrated that the SERO approach can minimize the transient process time despite any constraint at any time.","Title: Appointed Fixed Time Observer-Based Sliding Mode Control for a Quadrotor UAV Under External Disturbances\nYear: 2022\n\nAbstract: In this article, a study of the appointed fixed time control problem is presented for the quadrotor unmanned aerial vehicle attitude control system subject to external disturbances. Based on the appointed fixed time sliding mode variable and the adaptive technique, a novel sliding mode observer is first established to estimate the external disturbances within the appointed settling time. Subsequently, an appointed fixed time controller is proposed to track the desired attitude regardless of the initial states. To circumvent the singularity problem of the above controller, a novel improved adaptive sliding mode control law is designed by employing a nonlinear function and a modified sliding mode observer. Moreover, the chattering phenomenon is attenuated by replacing the sign function with hyperbolic tangent function. The appointed practical fixed time stability of the closed-loop attitude control system is proved and analyzed rigorously utilizing the Lyapunov theory. Finally, the effectiveness and fine performance of the proposed schemes are illustrated by numerical simulation results.","Title: Heterogeneous Graph Based Knowledge Tracing\nYear: 2022\n\nAbstract: Recent advances in on-line tutoring systems have brought on an increase in the research of Knowledge Tracing, which predicts the student's performance on coursework exercises over time. Previous researches, such as Bayesian Knowledge Tracing, Deep Knowledge Tracing (DKT) and qDKT, focused on either skill-level or question-level. As a result, those methods fail to take question-skill correlations into account. Inspired by Heterogeneous Graph Embedding (HGE), We propose a HGE-based knowledge tracing model. In this paper, a heterogeneous graph is built on skill information and question information, so as to capture the latent interactions between skill nodes and question nodes. In the proposed method, the knowledge tracing model can leverage more informations than previous methods. The experimental results show that the proposed method outperforms other state-of-the-art methods centered on either skills or questions.","Title: Velocity-Free State Feedback Fault-Tolerant Control for Satellite with Actuator and Sensor Faults\nYear: 2022\n\nAbstract: A velocity-free state feedback fault-tolerant control approach is proposed for the rigid satellite attitude stabilization problem subject to velocity-free measurements and actuator and sensor faults. First, multiplicative faults and additive faults are considered in the actuator and the sensor. The faults and system states are extended into a new augmented vector. Then, an improved sliding mode observer based on the augmented vector is presented to estimate unknown system states and actuator and sensor faults simultaneously. Next, a velocity-free state feedback attitude controller is designed based on the information from the observer. The controller compensates for the effects of actuator and sensor faults and asymptotically stabilizes the attitude. Finally, simulation results demonstrate the effectiveness of the proposed scheme.","Title: Knowledge-based formal modeling for CPPS in personalized intelligent manufacturing\nYear: 2021\n\nAbstract: Cyber Physical Production System (CPPS) is the product of the integration of CPS and production system. In view of the current CPPS model construction, especially in the application of multi-variety and small batch production in personalized production, CPPS is faced with the actual demand of reconstruction without downtime of the production line system. This article explores a reusable knowledge-...","Title: Predefined-time sliding manifold-based fixed-time attitude stabilization control of receiver aircraft with measurement noises\nYear: 2022\n\nAbstract: This paper investigates the challenging attitude stabilization control problem of receiver aircraft subject to uncertain and time-varying inertia, and wind disturbances in the aerial refueling task. An adaptive fixed-time control strategy is proposed. In this approach, a predefined-time sliding manifold is preliminarily developed. The main feature is that the settling time is not only independent of initial conditions but also can be simply and explicitly defined in advance through a single parameter in the sliding mode phase. On this basis, an adaptive control law is developed to guarantee that the trajectory of attitude can be stabilized into a small region containing the origin. The upper bound of unknown disturbance is not required to implement the proposed controller, which relaxes the restrictions of design for control systems. Moreover, the measuring noises are explicitly considered and addressed to make this approach still available in the noisy measurement environment. The Lyapunov stability analysis and numerical simulations verify the effectiveness of the proposed control strategy.","Title: Prescribed Time Attitude Tracking Control of Spacecraft With Arbitrary Disturbance\nYear: 2022\n\nAbstract: The problem of developing a controller for spacecraft to accomplish attitude maneuvers within a prescribed time, independently of the initial states, is addressed. A fixed-time disturbance observer-based control is established and proved to drive the error state to zero, in a prescribed time, in the presence of modeling and external uncertainties. The key advantages of the proposed control are that the gains are explicitly determined by the prescribed maneuvering time only and that it is also robust to instantaneous disturbances. The control approach enables an efficient design process for attitude control applications with time constraints. Simulation results are presented for the attitude control of a rigid spacecraft to verify the benefits of this control scheme."],"string":"[\n \"Title: Attitude Stabilization of Spacecrafts Under Actuator Saturation and Partial Loss of Control Effectiveness\\nYear: 2013\\n\\nAbstract: A practical solution is presented to the problem of fault tolerant attitude stabilization for a rigid spacecraft by using feedback from attitude orientation only. The attitude system, represented by modified Rodriguez parameters, is considered in the presence of external disturbances, uncertain inertia parameters, and actuator saturation. A low-cost control scheme is developed to compensate for the partial loss of actuator effectiveness fault. The derived controller not only has the capability to protect the control effort from actuator saturation but also guarantees all the signals in the closed-loop system to be uniformly ultimately bounded. Another feature of the approach is that the implementation of the controller does not require any rate sensor to measure angular velocity. An example is included to verify those highly desirable features in comparison with the conventional velocity-free control strategy.\",\n \"Title: Taxihailer: A Situation-Specific Taxi Pick-Up Points Recommendation System\\nYear: 2014\\n\\nAbstract: This demonstration presents TaxiHailer, a situation-specific recommendation system for passengers who are eager to find a taxi. Given a query with departure point, destination and time, it recommends pick-up points within a specified distance and ranked by potential waiting time. Unlike existing works, we consider three sets of features to build regression models, as well as Poisson process models for road segment clusters. We evaluate and choose the most proper models for each cluster under different situations. Also, TaxiHailer gives destination-aware recommendations for pick-up points with driving directions. We evaluate our recommendation results based on real GPS datasets.\",\n \"Title: Exponential convergence of an epidemic model with continuously distributed delays\\nYear: 2008\\n\\nAbstract: In this paper the convergence behavior of an epidemic model with continuously distributed delays is considered. Some new sufficient conditions are established to ensure that all the solutions of the epidemic model with permitted initial conditions converge exponentially to zero.\",\n \"Title: Competitive analysis of network load balancing\\nYear: 1997\\n\\nAbstract: This paper presents a theoretical analysis of the Load Balancing Problem (LBP) in a network of processing units. The performance objective is to minimize the makespan, i.e., the time spent to finish all jobs in a network of processing units. Because of the communication delay that results from the network topology, it is impossible to have a strategy which obtains the exact optimum under all load distributions. Instead, we measure the information efficiency of a load balancing policy by the worst case ratio of the solution (for each load distribution) of a load balancing policy to the optimal solution (for the same load distribution) assuming that processors have complete information about the load distribution over the network. This ratio is called the competitive ratio of this strategy [17, 24, 34]. In particular, a policy is called competitive if this ratio is bounded by a constant. As a first step, we discuss the centralized LBP, where all the processors have complete information of the load distribution over a network. Its solution serves as a benchmark to compare with realistic strategies, both in theoretical analysis, and experimental and simulational studies of distributed algorithms. We show that when jobs have different sizes, even with preemptive scheduling, LBP is NP–complete. When the jobs are of the same size, we give a polynomial algorithm, using network–flow techniques, which extends to approximate solutions for jobs of different sizes. We apply this benchmark solution in order to analyze the competitiveness for three network topologies: completely connected graphs, rings, and hierarchical complete k -ary trees. The constant competitive ratio results for complete network and hierarchical complete k -ary trees are applied to a study on the issues of network designs suitable for the LBP. We further discuss the problem for general networks with jobs of different sizes for slightly weaker results than those for the constant competitive ratio requirement. Finally, we comment on the related issues of job partitioning over parallel/distributed systems.\",\n \"Title: Photo-sketch synthesis and recognition based on subspace learning\\nYear: 2010\\n\\nAbstract: This paper aims to reducing difference between sketches and photos by synthesizing sketches from photos, and vice versa, and then performing sketch-sketch/photo-photo recognition with subspace learning based methods. Pseudo-sketch/pseudo-photo patches are synthesized with embedded hidden Markov model. Because these patches are assembled by averaging their overlapping area in most of the local strategy based methods, which leads to blurring effect to the resulted pseudo-sketch/pseudo-photo, we integrate the patches with image quilting. Experiments are carried out to demonstrate that the proposed method is effective to produce pseudo-sketch/pseudo-photo with high quality and achieve promising recognition results.\",\n \"Title: Existence and uniqueness of almost periodic solutions for a class of Hopfield neural networks with neutral delays\\nYear: 2009\\n\\nAbstract: In this work Hopfield neural networks with neutral delays are considered. Some sufficient conditions for the existence and uniqueness of almost periodic solutions are established by using the fixed point theorem and differential inequality techniques. The results of this work are new and complement previously known results.\",\n \"Title: A knot or not a knot? SETting the record 'straight' on proteins\\nYear: 2003\\n\\nAbstract: A novel knot found in the SET domain is examined in the light of five recent crystal structures and their descriptions in the literature. Using the algorithm of Taylor it was established that the backbone chain does not form a true knot. However, only two crosslinks corresponding to hydrogen-bonds were needed to form a knotted structure. Such loosely knotted structures formed by hydrogen-bonded crosslinks were assessed as lying between covalent crosslinks (such as disulphide bonds) and threaded-loops which are formed by close (unbonded) contacts between different parts of the chain. The term pseudo-knot was introduced (from the RNA field) to distinguish hydrogen-bonded 'knots'.\",\n \"Title: Attitude Tracking Control of Rigid Spacecraft With Actuator Misalignment and Fault\\nYear: 2013\\n\\nAbstract: A control scheme is developed to address the attitude tracking problem of a rigid spacecraft. A sufficient condition to accommodate actuator misalignment is presented. The controller asymptotically stabilizes the closed-loop attitude tracking system. The desired attitude trajectories can be tracked in finite time, even in the presence of uncertain inertia tensor, external disturbances, actuator fault, and actuator misalignment. The attitude tracking performance of the controller is evaluated through an illustrative example.\",\n \"Title: Adaptive frame rate up-conversion based on motion classification\\nYear: 2008\\n\\nAbstract: In this paper, a new technique on video frame rate up-conversion (FRUC) is presented by combining the adaptive motion classification (AMC) for image sequences and the mixed motion estimation (ME). In the proposed FRUC scheme, the AMC classifies sequences into global and local motion sequences (G/LMS) whose advantages are fully utilized. The mixture of bidirectional ME (BDME) and unidirectional ME (UDME) is proposed to avoid hole and ''background replace object'' phenomenon in interpolated frames. Moreover, the motion vectors (MVs) refinement technique is employed to smooth MVs field. For LMS, the motion detection is used to deal with static background and occlusions are located by using forward prediction of motion regions. Correctional sum of absolute difference (CSAD) which gives more precise physical MVs field than conventional SAD is applied during all of the ME process. Simulation results demonstrate the effectiveness of the proposed FRUC algorithm.\",\n \"Title: The general result of Gol'dberg's theorem concerning the growth of meromorphic solutions of algebraic differential equations\\nYear: 2009\\n\\nAbstract: In this paper, we give a general estimate result of Gol'dberg's theorem concerning finite growth order of meromorphic solutions of first-order algebraic differential equations by using the normal family theory. It is an extending result of corresponding theorem for Bergweiler and Barsegian and so on. We also give some examples to show that our result is sharp in special case.\",\n \"Title: Deterministic load balancing in computer networks\\nYear: 1990\\n\\nAbstract: The paper presents theoretical analysis of the deterministic complexity of the load balancing problem (LBP). Because of difficulty of the general problem, research in the area mostly restricts itself to probabilistic or approximation algorithms, or to the average behavior of a network. The paper provides deterministic analysis of the problem for general networks. It focuses on the worst-case complexity analysis of the problem. It shows certain cases of the LBP to be NP-complete. The paper also discusses situations closely related to computer networks, where there is a global view of load distribution in the network; it provides a polynomial algorithm for solving the load balancing problem in this network.\",\n \"Title: Visual Quality Assessment Of The Synthesized Sketch\\nYear: 2013\\n\\nAbstract: The existing image and video quality assessment algorithms are devoted to the degradation of quality caused by some artefacts or noise. We propose a novel image quality assessment for the synthesized sketch. When the sketch is synthesized from a photo, the quality of the resulted sketch is evaluated from two aspects. First, the synthesis of sketch may be affected by artefacts and therefore its visual quality should be measured with no-reference quality evaluation algorithm; secondly, the synthesized sketch should preserve discrimination information as the corresponding photo, and their similarity is also examined in the proposed method, with the photo as reference image. Because sketch synthesis aims at enhancing the performance of sketch-photo recognition and the purpose of sketch quality assessment is to determine whether the quality of the resulted sketch is sufficient for sketch-photo recognition, the proposed sketch quality assessment is associated with the recognition rate of sketch-photo recognition, that is to say, the rationality of the proposed method is judged according to the performance of sketch-photo recognition. A set of experiments are performed to prove that our method is more valid than the traditional methods.\",\n \"Title: A Face Recognition Scheme Based On Embedded Hidden Markov Model And Selective Ensemble Strategy\\nYear: 2009\\n\\nAbstract: As an effective method, the embedded hidden Markov model (E-HMM) has been widely used in pattern recognition. On applying the E-HMM to face recognition, the performance heavily depends on the selection of model parameters. Aiming at the problem of model selection, a selective ensemble of multi E-HMMs based face recognition algorithm is proposed. Experimental results illustrate that compared with the traditional E-HMM based face recognition algorithm the proposed method cannot only obtain better and more stable recognition performance, but also achieve higher generalization ability.\",\n \"Title: Adaptive fault tolerant control using integral sliding mode strategy with application to flexible spacecraft\\nYear: 2013\\n\\nAbstract: Adaptive-based integral sliding mode control scheme is developed to solve the actuator fault-tolerant compensation problem for linear time-invariant system in the presence of unknown actuator faults and external disturbances. A nonlinear integral-type sliding manifold is first presented that incorporates a virtual nominal control to achieve prescribed specifications of the perturbed system, and an adaptive sliding mode controller is constructed to automatically compensate for external disturbances and unknown time-invariant faults. It is shown that the proposed controller has the capability to guarantee that the resulting closed-loop system is asymptotically stable. Control design methodology is then extended to tackle with the unknown time-varying actuator faults. It is proved that any given level of gain attenuation from external disturbance/parametric estimation error to system output is achieved with the developed control law. The closed-loop performance of the new control solution derived here is evaluated extensively through numerical simulations in which the flexible spacecraft attitude control under both the external disturbances and actuator faults are considered.\",\n \"Title: Biview face recognition in the shape-texture domain\\nYear: 2013\\n\\nAbstract: Face recognition is one of the biometric identification methods with the highest potential. The existing face recognition algorithms relying on the texture information of face images are affected greatly by the variation of expression, scale and illumination. Whereas the algorithms based on the shape topology weaken the influence of illumination to some extent, but the impact of expression, scale and illumination on face recognition is still unsolved. To this end, we propose a new method for face recognition by integrating texture information with shape information, called biview face recognition algorithm. The texture models are constructed by using subspace learning methods and shape topologies are formed by building graphs for face images. The proposed biview face recognition method is compared with recognition algorithms merely based on texture or shape information. Experimental results of recognizing faces under the variation of illumination, expression and scale demonstrate that the performance of the proposed biview face recognition outperforms texture-based and shape-based algorithms.\",\n \"Title: A new approach for face recognition by sketches in photos\\nYear: 2009\\n\\nAbstract: Face recognition by sketches in photos remains a challenging task. Unlike the existing sketch-photo recognition methods, which convert a photo into sketch and then perform the sketch-photo recognition through sketch-sketch recognition, this paper devotes to synthesizing a photo from the sketch and transforming the sketch-photo recognition to photo-photo recognition to achieve better performance in mixture pattern recognition. The contribution of this paper mainly focuses on two aspects: (1) in view of that there are no many research findings of sketch-photo recognition based on the pseudo-photo synthesis and the existing methods require a large set of training samples, which is nearly impossible to achieve for the high cost of sketch acquisition, we make use of embedded hidden Markov model (EHMM), which can learn the nonlinearity of sketch-photo pair with less training samples, to produce pseudo-photos in terms of sketches; and (2) photos and sketches are divided into patches and pseudo-photo is generated by combining pseudo-photo patches, which makes pseudo-photo more recognizable. Experimental results demonstrate that the newly proposed method is effective to identify face sketches in photo set.\",\n \"Title: Predicting Users' Age Range In Micro-Blog Network\\nYear: 2013\\n\\nAbstract: In this report, we present our work on WISE 2013 Challenge Track II to predict the age range of Weibo users. In this challenge, a dataset consisting of Sina Weibo user information was presented. The goal of the challenge is to predict users age range. With personal information and original tweets for over one million users as training data, we analyze and process the dataset, and experiment a series of prediction methods including SVM, decision tree etc. The result shows that ensemble classifiers based on AdaBoost achieves the best prediction results in this challenge.\",\n \"Title: A survey of graph edit distance\\nYear: 2010\\n\\nAbstract: Inexact graph matching has been one of the significant research foci in the area of pattern analysis. As an important way to measure the similarity between pairwise graphs error-tolerantly, graph edit distance (GED) is the base of inexact graph matching. The research advance of GED is surveyed in order to provide a review of the existing literatures and offer some insights into the studies of GED. Since graphs may be attributed or non-attributed and the definition of costs for edit operations is various, the existing GED algorithms are categorized according to these two factors and described in detail. After these algorithms are analyzed and their limitations are identified, several promising directions for further research are proposed.\",\n \"Title: Adaptive Sliding Mode Fault Tolerant Attitude Tracking Control for Flexible Spacecraft Under Actuator Saturation.\\nYear: 2012\\n\\nAbstract: A novel fault tolerant attitude tracking control scheme is developed for flexible spacecraft with partial loss of actuator effectiveness fault. Neural networks are first introduced to account for system uncertainties, and an adaptive sliding mode controller is derived by using on-line updating law to estimate the bound of actuator fault such that any information of the fault is not required. To further address actuator saturation problem, a modified fault tolerant control law is then presented to ensure that the resulting control signal will never incur saturation. It is shown that the roll, pitch and yaw angle trajectories can globally asymptotically track the desired attitude in the face of faulty actuator, system uncertainties, external disturbances and even actuator saturation. A simulation example of a flexible spacecraft is given to illustrate the effectiveness of the proposed controller. © 2012 IEEE.\",\n \"Title: HMM-based graph edit distance for image indexing\\nYear: 2008\\n\\nAbstract: Most of the existing graph edit distance (GED) algorithms require cost functions which are difficult to be defined exactly. In this article, we propose a cost function free algorithm for computing GED. It only depends on the distribution of nodes rather than node or edge attributes in graphs. Hidden Markov model (HMM) is employed to model the distribution of feature points and thus dissimilarity measure of graphs can be posed as distance of HMMs. A fast algorithm of Kullback-Leibler Distance, suitable for computing the distance between two probability models, is adopted to compute the distance of HMMs. Experimental results demonstrate that the proposed GED algorithm can characterize the structure variety of graphs effectively and is available for clustering and indexing images of both rigid and nonrigid bodies. © 2008 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 18, 209–218, 2008\",\n \"Title: Image categorization: Graph edit distance + edge direction histogram\\nYear: 2008\\n\\nAbstract: This paper presents a novel algorithm for computing graph edit distance (GED) in image categorization. This algorithm is purely structural, i.e., it needs only connectivity structure of the graph and does not draw on node or edge attributes. There are two major contributions: (1) Introducing edge direction histogram (EDH) to characterize shape features of images. It is shown that GED can be employed as distance of EDHs. This algorithm is completely independent on cost function which is difficult to be defined exactly. (2) Computing distance of EDHs with earth mover distance (EMD) which takes neighborhood bins into account so as to compute distance of EDHs correctly. A set of experiments demonstrate that the newly presented algorithm is available for classifying and clustering images and is immune to the planar rotation of images. Compared with GED from spectral seriation, our algorithm can capture the structure change of graphs better and consume 12.79% time used by the former one. The average classification rate is 5% and average clustering rate is 25% higher than the spectral seriation method. © 2008 Elsevier Ltd. All rights reserved.\",\n \"Title: Finite-Time Attitude Tracking of Spacecraft With Fault-Tolerant Capability\\nYear: 2015\\n\\nAbstract: A finite-time attitude tracking control scheme is presented for rigid spacecraft subject to constant but unknown inertia and external disturbances. The controller, developed using sliding mode control technique, has great fault-tolerant capability to accommodate four types of actuator faults. Different from most of the existing works on attitude fault-tolerant control (FTC), the developed controller guarantees the desired attitude to be followed in finite time, which is critical for FTC systems. Moreover, the convergence time is an explicit parameter for designer’s choice. Thus, the controller design simply meets the finite-time requirement. The attitude-tracking performance is evaluated through a numerical example.\",\n \"Title: A Review on Recent Development of Spacecraft Attitude Fault Tolerant Control System.\\nYear: 2016\\n\\nAbstract: Motivated by several accidents, attitude control of a spacecraft subject to faults/failures has gained considerable attention in a wider range of aerospace engineering and academic communities. This paper is concerned with industrial practices and theoretical approaches for fault tolerant control (FTC) and fault detection and diagnosis (FDD) in spacecraft attitude control system. An overview on re...\",\n \"Title: Tracking Control of Robotic Manipulators With Uncertain Kinematics and Dynamics.\\nYear: 2016\\n\\nAbstract: This paper investigates a difficult problem of tracking control for robotic manipulations with guaranteed high accuracy. Uncertain kinematics, unknown torques including unknown gravitational torque, unknown friction torque, and uncertain dynamics induced by uncertain moment of inertia and disturbance, are addressed. The approach is developed in the framework of observer-based control design. Two s...\",\n \"Title: A Novel Disturbance Estimation Scheme for Formation Control of Ocean Surface Vessels.\\nYear: 2017\\n\\nAbstract: To achieve high-accuracy formation control of multiple ocean surface vessels, the external disturbance acting on each vessel should be accommodated. In this paper, a novel disturbance estimation scheme is presented to solve this problem, and to provide the future formation controller design with precise estimation information. This approach is developed in the theoretical framework of a terminal s...\",\n \"Title: NIOSIT: efficient data access for log-structured merge-tree style storage systems.\\nYear: 2017\\n\\nAbstract: Recent years, the log-structured merge-tree(LSM-tree) style storage has been widely adopted in distributed data storage systems(e.g. Bigtable and HBase) and commercial database systems(e.g. Ocean-Base, Cassandra, SQLite, etc.) to provide both large-volume storage capacity and high-performance data updates. Write operations become easier as the LSM-tree style storage avoids writing in place by updating a data copy in memory. However, read operations are affected as it requires an additional step during a data compaction to check if there exists the newest update of data record in memory, which brings many of costly empty reads in real usage since the volume of immutable data is pervasively far more massive than incremental delta data. To address this issue, we design a new network request processing mechanism to allow data access being processed in an auxiliary lightweight network communication IO thread. And a Bloom filter is incorporated with the network IO thread to effectively filter out the empty reads. We also analyze the efficiency advantage of the mechanism and introduce its detailed implementation based on the well-known OceanBase system. Experimental study using the YCSB benchmark demonstrates the proposed mechanism can significantly achieve 20 percent to 30 percent better performance than existing method.\",\n \"Title: A Fault-Tolerant Cooperative Positioning Approach for Multiple UAVs.\\nYear: 2017\\n\\nAbstract: Due to the increasing potentials and benefits of unmanned aerial vehicles (UAVs) in civil and surveillance applications, the cooperative fight using multiple UAVs has attracted more and more attention. However, fault-tolerant cooperative positioning for malfunctions in global positioning system (GPS) is one of the challenges that need to be addressed in various practical applications of UAVs. Motivated by solving this issue, this paper presents a reliable cooperative positioning approach for multiple UAVs cooperative fiight by using the geometric azimuth angle and the inclination angle relative to the reference UAVs. In this proposed cooperative positioning system (COPS), these relative angles are measured through radio direction finding equipped on UAVs. With the horizontal dilution of precision of the various reference UAVs carefully analyzed, the optimal reference UAVs set can be first selected. Based on a constant acceleration kinematic model, an extended Kalman filter is then employed to improve the positioning accuracy. A COPS/GPS faulttolerant navigation algorithm is finally developed to accommodate GPS fault. Simulation results are further presented to verify that the proposed algorithm can guarantee satisfactory navigation of the UAVs even when their GPS cannot work.\",\n \"Title: Efficient Multi-version Storage Engine for Main Memory Data Store.\\nYear: 2017\\n\\nAbstract: Multi-version storage engine is the fundamental component of modern main memory data store using the popular multiple version concurrency control (MVCC). The straightforward implementation of storage engine is to use a linked list to store multiple versions of an object. A read operation has to traverse the list for the specified version, which incurs pointer chasing. An optimization method implemented in HyPer is to store the current version in the object header, which is friendly to read the latest snapshot of data. However, a read operation still needs one extra pointer chasing in memory when accessing an object being updated. In this paper, we propose an efficient multi-version storage (EMS), a new storage engine for main memory data store. EMS embeds two latest versions in each object header, so that it can avoid the overhead of traversal of version list, especially in the update-intensive scenario. We present an implementation mechanism of widely used snapshot isolation level over EMS. The experimental results demonstrate that EMS outperforms the exiting multi-version storage engine of well-known main memory data stores in terms of throughput without excessive memory consumption.\",\n \"Title: A New Disturbance Attenuation Control Scheme for Quadrotor Unmanned Aerial Vehicles.\\nYear: 2017\\n\\nAbstract: This paper addresses a difficult problem of high-accuracy control for quadrotor unmanned aerial vehicles (UAVs) subject to external disturbance force and unknown disturbance torque. An observer-based full control scheme is presented. In the strategy, two observer-based estimators are first designed to estimate external disturbance force and torque, respectively. With the application of the precise...\",\n \"Title: A Data-Driven Clustering Approach for Fault Diagnosis.\\nYear: 2017\\n\\nAbstract: Clustering is an important approach in fault diagnosis. The dominant sets algorithm is a graph-based clustering algorithm, which defines the dominant set as a concept of a cluster. In this paper, we make an in-depth investigation of the dominant sets algorithm. As a result, we find that this algorithm is dependent on the similarity parameter in constructing the pairwise similarity matrix, and has the tendency to generate spherical clusters only. Based on the merits and drawbacks of this algorithm, we apply the histogram equalization transformation to the similarity matrices for the purpose of removing the influence of similarity parameters, and then use a density-based cluster expansion process to improve the clustering results. In experimental validation of the proposed algorithm, we use two criterions to evaluate the clustering results in order to arrive at convincing conclusions. Data clustering experiments on ten data sets and fault detection experiments on the Tennessee Eastman process demonstrate the effectiveness of the proposed algorithm.\",\n \"Title: An Intelligent Actuator Fault Reconstruction Scheme for Robotic Manipulators.\\nYear: 2018\\n\\nAbstract: This paper investigates a difficult problem of reconstructing actuator faults for robotic manipulators. An intelligent approach with fast reconstruction property is developed. This is achieved by using observer technique. This scheme is capable of precisely reconstructing the actual actuator fault. It is shown by Lyapunov stability analysis that the reconstruction error can converge to zero after ...\",\n \"Title: The development of an electronic control system for diesel-natural gas dual-fuel engine.\\nYear: 2018\\n\\nAbstract: To meet the actual application requirements for the diesel-natural gas dual-fuel engine refit, a new electronic control system for the dual-fuel engine is developed in the study. Specifically, an active mode switching board is developed to achieve flexible switching between the pure diesel mode and dual-fuel mode. A diesel nozzle physical simulator is developed to ensure that the original diesel electronic control unit does not trigger fault alarm when engine works in the dual-fuel mode. Moreover, a dual-fuel electronic control unit, which uses the high-speed and multicore TMS320F28M35 as its microcontroller unit, is additionally developed on the basis of the original diesel electronic control unit. The peak and hold current shape for the diesel nozzle is intelligently controlled by the software program in C2000 core. Experiments reveal that the developed electronic control system can select a proper working mode according to the engine operation condition and smoothly switch the working mode without any fault alarms.\",\n \"Title: Semi-Supervised Image Segmentation With Globalized Probability Of Boundary And Simple Linear Iterative Clustering\\nYear: 2017\\n\\nAbstract: Semi-supervised image segmentation in order to get a divided image from the tag portion. In this paper, a new semi-supervised image segmentation framework with global boundaries of probability (gPb) and simple linear iterative clustering (SLIC) is proposed. We use SLIC to perform image segmentation of super pixels, and we have constructed a graph, in which each vertex is a superpixel. Then we utilize gPb to define the edge weights for the graph. Semi-supervised learning based image (GSSL) method, and the use of labeled samples to calculate anchor anchor category, the category may be further determined by the anchor point of the sample. The results of the experiments certify the robustness and effectiveness of the method proposed in this paper.\",\n \"Title: A Graying Algorithm Based On Maximum Preserving Perception Features Of Images Whole Structure\\nYear: 2017\\n\\nAbstract: Although color images have gradually become the mainstream in image representation, gray images are still drawing more attention because of their small data amount and simple operation. Existing graying algorithms are difficult to ensure that the whole structure of images is not damaged and contrast difference information is consistent after dimensionality reduction. They cannot convey high quality gray images with human visual information. An optimization algorithm is proposed in this paper based on improving the objective energy function to retain these visual characteristics utmostly. The gray variance constraint is constructed to utmostly retain overall perceptual features of different gradients of the original color image regions, so as to ensure the stability of the global structure of the image. Let the difference between gray and color contrast be approximated to construct chromatic aberration of pixel pairs constraint term, and the contrast information is retained maximally. The final energy function is obtained by the linear combination of these two constraint terms. The iterative method is used to obtain the optimal channel weight coefficients which satisfy the minimum energy function and then get the result image. Finally, the experimental results show that the algorithm in this paper is basically consistent with the human visual perception of image contrast changes, and it can maintain the details and the global structure better.\",\n \"Title: Adaptive Quasi-Optimal Higher Order Sliding-Mode Control Without Gain Overestimation.\\nYear: 2018\\n\\nAbstract: This paper presents an adaptive quasi-optimal higher order sliding-mode control (HOSMC) scheme, which is able to avoid gain overestimation. The overall scheme consists of two elements: 1) a quasi-optimal control law that provides fast finite-time stabilization for a chain of integrators; and 2) an adaptive HOSMC with integration of the quasi-optimal control and the integral sliding-mode concept. T...\",\n \"Title: Adaptive attitude tracking control of a 3-degrees-of-freedom experimental helicopter with actuator dead-zone.\\nYear: 2019\\n\\nAbstract: In this article, the attitude tracking problem of a three rotational degrees-of-freedom experimental helicopter is studied based on adaptive fuzzy control method. The system unmodeled dynamics, external disturbances, and actuator dead-zone fault are taken into consideration simultaneously. The uncertainties and the faults are compensated by resorting to the fuzzy logic system. Based on adaptive backstepping technique and Lyapunov stability theory, state feedback controllers are designed such that the reference signals can be tracked by the elevation and pitch angles, respectively, within small errors. Finally, experimental results are given to demonstrate the reliability of the proposed method.\",\n \"Title: Observer-based control for robotic manipulations with uncertain kinematics and dynamics\\nYear: 2016\\n\\nAbstract: This paper investigates the problem of tracking control for robotic manipulations with high-accuracy guaranteed. Uncertain kinematics, unknown torques including unknown gravitational torque, unknown friction torque, and uncertain dynamics induced by uncertain moment of inertia and disturbance, are addressed. The approach is developed in the framework of observer-based control design. Two sliding mode observers are proposed to handle uncertain kinematics and to estimate unknown torques, respectively. Using the estimated information, a control law is then synthesized to guarantee that the desired trajectory can be followed after finite-time with zero tracking error. Experimental results are presented to show the performance of the proposed control approach.\",\n \"Title: An fast reconstruction approach for actuator fault in robot manipulators\\nYear: 2016\\n\\nAbstract: A theoretical framework of fast actuator fault reconstruction approach design is presented for robot manipulators in this paper. Taking advantages of terminal sliding mode observer (TSMO) technique, a TSMO-based reconstruction law is developed. Using this approach, it is shown that the actual actuator fault can be reconstructed within a finite time. Moreover, zero reconstruction error is guaranteed. Hence, a fast and precise reconstruction performance for actuator fault is provided. Another key advantage of this approach is that, it is independent on robot controller, mathematical model of actuator fault and even its time-profile. Simulation results are presented to verify the reconstruction performance of the proposed approach.\",\n \"Title: A Fusion Method For Node Importance Measurement In Complex Networks\\nYear: 2018\\n\\nAbstract: For the problem that closeness is difficult to effectively distinguish the importance of some nodes in complex networks, a new method of node importance measurement is proposed, which fuse the degree and closeness based on node re-ranking in segmentation. According to the network propagation dynamics model and Kendalls Tau coefficient, accuracy indicator and ranking stability indicator for evaluating measurement methods are given. Using the proposed method, simulations are carried out on Barabasi-Albert(BA) scale-free networks and ER random networks with different structures. The results show that compared with degree and closeness, fusion method not only has better measurement accuracy, but also has higher ranking stability.\",\n \"Title: Research of the Enhanced Anti-Xprobe2\\nYear: 2016\\n\\nAbstract: Anti-Xprobe2 defenses OS fingerprinting detection of Xprobe2 by camouflaging response packets. In this paper, aiming at that the separation of the detection data is achieved using a static data set in the original event separation module, which results in higher false alarm rate, the enhanced An-ti-Xprobe2 is proposed. Event separation module is improved, which is described by a finite state machine. The effectiveness of the proposed method is proved with experimental results.\",\n \"Title: A Structure Simple Controller for Satellite Attitude Tracking Maneuver.\\nYear: 2017\\n\\nAbstract: This paper addresses a difficult problem of designing a control approach with simple structure to perform attitude tracking maneuver for rigid satellites. The satellite is subject to external disturbance torque and uncertain inertia parameters. An observer-based estimation law is first proposed to reconstruct the uncertain dynamics. It is shown that such estimation can be achieved with zero estima...\",\n \"Title: Exponential Tracking Control of Robotic Manipulators With Uncertain Dynamics and Kinematics.\\nYear: 2019\\n\\nAbstract: This paper addresses a long-standing yet well documented open problem on task-space trajectory tracking control of robotic manipulators subject to both uncertain dynamics and uncertain kinematics. The main contribution is to establish a theoretical framework for designing an observer-based controller to achieve exponential tracking control. Two observers are designed for precisely estimating the u...\",\n \"Title: Further results on output-feedback stabilization of stochastic upper-triangular systems with state and input time-delays\\nYear: 2018\\n\\nAbstract: Based on stochastic time-delay system stability criterion and a homogeneous domination approach, the output-feedback stabilization problem for a class of more general stochastic upper-triangular systems with state and input time-delays has been solved in this paper. Firstly, the initial system is changed into an equivalent one with a designed scalar by introducing a set of coordinate transformations. After that, by designing an implementable homogeneous reduced-order observer, and tactfully selecting a suitable Lyapunov-Krasoviskii functional and a low gain scale, a delay-independent output-feedback controller is explicitly constructed. Finally, the globally asymptotically stability in probability of the closed-loop system is ensured by rigorous proof. The simulation results demonstrate the efficiency of the proposed design scheme.\",\n \"Title: Further results on output-feedback stabilization of stochastic upper-triangular systems with state and input time-delays\\nYear: 2018\\n\\nAbstract: Based on stochastic time-delay system stability criterion and a homogeneous domination approach, the output-feedback stabilization problem for a class of more general stochastic upper-triangular systems with state and input time-delays has been solved in this paper. Firstly, the initial system is changed into an equivalent one with a designed scalar by introducing a set of coordinate transformations. After that, by designing an implementable homogeneous reduced-order observer, and tactfully selecting a suitable Lyapunov-Krasoviskii functional and a low gain scale, a delay-independent output-feedback controller is explicitly constructed. Finally, the globally asymptotically stability in probability of the closed-loop system is ensured by rigorous proof. The simulation results demonstrate the efficiency of the proposed design scheme.\",\n \"Title: Reconfigurable Tolerant Control of Uncertain Mechanical Systems With Actuator Faults: A Sliding Mode Observer-Based Approach.\\nYear: 2018\\n\\nAbstract: This paper studies a key issue of developing reconfigurable fault-tolerant control to retain a nominal feedback controller and simultaneously handles actuator faults and system uncertainty, while the closed-loop system is stabilized with all control objectives achieved. A theoretical architecture of a reconfigurable control design is presented for a class of uncertain mechanical systems by using a...\",\n \"Title: Discrete Fireworks Algorithm for Clustering in Wireless Sensor Networks.\\nYear: 2018\\n\\nAbstract: Grouping the sensor nodes into clusters is an approach to save energy in wireless sensor networks (WSNs). We proposed a new solution to improve the performance of clustering based on a novel swarm intelligence algorithm. Firstly, the objective function for clustering optimization is defined. Secondly, discrete fireworks algorithm for clustering (DFWA-C) in WSNs is designed to calculate the optimal number of clusters and to find the cluster-heads. At last, simulation is conducted using the DFWA-C and relevant algorithms respectively. Results show that the proposed algorithm could obtain the number of clusters which is close to the theoretical optimal value, and can effectively reduce energy consumption to prolong the lifetime of WSNs.\",\n \"Title: Cooperative geometric localization for a ground target based on the relative distances by multiple UAVs\\nYear: 2019\\n\\nAbstract: Based on the locations of several unmanned aerial vehicles (UAVs) and their relative distances from a target, a ground target cooperative geometric localization method that is more effective than a traditional approach is proposed in this paper. First, an algorithm for determining the location of the target is described. The effectiveness and suitability of the proposed algorithm are then shown. Next, to investigate the location accuracy of the proposed method, the influence of three critical factors, namely, the flight altitude, UAV position errors, and measurement errors, is analyzed. Furthermore, for the required location accuracy, the feasible regions of these factors are determined based on their influence, and the location accuracy will satisfy the requirements if all factors are within the feasible regions. Finally, simulation results from the MATLAB/Simulink toolbox are presented to show the effectiveness of the proposed method and the availability of the feasible regions.\",\n \"Title: Accurate State-of-Charge Estimation Approach for Lithium-Ion Batteries by Gated Recurrent Unit With Ensemble Optimizer\\nYear: 2019\\n\\nAbstract: State-of-charge (SoC) estimation is indispensable for battery management systems (BMSs). Accurate SoC estimation can improve the efficiency of battery utilization, especially for electric vehicles (EVs). Several kinds of battery SoC estimation approaches have been developed, but a simple and efficient method for battery SoC estimation that can adapt to a variety of lithium-ion batteries is worth exploring. To this end, a recurrent neural network (RNN) model based on a gated recurrent unit (GRU) is presented for battery SoC estimation. The GRU-RNN model can rapidly learn its own parameters by means of an ensemble optimization method based on the Nadam and AdaMax optimizers. The Nadam optimizer is used in the model pre-training phase to find the minimum optimized value as soon as possible, and then the AdaMax optimizer is used in the model fine-tuning phase to further determine the model parameters. To validate the effectiveness and robustness of the proposed method, the GRU-RNN model was trained and tested with three kinds of dynamic loading profiles and compared with existing SoC estimation methods. The experimental results show that the proposed method dramatically reduces the model training time and increases estimation accuracy.\",\n \"Title: Semi-supervised Classification with Graph Structure Similarity and Extended Label Propagation\\nYear: 2019\\n\\nAbstract: Graph-based semi-supervised learning (GSSL) has attracted great attention over the past decade. However, there are still several open problems: 1) how to construct a graph that effectively represents the underlying structure of data and 2) how to incorporate label information of the labeled samples into a procedure of label propagation. Our solution mainly focuses on two aspects: 1) we propose a new graph construction technique by fusing local and global structural similarity (FLGSS). Based on an initial graph structure such as K-nearest neighbors (KNN), we utilize different types of link prediction algorithms to extract local and global graph structure information. These two types of structure information are fused into a graph structure that enhances the ability to represent the data correlation. 2) By incorporating the label correlation with feature similarity of samples, we propose an extended label propagation algorithm (ELP). Through experiments on three different types of datasets, it is shown that our method outperforms other widely used graph construction methods. The extended label inference algorithm achieves better classification results than some state-of-the-art methods. The proposed FLGSS method starts from KNN graph and two link prediction algorithms are performed to construct the graph. With the time complexity analysis, we theoretically deduce that the time complexity of FLGSS is not beyond that of KNN. Meanwhile, the time complexity of ELP remains the same as that of the traditional LP algorithm.\",\n \"Title: Neural Network Based Finite-Time Attitude Tracking Control of Spacecraft With Angular Velocity Sensor Failures and Actuator Saturation\\nYear: 2022\\n\\nAbstract: This article investigates the finite-time attitude tracking control problem of spacecraft with angular velocity sensor failures and actuator saturation. A model-free angular velocity observer is proposed first to reconstruct the unmeasurable angular velocity in finite time. Invoking the reconstructed information, a saturated finite-time attitude tracking controller is then designed via the fast te...\",\n \"Title: Design of Small Multiband Full-Screen Smartwatch Antenna for IoT Applications\\nYear: 2021\\n\\nAbstract: Smartwatch is a potential candidate for the Internet-of-Things (IoT) hub. However, the performance of smartwatch antennas is severely restricted by the smartwatch structure, especially when the antennas are designed by traditional methods. For adapting smartwatches to the role of IoT hub, a novel method of designing the multiband smartwatch antenna is presented in this article, aiming at increasin...\",\n \"Title: Centered Error Entropy Kalman Filter With Application To Satellite Attitude Determination\\nYear: 2021\\n\\nAbstract: Due to unavoidable factors, heavy-tailed noise appears in satellite attitude estimation. Traditional Kalman filter is prone to performance degradation and even filtering divergence when facing non-Gaussian noise. The existing robust algorithms have limited accuracy. To improve the attitude determination accuracy under non-Gaussian noise, we use the centered error entropy (CEE) criterion to derive a new filter named centered error entropy Kalman filter (CEEKF). CEEKF is formed by maximizing the CEE cost function. In the CEEKF algorithm, the prior state values are transmitted the same as the classical Kalman filter, and the posterior states are calculated by the fixed-point iteration method. The CEE EKF (CEE-EKF) algorithm is also derived to improve filtering accuracy in the case of the nonlinear system. We also give the convergence conditions of the iteration algorithm and the computational complexity analysis of CEEKF. The results of the two simulation examples validate the robustness of the algorithm we presented.\",\n \"Title: Fixed-time Trajectory Tracking Control of a Wheeled Mobile Robot\\nYear: 2021\\n\\nAbstract: This work investigates the fixed-time trajectory tracking control problem of a wheeled mobile robot (WMR) under external disturbances. Firstly, the dynamic error system of the WMR is transformed into a second-order attitude error subsystem and a third-order position error subsystem, respectively. Subsequently, the super-twisting-like algorithms are proposed for the trajectory tracking of the WMR, ...\",\n \"Title: A neural network learning-based global optimization approach for aero-engine transient control schedule\\nYear: 2022\\n\\nAbstract: Transient performance of aero-engine determines the maneuverability of aircraft. The optimal control schedule can tap the potential transient performance with subject to each constraint. Thus the transient time can be minimized with the optimal transient control schedule. This transient schedule is described by a strong constrained and nonlinear problem. It is therefore challenging to present an optimal method to achieve the best transient schedule. Motivated by solving this problem, a surrogate-assisted optimization framework is presented by using the learning capability of neural networks. It is achieved by presenting a sequential ensemble radial basis function (RBF) neural network-based optimization (SERO) algorithm. The advantage of the RBF neural network such as excellent prediction accuracy is taken and integrated into the surrogate-assisted optimization algorithm to improve the algorithm performance. With the application of the proposed scheme, the global optimization schedule is guaranteed for aero-engine transient control. Numerical validation is finally carried out by applying the presented optimization framework to a mixed-flow aero-engine transient control schedule. It is demonstrated that the SERO approach can minimize the transient process time despite any constraint at any time.\",\n \"Title: Appointed Fixed Time Observer-Based Sliding Mode Control for a Quadrotor UAV Under External Disturbances\\nYear: 2022\\n\\nAbstract: In this article, a study of the appointed fixed time control problem is presented for the quadrotor unmanned aerial vehicle attitude control system subject to external disturbances. Based on the appointed fixed time sliding mode variable and the adaptive technique, a novel sliding mode observer is first established to estimate the external disturbances within the appointed settling time. Subsequently, an appointed fixed time controller is proposed to track the desired attitude regardless of the initial states. To circumvent the singularity problem of the above controller, a novel improved adaptive sliding mode control law is designed by employing a nonlinear function and a modified sliding mode observer. Moreover, the chattering phenomenon is attenuated by replacing the sign function with hyperbolic tangent function. The appointed practical fixed time stability of the closed-loop attitude control system is proved and analyzed rigorously utilizing the Lyapunov theory. Finally, the effectiveness and fine performance of the proposed schemes are illustrated by numerical simulation results.\",\n \"Title: Heterogeneous Graph Based Knowledge Tracing\\nYear: 2022\\n\\nAbstract: Recent advances in on-line tutoring systems have brought on an increase in the research of Knowledge Tracing, which predicts the student's performance on coursework exercises over time. Previous researches, such as Bayesian Knowledge Tracing, Deep Knowledge Tracing (DKT) and qDKT, focused on either skill-level or question-level. As a result, those methods fail to take question-skill correlations into account. Inspired by Heterogeneous Graph Embedding (HGE), We propose a HGE-based knowledge tracing model. In this paper, a heterogeneous graph is built on skill information and question information, so as to capture the latent interactions between skill nodes and question nodes. In the proposed method, the knowledge tracing model can leverage more informations than previous methods. The experimental results show that the proposed method outperforms other state-of-the-art methods centered on either skills or questions.\",\n \"Title: Velocity-Free State Feedback Fault-Tolerant Control for Satellite with Actuator and Sensor Faults\\nYear: 2022\\n\\nAbstract: A velocity-free state feedback fault-tolerant control approach is proposed for the rigid satellite attitude stabilization problem subject to velocity-free measurements and actuator and sensor faults. First, multiplicative faults and additive faults are considered in the actuator and the sensor. The faults and system states are extended into a new augmented vector. Then, an improved sliding mode observer based on the augmented vector is presented to estimate unknown system states and actuator and sensor faults simultaneously. Next, a velocity-free state feedback attitude controller is designed based on the information from the observer. The controller compensates for the effects of actuator and sensor faults and asymptotically stabilizes the attitude. Finally, simulation results demonstrate the effectiveness of the proposed scheme.\",\n \"Title: Knowledge-based formal modeling for CPPS in personalized intelligent manufacturing\\nYear: 2021\\n\\nAbstract: Cyber Physical Production System (CPPS) is the product of the integration of CPS and production system. In view of the current CPPS model construction, especially in the application of multi-variety and small batch production in personalized production, CPPS is faced with the actual demand of reconstruction without downtime of the production line system. This article explores a reusable knowledge-...\",\n \"Title: Predefined-time sliding manifold-based fixed-time attitude stabilization control of receiver aircraft with measurement noises\\nYear: 2022\\n\\nAbstract: This paper investigates the challenging attitude stabilization control problem of receiver aircraft subject to uncertain and time-varying inertia, and wind disturbances in the aerial refueling task. An adaptive fixed-time control strategy is proposed. In this approach, a predefined-time sliding manifold is preliminarily developed. The main feature is that the settling time is not only independent of initial conditions but also can be simply and explicitly defined in advance through a single parameter in the sliding mode phase. On this basis, an adaptive control law is developed to guarantee that the trajectory of attitude can be stabilized into a small region containing the origin. The upper bound of unknown disturbance is not required to implement the proposed controller, which relaxes the restrictions of design for control systems. Moreover, the measuring noises are explicitly considered and addressed to make this approach still available in the noisy measurement environment. The Lyapunov stability analysis and numerical simulations verify the effectiveness of the proposed control strategy.\",\n \"Title: Prescribed Time Attitude Tracking Control of Spacecraft With Arbitrary Disturbance\\nYear: 2022\\n\\nAbstract: The problem of developing a controller for spacecraft to accomplish attitude maneuvers within a prescribed time, independently of the initial states, is addressed. A fixed-time disturbance observer-based control is established and proved to drive the error state to zero, in a prescribed time, in the presence of modeling and external uncertainties. The key advantages of the proposed control are that the gains are explicitly determined by the prescribed maneuvering time only and that it is also robust to instantaneous disturbances. The control approach enables an efficient design process for attitude control applications with time constraints. Simulation results are presented for the attitude control of a rigid spacecraft to verify the benefits of this control scheme.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":60,"string":"60"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13579,"string":"13,579"}}},{"rowIdx":13580,"cells":{"question_id":{"kind":"string","value":"S bastien Destercke"},"question":{"kind":"string","value":"Generate an abstract for the title \"A generic framework to include belief functions in preference handling and multi-criteria decision.\" using the following items: Modelling, Preferences, Decision Maker, Belief Functions, Inconsistencies."},"answer":{"kind":"string","value":"Modelling the preferences of a decision maker about alternatives having multiple criteria usually starts by collecting preference information (comparisons of alternatives, importance of criteria, …), which are then used to fit a preference model issued from some set of hypothesis (weighted average, CP-net, lexicographic orderings, AHP, …). In practice, this process often leads to inconsistencies that may be due to inaccurate information provided by the decision maker, who can be unsure about the provided information, or to a poor choice of hypothesis set, which can be too restrictive or not well adapted to the decision process. In this paper, we propose to use belief functions as a way to quantify and resolve such inconsistencies, notably by allowing the decision maker to express her/his certainty about the provided preferential information. Our framework is generic, in the sense that it does not assume a given set of hypothesis a priori, and is consistent with precise methods, in the sense that in the absence of uncertainty and inconsistencies in the information, precise models are ultimately retrieved."},"correct_answer":{"kind":"string","value":"Modelling the preferences of a decision maker about alternatives having multiple criteria usually starts by collecting preference information (comparisons of alternatives, importance of criteria, …), which are then used to fit a preference model issued from some set of hypothesis (weighted average, CP-net, lexicographic orderings, AHP, …). In practice, this process often leads to inconsistencies that may be due to inaccurate information provided by the decision maker, who can be unsure about the provided information, or to a poor choice of hypothesis set, which can be too restrictive or not well adapted to the decision process. In this paper, we propose to use belief functions as a way to quantify and resolve such inconsistencies, notably by allowing the decision maker to express her/his certainty about the provided preferential information. Our framework is generic, in the sense that it does not assume a given set of hypothesis a priori, and is consistent with precise methods, in the sense that in the absence of uncertainty and inconsistencies in the information, precise models are ultimately retrieved."},"documents":{"kind":"list like","value":["Title: F-boxes for filtering.\nYear: 2011\n\nAbstract: Selecting a particular summative kernel (i.e., formally equivalent to a probability distribution) when filtering a digital signal can be a difficult task. An obvious solution to this problem is to filter with multiple kernels rather than with a single one, but the computing cost of such a solution can quickly become prohibitive (especially in real-time applications). Another alternative, the one studied in this paper, is to consider kernels modeled by imprecise probabilistic representations. Considering such representations makes the use of numerical tools coming from imprecise probability theory possible, such as the Choquet integral, and allows one to work with multiple kernels without multiplying too much the number of required computations. In this paper, we propose to use the well-known p-box representation to filter a digital signal. We show that the use p-boxes allows making more precise inferences than those obtained with possibility distributions and clouds. We then discuss the practical aspect of computing a filtered signal with p-boxes, and finish by some experiments.","Title: Classification Trees Based on Belief Functions.\nYear: 2012\n\nAbstract: Decision tree classifiers are popular classification methods. In this paper, we extend to multi-class problems a decision tree method based on belief functions previously described for two-class problems only. We propose three possible extensions: combining multiple two-class trees together and directly extending the estimation of belief functions within the tree to the multi-class setting. We provide experiment results and compare them to usual decision trees.","Title: A New Contextual Discounting Rule For Lower Probabilities\nYear: 2010\n\nAbstract: Sources providing information about the value of a variable may not be totally reliable. In such a case, it is common in uncertainty theories to take account of this unreliability by a so-called discounting rule. A few discounting rules have been proposed in the framework of imprecise probability theory, but one of the drawback of those rules is that they do not preserve interesting properties (i.e. n-monotonicity) of lower probabilities. Another aspect that only a few of them consider is that source reliability is often dependent of the context, i.e. a source may be more reliable to identify some values than others. In such cases, it is useful to consider contextual discounting, where reliability information is dependent of the variable values. In this paper, we propose such a contextual discounting rule that also preserves some of the interesting mathematical properties a lower probability can have.","Title: Data reliability assessment in a data warehouse opened on the web\nYear: 2011\n\nAbstract: This paper presents an ontology-driven workflow that feeds and queries a data warehouse opened on the Web. Data are extracted from data tables in Web documents. As web documents are very heterogeneous in nature, a key issue in this workflow is the ability to assess the reliability of retrieved data. We first recall the main steps of our method to annotate and query Web data tables driven by a domain ontology. Then we propose an original method to assess Web data table reliability from a set of criteria by the means of evidence theory. Finally, we show how we extend the workflow to integrate the reliability assessment step.","Title: Independence and 2-monotonicity: nice to have, hard to keep\nYear: 2011\n\nAbstract: When using lower probabilities to model uncertainty about the value assumed by a variable, 2-monotonicity is an interesting property to satisfy, as it greatly facilitates further treatments (such as the computation of lower/upper expectation bounds). In this paper, we show that multivariate joint models induced from marginal ones by strong independence, epistemic independence or epistemic irrelevance do not usually preserve such a property, even if it is satisfied by all marginal models. We then propose a joint model outer-approximating those induced by strong and epistemic independence and study some of its properties.","Title: A Unified View of Some Representations of Imprecise Probabilities\nYear: 2006\n\nAbstract: Several methods for the practical representation of imprecise probabilities exist such as Ferson's p-boxes, possibility distributions, Neumaier's clouds, and random sets. in this paper some relationships existing between the four kinds of representations are discussed. A cloud as well as a p-box can be modelled as a pair of possibility distributions. We show that a generalized form of p-box is a special kind of belief function and also a special kind of cloud.","Title: Selecting source behavior in information fusion on the basis of consistency and specificity\nYear: 2013\n\nAbstract: Combining pieces of information provided by several sources without prior knowledge about the behavior of the sources is an old yet still important and rather open problem in belief function theory. In this paper, we propose a general approach to select the behavior of sources, based on two cornerstones of information fusion that are the notions of specificity and consistency. This approach is framed in a recently introduced and general fusion scheme that allows a wide range of assumptions on the sources. In the process, we are also led to generalize a recently introduced measure of conflict to all Boolean connectives. Eventually, we show that our approach generalizes some important existing information fusion strategies.","Title: Handling Bipolar Knowledge with Credal Sets\nYear: 2010\n\nAbstract: How to represent and handle bipolar information has recently received a lot of attention. Being bipolar means that the information has a positive and negative part. In this paper, we consider asymmetric bipolar information (i.e. situations where positive and negative information are unrelated and should be processed separately). We propose a framework to represent and handle it with so-called credal sets, i.e., with convex sets of probability distributions. We also provide some illustrative examples.","Title: Combining binary classifiers with imprecise probabilities\nYear: 2011\n\nAbstract: This paper proposes a simple framework to combine binary classifiers whose outputs are imprecise probabilities (or are transformed into some imprecise probabilities, e.g., by using confidence intervals). This combination comes down to solve linear programs describing constraints over events (here, subsets of classes). The number of constraints grows linearly with the number of classifiers, making the proposed framework tractable for problems involving a relatively large number of classes. After detailing the method, we provide some first experimental results illustrating the method interests.","Title: Relating Epistemic Irrelevance to Event Trees\nYear: 2008\n\nAbstract: We relate the epistemic irrelevance in Walley's behavioural theory of imprecise probabilities to the event-tree independence due to Shafer. In particular, we show that forward irrelevance is equivalent to event-tree independence in particular event trees, suitably generalised to allow for the fact that imprecise rather than precise probability models are attached to the nodes in the tree. This allows us to argue that in a theory of uncertain processes, the asymmetrical notion of epistemic irrelevance has a more important role to play than its more involved and symmetrical counterpart called epistemic independence.","Title: Making ontology-based knowledge and decision trees interact: an approach to enrich knowledge and increase expert confidence in data-driven models\nYear: 2010\n\nAbstract: When using data-driven models to make simulations and predictions in experimental sciences, it is essential for the domain expert to be confident about the predicted values. Increasing this confidence can be done by using interpretable models, so that the expert can follow the model reasoning pattern, and by integrating expert knowledge to the model itself. New pieces of useful formalised knowledge can then be integrated to an existing corpus while data-driven models are tuned according to the expert advice. In this paper, we propose a generic interactive procedure, relying on an ontology to model qualitative knowledge and on decision trees as a data-driven rule learning method. A case study based on data issued from multiple scientific papers in the field of cereal transformation illustrates the approach.","Title: A Consonant Approximation Of The Product Of Independent Consonant Random Sets\nYear: 2009\n\nAbstract: The belief structure resulting from the combination of consonant and independent marginal random sets is not, in general, consonant. Also, the complexity of such a structure grows exponentially with the number of combined random sets, making it quickly intractable for computations. In this paper, we propose a simple guaranteed consonant outer approximation of this structure. The complexity of this outer approximation does not increase with the number of marginal random sets (i.e., of dimensions), making it easier to handle in uncertainty propagation. Features and advantages of this outer approximation are then discussed, with the help of some illustrative examples.","Title: Generalised p-Boxes on Totally Ordered Spaces\nYear: 2008\n\nAbstract: Probability boxes are among the most simple and popular models used in imprecise probability theory, and many practical results concerning them exist in the literature. Nevertheless, little attention has been paid to their formal characterisation in the setting of Walley's behavioural theory of imprecise probabilities. This paper tries to remedy this situation by formalising, generalising and extending existing results as well as by giving new ones, within Walley's framework.","Title: Learning Decision Trees from Uncertain Data with an Evidential EM Approach\nYear: 2013\n\nAbstract: In real world applications, data are often uncertain or imperfect. In most classification approaches, they are transformed into precise data. However, this uncertainty is an information in itself which should be part of the learning process. Data uncertainty can take several forms: probabilities, (fuzzy)sets of possible values, expert assessments, etc. We therefore need a flexible and generic enough model to represent and treat this uncertainty, such as belief functions. Decision trees are well known classifiers which are usually learned from precise datasets. In this paper we propose a methodology to learn decision trees from uncertain data in the belief function framework. In the proposed method, the tree parameters are estimated through the maximization of an evidential likelihood function computed from belief functions, using the recently proposed E2M algorithm that extends the classical EM. Some promising experiments compare the obtained trees with classical CART decision trees.","Title: Possibilistic information fusion using maximal coherent subsets\nYear: 2009\n\nAbstract: When multiple sources provide information about the same unknown quantity, their fusion into a synthetic interpretable message is often a tricky problem, especially when sources are conflicting. In this paper, we propose to use possibility theory and the notion of maximal coherent subsets (MCSs), often used in logic-based representations, to build a fuzzy belief structure that will be instrumental both for extracting useful insight about various features of the information conveyed by the sources and for compressing this information into a unique possibility distribution. Extensions and properties of the basic fusion rule are also studied.","Title: Ontology-driven possibilistic reference fusion\nYear: 2010\n\nAbstract: It often happens that different references (i.e. data descriptions), possibly coming from different heterogeneous data sources, concern the same real world entity. In such cases, it is necessary: (i) to detect, through reconciliation methods, whether different data descriptions refer to the same real world entity and (ii) to fuse them into a unique representation. Here we assume the reference reconciliation is solved, and we propose a fusion method based on possibility theory, able to cope with uncertainty and with ontological knowledge. An implementation using W3C standards is provided. Rising from the fusion process, an ontology enrichment procedure is proposed to complete the global ontology.","Title: An extension of Universal Generating Function in Multi-State Systems Considering Epistemic Uncertainties.\nYear: 2013\n\nAbstract: Many practical methods and different approaches have been proposed to assess Multi-State Systems (MSS) reliability measures. The universal generating function (UGF) method, introduced in 1986, is known to be a very efficient way of evaluating the availability of different types of MSSs. In this paper, we propose an extension of the UGF method considering epistemic uncertainties. This extended meth...","Title: Can the Minimum Rule of Possibility Theory Be Extended to Belief Functions?\nYear: 2009\n\nAbstract: When merging belief functions, Dempster rule of combination is justified only when sources can be considered as independent. When dependencies are ill-known, it is usual to ask the merging operation to satisfy the property of idempotence, as this property ensures a cautious behaviour in the face of dependent sources. There are different strategies to find such rules for belief functions. One strategy is to rely on idempotent rules used in either more general or more specific frameworks and to respectively study their particularisation or extension to belief functions. In this paper, we try to extend the minimum rule of possibility theory to belief functions. We show that such an extension is not always possible, unless we accept the idea that the result of the fusion process can be a family of belief functions.","Title: Combining Imprecise Probability Masses with Maximal Coherent Subsets: Application to Ensemble Classification.\nYear: 2012\n\nAbstract: When working with sets of probabilities, basic information fusion operators quickly reach their limits: intersection becomes empty, while union results in a poorly informative model. An attractive means to overcome these limitations is to use maximal coherent subsets (MCS). However, identifying the maximal coherent subsets is generally NP-hard. Previous proposals advocating the use of MCS to merge probability sets have not provided efficient ways to perform this task. In this paper, we propose an efficient approach to do such a merging between imprecise probability masses, a popular model of probability sets, and test it on an ensemble classification problem.","Title: Using The Ols Algorithm To Build Interpretable Rule Bases: An Application To A Depollution Problem\nYear: 2007\n\nAbstract: One of the main advantages of fuzzy modeling is the ability to yield interpretable results. Amongst these modeling methods, the OLS algorithm is a mathematically robust technique that allows to induce a fuzzy rule base from a set of training data. It does so by using linear regression to select the most important rules. However, the original OLS algorithm only relies upon numerical accuracy, and doesn't take interpretability matters into account. Thus, we propose some modifications to the original method so that it builds interpretable rule bases.","Title: Extreme points of the credal sets generated by elementary comparative probabilities\nYear: 2013\n\nAbstract: When using convex probability sets (or, equivalently, lower previsions) as models of uncertainty, identifying extreme points can be useful to perform various computations or to use some algorithms. In general, sets induced by specific models such as possibility distributions, linear vacuous mixtures or 2-monotone measures may have extreme points easier to compute than generic convex sets. In this paper, we study extreme points of another specific model: comparative probability orderings between the elements of a finite space. We use these extreme points to study the properties of the lower probability induced by this set, and connect comparative probabilities with other uncertainty models.","Title: A Decision Rule for Imprecise Probabilities Based on Pair-Wise Comparison of Expectation Bounds\nYear: 2010\n\nAbstract: There are many ways to extend the classical expected utility decision rule to the case where uncertainty is described by (convex) probability sets. In this paper, we propose a simple new decision rule based on the pair-wise comparison of lower and upper expected bounds. We compare this rule to other rules proposed in the literature, showing that this new rule is both precise, computationally tractable and can help to boost the computation of other, more computationally demanding rules.","Title: How To Randomly Generate Mass Functions\nYear: 2013\n\nAbstract: As Dempster-Shafer theory spreads in different application fields, and as mass functions are involved in more and more complex systems, the need for algorithms randomly generating mass functions arises. Such algorithms can be used, for instance, to evaluate some statistical properties or to simulate the uncertainty in some systems (e.g., data base content, training sets). As such random generation is often perceived as secondary, most of the proposed algorithms use straightforward procedures whose sample statistical properties can be difficult to characterize. Thus, although such algorithms produce randomly generated mass functions, they do not always produce what could be expected from them (for example, uniform sampling in the set of all possible mass functions). In this paper, we briefly review some well-known algorithms, explaining why their statistical properties are hard to characterize. We then provide relatively simple algorithms and procedures to perform efficient random generation of mass functions whose sampling properties are controlled.","Title: Random Generation of Mass Functions: A Short Howto.\nYear: 2012\n\nAbstract: As Dempster-Shafer theory spreads in different applications fields involving complex systems, the need for algorithms randomly generating mass functions arises. As such random generation is often perceived as secondary, most pro-posed algorithms use procedures whose sample statistical properties are difficult to characterize. Thus, although they produce randomly generated mass functions, it is difficult to control the sample statistical laws. In this paper, we briefly review classical algorithms, explaining why their statistical properties are hard to characterize, and then provide simple procedures to perform efficient and controlled random generation.","Title: Correcting binary imprecise classifiers: local vs global approach\nYear: 2012\n\nAbstract: This paper proposes a simple strategy for combining binary classifiers with imprecise probabilities as outputs. Our combination strategy consists in computing a set of probability distributions by solving an optimization problem whose constraints depend on the classifiers outputs. However, the classifiers may provide assessments that are jointly incoherent, in which case the set of probability distributions satisfying all the constraints is empty. We study different correction strategies for restoring this consistency, by relaxing the constraints of the optimization problem so that it becomes feasible. In particular, we propose and compare a global strategy, where all constraints are relaxed to the same level, to a local strategy, where some constraints may be relaxed more than others. The local discounting strategy proves to give very good results compared both to single classifier approaches and to classifier combination schemes using a global correction scheme.","Title: Probability boxes on totally preordered spaces for multivariate modelling\nYear: 2011\n\nAbstract: A pair of lower and upper cumulative distribution functions, also called probability box or p-box, is among the most popular models used in imprecise probability theory. They arise naturally in expert elicitation, for instance in cases where bounds are specified on the quantiles of a random variable, or when quantiles are specified only at a finite number of points. Many practical and formal results concerning p-boxes already exist in the literature. In this paper, we provide new efficient tools to construct multivariate p-boxes and develop algorithms to draw inferences from them. For this purpose, we formalise and extend the theory of p-boxes using Walley's behavioural theory of imprecise probabilities, and heavily rely on its notion of natural extension and existing results about independence modeling. In particular, we allow p-boxes to be defined on arbitrary totally preordered spaces, hence thereby also admitting multivariate p-boxes via probability bounds over any collection of nested sets. We focus on the cases of independence (using the factorization property), and of unknown dependence (using the Frechet bounds), and we show that our approach extends the probabilistic arithmetic of Williamson and Downs. Two design problems-a damped oscillator, and a river dike-demonstrate the practical feasibility of our results.","Title: Evaluation of the naive evidential classifier (NEC): a comparison between its two variants based on a real agronomic application\nYear: 2012\n\nAbstract: We introduce the notion of naive evidential classifier. This classifier, which has a structure mirroring the naive Bayes classifier, is based on the Transferable Belief Model and uses mass assignments as its uncertainty model. This new method achieves more robust inferences, mainly by explicitly modeling imprecision when data are in little amount or are imprecise. After introducing the model and its inference process based on Smet's generalized Bayes theorem (GBT), we specify some possible methods to learn its parameters, based on the Imprecise Dirichlet Model (IDM) or on predictive belief functions. Some experimental results on an agronomic application are then given and evaluated.","Title: Numerical Sensitivity and Efficiency in the Treatment of Epistemic and Aleatory Uncertainty\nYear: 2007\n\nAbstract: The treatment of both aleatory and epistemic uncertainty by recent methods\noften requires an high computational effort. In this abstract, we propose a\nnumerical sampling method allowing to lighten the computational burden of\ntreating the information by means of so-called fuzzy random variables.","Title: A flexible bipolar querying approach with imprecise data and guaranteed results\nYear: 2011\n\nAbstract: In this paper, we propose an approach to query a database when the user preferences are bipolar (i.e., express both constraints and wishes about the desired result) and the data stored in the database are imprecise. Results are then completely ordered with respect to these bipolar preferences, giving priority to constraints over wishes. Additionally, we propose a treatment that allows us to guarantee that any query will return a result, even if no element satisfies all constraints specified by the user. Such a treatment may be useful when user's constraints are unrealistic (i.e., cannot be all satisfied simultaneously) and when the user desires a guaranteed result. The approach is illustrated on a real-world problem concerning the selection of optimal packaging for fresh fruits and vegetables.","Title: Decision Support System Using Flexible Query And Reliability Assessment - Application To Biodegradable And Biosourced Packaging Design\nYear: 2013\n\nAbstract: In decision tasks, imprecision and uncertainty can arise from many sources such as data uncertainty, data reliability, or the necessity to use intermediate (non-fully reliable) models. User preferences may also be included into the decision process, making it even more complex. This is particularly true for processes involving biological materials. Soft computing methods have the potential to be the kingpin of specialized software that can be integrated in decision support systems intended to solve the mentioned issues.This work presents three such methods and their implementation: (i)-simulation module with uncertain inputs and interval analysis, (ii)-reliability module allowing to rank candidates according to expert criteria, and (iii)-bipolar flexible querying module, that makes the difference between constraints and wishes in a data base query. A decision support system architecture is proposed and illustrated on a case study, dealing with biodegradable and biosourced packaging design.","Title: Idempotent conjunctive combination of belief functions: Extending the minimum rule of possibility theory.\nYear: 2011\n\nAbstract: When conjunctively merging two belief functions concerning a single variable but coming from different sources, Dempster rule of combination is justified only when information sources can be considered as independent. When dependencies between sources are ill-known, it is usual to require the property of idempotence for the merging of belief functions, as this property captures the possible redundancy of dependent sources. To study idempotent merging, different strategies can be followed. One strategy is to rely on idempotent rules used in either more general or more specific frameworks and to study, respectively, their particularization or extension to belief functions. In this paper, we study the feasibility of extending the idempotent fusion rule of possibility theory (the minimum) to belief functions. We first investigate how comparisons of information content, in the form of inclusion and least-commitment, can be exploited to relate idempotent merging in possibility theory to evidence theory. We reach the conclusion that unless we accept the idea that the result of the fusion process can be a family of belief functions, such an extension is not always possible. As handling such families seems impractical, we then turn our attention to a more quantitative criterion and consider those combinations that maximize the expected cardinality of the joint belief functions, among the least committed ones, taking advantage of the fact that the expected cardinality of a belief function only depends on its contour function.","Title: Building an interpretable fuzzy rule base from data using Orthogonal Least Squares---Application to a depollution problem\nYear: 2007\n\nAbstract: In many fields where human understanding plays a crucial role, such as bioprocesses, the capacity of extracting knowledge from data is of critical importance. Within this framework, fuzzy learning methods, if properly used, can greatly help human experts. Amongst these methods, the aim of orthogonal transformations, which have been proven to be mathematically robust, is to build rules from a set of training data and to select the most important ones by linear regression or rank revealing techniques. The OLS algorithm is a good representative of those methods. However, it was originally designed so that it only cared about numerical performance. Thus, we propose some modifications of the original method to take interpretability into account. After recalling the original algorithm, this paper presents the changes made to the original method, then discusses some results obtained from benchmark problems. Finally, the algorithm is applied to a real-world fault detection depollution problem.","Title: Revisiting the Notion of Conflicting Belief Functions.\nYear: 2012\n\nAbstract: The problem of conflict measurement between information sources knows a regain of interest. In most works related to this issue, Dempter's rule plays a central role. In this paper, we propose to revisit conflict from a different perspective. We do not make a priori assumption about dependencies and start from the definition of conflicting sets, studying its possible extensions to the framework of belief functions.","Title: Computing expectations with continuous p-boxes: Univariate case\nYear: 2009\n\nAbstract: Given an imprecise probabilistic model over a continuous space, computing lower/upper expectations is often computationally hard to achieve, even in simple cases. Because expectations are essential in decision making and risk analysis, tractable methods to compute them are crucial in many applications involving imprecise probabilistic models. We concentrate on p-boxes (a simple and popular model), and on the computation of lower expectations of non-monotone functions. This paper is devoted to the univariate case, that is where only one variable has uncertainty. We propose and compare two approaches: the first using general linear programming, and the second using the fact that p-boxes are special cases of random sets. We underline the complementarity of both approaches, as well as the differences.","Title: Transforming Probability Intervals into Other Uncertainty Models\nYear: 2007\n\nAbstract: Probability intervals are imprecise probability assignments over elementary events. They constitute a very convenient tool to model uncertain information : two common cases are confidence intervals on parameters of multinomial distributions built from sample data and expert opinions provided in terms of such intervals. In this paper, we study how probability intervals can be transformed into other uncertainty models such as possibility distributions, Ferson's p-boxes, random sets and Neumaier's clouds.","Title: A geometric and game-theoretic study of the conjunction of possibility measures.\nYear: 2015\n\nAbstract: We study the conjunction of possibility measures within imprecise probability theory.We identify conditions under which the conjunction remains a possibility measure.We provide a graphical way to check these conditions, using zero-sum games.We provide a graphical way to adjust possibility measures so their conjunction is a possibility measure.We identify conditions under which the conjunction is coherent. In this paper, we study the conjunction of possibility measures when they are interpreted as coherent upper probabilities, that is, as upper bounds for some set of probability measures. We identify conditions under which the minimum of two possibility measures remains a possibility measure. We provide graphical way to check these conditions, by means of a zero-sum game formulation of the problem. This also gives us a nice way to adjust the initial possibility measures so their minimum is guaranteed to be a possibility measure. Finally, we identify conditions under which the minimum of two possibility measures is a coherent upper probability, or in other words, conditions under which the minimum of two possibility measures is an exact upper bound for the intersection of the credal sets of those two possibility measures.","Title: Cautious Ordinal Classification by Binary Decomposition.\nYear: 2014\n\nAbstract: We study the problem of performing cautious inferences for an ordinal classification (a.k.a. ordinal regression) task, that is when the possible classes are totally ordered. By cautious inference, we mean that we may produce partial predictions when available information is insufficient to provide reliable precise ones. We do so by estimating probabilistic bounds instead of precise ones. These bounds induce a (convex) set of possible probabilistic models, from which we perform inferences. As the estimates or predictions for such models are usually computationally harder to obtain than for precise ones, we study the extension of two binary decomposition strategies that remain easy to obtain and computationally efficient to manipulate when shifting from precise to bounded estimates. We demonstrate the possible usefulness of such a cautious attitude on tests performed on benchmark data sets.","Title: A consistency-specificity trade-off to select source behavior in information fusion.\nYear: 2015\n\nAbstract: Combining pieces of information provided by several sources without or with little prior knowledge about the behavior of the sources is an old yet still important and rather open problem in the belief function theory. In this paper, we propose an approach to select the behavior of sources based on a very general and expressive fusion scheme, that has the important advantage of making clear the assumptions made about the sources. The selection process itself relies on two cornerstones that are the notions of specificity and consistency of a knowledge representation, and that we adapt to the considered fusion scheme. We illustrate our proposal on different examples and show that the proposed approach actually encompasses some important existing fusion strategies.","Title: Comments on \"A distance-based statistical analysis of fuzzy number-valued data\" by the SMIRE research group.\nYear: 2014\n\nAbstract: This paper is a fine review of various aspects related to the statistical handling of “ontic” random fuzzy sets by the means of appropriate distances. It is quite comprehensive and helpful, as it clarifies the status of fuzzy sets in such methods, explains the advantages of using a distance-based approach, specifies the pitfalls in which one should not fall when dealing with “ontic” random fuzzy sets and provides some illustration of practical computations. Not being a statistician but an occasional user of statistics, my discussion will mainly focus on this more practical aspect.","Title: Comments on \"Learning from imprecise and fuzzy observations: Data disambiguation through generalized loss minimization\" by Eyke Hüllermeier.\nYear: 2014\n\nAbstract: Eyke Hüllermeier provides a very convincing approach to learn from fuzzy data, both about the model and about the data themselves. In the process, he links the shape of fuzzy sets with classical loss functions, therefore providing strong theoretical links between fuzzy modeling and more classical machine learning approaches. This short note discusses various aspects of his proposal as well as possible extensions. I will first discuss the opportunity to consider more general uncertainty representations, before considering various alternatives to the proposed learning procedure. Finally, I will briefly discuss the differences I perceive about a loss-based and a likelihood-based approach.","Title: Nested Dichotomies with probability sets for multi-class classification.\nYear: 2014\n\nAbstract: Binary decomposition techniques transform a multi-class problem into several simpler binary problems. In such techniques, a classical issue is to ensure the consistency between the binary assessments of conditional probabilities. Nested dichotomies, which consider tree-shaped decomposition, do not suffer from this issue. Yet, a wrong probability estimate in the tree can strongly biase the results and provide wrong predictions. To overcome this issue, we consider in this paper imprecise nested dichotomies, in which binary probabilities become imprecise. We show in experiments that the approach has many advantages: it provides cautious inferences when only little information is available, and allows to make efficient computations with imprecise probabilities even when considering generic cost functions.","Title: Kolmogorov-Smirnov Test for Interval Data.\nYear: 2014\n\nAbstract: In this paper, we are interested in extending the classical Kolmogorov-Smirnov homogeneity test to compare two samples of interval-valued observed measurements. In such a case, the test result is interval-valued, and one major difficulty is to find the bounds of this set. We propose a very efficient computational method for approximating these bounds by using a p-box (pairs of upper and lower cumulative distributions) representation of the samples.","Title: A Note on Learning Dependence under Severe Uncertainty.\nYear: 2014\n\nAbstract: We propose two models, one continuous and one categorical, to learn about dependence between two random variables, given only limited joint observations, but assuming that the marginals are precisely known. The continuous model focuses on the Gaussian case, while the categorical model is generic. We illustrate the resulting statistical inferences on a simple example concerning the body mass index. Both methods can be extended easily to three or more random variables.","Title: Multilabel Prediction with Probability Sets: The Hamming Loss Case.\nYear: 2014\n\nAbstract: In this paper, we study how multilabel predictions can be obtained when our uncertainty is described by a convex set of probabilities. Such predictions, typically consisting of a set of potentially optimal decisions, are hard to make in large decision spaces such as the one considered in multilabel problems. However, we show that when considering the Hamming loss, an approximate prediction can be efficiently computed from label-wise information, as in the precise case. We also perform some first experiments showing the interest of performing partial predictions in the multilabel case.","Title: Toward an Axiomatic Definition of Conflict Between Belief Functions\nYear: 2013\n\nAbstract: Recently, the problem of measuring the conflict between two bodies of evidence represented by belief functions has known a regain of interest. In most works related to this issue, Dempster's rule plays a central role. In this paper, we propose to study the notion of conflict from a different perspective. We start by examining consistency and conflict on sets and extract from this settings basic properties that measures of consistency and conflict should have. We then extend this basic scheme to belief functions in different ways. In particular, we do not make any a priori assumption about sources (in)dependence and only consider such assumptions as possible additional information.","Title: Fuzzy belief structures viewed as classical belief structures: A practical viewpoint\nYear: 2010\n\nAbstract: Many authors have studied fuzzy belief structures, that is belief functions having fuzzy sets as focal elements. One of the main reason for this is that this structure offers a convenient way to mix probabilistic and fuzzy information. Still, one point on which authors often disagree is how information represented by fuzzy belief structures should be processed, i.e., how should be defined fusion operations, decision rules, uncertainty measures, uncertainty propagation, etc. for such representations. In this paper, we consider that fuzzy belief structures are mapped into classical belief structures encoding the same information, and propose to manipulate these latter structures. From a practical standpoint, it has the benefit that processing tools proper to belief structures can be used together with their interpretation, rather than having to consider a mix of probabilistic and fuzzy calculi, which can be harder to interpret.","Title: Modelling and predicting partial orders from pairwise belief functions\nYear: 2016\n\nAbstract: Abstract In this paper, we introduce a generic way to represent and manipulate pairwise information about partial orders (representing rankings, preferences, ...) with belief functions. We provide generic and practical tools to make inferences from this pairwise information and illustrate their use on the machine learning problems that are label ranking and multi-label prediction. Our approach differs from most other quantitative approaches handling complete or partial orders, in the sense that partial orders are here considered as primary objects and not as incomplete specifications of ideal but unknown complete orders.","Title: Online active learning of decision trees with evidential data\nYear: 2016\n\nAbstract: Learning from uncertain data has attracted increasing attention in recent years. In this paper, we propose a decision tree learning method that can not only handle uncertain data, but also reduce epistemic uncertainty by querying the most valuable uncertain instances within the learning procedure. Specifically, we use entropy intervals extracted from the evidential likelihood to query uncertain training instances when needed, with the goal to improve the selection of the splitting attribute. Experimental results under various conditions confirm the interest of the proposed approach. HighlightsActive belief decision trees are learned from uncertain data modelled by belief functions.A query strategy is proposed to query the most valuable uncertain instances while learning decision trees.To deal with evidential data, entropy intervals are extracted from the evidential likelihood.Experiments with UCI data illustrate the robustness of proposed approach to various kinds of uncertain data.","Title: An extension of the FURIA classification algorithm to low quality data through fuzzy rankings and its application to the early diagnosis of dyslexia.\nYear: 2016\n\nAbstract: An early detection and reeducation of dyslexic children is critical for their integration in the classroom. Parents and instructors can help the psychologist to detect potential cases of dyslexia before the children׳s writing age. Artificial intelligence tools can also assist in this task. Dyslexia symptoms are detected with tests whose results may be vague or ambiguous, thus machine learning techniques for low quality data are advised. In particular, in this paper it is suggested that a new extension to vague datasets of the classification algorithm FURIA (Fuzzy Unordered Rule Induction Algorithm) has advantages over other approaches in both the computational effort during the learning stage and the linguistic quality of the induced classification rules. The new approach is benchmarked with different test problems and compared to other artificial intelligence tools for dyslexia diagnosis in the literature.","Title: Cautious label ranking with label-wise decomposition.\nYear: 2015\n\nAbstract: •We propose to rank labels using label-wise information.•We consider the use of probability sets over ranks to produce cautious predictions.•We approximate the solution of the corresponding bilinear problem.•We perform some first experiment showing the potential benefits of the approach.","Title: Ranking of fuzzy intervals seen through the imprecise probabilistic lens\nYear: 2015\n\nAbstract: Within the fuzzy literature, the issue of ranking fuzzy intervals has been addressed by many authors, who proposed various solutions to the problem. Most of these solutions intend to find a total order on a given collection of fuzzy intervals. However, if one sees fuzzy intervals as descriptions of uncertain quantities, an alternative to rank them is to use ranking rules issued from the imprecise probabilistic literature. In this paper, we investigate ranking rules based on different statistical features (mean, median) and orderings, and relate the obtained (partial) orders to some classical proposals. In particular, we propose a generic expression of stochastic orderings, and then use it to systematically investigate extensions of the most usual stochastic orderings to fuzzy intervals. We also show some relations between those extensions, and explore their relation with existing fuzzy ranking proposals.","Title: A Decision Support System to design modified atmosphere packaging for fresh produce based on a bipolar flexible querying approach\nYear: 2015\n\nAbstract: We define a multi-criteria Decision Support System for designing fresh food packaging.A modified atmosphere packaging simulation module is included in the DSS.A flexible querying module handles imprecise data stored in a packaging database.Using the DSS the user will have only one trial to perform validation step. To design new packaging for fresh food, stakeholders of the food chain express their needs and requirements, according to some goals and objectives. These requirements can be gathered into two groups: (i) fresh food related characteristics and (ii) packaging intrinsic characteristics. Modified Atmosphere Packaging (MAP) is an efficient way to delay senescence and spoilage and thus to extend the very short shelf life of respiring products such as fresh fruits and vegetables. Consequently, packaging O2/CO2 permeabilities must fit the requirements of fresh fruits and vegetable as predicted by virtual MAP simulating tools. Beyond gas permeabilities, the choice of a packaging material for fresh produce includes numerous other factors such as the cost, availability, potential contaminants of raw materials, process ability, and waste management constraints. For instance, the user may have the following multi-criteria query for his/her product asking for a packaging with optimal gas permeabilities that guarantee product quality and optionally a transparent packaging material made from renewable resources with a cost for raw material less than 3 /kg. To help stakeholders taking a rational decision based on the expressed needs, a new multi-criteria Decision Support System (DSS) for designing biodegradable packaging for fresh produce has been built. In this paper we present the functional specification, the software architecture and the implementation of the developed tool. This tool includes (i) a MAP simulation module combining mass transfer models and respiration of the food, (ii) a multi-criteria flexible querying module which handles imprecise, uncertain and missing data stored in the database. We detail its operational functioning through a real life case study to determine the most satisfactory materials for apricots packaging.","Title: Multilabel predictions with sets of probabilities: The Hamming and ranking loss cases\nYear: 2015\n\nAbstract: In this paper, we study how multilabel predictions can be obtained when our uncertainty is described by a convex set of probabilities. Such predictions, typically consisting of a set of potentially optimal decisions, are hard to make in large decision spaces such as the one considered in multilabel problems. However, we show that when considering the Hamming or the ranking loss, outer-approximating predictions can be efficiently computed from label-wise information, as in the precise case. We also perform some first experiments showing the behaviour of the partial predictions obtained through these approximations. Such experiments also confirm that predictions become partial on those labels where the precise prediction is likely to make an error.","Title: Elicitation of a Utility from Uncertainty Equivalent Without Standard Gambles\nYear: 2015\n\nAbstract: In the context of decision under uncertainty, standard gambles are classically used to elicit a utility function on a set X of consequences. The utility of an element x in X is derived from the probability p for which a gamble giving the best outcome in X with probability p and the worst outcome in X otherwise, is indifferent to getting x for sure. In many situations, uncertainty that can be observed on the true value of X concerns only neighbour values. Uncertainty is then represented by a probability distribution whose support is an interval. In this case, standard gambles are unrealistic for the decision maker. We consider uncertainty represented by an equi-probability over an interval of X. This paper addresses the elicitation of a utility function on X by obtaining the certainty equivalent of an equi-probability over an interval of X. We show that not all utility models are suitable to accomplish this task.","Title: Optimal expert elicitation to reduce interval uncertainty.\nYear: 2015\n\nAbstract: Reducing uncertainty is an important problem in many applications such as risk and reliability analysis, system design, etc. In this paper, we study the problem of optimally querying experts to reduce interval uncertainty. Surprisingly, this problem has received little attention in the past, while similar issues in preference elicitation or social choice theory have witnessed a rising interest. We propose and discuss some solutions to determine optimal questions in a myopic way (one-at-a-time), and study the computational aspects of these solutions both in general and for some specific functions of practical interest. Finally, we illustrate the application of the approach in reliability analysis problems.","Title: Comparing System Reliabilities with Ill-Known Probabilities.\nYear: 2016\n\nAbstract: In reliability analysis, comparing system reliability is an essential task when designing safe systems. When the failure probabilities of the system components (assumed to be independent) are precisely known, this task is relatively simple to achieve, as system reliabilities are precise numbers. When failure probabilities are ill-known (known to lie in an interval) and we want to have guaranteed comparisons (i.e., declare a system more reliable than another when it is for any possible probability value), there are different ways to compare system reliabilities. We explore the computational problems posed by such extensions, providing first insights about their pros and cons.","Title: Easy and optimal queries to reduce set uncertainty.\nYear: 2017\n\nAbstract: •A setting for optimal expert elicitation to reduce set uncertainty is described.•The formal approach is justified by simulated experiments in reliability.•The approach proves to be efficient and computationally tractable.•Computations are particularly simple for specific functions of practical interest.","Title: A decision support system for eco-efficient biorefinery process comparison using a semantic approach.\nYear: 2016\n\nAbstract: •We define a decision support system for eco-efficient biorefinery process comparison.•Uncertainty is managed all along the pipeline for eco-design indicator computations.•Data extraction from textual documents is guided by a Termino-ontological resource.•Flexible structured database querying is guided by a Termino-ontological resource.","Title: Idempotent Conjunctive Combination of Belief Functions by Distance Minimization.\nYear: 2016\n\nAbstract: When combining multiple belief functions, designing a combination rule that selects the least informative belief function among those more informative than each of the combined ones is a difficult task. Such rules, commonly depicted as \"cautious\", are typically required to be idempotent, since when one is cautious, combining identical information should not lead to the reinforcement of some hypothesis. However, applying the least commitment principle using partial orders is in general not straightforward, mainly due to the non-uniqueness of solutions. Building upon previous work, this paper investigates the use of distances compatible with such partial orders to determine a unique solution to the combination problem. The obtained operators are conjunctive, idempotent and commutative, but lack associativity. They are, however, quasi-associative allowing sequential combinations at no extra complexity.","Title: Partial Data Querying Through Racing Algorithms.\nYear: 2018\n\nAbstract: The paper studies the problem of actively learning from instances characterized by imprecise features or imprecise class labels, where by actively learning we understand the possibility to query the precise value of imprecisely specified data. We differ from classical active learning by the fact that in the later, data are either fully precise or completely missing, while in our case they can be partially specified. Such situations can appear when sensor errors are important to encode, or when experts have only specified a subset of possible labels when tagging data. We provide a general active learning technique that can be applied in principle to any model. It is inspired from racing algorithms, in which several models are competing against each others. The main idea of our method is to identify the query that will be the most helpful in identifying the winning model in the competition. After discussing and formalizing the general ideas of our approach, we illustrate it by studying the particular case of binary SVM in the case of interval valued features and set-valued labels. The experimental results indicate that, in comparison to other baselines, racing algorithms provide a faster reduction of the uncertainty in the learning process, especially in the case of imprecise features.","Title: Unifying parameter learning and modelling complex systems with epistemic uncertainty using probability interval.\nYear: 2016\n\nAbstract: Knowledge regarding complex systems are heterogeneous and fragmented.modelling dynamic complex systems in the framework of dynamic credal networks.practical methodology coupling Dirichlet distributions with interval probabilities to incrementally build and update model parameters whatever source and format of knowledge.enables to take into account (1) stochastic and epistemic uncertainties pertaining to the system; (2) the confidence level on the different sources of information.illustrate the application of the methodology to the modelling of a simplified industrial case study. Modeling complex dynamical systems from heterogeneous pieces of knowledge varying in precision and reliability is a challenging task. We propose the combination of dynamical Bayesian networks and of imprecise probabilities to solve it. In order to limit the computational burden and to make interpretation easier, we also propose to encode pieces of (numerical) knowledge as probability intervals, which are then used in an imprecise Dirichlet model to update our knowledge. The idea is to obtain a model flexible enough so that it can easily cope with different uncertainties (i.e., stochastic and epistemic), integrate new pieces of knowledge as they arrive and be of limited computational complexity.","Title: Querying Partially Labelled Data to Improve a K-nn Classifier.\nYear: 2017\n\nAbstract: When learning from instances whose output labels may be partial, the problem of knowing which of these output labels should be made precise to improve the accuracy of predictions arises. This problem can be seen as the intersection of two tasks: the one of learning from partial labels and the one of active learning, where the goal is to provide the labels of additional instances to improve the model accuracy. In this paper, we propose querying strategies of partial labels for the well-known K-nn classifier. We propose different criteria of increasing complexity, using among other things the amount of ambiguity that partial labels introduce in the K-nn decision rule. We then show that our strategies usually outperform simple baseline schemes, and that more complex strategies provide a faster improvement of the model accuracies.","Title: K-Nearest Neighbour Classification for Interval-Valued Data.\nYear: 2017\n\nAbstract: This paper studies the problem of providing predictions with a K-nn approach when data have partial features given in the form of intervals. To do so, we adopt an optimistic approach to replace the ill-known values, that requires to compute sets of possible and necessary neighbours of an instance. We provide an easy way to compute such sets, as well as the decision rule that follows from them. Our approach is then compared to a simple imputation method in different scenarios, in order to identify those ones where it is advantageous.","Title: Reasons and Means to Model Preferences as Incomplete.\nYear: 2018\n\nAbstract: Literature involving preferences of artificial agents or human beings often assume their preferences can be represented using a complete transitive binary relation. Much has been written however on different models of preferences. We review some of the reasons that have been put forward to justify more complex modeling, and review some of the techniques that have been proposed to obtain models of such preferences.","Title: On extreme points of p-boxes and belief functions.\nYear: 2016\n\nAbstract: Within imprecise probability theory, the extreme points of convex probability sets have an important practical role (to perform inference on graphical models, to compute expectation bounds, …). This is especially true for sets presenting specific features that make them easy to manipulate in applications. This easiness is the reason why extreme points of such models (probability intervals, possibility distributions, …) have been well studied. Yet, imprecise cumulative distributions (a.k.a. p-boxes) constitute an important exception, as the characterization of their extreme points remain to be studied. This is what we do in this paper, where we characterize the maximal number of extreme points of a p-box, give a family of p-boxes that attains this number and show an algorithm that allows to compute the extreme points of a given p-box. To achieve all this, we also provide what we think to be a new characterization of extreme points of a belief function.","Title: On the median in imprecise ordinal problems.\nYear: 2017\n\nAbstract: When having to make a prediction under probabilistic uncertainty in ordinal problems, the median offers a number of interesting properties compared to other statistics such as the expected value. In particular, it does not depend on a particular metric defined over the elements, but still takes account of the ordinal nature of the data. It can also be shown to be the minimizer of the (L_1) loss function. In this paper, we show that similar results can be obtained when the uncertainty is described not by a single probability distribution, but by a convex set of those. In particular, we relate the lower and upper medians to the (L_1) loss function via the notion of lower and upper expectations (and extend these results to general quantiles). We also show that, using a different decision rule, the lower and upper median can be retrieved when assuming the cost to be strictly monotonic and symmetric, and nothing more. Finally, we run some tests to show the interest of using Median based predictions with convex sets of probabilities in ordinal regression problems.","Title: Comonotonicity for sets of probabilities.\nYear: 2017\n\nAbstract: Two variables are called comonotone when there is an increasing relation between them, in the sense that when one of them increases (decreases), so does the other one. This notion has been widely investigated in probability theory, and is related to copulas. This contribution studies how the notion of comonotonicity can be extended to an imprecise setting on discrete spaces, where probabilities are only known to belong to a convex set. We define comonotonicity for such sets and investigate its characterizations in terms of lower probabilities, as well as its connection with copulas. As this theoretical characterization can be tricky to apply to general lower probabilities, we also investigate specific models of practical importance. In particular, we provide some sufficient conditions for a comonotone belief function with fixed marginals to exist, and characterize comonotone bivariate p-boxes.","Title: The Costs of Indeterminacy: How to Determine Them?\nYear: 2017\n\nAbstract: Indeterminate classifiers are cautious models able to predict more than one class in case of high uncertainty. A problem that arises when using such classifiers is how to evaluate their performances. This problem has already been considered in the case where all prediction errors have equivalent costs (that we will refer as the “0/1 costs” or accuracy setting). The purpose of this paper is to stud...","Title: Cautious classification with nested dichotomies and imprecise probabilities.\nYear: 2017\n\nAbstract: In some applications of machine learning and information retrieval (e.g. medical diagnosis, image recognition, pre-classification...), it can be preferable to provide less informative but more reliable predictions. This can be done by making partial predictions in the form of class subsets when the available information is insufficient to provide a reliable unique class. Imprecise probabilistic approaches offer nice tools to learn models from which such cautious predictions can be produced. However, the learning and inference processes of such models are computationally harder than their precise counterparts. In this paper, we introduce and study a particular binary decomposition strategy, nested dichotomies, that offer computational advantages in both the learning (due to the binarization process) and the inference (due to the decomposition strategy) processes. We show with experiments that these computational advantages do not lower the performances of the classifiers, and can even improve them when the class space has some structure.","Title: Idempotent conjunctive and disjunctive combination of belief functions by distance minimization.\nYear: 2018\n\nAbstract: Idempotence is a desirable property when cautiousness is wanted in an information fusion process, since in this case combining identical information should not lead to the reinforcement of some hypothesis. Idempotent operators also guarantee that identical information items are not counted twice in the fusion process, a very important property in decentralized applications where the information origin cannot always be tracked (ad-hoc wireless networks are typical examples). In the theory of belief functions, a sound way to combine conjunctively multiple information items is to design a combination rule that selects the least informative element among a subset of belief functions more informative than each of the combined ones. In contrast, disjunctive rules can be retrieved by selecting the most informative element among a subset of belief functions less informative than each of the combined ones. One interest of such approaches is that they provide idempotent rules by construction.","Title: Classification by pairwise coupling of imprecise probabilities.\nYear: 2018\n\nAbstract: •We address combining imprecise classifiers in the framework of imprecise probabilities.•Classifier outputs are probability intervals, which are large for irrelevant classifiers.•The combination consists in computing probabilities consistent with the intervals.•This approach is particularly interesting when combining few classifiers.","Title: Density Estimation with Imprecise Kernels: Application to Classification.\nYear: 2018\n\nAbstract: In this paper, we explore the problem of estimating lower and upper densities from imprecisely defined families of parametric kernels. Such estimations allow to rely on a single bandwidth value, and we show that it provides good results on classification tasks when extending the naive Bayesian classifier.","Title: Technical Gestures Recognition by Set-Valued Hidden Markov Models with Prior Knowledge.\nYear: 2016\n\nAbstract: Hidden Markov models are popular tools for gesture recognition. Once the generative processes of gestures have been identified, an observation sequence is usually classified as the gesture having the highest likelihood, thus ignoring possible prior information. In this paper, we consider two potential improvements of such methods: the inclusion of prior information, and the possibility of considering convex sets of probabilities (in the likelihoods and the prior) to infer imprecise, but more reliable, predictions when information is insufficient. We apply the proposed approach to technical gestures, typically characterized by severe class imbalance. By modelling such imbalances as a prior information, we achieve more accurate results, while the imprecise quantification is shown to produce more reliable estimates.","Title: From Relations Between Sets to Relations Between Belief Functions.\nYear: 2018\n\nAbstract: In uncertainty theories, a common problem is to define how we can extend relations between sets (e.g., inclusion, ranking, consistency,...) to corresponding notions between uncertainty representations. Such definitions can then be used to perform the same operations as those that are done for sets: measuring information content, ordering alternatives or checking consistency, to name a few. In this paper, we propose a general way to extend set relations to belief functions, using constrained stochastic matrices to identify those belief functions in relation. We then study some properties of our proposal, as well as its relations with existing works focusing on peculiar relations.","Title: Pari-mutuel probabilities as an uncertainty model.\nYear: 2019\n\nAbstract: •We prove some properties of the Pari Mutuel Model as a coherent lower (and upper) probability.•We show that a Pari Mutuel Model is a particular case of a probability interval.•We give the maximal number of extreme points of the credal set associated with a Pari Mutuel Model and we give the form of those extreme points.•We give sufficient and necessary conditions for the Pari Mutuel Model to induce a belief function.•We show how to merge information given in terms of Pari Mutuel Models.","Title: A decision support system using multi-source scientific data, an ontological approach and soft computing - application to eco-efficient biorefinery\nYear: 2016\n\nAbstract: In decision tasks such as bioprocess efficiency comparison, scientific literature is a valuable source of data. This large number of scientific data is heterogeneously structured, mainly in textual format. Innovative tools able to integrate and treat constantly new information are required. In this context, the use of semantic web methods such as ontologies seems relevant to structure the experimental information. Imprecision and uncertainty can arise from data incompleteness and variability. This is particularly true for processes involving biological materials. Document reliability should also be considered. Soft computing methods have the potential to be the kingpin of specialized software that can be integrated in decision support systems (DSS) intended to solve these issues. This paper presents the implementation of a pipeline which permits to: (1) structure and integrate the experimental data of interest by using ontologies, (2) assess data source reliability, (3) compute and visualize indicators taking into account data imprecision.","Title: From set relations to belief function relations\nYear: 2019\n\nAbstract: In uncertainty theories, a common problem is to define how we can extend relations between sets (e.g., inclusion, ranking, consistency, …) to corresponding notions between uncertainty representations. Such definitions can then be used to perform the same operations as those that are done for sets: comparing information content, ordering alternatives or checking consistency, to name a few. In this paper, we propose a general way to extend set relations to belief functions, using constrained stochastic matrices to identify those belief functions in relation. We then study some properties of our proposal, as well as its connections with existing works focusing on specific relations.","Title: Imprecise Gaussian discriminant classification\nYear: 2019\n\nAbstract: •We robustify Gaussian discriminant analysis by considering sets of estimates.•We use near-ignorance priors to derive bounding boxes on mean estimates.•We discuss the computational issue for generic and diagonal covariance matrices.•We make a full experimental study showing the benefits of using imprecise estimates.•We make a first exploration of the benefits of using the model in non i.i.d. situations.","Title: Racing Trees To Query Partial Data\nYear: 2021\n\nAbstract: Dealing with partially known or missing data is a common problem in machine learning. This work is interested in the problem of querying the true value of data to improve the quality of the learned model, when those data are only partially known. This study is in the line of active learning, since we consider that the precise value of some partial data can be queried to reduce the uncertainty in the learning process, yet can consider any kind of partial data (not only entirely missing one). We propose a querying strategy based on the concept of racing algorithms in which several models are competing. The idea is to identify the query that will help the most to quickly decide the winning model in the competition. After discussing and formalizing the general ideas of our approach, we study the particular case of decision trees in case of interval-valued features and set-valued labels. The experimental results indicate that, in comparison with other baselines, the proposed approach significantly outperforms simpler strategies in the case of partially specified features, while it achieves similar performances in the case of partially specified labels.","Title: Multi-label Chaining with Imprecise Probabilities\nYear: 2021\n\nAbstract: We present two different strategies to extend the classical multi-label chaining approach to handle imprecise probability estimates. These estimates use convex sets of distributions (or credal sets) in order to describe our uncertainty rather than a precise one. The main reasons one could have for using such estimations are (1) to make cautious predictions (or no decision at all) when a high uncertainty is detected in the chaining and (2) to make better precise predictions by avoiding biases caused in early decisions in the chaining. We adapt both strategies to the case of the naive credal classifier, showing that this adaptations are computationally efficient. Our experimental results on missing labels, which investigate how reliable these predictions are in both approaches, indicate that our approaches produce relevant cautiousness on those hard-to-predict instances where the precise models fail.","Title: Inference And Decision In Credal Occupancy Grids: Use Case On Trajectory Planning\nYear: 2021\n\nAbstract: Occupancy grids are common tools used in robotics to represent the robot environment, and that may be used to plan trajectories, select additional measurements to acquire, etc. However, deriving information about those occupancy grids from sensor measurements often induce a lot of uncertainty, especially for grid elements that correspond to occluded or far away area from the robot. This means that occupancy information may be quite uncertain and imprecise at some places, while being very accurate at others. Modelling finely this occupancy information is essential to decide the optimal action the robot should take, but a refined modelling of uncertainty often implies a higher computational cost, a prohibitive feature for real-time applications. In this paper, we introduce the notion of credal occupancy grids, using the very general theory of imprecise probabilities to model occupancy uncertainty. We also show how one can perform efficient, real-time inferences with such a model, and show a use-case applying the model to an autonomous vehicle trajectory planning problem.","Title: Processing Multiple Distortion Models: a Comparative Study\nYear: 2021\n\nAbstract: When dealing with uncertain information, distortion or neighbourhood models are convenient practical tools, as they rely on very few parameters. In this paper, we study their behaviour when such models are combined and processed. More specifically, we study their behaviour when merging different distortion models quantifying uncertainty on the same quantity, and when manipulating distortion models defined over multiple variables.","Title: Logical and Evidential Inconsistencies Meet: First Steps\nYear: 2021\n\nAbstract: Measuring inconsistency has been and is still an active research topic in both logic and evidence theory. However, the two fields have developed distinct notions and measures of inconsistency, following different paths. In this paper, we attempt to build some first bridges between the two trends, suggesting some first means for one to enrich the other, and vice-versa.","Title: A Note About Entropy and Inconsistency in Evidence Theory\nYear: 2021\n\nAbstract: Information content is classically measured by entropy measures in probability theory, that can be interpreted as a measure of internal inconsistency of a probability distribution. While extensions of Shannon entropy have been proposed for quantifying information content of a belief function, other trends have been followed which rather focus on the notion of consistency between sets. Relying on previous general entropy measures of probability, we propose in this paper to establish some links between the different measures of internal inconsistency of a belief functions. We propose a general formulation which encompasses inconsistency measures derived from Shannon entropy as well as those derived from the N-consistency family of measures.","Title: Possibilistic preference elicitation by minimax regret.\nYear: 2021\n\nAbstract: Identifying the preferences of a given user through elicitation is a central part of multi-criteria decision aid (MCDA) or preference learning tasks. Two classical ways to perform this elicitation is to use either a robust or a Bayesian approach. However, both have their shortcoming: the robust approach has strong guarantees through very strong hypotheses, but cannot integrate uncertain information. While the Bayesian approach can integrate uncertainties, but sacrifices the previous guarantees and asks for stronger model assumptions. In this paper, we propose and test a method based on possibility theory, which keeps the guarantees of the robust approach without needing its strong hypotheses. Among other things, we show that it can detect user errors as well as model misspecification.","Title: Quantification of Credal Uncertainty in Machine Learning: A Critical Analysis and Empirical Comparison.\nYear: 2022\n\nAbstract: The representation and quantification of uncertainty has received increasing attention in machine learning in the recent past. The formalism of credal sets provides an interesting alternative in this regard, especially as it combines the representation of epistemic (lack of knowledge) and aleatoric (statistical) uncertainty in a rather natural way. In this paper, we elaborate on uncertainty measures for credal sets from the perspective of machine learning. More specifically, we provide an overview of proposals, discuss existing measures in a critical way, and also propose a new measure that is more tailored to the machine learning setting. Based on an experimental study, we conclude that theoretically well-justified measures also lead to better performance in practice. Besides, we corroborate the difficulty of the disaggregation problem, that is, of decomposing the amount of total uncertainty into aleatoric and epistemic uncertainty in a sound manner, thereby complementing theoretical findings with empirical evidence."],"string":"[\n \"Title: F-boxes for filtering.\\nYear: 2011\\n\\nAbstract: Selecting a particular summative kernel (i.e., formally equivalent to a probability distribution) when filtering a digital signal can be a difficult task. An obvious solution to this problem is to filter with multiple kernels rather than with a single one, but the computing cost of such a solution can quickly become prohibitive (especially in real-time applications). Another alternative, the one studied in this paper, is to consider kernels modeled by imprecise probabilistic representations. Considering such representations makes the use of numerical tools coming from imprecise probability theory possible, such as the Choquet integral, and allows one to work with multiple kernels without multiplying too much the number of required computations. In this paper, we propose to use the well-known p-box representation to filter a digital signal. We show that the use p-boxes allows making more precise inferences than those obtained with possibility distributions and clouds. We then discuss the practical aspect of computing a filtered signal with p-boxes, and finish by some experiments.\",\n \"Title: Classification Trees Based on Belief Functions.\\nYear: 2012\\n\\nAbstract: Decision tree classifiers are popular classification methods. In this paper, we extend to multi-class problems a decision tree method based on belief functions previously described for two-class problems only. We propose three possible extensions: combining multiple two-class trees together and directly extending the estimation of belief functions within the tree to the multi-class setting. We provide experiment results and compare them to usual decision trees.\",\n \"Title: A New Contextual Discounting Rule For Lower Probabilities\\nYear: 2010\\n\\nAbstract: Sources providing information about the value of a variable may not be totally reliable. In such a case, it is common in uncertainty theories to take account of this unreliability by a so-called discounting rule. A few discounting rules have been proposed in the framework of imprecise probability theory, but one of the drawback of those rules is that they do not preserve interesting properties (i.e. n-monotonicity) of lower probabilities. Another aspect that only a few of them consider is that source reliability is often dependent of the context, i.e. a source may be more reliable to identify some values than others. In such cases, it is useful to consider contextual discounting, where reliability information is dependent of the variable values. In this paper, we propose such a contextual discounting rule that also preserves some of the interesting mathematical properties a lower probability can have.\",\n \"Title: Data reliability assessment in a data warehouse opened on the web\\nYear: 2011\\n\\nAbstract: This paper presents an ontology-driven workflow that feeds and queries a data warehouse opened on the Web. Data are extracted from data tables in Web documents. As web documents are very heterogeneous in nature, a key issue in this workflow is the ability to assess the reliability of retrieved data. We first recall the main steps of our method to annotate and query Web data tables driven by a domain ontology. Then we propose an original method to assess Web data table reliability from a set of criteria by the means of evidence theory. Finally, we show how we extend the workflow to integrate the reliability assessment step.\",\n \"Title: Independence and 2-monotonicity: nice to have, hard to keep\\nYear: 2011\\n\\nAbstract: When using lower probabilities to model uncertainty about the value assumed by a variable, 2-monotonicity is an interesting property to satisfy, as it greatly facilitates further treatments (such as the computation of lower/upper expectation bounds). In this paper, we show that multivariate joint models induced from marginal ones by strong independence, epistemic independence or epistemic irrelevance do not usually preserve such a property, even if it is satisfied by all marginal models. We then propose a joint model outer-approximating those induced by strong and epistemic independence and study some of its properties.\",\n \"Title: A Unified View of Some Representations of Imprecise Probabilities\\nYear: 2006\\n\\nAbstract: Several methods for the practical representation of imprecise probabilities exist such as Ferson's p-boxes, possibility distributions, Neumaier's clouds, and random sets. in this paper some relationships existing between the four kinds of representations are discussed. A cloud as well as a p-box can be modelled as a pair of possibility distributions. We show that a generalized form of p-box is a special kind of belief function and also a special kind of cloud.\",\n \"Title: Selecting source behavior in information fusion on the basis of consistency and specificity\\nYear: 2013\\n\\nAbstract: Combining pieces of information provided by several sources without prior knowledge about the behavior of the sources is an old yet still important and rather open problem in belief function theory. In this paper, we propose a general approach to select the behavior of sources, based on two cornerstones of information fusion that are the notions of specificity and consistency. This approach is framed in a recently introduced and general fusion scheme that allows a wide range of assumptions on the sources. In the process, we are also led to generalize a recently introduced measure of conflict to all Boolean connectives. Eventually, we show that our approach generalizes some important existing information fusion strategies.\",\n \"Title: Handling Bipolar Knowledge with Credal Sets\\nYear: 2010\\n\\nAbstract: How to represent and handle bipolar information has recently received a lot of attention. Being bipolar means that the information has a positive and negative part. In this paper, we consider asymmetric bipolar information (i.e. situations where positive and negative information are unrelated and should be processed separately). We propose a framework to represent and handle it with so-called credal sets, i.e., with convex sets of probability distributions. We also provide some illustrative examples.\",\n \"Title: Combining binary classifiers with imprecise probabilities\\nYear: 2011\\n\\nAbstract: This paper proposes a simple framework to combine binary classifiers whose outputs are imprecise probabilities (or are transformed into some imprecise probabilities, e.g., by using confidence intervals). This combination comes down to solve linear programs describing constraints over events (here, subsets of classes). The number of constraints grows linearly with the number of classifiers, making the proposed framework tractable for problems involving a relatively large number of classes. After detailing the method, we provide some first experimental results illustrating the method interests.\",\n \"Title: Relating Epistemic Irrelevance to Event Trees\\nYear: 2008\\n\\nAbstract: We relate the epistemic irrelevance in Walley's behavioural theory of imprecise probabilities to the event-tree independence due to Shafer. In particular, we show that forward irrelevance is equivalent to event-tree independence in particular event trees, suitably generalised to allow for the fact that imprecise rather than precise probability models are attached to the nodes in the tree. This allows us to argue that in a theory of uncertain processes, the asymmetrical notion of epistemic irrelevance has a more important role to play than its more involved and symmetrical counterpart called epistemic independence.\",\n \"Title: Making ontology-based knowledge and decision trees interact: an approach to enrich knowledge and increase expert confidence in data-driven models\\nYear: 2010\\n\\nAbstract: When using data-driven models to make simulations and predictions in experimental sciences, it is essential for the domain expert to be confident about the predicted values. Increasing this confidence can be done by using interpretable models, so that the expert can follow the model reasoning pattern, and by integrating expert knowledge to the model itself. New pieces of useful formalised knowledge can then be integrated to an existing corpus while data-driven models are tuned according to the expert advice. In this paper, we propose a generic interactive procedure, relying on an ontology to model qualitative knowledge and on decision trees as a data-driven rule learning method. A case study based on data issued from multiple scientific papers in the field of cereal transformation illustrates the approach.\",\n \"Title: A Consonant Approximation Of The Product Of Independent Consonant Random Sets\\nYear: 2009\\n\\nAbstract: The belief structure resulting from the combination of consonant and independent marginal random sets is not, in general, consonant. Also, the complexity of such a structure grows exponentially with the number of combined random sets, making it quickly intractable for computations. In this paper, we propose a simple guaranteed consonant outer approximation of this structure. The complexity of this outer approximation does not increase with the number of marginal random sets (i.e., of dimensions), making it easier to handle in uncertainty propagation. Features and advantages of this outer approximation are then discussed, with the help of some illustrative examples.\",\n \"Title: Generalised p-Boxes on Totally Ordered Spaces\\nYear: 2008\\n\\nAbstract: Probability boxes are among the most simple and popular models used in imprecise probability theory, and many practical results concerning them exist in the literature. Nevertheless, little attention has been paid to their formal characterisation in the setting of Walley's behavioural theory of imprecise probabilities. This paper tries to remedy this situation by formalising, generalising and extending existing results as well as by giving new ones, within Walley's framework.\",\n \"Title: Learning Decision Trees from Uncertain Data with an Evidential EM Approach\\nYear: 2013\\n\\nAbstract: In real world applications, data are often uncertain or imperfect. In most classification approaches, they are transformed into precise data. However, this uncertainty is an information in itself which should be part of the learning process. Data uncertainty can take several forms: probabilities, (fuzzy)sets of possible values, expert assessments, etc. We therefore need a flexible and generic enough model to represent and treat this uncertainty, such as belief functions. Decision trees are well known classifiers which are usually learned from precise datasets. In this paper we propose a methodology to learn decision trees from uncertain data in the belief function framework. In the proposed method, the tree parameters are estimated through the maximization of an evidential likelihood function computed from belief functions, using the recently proposed E2M algorithm that extends the classical EM. Some promising experiments compare the obtained trees with classical CART decision trees.\",\n \"Title: Possibilistic information fusion using maximal coherent subsets\\nYear: 2009\\n\\nAbstract: When multiple sources provide information about the same unknown quantity, their fusion into a synthetic interpretable message is often a tricky problem, especially when sources are conflicting. In this paper, we propose to use possibility theory and the notion of maximal coherent subsets (MCSs), often used in logic-based representations, to build a fuzzy belief structure that will be instrumental both for extracting useful insight about various features of the information conveyed by the sources and for compressing this information into a unique possibility distribution. Extensions and properties of the basic fusion rule are also studied.\",\n \"Title: Ontology-driven possibilistic reference fusion\\nYear: 2010\\n\\nAbstract: It often happens that different references (i.e. data descriptions), possibly coming from different heterogeneous data sources, concern the same real world entity. In such cases, it is necessary: (i) to detect, through reconciliation methods, whether different data descriptions refer to the same real world entity and (ii) to fuse them into a unique representation. Here we assume the reference reconciliation is solved, and we propose a fusion method based on possibility theory, able to cope with uncertainty and with ontological knowledge. An implementation using W3C standards is provided. Rising from the fusion process, an ontology enrichment procedure is proposed to complete the global ontology.\",\n \"Title: An extension of Universal Generating Function in Multi-State Systems Considering Epistemic Uncertainties.\\nYear: 2013\\n\\nAbstract: Many practical methods and different approaches have been proposed to assess Multi-State Systems (MSS) reliability measures. The universal generating function (UGF) method, introduced in 1986, is known to be a very efficient way of evaluating the availability of different types of MSSs. In this paper, we propose an extension of the UGF method considering epistemic uncertainties. This extended meth...\",\n \"Title: Can the Minimum Rule of Possibility Theory Be Extended to Belief Functions?\\nYear: 2009\\n\\nAbstract: When merging belief functions, Dempster rule of combination is justified only when sources can be considered as independent. When dependencies are ill-known, it is usual to ask the merging operation to satisfy the property of idempotence, as this property ensures a cautious behaviour in the face of dependent sources. There are different strategies to find such rules for belief functions. One strategy is to rely on idempotent rules used in either more general or more specific frameworks and to respectively study their particularisation or extension to belief functions. In this paper, we try to extend the minimum rule of possibility theory to belief functions. We show that such an extension is not always possible, unless we accept the idea that the result of the fusion process can be a family of belief functions.\",\n \"Title: Combining Imprecise Probability Masses with Maximal Coherent Subsets: Application to Ensemble Classification.\\nYear: 2012\\n\\nAbstract: When working with sets of probabilities, basic information fusion operators quickly reach their limits: intersection becomes empty, while union results in a poorly informative model. An attractive means to overcome these limitations is to use maximal coherent subsets (MCS). However, identifying the maximal coherent subsets is generally NP-hard. Previous proposals advocating the use of MCS to merge probability sets have not provided efficient ways to perform this task. In this paper, we propose an efficient approach to do such a merging between imprecise probability masses, a popular model of probability sets, and test it on an ensemble classification problem.\",\n \"Title: Using The Ols Algorithm To Build Interpretable Rule Bases: An Application To A Depollution Problem\\nYear: 2007\\n\\nAbstract: One of the main advantages of fuzzy modeling is the ability to yield interpretable results. Amongst these modeling methods, the OLS algorithm is a mathematically robust technique that allows to induce a fuzzy rule base from a set of training data. It does so by using linear regression to select the most important rules. However, the original OLS algorithm only relies upon numerical accuracy, and doesn't take interpretability matters into account. Thus, we propose some modifications to the original method so that it builds interpretable rule bases.\",\n \"Title: Extreme points of the credal sets generated by elementary comparative probabilities\\nYear: 2013\\n\\nAbstract: When using convex probability sets (or, equivalently, lower previsions) as models of uncertainty, identifying extreme points can be useful to perform various computations or to use some algorithms. In general, sets induced by specific models such as possibility distributions, linear vacuous mixtures or 2-monotone measures may have extreme points easier to compute than generic convex sets. In this paper, we study extreme points of another specific model: comparative probability orderings between the elements of a finite space. We use these extreme points to study the properties of the lower probability induced by this set, and connect comparative probabilities with other uncertainty models.\",\n \"Title: A Decision Rule for Imprecise Probabilities Based on Pair-Wise Comparison of Expectation Bounds\\nYear: 2010\\n\\nAbstract: There are many ways to extend the classical expected utility decision rule to the case where uncertainty is described by (convex) probability sets. In this paper, we propose a simple new decision rule based on the pair-wise comparison of lower and upper expected bounds. We compare this rule to other rules proposed in the literature, showing that this new rule is both precise, computationally tractable and can help to boost the computation of other, more computationally demanding rules.\",\n \"Title: How To Randomly Generate Mass Functions\\nYear: 2013\\n\\nAbstract: As Dempster-Shafer theory spreads in different application fields, and as mass functions are involved in more and more complex systems, the need for algorithms randomly generating mass functions arises. Such algorithms can be used, for instance, to evaluate some statistical properties or to simulate the uncertainty in some systems (e.g., data base content, training sets). As such random generation is often perceived as secondary, most of the proposed algorithms use straightforward procedures whose sample statistical properties can be difficult to characterize. Thus, although such algorithms produce randomly generated mass functions, they do not always produce what could be expected from them (for example, uniform sampling in the set of all possible mass functions). In this paper, we briefly review some well-known algorithms, explaining why their statistical properties are hard to characterize. We then provide relatively simple algorithms and procedures to perform efficient random generation of mass functions whose sampling properties are controlled.\",\n \"Title: Random Generation of Mass Functions: A Short Howto.\\nYear: 2012\\n\\nAbstract: As Dempster-Shafer theory spreads in different applications fields involving complex systems, the need for algorithms randomly generating mass functions arises. As such random generation is often perceived as secondary, most pro-posed algorithms use procedures whose sample statistical properties are difficult to characterize. Thus, although they produce randomly generated mass functions, it is difficult to control the sample statistical laws. In this paper, we briefly review classical algorithms, explaining why their statistical properties are hard to characterize, and then provide simple procedures to perform efficient and controlled random generation.\",\n \"Title: Correcting binary imprecise classifiers: local vs global approach\\nYear: 2012\\n\\nAbstract: This paper proposes a simple strategy for combining binary classifiers with imprecise probabilities as outputs. Our combination strategy consists in computing a set of probability distributions by solving an optimization problem whose constraints depend on the classifiers outputs. However, the classifiers may provide assessments that are jointly incoherent, in which case the set of probability distributions satisfying all the constraints is empty. We study different correction strategies for restoring this consistency, by relaxing the constraints of the optimization problem so that it becomes feasible. In particular, we propose and compare a global strategy, where all constraints are relaxed to the same level, to a local strategy, where some constraints may be relaxed more than others. The local discounting strategy proves to give very good results compared both to single classifier approaches and to classifier combination schemes using a global correction scheme.\",\n \"Title: Probability boxes on totally preordered spaces for multivariate modelling\\nYear: 2011\\n\\nAbstract: A pair of lower and upper cumulative distribution functions, also called probability box or p-box, is among the most popular models used in imprecise probability theory. They arise naturally in expert elicitation, for instance in cases where bounds are specified on the quantiles of a random variable, or when quantiles are specified only at a finite number of points. Many practical and formal results concerning p-boxes already exist in the literature. In this paper, we provide new efficient tools to construct multivariate p-boxes and develop algorithms to draw inferences from them. For this purpose, we formalise and extend the theory of p-boxes using Walley's behavioural theory of imprecise probabilities, and heavily rely on its notion of natural extension and existing results about independence modeling. In particular, we allow p-boxes to be defined on arbitrary totally preordered spaces, hence thereby also admitting multivariate p-boxes via probability bounds over any collection of nested sets. We focus on the cases of independence (using the factorization property), and of unknown dependence (using the Frechet bounds), and we show that our approach extends the probabilistic arithmetic of Williamson and Downs. Two design problems-a damped oscillator, and a river dike-demonstrate the practical feasibility of our results.\",\n \"Title: Evaluation of the naive evidential classifier (NEC): a comparison between its two variants based on a real agronomic application\\nYear: 2012\\n\\nAbstract: We introduce the notion of naive evidential classifier. This classifier, which has a structure mirroring the naive Bayes classifier, is based on the Transferable Belief Model and uses mass assignments as its uncertainty model. This new method achieves more robust inferences, mainly by explicitly modeling imprecision when data are in little amount or are imprecise. After introducing the model and its inference process based on Smet's generalized Bayes theorem (GBT), we specify some possible methods to learn its parameters, based on the Imprecise Dirichlet Model (IDM) or on predictive belief functions. Some experimental results on an agronomic application are then given and evaluated.\",\n \"Title: Numerical Sensitivity and Efficiency in the Treatment of Epistemic and Aleatory Uncertainty\\nYear: 2007\\n\\nAbstract: The treatment of both aleatory and epistemic uncertainty by recent methods\\noften requires an high computational effort. In this abstract, we propose a\\nnumerical sampling method allowing to lighten the computational burden of\\ntreating the information by means of so-called fuzzy random variables.\",\n \"Title: A flexible bipolar querying approach with imprecise data and guaranteed results\\nYear: 2011\\n\\nAbstract: In this paper, we propose an approach to query a database when the user preferences are bipolar (i.e., express both constraints and wishes about the desired result) and the data stored in the database are imprecise. Results are then completely ordered with respect to these bipolar preferences, giving priority to constraints over wishes. Additionally, we propose a treatment that allows us to guarantee that any query will return a result, even if no element satisfies all constraints specified by the user. Such a treatment may be useful when user's constraints are unrealistic (i.e., cannot be all satisfied simultaneously) and when the user desires a guaranteed result. The approach is illustrated on a real-world problem concerning the selection of optimal packaging for fresh fruits and vegetables.\",\n \"Title: Decision Support System Using Flexible Query And Reliability Assessment - Application To Biodegradable And Biosourced Packaging Design\\nYear: 2013\\n\\nAbstract: In decision tasks, imprecision and uncertainty can arise from many sources such as data uncertainty, data reliability, or the necessity to use intermediate (non-fully reliable) models. User preferences may also be included into the decision process, making it even more complex. This is particularly true for processes involving biological materials. Soft computing methods have the potential to be the kingpin of specialized software that can be integrated in decision support systems intended to solve the mentioned issues.This work presents three such methods and their implementation: (i)-simulation module with uncertain inputs and interval analysis, (ii)-reliability module allowing to rank candidates according to expert criteria, and (iii)-bipolar flexible querying module, that makes the difference between constraints and wishes in a data base query. A decision support system architecture is proposed and illustrated on a case study, dealing with biodegradable and biosourced packaging design.\",\n \"Title: Idempotent conjunctive combination of belief functions: Extending the minimum rule of possibility theory.\\nYear: 2011\\n\\nAbstract: When conjunctively merging two belief functions concerning a single variable but coming from different sources, Dempster rule of combination is justified only when information sources can be considered as independent. When dependencies between sources are ill-known, it is usual to require the property of idempotence for the merging of belief functions, as this property captures the possible redundancy of dependent sources. To study idempotent merging, different strategies can be followed. One strategy is to rely on idempotent rules used in either more general or more specific frameworks and to study, respectively, their particularization or extension to belief functions. In this paper, we study the feasibility of extending the idempotent fusion rule of possibility theory (the minimum) to belief functions. We first investigate how comparisons of information content, in the form of inclusion and least-commitment, can be exploited to relate idempotent merging in possibility theory to evidence theory. We reach the conclusion that unless we accept the idea that the result of the fusion process can be a family of belief functions, such an extension is not always possible. As handling such families seems impractical, we then turn our attention to a more quantitative criterion and consider those combinations that maximize the expected cardinality of the joint belief functions, among the least committed ones, taking advantage of the fact that the expected cardinality of a belief function only depends on its contour function.\",\n \"Title: Building an interpretable fuzzy rule base from data using Orthogonal Least Squares---Application to a depollution problem\\nYear: 2007\\n\\nAbstract: In many fields where human understanding plays a crucial role, such as bioprocesses, the capacity of extracting knowledge from data is of critical importance. Within this framework, fuzzy learning methods, if properly used, can greatly help human experts. Amongst these methods, the aim of orthogonal transformations, which have been proven to be mathematically robust, is to build rules from a set of training data and to select the most important ones by linear regression or rank revealing techniques. The OLS algorithm is a good representative of those methods. However, it was originally designed so that it only cared about numerical performance. Thus, we propose some modifications of the original method to take interpretability into account. After recalling the original algorithm, this paper presents the changes made to the original method, then discusses some results obtained from benchmark problems. Finally, the algorithm is applied to a real-world fault detection depollution problem.\",\n \"Title: Revisiting the Notion of Conflicting Belief Functions.\\nYear: 2012\\n\\nAbstract: The problem of conflict measurement between information sources knows a regain of interest. In most works related to this issue, Dempter's rule plays a central role. In this paper, we propose to revisit conflict from a different perspective. We do not make a priori assumption about dependencies and start from the definition of conflicting sets, studying its possible extensions to the framework of belief functions.\",\n \"Title: Computing expectations with continuous p-boxes: Univariate case\\nYear: 2009\\n\\nAbstract: Given an imprecise probabilistic model over a continuous space, computing lower/upper expectations is often computationally hard to achieve, even in simple cases. Because expectations are essential in decision making and risk analysis, tractable methods to compute them are crucial in many applications involving imprecise probabilistic models. We concentrate on p-boxes (a simple and popular model), and on the computation of lower expectations of non-monotone functions. This paper is devoted to the univariate case, that is where only one variable has uncertainty. We propose and compare two approaches: the first using general linear programming, and the second using the fact that p-boxes are special cases of random sets. We underline the complementarity of both approaches, as well as the differences.\",\n \"Title: Transforming Probability Intervals into Other Uncertainty Models\\nYear: 2007\\n\\nAbstract: Probability intervals are imprecise probability assignments over elementary events. They constitute a very convenient tool to model uncertain information : two common cases are confidence intervals on parameters of multinomial distributions built from sample data and expert opinions provided in terms of such intervals. In this paper, we study how probability intervals can be transformed into other uncertainty models such as possibility distributions, Ferson's p-boxes, random sets and Neumaier's clouds.\",\n \"Title: A geometric and game-theoretic study of the conjunction of possibility measures.\\nYear: 2015\\n\\nAbstract: We study the conjunction of possibility measures within imprecise probability theory.We identify conditions under which the conjunction remains a possibility measure.We provide a graphical way to check these conditions, using zero-sum games.We provide a graphical way to adjust possibility measures so their conjunction is a possibility measure.We identify conditions under which the conjunction is coherent. In this paper, we study the conjunction of possibility measures when they are interpreted as coherent upper probabilities, that is, as upper bounds for some set of probability measures. We identify conditions under which the minimum of two possibility measures remains a possibility measure. We provide graphical way to check these conditions, by means of a zero-sum game formulation of the problem. This also gives us a nice way to adjust the initial possibility measures so their minimum is guaranteed to be a possibility measure. Finally, we identify conditions under which the minimum of two possibility measures is a coherent upper probability, or in other words, conditions under which the minimum of two possibility measures is an exact upper bound for the intersection of the credal sets of those two possibility measures.\",\n \"Title: Cautious Ordinal Classification by Binary Decomposition.\\nYear: 2014\\n\\nAbstract: We study the problem of performing cautious inferences for an ordinal classification (a.k.a. ordinal regression) task, that is when the possible classes are totally ordered. By cautious inference, we mean that we may produce partial predictions when available information is insufficient to provide reliable precise ones. We do so by estimating probabilistic bounds instead of precise ones. These bounds induce a (convex) set of possible probabilistic models, from which we perform inferences. As the estimates or predictions for such models are usually computationally harder to obtain than for precise ones, we study the extension of two binary decomposition strategies that remain easy to obtain and computationally efficient to manipulate when shifting from precise to bounded estimates. We demonstrate the possible usefulness of such a cautious attitude on tests performed on benchmark data sets.\",\n \"Title: A consistency-specificity trade-off to select source behavior in information fusion.\\nYear: 2015\\n\\nAbstract: Combining pieces of information provided by several sources without or with little prior knowledge about the behavior of the sources is an old yet still important and rather open problem in the belief function theory. In this paper, we propose an approach to select the behavior of sources based on a very general and expressive fusion scheme, that has the important advantage of making clear the assumptions made about the sources. The selection process itself relies on two cornerstones that are the notions of specificity and consistency of a knowledge representation, and that we adapt to the considered fusion scheme. We illustrate our proposal on different examples and show that the proposed approach actually encompasses some important existing fusion strategies.\",\n \"Title: Comments on \\\"A distance-based statistical analysis of fuzzy number-valued data\\\" by the SMIRE research group.\\nYear: 2014\\n\\nAbstract: This paper is a fine review of various aspects related to the statistical handling of “ontic” random fuzzy sets by the means of appropriate distances. It is quite comprehensive and helpful, as it clarifies the status of fuzzy sets in such methods, explains the advantages of using a distance-based approach, specifies the pitfalls in which one should not fall when dealing with “ontic” random fuzzy sets and provides some illustration of practical computations. Not being a statistician but an occasional user of statistics, my discussion will mainly focus on this more practical aspect.\",\n \"Title: Comments on \\\"Learning from imprecise and fuzzy observations: Data disambiguation through generalized loss minimization\\\" by Eyke Hüllermeier.\\nYear: 2014\\n\\nAbstract: Eyke Hüllermeier provides a very convincing approach to learn from fuzzy data, both about the model and about the data themselves. In the process, he links the shape of fuzzy sets with classical loss functions, therefore providing strong theoretical links between fuzzy modeling and more classical machine learning approaches. This short note discusses various aspects of his proposal as well as possible extensions. I will first discuss the opportunity to consider more general uncertainty representations, before considering various alternatives to the proposed learning procedure. Finally, I will briefly discuss the differences I perceive about a loss-based and a likelihood-based approach.\",\n \"Title: Nested Dichotomies with probability sets for multi-class classification.\\nYear: 2014\\n\\nAbstract: Binary decomposition techniques transform a multi-class problem into several simpler binary problems. In such techniques, a classical issue is to ensure the consistency between the binary assessments of conditional probabilities. Nested dichotomies, which consider tree-shaped decomposition, do not suffer from this issue. Yet, a wrong probability estimate in the tree can strongly biase the results and provide wrong predictions. To overcome this issue, we consider in this paper imprecise nested dichotomies, in which binary probabilities become imprecise. We show in experiments that the approach has many advantages: it provides cautious inferences when only little information is available, and allows to make efficient computations with imprecise probabilities even when considering generic cost functions.\",\n \"Title: Kolmogorov-Smirnov Test for Interval Data.\\nYear: 2014\\n\\nAbstract: In this paper, we are interested in extending the classical Kolmogorov-Smirnov homogeneity test to compare two samples of interval-valued observed measurements. In such a case, the test result is interval-valued, and one major difficulty is to find the bounds of this set. We propose a very efficient computational method for approximating these bounds by using a p-box (pairs of upper and lower cumulative distributions) representation of the samples.\",\n \"Title: A Note on Learning Dependence under Severe Uncertainty.\\nYear: 2014\\n\\nAbstract: We propose two models, one continuous and one categorical, to learn about dependence between two random variables, given only limited joint observations, but assuming that the marginals are precisely known. The continuous model focuses on the Gaussian case, while the categorical model is generic. We illustrate the resulting statistical inferences on a simple example concerning the body mass index. Both methods can be extended easily to three or more random variables.\",\n \"Title: Multilabel Prediction with Probability Sets: The Hamming Loss Case.\\nYear: 2014\\n\\nAbstract: In this paper, we study how multilabel predictions can be obtained when our uncertainty is described by a convex set of probabilities. Such predictions, typically consisting of a set of potentially optimal decisions, are hard to make in large decision spaces such as the one considered in multilabel problems. However, we show that when considering the Hamming loss, an approximate prediction can be efficiently computed from label-wise information, as in the precise case. We also perform some first experiments showing the interest of performing partial predictions in the multilabel case.\",\n \"Title: Toward an Axiomatic Definition of Conflict Between Belief Functions\\nYear: 2013\\n\\nAbstract: Recently, the problem of measuring the conflict between two bodies of evidence represented by belief functions has known a regain of interest. In most works related to this issue, Dempster's rule plays a central role. In this paper, we propose to study the notion of conflict from a different perspective. We start by examining consistency and conflict on sets and extract from this settings basic properties that measures of consistency and conflict should have. We then extend this basic scheme to belief functions in different ways. In particular, we do not make any a priori assumption about sources (in)dependence and only consider such assumptions as possible additional information.\",\n \"Title: Fuzzy belief structures viewed as classical belief structures: A practical viewpoint\\nYear: 2010\\n\\nAbstract: Many authors have studied fuzzy belief structures, that is belief functions having fuzzy sets as focal elements. One of the main reason for this is that this structure offers a convenient way to mix probabilistic and fuzzy information. Still, one point on which authors often disagree is how information represented by fuzzy belief structures should be processed, i.e., how should be defined fusion operations, decision rules, uncertainty measures, uncertainty propagation, etc. for such representations. In this paper, we consider that fuzzy belief structures are mapped into classical belief structures encoding the same information, and propose to manipulate these latter structures. From a practical standpoint, it has the benefit that processing tools proper to belief structures can be used together with their interpretation, rather than having to consider a mix of probabilistic and fuzzy calculi, which can be harder to interpret.\",\n \"Title: Modelling and predicting partial orders from pairwise belief functions\\nYear: 2016\\n\\nAbstract: Abstract In this paper, we introduce a generic way to represent and manipulate pairwise information about partial orders (representing rankings, preferences, ...) with belief functions. We provide generic and practical tools to make inferences from this pairwise information and illustrate their use on the machine learning problems that are label ranking and multi-label prediction. Our approach differs from most other quantitative approaches handling complete or partial orders, in the sense that partial orders are here considered as primary objects and not as incomplete specifications of ideal but unknown complete orders.\",\n \"Title: Online active learning of decision trees with evidential data\\nYear: 2016\\n\\nAbstract: Learning from uncertain data has attracted increasing attention in recent years. In this paper, we propose a decision tree learning method that can not only handle uncertain data, but also reduce epistemic uncertainty by querying the most valuable uncertain instances within the learning procedure. Specifically, we use entropy intervals extracted from the evidential likelihood to query uncertain training instances when needed, with the goal to improve the selection of the splitting attribute. Experimental results under various conditions confirm the interest of the proposed approach. HighlightsActive belief decision trees are learned from uncertain data modelled by belief functions.A query strategy is proposed to query the most valuable uncertain instances while learning decision trees.To deal with evidential data, entropy intervals are extracted from the evidential likelihood.Experiments with UCI data illustrate the robustness of proposed approach to various kinds of uncertain data.\",\n \"Title: An extension of the FURIA classification algorithm to low quality data through fuzzy rankings and its application to the early diagnosis of dyslexia.\\nYear: 2016\\n\\nAbstract: An early detection and reeducation of dyslexic children is critical for their integration in the classroom. Parents and instructors can help the psychologist to detect potential cases of dyslexia before the children׳s writing age. Artificial intelligence tools can also assist in this task. Dyslexia symptoms are detected with tests whose results may be vague or ambiguous, thus machine learning techniques for low quality data are advised. In particular, in this paper it is suggested that a new extension to vague datasets of the classification algorithm FURIA (Fuzzy Unordered Rule Induction Algorithm) has advantages over other approaches in both the computational effort during the learning stage and the linguistic quality of the induced classification rules. The new approach is benchmarked with different test problems and compared to other artificial intelligence tools for dyslexia diagnosis in the literature.\",\n \"Title: Cautious label ranking with label-wise decomposition.\\nYear: 2015\\n\\nAbstract: •We propose to rank labels using label-wise information.•We consider the use of probability sets over ranks to produce cautious predictions.•We approximate the solution of the corresponding bilinear problem.•We perform some first experiment showing the potential benefits of the approach.\",\n \"Title: Ranking of fuzzy intervals seen through the imprecise probabilistic lens\\nYear: 2015\\n\\nAbstract: Within the fuzzy literature, the issue of ranking fuzzy intervals has been addressed by many authors, who proposed various solutions to the problem. Most of these solutions intend to find a total order on a given collection of fuzzy intervals. However, if one sees fuzzy intervals as descriptions of uncertain quantities, an alternative to rank them is to use ranking rules issued from the imprecise probabilistic literature. In this paper, we investigate ranking rules based on different statistical features (mean, median) and orderings, and relate the obtained (partial) orders to some classical proposals. In particular, we propose a generic expression of stochastic orderings, and then use it to systematically investigate extensions of the most usual stochastic orderings to fuzzy intervals. We also show some relations between those extensions, and explore their relation with existing fuzzy ranking proposals.\",\n \"Title: A Decision Support System to design modified atmosphere packaging for fresh produce based on a bipolar flexible querying approach\\nYear: 2015\\n\\nAbstract: We define a multi-criteria Decision Support System for designing fresh food packaging.A modified atmosphere packaging simulation module is included in the DSS.A flexible querying module handles imprecise data stored in a packaging database.Using the DSS the user will have only one trial to perform validation step. To design new packaging for fresh food, stakeholders of the food chain express their needs and requirements, according to some goals and objectives. These requirements can be gathered into two groups: (i) fresh food related characteristics and (ii) packaging intrinsic characteristics. Modified Atmosphere Packaging (MAP) is an efficient way to delay senescence and spoilage and thus to extend the very short shelf life of respiring products such as fresh fruits and vegetables. Consequently, packaging O2/CO2 permeabilities must fit the requirements of fresh fruits and vegetable as predicted by virtual MAP simulating tools. Beyond gas permeabilities, the choice of a packaging material for fresh produce includes numerous other factors such as the cost, availability, potential contaminants of raw materials, process ability, and waste management constraints. For instance, the user may have the following multi-criteria query for his/her product asking for a packaging with optimal gas permeabilities that guarantee product quality and optionally a transparent packaging material made from renewable resources with a cost for raw material less than 3 /kg. To help stakeholders taking a rational decision based on the expressed needs, a new multi-criteria Decision Support System (DSS) for designing biodegradable packaging for fresh produce has been built. In this paper we present the functional specification, the software architecture and the implementation of the developed tool. This tool includes (i) a MAP simulation module combining mass transfer models and respiration of the food, (ii) a multi-criteria flexible querying module which handles imprecise, uncertain and missing data stored in the database. We detail its operational functioning through a real life case study to determine the most satisfactory materials for apricots packaging.\",\n \"Title: Multilabel predictions with sets of probabilities: The Hamming and ranking loss cases\\nYear: 2015\\n\\nAbstract: In this paper, we study how multilabel predictions can be obtained when our uncertainty is described by a convex set of probabilities. Such predictions, typically consisting of a set of potentially optimal decisions, are hard to make in large decision spaces such as the one considered in multilabel problems. However, we show that when considering the Hamming or the ranking loss, outer-approximating predictions can be efficiently computed from label-wise information, as in the precise case. We also perform some first experiments showing the behaviour of the partial predictions obtained through these approximations. Such experiments also confirm that predictions become partial on those labels where the precise prediction is likely to make an error.\",\n \"Title: Elicitation of a Utility from Uncertainty Equivalent Without Standard Gambles\\nYear: 2015\\n\\nAbstract: In the context of decision under uncertainty, standard gambles are classically used to elicit a utility function on a set X of consequences. The utility of an element x in X is derived from the probability p for which a gamble giving the best outcome in X with probability p and the worst outcome in X otherwise, is indifferent to getting x for sure. In many situations, uncertainty that can be observed on the true value of X concerns only neighbour values. Uncertainty is then represented by a probability distribution whose support is an interval. In this case, standard gambles are unrealistic for the decision maker. We consider uncertainty represented by an equi-probability over an interval of X. This paper addresses the elicitation of a utility function on X by obtaining the certainty equivalent of an equi-probability over an interval of X. We show that not all utility models are suitable to accomplish this task.\",\n \"Title: Optimal expert elicitation to reduce interval uncertainty.\\nYear: 2015\\n\\nAbstract: Reducing uncertainty is an important problem in many applications such as risk and reliability analysis, system design, etc. In this paper, we study the problem of optimally querying experts to reduce interval uncertainty. Surprisingly, this problem has received little attention in the past, while similar issues in preference elicitation or social choice theory have witnessed a rising interest. We propose and discuss some solutions to determine optimal questions in a myopic way (one-at-a-time), and study the computational aspects of these solutions both in general and for some specific functions of practical interest. Finally, we illustrate the application of the approach in reliability analysis problems.\",\n \"Title: Comparing System Reliabilities with Ill-Known Probabilities.\\nYear: 2016\\n\\nAbstract: In reliability analysis, comparing system reliability is an essential task when designing safe systems. When the failure probabilities of the system components (assumed to be independent) are precisely known, this task is relatively simple to achieve, as system reliabilities are precise numbers. When failure probabilities are ill-known (known to lie in an interval) and we want to have guaranteed comparisons (i.e., declare a system more reliable than another when it is for any possible probability value), there are different ways to compare system reliabilities. We explore the computational problems posed by such extensions, providing first insights about their pros and cons.\",\n \"Title: Easy and optimal queries to reduce set uncertainty.\\nYear: 2017\\n\\nAbstract: •A setting for optimal expert elicitation to reduce set uncertainty is described.•The formal approach is justified by simulated experiments in reliability.•The approach proves to be efficient and computationally tractable.•Computations are particularly simple for specific functions of practical interest.\",\n \"Title: A decision support system for eco-efficient biorefinery process comparison using a semantic approach.\\nYear: 2016\\n\\nAbstract: •We define a decision support system for eco-efficient biorefinery process comparison.•Uncertainty is managed all along the pipeline for eco-design indicator computations.•Data extraction from textual documents is guided by a Termino-ontological resource.•Flexible structured database querying is guided by a Termino-ontological resource.\",\n \"Title: Idempotent Conjunctive Combination of Belief Functions by Distance Minimization.\\nYear: 2016\\n\\nAbstract: When combining multiple belief functions, designing a combination rule that selects the least informative belief function among those more informative than each of the combined ones is a difficult task. Such rules, commonly depicted as \\\"cautious\\\", are typically required to be idempotent, since when one is cautious, combining identical information should not lead to the reinforcement of some hypothesis. However, applying the least commitment principle using partial orders is in general not straightforward, mainly due to the non-uniqueness of solutions. Building upon previous work, this paper investigates the use of distances compatible with such partial orders to determine a unique solution to the combination problem. The obtained operators are conjunctive, idempotent and commutative, but lack associativity. They are, however, quasi-associative allowing sequential combinations at no extra complexity.\",\n \"Title: Partial Data Querying Through Racing Algorithms.\\nYear: 2018\\n\\nAbstract: The paper studies the problem of actively learning from instances characterized by imprecise features or imprecise class labels, where by actively learning we understand the possibility to query the precise value of imprecisely specified data. We differ from classical active learning by the fact that in the later, data are either fully precise or completely missing, while in our case they can be partially specified. Such situations can appear when sensor errors are important to encode, or when experts have only specified a subset of possible labels when tagging data. We provide a general active learning technique that can be applied in principle to any model. It is inspired from racing algorithms, in which several models are competing against each others. The main idea of our method is to identify the query that will be the most helpful in identifying the winning model in the competition. After discussing and formalizing the general ideas of our approach, we illustrate it by studying the particular case of binary SVM in the case of interval valued features and set-valued labels. The experimental results indicate that, in comparison to other baselines, racing algorithms provide a faster reduction of the uncertainty in the learning process, especially in the case of imprecise features.\",\n \"Title: Unifying parameter learning and modelling complex systems with epistemic uncertainty using probability interval.\\nYear: 2016\\n\\nAbstract: Knowledge regarding complex systems are heterogeneous and fragmented.modelling dynamic complex systems in the framework of dynamic credal networks.practical methodology coupling Dirichlet distributions with interval probabilities to incrementally build and update model parameters whatever source and format of knowledge.enables to take into account (1) stochastic and epistemic uncertainties pertaining to the system; (2) the confidence level on the different sources of information.illustrate the application of the methodology to the modelling of a simplified industrial case study. Modeling complex dynamical systems from heterogeneous pieces of knowledge varying in precision and reliability is a challenging task. We propose the combination of dynamical Bayesian networks and of imprecise probabilities to solve it. In order to limit the computational burden and to make interpretation easier, we also propose to encode pieces of (numerical) knowledge as probability intervals, which are then used in an imprecise Dirichlet model to update our knowledge. The idea is to obtain a model flexible enough so that it can easily cope with different uncertainties (i.e., stochastic and epistemic), integrate new pieces of knowledge as they arrive and be of limited computational complexity.\",\n \"Title: Querying Partially Labelled Data to Improve a K-nn Classifier.\\nYear: 2017\\n\\nAbstract: When learning from instances whose output labels may be partial, the problem of knowing which of these output labels should be made precise to improve the accuracy of predictions arises. This problem can be seen as the intersection of two tasks: the one of learning from partial labels and the one of active learning, where the goal is to provide the labels of additional instances to improve the model accuracy. In this paper, we propose querying strategies of partial labels for the well-known K-nn classifier. We propose different criteria of increasing complexity, using among other things the amount of ambiguity that partial labels introduce in the K-nn decision rule. We then show that our strategies usually outperform simple baseline schemes, and that more complex strategies provide a faster improvement of the model accuracies.\",\n \"Title: K-Nearest Neighbour Classification for Interval-Valued Data.\\nYear: 2017\\n\\nAbstract: This paper studies the problem of providing predictions with a K-nn approach when data have partial features given in the form of intervals. To do so, we adopt an optimistic approach to replace the ill-known values, that requires to compute sets of possible and necessary neighbours of an instance. We provide an easy way to compute such sets, as well as the decision rule that follows from them. Our approach is then compared to a simple imputation method in different scenarios, in order to identify those ones where it is advantageous.\",\n \"Title: Reasons and Means to Model Preferences as Incomplete.\\nYear: 2018\\n\\nAbstract: Literature involving preferences of artificial agents or human beings often assume their preferences can be represented using a complete transitive binary relation. Much has been written however on different models of preferences. We review some of the reasons that have been put forward to justify more complex modeling, and review some of the techniques that have been proposed to obtain models of such preferences.\",\n \"Title: On extreme points of p-boxes and belief functions.\\nYear: 2016\\n\\nAbstract: Within imprecise probability theory, the extreme points of convex probability sets have an important practical role (to perform inference on graphical models, to compute expectation bounds, …). This is especially true for sets presenting specific features that make them easy to manipulate in applications. This easiness is the reason why extreme points of such models (probability intervals, possibility distributions, …) have been well studied. Yet, imprecise cumulative distributions (a.k.a. p-boxes) constitute an important exception, as the characterization of their extreme points remain to be studied. This is what we do in this paper, where we characterize the maximal number of extreme points of a p-box, give a family of p-boxes that attains this number and show an algorithm that allows to compute the extreme points of a given p-box. To achieve all this, we also provide what we think to be a new characterization of extreme points of a belief function.\",\n \"Title: On the median in imprecise ordinal problems.\\nYear: 2017\\n\\nAbstract: When having to make a prediction under probabilistic uncertainty in ordinal problems, the median offers a number of interesting properties compared to other statistics such as the expected value. In particular, it does not depend on a particular metric defined over the elements, but still takes account of the ordinal nature of the data. It can also be shown to be the minimizer of the (L_1) loss function. In this paper, we show that similar results can be obtained when the uncertainty is described not by a single probability distribution, but by a convex set of those. In particular, we relate the lower and upper medians to the (L_1) loss function via the notion of lower and upper expectations (and extend these results to general quantiles). We also show that, using a different decision rule, the lower and upper median can be retrieved when assuming the cost to be strictly monotonic and symmetric, and nothing more. Finally, we run some tests to show the interest of using Median based predictions with convex sets of probabilities in ordinal regression problems.\",\n \"Title: Comonotonicity for sets of probabilities.\\nYear: 2017\\n\\nAbstract: Two variables are called comonotone when there is an increasing relation between them, in the sense that when one of them increases (decreases), so does the other one. This notion has been widely investigated in probability theory, and is related to copulas. This contribution studies how the notion of comonotonicity can be extended to an imprecise setting on discrete spaces, where probabilities are only known to belong to a convex set. We define comonotonicity for such sets and investigate its characterizations in terms of lower probabilities, as well as its connection with copulas. As this theoretical characterization can be tricky to apply to general lower probabilities, we also investigate specific models of practical importance. In particular, we provide some sufficient conditions for a comonotone belief function with fixed marginals to exist, and characterize comonotone bivariate p-boxes.\",\n \"Title: The Costs of Indeterminacy: How to Determine Them?\\nYear: 2017\\n\\nAbstract: Indeterminate classifiers are cautious models able to predict more than one class in case of high uncertainty. A problem that arises when using such classifiers is how to evaluate their performances. This problem has already been considered in the case where all prediction errors have equivalent costs (that we will refer as the “0/1 costs” or accuracy setting). The purpose of this paper is to stud...\",\n \"Title: Cautious classification with nested dichotomies and imprecise probabilities.\\nYear: 2017\\n\\nAbstract: In some applications of machine learning and information retrieval (e.g. medical diagnosis, image recognition, pre-classification...), it can be preferable to provide less informative but more reliable predictions. This can be done by making partial predictions in the form of class subsets when the available information is insufficient to provide a reliable unique class. Imprecise probabilistic approaches offer nice tools to learn models from which such cautious predictions can be produced. However, the learning and inference processes of such models are computationally harder than their precise counterparts. In this paper, we introduce and study a particular binary decomposition strategy, nested dichotomies, that offer computational advantages in both the learning (due to the binarization process) and the inference (due to the decomposition strategy) processes. We show with experiments that these computational advantages do not lower the performances of the classifiers, and can even improve them when the class space has some structure.\",\n \"Title: Idempotent conjunctive and disjunctive combination of belief functions by distance minimization.\\nYear: 2018\\n\\nAbstract: Idempotence is a desirable property when cautiousness is wanted in an information fusion process, since in this case combining identical information should not lead to the reinforcement of some hypothesis. Idempotent operators also guarantee that identical information items are not counted twice in the fusion process, a very important property in decentralized applications where the information origin cannot always be tracked (ad-hoc wireless networks are typical examples). In the theory of belief functions, a sound way to combine conjunctively multiple information items is to design a combination rule that selects the least informative element among a subset of belief functions more informative than each of the combined ones. In contrast, disjunctive rules can be retrieved by selecting the most informative element among a subset of belief functions less informative than each of the combined ones. One interest of such approaches is that they provide idempotent rules by construction.\",\n \"Title: Classification by pairwise coupling of imprecise probabilities.\\nYear: 2018\\n\\nAbstract: •We address combining imprecise classifiers in the framework of imprecise probabilities.•Classifier outputs are probability intervals, which are large for irrelevant classifiers.•The combination consists in computing probabilities consistent with the intervals.•This approach is particularly interesting when combining few classifiers.\",\n \"Title: Density Estimation with Imprecise Kernels: Application to Classification.\\nYear: 2018\\n\\nAbstract: In this paper, we explore the problem of estimating lower and upper densities from imprecisely defined families of parametric kernels. Such estimations allow to rely on a single bandwidth value, and we show that it provides good results on classification tasks when extending the naive Bayesian classifier.\",\n \"Title: Technical Gestures Recognition by Set-Valued Hidden Markov Models with Prior Knowledge.\\nYear: 2016\\n\\nAbstract: Hidden Markov models are popular tools for gesture recognition. Once the generative processes of gestures have been identified, an observation sequence is usually classified as the gesture having the highest likelihood, thus ignoring possible prior information. In this paper, we consider two potential improvements of such methods: the inclusion of prior information, and the possibility of considering convex sets of probabilities (in the likelihoods and the prior) to infer imprecise, but more reliable, predictions when information is insufficient. We apply the proposed approach to technical gestures, typically characterized by severe class imbalance. By modelling such imbalances as a prior information, we achieve more accurate results, while the imprecise quantification is shown to produce more reliable estimates.\",\n \"Title: From Relations Between Sets to Relations Between Belief Functions.\\nYear: 2018\\n\\nAbstract: In uncertainty theories, a common problem is to define how we can extend relations between sets (e.g., inclusion, ranking, consistency,...) to corresponding notions between uncertainty representations. Such definitions can then be used to perform the same operations as those that are done for sets: measuring information content, ordering alternatives or checking consistency, to name a few. In this paper, we propose a general way to extend set relations to belief functions, using constrained stochastic matrices to identify those belief functions in relation. We then study some properties of our proposal, as well as its relations with existing works focusing on peculiar relations.\",\n \"Title: Pari-mutuel probabilities as an uncertainty model.\\nYear: 2019\\n\\nAbstract: •We prove some properties of the Pari Mutuel Model as a coherent lower (and upper) probability.•We show that a Pari Mutuel Model is a particular case of a probability interval.•We give the maximal number of extreme points of the credal set associated with a Pari Mutuel Model and we give the form of those extreme points.•We give sufficient and necessary conditions for the Pari Mutuel Model to induce a belief function.•We show how to merge information given in terms of Pari Mutuel Models.\",\n \"Title: A decision support system using multi-source scientific data, an ontological approach and soft computing - application to eco-efficient biorefinery\\nYear: 2016\\n\\nAbstract: In decision tasks such as bioprocess efficiency comparison, scientific literature is a valuable source of data. This large number of scientific data is heterogeneously structured, mainly in textual format. Innovative tools able to integrate and treat constantly new information are required. In this context, the use of semantic web methods such as ontologies seems relevant to structure the experimental information. Imprecision and uncertainty can arise from data incompleteness and variability. This is particularly true for processes involving biological materials. Document reliability should also be considered. Soft computing methods have the potential to be the kingpin of specialized software that can be integrated in decision support systems (DSS) intended to solve these issues. This paper presents the implementation of a pipeline which permits to: (1) structure and integrate the experimental data of interest by using ontologies, (2) assess data source reliability, (3) compute and visualize indicators taking into account data imprecision.\",\n \"Title: From set relations to belief function relations\\nYear: 2019\\n\\nAbstract: In uncertainty theories, a common problem is to define how we can extend relations between sets (e.g., inclusion, ranking, consistency, …) to corresponding notions between uncertainty representations. Such definitions can then be used to perform the same operations as those that are done for sets: comparing information content, ordering alternatives or checking consistency, to name a few. In this paper, we propose a general way to extend set relations to belief functions, using constrained stochastic matrices to identify those belief functions in relation. We then study some properties of our proposal, as well as its connections with existing works focusing on specific relations.\",\n \"Title: Imprecise Gaussian discriminant classification\\nYear: 2019\\n\\nAbstract: •We robustify Gaussian discriminant analysis by considering sets of estimates.•We use near-ignorance priors to derive bounding boxes on mean estimates.•We discuss the computational issue for generic and diagonal covariance matrices.•We make a full experimental study showing the benefits of using imprecise estimates.•We make a first exploration of the benefits of using the model in non i.i.d. situations.\",\n \"Title: Racing Trees To Query Partial Data\\nYear: 2021\\n\\nAbstract: Dealing with partially known or missing data is a common problem in machine learning. This work is interested in the problem of querying the true value of data to improve the quality of the learned model, when those data are only partially known. This study is in the line of active learning, since we consider that the precise value of some partial data can be queried to reduce the uncertainty in the learning process, yet can consider any kind of partial data (not only entirely missing one). We propose a querying strategy based on the concept of racing algorithms in which several models are competing. The idea is to identify the query that will help the most to quickly decide the winning model in the competition. After discussing and formalizing the general ideas of our approach, we study the particular case of decision trees in case of interval-valued features and set-valued labels. The experimental results indicate that, in comparison with other baselines, the proposed approach significantly outperforms simpler strategies in the case of partially specified features, while it achieves similar performances in the case of partially specified labels.\",\n \"Title: Multi-label Chaining with Imprecise Probabilities\\nYear: 2021\\n\\nAbstract: We present two different strategies to extend the classical multi-label chaining approach to handle imprecise probability estimates. These estimates use convex sets of distributions (or credal sets) in order to describe our uncertainty rather than a precise one. The main reasons one could have for using such estimations are (1) to make cautious predictions (or no decision at all) when a high uncertainty is detected in the chaining and (2) to make better precise predictions by avoiding biases caused in early decisions in the chaining. We adapt both strategies to the case of the naive credal classifier, showing that this adaptations are computationally efficient. Our experimental results on missing labels, which investigate how reliable these predictions are in both approaches, indicate that our approaches produce relevant cautiousness on those hard-to-predict instances where the precise models fail.\",\n \"Title: Inference And Decision In Credal Occupancy Grids: Use Case On Trajectory Planning\\nYear: 2021\\n\\nAbstract: Occupancy grids are common tools used in robotics to represent the robot environment, and that may be used to plan trajectories, select additional measurements to acquire, etc. However, deriving information about those occupancy grids from sensor measurements often induce a lot of uncertainty, especially for grid elements that correspond to occluded or far away area from the robot. This means that occupancy information may be quite uncertain and imprecise at some places, while being very accurate at others. Modelling finely this occupancy information is essential to decide the optimal action the robot should take, but a refined modelling of uncertainty often implies a higher computational cost, a prohibitive feature for real-time applications. In this paper, we introduce the notion of credal occupancy grids, using the very general theory of imprecise probabilities to model occupancy uncertainty. We also show how one can perform efficient, real-time inferences with such a model, and show a use-case applying the model to an autonomous vehicle trajectory planning problem.\",\n \"Title: Processing Multiple Distortion Models: a Comparative Study\\nYear: 2021\\n\\nAbstract: When dealing with uncertain information, distortion or neighbourhood models are convenient practical tools, as they rely on very few parameters. In this paper, we study their behaviour when such models are combined and processed. More specifically, we study their behaviour when merging different distortion models quantifying uncertainty on the same quantity, and when manipulating distortion models defined over multiple variables.\",\n \"Title: Logical and Evidential Inconsistencies Meet: First Steps\\nYear: 2021\\n\\nAbstract: Measuring inconsistency has been and is still an active research topic in both logic and evidence theory. However, the two fields have developed distinct notions and measures of inconsistency, following different paths. In this paper, we attempt to build some first bridges between the two trends, suggesting some first means for one to enrich the other, and vice-versa.\",\n \"Title: A Note About Entropy and Inconsistency in Evidence Theory\\nYear: 2021\\n\\nAbstract: Information content is classically measured by entropy measures in probability theory, that can be interpreted as a measure of internal inconsistency of a probability distribution. While extensions of Shannon entropy have been proposed for quantifying information content of a belief function, other trends have been followed which rather focus on the notion of consistency between sets. Relying on previous general entropy measures of probability, we propose in this paper to establish some links between the different measures of internal inconsistency of a belief functions. We propose a general formulation which encompasses inconsistency measures derived from Shannon entropy as well as those derived from the N-consistency family of measures.\",\n \"Title: Possibilistic preference elicitation by minimax regret.\\nYear: 2021\\n\\nAbstract: Identifying the preferences of a given user through elicitation is a central part of multi-criteria decision aid (MCDA) or preference learning tasks. Two classical ways to perform this elicitation is to use either a robust or a Bayesian approach. However, both have their shortcoming: the robust approach has strong guarantees through very strong hypotheses, but cannot integrate uncertain information. While the Bayesian approach can integrate uncertainties, but sacrifices the previous guarantees and asks for stronger model assumptions. In this paper, we propose and test a method based on possibility theory, which keeps the guarantees of the robust approach without needing its strong hypotheses. Among other things, we show that it can detect user errors as well as model misspecification.\",\n \"Title: Quantification of Credal Uncertainty in Machine Learning: A Critical Analysis and Empirical Comparison.\\nYear: 2022\\n\\nAbstract: The representation and quantification of uncertainty has received increasing attention in machine learning in the recent past. The formalism of credal sets provides an interesting alternative in this regard, especially as it combines the representation of epistemic (lack of knowledge) and aleatoric (statistical) uncertainty in a rather natural way. In this paper, we elaborate on uncertainty measures for credal sets from the perspective of machine learning. More specifically, we provide an overview of proposals, discuss existing measures in a critical way, and also propose a new measure that is more tailored to the machine learning setting. Based on an experimental study, we conclude that theoretically well-justified measures also lead to better performance in practice. Besides, we corroborate the difficulty of the disaggregation problem, that is, of decomposing the amount of total uncertainty into aleatoric and epistemic uncertainty in a sound manner, thereby complementing theoretical findings with empirical evidence.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":86,"string":"86"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13580,"string":"13,580"}}},{"rowIdx":13581,"cells":{"question_id":{"kind":"string","value":"Mihai Datcu"},"question":{"kind":"string","value":"Generate an abstract for the title \"Knowledge-driven image mining system for Big Earth Observation data fusion: GIS maps inclusion in active learning stage\" using the following items: 1. Knowledge-driven system \n2. Big Earth Observation data fusion \n3. Unsupervised clustering \n4. GIS data integration \n5. Computational optimization"},"answer":{"kind":"string","value":"In this paper, we present an accelerated knowledge-driven content-based information mining system for Big Earth Observation data fusion. The tool combines, at pixel level, the unsupervised clustering results of different number of features. The features, extracted from different EO raster image types and from existing GIS vector maps, are combined, in form of a BoW, with a user given semantic concepts in order to calculate the posterior probability that allows the final search. The inclusion of GIS data during the active learning, based on Bayesian networks, accelerate the definition processes of semantic labels and retrieve the related images with only a few user interactions. The inclusion of GIS data in conjunction with the recently introduced search algorithm have as a result a system which greatly optimizes the computational costs and over performs existing similar systems in various orders of magnitude."},"correct_answer":{"kind":"string","value":"In this paper, we present an accelerated knowledge-driven content-based information mining system for Big Earth Observation data fusion. The tool combines, at pixel level, the unsupervised clustering results of different number of features. The features, extracted from different EO raster image types and from existing GIS vector maps, are combined, in form of a BoW, with a user given semantic concepts in order to calculate the posterior probability that allows the final search. The inclusion of GIS data during the active learning, based on Bayesian networks, accelerate the definition processes of semantic labels and retrieve the related images with only a few user interactions. The inclusion of GIS data in conjunction with the recently introduced search algorithm have as a result a system which greatly optimizes the computational costs and over performs existing similar systems in various orders of magnitude."},"documents":{"kind":"list like","value":["Title: Algorithmic Relative Complexity.\nYear: 2011\n\nAbstract: Information content and compression are tightly related concepts that can be addressed through both classical and algorithmic information theories, on the basis of Shannon entropy and Kolmogorov complexity, respectively. The definition of several entities in Kolmogorov's framework relies upon ideas from classical information theory, and these two approaches share many common traits. In this work, we expand the relations between these two frameworks by introducing algorithmic cross-complexity and relative complexity, counterparts of the cross-entropy and relative entropy (or Kullback-Leibler divergence) found in Shannon's framework. We define the cross-complexity of an object x with respect to another object y as the amount of computational resources needed to specify x in terms of y, and the complexity of x related to y as the compression power which is lost when adopting such a description for x, compared to the shortest representation of x. Properties of analogous quantities in classical information theory hold for these new concepts. As these notions are incomputable, a suitable approximation based upon data compression is derived to enable the application to real data, yielding a divergence measure applicable to any pair of strings. Example applications are outlined, involving authorship attribution and satellite image classification, as well as a comparison to similar established techniques.","Title: Interactive learning and probabilistic retrieval in remote sensing image archives\nYear: 2000\n\nAbstract: We present a concept of interactive learning and probabilistic retrieval of user-specific cover types in a content-based remote sensing image archive. A cover type is incrementally defined via user-provided positive and negative examples. From these examples, we infer probabilities of the Bayesian network that link the user interests to a pre-extracted content index. Due to the stochastic nature of the cover type def- initions, the database system not only retrieves images according to the estimated coverage but also according to the accuracy of that estimation given the current state of learning. For the latter, we introduce the concept of separability. We expand on the steps of Bayesian inference to compute the application-free content index using a family of data models, and on the description of the stochastic link using hyperparameters. In particular, we focus on the interactive nature of our approach, which provides instantaneous feedback to the user in the form of an immediate update of the posterior map, and a very fast, approximate search in the archive. A java-based demonstrator using the presented concept of content-based access to a test archive of Landsat TM, X-SAR, and aerial images are available over the Internet (http://www.vision.ee.ethz.ch/~rsia/ClickBayes).","Title: A semantic framework for data retrieval in large remote sensing databases.\nYear: 2012\n\nAbstract: Remote sensing platforms acquire huge amounts of data every day. As a result, large archives of data have been created. In order to provide access to this data efficient search and retrieval methods have to be developed, such as image information mining systems. In this paper we present a framework for an image information mining system using the Latent Dirichlet Allocation text-mining algorithm to provide a high-level semantic model of data, the search being performed in the LDA model space.","Title: Parameter-free Clustering: Application to Fawns Detection\nYear: 2009\n\nAbstract: Many fawns and other wild animals are killed by mowing machines every year. To prevent them from being killed or injured, a sensor system is being developed to detect the fawns hidden in meadows under mowing. Beside a microwave radar system, two cameras (thermal Infrared and RGB) take a picture at the mower's current location. This contribution focuses on the compression-based algorithm that will be adopted to detect the locations containing a fawn hiding in the grass. such approach, being parameter-free, allows performing a fully unsupervised clustering by exploiting the intrinsic properties of data compression to estimate the amount of shared information between two images.","Title: SAR Image Categorization With Log Cumulants of the Fractional Fourier Transform Coefficients.\nYear: 2013\n\nAbstract: With the advent of high-resolution (HR) synthetic aperture radar (SAR) images from satellites like TerraSAR-X and TanDEM-X, interest is now on patch-oriented image categorization in contrast to the pixel-based classification in low-resolution SAR images. SAR image categorization requires the generation of a compact feature descriptor that accurately defines the content of the image patch under consideration. As phase information plays a critical role in SAR images, this paper proposes the use of a chirplet-derived transform, i.e., the fractional Fourier transform (FrFT), for generating a compact feature descriptor for single-look complex (SLC) SAR images. Representing a SAR signal in rotated joint time-frequency planes via the FrFT allows discovering the underlying backscattering phenomenon of the objects on the ground. SAR image projections on different planes of the joint time-frequency space using the FrFT provide a simple statistical response that is easier to analyze. The proposed method has been compared with a multiscale approach, i.e., Gabor filter banks, a second-order-statistics-based method (as gray-level co-occurrence matrices), and a spectral descriptor method. We demonstrate the suitability of the FrFT-based method for image categorization on the basis of backscattering behavior, whereas the Gabor-filter-bank-based method is found mainly suitable for images with a strong texture. This paper demonstrates enhancement in the separability for most of the considered categories when using logarithmic cumulants instead of linear moments for both the FrFT-based and Gabor-filter-bank-based methods. The experimental database consists of 2000 image patches (of size 200 x 200 pixels) extracted from SLC HR TerraSAR-X scenes.","Title: Identification and characterization of railway trains in high resolution TerraSAR-X images.\nYear: 2012\n\nAbstract: The recognition and analysis of moving targets in SAR images is a well-known topic in remote sensing. The public availability of high resolution SAR satellite images opened new perspectives for space-based small object recognition: traffic monitoring from space has matured into an established technique where the motion characteristics of ships and cars can be extracted routinely.A particular case of traffic monitoring is the search for moving trains in SAR images; here one can locate moving trains, determine their velocity, and identify the type and number of wagons. In the following, we concentrate on features to be extracted from TerraSAR-X images containing commuter trains in the area of Munich, Germany.","Title: A Latent Analysis of Earth Surface Dynamic Evolution Using Change Map Time Series.\nYear: 2013\n\nAbstract: With a continuous increase in the number of Earth Observation satellites, leading to the development of satellite image time series (SITS), the number of algorithms for land cover analysis and monitoring has greatly expanded. This paper offers a new perspective in dynamic classification for SITS. Four similarity measures (correlation coefficient, Kullback-Leibler divergence, conditional information, and normalized compression distance) based on consecutive image pairs from the data are employed. These measures employ linear dependences, statistical measures, and spatial relationships to compute radiometric, spectral, and texture changes that offer a description for the multitemporal behavior of the SITS. During this process, the original SITS is converted to a change map time series (CMTS), which removes the static information from the data set. The CMTS is analyzed using a latent Dirichlet allocation (LDA) model capable of discovering classes with semantic meaning based on the latent information hidden in the scene. This statistical method was originally used for text classification, thus requiring a word, document, corpus analogy with the elements inside the image. The experimental results were computed using 11 Landsat images over the city of Bucharest and surrounding areas. The LDA model enables us to discover a wide range of scene evolution classes based on the various dynamic behaviors of the land cover. The results are compared with the Corinne Land Cover map. However, this is not a validation method but one that adds static knowledge about the general usage of the analyzed area. In order to help the interpretation of the results, we use several studies on forms of relief, weather forecast, and very high resolution images that can explain the wide range of structures responsible for influencing the dynamic inside the resolution cell.","Title: Linear versus non-linear analysis of relevant scatterers in high resolution SAR images.\nYear: 2007\n\nAbstract: With the increase of Synthetic Aperture Radar (SAR) sensor resolution, SAR images could include a large variety of interesting real man-made structures. Therefore, a more detailed analysis and a finer description of SAR images of urban areas are needed for a better understanding of the scene.Nevertheless, recognizing scenes using high resolution SAR images requires the capability to identify relevant signal signatures (called also descriptors), depending on variable image acquisition geometry, arbitrary objects poses and configurations. Among feature extraction methods, we propose to use Principal Components Analysis (PCA) and/or Independent Components Analysis (ICA), in order to exploit deeper the nature of SAR signatures. In this paper, both a description of our work and a presentation of our preliminary classification performance results will be provided.","Title: Evaluation of Bayesian Despeckling and Texture Extraction Methods Based on Gauss-Markov and Auto-Binomial Gibbs Random Fields: Application to TerraSAR-X Data.\nYear: 2012\n\nAbstract: Speckle hinders information in synthetic aperture radar (SAR) images and makes automatic information extraction very difficult. The Bayesian approach allows us to perform the despeckling of an image while preserving its texture and structures. This model-based approach relies on a prior model of the scene. This paper presents an evaluation of two despeckling and texture extraction model-based meth...","Title: Multiscale and Dimensionality Behavior of ICA Components for Satellite Image Indexing\nYear: 2010\n\nAbstract: In this letter, our main objective is to demonstrate the capabilities of independent component analysis (ICA) for very high resolution satellite image characterization, particularly for geometrical structures contained in images from urban areas in comparison with natural landscapes. We propose a model based on ICA sources to index satellite images with a resolution of 0.6 m. An important part of this work is to study the effects of multiscaling and dimensionality behavior of ICA components. For evaluating the capabilities of our model, we design a classification method to distinguish images froma variety of natural and man-made scenes.","Title: A remote sensing image processing framework for damage assessment in a forest fire scenario.\nYear: 2012\n\nAbstract: In natural hazards management applications Earth Observation (EO) image processing methods are based on segmentation and classification. The result primary consists of thematic maps which are readily interpretable. We propose a complete EO image processing chain, which generates an end product with increased information content organized in thematic layers. The processing chain integrates four main components: image classification, identification of high anomaly areas relative to the entire scene context, spectral and texture change detection, and the integration of different information layers. The processing chain was tested in a fire management scenario, using a pair of Landsat5-TM images for the Pagami Creek forest fire which was active from August to October 2011.","Title: Data vaults: a database welcome to scientific file repositories\nYear: 2013\n\nAbstract: Efficient management and exploration of high-volume scientific file repositories have become pivotal for advancement in science. We propose to demonstrate the Data Vault, an extension of the database system architecture that transparently opens scientific file repositories for efficient in-database processing and exploration. The Data Vault facilitates science data analysis using high-level declarative languages, such as the traditional SQL and the novel array-oriented SciQL. Data of interest are loaded from the attached repository in a just-in-time manner without need for up-front data ingestion. The demo is built around concrete implementations of the Data Vault for two scientific use cases: seismic time series and Earth observation images. The seismic Data Vault uses the queries submitted by the audience to illustrate the internals of Data Vault functioning by revealing the mechanisms of dynamic query plan generation and on-demand external data ingestion. The image Data Vault shows an application view from the perspective of data mining researchers.","Title: Ratio-Detector-Based Feature Extraction for Very High Resolution SAR Image Patch Indexing\nYear: 2013\n\nAbstract: With the advent of very high resolution (VHR) synthetic aperture radar (SAR) images, local content description is becoming a critical issue for indexing. Conventional SAR image analysis techniques, like segmentation and pixel-level classification, are likely to fail as high-level semantic description should be considered for better discrimination. Therefore, we propose to use image-patch-based analysis method for SAR image interpretation. Inspired by ratio edge detector, in this letter, a new feature extraction method represented by the mean ratios in different directions is proposed for VHR SAR image content characterization. Based on the mean ratio, two simple yet powerful and robust features are proposed for SAR image patch indexing. One is the bag-of-word model using not only the basic statistics, i.e., local mean and variance, but also the mean ratios in different directions. The second one is an adaptation of the Weber local descriptor to SAR images by substituting the gradient with the ratio of mean differences in vertical and horizontal directions. To evaluate the proposed features, image patch indexing based on active learning using a SAR image database consisting of high-resolution TerraSAR-X patches is performed. Comparison with the state-of-the-art features, particularly texture features, has shown improved performance for SAR image categorization.","Title: Supervised learning on graphs of spatio-temporal similarity in satellite image sequences\nYear: 2007\n\nAbstract: Abstract: High resolution satellite image sequences are multidimensional signals composed of spatiotemporal patterns associated to numerous and various phenomena. Bayesian methods have been previously proposed in [8] to code the information contained in satellite image sequences in a graph representation using Bayesian methods. Based on such a representation, this paper further presents a supervised learning methodology of semantics associated to spatio-temporal patterns occurring in satellite image sequences. It enables the recognition and the probabilistic retrieval of similar events. Indeed, graphs are attached to statistical models for spatio-temporal processes, which at their turn describe physical changes in the observed scene. Therefore, we adjust a parametric model evaluating similarity types between graph patterns in order to represent user-specific semantics attached to spatio-temporal phenomena. The learning step is performed by the incremental definition of similarity types via user-provided spatio-temporal pattern examples attached to positive or/and negative semantics. From these examples, probabilities are inferred using a Bayesian network and a Dirichlet model. This enables to links user interest to a specific similarity model between graph patterns. According to the current state of learning, semantic posterior probabilities are updated for all possible graph patterns so that similar spatio-temporal phenomena,can be recognized and retrieved from the image sequence. Few experiments performed on a multi-spectral SPOT image sequence illustrate the proposed spatio-temporal recognition method.","Title: Authorship Analysis based on Data Compression.\nYear: 2014\n\nAbstract: •Compression-based similarity measures are effectively used in authorship analysis tasks.•Their high computational complexity makes it hard to employ them in practical applications.•We propose the Fast Compression Distance (FCD) for these specific tasks.•The FCD yields more accurate results and has a reduced computational complexity.•Satisfactory experiments are presented on datasets written in five different languages.","Title: TELEIOS: a database-powered virtual earth observatory\nYear: 2012\n\nAbstract: TELEIOS is a recent European project that addresses the need for scalable access to petabytes of Earth Observation data and the discovery and exploitation of knowledge that is hidden in them. TELEIOS builds on scientific database technologies (array databases, SciQL, data vaults) and Semantic Web technologies (stRDF and stSPARQL) implemented on top of a state of the art column store database system (MonetDB). We demonstrate a first prototype of the TELEIOS Virtual Earth Observatory (VEO) architecture, using a forest fire monitoring application as example.","Title: Building virtual earth observatories using ontologies and linked geospatial data\nYear: 2012\n\nAbstract: Advances in remote sensing technologies have enabled public and commercial organizations to send an ever-increasing number of satellites in orbit around Earth. As a result, Earth Observation (EO) data has been constantly increasing in volume in the last few years, and is currently reaching petabytes in many satellite archives. For example, the multi-mission data archive of the TELEIOS partner German Aerospace Center (DLR) is expected to reach 2PB next year, while ESA estimates that it will be archiving 20PB of data before the year 2020. As the volume of data in satellite archives has been increasing, so have the scientific and commercial applications of EO data. Nevertheless, it is estimated that up to 95% of the data present in existing archives has never been accessed, so the potential for increasing exploitation is very big.","Title: Modeling Trajectory Of Dynamic Clusters In Image Time-Series For Spatio-Temporal Reasoning\nYear: 2005\n\nAbstract: During the last decades, satellites have acquired incessantly high-resolution images of many Earth observation sites. New products have arisen from this intensive acquisition process: high-resolution satellite image time-series (SITS). They represent a large data volume with a rich information content and may open a broad range of new applications. This paper presents an information mining concept which enables a user to learn and retrieve spatio-temporal structures in SITS. The concept is based on a hierarchical Bayesian modeling of SITS information content which enables us to link the interest of a user to specific spatio-temporal structures. The hierarchy is composed of two inference steps: an unsupervised modeling of dynamic clusters resulting in a graph of trajectories, and an interactive learning procedure based on graphs which leads to the semantic labeling of spatio-temporal structures. Experiments performed on a SPOT image time-series demonstrate the concept capabilities.","Title: Modeling Trajectory Of Dynamic Clusters In Image Time-Series For Spatio-Temporal Reasoning\nYear: 2005\n\nAbstract: During the last decades, satellites have acquired incessantly high-resolution images of many Earth observation sites. New products have arisen from this intensive acquisition process: high-resolution satellite image time-series (SITS). They represent a large data volume with a rich information content and may open a broad range of new applications. This paper presents an information mining concept which enables a user to learn and retrieve spatio-temporal structures in SITS. The concept is based on a hierarchical Bayesian modeling of SITS information content which enables us to link the interest of a user to specific spatio-temporal structures. The hierarchy is composed of two inference steps: an unsupervised modeling of dynamic clusters resulting in a graph of trajectories, and an interactive learning procedure based on graphs which leads to the semantic labeling of spatio-temporal structures. Experiments performed on a SPOT image time-series demonstrate the concept capabilities.","Title: Semantic analysis of satellite image time series.\nYear: 2012\n\nAbstract: Large archives of satellite images have been created over time. The existence of these archives enables us to extract evolutions of the same of same geographic area over time, creating satellite image time series (SITS). as SITS represent an amount of information far grater than individual images. their analysis is complex and difficult. In this paper we propose a new unsupervised SITS analysis method based on the latent Dirichlet allocation (LDA) model, a hierarchical model originally developed for text analysis. In this model documents are represented as random mixture of latent topics, each topic being characterized by a distribution over words. This paper extends the use of LDA model for satellite image time series analysis by proposing a description language for SITS modeling according to the LDA model, and is applied on a SITS of 11 Landsat TM scenes acquired in 2007.","Title: Despeckling of TerraSAR-X Data Using Second-Generation Wavelets\nYear: 2010\n\nAbstract: This letter presents the despeckling of synthetic aperture radar (SAR) images within the bandelet and contourlet domains. A model-based approach is presented for the despeckling of SAR images. The speckle-reduced estimate is found using the first-order Bayesian inference, and the best model's parameters are estimated using the second-order Bayesian inference. Synthetic and real images are used for...","Title: SAR Target Analysis Based on Multiple-Sublook Decomposition: A Visual Exploration Approach.\nYear: 2012\n\nAbstract: The advent of submeter-resolution synthetic aperture radar (SAR) images from satellites such as TerraSAR-X has given a new dimension to SAR image understanding. Even though emphasis is always on discovering automatic means of target characterization, visual exploration of targets and objects is the first step in many applications. While considering the complex-valued SAR images, visual inspection ...","Title: Image information mining system evaluation using information-theoretic measures\nYear: 2005\n\nAbstract: During the last decade, the exponential increase of multimedia and remote sensing image archives, the fast expansion of the world wide web, and the high diversity of users have yielded concepts and systems for successful content-based image retrieval and image information mining. Image data information systems require both database and visual capabilities, but there is a gap between these systems. Database systems usually do not deal with multidimensional pictorial structures and vision systems do not provide database query functions. In terms of these points, the evaluation of content-based image retrieval systems became a focus of research interest. One can find several system evaluation approaches in literature, however, only few of them go beyond precision-recall graphs and do not allow a detailed evaluation of an interactive image retrieval system. Apart from the existing evaluation methodologies, we aim at the overall validation of our knowledge-driven content-based image information mining system. In this paper, an evaluation approach is demonstrated that is based on information-theoretic quantities to determine the information flow between system levels of different semantic abstraction and to analyze human-computer interactions.","Title: An Innovative Concept for Image Information Mining.\nYear: 2002\n\nAbstract: Information mining opens new perspectives and a huge potential for information extraction from large volumes of heterogeneous images and the correlation of this information with the goals of applications. We present a new concept and system for image information mining, based on modelling the causalities which link the image-signal contents to the objects and structures within interest of the users. The basic idea is to split the information representation into four steps: 1. image feature extraction using a library of algorithms so as to obtain a quasi-complete signal description 2. unsupervised grouping in a large number of clusters to be suitable for a large set of tasks 3. data reduction by parametric modelling the clusters 4. supervised leaming of user semantics, that is the level where, instead of being programmed, the systems is trained by a set of examples; thus the links from image contents to the users are created. The record of the sequence of links is a knowledge acquisition process, the system memorizes the user hypotheses. Step 4. is a man-machine dialogue, the information exchange is done using advanced visualization tools. The system learns what the users need. The system is presently prototyped for inclusion in a new generation of intelligent satellite ground segment systems, value adding tools in the area of geoinformation, and several applications in medicine and biometrics are also foreseen.","Title: Complexity Based Image Artifact Detection\nYear: 2008\n\nAbstract: Images may contain blemishes or artificial structures which come from the processing or directly from the sensors. These artifacts decrease the quality of the images and can lead to analysis and interpretation problems. We are interested in detecting these defects automatically. A visual analysis of such artifacts brings to the conclusion that they constitute a local modification of the complexity of the image. Therefore this paper will give a complexity based approach for the detection of these defects. This approach will distinguich two different methods: firstly the artifacts will be looked for respectively to given example, with a complexity comparison. This method will involve both direct and model based means. From the weaknesses of that method a second one will be deducted which will not be attached to the given examples and will take into consideration the local complexity context of the image. These methods will be applied in the domain of Earth Observation images.","Title: Categorization Based Relevance Feedback Search Engine For Earth Observation Images Repositories\nYear: 2006\n\nAbstract: Presently Earth Observation (EO) satellites acquire huge volumes of high resolution images very much over-passing the capacity of the users to access the information content of the acquired data. Thus, in addition to the existing methods for EO data and information extraction, new methods and tools are needed to explore and help to discover the information hidden in large EO image repositories. This article presents a categorisation based Relevance Feedback (RF) search engine for EO images repositories The developed method is presented as well results obtained for a SPOT5 satellite image database.","Title: Huber–Markov Model for Complex SAR Image Restoration\nYear: 2010\n\nAbstract: This letter presents the despeckling of single-look complex (SLC) synthetic aperture radar (SAR) images using nonquadratic regularization. The objective function consists of an image model, a gradient, and a prior model. The Huber-Markov random field (HMRF) models the prior. A numerical solution is achieved through extensions of half-quadratic regularization methods using complex-valued SAR data. ...","Title: Mutual information based measure for image content characterization\nYear: 2005\n\nAbstract: An image can be decomposed into different elementary descriptors depending on the observer interest. Similar techniques as used to understand words, regarded as molecules, formed by combining atoms, are proposed to describe images based on their information content. In this paper, we use primitive feature extraction and clustering to code the image information content. Our purpose is to describe the complexity of the information based on the combinational profile of the clustered primitive features using entropic measures like mutual information and Kullback-Leibler divergence. The developed method is demonstrated to asses image complexity for further applications to improve Earth Observation image analysis for sustainable humanitarian crisis response in risk reduction.","Title: Optimized lossless compression of remote sensing image files using splay trees\nYear: 2012\n\nAbstract: In this paper will be applied concepts like amortized complexity or self-adjustment to binary search trees. Motivation comes from the fact that the search trees have multiples drawbacks. It will be developed and analyzed the splay tree, a self-adjusting form of binary search trees.","Title: Sar Complex Image Analysis: A Gauss Markov And A Multiple Sub-Aperture Based Target Characterization\nYear: 2010\n\nAbstract: In this paper we discuss Gauss-Markov Random Field (GMRF) based on multiple sub-aperture decomposition method for the analysis of targets in complex-valued high-resolution SAR data. Gauss-Markov Random Field (GMRF) model with a quadratic energy function as a parametric analysis parameterizes the spectogram of the signal, whereas sub-aperture decomposition method exploits the holographic property of the spectrum at the cost of reducing resolution. This analysis helps to understand, characterize and analyze complex-valued SAR data and provides temptation to use complex-valued SAR data over detected data.","Title: Further results on dissimilarity spaces for hyperspectral images RF-CBIR\nYear: 2013\n\nAbstract: Content-Based Image Retrieval (CBIR) systems are powerful search tools in image databases that have been little applied to hyperspectral images. Relevance feedback (RF) is an iterative process that uses machine learning techniques and user's feedback to improve the CBIR systems performance. We pursued to expand previous research in hyperspectral CBIR systems built on dissimilarity functions defined either on spectral and spatial features extracted by spectral unmixing techniques, or on dictionaries extracted by dictionary-based compressors. These dissimilarity functions were not suitable for direct application in common machine learning techniques. We propose to use a RF general approach based on dissimilarity spaces which is more appropriate for the application of machine learning algorithms to the hyperspectral RF-CBIR. We validate the proposed RF method for hyperspectral CBIR systems over a real hyperspectral dataset.","Title: A Bayesian Multi-Class Image Content Retrieval\nYear: 2007\n\nAbstract: Modern imaging sensors, especially those aboard satellites, continuously deliver enormous amounts of data. The widespread of meter resolution images, is not only exploding the volumes of acquired data but also brings a new dimension in the image detail, thus growing the information content. These represent typical cases, where users need automated tools to discover, explore and explain the contents of large image databases. There is a strong need to build up applications that help the user in image interpretation task, applications that permit to query the archives in content based mode, without having to know all the information contained in the images at signal level. We propose in this article, a synergy between stochastic modelling, knowledge discovery, and semantic representation. To do that, we associate semantic labels to a combination of primitive image features. The user-defined semantic image content interpretation is linked with Bayesian networks to a completely unsupervised classification. This new paradigm for the interaction with EO archives can provide several applications for users coming from different domains, as change detection, agricultural field classification, environment monitoring, atmosphere effects or urbanization.","Title: Bayesian restoration of interferometric phase through biased anisotropic diffusion\nYear: 2009\n\nAbstract: In this paper a new Bayesian algorithm for interferometric phase restoration is presented. Based on a non-linear anisotropic extension of Orientation Diffusions, the inherent directionality of the fringe structure is introduced into its prior model. It also accounts for the periodicity of the phase representation. A fidelity term derived from the anisotropic metrics and the InSAR phase statistics deviates diffusion towards the acquired phase value. It acts as an adapted likelihood of the diffused phase. Hence, phase restoration is a trade-off between directionality and reconstruction fidelity, prior and likelihood. Results are provided on a High Resolution Spotlight scene acquired by TerraSAR-X.","Title: Scale Space Exploration For Mining Image Information Content.\nYear: 2002\n\nAbstract: Images are highly complex multidimensional signals, with rich and complicated information content. For this reason they are difficult to analyze with a specific automated approach. However, a hierarchical representation is helpful for understanding image content. In this paper, we describe an application of a scale-space clustering algorithm (melting) for exploration of image information content. Clustering by melting considers the feature space as a thermodynamical ensemble and groups the data by minimizing the free energy, having temperature as a scale parameter. We develop clustering by melting for multidimensional data, and propose and demonstrate a solution for the initialization of the algorithm. Due to the curse of dimensionality, for initialization of clusters we choose the initial clusters centers with an algorithm that performs fast cluster center estimation with low computation cost. We further analyze the information extracted by melting and propose a structure for information representation that enables exploration of image content. This structure is a tree in the scale space showing how the clusters merge. Implementation of the algorithm is through a multi-tree structure. With this structure, we can explore the image content as an information mining function, we obtain a more compact data structure, and we have maximum of information in scale space because we memorize the bifurcation points and the trajectories of the centers points in the scale space. The information encoded in the tree structure enables the fast reconstruction and exploration of the data cluster structure and the investigation of hierarchical sequences of image classifications. We demonstrate the effectiveness of the approach with examples using satellite multispectral image (SPOT 4) and Synthetic Aperture Radar (SAR) and Digital Elevation Models (DEM) derived from SAR interferometry (SRTM).","Title: A Comparative Study of Statistical Models for Multilook SAR Images\nYear: 2014\n\nAbstract: In this letter, we carry out a comparative study of statistical models for multilook synthetic aperture radar amplitude images. Ten state-of-the-art statistical models are selected for comparison. To achieve a fair evaluation, we estimate all model parameters using the method of log-cumulants and apply the method to an image pyramid with varying pixel spacing (and resolution). The pyramid is created by different image product generation options. In addition to pixel spacing and resolution, we also consider the homogeneity of a scene for performance evaluation and we apply three performance measures. Through this study, it was found out that some models perform well for all resolutions, while the performance of other models depends heavily on the image content.","Title: Stochastic Models Of Slc Hr Sar Images\nYear: 2007\n\nAbstract: The paper presents two algorithms for texture primitive feature extraction on Single Look Complex (SLC) and Polarimetric Synthetic Aperture Radar (PolSAR) SLC data. We assume the data to be modeled by a Gauss-Markov Random Field (GMRF): a complex GMRF model for characterizing the spatial correlation in SLC data and an extension of the model for inter-band correlation characterization.The complex GMRF characterizes the spatial relationship of a two-dimensional complex signal, i.e. SLC SAR data. The extended model characterizes the spatial interaction and the inter-band pixels correlation between the polarimetric complex channels. The Bayesian approach permits to deal with model fitting and selection in a direct way.The results are presented on a polarimetric E-SAR L band scene of Mannheim, Germany.","Title: Cascaded Active Learning For Object Retrieval Using Multiscale Coarse To Fine Analysis\nYear: 2011\n\nAbstract: In this paper, we describe an active learning scheme which performs coarse to fine testing using a multiscale patch-based representation of images to retrieve objects in large satellite image repositories. The proposed hierarchical top-down approach reduces step by step the size of the analysis window, eliminating each time large parts of the images considered as non-relevant. Unlike most object detection methods which requires large training sets and costly offline training, we use an active learning strategy to build a classifier at each level of the hierarchy and we propose an algorithm to propagate automatically the training examples from one level to the other.","Title: Salient Remote Sensing Image Segmentation Based on Rate-Distortion Measure\nYear: 2009\n\nAbstract: The ill-defined nature of the segmentation problem makes the selection of the optimal image partition difficult. One can characterize image segmentation as an attempt to find the best possible representation of a data set using a certain number of ldquoobjects.rdquo This can be regarded as data information compression, resulting in the distortion of the original values. Data sets are well represen...","Title: Parametric Modeling Of The Fractional Fourier Transform Coefficients For Complex-Valued Sar Image Categorization\nYear: 2013\n\nAbstract: In high-resolution (HR) and very-high resolution (VHR) synthetic aperture radar (SAR) images, focus is now on the patch-oriented image categorization in contrast to the pixelbased classification in low-resolution SAR images. SAR image categorization requires the generation of a compact feature descriptor that accurately defines the content of the image patch under consideration.In this paper we propose a parametric feature descriptor generated on the complex-valued SAR image within a transformation space. The fractional Fourier transform (FrFT), has been considered to transform the image pixels of the singlelook complex (SLC) SAR images in order to obtain a simpler statistical response. The real and imaginary components of the complex-valued FrFT coefficients have been modelled with generalized Gaussian distribution (GGD).The proposed feature descriptor is compared with a Wavelet-decomposition-based parametric feature descriptor; and with the FrFT-based and Gabor-filter-bank-based non-parametric feature descriptors. Categorization accuracy enhancement is demonstrated over several categories comprising of natural topologies. The experimental database consists of 2000 image patches (of size 200 x 200 pixels) extracted from SLC HR TerraSAR-X scenes.","Title: Gibbs Random Field Models for Model-Based Despeckling of SAR Images\nYear: 2010\n\nAbstract: Synthetic aperture radar (SAR) images are affected by multiplicative noise called speckle. This noise makes automatic image classification and image interpretation difficult. Thus, many methods have been developed to remove speckle from SAR images while preserving the useful information of the scene such as texture and geometry. In this letter, a comparison between three different despeckling meth...","Title: Semantic Map Generation from Satellite Images for Humanitarian Scenarios Applications\nYear: 2008\n\nAbstract: This paper demonstrates how knowledge driven methods and the associated data analysis algorithms are changing the paradigms of user-data interactions, providing an easier and wider access to the Earth Observation data. Some information theory based algorithms are proposed for anomaly and change detection on SPOT images, relative to a widespread humanitarian crisis scenario: floods. The outcomes of these algorithms define an informational representation of the image, revealing the spatial distribution of a particular theme. Using image analysis and interpretation, the multitude of features from a scene are classified into meaningful classes to create sematic maps.","Title: A fast compression-based similarity measure with applications to content-based image retrieval\nYear: 2012\n\nAbstract: Compression-based similarity measures are effectively employed in applications on diverse data types with a basically parameter-free approach. Nevertheless, there are problems in applying these techniques to medium-to-large datasets which have been seldom addressed. This paper proposes a similarity measure based on compression with dictionaries, the Fast Compression Distance (FCD), which reduces the complexity of these methods, without degradations in performance. On its basis a content-based color image retrieval system is defined, which can be compared to state-of-the-art methods based on invariant color features. Through the FCD a better understanding of compression-based techniques is achieved, by performing experiments on datasets which are larger than the ones analyzed so far in literature.","Title: The Model based Similarity Metric\nYear: 2007\n\nAbstract: This paper addresses the problem of building a compact representation of objects which enables to define a similarity measure. The key idea is to provide a mean to extract and represent the relevant information of an object. By integrating the Minimum Length Description to an informational similarity metric based on Kolmogorov complexity [1], we propose a new informational measure based on models.","Title: Cascade active learning for SAR image annotation\nYear: 2012\n\nAbstract: In this paper, a novel active learning approach and system incorporating multiple instance learning for SAR image mining and annotation is introduced. Based on a multiscale and hierarchial patch based image representation, a cascade classifier is learned at different levels. At each level of the hierarchy, a SVM classifier is trained based on active learning and the training sample propagation between different levels is achieved through Multiple Instance SVM (MI-SVM). Classification at the higher level is applied only to the positive patches obtained at the previous level, which can significantly reduce the burden of computation in the case of large data set. Performance has been evaluated through a large data set, which shows promising gain not only in accuracy but also in computation.","Title: Rate Distortion Based Detection of Artifacts in Earth Observation Images\nYear: 2008\n\nAbstract: Earth observation optical images may contain blemishes or artificial structures that come from processing or directly from the sensors. These artifacts decrease the quality of the images and can lead to analysis and interpretation problems. We are interested in detecting these defects automatically. We develop an approach based on an information-theoretic analysis of the image formation process, w...","Title: Spatial information retrieval from remote-sensing images. I. Information theoretical perspective\nYear: 1998\n\nAbstract: Automatic interpretation of remote-sensing (RS) images and the growing interest for query by image content from large remote-sensing image archives rely on the ability and robustness of information extraction from observed data. In Parts I and II of this article, the authors turn the attention to the modern Bayesian way of thinking and introduce a pragmatic approach to extract structural informati...","Title: Earth-Observation Image Retrieval Based on Content, Semantics, and Metadata\nYear: 2013\n\nAbstract: Advances in the image retrieval (IR) field have contributed to the elaboration of tools for interactive exploration and extraction of the images from huge archives associating the content of the images with semantic meaning. This paper presents an Earth-observation (EO) IR system based on enriched metadata, semantic annotations, and image content called EO retrieval. EO retrieval generates an EO-data model by using automatic feature extraction, processing the EO product metadata, and defining semantics, which later is fully exploited for supporting complex queries. In order to demonstrate the functionality of the system, we have created a semantic catalog of TerraSAR-X as application scenario. The database is composed of 39 high-resolution TerraSAR-X scenes comprising about 50 000 image patches (160 × 160 pixels) with their feature descriptors, 100 of metadata entries for each scene, and about 330 semantic annotations. Many query examples combining semantics, metadata, and image content for full exploitation of the image database are presented.","Title: Image Time Series Mining For Dynamic Scene Understanding\nYear: 2003\n\nAbstract: In this paper, a dynamic scene understanding concept is proposed and applied on multispectral image time series. Information mining enables the explorations and discovery of spatio temporal patterns localized in given spatio temporal windows.With this in mind, a hierarchical information representation is developed. It comprises different levels in which the data is modeled so that the relevant information is transmitted through the architecture, according to a query and to several assumptions made on the models employed. There are mainly four components : feature extraction, reduction of dimensionality, clustering and interactive exploration.","Title: Semantic Annotation of Satellite Images Using Latent Dirichlet Allocation\nYear: 2010\n\nAbstract: In this letter, we are interested in the annotation of large satellite images, using semantic concepts defined by the user. This annotation task combines a step of supervised classification of patches of the large image and the integration of the spatial information between these patches. Given a training set of images for each concept, learning is based on the latent Dirichlet allocation (LDA) mo...","Title: Expanding the Algorithmic Information Theory Frame for Applications to Earth Observation.\nYear: 2013\n\nAbstract: Recent years have witnessed an increased interest towards compression-based methods and their applications to remote sensing, as these have a data-driven and parameter-free approach and can be thus succesfully employed in several applications, especially in image information mining. This paper expands the algorithmic information theory frame, on which these methods are based. On the one hand, algorithms originally defined in the pattern matching domain are reformulated, allowing a better understanding of the available compression-based tools for remote sensing applications. On the other hand, the use of existing compression algorithms is proposed to store satellite images with added semantic value.","Title: Deep learning in very high resolution remote sensing image information mining communication concept.\nYear: 2012\n\nAbstract: This paper presents the image information mining based on a communication channel concept. The feature extraction algorithms encode the image, while an analysis of topic discovery will decode and send its content to the user in the shape of a semantic map. We consider this approach for a real meaning based semantic annotation of very high resolution remote sensing images. The scene content is described using a multi-level hierarchical information representation. Feature hierarchies are discovered considering that higher levels are formed by combining features from lower level. Such a level to level mapping defines our methodology as a deep learning process. The whole analysis can be divided in two major learning steps. The first one regards the Bayesian inference to extract objects and assign basic semantic to the image. The second step models the spatial interactions between the scene objects based on Latent Dirichlet Allocation, performing a high level semantic annotation. We used a WorldView2 image to exemplify the processing results.","Title: Enhancing complex interferograms by anisotropic diffusion\nYear: 2009\n\nAbstract: In this paper a new algorithm for interferometric phase restoration is presented. Firstly, a continuous framework for anisotropic phase diffusion is stated. A tensorial based metric allows directional control. The periodic continuous structure of the phase representation is accounted for. Secondly, this framework is adapted for interferometric phase filtering. Progressive re-estimation of directionality avoids directional bias. Isotropy and anisotropy are adaptively combined with a constant overall diffusion rhythm, so that the degree of regularization is the same regardless of the underlying topography. Robust estimation minimizes the spread of outliers. Results on both synthetic and TerraSAR-X data are provided.","Title: Wavelet-Based Despeckling of SAR Images Using Gauss–Markov Random Fields\nYear: 2007\n\nAbstract: In this paper, a wavelet-based speckle-removing algorithm is represented and tested on synthetic aperture radar (SAR) images. The SAR image is first transformed using a dyadic wavelet transform. The noise in the wavelet-transformed image is modeled as an additive signal-dependent noise with Gaussian distribution. The distribution of a noise-free image in a wavelet domain is modeled as a generalize...","Title: Coarse-to-Fine Approach for Urban Area Interpretation Using TerraSAR-X Data\nYear: 2010\n\nAbstract: With the launch of the German TerraSAR-X system in June 2007, a new generation of high-resolution spaceborne synthetic aperture radar (SAR) data is available, which should facilitate the interpretation of urban environments. Our overall objective in this letter is to provide a semiautomatic tool for urban area interpretation using SAR data. We propose in this letter to fuse different automatic obj...","Title: Information Theoretical Approach For Domain Ontology Exploration In Large Eo Image Archives\nYear: 2004\n\nAbstract: The interpretation of EO data requires not only data and/or information fusion for better understanding of Earth cover structures, but, at a higher level, needs the aggregation with existing bodies of knowledge specific to the application fields.","Title: Fusion of Real and Synthetic Images for Remote Sensing Scene Understanding\nYear: 1993\n\nAbstract: INTRODUCTIONThe last generation of Earth Observation Satellites is characterised by a multitude of sensortypes covering a large domain of spectral sensitivity and spatial resolution. The usual sensorsrange from visible, infrared to microwave spectrum, covering coherent or non-coherentradiation, with spatial resolution from meters to kilometres. In the mean time, the huge complexityof land-forms and land-cover types is reflected in the complexity of the recordedimages. As consequence the...","Title: Automatic Change Analysis in Satellite Images Using Binary Descriptors and Lloyd–Max Quantization\nYear: 2015\n\nAbstract: In this letter, we present a novel technique for unsupervised change analysis that leads to a method of ranking the changes that occur between two satellite images acquired at different moments of time. The proposed change analysis is based on binary descriptors and uses the Hamming distance as a similarity metric. In order to render a completely unsupervised solution, the obtained distances are further classified using vector quantization methods (i.e., Lloyd's algorithm for optimal quantization). The ultimate goal in the change analysis chain is to build change intensity maps that provide an overview of the severeness of changes in the area under analysis. In addition, the proposed analysis technique can be easily adapted for change detection by selecting only two levels for quantization. This discriminative method (i.e., between changed/unchanged zones) is compared with other previously developed techniques that use principal component analysis or Bayes theory as starting points for their analysis. The experiments are carried on Landsat images at a 30-m spatial resolution, covering an area of approximately 59×51 km2 over the surroundings of Bucharest, Romania, and containing multispectral information.","Title: FrFT-Based Scene Classification of Phase-Gradient InSAR Images and Effective Baseline Dependence\nYear: 2015\n\nAbstract: In the literature, scene recognition from interferometric synthetic aperture radar (InSAR) images has been mainly focused on the joint use of the backscatter intensity and the coherence between interferometric image pairs. However, the terrain height information residing in the interferometric phase requires further exploration for classification purposes. In this letter, taking the interferometric phase information into account together with the backscatter intensity, the whole complex- valued InSAR image is exploited for feature extraction. In addition, a new complex-valued phase-gradient InSAR (PGInSAR) image is defined. A fractional-Fourier-transform-based feature ex traction, which was proposed for the classification of single-look complex (SLC) SAR images, is adopted for InSAR and PGInSAR images. For patch-based classification, an image database is generated from bistatic pairs acquired from the same terrain with three different effective baselines. The supervised κ-nearest neighbor classification results show that InSAR outperforms SLC by 15%, whereas PGInSAR introduces further 10% improvement over InSAR or a total improvement of 27% over SLC. Moreover, PGInSAR is found to be more robust to effective baseline changes than InSAR, which makes PGInSAR a better candidate for feature extraction.","Title: A Comparative Study of Bag-of-Words and Bag-of-Topics Models of EO Image Patches\nYear: 2015\n\nAbstract: The large volume of detailed land cover features, provided by high resolution Earth observation (EO) images, has attracted considerable interest in the discovery of these features by learning systems. In this letter, we perform latent Dirichlet allocation on the bag of words (BoW) representation of collections of EO image patches to discover their semantic-level features, the so-called topics. To assess the discovered topics, the images are represented based on the occurrence of different topics, called bag of topics (BoT). The value added by BoT to the BoW model of image patches is then measured based on existing human annotations of the data. In our experiments, we compare the classification accuracy results of BoT and BoW representations of two different remote sensing image data sets, a multispectral optical data set and a synthetic-aperture-radar data set. Experimental results demonstrate that BoT can provide a compact and semantically meaningful representation of data; it either causes no significant reduction in the classification accuracy or increases the accuracy by a sufficient number of topics.","Title: Big, Linked and Open Data: Applications in the German Aerospace Center.\nYear: 2014\n\nAbstract: Earth Observation satellites acquire huge volumes of high resolution images continuously increasing the size of the archives and the variety of EO products. However, only a small part of this data is exploited. In this paper, we present how we take advantage of the TerraSAR-X images of the German Aerospace Center in order to build applications on top of EO data.","Title: On the statistical similarity of synthetic aperture radar images from COSMO-SKYMED and TerraSAR-X\nYear: 2014\n\nAbstract: The latest generation of synthetic aperture radar (SAR) instruments operating in X-band, that is, COSMO-SkyMed (CSK) and TerraSAR-X (TSX), are capable of providing images from coarse resolution to very high resolution. A lot of research effort has been invested in the study and understanding of images obtained from these satellites. However, there is still a huge scope of statistical understanding and comparison of data from both satellites. In this study, we demonstrate some striking similarities between medium resolution data obtained from CSK and TSX Stripmap mode images. Landcover unsupervised clustering using k-means is discussed to further justify our findings. Clustering is carried out using a feature descriptor based on log-cumulants of Gabor coefficients, which was recently proposed by us in earlier studies.","Title: Class evolution data analytics from sar image time series using information theory measures\nYear: 2014\n\nAbstract: In this paper we present the result of data analytics techniques applied to a database comprising of 32 SLC SM TerraSAR-X images, acquired over the area of Bucharest, Romania. The methodology follows a two step approach. The first stage consists of a coarse identification of potentially changed areas using a supervised learning image annotation tool with relevance feedback. Gabor texture features are used to describe image patches. The patch size is derived as a function of the resolution and pixel spacing of the data. In the second stage we apply an information theory strategy to refine the regions previously shown to exhibit class dynamics within the image stack, with pixel accuracy. Finally, a series of analytical indicators (absolute extent of areas affected by change, class evolution trends, inter-class correlations) are derived, in order to generate a predictive model for the selected test site.","Title: A new comprehensive approach for earth observation scene classification using joint image and text analysis\nYear: 2014\n\nAbstract: The amount of data available on the internet provides massive additional information for the Earth Observation (EO) imagery. Periodical news, various reports and measurements, pictures or online encyclopedias are just few examples of the existent information. Occasionally, this data offers new perspectives for EO image understanding and interpretation. However, current image analysis do not benefit from the advantage given by external sources. To overcome these drawbacks, the present paper proposes an approach that goes beyond traditional information mining by using a joint image and text analysis. Fast Compression Distance (FCD) is computed to measure the similarities inside a collection of very high resolution images and text files. The main purpose is to discover common patterns within the data, without any a priori assumption, parameter-free, relying on data compression-based techniques. A hierarchical clustering is performed in order to learn about the dependencies between different types of data.","Title: Despeckling and Information Extraction From SLC SAR Images\nYear: 2014\n\nAbstract: This paper presents an information extraction and image enhancement technique using single-look complex (SLC) synthetic aperture radar data. The novelty of this method is the proposed complex-domain despeckling stage. Tikhonov-like optimization is used for minimizing the cost function, which consists of a Gauss-Markov random field (GMRF) prior. The GMRF model is used for texture modeling. The texture parameters of the GMRF are estimated using the evidence maximization framework. The experimental results showed that despeckled SLC images have well-preserved textural features, structures, and point scatterers. The phase of the reconstructed image is well preserved and provides good-quality interferograms of high-resolution spotlight images.","Title: Locally Linear Salient Coding for image classification\nYear: 2014\n\nAbstract: Representing images with their descriptive features is the fundamental problem in CBIR. Feature coding as a key-step in feature description has attracted the attentions in recent years. Among the proposed coding strategies, Bag-of-Words (BoW) is the most widely used model. Recently saliency has been mentioned as the fundamental characteristic of BoW. Base on this idea, Salient Coding (SaC) has been introduced. Empirical studies show that SaC is not able to represent the global structure of data with small number of codewords. In this paper, we remedy this limitation by introducing Locally Linear Salient Coding (LLSaC). This method discovers the global structure of the data by exploiting the local linear reconstructions of the data points. This knowledge in addition to the salient responses, provided by SaC, helps to describe the structure of the data even with a few codewords. Experimental results show that LLSaC obtains state-of-the-art results on various data types such as multimedia and Earth Observation.","Title: Comparing the information extracted by feature descriptors from EO images using Huffman coding\nYear: 2014\n\nAbstract: Traditionally, images are understood based on their primitive features such as color, texture, and shape. The proposed feature extraction methods usually cover a range of primitive features. SIFT, for example, in addition to the shape-based information, extracts texture and color information to some extent. Thus, different descriptors may cover a common range of primitive features which we call information overlap. Selecting a set of feature descriptors with low information overlap allows more comprehensive understanding of the data by providing a broader range of new features. This article introduces a new method based on information theory for comparing various descriptors. The idea is to code each description of an image by Huffman coding. The distance between the coded descriptions are then measured using Levenshtein distance as the information overlap. Results show that the computed information overlap clearly describes the differences between the learning from different descriptions of Earth Observation images.","Title: Quantitative flood assessment: Case study of floods in Germany\nYear: 2014\n\nAbstract: In this paper, we present a quantitative analysis for a rapid mapping scenario that performs a damage assessment of the 2013 floods in Germany. The scenario is created using pre-disaster and post-disaster TerraSAR-X images and an automated annotation system. Our data set is tiled into patches and Gabor filters are used as a primitive feature method applied to each patch separately. An active learning system based on support vector machine is implemented in order to group the features into categories. Once all categories are identified, these are semantically annotated using reference data as ground truth. In our evaluation 7 categories were retrieved with their specific taxonomies defined using our previous hierarchical annotation scheme. We show that the system supports rapid mapping scenarios (e.g., floods, tsunami, earthquake, etc.) and interactive mapping generation. In addition, with the help of this system, quantitative assessment of disasters can be carried out.","Title: Farness preserving Non-negative matrix factorization\nYear: 2014\n\nAbstract: Dramatic growth in the volume of data made a compact and informative representation of the data highly demanded in computer vision, information retrieval, and pattern recognition. Non-negative Matrix Factorization (NMF) is used widely to provide parts-based representations by factorizing the data matrix into non-negative matrix factors. Since non-negativity constraint is not sufficient to achieve robust results, variants of NMF have been introduced to exploit the geometry of the data space. While these variants considered the local invariance based on the manifold assumption, we propose Farness preserving Non-negative Matrix Factorization (FNMF) to exploits the geometry of the data space by considering non-local invariance which is applicable to any data structure. FNMF adds a new constraint to enforce the far points (i.e., non-neighbors) in original space to stay far in the new space. Experiments on different kinds of data (e.g., Multimedia, Earth Observation) demonstrate that FNMF outperforms the other variants of NMF.","Title: Image information mining utilizing hierarchical segmentation\nYear: 2002\n\nAbstract: The Hierarchical Segmentation (HSEG) algorithm is an approach for producing high quality, hierarchically related image segmentations. The VisiMine image information mining system utilizes clustering and segmentation algorithms for reducing visual information in multispectral images to a manageable size. The project discussed herein seeks to enhance the VisiMine system through incorporating hierarchical segmentations from HSEG into the VisiMine system. I. INTRODUCTION The Hierarchical Segmentation (HSEG) algorithm is an approach for producing high quality, hierarchically related image segmentations. The HSEG algorithm is a hybrid of region growing and spectral clustering that produces a hierarchical set of image segmentations. Different image analysis applications often require different levels of image segmentation detail. Further, a particular application may require different levels of image segmentation detail depending on the image objects being segmented. Thus, a choice of several levels of image segmentation detail is often very useful. The HSEG algorithm automatically produces a hierarchical set of segmentations, based on detected convergence points. An analyst can select between segmentations at different levels in the segmentation hierarchy to produce the most appropriate segmentation for a particular application. The VisiMine image information mining system utilizes clustering and segmentation algorithms for reducing visual information in multispectral images to a manageable size. The indexing strategy extracts three levels of features: 1) pixel level, 2) polygon level, and 3) tile level features. The pixel level features are used for the extraction of higher-level features. Fast segmentation algorithms based on PDE methods are used to decompose image tiles into polygons. In the paper we compare the results of similarity searches utilizing features extracted using an enhanced version of CLARA algorithm (Clustering LARge Applications) in combination with statistical sampling with the results that use hierarchical image segmentation (HSEG) to derive class This project is supported by NASA's Intelligent Systems Program, NASA Headquarters Code R. boundaries for land cover. We present the results of polygon and tile similarity searches based on spectral and textural properties. The use of HSEG features improves the precision and recall of similarity searches. The interactive label training and retrieval based on Bayesian models also benefits from hierarchical image segmentation approach.","Title: Multisource data classification with dependence trees\nYear: 2002\n\nAbstract: Abstract—In order to apply a statistical approach to the classi- fication of multisource remote-sensing data, one of the main prob- lems to face lies in the estimation of probability distribution func- tions. This problem arises out of the difficulty of defining a common statistical model for such heterogeneous data. A possible solution is to adopt nonparametric approaches, which rely on the availability of training samples without any assumption about the related sta- tistical distributions. The purpose of this paper is to investigate the suitability of the concept of dependence trees for the integration of multisource information through estimation of probability distri- butions. First, this concept, introduced by Chow and Liu, is used to provide an approximation of a probability distribution defined in an,-dimensional space by a product of,probability dis- tributions defined in two-dimensional (2-D) spaces; this approxi- mation corresponds, in terms of graph theoretical interpretation, to a tree of dependence. For each land cover class, a dependence tree is generated by minimizing an appropriate closeness measure. Then, a nonparametric estimation of the second-order probability distributions is carried out through the Parzen window approach, based on the implementation of 2-D Gaussian kernels. In this way, it is possible to reduce the complexity of the estimation, while cap- turing a significant part of the interdependence among variables. A comparison with other multisource data fusion methods, namely, the multilayer perceptron (MLP) method, the -nearest neighbor ( -NN) method, and a Bayesian hierarchical classifier (BHC), is made. Experimental results obtained on multisensor [airborne the- matic mapper,(ATM) and synthetic aperture radar (SAR)] and multisource (experimental synthetic aperture radar (E-SAR) and a textural feature) data sets show that the proposed fusion method based on dependence trees is able to provide a classification accu- racy similar to those of the other methods considered, but with the advantage of a reduced computational load. Index Terms—Dependence trees, discrete probability distribu-","Title: Study and assessment of selected primitive features behaviour for SAR image description\nYear: 2012\n\nAbstract: The main purpose of this study is to define for Synthetic Aperture Radar (SAR) data the primitive feature parameters, the incidence angle, and the orbit direction which can be used further for indexing and querying in the EO systems. The evaluation is done on the high resolution SAR data and the interpretation is realized automatically. In this paper, we propose to study and asses the behavior of the primitive feature extracted methods for images of the same scene with two look angles covering the min-max range of the sensor and with ascending / descending orbit looking. The tests are done on TerraSAR-X products Stripmap and high resolution Spotlight, specially and radiometrically enhanced covering the area of Berlin (Germany) and Ottawa (Canada). To identify the optimal primitive features, incident angle, and orbit direction the Support Vector Machine and as a measure of the classification accuracy the precision/recall were considered.","Title: Automated interpretation of very-high resolution SAR images\nYear: 2012\n\nAbstract: Very-high resolution (VHR) synthetic aperture radar (SAR) images from the last generation satellites such as TerraSAR-X and TanDEM-X exhibits special characteristics, especially in the urban-areas. Consequently, attention is needed on special considerations while developing algorithms for SAR image processing and its applications for automated interpretation. With automatic interpretation we refer to the information extraction and characterization for image categorization, retrieval, segmentation, automated target recognition etc.. In this article we focus our attention to the problem of SAR image categorization. The interest in SAR image categorization in VHR SAR (on the contrary to the pixel-based classification in low-resolution SAR images) has increased with enhanced resolution providing opportunity to carry out a more detailed analysis of targets and objects. SAR image categorization requires generation of a compact feature descriptor which accurately define the image content. A feature descriptor can be generated using `parametric' or `non-parametric' approaches in `image' or `image within a transformation space'. The objective of this article is to review some selected techniques for this purpose in form of a methodological classification. Qualitative assessment of selected algorithms is presented.","Title: SBAS and PS measurement fusion for enhancing displacement measurements\nYear: 2012\n\nAbstract: A major drawback of classical SAR interferometry is its sensitivity to temporal and geometric effects, leading to a distortion of the final results. This problem can be solved by using multi-temporal techniques, like the Small Baseline Subset and Persistent Scatterers algorithms. Both use large datasets for monitoring land cover deformation, motion or for DEM generation. Despite the fact that using the Persistent Scatterers technique can lead to a maximization of the number of acquisitions used, the number of persistent points is greatly reduced by a decrease of the coherent values of the targets over time. In this article, the authors propose a method for the selection of temporal persistent points using image subsets that comply with the small baseline rule. This method is further used to increase the accuracy of deformation measurements over the subsets' time span.","Title: Assessment of Earth Observation data content based on data compression - application to settlements understanding\nYear: 2012\n\nAbstract: Urban areas around the world are rapidly changing in an unregulated manner and remote sensing is the most effective option for their monitoring and planning. Good modeling of urban areas means reliable translation of the scene semantics into an algorithmic language. The compression based image retrieval techniques are data driven. The intention of employing compression based image retrieval techniques is to exploit the compression properties of the objects and estimate the shared information between them. Fast compression distance (FCD) is the similarity metric used in a compression based image retrieval technique that can be applied on large datasets. FCD between any two objects can be computed using the sizes of their dictionaries (sequence of recurring patterns) extracted through compression with LZW algorithm and the intersection of their dictionaries. In this paper, it is proposed to assess high resolution Earth Observation data content based on data compression for understanding urban settlements.","Title: Query by example in Earth-Observation image archive using data compression-based approach\nYear: 2012\n\nAbstract: This paper presents an implementation of query by example in Earth Observation image archive using data compression-based approach. Data compression approach allows to exploit the compression properties of the objects and to estimate the shared information between them, this concept is extended to image retrieval for finding similar objects in the image archive. Our implementation is based on LZW algorithm for compressing the image content and extracting features of the images. The fast compression distance (FCD) is defined as a similarity metric in order to retrieve the most similar images. This tool is satisfactory implemented and tested using optical and SAR images.","Title: Latent Dirichlet Allocation for Spatial Analysis of Satellite Images\nYear: 2013\n\nAbstract: This paper describes research that seeks to supersede human inductive learning and reasoning in high-level scene understanding and content extraction. Searching for relevant knowledge with a semantic meaning consists mostly in visual human inspection of the data, regardless of the application. The method presented in this paper is an innovation in the field of information retrieval. It aims to discover latent semantic classes containing pairs of objects characterized by a certain spatial positioning. A hierarchical structure is recommended for the image content. This approach is based on a method initially developed for topics discovery in text, applied this time to invariant descriptors of image region or objects configurations. First, invariant spatial signatures are computed for pairs of objects, based on a measure of their interaction, as attributes for describing spatial arrangements inside the scene. Spatial visual words are then defined through a simple classification, extracting new patterns of similar object configurations. Further, the scene is modeled according to these new patterns (spatial visual words) using the latent Dirichlet allocation model into a finite mixture over an underlying set of topics. In the end, some statistics are done to achieve a better understanding of the spatial distributions inside the discovered semantic classes.","Title: Image time-series mining\nYear: 2004\n\nAbstract: A visual information mining concept is proposed for spatio-temporal patterns discovery in remotely sensed image time-series (ITS). An information theory framework is adopted to first model information content. It results in the inference of a relevant directed graph characterizing ITS. Then the user conjecture is modeled via visual information representations : similarity measures between sub-graphs, which represents spatio- temporal events are derived and included in an interactive learning and probabilistic retrieval procedure of user-specific event-types. The present concept for ITS mining is demonstrated on multitemporal SPOT data. During the last decades, the imaging satellite sensors have acquired huge quantities of data enabling the elaboration of spatio-temporal databases. Image time-series (ITS) contained in them are complex objects containing interesting informa- tion. However, our capability to access and store large volume of data has highly exceeded our capability to extract the relevant information from the data. Therefore, ITS information systems are needed to bridge the gap between database theory, which is not adapted to manipulate temporal and pictural multidimensional data, and computer vision, which does not offer query possibilities. To cover such a broad problematic, a visual information mining concept is proposed to help a user to explore and understand spatio-temporal behaviors present in ITS. The article begins by presenting unsupervised methods for coding the information content in the observations. Thus, the dynamics of the feature space (FS) is modeled which results in the inference of a relevant directed graph representation, characterizing ITS. It continues by describing sub-graphs, contained in the graph, with model parameters. Combinations of the parameters correspond to semantic descriptions of spatio-temporal events. In this perspective, a distance model between sub-graph parameters is developed to enable user specific similarity measurements. Then, a Bayesian interactive learning on graphs is presented to enable reasoning on ITS. The learning is performed by the incremental definition of a spatio-temporal pattern type via user-provided positive and negative sub-graph examples. In parallel, the interactive search of spatio-temporal events is guided by accuracy probabilistic measurements derived from the current state of learning via the user positive and negative semantic. Finally, experiments","Title: Object and topology extraction from remote sensing images\nYear: 2005\n\nAbstract: We present a complete processing line to generate an object based description of optical remote sensing (RS) images. A segmentation algorithm is used to generate a partition of regions and simplify the volume of data. Results are still at the pixel level. Based on topology analyses, a dynamical algorithm is proposed to retrieve, extract the segmented regions and encode them in a tree structure which describes their topological relations (adjacencies, inclusions). The overall collected information constitutes a consistent and independent database, generated efficiently on standard workstation. Many applications are possible, such as content based image retrieval, image description and compression, object classification or image-object fusion. A scenario is presented to emphasize interest of the method in the case of 3D visualization enhancement: image-objects are integrated on elevation data (digital elevation models, DEM) in order to generate more realistic rendering.","Title: Model-based despeckling and information extraction from SAR images\nYear: 2000\n\nAbstract: Basic textures as they appear, especially in high res- olution SAR images, are affected by multiplicative speckle noise and should be preserved by despeckling algorithms. Sharp edges between different regions and strong scatterers also must be pre- served. To despeckle images, we use a maximum a posteriori (MAP) estimation of the cross section, choosing between different prior models. The proposed approach uses a Gauss Markov random field (GMRF) model for textured areas and allows an adaptive neigh- borhood system for edge preservation between uniform areas. In order to obtain the best possible texture reconstruction, an expec- tation maximization algorithm is used to estimate the texture pa- rameters that provide the highest evidence. Borders between ho- mogeneous areas are detected with a stochastic region-growing al- gorithm, locally determining the neighborhood system of the Gauss Markov prior. Smoothed strong scatterers are found in the ratio image of the data and the filtering result and are replaced in the image. In this way, texture, edges between homogeneous regions, and strong scatterers are well reconstructed and preserved. Addi- tionally, the estimated model parameters can be used for further image interpretation methods.","Title: Introduction to the Special Section on Image Information Mining for Earth Observation Data\nYear: 2007\n\nAbstract: The eight papers in this special section describe some of the recent advances made in the field of image information mining (IIM) for Earth Observation in the development of tools, methods, and applications.","Title: Image Time-Series Data Mining Based on the Information-Bottleneck Principle\nYear: 2007\n\nAbstract: Satellite image time series (SITS) consist of a time se- quence of high-resolution spatial data. SITS may contain valuable information, but it may be deeply hidden. This paper addresses the problem of extracting relevant information from SITS based on the information-bottleneck principle. The method depends on suitable model selection, coupled with a rate-distortion analysis for determining the optimal number of clusters. We present how to use this method with the Gauss-Markov random fields and the autobinomial random fields model families in order to character- ize the spatio-temporal structures contained in SITS. Experimen- tal results on synthetic data and SITS from SPOT demonstrate the performance of the proposed methodology.","Title: A Similarity Metric for Retrieval of Compressed Objects: Application for Mining Satellite Image Time Series\nYear: 2008\n\nAbstract: This paper addresses the problem of building an index of compressed object databases. We introduce an informational similarity measure based on the coding length of two part codes. Then, we present a methodology for compressing the database by taking into account interobject redundancies and by using the informational similarity measure. The method produces an index included in the code of the data volume. This index is built such that it contains the minimal sufficient information to discriminate the data-volume objects. Then, we present an optimal two-part coder for compressing spatio-temporal events contained in satellite image time series (SITS). The two-part coder allows us to measure similarity and then to derive an optimal index of SITS spatio-temporal events. The resulting index is representative of the SITS information content and enables queries based on information content.","Title: Knowledge centred Earth Observation: Feature Extraction\nYear: 2007\n\nAbstract: Nowadays huge volumes of Earth Observation data are available and increasing every day. Users are faced with the problem of having to extract and interpret information in large volumes of data, with the additional problem of images being annotated only by simple descriptors. Therefore, in this paper we present the Knowledge-centered Earth Observation system which offers interactive probabilistic information mining, among other functionalities, and two Support Vector Machine based Feature Extraction services integrated in the system.","Title: Learning - unlearning for mining high resolution EO images\nYear: 2007\n\nAbstract: The last two decades showed an important development of satellite imagery with past and present satellites acquiring enormous volumes of data. Meanwhile, the quality of the acquired images increased permitting the recording of high resolution images (0.6divide2.5 meters/pixel) in multispectral bands. Thus, both the data volume and the information detail increase dramatically. Consequently, new methods and tools to access and interpret earth observation (EO) images are needed. The present paper presents a semantic search engine for high resolution (HR) EO images based on a hierarchical information model of satellite image contents. To face the potentially ambiguous meaning of image structures depending on their contextual understanding, the search engine uses Bayesian inference to learn categories and a support vector machine (SVM) classifier to assign semantics. The categories are grouping and memorising the semantics of image structures, facilitating their recognition in various contexts. Also the generation of categories helps learning from a small training data set (i.e. image examples); thus, the method is useful for the exploitation of very large data volumes. The concept has enhanced inferred power, therefore optimising the human machine communication (HMC), which is enhanced with learning / unlearning functions.","Title: TerraSAR-X: Complex Image Inversion for Feature Extraction\nYear: 2008\n\nAbstract: In this paper we present two algorithms for information extraction from Single Look Complex (SLC) Synthetic Aperture Radar (SAR) images. The first algorithm is based on Tikhonov regularization with Total Variation (TV) and a Point-Based Feature (PBF) term. Based on the equivalence of Tikhonov and the Bayesian estimate, the second algorithm is a Maximum A Posteriori (MAP) estimation with a complex-valued Gauss-Markov Random Field (GMRF) in addition to the TV prior. The first algorithm produces a despeckled image preserving fine details and texture. The second algorithm gives a denoised image and in addition the estimated feature parameter vector thetas.","Title: Geometrical and Topological Urban Areas Characterization using TerraSAR-X Data\nYear: 2008\n\nAbstract: With the launch of the German TerraSAR-X system in June 2007, a new generation of high-resolution spaceborne synthetic aperture radar (SAR) data is available; which should facilitate the interpretation of urban environments. This article proposes a new automatic tool for geometric and topological urban areas characterization. Our approach is divided into three main steps. First, a bright linear structures detector is applied to extract the geometrical information. Then, a graph-based spatial characterization is used to model the topological relationships between the different detected bright pixels (nodes). Next, a classification by examining the profile of the distributions of the angles between neighboring nodes, is performed in order to label the linked structures. Evaluations of the proposed approach in characterizing the geometry and topology of urban areas were carried out using TerraSAR-X data over three different cities: Las Vegas in the United States, Paris in France and Cairo in Egypt.","Title: Algorithmic Cross-Complexity and Relative Complexity\nYear: 2009\n\nAbstract: Information content and compression are tightly related concepts that can be addressed by classical and algorithmic information theory. Several entities in the latter have been defined relying upon notions of the former, such as entropy and mutual information, since the basic concepts of these two approaches present many common tracts. In this work we further expand this parallelism by defining the algorithmic versions of cross-entropy and relative entropy (or Kullback-Leiblerdivergence), two well-known concepts in classical information theory. We define the cross-complexity of an object x with respect to another object y as the amount of computational resources needed to specify x in terms of y, and the complexity of x related to y as the compression power which is lost when using such a description for x, with respect to its shortest representation. Since the main drawback of these concepts is their uncomputability, a suitable approximation based on data compression is derived for both and applied to real data. This allows us to improve the results obtained by similar previous methods which were intuitively defined.","Title: Wavelet-Based SAR Image Despeckling and Information Extraction, Using Particle Filter\nYear: 2009\n\nAbstract: This paper proposes a new-wavelet-based synthetic aperture radar (SAR) image despeckling algorithm using the sequential Monte Carlo method. A model-based Bayesian approach is proposed. This paper presents two methods for SAR image despeckling. The first method, called WGGPF, models a prior with Generalized Gaussian (GG) probability density function (pdf) and the second method, called WGMPF, models prior with a Generalized Gaussian Markov random field (GGMRF). The likelihood pdf is modeled using a Gaussian pdf. The GGMRF model is used because it enables texture parameter estimation. The prior is modeled using GG pdf, when texture parameters are not needed. A particle filter is used for drawing particles from the prior for different shape parameters of GG pdf. When the GGMRF prior is used, the particles are drawn from prior in order to estimate noise-free wavelet coefficients and for those coefficients the texture parameter is changed in order to obtain the best textural parameters. The texture parameters are changed for a predefined set of shape parameters of GGMRF. The particles with the highest weights represents the final noise-free estimate with corresponding textural parameters. The despeckling algorithms are compared with the state-of-the-art methods using synthetic and real SAR data. The experimental results show that the proposed despeckling algorithms efficiently remove noise and proposed methods are comparable with the state-of-the-art methods regarding objective measurements. The proposed WGMPF preserves textures of the real, high-resolution SAR images well.","Title: Satellite image artifacts detection based on complexity distortion theory\nYear: 2011\n\nAbstract: The artifacts detection is a step of data cleaning process. The classical approach is to predict or determine the existence of defects, to model it, and then design a method to detect and correct them. This classical approach is for specific artifacts. The approach presented in this work is using complexity distortion theory to implement a more generic method, thus, this work will aim at developing parameter free methods able to automatically detect artifacts in EO images. We use the Kolmogorov Structure Function as approximation to the Rate-distortion curve and examine how the artifacts can have the same structure.","Title: Mining large satellite image repositories using semi-supervised methods\nYear: 2011\n\nAbstract: The increasing number and resolution of earth observation (EO) imaging sensors has had a significant impact on both the acquired image data volume and the information content in images. There is consequently a strong need for highly efficient search tools for EO image databases and for search methods to automatically identify and recognize structures within EO images. In this paper, we present a concept for an earth observation image data mining system mixing an auto-annotation component with a category search engine which combines a generic image class search and an object detection feature. The proposed concept relies thus on three distinct components which are detailed successively: in the first part, we describe the auto-annotation component, in the second part, the generic category search engine and in the third part, the object detection tool. In the concluding part of the paper, we provide an insight into how these three components can be related to each other and used in a complementary way to arrive at a system which combines the advantages of both the auto-annotation systems and the category search engines.","Title: Algorithmic Information Theory-Based Analysis of Earth Observation Images: An Assessment\nYear: 2010\n\nAbstract: Earth observation image-understanding methodologies may be hindered by the assumed data models and the estimated parameters on which they are often heavily dependent. First, the definition of the parameters may negatively affect the quality of the analysis. The parameters could not be captured in all aspects, and those resulting superfluous or not accurately tuned may introduce nuisance in the data. Furthermore, the diversity of the data, as regards sensor type, spatial, spectral, and radiometric resolution, and the variety and regularity of the observed scenes make it difficult to establish enough valid and robust statistical models to describe them. This letter proposes algorithmic information theory-based analysis as a valid solution to overcome these limitations. We will present different applications on satellite images, i.e., clustering, classification, artifact detection, and image time series mining, showing the generalization power of these parameter-free data-driven methods based on the computational complexity analysis.","Title: System Design Considerations for Image Information Mining in Large Archives\nYear: 2010\n\nAbstract: Generally, the repository of a remote sensing system needs a capacity of hundreds of terabytes to archive original and processed data. There are several satellites whose downlink rates require ingesting over 100 GB into the archive each day. In order to process such a huge volume of data, to classify them, to index generated classes in the database, and to provide a quick access to stored information at real time, an image information mining system is developed. In this letter, we present the design of a system architecture for processing data from very large image archives.","Title: A survey of density estimation for sar images\nYear: 2012\n\nAbstract: Modeling of synthetic aperture radar (SAR) images has been an important topic of research since the inception of SAR satellites. Many theoretical and empirical models have been presented in literature to accurately model the amplitude SAR images. The method of parameters estimation of the probability density function (PDF) for selected models is another topic of research associated with modeling. Earlier the maximum-likelihood (ML) methodology or the methods of moments (MoM) were used for the parameter estimation. The method of logarithmic-cumulants (MoLC), which has been proposed for the parameter estimation for the PDF defined in R+, is now a very popular tool for the efficient parameter estimation for amplitude SAR images. In this article, we present a survey of some of the well-known PDFs proposed for SAR amplitude images by carrying out the parameter estimation with the MoLC method. The objective is to demonstrate that the statistical characterization of SAR images is strongly dependent upon the observed scene content. Instead of using a set of images, we carry out this study on set of object/texture categories on a larger data-base.","Title: Information Content of Very High Resolution SAR Images: Study of Feature Extraction and Imaging Parameters\nYear: 2013\n\nAbstract: In this paper, we propose to study the dependence of information extraction technique performance on synthetic aperture radar (SAR) imaging parameters and the selected primitive features (PFs). The evaluation is done on TerraSAR-X data, and the interpretation is realized automatically. In the first part of this paper (use case I), the following issues are analyzed: 1) finding the optimal TerraSAR-X products and their limits of variability and 2) retrieving the number of categories/classes that can be extracted from the TerraSAR-X images using the PFs (gray-level co-occurrence matrix, Gabor filters, quadrature mirror filters, and nonlinear short-time Fourier transform). In the second part of this paper (use case II), we investigate the invariance of the products with the orbit direction and incidence angle. On the one hand, the results show that using ascending looking is better than using descending looking with an average accuracy increase of 7%-8%, approximately. On the other hand, the classification accuracy for the incidence angle varies from a lower value of the incidence to an upper value of the incidence angle (depending on the sensor range) with 4%-5%. The test sites are Venice (Italy), Toulouse (France), Berlin (Germany), and Ottawa (Canada) and are covering as much as possible the huge diversity of modes, types, and geometric resolution configuration of the TerraSAR-X. For the evaluation of all these parameters (resolution, features, orbit looking, and incidence angle), the support-vector-machine classifier is considered. To evaluate the accuracy of the classification, the precision/recall metric is calculated. The first contribution of this paper is the evaluation of different PFs (proposed in the literature for different types of images) and adaptation of these for SAR images. These features are compared (based on the accuracy of the classification) for the first time for a multiresolution pyramid specially built for this purpose. During the evaluation,- all the classes were annotated, and a semantic meaning was defined for each class. The second main contribution of this paper is the evaluation of the dependence on the patch size, orbit direction, and incidence angle of the TerraSAR-X. This type of evaluation has not been systematically investigated so far. For the evaluation of the optimal patch, two different patch sizes were defined, with the constrained that the size on ground needs to cover a minimum of one object (e.g., 200 × 200 m on ground). This patch size depends also on the parameters of the data such as resolution and pixel spacing. The investigation of orbit looking and incidence angle is very important for indexing large data sets that has a higher variability of these two parameters. These parameters influence the accuracy of the classification (e.g., if the incidence angle is closer to the lower bounds or closer to the upper bound of the satellite sensor range).","Title: Multitemporal analysis of multisensor data: Information theoretical approaches\nYear: 2010\n\nAbstract: This paper presents two approaches for the analysis of multi-temporal and multisensory data analysis. The first approach focuses on a novel similarity measure, named Mixed Information, derived from information measures and assessment of its performance for multitemporal analysis in change detection. The second approach proposes the use of Kullback-Leibler divergence between marginal distributions, efficiently approximated by cumulant based expansion series. Comparison between mixed information and Kullback-Leibler divergence based change detection is performed to confirm the usefulness of joint metric for analyzing multitemproal and multisensory data. Experimental results obtained confirm justification of the mixed information in multitemporal analysis.","Title: Image information mining methods for exploring and understanding high resolution images\nYear: 2010\n\nAbstract: This paper discusses the basic paradigm of how image information mining methods work in the field of remote sensing. To this end, we compare our approaches to the approaches being used in the world of multimedia; then we discuss the annotation specifics of remote sensing data and describe the different types of remote sensing data that we are faced with today. We conclude with a description of algorithmic alternatives and compare several competing systems that are currently in use. Finally, we provide an outlook of what we expect in the near future.","Title: How many categories are in very high resolution SAR images?\nYear: 2013\n\nAbstract: In this paper, we propose to identify the number of categories that can be retrieved from a very high resolution SAR data. The evaluation is done on TerraSAR-X high resolution Spotlight data and the retrieved categories are semi-automatically annotated using as feature vector the Gabor filters; as a classifier the Support Vector Machine, and for ranking the suggested images the relevance feedback. The visualization of the tool was enhanced compared with our previous implementation in order to support the users in his/her approach to search the patches of interest in a large repository. Our dataset consist in 43 scenes that cover as much as possible all the regions over the world. A total of 352 categories are identified that contain urban and non-urban categories.","Title: The impact of rain, frost, seasonal cycle, and wind on sequences of high resolution urban SAR images\nYear: 2013\n\nAbstract: The TerraSAR-X satellite has been fully operational since nearly 6 years and has delivered a very large quantity of high resolution SAR images. Among these images one can find a number of repeated acquisitions of selected target areas taken with nearly identical imaging parameters. These image time series data often cover the full seasonal cycle of a target area and lend themselves well to small-scale change detection; however, in the case of urban areas, we have to be aware of the quantitative impact of rain, frost and wind on high resolution SAR images. In particular, images of Western European cities are characterized by construction work concentrating on single buildings within a fully built-up city and by public green space changes. A quantitative analysis of urban time series data has to discriminate between definite changes of the urban landscape and the transient impacts of rain, frost and wind. The critical issue is how to identify and characterize these transient phenomena.","Title: Measuring the semantic gap based on a communication channel model\nYear: 2013\n\nAbstract: The collected Earth Observation (EO) data volumes are increasing immensely. In the meantime, the need for retrieval of focused information for decision making is increasing. Due to the particular nature of EO sensors, recording signals very differently than humans perceptual system, the challenges raised by the semantic and sensory gaps are immensely amplified in designing retrieval methods for EO images. This article introduces a method based on communication channel model to quantify and measure the semantic gap, used to assess various feature descriptors for semantic annotation purposes. The approach uses Latent Dirichlet Allocation (LDA), considering images as the source and the semantic topics as the receiver. The parameters of LDA are estimated for computing the Mutual Information to assess latent semantics of feature space. We further introduce a method to measure the distance between humans' and computer's semantics. The results are validated using an SVM-based classifier for an annotated dataset.","Title: Contextual Descriptors for Scene Classes in Very High Resolution SAR Images\nYear: 2012\n\nAbstract: The new generation of spaceborne SAR instruments with meter or submeter resolution finds enormous applications for the observation of urban, industrial, in general of man-made scenes. Thus, targets are not any more observed in isolation, instead the groups of objects, e.g., house, bridge, and road, etc., need to be recognized in their spatial context. This paper proposes a feature extraction method for image patches in order to capture the spatial context. The method is based on the characteristics of the spectra of the SAR data, integrating radiometric, geometric, and texture properties of the SAR image patch. The method is demonstrated for TerraSAR-X High Resolution Spotlight data. To account for the spatial context in which a group of targets is located, it uses an image patch covering typically 200 × 200m2 of the scene. A comparative evaluation of our descriptors and grey-level co-occurrence matrix (GLCM) texture features has been performed over a database of 6916 patches. The method allowed for the robust recognition of over 30 different scene classes, with precision between 50% and 93%. Numerical results show that our method is able to discriminate between scene classes better than GLCM texture parameters.","Title: Scene recognition based on phase gradient InSAR images\nYear: 2014\n\nAbstract: Automated recognition of SAR images requires feature extraction from complex-valued data. In this work, complex-valued interferometric SAR (InSAR) images, which are mainly used to construct elevation models, are proposed for feature extraction. Feature extraction based on the log-cumulants of fractional Fourier transform (FrFT) coefficients has already been proposed in the literature and found to be quite successful for the classification of single-look complex (SLC) SAR images. Here, this method is applied to the complex-valued InSAR and newly introduced complex-valued phase gradient InSAR (PGInSAR) images. In order to evaluate the classification performance for SLC, InSAR and PGInSAR images, a database of bistatic TanDEM-X interferometric pairs is constructed, and a supervised KNN classification is performed. The overall classification accuracies for a 4% training set size show that the use of InSAR and PGInSAR images outperforms SLC images by 15% and 24%, respectively. In terms of individual class accuracies, the biggest improvement is observed in agricultural fields and mixed vegetation areas.","Title: New concepts for earth observation data catalogues\nYear: 2002\n\nAbstract: Conventional satellite data ground segments comprise sensor mode planning, data acquisition systems, data ingestion interfaces, processing capabilities, the catalogue of available data, a data archive, and interfaces for queries and data dissemination. In the present architecture of the systems to access satellite image archives, from the perspective of the information access, the catalogue plays the central role. The catalogue stores the meta information necessary to specify a query. The result of a query is given as a list of images and as quick looks. It is the user who has to browse visually over the quick looks to restrict the result of the query according to his interest. In order to have the possibility to query such catalogues by using the image content, we present a new concept based on an interactive learning of the image semantics, as given by the user.","Title: Semi-supervised learning and discovery of unkown structures among data: Application to satellite image annotation\nYear: 2009\n\nAbstract: In this paper, we present a semi-supervised method for auto-annotating image collections and discovering unknown structures among them. The approach relies on the existence of only a small training database of annotated examples. First, a fully-supervised algorithm using annotated samples is presented. Next, we introduce a semi-supervised procedure which allows us to incorporate unannotated samples and to infer the existence of unknown structures, that is, the existence of new image classes which are not represented in the training database. Finally, we present experimental results from a database of satellite images and briefly mention the possibility of reusing the presented approach as a basis for more complex systems such as Content Based Image Retrieval (CBIR) systems.","Title: Parametric versus non-parametric complex image analysis\nYear: 2009\n\nAbstract: In this paper we compare parametric and non-parametric method for the analysis of complex valued high-resolution SAR data. Gauss-Markov Random Field (GMRF) model with a quadratic energy function as a parametric analysis parameterizes the spectogram of the signal, whereas nonlinear short time Fourier transform (STFT) method, the method based on time frequency analysis (TFA) as a non-parametric approach exploits the signal's non-stationarity in the time-frequency domain for information extraction. This comparative analysis helps to understand, characterize and analyze complex valued SAR data.","Title: A Similarity Measure Using Smallest Context-Free Grammars\nYear: 2010\n\nAbstract: This work presents a new approximation for the Kolmogorov complexity of strings based on compression with smallest Context Free Grammars (CFG). If, for a given string, a dictionary containing its relevant patterns may be regarded as a model, a Context-Free Grammar may represent a generative model, with all of its rules (and as a consequence its own size) being meaningful. Thus, we define a new complexity approximation which takes into account the size of the string model, in a representation similar to the Minimum Description Length. These considerations result in the definition of a new compression-based similarity measure: its novelty lies in the fact that the impact of complexity overestimations, due to the limits that a real compressor has, can be accounted for and decreased.","Title: A Semi-Supervised Algorithm for Auto-Annotation and Unknown Structures Discovery in Satellite Image Databases\nYear: 2010\n\nAbstract: The increasing number and resolution of earth observation (EO) imaging sensors has had a significant impact on both the acquired image data volume and the information content in images. There is consequently a strong need for highly efficient search tools for EO image databases and for search methods to automatically identify and recognize structures within EO images. Content Based Image Retrieval (CBIR) and automatic image annotation systems have been designed to tackle the problem of image retrieval in large image databases. These two systems achieve a common goal, that is to learn the mapping function between low-level visual features and high-level image semantics. A setup, which has hardly been explored in annotating systems and which is the rule rather than the exception, is the case when the training database used to learn the mapping function is not exhaustive regarding semantic classes present in the images. This means that there exists unknown image classes for which there is no training examples in the training database. In this paper, we propose a semi-supervised method for auto-annotating satellite image databases and discovering unknown semantic image classes in these databases. The idea is to incorporate into the learning process the unannotated data which by definition contain the unknown image classes. The latter are considered to be latent structures in the data that appear when we train a hierarchical latent variable model with both the labeled and unlabeled data. We also show that, in our case, the use of unlabeled data leads to more reliable estimates regarding the model parameters. We present experimental results on a synthetic dataset, making a comparison of our algorithm with a semi-supervised Support Vector Machine (S3VM) on this dataset. We also demonstrate the effectiveness of our unknown image classes discovery procedure on a database of SPOT5 satellite images. We show that the results obtained on this database are rather positive since t- - he new structures detected correspond to semantic classes which are not represented in the training database.","Title: Texture estimation in sar images: The impact of scale and model order\nYear: 2010\n\nAbstract: This paper discusses methods and parameter settings that help to estimate texture in SAR images. In general, this is a difficult task for SAR images that are characterized by speckle noise and which span a wide range of pixel magnitudes. We applied Gauss Markov Random Field (GMRF) models and Enhanced Model Based Despeckling (EMBD) to 1 meter resolution amplitude images of the German TerraSAR-X mission. The results demonstrate that one can find appropriate parameter combinations that allow robust texture estimation even for different types of target areas.","Title: Assessment of dimensionality reduction based on communication channel model; application to immersive information visualization\nYear: 2013\n\nAbstract: We are dealing with large-scale high-dimensional image data sets requiring new approaches for data mining where visualization plays the main role. Dimension reduction (DR) techniques are widely used to visualize high-dimensional data. However, the information loss due to reducing the number of dimensions is the drawback of DRs. In this paper, we introduce a novel metric to assess the quality of DRs in terms of preserving the structure of data. We model the dimensionality reduction process as a communication channel model transferring data points from a high-dimensional space (input) to a lower one (output). In this model, a co-ranking matrix measures the degree of similarity between the input and the output. Mutual information (MI) and entropy defined over the co-ranking matrix measure the quality of the applied DR technique. We validate our method by reducing the dimension of SIFT and Weber descriptors extracted from Earth Observation (EO) optical images. In our experiments, Laplacian Eigenmaps (LE) and Stochastic Neighbor Embedding (SNE) act as DR techniques. The experimental results demonstrate that the DR technique with the largest MI and entropy preserves the structure of data better than the others.","Title: Pheromone averaging exploration algorithm\nYear: 2015\n\nAbstract: In this paper we introduce a novel spatial exploration and coverage algorithm based on reflex agents using a pheromone map as storage and communication medium. The algorithm proposed in this paper outperforms many of the popular reflex agent exploration algorithms in terms of exploration performance measured as the cumulative path length.","Title: Deep Learning Earth Observation Classification Using ImageNet Pretrained Networks\nYear: 2016\n\nAbstract: Deep learning methods such as convolutional neural networks (CNNs) can deliver highly accurate classification results when provided with large enough data sets and respective labels. However, using CNNs along with limited labeled data can be problematic, as this leads to extensive overfitting. In this letter, we propose a novel method by considering a pretrained CNN designed for tackling an entirely different classification problem, namely, the ImageNet challenge, and exploit it to extract an initial set of representations. The derived representations are then transferred into a supervised CNN classifier, along with their class labels, effectively training the system. Through this two-stage framework, we successfully deal with the limited-data problem in an end-to-end processing scheme. Comparative results over the UC Merced Land Use benchmark prove that our method significantly outperforms the previously best stated results, improving the overall accuracy from 83.1% up to 92.4%. Apart from statistical improvements, our method introduces a novel feature fusion algorithm that effectively tackles the large data dimensionality by using a simple and computationally efficient approach.","Title: The Semantic Gap: An Exploration of User and Computer Perspectives in Earth Observation Images.\nYear: 2015\n\nAbstract: Research on the semantic gap has considered differences between user and computer image interpretations and proposed methods to bridge it. These methods have been verified by comparing results to reference data or by measuring the degree of user acceptance. Although these methods result in a narrower semantic gap between computers and users, the resulting model for a specific user and search goal ...","Title: Discriminative Nonnegative Matrix Factorization for dimensionality reduction.\nYear: 2016\n\nAbstract: Nonnegative Matrix Factorization (NMF) has been widely used for different purposes such as feature learning, dictionary leaning and dimensionality reduction in data mining and computer vision. In t...","Title: Immersive visualization of visual data using nonnegative matrix factorization.\nYear: 2016\n\nAbstract: Over the last two decades, dimension reduction for visualization has gained a high amount of attention in visual data mining where the data is represented by high-dimensional features. Basically, this approach leads to an unbalanced and occluded distribution of visual data in display space, giving rise to difficulties in browsing the data. In this paper we propose an approach for the visualization of image collections in such a way as (1) images are not occluded by each other, and the provided space is used as much as possible; (2) the similar images are positioned close together; (3) an overview of data is feasible. To fulfill these requirements, we propose to use regularized Nonnegative Matrix Factorization (NMF) controlled by parameters to reduce the dimensionality of data. Experiments performed on optical and radar images confirm the flexibility of proposed method in visualizing large-scale visual data. Finally, an immersive 3D virtual environment is suggested, to visualize the images, to allow the user to navigate and explore the data.","Title: Comparison of Kullback-Leibler divergence approximation methods between Gaussian mixture models for satellite image retrieval\nYear: 2015\n\nAbstract: In many applications, such as image retrieval and change detection, we need to assess the similarity of two statistical models. As a distance measure between two probability density functions, Kullback-Leibler divergence is widely used for comparing two statistical models. Unfortunately, for some models such as Gaussian Mixture Model (GMM), Kullback-Leibler divergence has no analytically tractable formula. We have to resort to approximation methods. In this paper, we compare seven methods, namely Monte Carlo method, matched bond approximation, product of Gaussian, variation-al method, unscented transformation, Gaussian approximation, and min-Gaussian approximation, for approximating the Kullback-Leibler divergence between two Gaussian mixture models for satellite image retrieval. Two image retrieval experiments based on two publicly available datasets have been performed. The comparison is carried out in terms of both retrieval performance and computational time.","Title: Polarimetric Sar Data Feature Selection Using Measures Of Mutual Information\nYear: 2015\n\nAbstract: Several algorithms for polarimetric synthetic aperture radar (PolSAR) data indexing and classification were proposed in the state of the art literature. In particular, one of them computes powerful, compact feature descriptors composed of the first three logarithmic cumulants of the BiQuaternion Fractional Fourier Transform (BiQFrFT) coefficients of PolSAR patches. Since the BiQFrFT of each patch is computed at three different angles, the algorithm's result consists in nine complex-valued features (18 real-valued features) for single polarization images and in nine biquaternion-valued features (72 real-valued features) for fully polarimetric images. In this paper feature selection based on mutual information (MI) is employed to optimally select a subset of features, in order to improve the indexing performances and to minimize the classification error. The improved results are shown on two polarimetric images: a L-band PALSAR image over Danube's Delta, Romania and a C-band RadarSAT2 image over Braila, Romania.","Title: A Comparison Of Bag-Of-Words Method And Normalized Compression Distance For Satellite Image Retrieval\nYear: 2015\n\nAbstract: Recently, two improved methods have shown their advantages in browsing Earth Observation (EO) dataset. The first method is the Bag-of-Words (BoW) feature extraction method and the second is the Normalized Compression Distance (NCD) for assessing image similarity. However, they have not been compared so far for satellite image retrieval, which motivates this paper. Two retrieval experiments have been performed on a freely available optical image dataset and a SAR image dataset. Through these two experiments, we conclude that the BoW method performs generally better than NCD. Although it is a parameter-free solution for data mining, NCD only performs well for images with repetitive patterns like some homogeneous classes. In contrast, BoW method performs much far beyond that of NCD. In addition, NCD is computationally very expensive, which makes it infeasible to be applied in real applications. In contrast, BoW method is more realistic in practical applications in terms of both accuracy and computation.","Title: Interactive Feature Learning From Sar Image Patches\nYear: 2015\n\nAbstract: Feature learning algorithms aim to provide a compact and discriminative representation of complex datasets in order to increase the speed and accuracy of clustering or classification. In this paper, we propose a novel interactive feature learning approach which is mainly based on 3D interactive data visualization and Non-negative Matrix Factorization (NMF). Here, the data is visualized in a 3D interface to support human-data interaction. The user interactions are exploited in an NMF framework to learn a compact representation of the data. The conducted experiments on Synthetic Aperture Radar (SAR) images confirm the efficiency of the proposed approach.","Title: Gpu-Based Kernelized Locality-Sensitive Hashing For Satellite Image Retrieval\nYear: 2015\n\nAbstract: As the data acquisition capabilities of Earth observation (EO) satellites have been improved substantially in the past few years, large amount of high-resolution satellite images are downlinked continuously to ground stations. Such amount of data increases rapidly beyond the users' capability to access the images' content in reasonable time. Hence, automatic and fast interpretation of a large data volume is a computationally intensive task. Recently, approximate nearest neighbour search has been used for content-based image retrieval in sub-linear time. Kernelized locality sensitive hashing (KLSH) is a well-known approximate method, which has recently shown promising results for fast remote sensing image retrieval. This paper proposes a novel parallelization of KLSH using Graphical Processing Units (GPU), in order to perform fast parallel image retrieval. The proposed method was tested on high-dimensional feature vectors from two satellite-based image datasets, where an average speedup of 20 times was achieved.","Title: Semantic Interpretation Of Multi-Level Change Detection In Multi-Temporal Satellite Images\nYear: 2015\n\nAbstract: Satellite image time series are a valuable resource for enhancing land exploitation by respecting the natural cycles, analyzing urban expansion and its positive and negative effects, limiting the unhealthy rhythm of deforestation, understanding natural hazards and so on. In this context, understanding only the changes in multitemporal images is not sufficient. This paper aims to correlate multi-level change detection techniques with image semantic segmentation methods in order to build an hierarchy of changes for each semantic class. In this way, we are able to provide statistics regarding the levels of change suffered by a certain area. The methods are demonstrated with examples involving bi-temporal Landsat images.","Title: A Hierarchical Patch Clustering Method For High-Resolution Terrasar-X Images\nYear: 2015\n\nAbstract: In this paper, we present a Gaussian test-based hierarchical clustering method for high-resolution TerraSAR-X images. The purpose is to obtain homogeneous clusters. k-means is used to split image features to create a hierarchical structure. As image feature vectors usually fall into high dimensional feature space, we test different distance metrics, in order to try to tackle the curse of dimensionality problem. With prepared datasets, we evaluate the clustering results by defining a homogeneity percentage. The results show that by using Gabor texture feature, the Gaussian test-based hierarchical patch clustering method is able to obtain homogeneous clusters. Meanwhile, fractional distance or Minkowski distance performs better than Euclidean or Manhatten distance.","Title: Feature Space Dimensionality Reduction For The Optimization Of Visualization Methods\nYear: 2015\n\nAbstract: Visual data mining methods are of great importance in exploratory data analysis having a high potential for mining large databases. As the data feature space is generally ndimensional, visual data mining relies on dimensionality reduction techniques. This is the case for image feature spaces which can be visualized by giving each data point a location in a three dimensional space. This paper aims to present a comparative study of several dimensionality reduction methods considering as input image feature spaces, in order to detemine an optimal visualization method to illustrate the separation of the classes. At the beginning, to check the performance of the envisaged method, an artificial dataset consisting of random vectors describing six, 20-dimensional Gaussian distributions with spaced means and low variances was generated. Further, two real images datasets are used to evaluate the contributions of dimensionality reduction algorithms related to data visualization. The analysis focuses on the PCA, LDA and t-SNE dimensionality reduction techniques. Our tests are performed on images for which the computed features include the color histogram and Weber descriptors.","Title: Lucas Visual Browser: A Tool For Land Cover Visual Analytics\nYear: 2015\n\nAbstract: In this paper we present the LUCAS Visual Browser system, a tool for land cover visual analytics. The system implements different web technologies in a multilayer server-client architecture in order to allow the user to visually analyse land cover heterogeneous information. The information manage is composed of EO multispectral and SAR products along with the multitemporal in situ LUCAS surveys. The fusion of these data provides a very useful information during the EO scene interpretation process. Furthermore, the system offers interactive tools for the detection of optimal datasets for EO multitemporal image change detection, providing at the same time ground truth points for both, human and machine analysis.","Title: Information extraction from high resolution SAR data\nYear: 2002\n\nAbstract: The article aims at presenting the most recent developments and trends in information extraction from Synthetic Aperture Radar SAR data.","Title: Dictionary-Based Compact Data Representation for Very High Resolution Earth Observation Image Classification\nYear: 2015\n\nAbstract: In the context of fast growing data archives, with continuous changes in volume and diversity, information mining has proven to be a difficult, yet highly recommended task. The first and perhaps the most important part of the process is data representation for efficient and reliable image classification. This paper is presenting a new approach for describing the content of Earth Observation Very High Resolution images, by comparison with traditional representations based on specific features. The benefit of data compression is exploited in order to express the scene content in terms of dictionaries. The image is represented as a distribution of recurrent patterns, removing redundant information, but keeping all the explicit features, like spectral, texture and context. Further, a data domain analysis is performed using Support Vector Machine aiming to compare the influence of data representation to semantic scene annotation. WorldView2 data and a reference map are used for algorithm evaluation.","Title: Feature Extraction for Patch-Based Classification of Multispectral Earth Observation Images.\nYear: 2016\n\nAbstract: Recently, various patch-based approaches have emerged for high and very high resolution multispectral image classification and indexing. This comes as a consequence of the most important particularity of multispectral data: objects are represented using several spectral bands that equally influence the classification process. In this letter, by using a patch-based approach, we are aiming at extrac...","Title: Dimensionality Reduction for Visual Data Mining of Earth Observation Archives.\nYear: 2016\n\nAbstract: Modern knowledge discovery systems, empowered by visual data exploration techniques, enable the user to discover and understand the data content. Considering patch-level processing, the visual exploration of Earth Observation archives aims to identify groups of items sharing similar semantic content. Each patch is further represented by certain descriptors, i.e., spectral signatures or Weber local...","Title: The Earth-Observation Epitome: A New Interactive Value-Added Product.\nYear: 2016\n\nAbstract: Typical applications based on remote sensing imagery start with a visual inspection and selection of scenes and regions of interest to be purchased. In this letter, we propose a so-called Earth observation (EO) epitome as a novel value-added interactive product. Basically, it comprises several image tiles, primitive features, metadata entries, a quick look of the image tiles, and semantic annotati...","Title: Simplified Bistatic Sar Imaging With A Fixed Receiver And Terrasar-X As Transmitter Of Opportunity - First Results\nYear: 2016\n\nAbstract: This paper presents the first SAR imaging results obtained with a fixed ground-based system used in bistatic configuration having the TerraSAR-X satellite as transmitter of opportunity. The system's characteristics and signal processing flow are presented relative to the state of the art. Compared to previous works on bistatic SAR imaging where a significant amount of processing is dedicated to time/frequency synchronization between the satellite transmitter and ground receiver, we show that a bistatic SAR image with only meter-range geographic offset can be obtained using the state vectors from a monostatic image and minimal synchronization efforts consisting in acquisitions triggered by an amplitude threshold and stabilization of the local oscillator with a GPS-disciplined reference. We present the first bistatic image in ground geometry obtained over an area of Bucharest and compare it with a monostatic image refocused on the same grid.","Title: A Convolutional Deep Belief Network For Polarimetric Sar Data Feature Extraction\nYear: 2016\n\nAbstract: This paper proposes a custom convolutional deep belief network for polarimetric synthetic aperture radar (PolSAR) data feature extraction. The proposed architecture stands out through the interesting features it shows, starting with the fact that it is adapted to fully polarimetric SAR data. Then, the multilayer approach allows the stepwise discovery of higher-level features. The convolutional approach allows the discovery of local, spatially invariant features and makes the architecture scalable to fully sized PolSAR images. The network is trained in an unsupervised manner, without using labeled data and then it succeeds to extract powerful features from PolSAR patches. This fact is demonstrated by applying supervised and unsupervised classification algorithms on features extracted from patches of a fully polarimetric multi-look F-SAR image over Kaufbeuren airfield, Germany.","Title: Compound And Configurable Framework For Exploratory Earth Observation Data Analysis\nYear: 2016\n\nAbstract: The lack of a comprehensive solution for image information mining has often brought confusion and misunderstanding when Earth Observation data based application scenarios were addressed. Considering the variety of dedicated sensors available nowadays, the particularities of the recorded data raises serious issues when explored. Most of the proposed methodologies for data analysis integrate algorithms able to cope with single cases. In order to overcome this limitation, the present paper introduce a compound, configurable framework containing two processing levels, for feature extraction and image classification, that allows different settings depending on the application being handled. The design was proposed such that it facilitates the integration of several methods and algorithm for each level, including a module to serve for validation when reference data is available. The approach is not complete without the interaction with the user, therefore, a human-machine communication strategy was also developed. The validation was performed through a prototype system meeting all the criteria of the defined framework.","Title: Applications Of Sar Tomography On Persistent Scatterers Detection, Based On Beam-Forming Filtering\nYear: 2016\n\nAbstract: The main advantage of Synthetic Aperture Radar (SAR) Tomography over classical interferometry consists in capacity to detect the presence of multiple scatterers within the same resolution cell. In this paper we present an algorithm for detection of Persistent Scatterers (PS) based on the variation of reflectivity function in elevation direction reconstructed with Beam-Forming technique. Then, we extract the contribution of dominant scatterers from each resolution cell and reapply the detection algorithm to identify thr presence of secondary PSs in scene's pixels. The methods were implemented on a dataset of high-resolution Single Look Complex (SLC) images acquired by TerraSAR-X sensor, on which the residual component of interferometric phase was estimated and compensated locally.","Title: A Bistatic Sar Simulator For Ground-Based Fixed-Receiver Geometry\nYear: 2016\n\nAbstract: In this paper we present the early development of a SAR simulator for ground-based fixed-receiver bistatic geometry. Firstly, we describe the assumptions the simulator is based on, then a short presentation of the mathematical model that lays behind the simulator is given. Furthermore, we give some details about its implementation and finally, we present some results.","Title: A Dimensionality Reduction Approach For The Visualization Of The Cluster Space: A Trustworthiness Evaluation\nYear: 2016\n\nAbstract: The data mining systems solve the problem of handling Earth Observation archives counting on a feature vectors based description of the data. Increasing the dimensionality of the feature vectors would offer an effective perspective of the dataset's content. The modern systems provide visual exploration of data projecting their high-dimensional feature space in a 3-D space. The dimensionality reduction methods represent the main way to achieve such representation. Several dimensionality reduction methods have been proposed to identify the mapping, bot not all of them retain the same dataset properties. In order to compare their performance, the development of formal measures like \"Trustworthiness\" or the measures based on Co-ranking matrix was mandatory. These measures objectively evaluate the similarity between the structure detected in the original and the reduced space. In this paper we evaluate six dimensionality reduction methods using \"Trustworthiness\" and \"Continuity\" measures. In this regard three datasets have been used: an artificial one and two remote sensing datasets. Each of them have been described by a high-dimensional feature space.","Title: Visual Data Mining For Feature Space Exploration Using In-Situ Data\nYear: 2016\n\nAbstract: In this paper, we present the visualization of image databases based on their primitive features. Our approach is to have a visual navigation tool for allowing the exploration and exploitation of large image archives. The tool is able to project the content of a given image database based on the primitive feature space and to provide interaction between the final user and the huge amount of data. Land Use/Land Cover area frame statistical Survey in-situ data are used as test dataset. Bag-of-Words and Weber Local Descriptors are used as primitive features.","Title: Discovery of Semantic Relationships in PolSAR Images Using Latent Dirichlet Allocation.\nYear: 2017\n\nAbstract: We propose a multilevel semantics discovery approach for bridging the semantic gap when mining high-resolution polarimetric synthetic aperture radar (PolSAR) remote sensing images. First, an Entropy/Anisotropy/Alpha-Wishart classifier is employed to discover low-level semantics as classes representing the physical scattering properties of targets (e.g., low-entropy/surface scattering/high anisotro...","Title: Exploiting Deep Features for Remote Sensing Image Retrieval: A Systematic Investigation.\nYear: 2017\n\nAbstract: Remote sensing (RS) image retrieval based on visual content is of great significance for geological information mining. Over the past two decades, a large amount of research on this task has been carried out, which mainly focuses on the following three core issues of image retrieval: visual feature, similarity metric and relevance feedback. Along with the advance of these issues, the technology of RS image retrieval has been developed comparatively mature. However, due to the complexity and multiformity of high-resolution remote sensing (HRRS) images, there is still room for improvement in the current methods on HRRS data retrieval. In this paper, we analyze the three key aspects of retrieval and provide a comprehensive review on content-based RS image retrieval methods. Furthermore, for the goal to advance the state-of-the-art in HRRS image retrieval, we focus on the visual feature aspect and delve how to use powerful deep representations in this task. We conduct systematic investigation on evaluating factors that may affect the performance of deep features. By optimizing each factor, we acquire remarkable retrieval results on publicly available HRRS datasets. Finally, we explain the experimental phenomenon in detail and draw instructive conclusions according to our analysis. Our work can serve as a guiding role for the research of content-based RS image retrieval.","Title: Evaluation of Dimensionality Reduction Methods for Remote Sensing Images Using Classification and 3D Visualization.\nYear: 2017\n\nAbstract: Visual exploration is a natural way to understand the content of a data archive. If the data are multidimensional, the dimensionality reduction is an appropriate preprocessing step before the data visualization. In literature, two types of approaches are devoted to dimensionality reduction: feature selection and feature extraction algorithms. Both techniques intend to project a high dimensional space into a new one, with reduced dimensionality, preserving data inherent information. This paper aims to identify the similarity degree between low and high dimensional representations of a data archive using the optimal number of semantic classes as a criterion. This number is estimated based on the rate distortion theory being computed both before and after dimensionality reduction. The projection of the low dimensional space was obtained using one feature selection and six feature extraction methods.","Title: Artificial Generation of Big Data for Improving Image Classification: A Generative Adversarial Network Approach on SAR Data.\nYear: 2017\n\nAbstract: Very High Spatial Resolution (VHSR) large-scale SAR databases are still an unresolved issue in the RemoteSensing field. In this work, we propose such a dataset anduse it to explore patch-based classification in urban and periurban areas, considering 7 distinct semantic classes. In this context, we investigate the accuracy of large CNN classification models and pre-trained networks for SAR imaging systems.Furthermore, we propose a Generative Adversarial Network(GAN) for SAR image generation and test, whether thesynthetic data can actually improve classification accuracy.","Title: Image representation alternatives for the analysis of satellite image time series\nYear: 2017\n\nAbstract: Current satellite images and image time series provide us with detailed information about the state of our planet as well as about our technical infrastructure and human activities. These images allow us to learn more about local, regional, and global phenomena and events, including - if interpreted properly - their causes and effects. In particular, image time series provide specific information about the dynamics of many processes implicitly contained in our images that need to be unearthed and investigated in detail. A traditional approach towards this aim is to start with pixel-level or patch-level data analysis for pixel-based image analysis, followed, if necessary, by subsequent feature extraction, clustering, classification and semantic labelling in order to generate various types of change maps on different representation levels. The classification step can be supported by interactive human intervention, or by automated machine learning strategies to identify higher level objects and their spatial and temporal relationships. The detected relationships can then be formulated as parameterized rule sets that create higher-level descriptor sets of the content of the selected images, and of additional external data such as thematic maps or typical dynamics descriptions. As an innovative extension of this traditional concept, we propose a highly automated approach for application-adapted image content exploration and knowledge extraction. The reason for this strategy is the additional amount and the precision of semantic relationships and details that we can assign to an image time series once we know the final application field and how to embed and access image content within knowledge graphs.","Title: Temporal analysis of SAR imagery for permanent and evolving Earth land cover behavior assessment\nYear: 2017\n\nAbstract: In the era of constantly increasing Earth Observation (EO) data collections, information extraction and data analysis should be enhanced with a multi-temporal component enabled by the temporal resolution of satellite missions and create handy, yet powerful tools for those applications involving monitoring of land cover. The image time series, as results of the satellite revisiting period, gives you insights not only on a certain area, but also on its representation at different moments of time. In order to limit the issues that might arise due to irregular time sampling of multispectral data, the authors propose a Synthetic Aperture Radar (SAR) image time series for analysis. To this point, the main goal is to mine the satellite image time series (SITS) for understanding the temporal behaviour of an area in terms of evolution and persistency. The paper introduces an analytical approach, combining coherent and no coherent analysis of SAR SITS content. We propose the Latent Dirichlet Allocation model to extract categories of evolution from the SAR SITS and techniques which study statistical and coherent proprieties of the targets to identify the structures with stable electromagnetic characteristics over time, named Persistent Scatterers (PS). The obtained results indicate an evolutionary character hidden inside the persistent class. The results obtained on 30 ERS images encourages further analysis on Sentinel 1 data.","Title: Land cover change detection in Satellite Image Time Series using an active learning method\nYear: 2017\n\nAbstract: Some of the first Earth Observation (EO) missions date back to the 1970s. Over the time, large datasets of Satellite Image Time Series (SITS) have been used to identify and monitor land cover evolutions. The processing complexity increases proportionally to the time span of the Earth Observation (EO) series. Because of the SITS dataset complexity and variety of contained evolution patterns, most unsupervised classification methods fail to detect and isolate the user's evolution pattern of interest. This is usually caused by the discrepancy between automatically extracted low-level features and high-level meaning assigned by the user who searches for a specific evolution. In an effort to find a solution for this difficult task, this paper presents a SVM based active learning method for the extraction of specific evolution classes from SITS. Several experiments were conducted on a dataset composed of Landsat 4 TM (Thematic Mapper) and Landsat 5 TM acquisitions over Bucharest, Romania, in the time interval of 1984-1993.","Title: Land-cover evolution class analysis in Image Time Series of Landsat and Sentinel-2 based on Latent Dirichlet Allocation\nYear: 2017\n\nAbstract: Satellite Image Time Series (SITS) are widely used in monitoring the Earth's changes for various applications such as land-cover evolution analysis. In this paper, we propose an approach based on Latent Dirichlet Allocation (LDA) which considers spatial and spectral information to measure the land-cover changes in multispectral SITS. For our experiments, we focus on the vegetation dynamics of the Doñana National Park (in southwestern Spain) using a Landsat and a Sentinel-2 SITS dataset. The proposed approach represents each image by Normalized Difference Vegetation Index (NDVI) and tiles it into smaller patches. The patches are then modeled as Bag-of-Words (BoW) and LDA is applied to them in order to discover the latent structure of the image. The divergence between the latent structures of any two consecutive images is then considered as the measure of change. Results show that the changes measured by the proposed approach can represent the vegetation dynamics of the region of interest. Moreover, comparing the results obtained from the two datasets demonstrates that using high-level information allows the proposed approach to measure the changes independent of the sensor. This will support long-term monitoring through combining various available data.","Title: Multisensor Earth Observation Image Classification Based on a Multimodal Latent Dirichlet Allocation Model.\nYear: 2018\n\nAbstract: Many previous researches have already shown the advantages of multisensor land-cover classification. Here, we propose an innovative land-cover classification approach based on learning a joint latent model of synthetic aperture radar (SAR) and multispectral satellite images using multimodal latent Dirichlet allocation (mmLDA), a probabilistic generative model. It has already been successfully appl...","Title: Understanding Heterogeneous EO Datasets: A Framework for Semantic Representations.\nYear: 2018\n\nAbstract: Earth observation (EO) has become a valuable source of comprehensive, reliable, and persistent information for a wide number of applications. However, dealing with the complexity of land cover is sometimes difficult, as the variety of EO sensors reflects in the multitude of details recorded in several types of image data. Their properties dictate the category and nature of the perceptible land structures. The data heterogeneity hampers proper understanding, preventing the definition of universal procedures for content exploitation. The main shortcomings are due to the different human and sensor perception on objects, as well as to the lack of coincidence between visual elements and similarities obtained by computation. In order to bridge these sensory and semantic gaps, the paper presents a compound framework for EO image information extraction. The proposed approach acts like a common ground between the user's understanding, who is visually shortsighted to the visible domain, and the machines numerical interpretation of a much wider information. A hierarchical data representation is considered. At first, basic elements are automatically computed. Then, users can enforce their judgement on the data processing results until semantic structures are revealed. This procedure completes a user-machine knowledge transfer. The interaction is formalized as a dialogue, where communication is determined by a set of parameters guiding the computational process at each level of representation. The purpose is to maintain the data-driven observable connected to the level of semantics and to human awareness. The proposed concept offers flexibility and interoperability to users, allowing them to generate those results that best fit their application scenario. The experiments performed on different satellite images demonstrate the ability to increase the performances in case of semantic annotation by adjusting a set of parameters to the particularities of the analyzed data.","Title: On the Use of Normalized Compression Distances for Image Similarity Detection.\nYear: 2018\n\nAbstract: This paper investigates the usefulness of the normalized compression distance (NCD) for image similarity detection. Instead of the direct NCD between images, the paper considers the correlation between NCD based feature vectors extracted for each image. The vectors are derived by computing the NCD between the original image and sequences of translated (rotated) versions. Feature vectors for simple transforms (circular translations on horizontal, vertical, diagonal directions and rotations around image center) and several standard compressors are generated and tested in a very simple experiment of similarity detection between the original image and two filtered versions (median and moving average). The promising vector configurations (geometric transform, lossless compressor) are further tested for similarity detection on the 24 images of the Kodak set subject to some common image processing. While the direct computation of NCD fails to detect image similarity even in the case of simple median and moving average filtering in 3 x 3 windows, for certain transforms and compressors, the proposed approach appears to provide robustness at similarity detection against smoothing, lossy compression, contrast enhancement, noise addition and some robustness against geometrical transforms (scaling, cropping and rotation).","Title: Dialectical GAN for SAR Image Translation: From Sentinel-1 to TerraSAR-X.\nYear: 2018\n\nAbstract: With more and more SAR applications, the demand for enhanced high-quality SAR images has increased considerably. However, high-quality SAR images entail high costs, due to the limitations of current SAR devices and their image processing resources. To improve the quality of SAR images and to reduce the costs of their generation, we propose a Dialectical Generative Adversarial Network (Dialectical GAN) to generate high-quality SAR images. This method is based on the analysis of hierarchical SAR information and the dialectical structure of GAN frameworks. As a demonstration, a typical example will be shown, where a low-resolution SAR image (e.g., a Sentinel-1 image) with large ground coverage is translated into a high-resolution SAR image (e.g., a TerraSAR-X image). A new algorithm is proposed based on a network framework by combining conditional WGAN-GP (Wasserstein Generative Adversarial NetworkGradient Penalty) loss functions and Spatial Gram matrices under the rule of dialectics. Experimental results show that the SAR image translation works very well when we compare the results of our proposed method with the selected traditional methods.","Title: Moving Ship Velocity Estimation Using TanDEM-X Data Based on Subaperture Decomposition.\nYear: 2018\n\nAbstract: In this letter, a velocity estimation method for moving ships in synthetic aperture radar (SAR) images is proposed based on a subaperture decomposition technique. In contrast to traditional methods, our method needs only a few SAR imaging parameters besides the SAR image itself. The behavior of moving ships in subaperture images is theoretically analyzed, and the ship motion parameters in azimuth ...","Title: Joint SAR Image Time Series and PSInSAR Data Analytics: An LDA Based Approach.\nYear: 2018\n\nAbstract: Due to the constant increase in Earth Observation (EO) data collections, the monitoring of land cover is facilitated by the temporal diversity of the satellite images datasets. Due to the capacity of Synthetic Aperture Radar (SAR) sensors to operate independently of sunlight and weather conditions, SAR image time series offer the possibility to form a dataset with almost regular temporal sampling. This paper aims at mining the SAR image time series for an analysis of target's behavior from the perspective of both temporal evolution and coherence. The authors present a two-level analytical approach envisaging the assessment of global (related to perceivable structures on the ground) and local (related to changes occurred within a perceivable structure on the ground) evolution inside the scene. The Latent Dirichlet Allocation (LDA) model is implemented to identify the categories of evolution present in the analyzed scene, while the statistical and coherent proprieties of the dataset's images are exploited in order to identify the structures with stable electromagnetic response, the so-called Persistent Scatterers (PS). A comparative study of the two algorithms' classification results is conducted on ERS and Sentinel-1 data. At global scale, the results fit human perception, as most of the points which can be exploited for Persistent Scatterers Interferometry (PS-InSAR) are classified within the same class, referring to stable structures. At local scale, the LDA classification demands for an extended number of classes (defined through a perplexity-based analysis), enabling further differentiation inside the evolutional character of those stable structures. The comparison against the map of detected PS reveals which classes present higher temporal correlation, determining a stable evolutionary character, opening new perspectives for validation of both PS detection and SITS analysis algorithms.","Title: Bayesian estimation of generalized Gamma mixture model based on variational EM algorithm.\nYear: 2019\n\nAbstract: •We propose a VEM algorithm for the inference of the generalized Gamma mixture model (GMM) with all the closed-form update equations.•The help function is used to approximate the lower bound of the variational objective function in GMM.•With the proposed VEM algorithm, the effective number of components as well as all the other underlying parameters in GMM can be estimated automatically and simultaneously.•The results reveal that the proposed inference method is more efficient than the standard EM algorithm.•The experimental results show that the GMM is more appropriate for the asymmetric and heavy-tailed data than Gaussian mixture model.","Title: ICPR2018 Contest on Object Detection in Aerial Images (ODAI-18)\nYear: 2018\n\nAbstract: Object detection in aerial images plays a significant role in intelligent interpretation of aerial images. Hence many effective methods, especially the new-generation data-driven methods, have been developed for this task. Here, we hold the ODAI, a new contest that focused on object detection in aerial images, based on a new large-scale aerial image dataset called DOTA [1]. This contest contains over 3000 large-size images ( 4k×4k pixels), which cover 211,581 instances divided into 15 categories. Each instance is labeled by an arbitrary (8 d.o.f.) quadrilateral. Besides, we propose two tasks for this contest, named object detection with the horizontal bounding box (OD-HBB) and object detection with the oriented bounding box (OD-OBB). The contest was opened on February 7, 2018, and ended on April 30, 2018. A website is open to the public, which provides links to download data and evaluation server. We have totally received 60 registrations. There are 8 teams that have successfully submitted results on the OD-HBB task with the top mAP as 0.719, and 9 teams that have successfully submitted results on the OD-OBB task with the top mAP as 0.705. Through the contest, we hope to draw extensive attention from a wide range of communities and call for more future research and efforts for the task of object detection in aerial images.","Title: Pulse Radar with Field-Programmable Gate Array Range Compression for Real Time Displacement and Vibration Monitoring.\nYear: 2019\n\nAbstract: This paper aims to present the basic functionality of a radar platform for real time monitoring of displacement and vibration. The real time capabilities make the radar platform useful when live monitoring of targets is required. The system is based on the RF analog front-end of a USRP, and the range compression (time-domain cross-correlation) is implemented on the FPGA included in the USRP. Further processing is performed on the host computer to plot real time range profiles, displacements, vibration frequencies spectra and spectrograms (waterfall plots) for long term monitoring. The system is currently in experimental form and the present paper aims to prove its functionality. The precision of this system is estimated (using the 3 sigma approximation) at 0.6 mm for displacement measurements and 1.8 mm for vibration amplitude measurements.","Title: Multilayer Architecture for Heterogeneous Geospatial Data Analytics: Querying and Understanding EO Archives.\nYear: 2017\n\nAbstract: The constantly growing process of the Earth Observation (EO) data and their heterogeneity require new systems and tools for effectively querying and understanding the available data archives. In this paper, we present a tool for heterogeneous geospatial data analytics. The system implements different web technologies in a multilayer server-client architecture, allowing the user to visually analyze...","Title: Mining Multitemporal In Situ Heterogeneous Monitoring Information for the Assurance of Recorded Land Cover Changes.\nYear: 2017\n\nAbstract: We present a data mining methodology to filter and validate land cover change detections obtained from multitemporal in situ surveys. As in situ data we use the measurements from the European land use and coverage area frame survey (LUCAS), which provides images with standardized metadata about land cover and land use within the whole territory of the European Union. Multitemporal LUCAS surveys pr...","Title: Earth Observation Image Semantic Bias: A Collaborative User Annotation Approach.\nYear: 2017\n\nAbstract: Correctly annotated image datasets are important for developing and validating image mining methods. However, there is some doubt regarding the generalizability of the models trained and validated on available datasets. This is due to dataset biases, which occur when the same semantic label is used in different ways across datasets, and/or when identical object categories are labeled differently a...","Title: Quantitative Evaluation of the Feature Space Transformation Methods Used for Applications of Visual Semantic Clustering of EO Images.\nYear: 2017\n\nAbstract: Data visualization guides the process of indexing and retrieval, strengthening the link between low-level image features and high-level human understanding of image content. In this regard, we have described the semantic content of a multidimensional dataset using its descriptors to derive high-dimensional feature spaces. The dimensionality of these spaces is further reduced to three in order to p...","Title: Super-Resolution Multi-Look Detection in SAR Tomography.\nYear: 2018\n\nAbstract: Synthetic Aperture Radar (SAR) Tomography (TomoSAR) allows extending the 2-D focusing capabilities of SAR to the elevation direction, orthogonal to the azimuth and range. The multi-dimensional extension (along the time) also enables the monitoring of possible scatterer displacements. A key aspect of TomoSAR is the identification, in the presence of noise, of multiple persistent scatterers interfering within the same 2-D (azimuth range plane) pixel. To this aim, the use of multi-look has been shown to provide tangible improvements in the detection of single and double interfering persistent scatterers at the expense of a minor spatial resolution loss. Depending on the system acquisition characteristics, this operation may require also the detection of multiple scatterers interfering at distances lower than the Rayleigh resolution (super-resolution). In this work we further investigated the use of multi-look in TomoSAR for the detection of multiple scatterers located also below the Rayleigh resolution. A solution relying on the Capon filtering was first analyzed, due to its improved capabilities in the separation of the responses of multiple scatterers and sidelobe suppression. Moreover, in the framework of the Generalized Likelihood Ratio Test (GLRT), the single-look support based detection strategy recently proposed in the literature was extended to the multi-look case. Experimental results of tests carried out on two datasets acquired by TerraSAR-X and COSMO-SkyMED sensors are provided to show the performances of the proposed solution as well as the effects of the baseline span of the dataset for the detection capabilities of interfering scatterers.","Title: Ranking evolution maps for Satellite Image Time Series exploration: application to crustal deformation and environmental monitoring\nYear: 2019\n\nAbstract: Satellite Image Time Series (SITS) are large datasets containing spatiotemporal information about the surface of the Earth. In order to exploit the potential of such series, SITS analysis techniques have been designed for various applications such as earthquake monitoring, urban expansion assessment or glacier dynamic analysis. In this paper, we present an unsupervised technique for browsing SITS in preliminary explorations, before deciding whether to start deeper and more time consuming analyses. Such methods are lacking in today’s analyst toolbox, especially when it comes to stimulating the reuse of the ever growing list of available SITS. The method presented in this paper builds a summary of a SITS in the form of a set of maps depicting spatiotemporal phenomena. These maps are selected using an entropy-based ranking and a swap randomization technique. The approach is general and can handle either optical or radar SITS. As illustrated on both kinds of SITS, meaningful summaries capturing crustal deformation and environmental phenomena are produced. They can be computed on demand or precomputed once and stored together with the SITS for further usage.","Title: Semi-supervised Hierarchical Clustering for Semantic SAR Image Annotation.\nYear: 2016\n\nAbstract: In this paper, we propose a semi-automated hierarchical clustering and classification framework for synthetic aperture radar (SAR) image annotation. Our implementation of the framework allows the classification and annotation of image data ranging from scenes up to large satellite data archives. Our framework comprises three stages: 1) each image is cut into patches and each patch is transformed i...","Title: Application and Evaluation of a Hierarchical Patch Clustering Method for Remote Sensing Images.\nYear: 2016\n\nAbstract: In this paper, we apply and evaluate a modified Gaussian-test-based hierarchical clustering method for high-resolution satellite images. The purpose is to obtain homogeneous clusters within each hierarchy level which later allow the classification and annotation of image data ranging from single scenes up to large satellite data archives. After cutting a given image into small patches and feature ...","Title: Immersive Interactive SAR Image Representation Using Non-negative Matrix Factorization.\nYear: 2016\n\nAbstract: Earth observation (EO) images clustering is a challenging problem in data mining, where each image is represented by a high-dimensional feature vector. However, the feature vectors might not be appropriate to express the semantic content of images, which eventually lead to poor results in clustering and classification. To tackle this problem, we propose an interactive approach to generate compact ...","Title: SAR Image Land Cover Datasets for Classification Benchmarking of Temporal Changes.\nYear: 2018\n\nAbstract: The increased availability of high-resolution synthetic aperture radar (SAR) satellite images has led to new civil applications of these data. Among them is the systematic classification of land cover types based on the patterns of settlements or agriculture recorded by SAR imagers, in particular the identification and quantification of temporal changes. A systematic (re)classification shall allow...","Title: Improved image classification by proper patch size selection: TerraSAR-X vs. Sentinel-1A\nYear: 2016\n\nAbstract: When we perform image content classification by appending semantic labels to regularly cut image patches, we have to be sure that the selected patch size is a good choice for the images at hand. In the following, we look at SAR (Synthetic Aperture Radar) satellite images, and analyse the impact of the selected patch size on the attainable classification accuracy. For test images with precisely known ground truth, one can determine the true precision / recall performance of the applied classification method. In our case, we interactively trained a classifier system via active learning, and compared the resulting classification accuracy for high and medium resolution SAR images of different space borne instruments taken over urban areas, characterized by a high diversity of target categories. At a first glance, it turns out that the selected patch size does have a significant impact leading to a varying number of identified categories with strangely related confidence levels. A fundamental understanding of the relationships between the number of detected categories and their associated confidence levels requires detailed knowledge about SAR imaging, target characteristics, pixel size effects, radiometric image quality, the availability of appropriate semantic labels, and the selected active learning environment together with its image classification tool.","Title: Land Cover Semantic Annotation Derived from High-Resolution SAR Images.\nYear: 2016\n\nAbstract: Users of remote sensing images analyzing land cover characteristics are very much interested in classification schemes that define a consistent set of target categories. Up to now, a number of established classification schemes are mainly being used by interpreters of medium-resolution optical satellite images focusing on large-scale land cover. In contrast, we concentrate in this publication on t...","Title: Image disorder characterization based on rate distortion\nYear: 2005\n\nAbstract: Rate distortion theory is one of the areas of information transmission theory with important applications in multimodal signal processing, as for example image processing, information bottleneck and steganalysis. This article present an image characterization method based on rate distortion analysis in the feature space. This space is coded using clustering as vector quantization (k-means). Since image information usually cannot be coded by single clusters, because there are image regions corresponding to groups of clusters, the rate and distortion are specifically defined. The rate distortion curve is analyzed, extracting specific features for implementing a database image classification system.","Title: Classification With an Edge: Improving Semantic Image Segmentation with Boundary Detection.\nYear: 2016\n\nAbstract: We present an end-to-end trainable deep convolutional neural network (DCNN) for semantic segmentation with built-in awareness of semantically meaningful boundaries. Semantic segmentation is a fundamental remote sensing task, and most state-of-the-art methods rely on DCNNs as their workhorse. A major reason for their success is that deep networks learn to accumulate contextual information over very large receptive fields. However, this success comes at a cost, since the associated loss of effective spatial resolution washes out high-frequency details and leads to blurry object boundaries. Here, we propose to counter this effect by combining semantic segmentation with semantically informed edge detection, thus making class boundaries explicit in the model. First, we construct a comparatively simple, memory-efficient model by adding boundary detection to the segnet encoder-decoder architecture. Second, we also include boundary detection in fcn-type models and set up a high-end classifier ensemble. We show that boundary detection significantly improves semantic segmentation with CNNs in an end-to-end training scheme. Our best model achieves >90% overall accuracy on the ISPRS Vaihingen benchmark.","Title: New MPEG-7 Scalable Color Descriptor Based on Polar Coordinates for Multispectral Earth Observation Image Analysis.\nYear: 2017\n\nAbstract: Continuously expanding high-resolution and very high resolution multispectral image collections, provided by remote sensing satellites, require specific methods and techniques for data analysis and understanding. Even though there are several patch-based approaches for image classification and indexing, none of them are integrated within a standard. Having the goal to develop an MPEG-7 compliant d...","Title: CMIR-NET : A deep learning based model for cross-modal retrieval in remote sensing\nYear: 2019\n\nAbstract: •Cross-modal data retrieval.•Novel framework for designing a unified representation for different modalities of data.•Improved performance using the unified representation for both cross and uni-modal retrieval.•Extensive experiments performed on the benchmark DSRSID and UC-Merced with audio samples datasets.","Title: Unbiased Seamless SAR Image Change Detection Based on Normalized Compression Distance\nYear: 2019\n\nAbstract: Land cover changes may have very different nature, e.g., vegetation development, soil erosion, variation of humidity, or damage of buildings, only to enumerate few cases. In addition, synthetic aperture radar (SAR) observations are a doppelganger of the scene, imaging the scene signature rather than the scene itself. To overcome these challenges, SAR change detection methods generally adapt to the particular situations. We present seamless methods based on normalized compression distance (NCD) estimation. NCD is a similarity metric applied directly to the data, thus with no biases induced by feature estimators or classifiers. Since the diversity of changes is huge and extremely hard to derive typical classes, we introduce paradigm based both on an unsupervised and a supervised method. The change detection procedure mainly consists in dividing image dataset in patches, computing a collection of similarities for pairs of tiles formed differently in each case, and usage of this collection in unsupervised and supervised forms to generate a change map. Both the threshold based histogram, unsupervised method, and the k-NN classifier algorithm, supervised method, have a distinct flow to obtain the change map. To use the NCD operator according to our proposed methods, a speckle resistance test is involved. The experimental results for the two methodologies are computed using two TerraSAR-X images over Sendai and surrounding areas, Japan.","Title: Compact Ground-Based Interferometric Synthetic Aperture Radar: Short-range structural monitoring\nYear: 2019\n\nAbstract: Recently, structural monitoring by radar remote sensing has become more necessary for both economic and security reasons. Infrastructure monitoring with no incorporated deformation sensors (e.g., old generation water dams for which regulations did not impose monitoring capabilities) is usually performed by regular in situ topographic surveys. However, these surveys cannot be performed very often, and alternative methods are desirable. A feasible nonintrusive way to monitor such a structure is with interferometric synthetic aperture radar (SAR) data that can be acquired with monostatic/bistatic sensors.","Title: LUAI Challenge 2021 on Learning to Understand Aerial Images\nYear: 2021\n\nAbstract: This report summarizes the results of Learning to Understand Aerial Images (LUAI) 2021 challenge held on ICCV’2021, which focuses on object detection and semantic segmentation in aerial images. Using DOTA-v2.0 [7] and GID-15 [35] datasets, this challenge proposes three tasks for oriented object detection, horizontal object detection, and semantic segmentation of common categories in aerial images....","Title: Guided Deep Learning by Subaperture Decomposition: Ocean Patterns from SAR Imagery.\nYear: 2022\n\nAbstract: Spaceborne synthetic aperture radar can provide meters scale images of the ocean surface roughness day or night in nearly all weather conditions. This makes it a unique asset for many geophysical applications. Sentinel 1 SAR wave mode vignettes have made possible to capture many important oceanic and atmospheric phenomena since 2014. However, considering the amount of data provided, expanding applications requires a strategy to automatically process and extract geophysical parameters. In this study, we propose to apply subaperture decomposition as a preprocessing stage for SAR deep learning models. Our data centring approach surpassed the baseline by 0.7, obtaining state of the art on the TenGeoPSARwv data set. In addition, we empirically showed that subaperture decomposition could bring additional information over the original vignette, by rising the number of clusters for an unsupervised segmentation method. Overall, we encourage the development of data centring approaches, showing that, data preprocessing could bring significant performance improvements over existing deep learning models.","Title: Coreset of Hyperspectral Images on a Small Quantum Computer.\nYear: 2022\n\nAbstract: Machine Learning (ML) techniques are employed to analyze and process big Remote Sensing (RS) data, and one well-known ML technique is a Support Vector Machine (SVM). An SVM is a quadratic programming (QP) problem, and a D-Wave quantum annealer (D-Wave QA) promises to solve this QP problem more efficiently than a conventional computer. However, the D-Wave QA cannot solve directly the SVM due to its very few input qubits. Hence, we use a coreset (\"core of a dataset\") of given EO data for training an SVM on this small D-Wave QA. The coreset is a small, representative weighted subset of an original dataset, and any training models generate competitive classes by using the coreset in contrast to by using its original dataset. We measured the closeness between an original dataset and its coreset by employing a Kullback-Leibler (KL) divergence measure. Moreover, we trained the SVM on the coreset data by using both a D-Wave QA and a conventional method. We conclude that the coreset characterizes the original dataset with very small KL divergence measure. In addition, we present our KL divergence results for demonstrating the closeness between our original data and its coreset. As practical RS data, we use Hyperspectral Image (HSI) of Indian Pine, USA.","Title: Classification of Remote Sensing Images With Parameterized Quantum Gates\nYear: 2022\n\nAbstract: This letter studies how to program and assess a parameterized quantum circuit (PQC) for classifying Earth observation (EO) satellite images. In this exploratory study, we assess a PQC for classifying a two-label EO image dataset and compare it with a classic deep learning classifier. We use the PQC with an input space of only 17 quantum bits (qubits) due to the current limitations of quantum technology. As a real-world image for EO, we selected the Eurosat dataset obtained from multispectral Sentinel-2 images as a training dataset and a Sentinel-2 image of Berlin, Germany, as a test image. However, the high dimensionality of our images is incompatible with the PQC input domain of 17 qubits. Hence, we had to reduce the dimensionality of the input images for this two-label case to a vector with 16 elements; the 17th qubit remains reserved for storing label information. We employed a very deep convolutional network with an autoencoder as a technique for the dimensionality reduction of the input image, and we mapped the dimensionally reduced image onto 16 qubits by means of parameter thresholding. Then, we used a PQC to classify the two-label content of the dimensionally reduced Eurosat image dataset. A PQC classifies the Eurosat images with high accuracy as a classic deep learning method (and with even better accuracy in some instances). From our experiment, we derived and enhanced deeper insight into programming future gate-based quantum computers for many practical problems in EO.","Title: Attention-Driven Graph Convolution Network for Remote Sensing Image Retrieval\nYear: 2022\n\nAbstract: Graph convolution networks (GCNs) are useful in remote sensing (RS) image retrieval. It is found to be effective because, in a graph representation, the relative geometrical interactions between different regions (or segments) are appropriately captured, along with their region-wise features in their region adjacency graphs. Also, the attention mechanism has often been applied to the nodes to highlight the essential features in each node. In this regard, a significant amount of high-frequency information is missed since each image segment is effectively summarized within a single node. To account for this and increase the learning capacity, we propose to attend over the edge/adjacency matrix to highlight the interactions among meaningful regions that contribute to supervised learning from images. We exploit this novel edge attention mechanism together with node attention to highlight essential image context by allowing more importance to the meaningful neighboring regions that highlight a relevant node. We implement the proposed context-attended GCN framework for image retrieval on the benchmarked UC-Merced and the PatternNet datasets. We observe a notable improvement in the results compared to the state of the art.","Title: Near-Range Multipath Mitigation Methodology for Multistatic SAR Applications Using Matched-Adaptive Filters\nYear: 2022\n\nAbstract: This article presents a methodology based on matched-adaptive filters, which is used to mitigate the effect of near-range multipath in multistatic SAR systems and transponders. It is challenging to physically construct a bistatic receiver such that the reference signal is not leaked into the received signal, either via coupling in the circuitry or via reflections off objects in the vicinity of the receiver. Due to its much larger amplitude, the reference signal can easily mask near-range targets with its sidelobes. A similar signal degradation is observed in active transponders that are used for calibrating radar systems, when objects exist in their vicinity. In this article, we address these two issues: the coupling between the reference channel and the imaging channel, and the parasitic echoes present in the transponder response. A novel methodology is proposed that is capable of time-domain filtering the undesired components in real time. The novelty consists in combining matched and adaptive filters as a means of boosting performance and resolution estimation, resulting in an extremely accurate multipath elimination method. The proposed methodology is experimentally evaluated and optimized for each of the two aforementioned problems.","Title: Natural Embedding of the Stokes Parameters of Polarimetric Synthetic Aperture Radar Images in a Gate-Based Quantum Computer\nYear: 2022\n\nAbstract: Quantum algorithms are designed to process quantum data (quantum bits) in a gate-based quantum computer. They are proven rigorously that they reveal quantum advantages over conventional algorithms when their inputs are certain quantum data or some classical data mapped to quantum data. However, in a practical domain, data are classical in nature, and they are very big in dimension, size, and so on. Hence, there is a challenge to map (embed) classical data to quantum data, and even no quantum advantages of quantum algorithms are demonstrated over conventional ones when one processes the mapped classical data in a gate-based quantum computer. For the practical domain of earth observation (EO), due to the different sensors on remote-sensing platforms, we can map directly some types of EO data to quantum data. In particular, we have polarimetric synthetic aperture radar (PolSAR) images characterized by polarized beams. A polarized state of the polarized beam and a quantum bit are the Doppelganger of a physical state. We map them to each other, and we name this direct mapping a natural embedding, otherwise an artificial embedding. Furthermore, we process our naturally embedded data in a gate-based quantum computer by using a quantum algorithm regardless of its quantum advantages over conventional techniques; namely, we use the QML network as a quantum algorithm to prove that we naturally embedded our data in input qubits of a gate-based quantum computer. Therefore, we employed and directly processed PolSAR images in a QML network. Furthermore, we designed and provided a QML network with an additional layer of a neural network, namely, a hybrid quantum-classical network, and demonstrate how to program (via optimization and backpropagation) this hybrid quantum-classical network when employing and processing PolSAR images. In this work, we used a gate-based quantum computer offered by an IBM Quantum and a classical simulator for a gate-based quantum computer. Our contribution is that we provided very specific EO data with a natural embedding feature, the Doppelganger of quantum bits, and processed them in a hybrid quantum-classical network. More importantly, in the future, these PolSAR data can be processed by future quantum algorithms and future quantum computing platforms to obtain (or demonstrate) some quantum advantages over conventional techniques for EO problems.","Title: Cross-Bands Information Transfer to Offset Ambiguities and Atmospheric Phenomena for Multispectral Data Visualization\nYear: 2021\n\nAbstract: Visualization of multispectral images through band selection methods determines an information loss that in utmost cases proves to be critical for the adequate understanding of the represented scene. The R-G-B representation obtained by mapping the visual bands to the R, G, and B channels is highly used due to its great resemblance with the natural color one and aspects perceivable by the human eye. However, despite the similarity in terms of color code, ambiguities between classes such as water and vegetation or atmospheric phenomena like fog, clouds, and smoke that have been penetrated by other bands, remain visible and hinder the process of visualization of the Earth surface. This article presents a set of five different methods to offset the effects caused by ambiguities, fog, light clouds, and smoke by transferring relevant information between bands in order to visually reconstitute those parts of the image affected by atmospheric phenomena. The general concept shared by these methods implies a stacked autoencoder that successfully encompasses the information from all spectral bands into a latent representation used for visualization. Each proposed method is defined by different combination of input and error function formula. Spectral and polar coordinates features represent the possible options for the input, while formulas based on mean squared error or angular spectral distances determine the potential choices in terms of error function definition. The property of angular spectral distance and polar coordinates transformation to obtain illuminant invariant features determined their use in three out of five methods. We evaluate the methods through spectral signature graphical comparison and visual comparison related to the R-G-B representation. We conduct experiments on multiple Sentinel 2 full images.","Title: Earth Observation Semantic Data Mining: Latent Dirichlet Allocation-Based Approach\nYear: 2022\n\nAbstract: Recent advances in remote sensing technology have provided (very) high spatial resolution Earth Observation data with abundant latent semantic information. Conventional data processing algorithms are not capable of extracting the latent semantic information form these data and harness their full potential. As a result, semantic information discovery methods, based on data mining techniques, such as latent Dirichlet allocation and bag of visual words models, can discover the latent information. Despite their crucial rule, there are only a few studies in the field of semantic data mining for remote sensing applications. This article is focused on this shortage. Three different scenarios are used to evaluate the semantic information discovery in various remote sensing applications, including both optical and synthetic aperture radar (SAR) data with different spatial resolutions. In the first scenario, semantic discovery method correlated the semantic perception of the user and machine to correct and enhance the user defined Ground Truth map in very high-resolution RGB data. The potential of the semantic discovery is evaluated for wildfire affected area detection in Sentinel-2 data in the second scenario. Finally, in the third scenario, the semantic discovery method is utilized to detect the misclassifications as well as the patches with ambiguous or multiple semantic labels in a Sentinel-1 SAR patch-based benchmark dataset to enhance the robustness and accuracy of the annotation in the dataset. Our results in these three scenarios demonstrated the capability of the data-mining-based semantic information discovery methods for various remote sensing.","Title: A Framework for Interactive Visual Interpretation of Remote Sensing Data\nYear: 2022\n\nAbstract: Machine learning methods have shown tremendous success in understanding earth observation data; however, recently, there is a rising claim toward explainable machine learning approaches. Concerned researchers found interpretable visualizations to be greatly helpful in understanding how a model works. In this research, we propose a framework for interactive and interpretable visualization of remote sensing data using two machine learning models and an Elasticsearch (ES) database. Two explainable machine learning models, namely, bag-of-visual-words (BoVWs) and latent Dirichlet allocation (LDA) are chosen to model the data in an unsupervised manner and give a textual representation. The textualized remote sensing data are stored in an ES database. This framework offers several fast content-based search functionalities exploiting the full-text query capabilities of ES based on the respective representations and also offers an efficient storage mechanism for the data.","Title: Zero-Shot Cross-Modal Retrieval for Remote Sensing Images With Minimal Supervision\nYear: 2022\n\nAbstract: The performance of a deep-learning-based model primarily relies on the diversity and size of the training dataset. However, obtaining such a large amount of labeled data for practical remote sensing (RS) applications is expensive and labor-intensive. Training protocols have been previously proposed for few-shot learning (FSL) and zero-shot learning (ZSL). However, FSL is not compatible with handling unobserved class data at the inference phase, while ZSL requires many training samples of the seen classes. In this work, we propose a novel training protocol for image retrieval and name it as label-deficit zero-shot learning (LDZSL). We use this novel LDZSL training protocol for the challenging task of cross-sensor data retrieval in RS. This protocol uses very few labeled data samples of the seen classes during training and interprets unobserved class data samples at the inference phase. This strategy is critical as some data modalities are hard to annotate without domain experts. This work proposes a novel bilevel Siamese network to perform the LDZSL cross-sensor retrieval of multispectral and synthetic aperture radar (SAR) images. We use the available georeferenced SAR and multispectral data to domain align the embedding features of the two modalities. We experimentally demonstrate the proposed model's efficacy using the So2Sat dataset compared with the existing state-of-the-art models of the ZSL framework trained under a reduced training set. We also show the generalizability of the proposed model using a sketch-based image retrieval task. Experimental results on the Earth on the Canvas dataset exhibit comparative performance over the literature.","Title: Hybrid DNN-Dirichlet Anomaly Detection and Ranking: Case of Burned Areas Discovery\nYear: 2022\n\nAbstract: In the past decade, anomaly detection has experienced an expanding attraction in satellite data analysis. Monitoring wildfire dynamics plays a substantial part in global land management, i.e., to detect and determine the expansion of such areas, estimate the deterioration of forest regions, and assist the intervention plan. In this article, we proposed an approach for Sentinel-2 scenes that detects anomalies in burned area contexts using a rank-ordered method on a single post-event image. We adopted a self-supervised paradigm in learning image representations by training a deep convolutional model to differentiate in a series of geometric transformations. Dirichlet distributions are selected as priors to characterize the variability of the random multinomial distribution in multispectral data. Dirichlet precision parameters are computed from observed data and are used to construct a ranking function that quantifies the degree of anomaly in data based on the softmax responses given by the trained classifier. We evaluated the performance of the proposed method through a cumulative effort of two remote sensing tasks, namely, open-set detection (i.e., test datasets contain classes unseen at the training time) and location separation (i.e., test datasets include images from distinct spatial location than the training images). The experiments were performed on three different datasets, BigEarthNet, and two actual burned area Sentinel-2 datasets from predisposed zones to fire events, Australia and Bolivia. We achieved in all test datasets state-of-the-art performance, considering the substantial and diversified types of natural anomalies in multispectral data.","Title: Sar Super-Resolution Using Physics-Aware Adaptive Compressed Sensing\nYear: 2022\n\nAbstract: The resolution requirements of modern radar applications are increasing rapidly and cannot be fulfilled by the limited number of wide-band radar systems. Many approaches have been explored to solve this problem under the topic of super-resolution. In this paper, we propose a hybrid algorithm for resolution improvement, where we aim to combine the adaptability of deep neural networks with the reliability and expertise of traditional domain-specific SAR processing.","Title: Causality for Remote Sensing: An Exploratory Study\nYear: 2022\n\nAbstract: Causality is one of the most important topics in a Machine Learning (ML) research, and it gives insights beyond the dependency of data points. Causality is a very vital concept also for investigating the dynamic surface of our living planet. However, there are not many attempts for integrating a causal model in Remote Sensing (RS) methodologies. Hence, in this paper, we propose to use patch-based RS images and to represent each patch-based image by a single variable (e.g. entropy). Then we use a Structural Equation Model (SEM) to study their cause-effect relation. Moreover, the SEM is a simple causal model characterized by a Directed Acyclic Graph (DAG). Its nodes are causal variables, and its edges represent causal relationships among causal variables if and only if causal variables are dependent.","Title: Achieving Information Super-resolution for Sentinel-2 NDVI Through Gaussian Process Regression\nYear: 2022\n\nAbstract: Super-resolution is used to recover high resolution images from low resolution images. We use this concept in a slightly different context to achieve higher quality knowledge from low resolution satellite images. The technique involves transfer learning from high to low resolution images using a Gaussian Process Regression model. We use high resolution drone images to train the model. This technique is applied in three case studies to verify the consistency of results in case of NDVI computation. However, the same technique can be applied to obtain for other application of satellite images in plant vigor assessment.","Title: SAR Image Classification Using Mixed Spatial-Spectral Information and Pre-trained Convolutional Neural Networks\nYear: 2022\n\nAbstract: The recent technological advancements in remote sensing lead to an increased importance regarding the analysis of satellite data targeting security and surveillance tasks. Although the availability of data products is constantly augmented and the advances in Deep Learning technologies are constant, Synthetic Aperture Radar (SAR) image classification remains a challenge in the remote sensing domain because standard convolutional neural network-based architectures may encounter difficulties in recognizing objects that are characterized by similar texture, but different backscattering patterns. Moreover, training deep learning architectures requires a large volume of annotated data, which, in general, represents an obstacle, especially in the case of the remote sensing domain. This article addresses complex-valued SAR image classification through both spatial and Fourier-domain features, extracted by means of pretrained neural networks. While spatial features allow extracting knowl-edge regarding the structure and texture of the objects from intensity images, the physical properties of the objects are learned from radar spectrograms. In addition, we show that considering different polarizations of the SAR sensor, we are able to obtain better visual classifications. The experiments are conducted over Sentinel-1images, which are freely available for download under the Copernicus initiative."],"string":"[\n \"Title: Algorithmic Relative Complexity.\\nYear: 2011\\n\\nAbstract: Information content and compression are tightly related concepts that can be addressed through both classical and algorithmic information theories, on the basis of Shannon entropy and Kolmogorov complexity, respectively. The definition of several entities in Kolmogorov's framework relies upon ideas from classical information theory, and these two approaches share many common traits. In this work, we expand the relations between these two frameworks by introducing algorithmic cross-complexity and relative complexity, counterparts of the cross-entropy and relative entropy (or Kullback-Leibler divergence) found in Shannon's framework. We define the cross-complexity of an object x with respect to another object y as the amount of computational resources needed to specify x in terms of y, and the complexity of x related to y as the compression power which is lost when adopting such a description for x, compared to the shortest representation of x. Properties of analogous quantities in classical information theory hold for these new concepts. As these notions are incomputable, a suitable approximation based upon data compression is derived to enable the application to real data, yielding a divergence measure applicable to any pair of strings. Example applications are outlined, involving authorship attribution and satellite image classification, as well as a comparison to similar established techniques.\",\n \"Title: Interactive learning and probabilistic retrieval in remote sensing image archives\\nYear: 2000\\n\\nAbstract: We present a concept of interactive learning and probabilistic retrieval of user-specific cover types in a content-based remote sensing image archive. A cover type is incrementally defined via user-provided positive and negative examples. From these examples, we infer probabilities of the Bayesian network that link the user interests to a pre-extracted content index. Due to the stochastic nature of the cover type def- initions, the database system not only retrieves images according to the estimated coverage but also according to the accuracy of that estimation given the current state of learning. For the latter, we introduce the concept of separability. We expand on the steps of Bayesian inference to compute the application-free content index using a family of data models, and on the description of the stochastic link using hyperparameters. In particular, we focus on the interactive nature of our approach, which provides instantaneous feedback to the user in the form of an immediate update of the posterior map, and a very fast, approximate search in the archive. A java-based demonstrator using the presented concept of content-based access to a test archive of Landsat TM, X-SAR, and aerial images are available over the Internet (http://www.vision.ee.ethz.ch/~rsia/ClickBayes).\",\n \"Title: A semantic framework for data retrieval in large remote sensing databases.\\nYear: 2012\\n\\nAbstract: Remote sensing platforms acquire huge amounts of data every day. As a result, large archives of data have been created. In order to provide access to this data efficient search and retrieval methods have to be developed, such as image information mining systems. In this paper we present a framework for an image information mining system using the Latent Dirichlet Allocation text-mining algorithm to provide a high-level semantic model of data, the search being performed in the LDA model space.\",\n \"Title: Parameter-free Clustering: Application to Fawns Detection\\nYear: 2009\\n\\nAbstract: Many fawns and other wild animals are killed by mowing machines every year. To prevent them from being killed or injured, a sensor system is being developed to detect the fawns hidden in meadows under mowing. Beside a microwave radar system, two cameras (thermal Infrared and RGB) take a picture at the mower's current location. This contribution focuses on the compression-based algorithm that will be adopted to detect the locations containing a fawn hiding in the grass. such approach, being parameter-free, allows performing a fully unsupervised clustering by exploiting the intrinsic properties of data compression to estimate the amount of shared information between two images.\",\n \"Title: SAR Image Categorization With Log Cumulants of the Fractional Fourier Transform Coefficients.\\nYear: 2013\\n\\nAbstract: With the advent of high-resolution (HR) synthetic aperture radar (SAR) images from satellites like TerraSAR-X and TanDEM-X, interest is now on patch-oriented image categorization in contrast to the pixel-based classification in low-resolution SAR images. SAR image categorization requires the generation of a compact feature descriptor that accurately defines the content of the image patch under consideration. As phase information plays a critical role in SAR images, this paper proposes the use of a chirplet-derived transform, i.e., the fractional Fourier transform (FrFT), for generating a compact feature descriptor for single-look complex (SLC) SAR images. Representing a SAR signal in rotated joint time-frequency planes via the FrFT allows discovering the underlying backscattering phenomenon of the objects on the ground. SAR image projections on different planes of the joint time-frequency space using the FrFT provide a simple statistical response that is easier to analyze. The proposed method has been compared with a multiscale approach, i.e., Gabor filter banks, a second-order-statistics-based method (as gray-level co-occurrence matrices), and a spectral descriptor method. We demonstrate the suitability of the FrFT-based method for image categorization on the basis of backscattering behavior, whereas the Gabor-filter-bank-based method is found mainly suitable for images with a strong texture. This paper demonstrates enhancement in the separability for most of the considered categories when using logarithmic cumulants instead of linear moments for both the FrFT-based and Gabor-filter-bank-based methods. The experimental database consists of 2000 image patches (of size 200 x 200 pixels) extracted from SLC HR TerraSAR-X scenes.\",\n \"Title: Identification and characterization of railway trains in high resolution TerraSAR-X images.\\nYear: 2012\\n\\nAbstract: The recognition and analysis of moving targets in SAR images is a well-known topic in remote sensing. The public availability of high resolution SAR satellite images opened new perspectives for space-based small object recognition: traffic monitoring from space has matured into an established technique where the motion characteristics of ships and cars can be extracted routinely.A particular case of traffic monitoring is the search for moving trains in SAR images; here one can locate moving trains, determine their velocity, and identify the type and number of wagons. In the following, we concentrate on features to be extracted from TerraSAR-X images containing commuter trains in the area of Munich, Germany.\",\n \"Title: A Latent Analysis of Earth Surface Dynamic Evolution Using Change Map Time Series.\\nYear: 2013\\n\\nAbstract: With a continuous increase in the number of Earth Observation satellites, leading to the development of satellite image time series (SITS), the number of algorithms for land cover analysis and monitoring has greatly expanded. This paper offers a new perspective in dynamic classification for SITS. Four similarity measures (correlation coefficient, Kullback-Leibler divergence, conditional information, and normalized compression distance) based on consecutive image pairs from the data are employed. These measures employ linear dependences, statistical measures, and spatial relationships to compute radiometric, spectral, and texture changes that offer a description for the multitemporal behavior of the SITS. During this process, the original SITS is converted to a change map time series (CMTS), which removes the static information from the data set. The CMTS is analyzed using a latent Dirichlet allocation (LDA) model capable of discovering classes with semantic meaning based on the latent information hidden in the scene. This statistical method was originally used for text classification, thus requiring a word, document, corpus analogy with the elements inside the image. The experimental results were computed using 11 Landsat images over the city of Bucharest and surrounding areas. The LDA model enables us to discover a wide range of scene evolution classes based on the various dynamic behaviors of the land cover. The results are compared with the Corinne Land Cover map. However, this is not a validation method but one that adds static knowledge about the general usage of the analyzed area. In order to help the interpretation of the results, we use several studies on forms of relief, weather forecast, and very high resolution images that can explain the wide range of structures responsible for influencing the dynamic inside the resolution cell.\",\n \"Title: Linear versus non-linear analysis of relevant scatterers in high resolution SAR images.\\nYear: 2007\\n\\nAbstract: With the increase of Synthetic Aperture Radar (SAR) sensor resolution, SAR images could include a large variety of interesting real man-made structures. Therefore, a more detailed analysis and a finer description of SAR images of urban areas are needed for a better understanding of the scene.Nevertheless, recognizing scenes using high resolution SAR images requires the capability to identify relevant signal signatures (called also descriptors), depending on variable image acquisition geometry, arbitrary objects poses and configurations. Among feature extraction methods, we propose to use Principal Components Analysis (PCA) and/or Independent Components Analysis (ICA), in order to exploit deeper the nature of SAR signatures. In this paper, both a description of our work and a presentation of our preliminary classification performance results will be provided.\",\n \"Title: Evaluation of Bayesian Despeckling and Texture Extraction Methods Based on Gauss-Markov and Auto-Binomial Gibbs Random Fields: Application to TerraSAR-X Data.\\nYear: 2012\\n\\nAbstract: Speckle hinders information in synthetic aperture radar (SAR) images and makes automatic information extraction very difficult. The Bayesian approach allows us to perform the despeckling of an image while preserving its texture and structures. This model-based approach relies on a prior model of the scene. This paper presents an evaluation of two despeckling and texture extraction model-based meth...\",\n \"Title: Multiscale and Dimensionality Behavior of ICA Components for Satellite Image Indexing\\nYear: 2010\\n\\nAbstract: In this letter, our main objective is to demonstrate the capabilities of independent component analysis (ICA) for very high resolution satellite image characterization, particularly for geometrical structures contained in images from urban areas in comparison with natural landscapes. We propose a model based on ICA sources to index satellite images with a resolution of 0.6 m. An important part of this work is to study the effects of multiscaling and dimensionality behavior of ICA components. For evaluating the capabilities of our model, we design a classification method to distinguish images froma variety of natural and man-made scenes.\",\n \"Title: A remote sensing image processing framework for damage assessment in a forest fire scenario.\\nYear: 2012\\n\\nAbstract: In natural hazards management applications Earth Observation (EO) image processing methods are based on segmentation and classification. The result primary consists of thematic maps which are readily interpretable. We propose a complete EO image processing chain, which generates an end product with increased information content organized in thematic layers. The processing chain integrates four main components: image classification, identification of high anomaly areas relative to the entire scene context, spectral and texture change detection, and the integration of different information layers. The processing chain was tested in a fire management scenario, using a pair of Landsat5-TM images for the Pagami Creek forest fire which was active from August to October 2011.\",\n \"Title: Data vaults: a database welcome to scientific file repositories\\nYear: 2013\\n\\nAbstract: Efficient management and exploration of high-volume scientific file repositories have become pivotal for advancement in science. We propose to demonstrate the Data Vault, an extension of the database system architecture that transparently opens scientific file repositories for efficient in-database processing and exploration. The Data Vault facilitates science data analysis using high-level declarative languages, such as the traditional SQL and the novel array-oriented SciQL. Data of interest are loaded from the attached repository in a just-in-time manner without need for up-front data ingestion. The demo is built around concrete implementations of the Data Vault for two scientific use cases: seismic time series and Earth observation images. The seismic Data Vault uses the queries submitted by the audience to illustrate the internals of Data Vault functioning by revealing the mechanisms of dynamic query plan generation and on-demand external data ingestion. The image Data Vault shows an application view from the perspective of data mining researchers.\",\n \"Title: Ratio-Detector-Based Feature Extraction for Very High Resolution SAR Image Patch Indexing\\nYear: 2013\\n\\nAbstract: With the advent of very high resolution (VHR) synthetic aperture radar (SAR) images, local content description is becoming a critical issue for indexing. Conventional SAR image analysis techniques, like segmentation and pixel-level classification, are likely to fail as high-level semantic description should be considered for better discrimination. Therefore, we propose to use image-patch-based analysis method for SAR image interpretation. Inspired by ratio edge detector, in this letter, a new feature extraction method represented by the mean ratios in different directions is proposed for VHR SAR image content characterization. Based on the mean ratio, two simple yet powerful and robust features are proposed for SAR image patch indexing. One is the bag-of-word model using not only the basic statistics, i.e., local mean and variance, but also the mean ratios in different directions. The second one is an adaptation of the Weber local descriptor to SAR images by substituting the gradient with the ratio of mean differences in vertical and horizontal directions. To evaluate the proposed features, image patch indexing based on active learning using a SAR image database consisting of high-resolution TerraSAR-X patches is performed. Comparison with the state-of-the-art features, particularly texture features, has shown improved performance for SAR image categorization.\",\n \"Title: Supervised learning on graphs of spatio-temporal similarity in satellite image sequences\\nYear: 2007\\n\\nAbstract: Abstract: High resolution satellite image sequences are multidimensional signals composed of spatiotemporal patterns associated to numerous and various phenomena. Bayesian methods have been previously proposed in [8] to code the information contained in satellite image sequences in a graph representation using Bayesian methods. Based on such a representation, this paper further presents a supervised learning methodology of semantics associated to spatio-temporal patterns occurring in satellite image sequences. It enables the recognition and the probabilistic retrieval of similar events. Indeed, graphs are attached to statistical models for spatio-temporal processes, which at their turn describe physical changes in the observed scene. Therefore, we adjust a parametric model evaluating similarity types between graph patterns in order to represent user-specific semantics attached to spatio-temporal phenomena. The learning step is performed by the incremental definition of similarity types via user-provided spatio-temporal pattern examples attached to positive or/and negative semantics. From these examples, probabilities are inferred using a Bayesian network and a Dirichlet model. This enables to links user interest to a specific similarity model between graph patterns. According to the current state of learning, semantic posterior probabilities are updated for all possible graph patterns so that similar spatio-temporal phenomena,can be recognized and retrieved from the image sequence. Few experiments performed on a multi-spectral SPOT image sequence illustrate the proposed spatio-temporal recognition method.\",\n \"Title: Authorship Analysis based on Data Compression.\\nYear: 2014\\n\\nAbstract: •Compression-based similarity measures are effectively used in authorship analysis tasks.•Their high computational complexity makes it hard to employ them in practical applications.•We propose the Fast Compression Distance (FCD) for these specific tasks.•The FCD yields more accurate results and has a reduced computational complexity.•Satisfactory experiments are presented on datasets written in five different languages.\",\n \"Title: TELEIOS: a database-powered virtual earth observatory\\nYear: 2012\\n\\nAbstract: TELEIOS is a recent European project that addresses the need for scalable access to petabytes of Earth Observation data and the discovery and exploitation of knowledge that is hidden in them. TELEIOS builds on scientific database technologies (array databases, SciQL, data vaults) and Semantic Web technologies (stRDF and stSPARQL) implemented on top of a state of the art column store database system (MonetDB). We demonstrate a first prototype of the TELEIOS Virtual Earth Observatory (VEO) architecture, using a forest fire monitoring application as example.\",\n \"Title: Building virtual earth observatories using ontologies and linked geospatial data\\nYear: 2012\\n\\nAbstract: Advances in remote sensing technologies have enabled public and commercial organizations to send an ever-increasing number of satellites in orbit around Earth. As a result, Earth Observation (EO) data has been constantly increasing in volume in the last few years, and is currently reaching petabytes in many satellite archives. For example, the multi-mission data archive of the TELEIOS partner German Aerospace Center (DLR) is expected to reach 2PB next year, while ESA estimates that it will be archiving 20PB of data before the year 2020. As the volume of data in satellite archives has been increasing, so have the scientific and commercial applications of EO data. Nevertheless, it is estimated that up to 95% of the data present in existing archives has never been accessed, so the potential for increasing exploitation is very big.\",\n \"Title: Modeling Trajectory Of Dynamic Clusters In Image Time-Series For Spatio-Temporal Reasoning\\nYear: 2005\\n\\nAbstract: During the last decades, satellites have acquired incessantly high-resolution images of many Earth observation sites. New products have arisen from this intensive acquisition process: high-resolution satellite image time-series (SITS). They represent a large data volume with a rich information content and may open a broad range of new applications. This paper presents an information mining concept which enables a user to learn and retrieve spatio-temporal structures in SITS. The concept is based on a hierarchical Bayesian modeling of SITS information content which enables us to link the interest of a user to specific spatio-temporal structures. The hierarchy is composed of two inference steps: an unsupervised modeling of dynamic clusters resulting in a graph of trajectories, and an interactive learning procedure based on graphs which leads to the semantic labeling of spatio-temporal structures. Experiments performed on a SPOT image time-series demonstrate the concept capabilities.\",\n \"Title: Modeling Trajectory Of Dynamic Clusters In Image Time-Series For Spatio-Temporal Reasoning\\nYear: 2005\\n\\nAbstract: During the last decades, satellites have acquired incessantly high-resolution images of many Earth observation sites. New products have arisen from this intensive acquisition process: high-resolution satellite image time-series (SITS). They represent a large data volume with a rich information content and may open a broad range of new applications. This paper presents an information mining concept which enables a user to learn and retrieve spatio-temporal structures in SITS. The concept is based on a hierarchical Bayesian modeling of SITS information content which enables us to link the interest of a user to specific spatio-temporal structures. The hierarchy is composed of two inference steps: an unsupervised modeling of dynamic clusters resulting in a graph of trajectories, and an interactive learning procedure based on graphs which leads to the semantic labeling of spatio-temporal structures. Experiments performed on a SPOT image time-series demonstrate the concept capabilities.\",\n \"Title: Semantic analysis of satellite image time series.\\nYear: 2012\\n\\nAbstract: Large archives of satellite images have been created over time. The existence of these archives enables us to extract evolutions of the same of same geographic area over time, creating satellite image time series (SITS). as SITS represent an amount of information far grater than individual images. their analysis is complex and difficult. In this paper we propose a new unsupervised SITS analysis method based on the latent Dirichlet allocation (LDA) model, a hierarchical model originally developed for text analysis. In this model documents are represented as random mixture of latent topics, each topic being characterized by a distribution over words. This paper extends the use of LDA model for satellite image time series analysis by proposing a description language for SITS modeling according to the LDA model, and is applied on a SITS of 11 Landsat TM scenes acquired in 2007.\",\n \"Title: Despeckling of TerraSAR-X Data Using Second-Generation Wavelets\\nYear: 2010\\n\\nAbstract: This letter presents the despeckling of synthetic aperture radar (SAR) images within the bandelet and contourlet domains. A model-based approach is presented for the despeckling of SAR images. The speckle-reduced estimate is found using the first-order Bayesian inference, and the best model's parameters are estimated using the second-order Bayesian inference. Synthetic and real images are used for...\",\n \"Title: SAR Target Analysis Based on Multiple-Sublook Decomposition: A Visual Exploration Approach.\\nYear: 2012\\n\\nAbstract: The advent of submeter-resolution synthetic aperture radar (SAR) images from satellites such as TerraSAR-X has given a new dimension to SAR image understanding. Even though emphasis is always on discovering automatic means of target characterization, visual exploration of targets and objects is the first step in many applications. While considering the complex-valued SAR images, visual inspection ...\",\n \"Title: Image information mining system evaluation using information-theoretic measures\\nYear: 2005\\n\\nAbstract: During the last decade, the exponential increase of multimedia and remote sensing image archives, the fast expansion of the world wide web, and the high diversity of users have yielded concepts and systems for successful content-based image retrieval and image information mining. Image data information systems require both database and visual capabilities, but there is a gap between these systems. Database systems usually do not deal with multidimensional pictorial structures and vision systems do not provide database query functions. In terms of these points, the evaluation of content-based image retrieval systems became a focus of research interest. One can find several system evaluation approaches in literature, however, only few of them go beyond precision-recall graphs and do not allow a detailed evaluation of an interactive image retrieval system. Apart from the existing evaluation methodologies, we aim at the overall validation of our knowledge-driven content-based image information mining system. In this paper, an evaluation approach is demonstrated that is based on information-theoretic quantities to determine the information flow between system levels of different semantic abstraction and to analyze human-computer interactions.\",\n \"Title: An Innovative Concept for Image Information Mining.\\nYear: 2002\\n\\nAbstract: Information mining opens new perspectives and a huge potential for information extraction from large volumes of heterogeneous images and the correlation of this information with the goals of applications. We present a new concept and system for image information mining, based on modelling the causalities which link the image-signal contents to the objects and structures within interest of the users. The basic idea is to split the information representation into four steps: 1. image feature extraction using a library of algorithms so as to obtain a quasi-complete signal description 2. unsupervised grouping in a large number of clusters to be suitable for a large set of tasks 3. data reduction by parametric modelling the clusters 4. supervised leaming of user semantics, that is the level where, instead of being programmed, the systems is trained by a set of examples; thus the links from image contents to the users are created. The record of the sequence of links is a knowledge acquisition process, the system memorizes the user hypotheses. Step 4. is a man-machine dialogue, the information exchange is done using advanced visualization tools. The system learns what the users need. The system is presently prototyped for inclusion in a new generation of intelligent satellite ground segment systems, value adding tools in the area of geoinformation, and several applications in medicine and biometrics are also foreseen.\",\n \"Title: Complexity Based Image Artifact Detection\\nYear: 2008\\n\\nAbstract: Images may contain blemishes or artificial structures which come from the processing or directly from the sensors. These artifacts decrease the quality of the images and can lead to analysis and interpretation problems. We are interested in detecting these defects automatically. A visual analysis of such artifacts brings to the conclusion that they constitute a local modification of the complexity of the image. Therefore this paper will give a complexity based approach for the detection of these defects. This approach will distinguich two different methods: firstly the artifacts will be looked for respectively to given example, with a complexity comparison. This method will involve both direct and model based means. From the weaknesses of that method a second one will be deducted which will not be attached to the given examples and will take into consideration the local complexity context of the image. These methods will be applied in the domain of Earth Observation images.\",\n \"Title: Categorization Based Relevance Feedback Search Engine For Earth Observation Images Repositories\\nYear: 2006\\n\\nAbstract: Presently Earth Observation (EO) satellites acquire huge volumes of high resolution images very much over-passing the capacity of the users to access the information content of the acquired data. Thus, in addition to the existing methods for EO data and information extraction, new methods and tools are needed to explore and help to discover the information hidden in large EO image repositories. This article presents a categorisation based Relevance Feedback (RF) search engine for EO images repositories The developed method is presented as well results obtained for a SPOT5 satellite image database.\",\n \"Title: Huber–Markov Model for Complex SAR Image Restoration\\nYear: 2010\\n\\nAbstract: This letter presents the despeckling of single-look complex (SLC) synthetic aperture radar (SAR) images using nonquadratic regularization. The objective function consists of an image model, a gradient, and a prior model. The Huber-Markov random field (HMRF) models the prior. A numerical solution is achieved through extensions of half-quadratic regularization methods using complex-valued SAR data. ...\",\n \"Title: Mutual information based measure for image content characterization\\nYear: 2005\\n\\nAbstract: An image can be decomposed into different elementary descriptors depending on the observer interest. Similar techniques as used to understand words, regarded as molecules, formed by combining atoms, are proposed to describe images based on their information content. In this paper, we use primitive feature extraction and clustering to code the image information content. Our purpose is to describe the complexity of the information based on the combinational profile of the clustered primitive features using entropic measures like mutual information and Kullback-Leibler divergence. The developed method is demonstrated to asses image complexity for further applications to improve Earth Observation image analysis for sustainable humanitarian crisis response in risk reduction.\",\n \"Title: Optimized lossless compression of remote sensing image files using splay trees\\nYear: 2012\\n\\nAbstract: In this paper will be applied concepts like amortized complexity or self-adjustment to binary search trees. Motivation comes from the fact that the search trees have multiples drawbacks. It will be developed and analyzed the splay tree, a self-adjusting form of binary search trees.\",\n \"Title: Sar Complex Image Analysis: A Gauss Markov And A Multiple Sub-Aperture Based Target Characterization\\nYear: 2010\\n\\nAbstract: In this paper we discuss Gauss-Markov Random Field (GMRF) based on multiple sub-aperture decomposition method for the analysis of targets in complex-valued high-resolution SAR data. Gauss-Markov Random Field (GMRF) model with a quadratic energy function as a parametric analysis parameterizes the spectogram of the signal, whereas sub-aperture decomposition method exploits the holographic property of the spectrum at the cost of reducing resolution. This analysis helps to understand, characterize and analyze complex-valued SAR data and provides temptation to use complex-valued SAR data over detected data.\",\n \"Title: Further results on dissimilarity spaces for hyperspectral images RF-CBIR\\nYear: 2013\\n\\nAbstract: Content-Based Image Retrieval (CBIR) systems are powerful search tools in image databases that have been little applied to hyperspectral images. Relevance feedback (RF) is an iterative process that uses machine learning techniques and user's feedback to improve the CBIR systems performance. We pursued to expand previous research in hyperspectral CBIR systems built on dissimilarity functions defined either on spectral and spatial features extracted by spectral unmixing techniques, or on dictionaries extracted by dictionary-based compressors. These dissimilarity functions were not suitable for direct application in common machine learning techniques. We propose to use a RF general approach based on dissimilarity spaces which is more appropriate for the application of machine learning algorithms to the hyperspectral RF-CBIR. We validate the proposed RF method for hyperspectral CBIR systems over a real hyperspectral dataset.\",\n \"Title: A Bayesian Multi-Class Image Content Retrieval\\nYear: 2007\\n\\nAbstract: Modern imaging sensors, especially those aboard satellites, continuously deliver enormous amounts of data. The widespread of meter resolution images, is not only exploding the volumes of acquired data but also brings a new dimension in the image detail, thus growing the information content. These represent typical cases, where users need automated tools to discover, explore and explain the contents of large image databases. There is a strong need to build up applications that help the user in image interpretation task, applications that permit to query the archives in content based mode, without having to know all the information contained in the images at signal level. We propose in this article, a synergy between stochastic modelling, knowledge discovery, and semantic representation. To do that, we associate semantic labels to a combination of primitive image features. The user-defined semantic image content interpretation is linked with Bayesian networks to a completely unsupervised classification. This new paradigm for the interaction with EO archives can provide several applications for users coming from different domains, as change detection, agricultural field classification, environment monitoring, atmosphere effects or urbanization.\",\n \"Title: Bayesian restoration of interferometric phase through biased anisotropic diffusion\\nYear: 2009\\n\\nAbstract: In this paper a new Bayesian algorithm for interferometric phase restoration is presented. Based on a non-linear anisotropic extension of Orientation Diffusions, the inherent directionality of the fringe structure is introduced into its prior model. It also accounts for the periodicity of the phase representation. A fidelity term derived from the anisotropic metrics and the InSAR phase statistics deviates diffusion towards the acquired phase value. It acts as an adapted likelihood of the diffused phase. Hence, phase restoration is a trade-off between directionality and reconstruction fidelity, prior and likelihood. Results are provided on a High Resolution Spotlight scene acquired by TerraSAR-X.\",\n \"Title: Scale Space Exploration For Mining Image Information Content.\\nYear: 2002\\n\\nAbstract: Images are highly complex multidimensional signals, with rich and complicated information content. For this reason they are difficult to analyze with a specific automated approach. However, a hierarchical representation is helpful for understanding image content. In this paper, we describe an application of a scale-space clustering algorithm (melting) for exploration of image information content. Clustering by melting considers the feature space as a thermodynamical ensemble and groups the data by minimizing the free energy, having temperature as a scale parameter. We develop clustering by melting for multidimensional data, and propose and demonstrate a solution for the initialization of the algorithm. Due to the curse of dimensionality, for initialization of clusters we choose the initial clusters centers with an algorithm that performs fast cluster center estimation with low computation cost. We further analyze the information extracted by melting and propose a structure for information representation that enables exploration of image content. This structure is a tree in the scale space showing how the clusters merge. Implementation of the algorithm is through a multi-tree structure. With this structure, we can explore the image content as an information mining function, we obtain a more compact data structure, and we have maximum of information in scale space because we memorize the bifurcation points and the trajectories of the centers points in the scale space. The information encoded in the tree structure enables the fast reconstruction and exploration of the data cluster structure and the investigation of hierarchical sequences of image classifications. We demonstrate the effectiveness of the approach with examples using satellite multispectral image (SPOT 4) and Synthetic Aperture Radar (SAR) and Digital Elevation Models (DEM) derived from SAR interferometry (SRTM).\",\n \"Title: A Comparative Study of Statistical Models for Multilook SAR Images\\nYear: 2014\\n\\nAbstract: In this letter, we carry out a comparative study of statistical models for multilook synthetic aperture radar amplitude images. Ten state-of-the-art statistical models are selected for comparison. To achieve a fair evaluation, we estimate all model parameters using the method of log-cumulants and apply the method to an image pyramid with varying pixel spacing (and resolution). The pyramid is created by different image product generation options. In addition to pixel spacing and resolution, we also consider the homogeneity of a scene for performance evaluation and we apply three performance measures. Through this study, it was found out that some models perform well for all resolutions, while the performance of other models depends heavily on the image content.\",\n \"Title: Stochastic Models Of Slc Hr Sar Images\\nYear: 2007\\n\\nAbstract: The paper presents two algorithms for texture primitive feature extraction on Single Look Complex (SLC) and Polarimetric Synthetic Aperture Radar (PolSAR) SLC data. We assume the data to be modeled by a Gauss-Markov Random Field (GMRF): a complex GMRF model for characterizing the spatial correlation in SLC data and an extension of the model for inter-band correlation characterization.The complex GMRF characterizes the spatial relationship of a two-dimensional complex signal, i.e. SLC SAR data. The extended model characterizes the spatial interaction and the inter-band pixels correlation between the polarimetric complex channels. The Bayesian approach permits to deal with model fitting and selection in a direct way.The results are presented on a polarimetric E-SAR L band scene of Mannheim, Germany.\",\n \"Title: Cascaded Active Learning For Object Retrieval Using Multiscale Coarse To Fine Analysis\\nYear: 2011\\n\\nAbstract: In this paper, we describe an active learning scheme which performs coarse to fine testing using a multiscale patch-based representation of images to retrieve objects in large satellite image repositories. The proposed hierarchical top-down approach reduces step by step the size of the analysis window, eliminating each time large parts of the images considered as non-relevant. Unlike most object detection methods which requires large training sets and costly offline training, we use an active learning strategy to build a classifier at each level of the hierarchy and we propose an algorithm to propagate automatically the training examples from one level to the other.\",\n \"Title: Salient Remote Sensing Image Segmentation Based on Rate-Distortion Measure\\nYear: 2009\\n\\nAbstract: The ill-defined nature of the segmentation problem makes the selection of the optimal image partition difficult. One can characterize image segmentation as an attempt to find the best possible representation of a data set using a certain number of ldquoobjects.rdquo This can be regarded as data information compression, resulting in the distortion of the original values. Data sets are well represen...\",\n \"Title: Parametric Modeling Of The Fractional Fourier Transform Coefficients For Complex-Valued Sar Image Categorization\\nYear: 2013\\n\\nAbstract: In high-resolution (HR) and very-high resolution (VHR) synthetic aperture radar (SAR) images, focus is now on the patch-oriented image categorization in contrast to the pixelbased classification in low-resolution SAR images. SAR image categorization requires the generation of a compact feature descriptor that accurately defines the content of the image patch under consideration.In this paper we propose a parametric feature descriptor generated on the complex-valued SAR image within a transformation space. The fractional Fourier transform (FrFT), has been considered to transform the image pixels of the singlelook complex (SLC) SAR images in order to obtain a simpler statistical response. The real and imaginary components of the complex-valued FrFT coefficients have been modelled with generalized Gaussian distribution (GGD).The proposed feature descriptor is compared with a Wavelet-decomposition-based parametric feature descriptor; and with the FrFT-based and Gabor-filter-bank-based non-parametric feature descriptors. Categorization accuracy enhancement is demonstrated over several categories comprising of natural topologies. The experimental database consists of 2000 image patches (of size 200 x 200 pixels) extracted from SLC HR TerraSAR-X scenes.\",\n \"Title: Gibbs Random Field Models for Model-Based Despeckling of SAR Images\\nYear: 2010\\n\\nAbstract: Synthetic aperture radar (SAR) images are affected by multiplicative noise called speckle. This noise makes automatic image classification and image interpretation difficult. Thus, many methods have been developed to remove speckle from SAR images while preserving the useful information of the scene such as texture and geometry. In this letter, a comparison between three different despeckling meth...\",\n \"Title: Semantic Map Generation from Satellite Images for Humanitarian Scenarios Applications\\nYear: 2008\\n\\nAbstract: This paper demonstrates how knowledge driven methods and the associated data analysis algorithms are changing the paradigms of user-data interactions, providing an easier and wider access to the Earth Observation data. Some information theory based algorithms are proposed for anomaly and change detection on SPOT images, relative to a widespread humanitarian crisis scenario: floods. The outcomes of these algorithms define an informational representation of the image, revealing the spatial distribution of a particular theme. Using image analysis and interpretation, the multitude of features from a scene are classified into meaningful classes to create sematic maps.\",\n \"Title: A fast compression-based similarity measure with applications to content-based image retrieval\\nYear: 2012\\n\\nAbstract: Compression-based similarity measures are effectively employed in applications on diverse data types with a basically parameter-free approach. Nevertheless, there are problems in applying these techniques to medium-to-large datasets which have been seldom addressed. This paper proposes a similarity measure based on compression with dictionaries, the Fast Compression Distance (FCD), which reduces the complexity of these methods, without degradations in performance. On its basis a content-based color image retrieval system is defined, which can be compared to state-of-the-art methods based on invariant color features. Through the FCD a better understanding of compression-based techniques is achieved, by performing experiments on datasets which are larger than the ones analyzed so far in literature.\",\n \"Title: The Model based Similarity Metric\\nYear: 2007\\n\\nAbstract: This paper addresses the problem of building a compact representation of objects which enables to define a similarity measure. The key idea is to provide a mean to extract and represent the relevant information of an object. By integrating the Minimum Length Description to an informational similarity metric based on Kolmogorov complexity [1], we propose a new informational measure based on models.\",\n \"Title: Cascade active learning for SAR image annotation\\nYear: 2012\\n\\nAbstract: In this paper, a novel active learning approach and system incorporating multiple instance learning for SAR image mining and annotation is introduced. Based on a multiscale and hierarchial patch based image representation, a cascade classifier is learned at different levels. At each level of the hierarchy, a SVM classifier is trained based on active learning and the training sample propagation between different levels is achieved through Multiple Instance SVM (MI-SVM). Classification at the higher level is applied only to the positive patches obtained at the previous level, which can significantly reduce the burden of computation in the case of large data set. Performance has been evaluated through a large data set, which shows promising gain not only in accuracy but also in computation.\",\n \"Title: Rate Distortion Based Detection of Artifacts in Earth Observation Images\\nYear: 2008\\n\\nAbstract: Earth observation optical images may contain blemishes or artificial structures that come from processing or directly from the sensors. These artifacts decrease the quality of the images and can lead to analysis and interpretation problems. We are interested in detecting these defects automatically. We develop an approach based on an information-theoretic analysis of the image formation process, w...\",\n \"Title: Spatial information retrieval from remote-sensing images. I. Information theoretical perspective\\nYear: 1998\\n\\nAbstract: Automatic interpretation of remote-sensing (RS) images and the growing interest for query by image content from large remote-sensing image archives rely on the ability and robustness of information extraction from observed data. In Parts I and II of this article, the authors turn the attention to the modern Bayesian way of thinking and introduce a pragmatic approach to extract structural informati...\",\n \"Title: Earth-Observation Image Retrieval Based on Content, Semantics, and Metadata\\nYear: 2013\\n\\nAbstract: Advances in the image retrieval (IR) field have contributed to the elaboration of tools for interactive exploration and extraction of the images from huge archives associating the content of the images with semantic meaning. This paper presents an Earth-observation (EO) IR system based on enriched metadata, semantic annotations, and image content called EO retrieval. EO retrieval generates an EO-data model by using automatic feature extraction, processing the EO product metadata, and defining semantics, which later is fully exploited for supporting complex queries. In order to demonstrate the functionality of the system, we have created a semantic catalog of TerraSAR-X as application scenario. The database is composed of 39 high-resolution TerraSAR-X scenes comprising about 50 000 image patches (160 × 160 pixels) with their feature descriptors, 100 of metadata entries for each scene, and about 330 semantic annotations. Many query examples combining semantics, metadata, and image content for full exploitation of the image database are presented.\",\n \"Title: Image Time Series Mining For Dynamic Scene Understanding\\nYear: 2003\\n\\nAbstract: In this paper, a dynamic scene understanding concept is proposed and applied on multispectral image time series. Information mining enables the explorations and discovery of spatio temporal patterns localized in given spatio temporal windows.With this in mind, a hierarchical information representation is developed. It comprises different levels in which the data is modeled so that the relevant information is transmitted through the architecture, according to a query and to several assumptions made on the models employed. There are mainly four components : feature extraction, reduction of dimensionality, clustering and interactive exploration.\",\n \"Title: Semantic Annotation of Satellite Images Using Latent Dirichlet Allocation\\nYear: 2010\\n\\nAbstract: In this letter, we are interested in the annotation of large satellite images, using semantic concepts defined by the user. This annotation task combines a step of supervised classification of patches of the large image and the integration of the spatial information between these patches. Given a training set of images for each concept, learning is based on the latent Dirichlet allocation (LDA) mo...\",\n \"Title: Expanding the Algorithmic Information Theory Frame for Applications to Earth Observation.\\nYear: 2013\\n\\nAbstract: Recent years have witnessed an increased interest towards compression-based methods and their applications to remote sensing, as these have a data-driven and parameter-free approach and can be thus succesfully employed in several applications, especially in image information mining. This paper expands the algorithmic information theory frame, on which these methods are based. On the one hand, algorithms originally defined in the pattern matching domain are reformulated, allowing a better understanding of the available compression-based tools for remote sensing applications. On the other hand, the use of existing compression algorithms is proposed to store satellite images with added semantic value.\",\n \"Title: Deep learning in very high resolution remote sensing image information mining communication concept.\\nYear: 2012\\n\\nAbstract: This paper presents the image information mining based on a communication channel concept. The feature extraction algorithms encode the image, while an analysis of topic discovery will decode and send its content to the user in the shape of a semantic map. We consider this approach for a real meaning based semantic annotation of very high resolution remote sensing images. The scene content is described using a multi-level hierarchical information representation. Feature hierarchies are discovered considering that higher levels are formed by combining features from lower level. Such a level to level mapping defines our methodology as a deep learning process. The whole analysis can be divided in two major learning steps. The first one regards the Bayesian inference to extract objects and assign basic semantic to the image. The second step models the spatial interactions between the scene objects based on Latent Dirichlet Allocation, performing a high level semantic annotation. We used a WorldView2 image to exemplify the processing results.\",\n \"Title: Enhancing complex interferograms by anisotropic diffusion\\nYear: 2009\\n\\nAbstract: In this paper a new algorithm for interferometric phase restoration is presented. Firstly, a continuous framework for anisotropic phase diffusion is stated. A tensorial based metric allows directional control. The periodic continuous structure of the phase representation is accounted for. Secondly, this framework is adapted for interferometric phase filtering. Progressive re-estimation of directionality avoids directional bias. Isotropy and anisotropy are adaptively combined with a constant overall diffusion rhythm, so that the degree of regularization is the same regardless of the underlying topography. Robust estimation minimizes the spread of outliers. Results on both synthetic and TerraSAR-X data are provided.\",\n \"Title: Wavelet-Based Despeckling of SAR Images Using Gauss–Markov Random Fields\\nYear: 2007\\n\\nAbstract: In this paper, a wavelet-based speckle-removing algorithm is represented and tested on synthetic aperture radar (SAR) images. The SAR image is first transformed using a dyadic wavelet transform. The noise in the wavelet-transformed image is modeled as an additive signal-dependent noise with Gaussian distribution. The distribution of a noise-free image in a wavelet domain is modeled as a generalize...\",\n \"Title: Coarse-to-Fine Approach for Urban Area Interpretation Using TerraSAR-X Data\\nYear: 2010\\n\\nAbstract: With the launch of the German TerraSAR-X system in June 2007, a new generation of high-resolution spaceborne synthetic aperture radar (SAR) data is available, which should facilitate the interpretation of urban environments. Our overall objective in this letter is to provide a semiautomatic tool for urban area interpretation using SAR data. We propose in this letter to fuse different automatic obj...\",\n \"Title: Information Theoretical Approach For Domain Ontology Exploration In Large Eo Image Archives\\nYear: 2004\\n\\nAbstract: The interpretation of EO data requires not only data and/or information fusion for better understanding of Earth cover structures, but, at a higher level, needs the aggregation with existing bodies of knowledge specific to the application fields.\",\n \"Title: Fusion of Real and Synthetic Images for Remote Sensing Scene Understanding\\nYear: 1993\\n\\nAbstract: INTRODUCTIONThe last generation of Earth Observation Satellites is characterised by a multitude of sensortypes covering a large domain of spectral sensitivity and spatial resolution. The usual sensorsrange from visible, infrared to microwave spectrum, covering coherent or non-coherentradiation, with spatial resolution from meters to kilometres. In the mean time, the huge complexityof land-forms and land-cover types is reflected in the complexity of the recordedimages. As consequence the...\",\n \"Title: Automatic Change Analysis in Satellite Images Using Binary Descriptors and Lloyd–Max Quantization\\nYear: 2015\\n\\nAbstract: In this letter, we present a novel technique for unsupervised change analysis that leads to a method of ranking the changes that occur between two satellite images acquired at different moments of time. The proposed change analysis is based on binary descriptors and uses the Hamming distance as a similarity metric. In order to render a completely unsupervised solution, the obtained distances are further classified using vector quantization methods (i.e., Lloyd's algorithm for optimal quantization). The ultimate goal in the change analysis chain is to build change intensity maps that provide an overview of the severeness of changes in the area under analysis. In addition, the proposed analysis technique can be easily adapted for change detection by selecting only two levels for quantization. This discriminative method (i.e., between changed/unchanged zones) is compared with other previously developed techniques that use principal component analysis or Bayes theory as starting points for their analysis. The experiments are carried on Landsat images at a 30-m spatial resolution, covering an area of approximately 59×51 km2 over the surroundings of Bucharest, Romania, and containing multispectral information.\",\n \"Title: FrFT-Based Scene Classification of Phase-Gradient InSAR Images and Effective Baseline Dependence\\nYear: 2015\\n\\nAbstract: In the literature, scene recognition from interferometric synthetic aperture radar (InSAR) images has been mainly focused on the joint use of the backscatter intensity and the coherence between interferometric image pairs. However, the terrain height information residing in the interferometric phase requires further exploration for classification purposes. In this letter, taking the interferometric phase information into account together with the backscatter intensity, the whole complex- valued InSAR image is exploited for feature extraction. In addition, a new complex-valued phase-gradient InSAR (PGInSAR) image is defined. A fractional-Fourier-transform-based feature ex traction, which was proposed for the classification of single-look complex (SLC) SAR images, is adopted for InSAR and PGInSAR images. For patch-based classification, an image database is generated from bistatic pairs acquired from the same terrain with three different effective baselines. The supervised κ-nearest neighbor classification results show that InSAR outperforms SLC by 15%, whereas PGInSAR introduces further 10% improvement over InSAR or a total improvement of 27% over SLC. Moreover, PGInSAR is found to be more robust to effective baseline changes than InSAR, which makes PGInSAR a better candidate for feature extraction.\",\n \"Title: A Comparative Study of Bag-of-Words and Bag-of-Topics Models of EO Image Patches\\nYear: 2015\\n\\nAbstract: The large volume of detailed land cover features, provided by high resolution Earth observation (EO) images, has attracted considerable interest in the discovery of these features by learning systems. In this letter, we perform latent Dirichlet allocation on the bag of words (BoW) representation of collections of EO image patches to discover their semantic-level features, the so-called topics. To assess the discovered topics, the images are represented based on the occurrence of different topics, called bag of topics (BoT). The value added by BoT to the BoW model of image patches is then measured based on existing human annotations of the data. In our experiments, we compare the classification accuracy results of BoT and BoW representations of two different remote sensing image data sets, a multispectral optical data set and a synthetic-aperture-radar data set. Experimental results demonstrate that BoT can provide a compact and semantically meaningful representation of data; it either causes no significant reduction in the classification accuracy or increases the accuracy by a sufficient number of topics.\",\n \"Title: Big, Linked and Open Data: Applications in the German Aerospace Center.\\nYear: 2014\\n\\nAbstract: Earth Observation satellites acquire huge volumes of high resolution images continuously increasing the size of the archives and the variety of EO products. However, only a small part of this data is exploited. In this paper, we present how we take advantage of the TerraSAR-X images of the German Aerospace Center in order to build applications on top of EO data.\",\n \"Title: On the statistical similarity of synthetic aperture radar images from COSMO-SKYMED and TerraSAR-X\\nYear: 2014\\n\\nAbstract: The latest generation of synthetic aperture radar (SAR) instruments operating in X-band, that is, COSMO-SkyMed (CSK) and TerraSAR-X (TSX), are capable of providing images from coarse resolution to very high resolution. A lot of research effort has been invested in the study and understanding of images obtained from these satellites. However, there is still a huge scope of statistical understanding and comparison of data from both satellites. In this study, we demonstrate some striking similarities between medium resolution data obtained from CSK and TSX Stripmap mode images. Landcover unsupervised clustering using k-means is discussed to further justify our findings. Clustering is carried out using a feature descriptor based on log-cumulants of Gabor coefficients, which was recently proposed by us in earlier studies.\",\n \"Title: Class evolution data analytics from sar image time series using information theory measures\\nYear: 2014\\n\\nAbstract: In this paper we present the result of data analytics techniques applied to a database comprising of 32 SLC SM TerraSAR-X images, acquired over the area of Bucharest, Romania. The methodology follows a two step approach. The first stage consists of a coarse identification of potentially changed areas using a supervised learning image annotation tool with relevance feedback. Gabor texture features are used to describe image patches. The patch size is derived as a function of the resolution and pixel spacing of the data. In the second stage we apply an information theory strategy to refine the regions previously shown to exhibit class dynamics within the image stack, with pixel accuracy. Finally, a series of analytical indicators (absolute extent of areas affected by change, class evolution trends, inter-class correlations) are derived, in order to generate a predictive model for the selected test site.\",\n \"Title: A new comprehensive approach for earth observation scene classification using joint image and text analysis\\nYear: 2014\\n\\nAbstract: The amount of data available on the internet provides massive additional information for the Earth Observation (EO) imagery. Periodical news, various reports and measurements, pictures or online encyclopedias are just few examples of the existent information. Occasionally, this data offers new perspectives for EO image understanding and interpretation. However, current image analysis do not benefit from the advantage given by external sources. To overcome these drawbacks, the present paper proposes an approach that goes beyond traditional information mining by using a joint image and text analysis. Fast Compression Distance (FCD) is computed to measure the similarities inside a collection of very high resolution images and text files. The main purpose is to discover common patterns within the data, without any a priori assumption, parameter-free, relying on data compression-based techniques. A hierarchical clustering is performed in order to learn about the dependencies between different types of data.\",\n \"Title: Despeckling and Information Extraction From SLC SAR Images\\nYear: 2014\\n\\nAbstract: This paper presents an information extraction and image enhancement technique using single-look complex (SLC) synthetic aperture radar data. The novelty of this method is the proposed complex-domain despeckling stage. Tikhonov-like optimization is used for minimizing the cost function, which consists of a Gauss-Markov random field (GMRF) prior. The GMRF model is used for texture modeling. The texture parameters of the GMRF are estimated using the evidence maximization framework. The experimental results showed that despeckled SLC images have well-preserved textural features, structures, and point scatterers. The phase of the reconstructed image is well preserved and provides good-quality interferograms of high-resolution spotlight images.\",\n \"Title: Locally Linear Salient Coding for image classification\\nYear: 2014\\n\\nAbstract: Representing images with their descriptive features is the fundamental problem in CBIR. Feature coding as a key-step in feature description has attracted the attentions in recent years. Among the proposed coding strategies, Bag-of-Words (BoW) is the most widely used model. Recently saliency has been mentioned as the fundamental characteristic of BoW. Base on this idea, Salient Coding (SaC) has been introduced. Empirical studies show that SaC is not able to represent the global structure of data with small number of codewords. In this paper, we remedy this limitation by introducing Locally Linear Salient Coding (LLSaC). This method discovers the global structure of the data by exploiting the local linear reconstructions of the data points. This knowledge in addition to the salient responses, provided by SaC, helps to describe the structure of the data even with a few codewords. Experimental results show that LLSaC obtains state-of-the-art results on various data types such as multimedia and Earth Observation.\",\n \"Title: Comparing the information extracted by feature descriptors from EO images using Huffman coding\\nYear: 2014\\n\\nAbstract: Traditionally, images are understood based on their primitive features such as color, texture, and shape. The proposed feature extraction methods usually cover a range of primitive features. SIFT, for example, in addition to the shape-based information, extracts texture and color information to some extent. Thus, different descriptors may cover a common range of primitive features which we call information overlap. Selecting a set of feature descriptors with low information overlap allows more comprehensive understanding of the data by providing a broader range of new features. This article introduces a new method based on information theory for comparing various descriptors. The idea is to code each description of an image by Huffman coding. The distance between the coded descriptions are then measured using Levenshtein distance as the information overlap. Results show that the computed information overlap clearly describes the differences between the learning from different descriptions of Earth Observation images.\",\n \"Title: Quantitative flood assessment: Case study of floods in Germany\\nYear: 2014\\n\\nAbstract: In this paper, we present a quantitative analysis for a rapid mapping scenario that performs a damage assessment of the 2013 floods in Germany. The scenario is created using pre-disaster and post-disaster TerraSAR-X images and an automated annotation system. Our data set is tiled into patches and Gabor filters are used as a primitive feature method applied to each patch separately. An active learning system based on support vector machine is implemented in order to group the features into categories. Once all categories are identified, these are semantically annotated using reference data as ground truth. In our evaluation 7 categories were retrieved with their specific taxonomies defined using our previous hierarchical annotation scheme. We show that the system supports rapid mapping scenarios (e.g., floods, tsunami, earthquake, etc.) and interactive mapping generation. In addition, with the help of this system, quantitative assessment of disasters can be carried out.\",\n \"Title: Farness preserving Non-negative matrix factorization\\nYear: 2014\\n\\nAbstract: Dramatic growth in the volume of data made a compact and informative representation of the data highly demanded in computer vision, information retrieval, and pattern recognition. Non-negative Matrix Factorization (NMF) is used widely to provide parts-based representations by factorizing the data matrix into non-negative matrix factors. Since non-negativity constraint is not sufficient to achieve robust results, variants of NMF have been introduced to exploit the geometry of the data space. While these variants considered the local invariance based on the manifold assumption, we propose Farness preserving Non-negative Matrix Factorization (FNMF) to exploits the geometry of the data space by considering non-local invariance which is applicable to any data structure. FNMF adds a new constraint to enforce the far points (i.e., non-neighbors) in original space to stay far in the new space. Experiments on different kinds of data (e.g., Multimedia, Earth Observation) demonstrate that FNMF outperforms the other variants of NMF.\",\n \"Title: Image information mining utilizing hierarchical segmentation\\nYear: 2002\\n\\nAbstract: The Hierarchical Segmentation (HSEG) algorithm is an approach for producing high quality, hierarchically related image segmentations. The VisiMine image information mining system utilizes clustering and segmentation algorithms for reducing visual information in multispectral images to a manageable size. The project discussed herein seeks to enhance the VisiMine system through incorporating hierarchical segmentations from HSEG into the VisiMine system. I. INTRODUCTION The Hierarchical Segmentation (HSEG) algorithm is an approach for producing high quality, hierarchically related image segmentations. The HSEG algorithm is a hybrid of region growing and spectral clustering that produces a hierarchical set of image segmentations. Different image analysis applications often require different levels of image segmentation detail. Further, a particular application may require different levels of image segmentation detail depending on the image objects being segmented. Thus, a choice of several levels of image segmentation detail is often very useful. The HSEG algorithm automatically produces a hierarchical set of segmentations, based on detected convergence points. An analyst can select between segmentations at different levels in the segmentation hierarchy to produce the most appropriate segmentation for a particular application. The VisiMine image information mining system utilizes clustering and segmentation algorithms for reducing visual information in multispectral images to a manageable size. The indexing strategy extracts three levels of features: 1) pixel level, 2) polygon level, and 3) tile level features. The pixel level features are used for the extraction of higher-level features. Fast segmentation algorithms based on PDE methods are used to decompose image tiles into polygons. In the paper we compare the results of similarity searches utilizing features extracted using an enhanced version of CLARA algorithm (Clustering LARge Applications) in combination with statistical sampling with the results that use hierarchical image segmentation (HSEG) to derive class This project is supported by NASA's Intelligent Systems Program, NASA Headquarters Code R. boundaries for land cover. We present the results of polygon and tile similarity searches based on spectral and textural properties. The use of HSEG features improves the precision and recall of similarity searches. The interactive label training and retrieval based on Bayesian models also benefits from hierarchical image segmentation approach.\",\n \"Title: Multisource data classification with dependence trees\\nYear: 2002\\n\\nAbstract: Abstract—In order to apply a statistical approach to the classi- fication of multisource remote-sensing data, one of the main prob- lems to face lies in the estimation of probability distribution func- tions. This problem arises out of the difficulty of defining a common statistical model for such heterogeneous data. A possible solution is to adopt nonparametric approaches, which rely on the availability of training samples without any assumption about the related sta- tistical distributions. The purpose of this paper is to investigate the suitability of the concept of dependence trees for the integration of multisource information through estimation of probability distri- butions. First, this concept, introduced by Chow and Liu, is used to provide an approximation of a probability distribution defined in an,-dimensional space by a product of,probability dis- tributions defined in two-dimensional (2-D) spaces; this approxi- mation corresponds, in terms of graph theoretical interpretation, to a tree of dependence. For each land cover class, a dependence tree is generated by minimizing an appropriate closeness measure. Then, a nonparametric estimation of the second-order probability distributions is carried out through the Parzen window approach, based on the implementation of 2-D Gaussian kernels. In this way, it is possible to reduce the complexity of the estimation, while cap- turing a significant part of the interdependence among variables. A comparison with other multisource data fusion methods, namely, the multilayer perceptron (MLP) method, the -nearest neighbor ( -NN) method, and a Bayesian hierarchical classifier (BHC), is made. Experimental results obtained on multisensor [airborne the- matic mapper,(ATM) and synthetic aperture radar (SAR)] and multisource (experimental synthetic aperture radar (E-SAR) and a textural feature) data sets show that the proposed fusion method based on dependence trees is able to provide a classification accu- racy similar to those of the other methods considered, but with the advantage of a reduced computational load. Index Terms—Dependence trees, discrete probability distribu-\",\n \"Title: Study and assessment of selected primitive features behaviour for SAR image description\\nYear: 2012\\n\\nAbstract: The main purpose of this study is to define for Synthetic Aperture Radar (SAR) data the primitive feature parameters, the incidence angle, and the orbit direction which can be used further for indexing and querying in the EO systems. The evaluation is done on the high resolution SAR data and the interpretation is realized automatically. In this paper, we propose to study and asses the behavior of the primitive feature extracted methods for images of the same scene with two look angles covering the min-max range of the sensor and with ascending / descending orbit looking. The tests are done on TerraSAR-X products Stripmap and high resolution Spotlight, specially and radiometrically enhanced covering the area of Berlin (Germany) and Ottawa (Canada). To identify the optimal primitive features, incident angle, and orbit direction the Support Vector Machine and as a measure of the classification accuracy the precision/recall were considered.\",\n \"Title: Automated interpretation of very-high resolution SAR images\\nYear: 2012\\n\\nAbstract: Very-high resolution (VHR) synthetic aperture radar (SAR) images from the last generation satellites such as TerraSAR-X and TanDEM-X exhibits special characteristics, especially in the urban-areas. Consequently, attention is needed on special considerations while developing algorithms for SAR image processing and its applications for automated interpretation. With automatic interpretation we refer to the information extraction and characterization for image categorization, retrieval, segmentation, automated target recognition etc.. In this article we focus our attention to the problem of SAR image categorization. The interest in SAR image categorization in VHR SAR (on the contrary to the pixel-based classification in low-resolution SAR images) has increased with enhanced resolution providing opportunity to carry out a more detailed analysis of targets and objects. SAR image categorization requires generation of a compact feature descriptor which accurately define the image content. A feature descriptor can be generated using `parametric' or `non-parametric' approaches in `image' or `image within a transformation space'. The objective of this article is to review some selected techniques for this purpose in form of a methodological classification. Qualitative assessment of selected algorithms is presented.\",\n \"Title: SBAS and PS measurement fusion for enhancing displacement measurements\\nYear: 2012\\n\\nAbstract: A major drawback of classical SAR interferometry is its sensitivity to temporal and geometric effects, leading to a distortion of the final results. This problem can be solved by using multi-temporal techniques, like the Small Baseline Subset and Persistent Scatterers algorithms. Both use large datasets for monitoring land cover deformation, motion or for DEM generation. Despite the fact that using the Persistent Scatterers technique can lead to a maximization of the number of acquisitions used, the number of persistent points is greatly reduced by a decrease of the coherent values of the targets over time. In this article, the authors propose a method for the selection of temporal persistent points using image subsets that comply with the small baseline rule. This method is further used to increase the accuracy of deformation measurements over the subsets' time span.\",\n \"Title: Assessment of Earth Observation data content based on data compression - application to settlements understanding\\nYear: 2012\\n\\nAbstract: Urban areas around the world are rapidly changing in an unregulated manner and remote sensing is the most effective option for their monitoring and planning. Good modeling of urban areas means reliable translation of the scene semantics into an algorithmic language. The compression based image retrieval techniques are data driven. The intention of employing compression based image retrieval techniques is to exploit the compression properties of the objects and estimate the shared information between them. Fast compression distance (FCD) is the similarity metric used in a compression based image retrieval technique that can be applied on large datasets. FCD between any two objects can be computed using the sizes of their dictionaries (sequence of recurring patterns) extracted through compression with LZW algorithm and the intersection of their dictionaries. In this paper, it is proposed to assess high resolution Earth Observation data content based on data compression for understanding urban settlements.\",\n \"Title: Query by example in Earth-Observation image archive using data compression-based approach\\nYear: 2012\\n\\nAbstract: This paper presents an implementation of query by example in Earth Observation image archive using data compression-based approach. Data compression approach allows to exploit the compression properties of the objects and to estimate the shared information between them, this concept is extended to image retrieval for finding similar objects in the image archive. Our implementation is based on LZW algorithm for compressing the image content and extracting features of the images. The fast compression distance (FCD) is defined as a similarity metric in order to retrieve the most similar images. This tool is satisfactory implemented and tested using optical and SAR images.\",\n \"Title: Latent Dirichlet Allocation for Spatial Analysis of Satellite Images\\nYear: 2013\\n\\nAbstract: This paper describes research that seeks to supersede human inductive learning and reasoning in high-level scene understanding and content extraction. Searching for relevant knowledge with a semantic meaning consists mostly in visual human inspection of the data, regardless of the application. The method presented in this paper is an innovation in the field of information retrieval. It aims to discover latent semantic classes containing pairs of objects characterized by a certain spatial positioning. A hierarchical structure is recommended for the image content. This approach is based on a method initially developed for topics discovery in text, applied this time to invariant descriptors of image region or objects configurations. First, invariant spatial signatures are computed for pairs of objects, based on a measure of their interaction, as attributes for describing spatial arrangements inside the scene. Spatial visual words are then defined through a simple classification, extracting new patterns of similar object configurations. Further, the scene is modeled according to these new patterns (spatial visual words) using the latent Dirichlet allocation model into a finite mixture over an underlying set of topics. In the end, some statistics are done to achieve a better understanding of the spatial distributions inside the discovered semantic classes.\",\n \"Title: Image time-series mining\\nYear: 2004\\n\\nAbstract: A visual information mining concept is proposed for spatio-temporal patterns discovery in remotely sensed image time-series (ITS). An information theory framework is adopted to first model information content. It results in the inference of a relevant directed graph characterizing ITS. Then the user conjecture is modeled via visual information representations : similarity measures between sub-graphs, which represents spatio- temporal events are derived and included in an interactive learning and probabilistic retrieval procedure of user-specific event-types. The present concept for ITS mining is demonstrated on multitemporal SPOT data. During the last decades, the imaging satellite sensors have acquired huge quantities of data enabling the elaboration of spatio-temporal databases. Image time-series (ITS) contained in them are complex objects containing interesting informa- tion. However, our capability to access and store large volume of data has highly exceeded our capability to extract the relevant information from the data. Therefore, ITS information systems are needed to bridge the gap between database theory, which is not adapted to manipulate temporal and pictural multidimensional data, and computer vision, which does not offer query possibilities. To cover such a broad problematic, a visual information mining concept is proposed to help a user to explore and understand spatio-temporal behaviors present in ITS. The article begins by presenting unsupervised methods for coding the information content in the observations. Thus, the dynamics of the feature space (FS) is modeled which results in the inference of a relevant directed graph representation, characterizing ITS. It continues by describing sub-graphs, contained in the graph, with model parameters. Combinations of the parameters correspond to semantic descriptions of spatio-temporal events. In this perspective, a distance model between sub-graph parameters is developed to enable user specific similarity measurements. Then, a Bayesian interactive learning on graphs is presented to enable reasoning on ITS. The learning is performed by the incremental definition of a spatio-temporal pattern type via user-provided positive and negative sub-graph examples. In parallel, the interactive search of spatio-temporal events is guided by accuracy probabilistic measurements derived from the current state of learning via the user positive and negative semantic. Finally, experiments\",\n \"Title: Object and topology extraction from remote sensing images\\nYear: 2005\\n\\nAbstract: We present a complete processing line to generate an object based description of optical remote sensing (RS) images. A segmentation algorithm is used to generate a partition of regions and simplify the volume of data. Results are still at the pixel level. Based on topology analyses, a dynamical algorithm is proposed to retrieve, extract the segmented regions and encode them in a tree structure which describes their topological relations (adjacencies, inclusions). The overall collected information constitutes a consistent and independent database, generated efficiently on standard workstation. Many applications are possible, such as content based image retrieval, image description and compression, object classification or image-object fusion. A scenario is presented to emphasize interest of the method in the case of 3D visualization enhancement: image-objects are integrated on elevation data (digital elevation models, DEM) in order to generate more realistic rendering.\",\n \"Title: Model-based despeckling and information extraction from SAR images\\nYear: 2000\\n\\nAbstract: Basic textures as they appear, especially in high res- olution SAR images, are affected by multiplicative speckle noise and should be preserved by despeckling algorithms. Sharp edges between different regions and strong scatterers also must be pre- served. To despeckle images, we use a maximum a posteriori (MAP) estimation of the cross section, choosing between different prior models. The proposed approach uses a Gauss Markov random field (GMRF) model for textured areas and allows an adaptive neigh- borhood system for edge preservation between uniform areas. In order to obtain the best possible texture reconstruction, an expec- tation maximization algorithm is used to estimate the texture pa- rameters that provide the highest evidence. Borders between ho- mogeneous areas are detected with a stochastic region-growing al- gorithm, locally determining the neighborhood system of the Gauss Markov prior. Smoothed strong scatterers are found in the ratio image of the data and the filtering result and are replaced in the image. In this way, texture, edges between homogeneous regions, and strong scatterers are well reconstructed and preserved. Addi- tionally, the estimated model parameters can be used for further image interpretation methods.\",\n \"Title: Introduction to the Special Section on Image Information Mining for Earth Observation Data\\nYear: 2007\\n\\nAbstract: The eight papers in this special section describe some of the recent advances made in the field of image information mining (IIM) for Earth Observation in the development of tools, methods, and applications.\",\n \"Title: Image Time-Series Data Mining Based on the Information-Bottleneck Principle\\nYear: 2007\\n\\nAbstract: Satellite image time series (SITS) consist of a time se- quence of high-resolution spatial data. SITS may contain valuable information, but it may be deeply hidden. This paper addresses the problem of extracting relevant information from SITS based on the information-bottleneck principle. The method depends on suitable model selection, coupled with a rate-distortion analysis for determining the optimal number of clusters. We present how to use this method with the Gauss-Markov random fields and the autobinomial random fields model families in order to character- ize the spatio-temporal structures contained in SITS. Experimen- tal results on synthetic data and SITS from SPOT demonstrate the performance of the proposed methodology.\",\n \"Title: A Similarity Metric for Retrieval of Compressed Objects: Application for Mining Satellite Image Time Series\\nYear: 2008\\n\\nAbstract: This paper addresses the problem of building an index of compressed object databases. We introduce an informational similarity measure based on the coding length of two part codes. Then, we present a methodology for compressing the database by taking into account interobject redundancies and by using the informational similarity measure. The method produces an index included in the code of the data volume. This index is built such that it contains the minimal sufficient information to discriminate the data-volume objects. Then, we present an optimal two-part coder for compressing spatio-temporal events contained in satellite image time series (SITS). The two-part coder allows us to measure similarity and then to derive an optimal index of SITS spatio-temporal events. The resulting index is representative of the SITS information content and enables queries based on information content.\",\n \"Title: Knowledge centred Earth Observation: Feature Extraction\\nYear: 2007\\n\\nAbstract: Nowadays huge volumes of Earth Observation data are available and increasing every day. Users are faced with the problem of having to extract and interpret information in large volumes of data, with the additional problem of images being annotated only by simple descriptors. Therefore, in this paper we present the Knowledge-centered Earth Observation system which offers interactive probabilistic information mining, among other functionalities, and two Support Vector Machine based Feature Extraction services integrated in the system.\",\n \"Title: Learning - unlearning for mining high resolution EO images\\nYear: 2007\\n\\nAbstract: The last two decades showed an important development of satellite imagery with past and present satellites acquiring enormous volumes of data. Meanwhile, the quality of the acquired images increased permitting the recording of high resolution images (0.6divide2.5 meters/pixel) in multispectral bands. Thus, both the data volume and the information detail increase dramatically. Consequently, new methods and tools to access and interpret earth observation (EO) images are needed. The present paper presents a semantic search engine for high resolution (HR) EO images based on a hierarchical information model of satellite image contents. To face the potentially ambiguous meaning of image structures depending on their contextual understanding, the search engine uses Bayesian inference to learn categories and a support vector machine (SVM) classifier to assign semantics. The categories are grouping and memorising the semantics of image structures, facilitating their recognition in various contexts. Also the generation of categories helps learning from a small training data set (i.e. image examples); thus, the method is useful for the exploitation of very large data volumes. The concept has enhanced inferred power, therefore optimising the human machine communication (HMC), which is enhanced with learning / unlearning functions.\",\n \"Title: TerraSAR-X: Complex Image Inversion for Feature Extraction\\nYear: 2008\\n\\nAbstract: In this paper we present two algorithms for information extraction from Single Look Complex (SLC) Synthetic Aperture Radar (SAR) images. The first algorithm is based on Tikhonov regularization with Total Variation (TV) and a Point-Based Feature (PBF) term. Based on the equivalence of Tikhonov and the Bayesian estimate, the second algorithm is a Maximum A Posteriori (MAP) estimation with a complex-valued Gauss-Markov Random Field (GMRF) in addition to the TV prior. The first algorithm produces a despeckled image preserving fine details and texture. The second algorithm gives a denoised image and in addition the estimated feature parameter vector thetas.\",\n \"Title: Geometrical and Topological Urban Areas Characterization using TerraSAR-X Data\\nYear: 2008\\n\\nAbstract: With the launch of the German TerraSAR-X system in June 2007, a new generation of high-resolution spaceborne synthetic aperture radar (SAR) data is available; which should facilitate the interpretation of urban environments. This article proposes a new automatic tool for geometric and topological urban areas characterization. Our approach is divided into three main steps. First, a bright linear structures detector is applied to extract the geometrical information. Then, a graph-based spatial characterization is used to model the topological relationships between the different detected bright pixels (nodes). Next, a classification by examining the profile of the distributions of the angles between neighboring nodes, is performed in order to label the linked structures. Evaluations of the proposed approach in characterizing the geometry and topology of urban areas were carried out using TerraSAR-X data over three different cities: Las Vegas in the United States, Paris in France and Cairo in Egypt.\",\n \"Title: Algorithmic Cross-Complexity and Relative Complexity\\nYear: 2009\\n\\nAbstract: Information content and compression are tightly related concepts that can be addressed by classical and algorithmic information theory. Several entities in the latter have been defined relying upon notions of the former, such as entropy and mutual information, since the basic concepts of these two approaches present many common tracts. In this work we further expand this parallelism by defining the algorithmic versions of cross-entropy and relative entropy (or Kullback-Leiblerdivergence), two well-known concepts in classical information theory. We define the cross-complexity of an object x with respect to another object y as the amount of computational resources needed to specify x in terms of y, and the complexity of x related to y as the compression power which is lost when using such a description for x, with respect to its shortest representation. Since the main drawback of these concepts is their uncomputability, a suitable approximation based on data compression is derived for both and applied to real data. This allows us to improve the results obtained by similar previous methods which were intuitively defined.\",\n \"Title: Wavelet-Based SAR Image Despeckling and Information Extraction, Using Particle Filter\\nYear: 2009\\n\\nAbstract: This paper proposes a new-wavelet-based synthetic aperture radar (SAR) image despeckling algorithm using the sequential Monte Carlo method. A model-based Bayesian approach is proposed. This paper presents two methods for SAR image despeckling. The first method, called WGGPF, models a prior with Generalized Gaussian (GG) probability density function (pdf) and the second method, called WGMPF, models prior with a Generalized Gaussian Markov random field (GGMRF). The likelihood pdf is modeled using a Gaussian pdf. The GGMRF model is used because it enables texture parameter estimation. The prior is modeled using GG pdf, when texture parameters are not needed. A particle filter is used for drawing particles from the prior for different shape parameters of GG pdf. When the GGMRF prior is used, the particles are drawn from prior in order to estimate noise-free wavelet coefficients and for those coefficients the texture parameter is changed in order to obtain the best textural parameters. The texture parameters are changed for a predefined set of shape parameters of GGMRF. The particles with the highest weights represents the final noise-free estimate with corresponding textural parameters. The despeckling algorithms are compared with the state-of-the-art methods using synthetic and real SAR data. The experimental results show that the proposed despeckling algorithms efficiently remove noise and proposed methods are comparable with the state-of-the-art methods regarding objective measurements. The proposed WGMPF preserves textures of the real, high-resolution SAR images well.\",\n \"Title: Satellite image artifacts detection based on complexity distortion theory\\nYear: 2011\\n\\nAbstract: The artifacts detection is a step of data cleaning process. The classical approach is to predict or determine the existence of defects, to model it, and then design a method to detect and correct them. This classical approach is for specific artifacts. The approach presented in this work is using complexity distortion theory to implement a more generic method, thus, this work will aim at developing parameter free methods able to automatically detect artifacts in EO images. We use the Kolmogorov Structure Function as approximation to the Rate-distortion curve and examine how the artifacts can have the same structure.\",\n \"Title: Mining large satellite image repositories using semi-supervised methods\\nYear: 2011\\n\\nAbstract: The increasing number and resolution of earth observation (EO) imaging sensors has had a significant impact on both the acquired image data volume and the information content in images. There is consequently a strong need for highly efficient search tools for EO image databases and for search methods to automatically identify and recognize structures within EO images. In this paper, we present a concept for an earth observation image data mining system mixing an auto-annotation component with a category search engine which combines a generic image class search and an object detection feature. The proposed concept relies thus on three distinct components which are detailed successively: in the first part, we describe the auto-annotation component, in the second part, the generic category search engine and in the third part, the object detection tool. In the concluding part of the paper, we provide an insight into how these three components can be related to each other and used in a complementary way to arrive at a system which combines the advantages of both the auto-annotation systems and the category search engines.\",\n \"Title: Algorithmic Information Theory-Based Analysis of Earth Observation Images: An Assessment\\nYear: 2010\\n\\nAbstract: Earth observation image-understanding methodologies may be hindered by the assumed data models and the estimated parameters on which they are often heavily dependent. First, the definition of the parameters may negatively affect the quality of the analysis. The parameters could not be captured in all aspects, and those resulting superfluous or not accurately tuned may introduce nuisance in the data. Furthermore, the diversity of the data, as regards sensor type, spatial, spectral, and radiometric resolution, and the variety and regularity of the observed scenes make it difficult to establish enough valid and robust statistical models to describe them. This letter proposes algorithmic information theory-based analysis as a valid solution to overcome these limitations. We will present different applications on satellite images, i.e., clustering, classification, artifact detection, and image time series mining, showing the generalization power of these parameter-free data-driven methods based on the computational complexity analysis.\",\n \"Title: System Design Considerations for Image Information Mining in Large Archives\\nYear: 2010\\n\\nAbstract: Generally, the repository of a remote sensing system needs a capacity of hundreds of terabytes to archive original and processed data. There are several satellites whose downlink rates require ingesting over 100 GB into the archive each day. In order to process such a huge volume of data, to classify them, to index generated classes in the database, and to provide a quick access to stored information at real time, an image information mining system is developed. In this letter, we present the design of a system architecture for processing data from very large image archives.\",\n \"Title: A survey of density estimation for sar images\\nYear: 2012\\n\\nAbstract: Modeling of synthetic aperture radar (SAR) images has been an important topic of research since the inception of SAR satellites. Many theoretical and empirical models have been presented in literature to accurately model the amplitude SAR images. The method of parameters estimation of the probability density function (PDF) for selected models is another topic of research associated with modeling. Earlier the maximum-likelihood (ML) methodology or the methods of moments (MoM) were used for the parameter estimation. The method of logarithmic-cumulants (MoLC), which has been proposed for the parameter estimation for the PDF defined in R+, is now a very popular tool for the efficient parameter estimation for amplitude SAR images. In this article, we present a survey of some of the well-known PDFs proposed for SAR amplitude images by carrying out the parameter estimation with the MoLC method. The objective is to demonstrate that the statistical characterization of SAR images is strongly dependent upon the observed scene content. Instead of using a set of images, we carry out this study on set of object/texture categories on a larger data-base.\",\n \"Title: Information Content of Very High Resolution SAR Images: Study of Feature Extraction and Imaging Parameters\\nYear: 2013\\n\\nAbstract: In this paper, we propose to study the dependence of information extraction technique performance on synthetic aperture radar (SAR) imaging parameters and the selected primitive features (PFs). The evaluation is done on TerraSAR-X data, and the interpretation is realized automatically. In the first part of this paper (use case I), the following issues are analyzed: 1) finding the optimal TerraSAR-X products and their limits of variability and 2) retrieving the number of categories/classes that can be extracted from the TerraSAR-X images using the PFs (gray-level co-occurrence matrix, Gabor filters, quadrature mirror filters, and nonlinear short-time Fourier transform). In the second part of this paper (use case II), we investigate the invariance of the products with the orbit direction and incidence angle. On the one hand, the results show that using ascending looking is better than using descending looking with an average accuracy increase of 7%-8%, approximately. On the other hand, the classification accuracy for the incidence angle varies from a lower value of the incidence to an upper value of the incidence angle (depending on the sensor range) with 4%-5%. The test sites are Venice (Italy), Toulouse (France), Berlin (Germany), and Ottawa (Canada) and are covering as much as possible the huge diversity of modes, types, and geometric resolution configuration of the TerraSAR-X. For the evaluation of all these parameters (resolution, features, orbit looking, and incidence angle), the support-vector-machine classifier is considered. To evaluate the accuracy of the classification, the precision/recall metric is calculated. The first contribution of this paper is the evaluation of different PFs (proposed in the literature for different types of images) and adaptation of these for SAR images. These features are compared (based on the accuracy of the classification) for the first time for a multiresolution pyramid specially built for this purpose. During the evaluation,- all the classes were annotated, and a semantic meaning was defined for each class. The second main contribution of this paper is the evaluation of the dependence on the patch size, orbit direction, and incidence angle of the TerraSAR-X. This type of evaluation has not been systematically investigated so far. For the evaluation of the optimal patch, two different patch sizes were defined, with the constrained that the size on ground needs to cover a minimum of one object (e.g., 200 × 200 m on ground). This patch size depends also on the parameters of the data such as resolution and pixel spacing. The investigation of orbit looking and incidence angle is very important for indexing large data sets that has a higher variability of these two parameters. These parameters influence the accuracy of the classification (e.g., if the incidence angle is closer to the lower bounds or closer to the upper bound of the satellite sensor range).\",\n \"Title: Multitemporal analysis of multisensor data: Information theoretical approaches\\nYear: 2010\\n\\nAbstract: This paper presents two approaches for the analysis of multi-temporal and multisensory data analysis. The first approach focuses on a novel similarity measure, named Mixed Information, derived from information measures and assessment of its performance for multitemporal analysis in change detection. The second approach proposes the use of Kullback-Leibler divergence between marginal distributions, efficiently approximated by cumulant based expansion series. Comparison between mixed information and Kullback-Leibler divergence based change detection is performed to confirm the usefulness of joint metric for analyzing multitemproal and multisensory data. Experimental results obtained confirm justification of the mixed information in multitemporal analysis.\",\n \"Title: Image information mining methods for exploring and understanding high resolution images\\nYear: 2010\\n\\nAbstract: This paper discusses the basic paradigm of how image information mining methods work in the field of remote sensing. To this end, we compare our approaches to the approaches being used in the world of multimedia; then we discuss the annotation specifics of remote sensing data and describe the different types of remote sensing data that we are faced with today. We conclude with a description of algorithmic alternatives and compare several competing systems that are currently in use. Finally, we provide an outlook of what we expect in the near future.\",\n \"Title: How many categories are in very high resolution SAR images?\\nYear: 2013\\n\\nAbstract: In this paper, we propose to identify the number of categories that can be retrieved from a very high resolution SAR data. The evaluation is done on TerraSAR-X high resolution Spotlight data and the retrieved categories are semi-automatically annotated using as feature vector the Gabor filters; as a classifier the Support Vector Machine, and for ranking the suggested images the relevance feedback. The visualization of the tool was enhanced compared with our previous implementation in order to support the users in his/her approach to search the patches of interest in a large repository. Our dataset consist in 43 scenes that cover as much as possible all the regions over the world. A total of 352 categories are identified that contain urban and non-urban categories.\",\n \"Title: The impact of rain, frost, seasonal cycle, and wind on sequences of high resolution urban SAR images\\nYear: 2013\\n\\nAbstract: The TerraSAR-X satellite has been fully operational since nearly 6 years and has delivered a very large quantity of high resolution SAR images. Among these images one can find a number of repeated acquisitions of selected target areas taken with nearly identical imaging parameters. These image time series data often cover the full seasonal cycle of a target area and lend themselves well to small-scale change detection; however, in the case of urban areas, we have to be aware of the quantitative impact of rain, frost and wind on high resolution SAR images. In particular, images of Western European cities are characterized by construction work concentrating on single buildings within a fully built-up city and by public green space changes. A quantitative analysis of urban time series data has to discriminate between definite changes of the urban landscape and the transient impacts of rain, frost and wind. The critical issue is how to identify and characterize these transient phenomena.\",\n \"Title: Measuring the semantic gap based on a communication channel model\\nYear: 2013\\n\\nAbstract: The collected Earth Observation (EO) data volumes are increasing immensely. In the meantime, the need for retrieval of focused information for decision making is increasing. Due to the particular nature of EO sensors, recording signals very differently than humans perceptual system, the challenges raised by the semantic and sensory gaps are immensely amplified in designing retrieval methods for EO images. This article introduces a method based on communication channel model to quantify and measure the semantic gap, used to assess various feature descriptors for semantic annotation purposes. The approach uses Latent Dirichlet Allocation (LDA), considering images as the source and the semantic topics as the receiver. The parameters of LDA are estimated for computing the Mutual Information to assess latent semantics of feature space. We further introduce a method to measure the distance between humans' and computer's semantics. The results are validated using an SVM-based classifier for an annotated dataset.\",\n \"Title: Contextual Descriptors for Scene Classes in Very High Resolution SAR Images\\nYear: 2012\\n\\nAbstract: The new generation of spaceborne SAR instruments with meter or submeter resolution finds enormous applications for the observation of urban, industrial, in general of man-made scenes. Thus, targets are not any more observed in isolation, instead the groups of objects, e.g., house, bridge, and road, etc., need to be recognized in their spatial context. This paper proposes a feature extraction method for image patches in order to capture the spatial context. The method is based on the characteristics of the spectra of the SAR data, integrating radiometric, geometric, and texture properties of the SAR image patch. The method is demonstrated for TerraSAR-X High Resolution Spotlight data. To account for the spatial context in which a group of targets is located, it uses an image patch covering typically 200 × 200m2 of the scene. A comparative evaluation of our descriptors and grey-level co-occurrence matrix (GLCM) texture features has been performed over a database of 6916 patches. The method allowed for the robust recognition of over 30 different scene classes, with precision between 50% and 93%. Numerical results show that our method is able to discriminate between scene classes better than GLCM texture parameters.\",\n \"Title: Scene recognition based on phase gradient InSAR images\\nYear: 2014\\n\\nAbstract: Automated recognition of SAR images requires feature extraction from complex-valued data. In this work, complex-valued interferometric SAR (InSAR) images, which are mainly used to construct elevation models, are proposed for feature extraction. Feature extraction based on the log-cumulants of fractional Fourier transform (FrFT) coefficients has already been proposed in the literature and found to be quite successful for the classification of single-look complex (SLC) SAR images. Here, this method is applied to the complex-valued InSAR and newly introduced complex-valued phase gradient InSAR (PGInSAR) images. In order to evaluate the classification performance for SLC, InSAR and PGInSAR images, a database of bistatic TanDEM-X interferometric pairs is constructed, and a supervised KNN classification is performed. The overall classification accuracies for a 4% training set size show that the use of InSAR and PGInSAR images outperforms SLC images by 15% and 24%, respectively. In terms of individual class accuracies, the biggest improvement is observed in agricultural fields and mixed vegetation areas.\",\n \"Title: New concepts for earth observation data catalogues\\nYear: 2002\\n\\nAbstract: Conventional satellite data ground segments comprise sensor mode planning, data acquisition systems, data ingestion interfaces, processing capabilities, the catalogue of available data, a data archive, and interfaces for queries and data dissemination. In the present architecture of the systems to access satellite image archives, from the perspective of the information access, the catalogue plays the central role. The catalogue stores the meta information necessary to specify a query. The result of a query is given as a list of images and as quick looks. It is the user who has to browse visually over the quick looks to restrict the result of the query according to his interest. In order to have the possibility to query such catalogues by using the image content, we present a new concept based on an interactive learning of the image semantics, as given by the user.\",\n \"Title: Semi-supervised learning and discovery of unkown structures among data: Application to satellite image annotation\\nYear: 2009\\n\\nAbstract: In this paper, we present a semi-supervised method for auto-annotating image collections and discovering unknown structures among them. The approach relies on the existence of only a small training database of annotated examples. First, a fully-supervised algorithm using annotated samples is presented. Next, we introduce a semi-supervised procedure which allows us to incorporate unannotated samples and to infer the existence of unknown structures, that is, the existence of new image classes which are not represented in the training database. Finally, we present experimental results from a database of satellite images and briefly mention the possibility of reusing the presented approach as a basis for more complex systems such as Content Based Image Retrieval (CBIR) systems.\",\n \"Title: Parametric versus non-parametric complex image analysis\\nYear: 2009\\n\\nAbstract: In this paper we compare parametric and non-parametric method for the analysis of complex valued high-resolution SAR data. Gauss-Markov Random Field (GMRF) model with a quadratic energy function as a parametric analysis parameterizes the spectogram of the signal, whereas nonlinear short time Fourier transform (STFT) method, the method based on time frequency analysis (TFA) as a non-parametric approach exploits the signal's non-stationarity in the time-frequency domain for information extraction. This comparative analysis helps to understand, characterize and analyze complex valued SAR data.\",\n \"Title: A Similarity Measure Using Smallest Context-Free Grammars\\nYear: 2010\\n\\nAbstract: This work presents a new approximation for the Kolmogorov complexity of strings based on compression with smallest Context Free Grammars (CFG). If, for a given string, a dictionary containing its relevant patterns may be regarded as a model, a Context-Free Grammar may represent a generative model, with all of its rules (and as a consequence its own size) being meaningful. Thus, we define a new complexity approximation which takes into account the size of the string model, in a representation similar to the Minimum Description Length. These considerations result in the definition of a new compression-based similarity measure: its novelty lies in the fact that the impact of complexity overestimations, due to the limits that a real compressor has, can be accounted for and decreased.\",\n \"Title: A Semi-Supervised Algorithm for Auto-Annotation and Unknown Structures Discovery in Satellite Image Databases\\nYear: 2010\\n\\nAbstract: The increasing number and resolution of earth observation (EO) imaging sensors has had a significant impact on both the acquired image data volume and the information content in images. There is consequently a strong need for highly efficient search tools for EO image databases and for search methods to automatically identify and recognize structures within EO images. Content Based Image Retrieval (CBIR) and automatic image annotation systems have been designed to tackle the problem of image retrieval in large image databases. These two systems achieve a common goal, that is to learn the mapping function between low-level visual features and high-level image semantics. A setup, which has hardly been explored in annotating systems and which is the rule rather than the exception, is the case when the training database used to learn the mapping function is not exhaustive regarding semantic classes present in the images. This means that there exists unknown image classes for which there is no training examples in the training database. In this paper, we propose a semi-supervised method for auto-annotating satellite image databases and discovering unknown semantic image classes in these databases. The idea is to incorporate into the learning process the unannotated data which by definition contain the unknown image classes. The latter are considered to be latent structures in the data that appear when we train a hierarchical latent variable model with both the labeled and unlabeled data. We also show that, in our case, the use of unlabeled data leads to more reliable estimates regarding the model parameters. We present experimental results on a synthetic dataset, making a comparison of our algorithm with a semi-supervised Support Vector Machine (S3VM) on this dataset. We also demonstrate the effectiveness of our unknown image classes discovery procedure on a database of SPOT5 satellite images. We show that the results obtained on this database are rather positive since t- - he new structures detected correspond to semantic classes which are not represented in the training database.\",\n \"Title: Texture estimation in sar images: The impact of scale and model order\\nYear: 2010\\n\\nAbstract: This paper discusses methods and parameter settings that help to estimate texture in SAR images. In general, this is a difficult task for SAR images that are characterized by speckle noise and which span a wide range of pixel magnitudes. We applied Gauss Markov Random Field (GMRF) models and Enhanced Model Based Despeckling (EMBD) to 1 meter resolution amplitude images of the German TerraSAR-X mission. The results demonstrate that one can find appropriate parameter combinations that allow robust texture estimation even for different types of target areas.\",\n \"Title: Assessment of dimensionality reduction based on communication channel model; application to immersive information visualization\\nYear: 2013\\n\\nAbstract: We are dealing with large-scale high-dimensional image data sets requiring new approaches for data mining where visualization plays the main role. Dimension reduction (DR) techniques are widely used to visualize high-dimensional data. However, the information loss due to reducing the number of dimensions is the drawback of DRs. In this paper, we introduce a novel metric to assess the quality of DRs in terms of preserving the structure of data. We model the dimensionality reduction process as a communication channel model transferring data points from a high-dimensional space (input) to a lower one (output). In this model, a co-ranking matrix measures the degree of similarity between the input and the output. Mutual information (MI) and entropy defined over the co-ranking matrix measure the quality of the applied DR technique. We validate our method by reducing the dimension of SIFT and Weber descriptors extracted from Earth Observation (EO) optical images. In our experiments, Laplacian Eigenmaps (LE) and Stochastic Neighbor Embedding (SNE) act as DR techniques. The experimental results demonstrate that the DR technique with the largest MI and entropy preserves the structure of data better than the others.\",\n \"Title: Pheromone averaging exploration algorithm\\nYear: 2015\\n\\nAbstract: In this paper we introduce a novel spatial exploration and coverage algorithm based on reflex agents using a pheromone map as storage and communication medium. The algorithm proposed in this paper outperforms many of the popular reflex agent exploration algorithms in terms of exploration performance measured as the cumulative path length.\",\n \"Title: Deep Learning Earth Observation Classification Using ImageNet Pretrained Networks\\nYear: 2016\\n\\nAbstract: Deep learning methods such as convolutional neural networks (CNNs) can deliver highly accurate classification results when provided with large enough data sets and respective labels. However, using CNNs along with limited labeled data can be problematic, as this leads to extensive overfitting. In this letter, we propose a novel method by considering a pretrained CNN designed for tackling an entirely different classification problem, namely, the ImageNet challenge, and exploit it to extract an initial set of representations. The derived representations are then transferred into a supervised CNN classifier, along with their class labels, effectively training the system. Through this two-stage framework, we successfully deal with the limited-data problem in an end-to-end processing scheme. Comparative results over the UC Merced Land Use benchmark prove that our method significantly outperforms the previously best stated results, improving the overall accuracy from 83.1% up to 92.4%. Apart from statistical improvements, our method introduces a novel feature fusion algorithm that effectively tackles the large data dimensionality by using a simple and computationally efficient approach.\",\n \"Title: The Semantic Gap: An Exploration of User and Computer Perspectives in Earth Observation Images.\\nYear: 2015\\n\\nAbstract: Research on the semantic gap has considered differences between user and computer image interpretations and proposed methods to bridge it. These methods have been verified by comparing results to reference data or by measuring the degree of user acceptance. Although these methods result in a narrower semantic gap between computers and users, the resulting model for a specific user and search goal ...\",\n \"Title: Discriminative Nonnegative Matrix Factorization for dimensionality reduction.\\nYear: 2016\\n\\nAbstract: Nonnegative Matrix Factorization (NMF) has been widely used for different purposes such as feature learning, dictionary leaning and dimensionality reduction in data mining and computer vision. In t...\",\n \"Title: Immersive visualization of visual data using nonnegative matrix factorization.\\nYear: 2016\\n\\nAbstract: Over the last two decades, dimension reduction for visualization has gained a high amount of attention in visual data mining where the data is represented by high-dimensional features. Basically, this approach leads to an unbalanced and occluded distribution of visual data in display space, giving rise to difficulties in browsing the data. In this paper we propose an approach for the visualization of image collections in such a way as (1) images are not occluded by each other, and the provided space is used as much as possible; (2) the similar images are positioned close together; (3) an overview of data is feasible. To fulfill these requirements, we propose to use regularized Nonnegative Matrix Factorization (NMF) controlled by parameters to reduce the dimensionality of data. Experiments performed on optical and radar images confirm the flexibility of proposed method in visualizing large-scale visual data. Finally, an immersive 3D virtual environment is suggested, to visualize the images, to allow the user to navigate and explore the data.\",\n \"Title: Comparison of Kullback-Leibler divergence approximation methods between Gaussian mixture models for satellite image retrieval\\nYear: 2015\\n\\nAbstract: In many applications, such as image retrieval and change detection, we need to assess the similarity of two statistical models. As a distance measure between two probability density functions, Kullback-Leibler divergence is widely used for comparing two statistical models. Unfortunately, for some models such as Gaussian Mixture Model (GMM), Kullback-Leibler divergence has no analytically tractable formula. We have to resort to approximation methods. In this paper, we compare seven methods, namely Monte Carlo method, matched bond approximation, product of Gaussian, variation-al method, unscented transformation, Gaussian approximation, and min-Gaussian approximation, for approximating the Kullback-Leibler divergence between two Gaussian mixture models for satellite image retrieval. Two image retrieval experiments based on two publicly available datasets have been performed. The comparison is carried out in terms of both retrieval performance and computational time.\",\n \"Title: Polarimetric Sar Data Feature Selection Using Measures Of Mutual Information\\nYear: 2015\\n\\nAbstract: Several algorithms for polarimetric synthetic aperture radar (PolSAR) data indexing and classification were proposed in the state of the art literature. In particular, one of them computes powerful, compact feature descriptors composed of the first three logarithmic cumulants of the BiQuaternion Fractional Fourier Transform (BiQFrFT) coefficients of PolSAR patches. Since the BiQFrFT of each patch is computed at three different angles, the algorithm's result consists in nine complex-valued features (18 real-valued features) for single polarization images and in nine biquaternion-valued features (72 real-valued features) for fully polarimetric images. In this paper feature selection based on mutual information (MI) is employed to optimally select a subset of features, in order to improve the indexing performances and to minimize the classification error. The improved results are shown on two polarimetric images: a L-band PALSAR image over Danube's Delta, Romania and a C-band RadarSAT2 image over Braila, Romania.\",\n \"Title: A Comparison Of Bag-Of-Words Method And Normalized Compression Distance For Satellite Image Retrieval\\nYear: 2015\\n\\nAbstract: Recently, two improved methods have shown their advantages in browsing Earth Observation (EO) dataset. The first method is the Bag-of-Words (BoW) feature extraction method and the second is the Normalized Compression Distance (NCD) for assessing image similarity. However, they have not been compared so far for satellite image retrieval, which motivates this paper. Two retrieval experiments have been performed on a freely available optical image dataset and a SAR image dataset. Through these two experiments, we conclude that the BoW method performs generally better than NCD. Although it is a parameter-free solution for data mining, NCD only performs well for images with repetitive patterns like some homogeneous classes. In contrast, BoW method performs much far beyond that of NCD. In addition, NCD is computationally very expensive, which makes it infeasible to be applied in real applications. In contrast, BoW method is more realistic in practical applications in terms of both accuracy and computation.\",\n \"Title: Interactive Feature Learning From Sar Image Patches\\nYear: 2015\\n\\nAbstract: Feature learning algorithms aim to provide a compact and discriminative representation of complex datasets in order to increase the speed and accuracy of clustering or classification. In this paper, we propose a novel interactive feature learning approach which is mainly based on 3D interactive data visualization and Non-negative Matrix Factorization (NMF). Here, the data is visualized in a 3D interface to support human-data interaction. The user interactions are exploited in an NMF framework to learn a compact representation of the data. The conducted experiments on Synthetic Aperture Radar (SAR) images confirm the efficiency of the proposed approach.\",\n \"Title: Gpu-Based Kernelized Locality-Sensitive Hashing For Satellite Image Retrieval\\nYear: 2015\\n\\nAbstract: As the data acquisition capabilities of Earth observation (EO) satellites have been improved substantially in the past few years, large amount of high-resolution satellite images are downlinked continuously to ground stations. Such amount of data increases rapidly beyond the users' capability to access the images' content in reasonable time. Hence, automatic and fast interpretation of a large data volume is a computationally intensive task. Recently, approximate nearest neighbour search has been used for content-based image retrieval in sub-linear time. Kernelized locality sensitive hashing (KLSH) is a well-known approximate method, which has recently shown promising results for fast remote sensing image retrieval. This paper proposes a novel parallelization of KLSH using Graphical Processing Units (GPU), in order to perform fast parallel image retrieval. The proposed method was tested on high-dimensional feature vectors from two satellite-based image datasets, where an average speedup of 20 times was achieved.\",\n \"Title: Semantic Interpretation Of Multi-Level Change Detection In Multi-Temporal Satellite Images\\nYear: 2015\\n\\nAbstract: Satellite image time series are a valuable resource for enhancing land exploitation by respecting the natural cycles, analyzing urban expansion and its positive and negative effects, limiting the unhealthy rhythm of deforestation, understanding natural hazards and so on. In this context, understanding only the changes in multitemporal images is not sufficient. This paper aims to correlate multi-level change detection techniques with image semantic segmentation methods in order to build an hierarchy of changes for each semantic class. In this way, we are able to provide statistics regarding the levels of change suffered by a certain area. The methods are demonstrated with examples involving bi-temporal Landsat images.\",\n \"Title: A Hierarchical Patch Clustering Method For High-Resolution Terrasar-X Images\\nYear: 2015\\n\\nAbstract: In this paper, we present a Gaussian test-based hierarchical clustering method for high-resolution TerraSAR-X images. The purpose is to obtain homogeneous clusters. k-means is used to split image features to create a hierarchical structure. As image feature vectors usually fall into high dimensional feature space, we test different distance metrics, in order to try to tackle the curse of dimensionality problem. With prepared datasets, we evaluate the clustering results by defining a homogeneity percentage. The results show that by using Gabor texture feature, the Gaussian test-based hierarchical patch clustering method is able to obtain homogeneous clusters. Meanwhile, fractional distance or Minkowski distance performs better than Euclidean or Manhatten distance.\",\n \"Title: Feature Space Dimensionality Reduction For The Optimization Of Visualization Methods\\nYear: 2015\\n\\nAbstract: Visual data mining methods are of great importance in exploratory data analysis having a high potential for mining large databases. As the data feature space is generally ndimensional, visual data mining relies on dimensionality reduction techniques. This is the case for image feature spaces which can be visualized by giving each data point a location in a three dimensional space. This paper aims to present a comparative study of several dimensionality reduction methods considering as input image feature spaces, in order to detemine an optimal visualization method to illustrate the separation of the classes. At the beginning, to check the performance of the envisaged method, an artificial dataset consisting of random vectors describing six, 20-dimensional Gaussian distributions with spaced means and low variances was generated. Further, two real images datasets are used to evaluate the contributions of dimensionality reduction algorithms related to data visualization. The analysis focuses on the PCA, LDA and t-SNE dimensionality reduction techniques. Our tests are performed on images for which the computed features include the color histogram and Weber descriptors.\",\n \"Title: Lucas Visual Browser: A Tool For Land Cover Visual Analytics\\nYear: 2015\\n\\nAbstract: In this paper we present the LUCAS Visual Browser system, a tool for land cover visual analytics. The system implements different web technologies in a multilayer server-client architecture in order to allow the user to visually analyse land cover heterogeneous information. The information manage is composed of EO multispectral and SAR products along with the multitemporal in situ LUCAS surveys. The fusion of these data provides a very useful information during the EO scene interpretation process. Furthermore, the system offers interactive tools for the detection of optimal datasets for EO multitemporal image change detection, providing at the same time ground truth points for both, human and machine analysis.\",\n \"Title: Information extraction from high resolution SAR data\\nYear: 2002\\n\\nAbstract: The article aims at presenting the most recent developments and trends in information extraction from Synthetic Aperture Radar SAR data.\",\n \"Title: Dictionary-Based Compact Data Representation for Very High Resolution Earth Observation Image Classification\\nYear: 2015\\n\\nAbstract: In the context of fast growing data archives, with continuous changes in volume and diversity, information mining has proven to be a difficult, yet highly recommended task. The first and perhaps the most important part of the process is data representation for efficient and reliable image classification. This paper is presenting a new approach for describing the content of Earth Observation Very High Resolution images, by comparison with traditional representations based on specific features. The benefit of data compression is exploited in order to express the scene content in terms of dictionaries. The image is represented as a distribution of recurrent patterns, removing redundant information, but keeping all the explicit features, like spectral, texture and context. Further, a data domain analysis is performed using Support Vector Machine aiming to compare the influence of data representation to semantic scene annotation. WorldView2 data and a reference map are used for algorithm evaluation.\",\n \"Title: Feature Extraction for Patch-Based Classification of Multispectral Earth Observation Images.\\nYear: 2016\\n\\nAbstract: Recently, various patch-based approaches have emerged for high and very high resolution multispectral image classification and indexing. This comes as a consequence of the most important particularity of multispectral data: objects are represented using several spectral bands that equally influence the classification process. In this letter, by using a patch-based approach, we are aiming at extrac...\",\n \"Title: Dimensionality Reduction for Visual Data Mining of Earth Observation Archives.\\nYear: 2016\\n\\nAbstract: Modern knowledge discovery systems, empowered by visual data exploration techniques, enable the user to discover and understand the data content. Considering patch-level processing, the visual exploration of Earth Observation archives aims to identify groups of items sharing similar semantic content. Each patch is further represented by certain descriptors, i.e., spectral signatures or Weber local...\",\n \"Title: The Earth-Observation Epitome: A New Interactive Value-Added Product.\\nYear: 2016\\n\\nAbstract: Typical applications based on remote sensing imagery start with a visual inspection and selection of scenes and regions of interest to be purchased. In this letter, we propose a so-called Earth observation (EO) epitome as a novel value-added interactive product. Basically, it comprises several image tiles, primitive features, metadata entries, a quick look of the image tiles, and semantic annotati...\",\n \"Title: Simplified Bistatic Sar Imaging With A Fixed Receiver And Terrasar-X As Transmitter Of Opportunity - First Results\\nYear: 2016\\n\\nAbstract: This paper presents the first SAR imaging results obtained with a fixed ground-based system used in bistatic configuration having the TerraSAR-X satellite as transmitter of opportunity. The system's characteristics and signal processing flow are presented relative to the state of the art. Compared to previous works on bistatic SAR imaging where a significant amount of processing is dedicated to time/frequency synchronization between the satellite transmitter and ground receiver, we show that a bistatic SAR image with only meter-range geographic offset can be obtained using the state vectors from a monostatic image and minimal synchronization efforts consisting in acquisitions triggered by an amplitude threshold and stabilization of the local oscillator with a GPS-disciplined reference. We present the first bistatic image in ground geometry obtained over an area of Bucharest and compare it with a monostatic image refocused on the same grid.\",\n \"Title: A Convolutional Deep Belief Network For Polarimetric Sar Data Feature Extraction\\nYear: 2016\\n\\nAbstract: This paper proposes a custom convolutional deep belief network for polarimetric synthetic aperture radar (PolSAR) data feature extraction. The proposed architecture stands out through the interesting features it shows, starting with the fact that it is adapted to fully polarimetric SAR data. Then, the multilayer approach allows the stepwise discovery of higher-level features. The convolutional approach allows the discovery of local, spatially invariant features and makes the architecture scalable to fully sized PolSAR images. The network is trained in an unsupervised manner, without using labeled data and then it succeeds to extract powerful features from PolSAR patches. This fact is demonstrated by applying supervised and unsupervised classification algorithms on features extracted from patches of a fully polarimetric multi-look F-SAR image over Kaufbeuren airfield, Germany.\",\n \"Title: Compound And Configurable Framework For Exploratory Earth Observation Data Analysis\\nYear: 2016\\n\\nAbstract: The lack of a comprehensive solution for image information mining has often brought confusion and misunderstanding when Earth Observation data based application scenarios were addressed. Considering the variety of dedicated sensors available nowadays, the particularities of the recorded data raises serious issues when explored. Most of the proposed methodologies for data analysis integrate algorithms able to cope with single cases. In order to overcome this limitation, the present paper introduce a compound, configurable framework containing two processing levels, for feature extraction and image classification, that allows different settings depending on the application being handled. The design was proposed such that it facilitates the integration of several methods and algorithm for each level, including a module to serve for validation when reference data is available. The approach is not complete without the interaction with the user, therefore, a human-machine communication strategy was also developed. The validation was performed through a prototype system meeting all the criteria of the defined framework.\",\n \"Title: Applications Of Sar Tomography On Persistent Scatterers Detection, Based On Beam-Forming Filtering\\nYear: 2016\\n\\nAbstract: The main advantage of Synthetic Aperture Radar (SAR) Tomography over classical interferometry consists in capacity to detect the presence of multiple scatterers within the same resolution cell. In this paper we present an algorithm for detection of Persistent Scatterers (PS) based on the variation of reflectivity function in elevation direction reconstructed with Beam-Forming technique. Then, we extract the contribution of dominant scatterers from each resolution cell and reapply the detection algorithm to identify thr presence of secondary PSs in scene's pixels. The methods were implemented on a dataset of high-resolution Single Look Complex (SLC) images acquired by TerraSAR-X sensor, on which the residual component of interferometric phase was estimated and compensated locally.\",\n \"Title: A Bistatic Sar Simulator For Ground-Based Fixed-Receiver Geometry\\nYear: 2016\\n\\nAbstract: In this paper we present the early development of a SAR simulator for ground-based fixed-receiver bistatic geometry. Firstly, we describe the assumptions the simulator is based on, then a short presentation of the mathematical model that lays behind the simulator is given. Furthermore, we give some details about its implementation and finally, we present some results.\",\n \"Title: A Dimensionality Reduction Approach For The Visualization Of The Cluster Space: A Trustworthiness Evaluation\\nYear: 2016\\n\\nAbstract: The data mining systems solve the problem of handling Earth Observation archives counting on a feature vectors based description of the data. Increasing the dimensionality of the feature vectors would offer an effective perspective of the dataset's content. The modern systems provide visual exploration of data projecting their high-dimensional feature space in a 3-D space. The dimensionality reduction methods represent the main way to achieve such representation. Several dimensionality reduction methods have been proposed to identify the mapping, bot not all of them retain the same dataset properties. In order to compare their performance, the development of formal measures like \\\"Trustworthiness\\\" or the measures based on Co-ranking matrix was mandatory. These measures objectively evaluate the similarity between the structure detected in the original and the reduced space. In this paper we evaluate six dimensionality reduction methods using \\\"Trustworthiness\\\" and \\\"Continuity\\\" measures. In this regard three datasets have been used: an artificial one and two remote sensing datasets. Each of them have been described by a high-dimensional feature space.\",\n \"Title: Visual Data Mining For Feature Space Exploration Using In-Situ Data\\nYear: 2016\\n\\nAbstract: In this paper, we present the visualization of image databases based on their primitive features. Our approach is to have a visual navigation tool for allowing the exploration and exploitation of large image archives. The tool is able to project the content of a given image database based on the primitive feature space and to provide interaction between the final user and the huge amount of data. Land Use/Land Cover area frame statistical Survey in-situ data are used as test dataset. Bag-of-Words and Weber Local Descriptors are used as primitive features.\",\n \"Title: Discovery of Semantic Relationships in PolSAR Images Using Latent Dirichlet Allocation.\\nYear: 2017\\n\\nAbstract: We propose a multilevel semantics discovery approach for bridging the semantic gap when mining high-resolution polarimetric synthetic aperture radar (PolSAR) remote sensing images. First, an Entropy/Anisotropy/Alpha-Wishart classifier is employed to discover low-level semantics as classes representing the physical scattering properties of targets (e.g., low-entropy/surface scattering/high anisotro...\",\n \"Title: Exploiting Deep Features for Remote Sensing Image Retrieval: A Systematic Investigation.\\nYear: 2017\\n\\nAbstract: Remote sensing (RS) image retrieval based on visual content is of great significance for geological information mining. Over the past two decades, a large amount of research on this task has been carried out, which mainly focuses on the following three core issues of image retrieval: visual feature, similarity metric and relevance feedback. Along with the advance of these issues, the technology of RS image retrieval has been developed comparatively mature. However, due to the complexity and multiformity of high-resolution remote sensing (HRRS) images, there is still room for improvement in the current methods on HRRS data retrieval. In this paper, we analyze the three key aspects of retrieval and provide a comprehensive review on content-based RS image retrieval methods. Furthermore, for the goal to advance the state-of-the-art in HRRS image retrieval, we focus on the visual feature aspect and delve how to use powerful deep representations in this task. We conduct systematic investigation on evaluating factors that may affect the performance of deep features. By optimizing each factor, we acquire remarkable retrieval results on publicly available HRRS datasets. Finally, we explain the experimental phenomenon in detail and draw instructive conclusions according to our analysis. Our work can serve as a guiding role for the research of content-based RS image retrieval.\",\n \"Title: Evaluation of Dimensionality Reduction Methods for Remote Sensing Images Using Classification and 3D Visualization.\\nYear: 2017\\n\\nAbstract: Visual exploration is a natural way to understand the content of a data archive. If the data are multidimensional, the dimensionality reduction is an appropriate preprocessing step before the data visualization. In literature, two types of approaches are devoted to dimensionality reduction: feature selection and feature extraction algorithms. Both techniques intend to project a high dimensional space into a new one, with reduced dimensionality, preserving data inherent information. This paper aims to identify the similarity degree between low and high dimensional representations of a data archive using the optimal number of semantic classes as a criterion. This number is estimated based on the rate distortion theory being computed both before and after dimensionality reduction. The projection of the low dimensional space was obtained using one feature selection and six feature extraction methods.\",\n \"Title: Artificial Generation of Big Data for Improving Image Classification: A Generative Adversarial Network Approach on SAR Data.\\nYear: 2017\\n\\nAbstract: Very High Spatial Resolution (VHSR) large-scale SAR databases are still an unresolved issue in the RemoteSensing field. In this work, we propose such a dataset anduse it to explore patch-based classification in urban and periurban areas, considering 7 distinct semantic classes. In this context, we investigate the accuracy of large CNN classification models and pre-trained networks for SAR imaging systems.Furthermore, we propose a Generative Adversarial Network(GAN) for SAR image generation and test, whether thesynthetic data can actually improve classification accuracy.\",\n \"Title: Image representation alternatives for the analysis of satellite image time series\\nYear: 2017\\n\\nAbstract: Current satellite images and image time series provide us with detailed information about the state of our planet as well as about our technical infrastructure and human activities. These images allow us to learn more about local, regional, and global phenomena and events, including - if interpreted properly - their causes and effects. In particular, image time series provide specific information about the dynamics of many processes implicitly contained in our images that need to be unearthed and investigated in detail. A traditional approach towards this aim is to start with pixel-level or patch-level data analysis for pixel-based image analysis, followed, if necessary, by subsequent feature extraction, clustering, classification and semantic labelling in order to generate various types of change maps on different representation levels. The classification step can be supported by interactive human intervention, or by automated machine learning strategies to identify higher level objects and their spatial and temporal relationships. The detected relationships can then be formulated as parameterized rule sets that create higher-level descriptor sets of the content of the selected images, and of additional external data such as thematic maps or typical dynamics descriptions. As an innovative extension of this traditional concept, we propose a highly automated approach for application-adapted image content exploration and knowledge extraction. The reason for this strategy is the additional amount and the precision of semantic relationships and details that we can assign to an image time series once we know the final application field and how to embed and access image content within knowledge graphs.\",\n \"Title: Temporal analysis of SAR imagery for permanent and evolving Earth land cover behavior assessment\\nYear: 2017\\n\\nAbstract: In the era of constantly increasing Earth Observation (EO) data collections, information extraction and data analysis should be enhanced with a multi-temporal component enabled by the temporal resolution of satellite missions and create handy, yet powerful tools for those applications involving monitoring of land cover. The image time series, as results of the satellite revisiting period, gives you insights not only on a certain area, but also on its representation at different moments of time. In order to limit the issues that might arise due to irregular time sampling of multispectral data, the authors propose a Synthetic Aperture Radar (SAR) image time series for analysis. To this point, the main goal is to mine the satellite image time series (SITS) for understanding the temporal behaviour of an area in terms of evolution and persistency. The paper introduces an analytical approach, combining coherent and no coherent analysis of SAR SITS content. We propose the Latent Dirichlet Allocation model to extract categories of evolution from the SAR SITS and techniques which study statistical and coherent proprieties of the targets to identify the structures with stable electromagnetic characteristics over time, named Persistent Scatterers (PS). The obtained results indicate an evolutionary character hidden inside the persistent class. The results obtained on 30 ERS images encourages further analysis on Sentinel 1 data.\",\n \"Title: Land cover change detection in Satellite Image Time Series using an active learning method\\nYear: 2017\\n\\nAbstract: Some of the first Earth Observation (EO) missions date back to the 1970s. Over the time, large datasets of Satellite Image Time Series (SITS) have been used to identify and monitor land cover evolutions. The processing complexity increases proportionally to the time span of the Earth Observation (EO) series. Because of the SITS dataset complexity and variety of contained evolution patterns, most unsupervised classification methods fail to detect and isolate the user's evolution pattern of interest. This is usually caused by the discrepancy between automatically extracted low-level features and high-level meaning assigned by the user who searches for a specific evolution. In an effort to find a solution for this difficult task, this paper presents a SVM based active learning method for the extraction of specific evolution classes from SITS. Several experiments were conducted on a dataset composed of Landsat 4 TM (Thematic Mapper) and Landsat 5 TM acquisitions over Bucharest, Romania, in the time interval of 1984-1993.\",\n \"Title: Land-cover evolution class analysis in Image Time Series of Landsat and Sentinel-2 based on Latent Dirichlet Allocation\\nYear: 2017\\n\\nAbstract: Satellite Image Time Series (SITS) are widely used in monitoring the Earth's changes for various applications such as land-cover evolution analysis. In this paper, we propose an approach based on Latent Dirichlet Allocation (LDA) which considers spatial and spectral information to measure the land-cover changes in multispectral SITS. For our experiments, we focus on the vegetation dynamics of the Doñana National Park (in southwestern Spain) using a Landsat and a Sentinel-2 SITS dataset. The proposed approach represents each image by Normalized Difference Vegetation Index (NDVI) and tiles it into smaller patches. The patches are then modeled as Bag-of-Words (BoW) and LDA is applied to them in order to discover the latent structure of the image. The divergence between the latent structures of any two consecutive images is then considered as the measure of change. Results show that the changes measured by the proposed approach can represent the vegetation dynamics of the region of interest. Moreover, comparing the results obtained from the two datasets demonstrates that using high-level information allows the proposed approach to measure the changes independent of the sensor. This will support long-term monitoring through combining various available data.\",\n \"Title: Multisensor Earth Observation Image Classification Based on a Multimodal Latent Dirichlet Allocation Model.\\nYear: 2018\\n\\nAbstract: Many previous researches have already shown the advantages of multisensor land-cover classification. Here, we propose an innovative land-cover classification approach based on learning a joint latent model of synthetic aperture radar (SAR) and multispectral satellite images using multimodal latent Dirichlet allocation (mmLDA), a probabilistic generative model. It has already been successfully appl...\",\n \"Title: Understanding Heterogeneous EO Datasets: A Framework for Semantic Representations.\\nYear: 2018\\n\\nAbstract: Earth observation (EO) has become a valuable source of comprehensive, reliable, and persistent information for a wide number of applications. However, dealing with the complexity of land cover is sometimes difficult, as the variety of EO sensors reflects in the multitude of details recorded in several types of image data. Their properties dictate the category and nature of the perceptible land structures. The data heterogeneity hampers proper understanding, preventing the definition of universal procedures for content exploitation. The main shortcomings are due to the different human and sensor perception on objects, as well as to the lack of coincidence between visual elements and similarities obtained by computation. In order to bridge these sensory and semantic gaps, the paper presents a compound framework for EO image information extraction. The proposed approach acts like a common ground between the user's understanding, who is visually shortsighted to the visible domain, and the machines numerical interpretation of a much wider information. A hierarchical data representation is considered. At first, basic elements are automatically computed. Then, users can enforce their judgement on the data processing results until semantic structures are revealed. This procedure completes a user-machine knowledge transfer. The interaction is formalized as a dialogue, where communication is determined by a set of parameters guiding the computational process at each level of representation. The purpose is to maintain the data-driven observable connected to the level of semantics and to human awareness. The proposed concept offers flexibility and interoperability to users, allowing them to generate those results that best fit their application scenario. The experiments performed on different satellite images demonstrate the ability to increase the performances in case of semantic annotation by adjusting a set of parameters to the particularities of the analyzed data.\",\n \"Title: On the Use of Normalized Compression Distances for Image Similarity Detection.\\nYear: 2018\\n\\nAbstract: This paper investigates the usefulness of the normalized compression distance (NCD) for image similarity detection. Instead of the direct NCD between images, the paper considers the correlation between NCD based feature vectors extracted for each image. The vectors are derived by computing the NCD between the original image and sequences of translated (rotated) versions. Feature vectors for simple transforms (circular translations on horizontal, vertical, diagonal directions and rotations around image center) and several standard compressors are generated and tested in a very simple experiment of similarity detection between the original image and two filtered versions (median and moving average). The promising vector configurations (geometric transform, lossless compressor) are further tested for similarity detection on the 24 images of the Kodak set subject to some common image processing. While the direct computation of NCD fails to detect image similarity even in the case of simple median and moving average filtering in 3 x 3 windows, for certain transforms and compressors, the proposed approach appears to provide robustness at similarity detection against smoothing, lossy compression, contrast enhancement, noise addition and some robustness against geometrical transforms (scaling, cropping and rotation).\",\n \"Title: Dialectical GAN for SAR Image Translation: From Sentinel-1 to TerraSAR-X.\\nYear: 2018\\n\\nAbstract: With more and more SAR applications, the demand for enhanced high-quality SAR images has increased considerably. However, high-quality SAR images entail high costs, due to the limitations of current SAR devices and their image processing resources. To improve the quality of SAR images and to reduce the costs of their generation, we propose a Dialectical Generative Adversarial Network (Dialectical GAN) to generate high-quality SAR images. This method is based on the analysis of hierarchical SAR information and the dialectical structure of GAN frameworks. As a demonstration, a typical example will be shown, where a low-resolution SAR image (e.g., a Sentinel-1 image) with large ground coverage is translated into a high-resolution SAR image (e.g., a TerraSAR-X image). A new algorithm is proposed based on a network framework by combining conditional WGAN-GP (Wasserstein Generative Adversarial NetworkGradient Penalty) loss functions and Spatial Gram matrices under the rule of dialectics. Experimental results show that the SAR image translation works very well when we compare the results of our proposed method with the selected traditional methods.\",\n \"Title: Moving Ship Velocity Estimation Using TanDEM-X Data Based on Subaperture Decomposition.\\nYear: 2018\\n\\nAbstract: In this letter, a velocity estimation method for moving ships in synthetic aperture radar (SAR) images is proposed based on a subaperture decomposition technique. In contrast to traditional methods, our method needs only a few SAR imaging parameters besides the SAR image itself. The behavior of moving ships in subaperture images is theoretically analyzed, and the ship motion parameters in azimuth ...\",\n \"Title: Joint SAR Image Time Series and PSInSAR Data Analytics: An LDA Based Approach.\\nYear: 2018\\n\\nAbstract: Due to the constant increase in Earth Observation (EO) data collections, the monitoring of land cover is facilitated by the temporal diversity of the satellite images datasets. Due to the capacity of Synthetic Aperture Radar (SAR) sensors to operate independently of sunlight and weather conditions, SAR image time series offer the possibility to form a dataset with almost regular temporal sampling. This paper aims at mining the SAR image time series for an analysis of target's behavior from the perspective of both temporal evolution and coherence. The authors present a two-level analytical approach envisaging the assessment of global (related to perceivable structures on the ground) and local (related to changes occurred within a perceivable structure on the ground) evolution inside the scene. The Latent Dirichlet Allocation (LDA) model is implemented to identify the categories of evolution present in the analyzed scene, while the statistical and coherent proprieties of the dataset's images are exploited in order to identify the structures with stable electromagnetic response, the so-called Persistent Scatterers (PS). A comparative study of the two algorithms' classification results is conducted on ERS and Sentinel-1 data. At global scale, the results fit human perception, as most of the points which can be exploited for Persistent Scatterers Interferometry (PS-InSAR) are classified within the same class, referring to stable structures. At local scale, the LDA classification demands for an extended number of classes (defined through a perplexity-based analysis), enabling further differentiation inside the evolutional character of those stable structures. The comparison against the map of detected PS reveals which classes present higher temporal correlation, determining a stable evolutionary character, opening new perspectives for validation of both PS detection and SITS analysis algorithms.\",\n \"Title: Bayesian estimation of generalized Gamma mixture model based on variational EM algorithm.\\nYear: 2019\\n\\nAbstract: •We propose a VEM algorithm for the inference of the generalized Gamma mixture model (GMM) with all the closed-form update equations.•The help function is used to approximate the lower bound of the variational objective function in GMM.•With the proposed VEM algorithm, the effective number of components as well as all the other underlying parameters in GMM can be estimated automatically and simultaneously.•The results reveal that the proposed inference method is more efficient than the standard EM algorithm.•The experimental results show that the GMM is more appropriate for the asymmetric and heavy-tailed data than Gaussian mixture model.\",\n \"Title: ICPR2018 Contest on Object Detection in Aerial Images (ODAI-18)\\nYear: 2018\\n\\nAbstract: Object detection in aerial images plays a significant role in intelligent interpretation of aerial images. Hence many effective methods, especially the new-generation data-driven methods, have been developed for this task. Here, we hold the ODAI, a new contest that focused on object detection in aerial images, based on a new large-scale aerial image dataset called DOTA [1]. This contest contains over 3000 large-size images ( 4k×4k pixels), which cover 211,581 instances divided into 15 categories. Each instance is labeled by an arbitrary (8 d.o.f.) quadrilateral. Besides, we propose two tasks for this contest, named object detection with the horizontal bounding box (OD-HBB) and object detection with the oriented bounding box (OD-OBB). The contest was opened on February 7, 2018, and ended on April 30, 2018. A website is open to the public, which provides links to download data and evaluation server. We have totally received 60 registrations. There are 8 teams that have successfully submitted results on the OD-HBB task with the top mAP as 0.719, and 9 teams that have successfully submitted results on the OD-OBB task with the top mAP as 0.705. Through the contest, we hope to draw extensive attention from a wide range of communities and call for more future research and efforts for the task of object detection in aerial images.\",\n \"Title: Pulse Radar with Field-Programmable Gate Array Range Compression for Real Time Displacement and Vibration Monitoring.\\nYear: 2019\\n\\nAbstract: This paper aims to present the basic functionality of a radar platform for real time monitoring of displacement and vibration. The real time capabilities make the radar platform useful when live monitoring of targets is required. The system is based on the RF analog front-end of a USRP, and the range compression (time-domain cross-correlation) is implemented on the FPGA included in the USRP. Further processing is performed on the host computer to plot real time range profiles, displacements, vibration frequencies spectra and spectrograms (waterfall plots) for long term monitoring. The system is currently in experimental form and the present paper aims to prove its functionality. The precision of this system is estimated (using the 3 sigma approximation) at 0.6 mm for displacement measurements and 1.8 mm for vibration amplitude measurements.\",\n \"Title: Multilayer Architecture for Heterogeneous Geospatial Data Analytics: Querying and Understanding EO Archives.\\nYear: 2017\\n\\nAbstract: The constantly growing process of the Earth Observation (EO) data and their heterogeneity require new systems and tools for effectively querying and understanding the available data archives. In this paper, we present a tool for heterogeneous geospatial data analytics. The system implements different web technologies in a multilayer server-client architecture, allowing the user to visually analyze...\",\n \"Title: Mining Multitemporal In Situ Heterogeneous Monitoring Information for the Assurance of Recorded Land Cover Changes.\\nYear: 2017\\n\\nAbstract: We present a data mining methodology to filter and validate land cover change detections obtained from multitemporal in situ surveys. As in situ data we use the measurements from the European land use and coverage area frame survey (LUCAS), which provides images with standardized metadata about land cover and land use within the whole territory of the European Union. Multitemporal LUCAS surveys pr...\",\n \"Title: Earth Observation Image Semantic Bias: A Collaborative User Annotation Approach.\\nYear: 2017\\n\\nAbstract: Correctly annotated image datasets are important for developing and validating image mining methods. However, there is some doubt regarding the generalizability of the models trained and validated on available datasets. This is due to dataset biases, which occur when the same semantic label is used in different ways across datasets, and/or when identical object categories are labeled differently a...\",\n \"Title: Quantitative Evaluation of the Feature Space Transformation Methods Used for Applications of Visual Semantic Clustering of EO Images.\\nYear: 2017\\n\\nAbstract: Data visualization guides the process of indexing and retrieval, strengthening the link between low-level image features and high-level human understanding of image content. In this regard, we have described the semantic content of a multidimensional dataset using its descriptors to derive high-dimensional feature spaces. The dimensionality of these spaces is further reduced to three in order to p...\",\n \"Title: Super-Resolution Multi-Look Detection in SAR Tomography.\\nYear: 2018\\n\\nAbstract: Synthetic Aperture Radar (SAR) Tomography (TomoSAR) allows extending the 2-D focusing capabilities of SAR to the elevation direction, orthogonal to the azimuth and range. The multi-dimensional extension (along the time) also enables the monitoring of possible scatterer displacements. A key aspect of TomoSAR is the identification, in the presence of noise, of multiple persistent scatterers interfering within the same 2-D (azimuth range plane) pixel. To this aim, the use of multi-look has been shown to provide tangible improvements in the detection of single and double interfering persistent scatterers at the expense of a minor spatial resolution loss. Depending on the system acquisition characteristics, this operation may require also the detection of multiple scatterers interfering at distances lower than the Rayleigh resolution (super-resolution). In this work we further investigated the use of multi-look in TomoSAR for the detection of multiple scatterers located also below the Rayleigh resolution. A solution relying on the Capon filtering was first analyzed, due to its improved capabilities in the separation of the responses of multiple scatterers and sidelobe suppression. Moreover, in the framework of the Generalized Likelihood Ratio Test (GLRT), the single-look support based detection strategy recently proposed in the literature was extended to the multi-look case. Experimental results of tests carried out on two datasets acquired by TerraSAR-X and COSMO-SkyMED sensors are provided to show the performances of the proposed solution as well as the effects of the baseline span of the dataset for the detection capabilities of interfering scatterers.\",\n \"Title: Ranking evolution maps for Satellite Image Time Series exploration: application to crustal deformation and environmental monitoring\\nYear: 2019\\n\\nAbstract: Satellite Image Time Series (SITS) are large datasets containing spatiotemporal information about the surface of the Earth. In order to exploit the potential of such series, SITS analysis techniques have been designed for various applications such as earthquake monitoring, urban expansion assessment or glacier dynamic analysis. In this paper, we present an unsupervised technique for browsing SITS in preliminary explorations, before deciding whether to start deeper and more time consuming analyses. Such methods are lacking in today’s analyst toolbox, especially when it comes to stimulating the reuse of the ever growing list of available SITS. The method presented in this paper builds a summary of a SITS in the form of a set of maps depicting spatiotemporal phenomena. These maps are selected using an entropy-based ranking and a swap randomization technique. The approach is general and can handle either optical or radar SITS. As illustrated on both kinds of SITS, meaningful summaries capturing crustal deformation and environmental phenomena are produced. They can be computed on demand or precomputed once and stored together with the SITS for further usage.\",\n \"Title: Semi-supervised Hierarchical Clustering for Semantic SAR Image Annotation.\\nYear: 2016\\n\\nAbstract: In this paper, we propose a semi-automated hierarchical clustering and classification framework for synthetic aperture radar (SAR) image annotation. Our implementation of the framework allows the classification and annotation of image data ranging from scenes up to large satellite data archives. Our framework comprises three stages: 1) each image is cut into patches and each patch is transformed i...\",\n \"Title: Application and Evaluation of a Hierarchical Patch Clustering Method for Remote Sensing Images.\\nYear: 2016\\n\\nAbstract: In this paper, we apply and evaluate a modified Gaussian-test-based hierarchical clustering method for high-resolution satellite images. The purpose is to obtain homogeneous clusters within each hierarchy level which later allow the classification and annotation of image data ranging from single scenes up to large satellite data archives. After cutting a given image into small patches and feature ...\",\n \"Title: Immersive Interactive SAR Image Representation Using Non-negative Matrix Factorization.\\nYear: 2016\\n\\nAbstract: Earth observation (EO) images clustering is a challenging problem in data mining, where each image is represented by a high-dimensional feature vector. However, the feature vectors might not be appropriate to express the semantic content of images, which eventually lead to poor results in clustering and classification. To tackle this problem, we propose an interactive approach to generate compact ...\",\n \"Title: SAR Image Land Cover Datasets for Classification Benchmarking of Temporal Changes.\\nYear: 2018\\n\\nAbstract: The increased availability of high-resolution synthetic aperture radar (SAR) satellite images has led to new civil applications of these data. Among them is the systematic classification of land cover types based on the patterns of settlements or agriculture recorded by SAR imagers, in particular the identification and quantification of temporal changes. A systematic (re)classification shall allow...\",\n \"Title: Improved image classification by proper patch size selection: TerraSAR-X vs. Sentinel-1A\\nYear: 2016\\n\\nAbstract: When we perform image content classification by appending semantic labels to regularly cut image patches, we have to be sure that the selected patch size is a good choice for the images at hand. In the following, we look at SAR (Synthetic Aperture Radar) satellite images, and analyse the impact of the selected patch size on the attainable classification accuracy. For test images with precisely known ground truth, one can determine the true precision / recall performance of the applied classification method. In our case, we interactively trained a classifier system via active learning, and compared the resulting classification accuracy for high and medium resolution SAR images of different space borne instruments taken over urban areas, characterized by a high diversity of target categories. At a first glance, it turns out that the selected patch size does have a significant impact leading to a varying number of identified categories with strangely related confidence levels. A fundamental understanding of the relationships between the number of detected categories and their associated confidence levels requires detailed knowledge about SAR imaging, target characteristics, pixel size effects, radiometric image quality, the availability of appropriate semantic labels, and the selected active learning environment together with its image classification tool.\",\n \"Title: Land Cover Semantic Annotation Derived from High-Resolution SAR Images.\\nYear: 2016\\n\\nAbstract: Users of remote sensing images analyzing land cover characteristics are very much interested in classification schemes that define a consistent set of target categories. Up to now, a number of established classification schemes are mainly being used by interpreters of medium-resolution optical satellite images focusing on large-scale land cover. In contrast, we concentrate in this publication on t...\",\n \"Title: Image disorder characterization based on rate distortion\\nYear: 2005\\n\\nAbstract: Rate distortion theory is one of the areas of information transmission theory with important applications in multimodal signal processing, as for example image processing, information bottleneck and steganalysis. This article present an image characterization method based on rate distortion analysis in the feature space. This space is coded using clustering as vector quantization (k-means). Since image information usually cannot be coded by single clusters, because there are image regions corresponding to groups of clusters, the rate and distortion are specifically defined. The rate distortion curve is analyzed, extracting specific features for implementing a database image classification system.\",\n \"Title: Classification With an Edge: Improving Semantic Image Segmentation with Boundary Detection.\\nYear: 2016\\n\\nAbstract: We present an end-to-end trainable deep convolutional neural network (DCNN) for semantic segmentation with built-in awareness of semantically meaningful boundaries. Semantic segmentation is a fundamental remote sensing task, and most state-of-the-art methods rely on DCNNs as their workhorse. A major reason for their success is that deep networks learn to accumulate contextual information over very large receptive fields. However, this success comes at a cost, since the associated loss of effective spatial resolution washes out high-frequency details and leads to blurry object boundaries. Here, we propose to counter this effect by combining semantic segmentation with semantically informed edge detection, thus making class boundaries explicit in the model. First, we construct a comparatively simple, memory-efficient model by adding boundary detection to the segnet encoder-decoder architecture. Second, we also include boundary detection in fcn-type models and set up a high-end classifier ensemble. We show that boundary detection significantly improves semantic segmentation with CNNs in an end-to-end training scheme. Our best model achieves >90% overall accuracy on the ISPRS Vaihingen benchmark.\",\n \"Title: New MPEG-7 Scalable Color Descriptor Based on Polar Coordinates for Multispectral Earth Observation Image Analysis.\\nYear: 2017\\n\\nAbstract: Continuously expanding high-resolution and very high resolution multispectral image collections, provided by remote sensing satellites, require specific methods and techniques for data analysis and understanding. Even though there are several patch-based approaches for image classification and indexing, none of them are integrated within a standard. Having the goal to develop an MPEG-7 compliant d...\",\n \"Title: CMIR-NET : A deep learning based model for cross-modal retrieval in remote sensing\\nYear: 2019\\n\\nAbstract: •Cross-modal data retrieval.•Novel framework for designing a unified representation for different modalities of data.•Improved performance using the unified representation for both cross and uni-modal retrieval.•Extensive experiments performed on the benchmark DSRSID and UC-Merced with audio samples datasets.\",\n \"Title: Unbiased Seamless SAR Image Change Detection Based on Normalized Compression Distance\\nYear: 2019\\n\\nAbstract: Land cover changes may have very different nature, e.g., vegetation development, soil erosion, variation of humidity, or damage of buildings, only to enumerate few cases. In addition, synthetic aperture radar (SAR) observations are a doppelganger of the scene, imaging the scene signature rather than the scene itself. To overcome these challenges, SAR change detection methods generally adapt to the particular situations. We present seamless methods based on normalized compression distance (NCD) estimation. NCD is a similarity metric applied directly to the data, thus with no biases induced by feature estimators or classifiers. Since the diversity of changes is huge and extremely hard to derive typical classes, we introduce paradigm based both on an unsupervised and a supervised method. The change detection procedure mainly consists in dividing image dataset in patches, computing a collection of similarities for pairs of tiles formed differently in each case, and usage of this collection in unsupervised and supervised forms to generate a change map. Both the threshold based histogram, unsupervised method, and the k-NN classifier algorithm, supervised method, have a distinct flow to obtain the change map. To use the NCD operator according to our proposed methods, a speckle resistance test is involved. The experimental results for the two methodologies are computed using two TerraSAR-X images over Sendai and surrounding areas, Japan.\",\n \"Title: Compact Ground-Based Interferometric Synthetic Aperture Radar: Short-range structural monitoring\\nYear: 2019\\n\\nAbstract: Recently, structural monitoring by radar remote sensing has become more necessary for both economic and security reasons. Infrastructure monitoring with no incorporated deformation sensors (e.g., old generation water dams for which regulations did not impose monitoring capabilities) is usually performed by regular in situ topographic surveys. However, these surveys cannot be performed very often, and alternative methods are desirable. A feasible nonintrusive way to monitor such a structure is with interferometric synthetic aperture radar (SAR) data that can be acquired with monostatic/bistatic sensors.\",\n \"Title: LUAI Challenge 2021 on Learning to Understand Aerial Images\\nYear: 2021\\n\\nAbstract: This report summarizes the results of Learning to Understand Aerial Images (LUAI) 2021 challenge held on ICCV’2021, which focuses on object detection and semantic segmentation in aerial images. Using DOTA-v2.0 [7] and GID-15 [35] datasets, this challenge proposes three tasks for oriented object detection, horizontal object detection, and semantic segmentation of common categories in aerial images....\",\n \"Title: Guided Deep Learning by Subaperture Decomposition: Ocean Patterns from SAR Imagery.\\nYear: 2022\\n\\nAbstract: Spaceborne synthetic aperture radar can provide meters scale images of the ocean surface roughness day or night in nearly all weather conditions. This makes it a unique asset for many geophysical applications. Sentinel 1 SAR wave mode vignettes have made possible to capture many important oceanic and atmospheric phenomena since 2014. However, considering the amount of data provided, expanding applications requires a strategy to automatically process and extract geophysical parameters. In this study, we propose to apply subaperture decomposition as a preprocessing stage for SAR deep learning models. Our data centring approach surpassed the baseline by 0.7, obtaining state of the art on the TenGeoPSARwv data set. In addition, we empirically showed that subaperture decomposition could bring additional information over the original vignette, by rising the number of clusters for an unsupervised segmentation method. Overall, we encourage the development of data centring approaches, showing that, data preprocessing could bring significant performance improvements over existing deep learning models.\",\n \"Title: Coreset of Hyperspectral Images on a Small Quantum Computer.\\nYear: 2022\\n\\nAbstract: Machine Learning (ML) techniques are employed to analyze and process big Remote Sensing (RS) data, and one well-known ML technique is a Support Vector Machine (SVM). An SVM is a quadratic programming (QP) problem, and a D-Wave quantum annealer (D-Wave QA) promises to solve this QP problem more efficiently than a conventional computer. However, the D-Wave QA cannot solve directly the SVM due to its very few input qubits. Hence, we use a coreset (\\\"core of a dataset\\\") of given EO data for training an SVM on this small D-Wave QA. The coreset is a small, representative weighted subset of an original dataset, and any training models generate competitive classes by using the coreset in contrast to by using its original dataset. We measured the closeness between an original dataset and its coreset by employing a Kullback-Leibler (KL) divergence measure. Moreover, we trained the SVM on the coreset data by using both a D-Wave QA and a conventional method. We conclude that the coreset characterizes the original dataset with very small KL divergence measure. In addition, we present our KL divergence results for demonstrating the closeness between our original data and its coreset. As practical RS data, we use Hyperspectral Image (HSI) of Indian Pine, USA.\",\n \"Title: Classification of Remote Sensing Images With Parameterized Quantum Gates\\nYear: 2022\\n\\nAbstract: This letter studies how to program and assess a parameterized quantum circuit (PQC) for classifying Earth observation (EO) satellite images. In this exploratory study, we assess a PQC for classifying a two-label EO image dataset and compare it with a classic deep learning classifier. We use the PQC with an input space of only 17 quantum bits (qubits) due to the current limitations of quantum technology. As a real-world image for EO, we selected the Eurosat dataset obtained from multispectral Sentinel-2 images as a training dataset and a Sentinel-2 image of Berlin, Germany, as a test image. However, the high dimensionality of our images is incompatible with the PQC input domain of 17 qubits. Hence, we had to reduce the dimensionality of the input images for this two-label case to a vector with 16 elements; the 17th qubit remains reserved for storing label information. We employed a very deep convolutional network with an autoencoder as a technique for the dimensionality reduction of the input image, and we mapped the dimensionally reduced image onto 16 qubits by means of parameter thresholding. Then, we used a PQC to classify the two-label content of the dimensionally reduced Eurosat image dataset. A PQC classifies the Eurosat images with high accuracy as a classic deep learning method (and with even better accuracy in some instances). From our experiment, we derived and enhanced deeper insight into programming future gate-based quantum computers for many practical problems in EO.\",\n \"Title: Attention-Driven Graph Convolution Network for Remote Sensing Image Retrieval\\nYear: 2022\\n\\nAbstract: Graph convolution networks (GCNs) are useful in remote sensing (RS) image retrieval. It is found to be effective because, in a graph representation, the relative geometrical interactions between different regions (or segments) are appropriately captured, along with their region-wise features in their region adjacency graphs. Also, the attention mechanism has often been applied to the nodes to highlight the essential features in each node. In this regard, a significant amount of high-frequency information is missed since each image segment is effectively summarized within a single node. To account for this and increase the learning capacity, we propose to attend over the edge/adjacency matrix to highlight the interactions among meaningful regions that contribute to supervised learning from images. We exploit this novel edge attention mechanism together with node attention to highlight essential image context by allowing more importance to the meaningful neighboring regions that highlight a relevant node. We implement the proposed context-attended GCN framework for image retrieval on the benchmarked UC-Merced and the PatternNet datasets. We observe a notable improvement in the results compared to the state of the art.\",\n \"Title: Near-Range Multipath Mitigation Methodology for Multistatic SAR Applications Using Matched-Adaptive Filters\\nYear: 2022\\n\\nAbstract: This article presents a methodology based on matched-adaptive filters, which is used to mitigate the effect of near-range multipath in multistatic SAR systems and transponders. It is challenging to physically construct a bistatic receiver such that the reference signal is not leaked into the received signal, either via coupling in the circuitry or via reflections off objects in the vicinity of the receiver. Due to its much larger amplitude, the reference signal can easily mask near-range targets with its sidelobes. A similar signal degradation is observed in active transponders that are used for calibrating radar systems, when objects exist in their vicinity. In this article, we address these two issues: the coupling between the reference channel and the imaging channel, and the parasitic echoes present in the transponder response. A novel methodology is proposed that is capable of time-domain filtering the undesired components in real time. The novelty consists in combining matched and adaptive filters as a means of boosting performance and resolution estimation, resulting in an extremely accurate multipath elimination method. The proposed methodology is experimentally evaluated and optimized for each of the two aforementioned problems.\",\n \"Title: Natural Embedding of the Stokes Parameters of Polarimetric Synthetic Aperture Radar Images in a Gate-Based Quantum Computer\\nYear: 2022\\n\\nAbstract: Quantum algorithms are designed to process quantum data (quantum bits) in a gate-based quantum computer. They are proven rigorously that they reveal quantum advantages over conventional algorithms when their inputs are certain quantum data or some classical data mapped to quantum data. However, in a practical domain, data are classical in nature, and they are very big in dimension, size, and so on. Hence, there is a challenge to map (embed) classical data to quantum data, and even no quantum advantages of quantum algorithms are demonstrated over conventional ones when one processes the mapped classical data in a gate-based quantum computer. For the practical domain of earth observation (EO), due to the different sensors on remote-sensing platforms, we can map directly some types of EO data to quantum data. In particular, we have polarimetric synthetic aperture radar (PolSAR) images characterized by polarized beams. A polarized state of the polarized beam and a quantum bit are the Doppelganger of a physical state. We map them to each other, and we name this direct mapping a natural embedding, otherwise an artificial embedding. Furthermore, we process our naturally embedded data in a gate-based quantum computer by using a quantum algorithm regardless of its quantum advantages over conventional techniques; namely, we use the QML network as a quantum algorithm to prove that we naturally embedded our data in input qubits of a gate-based quantum computer. Therefore, we employed and directly processed PolSAR images in a QML network. Furthermore, we designed and provided a QML network with an additional layer of a neural network, namely, a hybrid quantum-classical network, and demonstrate how to program (via optimization and backpropagation) this hybrid quantum-classical network when employing and processing PolSAR images. In this work, we used a gate-based quantum computer offered by an IBM Quantum and a classical simulator for a gate-based quantum computer. Our contribution is that we provided very specific EO data with a natural embedding feature, the Doppelganger of quantum bits, and processed them in a hybrid quantum-classical network. More importantly, in the future, these PolSAR data can be processed by future quantum algorithms and future quantum computing platforms to obtain (or demonstrate) some quantum advantages over conventional techniques for EO problems.\",\n \"Title: Cross-Bands Information Transfer to Offset Ambiguities and Atmospheric Phenomena for Multispectral Data Visualization\\nYear: 2021\\n\\nAbstract: Visualization of multispectral images through band selection methods determines an information loss that in utmost cases proves to be critical for the adequate understanding of the represented scene. The R-G-B representation obtained by mapping the visual bands to the R, G, and B channels is highly used due to its great resemblance with the natural color one and aspects perceivable by the human eye. However, despite the similarity in terms of color code, ambiguities between classes such as water and vegetation or atmospheric phenomena like fog, clouds, and smoke that have been penetrated by other bands, remain visible and hinder the process of visualization of the Earth surface. This article presents a set of five different methods to offset the effects caused by ambiguities, fog, light clouds, and smoke by transferring relevant information between bands in order to visually reconstitute those parts of the image affected by atmospheric phenomena. The general concept shared by these methods implies a stacked autoencoder that successfully encompasses the information from all spectral bands into a latent representation used for visualization. Each proposed method is defined by different combination of input and error function formula. Spectral and polar coordinates features represent the possible options for the input, while formulas based on mean squared error or angular spectral distances determine the potential choices in terms of error function definition. The property of angular spectral distance and polar coordinates transformation to obtain illuminant invariant features determined their use in three out of five methods. We evaluate the methods through spectral signature graphical comparison and visual comparison related to the R-G-B representation. We conduct experiments on multiple Sentinel 2 full images.\",\n \"Title: Earth Observation Semantic Data Mining: Latent Dirichlet Allocation-Based Approach\\nYear: 2022\\n\\nAbstract: Recent advances in remote sensing technology have provided (very) high spatial resolution Earth Observation data with abundant latent semantic information. Conventional data processing algorithms are not capable of extracting the latent semantic information form these data and harness their full potential. As a result, semantic information discovery methods, based on data mining techniques, such as latent Dirichlet allocation and bag of visual words models, can discover the latent information. Despite their crucial rule, there are only a few studies in the field of semantic data mining for remote sensing applications. This article is focused on this shortage. Three different scenarios are used to evaluate the semantic information discovery in various remote sensing applications, including both optical and synthetic aperture radar (SAR) data with different spatial resolutions. In the first scenario, semantic discovery method correlated the semantic perception of the user and machine to correct and enhance the user defined Ground Truth map in very high-resolution RGB data. The potential of the semantic discovery is evaluated for wildfire affected area detection in Sentinel-2 data in the second scenario. Finally, in the third scenario, the semantic discovery method is utilized to detect the misclassifications as well as the patches with ambiguous or multiple semantic labels in a Sentinel-1 SAR patch-based benchmark dataset to enhance the robustness and accuracy of the annotation in the dataset. Our results in these three scenarios demonstrated the capability of the data-mining-based semantic information discovery methods for various remote sensing.\",\n \"Title: A Framework for Interactive Visual Interpretation of Remote Sensing Data\\nYear: 2022\\n\\nAbstract: Machine learning methods have shown tremendous success in understanding earth observation data; however, recently, there is a rising claim toward explainable machine learning approaches. Concerned researchers found interpretable visualizations to be greatly helpful in understanding how a model works. In this research, we propose a framework for interactive and interpretable visualization of remote sensing data using two machine learning models and an Elasticsearch (ES) database. Two explainable machine learning models, namely, bag-of-visual-words (BoVWs) and latent Dirichlet allocation (LDA) are chosen to model the data in an unsupervised manner and give a textual representation. The textualized remote sensing data are stored in an ES database. This framework offers several fast content-based search functionalities exploiting the full-text query capabilities of ES based on the respective representations and also offers an efficient storage mechanism for the data.\",\n \"Title: Zero-Shot Cross-Modal Retrieval for Remote Sensing Images With Minimal Supervision\\nYear: 2022\\n\\nAbstract: The performance of a deep-learning-based model primarily relies on the diversity and size of the training dataset. However, obtaining such a large amount of labeled data for practical remote sensing (RS) applications is expensive and labor-intensive. Training protocols have been previously proposed for few-shot learning (FSL) and zero-shot learning (ZSL). However, FSL is not compatible with handling unobserved class data at the inference phase, while ZSL requires many training samples of the seen classes. In this work, we propose a novel training protocol for image retrieval and name it as label-deficit zero-shot learning (LDZSL). We use this novel LDZSL training protocol for the challenging task of cross-sensor data retrieval in RS. This protocol uses very few labeled data samples of the seen classes during training and interprets unobserved class data samples at the inference phase. This strategy is critical as some data modalities are hard to annotate without domain experts. This work proposes a novel bilevel Siamese network to perform the LDZSL cross-sensor retrieval of multispectral and synthetic aperture radar (SAR) images. We use the available georeferenced SAR and multispectral data to domain align the embedding features of the two modalities. We experimentally demonstrate the proposed model's efficacy using the So2Sat dataset compared with the existing state-of-the-art models of the ZSL framework trained under a reduced training set. We also show the generalizability of the proposed model using a sketch-based image retrieval task. Experimental results on the Earth on the Canvas dataset exhibit comparative performance over the literature.\",\n \"Title: Hybrid DNN-Dirichlet Anomaly Detection and Ranking: Case of Burned Areas Discovery\\nYear: 2022\\n\\nAbstract: In the past decade, anomaly detection has experienced an expanding attraction in satellite data analysis. Monitoring wildfire dynamics plays a substantial part in global land management, i.e., to detect and determine the expansion of such areas, estimate the deterioration of forest regions, and assist the intervention plan. In this article, we proposed an approach for Sentinel-2 scenes that detects anomalies in burned area contexts using a rank-ordered method on a single post-event image. We adopted a self-supervised paradigm in learning image representations by training a deep convolutional model to differentiate in a series of geometric transformations. Dirichlet distributions are selected as priors to characterize the variability of the random multinomial distribution in multispectral data. Dirichlet precision parameters are computed from observed data and are used to construct a ranking function that quantifies the degree of anomaly in data based on the softmax responses given by the trained classifier. We evaluated the performance of the proposed method through a cumulative effort of two remote sensing tasks, namely, open-set detection (i.e., test datasets contain classes unseen at the training time) and location separation (i.e., test datasets include images from distinct spatial location than the training images). The experiments were performed on three different datasets, BigEarthNet, and two actual burned area Sentinel-2 datasets from predisposed zones to fire events, Australia and Bolivia. We achieved in all test datasets state-of-the-art performance, considering the substantial and diversified types of natural anomalies in multispectral data.\",\n \"Title: Sar Super-Resolution Using Physics-Aware Adaptive Compressed Sensing\\nYear: 2022\\n\\nAbstract: The resolution requirements of modern radar applications are increasing rapidly and cannot be fulfilled by the limited number of wide-band radar systems. Many approaches have been explored to solve this problem under the topic of super-resolution. In this paper, we propose a hybrid algorithm for resolution improvement, where we aim to combine the adaptability of deep neural networks with the reliability and expertise of traditional domain-specific SAR processing.\",\n \"Title: Causality for Remote Sensing: An Exploratory Study\\nYear: 2022\\n\\nAbstract: Causality is one of the most important topics in a Machine Learning (ML) research, and it gives insights beyond the dependency of data points. Causality is a very vital concept also for investigating the dynamic surface of our living planet. However, there are not many attempts for integrating a causal model in Remote Sensing (RS) methodologies. Hence, in this paper, we propose to use patch-based RS images and to represent each patch-based image by a single variable (e.g. entropy). Then we use a Structural Equation Model (SEM) to study their cause-effect relation. Moreover, the SEM is a simple causal model characterized by a Directed Acyclic Graph (DAG). Its nodes are causal variables, and its edges represent causal relationships among causal variables if and only if causal variables are dependent.\",\n \"Title: Achieving Information Super-resolution for Sentinel-2 NDVI Through Gaussian Process Regression\\nYear: 2022\\n\\nAbstract: Super-resolution is used to recover high resolution images from low resolution images. We use this concept in a slightly different context to achieve higher quality knowledge from low resolution satellite images. The technique involves transfer learning from high to low resolution images using a Gaussian Process Regression model. We use high resolution drone images to train the model. This technique is applied in three case studies to verify the consistency of results in case of NDVI computation. However, the same technique can be applied to obtain for other application of satellite images in plant vigor assessment.\",\n \"Title: SAR Image Classification Using Mixed Spatial-Spectral Information and Pre-trained Convolutional Neural Networks\\nYear: 2022\\n\\nAbstract: The recent technological advancements in remote sensing lead to an increased importance regarding the analysis of satellite data targeting security and surveillance tasks. Although the availability of data products is constantly augmented and the advances in Deep Learning technologies are constant, Synthetic Aperture Radar (SAR) image classification remains a challenge in the remote sensing domain because standard convolutional neural network-based architectures may encounter difficulties in recognizing objects that are characterized by similar texture, but different backscattering patterns. Moreover, training deep learning architectures requires a large volume of annotated data, which, in general, represents an obstacle, especially in the case of the remote sensing domain. This article addresses complex-valued SAR image classification through both spatial and Fourier-domain features, extracted by means of pretrained neural networks. While spatial features allow extracting knowl-edge regarding the structure and texture of the objects from intensity images, the physical properties of the objects are learned from radar spectrograms. In addition, we show that considering different polarizations of the SAR sensor, we are able to obtain better visual classifications. The experiments are conducted over Sentinel-1images, which are freely available for download under the Copernicus initiative.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":185,"string":"185"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13581,"string":"13,581"}}},{"rowIdx":13582,"cells":{"question_id":{"kind":"string","value":"Jorge Cardoso"},"question":{"kind":"string","value":"Generate an abstract for the title \"VR Book: A Tangible Interface for Smartphone-based Virtual Reality\" using the following items: 1. Tangible VR Book\n2. Marker-based solution\n3. Smartphone-based VR\n4. Object tracking\n5. User feedback"},"answer":{"kind":"string","value":"In this work we present three prototypes of the tangible VR Book: a visual marker-based solution to tangible interaction for smartphone-based Virtual Reality (VR). Smartphone-based VR represents a low barrier to entry in VR experiences given that many people nowadays own a smartphone device and that VR headsets for these devices are affordable and quick to set up when compared to desktop-based VR. Tangible interaction in smartphone-based VR has not been much explored, in our opinion, despite the fact that it can result in easy to use and engaging experiences. In this work, we explore a marker-based solution to object tracking that allows tangibles to be created in an easy and cheap way, maintaining the overall system accessible. We describe a design space for visual marker-based tangible interaction and three prototypes of a tangible VR Book that explore different aspects of the design space. We also present user feedback on their expectations regarding the interaction with the VR Book."},"correct_answer":{"kind":"string","value":"In this work we present three prototypes of the tangible VR Book: a visual marker-based solution to tangible interaction for smartphone-based Virtual Reality (VR). Smartphone-based VR represents a low barrier to entry in VR experiences given that many people nowadays own a smartphone device and that VR headsets for these devices are affordable and quick to set up when compared to desktop-based VR. Tangible interaction in smartphone-based VR has not been much explored, in our opinion, despite the fact that it can result in easy to use and engaging experiences. In this work, we explore a marker-based solution to object tracking that allows tangibles to be created in an easy and cheap way, maintaining the overall system accessible. We describe a design space for visual marker-based tangible interaction and three prototypes of a tangible VR Book that explore different aspects of the design space. We also present user feedback on their expectations regarding the interaction with the VR Book."},"documents":{"kind":"list like","value":["Title: Integrating Business Process And User Interface Models Using A Model-Driven Approach\nYear: 2009\n\nAbstract: Business services are complex entities that encompass descriptions about different aspects including business processes and user interfaces. Topically, the modeling of business services results on several correlated models. On the one hand, there is the need to keep these models apart in order to attain the levels of abstractions needed to model different aspects of a service. On the other hand, it is important to maintain the consistency and integrity of models. In this paper, we show how to maintain the consistency and integrity of models and present a use case for the integration of user interface design and business process models.","Title: Business Process Control-Flow Complexity: Metric, Evaluation, And Validation\nYear: 2008\n\nAbstract: Organizations are increasingly faced with the challenge of managing business processes, workflows, and recently, Web processes. One important aspect of business processes that has been overlooked is their complexity. High complexity in processes may result in poor understandability, errors, defects, and exceptions, leading processes to need more time to develop, test, and maintain. Therefore, excessive complexity should be avoided Business process measurement is the task of empirically and objectively assigning numbers to the properties of business processes in such a way so as to describe them. Desirable attributes to study and measure include complexity, cost, maintainability, and reliability. In our work, we will focus on investigating process complexity. We present and describe a metric to analyze the control-flow complexity of business processes. The metric is evaluated in terms of Weyuker's properties in order to guarantee that it qualifies as good and comprehensive. To test the validity of the metric, we describe the experiment we have carried out for empirically validating the metric.","Title: Workflow Quality of Service\nYear: 2002\n\nAbstract: Workflow management systems (WfMSs) have been used to support various types of business processes for more than a decade now. In e- commerce processes, suppliers and customers define a binding agreement or contract between the two parties, specifying quality of service (QoS) items such as products or services to be delivered, deadlines, quality of products, and cost of service. Management of such QoS directly impacts success of organizations participating in e-commerce. Organizations operating in modern markets require an excellent degree of quality of service management. Products and services must be available to customers with well-defined specifications. A good management of quality leads to the creation of quality products and services, which in turn fulfills customer expectations and achieves customer satisfaction. Therefore, when services or products are created or managed using workflow processes, the underlying WfMS must accept the specification, be able to predict, monitor, and control the QoS rendered to customers. To achieve these objectives the first step is to develop an adequate QoS model for workflow processes and develop methods to compute QoS.","Title: The Internet Of Services\nYear: 2009\n\nAbstract: The vision of the Internet of Services (IoS) can be seen as a new business model that can radically change the way we discover, select, invoke and interact with services. In the same way that the Internet can be viewed as a Web of Information, it can also be configured to support an Internet of Services. The IoS describes an approach that uses the Internet as a medium for selling and purchasing services. As a result, services become tradable entities. Service marketplaces, a logical location where providers and consumers are brought together to trade services and to engage in business interaction, are enabling the creation of platforms for the IoS vision. Thus, the IoS provides the business and technical base for advanced business models where service providers and consumers form business networks for service provision and consumption.","Title: Semantic E-Workflow Composition\nYear: 2003\n\nAbstract: Systems and infrastructures are currently being developed to support Web services. The main idea is to encapsulate an organization's functionality within an appropriate interface and advertise it as Web services. While in some cases Web services may be utilized in an isolated form, it is normal to expect Web services to be integrated as part of workflow processes. The composition of workflow processes that model e-service applications differs from the design of traditional workflows, in terms of the number of tasks (Web services) available to the composition process, in their heterogeneity, and in their autonomy. Therefore, two problems need to be solved: how to efficiently discover Web services—based on functional and operational requirements—and how to facilitate the interoperability of heterogeneous Web services. In this paper, we present a solution within the context of the emerging Semantic Web that includes use of ontologies to overcome some of the problem. We describe a prototype that has been implemented to illustrate how discovery and interoperability functions are achieved more efficiently.","Title: PuReWidgets: a programming toolkit for interactive public display applications\nYear: 2012\n\nAbstract: Interaction is repeatedly pointed out as a key enabling element towards more engaging and valuable public displays. Still, most digital public displays today do not support any interactive features. We argue that this is mainly due to the lack of efficient and clear abstractions that developers can use to incorporate interactivity into their applications. As a consequence, interaction represents a major overhead for developers, and users are faced with inconsistent interaction models across different displays. This paper describes the results of a study on interaction widgets for generalized interaction with public displays. We present PuReWidgets, a toolkit that supports multiple interaction mechanisms, automatically generated graphical interfaces, asynchronous events and concurrent interaction. This is an early effort towards the creation of a programming toolkit that developers can incorporate into their public display applications to support the interaction process across multiple display systems without considering the specifics of what interaction modality will be used on each particular display.","Title: Open Semantic Service Networks: Modeling and Analysis.\nYear: 2013\n\nAbstract: A new interesting research area is the representation and analysis of the networked economy using Open Semantic Service Networks (OSSN). OSSN are represented using the service description language USDL to model nodes and using the service relationship model OSSR to model edges. Nonetheless, in their current form USDL and OSSR do not provide constructs to capture the dynamic behavior of service networks. To bridge this gap, we used the General System Theory (GST) as a framework guiding the extension of USDL and OSSR to model dynamic OSSN. We evaluated the extensions made by applying USDL and OSSR to two distinct types of dynamic OSSN analysis: 1) evolutionary by using a Preferential Attachment (PA) and 2) analytical by using concepts from System Dynamics (SD). Results indicate that OSSN can constitute the first stepping stones toward the analysis of global service-based economies.","Title: Legosc - Mindstorms Nxt Robotics Programming For Artists\nYear: 2008\n\nAbstract: Robotics is an interesting but difficult area for digital artists who generally don't have much academic background on electronics or computer programming. Digital art students normally use high-level application to program their visual and sonorous installations. This paper presents LegOSC - a tool that allows the control of the Mindstorms NXT robots from any application that uses the Open Sound Control protocol which is implemented by most of those high-level applications. This allows artists to create works which incorporate robotic parts using the familiar programming environment.","Title: Using semantic web and service oriented technologies to build loosely coupled systems: SWOAT - a service and semantic web oriented architecture technology\nYear: 2007\n\nAbstract: The creation of loosely coupled and flexible applications has been a challenge faced by most organizations. This has been important because organization systems need to quickly respond and adapt to changes that occur in the business environment. In order to address these key issues, we implemented SWOAT, a 'Service and Semantic Web Oriented Architecture Technology' based middleware. Our system uses ontologies to semantically describe and formalize the information model of the organization, providing a global and integrated view over a set of database systems. It also allows interoperability with several systems using Web Services. Using ontologies and Web services, clients remain loosely coupled from data sources. As a result, data structures can be changed and moved without having to change all clients, internal or external to the organization.","Title: Poseidon: A Framework To Assist Web Process Design Based On Business Cases\nYear: 2006\n\nAbstract: Systems and infrastructures are currently being developed to support Web services and Web processes. One prominent solution to manage and coordinate Web services is the use of workflow technology. For more than two decades now, workflow management systems architectures, language specifications, and workflow analysis techniques have been extensively studied. While these areas of research have made the development of sophisticated workflow systems possible, one important research area that has been overlooked is the lifecycle of process application development. As a result, current process management systems support the analysis, enactment, and ad hoc design of workflows, but they lack the tools and methods to assist process design. The purpose of our study is to present a framework to assist and guide process analysts and designers in their task. The Poseidon framework includes a participative and an analytical design method. The participative phase uses a clean sheet approach and starts by constructing a business case table which captures all the business cases represented by a process. Afterward, an analytical design phase is followed where scheduling functions are derived from the business case table and are used to build the structure of a process.","Title: Academic and industrial research: do their approaches differ in adding semantics to web services?\nYear: 2004\n\nAbstract: Since the new terms, “Semantic Web” and “Web services”, have been introduced, researchers have followed two different roads. Following one road, academia has focused on developing a new set of languages to enable the automation of Web services execution and integration based on the Semantic Web. On the other road, industry has taken the lead to propose and develop technologies and infrastructures to support Web services and Web processes without, until recently, paying much attention to semantics. It is fundamental to analyze the trend that is being followed with regard to the “Semantic Web” and “Web services”. Therefore, two important questions need to be answered: “do the approaches taken by academia and industry differ in how they add semantics to Web services?” and “are their efforts converging or diverging?” This paper, based on a panel discussion at an international conference on Web services, which consisted of members of both academia and industry, addresses precisely these two questions.","Title: Evaluating the Process Control-Flow Complexity Measure\nYear: 2005\n\nAbstract: Process measurement is the task of empirically and objectivelyassigning numbers to the attributes of processes in such a wayas to describe them. We define process complexity as the degreeto which a process is difficult to analyze, understand or explain.One way to analyze a processý complexity is to use a process control-flow complexity measure. This measure analyzes the control- flow of processes and can be applied to both Web processes andworkflows. In this paper, we discuss how to evaluate the control- flow complexity measure to ensure that it can be qualify as a good and comprehensive one.","Title: Moving from syntactic to semantic organizations using JXML2OWL\nYear: 2008\n\nAbstract: Today's enterprises face critical needs in integrating disparate information spread over several data sources inside and even outside the organization. Most organizations already rely on XML standard to define their data models. Unfortunately, even when using XML to represent data, problems arise when it is necessary to integrate different data sources. Emerging Semantic Web technologies, such as ontologies, RDF, RDFS, and OWL, can play an important role in the semantic definition and integration of data. The purpose of our study is to present a framework to assist organizations to move from a syntactic data infrastructure defined in XML to a semantic data infrastructure using OWL. The framework supports mappings and fully automated instance transformation from syntactic data sources in XML format to a common shared global model defined by an ontology using Semantic Web technologies. The presented framework, JXML2OWL, allows organizations to automatically convert their XML data sources to a semantic model defined in OWL.","Title: Integrating HAD Organizational Data Assets using Semantic Web Technologies\nYear: 2007\n\nAbstract: To create value based on information and knowledge, organizations need to recognize that they are composed of various types of data assets. In this context, data integration has been proven to be a challenge due to the heterogeneity of the information systems involved. The Extensible Markup Language (XML) was one of most successful solutions developed to provide business-to-business integration. However, XML \"lacks semantics\". Thus problems arise when it is necessary to manipulate and integrate different XML data sources. Consequently, today's organizations are again shifting from a syntactic interoperability level to a semantic one. This paper presents a Semantic data Integration Middleware (SIM) which, when based on a single query, integrates data residing in different data sources. The middleware uses an ontology-based multi-source data extractor/wrapper approach to transform data into semantic knowledge.","Title: IntelliGEN: A Distributed Workflow System for Discovering Protein-Protein Interactions\nYear: 2003\n\nAbstract: A large genomics project involves a significant number of researchers and technicians performing dozens of tasks, either manual (e.g. performing laboratory experiments), computer assisted (e.g. looking for genes in the GENBANK database), or sometimes performed entirely automatically by the computer (e.g. sequence assembly). It has become apparent that managing such projects poses overwhelming problems and may lead to results of lower or even unacceptable quality, or possibly drastically increased project costs. In this paper, we present a design and an initial implementation of a distributed workflow system created to schedule and support activities in a genomics laboratory. The focus of the activities in the laboratory is the discovery of protein-protein interactions of fungi, specifically Neurospora crassa. We present our approach of developing, adapting and applying workflow technology in the genomics lab and illustrate it using one distinct part of a larger workflow to discover protein- protein interactions. Novel features of our system include the ability to monitor the quality and timeliness of the results and if necessary, suggesting and incorporating changes to the selected tasks and their scheduling. In the new millennium biology has gone through a paradigm shift with the advent of Genomics, the study of the structure, function, and evolution of whole genomes. The term \"genomics\" was coined in 1986, and institutional support for this new discipline began with the creation of the National Center for Human Genome Research (NCHGR) in 1989. The initial goal of this discipline was the determination of the entire DNA sequence of a human being (a human genome) and several related model organisms that have played a central role in Genetics, including the bacterium Escherichia coli, the yeast Saccharomyces cerevisiae, the worm Canerohabditis elegans, and the fruit fly Drosophila melanogaster. Sequencing the human genome and those of each model system were daunting distributed computing tasks which required collecting, storing, integrating, retrieving, and distributing 3 billion base pairs (bp) of sequence on the human genome alone. Many aspects of computer science were brought to bear to solve these problems (43, 81). What genomics promises is that with the availability of the complete genetic blueprints of","Title: Modeling Service Relationships for Service Networks.\nYear: 2013\n\nAbstract: The last decade has seen an increased interest in the study of networks in many fields of science. Examples are numerous, from sociology to biology, and to physical systems such as power grids. Nonetheless, the field of service networks has received less attention. Previous research has mainly tackled the modeling of single service systems and service compositions, often focusing only on studying temporal relationships between services. The objective of this paper is to propose a computational model to represent the various types of relationships which can be established between services systems to model service networks. This work acquires a particular importance since the study of service networks can bring new scientific discoveries on how service-based economies operate at a global scale.","Title: Discovering Semantic Web Services with and without a Common Ontology Commitment\nYear: 2006\n\nAbstract: This paper presents an algorithm to match a semantic Web service request against semantic Web service advertisements. The algorithm is to be used by systems to discover semantic Web services, such as the UDDI. Matching is based on the assessment of the similarity among semantic Web service properties, such as inputs and outputs. Semantic Web services have their inputs and outputs annotated or described by ontological concepts. The algorithm is able to match a semantic Web service request against advertisements that are annotated with concepts with and without a common ontology commitment. The similarity of inputs and outputs is evaluated based on concepts (classes), their semantic relations, and their common and distinguishing features (properties","Title: A weighted coupling metric for business process models\nYear: 2007\n\nAbstract: Various efforts recently aimed at the development of quality metrics for process models. In this paper, we propose a new notion of coupling, which has been used successfully in software engineering for many years. It extends other work by specifically incorporating the effects of different types of connectors used on a process model's coupling level.","Title: Semantic integration of web services and peertopeer networks to achieve fault-tolerance.\nYear: 2006\n\nAbstract: One fundamental property that critical Web services need to provide is a high level of availability. Along with the development of Web services, considerable technological advances are being made to use the semantic Web to achieve the automated processing and integration of data and applications. This paper describes the implementation of the Whisper architecture. This architecture semantically integrates Web services with a peer-to-peer infrastructure to increase service availability. Whisper achieves transparent fault-tolerance by automatically forwarding Web service requests to semantically equivalent peers that are dynamically located, selected, and invoked.","Title: Healthcare enterprise process development and integration\nYear: 2003\n\nAbstract: Healthcare enterprises involve complex processes that span diverse groups and organisations. These processes involve clinical and administrative tasks, large volumes of data, and large numbers of patients and personnel. The tasks can be performed either by humans or by automated systems. In the latter case, the tasks are supported by a variety of software applications and information systems which are very often heterogeneous, autonomous, and distributed. The development of systems to manage and automate these processes has increasingly played an important role in improving the efficiency of healthcare enterprises. In this paper we look at four healthcare and medical applications that involve investigative, clinical, and administrative functions. Based on these applications, we derive the requirements for developing enterprise applications that involve the coordination of a variety of tasks performed by humans, information systems, and legacy applications.","Title: Mapping between heterogeneous XML and OWL transaction representations in B2B integration\nYear: 2011\n\nAbstract: XML-based standards have been widely used to enable and ease Business-to-Business (B2B) integration. Examples of standards include cXML, CIDX and ebXML. While these XML-based standards are syntactic, contemporary organizations have available new means to structure their internal data representations using semantic descriptions, such as RDF(S) and OWL. This scenario poses an interesting challenge: ''How to reconcile external XML-based standards and internal OWL-based representations in B2B integration scenarios?'' In this paper, we present a conceptual approach, and its implementation, to integrate external syntactic data representations with organizational internal semantic data representations by using the notion of heterogeneous mappings which are established between the two types of representations. The application developed, B2BISS, enables an effective management of mappings. As the number of mappings stored in the repository increases over time, organizations can gradually rely on a semi-automatic to automatic B2B integration.","Title: Challenges of business service monitoring in the internet of services\nYear: 2008\n\nAbstract: The most industrialized countries have entered an era where their prosperity is largely created through a service economy. There is a transition from a manufacturing based economy to a service based economy. In the Internet of Services (IoS) vision services are seen as tradable goods. Business services are a major asset in this context. Since they are inherently different from e.g. web services, there is a need for a specialized monitoring approach. In this paper we describe the differences between e-services and business services, present an approach to describe business services, and discuss challenges regarding their monitoring requirements.","Title: Analysis and Validation of Control-Flow Complexity Measures with BPMN Process Models\nYear: 2009\n\nAbstract: Evaluating the complexity of business processes during the early stages of their development, primarily during the process modelling phase, provides organizations and stakeholders with process models which are easier to understand and easier to maintain. This presents advantages when carrying out evolution tasks in process models - key activities, given the current competitive market. In this work, we present the use and validation of the CFC metric to evaluate the complexity of business processes modelled with BPMN. The complexity of processes is evaluated from a control-flow perspective. An empirical evaluation has been carried out in order to demonstrate that the CFC metric can be useful when applied to BPMN models, providing information about their ease of maintenance.","Title: Business process quality metrics: log-based complexity of workflow patterns\nYear: 2007\n\nAbstract: We believe that analysis tools for BPM should provide other analytical capabilities besides verification. Namely, they should\n provide mechanisms to analyze the complexity of workflows. High complexity in workflows may result in poor understandability,\n errors, defects, and exceptions leading processes to need more time to develop, test, and maintain. Therefore, excessive complexity\n should be avoided. The major goal of this paper is to describe a quality metric to analyze the complexity of workflow patterns\n from a log-based perspective.","Title: Benchmarking a Semantic Web Service Architecture for Fault-tolerant B2B Integration\nYear: 2006\n\nAbstract: With the development and maturity of Service- Oriented Architectures (SOA) to support business-tobusiness transactions, organizations are implementing Web services to expose their public functionalities associated with internal systems and business processes. In many business processes, Web services need to provide is a high level of availability, since the globalization of the Internet enables business partners to easily switch to other competitors when services are not available. Along with the development of SOA, considerable technological advances are being made to use the semantic Web to achieve the automated processing and integration of data and applications. This paper describes the implementation and benchmarking of an architecture that semantically integrates Web services with a peer-to-peer infrastructure to increase service availability through fault-tolerance.","Title: Composing e-Workflows using Syntactic, Operational and Semantic Information\nYear: 2002\n\nAbstract: With the development and maturity of infrastructures that support e-services, we expect organizations to incorporate Web services as part of their business processes. While in some cases Web services may be utilized in an isolated form, it is natural to expect that Web services will be integrated as part of workflows. The composition of workflows to model e-service applications differs from the design of traditional workflows due to the number of Web services available during the composition process, their heterogeneity, and their autonomy. The composition process raises two main problems that need to be solved: how to efficiently discover Web services and how to facilitate their integration with the workflow host. To support, facilitate, and assist the composition of process, we present a solution based on ontologies. We have developed an algorithm and prototype that workflow systems and discovery mechanisms can use to find Web services with desired interfaces, operational metrics, and to assist designers in resolving heterogeneity issues among Web services. Our approach provides an important solution to enhance Web service discovery and integration.","Title: Measuring the Compliance of Processes with Reference Models\nYear: 2009\n\nAbstract: Reference models provide a set of generally accepted best practices to create efficient processes to be deployed inside organizations. However, a central challenge is to determine how these best practices are implemented in practice. One limitation of existing approaches for measuring compliance is the assumption that the compliance can be determined using the notion of process equivalence. Nonetheless, the use of equivalence algorithms is not adequate since two models can have different structures but one process can still be compliant with the other. This paper presents a new approach and algorithm which allow to measure the compliance of process models with reference models. We evaluate our approach by measuring the compliance of a model currently used by a German passenger airline with the IT Infrastructure Library (ITIL) reference model and by comparing our results with existing approaches.","Title: Semantic Web Processes: Semantics Enabled Annotation, Discovery, Composition and Orchestration of Web Scale Processes\nYear: 2003\n\nAbstract: This tutorial deals with the evolution of inter- Enterprise and Web scale process to support e-commerce and e-services. It taps into the promises of two of the hottest R&D and technology areas: Web services and the Semantic Web. It presents how applying semantics to each of the steps in the Semantic Web Process lifecycle can help address critical issues in reuse, integration and scalability.","Title: Mobile applications for open display networks: common design considerations\nYear: 2013\n\nAbstract: Mobile devices can be a powerful tool for interaction with public displays, but mobile applications supporting this form of interaction are not yet part of our everyday reality. There are no widely accepted abstractions, standards, or practices that may enable systematic interaction between mobile devices and public displays. We envision public displays to move away from a world of closed display networks to scenarios where mobile applications could allow people to interact with the myriad of displays they might encounter during their everyday trips. In this research, we study the key processes involved in this collaborative interaction between public shared displays and mobile applications. Based on the lessons learned from our own development and deployment of 3 applications, and also on the analysis of the interactive features described in the literature, we have identified 8 key processes that may shape this form of interaction: Discovery, Association, Presence Management, Exploration, Interface Migration, Controller, Media Upload and Media Download. The contribution of this work is the identification of these high-level processes and an elicitation of the main design considerations for display networks.","Title: Inter-enterprise System and Application Integration: A Reality Check\nYear: 2007\n\nAbstract: This paper structures the summary of the panel held at the 9th International Conference on Enterprise Information Systems, Funchal, Madeira, 12-16 June 2007 that addressed the following question: \"Are you still working on Inter-Enterprise System and Application Integration?\" The panel aggregated distinguished experts from the areas of process management, workflow, Web services, SOA, and Semantic Web.","Title: XML-based modeling and simulation: web service technologies and their synergy with simulation\nYear: 2002\n\nAbstract: The World Wide Web has had an huge influence on the computing field in general as well as simulation in particular (<i>e.g.</i>, Web-Based Simulation). A new wave of development based upon XML has started. Two of the most interesting aspects of this development are the Semantic Web and Web Services. This paper examines the synergy between Web service technology and simulation. In one direction, Web service processes can be simulated for the purpose of correcting/improving the design. In the other direction, simulation models/components can be built out of Web services. Work on seamlessly using simulation as a part of Web service composition and process design, as well as on using Web services to re-build the JSIM Web-based simulation environment is highlighted.","Title: Ontological Representation and Governance of Business Semantics in Compliant Service Networks.\nYear: 2013\n\nAbstract: The Internet would enable new ways for service innovation and trading, as well as for analysing the resulting value networks, with an unprecedented level of scale and dynamics. Yet most related economic activities remain of a largely brittle and manual nature. Service-oriented business implementations focus on operational aspects at the cost of value creation aspects such as quality and regulatory compliance. Indeed they enforce how to carry out a certain business in a prefixed non-adaptive manner rather than capturing the semantics of a business domain in a way that would enable service systems to adapt their role in changing value propositions. In this paper we set requirements for SDL-compliant business service semantics, and propose a method for their ontological representation and governance. We demonstrate an implementation of our approach in the context of service-oriented Information Governance.","Title: Distributed Contracting and Monitoring in the Internet of Services\nYear: 2009\n\nAbstract: The recent approval of the EU Services Directive is fostering the Internet of Services (IoS) and will promote the emergence of marketplaces for business and real-world services. From a research perspective, the IoS will require a new bread of technological infrastructures to support the concepts of business service description, contract management from various perspectives, end-to-end marketplaces, and business monitoring. The IoS is a vision referring to web service-based digital societies. When service hosting moves from best-effort provisioning to guaranteed service delivery, monitoring becomes a crucial point of proof for providers and consumers of such services. We present the uplifting of technical contract monitoring results to business effects based on the distributed service infrastructure developed in project THESEUS, use case TEXO.","Title: Cloud computing automation: integrating USDL and TOSCA\nYear: 2013\n\nAbstract: Standardization efforts to simplify the management of cloud applications are being conducted in isolation. The objective of this paper is to investigate to which extend two promising specifications, USDL and TOSCA, can be integrated to automate the lifecycle of cloud applications. In our approach, we selected a commercial SaaS CRM platform, modeled it using the service description language USDL, modeled its cloud deployment using TOSCA, and constructed a prototypical platform to integrate service selection with deployment. Our evaluation indicates that a high level of integration is possible. We were able to fully automatize the remote deployment of a cloud service after it was selected by a customer in a marketplace. Architectural decisions emerged during the construction of the platform and were related to global service identification and access, multi-layer routing, and dynamic binding.","Title: Service Engineering for Future Business Value Networks\nYear: 2008\n\nAbstract: Traditionally, business value networks have orchestrated human and technical resources that worked together to form relationships and to add value to a product or service. The Internet and the Web have extended traditional business networks by allowing also a web of different digital resources to work together to create value. Additionally, the increasing adoption of service-oriented architectures has allowed the creation of service ecosystems in which Web services are exposed and connected. The TEXO project proposes to combine these two trends to create what is called Future Business Value Networks which enable IT-supported value networks via service ecosystems. This paper addresses the main challenging issues that need to be explored to provide an integrated technical infrastructure to support this emerging new type of business networks.","Title: Web Services Discovery Utilizing Semantically Annotated WSDL\nYear: 2008\n\nAbstract: To make semantic Web services accessible to users, providers use registries to publish them. Unfortunately, the current registries use discovery mechanisms which are inefficient, as they do not support discovery based on the semantics of the services and thus lead to a considerable number of irrelevant matches. Semantic discovery and matching of services is a promising approach to address this challenge. This paper presents an algorithm to match a semantic Web service request described with SAWSDL against semantic Web service advertisements. The algorithm is novel in three fundamental aspects. First, the similarity among semantic Web service properties, such as inputs and outputs, is evaluated using Tversky's model which is based on concepts (classes), their semantic relationships, and their common and distinguishing features (properties). Second, the algorithm, not only takes into account services' inputs and outputs, but it also considers the functionality of services. Finally, the algorithm is able to match a semantic Web service request against advertisements that are annotated with concepts that are with or without a common ontological commitment. In other words, it can evaluate the similarity of concepts defined in the context of different ontologies.","Title: Open service networks: research directions\nYear: 2013\n\nAbstract: Service networks are a new research field which promises to increase our knowledge of service-based systems. Understanding what factors explain the evolution of global service networks can lead to more efficient and balanced infrastructures. The challenge consists in developing a novel perspective by connecting models representing services (e.g. cloud computing and governmental services) into networks. Theories, methods, and applications to be developed will enable, afterwards, to understand, describe, explain, analyse, predict, and control the structure and behaviour of global service networks.","Title: A Framework for Assessing Strategies and Technologies for Dynamic Packaging Applications in E-Tourism.\nYear: 2007\n\nAbstract: Dynamic packaging has been introduced as an innovative technology allowing for the automated online configuration and assembling of packaged travel products for individual customers. While dynamic packaging has been widely accepted by customers in UK and US markets, the possible strategic impacts enabled by dynamic packaging technologies are still uncertain. This article pro- vides a study of the strategic opportunities enabled by dynamic packaging, highlighting the key success factors. Current dynamic packaging applications of the three major online travel agencies are evaluated on the basis of the developed analysis framework. The results show that an appro- priate level of integration of tourism information systems is a key factor for further realizing the strategic opportunities of dynamic packaging. We then discuss Web services and semantics as a possible solution for overcoming the interoperability problems that (current) dynamic packaging applications face.","Title: An entropy-based uncertainty measure of process models\nYear: 2011\n\nAbstract: In managing business processes, the process uncertainty and variability are significant factors causing difficulties in prediction and decision making, which evokes and augments the importance and need of process measures for systematic analysis. We propose an entropy-based process measure to quantify the uncertainty of business process models. The proposed measure enables capturing the dynamic behavior of processes, in contrast to previous work which focused on providing measures for the static aspect of process models.","Title: What makes process models understandable?\nYear: 2007\n\nAbstract: Despite that formal and informal quality aspects are of significantimportance to business process modeling, there is only little empiricalwork reported on process model quality and its impact factors. Inthis paper we investigate understandability as a proxy for quality of processmodels and focus on its relations with personal and model characteristics.We used a questionnaire in classes at three European universitiesand generated several novel hypotheses from an exploratory data analysis.Furthermore, we interviewed practitioners to validate our findings.The results reveal that participants tend to exaggerate the differences inmodel understandability, that self-assessment of modeling competenceappears to be invalid, and that the number of arcs in models has animportant influence on understandability.","Title: Towards thresholds of control flow complexity measures for BPMN models\nYear: 2011\n\nAbstract: Business process models are considered to be a good mechanism for communication among stakeholders and are a key instrument in the analysis and design of information systems. It is therefore important to design business process models with a high level of quality, which can be discovered through measurement application. Several measurement initiatives exist in the literature, but these measures are only useful in real world decision making if we also have criteria with which to establish the goodness of models. We consider that measures with thresholds and decision criteria form indicators. Indicators allow us to make decisions by using the values of the measures which models should not exceed to ascertain whether the model is good in practice. In this paper we present the initial empirical results from which thresholds for the Control-Flow Complexity measure applied in BPMN models have been obtained according to the Bender method. Our findings reveal that there are different levels of understandability depending on the number of decision nodes: a very easily understandable model would have no more than 6 xor nodes, 1 or nodes and 1 and nodes, versus the 46 xor nodes, 14 or nodes and 7 and nodes which would constitute a model with a very difficult level of understandability.","Title: Web process and workflow path mining using the Multimethod approach\nYear: 2006\n\nAbstract: Workflow Management Systems (WfMS) provide a fundamental technological infrastructure to define and manage business processes efficiently. WfMS logs contain valuable data that can be used to discover and extract knowledge about the execution of workflows and processes. One piece of important and useful information that can be discovered is related to the prediction of the path that will be followed during the execution of a workflow. We call this type of discovery, path mining. In this paper, we present and describe how path mining can be achieved using different data mining techniques including the Multimethod approach.","Title: On a Quest for Good Process Models: The Cross-Connectivity Metric\nYear: 2008\n\nAbstract: Business process modeling is an important corporate activity, but the understanding of what constitutes good process models is rather limited. In this paper, we turn to the cognitive dimensions framework and identify the understanding of the structural relationship between any pair of model elements as a hard mental operation. Based on the weakest-link metaphor, we introduce the cross-connectivity metric that measures the strength of the links between process model elements. The definition of this new metric builds on the hypothesis that process models are easier understood and contain less errors if they have a high cross-connectivity. We undertake a thorough empirical evaluation to test this hypothesis and present our findings. The good performance of this novel metric underlines the importance of cognitive research for advancing the field of process model measurement.","Title: Network latency adaptive tempo in the Public Sound Objects system\nYear: 2005\n\nAbstract: In recent years Computer Network-Music has increasingly captured the attention of the Computer Music Community. With the advent of Internet communication, geographical displacement amongst the participants of a computer mediated music performance achieved world wide extension. However, when established over long distance networks, this form of musical communication has a fundamental problem: network latency (or net-delay) is an impediment for real-time collaboration. From a recent study, carried out by the authors, a relation between network latency tolerance and Music Tempo was established. This result emerged from an experiment, in which simulated network latency conditions were applied to the performance of different musicians playing jazz standard tunes.The Public Sound Objects (PSOs) project is web-based shared musical space, which has been an experimental framework to implement and test different approaches for on-line music communication. This paper describe features implemented in the latest version of the PSOs system, including the notion of a network-music instrument incorporating latency as a software function, by dynamically adapting its tempo to the communication delay measured in real-time.","Title: The Semantic Web Vision: Where Are We?\nYear: 2007\n\nAbstract: The full vision of the Semantic Web and its mainstream adoption is still five to 10 years away. To get a current snapshot of key trends and developments, University of Madeira researchers surveyed 627 Semantic Web researchers and practitioners to learn how they're using and interrelating semantic technologies.","Title: The Business Process Knowledge Framework\nYear: 2007\n\nAbstract: Organizations today are confronted with huge problems following and implementing their own business process models. On the one hand, due to a lack of planning and requirements analysis, process models are often unfeasible or difficult to execute in practice. On the other hand, process designers often ignore the importance of studying the different roles and their perspectives on a business process when constructing a process model. This leads to the deployment of process models that do not satisfy process stakeholders. This paper addresses such problems and proposes a business process knowledge framework as a possible solution. Our framework integrates three elements that we consider fundamental to model business processes: stakeholders' perspectives, knowledge types and views. We show how the business process framework can contribute to the improvement of the process knowledge acquisition phase of process design, and how it can support process knowledge communication to stakeholders.","Title: Enterprise Information Systems: 9th International Conference, ICEIS 2007, Funchal, Madeira, June 12-16, 2007, Revised Selected Papers\nYear: 2009\n\nAbstract: This books contains the best papers of the 9th International Conference on Enterprise Information Systems, ICEIS 2007, held in Funchal, Madeira, Portugal, in June 2007. The 6 invited papers are presented together with23 papers whichwere carefully reviewed and selected from 72 full papers accepted for presentation at the conference itself (out of 644 submissions). The topics covered are: databases and information systems integration; artificial intelligence and decision support systems; information systems analysis and specification;software agents and internet computing; and human-computer interaction.","Title: Soundserver: Data Sonification On-Demand for Computational Instances\nYear: 2004\n\nAbstract: The rapid accumulation of large collections of data has created the need for efficient and intelligent schemes for knowledge extraction and results analysis. The resulting information is typically visual- ized, but it may also be presented through audio techniques such as sonification. Sonification techniques become especially interest- ing when the client application runs on graphically limited devices such as mobile phones or PDAs (Personal Digital Assistants). In this paper we present an architecture for a sonification server that will be used in the Sound Data Mining project. In this project sound will be used to increase perception and present information extracted by spatial data mining techniques. The server is based on an audio synthesis engine and will relieve clients with little au- dio synthesis capabilities from the burden of sound processing. By providing sonification modules, this server can potentially be used on a variety of applications where sonification techniques are re- quired.","Title: Towards a Unified Service Description Language for the Internet of Services: Requirements and First Developments\nYear: 2010\n\nAbstract: Service-oriented Architectures (SOA) and Web services leverage the technical value of solutions in the areas of distributed systems and cross-enterprise integration. The emergence of Internet marketplaces for business services is driving the need to describe services, not only from a technical level, but also from a business and operational perspective. While, SOA and Web services reside in an IT layer, organizations owing Internet marketplaces are requiring advertising and trading business services which reside in a business layer. As a result, the gap between business and IT needs to be closed. This paper presents USDL (Unified Service Description Language), a specification language to describe services from a business, operational and technical perspective. USDL plays a major role in the Internet of Services to describe tradable services which are advertised in electronic marketplaces. The language has been tested using two service marketplaces as use cases.","Title: Tourism Information Aggregation Using an Ontology based approach\nYear: 2007\n\nAbstract: Aggregating related information, from different data sources, allows the creation of data repositories with very useful information. In the tourism domain, aggregating tourism products with related tourism attractions will add value to those products. The ability to create dynamic packages is another reason to aggregate tourism information. Defining an ontology, composed by the concepts to aggregate, is the first step to create tourism aggregation systems. In this paper we define the approach and the architecture that guides to the creation of aggregated solutions that provide valued tourism information and that allow the creation of dynamic packages.","Title: The Semantic Web: A mythical story or a solid reality?\nYear: 2007\n\nAbstract: The Semantic Web vision has drove hundreds of practitioners to research and develop a new bread of applications that could take the full potential of the Web to the next level. While there is a fairly clear understanding of where Web 1.0 and Web 2.0 stand now a day, the Current status and position of the Semantic Web, also known as Web 3.0, is not as clear and well defined. Therefore. in this paper we present a landscape that illustrates and captures the trends in the Semantic Web with the purpose of guiding future developments.","Title: Quality of service for workflows and web service processes\nYear: 2004\n\nAbstract: Workflow management systems (WfMSs) have been used to support various types of business processes for more than a decade now. In workflows or Web processes for e-commerce and Web service applications, suppliers and customers define a binding agreement or contract between the two parties, specifying quality of service (QoS) items such as products or services to be delivered, deadlines, quality of products, and cost of services. The management of QoS metrics directly impacts the success of organizations participating in e-commerce. Therefore, when services or products are created or managed using workflows or Web processes, the underlying workflow engine must accept the specifications and be able to estimate, monitor, and control the QoS rendered to customers. In this paper, we present a predictive QoS model that makes it possible to compute the quality of service for workflows automatically based on atomic task QoS attributes. We also present the implementation of our QoS model for the METEOR workflow system. We describe the components that have been changed or added, and discuss how they interact to enable the management of QoS.","Title: Process control-flow complexity metric: An empirical validation\nYear: 2006\n\nAbstract: Organizations are increasingly faced with the challenge of managing business processes, workflows, and, recently, Web processes. One important aspect of processes that has been overlooked is their complexity. High complexity in processes may result in bad understandability, errors, defects, and exceptions leading processes to need more time to develop, test, and maintain. Therefore, excessive complexity should be avoided. This paper describes an experiment designed to validate the Control-Flow Complexity (CFC) metric that we have proposed in our previous work. In order to demonstrate that our CFC metric serves the purpose it was defined for, we have carried out an empirical validation by means of a controlled experiment. The explanation of the steps followed to do the experiment, the results, and the conclusions obtained are the main objectives of this paper.","Title: About the Complexity of Teamwork and Collaboration Processes\nYear: 2005\n\nAbstract: Organizations across the globe are increasingly using teams to accomplish significant work and projects. Much of this work is also accomplished using technology tools to support their communication and collaborative efforts. As companies become increasingly multinational and distributed geographically, theformation of virtual teams has become a common practice. Workflow management systems are a specific type of systems that can be used to capture collaboration and group works processes and thus supports the creation of teamwork and enable collaboration. In some cases, collaboration and group work processes can become highly complex. High complexity in a process may result in bad understandability and more errors, defects, and exceptions leading processes to need more time to develop, test, and maintain. Therefore, excessive complexity should be avoided. The major goal of this paper is to discuss the need and requirementsfor the development of a measure to analyze the complexity of processes.","Title: M-GIS - Mobile and Interoperable Access to Geographic Information\nYear: 2004\n\nAbstract: This paper describes an architecture which can be used to access geographic information from mobile devices with limited display and processing characteristics, such as Personal Digital Assistants (PDA) or Mobile Phones. The information may come from different sources leading to an interoperable solution. M-GIS obtains geographic information from sources described using Geogxraphy Markup Language (GML) from one or more Web Feature Servers allowing an access independent from its format or physical location, as long as the information is according to the specification. Using XSLT, GML is transformed to a graphical format in Scalable Vector Graphics (SVG) which can then be manipulated in the mobile device. M-GIS follows a client/server architecture and the client application was developed using Java Mobile Information Device Profile technology. The results enable us to conclude that the mobile system, designed and developed with the use of open standards and representation formats to access geographic information, is a viable solution with some limitations. The main limitations of this approach have to do with the amount of information which the client can, at the moment, accommodate.","Title: Evolution and Overview of Linked USDL.\nYear: 2015\n\nAbstract: For more than 10 years, research on service descriptions has mainly studied software-based services and provided languages such as WSDL, OWL-S, WSMO for SOAP, and hREST for REST. Nonetheless, recent developments from service management (e.g., ITIL and COBIT) and cloud computing (e.g. Software-as-a-Service) have brought new requirements to service descriptions languages: the need to also model business services and account for the multi-faceted nature of services. Business-orientation, co-creation, pricing, legal aspects, and security issues are all elements which must also be part of service descriptions. While ontologies such as e(3) service and e(3) value provided a first modeling attempt to capture a business perspective, concerns on how to contract services and the agreements entailed by a contract also need to be taken into account. This has for the most part been disregarded by the e(3) family of ontologies. In this paper, we review the evolution and provide an overview of Linked USDL, a comprehensive language which provides a (multi-faceted) description to enable the commercialization of (business and technical) services over the web.","Title: Energy-Efficient Data Processing at Sweet Spot Frequencies.\nYear: 2014\n\nAbstract: The processing of Big Data often includes sorting as a basic operator. Indeed, it has been shown that many software applications spend up to 25% of their time sorting data. Moreover, for compute-bound applications, the most energy-efficient executions have shown to use a CPU speed lower than the maximum speed: the CPU sweet spot frequency. In this paper, we use these findings to run Big Data intensive applications in a more energy-efficient way. We give empirical evidence that data-intensive analytic tasks are more energy-efficient when CPU(s) operate(s) at sweet spots frequencies. Our approach uses a novel high-precision, fine-grained energy measurement infrastructure to investigate the energy (joules) consumed by different sorting algorithms. Our experiments show that algorithms can have different sweet spot frequencies for the same computational task. To leverage these findings, we describe how a new kind of self-adaptive software applications can be engineered to increase their energy-efficiency.","Title: Energy-Efficient Databases Using Sweet Spot Frequencies\nYear: 2014\n\nAbstract: Database management systems (DBMS) are typically tuned for high performance and scalability. Nevertheless, carbon footprint and energy efficiency are also becoming increasing concerns. Unfortunately, existing studies mainly present theoretical contributions but fall short on proposing practical techniques. These could be used by administrators or query optimizers to increase the energy efficiency of the DBMS. Thus, this paper explores the effect of so-called sweet spots, which are energy-efficient CPU frequencies, on the energy required to execute queries. From our findings, we derive the Sweet Spot Technique, which relies on identifying energy-efficient sweet spots and the optimal number of threads that minimizes energy consumption for a query or an entire database workload. The technique is simple and has a practical implementation leading to energy savings of up to 50% compared to using the nominal frequency and maximum number of threads.","Title: A Systematic Mapping Applied to MOOC's Study.\nYear: 2015\n\nAbstract: MOOC platforms are Web-based learning environments which allow a global participation on a large scale and with free access. The paradigm presents itself as a new teaching trend, changing the way education can be offered and funded worldwide. Many institutions are now investing in this teaching mode. However, since it is a web-based tool, the connection performance can impact both the way learning is experienced by the student as well as the operation of the platform. The âQuality of Experienceâ (QoE) concept has been widely used to refer to how users describe a service they have used while the âQuality of Serviceâ (QoS) concept deals with the technical performance parameters that are associated with the connection quality. This paper starts the process of developing a systematic mapping around MOOC platforms, and QoS and QoE concepts, aiming to provide an overview of the current state of research on these issues.","Title: Measuring Cortical Neurite-Dispersion and Perfusion in Preterm-Born Adolescents Using Multi-modal MRI\nYear: 2015\n\nAbstract: As a consequence of a global increase in rates of extremely preterm birth, predicting the long term impact of preterm birth has become an important focus of research. Cohorts of extremely preterm born subjects studied in the 1990s are now beginning to reach adulthood and the long term structural alterations of disrupted neurodevelopment in gestation can now be investigated, for instance with magnetic resonance (MR) imaging. Disruption to normal development as a result of preterm birth is likely to result in both cerebrovascular and microstructural differences compared to term-born controls. Of note, arterial spin labelled MRI provides a marker of cerebral blood flow, whilst multi-compartment diffusion models provide information on the cerebral microstructure, including that of the cortex. We apply these techniques to a cohort of 19 year-old adolescents consisting of both extremely-preterm and term-born individuals and investigate the structural and functional correlations of these MR modalities. Work of this type, revealing the long-term structural and functional differences in preterm cohorts, can help better inform on the likely outcomes of contemporary extremely preterm new-borns and provides an insight into the lifelong effects of preterm birth.","Title: Open Service Network Analysis\nYear: 2013\n\nAbstract: Understanding how services operate as part of large scale global networks, the related risks and gains of different network structures and their dynamics is becoming increasingly critical for society. Our vision and research agenda focuses on the particularly challenging task of building, analyzing, and reasoning about global service networks. This paper explains how Service Network Analysis (SNA) can be used to study and optimize the provisioning of complex services modeled as Open Semantic Service Networks (OSSN), a computer-understandable digital structure which represents connected and dependent services.","Title: On Black-Box Monitoring Techniques for Multi-Component Services\nYear: 2018\n\nAbstract: Despite the advantages of microservice and function-oriented architectures, there is an increase in complexity to monitor such highly dynamic systems. In this paper, we analyze two distinct methods to tackle the monitoring problem in a system with reduced instrumentation. Our goal is to understand the feasibility of such approach with one specific driver: simplicity. We aim to determine the extent to which it is possible to characterize the state of two generic tandem processes, using as little information as possible. To answer this question, we resorted to a simulation approach. Using a queue system, we simulated two services, that we could manipulate with distinct operation sets for each module. We used the total response time seen upstream of the system. Having this setup and metric, we applied two distinct methods to analyze the results. First, we used supervised machine learning algorithms to identify where the bottleneck is happening. Secondly, we used an exponential decomposition to identify the occupation in the two components in a more black-box fashion. Results show that both methodologies have their advantages and limitations. The separation of the signal more accurately identifies occupation in low occupied resources, but when a service is totally dominating the overall time, it lacks precision. The machine learning has a more stable error, but needs the training set. This study suggest that a black-box occupation approach with both techniques is possible and very useful.","Title: Response Time Characterization of Microservice-Based Systems\nYear: 2018\n\nAbstract: In pursuit of faster development cycles, companies have favored small decoupled services over monoliths. Following this trend, distributed systems made of microservices have grown in scale and complexity, giving rise to a new set of operational problems. Even though this paradigm simplifies development, deployment, management of individual services, it hinders system observability. In particular, performance monitoring and analysis becomes more challenging, especially for critical production systems that have grown organically, operate continuously, and cannot afford the availability cost of online benchmarking. Additionally, these systems are often very large and expensive, thus being bad candidates for full-scale development replicas. Creating models of services and systems for characterization and formal analysis can alleviate the aforementioned issues. Since performance, namely response time, is the main interest of this work, we focused on bottleneck detection and optimal resource scheduling. We propose a method for modeling production services as queuing systems from request traces. Additionally, we provide analytical tools for response time characterization and optimal resource allocation. Our results show that a simple queuing system with a single queue and multiple homogeneous servers has a small parameter space that can be estimated in production. The resulting model can be used to accurately predict response time distribution and the necessary number of instances to maintain a desired service level, under a given load.","Title: Efficient Failure Diagnosis of OpenStack Using Tempest.\nYear: 2018\n\nAbstract: While cloud computing continues to be popular in the IT world, companies offering cloud solutions are under pressure to provide the most reliable solutions to users. In this article, we describe an innovative approach to diagnose service failures in an OpenStack-based cloud using Tempest as a starting point.","Title: Service Engineering for the Internet of Services\nYear: 2008\n\nAbstract: The Internet and the Web have extended traditional business networks by allowing a Web of different digital resources to work together to create value for organizations. The most industrialized countries have entered a post-industrial era where their prosperity is largely created through a service economy. There is a clear transition from a manufacturing based economy to a service based economy. From the technological perspective, the development of Web-based infrastructures to Support and deliver services in this new economy raises a number of challenges. From a business perspective, there is the need to understand how value is created through set-vices. In this paper, we describe how we propose to address these two perspectives and realize the vision of the Internet of Services (IoS), where Web-based IT-supported service ecosystems form the base of service business value networks. This paper addresses the main challenging issues that need to be explored to provide an integrated technical and business infrastructure for the Internet of Service.","Title: COOPIS'10: PC co-chairs message\nYear: 2010\n\nAbstract: Welcome to the 18th International Conference on Cooperative Information Systems (CoopIS 2010). This year CoopIS was held in\n Crete, Greece, during October 27-29, 2010.\n \n \n Cooperative Information Systems (CIS) provide enterprises and communities of users with flexible, scalable and intelligent\n services in large-scale networking environments. Building a scalable cooperative information system requires technical breakthroughs\n to overcome tough challenges that traditional rigid distributed systems did not face.","Title: Anomaly Detection and Classification using Distributed Tracing and Deep Learning\nYear: 2019\n\nAbstract: Artificial Intelligence for IT Operations (AIOps) combines big data and machine learning to replace a broad range of IT Operations tasks including availability, performance, and monitoring of services. By exploiting log, tracing, metric, and network data, AIOps enable detection of faults and issues of services. The focus of this work is on detecting anomalies based on distributed tracing records that contain detailed information for the availability and the response time of the services. In large-scale distributed systems, where a service is deployed on heterogeneous hardware and has multiple scenarios of normal operation, it becomes challenging to detect such anomalous cases. We address the problem by proposing unsupervised, response time anomaly detection based on deep learning data modeling techniques; unsupervised dynamic error threshold approach; tolerance module for false positive reduction; and descriptive classification of the anomalies. The evaluation shows that the approach achieves high accuracy and solid performance in both, experimental testbed and large-scale production cloud.","Title: IAD: Indirect Anomalous VMMs Detection in the Cloud-Based Environment\nYear: 2021\n\nAbstract: Server virtualization in the form of virtual machines (VMs) with the use of a hypervisor or a Virtual Machine Monitor (VMM) is an essential part of cloud computing technology to provide infrastructure-as-a-service (IaaS). A fault or an anomaly in the VMM can propagate to the VMs hosted on it and ultimately affect the availability and reliability of the applications running on those VMs. Therefore, identifying and eventually resolving it quickly is highly important. However, anomalous VMM detection is a challenge in the cloud environment since the user does not have access to the VMM. This paper addresses this challenge of anomalous VMM detection in the cloud-based environment without having any knowledge or data from VMM by introducing a novel machine learning-based algorithm called IAD: Indirect Anomalous VMMs Detection. This algorithm solely uses the VM's resources utilization data hosted on those VMMs for the anomalous VMMs detection. The developed algorithm's accuracy was tested on four datasets comprising the synthetic and real and compared against four other popular algorithms, which can also be used to the described problem. It was found that the proposed IAD algorithm has an average F1-score of 83.7% averaged across four datasets, and also outperforms other algorithms by an average F1-score of 11%.","Title: QuLog: Data-Driven Approach for Log Instruction Quality Assessment\nYear: 2022\n\nAbstract: In the current IT world, developers write code while system operators run the code mostly as a black box. The connection between both worlds is typically established with log messages: the developer provides hints to the (unknown) operator, where the cause of an occurred issue is, and vice versa, the operator can report bugs during operation. To fulfil this purpose, developers write log instructions that are structured text commonly composed of a log level (e.g., “info”, “error”), static text (“IP {} cannot be reached”), and dynamic variables (e.g. IP {}). However, opposed to well-adopted coding practices, there are no widely adopted guidelines on how to write log instructions with good quality properties. For example, a developer may assign a high log level (e.g., “error”) for a trivial event that can confuse the operator and increase maintenance costs. Or the static text can be insufficient to hint at a specific issue. In this paper, we address the problem of log quality assessment and provide the first step towards its automation. We start with an in-depth analysis of quality log instruction properties in nine software systems and identify two quality properties: 1) correct log level assignment assessing the correctness of the log level, and 2) sufficient linguistic structure assessing the minimal richness of the static text necessary for verbose event description. Based on these findings, we developed a data-driven approach that adapts deep learning methods for each of the two properties. An extensive evaluation on large-scale open-source systems shows that our approach correctly assesses log level assignments with an accuracy of 0.88, and the sufficient linguistic structure with an F1 score of 0.99, outperforming the baselines. Our study highlights the potential of the data-driven methods in assessing log instructions quality.","Title: Failure Identification from Unstable Log Data using Deep Learning\nYear: 2022\n\nAbstract: The reliability of cloud platforms is of significant relevance because society increasingly relies on complex software systems running on the cloud. To improve it, cloud providers are automating various maintenance tasks, with failure identification frequently being considered. The precondition for automation is the availability of observability tools, with system logs commonly being used. The focus of this paper is log-based failure identification. This problem is challenging because of the instability of the log data and the incompleteness of the explicit logging failure coverage within the code. To address the two challenges, we present CLog as a method for failure identification. The key idea presented herein based is on our observation that by representing the log data as sequences of subprocesses instead of sequences of log events, the effect of the unstable log data is reduced. CLog introduces a novel subprocess extraction method that uses context-aware neural network and clustering methods to extract meaningful subprocesses. The direct modeling of log event contexts allows the identification of failures with respect to the abrupt context changes, addressing the challenge of insufficient logging failure coverage. Our experimental results demonstrate that the learned subprocesses representations reduce the instability in the input, allowing CLog to outperform the baselines on the failure identification subproblems - 1) failure detection by 9–24 % on F1 score and 2) failure type identification by 7% on the macro averaged F1 score. Further analysis shows the existent negative correlation between the instability in the input event sequences and the detection performance in a model-agnostic manner.","Title: Leveraging Log Instructions in Log-based Anomaly Detection\nYear: 2022\n\nAbstract: Artificial Intelligence for IT Operations (AIOps) describes the process of maintaining and operating large IT systems using diverse AI-enabled methods and tools for, e.g., anomaly detection and root cause analysis, to support the remediation, optimization, and automatic initiation of self-stabilizing IT activities. The core step of any AIOps workflow is anomaly detection, typically performed on high-volume heterogeneous data such as log messages (logs), metrics (e.g., CPU utilization), and distributed traces. In this paper, we propose a method for reliable and practical anomaly detection from system logs. It overcomes the common disadvantage of related works, i.e., the need for a large amount of manually labeled training data, by building an anomaly detection model with log instructions from the source code of 1000+ GitHub projects. The instructions from diverse systems contain rich and heterogenous information about many different normal and abnormal IT events and serve as a foundation for anomaly detection. The proposed method, named ADLILog, combines the log instructions and the data from the system of interest (target system) to learn a deep neural network model through a two-phase learning procedure. The experimental results show that ADLILog outperforms the related approaches by up to 60% on the F\n<inf xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">1</inf>\n score while satisfying core non-functional requirements for industrial deployments such as unsupervised design, efficient model updates, and small model sizes.","Title: Morphology-Preserving Autoregressive 3D Generative Modelling of the Brain\nYear: 2022\n\nAbstract: Human anatomy, morphology, and associated diseases can be studied using medical imaging data. However, access to medical imaging data is restricted by governance and privacy concerns, data ownership, and the cost of acquisition, thus limiting our ability to understand the human body. A possible solution to this issue is the creation of a model able to learn and then generate synthetic images of the human body conditioned on specific characteristics of relevance (e.g., age, sex, and disease status). Deep generative models, in the form of neural networks, have been recently used to create synthetic 2D images of natural scenes. Still, the ability to produce high-resolution 3D volumetric imaging data with correct anatomical morphology has been hampered by data scarcity and algorithmic and computational limitations. This work proposes a generative model that can be scaled to produce anatomically correct, high-resolution, and realistic images of the human brain, with the necessary quality to allow further downstream analyses. The ability to generate a potentially unlimited amount of data not only enables large-scale studies of human anatomy and pathology without jeopardizing patient privacy, but also significantly advances research in the field of anomaly detection, modality synthesis, learning under limited data, and fair and ethical Al. Code and trained models are available at: https://github.com/AmigoLab/SynthAnatomy."],"string":"[\n \"Title: Integrating Business Process And User Interface Models Using A Model-Driven Approach\\nYear: 2009\\n\\nAbstract: Business services are complex entities that encompass descriptions about different aspects including business processes and user interfaces. Topically, the modeling of business services results on several correlated models. On the one hand, there is the need to keep these models apart in order to attain the levels of abstractions needed to model different aspects of a service. On the other hand, it is important to maintain the consistency and integrity of models. In this paper, we show how to maintain the consistency and integrity of models and present a use case for the integration of user interface design and business process models.\",\n \"Title: Business Process Control-Flow Complexity: Metric, Evaluation, And Validation\\nYear: 2008\\n\\nAbstract: Organizations are increasingly faced with the challenge of managing business processes, workflows, and recently, Web processes. One important aspect of business processes that has been overlooked is their complexity. High complexity in processes may result in poor understandability, errors, defects, and exceptions, leading processes to need more time to develop, test, and maintain. Therefore, excessive complexity should be avoided Business process measurement is the task of empirically and objectively assigning numbers to the properties of business processes in such a way so as to describe them. Desirable attributes to study and measure include complexity, cost, maintainability, and reliability. In our work, we will focus on investigating process complexity. We present and describe a metric to analyze the control-flow complexity of business processes. The metric is evaluated in terms of Weyuker's properties in order to guarantee that it qualifies as good and comprehensive. To test the validity of the metric, we describe the experiment we have carried out for empirically validating the metric.\",\n \"Title: Workflow Quality of Service\\nYear: 2002\\n\\nAbstract: Workflow management systems (WfMSs) have been used to support various types of business processes for more than a decade now. In e- commerce processes, suppliers and customers define a binding agreement or contract between the two parties, specifying quality of service (QoS) items such as products or services to be delivered, deadlines, quality of products, and cost of service. Management of such QoS directly impacts success of organizations participating in e-commerce. Organizations operating in modern markets require an excellent degree of quality of service management. Products and services must be available to customers with well-defined specifications. A good management of quality leads to the creation of quality products and services, which in turn fulfills customer expectations and achieves customer satisfaction. Therefore, when services or products are created or managed using workflow processes, the underlying WfMS must accept the specification, be able to predict, monitor, and control the QoS rendered to customers. To achieve these objectives the first step is to develop an adequate QoS model for workflow processes and develop methods to compute QoS.\",\n \"Title: The Internet Of Services\\nYear: 2009\\n\\nAbstract: The vision of the Internet of Services (IoS) can be seen as a new business model that can radically change the way we discover, select, invoke and interact with services. In the same way that the Internet can be viewed as a Web of Information, it can also be configured to support an Internet of Services. The IoS describes an approach that uses the Internet as a medium for selling and purchasing services. As a result, services become tradable entities. Service marketplaces, a logical location where providers and consumers are brought together to trade services and to engage in business interaction, are enabling the creation of platforms for the IoS vision. Thus, the IoS provides the business and technical base for advanced business models where service providers and consumers form business networks for service provision and consumption.\",\n \"Title: Semantic E-Workflow Composition\\nYear: 2003\\n\\nAbstract: Systems and infrastructures are currently being developed to support Web services. The main idea is to encapsulate an organization's functionality within an appropriate interface and advertise it as Web services. While in some cases Web services may be utilized in an isolated form, it is normal to expect Web services to be integrated as part of workflow processes. The composition of workflow processes that model e-service applications differs from the design of traditional workflows, in terms of the number of tasks (Web services) available to the composition process, in their heterogeneity, and in their autonomy. Therefore, two problems need to be solved: how to efficiently discover Web services—based on functional and operational requirements—and how to facilitate the interoperability of heterogeneous Web services. In this paper, we present a solution within the context of the emerging Semantic Web that includes use of ontologies to overcome some of the problem. We describe a prototype that has been implemented to illustrate how discovery and interoperability functions are achieved more efficiently.\",\n \"Title: PuReWidgets: a programming toolkit for interactive public display applications\\nYear: 2012\\n\\nAbstract: Interaction is repeatedly pointed out as a key enabling element towards more engaging and valuable public displays. Still, most digital public displays today do not support any interactive features. We argue that this is mainly due to the lack of efficient and clear abstractions that developers can use to incorporate interactivity into their applications. As a consequence, interaction represents a major overhead for developers, and users are faced with inconsistent interaction models across different displays. This paper describes the results of a study on interaction widgets for generalized interaction with public displays. We present PuReWidgets, a toolkit that supports multiple interaction mechanisms, automatically generated graphical interfaces, asynchronous events and concurrent interaction. This is an early effort towards the creation of a programming toolkit that developers can incorporate into their public display applications to support the interaction process across multiple display systems without considering the specifics of what interaction modality will be used on each particular display.\",\n \"Title: Open Semantic Service Networks: Modeling and Analysis.\\nYear: 2013\\n\\nAbstract: A new interesting research area is the representation and analysis of the networked economy using Open Semantic Service Networks (OSSN). OSSN are represented using the service description language USDL to model nodes and using the service relationship model OSSR to model edges. Nonetheless, in their current form USDL and OSSR do not provide constructs to capture the dynamic behavior of service networks. To bridge this gap, we used the General System Theory (GST) as a framework guiding the extension of USDL and OSSR to model dynamic OSSN. We evaluated the extensions made by applying USDL and OSSR to two distinct types of dynamic OSSN analysis: 1) evolutionary by using a Preferential Attachment (PA) and 2) analytical by using concepts from System Dynamics (SD). Results indicate that OSSN can constitute the first stepping stones toward the analysis of global service-based economies.\",\n \"Title: Legosc - Mindstorms Nxt Robotics Programming For Artists\\nYear: 2008\\n\\nAbstract: Robotics is an interesting but difficult area for digital artists who generally don't have much academic background on electronics or computer programming. Digital art students normally use high-level application to program their visual and sonorous installations. This paper presents LegOSC - a tool that allows the control of the Mindstorms NXT robots from any application that uses the Open Sound Control protocol which is implemented by most of those high-level applications. This allows artists to create works which incorporate robotic parts using the familiar programming environment.\",\n \"Title: Using semantic web and service oriented technologies to build loosely coupled systems: SWOAT - a service and semantic web oriented architecture technology\\nYear: 2007\\n\\nAbstract: The creation of loosely coupled and flexible applications has been a challenge faced by most organizations. This has been important because organization systems need to quickly respond and adapt to changes that occur in the business environment. In order to address these key issues, we implemented SWOAT, a 'Service and Semantic Web Oriented Architecture Technology' based middleware. Our system uses ontologies to semantically describe and formalize the information model of the organization, providing a global and integrated view over a set of database systems. It also allows interoperability with several systems using Web Services. Using ontologies and Web services, clients remain loosely coupled from data sources. As a result, data structures can be changed and moved without having to change all clients, internal or external to the organization.\",\n \"Title: Poseidon: A Framework To Assist Web Process Design Based On Business Cases\\nYear: 2006\\n\\nAbstract: Systems and infrastructures are currently being developed to support Web services and Web processes. One prominent solution to manage and coordinate Web services is the use of workflow technology. For more than two decades now, workflow management systems architectures, language specifications, and workflow analysis techniques have been extensively studied. While these areas of research have made the development of sophisticated workflow systems possible, one important research area that has been overlooked is the lifecycle of process application development. As a result, current process management systems support the analysis, enactment, and ad hoc design of workflows, but they lack the tools and methods to assist process design. The purpose of our study is to present a framework to assist and guide process analysts and designers in their task. The Poseidon framework includes a participative and an analytical design method. The participative phase uses a clean sheet approach and starts by constructing a business case table which captures all the business cases represented by a process. Afterward, an analytical design phase is followed where scheduling functions are derived from the business case table and are used to build the structure of a process.\",\n \"Title: Academic and industrial research: do their approaches differ in adding semantics to web services?\\nYear: 2004\\n\\nAbstract: Since the new terms, “Semantic Web” and “Web services”, have been introduced, researchers have followed two different roads. Following one road, academia has focused on developing a new set of languages to enable the automation of Web services execution and integration based on the Semantic Web. On the other road, industry has taken the lead to propose and develop technologies and infrastructures to support Web services and Web processes without, until recently, paying much attention to semantics. It is fundamental to analyze the trend that is being followed with regard to the “Semantic Web” and “Web services”. Therefore, two important questions need to be answered: “do the approaches taken by academia and industry differ in how they add semantics to Web services?” and “are their efforts converging or diverging?” This paper, based on a panel discussion at an international conference on Web services, which consisted of members of both academia and industry, addresses precisely these two questions.\",\n \"Title: Evaluating the Process Control-Flow Complexity Measure\\nYear: 2005\\n\\nAbstract: Process measurement is the task of empirically and objectivelyassigning numbers to the attributes of processes in such a wayas to describe them. We define process complexity as the degreeto which a process is difficult to analyze, understand or explain.One way to analyze a processý complexity is to use a process control-flow complexity measure. This measure analyzes the control- flow of processes and can be applied to both Web processes andworkflows. In this paper, we discuss how to evaluate the control- flow complexity measure to ensure that it can be qualify as a good and comprehensive one.\",\n \"Title: Moving from syntactic to semantic organizations using JXML2OWL\\nYear: 2008\\n\\nAbstract: Today's enterprises face critical needs in integrating disparate information spread over several data sources inside and even outside the organization. Most organizations already rely on XML standard to define their data models. Unfortunately, even when using XML to represent data, problems arise when it is necessary to integrate different data sources. Emerging Semantic Web technologies, such as ontologies, RDF, RDFS, and OWL, can play an important role in the semantic definition and integration of data. The purpose of our study is to present a framework to assist organizations to move from a syntactic data infrastructure defined in XML to a semantic data infrastructure using OWL. The framework supports mappings and fully automated instance transformation from syntactic data sources in XML format to a common shared global model defined by an ontology using Semantic Web technologies. The presented framework, JXML2OWL, allows organizations to automatically convert their XML data sources to a semantic model defined in OWL.\",\n \"Title: Integrating HAD Organizational Data Assets using Semantic Web Technologies\\nYear: 2007\\n\\nAbstract: To create value based on information and knowledge, organizations need to recognize that they are composed of various types of data assets. In this context, data integration has been proven to be a challenge due to the heterogeneity of the information systems involved. The Extensible Markup Language (XML) was one of most successful solutions developed to provide business-to-business integration. However, XML \\\"lacks semantics\\\". Thus problems arise when it is necessary to manipulate and integrate different XML data sources. Consequently, today's organizations are again shifting from a syntactic interoperability level to a semantic one. This paper presents a Semantic data Integration Middleware (SIM) which, when based on a single query, integrates data residing in different data sources. The middleware uses an ontology-based multi-source data extractor/wrapper approach to transform data into semantic knowledge.\",\n \"Title: IntelliGEN: A Distributed Workflow System for Discovering Protein-Protein Interactions\\nYear: 2003\\n\\nAbstract: A large genomics project involves a significant number of researchers and technicians performing dozens of tasks, either manual (e.g. performing laboratory experiments), computer assisted (e.g. looking for genes in the GENBANK database), or sometimes performed entirely automatically by the computer (e.g. sequence assembly). It has become apparent that managing such projects poses overwhelming problems and may lead to results of lower or even unacceptable quality, or possibly drastically increased project costs. In this paper, we present a design and an initial implementation of a distributed workflow system created to schedule and support activities in a genomics laboratory. The focus of the activities in the laboratory is the discovery of protein-protein interactions of fungi, specifically Neurospora crassa. We present our approach of developing, adapting and applying workflow technology in the genomics lab and illustrate it using one distinct part of a larger workflow to discover protein- protein interactions. Novel features of our system include the ability to monitor the quality and timeliness of the results and if necessary, suggesting and incorporating changes to the selected tasks and their scheduling. In the new millennium biology has gone through a paradigm shift with the advent of Genomics, the study of the structure, function, and evolution of whole genomes. The term \\\"genomics\\\" was coined in 1986, and institutional support for this new discipline began with the creation of the National Center for Human Genome Research (NCHGR) in 1989. The initial goal of this discipline was the determination of the entire DNA sequence of a human being (a human genome) and several related model organisms that have played a central role in Genetics, including the bacterium Escherichia coli, the yeast Saccharomyces cerevisiae, the worm Canerohabditis elegans, and the fruit fly Drosophila melanogaster. Sequencing the human genome and those of each model system were daunting distributed computing tasks which required collecting, storing, integrating, retrieving, and distributing 3 billion base pairs (bp) of sequence on the human genome alone. Many aspects of computer science were brought to bear to solve these problems (43, 81). What genomics promises is that with the availability of the complete genetic blueprints of\",\n \"Title: Modeling Service Relationships for Service Networks.\\nYear: 2013\\n\\nAbstract: The last decade has seen an increased interest in the study of networks in many fields of science. Examples are numerous, from sociology to biology, and to physical systems such as power grids. Nonetheless, the field of service networks has received less attention. Previous research has mainly tackled the modeling of single service systems and service compositions, often focusing only on studying temporal relationships between services. The objective of this paper is to propose a computational model to represent the various types of relationships which can be established between services systems to model service networks. This work acquires a particular importance since the study of service networks can bring new scientific discoveries on how service-based economies operate at a global scale.\",\n \"Title: Discovering Semantic Web Services with and without a Common Ontology Commitment\\nYear: 2006\\n\\nAbstract: This paper presents an algorithm to match a semantic Web service request against semantic Web service advertisements. The algorithm is to be used by systems to discover semantic Web services, such as the UDDI. Matching is based on the assessment of the similarity among semantic Web service properties, such as inputs and outputs. Semantic Web services have their inputs and outputs annotated or described by ontological concepts. The algorithm is able to match a semantic Web service request against advertisements that are annotated with concepts with and without a common ontology commitment. The similarity of inputs and outputs is evaluated based on concepts (classes), their semantic relations, and their common and distinguishing features (properties\",\n \"Title: A weighted coupling metric for business process models\\nYear: 2007\\n\\nAbstract: Various efforts recently aimed at the development of quality metrics for process models. In this paper, we propose a new notion of coupling, which has been used successfully in software engineering for many years. It extends other work by specifically incorporating the effects of different types of connectors used on a process model's coupling level.\",\n \"Title: Semantic integration of web services and peertopeer networks to achieve fault-tolerance.\\nYear: 2006\\n\\nAbstract: One fundamental property that critical Web services need to provide is a high level of availability. Along with the development of Web services, considerable technological advances are being made to use the semantic Web to achieve the automated processing and integration of data and applications. This paper describes the implementation of the Whisper architecture. This architecture semantically integrates Web services with a peer-to-peer infrastructure to increase service availability. Whisper achieves transparent fault-tolerance by automatically forwarding Web service requests to semantically equivalent peers that are dynamically located, selected, and invoked.\",\n \"Title: Healthcare enterprise process development and integration\\nYear: 2003\\n\\nAbstract: Healthcare enterprises involve complex processes that span diverse groups and organisations. These processes involve clinical and administrative tasks, large volumes of data, and large numbers of patients and personnel. The tasks can be performed either by humans or by automated systems. In the latter case, the tasks are supported by a variety of software applications and information systems which are very often heterogeneous, autonomous, and distributed. The development of systems to manage and automate these processes has increasingly played an important role in improving the efficiency of healthcare enterprises. In this paper we look at four healthcare and medical applications that involve investigative, clinical, and administrative functions. Based on these applications, we derive the requirements for developing enterprise applications that involve the coordination of a variety of tasks performed by humans, information systems, and legacy applications.\",\n \"Title: Mapping between heterogeneous XML and OWL transaction representations in B2B integration\\nYear: 2011\\n\\nAbstract: XML-based standards have been widely used to enable and ease Business-to-Business (B2B) integration. Examples of standards include cXML, CIDX and ebXML. While these XML-based standards are syntactic, contemporary organizations have available new means to structure their internal data representations using semantic descriptions, such as RDF(S) and OWL. This scenario poses an interesting challenge: ''How to reconcile external XML-based standards and internal OWL-based representations in B2B integration scenarios?'' In this paper, we present a conceptual approach, and its implementation, to integrate external syntactic data representations with organizational internal semantic data representations by using the notion of heterogeneous mappings which are established between the two types of representations. The application developed, B2BISS, enables an effective management of mappings. As the number of mappings stored in the repository increases over time, organizations can gradually rely on a semi-automatic to automatic B2B integration.\",\n \"Title: Challenges of business service monitoring in the internet of services\\nYear: 2008\\n\\nAbstract: The most industrialized countries have entered an era where their prosperity is largely created through a service economy. There is a transition from a manufacturing based economy to a service based economy. In the Internet of Services (IoS) vision services are seen as tradable goods. Business services are a major asset in this context. Since they are inherently different from e.g. web services, there is a need for a specialized monitoring approach. In this paper we describe the differences between e-services and business services, present an approach to describe business services, and discuss challenges regarding their monitoring requirements.\",\n \"Title: Analysis and Validation of Control-Flow Complexity Measures with BPMN Process Models\\nYear: 2009\\n\\nAbstract: Evaluating the complexity of business processes during the early stages of their development, primarily during the process modelling phase, provides organizations and stakeholders with process models which are easier to understand and easier to maintain. This presents advantages when carrying out evolution tasks in process models - key activities, given the current competitive market. In this work, we present the use and validation of the CFC metric to evaluate the complexity of business processes modelled with BPMN. The complexity of processes is evaluated from a control-flow perspective. An empirical evaluation has been carried out in order to demonstrate that the CFC metric can be useful when applied to BPMN models, providing information about their ease of maintenance.\",\n \"Title: Business process quality metrics: log-based complexity of workflow patterns\\nYear: 2007\\n\\nAbstract: We believe that analysis tools for BPM should provide other analytical capabilities besides verification. Namely, they should\\n provide mechanisms to analyze the complexity of workflows. High complexity in workflows may result in poor understandability,\\n errors, defects, and exceptions leading processes to need more time to develop, test, and maintain. Therefore, excessive complexity\\n should be avoided. The major goal of this paper is to describe a quality metric to analyze the complexity of workflow patterns\\n from a log-based perspective.\",\n \"Title: Benchmarking a Semantic Web Service Architecture for Fault-tolerant B2B Integration\\nYear: 2006\\n\\nAbstract: With the development and maturity of Service- Oriented Architectures (SOA) to support business-tobusiness transactions, organizations are implementing Web services to expose their public functionalities associated with internal systems and business processes. In many business processes, Web services need to provide is a high level of availability, since the globalization of the Internet enables business partners to easily switch to other competitors when services are not available. Along with the development of SOA, considerable technological advances are being made to use the semantic Web to achieve the automated processing and integration of data and applications. This paper describes the implementation and benchmarking of an architecture that semantically integrates Web services with a peer-to-peer infrastructure to increase service availability through fault-tolerance.\",\n \"Title: Composing e-Workflows using Syntactic, Operational and Semantic Information\\nYear: 2002\\n\\nAbstract: With the development and maturity of infrastructures that support e-services, we expect organizations to incorporate Web services as part of their business processes. While in some cases Web services may be utilized in an isolated form, it is natural to expect that Web services will be integrated as part of workflows. The composition of workflows to model e-service applications differs from the design of traditional workflows due to the number of Web services available during the composition process, their heterogeneity, and their autonomy. The composition process raises two main problems that need to be solved: how to efficiently discover Web services and how to facilitate their integration with the workflow host. To support, facilitate, and assist the composition of process, we present a solution based on ontologies. We have developed an algorithm and prototype that workflow systems and discovery mechanisms can use to find Web services with desired interfaces, operational metrics, and to assist designers in resolving heterogeneity issues among Web services. Our approach provides an important solution to enhance Web service discovery and integration.\",\n \"Title: Measuring the Compliance of Processes with Reference Models\\nYear: 2009\\n\\nAbstract: Reference models provide a set of generally accepted best practices to create efficient processes to be deployed inside organizations. However, a central challenge is to determine how these best practices are implemented in practice. One limitation of existing approaches for measuring compliance is the assumption that the compliance can be determined using the notion of process equivalence. Nonetheless, the use of equivalence algorithms is not adequate since two models can have different structures but one process can still be compliant with the other. This paper presents a new approach and algorithm which allow to measure the compliance of process models with reference models. We evaluate our approach by measuring the compliance of a model currently used by a German passenger airline with the IT Infrastructure Library (ITIL) reference model and by comparing our results with existing approaches.\",\n \"Title: Semantic Web Processes: Semantics Enabled Annotation, Discovery, Composition and Orchestration of Web Scale Processes\\nYear: 2003\\n\\nAbstract: This tutorial deals with the evolution of inter- Enterprise and Web scale process to support e-commerce and e-services. It taps into the promises of two of the hottest R&D and technology areas: Web services and the Semantic Web. It presents how applying semantics to each of the steps in the Semantic Web Process lifecycle can help address critical issues in reuse, integration and scalability.\",\n \"Title: Mobile applications for open display networks: common design considerations\\nYear: 2013\\n\\nAbstract: Mobile devices can be a powerful tool for interaction with public displays, but mobile applications supporting this form of interaction are not yet part of our everyday reality. There are no widely accepted abstractions, standards, or practices that may enable systematic interaction between mobile devices and public displays. We envision public displays to move away from a world of closed display networks to scenarios where mobile applications could allow people to interact with the myriad of displays they might encounter during their everyday trips. In this research, we study the key processes involved in this collaborative interaction between public shared displays and mobile applications. Based on the lessons learned from our own development and deployment of 3 applications, and also on the analysis of the interactive features described in the literature, we have identified 8 key processes that may shape this form of interaction: Discovery, Association, Presence Management, Exploration, Interface Migration, Controller, Media Upload and Media Download. The contribution of this work is the identification of these high-level processes and an elicitation of the main design considerations for display networks.\",\n \"Title: Inter-enterprise System and Application Integration: A Reality Check\\nYear: 2007\\n\\nAbstract: This paper structures the summary of the panel held at the 9th International Conference on Enterprise Information Systems, Funchal, Madeira, 12-16 June 2007 that addressed the following question: \\\"Are you still working on Inter-Enterprise System and Application Integration?\\\" The panel aggregated distinguished experts from the areas of process management, workflow, Web services, SOA, and Semantic Web.\",\n \"Title: XML-based modeling and simulation: web service technologies and their synergy with simulation\\nYear: 2002\\n\\nAbstract: The World Wide Web has had an huge influence on the computing field in general as well as simulation in particular (<i>e.g.</i>, Web-Based Simulation). A new wave of development based upon XML has started. Two of the most interesting aspects of this development are the Semantic Web and Web Services. This paper examines the synergy between Web service technology and simulation. In one direction, Web service processes can be simulated for the purpose of correcting/improving the design. In the other direction, simulation models/components can be built out of Web services. Work on seamlessly using simulation as a part of Web service composition and process design, as well as on using Web services to re-build the JSIM Web-based simulation environment is highlighted.\",\n \"Title: Ontological Representation and Governance of Business Semantics in Compliant Service Networks.\\nYear: 2013\\n\\nAbstract: The Internet would enable new ways for service innovation and trading, as well as for analysing the resulting value networks, with an unprecedented level of scale and dynamics. Yet most related economic activities remain of a largely brittle and manual nature. Service-oriented business implementations focus on operational aspects at the cost of value creation aspects such as quality and regulatory compliance. Indeed they enforce how to carry out a certain business in a prefixed non-adaptive manner rather than capturing the semantics of a business domain in a way that would enable service systems to adapt their role in changing value propositions. In this paper we set requirements for SDL-compliant business service semantics, and propose a method for their ontological representation and governance. We demonstrate an implementation of our approach in the context of service-oriented Information Governance.\",\n \"Title: Distributed Contracting and Monitoring in the Internet of Services\\nYear: 2009\\n\\nAbstract: The recent approval of the EU Services Directive is fostering the Internet of Services (IoS) and will promote the emergence of marketplaces for business and real-world services. From a research perspective, the IoS will require a new bread of technological infrastructures to support the concepts of business service description, contract management from various perspectives, end-to-end marketplaces, and business monitoring. The IoS is a vision referring to web service-based digital societies. When service hosting moves from best-effort provisioning to guaranteed service delivery, monitoring becomes a crucial point of proof for providers and consumers of such services. We present the uplifting of technical contract monitoring results to business effects based on the distributed service infrastructure developed in project THESEUS, use case TEXO.\",\n \"Title: Cloud computing automation: integrating USDL and TOSCA\\nYear: 2013\\n\\nAbstract: Standardization efforts to simplify the management of cloud applications are being conducted in isolation. The objective of this paper is to investigate to which extend two promising specifications, USDL and TOSCA, can be integrated to automate the lifecycle of cloud applications. In our approach, we selected a commercial SaaS CRM platform, modeled it using the service description language USDL, modeled its cloud deployment using TOSCA, and constructed a prototypical platform to integrate service selection with deployment. Our evaluation indicates that a high level of integration is possible. We were able to fully automatize the remote deployment of a cloud service after it was selected by a customer in a marketplace. Architectural decisions emerged during the construction of the platform and were related to global service identification and access, multi-layer routing, and dynamic binding.\",\n \"Title: Service Engineering for Future Business Value Networks\\nYear: 2008\\n\\nAbstract: Traditionally, business value networks have orchestrated human and technical resources that worked together to form relationships and to add value to a product or service. The Internet and the Web have extended traditional business networks by allowing also a web of different digital resources to work together to create value. Additionally, the increasing adoption of service-oriented architectures has allowed the creation of service ecosystems in which Web services are exposed and connected. The TEXO project proposes to combine these two trends to create what is called Future Business Value Networks which enable IT-supported value networks via service ecosystems. This paper addresses the main challenging issues that need to be explored to provide an integrated technical infrastructure to support this emerging new type of business networks.\",\n \"Title: Web Services Discovery Utilizing Semantically Annotated WSDL\\nYear: 2008\\n\\nAbstract: To make semantic Web services accessible to users, providers use registries to publish them. Unfortunately, the current registries use discovery mechanisms which are inefficient, as they do not support discovery based on the semantics of the services and thus lead to a considerable number of irrelevant matches. Semantic discovery and matching of services is a promising approach to address this challenge. This paper presents an algorithm to match a semantic Web service request described with SAWSDL against semantic Web service advertisements. The algorithm is novel in three fundamental aspects. First, the similarity among semantic Web service properties, such as inputs and outputs, is evaluated using Tversky's model which is based on concepts (classes), their semantic relationships, and their common and distinguishing features (properties). Second, the algorithm, not only takes into account services' inputs and outputs, but it also considers the functionality of services. Finally, the algorithm is able to match a semantic Web service request against advertisements that are annotated with concepts that are with or without a common ontological commitment. In other words, it can evaluate the similarity of concepts defined in the context of different ontologies.\",\n \"Title: Open service networks: research directions\\nYear: 2013\\n\\nAbstract: Service networks are a new research field which promises to increase our knowledge of service-based systems. Understanding what factors explain the evolution of global service networks can lead to more efficient and balanced infrastructures. The challenge consists in developing a novel perspective by connecting models representing services (e.g. cloud computing and governmental services) into networks. Theories, methods, and applications to be developed will enable, afterwards, to understand, describe, explain, analyse, predict, and control the structure and behaviour of global service networks.\",\n \"Title: A Framework for Assessing Strategies and Technologies for Dynamic Packaging Applications in E-Tourism.\\nYear: 2007\\n\\nAbstract: Dynamic packaging has been introduced as an innovative technology allowing for the automated online configuration and assembling of packaged travel products for individual customers. While dynamic packaging has been widely accepted by customers in UK and US markets, the possible strategic impacts enabled by dynamic packaging technologies are still uncertain. This article pro- vides a study of the strategic opportunities enabled by dynamic packaging, highlighting the key success factors. Current dynamic packaging applications of the three major online travel agencies are evaluated on the basis of the developed analysis framework. The results show that an appro- priate level of integration of tourism information systems is a key factor for further realizing the strategic opportunities of dynamic packaging. We then discuss Web services and semantics as a possible solution for overcoming the interoperability problems that (current) dynamic packaging applications face.\",\n \"Title: An entropy-based uncertainty measure of process models\\nYear: 2011\\n\\nAbstract: In managing business processes, the process uncertainty and variability are significant factors causing difficulties in prediction and decision making, which evokes and augments the importance and need of process measures for systematic analysis. We propose an entropy-based process measure to quantify the uncertainty of business process models. The proposed measure enables capturing the dynamic behavior of processes, in contrast to previous work which focused on providing measures for the static aspect of process models.\",\n \"Title: What makes process models understandable?\\nYear: 2007\\n\\nAbstract: Despite that formal and informal quality aspects are of significantimportance to business process modeling, there is only little empiricalwork reported on process model quality and its impact factors. Inthis paper we investigate understandability as a proxy for quality of processmodels and focus on its relations with personal and model characteristics.We used a questionnaire in classes at three European universitiesand generated several novel hypotheses from an exploratory data analysis.Furthermore, we interviewed practitioners to validate our findings.The results reveal that participants tend to exaggerate the differences inmodel understandability, that self-assessment of modeling competenceappears to be invalid, and that the number of arcs in models has animportant influence on understandability.\",\n \"Title: Towards thresholds of control flow complexity measures for BPMN models\\nYear: 2011\\n\\nAbstract: Business process models are considered to be a good mechanism for communication among stakeholders and are a key instrument in the analysis and design of information systems. It is therefore important to design business process models with a high level of quality, which can be discovered through measurement application. Several measurement initiatives exist in the literature, but these measures are only useful in real world decision making if we also have criteria with which to establish the goodness of models. We consider that measures with thresholds and decision criteria form indicators. Indicators allow us to make decisions by using the values of the measures which models should not exceed to ascertain whether the model is good in practice. In this paper we present the initial empirical results from which thresholds for the Control-Flow Complexity measure applied in BPMN models have been obtained according to the Bender method. Our findings reveal that there are different levels of understandability depending on the number of decision nodes: a very easily understandable model would have no more than 6 xor nodes, 1 or nodes and 1 and nodes, versus the 46 xor nodes, 14 or nodes and 7 and nodes which would constitute a model with a very difficult level of understandability.\",\n \"Title: Web process and workflow path mining using the Multimethod approach\\nYear: 2006\\n\\nAbstract: Workflow Management Systems (WfMS) provide a fundamental technological infrastructure to define and manage business processes efficiently. WfMS logs contain valuable data that can be used to discover and extract knowledge about the execution of workflows and processes. One piece of important and useful information that can be discovered is related to the prediction of the path that will be followed during the execution of a workflow. We call this type of discovery, path mining. In this paper, we present and describe how path mining can be achieved using different data mining techniques including the Multimethod approach.\",\n \"Title: On a Quest for Good Process Models: The Cross-Connectivity Metric\\nYear: 2008\\n\\nAbstract: Business process modeling is an important corporate activity, but the understanding of what constitutes good process models is rather limited. In this paper, we turn to the cognitive dimensions framework and identify the understanding of the structural relationship between any pair of model elements as a hard mental operation. Based on the weakest-link metaphor, we introduce the cross-connectivity metric that measures the strength of the links between process model elements. The definition of this new metric builds on the hypothesis that process models are easier understood and contain less errors if they have a high cross-connectivity. We undertake a thorough empirical evaluation to test this hypothesis and present our findings. The good performance of this novel metric underlines the importance of cognitive research for advancing the field of process model measurement.\",\n \"Title: Network latency adaptive tempo in the Public Sound Objects system\\nYear: 2005\\n\\nAbstract: In recent years Computer Network-Music has increasingly captured the attention of the Computer Music Community. With the advent of Internet communication, geographical displacement amongst the participants of a computer mediated music performance achieved world wide extension. However, when established over long distance networks, this form of musical communication has a fundamental problem: network latency (or net-delay) is an impediment for real-time collaboration. From a recent study, carried out by the authors, a relation between network latency tolerance and Music Tempo was established. This result emerged from an experiment, in which simulated network latency conditions were applied to the performance of different musicians playing jazz standard tunes.The Public Sound Objects (PSOs) project is web-based shared musical space, which has been an experimental framework to implement and test different approaches for on-line music communication. This paper describe features implemented in the latest version of the PSOs system, including the notion of a network-music instrument incorporating latency as a software function, by dynamically adapting its tempo to the communication delay measured in real-time.\",\n \"Title: The Semantic Web Vision: Where Are We?\\nYear: 2007\\n\\nAbstract: The full vision of the Semantic Web and its mainstream adoption is still five to 10 years away. To get a current snapshot of key trends and developments, University of Madeira researchers surveyed 627 Semantic Web researchers and practitioners to learn how they're using and interrelating semantic technologies.\",\n \"Title: The Business Process Knowledge Framework\\nYear: 2007\\n\\nAbstract: Organizations today are confronted with huge problems following and implementing their own business process models. On the one hand, due to a lack of planning and requirements analysis, process models are often unfeasible or difficult to execute in practice. On the other hand, process designers often ignore the importance of studying the different roles and their perspectives on a business process when constructing a process model. This leads to the deployment of process models that do not satisfy process stakeholders. This paper addresses such problems and proposes a business process knowledge framework as a possible solution. Our framework integrates three elements that we consider fundamental to model business processes: stakeholders' perspectives, knowledge types and views. We show how the business process framework can contribute to the improvement of the process knowledge acquisition phase of process design, and how it can support process knowledge communication to stakeholders.\",\n \"Title: Enterprise Information Systems: 9th International Conference, ICEIS 2007, Funchal, Madeira, June 12-16, 2007, Revised Selected Papers\\nYear: 2009\\n\\nAbstract: This books contains the best papers of the 9th International Conference on Enterprise Information Systems, ICEIS 2007, held in Funchal, Madeira, Portugal, in June 2007. The 6 invited papers are presented together with23 papers whichwere carefully reviewed and selected from 72 full papers accepted for presentation at the conference itself (out of 644 submissions). The topics covered are: databases and information systems integration; artificial intelligence and decision support systems; information systems analysis and specification;software agents and internet computing; and human-computer interaction.\",\n \"Title: Soundserver: Data Sonification On-Demand for Computational Instances\\nYear: 2004\\n\\nAbstract: The rapid accumulation of large collections of data has created the need for efficient and intelligent schemes for knowledge extraction and results analysis. The resulting information is typically visual- ized, but it may also be presented through audio techniques such as sonification. Sonification techniques become especially interest- ing when the client application runs on graphically limited devices such as mobile phones or PDAs (Personal Digital Assistants). In this paper we present an architecture for a sonification server that will be used in the Sound Data Mining project. In this project sound will be used to increase perception and present information extracted by spatial data mining techniques. The server is based on an audio synthesis engine and will relieve clients with little au- dio synthesis capabilities from the burden of sound processing. By providing sonification modules, this server can potentially be used on a variety of applications where sonification techniques are re- quired.\",\n \"Title: Towards a Unified Service Description Language for the Internet of Services: Requirements and First Developments\\nYear: 2010\\n\\nAbstract: Service-oriented Architectures (SOA) and Web services leverage the technical value of solutions in the areas of distributed systems and cross-enterprise integration. The emergence of Internet marketplaces for business services is driving the need to describe services, not only from a technical level, but also from a business and operational perspective. While, SOA and Web services reside in an IT layer, organizations owing Internet marketplaces are requiring advertising and trading business services which reside in a business layer. As a result, the gap between business and IT needs to be closed. This paper presents USDL (Unified Service Description Language), a specification language to describe services from a business, operational and technical perspective. USDL plays a major role in the Internet of Services to describe tradable services which are advertised in electronic marketplaces. The language has been tested using two service marketplaces as use cases.\",\n \"Title: Tourism Information Aggregation Using an Ontology based approach\\nYear: 2007\\n\\nAbstract: Aggregating related information, from different data sources, allows the creation of data repositories with very useful information. In the tourism domain, aggregating tourism products with related tourism attractions will add value to those products. The ability to create dynamic packages is another reason to aggregate tourism information. Defining an ontology, composed by the concepts to aggregate, is the first step to create tourism aggregation systems. In this paper we define the approach and the architecture that guides to the creation of aggregated solutions that provide valued tourism information and that allow the creation of dynamic packages.\",\n \"Title: The Semantic Web: A mythical story or a solid reality?\\nYear: 2007\\n\\nAbstract: The Semantic Web vision has drove hundreds of practitioners to research and develop a new bread of applications that could take the full potential of the Web to the next level. While there is a fairly clear understanding of where Web 1.0 and Web 2.0 stand now a day, the Current status and position of the Semantic Web, also known as Web 3.0, is not as clear and well defined. Therefore. in this paper we present a landscape that illustrates and captures the trends in the Semantic Web with the purpose of guiding future developments.\",\n \"Title: Quality of service for workflows and web service processes\\nYear: 2004\\n\\nAbstract: Workflow management systems (WfMSs) have been used to support various types of business processes for more than a decade now. In workflows or Web processes for e-commerce and Web service applications, suppliers and customers define a binding agreement or contract between the two parties, specifying quality of service (QoS) items such as products or services to be delivered, deadlines, quality of products, and cost of services. The management of QoS metrics directly impacts the success of organizations participating in e-commerce. Therefore, when services or products are created or managed using workflows or Web processes, the underlying workflow engine must accept the specifications and be able to estimate, monitor, and control the QoS rendered to customers. In this paper, we present a predictive QoS model that makes it possible to compute the quality of service for workflows automatically based on atomic task QoS attributes. We also present the implementation of our QoS model for the METEOR workflow system. We describe the components that have been changed or added, and discuss how they interact to enable the management of QoS.\",\n \"Title: Process control-flow complexity metric: An empirical validation\\nYear: 2006\\n\\nAbstract: Organizations are increasingly faced with the challenge of managing business processes, workflows, and, recently, Web processes. One important aspect of processes that has been overlooked is their complexity. High complexity in processes may result in bad understandability, errors, defects, and exceptions leading processes to need more time to develop, test, and maintain. Therefore, excessive complexity should be avoided. This paper describes an experiment designed to validate the Control-Flow Complexity (CFC) metric that we have proposed in our previous work. In order to demonstrate that our CFC metric serves the purpose it was defined for, we have carried out an empirical validation by means of a controlled experiment. The explanation of the steps followed to do the experiment, the results, and the conclusions obtained are the main objectives of this paper.\",\n \"Title: About the Complexity of Teamwork and Collaboration Processes\\nYear: 2005\\n\\nAbstract: Organizations across the globe are increasingly using teams to accomplish significant work and projects. Much of this work is also accomplished using technology tools to support their communication and collaborative efforts. As companies become increasingly multinational and distributed geographically, theformation of virtual teams has become a common practice. Workflow management systems are a specific type of systems that can be used to capture collaboration and group works processes and thus supports the creation of teamwork and enable collaboration. In some cases, collaboration and group work processes can become highly complex. High complexity in a process may result in bad understandability and more errors, defects, and exceptions leading processes to need more time to develop, test, and maintain. Therefore, excessive complexity should be avoided. The major goal of this paper is to discuss the need and requirementsfor the development of a measure to analyze the complexity of processes.\",\n \"Title: M-GIS - Mobile and Interoperable Access to Geographic Information\\nYear: 2004\\n\\nAbstract: This paper describes an architecture which can be used to access geographic information from mobile devices with limited display and processing characteristics, such as Personal Digital Assistants (PDA) or Mobile Phones. The information may come from different sources leading to an interoperable solution. M-GIS obtains geographic information from sources described using Geogxraphy Markup Language (GML) from one or more Web Feature Servers allowing an access independent from its format or physical location, as long as the information is according to the specification. Using XSLT, GML is transformed to a graphical format in Scalable Vector Graphics (SVG) which can then be manipulated in the mobile device. M-GIS follows a client/server architecture and the client application was developed using Java Mobile Information Device Profile technology. The results enable us to conclude that the mobile system, designed and developed with the use of open standards and representation formats to access geographic information, is a viable solution with some limitations. The main limitations of this approach have to do with the amount of information which the client can, at the moment, accommodate.\",\n \"Title: Evolution and Overview of Linked USDL.\\nYear: 2015\\n\\nAbstract: For more than 10 years, research on service descriptions has mainly studied software-based services and provided languages such as WSDL, OWL-S, WSMO for SOAP, and hREST for REST. Nonetheless, recent developments from service management (e.g., ITIL and COBIT) and cloud computing (e.g. Software-as-a-Service) have brought new requirements to service descriptions languages: the need to also model business services and account for the multi-faceted nature of services. Business-orientation, co-creation, pricing, legal aspects, and security issues are all elements which must also be part of service descriptions. While ontologies such as e(3) service and e(3) value provided a first modeling attempt to capture a business perspective, concerns on how to contract services and the agreements entailed by a contract also need to be taken into account. This has for the most part been disregarded by the e(3) family of ontologies. In this paper, we review the evolution and provide an overview of Linked USDL, a comprehensive language which provides a (multi-faceted) description to enable the commercialization of (business and technical) services over the web.\",\n \"Title: Energy-Efficient Data Processing at Sweet Spot Frequencies.\\nYear: 2014\\n\\nAbstract: The processing of Big Data often includes sorting as a basic operator. Indeed, it has been shown that many software applications spend up to 25% of their time sorting data. Moreover, for compute-bound applications, the most energy-efficient executions have shown to use a CPU speed lower than the maximum speed: the CPU sweet spot frequency. In this paper, we use these findings to run Big Data intensive applications in a more energy-efficient way. We give empirical evidence that data-intensive analytic tasks are more energy-efficient when CPU(s) operate(s) at sweet spots frequencies. Our approach uses a novel high-precision, fine-grained energy measurement infrastructure to investigate the energy (joules) consumed by different sorting algorithms. Our experiments show that algorithms can have different sweet spot frequencies for the same computational task. To leverage these findings, we describe how a new kind of self-adaptive software applications can be engineered to increase their energy-efficiency.\",\n \"Title: Energy-Efficient Databases Using Sweet Spot Frequencies\\nYear: 2014\\n\\nAbstract: Database management systems (DBMS) are typically tuned for high performance and scalability. Nevertheless, carbon footprint and energy efficiency are also becoming increasing concerns. Unfortunately, existing studies mainly present theoretical contributions but fall short on proposing practical techniques. These could be used by administrators or query optimizers to increase the energy efficiency of the DBMS. Thus, this paper explores the effect of so-called sweet spots, which are energy-efficient CPU frequencies, on the energy required to execute queries. From our findings, we derive the Sweet Spot Technique, which relies on identifying energy-efficient sweet spots and the optimal number of threads that minimizes energy consumption for a query or an entire database workload. The technique is simple and has a practical implementation leading to energy savings of up to 50% compared to using the nominal frequency and maximum number of threads.\",\n \"Title: A Systematic Mapping Applied to MOOC's Study.\\nYear: 2015\\n\\nAbstract: MOOC platforms are Web-based learning environments which allow a global participation on a large scale and with free access. The paradigm presents itself as a new teaching trend, changing the way education can be offered and funded worldwide. Many institutions are now investing in this teaching mode. However, since it is a web-based tool, the connection performance can impact both the way learning is experienced by the student as well as the operation of the platform. The âQuality of Experienceâ (QoE) concept has been widely used to refer to how users describe a service they have used while the âQuality of Serviceâ (QoS) concept deals with the technical performance parameters that are associated with the connection quality. This paper starts the process of developing a systematic mapping around MOOC platforms, and QoS and QoE concepts, aiming to provide an overview of the current state of research on these issues.\",\n \"Title: Measuring Cortical Neurite-Dispersion and Perfusion in Preterm-Born Adolescents Using Multi-modal MRI\\nYear: 2015\\n\\nAbstract: As a consequence of a global increase in rates of extremely preterm birth, predicting the long term impact of preterm birth has become an important focus of research. Cohorts of extremely preterm born subjects studied in the 1990s are now beginning to reach adulthood and the long term structural alterations of disrupted neurodevelopment in gestation can now be investigated, for instance with magnetic resonance (MR) imaging. Disruption to normal development as a result of preterm birth is likely to result in both cerebrovascular and microstructural differences compared to term-born controls. Of note, arterial spin labelled MRI provides a marker of cerebral blood flow, whilst multi-compartment diffusion models provide information on the cerebral microstructure, including that of the cortex. We apply these techniques to a cohort of 19 year-old adolescents consisting of both extremely-preterm and term-born individuals and investigate the structural and functional correlations of these MR modalities. Work of this type, revealing the long-term structural and functional differences in preterm cohorts, can help better inform on the likely outcomes of contemporary extremely preterm new-borns and provides an insight into the lifelong effects of preterm birth.\",\n \"Title: Open Service Network Analysis\\nYear: 2013\\n\\nAbstract: Understanding how services operate as part of large scale global networks, the related risks and gains of different network structures and their dynamics is becoming increasingly critical for society. Our vision and research agenda focuses on the particularly challenging task of building, analyzing, and reasoning about global service networks. This paper explains how Service Network Analysis (SNA) can be used to study and optimize the provisioning of complex services modeled as Open Semantic Service Networks (OSSN), a computer-understandable digital structure which represents connected and dependent services.\",\n \"Title: On Black-Box Monitoring Techniques for Multi-Component Services\\nYear: 2018\\n\\nAbstract: Despite the advantages of microservice and function-oriented architectures, there is an increase in complexity to monitor such highly dynamic systems. In this paper, we analyze two distinct methods to tackle the monitoring problem in a system with reduced instrumentation. Our goal is to understand the feasibility of such approach with one specific driver: simplicity. We aim to determine the extent to which it is possible to characterize the state of two generic tandem processes, using as little information as possible. To answer this question, we resorted to a simulation approach. Using a queue system, we simulated two services, that we could manipulate with distinct operation sets for each module. We used the total response time seen upstream of the system. Having this setup and metric, we applied two distinct methods to analyze the results. First, we used supervised machine learning algorithms to identify where the bottleneck is happening. Secondly, we used an exponential decomposition to identify the occupation in the two components in a more black-box fashion. Results show that both methodologies have their advantages and limitations. The separation of the signal more accurately identifies occupation in low occupied resources, but when a service is totally dominating the overall time, it lacks precision. The machine learning has a more stable error, but needs the training set. This study suggest that a black-box occupation approach with both techniques is possible and very useful.\",\n \"Title: Response Time Characterization of Microservice-Based Systems\\nYear: 2018\\n\\nAbstract: In pursuit of faster development cycles, companies have favored small decoupled services over monoliths. Following this trend, distributed systems made of microservices have grown in scale and complexity, giving rise to a new set of operational problems. Even though this paradigm simplifies development, deployment, management of individual services, it hinders system observability. In particular, performance monitoring and analysis becomes more challenging, especially for critical production systems that have grown organically, operate continuously, and cannot afford the availability cost of online benchmarking. Additionally, these systems are often very large and expensive, thus being bad candidates for full-scale development replicas. Creating models of services and systems for characterization and formal analysis can alleviate the aforementioned issues. Since performance, namely response time, is the main interest of this work, we focused on bottleneck detection and optimal resource scheduling. We propose a method for modeling production services as queuing systems from request traces. Additionally, we provide analytical tools for response time characterization and optimal resource allocation. Our results show that a simple queuing system with a single queue and multiple homogeneous servers has a small parameter space that can be estimated in production. The resulting model can be used to accurately predict response time distribution and the necessary number of instances to maintain a desired service level, under a given load.\",\n \"Title: Efficient Failure Diagnosis of OpenStack Using Tempest.\\nYear: 2018\\n\\nAbstract: While cloud computing continues to be popular in the IT world, companies offering cloud solutions are under pressure to provide the most reliable solutions to users. In this article, we describe an innovative approach to diagnose service failures in an OpenStack-based cloud using Tempest as a starting point.\",\n \"Title: Service Engineering for the Internet of Services\\nYear: 2008\\n\\nAbstract: The Internet and the Web have extended traditional business networks by allowing a Web of different digital resources to work together to create value for organizations. The most industrialized countries have entered a post-industrial era where their prosperity is largely created through a service economy. There is a clear transition from a manufacturing based economy to a service based economy. From the technological perspective, the development of Web-based infrastructures to Support and deliver services in this new economy raises a number of challenges. From a business perspective, there is the need to understand how value is created through set-vices. In this paper, we describe how we propose to address these two perspectives and realize the vision of the Internet of Services (IoS), where Web-based IT-supported service ecosystems form the base of service business value networks. This paper addresses the main challenging issues that need to be explored to provide an integrated technical and business infrastructure for the Internet of Service.\",\n \"Title: COOPIS'10: PC co-chairs message\\nYear: 2010\\n\\nAbstract: Welcome to the 18th International Conference on Cooperative Information Systems (CoopIS 2010). This year CoopIS was held in\\n Crete, Greece, during October 27-29, 2010.\\n \\n \\n Cooperative Information Systems (CIS) provide enterprises and communities of users with flexible, scalable and intelligent\\n services in large-scale networking environments. Building a scalable cooperative information system requires technical breakthroughs\\n to overcome tough challenges that traditional rigid distributed systems did not face.\",\n \"Title: Anomaly Detection and Classification using Distributed Tracing and Deep Learning\\nYear: 2019\\n\\nAbstract: Artificial Intelligence for IT Operations (AIOps) combines big data and machine learning to replace a broad range of IT Operations tasks including availability, performance, and monitoring of services. By exploiting log, tracing, metric, and network data, AIOps enable detection of faults and issues of services. The focus of this work is on detecting anomalies based on distributed tracing records that contain detailed information for the availability and the response time of the services. In large-scale distributed systems, where a service is deployed on heterogeneous hardware and has multiple scenarios of normal operation, it becomes challenging to detect such anomalous cases. We address the problem by proposing unsupervised, response time anomaly detection based on deep learning data modeling techniques; unsupervised dynamic error threshold approach; tolerance module for false positive reduction; and descriptive classification of the anomalies. The evaluation shows that the approach achieves high accuracy and solid performance in both, experimental testbed and large-scale production cloud.\",\n \"Title: IAD: Indirect Anomalous VMMs Detection in the Cloud-Based Environment\\nYear: 2021\\n\\nAbstract: Server virtualization in the form of virtual machines (VMs) with the use of a hypervisor or a Virtual Machine Monitor (VMM) is an essential part of cloud computing technology to provide infrastructure-as-a-service (IaaS). A fault or an anomaly in the VMM can propagate to the VMs hosted on it and ultimately affect the availability and reliability of the applications running on those VMs. Therefore, identifying and eventually resolving it quickly is highly important. However, anomalous VMM detection is a challenge in the cloud environment since the user does not have access to the VMM. This paper addresses this challenge of anomalous VMM detection in the cloud-based environment without having any knowledge or data from VMM by introducing a novel machine learning-based algorithm called IAD: Indirect Anomalous VMMs Detection. This algorithm solely uses the VM's resources utilization data hosted on those VMMs for the anomalous VMMs detection. The developed algorithm's accuracy was tested on four datasets comprising the synthetic and real and compared against four other popular algorithms, which can also be used to the described problem. It was found that the proposed IAD algorithm has an average F1-score of 83.7% averaged across four datasets, and also outperforms other algorithms by an average F1-score of 11%.\",\n \"Title: QuLog: Data-Driven Approach for Log Instruction Quality Assessment\\nYear: 2022\\n\\nAbstract: In the current IT world, developers write code while system operators run the code mostly as a black box. The connection between both worlds is typically established with log messages: the developer provides hints to the (unknown) operator, where the cause of an occurred issue is, and vice versa, the operator can report bugs during operation. To fulfil this purpose, developers write log instructions that are structured text commonly composed of a log level (e.g., “info”, “error”), static text (“IP {} cannot be reached”), and dynamic variables (e.g. IP {}). However, opposed to well-adopted coding practices, there are no widely adopted guidelines on how to write log instructions with good quality properties. For example, a developer may assign a high log level (e.g., “error”) for a trivial event that can confuse the operator and increase maintenance costs. Or the static text can be insufficient to hint at a specific issue. In this paper, we address the problem of log quality assessment and provide the first step towards its automation. We start with an in-depth analysis of quality log instruction properties in nine software systems and identify two quality properties: 1) correct log level assignment assessing the correctness of the log level, and 2) sufficient linguistic structure assessing the minimal richness of the static text necessary for verbose event description. Based on these findings, we developed a data-driven approach that adapts deep learning methods for each of the two properties. An extensive evaluation on large-scale open-source systems shows that our approach correctly assesses log level assignments with an accuracy of 0.88, and the sufficient linguistic structure with an F1 score of 0.99, outperforming the baselines. Our study highlights the potential of the data-driven methods in assessing log instructions quality.\",\n \"Title: Failure Identification from Unstable Log Data using Deep Learning\\nYear: 2022\\n\\nAbstract: The reliability of cloud platforms is of significant relevance because society increasingly relies on complex software systems running on the cloud. To improve it, cloud providers are automating various maintenance tasks, with failure identification frequently being considered. The precondition for automation is the availability of observability tools, with system logs commonly being used. The focus of this paper is log-based failure identification. This problem is challenging because of the instability of the log data and the incompleteness of the explicit logging failure coverage within the code. To address the two challenges, we present CLog as a method for failure identification. The key idea presented herein based is on our observation that by representing the log data as sequences of subprocesses instead of sequences of log events, the effect of the unstable log data is reduced. CLog introduces a novel subprocess extraction method that uses context-aware neural network and clustering methods to extract meaningful subprocesses. The direct modeling of log event contexts allows the identification of failures with respect to the abrupt context changes, addressing the challenge of insufficient logging failure coverage. Our experimental results demonstrate that the learned subprocesses representations reduce the instability in the input, allowing CLog to outperform the baselines on the failure identification subproblems - 1) failure detection by 9–24 % on F1 score and 2) failure type identification by 7% on the macro averaged F1 score. Further analysis shows the existent negative correlation between the instability in the input event sequences and the detection performance in a model-agnostic manner.\",\n \"Title: Leveraging Log Instructions in Log-based Anomaly Detection\\nYear: 2022\\n\\nAbstract: Artificial Intelligence for IT Operations (AIOps) describes the process of maintaining and operating large IT systems using diverse AI-enabled methods and tools for, e.g., anomaly detection and root cause analysis, to support the remediation, optimization, and automatic initiation of self-stabilizing IT activities. The core step of any AIOps workflow is anomaly detection, typically performed on high-volume heterogeneous data such as log messages (logs), metrics (e.g., CPU utilization), and distributed traces. In this paper, we propose a method for reliable and practical anomaly detection from system logs. It overcomes the common disadvantage of related works, i.e., the need for a large amount of manually labeled training data, by building an anomaly detection model with log instructions from the source code of 1000+ GitHub projects. The instructions from diverse systems contain rich and heterogenous information about many different normal and abnormal IT events and serve as a foundation for anomaly detection. The proposed method, named ADLILog, combines the log instructions and the data from the system of interest (target system) to learn a deep neural network model through a two-phase learning procedure. The experimental results show that ADLILog outperforms the related approaches by up to 60% on the F\\n<inf xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">1</inf>\\n score while satisfying core non-functional requirements for industrial deployments such as unsupervised design, efficient model updates, and small model sizes.\",\n \"Title: Morphology-Preserving Autoregressive 3D Generative Modelling of the Brain\\nYear: 2022\\n\\nAbstract: Human anatomy, morphology, and associated diseases can be studied using medical imaging data. However, access to medical imaging data is restricted by governance and privacy concerns, data ownership, and the cost of acquisition, thus limiting our ability to understand the human body. A possible solution to this issue is the creation of a model able to learn and then generate synthetic images of the human body conditioned on specific characteristics of relevance (e.g., age, sex, and disease status). Deep generative models, in the form of neural networks, have been recently used to create synthetic 2D images of natural scenes. Still, the ability to produce high-resolution 3D volumetric imaging data with correct anatomical morphology has been hampered by data scarcity and algorithmic and computational limitations. This work proposes a generative model that can be scaled to produce anatomically correct, high-resolution, and realistic images of the human brain, with the necessary quality to allow further downstream analyses. The ability to generate a potentially unlimited amount of data not only enables large-scale studies of human anatomy and pathology without jeopardizing patient privacy, but also significantly advances research in the field of anomaly detection, modality synthesis, learning under limited data, and fair and ethical Al. Code and trained models are available at: https://github.com/AmigoLab/SynthAnatomy.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":72,"string":"72"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13582,"string":"13,582"}}},{"rowIdx":13583,"cells":{"question_id":{"kind":"string","value":"Hans Joachim Bungartz"},"question":{"kind":"string","value":"Generate an abstract for the title \"Efficient Global Element Indexing for Parallel Adaptive Flow Solvers.\" using the following items: Grid-based solvers, partial differential equations (PDE), matrices, global indexing, bottleneck."},"answer":{"kind":"string","value":"Many grid-based solvers for partial differential equations (PDE) assemble matrices explicitly for discretizing the underlying PDE operators and/or for the underlying (non-) linear systems of equations. Often, the data structures or solver packages require a consecutive global numbering of the degrees of freedom across the boundaries of different parallel subdomains. Straightfor- ward approaches to realize this global indexing in parallel frequently result in serial parts of the assembling algorithms which causes a considerable bottleneck, in particular in large-scale applications."},"correct_answer":{"kind":"string","value":"Many grid-based solvers for partial differential equations (PDE) assemble matrices explicitly for discretizing the underlying PDE operators and/or for the underlying (non-) linear systems of equations. Often, the data structures or solver packages require a consecutive global numbering of the degrees of freedom across the boundaries of different parallel subdomains. Straightfor- ward approaches to realize this global indexing in parallel frequently result in serial parts of the assembling algorithms which causes a considerable bottleneck, in particular in large-scale applications."},"documents":{"kind":"list like","value":["Title: A high-level Fortran interface to parallel matrix algebra\nYear: 2012\n\nAbstract: Scientific software often relies on complex vector- and matrix algebra, sometimes combined with more elaborate linear algebra problems, such as systems of linear equations or eigenvalue problems. Its efficient parallel computation is well studied, and there is a variety of parallel libraries for this task available. However, the user still has to cope with many technical details, such as library-specific data distribution or complicated routine calls, which, if not carefully used, easily result in confusing, difficult to maintain code. This paper presents a Fortran 90 programming interface and its library implementation, which specifies easy-to-use distributed data objects, combined with a set of linear algebra operators. The interface allows matrix algebra to be expressed in a high-level mathematical notation, while operating internally with parallel routines. Thus, linear algebra can be expressed in a clear, comprehensible way, while low-level tasks, such as data organization and communication, are hidden from the user. We evaluate this approach with an existing code from relativistic quantum chemistry, which includes complex operations with dense- and diagonal matrices, solving the generalized eigenvalue problem and several domain specific operations. The results show that the required relativistic transformations can be implemented quickly, while producing clear, light-weight code which scales up to 81 processor cores with efficiency of about 50% for our test cases.","Title: Fast GPGPU Data Rearrangement Kernels using CUDA\nYear: 2010\n\nAbstract: Many high performance-computing algorithms are bandwidth limited, hence the\nneed for optimal data rearrangement kernels as well as their easy integration\ninto the rest of the application. In this work, we have built a CUDA library of\nfast kernels for a set of data rearrangement operations. In particular, we have\nbuilt generic kernels for rearranging m dimensional data into n dimensions,\nincluding Permute, Reorder, Interlace/De-interlace, etc. We have also built\nkernels for generic Stencil computations on a two-dimensional data using\ntemplates and functors that allow application developers to rapidly build\ncustomized high performance kernels. All the kernels built achieve or surpass\nbest-known performance in terms of bandwidth utilization.","Title: Hybrid molecular-continuum methods: From prototypes to coupling software\nYear: 2014\n\nAbstract: In this contribution, we review software requirements in hybrid molecular-continuum simulations. For this purpose, we analyze a prototype implementation which combines two frameworks-the Molecular Dynamics framework MarDyn and the framework Peano for spatially adaptive mesh-based simulations-and point out particular challenges of a general coupling software. Based on this analysis, we discuss the software design of our recently published coupling tool. We explain details on its overall structure and show how the challenges that arise in respective couplings are resolved by the software.","Title: Numerical Simulation of Particle Transport in a Drift Ratchet\nYear: 2008\n\nAbstract: The directed transport of microparticles depending on their size is the basis for particle sorting methods that are of utmost importance in, for example, life sciences. A drift ratchet is a Brownian motor that allows for such a directed transport. Hereby, the particle motion is induced by a combination of the Brownian motion and asymmetries stemming, for example, from the domain's geometry, electrical fields, or transient pressure boundary conditions. We simulate a particular drift ratchet which consists of a matrix of pores with asymmetrically oscillating diameter wherein a fluid with suspended particles is pumped forward and backward, and where the particles' long-term transport direction depends on their size. Thus, this setup allows for continuous and parallel particle separation, which has been shown experimentally. However, for a deeper understanding and for an optimized parameters' choice, further investigations, i.e., simulations, are necessary. In this paper, we present first results achieved with our parallel three-dimensional simulation codes applied to a still simplified scenario. This simplification is necessary to isolate different phenomena (e.g., asymmetries and Brownian motion) to check their relevance for the particle transport. The simulation codes are based on (adaptive) Cartesian grids in combination with finite volume and finite element discretizations. Cartesian grids allow for a very efficient implementation of the solver algorithms and an efficient balanced parallelization via domain decomposition. The achieved simulation results show the effectiveness of our approach and give some strong hints on a directed particle transport already with the simplified model we used here.","Title: Option pricing with a direct adaptive sparse grid approach\nYear: 2012\n\nAbstract: We present an adaptive sparse grid algorithm for the solution of the Black-Scholes equation for option pricing, using the finite element method. Sparse grids enable us to deal with higher-dimensional problems better than full grids. In contrast to common approaches that are based on the combination technique, which combines different solutions on anisotropic coarse full grids, the direct sparse grid approach allows for local adaptive refinement. When dealing with non-smooth payoff functions, this reduces the computational effort significantly. In this paper, we introduce the spatially adaptive discretization of the Black-Scholes equation with sparse grids and describe the algorithmic structure of the numerical solver. We present several strategies for adaptive refinement, evaluate them for different dimensionalities, and demonstrate their performance showing numerical results.","Title: Adaptive Sparse Grid Techniques for Data Mining.\nYear: 2006\n\nAbstract: It was shown in [GaGT01] that the task of classification in data mining can be tackled by employing ansatz functions associated to grid points in the (often high dimensional) feature-space rather than using data-centered ansatz functions. To cope with the curse of dimensionality, sparse grids have been used. Based on this approach we propose an efficient finite-element-like discretization technique for classification instead of the combination technique used in [GaGT01]. The main goal of our method is to make use of adaptivity to further reduce the number of grid points needed. Employing adaptivity in classification is reasonable as the target function contains smooth regions as well as rough ones. Regarding implementational issues we present an algorithm for the fast multiplication of the vector of unknowns with the coefficient matrix. We give an example for the adaptive selection of grid points and show that special care has to be taken regarding the boundary values, as adaptive techniques commonly used for solving PDEs are not optimal here. Results for some typical classification tasks, including a problem from the UCI repository, are presented.","Title: A parallel adaptive cartesian PDE solver using space–filling curves\nYear: 2006\n\nAbstract: In this paper, we present a parallel multigrid PDE solver working on adaptive hierarchical cartesian grids. The presentation is restricted to the linear elliptic operator of second order, but extensions are possible and have already been realised as prototypes. Within the solver the handling of the vertices and the degrees of freedom associated to them is implemented solely using stacks and iterates of a Peano space–filling curve. Thus, due to the structuredness of the grid, two administrative bits per vertex are sufficient to store both geometry and grid refinement information. The implementation and parallel extension, using a space–filling curve to obtain a load balanced domain decomposition, will be formalised. In view of the fact that we are using a multigrid solver of linear complexity $\\mathcal{O}(n)$, it has to be ensured that communication cost and, hence, the parallel algorithm's overall complexity do not exceed this linear behaviour.","Title: DaStGen--A Data Structure Generator for Parallel C++ HPC Software\nYear: 2008\n\nAbstract: Simulation codes often suffer from high memory requirements. This holds in particular if they are memory-bounded, and, with multicore systems coming up, the problem will become even worse as more and more cores have to share the memory connections. To optimise data structures with respect to memory manually is error-prone and cumbersome.This paper presents the tool DaStGen translating classes declared in C++ syntax and augmented by new keywords into plain C++ code. The tool automates the record optimisation, as it analyses the potential range of each attribute, and as the user can restrict this range further. Herefrom, the generated code stores multiple attributes within one single primitive type. Furthermore, the tool derives user-defined MPI data types for each class. Using the tool reduces any algorithm's memory footprint, it speeds up memory-bounded applications such as CFD codes, and it hides technical details of MPI applications from the programmer.","Title: Computational Steering Of Complex Flow Simulations\nYear: 2018\n\nAbstract: Computational Steering, the combination of a simulation back-end with a visualisation front-end, offers great possibilities to exploit and optimise scenarios in engineering applications. Due to its interactivity, it requires fast grid generation, simulation, and visualisation and, therefore, mostly has to rely on coarse and inaccurate simulations typically performed on rather small interactive computing facilities and not on much more powerful high-performance computing architectures operated in batch-mode. This paper presents a steering environment that intends to bring these two worlds the interactive and the classical HPC world together in an integrated way. The environment consists of efficient fluid dynamics simulation codes and a steering and visualisation framework providing a user interface, communication methods for distributed steering, and parallel visualisation tools. The gap between steering and HPC is bridged by a hierarchical approach that performs fast interactive simulations for many scenario variants increasing the accuracy via hierarchical refinements in dependence of the time the user wants to wait. Finally, the user can trigger large simulations for selected setups on an HPC architecture exploiting the pre-computations already done on the interactive system.","Title: Computational science and engineering: a new master's program at the Technische Universität München\nYear: 2003\n\nAbstract: In the winter semester of 2001/2002, the Technische Universität München (TUM) started its new master's program computational science and engineering (CSE) as a joint initiative of seven faculties. It is the objective of this contribution to outline the underlying ideas and concepts and their curricular implementation as well as to emphasize some features that are probably non-standard in other comparable programs.","Title: A sparse-grid-based out-of-sample extension for dimensionality reduction and clustering with laplacian eigenmaps\nYear: 2011\n\nAbstract: Spectral graph theoretic methods such as Laplacian Eigenmaps are among the most popular algorithms for manifold learning and clustering. One drawback of these methods is, however, that they do not provide a natural out-of-sample extension. They only provide an embedding for the given training data. We propose to use sparse grid functions to approximate the eigenfunctions of the Laplace-Beltrami operator. We then have an explicit mapping between ambient and latent space. Thus, out-of-sample points can be mapped as well. We present results for synthetic and real-world examples to support the effectiveness of the sparse-grid-based explicit mapping.","Title: 591 TFLOPS Multi-trillion Particles Simulation on SuperMUC.\nYear: 2013\n\nAbstract: Anticipating large-scale molecular dynamics simulations (MD) in nano-fluidics, we conduct performance and scalability studies of an optimized version of the code ls1 mardyn. We present our implementation requiring only 32 Bytes per molecule, which allows us to run the, to our knowledge, largest MD simulation to date. Our optimizations tailored to the Intel Sandy Bridge processor are explained, including vectorization as well as shared-memory parallelization to make use of Hyperthreading. Finally we present results for weak and strong scaling experiments on up to 146016 Cores of SuperMUC at the Leibniz Supercomputing Centre, achieving a speed-up of 133k times which corresponds to an absolute performance of 591.2 TFLOPS.","Title: Teaching high-performance computing on a high-performance cluster\nYear: 2005\n\nAbstract: The university education in parallel and high-performance computing often suffers from a significant gap between the effects and potential performance taught in the lectures on the one hand and those practically experienced in exercises or lab courses on the other hand. With a small number of processors, the results obtained are often hardly convincing; however, supercomputers are rarely accessible to students doing their first steps in parallel programming. In this contribution, we present our experiences of how a state-of-the-art mid-size Linux cluster, bought and operated on a department level primarily for education and algorithm development purposes, can be used for teaching a large variety of HPC aspects. Special focus is put on the effects of such an approach on the intensity and sustainability of learning.","Title: Learning by doing: software projects in CSE education\nYear: 2006\n\nAbstract: Software development is one of the main routine activities in Computational Science and Engineering (CSE). Nevertheless, there is a huge gap between software engineering techniques available and established today in most fields where mainstream software is developed on the one hand and the typical extent of their application in a CSE context on the other hand. CSE curricula often reflect this tendency by not including software engineering topics adequately. This contribution reports experiences with a new course format called “student project” in the CSE master's program at TU München. There, for about half a year, a group of 4-8 students cooperate on a software development project – this time dealing with molecular dynamics. Although it is one objective to get a well performing code, the project's focus is on the consequent application of software engineering and project management practices.","Title: Spatially adaptive sparse grids for high-dimensional data-driven problems\nYear: 2010\n\nAbstract: Sparse grids allow one to employ grid-based discretization methods in data-driven problems. We present an extension of the classical sparse grid approach that allows us to tackle high-dimensional problems by spatially adaptive refinement, modified ansatz functions, and efficient regularization techniques. The competitiveness of this method is shown for typical benchmark problems with up to 166 dimensions for classification in data mining, pointing out properties of sparse grids in this context. To gain insight into the adaptive refinement and to examine the scope for further improvements, the approximation of non-smooth indicator functions with adaptive sparse grids has been studied as a model problem. As an example for an improved adaptive grid refinement, we present results for an edge-detection strategy.","Title: Adaptive Sparse Grid Classification Using Grid Environments\nYear: 2007\n\nAbstract: Common techniques tackling the task of classification in data mining employ ansatz functions associated to training data points to fit the data as well as possible. Instead, the feature space can be discretized and ansatz functions centered on grid points can be used. This allows for classification algorithms scaling only linearly in the number of training data points, enabling to learn from data sets with millions of data points. As the curse of dimensionality prohibits the use of standard grids, sparse grids have to be used.Adaptive sparse grids allow to get a trade-off between both worlds by refining in rough regions of the target function rather than in smooth ones. We present new results for some typical classification tasks and show first observations of dimension adaptivity. As the study of the critical parameters during development involves many computations for different parameter values, we used a grid environment which we present.","Title: Towards High-Dimensional Computational Steering of Precomputed Simulation Data using Sparse Grids\nYear: 2011\n\nAbstract: With the ever-increasing complexity, accuracy, dimensionality, and size of simulations, a step in the direction of data-intensive scientific discovery becomes necessary. Parameter-dependent simulations are an example of such a data-intensive tasks: The researcher, who is interested in the dependency of the simulation's result on a set of input parameters, changes essential parameters and wants to immediately see the effect of the changes in a visual environment. In this scenario, an interactive exploration is not possible due to the long execution time needed by even a single simulation corresponding to one parameter combination and the overall large number of parameter combinations which could be of interest.","Title: Multivariate quadrature on adaptive sparse grids\nYear: 2003\n\nAbstract: In this paper, we study the potential of adaptive sparse grids for multivariate numerical quadrature in the moderate or high dimensional case, i.e. for a number of dimensions beyond three and up to several hundreds. There, conventional methods typically suffer from the curse of dimension or are unsatisfactory with respect to accuracy. Our sparse grid approach, based upon a direct higher order discretization on the sparse grid, overcomes this dilemma to some extent, and introduces additional flexibility with respect to both the order of the 1 D quadrature rule applied (in the sense of Smolyak's tensor product decomposition) and the placement of grid points. The presented algorithm is applied to some test problems and compared with other existing methods.","Title: Clustering based on density estimation with sparse grids\nYear: 2012\n\nAbstract: We present a density-based clustering method. The clusters are determined by splitting a similarity graph of the data into connected components. The splitting is accomplished by removing vertices of the graph at which an estimated density function of the data evaluates to values below a threshold. The density function is approximated on a sparse grid in order to make the method feasible in higher-dimensional settings and scalable in the number of data points. With benchmark examples we show that our method is competitive with other modern clustering methods. Furthermore, we consider a real-world example where we cluster nodes of a finite element model of a Chevrolet pick-up truck with respect to the displacements of the nodes during a frontal crash.","Title: Localized Discrete Empirical Interpolation Method.\nYear: 2014\n\nAbstract: This paper presents a new approach to construct more efficient reduced-order models for nonlinear partial differential equations with proper orthogonal decomposition and the discrete empirical interpolation method (DEIM). Whereas DEIM projects the nonlinear term onto one global subspace, our localized discrete empirical interpolation method (LDEIM) computes several local subspaces, each tailored to a particular region of characteristic system behavior. Then, depending on the current state of the system, LDEIM selects an appropriate local subspace for the approximation of the nonlinear term. In this way, the dimensions of the local DEIM subspaces, and thus the computational costs, remain low even though the system might exhibit a wide range of behaviors as it passes through different regimes. LDEIM uses machine learning methods in the offline computational phase to discover these regions via clustering. Local DEIM approximations are then computed for each cluster. In the online computational phase, machine-learning-based classification procedures select one of these local subspaces adaptively as the computation proceeds. The classification can be achieved using either the system parameters or a low-dimensional representation of the current state of the system obtained via feature extraction. The LDEIM approach is demonstrated for a reacting flow example of an H-2-Air flame. In this example, where the system state has a strong nonlinear dependence on the parameters, the LDEIM provides speedups of two orders of magnitude over standard DEIM.","Title: Towards a Navier Stokes-Darcy Upscaling Based on Permeability Tensor Computation.\nYear: 2012\n\nAbstract: The micro scale simulation of CO2 sequestration involves complex, porous-like geometries. For the generation of such geometries, we present two approaches: In 2D, we construct a fractured domain by channel networks. In 3D, we approximate sand grain-like scenarios by dense sphere packings. The flow through these structures is simulated with the incompressible Navier-Stokes solver of the PDE framework Peano. Using an upscaling scheme, the results of the micro scale are used as input data for a Darcy solver on the coarse scales. The coupling concept and the scenario generators are presented together with first simulation results showing the validity of the approach.","Title: Free-Surface Lattice-Boltzmann Simulation On Many-Core Architectures\nYear: 2011\n\nAbstract: Current advances in many-core technologies demand simulation algorithms suited for the corresponding architectures while with regard to the respective increase of computational power, real-time and interactive simulations become possible and desirable. We present an OpenCL implementation of a Lattice-Boltzmann-based free-surface solver for GPU architectures. The massively parallel execution especially requires special techniques to keep the interface region consistent, which is here addressed by a novel multipass method. We further compare different memory layouts according to their performance for both a basic driven cavity implementation and the free-surface method, pointing out the capabilities of our implementation in real-time and interactive scenarios, and shortly present visualizations of the flow, obtained in real-time.","Title: A note on the complexity of solving Poisson's equation for spaces of bounded mixed derivatives\nYear: 1999\n\nAbstract: We study the complexity of solving the d -dimensional Poisson equation on ]0, 1[ d . We restrict ourselves to cases where the solution u lies in some space of functions of bounded mixed derivatives (with respect to the L ∞ - or the L 2 -norm) up to ∂ 2 d /∏ d j =1 ∂ x 2 j . An upper bound for the complexity of computing a solution of some prescribed accuracy ε with respect to the energy norm is given, which is proportional to ε −1 . We show this result in a constructive manner by proposing a finite element method in a special sparse grid space, which is obtained by an a priori grid optimization process based on the energy norm. Thus, the result of this paper is twofold: First, from a theoretical point of view concerning the complexity of solving elliptic PDEs, a strong tractability result of the order O ( ε −1 ) is given, and, second, we provide a practically usable hierarchical basis finite element method of this complexity O ( ε −1 ), i.e., without logarithmic terms growing exponentially in d , at least for our sparse grid setting with its underlying smoothness requirements.","Title: Many-core architectures boost the pricing of basket options on adaptive sparse grids.\nYear: 2013\n\nAbstract: In this work, we present a highly scalable approach for numerically solving the Black-Scholes PDE in order to price basket options. Our method is based on a spatially adaptive sparse-grid discretization with finite elements. Since we cannot unleash the compute capabilities of modern many-core chips such as GPUs using the complexity-optimal Up-Down method, we implemented an embarrassingly parallel direct method. This operator is paired with a distributed memory parallelization using MPI and we achieved very good scalability results compared to the standard Up-Down approach. Since we exploit all levels of the operator's parallelism, we are able to achieve nearly perfect strong scaling for the Black-Scholes solver. Our results show that typical problem sizes (5 dimensional basket options), require at least 4 NVIDIA K20X Kepler GPUs (inside a Cray XK7) in order to be faster than the Up-Down scheme running on 16 Intel Sandy Bridge cores (one box). On a Cray XK7 machine we outperform our highly parallel Up-Down implementation by 55X with respect to time to solution. Both results emphasize the competitiveness of our proposed operator.","Title: Semi-Coarsening In Space And Time For The Hierarchical Transformation Multigrid Method\nYear: 2012\n\nAbstract: We extend the hierarchical transformation multigrid (HT-MG) method with semi-coarsening in space and time in order to tackle anisotropic problems. Semi-coarsening allows us to smooth on anisotropic grids and, thus to employ more advanced multigrid cycles, e. g. the so-called Q-cycle. Our numerical examples show that we can tackle anisotropic problems with the HT-MG method and the Q-cycle. We do not only apply the multigrid method in space directions but also in time directions where the Q-cycle is beneficial again.","Title: A highly parallel Black–Scholes solver based on adaptive sparse grids\nYear: 2012\n\nAbstract: In this paper, we present a highly efficient approach for numerically solving the Black–Scholes equation in order to price European and American basket options. Therefore, hardware features of contemporary high performance computer architectures such as non-uniform memory access and hardware-threading are exploited by a hybrid parallelization using MPI and OpenMP which is able to drastically reduce the computing time. In this way, we achieve very good speed-ups and are able to price baskets with up to six underlyings. Our approach is based on a sparse grid discretization with finite elements and makes use of a sophisticated adaption. The resulting linear system is solved by a conjugate gradient method that uses a parallel operator for applying the system matrix implicitly. Since we exploit all levels of the operator's parallelism, we are able to benefit from the compute power of more than 100 cores. Several numerical examples as well as an analysis of the performance for different computer architectures are provided.","Title: Image Segmentation with Adaptive Sparse Grids\nYear: 2013\n\nAbstract: We present a novel adaptive sparse grid method for unsupervised image segmentation. The method is based on spectral clustering. The use of adaptive sparse grids achieves that the dimensions of the involved eigensystem do not depend on the number of pixels. In contrast to classical spectral clustering, our sparse-grid variant is therefore able to segment larger images. We evaluate the method on real-world images from the Berkeley Segmentation Dataset. The results indicate that images with 150,000 pixels can be segmented by solving an eigenvalue system of dimensions 500 × 500 instead of 150, 000 × 150, 000.","Title: Cluster optimization and parallelization of simulations with dynamically adaptive grids\nYear: 2013\n\nAbstract: The present paper studies solvers for partial differential equations that work on dynamically adaptive grids stemming from spacetrees. Due to the underlying tree formalism, such grids efficiently can be decomposed into connected grid regions (clusters) on-the-fly. A graph on those clusters classified according to their grid invariancy, workload, multi-core affinity, and further meta data represents the inter-cluster communication. While stationary clusters already can be handled more efficiently than their dynamic counterparts, we propose to treat them as atomic grid entities and introduce a skip mechanism that allows the grid traversal to omit those regions completely. The communication graph ensures that the cluster data nevertheless are kept consistent, and several shared memory parallelization strategies are feasible. A hyperbolic benchmark that has to remesh selected mesh regions iteratively to preserve conforming tessellations acts as benchmark for the present work. We discuss runtime improvements resulting from the skip mechanism and the implications on shared memory performance and load balancing.","Title: Extending a highly parallel data mining algorithm to the intel ® many integrated core architecture\nYear: 2011\n\nAbstract: Extracting knowledge from vast datasets is a major challenge in data-driven applications, such as classification and regression, which are mostly compute bound. In this paper, we extend our SG++ algorithm to the Intel® Many Integrated Core Architecture (Intel® MIC Architecture). The ease of porting an application to Intel MIC Architecture is shown: porting existing SSE code is very easy and straightforward. We evaluate the current prototype pre-release coprocessor board codenamed Intel® \"Knights Ferry\". We utilize the pragma-based offloading programming model offered by the Intel® Composer XE for Intel MIC Architecture, generating both the host and the coprocessor code. We compare the achieved performance with an NVIDIA C2050 accelerator and show that the pre-release Knights Ferry coprocessor delivers better performance than the C2050 and exceeds the C2050 when comparing the productivity aspect of implementing algorithms for the coprocessors.","Title: Shared Memory Parallelization Of Fully-Adaptive Simulations Using A Dynamic Tree-Split And -Join Approach\nYear: 2012\n\nAbstract: In this work we present an approach for the parallelization of hyperbolic simulations on shared-memory architectures running on fully-adaptive grids.We tackle the parallelization problem with a dynamic sub-tree split-and join-approach by running computations on those split sub-trees in parallel using lightweight tasks. The traversal of sub-trees created by tree-splittings is built upon an inherently cache efficient approach for solving hyperbolic PDEs on dynamically adaptive triangular grids using a Sierpinski space filling curve. Our communication scheme among sub-trees stores the exchange-data to/from adjacent sub-trees in a consecutive memory area which is further utilized for an improved run-length-encoded data exchange.To give results for a concrete scenario, we implemented a solver for the shallow water equations which demands for fully-adaptive grid refinement and coarsening after each time-step. Our results give detailed statistics about optimization of the split size, parallelization overhead and also strong scalability results for a simulation running on multi-socket Intel and AMD architectures.","Title: Invasive Compute Balancing for Applications with Hybrid Parallelization\nYear: 2013\n\nAbstract: Achieving high scalability for dynamical adaptive algorithms in HPC is a non-trivial task. In this contex, explicit data migration is typically applied which comes with several challenges such as criteria when and where to migrate data or additional bandwidth requirements. In our work, we use an alternative to data-migration, namely compute-migration on cache-coherent memory systems. We focus on migrating computational power for simulations with dynamical adaptive grids with invasive paradigm. An extension to our existing core-distribution scheduler is presented, distributing the cores depending on the requirements specified by each parallel program instance. In addition, a programming pattern for the hybrid parallel application code is presented to overcome deadlock issues created by applying invasive computing. We validate our approach with benchmark computations for a simulation with artificial workload and furthermore for a realistic scenario based on dynamical adaptive shallow water simulations. Those benchmarks are conducted within a hybrid parallelization environment on a 40-core HPC node. On this shared-memory system, our invasive approach results in faster execution times and higher hardware utilization than the non-invasive approach. Our approach can be applied in general to similar classes of simulations on dynamical adaptive grids.","Title: Space Tree Structures for PDE Software\nYear: 2002\n\nAbstract: In this paper, we study the potential of space trees (boundary extended octrees for an arbitrary number of dimensions) in the context of software for the numerical solution of PDEs. The main advantage of the approach presented is the fact that the underlying geometry's resolution can be decoupled from the computational grid's resolution, although both are organized within the same data structure. This allows us to solve the PDE on a quite coarse orthogonal grid at an accuracy corresponding to a much finer resolution. We show how fast (multigrid) solvers based on the nested dissection principle can be directly implemented on a space tree. Furthermore, we discuss the use of this hierarchical concept as the common data basis for the partitioned solution of coupled problems like fluid-structure interactions, e.g., and we address its suitability for an integration of simulation software.","Title: Fast Insight into High-Dimensional Parametrized Simulation Data\nYear: 2012\n\nAbstract: Numerical simulation has become an inevitable tool in most industrial product development processes with simulations being used to understand the influence of design decisions (parameter configurations) on the structure and properties of the product. However, in order to allow the engineer to thoroughly explore the design space and fine-tune parameters, many -- usually very time-consuming -- simulation runs are necessary. Additionally, this results in a huge amount of data that cannot be analyzed in an efficient way without the support of appropriate tools. In this paper, we address the two-fold problem: First, instantly provide simulation results if the parameter configuration is changed, and, second, identify specific areas of the design space with concentrated change and thus importance. We propose the use of a hierarchical approach based on sparse grid interpolation or regression which acts as an efficient and cheap substitute for the simulation. Furthermore, we develop new visual representations based on the derivative information contained inherently in the hierarchical basis. They intuitively let a user identify interesting parameter regions even in higher-dimensional settings. This workflow is combined in an interactive visualization and exploration framework. We discuss examples from different fields of computational science and engineering and show how our sparse-grid-based techniques make parameter dependencies apparent and how they can be used to fine-tune parameter configurations.","Title: ls1 mardyn: The massively parallel molecular dynamics code for large systems.\nYear: 2014\n\nAbstract: The molecular dynamics simulation code ls1 mardyn is presented. It is a highly scalable code, optimized for massively parallel execution on supercomputing architectures and currently holds the world record for the largest molecular simulation with over four trillion particles. It enables the application of pair potentials to length and time scales that were previously out of scope for molecular dynamics simulation. With an efficient dynamic load balancing scheme, it delivers high scalability even for challenging heterogeneous configurations. Presently, multicenter rigid potential models based on Lennard-Jones sites, point charges, and higher-order polarities are supported. Due to its modular design, ls1 mardyn can be extended to new physical models, methods, and algorithms, allowing future users to tailor it to suit their respective needs. Possible applications include scenarios with complex geometries, such as fluids at interfaces, as well as nonequilibrium molecular dynamics simulation of heat and mass transfer.","Title: GPU Optimization of Pseudo Random Number Generators for Random Ordinary Differential Equations.\nYear: 2014\n\nAbstract: Solving differential equations with stochastic terms involves a massive use of pseudo random numbers. We present an application for the simulation of wireframe buildings under stochastic earthquake excitation. The inherent potential for vectorization of the application is used to its full extent on GPU accelerator hardware. A representative set of pseudo random number gen- erators for uniformly and normally distributed pseudo random numbers has been implemented, optimized, and benchmarked. The resulting optimized variants outperform standard library implementations on GPUs. The techniques and improvements shown in this contribution using the Kanai-Tajimi model can be generalized to other random differential equations or stochastic models as well as other accelerators.","Title: Rendering of Feature-rich Dynamically Changing Volumetric Datasets on GPU.\nYear: 2014\n\nAbstract: Interactive photo-realistic representation of dynamic liquid volumes is a challenging task for today's GPUs and state-of-the-art visualization algorithms. Methods of the last two decades consider either static volumetric datasets applying several optimizations for volume casting, or dynamic volumetric datasets with rough approximations to realistic rendering. Nevertheless, accurate real-time visualization of dynamic datasets is crucial in areas of scientific visualization as well as areas demanding for accurate rendering of feature-rich datasets.","Title: EXAHD: An Exa-scalable Two-Level Sparse Grid Approach for Higher-Dimensional Problems in Plasma Physics and Beyond.\nYear: 2014\n\nAbstract: High-dimensional problems pose a challenge for tomorrow's supercomputing. Problems that require the joint discretization of more dimensions than space and time are among the most compute-hungry ones and thus standard candidates for exascale computing and even beyond. This project tackles such problems by a hierarchical extrapolation approach, the sparse grid combination technique. The method not only enables their treatment in the first place. The hierarchical approach also provides novel ways to deal with central problems in high-performance computing such as scalability and resilience: Global communication can be avoided and reduced to a small subset, and faults can be compensated for without the need for recomputations or checkpoint-restart. As an exemplary prototype for high-dimensional problems, turbulence simulations in plasma physics are studied.","Title: First experiences with group projects in CSE education\nYear: 2006\n\nAbstract: The various educational workshops at CSE-related conferences reveal that there is an ongoing open discussion on how to define the relevant aspects of this discipline, on how to integrate the identified topics into the curricula, and, of course, on how to design suitable course formats -- or briefly, how to teach CSE in an appropriate way. One question more and more addressed in such discussions is to what extent and how methods established in software engineering can or even must be adopted. A second evergreen issue (not restricted to CSE programs, of course) is whether soft skills such as teamwork, project management, or leadership should be taught in special courses or better imparted within suitable CSE-related modules. In this contribution, we report our experiences with project-based and software-focused CSE education. The discussion of expectations and outcomes covers both the software-related issues and the main project-related aspects.","Title: From GPGPU to Many-Core: Nvidia Fermi and Intel Many Integrated Core Architecture\nYear: 2012\n\nAbstract: Comparing the architectures and performance levels of an Nvidia Fermi accelerator with an Intel MIC Architecture coprocessor demonstrates the benefit of the coprocessor for bringing highly parallel applications into, or even beyond, GPGPU performance regions.","Title: Computational Science and Engineering Education: SIAM's Perspective\nYear: 2009\n\nAbstract: This year's SIAM CSE conference featured a structured program on education that sought to attract attention to and sustained interest in this essential topic.","Title: A discriminative approach to polyphonic piano note transcription using supervised non-negative matrix factorization\nYear: 2013\n\nAbstract: We introduce a novel method for the transcription of polyphonic piano music by discriminative training of support vector machines (SVMs). As features, we use pitch activations computed by supervised non-negative matrix factorization from low-level spectral features. Different approaches to low-level feature extraction, NMF dictionary learning and activation feature extraction are analyzed in a large-scale evaluation on eight hours of piano music including synthesized and real recordings. We conclude that the proposed method delivers state-of-the-art results and clearly outperforms SVMs using simple spectral features.","Title: Data mining on vast data sets as a cluster system benchmark\nYear: 2016\n\nAbstract: Comparing different accelerated cluster architectures by a single application is a tough piece of work because this application has to be optimized with respect to platform-dependent features. In this work, we demonstrate such an optimization for a data mining algorithm which solves regression and classification problems on vast data sets. Our technique is based on least squares regression, and its major component is the iterative matrix-free solution of a linear system of equations. By processing data sets ranging from several hundreds of thousands instances to multi-million data points in strong-scaling and weak-scaling settings, we are able to estimate the amount of parallelism needed to unleash the performance of classic CPU-based machines and clusters employing Intel Xeon Phi coprocessors and NVIDIA Kepler GPUs. Only in strong-scaling experiments, GPUs and coprocessors suffer from their tremendous amount of needed parallelism and get outperformed by dual socket Intel Sandy Bridge nodes at large scale more than 64 nodes/accelerators. However, in weak-scaling scenarios, a speed-up larger than 2X over an entire CPU node can be achieved by a single accelerator. Copyright © 2015 John Wiley & Sons, Ltd.","Title: MaMiCo: Software design for parallel molecular-continuum flow simulations.\nYear: 2016\n\nAbstract: The macro–micro-coupling tool (MaMiCo) was developed to ease the development of and modularize molecular-continuum simulations, retaining sequential and parallel performance. We demonstrate the functionality and performance of MaMiCo by coupling the spatially adaptive Lattice Boltzmann framework waLBerla with four molecular dynamics (MD) codes: the light-weight Lennard-Jones-based implementation SimpleMD, the node-level optimized software ls1 mardyn, and the community codes ESPResSo and LAMMPS. We detail interface implementations to connect each solver with MaMiCo. The coupling for each waLBerla-MD setup is validated in three-dimensional channel flow simulations which are solved by means of a state-based coupling method. We provide sequential and strong scaling measurements for the four molecular-continuum simulations. The overhead of MaMiCo is found to come at 10%–20% of the total (MD) runtime. The measurements further show that scalability of the hybrid simulations is reached on up to 500 Intel SandyBridge, and more than 1000 AMD Bulldozer compute cores.","Title: Block-structured grids for Eulerian gyrokinetic simulations.\nYear: 2016\n\nAbstract: In order to predict the turbulent transport in magnetic fusion experiments, global (i.e., full-torus) gyrokinetic simulations are often carried out. In this context, one frequently encounters situations in which the plasma temperature varies by large factors across the radial simulation domain. In grid-based Eulerian codes, this enforces the use of a very large number of grid points in the two-dimensional velocity space, and, thus, an enormous computational effort. To minimize the computational requirements, one may employ block-structured grids, adapted to the radial changes of the temperature. As the block-structured grids rely on a general approach, they can be applied to different Eulerian gyrokinetic implementations. In this paper, we explain the construction and implementation of such grids in the gyrokinetic code GENE, F. Jenko et al. (2000), and present corresponding simulation results.","Title: Mass-spring systems for simulating mitral valve repair using 3D ultrasound images.\nYear: 2015\n\nAbstract: Mitral valve (MV) diseases are among the most common types of heart diseases, while heart diseases are the most common cause of death worldwide. MV repair surgery is connected to higher survival rates and fewer complications than the total replacement of the MV, but MV repair requires extensive patient-specific therapy planning. The simulation of MV repair with a patient-specific model could help to optimize surgery results and make MV repair available to more patients. However, current patient-specific simulations are difficult to transfer to clinical application because of time-constraints or prohibitive requirements on the resolution of the image data. As one possible solution to the problem of patient-specific MV modeling, we present a mass-spring MV model based on 3D transesophageal echocardiographic (TEE) images already routinely acquired for MV repair therapy planning. Our novel approach to the rest-length estimation of springs allows us to model the global support of the MV leaflets through the chordae tendinae without the need for high-resolution image data. The model is used to simulate MV annuloplasty for five patients undergoing MV repair, and the simulated results are compared to post-surgical TEE images. The comparison shows that our model is able to provide a qualitative estimate of annuloplasty surgery. In addition, the data suggests that the model might also be applied to simulating the implantation of artificial chordae. (C) 2015 Elsevier Ltd. All rights reserved.","Title: Reduced models for sparse grid discretizations of the multi-asset Black-Scholes equation\nYear: 2015\n\nAbstract: This work presents reduced models for pricing basket options with the Black-Scholes and the Heston model. Basket options lead to multi-dimensional partial differential equations (PDEs) that quickly become computationally infeasible to discretize on full tensor grids. We therefore rely on sparse grid discretizations of the PDEs, which allow us to cope with the curse of dimensionality to some extent. We then derive reduced models with proper orthogonal decomposition. Our numerical results with the Black-Scholes model show that sufficiently accurate results are achieved while gaining speedups between 80 and 160 compared to the high-fidelity sparse grid model for 2-, 3-, and 4-asset options. For the Heston model, results are presented for a single-asset option that leads to a two-dimensional pricing problem, where we achieve significant speedups with our model reduction approach based on high-fidelity sparse grid models.","Title: A Multigrid Method for Adaptive Sparse Grids\nYear: 2015\n\nAbstract: Sparse grids have become an important tool to reduce the number of degrees of freedom of discretizations of moderately high-dimensional partial differential equations; however, the reduction in degrees of freedom comes at the cost of an almost dense and unconventionally structured system of linear equations. To guarantee overall efficiency of the sparse grid approach, special linear solvers are required. We present a multigrid method that exploits the sparse grid structure to achieve an optimal runtime that scales linearly with the number of sparse grid points. Our approach is based on a novel decomposition of the right-hand sides of the coarse grid equations that leads to a reformulation in so-called auxiliary coefficients. With these auxiliary coefficients, the right-hand sides can be represented in a nodal point basis on low-dimensional full grids. Our proposed multigrid method directly operates in this auxiliary coefficient representation, circumventing most of the computationally cumbersome sparse grid structure. Numerical results on nonadaptive and spatially adaptive sparse grids confirm that the runtime of our method scales linearly with the number of sparse grid points and they indicate that the obtained convergence factors are bounded independently of the mesh width.","Title: Dynamically adaptive Lattice Boltzmann simulation of shallow water flows with the Peano framework\nYear: 2015\n\nAbstract: We present a dynamically adaptive Lattice Boltzmann (LB) implementation for solving the shallow water equations (SWEs). Our implementation extends an existing LB component of the Peano framework. We revise the modular design with respect to the incorporation of new simulation aspects and LB models. The basic SWE-LB implementation is validated in different breaking dam scenarios. We further provide a numerical study on stability of the MRT collision operator used in our simulations.","Title: Modelling social identification and helping in evacuation simulation.\nYear: 2016\n\nAbstract: •An algorithmic formulation of empirical findings from social psychology is presented.•The model is qualitatively validated against data from London bombings.•With uncertainty quantification the impact of the model parameters is analysed.•Simulations with the model can lead to a better estimation of evacuation times.","Title: Invasive Compute Balancing for Applications with Shared and Hybrid Parallelization\nYear: 2015\n\nAbstract: Abstract Achieving high scalability with dynamically adaptive algorithms in high-performance computing (HPC) is a non-trivial task. The invasive paradigm using compute migration represents an efficient alternative to classical data migration approaches for such algorithms in HPC. We present a core-distribution scheduler which realizes the migration of computational power by distributing the cores depending on the requirements specified by one or more parallel program instances. We validate our approach with different benchmark suites for simulations with artificial workload as well as applications based on dynamically adaptive shallow water simulations, and investigate concurrently executed adaptivity parameter studies on realistic Tsunami simulations. The invasive approach results in significantly faster overall execution times and higher hardware utilization than alternative approaches. A dynamic resource management is therefore mandatory for a more efficient execution of scenarios similar to our simulations, e.g. several Tsunami simulations in urgent computing, to overcome strong scalability challenges in the area of HPC. The optimizations obtained by invasive migration of cores can be generalized to similar classes of algorithms with dynamic resource requirements.","Title: Invasive computing in HPC with X10\nYear: 2013\n\nAbstract: High performance computing with thousands of cores relies on distributed memory due to memory consistency reasons. The resource management on such systems usually relies on static assignment of resources at the start of each application. Such a static scheduling is incapable of starting applications with required resources being used by others since a reduction of resources assigned to applications without stopping them is not possible. This lack of dynamic adaptive scheduling leads to idling resources until the remaining amount of requested resources gets available. Additionally, applications with changing resource requirements lead to idling or less efficiently used resources. The invasive computing paradigm suggests dynamic resource scheduling and applications able to dynamically adapt to changing resource requirements. As a case study, we developed an invasive resource manager as well as a multigrid with dynamically changing resource demands. Such a multigrid has changing scalability behavior during its execution and requires data migration upon reallocation due to distributed memory systems. To counteract the additional complexity introduced by the additional interfaces, e. g. for data migration, we use the X10 programming language for improved programmability. Our results show improved application throughput and the dynamic adaptivity. In addition, we show our extension for the distributed arrays of X10 to support data migration.","Title: Optimized Force Calculation in Molecular Dynamics Simulations for the Intel Xeon Phi.\nYear: 2015\n\nAbstract: We provide details on the shared-memory parallelization for manycore architectures of the molecular dynamics framework ls1-mardyn, including an optimization of the SIMD vectorization for multi-centered molecules. The novel shared-memory parallelization scheme allows to retain Newton's third law optimization and exhibits very good scaling on many-core devices such as a full Xeon Phi card running 240 threads. The Xeon Phi can thus be exploited and delivers comparable performance as IvyBridge nodes in our experiments.","Title: Research and Education in Computational Science and Engineering.\nYear: 2018\n\nAbstract: This report presents challenges, opportunities, and directions for computational science and engineering (CSE) research and education for the next decade. Over the past two decades the field of CSE has penetrated both basic and applied research in academia, industry, and laboratories to advance discovery, optimize systems, support decision-makers, and educate the scientific and engineering workforce. Informed by centuries of theory and experiment, CSE performs computational experiments to answer questions that neither theory nor experiment, alone is equipped to answer. CSE provides scientists and engineers with algorithmic inventions and software systems that, transcend disciplines and scales. CSE brings the power of parallelism to bear on troves of data. Mathematics-based advanced computing has become a prevalent means of discovery and innovation in essentially all areas of science, engineering, technology, and society, and the CSE community is at the core of this transformation. However, a combination of disruptive developments-including the architectural complexity of extreme-scale computing, the data revolution and increased attention to data-driven discovery, and the specialization required to follow the applications to new frontiers-is redefining the scope and reach of the CSE endeavor. With these many current and expanding opportunities for the CSE field, there is a growing demand for CSE graduates and a need to expand CSE educational offerings. This need includes CSE programs at both the undergraduate and graduate levels, as well as continuing education and professional development programs, exploiting the synergy between computational science and data science. Yet, as institutions consider new and evolving educational programs, it is essential to consider the broader research challenges and opportunities that provide the context for CSE education and workforce development.","Title: A parallel arbitrary-order accurate AMR algorithm for the scalar advection-diffusion equation.\nYear: 2016\n\nAbstract: We present a numerical method for solving the scalar advection-diffusion equation using adaptive mesh refinement. Our solver has three unique characteristics: (1) it supports arbitrary-order accuracy in space; (2) it allows different discretizations for the velocity and scalar advected quantity; (3) it combines the method of characteristics with an integral equation formulation; and (4) it supports shared and distributed memory architectures. In particular, our solver is based on a second-order accurate, unconditionally stable, semi-Lagrangian scheme combined with a spatially-adaptive Chebyshev octree for discretization. We study the convergence, single-node performance, strong scaling, and weak scaling of our scheme for several challenging flows that cannot be resolved efficiently without using high-order accurate discretizations. For example, we consider problems for which switching from 4th order to 14th order approximation results in two orders of magnitude speedups for a computation in which we keep the target accuracy in the solution fixed. For our largest run, we solve a problem with one billion unknowns on a tree with maximum depth equal to 10 and using 14th-order elements on 16,384 ×86 cores on the \\\"STAMPEDE\\\" system at the Texas Advanced Computing Center.","Title: Infrastructure and API Extensions for Elastic Execution of MPI Applications.\nYear: 2016\n\nAbstract: Dynamic Processes support was added to MPI in version 2.0 of the standard. This feature of MPI has not been widely used by application developers in part due to the performance cost and limitations of the spawn operation. In this paper, we propose an extension to MPI that consists of four new operations. These operations allow an application to be initialized in an elastic mode of execution and enter an adaptation window when necessary, where resources are incorporated into or released from the application's world communicator. A prototype solution based on the MPICH library and the SLURM resource manager is presented and evaluated alongside an elastic scientific application that makes use of the new MPI extensions. The cost of these new operations is shown to be negligible due mainly to the latency hiding design, leaving the application's time for data redistribution as the only significant performance cost.","Title: Evaluation of an Efficient Stack-RLE Clustering Concept for Dynamically Adaptive Grids.\nYear: 2016\n\nAbstract: One approach for tackling the challenge of efficient implementations for parallel PDE simulations on dynamically changing grids is the usage of space-filling curves (SFCs). While SFC algorithms possess advantageous properties such as low memory requirements and close-to-optimal partitioning approaches with linear complexity, they require efficient communication strategies for keeping and utilizing the connectivity information, in particular for dynamically changing grids. Our approach is to use a sparse communication graph to store the connectivity information and to transfer data blockwise. This permits efficient generation of multiple partitions per memory context (denoted by cl us tering),which-in combination with a run-length encoding (RLE)-directly leads to elegant solutions for shared, distributed, and hybrid parallelization and allows cluster-based optimizations. While previous work focused on specific aspects, we present in this paper an overall compact summary of the stack-RLE clustering approach complete with aspects of the vertex-based communication that facilitate understanding the approach. The central contribution of this work is the proof of suitability of the stack-RLE clustering approach for an efficient realization of different, relevant building blocks of scientific computing methodology and real-life computer science and engineering (CSE) applications: We show 95% strong scalability for small-scale scalability benchmarks on 512 cores and weak scalability of over 90% on 8192 cores for finite-volume solvers and changing grid structure in every time step; optimizations of simulation data backends by writer tasks; comparisons of analytical benchmarks to analyze the adaptivity criteria; and a tsunami simulation as a representative real-world showcase of a wave propagation for our approach which reduces the overall workload by 95% for parallel fully adaptive mesh refinement and, based on a comparison with SFC-ordered regular grid cells, reduces the computation time by a factor of 7.6 with improved results and a factor of 62.2 with results of similar accuracy of buoy station data.","Title: Steady-State Anderson Accelerated Coupling Of Lattice Boltzmann And Navier-Stokes Solvers\nYear: 2016\n\nAbstract: We present an Anderson acceleration-based approach to spatially couple three-dimensional Lattice Boltzmann and Navier-Stokes (LBNS) flow simulations. This allows to locally exploit the computational features of both fluid flow solver approaches to the fullest extent and yields enhanced control to match the LB and NS degrees of freedom within the LBNS overlap layer. Designed for parallel Schwarz coupling, the Anderson acceleration allows for the simultaneous execution of both Lattice Boltzmann and Navier-Stokes solver. We detail our coupling methodology, validate it, and study convergence and accuracy of the Anderson accelerated coupling, considering three steady-state scenarios: plane channel flow, flow around a sphere and channel flow across a porous structure. We find that the Anderson accelerated coupling yields a speed-up (in terms of iteration steps) of up to 40% in the considered scenarios, compared to strictly sequential Schwarz coupling.","Title: Load Balancing for Molecular Dynamics Simulations on Heterogeneous Architectures\nYear: 2016\n\nAbstract: Upcoming exascale compute systems are expected to be built from heterogeneous hardware architectures. According to this trend, there exist various methods to handle clusters composed of CPUs, GPUs or other accelerators. Most of these assume that each node has the same structure-for example a dual socket system with an accelerator (GPU or Xeon Phi). The workload is then distributed homogeneously among the nodes. However, not all clusters fulfill this requirement. They might contain different partitions with and without accelerators. Furthermore, depending on the underlying problem to be solved, accelerator cards may perform better in native mode compared to offloading. Besides, various aspects such as cooling may influence the performance of individual nodes. It therefore cannot always be assumed, that the structure and performance of each node and hence the performance of every MPI rank is the same. In this contribution, we apply a k-d tree decomposition method to balance load on heterogeneous compute clusters. The algorithm is incorporated into the molecular dynamics simulation program ls1 mardyn. We present performance results for simulations executed on hybrid AMD Bulldozer-Intel Sandy Bridge, Intel Westmere-Intel Sandy Bridge and Intel Xeon-Intel Xeon Phi-architectures. The only prerequisite for the proposed algorithm is a cost estimation for different decompositions. It is hence expected to be applicable to a variety of n-body scenarios, for which a domain decomposition is possible.","Title: Block-structured grids in full velocity space for Eulerian gyrokinetic simulations.\nYear: 2017\n\nAbstract: Global, i.e., full-torus, gyrokinetic simulations play an important role in exploring plasma microturbulence in magnetic fusion devices with strong radial variations. In the presence of steep temperature profiles, grid-based Eulerian approaches can become quite challenging as the correspondingly varying velocity space structures need to be accommodated and sufficiently resolved. A rigid velocity space grid then requires a very high number of discretization nodes resulting in enormous computational costs. To tackle this issue and reduce the computational demands, we introduce block-structured grids in the all velocity space dimensions. The construction of these grids is based on a general approach, making them suitable for various Eulerian implementations. In the current study, we explain the rationale behind the presented approach, detail the implementation, and provide simulation results obtained with the block-structured grids. The achieved reduction in the number of computational nodes depends on the temperature profile and simulation scenario provided. In the test cases at hand, about ten times fewer grid points are required for nonlinear simulations performed with block-structured grids in the plasma turbulence code GENE (http://genecode.org). With the speed-up found to scale almost exactly reciprocal to the number of grid points, the new implementation greatly reduces the computational costs and therefore opens new possibilities for applications of this software package.","Title: High performance shallow water kernels for parallel overland flow simulations based on FullSWOF2D.\nYear: 2017\n\nAbstract: We describe code optimization and parallelization procedures applied to the sequential overland flow solver FullSWOF2D. Major difficulties when simulating overland flows comprise dealing with high resolution datasets of large scale areas which either cannot be computed on a single node either due to limited amount of memory or due to too many (time step) iterations resulting from the CFL condition. We address these issues in terms of two major contributions. First, we demonstrate a generic step-by-step transformation of the second order finite volume scheme in FullSWOF2D towards MPI parallelization. Second, the computational kernels are optimized by the use of templates and a portable vectorization approach. We discuss the load imbalance of the flux computation due to dry and wet cells and propose a solution using an efficient cell counting approach. Finally, scalability results are shown for different test scenarios along with a flood simulation benchmark using the Shaheen II supercomputer.","Title: A Large-Scale Malleable Tsunami Simulation Realized on an Elastic MPI Infrastructure.\nYear: 2017\n\nAbstract: Realization of resource awareness and elasticity in hardware and software is an answer to many problems and challenges we are facing in High Performance Computing (HPC) today. Resource utilization inefficiency is a real problem in current HPC systems due to the current static, inflexible resource assignment configuration. One way to resolve this problem is to change the static resource assignment setting---by introducing runtime resource elasticity, which requires both malleability in software implementation and support for runtime resource adaptation in the system infrastructure. In this paper, we show a successful implementation of a malleable tsunami simulation realized on an elastic MPI infrastructure we previously proposed. We also prove that introducing malleability to such a tightly coupled parallel application can be beneficial.","Title: A Massively-Parallel, Fault-Tolerant Solver for High-Dimensional PDEs.\nYear: 2016\n\nAbstract: We investigate the effect of hard faults on a massively-parallel implementation of the Sparse Grid Combination Technique (SGCT), an efficient numerical approach for the solution of high-dimensional time-dependent PDEs. The SGCT allows us to increase the spatial resolution of a solver to a level that is out of scope with classical discretization schemes due to the curse of dimensionality. We exploit the inherent data redundancy of this algorithm to obtain a scalable and fault-tolerant implementation without the need of checkpointing or process replication. It is a lossy approach that can guarantee convergence for a large number of faults and a wide range of applications. We present first results using our fault simulation framework - and the first convergence and scalability results with simulated faults and algorithm-based fault tolerance for PDEs in more than three dimensions.","Title: A highly scalable, algorithm-based fault-tolerant solver for gyrokinetic plasma simulations\nYear: 2017\n\nAbstract: With future exascale computers expected to have millions of compute units distributed among thousands of nodes, system faults are predicted to become more frequent. Fault tolerance will thus play a key role in HPC at this scale. In this paper we focus on solving the 5-dimensional gyrokinetic Vlasov-Maxwell equations using the application code GENE as it represents a high-dimensional and resource-intensive problem which is a natural candidate for exascale computing. We discuss the Fault-Tolerant Combination Technique, a resilient version of the Combination Technique, a method to increase the discretization resolution of existing PDE solvers. For the first time, we present an efficient, scalable and fault-tolerant implementation of this algorithm for plasma physics simulations based on a manager-worker model and test it under very realistic and pessimistic environments with simulated faults. We show that the Fault-Tolerant Combination Technique - an algorithm-based forward recovery method - can tolerate a large number of faults with a low overhead and at an acceptable loss in accuracy. Our parallel experiments with up to 32k cores show good scalability at a relative parallel efficiency of 93.61%. We conclude that algorithm-based solutions to fault tolerance are attractive for this type of problems.","Title: Challenges in Creating a Sustainable Generic Research Data Infrastructure.\nYear: 2017\n\nAbstract: Research data management is of the utmost importance in a world where research data is created with an ever increasing amount and rate and with a high variety across all scientific disciplines. This paper especially discusses software engineering challenges stemming from creating a long-living software system. It aims at providing a reference implementation for a federated research data infrastructure including interconnected individual repositories for communities and an overarching search based on metadata. The challenges involve a high variety of evolving requirements, the management and development of the distributed and federated infrastructure that are based on existing components, the piloting within the use cases, the efficient training of users, and how to enable the future sustainable operation.","Title: A Holistic Scalable Implementation Approach Of The Lattice Boltzmann Method For Cpu/Gpu Heterogeneous Clusters\nYear: 2017\n\nAbstract: Heterogeneous clusters are a widely utilized class of supercomputers assembled from different types of computing devices, for instance CPUs and GPUs, providing a huge computational potential. Programming them in a scalable way exploiting the maximal performance introduces numerous challenges such as optimizations for different computing devices, dealing with multiple levels of parallelism, the application of different programming models, work distribution, and hiding of communication with computation. We utilize the lattice Boltzmann method for fluid flow as a representative of a scientific computing application and develop a holistic implementation for large-scale CPU/GPU heterogeneous clusters. We review and combine a set of best practices and techniques ranging from optimizations for the particular computing devices to the orchestration of tens of thousands of CPU cores and thousands of GPUs. Eventually, we come up with an implementation using all the available computational resources for the lattice Boltzmann method operators. Our approach shows excellent scalability behavior making it future-proof for heterogeneous clusters of the upcoming architectures on the exaFLOPS scale. Parallel efficiencies of more than 90% are achieved leading to 2604.72 GLUPS utilizing 24,576 CPU cores and 2048 GPUs of the CPU/GPU heterogeneous cluster Piz Daint and computing more than 6.8 x 10(9) lattice cells.","Title: Nonintrusive Uncertainty Analysis of Fluid-Structure Interaction with Spatially Adaptive Sparse Grids and Polynomial Chaos Expansion.\nYear: 2018\n\nAbstract: The propagation of uncertainty in physical parameters of fluid-structure interaction problems is a challenging task-both mathematically and in terms of computational workload. In this paper, we employ nonintrusive polynomial chaos expansion and model the uncertainty in five independent input parameters that characterize both fluid and structure. We propose a novel methodology to compute the expansion's coefficients using spatially adaptive sparse grids and products of one-dimensional integrals, which exploits the tensor structure of both sparse grids and probabilistic space. Furthermore, with spatial adaptivity and modified basis functions, we keep the number of sparse grid points small. We test our approach in two test cases: (i) an elastic vertical flap in a channel flow and (ii) a computationally challenging, well-established benchmark. The outputs of interest are the x-deflection and total force on the structure. In the first test case, we consider six implementations of our methodology and two established methods based on sparse grid quadrature. We evaluate the performance (in terms of number of runs of the numerical solver) and accuracy of all eight methods. The results show that one variant of our approach outperforms all the other implementations. We apply our best variant to the benchmark scenario and address the high computational demands of the resulting problem using three levels of parallelism. Importantly, our approach is not restricted to fluid-structure interaction problems. We can address a broad spectrum of computationally expensive problems, provided that sparse grid approximations can be employed.","Title: Optimization of a Sparse Grid-Based Data Mining Kernel for Architectures Using AVX-512\nYear: 2018\n\nAbstract: Sparse grids have already been successfully used in various high-performance computing (HPC) applications, including data mining. In this article, we take a legacy classification kernel previously optimized for the AVX2 instruction set and investigate the benefits of using the newer AVX-S12-based multi-and many-core architectures. In particular, the Knights Landing (KNL) processor is used to study the possible performance gains of the code. Not all kernels benefit equally from such architectures, therefore choices in optimization steps and KNL cluster and memory modes need to be filtered through the lens of the code implementation at hand. With a less traditional approach of manual vectorization through instruction-level intrinsics, our kernel provides a differently faceted look into the optimization process. Observations stem from results obtained for node-and cluster-level classification simulations with up to 2\n<sup xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">^</sup>\n28 multidimensional training data points, using the CooLMUC-3cluster of the Leibniz Supercomputing Center (LRZ) in Garching, Germany.","Title: Scheduling parallel eigenvalue computations in a quantum chemistry code\nYear: 2010\n\nAbstract: The application of High Performance Computing to Quantum Chemical (QC) calculations faces many challenges. A central step is the solution of the generalized eigenvalue problem of a Hamilton matrix. Although in many cases its execution time is small relative to other numerical tasks, its complexity of O(N3) is higher, thus more significant in larger applications. For parallel QC codes, it therefore is advantageous to have a scalable solver for this step. We investigate the case where the symmetry of a molecule leads to a block-diagonal matrix structure, which complicates an efficient use of available parallel eigensolvers. We present a technique which employs a malleable parallel task scheduling (MPTS) algorithm to schedule instances of sequential and parallel eigensolver routines from LAPACK and ScaLAPACK. In this way, an efficient use of hardware resources is guaranteed while overall scalability is facilitated. Finally, we evaluate the proposed technique for electronic structure calculations of real chemical systems. For the systems considered, the performance was improved by factors of up to 8.4, compared to the previously used, nonmalleable parallel scheduling approach.","Title: TweTriS: Twenty trillion-atom simulation\nYear: 2019\n\nAbstract: AbstractSignificant improvements are presented for the molecular dynamics code ls1 mardyn — a linked cell-based code for simulating a large number of small, rigid molecules with application areas in chemical engineering. The changes consist of a redesign of the SIMD vectorization via wrappers, MPI improvements and a software redesign to allow memory-efficient execution with the production trunk to increase portability and extensibility. Two novel, memory-efficient OpenMP schemes for the linked cell-based force calculation are presented, which are able to retain Newton’s third law optimization. Comparisons to well-optimized Verlet list-based codes, such as LAMMPS and GROMACS, demonstrate the viability of the linked cell-based approach. The present version of ls1 mardyn is used to run simulations on entire supercomputers, maximizing the number of sampled atoms. Compared to the preceding version of ls1 mardyn on the entire set of 9216 nodes of SuperMUC, Phase 1, 27% more atoms are simulated. Weak scaling performance is increased by up to 40% and strong scaling performance by up to more than 220%. On Hazel Hen, strong scaling efficiency of up to 81% and 189 billion molecule updates per second is attained, when scaling from 8 to 7168 nodes. Moreover, a total of 20 trillion atoms is simulated at up to 88% weak scaling efficiency running at up to 1.33 PFLOPS. This represents a fivefold increase in terms of the number of atoms simulated to date.","Title: AutoPas: Auto-Tuning for Particle Simulations\nYear: 2019\n\nAbstract: The C++ library AutoPas aims at delivering optimal node-level performance for particle simulations. This paper describes the internally implemented algorithms, and how the library uses auto-tuning to dynamically select their optimal combination at run-time. Results are presented, which show that all available algorithms and configuration options have their specific advantages. To demonstrate the library's capabilities in relevant application settings, it has been integrated into the software package ls1 mardyn. An example of a realistic molecular dynamics simulation from thermodynamics is shown in which AutoPas detects a change in the best possible algorithm configuration. It adapts the simulation algorithm accordingly, sustaining optimal performance without additional user input.","Title: A Generalized Spatially Adaptive Sparse Grid Combination Technique With Dimension-Wise Refinement\nYear: 2021\n\nAbstract: Today, high-dimensional calculations can be found in almost all scientific disciplines. The application of machine learning and uncertainty quantification methods are common examples where high-dimensional problems appear. Typically, these problems are computationally expensive or even infeasible on current machines due to the curse of dimensionality. The sparse grid combination technique is one method to mitigate this effect, but it still does not generate optimal grids for many application scenarios. In such cases, adaptivity strategies are applied to further optimize the grid generation. Generally, adaptive grid generation strategies can be be classified as spatially or dimensionally adaptive. One of the most prominent examples is the dimension-adaptive combination technique, which is easy to implement and suitable for cases where dimensions contribute in different magnitudes to the accuracy of the solution. Unfortunately, spatial adaptivity is not possible in the standard combination technique due to the required regular structure of the grids. We therefore propose a new algorithmic variant of the combination technique that is based on the combination of rectilinear grids which can adapt themselves locally to the target function using 1-dimensional refinements. We further increase the efficiency by adjusting point levels with tree rebalancing. Results for numerical quadrature and interpolation show that we can significantly improve upon the standard combination technique and compete with or even surpass common spatially adaptive implementations with sparse grids. At the same time our method keeps the black-box property of the combination technique which makes it possible to apply it to black-box solvers.","Title: N ways to simulate short-range particle systems: Automated algorithm selection with the node-level library AutoPas\nYear: 2022\n\nAbstract: AutoPas is an open-source C++ library delivering optimal node-level performance by providing the ideal algorithmic configuration for an arbitrary scenario in a given short-range particle simulation. It acts as a black-box container, internally implementing an extensive set of algorithms, parallelization strategies, and optimizations that are combined dynamically according to the state of the simulation via auto-tuning. This paper gives an overview of the high-level user perspective, as well as the internal view, covering the implemented techniques and features. The library is showcased by incorporating it into the codes LAMMPS and ls1 mardyn, and by investigating various applications. We further outline node-level shared-memory performance and scalability of our auto-tuning software which is comparable to LAMMPS. (C) 2021 Elsevier B.V. All rights reserved."],"string":"[\n \"Title: A high-level Fortran interface to parallel matrix algebra\\nYear: 2012\\n\\nAbstract: Scientific software often relies on complex vector- and matrix algebra, sometimes combined with more elaborate linear algebra problems, such as systems of linear equations or eigenvalue problems. Its efficient parallel computation is well studied, and there is a variety of parallel libraries for this task available. However, the user still has to cope with many technical details, such as library-specific data distribution or complicated routine calls, which, if not carefully used, easily result in confusing, difficult to maintain code. This paper presents a Fortran 90 programming interface and its library implementation, which specifies easy-to-use distributed data objects, combined with a set of linear algebra operators. The interface allows matrix algebra to be expressed in a high-level mathematical notation, while operating internally with parallel routines. Thus, linear algebra can be expressed in a clear, comprehensible way, while low-level tasks, such as data organization and communication, are hidden from the user. We evaluate this approach with an existing code from relativistic quantum chemistry, which includes complex operations with dense- and diagonal matrices, solving the generalized eigenvalue problem and several domain specific operations. The results show that the required relativistic transformations can be implemented quickly, while producing clear, light-weight code which scales up to 81 processor cores with efficiency of about 50% for our test cases.\",\n \"Title: Fast GPGPU Data Rearrangement Kernels using CUDA\\nYear: 2010\\n\\nAbstract: Many high performance-computing algorithms are bandwidth limited, hence the\\nneed for optimal data rearrangement kernels as well as their easy integration\\ninto the rest of the application. In this work, we have built a CUDA library of\\nfast kernels for a set of data rearrangement operations. In particular, we have\\nbuilt generic kernels for rearranging m dimensional data into n dimensions,\\nincluding Permute, Reorder, Interlace/De-interlace, etc. We have also built\\nkernels for generic Stencil computations on a two-dimensional data using\\ntemplates and functors that allow application developers to rapidly build\\ncustomized high performance kernels. All the kernels built achieve or surpass\\nbest-known performance in terms of bandwidth utilization.\",\n \"Title: Hybrid molecular-continuum methods: From prototypes to coupling software\\nYear: 2014\\n\\nAbstract: In this contribution, we review software requirements in hybrid molecular-continuum simulations. For this purpose, we analyze a prototype implementation which combines two frameworks-the Molecular Dynamics framework MarDyn and the framework Peano for spatially adaptive mesh-based simulations-and point out particular challenges of a general coupling software. Based on this analysis, we discuss the software design of our recently published coupling tool. We explain details on its overall structure and show how the challenges that arise in respective couplings are resolved by the software.\",\n \"Title: Numerical Simulation of Particle Transport in a Drift Ratchet\\nYear: 2008\\n\\nAbstract: The directed transport of microparticles depending on their size is the basis for particle sorting methods that are of utmost importance in, for example, life sciences. A drift ratchet is a Brownian motor that allows for such a directed transport. Hereby, the particle motion is induced by a combination of the Brownian motion and asymmetries stemming, for example, from the domain's geometry, electrical fields, or transient pressure boundary conditions. We simulate a particular drift ratchet which consists of a matrix of pores with asymmetrically oscillating diameter wherein a fluid with suspended particles is pumped forward and backward, and where the particles' long-term transport direction depends on their size. Thus, this setup allows for continuous and parallel particle separation, which has been shown experimentally. However, for a deeper understanding and for an optimized parameters' choice, further investigations, i.e., simulations, are necessary. In this paper, we present first results achieved with our parallel three-dimensional simulation codes applied to a still simplified scenario. This simplification is necessary to isolate different phenomena (e.g., asymmetries and Brownian motion) to check their relevance for the particle transport. The simulation codes are based on (adaptive) Cartesian grids in combination with finite volume and finite element discretizations. Cartesian grids allow for a very efficient implementation of the solver algorithms and an efficient balanced parallelization via domain decomposition. The achieved simulation results show the effectiveness of our approach and give some strong hints on a directed particle transport already with the simplified model we used here.\",\n \"Title: Option pricing with a direct adaptive sparse grid approach\\nYear: 2012\\n\\nAbstract: We present an adaptive sparse grid algorithm for the solution of the Black-Scholes equation for option pricing, using the finite element method. Sparse grids enable us to deal with higher-dimensional problems better than full grids. In contrast to common approaches that are based on the combination technique, which combines different solutions on anisotropic coarse full grids, the direct sparse grid approach allows for local adaptive refinement. When dealing with non-smooth payoff functions, this reduces the computational effort significantly. In this paper, we introduce the spatially adaptive discretization of the Black-Scholes equation with sparse grids and describe the algorithmic structure of the numerical solver. We present several strategies for adaptive refinement, evaluate them for different dimensionalities, and demonstrate their performance showing numerical results.\",\n \"Title: Adaptive Sparse Grid Techniques for Data Mining.\\nYear: 2006\\n\\nAbstract: It was shown in [GaGT01] that the task of classification in data mining can be tackled by employing ansatz functions associated to grid points in the (often high dimensional) feature-space rather than using data-centered ansatz functions. To cope with the curse of dimensionality, sparse grids have been used. Based on this approach we propose an efficient finite-element-like discretization technique for classification instead of the combination technique used in [GaGT01]. The main goal of our method is to make use of adaptivity to further reduce the number of grid points needed. Employing adaptivity in classification is reasonable as the target function contains smooth regions as well as rough ones. Regarding implementational issues we present an algorithm for the fast multiplication of the vector of unknowns with the coefficient matrix. We give an example for the adaptive selection of grid points and show that special care has to be taken regarding the boundary values, as adaptive techniques commonly used for solving PDEs are not optimal here. Results for some typical classification tasks, including a problem from the UCI repository, are presented.\",\n \"Title: A parallel adaptive cartesian PDE solver using space–filling curves\\nYear: 2006\\n\\nAbstract: In this paper, we present a parallel multigrid PDE solver working on adaptive hierarchical cartesian grids. The presentation is restricted to the linear elliptic operator of second order, but extensions are possible and have already been realised as prototypes. Within the solver the handling of the vertices and the degrees of freedom associated to them is implemented solely using stacks and iterates of a Peano space–filling curve. Thus, due to the structuredness of the grid, two administrative bits per vertex are sufficient to store both geometry and grid refinement information. The implementation and parallel extension, using a space–filling curve to obtain a load balanced domain decomposition, will be formalised. In view of the fact that we are using a multigrid solver of linear complexity $\\\\mathcal{O}(n)$, it has to be ensured that communication cost and, hence, the parallel algorithm's overall complexity do not exceed this linear behaviour.\",\n \"Title: DaStGen--A Data Structure Generator for Parallel C++ HPC Software\\nYear: 2008\\n\\nAbstract: Simulation codes often suffer from high memory requirements. This holds in particular if they are memory-bounded, and, with multicore systems coming up, the problem will become even worse as more and more cores have to share the memory connections. To optimise data structures with respect to memory manually is error-prone and cumbersome.This paper presents the tool DaStGen translating classes declared in C++ syntax and augmented by new keywords into plain C++ code. The tool automates the record optimisation, as it analyses the potential range of each attribute, and as the user can restrict this range further. Herefrom, the generated code stores multiple attributes within one single primitive type. Furthermore, the tool derives user-defined MPI data types for each class. Using the tool reduces any algorithm's memory footprint, it speeds up memory-bounded applications such as CFD codes, and it hides technical details of MPI applications from the programmer.\",\n \"Title: Computational Steering Of Complex Flow Simulations\\nYear: 2018\\n\\nAbstract: Computational Steering, the combination of a simulation back-end with a visualisation front-end, offers great possibilities to exploit and optimise scenarios in engineering applications. Due to its interactivity, it requires fast grid generation, simulation, and visualisation and, therefore, mostly has to rely on coarse and inaccurate simulations typically performed on rather small interactive computing facilities and not on much more powerful high-performance computing architectures operated in batch-mode. This paper presents a steering environment that intends to bring these two worlds the interactive and the classical HPC world together in an integrated way. The environment consists of efficient fluid dynamics simulation codes and a steering and visualisation framework providing a user interface, communication methods for distributed steering, and parallel visualisation tools. The gap between steering and HPC is bridged by a hierarchical approach that performs fast interactive simulations for many scenario variants increasing the accuracy via hierarchical refinements in dependence of the time the user wants to wait. Finally, the user can trigger large simulations for selected setups on an HPC architecture exploiting the pre-computations already done on the interactive system.\",\n \"Title: Computational science and engineering: a new master's program at the Technische Universität München\\nYear: 2003\\n\\nAbstract: In the winter semester of 2001/2002, the Technische Universität München (TUM) started its new master's program computational science and engineering (CSE) as a joint initiative of seven faculties. It is the objective of this contribution to outline the underlying ideas and concepts and their curricular implementation as well as to emphasize some features that are probably non-standard in other comparable programs.\",\n \"Title: A sparse-grid-based out-of-sample extension for dimensionality reduction and clustering with laplacian eigenmaps\\nYear: 2011\\n\\nAbstract: Spectral graph theoretic methods such as Laplacian Eigenmaps are among the most popular algorithms for manifold learning and clustering. One drawback of these methods is, however, that they do not provide a natural out-of-sample extension. They only provide an embedding for the given training data. We propose to use sparse grid functions to approximate the eigenfunctions of the Laplace-Beltrami operator. We then have an explicit mapping between ambient and latent space. Thus, out-of-sample points can be mapped as well. We present results for synthetic and real-world examples to support the effectiveness of the sparse-grid-based explicit mapping.\",\n \"Title: 591 TFLOPS Multi-trillion Particles Simulation on SuperMUC.\\nYear: 2013\\n\\nAbstract: Anticipating large-scale molecular dynamics simulations (MD) in nano-fluidics, we conduct performance and scalability studies of an optimized version of the code ls1 mardyn. We present our implementation requiring only 32 Bytes per molecule, which allows us to run the, to our knowledge, largest MD simulation to date. Our optimizations tailored to the Intel Sandy Bridge processor are explained, including vectorization as well as shared-memory parallelization to make use of Hyperthreading. Finally we present results for weak and strong scaling experiments on up to 146016 Cores of SuperMUC at the Leibniz Supercomputing Centre, achieving a speed-up of 133k times which corresponds to an absolute performance of 591.2 TFLOPS.\",\n \"Title: Teaching high-performance computing on a high-performance cluster\\nYear: 2005\\n\\nAbstract: The university education in parallel and high-performance computing often suffers from a significant gap between the effects and potential performance taught in the lectures on the one hand and those practically experienced in exercises or lab courses on the other hand. With a small number of processors, the results obtained are often hardly convincing; however, supercomputers are rarely accessible to students doing their first steps in parallel programming. In this contribution, we present our experiences of how a state-of-the-art mid-size Linux cluster, bought and operated on a department level primarily for education and algorithm development purposes, can be used for teaching a large variety of HPC aspects. Special focus is put on the effects of such an approach on the intensity and sustainability of learning.\",\n \"Title: Learning by doing: software projects in CSE education\\nYear: 2006\\n\\nAbstract: Software development is one of the main routine activities in Computational Science and Engineering (CSE). Nevertheless, there is a huge gap between software engineering techniques available and established today in most fields where mainstream software is developed on the one hand and the typical extent of their application in a CSE context on the other hand. CSE curricula often reflect this tendency by not including software engineering topics adequately. This contribution reports experiences with a new course format called “student project” in the CSE master's program at TU München. There, for about half a year, a group of 4-8 students cooperate on a software development project – this time dealing with molecular dynamics. Although it is one objective to get a well performing code, the project's focus is on the consequent application of software engineering and project management practices.\",\n \"Title: Spatially adaptive sparse grids for high-dimensional data-driven problems\\nYear: 2010\\n\\nAbstract: Sparse grids allow one to employ grid-based discretization methods in data-driven problems. We present an extension of the classical sparse grid approach that allows us to tackle high-dimensional problems by spatially adaptive refinement, modified ansatz functions, and efficient regularization techniques. The competitiveness of this method is shown for typical benchmark problems with up to 166 dimensions for classification in data mining, pointing out properties of sparse grids in this context. To gain insight into the adaptive refinement and to examine the scope for further improvements, the approximation of non-smooth indicator functions with adaptive sparse grids has been studied as a model problem. As an example for an improved adaptive grid refinement, we present results for an edge-detection strategy.\",\n \"Title: Adaptive Sparse Grid Classification Using Grid Environments\\nYear: 2007\\n\\nAbstract: Common techniques tackling the task of classification in data mining employ ansatz functions associated to training data points to fit the data as well as possible. Instead, the feature space can be discretized and ansatz functions centered on grid points can be used. This allows for classification algorithms scaling only linearly in the number of training data points, enabling to learn from data sets with millions of data points. As the curse of dimensionality prohibits the use of standard grids, sparse grids have to be used.Adaptive sparse grids allow to get a trade-off between both worlds by refining in rough regions of the target function rather than in smooth ones. We present new results for some typical classification tasks and show first observations of dimension adaptivity. As the study of the critical parameters during development involves many computations for different parameter values, we used a grid environment which we present.\",\n \"Title: Towards High-Dimensional Computational Steering of Precomputed Simulation Data using Sparse Grids\\nYear: 2011\\n\\nAbstract: With the ever-increasing complexity, accuracy, dimensionality, and size of simulations, a step in the direction of data-intensive scientific discovery becomes necessary. Parameter-dependent simulations are an example of such a data-intensive tasks: The researcher, who is interested in the dependency of the simulation's result on a set of input parameters, changes essential parameters and wants to immediately see the effect of the changes in a visual environment. In this scenario, an interactive exploration is not possible due to the long execution time needed by even a single simulation corresponding to one parameter combination and the overall large number of parameter combinations which could be of interest.\",\n \"Title: Multivariate quadrature on adaptive sparse grids\\nYear: 2003\\n\\nAbstract: In this paper, we study the potential of adaptive sparse grids for multivariate numerical quadrature in the moderate or high dimensional case, i.e. for a number of dimensions beyond three and up to several hundreds. There, conventional methods typically suffer from the curse of dimension or are unsatisfactory with respect to accuracy. Our sparse grid approach, based upon a direct higher order discretization on the sparse grid, overcomes this dilemma to some extent, and introduces additional flexibility with respect to both the order of the 1 D quadrature rule applied (in the sense of Smolyak's tensor product decomposition) and the placement of grid points. The presented algorithm is applied to some test problems and compared with other existing methods.\",\n \"Title: Clustering based on density estimation with sparse grids\\nYear: 2012\\n\\nAbstract: We present a density-based clustering method. The clusters are determined by splitting a similarity graph of the data into connected components. The splitting is accomplished by removing vertices of the graph at which an estimated density function of the data evaluates to values below a threshold. The density function is approximated on a sparse grid in order to make the method feasible in higher-dimensional settings and scalable in the number of data points. With benchmark examples we show that our method is competitive with other modern clustering methods. Furthermore, we consider a real-world example where we cluster nodes of a finite element model of a Chevrolet pick-up truck with respect to the displacements of the nodes during a frontal crash.\",\n \"Title: Localized Discrete Empirical Interpolation Method.\\nYear: 2014\\n\\nAbstract: This paper presents a new approach to construct more efficient reduced-order models for nonlinear partial differential equations with proper orthogonal decomposition and the discrete empirical interpolation method (DEIM). Whereas DEIM projects the nonlinear term onto one global subspace, our localized discrete empirical interpolation method (LDEIM) computes several local subspaces, each tailored to a particular region of characteristic system behavior. Then, depending on the current state of the system, LDEIM selects an appropriate local subspace for the approximation of the nonlinear term. In this way, the dimensions of the local DEIM subspaces, and thus the computational costs, remain low even though the system might exhibit a wide range of behaviors as it passes through different regimes. LDEIM uses machine learning methods in the offline computational phase to discover these regions via clustering. Local DEIM approximations are then computed for each cluster. In the online computational phase, machine-learning-based classification procedures select one of these local subspaces adaptively as the computation proceeds. The classification can be achieved using either the system parameters or a low-dimensional representation of the current state of the system obtained via feature extraction. The LDEIM approach is demonstrated for a reacting flow example of an H-2-Air flame. In this example, where the system state has a strong nonlinear dependence on the parameters, the LDEIM provides speedups of two orders of magnitude over standard DEIM.\",\n \"Title: Towards a Navier Stokes-Darcy Upscaling Based on Permeability Tensor Computation.\\nYear: 2012\\n\\nAbstract: The micro scale simulation of CO2 sequestration involves complex, porous-like geometries. For the generation of such geometries, we present two approaches: In 2D, we construct a fractured domain by channel networks. In 3D, we approximate sand grain-like scenarios by dense sphere packings. The flow through these structures is simulated with the incompressible Navier-Stokes solver of the PDE framework Peano. Using an upscaling scheme, the results of the micro scale are used as input data for a Darcy solver on the coarse scales. The coupling concept and the scenario generators are presented together with first simulation results showing the validity of the approach.\",\n \"Title: Free-Surface Lattice-Boltzmann Simulation On Many-Core Architectures\\nYear: 2011\\n\\nAbstract: Current advances in many-core technologies demand simulation algorithms suited for the corresponding architectures while with regard to the respective increase of computational power, real-time and interactive simulations become possible and desirable. We present an OpenCL implementation of a Lattice-Boltzmann-based free-surface solver for GPU architectures. The massively parallel execution especially requires special techniques to keep the interface region consistent, which is here addressed by a novel multipass method. We further compare different memory layouts according to their performance for both a basic driven cavity implementation and the free-surface method, pointing out the capabilities of our implementation in real-time and interactive scenarios, and shortly present visualizations of the flow, obtained in real-time.\",\n \"Title: A note on the complexity of solving Poisson's equation for spaces of bounded mixed derivatives\\nYear: 1999\\n\\nAbstract: We study the complexity of solving the d -dimensional Poisson equation on ]0, 1[ d . We restrict ourselves to cases where the solution u lies in some space of functions of bounded mixed derivatives (with respect to the L ∞ - or the L 2 -norm) up to ∂ 2 d /∏ d j =1 ∂ x 2 j . An upper bound for the complexity of computing a solution of some prescribed accuracy ε with respect to the energy norm is given, which is proportional to ε −1 . We show this result in a constructive manner by proposing a finite element method in a special sparse grid space, which is obtained by an a priori grid optimization process based on the energy norm. Thus, the result of this paper is twofold: First, from a theoretical point of view concerning the complexity of solving elliptic PDEs, a strong tractability result of the order O ( ε −1 ) is given, and, second, we provide a practically usable hierarchical basis finite element method of this complexity O ( ε −1 ), i.e., without logarithmic terms growing exponentially in d , at least for our sparse grid setting with its underlying smoothness requirements.\",\n \"Title: Many-core architectures boost the pricing of basket options on adaptive sparse grids.\\nYear: 2013\\n\\nAbstract: In this work, we present a highly scalable approach for numerically solving the Black-Scholes PDE in order to price basket options. Our method is based on a spatially adaptive sparse-grid discretization with finite elements. Since we cannot unleash the compute capabilities of modern many-core chips such as GPUs using the complexity-optimal Up-Down method, we implemented an embarrassingly parallel direct method. This operator is paired with a distributed memory parallelization using MPI and we achieved very good scalability results compared to the standard Up-Down approach. Since we exploit all levels of the operator's parallelism, we are able to achieve nearly perfect strong scaling for the Black-Scholes solver. Our results show that typical problem sizes (5 dimensional basket options), require at least 4 NVIDIA K20X Kepler GPUs (inside a Cray XK7) in order to be faster than the Up-Down scheme running on 16 Intel Sandy Bridge cores (one box). On a Cray XK7 machine we outperform our highly parallel Up-Down implementation by 55X with respect to time to solution. Both results emphasize the competitiveness of our proposed operator.\",\n \"Title: Semi-Coarsening In Space And Time For The Hierarchical Transformation Multigrid Method\\nYear: 2012\\n\\nAbstract: We extend the hierarchical transformation multigrid (HT-MG) method with semi-coarsening in space and time in order to tackle anisotropic problems. Semi-coarsening allows us to smooth on anisotropic grids and, thus to employ more advanced multigrid cycles, e. g. the so-called Q-cycle. Our numerical examples show that we can tackle anisotropic problems with the HT-MG method and the Q-cycle. We do not only apply the multigrid method in space directions but also in time directions where the Q-cycle is beneficial again.\",\n \"Title: A highly parallel Black–Scholes solver based on adaptive sparse grids\\nYear: 2012\\n\\nAbstract: In this paper, we present a highly efficient approach for numerically solving the Black–Scholes equation in order to price European and American basket options. Therefore, hardware features of contemporary high performance computer architectures such as non-uniform memory access and hardware-threading are exploited by a hybrid parallelization using MPI and OpenMP which is able to drastically reduce the computing time. In this way, we achieve very good speed-ups and are able to price baskets with up to six underlyings. Our approach is based on a sparse grid discretization with finite elements and makes use of a sophisticated adaption. The resulting linear system is solved by a conjugate gradient method that uses a parallel operator for applying the system matrix implicitly. Since we exploit all levels of the operator's parallelism, we are able to benefit from the compute power of more than 100 cores. Several numerical examples as well as an analysis of the performance for different computer architectures are provided.\",\n \"Title: Image Segmentation with Adaptive Sparse Grids\\nYear: 2013\\n\\nAbstract: We present a novel adaptive sparse grid method for unsupervised image segmentation. The method is based on spectral clustering. The use of adaptive sparse grids achieves that the dimensions of the involved eigensystem do not depend on the number of pixels. In contrast to classical spectral clustering, our sparse-grid variant is therefore able to segment larger images. We evaluate the method on real-world images from the Berkeley Segmentation Dataset. The results indicate that images with 150,000 pixels can be segmented by solving an eigenvalue system of dimensions 500 × 500 instead of 150, 000 × 150, 000.\",\n \"Title: Cluster optimization and parallelization of simulations with dynamically adaptive grids\\nYear: 2013\\n\\nAbstract: The present paper studies solvers for partial differential equations that work on dynamically adaptive grids stemming from spacetrees. Due to the underlying tree formalism, such grids efficiently can be decomposed into connected grid regions (clusters) on-the-fly. A graph on those clusters classified according to their grid invariancy, workload, multi-core affinity, and further meta data represents the inter-cluster communication. While stationary clusters already can be handled more efficiently than their dynamic counterparts, we propose to treat them as atomic grid entities and introduce a skip mechanism that allows the grid traversal to omit those regions completely. The communication graph ensures that the cluster data nevertheless are kept consistent, and several shared memory parallelization strategies are feasible. A hyperbolic benchmark that has to remesh selected mesh regions iteratively to preserve conforming tessellations acts as benchmark for the present work. We discuss runtime improvements resulting from the skip mechanism and the implications on shared memory performance and load balancing.\",\n \"Title: Extending a highly parallel data mining algorithm to the intel ® many integrated core architecture\\nYear: 2011\\n\\nAbstract: Extracting knowledge from vast datasets is a major challenge in data-driven applications, such as classification and regression, which are mostly compute bound. In this paper, we extend our SG++ algorithm to the Intel® Many Integrated Core Architecture (Intel® MIC Architecture). The ease of porting an application to Intel MIC Architecture is shown: porting existing SSE code is very easy and straightforward. We evaluate the current prototype pre-release coprocessor board codenamed Intel® \\\"Knights Ferry\\\". We utilize the pragma-based offloading programming model offered by the Intel® Composer XE for Intel MIC Architecture, generating both the host and the coprocessor code. We compare the achieved performance with an NVIDIA C2050 accelerator and show that the pre-release Knights Ferry coprocessor delivers better performance than the C2050 and exceeds the C2050 when comparing the productivity aspect of implementing algorithms for the coprocessors.\",\n \"Title: Shared Memory Parallelization Of Fully-Adaptive Simulations Using A Dynamic Tree-Split And -Join Approach\\nYear: 2012\\n\\nAbstract: In this work we present an approach for the parallelization of hyperbolic simulations on shared-memory architectures running on fully-adaptive grids.We tackle the parallelization problem with a dynamic sub-tree split-and join-approach by running computations on those split sub-trees in parallel using lightweight tasks. The traversal of sub-trees created by tree-splittings is built upon an inherently cache efficient approach for solving hyperbolic PDEs on dynamically adaptive triangular grids using a Sierpinski space filling curve. Our communication scheme among sub-trees stores the exchange-data to/from adjacent sub-trees in a consecutive memory area which is further utilized for an improved run-length-encoded data exchange.To give results for a concrete scenario, we implemented a solver for the shallow water equations which demands for fully-adaptive grid refinement and coarsening after each time-step. Our results give detailed statistics about optimization of the split size, parallelization overhead and also strong scalability results for a simulation running on multi-socket Intel and AMD architectures.\",\n \"Title: Invasive Compute Balancing for Applications with Hybrid Parallelization\\nYear: 2013\\n\\nAbstract: Achieving high scalability for dynamical adaptive algorithms in HPC is a non-trivial task. In this contex, explicit data migration is typically applied which comes with several challenges such as criteria when and where to migrate data or additional bandwidth requirements. In our work, we use an alternative to data-migration, namely compute-migration on cache-coherent memory systems. We focus on migrating computational power for simulations with dynamical adaptive grids with invasive paradigm. An extension to our existing core-distribution scheduler is presented, distributing the cores depending on the requirements specified by each parallel program instance. In addition, a programming pattern for the hybrid parallel application code is presented to overcome deadlock issues created by applying invasive computing. We validate our approach with benchmark computations for a simulation with artificial workload and furthermore for a realistic scenario based on dynamical adaptive shallow water simulations. Those benchmarks are conducted within a hybrid parallelization environment on a 40-core HPC node. On this shared-memory system, our invasive approach results in faster execution times and higher hardware utilization than the non-invasive approach. Our approach can be applied in general to similar classes of simulations on dynamical adaptive grids.\",\n \"Title: Space Tree Structures for PDE Software\\nYear: 2002\\n\\nAbstract: In this paper, we study the potential of space trees (boundary extended octrees for an arbitrary number of dimensions) in the context of software for the numerical solution of PDEs. The main advantage of the approach presented is the fact that the underlying geometry's resolution can be decoupled from the computational grid's resolution, although both are organized within the same data structure. This allows us to solve the PDE on a quite coarse orthogonal grid at an accuracy corresponding to a much finer resolution. We show how fast (multigrid) solvers based on the nested dissection principle can be directly implemented on a space tree. Furthermore, we discuss the use of this hierarchical concept as the common data basis for the partitioned solution of coupled problems like fluid-structure interactions, e.g., and we address its suitability for an integration of simulation software.\",\n \"Title: Fast Insight into High-Dimensional Parametrized Simulation Data\\nYear: 2012\\n\\nAbstract: Numerical simulation has become an inevitable tool in most industrial product development processes with simulations being used to understand the influence of design decisions (parameter configurations) on the structure and properties of the product. However, in order to allow the engineer to thoroughly explore the design space and fine-tune parameters, many -- usually very time-consuming -- simulation runs are necessary. Additionally, this results in a huge amount of data that cannot be analyzed in an efficient way without the support of appropriate tools. In this paper, we address the two-fold problem: First, instantly provide simulation results if the parameter configuration is changed, and, second, identify specific areas of the design space with concentrated change and thus importance. We propose the use of a hierarchical approach based on sparse grid interpolation or regression which acts as an efficient and cheap substitute for the simulation. Furthermore, we develop new visual representations based on the derivative information contained inherently in the hierarchical basis. They intuitively let a user identify interesting parameter regions even in higher-dimensional settings. This workflow is combined in an interactive visualization and exploration framework. We discuss examples from different fields of computational science and engineering and show how our sparse-grid-based techniques make parameter dependencies apparent and how they can be used to fine-tune parameter configurations.\",\n \"Title: ls1 mardyn: The massively parallel molecular dynamics code for large systems.\\nYear: 2014\\n\\nAbstract: The molecular dynamics simulation code ls1 mardyn is presented. It is a highly scalable code, optimized for massively parallel execution on supercomputing architectures and currently holds the world record for the largest molecular simulation with over four trillion particles. It enables the application of pair potentials to length and time scales that were previously out of scope for molecular dynamics simulation. With an efficient dynamic load balancing scheme, it delivers high scalability even for challenging heterogeneous configurations. Presently, multicenter rigid potential models based on Lennard-Jones sites, point charges, and higher-order polarities are supported. Due to its modular design, ls1 mardyn can be extended to new physical models, methods, and algorithms, allowing future users to tailor it to suit their respective needs. Possible applications include scenarios with complex geometries, such as fluids at interfaces, as well as nonequilibrium molecular dynamics simulation of heat and mass transfer.\",\n \"Title: GPU Optimization of Pseudo Random Number Generators for Random Ordinary Differential Equations.\\nYear: 2014\\n\\nAbstract: Solving differential equations with stochastic terms involves a massive use of pseudo random numbers. We present an application for the simulation of wireframe buildings under stochastic earthquake excitation. The inherent potential for vectorization of the application is used to its full extent on GPU accelerator hardware. A representative set of pseudo random number gen- erators for uniformly and normally distributed pseudo random numbers has been implemented, optimized, and benchmarked. The resulting optimized variants outperform standard library implementations on GPUs. The techniques and improvements shown in this contribution using the Kanai-Tajimi model can be generalized to other random differential equations or stochastic models as well as other accelerators.\",\n \"Title: Rendering of Feature-rich Dynamically Changing Volumetric Datasets on GPU.\\nYear: 2014\\n\\nAbstract: Interactive photo-realistic representation of dynamic liquid volumes is a challenging task for today's GPUs and state-of-the-art visualization algorithms. Methods of the last two decades consider either static volumetric datasets applying several optimizations for volume casting, or dynamic volumetric datasets with rough approximations to realistic rendering. Nevertheless, accurate real-time visualization of dynamic datasets is crucial in areas of scientific visualization as well as areas demanding for accurate rendering of feature-rich datasets.\",\n \"Title: EXAHD: An Exa-scalable Two-Level Sparse Grid Approach for Higher-Dimensional Problems in Plasma Physics and Beyond.\\nYear: 2014\\n\\nAbstract: High-dimensional problems pose a challenge for tomorrow's supercomputing. Problems that require the joint discretization of more dimensions than space and time are among the most compute-hungry ones and thus standard candidates for exascale computing and even beyond. This project tackles such problems by a hierarchical extrapolation approach, the sparse grid combination technique. The method not only enables their treatment in the first place. The hierarchical approach also provides novel ways to deal with central problems in high-performance computing such as scalability and resilience: Global communication can be avoided and reduced to a small subset, and faults can be compensated for without the need for recomputations or checkpoint-restart. As an exemplary prototype for high-dimensional problems, turbulence simulations in plasma physics are studied.\",\n \"Title: First experiences with group projects in CSE education\\nYear: 2006\\n\\nAbstract: The various educational workshops at CSE-related conferences reveal that there is an ongoing open discussion on how to define the relevant aspects of this discipline, on how to integrate the identified topics into the curricula, and, of course, on how to design suitable course formats -- or briefly, how to teach CSE in an appropriate way. One question more and more addressed in such discussions is to what extent and how methods established in software engineering can or even must be adopted. A second evergreen issue (not restricted to CSE programs, of course) is whether soft skills such as teamwork, project management, or leadership should be taught in special courses or better imparted within suitable CSE-related modules. In this contribution, we report our experiences with project-based and software-focused CSE education. The discussion of expectations and outcomes covers both the software-related issues and the main project-related aspects.\",\n \"Title: From GPGPU to Many-Core: Nvidia Fermi and Intel Many Integrated Core Architecture\\nYear: 2012\\n\\nAbstract: Comparing the architectures and performance levels of an Nvidia Fermi accelerator with an Intel MIC Architecture coprocessor demonstrates the benefit of the coprocessor for bringing highly parallel applications into, or even beyond, GPGPU performance regions.\",\n \"Title: Computational Science and Engineering Education: SIAM's Perspective\\nYear: 2009\\n\\nAbstract: This year's SIAM CSE conference featured a structured program on education that sought to attract attention to and sustained interest in this essential topic.\",\n \"Title: A discriminative approach to polyphonic piano note transcription using supervised non-negative matrix factorization\\nYear: 2013\\n\\nAbstract: We introduce a novel method for the transcription of polyphonic piano music by discriminative training of support vector machines (SVMs). As features, we use pitch activations computed by supervised non-negative matrix factorization from low-level spectral features. Different approaches to low-level feature extraction, NMF dictionary learning and activation feature extraction are analyzed in a large-scale evaluation on eight hours of piano music including synthesized and real recordings. We conclude that the proposed method delivers state-of-the-art results and clearly outperforms SVMs using simple spectral features.\",\n \"Title: Data mining on vast data sets as a cluster system benchmark\\nYear: 2016\\n\\nAbstract: Comparing different accelerated cluster architectures by a single application is a tough piece of work because this application has to be optimized with respect to platform-dependent features. In this work, we demonstrate such an optimization for a data mining algorithm which solves regression and classification problems on vast data sets. Our technique is based on least squares regression, and its major component is the iterative matrix-free solution of a linear system of equations. By processing data sets ranging from several hundreds of thousands instances to multi-million data points in strong-scaling and weak-scaling settings, we are able to estimate the amount of parallelism needed to unleash the performance of classic CPU-based machines and clusters employing Intel Xeon Phi coprocessors and NVIDIA Kepler GPUs. Only in strong-scaling experiments, GPUs and coprocessors suffer from their tremendous amount of needed parallelism and get outperformed by dual socket Intel Sandy Bridge nodes at large scale more than 64 nodes/accelerators. However, in weak-scaling scenarios, a speed-up larger than 2X over an entire CPU node can be achieved by a single accelerator. Copyright © 2015 John Wiley & Sons, Ltd.\",\n \"Title: MaMiCo: Software design for parallel molecular-continuum flow simulations.\\nYear: 2016\\n\\nAbstract: The macro–micro-coupling tool (MaMiCo) was developed to ease the development of and modularize molecular-continuum simulations, retaining sequential and parallel performance. We demonstrate the functionality and performance of MaMiCo by coupling the spatially adaptive Lattice Boltzmann framework waLBerla with four molecular dynamics (MD) codes: the light-weight Lennard-Jones-based implementation SimpleMD, the node-level optimized software ls1 mardyn, and the community codes ESPResSo and LAMMPS. We detail interface implementations to connect each solver with MaMiCo. The coupling for each waLBerla-MD setup is validated in three-dimensional channel flow simulations which are solved by means of a state-based coupling method. We provide sequential and strong scaling measurements for the four molecular-continuum simulations. The overhead of MaMiCo is found to come at 10%–20% of the total (MD) runtime. The measurements further show that scalability of the hybrid simulations is reached on up to 500 Intel SandyBridge, and more than 1000 AMD Bulldozer compute cores.\",\n \"Title: Block-structured grids for Eulerian gyrokinetic simulations.\\nYear: 2016\\n\\nAbstract: In order to predict the turbulent transport in magnetic fusion experiments, global (i.e., full-torus) gyrokinetic simulations are often carried out. In this context, one frequently encounters situations in which the plasma temperature varies by large factors across the radial simulation domain. In grid-based Eulerian codes, this enforces the use of a very large number of grid points in the two-dimensional velocity space, and, thus, an enormous computational effort. To minimize the computational requirements, one may employ block-structured grids, adapted to the radial changes of the temperature. As the block-structured grids rely on a general approach, they can be applied to different Eulerian gyrokinetic implementations. In this paper, we explain the construction and implementation of such grids in the gyrokinetic code GENE, F. Jenko et al. (2000), and present corresponding simulation results.\",\n \"Title: Mass-spring systems for simulating mitral valve repair using 3D ultrasound images.\\nYear: 2015\\n\\nAbstract: Mitral valve (MV) diseases are among the most common types of heart diseases, while heart diseases are the most common cause of death worldwide. MV repair surgery is connected to higher survival rates and fewer complications than the total replacement of the MV, but MV repair requires extensive patient-specific therapy planning. The simulation of MV repair with a patient-specific model could help to optimize surgery results and make MV repair available to more patients. However, current patient-specific simulations are difficult to transfer to clinical application because of time-constraints or prohibitive requirements on the resolution of the image data. As one possible solution to the problem of patient-specific MV modeling, we present a mass-spring MV model based on 3D transesophageal echocardiographic (TEE) images already routinely acquired for MV repair therapy planning. Our novel approach to the rest-length estimation of springs allows us to model the global support of the MV leaflets through the chordae tendinae without the need for high-resolution image data. The model is used to simulate MV annuloplasty for five patients undergoing MV repair, and the simulated results are compared to post-surgical TEE images. The comparison shows that our model is able to provide a qualitative estimate of annuloplasty surgery. In addition, the data suggests that the model might also be applied to simulating the implantation of artificial chordae. (C) 2015 Elsevier Ltd. All rights reserved.\",\n \"Title: Reduced models for sparse grid discretizations of the multi-asset Black-Scholes equation\\nYear: 2015\\n\\nAbstract: This work presents reduced models for pricing basket options with the Black-Scholes and the Heston model. Basket options lead to multi-dimensional partial differential equations (PDEs) that quickly become computationally infeasible to discretize on full tensor grids. We therefore rely on sparse grid discretizations of the PDEs, which allow us to cope with the curse of dimensionality to some extent. We then derive reduced models with proper orthogonal decomposition. Our numerical results with the Black-Scholes model show that sufficiently accurate results are achieved while gaining speedups between 80 and 160 compared to the high-fidelity sparse grid model for 2-, 3-, and 4-asset options. For the Heston model, results are presented for a single-asset option that leads to a two-dimensional pricing problem, where we achieve significant speedups with our model reduction approach based on high-fidelity sparse grid models.\",\n \"Title: A Multigrid Method for Adaptive Sparse Grids\\nYear: 2015\\n\\nAbstract: Sparse grids have become an important tool to reduce the number of degrees of freedom of discretizations of moderately high-dimensional partial differential equations; however, the reduction in degrees of freedom comes at the cost of an almost dense and unconventionally structured system of linear equations. To guarantee overall efficiency of the sparse grid approach, special linear solvers are required. We present a multigrid method that exploits the sparse grid structure to achieve an optimal runtime that scales linearly with the number of sparse grid points. Our approach is based on a novel decomposition of the right-hand sides of the coarse grid equations that leads to a reformulation in so-called auxiliary coefficients. With these auxiliary coefficients, the right-hand sides can be represented in a nodal point basis on low-dimensional full grids. Our proposed multigrid method directly operates in this auxiliary coefficient representation, circumventing most of the computationally cumbersome sparse grid structure. Numerical results on nonadaptive and spatially adaptive sparse grids confirm that the runtime of our method scales linearly with the number of sparse grid points and they indicate that the obtained convergence factors are bounded independently of the mesh width.\",\n \"Title: Dynamically adaptive Lattice Boltzmann simulation of shallow water flows with the Peano framework\\nYear: 2015\\n\\nAbstract: We present a dynamically adaptive Lattice Boltzmann (LB) implementation for solving the shallow water equations (SWEs). Our implementation extends an existing LB component of the Peano framework. We revise the modular design with respect to the incorporation of new simulation aspects and LB models. The basic SWE-LB implementation is validated in different breaking dam scenarios. We further provide a numerical study on stability of the MRT collision operator used in our simulations.\",\n \"Title: Modelling social identification and helping in evacuation simulation.\\nYear: 2016\\n\\nAbstract: •An algorithmic formulation of empirical findings from social psychology is presented.•The model is qualitatively validated against data from London bombings.•With uncertainty quantification the impact of the model parameters is analysed.•Simulations with the model can lead to a better estimation of evacuation times.\",\n \"Title: Invasive Compute Balancing for Applications with Shared and Hybrid Parallelization\\nYear: 2015\\n\\nAbstract: Abstract Achieving high scalability with dynamically adaptive algorithms in high-performance computing (HPC) is a non-trivial task. The invasive paradigm using compute migration represents an efficient alternative to classical data migration approaches for such algorithms in HPC. We present a core-distribution scheduler which realizes the migration of computational power by distributing the cores depending on the requirements specified by one or more parallel program instances. We validate our approach with different benchmark suites for simulations with artificial workload as well as applications based on dynamically adaptive shallow water simulations, and investigate concurrently executed adaptivity parameter studies on realistic Tsunami simulations. The invasive approach results in significantly faster overall execution times and higher hardware utilization than alternative approaches. A dynamic resource management is therefore mandatory for a more efficient execution of scenarios similar to our simulations, e.g. several Tsunami simulations in urgent computing, to overcome strong scalability challenges in the area of HPC. The optimizations obtained by invasive migration of cores can be generalized to similar classes of algorithms with dynamic resource requirements.\",\n \"Title: Invasive computing in HPC with X10\\nYear: 2013\\n\\nAbstract: High performance computing with thousands of cores relies on distributed memory due to memory consistency reasons. The resource management on such systems usually relies on static assignment of resources at the start of each application. Such a static scheduling is incapable of starting applications with required resources being used by others since a reduction of resources assigned to applications without stopping them is not possible. This lack of dynamic adaptive scheduling leads to idling resources until the remaining amount of requested resources gets available. Additionally, applications with changing resource requirements lead to idling or less efficiently used resources. The invasive computing paradigm suggests dynamic resource scheduling and applications able to dynamically adapt to changing resource requirements. As a case study, we developed an invasive resource manager as well as a multigrid with dynamically changing resource demands. Such a multigrid has changing scalability behavior during its execution and requires data migration upon reallocation due to distributed memory systems. To counteract the additional complexity introduced by the additional interfaces, e. g. for data migration, we use the X10 programming language for improved programmability. Our results show improved application throughput and the dynamic adaptivity. In addition, we show our extension for the distributed arrays of X10 to support data migration.\",\n \"Title: Optimized Force Calculation in Molecular Dynamics Simulations for the Intel Xeon Phi.\\nYear: 2015\\n\\nAbstract: We provide details on the shared-memory parallelization for manycore architectures of the molecular dynamics framework ls1-mardyn, including an optimization of the SIMD vectorization for multi-centered molecules. The novel shared-memory parallelization scheme allows to retain Newton's third law optimization and exhibits very good scaling on many-core devices such as a full Xeon Phi card running 240 threads. The Xeon Phi can thus be exploited and delivers comparable performance as IvyBridge nodes in our experiments.\",\n \"Title: Research and Education in Computational Science and Engineering.\\nYear: 2018\\n\\nAbstract: This report presents challenges, opportunities, and directions for computational science and engineering (CSE) research and education for the next decade. Over the past two decades the field of CSE has penetrated both basic and applied research in academia, industry, and laboratories to advance discovery, optimize systems, support decision-makers, and educate the scientific and engineering workforce. Informed by centuries of theory and experiment, CSE performs computational experiments to answer questions that neither theory nor experiment, alone is equipped to answer. CSE provides scientists and engineers with algorithmic inventions and software systems that, transcend disciplines and scales. CSE brings the power of parallelism to bear on troves of data. Mathematics-based advanced computing has become a prevalent means of discovery and innovation in essentially all areas of science, engineering, technology, and society, and the CSE community is at the core of this transformation. However, a combination of disruptive developments-including the architectural complexity of extreme-scale computing, the data revolution and increased attention to data-driven discovery, and the specialization required to follow the applications to new frontiers-is redefining the scope and reach of the CSE endeavor. With these many current and expanding opportunities for the CSE field, there is a growing demand for CSE graduates and a need to expand CSE educational offerings. This need includes CSE programs at both the undergraduate and graduate levels, as well as continuing education and professional development programs, exploiting the synergy between computational science and data science. Yet, as institutions consider new and evolving educational programs, it is essential to consider the broader research challenges and opportunities that provide the context for CSE education and workforce development.\",\n \"Title: A parallel arbitrary-order accurate AMR algorithm for the scalar advection-diffusion equation.\\nYear: 2016\\n\\nAbstract: We present a numerical method for solving the scalar advection-diffusion equation using adaptive mesh refinement. Our solver has three unique characteristics: (1) it supports arbitrary-order accuracy in space; (2) it allows different discretizations for the velocity and scalar advected quantity; (3) it combines the method of characteristics with an integral equation formulation; and (4) it supports shared and distributed memory architectures. In particular, our solver is based on a second-order accurate, unconditionally stable, semi-Lagrangian scheme combined with a spatially-adaptive Chebyshev octree for discretization. We study the convergence, single-node performance, strong scaling, and weak scaling of our scheme for several challenging flows that cannot be resolved efficiently without using high-order accurate discretizations. For example, we consider problems for which switching from 4th order to 14th order approximation results in two orders of magnitude speedups for a computation in which we keep the target accuracy in the solution fixed. For our largest run, we solve a problem with one billion unknowns on a tree with maximum depth equal to 10 and using 14th-order elements on 16,384 ×86 cores on the \\\\\\\"STAMPEDE\\\\\\\" system at the Texas Advanced Computing Center.\",\n \"Title: Infrastructure and API Extensions for Elastic Execution of MPI Applications.\\nYear: 2016\\n\\nAbstract: Dynamic Processes support was added to MPI in version 2.0 of the standard. This feature of MPI has not been widely used by application developers in part due to the performance cost and limitations of the spawn operation. In this paper, we propose an extension to MPI that consists of four new operations. These operations allow an application to be initialized in an elastic mode of execution and enter an adaptation window when necessary, where resources are incorporated into or released from the application's world communicator. A prototype solution based on the MPICH library and the SLURM resource manager is presented and evaluated alongside an elastic scientific application that makes use of the new MPI extensions. The cost of these new operations is shown to be negligible due mainly to the latency hiding design, leaving the application's time for data redistribution as the only significant performance cost.\",\n \"Title: Evaluation of an Efficient Stack-RLE Clustering Concept for Dynamically Adaptive Grids.\\nYear: 2016\\n\\nAbstract: One approach for tackling the challenge of efficient implementations for parallel PDE simulations on dynamically changing grids is the usage of space-filling curves (SFCs). While SFC algorithms possess advantageous properties such as low memory requirements and close-to-optimal partitioning approaches with linear complexity, they require efficient communication strategies for keeping and utilizing the connectivity information, in particular for dynamically changing grids. Our approach is to use a sparse communication graph to store the connectivity information and to transfer data blockwise. This permits efficient generation of multiple partitions per memory context (denoted by cl us tering),which-in combination with a run-length encoding (RLE)-directly leads to elegant solutions for shared, distributed, and hybrid parallelization and allows cluster-based optimizations. While previous work focused on specific aspects, we present in this paper an overall compact summary of the stack-RLE clustering approach complete with aspects of the vertex-based communication that facilitate understanding the approach. The central contribution of this work is the proof of suitability of the stack-RLE clustering approach for an efficient realization of different, relevant building blocks of scientific computing methodology and real-life computer science and engineering (CSE) applications: We show 95% strong scalability for small-scale scalability benchmarks on 512 cores and weak scalability of over 90% on 8192 cores for finite-volume solvers and changing grid structure in every time step; optimizations of simulation data backends by writer tasks; comparisons of analytical benchmarks to analyze the adaptivity criteria; and a tsunami simulation as a representative real-world showcase of a wave propagation for our approach which reduces the overall workload by 95% for parallel fully adaptive mesh refinement and, based on a comparison with SFC-ordered regular grid cells, reduces the computation time by a factor of 7.6 with improved results and a factor of 62.2 with results of similar accuracy of buoy station data.\",\n \"Title: Steady-State Anderson Accelerated Coupling Of Lattice Boltzmann And Navier-Stokes Solvers\\nYear: 2016\\n\\nAbstract: We present an Anderson acceleration-based approach to spatially couple three-dimensional Lattice Boltzmann and Navier-Stokes (LBNS) flow simulations. This allows to locally exploit the computational features of both fluid flow solver approaches to the fullest extent and yields enhanced control to match the LB and NS degrees of freedom within the LBNS overlap layer. Designed for parallel Schwarz coupling, the Anderson acceleration allows for the simultaneous execution of both Lattice Boltzmann and Navier-Stokes solver. We detail our coupling methodology, validate it, and study convergence and accuracy of the Anderson accelerated coupling, considering three steady-state scenarios: plane channel flow, flow around a sphere and channel flow across a porous structure. We find that the Anderson accelerated coupling yields a speed-up (in terms of iteration steps) of up to 40% in the considered scenarios, compared to strictly sequential Schwarz coupling.\",\n \"Title: Load Balancing for Molecular Dynamics Simulations on Heterogeneous Architectures\\nYear: 2016\\n\\nAbstract: Upcoming exascale compute systems are expected to be built from heterogeneous hardware architectures. According to this trend, there exist various methods to handle clusters composed of CPUs, GPUs or other accelerators. Most of these assume that each node has the same structure-for example a dual socket system with an accelerator (GPU or Xeon Phi). The workload is then distributed homogeneously among the nodes. However, not all clusters fulfill this requirement. They might contain different partitions with and without accelerators. Furthermore, depending on the underlying problem to be solved, accelerator cards may perform better in native mode compared to offloading. Besides, various aspects such as cooling may influence the performance of individual nodes. It therefore cannot always be assumed, that the structure and performance of each node and hence the performance of every MPI rank is the same. In this contribution, we apply a k-d tree decomposition method to balance load on heterogeneous compute clusters. The algorithm is incorporated into the molecular dynamics simulation program ls1 mardyn. We present performance results for simulations executed on hybrid AMD Bulldozer-Intel Sandy Bridge, Intel Westmere-Intel Sandy Bridge and Intel Xeon-Intel Xeon Phi-architectures. The only prerequisite for the proposed algorithm is a cost estimation for different decompositions. It is hence expected to be applicable to a variety of n-body scenarios, for which a domain decomposition is possible.\",\n \"Title: Block-structured grids in full velocity space for Eulerian gyrokinetic simulations.\\nYear: 2017\\n\\nAbstract: Global, i.e., full-torus, gyrokinetic simulations play an important role in exploring plasma microturbulence in magnetic fusion devices with strong radial variations. In the presence of steep temperature profiles, grid-based Eulerian approaches can become quite challenging as the correspondingly varying velocity space structures need to be accommodated and sufficiently resolved. A rigid velocity space grid then requires a very high number of discretization nodes resulting in enormous computational costs. To tackle this issue and reduce the computational demands, we introduce block-structured grids in the all velocity space dimensions. The construction of these grids is based on a general approach, making them suitable for various Eulerian implementations. In the current study, we explain the rationale behind the presented approach, detail the implementation, and provide simulation results obtained with the block-structured grids. The achieved reduction in the number of computational nodes depends on the temperature profile and simulation scenario provided. In the test cases at hand, about ten times fewer grid points are required for nonlinear simulations performed with block-structured grids in the plasma turbulence code GENE (http://genecode.org). With the speed-up found to scale almost exactly reciprocal to the number of grid points, the new implementation greatly reduces the computational costs and therefore opens new possibilities for applications of this software package.\",\n \"Title: High performance shallow water kernels for parallel overland flow simulations based on FullSWOF2D.\\nYear: 2017\\n\\nAbstract: We describe code optimization and parallelization procedures applied to the sequential overland flow solver FullSWOF2D. Major difficulties when simulating overland flows comprise dealing with high resolution datasets of large scale areas which either cannot be computed on a single node either due to limited amount of memory or due to too many (time step) iterations resulting from the CFL condition. We address these issues in terms of two major contributions. First, we demonstrate a generic step-by-step transformation of the second order finite volume scheme in FullSWOF2D towards MPI parallelization. Second, the computational kernels are optimized by the use of templates and a portable vectorization approach. We discuss the load imbalance of the flux computation due to dry and wet cells and propose a solution using an efficient cell counting approach. Finally, scalability results are shown for different test scenarios along with a flood simulation benchmark using the Shaheen II supercomputer.\",\n \"Title: A Large-Scale Malleable Tsunami Simulation Realized on an Elastic MPI Infrastructure.\\nYear: 2017\\n\\nAbstract: Realization of resource awareness and elasticity in hardware and software is an answer to many problems and challenges we are facing in High Performance Computing (HPC) today. Resource utilization inefficiency is a real problem in current HPC systems due to the current static, inflexible resource assignment configuration. One way to resolve this problem is to change the static resource assignment setting---by introducing runtime resource elasticity, which requires both malleability in software implementation and support for runtime resource adaptation in the system infrastructure. In this paper, we show a successful implementation of a malleable tsunami simulation realized on an elastic MPI infrastructure we previously proposed. We also prove that introducing malleability to such a tightly coupled parallel application can be beneficial.\",\n \"Title: A Massively-Parallel, Fault-Tolerant Solver for High-Dimensional PDEs.\\nYear: 2016\\n\\nAbstract: We investigate the effect of hard faults on a massively-parallel implementation of the Sparse Grid Combination Technique (SGCT), an efficient numerical approach for the solution of high-dimensional time-dependent PDEs. The SGCT allows us to increase the spatial resolution of a solver to a level that is out of scope with classical discretization schemes due to the curse of dimensionality. We exploit the inherent data redundancy of this algorithm to obtain a scalable and fault-tolerant implementation without the need of checkpointing or process replication. It is a lossy approach that can guarantee convergence for a large number of faults and a wide range of applications. We present first results using our fault simulation framework - and the first convergence and scalability results with simulated faults and algorithm-based fault tolerance for PDEs in more than three dimensions.\",\n \"Title: A highly scalable, algorithm-based fault-tolerant solver for gyrokinetic plasma simulations\\nYear: 2017\\n\\nAbstract: With future exascale computers expected to have millions of compute units distributed among thousands of nodes, system faults are predicted to become more frequent. Fault tolerance will thus play a key role in HPC at this scale. In this paper we focus on solving the 5-dimensional gyrokinetic Vlasov-Maxwell equations using the application code GENE as it represents a high-dimensional and resource-intensive problem which is a natural candidate for exascale computing. We discuss the Fault-Tolerant Combination Technique, a resilient version of the Combination Technique, a method to increase the discretization resolution of existing PDE solvers. For the first time, we present an efficient, scalable and fault-tolerant implementation of this algorithm for plasma physics simulations based on a manager-worker model and test it under very realistic and pessimistic environments with simulated faults. We show that the Fault-Tolerant Combination Technique - an algorithm-based forward recovery method - can tolerate a large number of faults with a low overhead and at an acceptable loss in accuracy. Our parallel experiments with up to 32k cores show good scalability at a relative parallel efficiency of 93.61%. We conclude that algorithm-based solutions to fault tolerance are attractive for this type of problems.\",\n \"Title: Challenges in Creating a Sustainable Generic Research Data Infrastructure.\\nYear: 2017\\n\\nAbstract: Research data management is of the utmost importance in a world where research data is created with an ever increasing amount and rate and with a high variety across all scientific disciplines. This paper especially discusses software engineering challenges stemming from creating a long-living software system. It aims at providing a reference implementation for a federated research data infrastructure including interconnected individual repositories for communities and an overarching search based on metadata. The challenges involve a high variety of evolving requirements, the management and development of the distributed and federated infrastructure that are based on existing components, the piloting within the use cases, the efficient training of users, and how to enable the future sustainable operation.\",\n \"Title: A Holistic Scalable Implementation Approach Of The Lattice Boltzmann Method For Cpu/Gpu Heterogeneous Clusters\\nYear: 2017\\n\\nAbstract: Heterogeneous clusters are a widely utilized class of supercomputers assembled from different types of computing devices, for instance CPUs and GPUs, providing a huge computational potential. Programming them in a scalable way exploiting the maximal performance introduces numerous challenges such as optimizations for different computing devices, dealing with multiple levels of parallelism, the application of different programming models, work distribution, and hiding of communication with computation. We utilize the lattice Boltzmann method for fluid flow as a representative of a scientific computing application and develop a holistic implementation for large-scale CPU/GPU heterogeneous clusters. We review and combine a set of best practices and techniques ranging from optimizations for the particular computing devices to the orchestration of tens of thousands of CPU cores and thousands of GPUs. Eventually, we come up with an implementation using all the available computational resources for the lattice Boltzmann method operators. Our approach shows excellent scalability behavior making it future-proof for heterogeneous clusters of the upcoming architectures on the exaFLOPS scale. Parallel efficiencies of more than 90% are achieved leading to 2604.72 GLUPS utilizing 24,576 CPU cores and 2048 GPUs of the CPU/GPU heterogeneous cluster Piz Daint and computing more than 6.8 x 10(9) lattice cells.\",\n \"Title: Nonintrusive Uncertainty Analysis of Fluid-Structure Interaction with Spatially Adaptive Sparse Grids and Polynomial Chaos Expansion.\\nYear: 2018\\n\\nAbstract: The propagation of uncertainty in physical parameters of fluid-structure interaction problems is a challenging task-both mathematically and in terms of computational workload. In this paper, we employ nonintrusive polynomial chaos expansion and model the uncertainty in five independent input parameters that characterize both fluid and structure. We propose a novel methodology to compute the expansion's coefficients using spatially adaptive sparse grids and products of one-dimensional integrals, which exploits the tensor structure of both sparse grids and probabilistic space. Furthermore, with spatial adaptivity and modified basis functions, we keep the number of sparse grid points small. We test our approach in two test cases: (i) an elastic vertical flap in a channel flow and (ii) a computationally challenging, well-established benchmark. The outputs of interest are the x-deflection and total force on the structure. In the first test case, we consider six implementations of our methodology and two established methods based on sparse grid quadrature. We evaluate the performance (in terms of number of runs of the numerical solver) and accuracy of all eight methods. The results show that one variant of our approach outperforms all the other implementations. We apply our best variant to the benchmark scenario and address the high computational demands of the resulting problem using three levels of parallelism. Importantly, our approach is not restricted to fluid-structure interaction problems. We can address a broad spectrum of computationally expensive problems, provided that sparse grid approximations can be employed.\",\n \"Title: Optimization of a Sparse Grid-Based Data Mining Kernel for Architectures Using AVX-512\\nYear: 2018\\n\\nAbstract: Sparse grids have already been successfully used in various high-performance computing (HPC) applications, including data mining. In this article, we take a legacy classification kernel previously optimized for the AVX2 instruction set and investigate the benefits of using the newer AVX-S12-based multi-and many-core architectures. In particular, the Knights Landing (KNL) processor is used to study the possible performance gains of the code. Not all kernels benefit equally from such architectures, therefore choices in optimization steps and KNL cluster and memory modes need to be filtered through the lens of the code implementation at hand. With a less traditional approach of manual vectorization through instruction-level intrinsics, our kernel provides a differently faceted look into the optimization process. Observations stem from results obtained for node-and cluster-level classification simulations with up to 2\\n<sup xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">^</sup>\\n28 multidimensional training data points, using the CooLMUC-3cluster of the Leibniz Supercomputing Center (LRZ) in Garching, Germany.\",\n \"Title: Scheduling parallel eigenvalue computations in a quantum chemistry code\\nYear: 2010\\n\\nAbstract: The application of High Performance Computing to Quantum Chemical (QC) calculations faces many challenges. A central step is the solution of the generalized eigenvalue problem of a Hamilton matrix. Although in many cases its execution time is small relative to other numerical tasks, its complexity of O(N3) is higher, thus more significant in larger applications. For parallel QC codes, it therefore is advantageous to have a scalable solver for this step. We investigate the case where the symmetry of a molecule leads to a block-diagonal matrix structure, which complicates an efficient use of available parallel eigensolvers. We present a technique which employs a malleable parallel task scheduling (MPTS) algorithm to schedule instances of sequential and parallel eigensolver routines from LAPACK and ScaLAPACK. In this way, an efficient use of hardware resources is guaranteed while overall scalability is facilitated. Finally, we evaluate the proposed technique for electronic structure calculations of real chemical systems. For the systems considered, the performance was improved by factors of up to 8.4, compared to the previously used, nonmalleable parallel scheduling approach.\",\n \"Title: TweTriS: Twenty trillion-atom simulation\\nYear: 2019\\n\\nAbstract: AbstractSignificant improvements are presented for the molecular dynamics code ls1 mardyn — a linked cell-based code for simulating a large number of small, rigid molecules with application areas in chemical engineering. The changes consist of a redesign of the SIMD vectorization via wrappers, MPI improvements and a software redesign to allow memory-efficient execution with the production trunk to increase portability and extensibility. Two novel, memory-efficient OpenMP schemes for the linked cell-based force calculation are presented, which are able to retain Newton’s third law optimization. Comparisons to well-optimized Verlet list-based codes, such as LAMMPS and GROMACS, demonstrate the viability of the linked cell-based approach. The present version of ls1 mardyn is used to run simulations on entire supercomputers, maximizing the number of sampled atoms. Compared to the preceding version of ls1 mardyn on the entire set of 9216 nodes of SuperMUC, Phase 1, 27% more atoms are simulated. Weak scaling performance is increased by up to 40% and strong scaling performance by up to more than 220%. On Hazel Hen, strong scaling efficiency of up to 81% and 189 billion molecule updates per second is attained, when scaling from 8 to 7168 nodes. Moreover, a total of 20 trillion atoms is simulated at up to 88% weak scaling efficiency running at up to 1.33 PFLOPS. This represents a fivefold increase in terms of the number of atoms simulated to date.\",\n \"Title: AutoPas: Auto-Tuning for Particle Simulations\\nYear: 2019\\n\\nAbstract: The C++ library AutoPas aims at delivering optimal node-level performance for particle simulations. This paper describes the internally implemented algorithms, and how the library uses auto-tuning to dynamically select their optimal combination at run-time. Results are presented, which show that all available algorithms and configuration options have their specific advantages. To demonstrate the library's capabilities in relevant application settings, it has been integrated into the software package ls1 mardyn. An example of a realistic molecular dynamics simulation from thermodynamics is shown in which AutoPas detects a change in the best possible algorithm configuration. It adapts the simulation algorithm accordingly, sustaining optimal performance without additional user input.\",\n \"Title: A Generalized Spatially Adaptive Sparse Grid Combination Technique With Dimension-Wise Refinement\\nYear: 2021\\n\\nAbstract: Today, high-dimensional calculations can be found in almost all scientific disciplines. The application of machine learning and uncertainty quantification methods are common examples where high-dimensional problems appear. Typically, these problems are computationally expensive or even infeasible on current machines due to the curse of dimensionality. The sparse grid combination technique is one method to mitigate this effect, but it still does not generate optimal grids for many application scenarios. In such cases, adaptivity strategies are applied to further optimize the grid generation. Generally, adaptive grid generation strategies can be be classified as spatially or dimensionally adaptive. One of the most prominent examples is the dimension-adaptive combination technique, which is easy to implement and suitable for cases where dimensions contribute in different magnitudes to the accuracy of the solution. Unfortunately, spatial adaptivity is not possible in the standard combination technique due to the required regular structure of the grids. We therefore propose a new algorithmic variant of the combination technique that is based on the combination of rectilinear grids which can adapt themselves locally to the target function using 1-dimensional refinements. We further increase the efficiency by adjusting point levels with tree rebalancing. Results for numerical quadrature and interpolation show that we can significantly improve upon the standard combination technique and compete with or even surpass common spatially adaptive implementations with sparse grids. At the same time our method keeps the black-box property of the combination technique which makes it possible to apply it to black-box solvers.\",\n \"Title: N ways to simulate short-range particle systems: Automated algorithm selection with the node-level library AutoPas\\nYear: 2022\\n\\nAbstract: AutoPas is an open-source C++ library delivering optimal node-level performance by providing the ideal algorithmic configuration for an arbitrary scenario in a given short-range particle simulation. It acts as a black-box container, internally implementing an extensive set of algorithms, parallelization strategies, and optimizations that are combined dynamically according to the state of the simulation via auto-tuning. This paper gives an overview of the high-level user perspective, as well as the internal view, covering the implemented techniques and features. The library is showcased by incorporating it into the codes LAMMPS and ls1 mardyn, and by investigating various applications. We further outline node-level shared-memory performance and scalability of our auto-tuning software which is comparable to LAMMPS. (C) 2021 Elsevier B.V. All rights reserved.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":72,"string":"72"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13583,"string":"13,583"}}},{"rowIdx":13584,"cells":{"question_id":{"kind":"string","value":"Eamonn J Keogh"},"question":{"kind":"string","value":"Generate an abstract for the title \"Matrix Profile VI: Meaningful Multidimensional Motif Discovery\" using the following items: Time series motifs, exploratory data mining, multidimensional motifs, mSTAMP algorithm, diverse applications."},"answer":{"kind":"string","value":"Time series motifs are approximately repeating patterns in real-valued time series data. They are useful for exploratory data mining and are often used as inputs for various time series clustering, classification, segmentation, rule discovery, and visualization algorithms. Since the introduction of the first motif discovery algorithm for univariate time series in 2002, multiple efforts have been made to generalize motifs to the multidimensional case. In this work, we show that these efforts, which typically attempt to find motifs on all dimensions, will not produce meaningful motifs except in the most contrived situations. We explain this finding and introduce mSTAMP, an algorithm that allows meaningful discovery of multidimensional motifs. Beyond producing objectively and subjectively meaningful results, our algorithm has a host of additional advantages, including being much faster, requiring fewer parameters and supporting streaming data. We demonstrate the utility of our mSTAMP-based motif discovery framework on domains as diverse as audio processing, industry, and sports analytics."},"correct_answer":{"kind":"string","value":"Time series motifs are approximately repeating patterns in real-valued time series data. They are useful for exploratory data mining and are often used as inputs for various time series clustering, classification, segmentation, rule discovery, and visualization algorithms. Since the introduction of the first motif discovery algorithm for univariate time series in 2002, multiple efforts have been made to generalize motifs to the multidimensional case. In this work, we show that these efforts, which typically attempt to find motifs on all dimensions, will not produce meaningful motifs except in the most contrived situations. We explain this finding and introduce mSTAMP, an algorithm that allows meaningful discovery of multidimensional motifs. Beyond producing objectively and subjectively meaningful results, our algorithm has a host of additional advantages, including being much faster, requiring fewer parameters and supporting streaming data. We demonstrate the utility of our mSTAMP-based motif discovery framework on domains as diverse as audio processing, industry, and sports analytics."},"documents":{"kind":"list like","value":["Title: Beyond one billion time series: indexing and mining very large time series collections with i SAX2+.\nYear: 2014\n\nAbstract: There is an increasingly pressing need, by several applications in diverse domains, for developing techniques able to index and mine very large collections of time series. Examples of such applications come from astronomy, biology, the web, and other domains. It is not unusual for these applications to involve numbers of time series in the order of hundreds of millions to billions. However, all relevant techniques that have been proposed in the literature so far have not considered any data collections much larger than one-million time series. In this paper, we describe \\(i\\)SAX 2.0 and its improvements, \\(i\\)SAX 2.0 Clustered and \\(i\\)SAX2+, three methods designed for indexing and mining truly massive collections of time series. We show that the main bottleneck in mining such massive datasets is the time taken to build the index, and we thus introduce a novel bulk loading mechanism, the first of this kind specifically tailored to a time series index. We show how our methods allows mining on datasets that would otherwise be completely untenable, including the first published experiments to index one billion time series, and experiments in mining massive data from domains as diverse as entomology, DNA and web-scale image collections.","Title: Iterative Deepening Dynamic Time Warping for Time Series\nYear: 2002\n\nAbstract: this paper is organized as follows. Section 2 provides more detailedmotivation for IDDTW and a review of the classic DTW algorithm. Section 3 introducesthe IDDTW algorithm. In section 4 we experimentally compare IDDTW to DTW andEuclidean distance on real and synthetic datasets. Section 5 contains a discussion ofrelated work. Finally section 6 contains our conclusions and directions for future work","Title: CID: an efficient complexity-invariant distance for time series\nYear: 2014\n\nAbstract: The ubiquity of time series data across almost all human endeavors has produced a great interest in time series data mining in the last decade. While dozens of classification algorithms have been applied to time series, recent empirical evidence strongly suggests that simple nearest neighbor classification is exceptionally difficult to beat. The choice of distance measure used by the nearest neighbor algorithm is important, and depends on the invariances required by the domain. For example, motion capture data typically requires invariance to warping, and cardiology data requires invariance to the baseline (the mean value). Similarly, recent work suggests that for time series clustering, the choice of clustering algorithm is much less important than the choice of distance measure used.In this work we make a somewhat surprising claim. There is an invariance that the community seems to have missed, complexity invariance. Intuitively, the problem is that in many domains the different classes may have different complexities, and pairs of complex objects, even those which subjectively may seem very similar to the human eye, tend to be further apart under current distance measures than pairs of simple objects. This fact introduces errors in nearest neighbor classification, where some complex objects may be incorrectly assigned to a simpler class. Similarly, for clustering this effect can introduce errors by \"suggesting\" to the clustering algorithm that subjectively similar, but complex objects belong in a sparser and larger diameter cluster than is truly warranted.We introduce the first complexity-invariant distance measure for time series, and show that it generally produces significant improvements in classification and clustering accuracy. We further show that this improvement does not compromise efficiency, since we can lower bound the measure and use a modification of triangular inequality, thus making use of most existing indexing and data mining algorithms. We evaluate our ideas with the largest and most comprehensive set of time series mining experiments ever attempted in a single work, and show that complexity-invariant distance measures can produce improvements in classification and clustering in the vast majority of cases.","Title: Dimensionality Reduction for Fast Similarity Search in Large Time Series Databases.\nYear: 2001\n\nAbstract: The problem of similarity search in large time series databases has attracted muchattention recently. It is a non-trivial problem because of the inherent high dimensionality of thedata. The most promising solutions involve first performing dimensionality reduction on the data,and then indexing the reduced data with a spatial access method. Three major dimensionalityreduction techniques have been proposed, Singular Value Decomposition (SVD), the DiscreteFourier transform (DFT), and more...","Title: Derivative Dynamic Time Warping\nYear: 2001\n\nAbstract: this paper we address both these problems by introducing a modification of DTW.The crucial difference is in the features we consider when attempting to find the correctwarping. Rather than use the raw data, we consider only the (estimated) local derivativesof the data","Title: Efficiently Finding Arbitrarily Scaled Patterns in Massive Time Series Databases\nYear: 2003\n\nAbstract: The problem of efficiently finding patterns in massive time series databases has attracted great interest, and, at least for the Euclidean distance measure, may now be regarded as a solved problem. However in recent years there has been an increasing awareness that Euclidean distance is inappropriate for many real world applications. The limitations of Euclidean distance stems from the fact that it is very sensitive to distortions in the time axis. A partial solution to this problem, Dynamic Time Warping (DTW), aligns the time axis before calculating the Euclidean distance. However, DTW can only address the problem of local scaling. As we demonstrate in this work, uniform scaling may be just as important in many domains, including applications as diverse as bioinformatics, space telemetry monitoring and motion editing for computer animation. In this work, we demonstrate a novel technique to speed up similarity search under uniform scaling. As we will demonstrate, our technique is simple and intuitive, and can achieve a speedup of 2 to 3 orders of magnitude under realistic settings.","Title: Exact Discovery of Time Series Motifs\nYear: 2009\n\nAbstract: Time series motifs are pairs of individual time series, or subsequences of a longer time series, which are very similar to each other. As with their discrete analogues in computational biology, this similarity hints at structure which has been conserved for some reason and may therefore be of interest. Since the formalism of time series motifs in 2002, dozens of researchers have used them for diverse applications in many different domains. Because the obvious algorithm for computing motifs is quadratic in the number of items, more than a dozen approximate algorithms to discover motifs have been proposed in the literature. In this work, for the first time, we show a tractable exact algorithm to find time series motifs. As we shall show through extensive experiments, our algorithm is up to three orders of magnitude faster than brute-force search in large datasets. We further show that our algorithm is fast enough to be used as a subroutine in higher level data mining algorithms for anytime classification, near-duplicate detection and summarization, and we consider detailed case studies in domains as diverse as electroencephalograph interpretation and entomological telemetry data mining.","Title: Parameter-Free Audio Motif Discovery in Large Data Archives\nYear: 2013\n\nAbstract: The discovery of repeated structure, i.e. motifs/near-duplicates, is often the first step in exploratory data mining. As such, the last decade has seen extensive research efforts in motif discovery algorithms for text, DNA, time series, protein sequences, graphs, images, and video. Surprisingly, there has been less attention devoted to finding repeated patterns in audio sequences, in spite of their ubiquity in science and entertainment. While there is significant work for the special case of motifs in music, virtually all this work makes many assumptions about data (often to the point of being genre specific) and thus these algorithms do not generalize to audio sequences containing animal vocalizations, industrial processes, or a host of other domains that we may wish to explore. In this work we introduce a novel technique for finding audio motifs. Our method does not require any domain-specific tuning and is essentially parameter-free. We demonstrate our algorithm on very diverse domains, finding audio motifs in laboratory mice vocalizations, wild animal sounds, music, and human speech. Our experiments demonstrate that our ideas are effective in discovering objectively correct or subjectively plausible motifs. Moreover, we show our novel probabilistic early abandoning approach is efficient, being two to three orders of magnitude faster than brute-force search, and thus faster than real-time for most problems.","Title: Searching historical manuscripts for near-duplicate figures\nYear: 2011\n\nAbstract: In the next decade a majority of all the books ever published will be digitized and online. Naturally, most of the data in historical manuscripts is text, but there is also a large amount devoted to images. This observation is responsible for the dramatic increase in interest in query-by-content systems for historical documents. While querying/indexing systems can be useful, we believe that this domain is finally ready for unsupervised discovery of patterns. With this in mind, we introduce an efficient and scalable technique that can detect approximately repeated occurrences of images both within and between historical texts. We demonstrate that this ability to find repeated shapes allows us to do automatic annotation of manuscripts. We show the utility of our technique on datasets dating back to the fourteenth century.","Title: Quantitative Analysis of Live-Cell Growth at the Shoot Apex of Arabidopsis thaliana: Algorithms for Feature Measurement and Temporal Alignment\nYear: 2013\n\nAbstract: Study of the molecular control of organ growth requires establishment of the causal relationship between gene expression and cell behaviors. We seek to understand this relationship at the shoot apical meristem (SAM) of model plant Arabidopsis thaliana. This requires the spatial mapping and temporal alignment of different functional domains into a single template. Live-cell imaging techniques allow us to observe real-time organ primordia growth and gene expression dynamics at cellular resolution. In this paper, we propose a framework for the measurement of growth features at the 3D reconstructed surface of organ primordia, as well as algorithms for robust time alignment of primordia. We computed areas and deformation values from reconstructed 3D surfaces of individual primordia from live-cell imaging data. Based on these growth measurements, we applied a multiple feature landscape matching (LAM-M) algorithm to ensure a reliable temporal alignment of multiple primordia. Although the original landscape matching (LAM) algorithm motivated our alignment approach, it sometimes fails to properly align growth curves in the presence of high noise/distortion. To overcome this shortcoming, we modified the cost function to consider the landscape of the corresponding growth features. We also present an alternate parameter-free growth alignment algorithm which performs as well as LAM-M for high-quality data, but is more robust to the presence of outliers or noise. Results on primordia and guppy evolutionary growth data show that the proposed alignment framework performs at least as well as the LAM algorithm in the general case, and significantly better in the case of increased noise.","Title: Supporting exact indexing of arbitrarily rotated shapes and periodic time series under Euclidean and warping distance measures\nYear: 2009\n\nAbstract: Shape matching and indexing is important topic in its own right, and is a fundamental subroutine in most shape data mining\n algorithms. Given the ubiquity of shape, shape matching is an important problem with applications in domains as diverse as\n biometrics, industry, medicine, zoology and anthropology. The distance/similarity measure for used for shape matching must\n be invariant to many distortions, including scale, offset, noise, articulation, partial occlusion, etc. Most of these distortions\n are relatively easy to handle, either in the representation of the data or in the similarity measure used. However, rotation\n invariance is noted in the literature as being an especially difficult challenge. Current approaches typically try to achieve\n rotation invariance in the representation of the data, at the expense of discrimination ability, or in the distance measure,\n at the expense of efficiency. In this work, we show that we can take the slow but accurate approaches and dramatically speed\n them up. On real world problems our technique can take current approaches and make them four orders of magnitude faster without\n false dismissals. Moreover, our technique can be used with any of the dozens of existing shape representations and with all\n the most popular distance measures including Euclidean distance, dynamic time warping and Longest Common Subsequence. We further\n show that our indexing technique can be used to index star light curves, an important type of astronomical data, without modification.","Title: On the need for time series data mining benchmarks: a survey and empirical demonstration\nYear: 2003\n\nAbstract: In the last decade there has been an explosion of interest in mining time series data. Literally hundreds of papers have introduced new algorithms to index, classify, cluster and segment time series. In this work we make the following claim. Much of this work has very little utility because the contribution made (speed in the case of indexing, accuracy in the case of classification and clustering, model accuracy in the case of segmentation) offer an amount of \"improvement\" that would have been completely dwarfed by the variance that would have been observed by testing on many real world datasets, or the variance that would have been observed by changing minor (unstated) implementation details.To illustrate our point, we have undertaken the most exhaustive set of time series experiments ever attempted, re-implementing the contribution of more than two dozen papers, and testing them on 50 real world, highly diverse datasets. Our empirical results strongly support our assertion, and suggest the need for a set of time series benchmarks and more careful empirical evaluation in the data mining community.","Title: A novel bit level time series representation with implication of similarity search and clustering\nYear: 2005\n\nAbstract: Because time series are a ubiquitous and increasingly prevalent type of data, there has been much research effort devoted to time series data mining recently. As with all data mining problems, the key to effective and scalable algorithms is choosing the right representation of the data. Many high level representations of time series have been proposed for data mining. In this work, we introduce a new technique based on a bit level approximation of the data. The representation has several important advantages over existing techniques. One unique advantage is that it allows raw data to be directly compared to the reduced representation, while still guaranteeing lower bounds to Euclidean distance. This fact can be exploited to produce faster exact algorithms for similarly search. In addition, we demonstrate that our new representation allows time series clustering to scale to much larger datasets.","Title: Scaling up Dynamic Time Warping to Massive Dataset\nYear: 1999\n\nAbstract: There has been much recent interest in adapting da ta mining algorithms to time series databases. Many of these algorithms nee d to compare time series. Typically some variation or extension of Euclidean distance is used. However, as we demonstrate in this paper, Euclidean distance can b e an extremely brittle distance measure. Dynamic time warping (DTW) has been suggested as a technique to allow more robust distance calculations, however it is co mputationally expensive. In this paper we introduce a modification of DTW which operates on a higher level abstraction of the data, in particular, a piecewise linear representation. We demonstrate that our approach allows us to outperfo rm DTW by one to three orders of magnitude. We experimentally evaluate our approach on medical, astronomical and sign language data.","Title: A Bit Level Representation for Time Series Data Mining with Shape Based Similarity\nYear: 2006\n\nAbstract: Clipping is the process of transforming a real valued series into a sequence of bits representing whether each data is above or below the average. In this paper, we argue that clipping is a useful and flexible transformation for the exploratory analysis of large time dependent data sets. We demonstrate how time series stored as bits can be very efficiently compressed and manipulated and that, under some assumptions, the discriminatory power with clipped series is asymptotically equivalent to that achieved with the raw data. Unlike other transformations, clipped series can be compared directly to the raw data series. We show that this means we can form a tight lower bounding metric for Euclidean and Dynamic Time Warping distance and hence efficiently query by content. Clipped data can be used in conjunction with a host of algorithms and statistical tests that naturally follow from the binary nature of the data. A series of experiments illustrate how clipped series can be used in increasingly complex ways to achieve better results than other popular representations. The usefulness of the proposed representation is demonstrated by the fact that the results with clipped data are consistently better than those achieved with a Wavelet or Discrete Fourier Transformation at the same compression ratio for both clustering and query by content. The flexibility of the representation is shown by the fact that we can take advantage of a variable Run Length Encoding of clipped series to define an approximation of the Kolmogorov complexity and hence perform Kolmogorov based clustering.","Title: Integrating Lite-Weight but Ubiquitous Data Mining into GUI Operating Systems\nYear: 2005\n\nAbstract: Most visualization tools introduced in the literature are specialized for a particular task. In this work, we introduce a novel framework which allows visualization to take place in the background of normal day to day operations of any GUI based operating system such as MS Windows, OS X or Linux. Our system works by replacing the standard file icons with automatically generated icons that reflect the contents of the files in a principled way. We call such icons Intelligent Icons. While there is little utility in examining an individual icon, examining groups of them provides a greater possibility of unexpected and serendipitous discoveries. The utility of Intelligent Icons can be further enhanced by arranging them on the screen in a way that reflects their similarity/differences. We demonstrate the utility of our approach on data as diverse as DNA, text files, electrocardiograms, and Space Shuttle telemetry. In addition we show that our system is unique in also supporting fast and intuitive similarity search.","Title: Ensemble-index: a new approach to indexing large databases\nYear: 2001\n\nAbstract: The problem of similarity search (query-by-content) has attracted much research interest. It is a difficult problem because of the inherently high dimensionality of the data. The most promising solutions involve performing dimensionality reduction on the data, then indexing the reduced data with a multidimensional index structure. Many dimensionality reduction techniques have been proposed, including Singular Value Decomposition (SVD), the Discrete Fourier Transform (DFT), the Discrete Wavelet Transform (DWT) and Piecewise Polynomial Approximation. In this work, we introduce a novel framework for using ensembles of two or more representations for more efficient indexing. The basic idea is that instead of committing to a single representation for an entire dataset, different representations are chosen for indexing different parts of the database. The representations are chosen based upon a local view of the database. For example, sections of the data that can achieve a high fidelity representation with wavelets are indexed as wavelets, but highly spectral sections of the data are indexed using the Fourier transform. At query time, it is necessary to search several small heterogeneous indices, rather than one large homogeneous index. As we will theoretically and empirically demonstrate this results in much faster query response times.","Title: A Simple Dimensionality Reduction Technique for Fast Similarity Search in Large Time Series Databases\nYear: 2000\n\nAbstract: We address the problem of similarity search in large time series databases. We introduce a novel-dimensionality reduction technique that supports an indexing algorithm that is more than an order of magnitude faster than the previous best known method. In addition to being much faster our approach has numerous other advantages. It is simple to understand and implement, allows more flexible distance measures including weighted Euclidean queries and the index can be built in linear time. We call our approach PCA-indexing (Piece-wise Constant Approximation) and experimentally validate it on space telemetry, financial, astronomical, medical and synthetic data.","Title: A Practical Tool for Visualizing and Data Mining Medical Time Series\nYear: 2005\n\nAbstract: The increasing interest in time series data mining has had surprisingly little impact on real world medical applications. Practitioners who work with time series on a daily basis rarely take advantage of the wealth of tools that the data mining community has made available. In this work, we attempt to address this problem by introducing a parameter-light tool that allows users to efficiently navigate through large collections of time series. Our approach extracts features from a time series of arbitrary length and uses information about the relative frequency of these features to color a bitmap in a principled way. By visualizing the similarities and differences within a collection of bitmaps, a user can quickly discover clusters, anomalies, and other regularities within the data collection. We demonstrate the utility of our approach with a set of comprehensive experiments on real datasets from a variety of medical domains.","Title: Indexing large human-motion databases\nYear: 2004\n\nAbstract: Data-driven animation has become the industry standard for computer games and many animated movies and special effects. In particular, motion capture data recorded from live actors, is the most promising approach offered thus far for animating realistic human characters. However, the manipulation of such data for general use and re-use is not yet a solved problem. Many of the existing techniques dealing with editing motion rely on indexing for annotation, segmentation, and re-ordering of the data. Euclidean distance is inappropriate for solving these indexing problems because of the inherent variability found in human motion. The limitations of Euclidean distance stems from the fact that it is very sensitive to distortions in the time axis. A partial solution to this problem, Dynamic Time Warping (DTW), aligns the time axis before calculating the Euclidean distance. However, DTW can only address the problem of local scaling. As we demonstrate in this paper, global or uniform scaling is just as important in the indexing of human motion. We propose a novel technique to speed up similarity search under uniform scaling, based on bounding envelopes. Our technique is intuitive and simple to implement. We describe algorithms that make use of this technique, we perform an experimental analysis with real datasets, and we evaluate it in the context of a motion capture processing system. The results demonstrate the utility of our approach, and show that we can achieve orders of magnitude of speedup over the brute force approach, the only alternative solution currently available.","Title: Efficiently finding unusual shapes in large image databases\nYear: 2008\n\nAbstract: Among the visual features of multimedia content, shape is of particular interest because humans can often recognize objects solely on the basis of shape. Over the past three decades, there has been a great deal of research on shape analysis, focusing mostly on shape indexing, clustering, and classification. In this work, we introduce the new problem of finding shape discords, the most unusual shapes in a collection. We motivate the problem by considering the utility of shape discords in diverse domains including zoology, microscopy, anthropology, and medicine. While the brute force search algorithm has quadratic time complexity, we avoid this untenable lethargy by using locality-sensitive hashing to estimate similarity between shapes which enables us to reorder the search more efficiently and thus extract the maximum benefit from an admissible pruning strategy we introduce. An extensive experimental evaluation demonstrates that our approach is empirically linear in time.","Title: Annotating Historical Archives of Images\nYear: 2010\n\nAbstract: Recent programs like the Million Book Project and Google Print Library Project have archived several million books in digital format, and within a few years a significant fraction of world's books will be online. While the majority of the data will naturally be text, there will also be tens of millions of pages of images. Many of these images will defy automation annotation for the foreseeable future, but a considerable fraction of the images may be amiable to automatic annotation by algorithms that can link the historical image with a modern contemporary, with its attendant metatags. To perform this linking, there must be a suitable distance measure that appropriately combines the relevant features of shape, color, texture and text. However, the best combination of these features will vary from application to application and even from one manuscript to another. In this work, the authors propose a simple technique to learn the distance measure by perturbing the training set in a principled way.","Title: A Grid-Based Index Method for Time Warping Distance\nYear: 2004\n\nAbstract: Recently DTW (dynamic time warping) has been recognized as the most robust distance function to measure the similarity between\n two time series, and this fact has spawned a flurry of research on this topic. Most indexing methods proposed for DTW are\n based on the R-tree structure. Because of high dimensionality and loose lower bounds for time warping distance, the pruning\n power of these tree structures are quite weak, resulting in inefficient search. In this paper, we propose a dimensionality\n reduction method motivated by observations about the inherent character of each time series. A very compact index file is\n constructed. By scanning the index file, we can get a very small candidate set, so that the number of page access is dramatically\n reduced. We demonstrate the effectiveness of our approach on real and synthetic datasets.","Title: Fast similarity search in the presence of longitudinal scaling in time series databases\nYear: 1997\n\nAbstract: Abstract: The problem of finding patterns of interest in time series databases (query by content) is an important one, with applications in virtually every field of science. A variety of approaches have been suggested. These approaches are robust to noise, offset translation, and amplitude scaling to varying degrees. However, they are all extremely sensitive to scaling in the time axis (longitudinal scaling). We present a method for similarity search that is robust to scaling in the time axis, in addition to noise, offset translation, and amplitude scaling. The method has been tested on medical, financial, space telemetry and artificial data. Furthermore the method is exceptionally fast, with the predicted 2 to 4 orders of magnitude speedup actually observed. The method uses a piecewise linear representation of the original data. We also introduce a new algorithm which both decides the optimal number of linear segments to use, and produces the actual linear representation.","Title: LB_Keogh supports exact indexing of shapes under rotation invariance with arbitrary representations and distance measures\nYear: 2006\n\nAbstract: The matching of two-dimensional shapes is an important problem with applications in domains as diverse as biometrics, industry, medicine and anthropology. The distance measure used must be invariant to many distortions, including scale, offset, noise, partial occlusion, etc. Most of these distortions are relatively easy to handle, either in the representation of the data or in the similarity measure used. However rotation invariance seems to be uniquely difficult. Current approaches typically try to achieve rotation invariance in the representation of the data, at the expense of discrimination ability, or in the distance measure, at the expense of efficiency. In this work we show that we can take the slow but accurate approaches and dramatically speed them up. On real world problems our technique can take current approaches and make them four orders of magnitude faster, without false dismissals. Moreover, our technique can be used with any of the dozens of existing shape representations and with all the most popular distance measures including Euclidean distance, Dynamic Time Warping and Longest Common Subsequence.","Title: Flying Insect Classification with Inexpensive Sensors.\nYear: 2014\n\nAbstract: The ability to use inexpensive, noninvasive sensors to accurately classify flying insects would have significant implications for entomological research, and allow for the development of many useful applications in vector control for both medical and agricultural entomology. Given this, the last sixty years have seen many research efforts on this task. To date, however, none of this research has had a lasting impact. In this work, we explain this lack of progress. We attribute the stagnation on this problem to several factors, including the use of acoustic sensing devices, the overreliance on the single feature of wingbeat frequency, and the attempts to learn complex models with relatively little data. In contrast, we show that pseudo-acoustic optical sensors can produce vastly superior data, that we can exploit additional features, both intrinsic and extrinsic to the insect’s flight behavior, and that a Bayesian classification approach allows us to efficiently learn classification models that are very robust to overfitting. We demonstrate our findings with large scale experiments, as measured both by the number of insects and the number of species considered.","Title: Finding the most unusual time series subsequence: algorithms and applications\nYear: 2007\n\nAbstract: In this work we introduce the new problem of finding time seriesdiscords. Time series discords are subsequences of longer time series that are maximally different to all the rest of the time series subsequences. They thus capture the sense of the most unusual subsequence within a time series. While discords have many uses for data mining, they are particularly attractive as anomaly detectors because they only require one intuitive parameter (the length of the subsequence) unlike most anomaly detection algorithms that typically require many parameters. While the brute force algorithm to discover time series discords is quadratic in the length of the time series, we show a simple algorithm that is three to four orders of magnitude faster than brute force, while guaranteed to produce identical results. We evaluate our work with a comprehensive set of experiments on diverse data sources including electrocardiograms, space telemetry, respiration physiology, anthropological and video datasets.","Title: A decade of progress in indexing and mining large time series databases\nYear: 2006\n\nAbstract: Time series data is ubiquitous; large volumes of time series data are routinely created in scientific, industrial, entertainment, medical and biological domains. Examples include gene expression data, electrocardiograms, electroencephalograms, gait analysis, stock market quotes, space telemetry etc. Although statisticians have worked with time series for more than a century, many of their techniques hold little utility for researchers working with massive time series databases.A decade ago, a seminal paper by Faloutsos, Ranganathan, Manolopoulos appeared in SIGMOD. The paper, Fast Subsequence Matching in Time-Series Databases, has spawned at least a thousand references and extensions in the database/ data mining and information retrieval communities. This tutorial will summarize the decade of progress since this influential paper appeared.","Title: MDL-based time series clustering.\nYear: 2012\n\nAbstract: Time series data are pervasive across all human endeavors, and clustering is arguably the most fundamental data mining application. Given this, it is somewhat surprising that the problem of time series clustering from a single stream remains largely unsolved. Most work on time series clustering considers the clustering of individual time series that have been carefully extracted from their original context, for example, gene expression profiles, individual heartbeats, or individual gait cycles. The few attempts at clustering time series streams have been shown to be objectively incorrect in some cases, and in other cases shown to work only on the most contrived synthetic datasets by carefully adjusting a large set of parameters. In this work, we make two fundamental contributions that allow for the first time, the meaningful clustering of subsequences from a time series stream. First, we show that the problem definition for time series clustering from streams currently used is inherently flawed, and a new definition is necessary. Second, we show that the minimum description length framework offers an efficient, effective, and essentially parameter-free method for time series clustering. We show that our method produces objectively correct results on a wide variety of datasets from medicine, speech recognition, zoology, gesture recognition, and industrial process analyses.","Title: Finding surprising patterns in a time series database in linear time and space\nYear: 2002\n\nAbstract: The problem of finding a specified pattern in a time series database (i.e. query by content) has received much attention and is now a relatively mature field. In contrast, the important problem of enumerating all surprising or interesting patterns has received far less attention. This problem requires a meaningful definition of \"surprise\", and an efficient search technique. All previous attempts at finding surprising patterns in time series use a very limited notion of surprise, and/or do not scale to massive datasets. To overcome these limitations we introduce a novel technique that defines a pattern surprising if the frequency of its occurrence differs substantially from that expected by chance, given some previously seen data.","Title: HOT SAX: Efficiently Finding the Most Unusual Time Series Subsequence\nYear: 2005\n\nAbstract: In this work, we introduce the new problem of finding time series discords. Time series discords are subsequences of a longer time series that are maximally different to all the rest of the time series subsequences. They thus capture the sense of the most unusual subsequence within a time series. Time series discords have many uses for data mining, including improving the quality of clustering, data cleaning, summarization, and anomaly detection. As we will show, discords are particularly attractive as anomaly detectors because they only require one intuitive parameter (the length of the subsequence) unlike most anomaly detection algorithms that typically require many parameters. We evaluate our work with a comprehensive set of experiments. In particular, we demonstrate the utility of discords with objective experiments on domains as diverse as Space Shuttle telemetry monitoring, medicine, surveillance, and industry, and we demonstrate the effectiveness of our discord discovery algorithm with more than one million experiments, on 82 different datasets from diverse domains.","Title: The Asymmetric Approximate Anytime Join: A New Primitive with Applications to Data Mining\nYear: 2008\n\nAbstract: It has long been noted that many data mining algorithms can be built on top of join algorithms. This has lead to a wealth of recent work on efficiently supporting such joins with various indexing techniques. However, there are many applications which are characterized by two special conditions, firstly the two datasets to be joined are of radically different sizes, a situation we call an asymmetric join. Secondly, the two datasets are not, and possibly can not be indexed for some reason. In such circumstances the time complexity is proportional to the product of the number of objects in each of the two datasets, an untenable proposition in most cases. In this work we make two contributions to mitigate this situation. We argue that for many applications, an exact solution to the problem is not required, and we show that by framing the problem as an anytime algorithm we can extract most of the benefit of a join in a small fraction of the time taken by the full algorithm. In situations where the exact answer is required, we show that we can quickly index just the smaller dataset on the fly, and greatly speed up the exact computation. We motivate and empirically confirm the utility of our ideas with case studies on problems as diverse as batch classification, anomaly detection and annotation of historical manuscripts.","Title: Locally adaptive dimensionality reduction for indexing large time series databases\nYear: 2002\n\nAbstract: Similarity search in large time series databases has attracted much research interest recently. It is a difficult problem because of the typically high dimensionality of the data. The most promising solutions involve performing dimensionality reduction on the data, then indexing the reduced data with a multidimensional index structure. Many dimensionality reduction techniques have been proposed, including Singular Value Decomposition (SVD), the Discrete Fourier transform (DFT), and the Discrete Wavelet Transform (DWT). In this article, we introduce a new dimensionality reduction technique, which we call Adaptive Piecewise Constant Approximation (APCA). While previous techniques (e.g., SVD, DFT and DWT) choose a common representation for all the items in the database that minimizes the global reconstruction error, APCA approximates each time series by a set of constant value segments of varying lengths such that their individual reconstruction errors are minimal. We show how APCA can be indexed using a multidimensional index structure. We propose two distance measures in the indexed space that exploit the high fidelity of APCA for fast searching: a lower bounding Euclidean distance approximation, and a non-lower-bounding, but very tight, Euclidean distance approximation, and show how they can support fast exact searching and even faster approximate searching on the same index structure. We theoretically and empirically compare APCA to all the other techniques and demonstrate its superiority.","Title: An Augmented Visual Query Mechanism for Finding Patterns in Time Series Data\nYear: 2002\n\nAbstract: Relatively few query tools exist for data exploration and pattern identification in time series data sets. In previous work we introduced Time-boxes. Timeboxes are rectangular, direct-manipulation queries for studying time-series datasets. We demonstrated how Timeboxes can be used to support interactive exploration via dynamic queries, along with overviews of query results and drag-and-drop support for query-by-example. In this paper, we extend our work by introducing Variable Time Timeboxes (VTT). VTTs are a natural generalization of Timeboxes, which permit the specification of queries that allow a degree of uncertainty in the time axis. We carefully motivate the need for these more expressive queries, and demonstrate the utility of our approach on several data sets.","Title: Mining Historical Documents for Near-Duplicate Figures\nYear: 2011\n\nAbstract: The increasing interest in archiving all of humankind's cultural artifacts has resulted in the digitization of millions of books, and soon a significant fraction of the world's books will be online. Most of the data in historical manuscripts is text, but there is also a significant fraction devoted to images. This fact has driven much of the recent increase in interest in query-by-content systems for images. While querying/indexing systems can undoubtedly be useful, we believe that the historical manuscript domain is finally ripe for true unsupervised discovery of patterns and regularities. To this end, we introduce an efficient and scalable system which can detect approximately repeated occurrences of shape patterns both within and between historical texts. We show that this ability to find repeated shapes allows automatic annotation of manuscripts, and allows users to trace the evolution of ideas. We demonstrate our ideas on datasets of scientific and cultural manuscripts dating back to the fourteenth century.","Title: An Enhanced Representation of Time Series Which Allows Fast and Accurate Classification, Clustering and Relevance Feedback\nYear: 1998\n\nAbstract: We introduce an extended representation of time series that allows fast, accurate classification and clustering in addition to the ability to explore time series data in a relevance feedback framework. The representation consists of piece- wise linear segments to represent shape and a weight vector that contains the relative importance of each individual linear segment. In the classification context, the weights are learned automatically as part of the training cycle. In the relevance feedback context, the weights are determined by an interactive and iterative process in which users rate various choices presented to them. Our representation allows a user to define a variety of similarity measures that can be tailored to specific domains. We demonstrate our approach on space telemetry, medical and synthetic data.","Title: Making Time-Series Classification More Accurate Using Learned Constraints\nYear: 2004\n\nAbstract: It has long been known that Dynamic Time Warping (DTW) is superior to Euclidean distance for classification and clustering of time series. However, until lately, most research has utilized Euclidean distance because it is more efficiently calculated. A recently introduced technique that greatly mitigates DTWs demanding CPU time has sparked a flurry of research activity. However, the technique and its many extensions still only allow DTW to be applied to moderately large datasets. In addition, almost all of the research on DTW has focused exclusively on speeding up its calculation; there has been little work done on improving its accuracy. In this work, we target the accuracy aspect of DTW performance and introduce a new framework that learns arbitrary constraints on the warping path of the DTW calculation. Apart from improving the accuracy of classification, our technique as a side effect speeds up DTW by a wide margin as well. We show the utility of our approach on datasets from diverse domains and demonstrate significant gains in accuracy and efficiency.","Title: Classification Of Multi-Dimensional Streaming Time Series By Weighting Each Classifier'S Track Record\nYear: 2013\n\nAbstract: Extensive research on time series classification in the last decade has produced fast and accurate algorithms for the single-dimensional case. However, the increasing prevalence of inexpensive sensors has reinforced the need for algorithms to handle multi-dimensional time series. For example, modern smartphones have at least a dozen sensors capable of producing streaming time series, and hospital-based (and increasingly, home-based) medical devices can produce time series streams from more than twenty sensors. The two most common ways to generalize from single to multi-dimensional data are to use all the streams or just the single best stream as determined at training time. However, as we show here, both approaches can be very brittle. Moreover, neither approach exploits the observation that different sensors may be considered \"experts\" on different classes. In this work, we introduce a novel framework for multi-dimensional time series classification that weights the class prediction from each time series stream. These weights are based not only on each stream's previous track record on the class it is currently predicting, but also on the distance from the unlabeled object. As we demonstrate with extensive experiments on real data, our method is more accurate than current approaches and particularly robust in the face of concept drift or sensor noise.","Title: SIGKDD demo: sensors and software to allow computational entomology, an emerging application of data mining\nYear: 2011\n\nAbstract: The history of humankind is intimately connected to insects. Insect borne diseases kill a million people and destroy tens of billions of dollars worth of crops annually. However, at the same time, beneficial insects pollinate the majority of crop species, and it has been estimated that approximately one third of all food consumed by humans is directly pollinated by bees alone. Given the importance of insects in human affairs, it is somewhat surprising that computer science has not had a larger impact in entomology. We believe that recent advances in sensor technology are beginning change this, and a new field of Computational Entomology will emerge. We will demonstrate an inexpensive sensor that allows us to capture data from flying insects, and the software that allows us to analyze the data. Moreover, we will distribute both the sensors and software for free, to parties willing to take part in a crowdsourcing project on insect classification.","Title: Semi-supervised time series classification\nYear: 2006\n\nAbstract: The problem of time series classification has attracted great interest in the last decade. However current research assumes the existence of large amounts of labeled training data. In reality, such data may be very difficult or expensive to obtain. For example, it may require the time and expertise of cardiologists, space launch technicians, or other domain specialists. As in many other domains, there are often copious amounts of unlabeled data available. For example, the PhysioBank archive contains gigabytes of ECG data. In this work we propose a semi- supervised technique for building time series classifiers. While such algorithms are well known in text domains, we will show that special considerations must be made to make them both efficient and effective for the time series domain. We evaluate our work with a comprehensive set of experiments on diverse data sources including electrocardiograms, handwritten documents, manufacturing, and video datasets. The experimental results demonstrate that our approach requires only a handful of labeled examples to construct accurate classifiers.","Title: Towards parameter-free data mining\nYear: 2004\n\nAbstract: Most data mining algorithms require the setting of many input parameters. Two main dangers of working with parameter-laden algorithms are the following. First, incorrect settings may cause an algorithm to fail in finding the true patterns. Second, a perhaps more insidious problem is that the algorithm may report spurious patterns that do not really exist, or greatly overestimate the significance of the reported patterns. This is especially likely when the user fails to understand the role of parameters in the data mining process.Data mining algorithms should have as few parameters as possible, ideally none. A parameter-free algorithm would limit our ability to impose our prejudices, expectations, and presumptions on the problem at hand, and would let the data itself speak to us. In this work, we show that recent results in bioinformatics and computational theory hold great promise for a parameter-free data-mining paradigm. The results are motivated by observations in Kolmogorov complexity theory. However, as a practical matter, they can be implemented using any off-the-shelf compression algorithm with the addition of just a dozen or so lines of code. We will show that this approach is competitive or superior to the state-of-the-art approaches in anomaly/interestingness detection, classification, and clustering with empirical tests on time series/DNA/text/video datasets.","Title: Discovering the Intrinsic Cardinality and Dimensionality of Time Series Using MDL\nYear: 2011\n\nAbstract: Most algorithms for mining or indexing time series data do not operate directly on the original data, but instead they consider alternative representations that include transforms, quantization, approximation, and multi-resolution abstractions. Choosing the best representation and abstraction level for a given task/dataset is arguably the most critical step in time series data mining. In this paper, we investigate techniques to discover the natural intrinsic representation model, dimensionality and alphabet cardinality of a time series. The ability to discover these intrinsic features has implications beyond selecting the best parameters for particular algorithms, as characterizing data in such a manner is useful in its own right and an important sub-routine in algorithms for classification, clustering and outlier discovery. We will frame the discovery of these intrinsic features in the Minimal Description Length (MDL) framework. Extensive empirical tests show that our method is simpler, more general and significantly more accurate than previous methods, and has the important advantage of being essentially parameter-free.","Title: Intelligent Icons: Integrating Lite-Weight Data Mining and Visualization into GUI Operating Systems\nYear: 2006\n\nAbstract: The vast majority of visualization tools introduced so far are specialized pieces of software that run explicitly on a particular dataset at a particular time for a particular purpose. In this work we introduce a novel framework for allowing visualization to take place in the background of normal day-to-day operation of any GUI based operation system. Our system works by replacing the standard file icons with automatically created icons that reflect the contents of the files in a principled way. We call such icons INTELLIGENT ICONS. The utility of Intelligent Icons is further enhanced by arranging them in a way that reflects their similarity/differences. We demonstrate the utility of our approach on diverse applications.","Title: An Indexing Scheme for Fast Similarity Search in Large Time Series Databases\nYear: 1999\n\nAbstract: We address the problem of similarity search in large time series databases. We introduce a novel indexing algorithm that allows faster retrieval. The index is formed by creating bins that contain time series subsequences of approximately the same shape. For each bin, we can quickly calculate a lower-bound on the distance between a given query and the most similar element of the bin. This bound allows us to search the bins in best first order, and to prune some bins from the search space without having to examine the contents. Additional speedup is obtained by optimizing the data within the bins such that we can avoid having to compare the query to every item in the bin. We call our approach STB-indexing and experimentally validate it on space telemetry, medical and synthetic data, demonstrating approximately an order of magnitude speed-up.","Title: A compression‐based distance measure for texture\nYear: 2010\n\nAbstract: The analysis of texture is an important subroutine in application areas as diverse as biology, medicine, robotics, and forensic science. While the last three decades have seen extensive research in algorithms to measure texture similarity, almost all existing methods require the careful setting of many parameters. There are many problems associated with a lot of parameters, the most obvious of which is that with many parameters to fit, it is very difficult to avoid overfitting. In this work, we propose to extend recent advances in Kolmogorov complexity‐based similarity measures to texture matching problems. These Kolmogorov‐based methods have been shown to be very useful in intrinsically discrete domains such as DNA, protein sequences, MIDI music, and natural languages; however, they are not well defined for real‐valued data. To address this, we introduce a very simple idea, the Campana‐Keogh (CK) video compression‐based method for texture measures. These measures utilize video compressors to approximate the Kolmogorov complexity. Using the parameter‐free CK method, we novely utilize lossy compression to create an efficient and robust parameter‐lite texture similarity measure: the CK‐1 distance measure. We demonstrate the utility of our measure with extensive empirical evaluations on real‐world case studies drawn from nematology, arachnology, entomology, medicine, forensics, texture analysis benchmarks, and many other domains. Copyright © 2010 Wiley Periodicals, Inc. Statistical Analysis and Data Mining 3: 381‐398, 2010 © 2010 Wiley Periodicals, Inc.","Title: Autocannibalistic and Anyspace Indexing Algorithms with Application to Sensor Data Mining\nYear: 2009\n\nAbstract: ABSTRACT Efficient indexing ,is at ,the heart of many ,data mining algorithms. A simple and extremely effective algorithm for indexing under any metric space was introduced in 1991 by Orchard. Orchard’s algorithm ,has ,not received ,much attention in the ,data mining ,and ,database ,community because of a fatal flaw; it requires quadratic space. In this work,we show ,that we can ,produce a reduced ,version of Orchard’s algorithm that requires much less space, but produces,nearly identical speedup. We achieve ,this by casting the algorithm in an anyspace framework, allowing deployed applications to take as much,of an,index as their main memory/sensor can afford. As we shall demonstrate, this ability to create an ,anyspace algorithm also allows us tocreate auto-cannibalistic algorithms. Auto-cannibalistic algorithms are algorithms which ,initially require a certain amount of space to index or classify data, but if unexpected circumstances require them to store additional information, they can dynamically ,delete parts of themselves ,to make room for the new data. We demonstrate,the utility of auto- cannibalistic algorithms ,in a ,fielded project on insect monitoring with low power sensors, and a simple autonomous,robot application. Keywords Indexing, Anyspace Algorithms, Data Mining, Sensors.","Title: An Online Algorithm for Segmenting Time Series\nYear: 2001\n\nAbstract: In recent years, there has been an explosion of interest in mining time-series databases. As with most computer science problems, representation of the data is the key to efficient and effective solutions. One of the most commonly used representations is piecewise linear approximation. This representation has been used by various researchers to support clustering, classification, indexing and association rule mining of time-series data. A variety of algorithms have been proposed to obtain this representation, with several algorithms having been independently rediscovered several times. In this paper, we undertake the first extensive review and empirical comparison of all proposed techniques. We show that all these algorithms have fatal flaws from a data-mining perspective. We introduce a novel algorithm that we empirically show to be superior to all others in the literature","Title: Data mining and information retrieval in time series/multimedia databases\nYear: 2006\n\nAbstract: Time series and multimedia data are ubiquitous; large volumes of such data are routinely created in scientific, industrial, entertainment, medical and biological domains. Examples include gene expression data, X-rays, electrocardiograms, electroencephalograms, gait analysis, stock market quotes, space telemetry etc.A decade ago, a seminal paper by Faloutsos, Ranganathan, Manolopoulos appeared in SIGMOD [1]. The paper, Fast Subsequence Matching in Time-Series Databases, has spawned at least a thousand references and extensions in the database/data mining and information retrieval communities. This tutorial will summarize the decade of progress since this influential paper appeared.","Title: How to Do Good Data Mining Research and Get it Published in Top Venues\nYear: 2010\n\nAbstract: While ICDM has traditionally enjoyed an unusually high quality of reviewing, there is no doubt that publishing in ICDM is very challenging. In this tutorial Dr. Keogh will demonstrate some simple ideas to enhance the probability of success in getting your paper published in a top data mining conference, and after the work is published, getting it highly cited.","Title: Compression-based data mining of sequential data\nYear: 2007\n\nAbstract: The vast majority of data mining algorithms require the setting of many input parameters. The dangers of working with parameter-laden algorithms are twofold. First, incorrect settings may cause an algorithm to fail in finding the true patterns. Second, a perhaps more insidious problem is that the algorithm may report spurious patterns that do not really exist, or greatly overestimate the significance of the reported patterns. This is especially likely when the user fails to understand the role of parameters in the data mining process. Data mining algorithms should have as few parameters as possible. A parameter-light algorithm would limit our ability to impose our prejudices, expectations, and presumptions on the problem at hand, and would let the data itself speak to us. In this work, we show that recent results in bioinformatics, learning, and computational theory hold great promise for a parameter-light data-mining paradigm. The results are strongly connected to Kolmogorov complexity theory. However, as a practical matter, they can be implemented using any off-the-shelf compression algorithm with the addition of just a dozen lines of code. We will show that this approach is competitive or superior to many of the state-of-the-art approaches in anomaly/interestingness detection, classification, and clustering with empirical tests on time series/DNA/text/XML/video datasets. As a further evidence of the advantages of our method, we will demonstrate its effectiveness to solve a real world classification problem in recommending printing services and products.","Title: Relevance Feedback Retrieval of Time Series Data\nYear: 1999\n\nAbstract: ABSTRACT There has been much ,recent interest in retrieval of time ,series data. Earlier work has used a fixed similarity metric (e.g., Euclidean distance) to determine ,the similarity between ,a userspecified query and items in the database. Here, we describe a novel approach,to retrieval of time series data by using relevance feedback,from the user to adjust ,the similarity metric. This is important because the Euclidean distance metric does not capture many notions of similarity between time series. In particular, Euclidean distance is sensitive to various ,“distortions” such as offset translation, amplitude scaling, etc. Depending on the domain and the user, one may wish a query to be sensitive or insensitive to these ,distortions to varying ,degrees. This paper addresses this problem ,by introducing ,a profile that encodes the user's subjective notion of similarity in a domain. These profiles can be learned ,continuously from interaction with the user. We further show,how,the user profile may,be embedded ,in a,system that uses relevance feedback ,to modify ,the query in a ,manner analogous to the familiar text retrieval algorithms. Keywords Time series, multimedia data, relevance feedback, modeling user subjectivity.","Title: Iterative Incremental Clustering of Time Series\nYear: 2004\n\nAbstract: We present a novel anytime version of partitional clustering algorithm, such as k-Means and EM, for time series. The algorithm works by lever-aging off the multi-resolution property of wavelets. The dilemma of choosing the initial centers is mitigated by initializing the centers at each approximation level, using the final centers returned by the coarser representations. In addition to casting the clustering algorithms as anytime algorithms, this approach has two other very desirable properties. By working at lower dimensionalities we can efficiently avoid local minima. Therefore, the quality of the clustering is usually better than the batch algorithm. In addition, even if the algorithm is run to completion, our approach is much faster than its batch counterpart. We explain, and empirically demonstrate these surprising and desirable properties with comprehensive experiments on several publicly available real data sets. We further demonstrate that our approach can be generalized to a framework of much broader range of algorithms or data mining problems.","Title: Finding Unusual Medical Time-Series Subsequences: Algorithms and Applications\nYear: 2006\n\nAbstract: In this work, we introduce the new problem of finding time series discords. Time series discords are subsequences of longer time series that are maximally different to all the rest of the time series subsequences. They thus capture the sense of the most unusual subsequence within a time series. While discords have many uses for data mining, they are particularly attractive as anomaly detectors because they only require one intuitive parameter (the length of the subsequence), unlike most anomaly detection algorithms that typically require many parameters. While the brute force algorithm to discover time series discords is quadratic in the length of the time series, we show a simple algorithm that is three to four orders of magnitude faster than brute force, while guaranteed to produce identical results. We evaluate our work with a comprehensive set of experiments on electrocardiograms and other medical datasets","Title: Three Myths about Dynamic Time Warping Data Mining\nYear: 2005\n\nAbstract: The Dynamic Time Warping (DTW) distance measure is a technique that has long been known in speech recognition community. It allows a non-linear mapping of one signal to another by minimizing the distance between the two. A decade ago, DTW was introduced into Data Mining community as a utility for various tasks for time series problems including classification, clustering, and anomaly detection. The technique has flourished, particularly in the last three years, and has been applied to a variety of problems in various disciplines. In spite of DTW's great success, there are still several persistent \"myths\" about it. These myths have caused confusion and led to much wasted research effort. In this work, we will dispel these myths with the most comprehensive set of time series experiments ever conducted.","Title: Establishing the provenance of historical manuscripts with a novel distance measure\nYear: 2015\n\nAbstract: The recent digitization of more than 20 million books has been led by initiatives from countries wishing to preserve their cultural heritage and by several commercial endeavors, including the Google Print Library Project. It is expected that within a few years a significant fraction of the world's books will be online. However, for millions of complete books and tens of millions of loose pages, the provenance of the manuscripts may be completely unknown or disputed, thus denying historians an understanding of the context in which the content was created. In a handful of cases, it may be possible for experts to regain the provenance by examining linguistic, cultural and/or stylistic clues. However, such experts are a rarity and these investigations are time-consuming and expensive. One technique used by experts to establish provenance is the examination of the ornate initial letters appearing in the questioned manuscript. By comparing the initial letters in the manuscript to annotated initial letters whose origin is known, the provenance can be determined. In this work, we show for the first time that we can reproduce this ability with a computer algorithm. We use a recently introduced technique to measure texture similarity and show that it can recognize initial letters with an accuracy that rivals or exceeds human performance. A brute force implementation of this measure would require several months to process a single large book; however, we introduce a novel lower bound that allows us to process the books in hours or minutes.","Title: Dynamic Time Warping Averaging of Time Series Allows Faster and More Accurate Classification\nYear: 2014\n\nAbstract: Recent years have seen significant progress in improving both the efficiency and effectiveness of time series classification. However, because the best solution is typically the Nearest Neighbor algorithm with the relatively expensive Dynamic Time Warping as the distance measure, successful deployments on resource constrained devices remain elusive. Moreover, the recent explosion of interest in wearable devices, which typically have limited computational resources, has created a growing need for very efficient classification algorithms. A commonly used technique to glean the benefits of the Nearest Neighbor algorithm, without inheriting its undesirable time complexity, is to use the Nearest Centroid algorithm. However, because of the unique properties of (most) time series data, the centroid typically does not resemble any of the instances, an unintuitive and underappreciated fact. In this work we show that we can exploit a recent result to allow meaningful averaging of 'warped' times series, and that this result allows us to create ultra-efficient Nearest 'Centroid' classifiers that are at least as accurate as their more lethargic Nearest Neighbor cousins.","Title: Generating Synthetic Data to Allow Learning from a Single Exemplar per Class.\nYear: 2014\n\nAbstract: Recent years have seen an explosion in the volume of historical documents placed online. The individuality of fonts combined with the degradation suffered by century old manuscripts means that Optical Character Recognition Systems do not work well here. As human transcription is prohibitively expensive, recent efforts focused on human/computer cooperative transcription: a human annotates a small fraction of a text to provide labeled data for recognition algorithms. Such a system naturally begs the question of how much data must the human label? In this work we show that we can do well even if the human labels only a single instance from each class. We achieve this good result using two novel observations: we can leverage off a recently introduced parameter-free distance measure, improving it by taking into account the \"complexity\" of the glyphs being compared; we can estimate this complexity using synthetic but plausible instances made from the single training instance. We demonstrate the utility of our observations on diverse historical manuscripts.","Title: Accelerating the dynamic time warping distance measure using logarithmetic arithmetic\nYear: 2014\n\nAbstract: This paper describes an application-specific embedded processor with instruction set extensions (ISEs) for the Dynamic Time Warping (DTW) distance measure, which is widely used in time series similarity search. The ISEs in this paper are implemented using a form of logarithmic arithmetic that offers significant performance and power/energy advantages compared to more traditional floating-point operations.","Title: Time Series Epenthesis: Clustering Time Series Streams Requires Ignoring Some Data\nYear: 2011\n\nAbstract: Given the pervasiveness of time series data in all human endeavors, and the ubiquity of clustering as a data mining application, it is somewhat surprising that the problem of time series clustering from a single stream remains largely unsolved. Most work on time series clustering considers the clustering of individual time series, e.g., gene expression profiles, individual heartbeats or individual gait cycles. The few attempts at clustering time series streams have been shown to be objectively incorrect in some cases, and in other cases shown to work only on the most contrived datasets by carefully adjusting a large set of parameters. In this work, we make two fundamental contributions. First, we show that the problem definition for time series clustering from streams currently used is inherently flawed, and a new definition is necessary. Second, we show that the Minimum Description Length (MDL) framework offers an efficient, effective and essentially parameter-free method for time series clustering. We show that our method produces objectively correct results on a wide variety of datasets from medicine, zoology and industrial process analyses.","Title: Clustering of Symbols Using Minimal Description Length\nYear: 2013\n\nAbstract: The clustering of glyphs (individual letters/characters/symbols) is typically the first step in document processing algorithms and a critical enabling technology for most historical document indexing techniques. In this work, we take a step back from current domain/language specialized research efforts to consider the problem from an agnostic perspective. In particular, we claim that, independent of the distance measure used, any method that attempts to cluster all the data is almost certainly doomed to failure. We explain this observation, and introduce a clustering method based on Minimum Description Length (MDL) that can overcome it.","Title: Classification of streaming time series under more realistic assumptions\nYear: 2016\n\nAbstract: Much of the vast literature on time series classification makes several assumptions about data and the algorithm's eventual deployment that are almost certainly unwarranted. For example, many research efforts assume that the beginning and ending points of the pattern of interest can be correctly identified, during both the training phase and later deployment. Another example is the common assumption that queries will be made at a constant rate that is known ahead of time, thus computational resources can be exactly budgeted. In this work, we argue that these assumptions are unjustified, and this has in many cases led to unwarranted optimism about the performance of the proposed algorithms. As we shall show, the task of correctly extracting individual gait cycles, heartbeats, gestures, behaviors, etc., is generally much more difficult than the task of actually classifying those patterns. Likewise, gesture classification systems deployed on a device such as Google Glass may issue queries at frequencies that range over an order of magnitude, making it difficult to plan computational resources. We propose to mitigate these problems by introducing an alignment-free time series classification framework. The framework requires only very weakly annotated data, such as \\\"in this ten minutes of data, we see mostly normal heartbeats$$\\\\ldots $$¿,\\\" and by generalizing the classic machine learning idea of data editing to streaming/continuous data, allows us to build robust, fast and accurate anytime classifiers. We demonstrate on several diverse real-world problems that beyond removing unwarranted assumptions and requiring essentially no human intervention, our framework is both extremely fast and significantly more accurate than current state-of-the-art approaches.","Title: Accelerating the discovery of unsupervised-shapelets\nYear: 2016\n\nAbstract: Over the past decade, time series clustering has become an increasingly important research topic in data mining community. Most existing methods for time series clustering rely on distances calculated from the raw data using the Euclidean distance or Dynamic Time Warping distance as the distance measure. However, the presence of significant noise, dropouts, or extraneous data can greatly limit the accuracy of clustering in this domain. Moreover, for most real world problems, we cannot expect objects from the same class to be equal in length. As a consequence, most work on time series clustering only considers the clustering of individual time series “behaviors,” e.g., individual heart beats or individual gait cycles, and contrives the time series in some way to make them all equal in length. However, automatically formatting the data in such a way is often a harder problem than the clustering itself. In this work, we show that by using only some local patterns and deliberately ignoring the rest of the data, we can mitigate the above problems and cluster time series of different lengths, e.g., cluster one heartbeat with multiple heartbeats. To achieve this, we exploit and extend a recently introduced concept in time series data mining called . Unlike existing work, our work demonstrates the unintuitive fact that shapelets can be learned from time series. We show, with extensive empirical evaluation in diverse domains, that our method is more accurate than existing methods. Moreover, in addition to accurate clustering results, we show that our work also has the potential to give insight into the domains to which it is applied. While a brute-force algorithm to discover shapelets in an unsupervised way could be untenably slow, we introduce two novel optimization procedures to significantly speed up the unsupervised-shapelet discovery process and allow it to be cast as an anytime algorithm.","Title: A general framework for never-ending learning from time series streams\nYear: 2015\n\nAbstract: Time series classification has been an active area of research in the data mining community for over a decade, and significant progress has been made in the tractability and accuracy of learning. However, virtually all work assumes a one-time training session in which labeled examples of all the concepts to be learned are provided. This assumption may be valid in a handful of situations, but it does not hold in most medical and scientific applications where we initially may have only the vaguest understanding of what concepts can be learned. Based on this observation, we propose a never-ending learning framework for time series in which an agent examines an unbounded stream of data and occasionally asks a teacher (which may be a human or an algorithm) for a label. We demonstrate the utility of our ideas with experiments that consider real-world problems in domains as diverse as medicine, entomology, wildlife monitoring, and human behavior analyses.","Title: Faster and more accurate classification of time series by exploiting a novel dynamic time warping averaging algorithm.\nYear: 2016\n\nAbstract: A concerted research effort over the past two decades has heralded significant improvements in both the efficiency and effectiveness of time series classification. The consensus that has emerged in the community is that the best solution is a surprisingly simple one. In virtually all domains, the most accurate classifier is the nearest neighbor algorithm with dynamic time warping as the distance measure. The time complexity of dynamic time warping means that successful deployments on resource-constrained devices remain elusive. Moreover, the recent explosion of interest in wearable computing devices, which typically have limited computational resources, has greatly increased the need for very efficient classification algorithms. A classic technique to obtain the benefits of the nearest neighbor algorithm, without inheriting its undesirable time and space complexity, is to use the nearest centroid algorithm. Unfortunately, the unique properties of (most) time series data mean that the centroid typically does not resemble any of the instances, an unintuitive and underappreciated fact. In this paper we demonstrate that we can exploit a recent result by Petitjean et al. to allow meaningful averaging of “warped” time series, which then allows us to create super-efficient nearest “centroid” classifiers that are at least as accurate as their more computationally challenged nearest neighbor relatives. We demonstrate empirically the utility of our approach by comparing it to all the appropriate strawmen algorithms on the ubiquitous UCR Benchmarks and with a case study in supporting insect classification on resource-constrained sensors.","Title: Discovery of Meaningful Rules in Time Series\nYear: 2015\n\nAbstract: The ability to make predictions about future events is at the heart of much of science; so, it is not surprising that prediction has been a topic of great interest in the data mining community for the last decade. Most of the previous work has attempted to predict the future based on the current value of a stream. However, for many problems the actual values are irrelevant, whereas the shape of the current time series pattern may foretell the future. The handful of research efforts that consider this variant of the problem have met with limited success. In particular, it is now understood that most of these efforts allow the discovery of spurious rules. We believe the reason why rule discovery in real-valued time series has failed thus far is because most efforts have more or less indiscriminately applied the ideas of symbolic stream rule discovery to real-valued rule discovery. In this work, we show why these ideas are not directly suitable for rule discovery in time series. Beyond our novel definitions/representations, which allow for meaningful and extendable specifications of rules, we further show novel algorithms that allow us to quickly discover high quality rules in very large datasets that accurately predict the occurrence of future events.","Title: Efficient Long-Term Degradation Profiling in Time Series for Complex Physical Systems\nYear: 2015\n\nAbstract: The long term operation of physical systems inevitably leads to their wearing out, and may cause degradations in performance or the unexpected failure of the entire system. To reduce the possibility of such unanticipated failures, the system must be monitored for tell-tale symptoms of degradation that are suggestive of imminent failure. In this work, we introduce a novel time series analysis technique that allows the decomposition of the time series into trend and fluctuation components, providing the monitoring software with actionable information about the changes of the system's behavior over time. We analyze the underlying problem and formulate it to a Quadratic Programming (QP) problem that can be solved with existing QP-solvers. However, when the profiling resolution is high, as generally required by real-world applications, such a decomposition becomes intractable to general QP-solvers. To speed up the problem solving, we further transform the problem and present a novel QP formulation, Non-negative QP, for the problem and demonstrate a tractable solution that bypasses the use of slow general QP-solvers. We demonstrate our ideas on both synthetic and real datasets, showing that our method allows us to accurately extract the degradation phenomenon of time series. We further demonstrate the generality of our ideas by applying them beyond classic machine prognostics to problems in identifying the influence of news events on currency exchange rates and stock prices. We fully implement our profiling system and deploy it into several physical systems, such as chemical plants and nuclear power plants, and it greatly helps detect the degradation phenomenon, and diagnose the corresponding components.","Title: Accelerating Dynamic Time Warping Clustering with a Novel Admissible Pruning Strategy\nYear: 2015\n\nAbstract: Clustering time series is a useful operation in its own right, and an important subroutine in many higher-level data mining analyses, including data editing for classifiers, summarization, and outlier detection. While it has been noted that the general superiority of Dynamic Time Warping (DTW) over Euclidean Distance for similarity search diminishes as we consider ever larger datasets, as we shall show, the same is not true for clustering. Thus, clustering time series under DTW remains a computationally challenging task. In this work, we address this lethargy in two ways. We propose a novel pruning strategy that exploits both upper and lower bounds to prune off a large fraction of the expensive distance calculations. This pruning strategy is admissible; giving us provably identical results to the brute force algorithm, but is at least an order of magnitude faster. For datasets where even this level of speedup is inadequate, we show that we can use a simple heuristic to order the unavoidable calculations in a most-useful-first ordering, thus casting the clustering as an anytime algorithm. We demonstrate the utility of our ideas with both single and multidimensional case studies in the domains of astronomy, speech physiology, medicine and entomology.","Title: Scalable Clustering of Time Series with U-Shapelets.\nYear: 2015\n\nAbstract: A recently introduced primitive for time series data mining, unsupervised shapelets (u-shapelets), has demonstrated significant potential for time series clustering. In contrast to approaches that consider the entire time series to compute pairwise similarities, the u-shapelets technique allows considering only relevant subsequences of time series. Moreover, u-shapelets allow us to bypass the apparent chicken-and-egg paradox of defining relevant with reference to the clustering itself. U-shapelets have several advantages over rival methods. First, they are defined even when the time series are of different lengths; for example, they allow clustering datasets containing a mixture of single heartbeats and multi-beat ECG recordings. Second, u-shapelets mitigate sensitivity to irrelevant data such as noise, spikes, dropouts, etc. Finally, u-shapelets demonstrated ability to provide additional insights into the data. Unfortunately, the state-ofthe-art algorithms for u-shapelets search are intractable and so their advantages have only been demonstrated on tiny datasets. We propose a simple approach to speed up a ushapelet discovery by two orders of magnitude, without any significant loss in clustering quality.","Title: On the Non-Trivial Generalization of Dynamic Time Warping to the Multi-Dimensional Case.\nYear: 2015\n\nAbstract: In the last decade, Dynamic Time Warping (DTW) has emerged as the distance measure of choice for virtually all time series data mining applications. This is the result of significant progress in improving DTW’s efficiency, and multiple empirical studies showing that DTW-based classifiers at least equal the accuracy of all their rivals across dozens of datasets. Thus far, most of the research has considered only the one-dimensional case, with practitioners generalizing to the multi-dimensional case in one of two ways. In general, it appears the community believes either that the two ways are equivalent, or that the choice is irrelevant. In this work, we show that this is not the case. The two most commonly used multidimensional DTW methods can produce different classifications, and neither one dominates over the other. This seems to suggest that one should learn the best method for a particular application. However, we will show that this is not necessary; a simple, principled rule can be used on a case-by-case basis to predict which of the two methods we should give credence to. Our method allows us to ensure that classification results are at least as accurate as the better of the two rival methods, and in many cases, our method is strictly more accurate. We demonstrate our ideas with the most extensive set of multi-dimensional time series classification experiments ever attempted. Keywords—Dynamic Time Warping; Classification","Title: Extracting Optimal Performance from Dynamic Time Warping.\nYear: 2016\n\nAbstract: Dynamic Time Warping (DTW) is a distance measure that compares two time series after optimally aligning them. DTW is being used for decades in thousands of academic and industrial projects despite the very expensive computational complexity, O(n2). These applications include data mining, image processing, signal processing, robotics and computer graphics among many others. In spite of all this research effort, there are many myths and misunderstanding about DTW in the literature, for example \\\"it is too slow to be useful\\\" or \\\"the warping window size does not matter much.\\\" In this tutorial, we correct these misunderstandings and we summarize the research efforts in optimizing both the efficiency and effectiveness of both the basic DTW algorithm, and of the higher-level algorithms that exploit DTW such as similarity search, clustering and classification. We will discuss variants of DTW such as constrained DTW, multidimensional DTW and asynchronous DTW, and optimization techniques such as lower bounding, early abandoning, run-length encoding, bounded approximation and hardware optimization. We will discuss a multitude of application areas including physiological monitoring, social media mining, activity recognition and animal sound processing. The optimization techniques are generalizable to other domains on various data types and problems.","Title: A General Framework for Density Based Time Series Clustering Exploiting a Novel Admissible Pruning Strategy.\nYear: 2016\n\nAbstract: Time Series Clustering is an important subroutine in many higher-level data mining analyses, including data editing for classifiers, summarization, and outlier detection. It is well known that for similarity search the superiority of Dynamic Time Warping (DTW) over Euclidean distance gradually diminishes as we consider ever larger datasets. However, as we shall show, the same is not true for clustering. Clustering time series under DTW remains a computationally expensive operation. In this work, we address this issue in two ways. We propose a novel pruning strategy that exploits both the upper and lower bounds to prune off a very large fraction of the expensive distance calculations. This pruning strategy is admissible and gives us provably identical results to the brute force algorithm, but is at least an order of magnitude faster. For datasets where even this level of speedup is inadequate, we show that we can use a simple heuristic to order the unavoidable calculations in a most-useful-first ordering, thus casting the clustering into an anytime framework. We demonstrate the utility of our ideas with both single and multidimensional case studies in the domains of astronomy, speech physiology, medicine and entomology. In addition, we show the generality of our clustering framework to other domains by efficiently obtaining semantically significant clusters in protein sequences using the Edit Distance, the discrete data analogue of DTW.","Title: Reliable early classification of time series based on discriminating the classes over time.\nYear: 2017\n\nAbstract: The goal of early classification of time series is to predict the class value of a sequence early in time, when its full length is not yet available. This problem arises naturally in many contexts where the data is collected over time and the label predictions have to be made as soon as possible. In this work, a method based on probabilistic classifiers is proposed for the problem of early classification of time series. An important feature of this method is that, in its learning stage, it discovers the timestamps in which the prediction accuracy for each class begins to surpass a pre-defined threshold. This threshold is defined as a percentage of the accuracy that would be obtained if the full series were available, and it is defined by the user. The class predictions for new time series will only be made in these timestamps or later. Furthermore, when applying the model to a new time series, a class label will only be provided if the difference between the two largest predicted class probabilities is higher than or equal to a certain threshold, which is calculated in the training step. The proposal is validated on 45 benchmark time series databases and compared with several state-of-the-art methods, and obtains superior results in both earliness and accuracy. In addition, we show the practical applicability of our method for a real-world problem: the detection and identification of bird calls in a biodiversity survey scenario.","Title: Generalizing DTW to the multi-dimensional case requires an adaptive approach.\nYear: 2017\n\nAbstract: In recent years Dynamic Time Warping (DTW) has emerged as the distance measure of choice for virtually all time series data mining applications. For example, virtually all applications that process data from wearable devices use DTW as a core sub-routine. This is the result of significant progress in improving DTW's efficiency, together with multiple empirical studies showing that DTW-based classifiers at least equal (and generally surpass) the accuracy of all their rivals across dozens of datasets. Thus far, most of the research has considered only the one-dimensional case, with practitioners generalizing to the multi-dimensional case in one of two ways, dependent or independent warping. In general, it appears the community believes either that the two ways are equivalent, or that the choice is irrelevant. In this work, we show that this is not the case. The two most commonly used multi-dimensional DTW methods can produce different classifications, and neither one dominates over the other. This seems to suggest that one should learn the best method for a particular application. However, we will show that this is not necessary; a simple, principled rule can be used on a case-by-case basis to predict which of the two methods we should trust at the time of classification. Our method allows us to ensure that classification results are at least as accurate as the better of the two rival methods, and, in many cases, our method is significantly more accurate. We demonstrate our ideas with the most extensive set of multi-dimensional time series classification experiments ever attempted.","Title: Matrix Profile II: Exploiting a Novel Algorithm and GPUs to Break the One Hundred Million Barrier for Time Series Motifs and Joins\nYear: 2016\n\nAbstract: Time series motifs have been in the literature for about fifteen years, but have only recently begun to receive significant attention in the research community. This is perhaps due to the growing realization that they implicitly offer solutions to a host of time series problems, including rule discovery, anomaly detection, density estimation, semantic segmentation, etc. Recent work has improved the scalability to the point where exact motifs can be computed on datasets with up to a million data points in tenable time. However, in some domains, for example seismology, there is an insatiable need to address even larger datasets. In this work we show that a combination of a novel algorithm and a high-performance GPU allows us to significantly improve the scalability of motif discovery. We demonstrate the scalability of our ideas by finding the full set of exact motifs on a dataset with one hundred million subsequences, by far the largest dataset ever mined for time series motifs. Furthermore, we demonstrate that our algorithm can produce actionable insights in seismology and other domains.","Title: Matrix Profile I: All Pairs Similarity Joins For Time Series: A Unifying View That Includes Motifs, Discords And Shapelets\nYear: 2016\n\nAbstract: The all-pairs-similarity-search (or similarity join) problem has been extensively studied for text and a handful of other datatypes. However, surprisingly little progress has been made on similarity joins for time series subsequences. The lack of progress probably stems from the daunting nature of the problem. For even modest sized datasets the obvious nested-loop algorithm can take months, and the typical speed-up techniques in this domain (i.e., indexing, lower-bounding, triangular-inequality pruning and early abandoning) at best produce one or two orders of magnitude speedup. In this work we introduce a novel scalable algorithm for time series subsequence all-pairs-similarity-search. For exceptionally large datasets, the algorithm can be trivially cast as an anytime algorithm and produce high-quality approximate solutions in reasonable time. The exact similarity join algorithm computes the answer to the time series motif and time series discord problem as a side-effect, and our algorithm incidentally provides the fastest known algorithm for both these extensively-studied problems. We demonstrate the utility of our ideas for two time series data mining problems, including motif discovery and novelty discovery.","Title: Matrix Profile III: The Matrix Profile Allows Visualization of Salient Subsequences in Massive Time Series\nYear: 2016\n\nAbstract: Multidimensional Scaling (MDS) is one of the most versatile tools used for exploratory data mining. It allows a first glimpse of possible structure in the data, which can inform the choice of analyses used. Its uses are multiple. It can give the user an idea as to the cluster ability or linear separability of the data. It can help spot outliers, or can hint at the intrinsic dimensionality of the data. Moreover, it can sometimes reveal unexpected latent dimensions in the data. With all these uses, MDS is increasingly used in areas as diverse as marketing, medicine, genetics, music and linguistics. One of the strengths of MDS is that it is essentially agnostic to data type, as we can use any distance measure to create the distance matrix, which is the only required input to the MDS algorithm. In spite of this generality, we make the following claim. MDS is not (well) defined for an increasingly important data type, time series subsequences. In this work we explain why this is the case, and we propose a scalable solution. We demonstrate the utility of our ideas on several diverse real-world datasets. At the core of our approach is a novel Minimum Description Length (MDL) subsequence extraction algorithm. Beyond MDS visualization, this subsequence extraction subroutine may be a useful tool in its own right.","Title: The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances.\nYear: 2017\n\nAbstract: In the last 5 years there have been a large number of new time series classification algorithms proposed in the literature. These algorithms have been evaluated on subsets of the 47 data sets in the University of California, Riverside time series classification archive. The archive has recently been expanded to 85 data sets, over half of which have been donated by researchers at the University of East Anglia. Aspects of previous evaluations have made comparisons between algorithms difficult. For example, several different programming languages have been used, experiments involved a single train/test split and some used normalised data whilst others did not. The relaunch of the archive provides a timely opportunity to thoroughly evaluate algorithms on a larger number of datasets. We have implemented 18 recently proposed algorithms in a common Java framework and compared them against two standard benchmark classifiers (and each other) by performing 100 resampling experiments on each of the 85 datasets. We use these results to test several hypotheses relating to whether the algorithms are significantly more accurate than the benchmarks and each other. Our results indicate that only nine of these algorithms are significantly more accurate than both benchmarks and that one classifier, the collective of transformation ensembles, is significantly more accurate than all of the others. All of our experiments and results are reproducible: we release all of our code, results and experimental details and we hope these experiments form the basis for more robust testing of new algorithms in the future.","Title: Searching Time Series with Invariance to Large Amounts of Uniform Scaling\nYear: 2017\n\nAbstract: Similarity search is arguably the most important primitive in time series data mining. Recent research has made significant progress on fast algorithms for time series similarity search under Dynamic Time Warping (DTW) and Uniform Scaling (US) distance measures. However, the current state-of-the-art algorithms cannot support greater amounts of rescaling in many practical applications. In this paper, we introduce a novel lower bound, LB\n<sub xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">new</sub>\n, to allow efficient search even in domains that exhibit more than a factor-of-two variability in scale. The effectiveness of our idea is validated on various large-scale real datasets from commercial important domains.","Title: Query Suggestion to allow Intuitive Interactive Search in Multidimensional Time Series.\nYear: 2017\n\nAbstract: In recent years, the research community, inspired by its success in dealing with single-dimensional time series, has turned its attention to dealing with multidimensional time series. There are now a plethora of techniques for indexing, classification, and clustering of multidimensional time series. However, we argue that the difficulty of exploratory search in large multidimensional time series remains underappreciated. In essence, the problem reduces to the \\\"chicken-and-egg\\\" paradox that it is difficult to produce a meaningful query without knowing the best subset of dimensions to use, but finding the best subset of dimensions is itself query dependent. In this work we propose a solution to this problem. We introduce an algorithm that runs in the background, observing the user's search interactions. When appropriate, our algorithm suggests to the user a dimension that could be added or deleted to improve the user's satisfaction with the query. These query dependent suggestions may be useful to the user, even if she does not act on them (by reissuing the query), as they can hint at unexpected relationships or redundancies between the dimensions of the data. We evaluate our algorithm on several real-world datasets in medical, human activity, and industrial domains, showing that it produces subjectively sensible and objectively superior results.","Title: Matrix profile IV: using weakly labeled time series to predict outcomes\nYear: 2017\n\nAbstract: AbstractIn academic settings over the last decade, there has been significant progress in time series classification. However, much of this work makes assumptions that are simply unrealistic for deployed industrial applications. Examples of these unrealistic assumptions include the following: assuming that data subsequences have a single fixed-length, are precisely extracted from the data, and are correctly labeled according to their membership in a set of equal-size classes. In real-world industrial settings, these patterns can be of different lengths, the class annotations may only belong to a general region of the data, may contain errors, and finally, the class distribution is typically highly skewed. Can we learn from such weakly labeled data? In this work, we introduce SDTS, a scalable algorithm that can learn in such challenging settings. We demonstrate the utility of our ideas by learning from diverse datasets with millions of datapoints. As we shall demonstrate, our domain-agnostic parameter-free algorithm can be competitive with domain-specific algorithms used in neuroscience and entomology, even when those algorithms have been tuned by domain experts to incorporate domain knowledge.","Title: Matrix Profile V: A Generic Technique to Incorporate Domain Knowledge into Motif Discovery.\nYear: 2017\n\nAbstract: Time series motif discovery has emerged as perhaps the most used primitive for time series data mining, and has seen applications to domains as diverse as robotics, medicine and climatology. There has been recent significant progress on the scalability of motif discovery. However, we believe that the current definitions of motif discovery are limited, and can create a mismatch between the user's intent/expectations, and the motif discovery search outcomes. In this work, we explain the reasons behind these issues, and introduce a novel and general framework to address them. Our ideas can be used with current state-of-the-art algorithms with virtually no time or space overhead, and are fast enough to allow real-time interaction and hypotheses testing on massive datasets. We demonstrate the utility of our ideas on domains as diverse as seismology and epileptic seizure monitoring.","Title: Matrix Profile VII: Time Series Chains: A New Primitive for Time Series Data Mining (Best Student Paper Award).\nYear: 2017\n\nAbstract: Since their introduction over a decade ago, time series motifs have become a fundamental tool for time series analytics, finding diverse uses in dozens of domains. In this work we introduce Time Series Chains, which are related to, but distinct from, time series motifs. Informally, time series chains are a temporally ordered set of subsequence patterns, such that each pattern is similar to the pattern that preceded it, but the first and last patterns are arbitrarily dissimilar. In the discrete space, this is similar to extracting the text chain hit, hot, dot, dog from a paragraph. The first and last words have nothing in common, yet they are connected by a chain of words with a small mutual difference. Time series chains can capture the evolution of systems, and help predict the future. As such, they potentially have implications for prognostics. In this work, we introduce a robust definition of time series chains, and a scalable algorithm that allows us to discover them in massive datasets.","Title: Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels\nYear: 2017\n\nAbstract: Unsupervised semantic segmentation in the time series domain is a much-studied problem due to its potential to detect unexpected regularities and regimes in poorly understood data. However, the current techniques have several shortcomings, which have limited the adoption of time series semantic segmentation beyond academic settings for three primary reasons. First, most methods require setting/learning many parameters and thus may have problems generalizing to novel situations. Second, most methods implicitly assume that all the data is segmentable, and have difficulty when that assumption is unwarranted. Finally, most research efforts have been confined to the batch case, but online segmentation is clearly more useful and actionable. To address these issues, we present an algorithm which is domain agnostic, has only one easily determined parameter, and can handle data streaming at a high rate. In this context, we test our algorithm on the largest and most diverse collection of time series datasets ever considered, and demonstrate our algorithm's superiority over current solutions. Furthermore, we are the first to show that semantic segmentation may be possible at superhuman performance levels.","Title: Judicious Setting Of Dynamic Time Warping'S Window Width Allows More Accurate Classification Of Time Series\nYear: 2017\n\nAbstract: While the Dynamic Time Warping (DTW) -based Nearest-Neighbor Classification algorithm is regarded as a strong baseline for time series classification, in recent years there has been a plethora of algorithms that have claimed to be able to improve upon its accuracy in the general case. Many of these proposed ideas sacrifice the simplicity of implementation that DTW-based classifiers offer for rather modest gains. Nevertheless, there are clearly times when even a small improvement could make a large difference in an important medical or financial domain. In this work, we make an unexpected claim; an underappreciated \"low hanging fruit\" in optimizing DTW's performance can produce improvements that make it an even stronger baseline, closing most or all the improvement gap of the more sophisticated methods. We show that the method currently used to learn DTW's only parameter, the maximum amount of warping allowed, is likely to give the wrong answer for small training sets. We introduce a simple method to mitigate the small training set issue by creating synthetic exemplars to help learn the parameter. We evaluate our ideas on the UCR Time Series Archive and a case study in fall classification, and demonstrate that our algorithm produces significant improvement in classification accuracy.","Title: Exploiting a novel algorithm and GPUs to break the ten quadrillion pairwise comparisons barrier for time series motifs and joins.\nYear: 2018\n\nAbstract: Time series motifs are approximately repeated subsequences found within a longer time series. They have been in the literature since 2002, but recently they have begun to receive significant attention in research and industrial communities. This is perhaps due to the growing realization that they implicitly offer solutions to a host of time series problems, including rule discovery, anomaly detection, density estimation, semantic segmentation, summarization, etc. Recent work has improved the scalability so exact motifs can be computed on datasets with up to a million data points in tenable time. However, in some domains, for example seismology or climatology, there is an immediate need to address even larger datasets. In this work, we demonstrate that a combination of a novel algorithm and a high-performance GPU allows us to significantly improve the scalability of motif discovery. We demonstrate the scalability of our ideas by finding the full set of exact motifs on a dataset with one hundred and forty-three million subsequences, which is by far the largest dataset ever mined for time series motifs/joins; it requires ten quadrillion pairwise comparisons. Furthermore, we demonstrate that our algorithm can produce actionable insights into seismology and ethology.","Title: Time series joins, motifs, discords and shapelets: a unifying view that exploits the matrix profile.\nYear: 2018\n\nAbstract: The last decade has seen a flurry of research on (or ) for text, DNA and a handful of other datatypes, and these systems have been applied to many diverse data mining problems. However, there has been surprisingly little progress made on similarity joins for . The lack of progress probably stems from the daunting nature of the problem. For even modest sized datasets the obvious nested-loop algorithm can take months, and the typical speed-up techniques in this domain (i.e., indexing, lower-bounding, triangular-inequality pruning and early abandoning) at best produce only one or two orders of magnitude speedup. In this work we introduce a novel scalable algorithm for time series subsequence all-pairs-similarity-search. For exceptionally large datasets, the algorithm can be trivially cast as an anytime algorithm and produce high-quality approximate solutions in reasonable time and/or be accelerated by a trivial porting to a GPU framework. The exact similarity join algorithm computes the answer to the and problem as a side-effect, and our algorithm incidentally provides the fastest known algorithm for both these extensively-studied problems. We demonstrate the utility of our ideas for many time series data mining problems, including motif discovery, novelty discovery, shapelet discovery, semantic segmentation, density estimation, and contrast set mining. Moreover, we demonstrate the utility of our ideas on domains as diverse as seismology, music processing, bioinformatics, human activity monitoring, electrical power-demand monitoring and medicine.","Title: Admissible Time Series Motif Discovery with Missing Data.\nYear: 2018\n\nAbstract: The discovery of time series motifs has emerged as one of the most useful primitives in time series data mining. Researchers have shown its utility for exploratory data mining, summarization, visualization, segmentation, classification, clustering, and rule discovery. Although there has been more than a decade of extensive research, there is still no technique to allow the discovery of time series motifs in the presence of missing data, despite the well-documented ubiquity of missing data in scientific, industrial, and medical datasets. In this work, we introduce a technique for motif discovery in the presence of missing data. We formally prove that our method is admissible, producing no false negatives. We also show that our method can piggy-back off the fastest known motif discovery method with a small constant factor time/space overhead. We will demonstrate our approach on diverse datasets with varying amounts of missing data","Title: VALMOD: A Suite for Easy and Exact Detection of Variable Length Motifs in Data Series.\nYear: 2018\n\nAbstract: Data series motif discovery represents one of the most useful primitives for data series mining, with applications to many domains, such as robotics, entomology, seismology, medicine, and climatology, and others. The state-of-the-art motif discovery tools still require the user to provide the motif length. Yet, in several cases, the choice of motif length is critical for their detection. Unfortunately, the obvious brute-force solution, which tests all lengths within a given range, is computationally untenable, and does not provide any support for ranking motifs at different resolutions (i.e., lengths). We demonstrate VALMOD, our scalable motif discovery algorithm that efficiently finds all motifs in a given range of lengths, and outputs a length-invariant ranking of motifs. Furthermore, we support the analysis process by means of a newly proposed meta-data structure that helps the user to select the most promising pattern length. This demo aims at illustrating in detail the steps of the proposed approach, showcasing how our algorithm and corresponding graphical insights enable users to efficiently identify the correct motifs.","Title: Matrix Profile X: VALMOD - Scalable Discovery of Variable-Length Motifs in Data Series.\nYear: 2018\n\nAbstract: In the last fifteen years, data series motif discovery has emerged as one of the most useful primitives for data series mining, with applications to many domains, including robotics, entomology, seismology, medicine, and climatology. Nevertheless, the state-of-the-art motif discovery tools still require the user to provide the motif length. Yet, in at least some cases, the choice of motif length is critical and unforgiving. Unfortunately, the obvious brute-force solution, which tests all lengths within a given range, is computationally untenable. In this work, we introduce VALMOD, an exact and scalable motif discovery algorithm that efficiently finds all motifs in a given range of lengths. We evaluate our approach with five diverse real datasets, and demonstrate that it is up to 20 times faster than the state-of-the-art. Our results also show that removing the unrealistic assumption that the user knows the correct length, can often produce more intuitive and actionable results, which could have been missed otherwise.","Title: Optimizing dynamic time warping's window width for time series data mining applications.\nYear: 2018\n\nAbstract: Dynamic Time Warping (DTW) is a highly competitive distance measure for most time series data mining problems. Obtaining the best performance from DTW requires setting its only parameter, the maximum amount of warping (). In the supervised case with ample data, is typically set by cross-validation in the training stage. However, this method is likely to yield suboptimal results for small training sets. For the unsupervised case, learning via cross-validation is not possible because we do not have access to labeled data. Many practitioners have thus resorted to assuming that “”, and they use the largest value of permitted by the computational resources. However, as we will show, in most circumstances, this is a naïve approach that produces inferior clusterings. Moreover, the best warping window width is generally non-transferable between the two tasks, i.e., for a single dataset, practitioners cannot simply apply the best learned for classification on clustering or vice versa. In addition, we will demonstrate that the appropriate amount of warping not only depends on the data structure, but also on the dataset . Thus, even if a practitioner knows the best setting for a given dataset, they will likely be at a lost if they apply that setting on a bigger size version of that data. All these issues seem largely unknown or at least unappreciated in the community. In this work, we demonstrate the importance of setting DTW’s warping window width correctly, and we also propose novel methods to learn this parameter in both supervised and unsupervised settings. The algorithms we propose to learn can produce significant improvements in classification accuracy and clustering quality. We demonstrate the correctness of our novel observations and the utility of our ideas by testing them with more than one hundred publicly available datasets. Our forceful results allow us to make a perhaps unexpected claim; an underappreciated “low hanging fruit” in optimizing DTW’s performance can produce improvements that make it an even stronger baseline, closing most or all the improvement gap of the more sophisticated methods proposed in recent years.","Title: The UCR Time Series Archive.\nYear: 2018\n\nAbstract: The UCR Time Series Archive - introduced in 2002, has become an important resource in the time series data mining community, with at least one thousand published papers making use of at least one dataset from the archive. The original incarnation of the archive had sixteen datasets but since that time, it has gone through periodic expansions. The last expansion took place in the summer of 2015 when the archive grew from 45 datasets to 85 datasets. This paper introduces and will focus on the new data expansion from 85 to 128 datasets. Beyond expanding this valuable resource, this paper offers pragmatic advice to anyone who may wish to evaluate a new algorithm on the archive. Finally, this paper makes a novel and yet actionable claim: of the hundreds of papers that show an improvement over the standard baseline (1-Nearest Neighbor classification), a large fraction may be misattributing the reasons for their improvement. Moreover, they may have been able to achieve the same improvement with a much simpler modification, requiring just a single line of code.","Title: Representation Learning by Reconstructing Neighborhoods.\nYear: 2018\n\nAbstract: Since its introduction, unsupervised representation learning has attracted a lot of attention from the research community, as it is demonstrated to be highly effective and easy-to-apply in tasks such as dimension reduction, clustering, visualization, information retrieval, and semi-supervised learning. In this work, we propose a novel unsupervised representation learning framework called neighbor-encoder, in which domain knowledge can be easily incorporated into the learning process without modifying the general encoder-decoder architecture of the classic autoencoder.In contrast to autoencoder, which reconstructs the input data itself, neighbor-encoder reconstructs the input datau0027s neighbors. As the proposed representation learning problem is essentially a neighbor reconstruction problem, domain knowledge can be easily incorporated in the form of an appropriate definition of similarity between objects. Based on that observation, our framework can leverage any off-the-shelf similarity search algorithms or side information to find the neighbor of an input object. Applications of other algorithms (e.g., association rule mining) in our framework are also possible, given that the appropriate definition of neighbor can vary in different contexts. We have demonstrated the effectiveness of our framework in many diverse domains, including images, text, and time series, and for various data mining tasks including classification, clustering, and visualization. Experimental results show that neighbor-encoder not only outperforms autoencoder in most of the scenarios we consider, but also achieves the state-of-the-art performance on text document clustering.","Title: The UEA multivariate time series classification archive, 2018.\nYear: 2018\n\nAbstract: In 2002, the UCR time series classification archive was first released with sixteen datasets. It gradually expanded, until 2015 when it increased in size from 45 datasets to 85 datasets. In October 2018 more datasets were added, bringing the total to 128. The new archive contains a wide range of problems, including variable length series, but it still only contains univariate time series classification problems. One of the motivations for introducing the archive was to encourage researchers to perform a more rigorous evaluation of newly proposed time series classification (TSC) algorithms. It has worked: most recent research into TSC uses all 85 datasets to evaluate algorithmic advances. Research into multivariate time series classification, where more than one series are associated with each class label, is in a position where univariate TSC research was a decade ago. Algorithms are evaluated using very few datasets and claims of improvement are not based on statistical comparisons. We aim to address this problem by forming the first iteration of the MTSC archive, to be hosted at the website www.timeseriesclassification.com. Like the univariate archive, this formulation was a collaborative effort between researchers at the University of East Anglia (UEA) and the University of California, Riverside (UCR). The 2018 vintage consists of 30 datasets with a wide range of cases, dimensions and series lengths. For this first iteration of the archive we format all data to be of equal length, include no series with missing data and provide train/test splits.","Title: Time Series Classification To Improve Poultry Welfare\nYear: 2018\n\nAbstract: Poultry farms are an important contributor to the human food chain. Worldwide, humankind keeps an enormous number of domesticated birds (e.g. chickens) for their eggs and their meat, providing rich sources of low-fat protein. However, around the world, there have been growing concerns about the quality of life for the livestock in poultry farms; and increasingly vocal demands for improved standards of animal welfare. Recent advances in sensing technologies and machine learning allow the possibility of automatically assessing the health of some individual birds, and employing the lessons learned to improve the welfare for all birds. This task superficially appears to be easy, given the dramatic progress in recent years in classifying human behaviors, and given that human behaviors are presumably more complex. However, as we shall demonstrate, classifying chicken behaviors poses several unique challenges, chief among which is creating a generalizable \"dictionary\" of behaviors from sparse and noisy data. In this work we introduce a novel time series dictionary learning algorithm that can robustly learn from weakly labeled data sources.","Title: Matrix Profile XIII: Time Series Snippets: A New Primitive for Time Series Data Mining\nYear: 2018\n\nAbstract: Perhaps the most basic query made by a data analyst confronting a new data source is \"Show me some representative/typical data.\" Answering this question is trivial in many domains, but surprisingly, it is very difficult in large time series datasets. The major difficulty is not time or space complexity, but defining what it means to be representative data in this domain. In this work, we show that the obvious candidate definitions: motifs, shapelets, cluster centers, random samples etc., are all poor choices. Thus motivated, we introduce time series snippets, a novel representation of typical time series subsequences. Beyond their utility for visualizing and summarizing massive time series collections, we show that time series snippets have utility for high-level comparison of large time series collections.","Title: Prefix And Suffix Invariant Dynamic Time Warping\nYear: 2016\n\nAbstract: While there exist a plethora of classification algorithms for most data types, there is an increasing acceptance that the unique properties of time series mean that the combination of nearest neighbor classifiers and Dynamic Time Warping (DTW) is very competitive across a host of domains, from medicine to astronomy to environmental sensors. While there has been significant progress in improving the efficiency and effectiveness of DTW in recent years, in this work we demonstrate that an underappreciated issue can significantly degrade the accuracy of DTW in real-world deployments. This issue has probably escaped the attention of the very active time series research community because of its reliance on static highly contrived benchmark datasets, rather than real world dynamic datasets where the problem tends to manifest itself. In essence, the issue is that DTW's eponymous invariance to warping is only true for the main \"body\" of the two time series being compared. However, for the \"head\" and \"tail\" of the time series, the DTW algorithm affords no warping invariance. The effect of this is that tiny differences at the beginning or end of the time series (which may be either consequential or simply the result of poor \"cropping\") will tend to contribute disproportionally to the estimated similarity, producing incorrect classifications. In this work, we show that this effect is real, and reduces the performance of the algorithm. We further show that we can fix the issue with a subtle redesign of the DTW algorithm, and that we can learn an appropriate setting for the extra parameter we introduced. We further demonstrate that our generalization is amiable to all the optimizations that make DTW tractable for large datasets.","Title: Speeding up similarity search under dynamic time warping by pruning unpromising alignments.\nYear: 2018\n\nAbstract: Similarity search is the core procedure for several time series mining tasks. While different distance measures can be used for this purpose, there is clear evidence that the Dynamic Time Warping (DTW) is the most suitable distance function for a wide range of application domains. Despite its quadratic complexity, research efforts have proposed a significant number of pruning methods to speed up the similarity search under DTW. However, the search may still take a considerable amount of time depending on the parameters of the search, such as the length of the query and the warping window width. The main reason is that the current techniques for speeding up the similarity search focus on avoiding the costly distance calculation between as many pairs of time series as possible. Nevertheless, the few pairs of subsequences that were not discarded by the pruning techniques can represent a significant part of the entire search time. In this work, we adapt a recently proposed algorithm to improve the internal efficiency of the DTW calculation. Our method can speed up the UCR suite, considered the current fastest tool for similarity search under DTW. More important, the longer the time needed for the search, the higher the speedup ratio achieved by our method. We demonstrate that our method performs similarly to UCR suite for small queries and narrow warping constraints. However, it performs up to five times faster for long queries and large warping windows.","Title: Generalized Dynamic Time Warping: Unleashing the Warping Power Hidden in Point-Wise Distances\nYear: 2018\n\nAbstract: Domain-specific distances preferred by analysts for exploring similarities among time series tend to be \"point-to-point\" distances. Unfortunately, this point-wise nature limits their ability to perform meaningful comparisons between sequences of different lengths and with temporal mis-alignments. Analysts instead need \"elastic\" alignment tools such as Dynamic Time Warping (DTW) to perform such flexible comparisons. However, the existing alignment tools are limited in that they do not incorporate diverse distances. To address this shortcoming, our work introduces the first conceptual framework called Generalized Dynamic Time Warping (GDTW) that supports now alignment (warping) of a large array of domain-specific distances in a uniform manner. While the classic DTW and its prior extensions focus on the Euclidean Distance, our GDTW is the first method that generalizes the ubiquitous DTW and \"extends\" its warping capabilities to a rich diversity of point-to-point distances. Based on our GDTW paradigm that preserves the efficiency of the dynamic programming paradigm of DTW, we design an abstraction that implemented by our GDTW Design Tool enables analysts to \"warp\" new distances with little programming effort. Through extensive evaluation studies on 85 real public domain benchmark datasets, we show that our newly warped distances offer higher classification accuracy than the previously available distances for the majority of these datasets. Further, our case study on heart arrhythmia data illustrates the utility of the new distances enabled by our GDTW warping methodology.","Title: Time Series Classification: Lessons Learned In The (Literal) Field While Studying Chicken Behavior\nYear: 2019\n\nAbstract: Poultry farms are a major contributor to the human food chain. However, around the world, there have been growing concerns about the quality of life for the livestock in poultry farms; and increasingly vocal demands for improved standards of animal welfare. Recent advances in sensing technologies and machine learning allow the possibility of monitoring birds, and employing the lessons learned to improve the welfare for all birds. This task superficially appears to be easy, yet, studying behavioral patterns involves collecting enormous amounts of data, justifying the term Big Data. Before the big data can be used for analytical purposes to tease out meaningful, well-conserved behavioral patterns, the collected data needs to be preprocessed. The pre-processing refers to processes for cleansing and preparing data so that it is in the format ready to be analyzed by downstream algorithms, such as classification and clustering algorithms. However, as we shall demonstrate, efficient preprocessing of chicken big data is both non-trivial and crucial towards success of further analytics.","Title: Fastdtw Is Approximate And Generally Slower Than The Algorithm It Approximates (Extended Abstract)\nYear: 2021\n\nAbstract: Many time series data mining problems can be solved with repeated use of distance measure. Examples of such tasks include similarity search, clustering, classification, anomaly detection and segmentation. For over two decades it has been known that the Dynamic Time Warping (DTW) distance measure is the best measure to use for most tasks, in most domains. Because the classic DTW algorithm has quadratic time complexity, many ideas have been introduced to reduce its amortized time, or to quickly approximate it. One of the most cited approximate approaches is FastDTW. The FastDTW algorithm has well over a thousand citations and has been explicitly used in several hundred research efforts. In this work, we make a surprising claim. In any realistic data mining application, the approximate FastDTW is much slower than the exact DTW. This fact clearly has implications for the community that uses this algorithm: allowing it to address much larger datasets, get exact results, and do so in less time.","Title: Shape-Based Telemetry Approach for Distracted Driving Behavior Detection\nYear: 2021\n\nAbstract: Nearly nine in ten U.S. drivers exhibit distracted driving behavior, and such abnormal driving behavior is the leading cause of many fatal crashes. The prior art in distracted driving behavior detection focused on an ego vehicle sensing its driver. However, such approaches do not address the situation when other nearby vehicles behave abnormally (e.g., a distracted driver follows the ego vehicle). Modern cars have in-vehicle sensors to observe the lane-level driving behavior of others (e.g., preceding/following vehicles). We hypothesize that time series analysis of such data could reveal the nearby distracted drivers. In this work, we tackle this issue and analyze the lane-level driving behavior with Contrast Profile, a recently developed time-series tool, to detect distracted drivers. The Contrast Profile can examine archived driving data and infer the shape of the most discriminative driving behavior while and before being distracted. We then use such shape-based knowledge to detect distracted driving behavior in real time. Our preliminary results suggest that the proposed shape-based approach can identify the distracted driving behavior with about 90% accuracy under different settings.","Title: Matrix Profile XXIII: Contrast Profile: A Novel Time Series Primitive that Allows Real World Classification\nYear: 2021\n\nAbstract: Time series data remains a perennially important datatype considered in data mining. In the last decade there has been an increasing realization that time series data can best understood by reasoning about time series subsequences on the basis of their similarity to other subsequences: the two most familiar such time series concepts being motifs and discords. Time series motifs refer to two particularly close subsequences, whereas time series discords indicate subsequences that are far from their nearest neighbors. However, we argue that it can sometimes be useful to simultaneously reason about a subsequence's closeness to certain data and its distance to other data. In this work we introduce a novel primitive called the Contrast Profile that allows us to efficiently compute such a definition in a principled way. As we will show, the Contrast Profile has many downstream uses, including anomaly detection, data exploration, and preprocessing unstructured data for classification."],"string":"[\n \"Title: Beyond one billion time series: indexing and mining very large time series collections with i SAX2+.\\nYear: 2014\\n\\nAbstract: There is an increasingly pressing need, by several applications in diverse domains, for developing techniques able to index and mine very large collections of time series. Examples of such applications come from astronomy, biology, the web, and other domains. It is not unusual for these applications to involve numbers of time series in the order of hundreds of millions to billions. However, all relevant techniques that have been proposed in the literature so far have not considered any data collections much larger than one-million time series. In this paper, we describe \\\\(i\\\\)SAX 2.0 and its improvements, \\\\(i\\\\)SAX 2.0 Clustered and \\\\(i\\\\)SAX2+, three methods designed for indexing and mining truly massive collections of time series. We show that the main bottleneck in mining such massive datasets is the time taken to build the index, and we thus introduce a novel bulk loading mechanism, the first of this kind specifically tailored to a time series index. We show how our methods allows mining on datasets that would otherwise be completely untenable, including the first published experiments to index one billion time series, and experiments in mining massive data from domains as diverse as entomology, DNA and web-scale image collections.\",\n \"Title: Iterative Deepening Dynamic Time Warping for Time Series\\nYear: 2002\\n\\nAbstract: this paper is organized as follows. Section 2 provides more detailedmotivation for IDDTW and a review of the classic DTW algorithm. Section 3 introducesthe IDDTW algorithm. In section 4 we experimentally compare IDDTW to DTW andEuclidean distance on real and synthetic datasets. Section 5 contains a discussion ofrelated work. Finally section 6 contains our conclusions and directions for future work\",\n \"Title: CID: an efficient complexity-invariant distance for time series\\nYear: 2014\\n\\nAbstract: The ubiquity of time series data across almost all human endeavors has produced a great interest in time series data mining in the last decade. While dozens of classification algorithms have been applied to time series, recent empirical evidence strongly suggests that simple nearest neighbor classification is exceptionally difficult to beat. The choice of distance measure used by the nearest neighbor algorithm is important, and depends on the invariances required by the domain. For example, motion capture data typically requires invariance to warping, and cardiology data requires invariance to the baseline (the mean value). Similarly, recent work suggests that for time series clustering, the choice of clustering algorithm is much less important than the choice of distance measure used.In this work we make a somewhat surprising claim. There is an invariance that the community seems to have missed, complexity invariance. Intuitively, the problem is that in many domains the different classes may have different complexities, and pairs of complex objects, even those which subjectively may seem very similar to the human eye, tend to be further apart under current distance measures than pairs of simple objects. This fact introduces errors in nearest neighbor classification, where some complex objects may be incorrectly assigned to a simpler class. Similarly, for clustering this effect can introduce errors by \\\"suggesting\\\" to the clustering algorithm that subjectively similar, but complex objects belong in a sparser and larger diameter cluster than is truly warranted.We introduce the first complexity-invariant distance measure for time series, and show that it generally produces significant improvements in classification and clustering accuracy. We further show that this improvement does not compromise efficiency, since we can lower bound the measure and use a modification of triangular inequality, thus making use of most existing indexing and data mining algorithms. We evaluate our ideas with the largest and most comprehensive set of time series mining experiments ever attempted in a single work, and show that complexity-invariant distance measures can produce improvements in classification and clustering in the vast majority of cases.\",\n \"Title: Dimensionality Reduction for Fast Similarity Search in Large Time Series Databases.\\nYear: 2001\\n\\nAbstract: The problem of similarity search in large time series databases has attracted muchattention recently. It is a non-trivial problem because of the inherent high dimensionality of thedata. The most promising solutions involve first performing dimensionality reduction on the data,and then indexing the reduced data with a spatial access method. Three major dimensionalityreduction techniques have been proposed, Singular Value Decomposition (SVD), the DiscreteFourier transform (DFT), and more...\",\n \"Title: Derivative Dynamic Time Warping\\nYear: 2001\\n\\nAbstract: this paper we address both these problems by introducing a modification of DTW.The crucial difference is in the features we consider when attempting to find the correctwarping. Rather than use the raw data, we consider only the (estimated) local derivativesof the data\",\n \"Title: Efficiently Finding Arbitrarily Scaled Patterns in Massive Time Series Databases\\nYear: 2003\\n\\nAbstract: The problem of efficiently finding patterns in massive time series databases has attracted great interest, and, at least for the Euclidean distance measure, may now be regarded as a solved problem. However in recent years there has been an increasing awareness that Euclidean distance is inappropriate for many real world applications. The limitations of Euclidean distance stems from the fact that it is very sensitive to distortions in the time axis. A partial solution to this problem, Dynamic Time Warping (DTW), aligns the time axis before calculating the Euclidean distance. However, DTW can only address the problem of local scaling. As we demonstrate in this work, uniform scaling may be just as important in many domains, including applications as diverse as bioinformatics, space telemetry monitoring and motion editing for computer animation. In this work, we demonstrate a novel technique to speed up similarity search under uniform scaling. As we will demonstrate, our technique is simple and intuitive, and can achieve a speedup of 2 to 3 orders of magnitude under realistic settings.\",\n \"Title: Exact Discovery of Time Series Motifs\\nYear: 2009\\n\\nAbstract: Time series motifs are pairs of individual time series, or subsequences of a longer time series, which are very similar to each other. As with their discrete analogues in computational biology, this similarity hints at structure which has been conserved for some reason and may therefore be of interest. Since the formalism of time series motifs in 2002, dozens of researchers have used them for diverse applications in many different domains. Because the obvious algorithm for computing motifs is quadratic in the number of items, more than a dozen approximate algorithms to discover motifs have been proposed in the literature. In this work, for the first time, we show a tractable exact algorithm to find time series motifs. As we shall show through extensive experiments, our algorithm is up to three orders of magnitude faster than brute-force search in large datasets. We further show that our algorithm is fast enough to be used as a subroutine in higher level data mining algorithms for anytime classification, near-duplicate detection and summarization, and we consider detailed case studies in domains as diverse as electroencephalograph interpretation and entomological telemetry data mining.\",\n \"Title: Parameter-Free Audio Motif Discovery in Large Data Archives\\nYear: 2013\\n\\nAbstract: The discovery of repeated structure, i.e. motifs/near-duplicates, is often the first step in exploratory data mining. As such, the last decade has seen extensive research efforts in motif discovery algorithms for text, DNA, time series, protein sequences, graphs, images, and video. Surprisingly, there has been less attention devoted to finding repeated patterns in audio sequences, in spite of their ubiquity in science and entertainment. While there is significant work for the special case of motifs in music, virtually all this work makes many assumptions about data (often to the point of being genre specific) and thus these algorithms do not generalize to audio sequences containing animal vocalizations, industrial processes, or a host of other domains that we may wish to explore. In this work we introduce a novel technique for finding audio motifs. Our method does not require any domain-specific tuning and is essentially parameter-free. We demonstrate our algorithm on very diverse domains, finding audio motifs in laboratory mice vocalizations, wild animal sounds, music, and human speech. Our experiments demonstrate that our ideas are effective in discovering objectively correct or subjectively plausible motifs. Moreover, we show our novel probabilistic early abandoning approach is efficient, being two to three orders of magnitude faster than brute-force search, and thus faster than real-time for most problems.\",\n \"Title: Searching historical manuscripts for near-duplicate figures\\nYear: 2011\\n\\nAbstract: In the next decade a majority of all the books ever published will be digitized and online. Naturally, most of the data in historical manuscripts is text, but there is also a large amount devoted to images. This observation is responsible for the dramatic increase in interest in query-by-content systems for historical documents. While querying/indexing systems can be useful, we believe that this domain is finally ready for unsupervised discovery of patterns. With this in mind, we introduce an efficient and scalable technique that can detect approximately repeated occurrences of images both within and between historical texts. We demonstrate that this ability to find repeated shapes allows us to do automatic annotation of manuscripts. We show the utility of our technique on datasets dating back to the fourteenth century.\",\n \"Title: Quantitative Analysis of Live-Cell Growth at the Shoot Apex of Arabidopsis thaliana: Algorithms for Feature Measurement and Temporal Alignment\\nYear: 2013\\n\\nAbstract: Study of the molecular control of organ growth requires establishment of the causal relationship between gene expression and cell behaviors. We seek to understand this relationship at the shoot apical meristem (SAM) of model plant Arabidopsis thaliana. This requires the spatial mapping and temporal alignment of different functional domains into a single template. Live-cell imaging techniques allow us to observe real-time organ primordia growth and gene expression dynamics at cellular resolution. In this paper, we propose a framework for the measurement of growth features at the 3D reconstructed surface of organ primordia, as well as algorithms for robust time alignment of primordia. We computed areas and deformation values from reconstructed 3D surfaces of individual primordia from live-cell imaging data. Based on these growth measurements, we applied a multiple feature landscape matching (LAM-M) algorithm to ensure a reliable temporal alignment of multiple primordia. Although the original landscape matching (LAM) algorithm motivated our alignment approach, it sometimes fails to properly align growth curves in the presence of high noise/distortion. To overcome this shortcoming, we modified the cost function to consider the landscape of the corresponding growth features. We also present an alternate parameter-free growth alignment algorithm which performs as well as LAM-M for high-quality data, but is more robust to the presence of outliers or noise. Results on primordia and guppy evolutionary growth data show that the proposed alignment framework performs at least as well as the LAM algorithm in the general case, and significantly better in the case of increased noise.\",\n \"Title: Supporting exact indexing of arbitrarily rotated shapes and periodic time series under Euclidean and warping distance measures\\nYear: 2009\\n\\nAbstract: Shape matching and indexing is important topic in its own right, and is a fundamental subroutine in most shape data mining\\n algorithms. Given the ubiquity of shape, shape matching is an important problem with applications in domains as diverse as\\n biometrics, industry, medicine, zoology and anthropology. The distance/similarity measure for used for shape matching must\\n be invariant to many distortions, including scale, offset, noise, articulation, partial occlusion, etc. Most of these distortions\\n are relatively easy to handle, either in the representation of the data or in the similarity measure used. However, rotation\\n invariance is noted in the literature as being an especially difficult challenge. Current approaches typically try to achieve\\n rotation invariance in the representation of the data, at the expense of discrimination ability, or in the distance measure,\\n at the expense of efficiency. In this work, we show that we can take the slow but accurate approaches and dramatically speed\\n them up. On real world problems our technique can take current approaches and make them four orders of magnitude faster without\\n false dismissals. Moreover, our technique can be used with any of the dozens of existing shape representations and with all\\n the most popular distance measures including Euclidean distance, dynamic time warping and Longest Common Subsequence. We further\\n show that our indexing technique can be used to index star light curves, an important type of astronomical data, without modification.\",\n \"Title: On the need for time series data mining benchmarks: a survey and empirical demonstration\\nYear: 2003\\n\\nAbstract: In the last decade there has been an explosion of interest in mining time series data. Literally hundreds of papers have introduced new algorithms to index, classify, cluster and segment time series. In this work we make the following claim. Much of this work has very little utility because the contribution made (speed in the case of indexing, accuracy in the case of classification and clustering, model accuracy in the case of segmentation) offer an amount of \\\"improvement\\\" that would have been completely dwarfed by the variance that would have been observed by testing on many real world datasets, or the variance that would have been observed by changing minor (unstated) implementation details.To illustrate our point, we have undertaken the most exhaustive set of time series experiments ever attempted, re-implementing the contribution of more than two dozen papers, and testing them on 50 real world, highly diverse datasets. Our empirical results strongly support our assertion, and suggest the need for a set of time series benchmarks and more careful empirical evaluation in the data mining community.\",\n \"Title: A novel bit level time series representation with implication of similarity search and clustering\\nYear: 2005\\n\\nAbstract: Because time series are a ubiquitous and increasingly prevalent type of data, there has been much research effort devoted to time series data mining recently. As with all data mining problems, the key to effective and scalable algorithms is choosing the right representation of the data. Many high level representations of time series have been proposed for data mining. In this work, we introduce a new technique based on a bit level approximation of the data. The representation has several important advantages over existing techniques. One unique advantage is that it allows raw data to be directly compared to the reduced representation, while still guaranteeing lower bounds to Euclidean distance. This fact can be exploited to produce faster exact algorithms for similarly search. In addition, we demonstrate that our new representation allows time series clustering to scale to much larger datasets.\",\n \"Title: Scaling up Dynamic Time Warping to Massive Dataset\\nYear: 1999\\n\\nAbstract: There has been much recent interest in adapting da ta mining algorithms to time series databases. Many of these algorithms nee d to compare time series. Typically some variation or extension of Euclidean distance is used. However, as we demonstrate in this paper, Euclidean distance can b e an extremely brittle distance measure. Dynamic time warping (DTW) has been suggested as a technique to allow more robust distance calculations, however it is co mputationally expensive. In this paper we introduce a modification of DTW which operates on a higher level abstraction of the data, in particular, a piecewise linear representation. We demonstrate that our approach allows us to outperfo rm DTW by one to three orders of magnitude. We experimentally evaluate our approach on medical, astronomical and sign language data.\",\n \"Title: A Bit Level Representation for Time Series Data Mining with Shape Based Similarity\\nYear: 2006\\n\\nAbstract: Clipping is the process of transforming a real valued series into a sequence of bits representing whether each data is above or below the average. In this paper, we argue that clipping is a useful and flexible transformation for the exploratory analysis of large time dependent data sets. We demonstrate how time series stored as bits can be very efficiently compressed and manipulated and that, under some assumptions, the discriminatory power with clipped series is asymptotically equivalent to that achieved with the raw data. Unlike other transformations, clipped series can be compared directly to the raw data series. We show that this means we can form a tight lower bounding metric for Euclidean and Dynamic Time Warping distance and hence efficiently query by content. Clipped data can be used in conjunction with a host of algorithms and statistical tests that naturally follow from the binary nature of the data. A series of experiments illustrate how clipped series can be used in increasingly complex ways to achieve better results than other popular representations. The usefulness of the proposed representation is demonstrated by the fact that the results with clipped data are consistently better than those achieved with a Wavelet or Discrete Fourier Transformation at the same compression ratio for both clustering and query by content. The flexibility of the representation is shown by the fact that we can take advantage of a variable Run Length Encoding of clipped series to define an approximation of the Kolmogorov complexity and hence perform Kolmogorov based clustering.\",\n \"Title: Integrating Lite-Weight but Ubiquitous Data Mining into GUI Operating Systems\\nYear: 2005\\n\\nAbstract: Most visualization tools introduced in the literature are specialized for a particular task. In this work, we introduce a novel framework which allows visualization to take place in the background of normal day to day operations of any GUI based operating system such as MS Windows, OS X or Linux. Our system works by replacing the standard file icons with automatically generated icons that reflect the contents of the files in a principled way. We call such icons Intelligent Icons. While there is little utility in examining an individual icon, examining groups of them provides a greater possibility of unexpected and serendipitous discoveries. The utility of Intelligent Icons can be further enhanced by arranging them on the screen in a way that reflects their similarity/differences. We demonstrate the utility of our approach on data as diverse as DNA, text files, electrocardiograms, and Space Shuttle telemetry. In addition we show that our system is unique in also supporting fast and intuitive similarity search.\",\n \"Title: Ensemble-index: a new approach to indexing large databases\\nYear: 2001\\n\\nAbstract: The problem of similarity search (query-by-content) has attracted much research interest. It is a difficult problem because of the inherently high dimensionality of the data. The most promising solutions involve performing dimensionality reduction on the data, then indexing the reduced data with a multidimensional index structure. Many dimensionality reduction techniques have been proposed, including Singular Value Decomposition (SVD), the Discrete Fourier Transform (DFT), the Discrete Wavelet Transform (DWT) and Piecewise Polynomial Approximation. In this work, we introduce a novel framework for using ensembles of two or more representations for more efficient indexing. The basic idea is that instead of committing to a single representation for an entire dataset, different representations are chosen for indexing different parts of the database. The representations are chosen based upon a local view of the database. For example, sections of the data that can achieve a high fidelity representation with wavelets are indexed as wavelets, but highly spectral sections of the data are indexed using the Fourier transform. At query time, it is necessary to search several small heterogeneous indices, rather than one large homogeneous index. As we will theoretically and empirically demonstrate this results in much faster query response times.\",\n \"Title: A Simple Dimensionality Reduction Technique for Fast Similarity Search in Large Time Series Databases\\nYear: 2000\\n\\nAbstract: We address the problem of similarity search in large time series databases. We introduce a novel-dimensionality reduction technique that supports an indexing algorithm that is more than an order of magnitude faster than the previous best known method. In addition to being much faster our approach has numerous other advantages. It is simple to understand and implement, allows more flexible distance measures including weighted Euclidean queries and the index can be built in linear time. We call our approach PCA-indexing (Piece-wise Constant Approximation) and experimentally validate it on space telemetry, financial, astronomical, medical and synthetic data.\",\n \"Title: A Practical Tool for Visualizing and Data Mining Medical Time Series\\nYear: 2005\\n\\nAbstract: The increasing interest in time series data mining has had surprisingly little impact on real world medical applications. Practitioners who work with time series on a daily basis rarely take advantage of the wealth of tools that the data mining community has made available. In this work, we attempt to address this problem by introducing a parameter-light tool that allows users to efficiently navigate through large collections of time series. Our approach extracts features from a time series of arbitrary length and uses information about the relative frequency of these features to color a bitmap in a principled way. By visualizing the similarities and differences within a collection of bitmaps, a user can quickly discover clusters, anomalies, and other regularities within the data collection. We demonstrate the utility of our approach with a set of comprehensive experiments on real datasets from a variety of medical domains.\",\n \"Title: Indexing large human-motion databases\\nYear: 2004\\n\\nAbstract: Data-driven animation has become the industry standard for computer games and many animated movies and special effects. In particular, motion capture data recorded from live actors, is the most promising approach offered thus far for animating realistic human characters. However, the manipulation of such data for general use and re-use is not yet a solved problem. Many of the existing techniques dealing with editing motion rely on indexing for annotation, segmentation, and re-ordering of the data. Euclidean distance is inappropriate for solving these indexing problems because of the inherent variability found in human motion. The limitations of Euclidean distance stems from the fact that it is very sensitive to distortions in the time axis. A partial solution to this problem, Dynamic Time Warping (DTW), aligns the time axis before calculating the Euclidean distance. However, DTW can only address the problem of local scaling. As we demonstrate in this paper, global or uniform scaling is just as important in the indexing of human motion. We propose a novel technique to speed up similarity search under uniform scaling, based on bounding envelopes. Our technique is intuitive and simple to implement. We describe algorithms that make use of this technique, we perform an experimental analysis with real datasets, and we evaluate it in the context of a motion capture processing system. The results demonstrate the utility of our approach, and show that we can achieve orders of magnitude of speedup over the brute force approach, the only alternative solution currently available.\",\n \"Title: Efficiently finding unusual shapes in large image databases\\nYear: 2008\\n\\nAbstract: Among the visual features of multimedia content, shape is of particular interest because humans can often recognize objects solely on the basis of shape. Over the past three decades, there has been a great deal of research on shape analysis, focusing mostly on shape indexing, clustering, and classification. In this work, we introduce the new problem of finding shape discords, the most unusual shapes in a collection. We motivate the problem by considering the utility of shape discords in diverse domains including zoology, microscopy, anthropology, and medicine. While the brute force search algorithm has quadratic time complexity, we avoid this untenable lethargy by using locality-sensitive hashing to estimate similarity between shapes which enables us to reorder the search more efficiently and thus extract the maximum benefit from an admissible pruning strategy we introduce. An extensive experimental evaluation demonstrates that our approach is empirically linear in time.\",\n \"Title: Annotating Historical Archives of Images\\nYear: 2010\\n\\nAbstract: Recent programs like the Million Book Project and Google Print Library Project have archived several million books in digital format, and within a few years a significant fraction of world's books will be online. While the majority of the data will naturally be text, there will also be tens of millions of pages of images. Many of these images will defy automation annotation for the foreseeable future, but a considerable fraction of the images may be amiable to automatic annotation by algorithms that can link the historical image with a modern contemporary, with its attendant metatags. To perform this linking, there must be a suitable distance measure that appropriately combines the relevant features of shape, color, texture and text. However, the best combination of these features will vary from application to application and even from one manuscript to another. In this work, the authors propose a simple technique to learn the distance measure by perturbing the training set in a principled way.\",\n \"Title: A Grid-Based Index Method for Time Warping Distance\\nYear: 2004\\n\\nAbstract: Recently DTW (dynamic time warping) has been recognized as the most robust distance function to measure the similarity between\\n two time series, and this fact has spawned a flurry of research on this topic. Most indexing methods proposed for DTW are\\n based on the R-tree structure. Because of high dimensionality and loose lower bounds for time warping distance, the pruning\\n power of these tree structures are quite weak, resulting in inefficient search. In this paper, we propose a dimensionality\\n reduction method motivated by observations about the inherent character of each time series. A very compact index file is\\n constructed. By scanning the index file, we can get a very small candidate set, so that the number of page access is dramatically\\n reduced. We demonstrate the effectiveness of our approach on real and synthetic datasets.\",\n \"Title: Fast similarity search in the presence of longitudinal scaling in time series databases\\nYear: 1997\\n\\nAbstract: Abstract: The problem of finding patterns of interest in time series databases (query by content) is an important one, with applications in virtually every field of science. A variety of approaches have been suggested. These approaches are robust to noise, offset translation, and amplitude scaling to varying degrees. However, they are all extremely sensitive to scaling in the time axis (longitudinal scaling). We present a method for similarity search that is robust to scaling in the time axis, in addition to noise, offset translation, and amplitude scaling. The method has been tested on medical, financial, space telemetry and artificial data. Furthermore the method is exceptionally fast, with the predicted 2 to 4 orders of magnitude speedup actually observed. The method uses a piecewise linear representation of the original data. We also introduce a new algorithm which both decides the optimal number of linear segments to use, and produces the actual linear representation.\",\n \"Title: LB_Keogh supports exact indexing of shapes under rotation invariance with arbitrary representations and distance measures\\nYear: 2006\\n\\nAbstract: The matching of two-dimensional shapes is an important problem with applications in domains as diverse as biometrics, industry, medicine and anthropology. The distance measure used must be invariant to many distortions, including scale, offset, noise, partial occlusion, etc. Most of these distortions are relatively easy to handle, either in the representation of the data or in the similarity measure used. However rotation invariance seems to be uniquely difficult. Current approaches typically try to achieve rotation invariance in the representation of the data, at the expense of discrimination ability, or in the distance measure, at the expense of efficiency. In this work we show that we can take the slow but accurate approaches and dramatically speed them up. On real world problems our technique can take current approaches and make them four orders of magnitude faster, without false dismissals. Moreover, our technique can be used with any of the dozens of existing shape representations and with all the most popular distance measures including Euclidean distance, Dynamic Time Warping and Longest Common Subsequence.\",\n \"Title: Flying Insect Classification with Inexpensive Sensors.\\nYear: 2014\\n\\nAbstract: The ability to use inexpensive, noninvasive sensors to accurately classify flying insects would have significant implications for entomological research, and allow for the development of many useful applications in vector control for both medical and agricultural entomology. Given this, the last sixty years have seen many research efforts on this task. To date, however, none of this research has had a lasting impact. In this work, we explain this lack of progress. We attribute the stagnation on this problem to several factors, including the use of acoustic sensing devices, the overreliance on the single feature of wingbeat frequency, and the attempts to learn complex models with relatively little data. In contrast, we show that pseudo-acoustic optical sensors can produce vastly superior data, that we can exploit additional features, both intrinsic and extrinsic to the insect’s flight behavior, and that a Bayesian classification approach allows us to efficiently learn classification models that are very robust to overfitting. We demonstrate our findings with large scale experiments, as measured both by the number of insects and the number of species considered.\",\n \"Title: Finding the most unusual time series subsequence: algorithms and applications\\nYear: 2007\\n\\nAbstract: In this work we introduce the new problem of finding time seriesdiscords. Time series discords are subsequences of longer time series that are maximally different to all the rest of the time series subsequences. They thus capture the sense of the most unusual subsequence within a time series. While discords have many uses for data mining, they are particularly attractive as anomaly detectors because they only require one intuitive parameter (the length of the subsequence) unlike most anomaly detection algorithms that typically require many parameters. While the brute force algorithm to discover time series discords is quadratic in the length of the time series, we show a simple algorithm that is three to four orders of magnitude faster than brute force, while guaranteed to produce identical results. We evaluate our work with a comprehensive set of experiments on diverse data sources including electrocardiograms, space telemetry, respiration physiology, anthropological and video datasets.\",\n \"Title: A decade of progress in indexing and mining large time series databases\\nYear: 2006\\n\\nAbstract: Time series data is ubiquitous; large volumes of time series data are routinely created in scientific, industrial, entertainment, medical and biological domains. Examples include gene expression data, electrocardiograms, electroencephalograms, gait analysis, stock market quotes, space telemetry etc. Although statisticians have worked with time series for more than a century, many of their techniques hold little utility for researchers working with massive time series databases.A decade ago, a seminal paper by Faloutsos, Ranganathan, Manolopoulos appeared in SIGMOD. The paper, Fast Subsequence Matching in Time-Series Databases, has spawned at least a thousand references and extensions in the database/ data mining and information retrieval communities. This tutorial will summarize the decade of progress since this influential paper appeared.\",\n \"Title: MDL-based time series clustering.\\nYear: 2012\\n\\nAbstract: Time series data are pervasive across all human endeavors, and clustering is arguably the most fundamental data mining application. Given this, it is somewhat surprising that the problem of time series clustering from a single stream remains largely unsolved. Most work on time series clustering considers the clustering of individual time series that have been carefully extracted from their original context, for example, gene expression profiles, individual heartbeats, or individual gait cycles. The few attempts at clustering time series streams have been shown to be objectively incorrect in some cases, and in other cases shown to work only on the most contrived synthetic datasets by carefully adjusting a large set of parameters. In this work, we make two fundamental contributions that allow for the first time, the meaningful clustering of subsequences from a time series stream. First, we show that the problem definition for time series clustering from streams currently used is inherently flawed, and a new definition is necessary. Second, we show that the minimum description length framework offers an efficient, effective, and essentially parameter-free method for time series clustering. We show that our method produces objectively correct results on a wide variety of datasets from medicine, speech recognition, zoology, gesture recognition, and industrial process analyses.\",\n \"Title: Finding surprising patterns in a time series database in linear time and space\\nYear: 2002\\n\\nAbstract: The problem of finding a specified pattern in a time series database (i.e. query by content) has received much attention and is now a relatively mature field. In contrast, the important problem of enumerating all surprising or interesting patterns has received far less attention. This problem requires a meaningful definition of \\\"surprise\\\", and an efficient search technique. All previous attempts at finding surprising patterns in time series use a very limited notion of surprise, and/or do not scale to massive datasets. To overcome these limitations we introduce a novel technique that defines a pattern surprising if the frequency of its occurrence differs substantially from that expected by chance, given some previously seen data.\",\n \"Title: HOT SAX: Efficiently Finding the Most Unusual Time Series Subsequence\\nYear: 2005\\n\\nAbstract: In this work, we introduce the new problem of finding time series discords. Time series discords are subsequences of a longer time series that are maximally different to all the rest of the time series subsequences. They thus capture the sense of the most unusual subsequence within a time series. Time series discords have many uses for data mining, including improving the quality of clustering, data cleaning, summarization, and anomaly detection. As we will show, discords are particularly attractive as anomaly detectors because they only require one intuitive parameter (the length of the subsequence) unlike most anomaly detection algorithms that typically require many parameters. We evaluate our work with a comprehensive set of experiments. In particular, we demonstrate the utility of discords with objective experiments on domains as diverse as Space Shuttle telemetry monitoring, medicine, surveillance, and industry, and we demonstrate the effectiveness of our discord discovery algorithm with more than one million experiments, on 82 different datasets from diverse domains.\",\n \"Title: The Asymmetric Approximate Anytime Join: A New Primitive with Applications to Data Mining\\nYear: 2008\\n\\nAbstract: It has long been noted that many data mining algorithms can be built on top of join algorithms. This has lead to a wealth of recent work on efficiently supporting such joins with various indexing techniques. However, there are many applications which are characterized by two special conditions, firstly the two datasets to be joined are of radically different sizes, a situation we call an asymmetric join. Secondly, the two datasets are not, and possibly can not be indexed for some reason. In such circumstances the time complexity is proportional to the product of the number of objects in each of the two datasets, an untenable proposition in most cases. In this work we make two contributions to mitigate this situation. We argue that for many applications, an exact solution to the problem is not required, and we show that by framing the problem as an anytime algorithm we can extract most of the benefit of a join in a small fraction of the time taken by the full algorithm. In situations where the exact answer is required, we show that we can quickly index just the smaller dataset on the fly, and greatly speed up the exact computation. We motivate and empirically confirm the utility of our ideas with case studies on problems as diverse as batch classification, anomaly detection and annotation of historical manuscripts.\",\n \"Title: Locally adaptive dimensionality reduction for indexing large time series databases\\nYear: 2002\\n\\nAbstract: Similarity search in large time series databases has attracted much research interest recently. It is a difficult problem because of the typically high dimensionality of the data. The most promising solutions involve performing dimensionality reduction on the data, then indexing the reduced data with a multidimensional index structure. Many dimensionality reduction techniques have been proposed, including Singular Value Decomposition (SVD), the Discrete Fourier transform (DFT), and the Discrete Wavelet Transform (DWT). In this article, we introduce a new dimensionality reduction technique, which we call Adaptive Piecewise Constant Approximation (APCA). While previous techniques (e.g., SVD, DFT and DWT) choose a common representation for all the items in the database that minimizes the global reconstruction error, APCA approximates each time series by a set of constant value segments of varying lengths such that their individual reconstruction errors are minimal. We show how APCA can be indexed using a multidimensional index structure. We propose two distance measures in the indexed space that exploit the high fidelity of APCA for fast searching: a lower bounding Euclidean distance approximation, and a non-lower-bounding, but very tight, Euclidean distance approximation, and show how they can support fast exact searching and even faster approximate searching on the same index structure. We theoretically and empirically compare APCA to all the other techniques and demonstrate its superiority.\",\n \"Title: An Augmented Visual Query Mechanism for Finding Patterns in Time Series Data\\nYear: 2002\\n\\nAbstract: Relatively few query tools exist for data exploration and pattern identification in time series data sets. In previous work we introduced Time-boxes. Timeboxes are rectangular, direct-manipulation queries for studying time-series datasets. We demonstrated how Timeboxes can be used to support interactive exploration via dynamic queries, along with overviews of query results and drag-and-drop support for query-by-example. In this paper, we extend our work by introducing Variable Time Timeboxes (VTT). VTTs are a natural generalization of Timeboxes, which permit the specification of queries that allow a degree of uncertainty in the time axis. We carefully motivate the need for these more expressive queries, and demonstrate the utility of our approach on several data sets.\",\n \"Title: Mining Historical Documents for Near-Duplicate Figures\\nYear: 2011\\n\\nAbstract: The increasing interest in archiving all of humankind's cultural artifacts has resulted in the digitization of millions of books, and soon a significant fraction of the world's books will be online. Most of the data in historical manuscripts is text, but there is also a significant fraction devoted to images. This fact has driven much of the recent increase in interest in query-by-content systems for images. While querying/indexing systems can undoubtedly be useful, we believe that the historical manuscript domain is finally ripe for true unsupervised discovery of patterns and regularities. To this end, we introduce an efficient and scalable system which can detect approximately repeated occurrences of shape patterns both within and between historical texts. We show that this ability to find repeated shapes allows automatic annotation of manuscripts, and allows users to trace the evolution of ideas. We demonstrate our ideas on datasets of scientific and cultural manuscripts dating back to the fourteenth century.\",\n \"Title: An Enhanced Representation of Time Series Which Allows Fast and Accurate Classification, Clustering and Relevance Feedback\\nYear: 1998\\n\\nAbstract: We introduce an extended representation of time series that allows fast, accurate classification and clustering in addition to the ability to explore time series data in a relevance feedback framework. The representation consists of piece- wise linear segments to represent shape and a weight vector that contains the relative importance of each individual linear segment. In the classification context, the weights are learned automatically as part of the training cycle. In the relevance feedback context, the weights are determined by an interactive and iterative process in which users rate various choices presented to them. Our representation allows a user to define a variety of similarity measures that can be tailored to specific domains. We demonstrate our approach on space telemetry, medical and synthetic data.\",\n \"Title: Making Time-Series Classification More Accurate Using Learned Constraints\\nYear: 2004\\n\\nAbstract: It has long been known that Dynamic Time Warping (DTW) is superior to Euclidean distance for classification and clustering of time series. However, until lately, most research has utilized Euclidean distance because it is more efficiently calculated. A recently introduced technique that greatly mitigates DTWs demanding CPU time has sparked a flurry of research activity. However, the technique and its many extensions still only allow DTW to be applied to moderately large datasets. In addition, almost all of the research on DTW has focused exclusively on speeding up its calculation; there has been little work done on improving its accuracy. In this work, we target the accuracy aspect of DTW performance and introduce a new framework that learns arbitrary constraints on the warping path of the DTW calculation. Apart from improving the accuracy of classification, our technique as a side effect speeds up DTW by a wide margin as well. We show the utility of our approach on datasets from diverse domains and demonstrate significant gains in accuracy and efficiency.\",\n \"Title: Classification Of Multi-Dimensional Streaming Time Series By Weighting Each Classifier'S Track Record\\nYear: 2013\\n\\nAbstract: Extensive research on time series classification in the last decade has produced fast and accurate algorithms for the single-dimensional case. However, the increasing prevalence of inexpensive sensors has reinforced the need for algorithms to handle multi-dimensional time series. For example, modern smartphones have at least a dozen sensors capable of producing streaming time series, and hospital-based (and increasingly, home-based) medical devices can produce time series streams from more than twenty sensors. The two most common ways to generalize from single to multi-dimensional data are to use all the streams or just the single best stream as determined at training time. However, as we show here, both approaches can be very brittle. Moreover, neither approach exploits the observation that different sensors may be considered \\\"experts\\\" on different classes. In this work, we introduce a novel framework for multi-dimensional time series classification that weights the class prediction from each time series stream. These weights are based not only on each stream's previous track record on the class it is currently predicting, but also on the distance from the unlabeled object. As we demonstrate with extensive experiments on real data, our method is more accurate than current approaches and particularly robust in the face of concept drift or sensor noise.\",\n \"Title: SIGKDD demo: sensors and software to allow computational entomology, an emerging application of data mining\\nYear: 2011\\n\\nAbstract: The history of humankind is intimately connected to insects. Insect borne diseases kill a million people and destroy tens of billions of dollars worth of crops annually. However, at the same time, beneficial insects pollinate the majority of crop species, and it has been estimated that approximately one third of all food consumed by humans is directly pollinated by bees alone. Given the importance of insects in human affairs, it is somewhat surprising that computer science has not had a larger impact in entomology. We believe that recent advances in sensor technology are beginning change this, and a new field of Computational Entomology will emerge. We will demonstrate an inexpensive sensor that allows us to capture data from flying insects, and the software that allows us to analyze the data. Moreover, we will distribute both the sensors and software for free, to parties willing to take part in a crowdsourcing project on insect classification.\",\n \"Title: Semi-supervised time series classification\\nYear: 2006\\n\\nAbstract: The problem of time series classification has attracted great interest in the last decade. However current research assumes the existence of large amounts of labeled training data. In reality, such data may be very difficult or expensive to obtain. For example, it may require the time and expertise of cardiologists, space launch technicians, or other domain specialists. As in many other domains, there are often copious amounts of unlabeled data available. For example, the PhysioBank archive contains gigabytes of ECG data. In this work we propose a semi- supervised technique for building time series classifiers. While such algorithms are well known in text domains, we will show that special considerations must be made to make them both efficient and effective for the time series domain. We evaluate our work with a comprehensive set of experiments on diverse data sources including electrocardiograms, handwritten documents, manufacturing, and video datasets. The experimental results demonstrate that our approach requires only a handful of labeled examples to construct accurate classifiers.\",\n \"Title: Towards parameter-free data mining\\nYear: 2004\\n\\nAbstract: Most data mining algorithms require the setting of many input parameters. Two main dangers of working with parameter-laden algorithms are the following. First, incorrect settings may cause an algorithm to fail in finding the true patterns. Second, a perhaps more insidious problem is that the algorithm may report spurious patterns that do not really exist, or greatly overestimate the significance of the reported patterns. This is especially likely when the user fails to understand the role of parameters in the data mining process.Data mining algorithms should have as few parameters as possible, ideally none. A parameter-free algorithm would limit our ability to impose our prejudices, expectations, and presumptions on the problem at hand, and would let the data itself speak to us. In this work, we show that recent results in bioinformatics and computational theory hold great promise for a parameter-free data-mining paradigm. The results are motivated by observations in Kolmogorov complexity theory. However, as a practical matter, they can be implemented using any off-the-shelf compression algorithm with the addition of just a dozen or so lines of code. We will show that this approach is competitive or superior to the state-of-the-art approaches in anomaly/interestingness detection, classification, and clustering with empirical tests on time series/DNA/text/video datasets.\",\n \"Title: Discovering the Intrinsic Cardinality and Dimensionality of Time Series Using MDL\\nYear: 2011\\n\\nAbstract: Most algorithms for mining or indexing time series data do not operate directly on the original data, but instead they consider alternative representations that include transforms, quantization, approximation, and multi-resolution abstractions. Choosing the best representation and abstraction level for a given task/dataset is arguably the most critical step in time series data mining. In this paper, we investigate techniques to discover the natural intrinsic representation model, dimensionality and alphabet cardinality of a time series. The ability to discover these intrinsic features has implications beyond selecting the best parameters for particular algorithms, as characterizing data in such a manner is useful in its own right and an important sub-routine in algorithms for classification, clustering and outlier discovery. We will frame the discovery of these intrinsic features in the Minimal Description Length (MDL) framework. Extensive empirical tests show that our method is simpler, more general and significantly more accurate than previous methods, and has the important advantage of being essentially parameter-free.\",\n \"Title: Intelligent Icons: Integrating Lite-Weight Data Mining and Visualization into GUI Operating Systems\\nYear: 2006\\n\\nAbstract: The vast majority of visualization tools introduced so far are specialized pieces of software that run explicitly on a particular dataset at a particular time for a particular purpose. In this work we introduce a novel framework for allowing visualization to take place in the background of normal day-to-day operation of any GUI based operation system. Our system works by replacing the standard file icons with automatically created icons that reflect the contents of the files in a principled way. We call such icons INTELLIGENT ICONS. The utility of Intelligent Icons is further enhanced by arranging them in a way that reflects their similarity/differences. We demonstrate the utility of our approach on diverse applications.\",\n \"Title: An Indexing Scheme for Fast Similarity Search in Large Time Series Databases\\nYear: 1999\\n\\nAbstract: We address the problem of similarity search in large time series databases. We introduce a novel indexing algorithm that allows faster retrieval. The index is formed by creating bins that contain time series subsequences of approximately the same shape. For each bin, we can quickly calculate a lower-bound on the distance between a given query and the most similar element of the bin. This bound allows us to search the bins in best first order, and to prune some bins from the search space without having to examine the contents. Additional speedup is obtained by optimizing the data within the bins such that we can avoid having to compare the query to every item in the bin. We call our approach STB-indexing and experimentally validate it on space telemetry, medical and synthetic data, demonstrating approximately an order of magnitude speed-up.\",\n \"Title: A compression‐based distance measure for texture\\nYear: 2010\\n\\nAbstract: The analysis of texture is an important subroutine in application areas as diverse as biology, medicine, robotics, and forensic science. While the last three decades have seen extensive research in algorithms to measure texture similarity, almost all existing methods require the careful setting of many parameters. There are many problems associated with a lot of parameters, the most obvious of which is that with many parameters to fit, it is very difficult to avoid overfitting. In this work, we propose to extend recent advances in Kolmogorov complexity‐based similarity measures to texture matching problems. These Kolmogorov‐based methods have been shown to be very useful in intrinsically discrete domains such as DNA, protein sequences, MIDI music, and natural languages; however, they are not well defined for real‐valued data. To address this, we introduce a very simple idea, the Campana‐Keogh (CK) video compression‐based method for texture measures. These measures utilize video compressors to approximate the Kolmogorov complexity. Using the parameter‐free CK method, we novely utilize lossy compression to create an efficient and robust parameter‐lite texture similarity measure: the CK‐1 distance measure. We demonstrate the utility of our measure with extensive empirical evaluations on real‐world case studies drawn from nematology, arachnology, entomology, medicine, forensics, texture analysis benchmarks, and many other domains. Copyright © 2010 Wiley Periodicals, Inc. Statistical Analysis and Data Mining 3: 381‐398, 2010 © 2010 Wiley Periodicals, Inc.\",\n \"Title: Autocannibalistic and Anyspace Indexing Algorithms with Application to Sensor Data Mining\\nYear: 2009\\n\\nAbstract: ABSTRACT Efficient indexing ,is at ,the heart of many ,data mining algorithms. A simple and extremely effective algorithm for indexing under any metric space was introduced in 1991 by Orchard. Orchard’s algorithm ,has ,not received ,much attention in the ,data mining ,and ,database ,community because of a fatal flaw; it requires quadratic space. In this work,we show ,that we can ,produce a reduced ,version of Orchard’s algorithm that requires much less space, but produces,nearly identical speedup. We achieve ,this by casting the algorithm in an anyspace framework, allowing deployed applications to take as much,of an,index as their main memory/sensor can afford. As we shall demonstrate, this ability to create an ,anyspace algorithm also allows us tocreate auto-cannibalistic algorithms. Auto-cannibalistic algorithms are algorithms which ,initially require a certain amount of space to index or classify data, but if unexpected circumstances require them to store additional information, they can dynamically ,delete parts of themselves ,to make room for the new data. We demonstrate,the utility of auto- cannibalistic algorithms ,in a ,fielded project on insect monitoring with low power sensors, and a simple autonomous,robot application. Keywords Indexing, Anyspace Algorithms, Data Mining, Sensors.\",\n \"Title: An Online Algorithm for Segmenting Time Series\\nYear: 2001\\n\\nAbstract: In recent years, there has been an explosion of interest in mining time-series databases. As with most computer science problems, representation of the data is the key to efficient and effective solutions. One of the most commonly used representations is piecewise linear approximation. This representation has been used by various researchers to support clustering, classification, indexing and association rule mining of time-series data. A variety of algorithms have been proposed to obtain this representation, with several algorithms having been independently rediscovered several times. In this paper, we undertake the first extensive review and empirical comparison of all proposed techniques. We show that all these algorithms have fatal flaws from a data-mining perspective. We introduce a novel algorithm that we empirically show to be superior to all others in the literature\",\n \"Title: Data mining and information retrieval in time series/multimedia databases\\nYear: 2006\\n\\nAbstract: Time series and multimedia data are ubiquitous; large volumes of such data are routinely created in scientific, industrial, entertainment, medical and biological domains. Examples include gene expression data, X-rays, electrocardiograms, electroencephalograms, gait analysis, stock market quotes, space telemetry etc.A decade ago, a seminal paper by Faloutsos, Ranganathan, Manolopoulos appeared in SIGMOD [1]. The paper, Fast Subsequence Matching in Time-Series Databases, has spawned at least a thousand references and extensions in the database/data mining and information retrieval communities. This tutorial will summarize the decade of progress since this influential paper appeared.\",\n \"Title: How to Do Good Data Mining Research and Get it Published in Top Venues\\nYear: 2010\\n\\nAbstract: While ICDM has traditionally enjoyed an unusually high quality of reviewing, there is no doubt that publishing in ICDM is very challenging. In this tutorial Dr. Keogh will demonstrate some simple ideas to enhance the probability of success in getting your paper published in a top data mining conference, and after the work is published, getting it highly cited.\",\n \"Title: Compression-based data mining of sequential data\\nYear: 2007\\n\\nAbstract: The vast majority of data mining algorithms require the setting of many input parameters. The dangers of working with parameter-laden algorithms are twofold. First, incorrect settings may cause an algorithm to fail in finding the true patterns. Second, a perhaps more insidious problem is that the algorithm may report spurious patterns that do not really exist, or greatly overestimate the significance of the reported patterns. This is especially likely when the user fails to understand the role of parameters in the data mining process. Data mining algorithms should have as few parameters as possible. A parameter-light algorithm would limit our ability to impose our prejudices, expectations, and presumptions on the problem at hand, and would let the data itself speak to us. In this work, we show that recent results in bioinformatics, learning, and computational theory hold great promise for a parameter-light data-mining paradigm. The results are strongly connected to Kolmogorov complexity theory. However, as a practical matter, they can be implemented using any off-the-shelf compression algorithm with the addition of just a dozen lines of code. We will show that this approach is competitive or superior to many of the state-of-the-art approaches in anomaly/interestingness detection, classification, and clustering with empirical tests on time series/DNA/text/XML/video datasets. As a further evidence of the advantages of our method, we will demonstrate its effectiveness to solve a real world classification problem in recommending printing services and products.\",\n \"Title: Relevance Feedback Retrieval of Time Series Data\\nYear: 1999\\n\\nAbstract: ABSTRACT There has been much ,recent interest in retrieval of time ,series data. Earlier work has used a fixed similarity metric (e.g., Euclidean distance) to determine ,the similarity between ,a userspecified query and items in the database. Here, we describe a novel approach,to retrieval of time series data by using relevance feedback,from the user to adjust ,the similarity metric. This is important because the Euclidean distance metric does not capture many notions of similarity between time series. In particular, Euclidean distance is sensitive to various ,“distortions” such as offset translation, amplitude scaling, etc. Depending on the domain and the user, one may wish a query to be sensitive or insensitive to these ,distortions to varying ,degrees. This paper addresses this problem ,by introducing ,a profile that encodes the user's subjective notion of similarity in a domain. These profiles can be learned ,continuously from interaction with the user. We further show,how,the user profile may,be embedded ,in a,system that uses relevance feedback ,to modify ,the query in a ,manner analogous to the familiar text retrieval algorithms. Keywords Time series, multimedia data, relevance feedback, modeling user subjectivity.\",\n \"Title: Iterative Incremental Clustering of Time Series\\nYear: 2004\\n\\nAbstract: We present a novel anytime version of partitional clustering algorithm, such as k-Means and EM, for time series. The algorithm works by lever-aging off the multi-resolution property of wavelets. The dilemma of choosing the initial centers is mitigated by initializing the centers at each approximation level, using the final centers returned by the coarser representations. In addition to casting the clustering algorithms as anytime algorithms, this approach has two other very desirable properties. By working at lower dimensionalities we can efficiently avoid local minima. Therefore, the quality of the clustering is usually better than the batch algorithm. In addition, even if the algorithm is run to completion, our approach is much faster than its batch counterpart. We explain, and empirically demonstrate these surprising and desirable properties with comprehensive experiments on several publicly available real data sets. We further demonstrate that our approach can be generalized to a framework of much broader range of algorithms or data mining problems.\",\n \"Title: Finding Unusual Medical Time-Series Subsequences: Algorithms and Applications\\nYear: 2006\\n\\nAbstract: In this work, we introduce the new problem of finding time series discords. Time series discords are subsequences of longer time series that are maximally different to all the rest of the time series subsequences. They thus capture the sense of the most unusual subsequence within a time series. While discords have many uses for data mining, they are particularly attractive as anomaly detectors because they only require one intuitive parameter (the length of the subsequence), unlike most anomaly detection algorithms that typically require many parameters. While the brute force algorithm to discover time series discords is quadratic in the length of the time series, we show a simple algorithm that is three to four orders of magnitude faster than brute force, while guaranteed to produce identical results. We evaluate our work with a comprehensive set of experiments on electrocardiograms and other medical datasets\",\n \"Title: Three Myths about Dynamic Time Warping Data Mining\\nYear: 2005\\n\\nAbstract: The Dynamic Time Warping (DTW) distance measure is a technique that has long been known in speech recognition community. It allows a non-linear mapping of one signal to another by minimizing the distance between the two. A decade ago, DTW was introduced into Data Mining community as a utility for various tasks for time series problems including classification, clustering, and anomaly detection. The technique has flourished, particularly in the last three years, and has been applied to a variety of problems in various disciplines. In spite of DTW's great success, there are still several persistent \\\"myths\\\" about it. These myths have caused confusion and led to much wasted research effort. In this work, we will dispel these myths with the most comprehensive set of time series experiments ever conducted.\",\n \"Title: Establishing the provenance of historical manuscripts with a novel distance measure\\nYear: 2015\\n\\nAbstract: The recent digitization of more than 20 million books has been led by initiatives from countries wishing to preserve their cultural heritage and by several commercial endeavors, including the Google Print Library Project. It is expected that within a few years a significant fraction of the world's books will be online. However, for millions of complete books and tens of millions of loose pages, the provenance of the manuscripts may be completely unknown or disputed, thus denying historians an understanding of the context in which the content was created. In a handful of cases, it may be possible for experts to regain the provenance by examining linguistic, cultural and/or stylistic clues. However, such experts are a rarity and these investigations are time-consuming and expensive. One technique used by experts to establish provenance is the examination of the ornate initial letters appearing in the questioned manuscript. By comparing the initial letters in the manuscript to annotated initial letters whose origin is known, the provenance can be determined. In this work, we show for the first time that we can reproduce this ability with a computer algorithm. We use a recently introduced technique to measure texture similarity and show that it can recognize initial letters with an accuracy that rivals or exceeds human performance. A brute force implementation of this measure would require several months to process a single large book; however, we introduce a novel lower bound that allows us to process the books in hours or minutes.\",\n \"Title: Dynamic Time Warping Averaging of Time Series Allows Faster and More Accurate Classification\\nYear: 2014\\n\\nAbstract: Recent years have seen significant progress in improving both the efficiency and effectiveness of time series classification. However, because the best solution is typically the Nearest Neighbor algorithm with the relatively expensive Dynamic Time Warping as the distance measure, successful deployments on resource constrained devices remain elusive. Moreover, the recent explosion of interest in wearable devices, which typically have limited computational resources, has created a growing need for very efficient classification algorithms. A commonly used technique to glean the benefits of the Nearest Neighbor algorithm, without inheriting its undesirable time complexity, is to use the Nearest Centroid algorithm. However, because of the unique properties of (most) time series data, the centroid typically does not resemble any of the instances, an unintuitive and underappreciated fact. In this work we show that we can exploit a recent result to allow meaningful averaging of 'warped' times series, and that this result allows us to create ultra-efficient Nearest 'Centroid' classifiers that are at least as accurate as their more lethargic Nearest Neighbor cousins.\",\n \"Title: Generating Synthetic Data to Allow Learning from a Single Exemplar per Class.\\nYear: 2014\\n\\nAbstract: Recent years have seen an explosion in the volume of historical documents placed online. The individuality of fonts combined with the degradation suffered by century old manuscripts means that Optical Character Recognition Systems do not work well here. As human transcription is prohibitively expensive, recent efforts focused on human/computer cooperative transcription: a human annotates a small fraction of a text to provide labeled data for recognition algorithms. Such a system naturally begs the question of how much data must the human label? In this work we show that we can do well even if the human labels only a single instance from each class. We achieve this good result using two novel observations: we can leverage off a recently introduced parameter-free distance measure, improving it by taking into account the \\\"complexity\\\" of the glyphs being compared; we can estimate this complexity using synthetic but plausible instances made from the single training instance. We demonstrate the utility of our observations on diverse historical manuscripts.\",\n \"Title: Accelerating the dynamic time warping distance measure using logarithmetic arithmetic\\nYear: 2014\\n\\nAbstract: This paper describes an application-specific embedded processor with instruction set extensions (ISEs) for the Dynamic Time Warping (DTW) distance measure, which is widely used in time series similarity search. The ISEs in this paper are implemented using a form of logarithmic arithmetic that offers significant performance and power/energy advantages compared to more traditional floating-point operations.\",\n \"Title: Time Series Epenthesis: Clustering Time Series Streams Requires Ignoring Some Data\\nYear: 2011\\n\\nAbstract: Given the pervasiveness of time series data in all human endeavors, and the ubiquity of clustering as a data mining application, it is somewhat surprising that the problem of time series clustering from a single stream remains largely unsolved. Most work on time series clustering considers the clustering of individual time series, e.g., gene expression profiles, individual heartbeats or individual gait cycles. The few attempts at clustering time series streams have been shown to be objectively incorrect in some cases, and in other cases shown to work only on the most contrived datasets by carefully adjusting a large set of parameters. In this work, we make two fundamental contributions. First, we show that the problem definition for time series clustering from streams currently used is inherently flawed, and a new definition is necessary. Second, we show that the Minimum Description Length (MDL) framework offers an efficient, effective and essentially parameter-free method for time series clustering. We show that our method produces objectively correct results on a wide variety of datasets from medicine, zoology and industrial process analyses.\",\n \"Title: Clustering of Symbols Using Minimal Description Length\\nYear: 2013\\n\\nAbstract: The clustering of glyphs (individual letters/characters/symbols) is typically the first step in document processing algorithms and a critical enabling technology for most historical document indexing techniques. In this work, we take a step back from current domain/language specialized research efforts to consider the problem from an agnostic perspective. In particular, we claim that, independent of the distance measure used, any method that attempts to cluster all the data is almost certainly doomed to failure. We explain this observation, and introduce a clustering method based on Minimum Description Length (MDL) that can overcome it.\",\n \"Title: Classification of streaming time series under more realistic assumptions\\nYear: 2016\\n\\nAbstract: Much of the vast literature on time series classification makes several assumptions about data and the algorithm's eventual deployment that are almost certainly unwarranted. For example, many research efforts assume that the beginning and ending points of the pattern of interest can be correctly identified, during both the training phase and later deployment. Another example is the common assumption that queries will be made at a constant rate that is known ahead of time, thus computational resources can be exactly budgeted. In this work, we argue that these assumptions are unjustified, and this has in many cases led to unwarranted optimism about the performance of the proposed algorithms. As we shall show, the task of correctly extracting individual gait cycles, heartbeats, gestures, behaviors, etc., is generally much more difficult than the task of actually classifying those patterns. Likewise, gesture classification systems deployed on a device such as Google Glass may issue queries at frequencies that range over an order of magnitude, making it difficult to plan computational resources. We propose to mitigate these problems by introducing an alignment-free time series classification framework. The framework requires only very weakly annotated data, such as \\\\\\\"in this ten minutes of data, we see mostly normal heartbeats$$\\\\\\\\ldots $$¿,\\\\\\\" and by generalizing the classic machine learning idea of data editing to streaming/continuous data, allows us to build robust, fast and accurate anytime classifiers. We demonstrate on several diverse real-world problems that beyond removing unwarranted assumptions and requiring essentially no human intervention, our framework is both extremely fast and significantly more accurate than current state-of-the-art approaches.\",\n \"Title: Accelerating the discovery of unsupervised-shapelets\\nYear: 2016\\n\\nAbstract: Over the past decade, time series clustering has become an increasingly important research topic in data mining community. Most existing methods for time series clustering rely on distances calculated from the raw data using the Euclidean distance or Dynamic Time Warping distance as the distance measure. However, the presence of significant noise, dropouts, or extraneous data can greatly limit the accuracy of clustering in this domain. Moreover, for most real world problems, we cannot expect objects from the same class to be equal in length. As a consequence, most work on time series clustering only considers the clustering of individual time series “behaviors,” e.g., individual heart beats or individual gait cycles, and contrives the time series in some way to make them all equal in length. However, automatically formatting the data in such a way is often a harder problem than the clustering itself. In this work, we show that by using only some local patterns and deliberately ignoring the rest of the data, we can mitigate the above problems and cluster time series of different lengths, e.g., cluster one heartbeat with multiple heartbeats. To achieve this, we exploit and extend a recently introduced concept in time series data mining called . Unlike existing work, our work demonstrates the unintuitive fact that shapelets can be learned from time series. We show, with extensive empirical evaluation in diverse domains, that our method is more accurate than existing methods. Moreover, in addition to accurate clustering results, we show that our work also has the potential to give insight into the domains to which it is applied. While a brute-force algorithm to discover shapelets in an unsupervised way could be untenably slow, we introduce two novel optimization procedures to significantly speed up the unsupervised-shapelet discovery process and allow it to be cast as an anytime algorithm.\",\n \"Title: A general framework for never-ending learning from time series streams\\nYear: 2015\\n\\nAbstract: Time series classification has been an active area of research in the data mining community for over a decade, and significant progress has been made in the tractability and accuracy of learning. However, virtually all work assumes a one-time training session in which labeled examples of all the concepts to be learned are provided. This assumption may be valid in a handful of situations, but it does not hold in most medical and scientific applications where we initially may have only the vaguest understanding of what concepts can be learned. Based on this observation, we propose a never-ending learning framework for time series in which an agent examines an unbounded stream of data and occasionally asks a teacher (which may be a human or an algorithm) for a label. We demonstrate the utility of our ideas with experiments that consider real-world problems in domains as diverse as medicine, entomology, wildlife monitoring, and human behavior analyses.\",\n \"Title: Faster and more accurate classification of time series by exploiting a novel dynamic time warping averaging algorithm.\\nYear: 2016\\n\\nAbstract: A concerted research effort over the past two decades has heralded significant improvements in both the efficiency and effectiveness of time series classification. The consensus that has emerged in the community is that the best solution is a surprisingly simple one. In virtually all domains, the most accurate classifier is the nearest neighbor algorithm with dynamic time warping as the distance measure. The time complexity of dynamic time warping means that successful deployments on resource-constrained devices remain elusive. Moreover, the recent explosion of interest in wearable computing devices, which typically have limited computational resources, has greatly increased the need for very efficient classification algorithms. A classic technique to obtain the benefits of the nearest neighbor algorithm, without inheriting its undesirable time and space complexity, is to use the nearest centroid algorithm. Unfortunately, the unique properties of (most) time series data mean that the centroid typically does not resemble any of the instances, an unintuitive and underappreciated fact. In this paper we demonstrate that we can exploit a recent result by Petitjean et al. to allow meaningful averaging of “warped” time series, which then allows us to create super-efficient nearest “centroid” classifiers that are at least as accurate as their more computationally challenged nearest neighbor relatives. We demonstrate empirically the utility of our approach by comparing it to all the appropriate strawmen algorithms on the ubiquitous UCR Benchmarks and with a case study in supporting insect classification on resource-constrained sensors.\",\n \"Title: Discovery of Meaningful Rules in Time Series\\nYear: 2015\\n\\nAbstract: The ability to make predictions about future events is at the heart of much of science; so, it is not surprising that prediction has been a topic of great interest in the data mining community for the last decade. Most of the previous work has attempted to predict the future based on the current value of a stream. However, for many problems the actual values are irrelevant, whereas the shape of the current time series pattern may foretell the future. The handful of research efforts that consider this variant of the problem have met with limited success. In particular, it is now understood that most of these efforts allow the discovery of spurious rules. We believe the reason why rule discovery in real-valued time series has failed thus far is because most efforts have more or less indiscriminately applied the ideas of symbolic stream rule discovery to real-valued rule discovery. In this work, we show why these ideas are not directly suitable for rule discovery in time series. Beyond our novel definitions/representations, which allow for meaningful and extendable specifications of rules, we further show novel algorithms that allow us to quickly discover high quality rules in very large datasets that accurately predict the occurrence of future events.\",\n \"Title: Efficient Long-Term Degradation Profiling in Time Series for Complex Physical Systems\\nYear: 2015\\n\\nAbstract: The long term operation of physical systems inevitably leads to their wearing out, and may cause degradations in performance or the unexpected failure of the entire system. To reduce the possibility of such unanticipated failures, the system must be monitored for tell-tale symptoms of degradation that are suggestive of imminent failure. In this work, we introduce a novel time series analysis technique that allows the decomposition of the time series into trend and fluctuation components, providing the monitoring software with actionable information about the changes of the system's behavior over time. We analyze the underlying problem and formulate it to a Quadratic Programming (QP) problem that can be solved with existing QP-solvers. However, when the profiling resolution is high, as generally required by real-world applications, such a decomposition becomes intractable to general QP-solvers. To speed up the problem solving, we further transform the problem and present a novel QP formulation, Non-negative QP, for the problem and demonstrate a tractable solution that bypasses the use of slow general QP-solvers. We demonstrate our ideas on both synthetic and real datasets, showing that our method allows us to accurately extract the degradation phenomenon of time series. We further demonstrate the generality of our ideas by applying them beyond classic machine prognostics to problems in identifying the influence of news events on currency exchange rates and stock prices. We fully implement our profiling system and deploy it into several physical systems, such as chemical plants and nuclear power plants, and it greatly helps detect the degradation phenomenon, and diagnose the corresponding components.\",\n \"Title: Accelerating Dynamic Time Warping Clustering with a Novel Admissible Pruning Strategy\\nYear: 2015\\n\\nAbstract: Clustering time series is a useful operation in its own right, and an important subroutine in many higher-level data mining analyses, including data editing for classifiers, summarization, and outlier detection. While it has been noted that the general superiority of Dynamic Time Warping (DTW) over Euclidean Distance for similarity search diminishes as we consider ever larger datasets, as we shall show, the same is not true for clustering. Thus, clustering time series under DTW remains a computationally challenging task. In this work, we address this lethargy in two ways. We propose a novel pruning strategy that exploits both upper and lower bounds to prune off a large fraction of the expensive distance calculations. This pruning strategy is admissible; giving us provably identical results to the brute force algorithm, but is at least an order of magnitude faster. For datasets where even this level of speedup is inadequate, we show that we can use a simple heuristic to order the unavoidable calculations in a most-useful-first ordering, thus casting the clustering as an anytime algorithm. We demonstrate the utility of our ideas with both single and multidimensional case studies in the domains of astronomy, speech physiology, medicine and entomology.\",\n \"Title: Scalable Clustering of Time Series with U-Shapelets.\\nYear: 2015\\n\\nAbstract: A recently introduced primitive for time series data mining, unsupervised shapelets (u-shapelets), has demonstrated significant potential for time series clustering. In contrast to approaches that consider the entire time series to compute pairwise similarities, the u-shapelets technique allows considering only relevant subsequences of time series. Moreover, u-shapelets allow us to bypass the apparent chicken-and-egg paradox of defining relevant with reference to the clustering itself. U-shapelets have several advantages over rival methods. First, they are defined even when the time series are of different lengths; for example, they allow clustering datasets containing a mixture of single heartbeats and multi-beat ECG recordings. Second, u-shapelets mitigate sensitivity to irrelevant data such as noise, spikes, dropouts, etc. Finally, u-shapelets demonstrated ability to provide additional insights into the data. Unfortunately, the state-ofthe-art algorithms for u-shapelets search are intractable and so their advantages have only been demonstrated on tiny datasets. We propose a simple approach to speed up a ushapelet discovery by two orders of magnitude, without any significant loss in clustering quality.\",\n \"Title: On the Non-Trivial Generalization of Dynamic Time Warping to the Multi-Dimensional Case.\\nYear: 2015\\n\\nAbstract: In the last decade, Dynamic Time Warping (DTW) has emerged as the distance measure of choice for virtually all time series data mining applications. This is the result of significant progress in improving DTW’s efficiency, and multiple empirical studies showing that DTW-based classifiers at least equal the accuracy of all their rivals across dozens of datasets. Thus far, most of the research has considered only the one-dimensional case, with practitioners generalizing to the multi-dimensional case in one of two ways. In general, it appears the community believes either that the two ways are equivalent, or that the choice is irrelevant. In this work, we show that this is not the case. The two most commonly used multidimensional DTW methods can produce different classifications, and neither one dominates over the other. This seems to suggest that one should learn the best method for a particular application. However, we will show that this is not necessary; a simple, principled rule can be used on a case-by-case basis to predict which of the two methods we should give credence to. Our method allows us to ensure that classification results are at least as accurate as the better of the two rival methods, and in many cases, our method is strictly more accurate. We demonstrate our ideas with the most extensive set of multi-dimensional time series classification experiments ever attempted. Keywords—Dynamic Time Warping; Classification\",\n \"Title: Extracting Optimal Performance from Dynamic Time Warping.\\nYear: 2016\\n\\nAbstract: Dynamic Time Warping (DTW) is a distance measure that compares two time series after optimally aligning them. DTW is being used for decades in thousands of academic and industrial projects despite the very expensive computational complexity, O(n2). These applications include data mining, image processing, signal processing, robotics and computer graphics among many others. In spite of all this research effort, there are many myths and misunderstanding about DTW in the literature, for example \\\\\\\"it is too slow to be useful\\\\\\\" or \\\\\\\"the warping window size does not matter much.\\\\\\\" In this tutorial, we correct these misunderstandings and we summarize the research efforts in optimizing both the efficiency and effectiveness of both the basic DTW algorithm, and of the higher-level algorithms that exploit DTW such as similarity search, clustering and classification. We will discuss variants of DTW such as constrained DTW, multidimensional DTW and asynchronous DTW, and optimization techniques such as lower bounding, early abandoning, run-length encoding, bounded approximation and hardware optimization. We will discuss a multitude of application areas including physiological monitoring, social media mining, activity recognition and animal sound processing. The optimization techniques are generalizable to other domains on various data types and problems.\",\n \"Title: A General Framework for Density Based Time Series Clustering Exploiting a Novel Admissible Pruning Strategy.\\nYear: 2016\\n\\nAbstract: Time Series Clustering is an important subroutine in many higher-level data mining analyses, including data editing for classifiers, summarization, and outlier detection. It is well known that for similarity search the superiority of Dynamic Time Warping (DTW) over Euclidean distance gradually diminishes as we consider ever larger datasets. However, as we shall show, the same is not true for clustering. Clustering time series under DTW remains a computationally expensive operation. In this work, we address this issue in two ways. We propose a novel pruning strategy that exploits both the upper and lower bounds to prune off a very large fraction of the expensive distance calculations. This pruning strategy is admissible and gives us provably identical results to the brute force algorithm, but is at least an order of magnitude faster. For datasets where even this level of speedup is inadequate, we show that we can use a simple heuristic to order the unavoidable calculations in a most-useful-first ordering, thus casting the clustering into an anytime framework. We demonstrate the utility of our ideas with both single and multidimensional case studies in the domains of astronomy, speech physiology, medicine and entomology. In addition, we show the generality of our clustering framework to other domains by efficiently obtaining semantically significant clusters in protein sequences using the Edit Distance, the discrete data analogue of DTW.\",\n \"Title: Reliable early classification of time series based on discriminating the classes over time.\\nYear: 2017\\n\\nAbstract: The goal of early classification of time series is to predict the class value of a sequence early in time, when its full length is not yet available. This problem arises naturally in many contexts where the data is collected over time and the label predictions have to be made as soon as possible. In this work, a method based on probabilistic classifiers is proposed for the problem of early classification of time series. An important feature of this method is that, in its learning stage, it discovers the timestamps in which the prediction accuracy for each class begins to surpass a pre-defined threshold. This threshold is defined as a percentage of the accuracy that would be obtained if the full series were available, and it is defined by the user. The class predictions for new time series will only be made in these timestamps or later. Furthermore, when applying the model to a new time series, a class label will only be provided if the difference between the two largest predicted class probabilities is higher than or equal to a certain threshold, which is calculated in the training step. The proposal is validated on 45 benchmark time series databases and compared with several state-of-the-art methods, and obtains superior results in both earliness and accuracy. In addition, we show the practical applicability of our method for a real-world problem: the detection and identification of bird calls in a biodiversity survey scenario.\",\n \"Title: Generalizing DTW to the multi-dimensional case requires an adaptive approach.\\nYear: 2017\\n\\nAbstract: In recent years Dynamic Time Warping (DTW) has emerged as the distance measure of choice for virtually all time series data mining applications. For example, virtually all applications that process data from wearable devices use DTW as a core sub-routine. This is the result of significant progress in improving DTW's efficiency, together with multiple empirical studies showing that DTW-based classifiers at least equal (and generally surpass) the accuracy of all their rivals across dozens of datasets. Thus far, most of the research has considered only the one-dimensional case, with practitioners generalizing to the multi-dimensional case in one of two ways, dependent or independent warping. In general, it appears the community believes either that the two ways are equivalent, or that the choice is irrelevant. In this work, we show that this is not the case. The two most commonly used multi-dimensional DTW methods can produce different classifications, and neither one dominates over the other. This seems to suggest that one should learn the best method for a particular application. However, we will show that this is not necessary; a simple, principled rule can be used on a case-by-case basis to predict which of the two methods we should trust at the time of classification. Our method allows us to ensure that classification results are at least as accurate as the better of the two rival methods, and, in many cases, our method is significantly more accurate. We demonstrate our ideas with the most extensive set of multi-dimensional time series classification experiments ever attempted.\",\n \"Title: Matrix Profile II: Exploiting a Novel Algorithm and GPUs to Break the One Hundred Million Barrier for Time Series Motifs and Joins\\nYear: 2016\\n\\nAbstract: Time series motifs have been in the literature for about fifteen years, but have only recently begun to receive significant attention in the research community. This is perhaps due to the growing realization that they implicitly offer solutions to a host of time series problems, including rule discovery, anomaly detection, density estimation, semantic segmentation, etc. Recent work has improved the scalability to the point where exact motifs can be computed on datasets with up to a million data points in tenable time. However, in some domains, for example seismology, there is an insatiable need to address even larger datasets. In this work we show that a combination of a novel algorithm and a high-performance GPU allows us to significantly improve the scalability of motif discovery. We demonstrate the scalability of our ideas by finding the full set of exact motifs on a dataset with one hundred million subsequences, by far the largest dataset ever mined for time series motifs. Furthermore, we demonstrate that our algorithm can produce actionable insights in seismology and other domains.\",\n \"Title: Matrix Profile I: All Pairs Similarity Joins For Time Series: A Unifying View That Includes Motifs, Discords And Shapelets\\nYear: 2016\\n\\nAbstract: The all-pairs-similarity-search (or similarity join) problem has been extensively studied for text and a handful of other datatypes. However, surprisingly little progress has been made on similarity joins for time series subsequences. The lack of progress probably stems from the daunting nature of the problem. For even modest sized datasets the obvious nested-loop algorithm can take months, and the typical speed-up techniques in this domain (i.e., indexing, lower-bounding, triangular-inequality pruning and early abandoning) at best produce one or two orders of magnitude speedup. In this work we introduce a novel scalable algorithm for time series subsequence all-pairs-similarity-search. For exceptionally large datasets, the algorithm can be trivially cast as an anytime algorithm and produce high-quality approximate solutions in reasonable time. The exact similarity join algorithm computes the answer to the time series motif and time series discord problem as a side-effect, and our algorithm incidentally provides the fastest known algorithm for both these extensively-studied problems. We demonstrate the utility of our ideas for two time series data mining problems, including motif discovery and novelty discovery.\",\n \"Title: Matrix Profile III: The Matrix Profile Allows Visualization of Salient Subsequences in Massive Time Series\\nYear: 2016\\n\\nAbstract: Multidimensional Scaling (MDS) is one of the most versatile tools used for exploratory data mining. It allows a first glimpse of possible structure in the data, which can inform the choice of analyses used. Its uses are multiple. It can give the user an idea as to the cluster ability or linear separability of the data. It can help spot outliers, or can hint at the intrinsic dimensionality of the data. Moreover, it can sometimes reveal unexpected latent dimensions in the data. With all these uses, MDS is increasingly used in areas as diverse as marketing, medicine, genetics, music and linguistics. One of the strengths of MDS is that it is essentially agnostic to data type, as we can use any distance measure to create the distance matrix, which is the only required input to the MDS algorithm. In spite of this generality, we make the following claim. MDS is not (well) defined for an increasingly important data type, time series subsequences. In this work we explain why this is the case, and we propose a scalable solution. We demonstrate the utility of our ideas on several diverse real-world datasets. At the core of our approach is a novel Minimum Description Length (MDL) subsequence extraction algorithm. Beyond MDS visualization, this subsequence extraction subroutine may be a useful tool in its own right.\",\n \"Title: The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances.\\nYear: 2017\\n\\nAbstract: In the last 5 years there have been a large number of new time series classification algorithms proposed in the literature. These algorithms have been evaluated on subsets of the 47 data sets in the University of California, Riverside time series classification archive. The archive has recently been expanded to 85 data sets, over half of which have been donated by researchers at the University of East Anglia. Aspects of previous evaluations have made comparisons between algorithms difficult. For example, several different programming languages have been used, experiments involved a single train/test split and some used normalised data whilst others did not. The relaunch of the archive provides a timely opportunity to thoroughly evaluate algorithms on a larger number of datasets. We have implemented 18 recently proposed algorithms in a common Java framework and compared them against two standard benchmark classifiers (and each other) by performing 100 resampling experiments on each of the 85 datasets. We use these results to test several hypotheses relating to whether the algorithms are significantly more accurate than the benchmarks and each other. Our results indicate that only nine of these algorithms are significantly more accurate than both benchmarks and that one classifier, the collective of transformation ensembles, is significantly more accurate than all of the others. All of our experiments and results are reproducible: we release all of our code, results and experimental details and we hope these experiments form the basis for more robust testing of new algorithms in the future.\",\n \"Title: Searching Time Series with Invariance to Large Amounts of Uniform Scaling\\nYear: 2017\\n\\nAbstract: Similarity search is arguably the most important primitive in time series data mining. Recent research has made significant progress on fast algorithms for time series similarity search under Dynamic Time Warping (DTW) and Uniform Scaling (US) distance measures. However, the current state-of-the-art algorithms cannot support greater amounts of rescaling in many practical applications. In this paper, we introduce a novel lower bound, LB\\n<sub xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">new</sub>\\n, to allow efficient search even in domains that exhibit more than a factor-of-two variability in scale. The effectiveness of our idea is validated on various large-scale real datasets from commercial important domains.\",\n \"Title: Query Suggestion to allow Intuitive Interactive Search in Multidimensional Time Series.\\nYear: 2017\\n\\nAbstract: In recent years, the research community, inspired by its success in dealing with single-dimensional time series, has turned its attention to dealing with multidimensional time series. There are now a plethora of techniques for indexing, classification, and clustering of multidimensional time series. However, we argue that the difficulty of exploratory search in large multidimensional time series remains underappreciated. In essence, the problem reduces to the \\\\\\\"chicken-and-egg\\\\\\\" paradox that it is difficult to produce a meaningful query without knowing the best subset of dimensions to use, but finding the best subset of dimensions is itself query dependent. In this work we propose a solution to this problem. We introduce an algorithm that runs in the background, observing the user's search interactions. When appropriate, our algorithm suggests to the user a dimension that could be added or deleted to improve the user's satisfaction with the query. These query dependent suggestions may be useful to the user, even if she does not act on them (by reissuing the query), as they can hint at unexpected relationships or redundancies between the dimensions of the data. We evaluate our algorithm on several real-world datasets in medical, human activity, and industrial domains, showing that it produces subjectively sensible and objectively superior results.\",\n \"Title: Matrix profile IV: using weakly labeled time series to predict outcomes\\nYear: 2017\\n\\nAbstract: AbstractIn academic settings over the last decade, there has been significant progress in time series classification. However, much of this work makes assumptions that are simply unrealistic for deployed industrial applications. Examples of these unrealistic assumptions include the following: assuming that data subsequences have a single fixed-length, are precisely extracted from the data, and are correctly labeled according to their membership in a set of equal-size classes. In real-world industrial settings, these patterns can be of different lengths, the class annotations may only belong to a general region of the data, may contain errors, and finally, the class distribution is typically highly skewed. Can we learn from such weakly labeled data? In this work, we introduce SDTS, a scalable algorithm that can learn in such challenging settings. We demonstrate the utility of our ideas by learning from diverse datasets with millions of datapoints. As we shall demonstrate, our domain-agnostic parameter-free algorithm can be competitive with domain-specific algorithms used in neuroscience and entomology, even when those algorithms have been tuned by domain experts to incorporate domain knowledge.\",\n \"Title: Matrix Profile V: A Generic Technique to Incorporate Domain Knowledge into Motif Discovery.\\nYear: 2017\\n\\nAbstract: Time series motif discovery has emerged as perhaps the most used primitive for time series data mining, and has seen applications to domains as diverse as robotics, medicine and climatology. There has been recent significant progress on the scalability of motif discovery. However, we believe that the current definitions of motif discovery are limited, and can create a mismatch between the user's intent/expectations, and the motif discovery search outcomes. In this work, we explain the reasons behind these issues, and introduce a novel and general framework to address them. Our ideas can be used with current state-of-the-art algorithms with virtually no time or space overhead, and are fast enough to allow real-time interaction and hypotheses testing on massive datasets. We demonstrate the utility of our ideas on domains as diverse as seismology and epileptic seizure monitoring.\",\n \"Title: Matrix Profile VII: Time Series Chains: A New Primitive for Time Series Data Mining (Best Student Paper Award).\\nYear: 2017\\n\\nAbstract: Since their introduction over a decade ago, time series motifs have become a fundamental tool for time series analytics, finding diverse uses in dozens of domains. In this work we introduce Time Series Chains, which are related to, but distinct from, time series motifs. Informally, time series chains are a temporally ordered set of subsequence patterns, such that each pattern is similar to the pattern that preceded it, but the first and last patterns are arbitrarily dissimilar. In the discrete space, this is similar to extracting the text chain hit, hot, dot, dog from a paragraph. The first and last words have nothing in common, yet they are connected by a chain of words with a small mutual difference. Time series chains can capture the evolution of systems, and help predict the future. As such, they potentially have implications for prognostics. In this work, we introduce a robust definition of time series chains, and a scalable algorithm that allows us to discover them in massive datasets.\",\n \"Title: Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels\\nYear: 2017\\n\\nAbstract: Unsupervised semantic segmentation in the time series domain is a much-studied problem due to its potential to detect unexpected regularities and regimes in poorly understood data. However, the current techniques have several shortcomings, which have limited the adoption of time series semantic segmentation beyond academic settings for three primary reasons. First, most methods require setting/learning many parameters and thus may have problems generalizing to novel situations. Second, most methods implicitly assume that all the data is segmentable, and have difficulty when that assumption is unwarranted. Finally, most research efforts have been confined to the batch case, but online segmentation is clearly more useful and actionable. To address these issues, we present an algorithm which is domain agnostic, has only one easily determined parameter, and can handle data streaming at a high rate. In this context, we test our algorithm on the largest and most diverse collection of time series datasets ever considered, and demonstrate our algorithm's superiority over current solutions. Furthermore, we are the first to show that semantic segmentation may be possible at superhuman performance levels.\",\n \"Title: Judicious Setting Of Dynamic Time Warping'S Window Width Allows More Accurate Classification Of Time Series\\nYear: 2017\\n\\nAbstract: While the Dynamic Time Warping (DTW) -based Nearest-Neighbor Classification algorithm is regarded as a strong baseline for time series classification, in recent years there has been a plethora of algorithms that have claimed to be able to improve upon its accuracy in the general case. Many of these proposed ideas sacrifice the simplicity of implementation that DTW-based classifiers offer for rather modest gains. Nevertheless, there are clearly times when even a small improvement could make a large difference in an important medical or financial domain. In this work, we make an unexpected claim; an underappreciated \\\"low hanging fruit\\\" in optimizing DTW's performance can produce improvements that make it an even stronger baseline, closing most or all the improvement gap of the more sophisticated methods. We show that the method currently used to learn DTW's only parameter, the maximum amount of warping allowed, is likely to give the wrong answer for small training sets. We introduce a simple method to mitigate the small training set issue by creating synthetic exemplars to help learn the parameter. We evaluate our ideas on the UCR Time Series Archive and a case study in fall classification, and demonstrate that our algorithm produces significant improvement in classification accuracy.\",\n \"Title: Exploiting a novel algorithm and GPUs to break the ten quadrillion pairwise comparisons barrier for time series motifs and joins.\\nYear: 2018\\n\\nAbstract: Time series motifs are approximately repeated subsequences found within a longer time series. They have been in the literature since 2002, but recently they have begun to receive significant attention in research and industrial communities. This is perhaps due to the growing realization that they implicitly offer solutions to a host of time series problems, including rule discovery, anomaly detection, density estimation, semantic segmentation, summarization, etc. Recent work has improved the scalability so exact motifs can be computed on datasets with up to a million data points in tenable time. However, in some domains, for example seismology or climatology, there is an immediate need to address even larger datasets. In this work, we demonstrate that a combination of a novel algorithm and a high-performance GPU allows us to significantly improve the scalability of motif discovery. We demonstrate the scalability of our ideas by finding the full set of exact motifs on a dataset with one hundred and forty-three million subsequences, which is by far the largest dataset ever mined for time series motifs/joins; it requires ten quadrillion pairwise comparisons. Furthermore, we demonstrate that our algorithm can produce actionable insights into seismology and ethology.\",\n \"Title: Time series joins, motifs, discords and shapelets: a unifying view that exploits the matrix profile.\\nYear: 2018\\n\\nAbstract: The last decade has seen a flurry of research on (or ) for text, DNA and a handful of other datatypes, and these systems have been applied to many diverse data mining problems. However, there has been surprisingly little progress made on similarity joins for . The lack of progress probably stems from the daunting nature of the problem. For even modest sized datasets the obvious nested-loop algorithm can take months, and the typical speed-up techniques in this domain (i.e., indexing, lower-bounding, triangular-inequality pruning and early abandoning) at best produce only one or two orders of magnitude speedup. In this work we introduce a novel scalable algorithm for time series subsequence all-pairs-similarity-search. For exceptionally large datasets, the algorithm can be trivially cast as an anytime algorithm and produce high-quality approximate solutions in reasonable time and/or be accelerated by a trivial porting to a GPU framework. The exact similarity join algorithm computes the answer to the and problem as a side-effect, and our algorithm incidentally provides the fastest known algorithm for both these extensively-studied problems. We demonstrate the utility of our ideas for many time series data mining problems, including motif discovery, novelty discovery, shapelet discovery, semantic segmentation, density estimation, and contrast set mining. Moreover, we demonstrate the utility of our ideas on domains as diverse as seismology, music processing, bioinformatics, human activity monitoring, electrical power-demand monitoring and medicine.\",\n \"Title: Admissible Time Series Motif Discovery with Missing Data.\\nYear: 2018\\n\\nAbstract: The discovery of time series motifs has emerged as one of the most useful primitives in time series data mining. Researchers have shown its utility for exploratory data mining, summarization, visualization, segmentation, classification, clustering, and rule discovery. Although there has been more than a decade of extensive research, there is still no technique to allow the discovery of time series motifs in the presence of missing data, despite the well-documented ubiquity of missing data in scientific, industrial, and medical datasets. In this work, we introduce a technique for motif discovery in the presence of missing data. We formally prove that our method is admissible, producing no false negatives. We also show that our method can piggy-back off the fastest known motif discovery method with a small constant factor time/space overhead. We will demonstrate our approach on diverse datasets with varying amounts of missing data\",\n \"Title: VALMOD: A Suite for Easy and Exact Detection of Variable Length Motifs in Data Series.\\nYear: 2018\\n\\nAbstract: Data series motif discovery represents one of the most useful primitives for data series mining, with applications to many domains, such as robotics, entomology, seismology, medicine, and climatology, and others. The state-of-the-art motif discovery tools still require the user to provide the motif length. Yet, in several cases, the choice of motif length is critical for their detection. Unfortunately, the obvious brute-force solution, which tests all lengths within a given range, is computationally untenable, and does not provide any support for ranking motifs at different resolutions (i.e., lengths). We demonstrate VALMOD, our scalable motif discovery algorithm that efficiently finds all motifs in a given range of lengths, and outputs a length-invariant ranking of motifs. Furthermore, we support the analysis process by means of a newly proposed meta-data structure that helps the user to select the most promising pattern length. This demo aims at illustrating in detail the steps of the proposed approach, showcasing how our algorithm and corresponding graphical insights enable users to efficiently identify the correct motifs.\",\n \"Title: Matrix Profile X: VALMOD - Scalable Discovery of Variable-Length Motifs in Data Series.\\nYear: 2018\\n\\nAbstract: In the last fifteen years, data series motif discovery has emerged as one of the most useful primitives for data series mining, with applications to many domains, including robotics, entomology, seismology, medicine, and climatology. Nevertheless, the state-of-the-art motif discovery tools still require the user to provide the motif length. Yet, in at least some cases, the choice of motif length is critical and unforgiving. Unfortunately, the obvious brute-force solution, which tests all lengths within a given range, is computationally untenable. In this work, we introduce VALMOD, an exact and scalable motif discovery algorithm that efficiently finds all motifs in a given range of lengths. We evaluate our approach with five diverse real datasets, and demonstrate that it is up to 20 times faster than the state-of-the-art. Our results also show that removing the unrealistic assumption that the user knows the correct length, can often produce more intuitive and actionable results, which could have been missed otherwise.\",\n \"Title: Optimizing dynamic time warping's window width for time series data mining applications.\\nYear: 2018\\n\\nAbstract: Dynamic Time Warping (DTW) is a highly competitive distance measure for most time series data mining problems. Obtaining the best performance from DTW requires setting its only parameter, the maximum amount of warping (). In the supervised case with ample data, is typically set by cross-validation in the training stage. However, this method is likely to yield suboptimal results for small training sets. For the unsupervised case, learning via cross-validation is not possible because we do not have access to labeled data. Many practitioners have thus resorted to assuming that “”, and they use the largest value of permitted by the computational resources. However, as we will show, in most circumstances, this is a naïve approach that produces inferior clusterings. Moreover, the best warping window width is generally non-transferable between the two tasks, i.e., for a single dataset, practitioners cannot simply apply the best learned for classification on clustering or vice versa. In addition, we will demonstrate that the appropriate amount of warping not only depends on the data structure, but also on the dataset . Thus, even if a practitioner knows the best setting for a given dataset, they will likely be at a lost if they apply that setting on a bigger size version of that data. All these issues seem largely unknown or at least unappreciated in the community. In this work, we demonstrate the importance of setting DTW’s warping window width correctly, and we also propose novel methods to learn this parameter in both supervised and unsupervised settings. The algorithms we propose to learn can produce significant improvements in classification accuracy and clustering quality. We demonstrate the correctness of our novel observations and the utility of our ideas by testing them with more than one hundred publicly available datasets. Our forceful results allow us to make a perhaps unexpected claim; an underappreciated “low hanging fruit” in optimizing DTW’s performance can produce improvements that make it an even stronger baseline, closing most or all the improvement gap of the more sophisticated methods proposed in recent years.\",\n \"Title: The UCR Time Series Archive.\\nYear: 2018\\n\\nAbstract: The UCR Time Series Archive - introduced in 2002, has become an important resource in the time series data mining community, with at least one thousand published papers making use of at least one dataset from the archive. The original incarnation of the archive had sixteen datasets but since that time, it has gone through periodic expansions. The last expansion took place in the summer of 2015 when the archive grew from 45 datasets to 85 datasets. This paper introduces and will focus on the new data expansion from 85 to 128 datasets. Beyond expanding this valuable resource, this paper offers pragmatic advice to anyone who may wish to evaluate a new algorithm on the archive. Finally, this paper makes a novel and yet actionable claim: of the hundreds of papers that show an improvement over the standard baseline (1-Nearest Neighbor classification), a large fraction may be misattributing the reasons for their improvement. Moreover, they may have been able to achieve the same improvement with a much simpler modification, requiring just a single line of code.\",\n \"Title: Representation Learning by Reconstructing Neighborhoods.\\nYear: 2018\\n\\nAbstract: Since its introduction, unsupervised representation learning has attracted a lot of attention from the research community, as it is demonstrated to be highly effective and easy-to-apply in tasks such as dimension reduction, clustering, visualization, information retrieval, and semi-supervised learning. In this work, we propose a novel unsupervised representation learning framework called neighbor-encoder, in which domain knowledge can be easily incorporated into the learning process without modifying the general encoder-decoder architecture of the classic autoencoder.In contrast to autoencoder, which reconstructs the input data itself, neighbor-encoder reconstructs the input datau0027s neighbors. As the proposed representation learning problem is essentially a neighbor reconstruction problem, domain knowledge can be easily incorporated in the form of an appropriate definition of similarity between objects. Based on that observation, our framework can leverage any off-the-shelf similarity search algorithms or side information to find the neighbor of an input object. Applications of other algorithms (e.g., association rule mining) in our framework are also possible, given that the appropriate definition of neighbor can vary in different contexts. We have demonstrated the effectiveness of our framework in many diverse domains, including images, text, and time series, and for various data mining tasks including classification, clustering, and visualization. Experimental results show that neighbor-encoder not only outperforms autoencoder in most of the scenarios we consider, but also achieves the state-of-the-art performance on text document clustering.\",\n \"Title: The UEA multivariate time series classification archive, 2018.\\nYear: 2018\\n\\nAbstract: In 2002, the UCR time series classification archive was first released with sixteen datasets. It gradually expanded, until 2015 when it increased in size from 45 datasets to 85 datasets. In October 2018 more datasets were added, bringing the total to 128. The new archive contains a wide range of problems, including variable length series, but it still only contains univariate time series classification problems. One of the motivations for introducing the archive was to encourage researchers to perform a more rigorous evaluation of newly proposed time series classification (TSC) algorithms. It has worked: most recent research into TSC uses all 85 datasets to evaluate algorithmic advances. Research into multivariate time series classification, where more than one series are associated with each class label, is in a position where univariate TSC research was a decade ago. Algorithms are evaluated using very few datasets and claims of improvement are not based on statistical comparisons. We aim to address this problem by forming the first iteration of the MTSC archive, to be hosted at the website www.timeseriesclassification.com. Like the univariate archive, this formulation was a collaborative effort between researchers at the University of East Anglia (UEA) and the University of California, Riverside (UCR). The 2018 vintage consists of 30 datasets with a wide range of cases, dimensions and series lengths. For this first iteration of the archive we format all data to be of equal length, include no series with missing data and provide train/test splits.\",\n \"Title: Time Series Classification To Improve Poultry Welfare\\nYear: 2018\\n\\nAbstract: Poultry farms are an important contributor to the human food chain. Worldwide, humankind keeps an enormous number of domesticated birds (e.g. chickens) for their eggs and their meat, providing rich sources of low-fat protein. However, around the world, there have been growing concerns about the quality of life for the livestock in poultry farms; and increasingly vocal demands for improved standards of animal welfare. Recent advances in sensing technologies and machine learning allow the possibility of automatically assessing the health of some individual birds, and employing the lessons learned to improve the welfare for all birds. This task superficially appears to be easy, given the dramatic progress in recent years in classifying human behaviors, and given that human behaviors are presumably more complex. However, as we shall demonstrate, classifying chicken behaviors poses several unique challenges, chief among which is creating a generalizable \\\"dictionary\\\" of behaviors from sparse and noisy data. In this work we introduce a novel time series dictionary learning algorithm that can robustly learn from weakly labeled data sources.\",\n \"Title: Matrix Profile XIII: Time Series Snippets: A New Primitive for Time Series Data Mining\\nYear: 2018\\n\\nAbstract: Perhaps the most basic query made by a data analyst confronting a new data source is \\\"Show me some representative/typical data.\\\" Answering this question is trivial in many domains, but surprisingly, it is very difficult in large time series datasets. The major difficulty is not time or space complexity, but defining what it means to be representative data in this domain. In this work, we show that the obvious candidate definitions: motifs, shapelets, cluster centers, random samples etc., are all poor choices. Thus motivated, we introduce time series snippets, a novel representation of typical time series subsequences. Beyond their utility for visualizing and summarizing massive time series collections, we show that time series snippets have utility for high-level comparison of large time series collections.\",\n \"Title: Prefix And Suffix Invariant Dynamic Time Warping\\nYear: 2016\\n\\nAbstract: While there exist a plethora of classification algorithms for most data types, there is an increasing acceptance that the unique properties of time series mean that the combination of nearest neighbor classifiers and Dynamic Time Warping (DTW) is very competitive across a host of domains, from medicine to astronomy to environmental sensors. While there has been significant progress in improving the efficiency and effectiveness of DTW in recent years, in this work we demonstrate that an underappreciated issue can significantly degrade the accuracy of DTW in real-world deployments. This issue has probably escaped the attention of the very active time series research community because of its reliance on static highly contrived benchmark datasets, rather than real world dynamic datasets where the problem tends to manifest itself. In essence, the issue is that DTW's eponymous invariance to warping is only true for the main \\\"body\\\" of the two time series being compared. However, for the \\\"head\\\" and \\\"tail\\\" of the time series, the DTW algorithm affords no warping invariance. The effect of this is that tiny differences at the beginning or end of the time series (which may be either consequential or simply the result of poor \\\"cropping\\\") will tend to contribute disproportionally to the estimated similarity, producing incorrect classifications. In this work, we show that this effect is real, and reduces the performance of the algorithm. We further show that we can fix the issue with a subtle redesign of the DTW algorithm, and that we can learn an appropriate setting for the extra parameter we introduced. We further demonstrate that our generalization is amiable to all the optimizations that make DTW tractable for large datasets.\",\n \"Title: Speeding up similarity search under dynamic time warping by pruning unpromising alignments.\\nYear: 2018\\n\\nAbstract: Similarity search is the core procedure for several time series mining tasks. While different distance measures can be used for this purpose, there is clear evidence that the Dynamic Time Warping (DTW) is the most suitable distance function for a wide range of application domains. Despite its quadratic complexity, research efforts have proposed a significant number of pruning methods to speed up the similarity search under DTW. However, the search may still take a considerable amount of time depending on the parameters of the search, such as the length of the query and the warping window width. The main reason is that the current techniques for speeding up the similarity search focus on avoiding the costly distance calculation between as many pairs of time series as possible. Nevertheless, the few pairs of subsequences that were not discarded by the pruning techniques can represent a significant part of the entire search time. In this work, we adapt a recently proposed algorithm to improve the internal efficiency of the DTW calculation. Our method can speed up the UCR suite, considered the current fastest tool for similarity search under DTW. More important, the longer the time needed for the search, the higher the speedup ratio achieved by our method. We demonstrate that our method performs similarly to UCR suite for small queries and narrow warping constraints. However, it performs up to five times faster for long queries and large warping windows.\",\n \"Title: Generalized Dynamic Time Warping: Unleashing the Warping Power Hidden in Point-Wise Distances\\nYear: 2018\\n\\nAbstract: Domain-specific distances preferred by analysts for exploring similarities among time series tend to be \\\"point-to-point\\\" distances. Unfortunately, this point-wise nature limits their ability to perform meaningful comparisons between sequences of different lengths and with temporal mis-alignments. Analysts instead need \\\"elastic\\\" alignment tools such as Dynamic Time Warping (DTW) to perform such flexible comparisons. However, the existing alignment tools are limited in that they do not incorporate diverse distances. To address this shortcoming, our work introduces the first conceptual framework called Generalized Dynamic Time Warping (GDTW) that supports now alignment (warping) of a large array of domain-specific distances in a uniform manner. While the classic DTW and its prior extensions focus on the Euclidean Distance, our GDTW is the first method that generalizes the ubiquitous DTW and \\\"extends\\\" its warping capabilities to a rich diversity of point-to-point distances. Based on our GDTW paradigm that preserves the efficiency of the dynamic programming paradigm of DTW, we design an abstraction that implemented by our GDTW Design Tool enables analysts to \\\"warp\\\" new distances with little programming effort. Through extensive evaluation studies on 85 real public domain benchmark datasets, we show that our newly warped distances offer higher classification accuracy than the previously available distances for the majority of these datasets. Further, our case study on heart arrhythmia data illustrates the utility of the new distances enabled by our GDTW warping methodology.\",\n \"Title: Time Series Classification: Lessons Learned In The (Literal) Field While Studying Chicken Behavior\\nYear: 2019\\n\\nAbstract: Poultry farms are a major contributor to the human food chain. However, around the world, there have been growing concerns about the quality of life for the livestock in poultry farms; and increasingly vocal demands for improved standards of animal welfare. Recent advances in sensing technologies and machine learning allow the possibility of monitoring birds, and employing the lessons learned to improve the welfare for all birds. This task superficially appears to be easy, yet, studying behavioral patterns involves collecting enormous amounts of data, justifying the term Big Data. Before the big data can be used for analytical purposes to tease out meaningful, well-conserved behavioral patterns, the collected data needs to be preprocessed. The pre-processing refers to processes for cleansing and preparing data so that it is in the format ready to be analyzed by downstream algorithms, such as classification and clustering algorithms. However, as we shall demonstrate, efficient preprocessing of chicken big data is both non-trivial and crucial towards success of further analytics.\",\n \"Title: Fastdtw Is Approximate And Generally Slower Than The Algorithm It Approximates (Extended Abstract)\\nYear: 2021\\n\\nAbstract: Many time series data mining problems can be solved with repeated use of distance measure. Examples of such tasks include similarity search, clustering, classification, anomaly detection and segmentation. For over two decades it has been known that the Dynamic Time Warping (DTW) distance measure is the best measure to use for most tasks, in most domains. Because the classic DTW algorithm has quadratic time complexity, many ideas have been introduced to reduce its amortized time, or to quickly approximate it. One of the most cited approximate approaches is FastDTW. The FastDTW algorithm has well over a thousand citations and has been explicitly used in several hundred research efforts. In this work, we make a surprising claim. In any realistic data mining application, the approximate FastDTW is much slower than the exact DTW. This fact clearly has implications for the community that uses this algorithm: allowing it to address much larger datasets, get exact results, and do so in less time.\",\n \"Title: Shape-Based Telemetry Approach for Distracted Driving Behavior Detection\\nYear: 2021\\n\\nAbstract: Nearly nine in ten U.S. drivers exhibit distracted driving behavior, and such abnormal driving behavior is the leading cause of many fatal crashes. The prior art in distracted driving behavior detection focused on an ego vehicle sensing its driver. However, such approaches do not address the situation when other nearby vehicles behave abnormally (e.g., a distracted driver follows the ego vehicle). Modern cars have in-vehicle sensors to observe the lane-level driving behavior of others (e.g., preceding/following vehicles). We hypothesize that time series analysis of such data could reveal the nearby distracted drivers. In this work, we tackle this issue and analyze the lane-level driving behavior with Contrast Profile, a recently developed time-series tool, to detect distracted drivers. The Contrast Profile can examine archived driving data and infer the shape of the most discriminative driving behavior while and before being distracted. We then use such shape-based knowledge to detect distracted driving behavior in real time. Our preliminary results suggest that the proposed shape-based approach can identify the distracted driving behavior with about 90% accuracy under different settings.\",\n \"Title: Matrix Profile XXIII: Contrast Profile: A Novel Time Series Primitive that Allows Real World Classification\\nYear: 2021\\n\\nAbstract: Time series data remains a perennially important datatype considered in data mining. In the last decade there has been an increasing realization that time series data can best understood by reasoning about time series subsequences on the basis of their similarity to other subsequences: the two most familiar such time series concepts being motifs and discords. Time series motifs refer to two particularly close subsequences, whereas time series discords indicate subsequences that are far from their nearest neighbors. However, we argue that it can sometimes be useful to simultaneously reason about a subsequence's closeness to certain data and its distance to other data. In this work we introduce a novel primitive called the Contrast Profile that allows us to efficiently compute such a definition in a principled way. As we will show, the Contrast Profile has many downstream uses, including anomaly detection, data exploration, and preprocessing unstructured data for classification.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":102,"string":"102"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13584,"string":"13,584"}}},{"rowIdx":13585,"cells":{"question_id":{"kind":"string","value":"Chang Hun Kim"},"question":{"kind":"string","value":"Generate an abstract for the title \"Discrete differential error metric for surface simplification\" using the following items: 1. surface simplification\n2. discrete differential error metric\n3. geometric error\n4. local differential geometry\n5. preservation of sharp feature regions"},"answer":{"kind":"string","value":"In this paper we propose a new discrete differential error metric for surface simplification. Many surface simplification algorithms have been developed in order to produce rapidly high quality approximations of polygonal models, and the quadric error metric based on the distance error is the most popular and successful error metric so far Even though such distance based error metrics give visually pleasing results with a reasonably fast speed, it is hard to measure an accurate geometric error on a highly curved and thin region since the error measured by the distance metric on such a region is usually small and causes a loss of visually important features. To overcome such a drawback, we define a new error metric based on the theory of local differential geometry in such a way that the first and the second order discrete differentials approximated locally on a discrete polygonal surface are integrated into the usual distance error metric. The benefits of our error metric are preservation of sharp feature regions after a drastic simplification, small geometric errors, and fast computation comparable to the existing methods."},"correct_answer":{"kind":"string","value":"In this paper we propose a new discrete differential error metric for surface simplification. Many surface simplification algorithms have been developed in order to produce rapidly high quality approximations of polygonal models, and the quadric error metric based on the distance error is the most popular and successful error metric so far Even though such distance based error metrics give visually pleasing results with a reasonably fast speed, it is hard to measure an accurate geometric error on a highly curved and thin region since the error measured by the distance metric on such a region is usually small and causes a loss of visually important features. To overcome such a drawback, we define a new error metric based on the theory of local differential geometry in such a way that the first and the second order discrete differentials approximated locally on a discrete polygonal surface are integrated into the usual distance error metric. The benefits of our error metric are preservation of sharp feature regions after a drastic simplification, small geometric errors, and fast computation comparable to the existing methods."},"documents":{"kind":"list like","value":["Title: Discontinuous fluids\nYear: 2005\n\nAbstract: At interfaces between different fluids, properties such as density, viscosity, and molecular cohesion are discontinuous. To animate small-scale details of incompressible viscous multi-phase fluids realistically, we focus on the discontinuities in the state variables that express these properties. Surface tension of both free and bubble surfaces is modeled using the jump condition in the pressure field; and discontinuities in the velocity gradient field. driven by viscosity differences, are also considered. To obtain derivatives of the pressure and velocity fields with sub-grid accuracy, they are extrapolated across interfaces using continuous variables based on physical properties. The numerical methods that we present are easy to implement and do not impact the performance of existing solvers. Small-scale fluid motions, such as capillary instability, breakup of liquid sheets, and bubbly water can all be successfully animated.","Title: Bubbles alive\nYear: 2008\n\nAbstract: We propose a hybrid method for simulating multiphase fluids such as bubbly water. The appearance of subgrid visual details is improved by incorporating a new bubble model based on smoothed particle hydrodynamics (SPH) into an Eulerian grid-based simulation that handles background flows of large bodies of water and air. To overcome the difficulty in simulating small bubbles in the context of the multiphase flows on a coarse grid, we heuristically model the interphase properties of water and air by means of the interactions between bubble particles. As a result, we can animate lively motion of bubbly water with small scale details efficiently.","Title: Progressive reconstruction of 3D objects from a single free-hand line drawing\nYear: 2003\n\nAbstract: This paper presents a progressive algorithm by which we can narrow down the searching domain for face identification and reconstruct various 3D objects from a single freehand line sketch drawing in short time. The sketch drawings served as an input for the reconstruction process in this paper are obtained from an inaccurate freehand sketch of a 3D wireframe object, which is the edge–vertex graph without any removal of hidden lines. The algorithm is executed through two stages. First, at the face identification stage, every possible faces are generated from the sketch and they are classified into implausible faces, basis faces and minimal faces by means of the geometrical and topological constraints in order to reduce the searching space. The proposed algorithm searches minimal space of the faces so that the actual faces of the given object could be identified in short time. Second, at the object reconstruction stage, we calculate a 3D structure progressively by optimizing the vertices coordinates of the object in accordance with the sketching order of faces. This progressive method reconstructs the most plausible 3D object in a short time by applying the 3D constraints that are derived from the relationship between an object and its sketch drawing in the optimization process. In addition, this method makes it possible for designers to change their viewpoints while sketching. In this paper, we are describing the progressive reconstruction algorithm, and giving the results of the implementation.","Title: Divergence-constrained moving least squares for fluid simulation\nYear: 2008\n\nAbstract: Developing suitable interpolation methods to simulate dynamic motions of continuous materials such as fluids is an important problem. In this paper, we propose a novel method to enforce the divergence condition to the interpolated velocity field by moving least squares (MLS), by means of the diffusive derivatives and moving divergence constraints that allow the practical use and easy implementation. As results, we present the velocity interpolation examples and a fluid-like particle simulation method to show the meaningful potential of our method as a tool of physical interpolation for fluid simulations. Copyright © 2008 John Wiley & Sons, Ltd.","Title: User-Guided 3d su-muk painting\nYear: 2005\n\nAbstract: We present a technique for rendering animated 3D models in a Su-Muk painting style with user's guide. First, a user can sketch directly over 3D models by varying ink and water values. And we simulate the behavior of ink-water directly on 3D model for consistent shade information over 3D models. After simulation, ink and water are spread over a 3D surface appropriately. Second, to achieve a real hand-crafted look of Su-Muk painting, ink-water behavior is simulated again on a 2D screen by using overall ink-water shade information from previous output of 3D ink-water simulation. We demonstrate some images for Su-Muk painting with our user-guided drawing system.","Title: Progressive Compression of Geometry Information with Smooth Intermediate Meshes\nYear: 2007\n\nAbstract: We present a new geometry compression algorithm for manifold 3D meshes based on octree coding. For a given mesh, regular volume grids are built with an adaptive octree. For each grid point, a binary sign, which indicates inside or outside of the mesh, is generated based on the distance to the mesh. In each leaf cell having a vertex, a least square fitting plane is created for a localized geometry range with signs. Finally, quantized geometry information is locally encoded. We demonstrate that the octree with signs can be used to predict the vertex positions. As a result, the proposed method generates competitive bitrates compared to the current state-of-art progressive geometry coder. Our method also shows better rate-distortion performance during decompression or transmission with improved smoothness.","Title: Primitive Trees for Precomputed Distance Queries.\nYear: 2013\n\nAbstract: We propose the primitive tree, a novel and compact space-partition method that samples and reconstructs a distance field with high accuracy, even for regions far from the surfaces. The primitive tree is based on the octree and stores the indices of the nearest primitives in its leaf nodes. Most previous approaches have involved a trade-off between accuracy and speed in distance queries, but our method can improve both aspects simultaneously. In addition, our method can sample unsigned distance fields effectively, even for self-intersecting and nonmanifold models. We present test results showing that our method can sample and represent large scenes, with more than ten million triangles, rapidly and accurately.","Title: Automatic player behavior analysis system using trajectory data in a massive multiplayer online game\nYear: 2013\n\nAbstract: This paper presents a new automated behavior analysis system using a trajectory clustering method for massive multiplayer online games (MMOGs). The description of a player's behavior is useful information in MMOG development, but the monitoring and evaluation cost of player behavior is expensive. In this paper, we suggest an automated behavior analysis system using simple trajectory data with few monitoring and evaluation costs. We used hierarchical classification first, then applied an extended density based clustering algorithm for behavior analysis. We show the usefulness of our system using trajectory data from the commercial MMOG World of Warcraft (WOW). The results show that the proposed system can analyze player behavior and automatically generate insights on players' experience from simple trajectory data.","Title: Interchangeable SPH and level set method in multiphase fluids\nYear: 2009\n\nAbstract: Subgrid-scale fluid is difficult to represent realistically in a grid-based fluid simulation. We show how to describe such small-scale details effectively, even on a coarse grid, by using escaped particles. The simulation of these particles with SPH (smooth particle hydrodynamics) allows the illustration of dynamic and realistic animation of fluids. Particles modeled by SPH have a force which leads them to merge if they are within a certain range. This reduces the accuracy of a simulation. Consequently, aggregated particles which form volumes large enough to be described by the level set method will be simulated inefficiently by particles. We address this problem with a new method in which details too small for the grid are represented by particles, while the level set method with a grid is used to describe merged particles on the grid.","Title: Hardware-accelerated real-time rendering for 3D Sumi-e painting\nYear: 2003\n\nAbstract: This paper presents a method for real-time 3D Sumi-e rendering using normal graphics hardware. Sumi-e is one of the traditional oriental painting styles. Most research on Sumi-e paintings has focused on 2D or 2.5D Sumi-e brushwork simulation. On these systems, complicated user's hand drawing is required to generate the image of Sumi-e effects, and it can render the 2D or 2.5D Sumi-e images only. We present an automated rendering system for 3D image of Sumi-e painting. It uses 3D common object as an input data and does not need any additional input of user brushwork. Especially for the real-time rendering, hardware-accelerated algorithm for Sumi-e rendering is newly suggested in our system. It is designed with efficiency for customer level graphics hardware. The results of this paper show that the features of traditional Sumi-e painting are successfully modeled and that 3D Sumi-e painting is rendered in real-time effectively.","Title: Analytic solutions of integral moving least squares for polygon soups.\nYear: 2012\n\nAbstract: This paper presents analytic solutions to the integral moving least squares (MLS) equations originally proposed by Shen et al. by choosing another specific weighting function that renders the numerator in the MLS equation unitless. In addition, we analyze the original method to show that their approximation surfaces (i.e., enveloping surfaces with nonzero values in the weighting function) often form zero isosurfaces near concavities behind the triangle-soup models. This paper also presents error terms for the integral MLS formulations against signed distance fields. Based on our analytic solutions, we show that our method provides both interpolation and approximation surfaces faster and more efficiently. Because our method computes solutions for integral MLS equations directly, it does not rely on numerical steps that might have numerical-accuracy issues. In particular, unlike the original method that deals with incorrect approximation surfaces by iteratively adjusting parameters, this paper proposes faster and more efficient approximations to surfaces without needing iterative routines. We also present computational efficiency comparisons, in which our method is 15-fold faster in computing integrations, even with conservative assumptions. Finally, we show that the surface normal vectors on the implicit surfaces formed by our analytic solutions are identical to the angle-weighted pseudonormal vectors.","Title: System for evaluating usability and user experience by analyzing repeated patterns\nYear: 2013\n\nAbstract: In this paper, a new system for evaluating interface usability through the analysis of repeated patterns is proposed. The system can be a valuable tool for verifying interfaces and in evaluating their usability by users, both of which are necessary stages in the development and operation of software. This paper concentrates on the repeated patterns that occur when users use an interface. Extracting these repeated patterns and analyzing them could enhance the development and usability of interfaces. Through experiments that applied the proposed system to several kinds of software, it was confirmed that problems with interfaces can be understood, and usability can be improved without requiring complicated analyses of user logs.","Title: Fast Reconstruction of 3D Objects from Single Free-Hand Line Drawing\nYear: 2001\n\nAbstract: This paper proposes an efficient algorithm that not only can narrow down the search domain of face identification but also can reconstruct various 3D objects from a single free-hand line drawing. The algorithm is executed in two stages. In the face identification stage, we generate and classify potential faces into implausible, basis and minimal faces by using geometrical and topological constraints to reduce search space. The proposed algorithm searches the space of minimal faces only to identify actual faces of an object fast. In the object reconstruction stage, we introduce 3D regularities and quadric face regularities to reconstruct 3D object accurately. Furthermore, the proposed method can be applied to a wide scope of general objects containing flat and quadric faces. The experimental results show that the proposed method identifies faces much faster than previous ones and efficiently reconstructs various objects from a single free-hand line drawing.","Title: Simulation of swirling bubbly water using bubble particles\nYear: 2009\n\nAbstract: The effect of surface tension is dynamically and realistically represented within a multiphase fluid simulation. Air bubbles are seeded with ‘bubble particles’ which move randomly. These molecule-like movements modify the surface of the air bubbles and generate turbulence in the water. The surface tension between air bubble and water, determined by the composition of the water, remains constant regardless of the size of the bubble, while external forces cause unstable fluid motion as the surface tension strives to remain constant, bubbles split and merge. The bubble particles can also compute for the numerical dissipation usually experienced in grid-based fluid simulations, by restoring the lost volume of individual bubbles. The realistic tearing of bubble surfaces is shown in a range of examples.","Title: Multiresolution recognition of unconstrained handwritten numerals with wavelet transform and multilayer cluster neural network\nYear: 1996\n\nAbstract: In this paper, we propose a new scheme for multiresolution recognition of unconstrained handwritten numerals using wavelet transform and a simple multilayer cluster neural network. The proposed scheme consists of two stages: a feature extraction stage for extracting multiresolution features with wavelet transform, and a classification stage for classifying unconstrained handwritten numerals with a simple multilayer cluster neural network. In order to verify the performance of the proposed scheme, experiments with unconstrained handwritten numeral database of Concordia University of Canada, Electro-Technical Laboratory of Japan, and Electronics and Telecommunications Research Institute of Korea were performed. The error rates were 3.20%, 0.83%, and 0.75%, respectively. These results showed that the proposed scheme is very robust in terms of various writing styles and sizes.","Title: Reconstruction of Non-manifold Objects from Two Orthographic Views\nYear: 1998\n\nAbstract: Since the two-view orthographic representation of a 3D object is ambiguous, it requires a numerous amount of combinatorial searches in the process of reconstruction. This paper presents an efficient algorithm for reconstructing polyhedral 3D objects from two-view drawings. The main feature of the algorithm is to improve the reconstruction process speed. First, the partially constructed objects are reconstructed from the restricted candidate faces corresponding to each area on the two-view drawings. Then the complete objects are obtained from the partially constructed objects by adding perpendicular faces with geometrical validity. By limiting the number of candidate faces corresponding to areas only, the combinatorial search space can be considerably reduced. In addition, the reconstruction finds the most plausible 3D object that human observers are most likely to select first among the given multiple solutions. Several examples from a working implementation are given to demonstrate the completeness of the algorithm.","Title: Polygonal space carving with geometric anti-aliasing\nYear: 2004\n\nAbstract: This paper proposes an algorithm that solves the shape recovery problem from N arbitrary images. By introducing a polygonal carving technique, the proposed algorithm can reconstruct the image-consistent polygonal shape that is patched by input images. This algorithm eliminates the invalid vertices and polygons from the initial polygonal grid space according to the color variance that represents their image consistency. The carved shape is refined by moving the outlier vertices on the boundary of each image. The final reconstructed shape faithfully accounts for the input images, and its textured appearance reflects the similar color property of the target object.","Title: Self-correctional 3D shape reconstruction from a single freehand line drawing\nYear: 2003\n\nAbstract: The goal of sketch reconstruction is to take an inaccurate, 2D edge-vertex graph (i.e., sketch drawing) as input and reconstruct a 3D shape as output. However, traditional reconstruction methods based on image regularities tend to produce a distorted 3D shape. In part, this distortion is due to the inherent inaccuracies in the sketch, but it also relates to the failure to accurately distinguish between important and less important regularities. We propose a new self-correctional reconstruction algorithm that can progressively produce refined versions of sketch reconstructions. The algorithm corrects the shape and the drawing simultaneously using geometric error metrics. The proposed algorithm can minimize the distortion of the shape by adding 3D regularities to the image regularities. The self-correctional algorithm for minimizing the distortion of sketch reconstruction is discussed, and the experimental results show that the proposed method efficiently reconstructs more accurate 3D objects than previous ones.","Title: Adaptive synthesis of distance fields.\nYear: 2012\n\nAbstract: We address the computational resource requirements of 3D example-based synthesis with an adaptive synthesis technique that uses a tree-based synthesis map. A signed-distance field (SDF) is determined for the 3D exemplars, and then new models can be synthesized as SDFs by neighborhood matching. Unlike voxel synthesis approach, our input is posed in the real domain to preserve maximum detail. In comparison to straightforward extensions to the existing volume texture synthesis approach, we made several improvements in terms of memory requirements, computation times, and synthesis quality. The inherent parallelism in this method makes it suitable for a multicore CPU. Results show that computation times and memory requirements are very much reduced, and large synthesized scenes exhibit fine details which mimic the exemplars.","Title: Direct reconstruction of a displaced subdivision surface from unorganized points\nYear: 2002\n\nAbstract: In this paper we propose a new mesh reconstruction algorithm that produces a displaced subdivision mesh directly from unorganized points. The displaced subdivision surface is a new mesh representation that defines a detailed mesh with a displacement map over a smooth domain surface. This mesh representation has several benefits - compact mesh size, piecewise regular connectivity - to overcome limitations of irregular mesh produced by ordinary mesh reconstruction scheme, but the original displaced subdivision surface generation algorithm needs an explicit polygonal mesh to be converted. Our approach is producing displaced subdivision surface directly from input points during the mesh reconstruction process. The main ideas of our algorithm are building initial coarse control mesh by the shrink-wrapping like projection and sampling fine surface detail from unorganized points along the each limit vertex normal without any connectivity information of given points. We employ an existing subdivision surface fitting scheme to generate a parametric domain surface, and suggest a surface detail sampling scheme that determines a valid sampling triangle which can be made with combinations of input points. We show several reconstruction examples and applications to show the validity of suggested sampling technique and benefits of the result like multiresolution modeling.","Title: A Particle-Grid Method for Opaque Ice Formation.\nYear: 2013\n\nAbstract: This paper presents a particle-grid method to simulate the generation of opaque ice which has air bubbles in it. Water temperature is diffused over a grid, and the exchange of dissolved air between ice and water particles is simulated. We render a particle as an air bubble if it has sufficient air. Otherwise, it is treated as a cloudy volume by distributing air into dissolved air field when the final state has been reached. In addition, our method includes a model in which heat transfer rate may change across the grid. Unlike previous models which could generate an ice volume of only fixed shapes, our approach uses signed distance function (SDF) to generate opaque ice volumes stored in containers of various geometric shapes and can render needle-shaped or egg-shaped bubbles.","Title: Controllable multi-phase smoke with lagrangian particles\nYear: 2006\n\nAbstract: This paper describes a method for controlling the multi-phase smoke animation that uses Lagrangian particles. Previous methods need several density fields to simulate different types of smoke. We use a single field, which is more like a natural phenomenon to obtain the interactive motions of fluid. Also, whereas existing methods which apply control forces to cells only, we define particle forces which enable each particle to move independently towards to target shape. Additionally, we set up internal target forces which distribute the particles uniformly within the target shape, no that it is represented more precisely.","Title: Simulation of Morphology Changes in Drying Leaves.\nYear: 2013\n\nAbstract: We introduce a biologically motivated simulation technique for the realistic shape deformation of drying leaves. In contrast to skeleton-based leaf deformation, our approach simulates the whole leaf surface to capture the fine details of desiccated leaves. We represent a leaf as a triangulated double-layer structure that consists of a Delaunay triangulation discretized along the vein structure and its corresponding Voronoi diagram. This structure can generate not only sharp creases along leaf veins, but also the complicated curling and crumpling on the leaf surface. The loss of water is the major factor that controls the inhomogeneous shrinkage of drying leaves. The proposed osmotic water flow successfully models the gradual changes of dehydrated regions advancing towards the veins. We demonstrate the robustness of our method by comparing a sequence of simulated morphology changes with photographs of real leaves.","Title: Combustion Waves on the Point Set Surface.\nYear: 2013\n\nAbstract: This paper introduces a combustion model of heat transfer and fuel consumption for the propagation of a fire front on a point cloud surface. The heat transfer includes the heat advection by the airflow as well as diffusion, chemical reaction, and heat loss to generate complex, but controllable heat flows with a designed airflow velocity. For the stable heat advection, we solve a semi-Lagrangian method on point samples using discrete exponential maps to trace the position from which the wind blows while preserving the geodesic distance. We also propose angular Voronoi weights for a discrete Laplace-Beltrami operator that shows better isotropic diffusion on the inhomogeneous distribution of point clouds than the cotangent or moving least-squares schemes. We demonstrate a diversity of burning scenarios by incorporating factors affecting the fire spreading such as buoyancy and object geometries in the airflow velocity fields, or by synthesizing patterns.","Title: Finding ridges and valleys in a discrete surface using a modified MLS approximation\nYear: 2005\n\nAbstract: Implicit surface fitting is a promising approach to finding ridges and valleys in discrete surfaces, but existing techniques are time-consuming and rely on user-supplied tuning parameters. We use a modified MLS (moving-least-squares) approximation technique to estimate the local differential information near a vertex by means of an approximating surface. Ridge and valley vertices are easily detected as zero-crossings, and can then be connected along the direction of principal curvature. Our method, demonstrated on several large meshed models, produces a good fit which leads to improved visualization. It does not oscillate and is quick to compute.","Title: Finding ridges and valleys in a discrete surface using a modified MLS approximation [Computer-Aided Design 37 (14) 1533–1542]\nYear: 2006\n\nAbstract: Implicit surface fitting is a promising approach to finding ridges and valleys in discrete surfaces, but existing techniques are time-consuming and rely on user-supplied tuning parameters. We use a modified MLS (moving-least-squares) approximation technique to estimate the local differential information near a vertex by means of an approximating surface. Ridge and valley vertices are easily detected as zero-crossings, and can then be connected along the direction of principal curvature. Our method, demonstrated on several large meshed models, produces a good fit which leads to improved visualization. It does not oscillate and is quick to compute. (c) 2005 Elsevier Ltd. All rights reserved.","Title: Controlling shapes of air bubbles in a multi-phase fluid simulation\nYear: 2012\n\nAbstract: Controlling shapes is a challenging problem in a multi-phase fluid simulation. Bubble particles enable the details of air bubbles to be represented within a simulation based on an Euler grid. We control the target shapes of bubbles by the gradient vectors of the signed distance field and attraction forces associated with control particles. Our hybrid approach enables to simulate physically plausible movements of bubbles while preserving the details of a target shape. Furthermore, we control the paths of moving bubbles using user-defined curves and the shape of an air bubbles by drag force. An accurate model of the drag force near the fluid surface means that bubbles have realistic ellipsoidal shapes.","Title: A feature-preserved simplification for autonomous facial animation from 3D scan data\nYear: 2003\n\nAbstract: We propose a new simplification algorithm of facial models for animation. For the facial animation, the models are often simplified from complex scan data based on geometric features, but it leads to decrease the quality and such features are easily noticed by human perception. For example, a lip line and eyebrows easily lose their details by geometry-based simplification. In this paper, facial features are extracted using an image processing of a 2D texture image and the curvature analysis of the 3D geometry, which improves the details around the feature areas of the facial model. Especially if lip contact line is simplified to one or two edges, it may not be proper for lip animation. Finally, we will show that our simplified model can produce as good as a facial animation as the one from the original model.","Title: Target-driven liquid animation with interfacial discontinuities\nYear: 2007\n\nAbstract: We propose a novel method of controlling a multi-phase fluid so that it flows into a target shape in a natural way. To preserve the sharp detail of the target shape, we represent it as an implicit function and construct the level-set of that function. Previous approaches add the target-driven control force as an external term, which then becomes attenuated during the velocity projection step, making the convergence process unstable and causing sharp detail to be lost from the target shape. But we calculate the force on the fluid from the pressure discontinuity at the interface between phases, and integrate the control force into the projection step so as to preserve its effect. The control force is calculated using an enhanced version of the ghost fluid method (GFM), which guarantees that the fluid flows from the source shape and converges into the target shape, while achieving a more natural animation than other approaches. Our control force is merged during the projection step avoiding the need for a post-optimization process to eliminate divergence at the liquid interface. This makes our method easy to implement using existing fluid engines and it incurs little computational overhead. Experimental results show the accuracy and robustness of this technique. Copyright © 2007 John Wiley & Sons, Ltd.","Title: Sports equipment based motion deformation\nYear: 2005\n\nAbstract: We propose a novel scheme for generating the locomotion of sports equipment along with motion deformation. To accomplish this, we search the event frames in order to find the points where the character is in contact with the sports equipment (contact points) or not (split points). To find these points, it is necessary to locate the corresponding event points in the signal using Discrete Fourier Transformation. This technique generates a left and right hand-cross analysis and sports equipment locomotion adapted to the character motion. Also, the locomotion of the sports equipment is changed by adjusting its properties, and this may require the posture of the character to be modified. In conclusion, we are able to deform the motion of both the sports equipment and the character simply by adjusting the properties of the sports equipment. Our scheme can be used for the motion deformation of any character with an associated sports equipment.","Title: Design and evaluation of a hybrid display system for motion-following tasks\nYear: 2007\n\nAbstract: Hybrid display systems are those that combine different types of displays to exploit the complementary characteristics of the constituent display systems. In this paper, we introduce a hybrid system that combines a stereoscopic optical see-through head-mounted display (HMD) and a large projection display for an application in a multi-user ship painting training scenario. The proposed hybrid system's projection display provides a large FOV and a physical metaphor to the ship surface with natural depth perception, while the HMD provides personal and unoccluded display of the motion training guides. To quantify its effectiveness, we conducted a human subject experiment, comparing the subject's motion following task performance among three different display systems: large projection display, head-mounted display, and hybrid. The preliminary results obtained from the experiment has shown that given the same FOV, the hybrid system performed, despite problems with registration between the real and virtual worlds, up to par with the head-mounted display, and better than the projection display. Thus, it is expected that the hybrid display will result in higher task performance with the larger FOV factor available.","Title: Adaptive fuzzy logic for focusing in optical disk player\nYear: 2008\n\nAbstract: For the nonlinear behaviors of the actuator and servo system of the optical disk player, a fuzzy control is introduced. Focus servo controls the laser beam to focus on the playing surface of the disc by vertical movement of the objective lens. The change of the distance between the objective lens and the disc must be within the allowable focus error to stay in focus. This must be maintained even with a slightly warped or uneven disc and in a portable player. This paper will observe on how the focusing is controlled in an optical disk player and will evaluate algorithm based on the model using Adaptive Fuzzy Logic.","Title: Progressive 3D Reconstruction from a Sketch Drawing\nYear: 2001\n\nAbstract: This paper presents a progressive algorithm for reconstructing a 3D structure from a given 2D sketch drawing (edge-vertex graph without hidden line removal) according to the user's sketch order. While previous methods reconstruct a 3D structure at once, the proposed method progressively calculate a 3D structure by optimizing the coordinates of vertices of an object according to the sketch order. The progressive method reconstructs the most plausible 3D object quickly by applying 3D constraints that are derived from the relationship between the object and the sketch drawing in the optimization process. Furthermore, it allows the user to change viewpoint during sketching, and also minimize distortion of an object by refining inaccurate sketch drawings. The progressive reconstruction algorithm is discussed, and examples from a working implementation are given.","Title: Extracting feature lines from unstructured meshes using a model feature map\nYear: 2006\n\nAbstract: We propose a new method of extracting feature lines from a 3D scan model. Given an unstructured 3D triangle mesh containing texture information, we use model feature map (MFM), which is a 2D map that abstracts the variation of texture and curvature in the 3D model. The MFM is created from both a texture map and a curvature map, which are produced separately by edge-detection to locate line features. The MFM can be edited interactively using standard image-processing tools. We demonstrate the technique on several data sets, including, but not limited to facial data.","Title: Systematic Reconstruction Of 3d Curvilinear Objects From Two-View Drawings\nYear: 1999\n\nAbstract: This paper presents an efficient algorithm for reconstructing curvilinear 3D objects from two-view drawings. The main features of the algorithm are to improve the speed of the reconstruction process and to generate a sequence of solutions. First, the partially constructed objects are reconstructed from the restricted candidate faces corresponding to each area in the two-view drawings. Then the complete objects are obtained from the partially constructed objects by adding perpendicular faces with geometric validity. By limiting the number of candidate faces corresponding to areas only, the computational time of the combinatorial search process is considerably reduced. Second, to find more plausible solutions prior to unusual ones, the algorithm generates the sequence number to multiple solutions based on a plausibility function that maps the proposed solution to the designer's intention in the original drawings. Furthermore, the proposed method of reconstruction supports a wide scope of general (manifold and non-manifold) 3D objects containing cylindrical, conical, spherical, and toroidal faces. Several examples from a working implementation are given to demonstrate the completeness of the algorithm. (C) 1999 Elsevier Science Ltd. All rights reserved.","Title: Live path: adaptive agent navigation in the interactive virtual world\nYear: 2010\n\nAbstract: We present a novel approach to adaptive navigation in the interactive virtual world by using data from the user. Our method constructs automatically a navigation mesh that provides new paths for agents by referencing the user movements. To acquire accurate data samples from all the user data in the interactive world, we use the following techniques: an agent of interest (AOI), a region of interest (ROI) map, and a discretized path graph (DPG). Our method enables adaptive changes to the virtual world over time and provides user-preferred path weights for smart-agent path planning. We have tested the usefulness of our algorithm with several example scenarios from interactive worlds such as video games. In practice, our framework can be applied easily to any type of navigation in an interactive world. In addition, it may prove useful for solving previous pathfinding problems in static navigation planning.","Title: CUDA-based Signed Distance Field Calculation for Adaptive Grids\nYear: 2010\n\nAbstract: We present a simple yet robust signed distance field (SDF) generator based on recent GPU architectures. In our approach, the squared Euclidean distance is calculated for each triangle face in parallel, and then an optimized stream reduction process is used to find the shortest distance. During this process, the stream reduction operation acts like a parallel binary space-searching routine for each level. This process uses computations and memory bandwidth efficiently because of the massive number of CUDA threads. Signs are then determined by calculating angle-weighted pseudonormals. Unlike some previous SDF approaches that only calculate the SDF near the surface or within the bounding box, our method can calculate the SDF adaptively so that there are no limitations on proximity or regularity. We also compare our GPU-based results with a kd-tree based single CPU approach for a 3D geometry synthesis application.","Title: CDP Servo System Control using Fuzzy Logic Control\nYear: 2007\n\nAbstract: Several servo mechanisms, such as focusing, spindling, tracking and sledding, are continuously balanced to make compact disc player (CDP) to play audio or video contents. However, due to the inconsistent uneven surface of the disc, the speed control of a spindle motor along the track radius, and unexpected random shock; the servo mechanism is very complicatedly driven in non-linear method. This paper will discuss on how to approach this method, how to develop and how to evaluate each algorithm based on the model of each servo mechanism using Fuzzy Logic Control1.","Title: Extraction of ridges-valleys for feature-preserving simplification of polygonal models\nYear: 2006\n\nAbstract: We propose a new method for simplifying polygonal models while retaining salient features. We extract ridge and valley features defined by means of curvature derivatives on the model surface. We combine this extraction with a simplification technique so as to construct coarse models with features preserved. Experimental results have better quality and smaller geometric error than these of previous methods.","Title: Adaptive space carving with texture mapping\nYear: 2005\n\nAbstract: Space carving reconstructs a 3D object from multiple images, but existing algorithms rely on a regular grid which makes poor use of memory. By using the image information, adaptive space carving uses a recursively generated structure which reduces memory requirements and thus allows a finer grid. After reconstruction, models are triangulated to facilitate texture mapping. Experimental results show the enhanced appearance of models reconstructed in this way.","Title: Surface simplification using a discrete curvature norm\nYear: 2002\n\nAbstract: This paper proposes a mesh simplification algorithm using a discrete curvature norm. Most of the simplification algorithms are using a distance metric to date. The distance metric is very efficient to measure geometric error, but it is difficult to distinguish important shape features such as a high-curvature region even though it has a small distance metric. We suggest a discrete curvature norm to measure geometric error for such features. During simplification the new vertex resulted from an edge collapse takes a position using a butterfly subdivision mask to minimize geometric error. This paper shows that simplification results have smaller geometric errors than previous works, when a discrete curvature norm and a distance metric are together applied to its criterion.","Title: Shrinkage, wrinkling and ablation of burning cloth and paper\nYear: 2011\n\nAbstract: The burning of a sheet of cellulose-based material, such as paper or cloth, involves uneven shrinkage which causes wrinkling. We simulate this geometrically complicated phenomenon by modeling the effects of heat transfer, shrinkage and partial ablation on a thin shell. A strain-limitation technique is applied to a two-layer structure of springs arranged as a body-centered square. Although this structure is over-constrained, convergence can be achieved using a new successive fast projection method. We also remesh the shells dynamically to deal with the topological changes that occur as regions burn away.","Title: Reconstructing 3D models with algebraic curved surfaces from three-view drawings\nYear: 1996\n\nAbstract: A system that reconstructs 3D models with algebraic curved surfaces from three orthographic views is described. In this paper, we present two features. First, we can directly reconstruct 3D models with algebraic curved surfaces without using a method of polyhedron approximation by introducing two kinds of supplementary lines: silhouette lines and tangency lines. Second, we develop new rules and heuristics to handle curved surfaces, used in combinatorial search of faces. Rules work to reduce the processing time, and heuristics are aiming at simulation of the way human interprets the three orthographic views. Soundness and efficiency of the proposed method are also proved experimentally","Title: Evolving Physically Simulated Flying Creatures for Efficient Cruising\nYear: 2006\n\nAbstract: The body-brain coevolution of aerial life forms has not been developed as far as aquatic or terrestrial locomotion in the field of artificial life. We are studying physically simulated 3D flying creatures by evolving both wing shapes and their controllers. A creature's wing is modeled as a number of articulated cylinders, connected by triangular films (patagia). The wing structure and its motor controllers for cruising flight are generated by an evolutionary algorithm within a simulated aerodynamic environment. The most energy-efficient cruising speed and the lift and drag coefficients of each flier are calculated from its morphological characteristics and used in the fitness evaluation. To observe a wide range of creature size, the evolution is run separately for creatures categorized into three species by body weight. The resulting creatures vary in size from pigeons to pterosaurs, with various wing configurations. We discuss the characteristics of shape and motion of the evolved creatures, including flight stability and Strouhal number.","Title: RViz: a toolkit for real domain data visualization\nYear: 2015\n\nAbstract: In computational science and computer graphics, there is a strong requirement to represent and visualize information in the real domain, and many visualization data structures and algorithms have been proposed to achieve this aim. Unfortunately, the dataflow model that is often selected to address this issue in visualization systems is not flexible enough to visualize newly invented data structures and algorithms because this scheme can accept only specific data structures. To address this problem, we propose a new visualization tool, RViz, which is independent of the input information data structures. Since there is no requirement for additional efforts to manage the flow networks and the interface to abstracted information is simple in RViz, any scientific information visualization algorithms are easier to implement than the dataflow model. In this paper, we provide case studies in which we have successfully implemented new data structures and related algorithms using RViz, including geometry synthesis, distance field representation, and implicit surface reconstruction. Through these cases, we show how RViz helps users visualize and understand any hidden insights in input information.","Title: Controlling liquids using pressure jump\nYear: 2006\n\nAbstract: This sketch presents a method to control liquids so that they flow into a target shape in a natural way. To avoid the oscillation and preserve the sharp detail of the target shape, we integrate the target-driven force into the projection step. So we require no post-optimization process for divergence-free condition at the liquid interface. Additionally, the target-driven force is calculated by enhanced ghost fluid method considering the pressure discontinuity around the liquid interface.","Title: Generating flying creatures using body-brain co-evolution\nYear: 2003\n\nAbstract: This paper describes a system that produces double-winged flying creatures using body-brain co-evolution without need of complex flapping flight aerodynamics. While artificial life techniques have been used to create a variety of virtual creatures, little work has explored flapping-winged creatures for the difficulty of genetic encoding problem of wings with limited geometric primitives as well as flapping-wing aerodynamics. Despite of the simplicity of system, our result shows aesthetical looking and organic flapping flight locomotions. The restricted list structure is used in genotype encoding for morphological symmetry of creatures and is more easily handled than other data structures. The creatures evolved by this system have two symmetric flapping wings consisting of continuous triangular patches and show various looking and locomotion such as wings of birds, butterflies and bats or even imaginary wings of a dragon and pterosaurs.","Title: Physics-inspired controllable flame animation.\nYear: 2016\n\nAbstract: We propose a novel method conceptualized from the properties of physics where in particular the shape of a flame is determined by temperature that enables a control mechanism for the intuitive shaping of a flame. We focused on a trade-off issue from computer graphics whereby the turbulent flow that expresses the characteristics of the flame has a tendency to shift continuously, whereas the velocity constraints that contain a fluid within a target shape have a tendency to force movement in a particular direction. Trade-off made it difficult for animation designers to maintain a flame within the intended target shape. This paper resolves the issue by enabling the flame to be controlled without any velocity constraints by using the following two techniques: First, we model the temperature and force of the explosion generated by the combustion of explosive gaseous fuel and apply it to certain regions. Second, we expand the space of the interface between the fuel and the burned products, classifying that space into four regions and controlling the target shape of the flame by delicate adjustments to the temperature in each region. Experiments show that the flame maintains the appearance of dynamic movement while preserving the detailed 3D shapes specified by the scene designers.","Title: Subtle features of ice with cloudy effects and scratches from collision damage\nYear: 2016\n\nAbstract: AbstractWe propose a simulation framework for expressing cloudy effects and scratches on ice caused by the collision of other objects. Cloudy effects are created and diffused in proportion to the collision on the ice surfaces calculated by a combination of a grid-projection technique and the boundary particle method. To prevent dissipation during the diffusion process, a geodesic distance is used as a constraint. Scratches are modeled directionally by analyzing the density gradient of cloudy effects and rendered using needle-shaped ellipsoids. Experiments provide detailed expression of cloudy effects and scratches from collision damage. Copyright © 2016 John Wiley & Sons, Ltd.","Title: Let's be a virtual juggler.\nYear: 2016\n\nAbstract: Juggling, which uses both hands to keep several objects in the air at once, is admired by anyone who sees it. However, skillful real-world juggling requires long, hard practice. Therefore, we propose an interesting method to enable anyone to juggle skillfully in the virtual world. In the real world, the human motion has to follow the motion of the moving objects; in the virtual world, the objects' motion can be adjusted together with the human motion. By using this freedom, we have generated a juggling avatar that can follow the user's motion. The user simply makes juggling-like motions in front of a motion sensor. Our system then searches for juggling motions that closely match the user's motions and connects them smoothly. We then generate moving objects that both satisfy the laws of physics and are synchronized with the synthesized motion of the avatar. In this way, we can generate a variety of juggling animations by an avatar in real time. Copyright © 2016 John Wiley & Sons, Ltd.","Title: Visual simulation of fire-flakes synchronized with flame.\nYear: 2017\n\nAbstract: We propose a framework for generating, animating, and controlling fire-flakes to correspond with the movements of flame. Not only do fire-flakes themselves display a distinctively complex and elaborate movement, but also they are heavily influenced by where the heat transforms into flame and how the flame moves. It is difficult for designers to generate and simulate synchronous movement of fire-flakes and flame because the movement of flame tends to be chaotic and constantly changing. We provide a novel framework that can improve on the current processes both effectively and efficiently. We first propose a fire-flake motion model that simulates movement harmoniously with the movement of flame. We then introduce automatic generation of fire-flakes by analyzing temperature fields and the velocity fields input from the fire simulation. Lastly, our sample-based control method enables the formation of a target shape by gathering fire-flakes together into one location while maintaining their inherent movement without external forces. We also provide various parameters that the designer can use to control the effects of the forces determining the amount, velocity, and movement of fire-flakes. To the best of our knowledge, this work is the first visual simulation method to generate fire-flakes in coordination with flame, and we believe it will contribute not only to research, but also to the animation industry.","Title: Incorporating particle motion into an ADF for fast coupling of fluids with rigid and deformable solids.\nYear: 2017\n\nAbstract: We present a new method for the fast simulation of interactions between fluids and solids by incorporating particle-based water flow into an adaptive signed distance field. In some previous methods, the motion of every water particle is checked when simulating the collision with the solid in the coupling process, and the computational cost becomes very great as the number of particles increases. If only the particles on the leaf nodes surrounding the solid are considered, this reduces the computational cost of collision detection, but some collisions may not be detected. This may lead to the tunneling artifact, in which particles with high velocities skip across the layer of leaf nodes. This paper addresses the problem by (i) considering particles only on the leaf nodes in the adaptive structure to improve the processing time required for the water-solid coupling and (ii) considering the water flow to avoid the tunneling artifact by incorporating particle motion into the tree structure of the adaptive signed distance field. Our method can be computed in parallel, and experimental results show that it outperforms previous methods while producing animations that are largely free of artifacts. Copyright (c) 2016 John Wiley & Sons, Ltd.","Title: Visual simulation of rapidly freezing water based on crystallization.\nYear: 2017\n\nAbstract: We propose a physics-inspired simulation framework that expresses visual effects of flowing water frozen in glaze or directional icicles. The proposed ice model considers the direction of the water flow, which affects the growth of icicles. Water dynamics are computed using a conventional particle-based simulation. Ice glaze and directional icicles are generated by incorporating our freezing solver. To determine whether a water particle is converted into ice or remains liquid, we compute the nucleation energy based on the humidity and water flow. The humidity is approximated as a virtual water film on object surfaces. The water flow is incorporated by introducing a growth direction vector to guide the direction of icicle growth. Ice-generating regions can be controlled using 3-D modeling tools such as Autodesk Maya or 3DS Max. Experiments showed that a realistic ice glaze was created on the surfaces of objects and that icicles grew in the direction of the water flow.","Title: A hybrid SDF for the detailed representation of liquid-solid mixed surfaces\nYear: 2015\n\nAbstract: We propose a hybrid signed distance field SDF method for reconstructing the detailed surface of a model as it changes from a solid state to a liquid state. Previous particle-based fluid simulations suffer from a noisy surface problem when the particles are distributed irregularly. If a smoothing scheme is applied to reduce the problem, sharp and detailed features can be lost by over-smoothing artifacts. Our method constructs a hybrid SDF by combining level-set values from the solid and liquid parts of the object. This makes it possible to represent the detailed features and smooth surfaces of an object when both solid and liquid parts are mixed in that object. In addition, the concept of a guiding shape is proposed, which uses a coordinate-warping technique to query the level-set values quickly. The guiding shape is constructed from the object before the simulation begins and some parts of it become liquid. To track the details of the initial solid shape and preserve it, the transformation of the guiding shape is accumulated while the phase-shift is in progress. By warping the coordinates of this accumulated transformation of the guiding shape, the level-set values of the solid part can be acquired very quickly. Copyright © 2015John Wiley & Sons, Ltd.","Title: Efficient Representation of Detailed Foam Waves by Incorporating Projective Space.\nYear: 2017\n\nAbstract: We propose an efficient framework to realistically simulate foam effects in which 3D water particles from a base water solver are first projected onto 2D screen space in order to reduce computational complexity of finding foam particles. Because foam effects are often created primarily in fast and complicated water flows, we analyze acceleration and curvature values to identify the areas exhibitin...","Title: Realistic Visual Simulation of Water Effects in Response to Human Motion using a Depth Camera.\nYear: 2017\n\nAbstract: In this study, we propose a new method for simulating water responding to human motion. Motion data obtained from motion-capture devices are represented as a jointed skeleton, which interacts with the velocity field in the water simulation. To integrate the motion data into the water simulation space, it is necessary to establish a mapping relationship between two fields with different properties. However, there can be severe numerical instability if the mapping breaks down, with the realism of the human-water interaction being adversely affected. To address this problem, our method extends the joint velocity mapped to each grid point to neighboring nodes. We refine these extended velocities to enable increased robustness in the water solver. Our experimental results demonstrate that water animation can be made to respond to human motions such as walking and jumping.","Title: An optimized real time algorithm for window frost formation suited to mobile devices.\nYear: 2018\n\nAbstract: We propose a real time simulation for window frost formation on mobile devices that uses both particles and grids. Previous ice formation methods made heavy demands on both memory and computational capacity because they were designed for a desktop environment. In this paper, a frost skeleton grows around a location touched by the user using particles, and the ice surfaces are constructed using a grid. Using a nonlattice random-walk technique, the frost skeleton grows freely and naturally. A hash grid technique is used to search efficiently for neighbor particles during the crystallization process. Finally, some 2.5D details are added to the ice skeleton by adjusting the height of the grid vertices around the skeleton. Experiments show that our method creates realistic frost in real time. Our method can be used to express ice formation effects in touch-based mobile device applications such as weather forecasts or games.","Title: Predicting the Currency Market in Online Gaming via Lexicon-Based Analysis on Its Online Forum.\nYear: 2017\n\nAbstract: Transactions involving virtual currencies are becoming increasingly common, including those in online games. In response, predicting the market price of a virtual currency is an important task for all involved, but it has not yet attracted much attention from researchers. This paper presents user opinions from online forums in a massive multiplayer online game (MMOG) setting widely used around the world. We propose a method for predicting the next-day rise and fall of the currency used in an MMOG environment. Based on analysis of online forum users' opinions, we predict daily fluctuations in the price of a currency used in an MMOG setting. Focusing specifically on the World of Warcraft game, one of the most widely used MMOGs, we demonstrate the feasibility of predicting the fluctuation in value of virtual currencies used in this game community.","Title: Surface simplification with semantic features using texture and curvature maps\nYear: 2005\n\nAbstract: We propose a polygonal surface simplification algorithm that can preserve semantic features without user control. The semantic features of a model are important for human perception, which are insensitive to small geometric errors. Using an edge detects: Its three kinds of maps are employed to extract these features. First, an image map is generated boundary lines represent changes of chroma in the texture image by using edge detector. Second, the discrete curvatures at 3D vertices are mapped to the curvature map, and their data is also analyzed by an edge detector. Finally, a feature map is generated by combining the image and curvature maps. By finding areas of the 2D map that correspond to areas of the 3D model, semantic features can be preserved after simplification. We demonstrate this experimentally.","Title: Shrubbery-shell inspired 3D model stylization\nYear: 2019\n\nAbstract: •Present a modified particle-based branching method for automatically generating a shrubbery pattern that conforms to the input surface curvature that is applied directly to the 3D input mesh without requiring any basic tree element from existing sources.•Branching patterns can be controlled by the user’s brush on a partial area of input mesh, where the branch shapes reflects the brush trajectory.•Controllability for both global and local tree-branch shape by providing a set of parameters for particle-based growth algorithm.","Title: Particle-Based Dynamic Water Drops With High Surface Tension In Real Time\nYear: 2021\n\nAbstract: Surface tension has a great influence on the shape of the fluid interface, and is an important physical characteristic in expressing not only liquids but also liquid metals such as mercury and gallium. In the field of physics-based particle fluid simulations, it is a challenging problem to express the high surface tension generated by fluid-air or fluid-solid interaction in real time. The main reasons for this are (1) The magnitude of the force that can be stably expressed in real-time fluid simulation is limited, so when the magnitude of the surface tension increases at a large time-step, the simulation stability decreases, and (2) If we use a small time-step, a stronger force can be expressed. However, it becomes difficult to operate in real time because the computational cost increases. Techniques were proposed to solve this problem for a few specific scenes, but there has not yet been a general approach that can reliably express high surface tension in various scenarios. In this paper, we propose a real-time particle-based fluid simulation framework that can efficiently and stably express high surface tension. Unlike the previous methods, we newly model the surface tension so that the strong surface tension force generated in the droplet area with a large curvature is applied evenly in the normal and tangent directions regardless of the size of the droplet. We also propose new pressure constraints that converge quickly and accurately using this force. Our method can be effectively used in various physics-based simulation scenarios because it can easily express and control surface tension effects that appear in materials such as liquid metal as well as water.","Title: Automatic Chinese Meme Generation Using Deep Neural Networks\nYear: 2021\n\nAbstract: Internet memes have become widely used by people for online communication and interaction, particularly through social media. Interest in meme-generation research has been increasing rapidly. In this study, we address the problem of meme generation as an image captioning task, which uses an encoder-decoder architecture to generate Chinese meme texts that match image content. First, to train the model on the characteristics of Chinese memes, we collected a dataset of 3,000 meme images with 30,000 corresponding humorous Chinese meme texts. Second, we introduced a Chinese meme generation system that can generate humorous and relevant texts from any given image. Our system used a pre-trained ResNet-50 for image feature extraction and a state-of-the-art transformer-based GPT-2 model to generate Chinese meme texts. Finally, we combined the generated text and images to form common image memes. We performed qualitative evaluations of the generated Chinese meme texts through different user studies. The evaluation results revealed that the Chinese memes generated by our model were indistinguishable from real ones."],"string":"[\n \"Title: Discontinuous fluids\\nYear: 2005\\n\\nAbstract: At interfaces between different fluids, properties such as density, viscosity, and molecular cohesion are discontinuous. To animate small-scale details of incompressible viscous multi-phase fluids realistically, we focus on the discontinuities in the state variables that express these properties. Surface tension of both free and bubble surfaces is modeled using the jump condition in the pressure field; and discontinuities in the velocity gradient field. driven by viscosity differences, are also considered. To obtain derivatives of the pressure and velocity fields with sub-grid accuracy, they are extrapolated across interfaces using continuous variables based on physical properties. The numerical methods that we present are easy to implement and do not impact the performance of existing solvers. Small-scale fluid motions, such as capillary instability, breakup of liquid sheets, and bubbly water can all be successfully animated.\",\n \"Title: Bubbles alive\\nYear: 2008\\n\\nAbstract: We propose a hybrid method for simulating multiphase fluids such as bubbly water. The appearance of subgrid visual details is improved by incorporating a new bubble model based on smoothed particle hydrodynamics (SPH) into an Eulerian grid-based simulation that handles background flows of large bodies of water and air. To overcome the difficulty in simulating small bubbles in the context of the multiphase flows on a coarse grid, we heuristically model the interphase properties of water and air by means of the interactions between bubble particles. As a result, we can animate lively motion of bubbly water with small scale details efficiently.\",\n \"Title: Progressive reconstruction of 3D objects from a single free-hand line drawing\\nYear: 2003\\n\\nAbstract: This paper presents a progressive algorithm by which we can narrow down the searching domain for face identification and reconstruct various 3D objects from a single freehand line sketch drawing in short time. The sketch drawings served as an input for the reconstruction process in this paper are obtained from an inaccurate freehand sketch of a 3D wireframe object, which is the edge–vertex graph without any removal of hidden lines. The algorithm is executed through two stages. First, at the face identification stage, every possible faces are generated from the sketch and they are classified into implausible faces, basis faces and minimal faces by means of the geometrical and topological constraints in order to reduce the searching space. The proposed algorithm searches minimal space of the faces so that the actual faces of the given object could be identified in short time. Second, at the object reconstruction stage, we calculate a 3D structure progressively by optimizing the vertices coordinates of the object in accordance with the sketching order of faces. This progressive method reconstructs the most plausible 3D object in a short time by applying the 3D constraints that are derived from the relationship between an object and its sketch drawing in the optimization process. In addition, this method makes it possible for designers to change their viewpoints while sketching. In this paper, we are describing the progressive reconstruction algorithm, and giving the results of the implementation.\",\n \"Title: Divergence-constrained moving least squares for fluid simulation\\nYear: 2008\\n\\nAbstract: Developing suitable interpolation methods to simulate dynamic motions of continuous materials such as fluids is an important problem. In this paper, we propose a novel method to enforce the divergence condition to the interpolated velocity field by moving least squares (MLS), by means of the diffusive derivatives and moving divergence constraints that allow the practical use and easy implementation. As results, we present the velocity interpolation examples and a fluid-like particle simulation method to show the meaningful potential of our method as a tool of physical interpolation for fluid simulations. Copyright © 2008 John Wiley & Sons, Ltd.\",\n \"Title: User-Guided 3d su-muk painting\\nYear: 2005\\n\\nAbstract: We present a technique for rendering animated 3D models in a Su-Muk painting style with user's guide. First, a user can sketch directly over 3D models by varying ink and water values. And we simulate the behavior of ink-water directly on 3D model for consistent shade information over 3D models. After simulation, ink and water are spread over a 3D surface appropriately. Second, to achieve a real hand-crafted look of Su-Muk painting, ink-water behavior is simulated again on a 2D screen by using overall ink-water shade information from previous output of 3D ink-water simulation. We demonstrate some images for Su-Muk painting with our user-guided drawing system.\",\n \"Title: Progressive Compression of Geometry Information with Smooth Intermediate Meshes\\nYear: 2007\\n\\nAbstract: We present a new geometry compression algorithm for manifold 3D meshes based on octree coding. For a given mesh, regular volume grids are built with an adaptive octree. For each grid point, a binary sign, which indicates inside or outside of the mesh, is generated based on the distance to the mesh. In each leaf cell having a vertex, a least square fitting plane is created for a localized geometry range with signs. Finally, quantized geometry information is locally encoded. We demonstrate that the octree with signs can be used to predict the vertex positions. As a result, the proposed method generates competitive bitrates compared to the current state-of-art progressive geometry coder. Our method also shows better rate-distortion performance during decompression or transmission with improved smoothness.\",\n \"Title: Primitive Trees for Precomputed Distance Queries.\\nYear: 2013\\n\\nAbstract: We propose the primitive tree, a novel and compact space-partition method that samples and reconstructs a distance field with high accuracy, even for regions far from the surfaces. The primitive tree is based on the octree and stores the indices of the nearest primitives in its leaf nodes. Most previous approaches have involved a trade-off between accuracy and speed in distance queries, but our method can improve both aspects simultaneously. In addition, our method can sample unsigned distance fields effectively, even for self-intersecting and nonmanifold models. We present test results showing that our method can sample and represent large scenes, with more than ten million triangles, rapidly and accurately.\",\n \"Title: Automatic player behavior analysis system using trajectory data in a massive multiplayer online game\\nYear: 2013\\n\\nAbstract: This paper presents a new automated behavior analysis system using a trajectory clustering method for massive multiplayer online games (MMOGs). The description of a player's behavior is useful information in MMOG development, but the monitoring and evaluation cost of player behavior is expensive. In this paper, we suggest an automated behavior analysis system using simple trajectory data with few monitoring and evaluation costs. We used hierarchical classification first, then applied an extended density based clustering algorithm for behavior analysis. We show the usefulness of our system using trajectory data from the commercial MMOG World of Warcraft (WOW). The results show that the proposed system can analyze player behavior and automatically generate insights on players' experience from simple trajectory data.\",\n \"Title: Interchangeable SPH and level set method in multiphase fluids\\nYear: 2009\\n\\nAbstract: Subgrid-scale fluid is difficult to represent realistically in a grid-based fluid simulation. We show how to describe such small-scale details effectively, even on a coarse grid, by using escaped particles. The simulation of these particles with SPH (smooth particle hydrodynamics) allows the illustration of dynamic and realistic animation of fluids. Particles modeled by SPH have a force which leads them to merge if they are within a certain range. This reduces the accuracy of a simulation. Consequently, aggregated particles which form volumes large enough to be described by the level set method will be simulated inefficiently by particles. We address this problem with a new method in which details too small for the grid are represented by particles, while the level set method with a grid is used to describe merged particles on the grid.\",\n \"Title: Hardware-accelerated real-time rendering for 3D Sumi-e painting\\nYear: 2003\\n\\nAbstract: This paper presents a method for real-time 3D Sumi-e rendering using normal graphics hardware. Sumi-e is one of the traditional oriental painting styles. Most research on Sumi-e paintings has focused on 2D or 2.5D Sumi-e brushwork simulation. On these systems, complicated user's hand drawing is required to generate the image of Sumi-e effects, and it can render the 2D or 2.5D Sumi-e images only. We present an automated rendering system for 3D image of Sumi-e painting. It uses 3D common object as an input data and does not need any additional input of user brushwork. Especially for the real-time rendering, hardware-accelerated algorithm for Sumi-e rendering is newly suggested in our system. It is designed with efficiency for customer level graphics hardware. The results of this paper show that the features of traditional Sumi-e painting are successfully modeled and that 3D Sumi-e painting is rendered in real-time effectively.\",\n \"Title: Analytic solutions of integral moving least squares for polygon soups.\\nYear: 2012\\n\\nAbstract: This paper presents analytic solutions to the integral moving least squares (MLS) equations originally proposed by Shen et al. by choosing another specific weighting function that renders the numerator in the MLS equation unitless. In addition, we analyze the original method to show that their approximation surfaces (i.e., enveloping surfaces with nonzero values in the weighting function) often form zero isosurfaces near concavities behind the triangle-soup models. This paper also presents error terms for the integral MLS formulations against signed distance fields. Based on our analytic solutions, we show that our method provides both interpolation and approximation surfaces faster and more efficiently. Because our method computes solutions for integral MLS equations directly, it does not rely on numerical steps that might have numerical-accuracy issues. In particular, unlike the original method that deals with incorrect approximation surfaces by iteratively adjusting parameters, this paper proposes faster and more efficient approximations to surfaces without needing iterative routines. We also present computational efficiency comparisons, in which our method is 15-fold faster in computing integrations, even with conservative assumptions. Finally, we show that the surface normal vectors on the implicit surfaces formed by our analytic solutions are identical to the angle-weighted pseudonormal vectors.\",\n \"Title: System for evaluating usability and user experience by analyzing repeated patterns\\nYear: 2013\\n\\nAbstract: In this paper, a new system for evaluating interface usability through the analysis of repeated patterns is proposed. The system can be a valuable tool for verifying interfaces and in evaluating their usability by users, both of which are necessary stages in the development and operation of software. This paper concentrates on the repeated patterns that occur when users use an interface. Extracting these repeated patterns and analyzing them could enhance the development and usability of interfaces. Through experiments that applied the proposed system to several kinds of software, it was confirmed that problems with interfaces can be understood, and usability can be improved without requiring complicated analyses of user logs.\",\n \"Title: Fast Reconstruction of 3D Objects from Single Free-Hand Line Drawing\\nYear: 2001\\n\\nAbstract: This paper proposes an efficient algorithm that not only can narrow down the search domain of face identification but also can reconstruct various 3D objects from a single free-hand line drawing. The algorithm is executed in two stages. In the face identification stage, we generate and classify potential faces into implausible, basis and minimal faces by using geometrical and topological constraints to reduce search space. The proposed algorithm searches the space of minimal faces only to identify actual faces of an object fast. In the object reconstruction stage, we introduce 3D regularities and quadric face regularities to reconstruct 3D object accurately. Furthermore, the proposed method can be applied to a wide scope of general objects containing flat and quadric faces. The experimental results show that the proposed method identifies faces much faster than previous ones and efficiently reconstructs various objects from a single free-hand line drawing.\",\n \"Title: Simulation of swirling bubbly water using bubble particles\\nYear: 2009\\n\\nAbstract: The effect of surface tension is dynamically and realistically represented within a multiphase fluid simulation. Air bubbles are seeded with ‘bubble particles’ which move randomly. These molecule-like movements modify the surface of the air bubbles and generate turbulence in the water. The surface tension between air bubble and water, determined by the composition of the water, remains constant regardless of the size of the bubble, while external forces cause unstable fluid motion as the surface tension strives to remain constant, bubbles split and merge. The bubble particles can also compute for the numerical dissipation usually experienced in grid-based fluid simulations, by restoring the lost volume of individual bubbles. The realistic tearing of bubble surfaces is shown in a range of examples.\",\n \"Title: Multiresolution recognition of unconstrained handwritten numerals with wavelet transform and multilayer cluster neural network\\nYear: 1996\\n\\nAbstract: In this paper, we propose a new scheme for multiresolution recognition of unconstrained handwritten numerals using wavelet transform and a simple multilayer cluster neural network. The proposed scheme consists of two stages: a feature extraction stage for extracting multiresolution features with wavelet transform, and a classification stage for classifying unconstrained handwritten numerals with a simple multilayer cluster neural network. In order to verify the performance of the proposed scheme, experiments with unconstrained handwritten numeral database of Concordia University of Canada, Electro-Technical Laboratory of Japan, and Electronics and Telecommunications Research Institute of Korea were performed. The error rates were 3.20%, 0.83%, and 0.75%, respectively. These results showed that the proposed scheme is very robust in terms of various writing styles and sizes.\",\n \"Title: Reconstruction of Non-manifold Objects from Two Orthographic Views\\nYear: 1998\\n\\nAbstract: Since the two-view orthographic representation of a 3D object is ambiguous, it requires a numerous amount of combinatorial searches in the process of reconstruction. This paper presents an efficient algorithm for reconstructing polyhedral 3D objects from two-view drawings. The main feature of the algorithm is to improve the reconstruction process speed. First, the partially constructed objects are reconstructed from the restricted candidate faces corresponding to each area on the two-view drawings. Then the complete objects are obtained from the partially constructed objects by adding perpendicular faces with geometrical validity. By limiting the number of candidate faces corresponding to areas only, the combinatorial search space can be considerably reduced. In addition, the reconstruction finds the most plausible 3D object that human observers are most likely to select first among the given multiple solutions. Several examples from a working implementation are given to demonstrate the completeness of the algorithm.\",\n \"Title: Polygonal space carving with geometric anti-aliasing\\nYear: 2004\\n\\nAbstract: This paper proposes an algorithm that solves the shape recovery problem from N arbitrary images. By introducing a polygonal carving technique, the proposed algorithm can reconstruct the image-consistent polygonal shape that is patched by input images. This algorithm eliminates the invalid vertices and polygons from the initial polygonal grid space according to the color variance that represents their image consistency. The carved shape is refined by moving the outlier vertices on the boundary of each image. The final reconstructed shape faithfully accounts for the input images, and its textured appearance reflects the similar color property of the target object.\",\n \"Title: Self-correctional 3D shape reconstruction from a single freehand line drawing\\nYear: 2003\\n\\nAbstract: The goal of sketch reconstruction is to take an inaccurate, 2D edge-vertex graph (i.e., sketch drawing) as input and reconstruct a 3D shape as output. However, traditional reconstruction methods based on image regularities tend to produce a distorted 3D shape. In part, this distortion is due to the inherent inaccuracies in the sketch, but it also relates to the failure to accurately distinguish between important and less important regularities. We propose a new self-correctional reconstruction algorithm that can progressively produce refined versions of sketch reconstructions. The algorithm corrects the shape and the drawing simultaneously using geometric error metrics. The proposed algorithm can minimize the distortion of the shape by adding 3D regularities to the image regularities. The self-correctional algorithm for minimizing the distortion of sketch reconstruction is discussed, and the experimental results show that the proposed method efficiently reconstructs more accurate 3D objects than previous ones.\",\n \"Title: Adaptive synthesis of distance fields.\\nYear: 2012\\n\\nAbstract: We address the computational resource requirements of 3D example-based synthesis with an adaptive synthesis technique that uses a tree-based synthesis map. A signed-distance field (SDF) is determined for the 3D exemplars, and then new models can be synthesized as SDFs by neighborhood matching. Unlike voxel synthesis approach, our input is posed in the real domain to preserve maximum detail. In comparison to straightforward extensions to the existing volume texture synthesis approach, we made several improvements in terms of memory requirements, computation times, and synthesis quality. The inherent parallelism in this method makes it suitable for a multicore CPU. Results show that computation times and memory requirements are very much reduced, and large synthesized scenes exhibit fine details which mimic the exemplars.\",\n \"Title: Direct reconstruction of a displaced subdivision surface from unorganized points\\nYear: 2002\\n\\nAbstract: In this paper we propose a new mesh reconstruction algorithm that produces a displaced subdivision mesh directly from unorganized points. The displaced subdivision surface is a new mesh representation that defines a detailed mesh with a displacement map over a smooth domain surface. This mesh representation has several benefits - compact mesh size, piecewise regular connectivity - to overcome limitations of irregular mesh produced by ordinary mesh reconstruction scheme, but the original displaced subdivision surface generation algorithm needs an explicit polygonal mesh to be converted. Our approach is producing displaced subdivision surface directly from input points during the mesh reconstruction process. The main ideas of our algorithm are building initial coarse control mesh by the shrink-wrapping like projection and sampling fine surface detail from unorganized points along the each limit vertex normal without any connectivity information of given points. We employ an existing subdivision surface fitting scheme to generate a parametric domain surface, and suggest a surface detail sampling scheme that determines a valid sampling triangle which can be made with combinations of input points. We show several reconstruction examples and applications to show the validity of suggested sampling technique and benefits of the result like multiresolution modeling.\",\n \"Title: A Particle-Grid Method for Opaque Ice Formation.\\nYear: 2013\\n\\nAbstract: This paper presents a particle-grid method to simulate the generation of opaque ice which has air bubbles in it. Water temperature is diffused over a grid, and the exchange of dissolved air between ice and water particles is simulated. We render a particle as an air bubble if it has sufficient air. Otherwise, it is treated as a cloudy volume by distributing air into dissolved air field when the final state has been reached. In addition, our method includes a model in which heat transfer rate may change across the grid. Unlike previous models which could generate an ice volume of only fixed shapes, our approach uses signed distance function (SDF) to generate opaque ice volumes stored in containers of various geometric shapes and can render needle-shaped or egg-shaped bubbles.\",\n \"Title: Controllable multi-phase smoke with lagrangian particles\\nYear: 2006\\n\\nAbstract: This paper describes a method for controlling the multi-phase smoke animation that uses Lagrangian particles. Previous methods need several density fields to simulate different types of smoke. We use a single field, which is more like a natural phenomenon to obtain the interactive motions of fluid. Also, whereas existing methods which apply control forces to cells only, we define particle forces which enable each particle to move independently towards to target shape. Additionally, we set up internal target forces which distribute the particles uniformly within the target shape, no that it is represented more precisely.\",\n \"Title: Simulation of Morphology Changes in Drying Leaves.\\nYear: 2013\\n\\nAbstract: We introduce a biologically motivated simulation technique for the realistic shape deformation of drying leaves. In contrast to skeleton-based leaf deformation, our approach simulates the whole leaf surface to capture the fine details of desiccated leaves. We represent a leaf as a triangulated double-layer structure that consists of a Delaunay triangulation discretized along the vein structure and its corresponding Voronoi diagram. This structure can generate not only sharp creases along leaf veins, but also the complicated curling and crumpling on the leaf surface. The loss of water is the major factor that controls the inhomogeneous shrinkage of drying leaves. The proposed osmotic water flow successfully models the gradual changes of dehydrated regions advancing towards the veins. We demonstrate the robustness of our method by comparing a sequence of simulated morphology changes with photographs of real leaves.\",\n \"Title: Combustion Waves on the Point Set Surface.\\nYear: 2013\\n\\nAbstract: This paper introduces a combustion model of heat transfer and fuel consumption for the propagation of a fire front on a point cloud surface. The heat transfer includes the heat advection by the airflow as well as diffusion, chemical reaction, and heat loss to generate complex, but controllable heat flows with a designed airflow velocity. For the stable heat advection, we solve a semi-Lagrangian method on point samples using discrete exponential maps to trace the position from which the wind blows while preserving the geodesic distance. We also propose angular Voronoi weights for a discrete Laplace-Beltrami operator that shows better isotropic diffusion on the inhomogeneous distribution of point clouds than the cotangent or moving least-squares schemes. We demonstrate a diversity of burning scenarios by incorporating factors affecting the fire spreading such as buoyancy and object geometries in the airflow velocity fields, or by synthesizing patterns.\",\n \"Title: Finding ridges and valleys in a discrete surface using a modified MLS approximation\\nYear: 2005\\n\\nAbstract: Implicit surface fitting is a promising approach to finding ridges and valleys in discrete surfaces, but existing techniques are time-consuming and rely on user-supplied tuning parameters. We use a modified MLS (moving-least-squares) approximation technique to estimate the local differential information near a vertex by means of an approximating surface. Ridge and valley vertices are easily detected as zero-crossings, and can then be connected along the direction of principal curvature. Our method, demonstrated on several large meshed models, produces a good fit which leads to improved visualization. It does not oscillate and is quick to compute.\",\n \"Title: Finding ridges and valleys in a discrete surface using a modified MLS approximation [Computer-Aided Design 37 (14) 1533–1542]\\nYear: 2006\\n\\nAbstract: Implicit surface fitting is a promising approach to finding ridges and valleys in discrete surfaces, but existing techniques are time-consuming and rely on user-supplied tuning parameters. We use a modified MLS (moving-least-squares) approximation technique to estimate the local differential information near a vertex by means of an approximating surface. Ridge and valley vertices are easily detected as zero-crossings, and can then be connected along the direction of principal curvature. Our method, demonstrated on several large meshed models, produces a good fit which leads to improved visualization. It does not oscillate and is quick to compute. (c) 2005 Elsevier Ltd. All rights reserved.\",\n \"Title: Controlling shapes of air bubbles in a multi-phase fluid simulation\\nYear: 2012\\n\\nAbstract: Controlling shapes is a challenging problem in a multi-phase fluid simulation. Bubble particles enable the details of air bubbles to be represented within a simulation based on an Euler grid. We control the target shapes of bubbles by the gradient vectors of the signed distance field and attraction forces associated with control particles. Our hybrid approach enables to simulate physically plausible movements of bubbles while preserving the details of a target shape. Furthermore, we control the paths of moving bubbles using user-defined curves and the shape of an air bubbles by drag force. An accurate model of the drag force near the fluid surface means that bubbles have realistic ellipsoidal shapes.\",\n \"Title: A feature-preserved simplification for autonomous facial animation from 3D scan data\\nYear: 2003\\n\\nAbstract: We propose a new simplification algorithm of facial models for animation. For the facial animation, the models are often simplified from complex scan data based on geometric features, but it leads to decrease the quality and such features are easily noticed by human perception. For example, a lip line and eyebrows easily lose their details by geometry-based simplification. In this paper, facial features are extracted using an image processing of a 2D texture image and the curvature analysis of the 3D geometry, which improves the details around the feature areas of the facial model. Especially if lip contact line is simplified to one or two edges, it may not be proper for lip animation. Finally, we will show that our simplified model can produce as good as a facial animation as the one from the original model.\",\n \"Title: Target-driven liquid animation with interfacial discontinuities\\nYear: 2007\\n\\nAbstract: We propose a novel method of controlling a multi-phase fluid so that it flows into a target shape in a natural way. To preserve the sharp detail of the target shape, we represent it as an implicit function and construct the level-set of that function. Previous approaches add the target-driven control force as an external term, which then becomes attenuated during the velocity projection step, making the convergence process unstable and causing sharp detail to be lost from the target shape. But we calculate the force on the fluid from the pressure discontinuity at the interface between phases, and integrate the control force into the projection step so as to preserve its effect. The control force is calculated using an enhanced version of the ghost fluid method (GFM), which guarantees that the fluid flows from the source shape and converges into the target shape, while achieving a more natural animation than other approaches. Our control force is merged during the projection step avoiding the need for a post-optimization process to eliminate divergence at the liquid interface. This makes our method easy to implement using existing fluid engines and it incurs little computational overhead. Experimental results show the accuracy and robustness of this technique. Copyright © 2007 John Wiley & Sons, Ltd.\",\n \"Title: Sports equipment based motion deformation\\nYear: 2005\\n\\nAbstract: We propose a novel scheme for generating the locomotion of sports equipment along with motion deformation. To accomplish this, we search the event frames in order to find the points where the character is in contact with the sports equipment (contact points) or not (split points). To find these points, it is necessary to locate the corresponding event points in the signal using Discrete Fourier Transformation. This technique generates a left and right hand-cross analysis and sports equipment locomotion adapted to the character motion. Also, the locomotion of the sports equipment is changed by adjusting its properties, and this may require the posture of the character to be modified. In conclusion, we are able to deform the motion of both the sports equipment and the character simply by adjusting the properties of the sports equipment. Our scheme can be used for the motion deformation of any character with an associated sports equipment.\",\n \"Title: Design and evaluation of a hybrid display system for motion-following tasks\\nYear: 2007\\n\\nAbstract: Hybrid display systems are those that combine different types of displays to exploit the complementary characteristics of the constituent display systems. In this paper, we introduce a hybrid system that combines a stereoscopic optical see-through head-mounted display (HMD) and a large projection display for an application in a multi-user ship painting training scenario. The proposed hybrid system's projection display provides a large FOV and a physical metaphor to the ship surface with natural depth perception, while the HMD provides personal and unoccluded display of the motion training guides. To quantify its effectiveness, we conducted a human subject experiment, comparing the subject's motion following task performance among three different display systems: large projection display, head-mounted display, and hybrid. The preliminary results obtained from the experiment has shown that given the same FOV, the hybrid system performed, despite problems with registration between the real and virtual worlds, up to par with the head-mounted display, and better than the projection display. Thus, it is expected that the hybrid display will result in higher task performance with the larger FOV factor available.\",\n \"Title: Adaptive fuzzy logic for focusing in optical disk player\\nYear: 2008\\n\\nAbstract: For the nonlinear behaviors of the actuator and servo system of the optical disk player, a fuzzy control is introduced. Focus servo controls the laser beam to focus on the playing surface of the disc by vertical movement of the objective lens. The change of the distance between the objective lens and the disc must be within the allowable focus error to stay in focus. This must be maintained even with a slightly warped or uneven disc and in a portable player. This paper will observe on how the focusing is controlled in an optical disk player and will evaluate algorithm based on the model using Adaptive Fuzzy Logic.\",\n \"Title: Progressive 3D Reconstruction from a Sketch Drawing\\nYear: 2001\\n\\nAbstract: This paper presents a progressive algorithm for reconstructing a 3D structure from a given 2D sketch drawing (edge-vertex graph without hidden line removal) according to the user's sketch order. While previous methods reconstruct a 3D structure at once, the proposed method progressively calculate a 3D structure by optimizing the coordinates of vertices of an object according to the sketch order. The progressive method reconstructs the most plausible 3D object quickly by applying 3D constraints that are derived from the relationship between the object and the sketch drawing in the optimization process. Furthermore, it allows the user to change viewpoint during sketching, and also minimize distortion of an object by refining inaccurate sketch drawings. The progressive reconstruction algorithm is discussed, and examples from a working implementation are given.\",\n \"Title: Extracting feature lines from unstructured meshes using a model feature map\\nYear: 2006\\n\\nAbstract: We propose a new method of extracting feature lines from a 3D scan model. Given an unstructured 3D triangle mesh containing texture information, we use model feature map (MFM), which is a 2D map that abstracts the variation of texture and curvature in the 3D model. The MFM is created from both a texture map and a curvature map, which are produced separately by edge-detection to locate line features. The MFM can be edited interactively using standard image-processing tools. We demonstrate the technique on several data sets, including, but not limited to facial data.\",\n \"Title: Systematic Reconstruction Of 3d Curvilinear Objects From Two-View Drawings\\nYear: 1999\\n\\nAbstract: This paper presents an efficient algorithm for reconstructing curvilinear 3D objects from two-view drawings. The main features of the algorithm are to improve the speed of the reconstruction process and to generate a sequence of solutions. First, the partially constructed objects are reconstructed from the restricted candidate faces corresponding to each area in the two-view drawings. Then the complete objects are obtained from the partially constructed objects by adding perpendicular faces with geometric validity. By limiting the number of candidate faces corresponding to areas only, the computational time of the combinatorial search process is considerably reduced. Second, to find more plausible solutions prior to unusual ones, the algorithm generates the sequence number to multiple solutions based on a plausibility function that maps the proposed solution to the designer's intention in the original drawings. Furthermore, the proposed method of reconstruction supports a wide scope of general (manifold and non-manifold) 3D objects containing cylindrical, conical, spherical, and toroidal faces. Several examples from a working implementation are given to demonstrate the completeness of the algorithm. (C) 1999 Elsevier Science Ltd. All rights reserved.\",\n \"Title: Live path: adaptive agent navigation in the interactive virtual world\\nYear: 2010\\n\\nAbstract: We present a novel approach to adaptive navigation in the interactive virtual world by using data from the user. Our method constructs automatically a navigation mesh that provides new paths for agents by referencing the user movements. To acquire accurate data samples from all the user data in the interactive world, we use the following techniques: an agent of interest (AOI), a region of interest (ROI) map, and a discretized path graph (DPG). Our method enables adaptive changes to the virtual world over time and provides user-preferred path weights for smart-agent path planning. We have tested the usefulness of our algorithm with several example scenarios from interactive worlds such as video games. In practice, our framework can be applied easily to any type of navigation in an interactive world. In addition, it may prove useful for solving previous pathfinding problems in static navigation planning.\",\n \"Title: CUDA-based Signed Distance Field Calculation for Adaptive Grids\\nYear: 2010\\n\\nAbstract: We present a simple yet robust signed distance field (SDF) generator based on recent GPU architectures. In our approach, the squared Euclidean distance is calculated for each triangle face in parallel, and then an optimized stream reduction process is used to find the shortest distance. During this process, the stream reduction operation acts like a parallel binary space-searching routine for each level. This process uses computations and memory bandwidth efficiently because of the massive number of CUDA threads. Signs are then determined by calculating angle-weighted pseudonormals. Unlike some previous SDF approaches that only calculate the SDF near the surface or within the bounding box, our method can calculate the SDF adaptively so that there are no limitations on proximity or regularity. We also compare our GPU-based results with a kd-tree based single CPU approach for a 3D geometry synthesis application.\",\n \"Title: CDP Servo System Control using Fuzzy Logic Control\\nYear: 2007\\n\\nAbstract: Several servo mechanisms, such as focusing, spindling, tracking and sledding, are continuously balanced to make compact disc player (CDP) to play audio or video contents. However, due to the inconsistent uneven surface of the disc, the speed control of a spindle motor along the track radius, and unexpected random shock; the servo mechanism is very complicatedly driven in non-linear method. This paper will discuss on how to approach this method, how to develop and how to evaluate each algorithm based on the model of each servo mechanism using Fuzzy Logic Control1.\",\n \"Title: Extraction of ridges-valleys for feature-preserving simplification of polygonal models\\nYear: 2006\\n\\nAbstract: We propose a new method for simplifying polygonal models while retaining salient features. We extract ridge and valley features defined by means of curvature derivatives on the model surface. We combine this extraction with a simplification technique so as to construct coarse models with features preserved. Experimental results have better quality and smaller geometric error than these of previous methods.\",\n \"Title: Adaptive space carving with texture mapping\\nYear: 2005\\n\\nAbstract: Space carving reconstructs a 3D object from multiple images, but existing algorithms rely on a regular grid which makes poor use of memory. By using the image information, adaptive space carving uses a recursively generated structure which reduces memory requirements and thus allows a finer grid. After reconstruction, models are triangulated to facilitate texture mapping. Experimental results show the enhanced appearance of models reconstructed in this way.\",\n \"Title: Surface simplification using a discrete curvature norm\\nYear: 2002\\n\\nAbstract: This paper proposes a mesh simplification algorithm using a discrete curvature norm. Most of the simplification algorithms are using a distance metric to date. The distance metric is very efficient to measure geometric error, but it is difficult to distinguish important shape features such as a high-curvature region even though it has a small distance metric. We suggest a discrete curvature norm to measure geometric error for such features. During simplification the new vertex resulted from an edge collapse takes a position using a butterfly subdivision mask to minimize geometric error. This paper shows that simplification results have smaller geometric errors than previous works, when a discrete curvature norm and a distance metric are together applied to its criterion.\",\n \"Title: Shrinkage, wrinkling and ablation of burning cloth and paper\\nYear: 2011\\n\\nAbstract: The burning of a sheet of cellulose-based material, such as paper or cloth, involves uneven shrinkage which causes wrinkling. We simulate this geometrically complicated phenomenon by modeling the effects of heat transfer, shrinkage and partial ablation on a thin shell. A strain-limitation technique is applied to a two-layer structure of springs arranged as a body-centered square. Although this structure is over-constrained, convergence can be achieved using a new successive fast projection method. We also remesh the shells dynamically to deal with the topological changes that occur as regions burn away.\",\n \"Title: Reconstructing 3D models with algebraic curved surfaces from three-view drawings\\nYear: 1996\\n\\nAbstract: A system that reconstructs 3D models with algebraic curved surfaces from three orthographic views is described. In this paper, we present two features. First, we can directly reconstruct 3D models with algebraic curved surfaces without using a method of polyhedron approximation by introducing two kinds of supplementary lines: silhouette lines and tangency lines. Second, we develop new rules and heuristics to handle curved surfaces, used in combinatorial search of faces. Rules work to reduce the processing time, and heuristics are aiming at simulation of the way human interprets the three orthographic views. Soundness and efficiency of the proposed method are also proved experimentally\",\n \"Title: Evolving Physically Simulated Flying Creatures for Efficient Cruising\\nYear: 2006\\n\\nAbstract: The body-brain coevolution of aerial life forms has not been developed as far as aquatic or terrestrial locomotion in the field of artificial life. We are studying physically simulated 3D flying creatures by evolving both wing shapes and their controllers. A creature's wing is modeled as a number of articulated cylinders, connected by triangular films (patagia). The wing structure and its motor controllers for cruising flight are generated by an evolutionary algorithm within a simulated aerodynamic environment. The most energy-efficient cruising speed and the lift and drag coefficients of each flier are calculated from its morphological characteristics and used in the fitness evaluation. To observe a wide range of creature size, the evolution is run separately for creatures categorized into three species by body weight. The resulting creatures vary in size from pigeons to pterosaurs, with various wing configurations. We discuss the characteristics of shape and motion of the evolved creatures, including flight stability and Strouhal number.\",\n \"Title: RViz: a toolkit for real domain data visualization\\nYear: 2015\\n\\nAbstract: In computational science and computer graphics, there is a strong requirement to represent and visualize information in the real domain, and many visualization data structures and algorithms have been proposed to achieve this aim. Unfortunately, the dataflow model that is often selected to address this issue in visualization systems is not flexible enough to visualize newly invented data structures and algorithms because this scheme can accept only specific data structures. To address this problem, we propose a new visualization tool, RViz, which is independent of the input information data structures. Since there is no requirement for additional efforts to manage the flow networks and the interface to abstracted information is simple in RViz, any scientific information visualization algorithms are easier to implement than the dataflow model. In this paper, we provide case studies in which we have successfully implemented new data structures and related algorithms using RViz, including geometry synthesis, distance field representation, and implicit surface reconstruction. Through these cases, we show how RViz helps users visualize and understand any hidden insights in input information.\",\n \"Title: Controlling liquids using pressure jump\\nYear: 2006\\n\\nAbstract: This sketch presents a method to control liquids so that they flow into a target shape in a natural way. To avoid the oscillation and preserve the sharp detail of the target shape, we integrate the target-driven force into the projection step. So we require no post-optimization process for divergence-free condition at the liquid interface. Additionally, the target-driven force is calculated by enhanced ghost fluid method considering the pressure discontinuity around the liquid interface.\",\n \"Title: Generating flying creatures using body-brain co-evolution\\nYear: 2003\\n\\nAbstract: This paper describes a system that produces double-winged flying creatures using body-brain co-evolution without need of complex flapping flight aerodynamics. While artificial life techniques have been used to create a variety of virtual creatures, little work has explored flapping-winged creatures for the difficulty of genetic encoding problem of wings with limited geometric primitives as well as flapping-wing aerodynamics. Despite of the simplicity of system, our result shows aesthetical looking and organic flapping flight locomotions. The restricted list structure is used in genotype encoding for morphological symmetry of creatures and is more easily handled than other data structures. The creatures evolved by this system have two symmetric flapping wings consisting of continuous triangular patches and show various looking and locomotion such as wings of birds, butterflies and bats or even imaginary wings of a dragon and pterosaurs.\",\n \"Title: Physics-inspired controllable flame animation.\\nYear: 2016\\n\\nAbstract: We propose a novel method conceptualized from the properties of physics where in particular the shape of a flame is determined by temperature that enables a control mechanism for the intuitive shaping of a flame. We focused on a trade-off issue from computer graphics whereby the turbulent flow that expresses the characteristics of the flame has a tendency to shift continuously, whereas the velocity constraints that contain a fluid within a target shape have a tendency to force movement in a particular direction. Trade-off made it difficult for animation designers to maintain a flame within the intended target shape. This paper resolves the issue by enabling the flame to be controlled without any velocity constraints by using the following two techniques: First, we model the temperature and force of the explosion generated by the combustion of explosive gaseous fuel and apply it to certain regions. Second, we expand the space of the interface between the fuel and the burned products, classifying that space into four regions and controlling the target shape of the flame by delicate adjustments to the temperature in each region. Experiments show that the flame maintains the appearance of dynamic movement while preserving the detailed 3D shapes specified by the scene designers.\",\n \"Title: Subtle features of ice with cloudy effects and scratches from collision damage\\nYear: 2016\\n\\nAbstract: AbstractWe propose a simulation framework for expressing cloudy effects and scratches on ice caused by the collision of other objects. Cloudy effects are created and diffused in proportion to the collision on the ice surfaces calculated by a combination of a grid-projection technique and the boundary particle method. To prevent dissipation during the diffusion process, a geodesic distance is used as a constraint. Scratches are modeled directionally by analyzing the density gradient of cloudy effects and rendered using needle-shaped ellipsoids. Experiments provide detailed expression of cloudy effects and scratches from collision damage. Copyright © 2016 John Wiley & Sons, Ltd.\",\n \"Title: Let's be a virtual juggler.\\nYear: 2016\\n\\nAbstract: Juggling, which uses both hands to keep several objects in the air at once, is admired by anyone who sees it. However, skillful real-world juggling requires long, hard practice. Therefore, we propose an interesting method to enable anyone to juggle skillfully in the virtual world. In the real world, the human motion has to follow the motion of the moving objects; in the virtual world, the objects' motion can be adjusted together with the human motion. By using this freedom, we have generated a juggling avatar that can follow the user's motion. The user simply makes juggling-like motions in front of a motion sensor. Our system then searches for juggling motions that closely match the user's motions and connects them smoothly. We then generate moving objects that both satisfy the laws of physics and are synchronized with the synthesized motion of the avatar. In this way, we can generate a variety of juggling animations by an avatar in real time. Copyright © 2016 John Wiley & Sons, Ltd.\",\n \"Title: Visual simulation of fire-flakes synchronized with flame.\\nYear: 2017\\n\\nAbstract: We propose a framework for generating, animating, and controlling fire-flakes to correspond with the movements of flame. Not only do fire-flakes themselves display a distinctively complex and elaborate movement, but also they are heavily influenced by where the heat transforms into flame and how the flame moves. It is difficult for designers to generate and simulate synchronous movement of fire-flakes and flame because the movement of flame tends to be chaotic and constantly changing. We provide a novel framework that can improve on the current processes both effectively and efficiently. We first propose a fire-flake motion model that simulates movement harmoniously with the movement of flame. We then introduce automatic generation of fire-flakes by analyzing temperature fields and the velocity fields input from the fire simulation. Lastly, our sample-based control method enables the formation of a target shape by gathering fire-flakes together into one location while maintaining their inherent movement without external forces. We also provide various parameters that the designer can use to control the effects of the forces determining the amount, velocity, and movement of fire-flakes. To the best of our knowledge, this work is the first visual simulation method to generate fire-flakes in coordination with flame, and we believe it will contribute not only to research, but also to the animation industry.\",\n \"Title: Incorporating particle motion into an ADF for fast coupling of fluids with rigid and deformable solids.\\nYear: 2017\\n\\nAbstract: We present a new method for the fast simulation of interactions between fluids and solids by incorporating particle-based water flow into an adaptive signed distance field. In some previous methods, the motion of every water particle is checked when simulating the collision with the solid in the coupling process, and the computational cost becomes very great as the number of particles increases. If only the particles on the leaf nodes surrounding the solid are considered, this reduces the computational cost of collision detection, but some collisions may not be detected. This may lead to the tunneling artifact, in which particles with high velocities skip across the layer of leaf nodes. This paper addresses the problem by (i) considering particles only on the leaf nodes in the adaptive structure to improve the processing time required for the water-solid coupling and (ii) considering the water flow to avoid the tunneling artifact by incorporating particle motion into the tree structure of the adaptive signed distance field. Our method can be computed in parallel, and experimental results show that it outperforms previous methods while producing animations that are largely free of artifacts. Copyright (c) 2016 John Wiley & Sons, Ltd.\",\n \"Title: Visual simulation of rapidly freezing water based on crystallization.\\nYear: 2017\\n\\nAbstract: We propose a physics-inspired simulation framework that expresses visual effects of flowing water frozen in glaze or directional icicles. The proposed ice model considers the direction of the water flow, which affects the growth of icicles. Water dynamics are computed using a conventional particle-based simulation. Ice glaze and directional icicles are generated by incorporating our freezing solver. To determine whether a water particle is converted into ice or remains liquid, we compute the nucleation energy based on the humidity and water flow. The humidity is approximated as a virtual water film on object surfaces. The water flow is incorporated by introducing a growth direction vector to guide the direction of icicle growth. Ice-generating regions can be controlled using 3-D modeling tools such as Autodesk Maya or 3DS Max. Experiments showed that a realistic ice glaze was created on the surfaces of objects and that icicles grew in the direction of the water flow.\",\n \"Title: A hybrid SDF for the detailed representation of liquid-solid mixed surfaces\\nYear: 2015\\n\\nAbstract: We propose a hybrid signed distance field SDF method for reconstructing the detailed surface of a model as it changes from a solid state to a liquid state. Previous particle-based fluid simulations suffer from a noisy surface problem when the particles are distributed irregularly. If a smoothing scheme is applied to reduce the problem, sharp and detailed features can be lost by over-smoothing artifacts. Our method constructs a hybrid SDF by combining level-set values from the solid and liquid parts of the object. This makes it possible to represent the detailed features and smooth surfaces of an object when both solid and liquid parts are mixed in that object. In addition, the concept of a guiding shape is proposed, which uses a coordinate-warping technique to query the level-set values quickly. The guiding shape is constructed from the object before the simulation begins and some parts of it become liquid. To track the details of the initial solid shape and preserve it, the transformation of the guiding shape is accumulated while the phase-shift is in progress. By warping the coordinates of this accumulated transformation of the guiding shape, the level-set values of the solid part can be acquired very quickly. Copyright © 2015John Wiley & Sons, Ltd.\",\n \"Title: Efficient Representation of Detailed Foam Waves by Incorporating Projective Space.\\nYear: 2017\\n\\nAbstract: We propose an efficient framework to realistically simulate foam effects in which 3D water particles from a base water solver are first projected onto 2D screen space in order to reduce computational complexity of finding foam particles. Because foam effects are often created primarily in fast and complicated water flows, we analyze acceleration and curvature values to identify the areas exhibitin...\",\n \"Title: Realistic Visual Simulation of Water Effects in Response to Human Motion using a Depth Camera.\\nYear: 2017\\n\\nAbstract: In this study, we propose a new method for simulating water responding to human motion. Motion data obtained from motion-capture devices are represented as a jointed skeleton, which interacts with the velocity field in the water simulation. To integrate the motion data into the water simulation space, it is necessary to establish a mapping relationship between two fields with different properties. However, there can be severe numerical instability if the mapping breaks down, with the realism of the human-water interaction being adversely affected. To address this problem, our method extends the joint velocity mapped to each grid point to neighboring nodes. We refine these extended velocities to enable increased robustness in the water solver. Our experimental results demonstrate that water animation can be made to respond to human motions such as walking and jumping.\",\n \"Title: An optimized real time algorithm for window frost formation suited to mobile devices.\\nYear: 2018\\n\\nAbstract: We propose a real time simulation for window frost formation on mobile devices that uses both particles and grids. Previous ice formation methods made heavy demands on both memory and computational capacity because they were designed for a desktop environment. In this paper, a frost skeleton grows around a location touched by the user using particles, and the ice surfaces are constructed using a grid. Using a nonlattice random-walk technique, the frost skeleton grows freely and naturally. A hash grid technique is used to search efficiently for neighbor particles during the crystallization process. Finally, some 2.5D details are added to the ice skeleton by adjusting the height of the grid vertices around the skeleton. Experiments show that our method creates realistic frost in real time. Our method can be used to express ice formation effects in touch-based mobile device applications such as weather forecasts or games.\",\n \"Title: Predicting the Currency Market in Online Gaming via Lexicon-Based Analysis on Its Online Forum.\\nYear: 2017\\n\\nAbstract: Transactions involving virtual currencies are becoming increasingly common, including those in online games. In response, predicting the market price of a virtual currency is an important task for all involved, but it has not yet attracted much attention from researchers. This paper presents user opinions from online forums in a massive multiplayer online game (MMOG) setting widely used around the world. We propose a method for predicting the next-day rise and fall of the currency used in an MMOG environment. Based on analysis of online forum users' opinions, we predict daily fluctuations in the price of a currency used in an MMOG setting. Focusing specifically on the World of Warcraft game, one of the most widely used MMOGs, we demonstrate the feasibility of predicting the fluctuation in value of virtual currencies used in this game community.\",\n \"Title: Surface simplification with semantic features using texture and curvature maps\\nYear: 2005\\n\\nAbstract: We propose a polygonal surface simplification algorithm that can preserve semantic features without user control. The semantic features of a model are important for human perception, which are insensitive to small geometric errors. Using an edge detects: Its three kinds of maps are employed to extract these features. First, an image map is generated boundary lines represent changes of chroma in the texture image by using edge detector. Second, the discrete curvatures at 3D vertices are mapped to the curvature map, and their data is also analyzed by an edge detector. Finally, a feature map is generated by combining the image and curvature maps. By finding areas of the 2D map that correspond to areas of the 3D model, semantic features can be preserved after simplification. We demonstrate this experimentally.\",\n \"Title: Shrubbery-shell inspired 3D model stylization\\nYear: 2019\\n\\nAbstract: •Present a modified particle-based branching method for automatically generating a shrubbery pattern that conforms to the input surface curvature that is applied directly to the 3D input mesh without requiring any basic tree element from existing sources.•Branching patterns can be controlled by the user’s brush on a partial area of input mesh, where the branch shapes reflects the brush trajectory.•Controllability for both global and local tree-branch shape by providing a set of parameters for particle-based growth algorithm.\",\n \"Title: Particle-Based Dynamic Water Drops With High Surface Tension In Real Time\\nYear: 2021\\n\\nAbstract: Surface tension has a great influence on the shape of the fluid interface, and is an important physical characteristic in expressing not only liquids but also liquid metals such as mercury and gallium. In the field of physics-based particle fluid simulations, it is a challenging problem to express the high surface tension generated by fluid-air or fluid-solid interaction in real time. The main reasons for this are (1) The magnitude of the force that can be stably expressed in real-time fluid simulation is limited, so when the magnitude of the surface tension increases at a large time-step, the simulation stability decreases, and (2) If we use a small time-step, a stronger force can be expressed. However, it becomes difficult to operate in real time because the computational cost increases. Techniques were proposed to solve this problem for a few specific scenes, but there has not yet been a general approach that can reliably express high surface tension in various scenarios. In this paper, we propose a real-time particle-based fluid simulation framework that can efficiently and stably express high surface tension. Unlike the previous methods, we newly model the surface tension so that the strong surface tension force generated in the droplet area with a large curvature is applied evenly in the normal and tangent directions regardless of the size of the droplet. We also propose new pressure constraints that converge quickly and accurately using this force. Our method can be effectively used in various physics-based simulation scenarios because it can easily express and control surface tension effects that appear in materials such as liquid metal as well as water.\",\n \"Title: Automatic Chinese Meme Generation Using Deep Neural Networks\\nYear: 2021\\n\\nAbstract: Internet memes have become widely used by people for online communication and interaction, particularly through social media. Interest in meme-generation research has been increasing rapidly. In this study, we address the problem of meme generation as an image captioning task, which uses an encoder-decoder architecture to generate Chinese meme texts that match image content. First, to train the model on the characteristics of Chinese memes, we collected a dataset of 3,000 meme images with 30,000 corresponding humorous Chinese meme texts. Second, we introduced a Chinese meme generation system that can generate humorous and relevant texts from any given image. Our system used a pre-trained ResNet-50 for image feature extraction and a state-of-the-art transformer-based GPT-2 model to generate Chinese meme texts. Finally, we combined the generated text and images to form common image memes. We performed qualitative evaluations of the generated Chinese meme texts through different user studies. The evaluation results revealed that the Chinese memes generated by our model were indistinguishable from real ones.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":62,"string":"62"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13585,"string":"13,585"}}},{"rowIdx":13586,"cells":{"question_id":{"kind":"string","value":"Lixin Fan"},"question":{"kind":"string","value":"Generate an abstract for the title \"Augmented and interactive video playback based on global camera pose\" using the following items: 1. Video playback system\n2. Field of view expansion\n3. Augmented reality\n4. Geo-metadata extraction\n5. Point-of-interest (POIs) data"},"answer":{"kind":"string","value":"This paper proposes a video playback system that allows user to expend the field of view to surrounding environments that are not visible in the original video frame, arbitrarily change the viewing angles, and see the superimposed point-of-interest (POIs) data in an augmented reality manner during the video playback. The processing consists of two main steps: in the first step, client uploads a video to the GeoVideo Engine, and then the GeoVideo Engine extracts the geo-metadata and returns them back to the client; in the second step, client requests POIs from server, and then the client renders the video with POIs."},"correct_answer":{"kind":"string","value":"This paper proposes a video playback system that allows user to expend the field of view to surrounding environments that are not visible in the original video frame, arbitrarily change the viewing angles, and see the superimposed point-of-interest (POIs) data in an augmented reality manner during the video playback. The processing consists of two main steps: in the first step, client uploads a video to the GeoVideo Engine, and then the GeoVideo Engine extracts the geo-metadata and returns them back to the client; in the second step, client requests POIs from server, and then the client renders the video with POIs."},"documents":{"kind":"list like","value":["Title: An efficient parallel algorithm for hexagonal-based fractal image compression\nYear: 2007\n\nAbstract: A two-level fast search algorithm to reduce the encoding time for hexagonal-based fractal image compression is presented. The design of the sequential algorithm is based on the distribution of matched domains in a given image. The first search level previews various portions of the image and identifies promising domains among all possible domains. The second search level picks out domain blocks in the image portion where the first level gives positive results, and compares them with a given range block for encoding. The algorithm is parallelized by a dynamic range distribution scheme to achieve load balancing. Experimental results show that by running the parallelized encoding algorithm on multiple processors, the encoding time is drastically reduced while the quality of image reconstruction is retained. A speed-up of about 9 can be obtained by using 13 processors.","Title: Wavelet diffusion for document image denoising\nYear: 2003\n\nAbstract: To distinguish image features from high level noises, wepropose to exploit the spatial correlations between waveletcoefficients by replacing the thresholding process with a diffusionprocess. The proposed smoothing method is more robustto noise when applied to highly corrupted documentimages.","Title: Adaptive image restoration using structure tensors\nYear: 2002\n\nAbstract: This paper proposes a structure tensor based image restoration algorithm. Structure tensors capture the structural and textural distribution of similar pixels at each site. This property adapts the smoothing process to the contrast, orientation and spatial size of local image structures.","Title: Pedestrian registration in static images with unconstrained background\nYear: 2003\n\nAbstract: This paper introduces a human body contour registration method for static pedestrian images with unconstrained backgrounds. By using a statistical compound model to impose structural and textural constraints on valid pedestrian appearances, the matching process is robust to image clutter. Experimental results show that the proposed method register pedestrian contours in complex backgrounds effectively.","Title: On-line Object Reconstruction and Tracking for 3D Interaction\nYear: 2012\n\nAbstract: This paper presents a flexible and easy-to-use tracking method for 3D interaction. The method reconstructs points of a user-specified object from a video sequence, and recovers the 6 degrees of freedom (DOF) camera pose and position relative to the reconstructed points in each video frame. As opposed to most existing 3D object tracking methods, the proposed method does not need any off-line modeling or training process. Instead, it first segments the object from the background, then reconstructs and tracks the object using the Visual Simultaneous Localization And Mapping (VSLAM) techniques. To our knowledge, there are no existing works investigating this kind of on line reconstruction and tracking of moving objects. The proposed method employs the adapted pyramidal Lucas-Kanade tracker to increase the stability and the robustness of the tracking when dealing with a lightly textured or fast moving object. Experiments show that fast, accurate, stable and robust tracking can be achieved in everyday environment. Moreover, a simple stereo initialization approach is adopted to minimize user intervention. All these attributes conspire to make the method an adequate tool for some interaction applications. As a concrete example, an interactive 3D scene displaying system is demonstrated.","Title: Head-tracking virtual 3-D display for mobile devices.\nYear: 2012\n\nAbstract: Computer vision enables camera data to be utilized in user interfaces to analyze the 3-D context and automatically detect the user intentions. Using cameras as an input modality provides single-handed operations in which the users' actions are recognized without interactions with the screen or keypad. In this context, we have constructed a real-time mobile application prototype where the user's position and gaze is determined in real time, a technique that enables the display of true three-dimensional objects even on a typical 2-D LCD screen. We have defined a series of interaction methods where the user's motion and camera input realistically control the viewpoint on a 3-D scene. The head movement and gaze can be used to interact with hidden objects in a natural manner just by looking at them. We provide a description of the embedded implementation at a system-level where we highlight the application development challenges and trade-offs that need to be dealt with battery powered mobile devices. The implementation includes a parallel pipeline that reduces the latencies of the application.","Title: Animation pyramid for efficient image database browsing\nYear: 2010\n\nAbstract: This paper introduces an image database organization method as well as dedicated user interface (UI) for efficient browsing of large image databases. The proposed method and UI are particularly useful for image browsing on small screens of mobile devices.","Title: Semantic Parsing of Street Scene Images Using 3D LiDAR Point Cloud\nYear: 2013\n\nAbstract: In this paper we propose a novel street scene semantic parsing framework, which takes advantage of 3D point clouds captured by a high-definition LiDAR laser scanner. Local 3D geometrical features extracted from subsets of point clouds are classified by trained boosted decision trees and then corresponding image segments are labeled with semantic classes e.g. buildings, road, sky etc. In contrast to existing image-based scene parsing approaches, the proposed 3D LiDAR point cloud based approach is robust to varying imaging conditions such as lighting and urban structures. The proposed method is evaluated both quantitatively and qualitatively on three challenging NAVTEQ True databases and robust scene parsing results are reported.","Title: Efficient Random Sampling for Nonrigid Feature Matching\nYear: 2009\n\nAbstract: This paper aims to match two sets of nonrigid feature points using random sampling methods. By exploiting the principle eigenvector of corres pondence-model-linkage, an adaptive sampling method is devised to efficiently deal with non-rigid matching problems.","Title: A Feature-Based Object Tracking Method Using Online Template Switching and Feature Adaptation.\nYear: 2011\n\nAbstract: This paper proposes a robust tracking method which stores representative object appearances as candidate templates during tracking, and selects the best template to match new frames. This online template adding and switching strategy, in one aspect, keeps update with new object appearances. In yet another aspect, it is resilient to misaligned templates and alleviates the drifting problem. The method is able to achieve real time speed on a laptop PC, and the tracking is robust to significant image variations including illumination changes, deformation, pose changes and occlusions.","Title: A diffusion process for wavelet-transform-based image denoising and its application to document image binarization\nYear: 2003\n\nAbstract: To distinguish image features from high level noises, we propose to exploit the spatial correlations between wavelet coefficients by replacing the thresholding process with a diffusion process. The proposed smoothing method is more robust to noise when applied to highly corrupted images.","Title: Model-based varying pose face detection and facial feature registration in video images\nYear: 2000\n\nAbstract: This paper presents an automatic method for simultaneous human face detection and facial feature registration from colour video images. At the first stage, we use a skin colour Gaussian model to identify possible face locations under varying pose. Secondly, we compare image patterns with a varying pose face model in terms of shape and texture differences, using a combined feature-texture similarity measure (FTSM). False detections from the first stage are eliminated by setting an appropriate FTSM threshold. Moreover, one can also register the facial features (eyes, nose and mouth) by aligning a prototype face with the unknown pose faces. Experimental results show that the proposed method can achieve reliable face detection and feature registration under various conditions, including different poses, face appearances, and lighting conditions.","Title: Model-based varying pose face detection and facial feature registration in colour images\nYear: 2003\n\nAbstract: This paper presents a hybrid method for varying pose face detection and facial feature registration from colour images. At the first stage, we use a skin-colour Gaussian model to identify possible varying pose face regions. At the second stage, we compare each face candidates with a varying pose face model using a combined feature-texture similarity measure (FTSM). False detections from the first stage are eliminated by setting an appropriate FTSM threshold. Also, one can register the facial features (eyes, nose and mouth) by aligning a prototype face with the unknown pose faces. Experimental results show that the proposed method can achieve reliable face detection and feature registration under various conditions, including different poses, face appearances, and lighting conditions.","Title: Graph Partition Based Bundle Adjustment for Structured Dataset\nYear: 2011\n\nAbstract: Bundle adjustment has been considered as one of the most important components in many visual tasks such as 3D reconstruction, photo grammetry, visual SLAM, etc. Unfortunately, both time and space complexity of this adjustment prevent it from being directly applied to large scale datasets. This paper presents a sub mapping method, which partitions a large scale dataset into disjointed subsets and adjusts them one by one or in parallel. Pair-wise sub maps are then \"stitched\" together by applying a similarity transformation. Both simulations and real applications show that our method scales well. Also some basic questions of this sub mapping method including map size, map fusion and global consistency are discussed.","Title: Adaptive Sample Consensus for Efficient Random Optimization\nYear: 2009\n\nAbstract: This paper approaches random optimization problem with adaptive sampling, which exploits knowledge about data structure obtained from historical samples. The proposal distribution is adaptive so that it invests more searching efforts on high likelihood regions. In this way, the probability of reaching the global optimum is improved. The method demonstrates improved performance as compared with standard RANSAC and related adaptive methods, for line/plane/ellipse fitting and pose estimation problems.","Title: Supervised Binary Hash Code Learning with Jensen Shannon Divergence\nYear: 2013\n\nAbstract: This paper proposes to learn binary hash codes within a statistical learning framework, in which an upper bound of the probability of Bayes decision errors is derived for different forms of hash functions and a rigorous proof of the convergence of the upper bound is presented. Consequently, minimizing such an upper bound leads to consistent performance improvements of existing hash code learning algorithms, regardless of whether original algorithms are unsupervised or supervised. This paper also illustrates a fast hash coding method that exploits simple binary tests to achieve orders of magnitude improvement in coding speed as compared to projection based methods.","Title: In-Situ Web Mashup and Hypermedia Support for Mobile Vision-Based AR Applications\nYear: 2013\n\nAbstract: Recently, mobile augmented reality (AR) applications are becoming popular and allow normal users to see an enhanced real world through their mobile camera. Creating AR content however is tedious and traditionally done on desktop environments by professionals, with extensive knowledge and/or even programming skills required. In this paper, we introduce a complete mobile approach for creating vision-based AR in both indoor and outdoor environment. Using hyperlinks, Web mashups are built to dynamically augment the physical world by normal users without programing skills. This paper presents the technical challenges, design, and the implementation with some technical discussions to improve the user experience.","Title: Hill climbing algorithm for random sample consensus methods\nYear: 2007\n\nAbstract: We propose a modification of RANSAC that performs guided search of the sample space. The sampling is applicable to any of the sample consensus methods, such as MAPSAC or MLESAC. We give simulation results which show that the new method can reduce the number of required iterations to find a good model by orders of magnitude.","Title: Intra-class variation, affine transformation and background clutter: towards robust image matching\nYear: 2004\n\nAbstract: Image matching is a fundamental computer vision problem that includes many scenarios, such as varying the view scene matching, feature selection and registration, object recognition, and general object class matching. This article presents a unified framework and working algorithm for these different matching scenarios. The proposed feature-based image matching method demonstrates excellent robustness to significant geometrical transformation, intra-class variation and background clutter which are usually presented in different matching scenarios.","Title: Adaptive image-smoothing using a coplanar matrix and its application to document image binarization\nYear: 2003\n\nAbstract: Abstract. For document images corrupted by vari-ous kinds of noise, direct binarization images may be severely blurred and degraded. A common treatment for this problem is to pre-smooth input images using noise-suppressing filters. This article proposes an image-smoothing method used for prefiltering the document im-age binarization. Conceptually, we propose that the in-fluence range of each pixel a ecting its neighbors should depend on local image statistics. Technically, we sug-gest using coplanar matrices to capture the structural and textural distribution of similar pixels at each site. This property adapts the smoothing process to the con-trast, orientation, and spatial size of local image struc-tures. Experimental results demonstrate the e ectiveness of the proposed method, which compares favorably with existing methods in reducing noise and preserving im-age features. In addition, due to the adaptive nature of the similar pixel definition, the proposed filter output is more robust regarding di erent noise levels than existing methods.","Title: A feature-based object tracking approach for realtime image processing on mobile devices\nYear: 2010\n\nAbstract: In this paper we present a robust object tracking approach which is suitable for real-time image processing on mobile devices. Challenging mobile environments render traditional color-based tracking methods useless. Many online learning tracking methods are too computationally complex to be used for real-time mobile applications, which only have access to limited computational resource and memory storage. The proposed method takes advantage of local feature to deal with rapid camera motion, and employs an online feature updating scheme to cope with variation in object appearances. The method is also computationally lightweight, being able to support real-time image processing on mobile devices.","Title: Finger Tracking For Gestural Interaction In Mobile Devices\nYear: 2013\n\nAbstract: In this paper we propose a finger tracking system that is suitable for gesture recognition in mobile devices. The initialisation of the system does not require the use of any I/O devices. The user covers the camera lense with his or her hand and then takes it to the operating distance. The statistical models used for hand segmentation are initialised from the first frames after the hand is removed from the lense. The hand segmentation does not need to be perfect because we do not use the hand contour in the recognition. In our method the fingertips are found using template matching. However, the template matching produces false detections. These false detections are pruned by searching a path from the fingertip to the estimated hand centre and discarding the paths that do not meet a predefined criteria. We evaluate the performance of the method against a fingertip detector proposed by Baldauf et al. [2] by using seven test subjects who initialise the system and then wave their hand in front of the camera. In testing we use one handheld USB camera that matches the image quality of most recent front cameras in mobile phones.","Title: Edge-preserving prefiltering for document image binarization\nYear: 2001\n\nAbstract: We propose a novel coplanar filter, which exploits the coplanarity of the gray-level distribution of neighboring pixels, to pre-filter document images. Experiments show that the proposed filter exhibits the following desired properties for document image binarization: (1) impulsive noise removal; (2) piecewise smoothing; (3) sharp edge preservation","Title: PCA-based structure refinement for reconstruction of urban scene\nYear: 2010\n\nAbstract: There is plenty of structured information (such as lines and planes) in urban scenes. Considering this, we propose a new method for making use of this information to enhance the reconstruction of urban outdoor scenes. Structured information (collinearity and coplanarity) is extracted from images by performing line detection and color image segmentation, which is used as hypothetic constraints of the 3D structure. In refining stage, we first build PCA subspaces for each structured components (collinear and coplanar point sets), during which the former hypothetic structure information is further inspected by the initial 3D structure. Then we iteratively update the structure through EM estimation. Experiments show that this method effectively improves the accuracy and robustness of reconstruction of urban scenes.","Title: Binarizing document image using coplanar prefilter\nYear: 2001\n\nAbstract: We propose a novel coplanar filter, which exploits the coplanarity of gray-level distribution of neighboring pixels, to pre-filter the document images. Experiments show that the proposed filter exhibits the following desired properties for document image binarization: 1) impulsive noise removal, 2) piecewise smoothing, and 3) sharp edge preservation","Title: Revisiting the PnP Problem with a GPS\nYear: 2009\n\nAbstract: This paper revisits the pose estimation from point correspondences problem to properly exploit data provided by a GPS. In practice, the location given by the GPS is only a noisy estimate, and some point correspondences may be erroneous. Our method therefore starts from the GPS location estimate to progressively refine the full pose estimate by hypothesizing correct correspondences. We show how the GPS location estimate and the choice of a first random correspondence dramatically reduce the possibility for a second correspondence, which in turn constrains even more the remaining possible correspondences. This results in an efficient sampling of the solution space. Experimental results on a large 3D scene show that our method outperforms standard approaches and a recent related method [1] in terms of accuracy and robustness.","Title: Robust Scale Estimation from Ensemble Inlier Sets for Random Sample Consensus Methods\nYear: 2008\n\nAbstract: This paper proposes a RANSAC modification that performs automatic estimation of the scale of inlier noise. The scale estimation takes advantage of accumulated inlier sets from all proposed models. It is shown that the proposed method gives robust results in case of high outlier ratio data, in spite that no user specified threshold is needed. The method also improves sampling efficiency, without requiring any auxiliary information other than the data to be modeled.","Title: Real-time object tracking on mobile phones\nYear: 2011\n\nAbstract: For a mobile landmark recognition and Augmented Reality system, real-time object tracking is crucial for the user experience. In this paper, an effective natural feature (low dimensional Haar-like feature) based method is proposed for real-time object tracking on mobile phones. This method can cope with image variations and alleviate the drift problem. The main idea is using an online two template switching strategy for estimating the geometrical transformation. Experiments on NOKIA N900 phone show the good performance of the proposed method and support our claim that our tracking method is effective and robust.","Title: Object Recognition In 3d Point Cloud Of Urban Street Scene\nYear: 2014\n\nAbstract: In this paper we present a novel street scene semantic recognition framework, which takes advantage of 3D point clouds captured by a high-definition LiDAR laser scanner. An important problem in object recognition is the need for sufficient labeled training data to learn robust classifiers. In this paper we show how to significantly reduce the need for manually labeled training data by reduction of scene complexity using non-supervised ground and building segmentation. Our system first automatically segments grounds point cloud, this is because the ground connects almost all other objects and we will use a connect component based algorithm to oversegment the point clouds. Then, using binary range image processing building facades will be detected. Remained point cloud will grouped into voxels which are then transformed to super voxels. Local 3D features extracted from super voxels are classified by trained boosted decision trees and labeled with semantic classes e.g. tree, pedestrian, car, etc. The proposed method is evaluated both quantitatively and qualitatively on a challenging fixed-position Terrestrial Laser Scanning (TLS) Velodyne data set and two Mobile Laser Scanning (MLS), Paris-rue-Madam and NAVTEQ True databases. Robust scene parsing results are reported.","Title: Online Learning Of Binary Feature Indexing For Real-Time Slam Relocalization\nYear: 2014\n\nAbstract: In this paper, we propose an indexing method for approximate nearest neighbor search of binary features. Being different from the popular Locality Sensitive Hashing (LSH), the proposed method construct the hash keys by an online learning process instead of pure randomness. In the learning process, the hash keys are constructed with the aim of obtaining uniform hash buckets and high collision rates, which makes the method more efficient on approximate nearest neighbor search than LSH. By distributing the online learning into the simultaneous localization and mapping (SLAM) process, we successfully apply the method to SLAM relocalization. Experiments show that camera poses can be successfully recovered in real time even there are tens of thousands of landmarks in the map.","Title: Simple Octree Solution for Multi-resolution LiDAR Processing and Visualisation\nYear: 2014\n\nAbstract: LiDAR (light detection and ranging) 3D points (both terrestrial and airborne) have recently become a prevalent type of geospatial data. However, this large-scale data poses significant technical challenges for efficient access and visualization. One approach is to generate points with different density levels, i.e. Multi-resolution data sets that can be used for various purposes. This paper presents a simple static tree-based approach and discusses implementation alternatives in favour of speeding up tree building and point sampling processes. This approach is generic and can be applied to other geographic data types as well.","Title: Street view cross-sourced point cloud matching and registration\nYear: 2014\n\nAbstract: Object registration has been widely discussed with the development of various range sensing technologies. In most work, however, the point clouds of reference and target are generated by the same technology, such as a Kinect range camera, LiDAR sensor, or Structure from Motion technique. Cases in which reference and target point clouds are generated by different technologies are rarely discussed. Due to the significant differences across various point cloud data in terms of point cloud density, sensing noise, scale, occlusion etc., object registration between such different point clouds becomes extremely difficult. In this study, we address for the first time an even more challenging case in which the differently-sourced point clouds are acquired from a real street view. One is generated on the basis of an image sequence through the SfM process, and the other is produced directly by the LiDAR system. We propose a two-stage matching and registration algorithm to achieve object registration between these two different point clouds. The experiments are based on real building object point cloud data and demonstrate the effectiveness and efficiency of the proposed solution. The newly proposed solution can be further developed to contribute to several related applications, such as Location Based Service.","Title: A 3D map augmented photo gallery application on mobile device\nYear: 2014\n\nAbstract: This paper proposes a 3D map augmented photo gallery mobile application that allows user to virtually transit from 2D image space to the 3D map space, to expand the field of view to surrounding environments that are not visible in the original image, and to change viewing angles among different global registered images during the image browsing. The processing of images consists of two main steps: in the first step, the client application uploads an image to the GeoImage Engine which extracts the geo-metadata and returns them back to the client; in the second step, the client application requests augmented content from server, and then renders 3D view of images on the screen of mobile devices.","Title: Efficient pose tracking on mobile phones with 3D points grouping\nYear: 2014\n\nAbstract: With the rapid growth of computational capability and popularity of mobile phones, Mobile Augmented Reality (MAR) in large scale 3D scenes becomes an emerging field in recent years. The core of MAR is to continuously compute a precise 6 Degree-of-Freedom (DOF) camera pose for each frame, i.e. localization. However, as a crucial part of localization, the 2D-3D points matching is usually inefficient due to the usage of traditional features, e.g. SIFT, and the large number of 3D points candidates. This paper aims to tackle this problem by designing an efficient 6DOF pose tracking system on mobile phone. In this system, binary features are used in both offline sparse reconstruction and online tracking, while a PCA-based 3D points partition method is proposed to reduce the searching space of 2D-3D points matching, making it capable to achieve a low computational cost. Experiments on a NOKIA N900 smartphone show that our system could efficiently and robustly estimate the 6DOF camera pose.","Title: Fast Localization in Large-Scale Environments Using Supervised Indexing of Binary Features\nYear: 2016\n\nAbstract: The essence of image-based localization lies in matching 2D key points in the query image and 3D points in the database. State-of-the-art methods mostly employ sophisticated key point detectors and feature descriptors, e.g., Difference of Gaussian (DoG) and Scale Invariant Feature Transform (SIFT), to ensure robust matching. While a high registration rate is attained, the registration speed is impeded by the expensive key point detection and the descriptor extraction. In this paper, we propose to use efficient key point detectors along with binary feature descriptors, since the extraction of such binary features is extremely fast. The naive usage of binary features, however, does not lend itself to significant speedup of localization, since existing indexing approaches, such as hierarchical clustering trees and locality sensitive hashing, are not efficient enough in indexing binary features and matching binary features turns out to be much slower than matching SIFT features. To overcome this, we propose a much more efficient indexing approach for approximate nearest neighbor search of binary features. This approach resorts to randomized trees that are constructed in a supervised training process by exploiting the label information derived from that multiple features correspond to a common 3D point. In the tree construction process, node tests are selected in a way such that trees have uniform leaf sizes and low error rates, which are two desired properties for efficient approximate nearest neighbor search. To further improve the search efficiency, a probabilistic priority search strategy is adopted. Apart from the label information, this strategy also uses non-binary pixel intensity differences available in descriptor extraction. By using the proposed indexing approach, matching binary features is no longer much slower but slightly faster than matching SIFT features. Consequently, the overall localization speed is significantly improved due to the much faster key p- int detection and descriptor extraction. It is empirically demonstrated that the localization speed is improved by an order of magnitude as compared with state-of-the-art methods, while comparable registration rate and localization accuracy are still maintained.","Title: A weakly supervised geodesic level set framework for interactive image segmentation\nYear: 2015\n\nAbstract: Interactive image segmentation is growingly useful for selecting objects of interest in images, facilitating spatially localized media manipulation especially on touch screen devices. We present a robust and efficient approach for segmenting image with less and intuitive user interaction. Our approach combines geodesic distance information with the flexibility of level set methods in energy minimization, leveraging the complementary strengths of each to promote accurate boundary placement and strong region connectivity while requiring less user interaction. We harness weakly supervised segment annotation to maximize the user-provided prior knowledge. This leads to a seed generation algorithm which enables image object segmentation without user-provided background seeds. We demonstrate that our approach is less sensitive to seed placement and better at edge localization, whilst requiring less user interaction, compared with the state-of-the-art methods.","Title: Using the Similarity Measure between Intuitionistic Fuzzy Sets for the Application on Pattern Recognitions\nYear: 2015\n\nAbstract: AbstractThe measurement of uncertainty is an important topic for the theories dealing with uncertainty. The definition of similarity measure between two IFSs is one of the most interesting topics in IFSs theory. A similarity measure is defined to compare the information carried by IFSs. Many similarity measures have been proposed. A few of them come from the well-known distance measures. In this work, a new similarity measure between IFSs was proposed by the consideration of the information carried by the membership degree, the non-membership degree, and hesitancy degree in intuitionistic fuzzy sets IFSs. To demonstrate the efficiency of the proposed similarity measure, various similarity measures between IFSs were compared with the proposed similarity measure between IFSs by numerical examples. The compared results demonstrated that the new similarity measure is reasonable and has stronger discrimination among them. Finally, the similarity measure was applied to pattern recognition and medical diagnosis. Two illustrative examples were provided to show the effectiveness of the pattern recognition and medical diagnosis.","Title: Automated super-voxel based features classification of urban environments by integrating 3D point cloud and image content\nYear: 2015\n\nAbstract: In this paper we present a novel street scene semantic recognition framework, which takes advantage of 3D point cloud captured by a high definition LiDAR laser scanner. An important problem in object recognition is the need for sufficient labeled training data to learn robust classifiers. We show how to significantly reduce the need for manually labeled training data by reduction of scene complexity using non-supervised ground and building segmentation. Our system first automatically segments grounds point cloud. Then, using binary range image processing building facades will be detected. Remained point cloud will grouped into voxels which are then transformed to super voxels. Local 3D features extracted from super voxels are classified by trained boosted decision trees and labeled with semantic classes e.g. tree, pedestrian, car. Given labeled 3D points cloud and 2D image with known viewing camera pose, the proposed association module aligned collections of 3D points to the groups of 2D image pixel to parsing 2D cubic images.","Title: Robust interactive image segmentation with weak supervision for mobile touch screen devices\nYear: 2015\n\nAbstract: In this paper, we present a robust and efficient approach for segmenting images with less and intuitive user interaction, particularly targeted for mobile touch screen devices. Our approach combines geodesic distance information with the flexibility of level set methods in energy minimization, leveraging the complementary strengths of each to promote accurate boundary placement and strong region connectivity while requiring less user interaction. To maximize the user-provided prior knowledge, we further propose a weakly supervised seed generation algorithm which enables image object segmentation without user-provided background seeds. Our approach provides a practical solution for visual object cutout on mobile touch screen devices, facilitating various media manipulation applications. We describe such a use case to selectively create oil painting effects on images. We demonstrate that our approach is less sensitive to seed placement and better at edge localization, whilst requiring less user interaction, compared with the state-of-the-art methods.","Title: Accurate Robust and Efficient Error Estimation for Decision Trees.\nYear: 2016\n\nAbstract: This paper illustrates a novel approach to the estimation of generalization error of decision tree classifiers. We set out the study of decision tree errors in the context of consistency analysis theory, which proved that the Bayes error can be achieved only if when the number of data samples thrown into each leaf node goes to infinity. For the more challenging and practical case where the sample size is finite or small, a novel sampling error term is introduced in this paper to cope with the small sample problem effectively and efficiently. Extensive experimental results show that the proposed error estimate is superior to the well-known K-fold cross validation methods in terms of robustness and accuracy. Moreover it is orders of magnitudes more efficient than cross validation methods.","Title: Global Scale Integral Volumes.\nYear: 2016\n\nAbstract: Integral volume is an important image representation technique, which is useful in many computer vision applications. Processing integral volumes for large scale 3D datasets is challenging due to high memory requirements. The difficulties lie in efficiently computing, storing, querying and updating the integral volume values. In this work, we address the above problems and present a novel solution for processing integral volumes for large scale 3D datasets efficiently. We propose an octree-based method where the worst-case complexity for querying the integral volume of arbitrary regions is (mathcal {O}(log {}n)), here n is the number of nodes in the octree. We evaluate our proposed method on multi-resolution LiDAR point cloud data. Our work can serve as a tool to fast extract features from large scale 3D datasets, which can be beneficial for computer vision applications.","Title: Weighted fuzzy track association method based on Dempster-Shafer theory in distributed sensor networks.\nYear: 2016\n\nAbstract: The uncertainty problem in sensor track to local track association is a difficult problem in distributed sensor networks, particularly when there is a big difference of sensors’ tracking performance. To solve this problem, a weighted fuzzy track association (FTA) method based on Dempster–Shafer theory is proposed. In the proposed method, five characteristics of sensor tracks from different sensors are established, and meanwhile their belief functions are defined to determine the corresponding beliefs. Considering the different effects of sensor tracks on track association, the reliabilities of sensor tracks are further presented and their magnitudes can be calculated by the combination belief function defined. Then, these reliabilities are used to reconstruct the fuzzy association degrees by the FTA method. The proposed method has an advantage that it can dynamically allocate the weight of each sensor track in association decision according to its characteristics. The performance of the proposed method is evaluated by using two experiments with simulation data in manoeuvring and uniform situations. It is found to be better than those of other two track association methods in tracking accuracy.","Title: Comprehensive Automated 3D Urban Environment Modelling Using Terrestrial Laser Scanning Point Cloud\nYear: 2016\n\nAbstract: In this paper we present a novel street scene modelling framework, which takes advantage of 3D point cloud captured by a high definition LiDAR laser scanner. We propose an automatic and robust approach to detect, segment and classify urban objects from point clouds hence reconstructing a comprehensive 3D urban environment model. Our system first automatically segments grounds point cloud. Then building facades will be detected by using binary range image processing. Remained point cloud will be grouped into voxels and subsequently transformed into super voxels. Local 3D features are extracted from super voxels and classified by trained boosted decision trees with semantic classes e.g. tree, pedestrian, and car. Given labeled point cloud the proposed algorithm reconstructs the realistic model in two phases. Firstly building facades will be rendered by ShadVis algorithm. In the second step we apply a novel and fast method for fitting the solid predefined template mesh models to non-building labeled point cloud. The proposed method is evaluated both quantitatively and qualitatively on a challenging TLS NAVTEQ True databases.","Title: Deep Epitome for Unravelling Generalized Hamming Network: A Fuzzy Logic Interpretation of Deep Learning.\nYear: 2017\n\nAbstract: This paper gives a rigorous analysis of trained Generalized Hamming Networks(GHN) proposed by Fan (2017) and discloses an interesting finding about GHNs, i.e., stacked convolution layers in a GHN is equivalent to a single yet wide convolution layer. The revealed equivalence, on the theoretical side, can be regarded as a constructive manifestation of the universal approximation theorem Cybenko(1989); Hornik (1991). In practice, it has profound and multi-fold implications. For network visualization, the constructed deep epitomes at each layer provide a visualization of network internal representation that does not rely on the input data. Moreover, deep epitomes allows the direct extraction of features in just one step, without resorting to regularized optimizations used in existing visualization tools.","Title: Urban 3D segmentation and modelling from street view images and LiDAR point clouds.\nYear: 2017\n\nAbstract: 3D urban maps with semantic labels and metric information are not only essential for the next generation robots such autonomous vehicles and city drones, but also help to visualize and augment local environment in mobile user applications. The machine vision challenge is to generate accurate urban maps from existing data with minimal manual annotation. In this work, we propose a novel methodology that takes GPS registered LiDAR (Light Detection And Ranging) point clouds and street view images as inputs and creates semantic labels for the 3D points clouds using a hybrid of rule-based parsing and learning-based labelling that combine point cloud and photometric features. The rule-based parsing boosts segmentation of simple and large structures such as street surfaces and building facades that span almost 75% of the point cloud data. For more complex structures, such as cars, trees and pedestrians, we adopt boosted decision trees that exploit both structure (LiDAR) and photometric (street view) features. We provide qualitative examples of our methodology in 3D visualization where we construct parametric graphical models from labelled data and in 2D image segmentation where 3D labels are back projected to the street view images. In quantitative evaluation we report classification accuracy and computing times and compare results to competing methods with three popular databases: NAVTEQ True, Paris-Rue-Madame and TLS (terrestrial laser scanned) Velodyne.","Title: Real-time SLAM relocalization with online learning of binary feature indexing.\nYear: 2017\n\nAbstract: A visual simultaneous localization and mapping (SLAM) system usually contains a relocalization module to recover the camera pose after tracking failure. The core of this module is to establish correspondences between map points and key points in the image, which is typically achieved by local image feature matching. Since recently emerged binary features have orders of magnitudes higher extraction speed than traditional features such as scale invariant feature transform, they can be applied to develop a real-time relocalization module once an efficient method of binary feature matching is provided. In this paper, we propose such a method by indexing binary features with hashing. Being different from the popular locality sensitive hashing, the proposed method constructs the hash keys by an online learning process instead of pure randomness. Specifically, the hash keys are trained with the aim of attaining uniform hash buckets and high collision rates of matched feature pairs, which makes the method more efficient on approximate nearest neighbor search. By distributing the online learning into the simultaneous localization and mapping process, we successfully apply the method to SLAM relocalization. Experiments show that camera poses can be recovered in real time even when there are tens of thousands of landmarks in the map.","Title: Memory-Efficient Deep Salient Object Segmentation Networks on Gridized Superpixels\nYear: 2018\n\nAbstract: Computer vision algorithms with pixel-wise labeling tasks, such as semantic segmentation and salient object detection, have gone through a significant accuracy increase with the incorporation of deep learning. Deep segmentation methods slightly modify and fine-tune pre-trained networks that have hundreds of millions of parameters. In this work, we question the need of having such memory demanding networks for a reasonable performance in salient object segmentation. To this end, we propose a way to learn a memory-efficient network from scratch by training it only on salient object detection datasets. Our method encodes images to gridized superpixels that preserve both the object boundaries and the connectivity rules of regular pixels. This representation allows us to use convolutional neural networks that operate on regular grids. By using these encoded images, we train a memory-efficient network using only 0.048% of the number of parameters that a majority of other deep salient object detection networks have. Our method shows comparable accuracy with the state-of-the-art deep salient object detection methods and provides a much more memory-efficient alternative to them. Due to its easy deployment and small size, such a network is preferable for applications in memory limited IoT devices.","Title: A Theoretical Investigation of Graph Degree as an Unsupervised Normality Measure.\nYear: 2018\n\nAbstract: For a graph representation of a dataset, a straightforward normality measure for a sample can be its graph degree. Considering a weighted graph, this corresponds to sum of rows in a similarity matrix. The measure is intuitive given the abnormal samples are usually rare and they are dissimilar to the rest of the data. In order to have an in-depth theoretical understanding, in this manuscript, we investigate the graph degree in spectral graph clustering based and kernel based point of views and draw connections to a recent kernel method for the two sample problem. We show that our analyses guide us to choose fully-connected graphs whose edge weights are calculated via universal kernels. We show that a simple graph degree based unsupervised anomaly detection method with the above properties, achieves leading accuracy compared to other unsupervised anomaly detection methods on average over 10 widely used datasets. We also provide an extensive analysis on the effect of the kernel parameter on the methodu0027s accuracy.","Title: Depth Masked Discriminative Correlation Filter\nYear: 2018\n\nAbstract: Depth information provides a strong cue for occlusion detection and handling, but has been largely omitted in generic object tracking until recently due to lack of suitable benchmark datasets and applications. In this work, we propose a Depth Masked Discriminative Correlation Filter (DM-DCF) which adopts novel depth segmentation based occlusion detection that stops correlation filter updating and depth masking which adaptively adjusts the spatial support for correlation filter. In Princeton RGBD Tracking Benchmark, our DM-DCF is among the state-of-the-art in overall ranking and the winner on multiple categories. Moreover, since it is based on DCF, \"DM-DCF\" runs an order of magnitude faster than its competitors making it suitable for time constrained applications.","Title: Object Detection in Equirectangular Panorama.\nYear: 2018\n\nAbstract: We introduce a high-resolution equirectangular panorama (aka 360-degree, virtual reality, VR) dataset for object detection and propose a multi-projection variant of the YOLO detector. The main challenges with equirectangular panorama images are i) the lack of annotated training data, ii) high-resolution imagery and iii) severe geometric distortions of objects near the panorama projection poles. In this work, we solve the challenges by I) using training examples available in the “conventional datasets” (ImageNet and COCO), II) employing only low resolution images that require only moderate GPU computing power and memory, and III) our multi-projection YOLO handles projection distortions by making multiple stereographic sub-projections. In our experiments, YOLO outperforms the other state-of-the-art detector, Faster R-CNN, and our multi-projection YOLO achieves the best accuracy with low-resolution input.","Title: Deep Epitome for Unravelling Generalized Hamming Network\nYear: 2018\n\nAbstract: This paper gives a rigorous analysis of trained Generalized Hamming Networks (GHN) proposed by [9] and discloses an interesting finding about GHNs, i.e., stacked convolution layers in a GHN is equivalent to a single yet wide convolution layer. The revealed equivalence, on the theoretical side, can be regarded as a constructive manifestation of the universal approximation theorem [7], [16]. In practice, it has profound and multi-fold implications. For network visualization, the constructed deep epitomes at each layer provide a visualization of network internal representation that does not rely on the input data. Moreover, deep epitomes allows the direct extraction of features in just one step, without resorting to regularized optimizations used in existing visualization tools.","Title: Automatic Feature Extraction For Wide-Angle And Fish-Eye Camera Calibration\nYear: 2018\n\nAbstract: The increase need to have 360 degree perception of world environment has led to the combination of various cameras in the varieties of narrow and wide-angle field of view. Current state of art methods do not provide an automatic means for fish-eye calibration, which is indispensable in an industrial environment where many lenses are to be calibrated in a relative short time. As automatic feature extraction is the key issue for fully automatic calibration framework, we address this issue to remove any form of human intervention during the calibration process.Unlike state-of-the-art methods, the proposed framework is completely automatic and not prone to detection errors. We evaluate the proposed feature extraction using state-of-the-art generic calibration model with both real and synthetic images.","Title: Generic Calibration Of Cameras With Non-Parallel Optical Elements\nYear: 2018\n\nAbstract: Multiple cameras are increasingly prevalent for autonomous driving, and the increase need to have 360 degree perception of world space has led to the combination of various cameras in the varieties of narrow-angle and wide-angle field of view. This has led to issues regarding the quality of these optics as a result of bad and cheap designs. Intrinsic calibration is indispensable for accurate perception of the environment in order for it to make accurate decisions such as camera pose estimation and 3-D reconstruction.In this work, we propose a lens distortion model that has been motivated by unintentional tilt in the optical lens system. The proposed distortion model has been added to current state of art generic method of Kannala to form an extended model. To our knowledge, this is the first time that the idea of tilt distortion has been introduced for wide-angle view and fish-eye cameras. We show improved result of 4 to 13% from experiments.","Title: Happy Travelers Take Big Pictures: A Psychological Study with Machine Learning and Big Data.\nYear: 2017\n\nAbstract: In psychology, theory-driven researches are usually conducted with extensive laboratory experiments, yet rarely tested or disproved with big data. In this paper, we make use of 418K travel photos with traveler ratings to test the influential broaden-and-build theory, that suggests positive emotions broaden oneu0027s visual attention. The core hypothesis examined in this study is that positive emotion is associated with a wider attention, hence highly-rated sites would trigger wide-angle photographs. By analyzing travel photos, we find a strong correlation between a preference for wide-angle photos and the high rating of tourist sites on TripAdvisor. We are able to carry out this analysis through the use of deep learning algorithms to classify the photos into wide and narrow angles, and present this study as an exemplar of how big data and deep learning can be used to test laboratory findings in the wild.","Title: Proceedings of AAAI 2019 Workshop on Network Interpretability for Deep Learning.\nYear: 2019\n\nAbstract: This is the Proceedings of AAAI 2019 Workshop on Network Interpretability for Deep Learning","Title: A Universal Logic Operator for Interpretable Deep Convolution Networks.\nYear: 2019\n\nAbstract: Explaining neural network computation in terms of probabilistic/fuzzy logical operations has attracted much attention due to its simplicity and high interpretability. Different choices of logical operators such as AND, OR and XOR give rise to another dimension for network optimization, and in this paper, we study the open problem of learning a universal logical operator without prescribing to any logical operations manually. Insightful observations along this exploration furnish deep convolution networks with a novel logical interpretation.","Title: Digital Passport: A Novel Technological Strategy for Intellectual Property Protection of Convolutional Neural Networks.\nYear: 2019\n\nAbstract: In order to prevent deep neural networks from being infringed by unauthorized parties, we propose a generic solution which embeds a designated digital passport into a network, and subsequently, either paralyzes the network functionalities for unauthorized usages or maintain its functionalities in the presence of a verified passport. Such a desired network behavior is successfully demonstrated in a number of implementation schemes, which provide reliable, preventive and timely protections against tens of thousands of fake-passport deceptions. Extensive experiments also show that the deep neural network performance under unauthorized usages deteriorate significantly (e.g. with 33% to 82% reductions of CIFAR10 classification accuracies), while networks endorsed with valid passports remain intact.","Title: DeepIPR: Deep Neural Network Ownership Verification With Passports\nYear: 2022\n\nAbstract: With substantial amount of time, resources and human (team) efforts invested to explore and develop successful deep neural networks (DNN), there emerges an urgent need to protect these inventions from being illegally copied, redistributed, or abused without respecting the intellectual properties of legitimate owners. Following recent progresses along this line, we investigate a number of watermark-based DNN ownership verification methods in the face of ambiguity attacks, which aim to cast doubts on the ownership verification by forging counterfeit watermarks. It is shown that ambiguity attacks pose serious threats to existing DNN watermarking methods. As remedies to the above-mentioned loophole, this paper proposes novel \n<italic xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">passport</i>\n-based DNN ownership verification schemes which are both \n<italic xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">robust to network modifications</i>\n and \n<italic xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">resilient to ambiguity attacks</i>\n. The gist of embedding digital passports is to design and train DNN models in a way such that, the DNN inference performance of an original task will be significantly \n<italic xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">deteriorated due to forged passports</i>\n. In other words, genuine passports are not only verified by looking for the predefined signatures, but also reasserted by the \n<italic xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">unyielding DNN model inference performances</i>\n. Extensive experimental results justify the effectiveness of the proposed passport-based DNN ownership verification schemes. Code is available at \n<uri xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">https://github.com/kamwoh/DeepIPR</uri>\n.","Title: ICDAR 2021 Competition on Integrated Circuit Text Spotting and Aesthetic Assessment\nYear: 2021\n\nAbstract: With hundreds of thousands of electronic chip components are being manufactured every day, chip manufacturers have seen an increasing demand in seeking a more efficient and effective way of inspecting the quality of printed texts on chip components. The major problem that deters this area of research is the lacking of realistic text on chips datasets to act as a strong foundation. Hence, a text on chips dataset, ICText is used as the main target for the proposed Robust Reading Challenge on Integrated Circuit Text Spotting and Aesthetic Assessment (RRC-ICText) 2021 to encourage the research on this problem. Throughout the entire competition, we have received a total of 233 submissions from 10 unique teams/individuals. Details of the competition and submission results are presented in this report.","Title: L2RS: A Learning-to-Rescore Mechanism for Hybrid Speech Recognition\nYear: 2021\n\nAbstract: ABSTRACTThis paper aims to advance the performance of industrial ASR systems by exploring a more effective method for N-best rescoring, a critical step that greatly affects the final recognition accuracy. Existing rescoring approaches suffer the following issues: (i) limited performance since they optimize an unnecessarily harder problem, namely predicting accurate grammatical legitimacy scores of the N-best hypotheses rather than directly predicting their partial orders regarding a specific acoustic input; (ii) hard to incorporate various information by advanced natural language processing (NLP) models such as BERT to achieve a comprehensive evaluation of each N-best candidate. To relieve the above drawbacks, we propose a simple yet effective mechanism, Learning-to-Rescore (L2RS), to empower ASR systems with state-of-the-art information retrieval (IR) techniques. Specifically, L2RS utilizes a wide range of textual information from the state-of-the-art NLP models and automatically deciding their weights to directly learn the ranking order of each N-best hypothesis with respect to a specific acoustic input. We incorporate various features including BERT sentence embeddings, the topic vectors, and perplexity scores produced by an n-gram language model (LM), topic modeling LM, BERT, and RNNLM to train the rescoring model. Experimental results on a public dataset show that L2RS outperforms not only traditional rescoring methods but also its deep neural network counterparts by a substantial margin of 20.85% in terms of [email protected] The L2RS toolkit has been successfully deployed for many online commercial services in WeBank Co., Ltd, China's leading digital bank. The efficacy and applicability of L2RS are validated by real-life online customer datasets.","Title: A Health-Friendly Speaker Verification System Supporting Mask Wearing\nYear: 2021\n\nAbstract: We demonstrate a health-friendly speaker verification system for voice-based identity verification on mobile devices. The system is built upon a speech processing module, a ResNet-based local acoustic feature extractor and a multi head attention-based embedding layer, and is optimized under an additive margin softmax loss for discriminative speaker verification. It is shown that the system achieves superior performance no matter whether there is mask wearing or not. This characteristic is important for speaker verification services operating in regions affected by the raging coronavirus pneumonia. With this demonstrationl, the audience will have an in-depth experience of how the accuracy of bio-metric verification and the personal health are simultaneously ensured. We wish that this demonstration would boost the development of next-generation bio-metric verification technologies.","Title: Protect, show, attend and tell: Empowering image captioning models with ownership protection\nYear: 2022\n\nAbstract: •We propose a key-based strategy that provides reliable, preventive and timely IP protection for image captioning task.•We empirically show the effectiveness of our approach against various attacks and prove the ownership of the model.•To the best of our knowledge, we are the first to propose IP protection on image captioning model.","Title: You Don't Know My Favorite Color: Preventing Dialogue Representations from Revealing Speakers' Private Personas\nYear: 2022\n\nAbstract: Social chatbots, also known as chit-chat chatbots, evolve rapidly with large pretrained language models. Despite the huge progress, privacy concerns have arisen recently: training data of large language models can be extracted via model inversion attacks. On the other hand, the datasets used for training chatbots contain many private conversations between two individuals. In this work, we further investigate the privacy leakage of the hidden states of chatbots trained by language modeling which has not been well studied yet. We show that speakers' personas can be inferred through a simple neural network with high accuracy. To this end, we propose effective defense objectives to protect persona leakage from hidden states. We conduct extensive experiments to demonstrate that our proposed defense objectives can greatly reduce the attack accuracy from 37.6% to 0.5%. Meanwhile, the proposed objectives preserve language models' powerful generation ability."],"string":"[\n \"Title: An efficient parallel algorithm for hexagonal-based fractal image compression\\nYear: 2007\\n\\nAbstract: A two-level fast search algorithm to reduce the encoding time for hexagonal-based fractal image compression is presented. The design of the sequential algorithm is based on the distribution of matched domains in a given image. The first search level previews various portions of the image and identifies promising domains among all possible domains. The second search level picks out domain blocks in the image portion where the first level gives positive results, and compares them with a given range block for encoding. The algorithm is parallelized by a dynamic range distribution scheme to achieve load balancing. Experimental results show that by running the parallelized encoding algorithm on multiple processors, the encoding time is drastically reduced while the quality of image reconstruction is retained. A speed-up of about 9 can be obtained by using 13 processors.\",\n \"Title: Wavelet diffusion for document image denoising\\nYear: 2003\\n\\nAbstract: To distinguish image features from high level noises, wepropose to exploit the spatial correlations between waveletcoefficients by replacing the thresholding process with a diffusionprocess. The proposed smoothing method is more robustto noise when applied to highly corrupted documentimages.\",\n \"Title: Adaptive image restoration using structure tensors\\nYear: 2002\\n\\nAbstract: This paper proposes a structure tensor based image restoration algorithm. Structure tensors capture the structural and textural distribution of similar pixels at each site. This property adapts the smoothing process to the contrast, orientation and spatial size of local image structures.\",\n \"Title: Pedestrian registration in static images with unconstrained background\\nYear: 2003\\n\\nAbstract: This paper introduces a human body contour registration method for static pedestrian images with unconstrained backgrounds. By using a statistical compound model to impose structural and textural constraints on valid pedestrian appearances, the matching process is robust to image clutter. Experimental results show that the proposed method register pedestrian contours in complex backgrounds effectively.\",\n \"Title: On-line Object Reconstruction and Tracking for 3D Interaction\\nYear: 2012\\n\\nAbstract: This paper presents a flexible and easy-to-use tracking method for 3D interaction. The method reconstructs points of a user-specified object from a video sequence, and recovers the 6 degrees of freedom (DOF) camera pose and position relative to the reconstructed points in each video frame. As opposed to most existing 3D object tracking methods, the proposed method does not need any off-line modeling or training process. Instead, it first segments the object from the background, then reconstructs and tracks the object using the Visual Simultaneous Localization And Mapping (VSLAM) techniques. To our knowledge, there are no existing works investigating this kind of on line reconstruction and tracking of moving objects. The proposed method employs the adapted pyramidal Lucas-Kanade tracker to increase the stability and the robustness of the tracking when dealing with a lightly textured or fast moving object. Experiments show that fast, accurate, stable and robust tracking can be achieved in everyday environment. Moreover, a simple stereo initialization approach is adopted to minimize user intervention. All these attributes conspire to make the method an adequate tool for some interaction applications. As a concrete example, an interactive 3D scene displaying system is demonstrated.\",\n \"Title: Head-tracking virtual 3-D display for mobile devices.\\nYear: 2012\\n\\nAbstract: Computer vision enables camera data to be utilized in user interfaces to analyze the 3-D context and automatically detect the user intentions. Using cameras as an input modality provides single-handed operations in which the users' actions are recognized without interactions with the screen or keypad. In this context, we have constructed a real-time mobile application prototype where the user's position and gaze is determined in real time, a technique that enables the display of true three-dimensional objects even on a typical 2-D LCD screen. We have defined a series of interaction methods where the user's motion and camera input realistically control the viewpoint on a 3-D scene. The head movement and gaze can be used to interact with hidden objects in a natural manner just by looking at them. We provide a description of the embedded implementation at a system-level where we highlight the application development challenges and trade-offs that need to be dealt with battery powered mobile devices. The implementation includes a parallel pipeline that reduces the latencies of the application.\",\n \"Title: Animation pyramid for efficient image database browsing\\nYear: 2010\\n\\nAbstract: This paper introduces an image database organization method as well as dedicated user interface (UI) for efficient browsing of large image databases. The proposed method and UI are particularly useful for image browsing on small screens of mobile devices.\",\n \"Title: Semantic Parsing of Street Scene Images Using 3D LiDAR Point Cloud\\nYear: 2013\\n\\nAbstract: In this paper we propose a novel street scene semantic parsing framework, which takes advantage of 3D point clouds captured by a high-definition LiDAR laser scanner. Local 3D geometrical features extracted from subsets of point clouds are classified by trained boosted decision trees and then corresponding image segments are labeled with semantic classes e.g. buildings, road, sky etc. In contrast to existing image-based scene parsing approaches, the proposed 3D LiDAR point cloud based approach is robust to varying imaging conditions such as lighting and urban structures. The proposed method is evaluated both quantitatively and qualitatively on three challenging NAVTEQ True databases and robust scene parsing results are reported.\",\n \"Title: Efficient Random Sampling for Nonrigid Feature Matching\\nYear: 2009\\n\\nAbstract: This paper aims to match two sets of nonrigid feature points using random sampling methods. By exploiting the principle eigenvector of corres pondence-model-linkage, an adaptive sampling method is devised to efficiently deal with non-rigid matching problems.\",\n \"Title: A Feature-Based Object Tracking Method Using Online Template Switching and Feature Adaptation.\\nYear: 2011\\n\\nAbstract: This paper proposes a robust tracking method which stores representative object appearances as candidate templates during tracking, and selects the best template to match new frames. This online template adding and switching strategy, in one aspect, keeps update with new object appearances. In yet another aspect, it is resilient to misaligned templates and alleviates the drifting problem. The method is able to achieve real time speed on a laptop PC, and the tracking is robust to significant image variations including illumination changes, deformation, pose changes and occlusions.\",\n \"Title: A diffusion process for wavelet-transform-based image denoising and its application to document image binarization\\nYear: 2003\\n\\nAbstract: To distinguish image features from high level noises, we propose to exploit the spatial correlations between wavelet coefficients by replacing the thresholding process with a diffusion process. The proposed smoothing method is more robust to noise when applied to highly corrupted images.\",\n \"Title: Model-based varying pose face detection and facial feature registration in video images\\nYear: 2000\\n\\nAbstract: This paper presents an automatic method for simultaneous human face detection and facial feature registration from colour video images. At the first stage, we use a skin colour Gaussian model to identify possible face locations under varying pose. Secondly, we compare image patterns with a varying pose face model in terms of shape and texture differences, using a combined feature-texture similarity measure (FTSM). False detections from the first stage are eliminated by setting an appropriate FTSM threshold. Moreover, one can also register the facial features (eyes, nose and mouth) by aligning a prototype face with the unknown pose faces. Experimental results show that the proposed method can achieve reliable face detection and feature registration under various conditions, including different poses, face appearances, and lighting conditions.\",\n \"Title: Model-based varying pose face detection and facial feature registration in colour images\\nYear: 2003\\n\\nAbstract: This paper presents a hybrid method for varying pose face detection and facial feature registration from colour images. At the first stage, we use a skin-colour Gaussian model to identify possible varying pose face regions. At the second stage, we compare each face candidates with a varying pose face model using a combined feature-texture similarity measure (FTSM). False detections from the first stage are eliminated by setting an appropriate FTSM threshold. Also, one can register the facial features (eyes, nose and mouth) by aligning a prototype face with the unknown pose faces. Experimental results show that the proposed method can achieve reliable face detection and feature registration under various conditions, including different poses, face appearances, and lighting conditions.\",\n \"Title: Graph Partition Based Bundle Adjustment for Structured Dataset\\nYear: 2011\\n\\nAbstract: Bundle adjustment has been considered as one of the most important components in many visual tasks such as 3D reconstruction, photo grammetry, visual SLAM, etc. Unfortunately, both time and space complexity of this adjustment prevent it from being directly applied to large scale datasets. This paper presents a sub mapping method, which partitions a large scale dataset into disjointed subsets and adjusts them one by one or in parallel. Pair-wise sub maps are then \\\"stitched\\\" together by applying a similarity transformation. Both simulations and real applications show that our method scales well. Also some basic questions of this sub mapping method including map size, map fusion and global consistency are discussed.\",\n \"Title: Adaptive Sample Consensus for Efficient Random Optimization\\nYear: 2009\\n\\nAbstract: This paper approaches random optimization problem with adaptive sampling, which exploits knowledge about data structure obtained from historical samples. The proposal distribution is adaptive so that it invests more searching efforts on high likelihood regions. In this way, the probability of reaching the global optimum is improved. The method demonstrates improved performance as compared with standard RANSAC and related adaptive methods, for line/plane/ellipse fitting and pose estimation problems.\",\n \"Title: Supervised Binary Hash Code Learning with Jensen Shannon Divergence\\nYear: 2013\\n\\nAbstract: This paper proposes to learn binary hash codes within a statistical learning framework, in which an upper bound of the probability of Bayes decision errors is derived for different forms of hash functions and a rigorous proof of the convergence of the upper bound is presented. Consequently, minimizing such an upper bound leads to consistent performance improvements of existing hash code learning algorithms, regardless of whether original algorithms are unsupervised or supervised. This paper also illustrates a fast hash coding method that exploits simple binary tests to achieve orders of magnitude improvement in coding speed as compared to projection based methods.\",\n \"Title: In-Situ Web Mashup and Hypermedia Support for Mobile Vision-Based AR Applications\\nYear: 2013\\n\\nAbstract: Recently, mobile augmented reality (AR) applications are becoming popular and allow normal users to see an enhanced real world through their mobile camera. Creating AR content however is tedious and traditionally done on desktop environments by professionals, with extensive knowledge and/or even programming skills required. In this paper, we introduce a complete mobile approach for creating vision-based AR in both indoor and outdoor environment. Using hyperlinks, Web mashups are built to dynamically augment the physical world by normal users without programing skills. This paper presents the technical challenges, design, and the implementation with some technical discussions to improve the user experience.\",\n \"Title: Hill climbing algorithm for random sample consensus methods\\nYear: 2007\\n\\nAbstract: We propose a modification of RANSAC that performs guided search of the sample space. The sampling is applicable to any of the sample consensus methods, such as MAPSAC or MLESAC. We give simulation results which show that the new method can reduce the number of required iterations to find a good model by orders of magnitude.\",\n \"Title: Intra-class variation, affine transformation and background clutter: towards robust image matching\\nYear: 2004\\n\\nAbstract: Image matching is a fundamental computer vision problem that includes many scenarios, such as varying the view scene matching, feature selection and registration, object recognition, and general object class matching. This article presents a unified framework and working algorithm for these different matching scenarios. The proposed feature-based image matching method demonstrates excellent robustness to significant geometrical transformation, intra-class variation and background clutter which are usually presented in different matching scenarios.\",\n \"Title: Adaptive image-smoothing using a coplanar matrix and its application to document image binarization\\nYear: 2003\\n\\nAbstract: Abstract. For document images corrupted by vari-ous kinds of noise, direct binarization images may be severely blurred and degraded. A common treatment for this problem is to pre-smooth input images using noise-suppressing filters. This article proposes an image-smoothing method used for prefiltering the document im-age binarization. Conceptually, we propose that the in-fluence range of each pixel a ecting its neighbors should depend on local image statistics. Technically, we sug-gest using coplanar matrices to capture the structural and textural distribution of similar pixels at each site. This property adapts the smoothing process to the con-trast, orientation, and spatial size of local image struc-tures. Experimental results demonstrate the e ectiveness of the proposed method, which compares favorably with existing methods in reducing noise and preserving im-age features. In addition, due to the adaptive nature of the similar pixel definition, the proposed filter output is more robust regarding di erent noise levels than existing methods.\",\n \"Title: A feature-based object tracking approach for realtime image processing on mobile devices\\nYear: 2010\\n\\nAbstract: In this paper we present a robust object tracking approach which is suitable for real-time image processing on mobile devices. Challenging mobile environments render traditional color-based tracking methods useless. Many online learning tracking methods are too computationally complex to be used for real-time mobile applications, which only have access to limited computational resource and memory storage. The proposed method takes advantage of local feature to deal with rapid camera motion, and employs an online feature updating scheme to cope with variation in object appearances. The method is also computationally lightweight, being able to support real-time image processing on mobile devices.\",\n \"Title: Finger Tracking For Gestural Interaction In Mobile Devices\\nYear: 2013\\n\\nAbstract: In this paper we propose a finger tracking system that is suitable for gesture recognition in mobile devices. The initialisation of the system does not require the use of any I/O devices. The user covers the camera lense with his or her hand and then takes it to the operating distance. The statistical models used for hand segmentation are initialised from the first frames after the hand is removed from the lense. The hand segmentation does not need to be perfect because we do not use the hand contour in the recognition. In our method the fingertips are found using template matching. However, the template matching produces false detections. These false detections are pruned by searching a path from the fingertip to the estimated hand centre and discarding the paths that do not meet a predefined criteria. We evaluate the performance of the method against a fingertip detector proposed by Baldauf et al. [2] by using seven test subjects who initialise the system and then wave their hand in front of the camera. In testing we use one handheld USB camera that matches the image quality of most recent front cameras in mobile phones.\",\n \"Title: Edge-preserving prefiltering for document image binarization\\nYear: 2001\\n\\nAbstract: We propose a novel coplanar filter, which exploits the coplanarity of the gray-level distribution of neighboring pixels, to pre-filter document images. Experiments show that the proposed filter exhibits the following desired properties for document image binarization: (1) impulsive noise removal; (2) piecewise smoothing; (3) sharp edge preservation\",\n \"Title: PCA-based structure refinement for reconstruction of urban scene\\nYear: 2010\\n\\nAbstract: There is plenty of structured information (such as lines and planes) in urban scenes. Considering this, we propose a new method for making use of this information to enhance the reconstruction of urban outdoor scenes. Structured information (collinearity and coplanarity) is extracted from images by performing line detection and color image segmentation, which is used as hypothetic constraints of the 3D structure. In refining stage, we first build PCA subspaces for each structured components (collinear and coplanar point sets), during which the former hypothetic structure information is further inspected by the initial 3D structure. Then we iteratively update the structure through EM estimation. Experiments show that this method effectively improves the accuracy and robustness of reconstruction of urban scenes.\",\n \"Title: Binarizing document image using coplanar prefilter\\nYear: 2001\\n\\nAbstract: We propose a novel coplanar filter, which exploits the coplanarity of gray-level distribution of neighboring pixels, to pre-filter the document images. Experiments show that the proposed filter exhibits the following desired properties for document image binarization: 1) impulsive noise removal, 2) piecewise smoothing, and 3) sharp edge preservation\",\n \"Title: Revisiting the PnP Problem with a GPS\\nYear: 2009\\n\\nAbstract: This paper revisits the pose estimation from point correspondences problem to properly exploit data provided by a GPS. In practice, the location given by the GPS is only a noisy estimate, and some point correspondences may be erroneous. Our method therefore starts from the GPS location estimate to progressively refine the full pose estimate by hypothesizing correct correspondences. We show how the GPS location estimate and the choice of a first random correspondence dramatically reduce the possibility for a second correspondence, which in turn constrains even more the remaining possible correspondences. This results in an efficient sampling of the solution space. Experimental results on a large 3D scene show that our method outperforms standard approaches and a recent related method [1] in terms of accuracy and robustness.\",\n \"Title: Robust Scale Estimation from Ensemble Inlier Sets for Random Sample Consensus Methods\\nYear: 2008\\n\\nAbstract: This paper proposes a RANSAC modification that performs automatic estimation of the scale of inlier noise. The scale estimation takes advantage of accumulated inlier sets from all proposed models. It is shown that the proposed method gives robust results in case of high outlier ratio data, in spite that no user specified threshold is needed. The method also improves sampling efficiency, without requiring any auxiliary information other than the data to be modeled.\",\n \"Title: Real-time object tracking on mobile phones\\nYear: 2011\\n\\nAbstract: For a mobile landmark recognition and Augmented Reality system, real-time object tracking is crucial for the user experience. In this paper, an effective natural feature (low dimensional Haar-like feature) based method is proposed for real-time object tracking on mobile phones. This method can cope with image variations and alleviate the drift problem. The main idea is using an online two template switching strategy for estimating the geometrical transformation. Experiments on NOKIA N900 phone show the good performance of the proposed method and support our claim that our tracking method is effective and robust.\",\n \"Title: Object Recognition In 3d Point Cloud Of Urban Street Scene\\nYear: 2014\\n\\nAbstract: In this paper we present a novel street scene semantic recognition framework, which takes advantage of 3D point clouds captured by a high-definition LiDAR laser scanner. An important problem in object recognition is the need for sufficient labeled training data to learn robust classifiers. In this paper we show how to significantly reduce the need for manually labeled training data by reduction of scene complexity using non-supervised ground and building segmentation. Our system first automatically segments grounds point cloud, this is because the ground connects almost all other objects and we will use a connect component based algorithm to oversegment the point clouds. Then, using binary range image processing building facades will be detected. Remained point cloud will grouped into voxels which are then transformed to super voxels. Local 3D features extracted from super voxels are classified by trained boosted decision trees and labeled with semantic classes e.g. tree, pedestrian, car, etc. The proposed method is evaluated both quantitatively and qualitatively on a challenging fixed-position Terrestrial Laser Scanning (TLS) Velodyne data set and two Mobile Laser Scanning (MLS), Paris-rue-Madam and NAVTEQ True databases. Robust scene parsing results are reported.\",\n \"Title: Online Learning Of Binary Feature Indexing For Real-Time Slam Relocalization\\nYear: 2014\\n\\nAbstract: In this paper, we propose an indexing method for approximate nearest neighbor search of binary features. Being different from the popular Locality Sensitive Hashing (LSH), the proposed method construct the hash keys by an online learning process instead of pure randomness. In the learning process, the hash keys are constructed with the aim of obtaining uniform hash buckets and high collision rates, which makes the method more efficient on approximate nearest neighbor search than LSH. By distributing the online learning into the simultaneous localization and mapping (SLAM) process, we successfully apply the method to SLAM relocalization. Experiments show that camera poses can be successfully recovered in real time even there are tens of thousands of landmarks in the map.\",\n \"Title: Simple Octree Solution for Multi-resolution LiDAR Processing and Visualisation\\nYear: 2014\\n\\nAbstract: LiDAR (light detection and ranging) 3D points (both terrestrial and airborne) have recently become a prevalent type of geospatial data. However, this large-scale data poses significant technical challenges for efficient access and visualization. One approach is to generate points with different density levels, i.e. Multi-resolution data sets that can be used for various purposes. This paper presents a simple static tree-based approach and discusses implementation alternatives in favour of speeding up tree building and point sampling processes. This approach is generic and can be applied to other geographic data types as well.\",\n \"Title: Street view cross-sourced point cloud matching and registration\\nYear: 2014\\n\\nAbstract: Object registration has been widely discussed with the development of various range sensing technologies. In most work, however, the point clouds of reference and target are generated by the same technology, such as a Kinect range camera, LiDAR sensor, or Structure from Motion technique. Cases in which reference and target point clouds are generated by different technologies are rarely discussed. Due to the significant differences across various point cloud data in terms of point cloud density, sensing noise, scale, occlusion etc., object registration between such different point clouds becomes extremely difficult. In this study, we address for the first time an even more challenging case in which the differently-sourced point clouds are acquired from a real street view. One is generated on the basis of an image sequence through the SfM process, and the other is produced directly by the LiDAR system. We propose a two-stage matching and registration algorithm to achieve object registration between these two different point clouds. The experiments are based on real building object point cloud data and demonstrate the effectiveness and efficiency of the proposed solution. The newly proposed solution can be further developed to contribute to several related applications, such as Location Based Service.\",\n \"Title: A 3D map augmented photo gallery application on mobile device\\nYear: 2014\\n\\nAbstract: This paper proposes a 3D map augmented photo gallery mobile application that allows user to virtually transit from 2D image space to the 3D map space, to expand the field of view to surrounding environments that are not visible in the original image, and to change viewing angles among different global registered images during the image browsing. The processing of images consists of two main steps: in the first step, the client application uploads an image to the GeoImage Engine which extracts the geo-metadata and returns them back to the client; in the second step, the client application requests augmented content from server, and then renders 3D view of images on the screen of mobile devices.\",\n \"Title: Efficient pose tracking on mobile phones with 3D points grouping\\nYear: 2014\\n\\nAbstract: With the rapid growth of computational capability and popularity of mobile phones, Mobile Augmented Reality (MAR) in large scale 3D scenes becomes an emerging field in recent years. The core of MAR is to continuously compute a precise 6 Degree-of-Freedom (DOF) camera pose for each frame, i.e. localization. However, as a crucial part of localization, the 2D-3D points matching is usually inefficient due to the usage of traditional features, e.g. SIFT, and the large number of 3D points candidates. This paper aims to tackle this problem by designing an efficient 6DOF pose tracking system on mobile phone. In this system, binary features are used in both offline sparse reconstruction and online tracking, while a PCA-based 3D points partition method is proposed to reduce the searching space of 2D-3D points matching, making it capable to achieve a low computational cost. Experiments on a NOKIA N900 smartphone show that our system could efficiently and robustly estimate the 6DOF camera pose.\",\n \"Title: Fast Localization in Large-Scale Environments Using Supervised Indexing of Binary Features\\nYear: 2016\\n\\nAbstract: The essence of image-based localization lies in matching 2D key points in the query image and 3D points in the database. State-of-the-art methods mostly employ sophisticated key point detectors and feature descriptors, e.g., Difference of Gaussian (DoG) and Scale Invariant Feature Transform (SIFT), to ensure robust matching. While a high registration rate is attained, the registration speed is impeded by the expensive key point detection and the descriptor extraction. In this paper, we propose to use efficient key point detectors along with binary feature descriptors, since the extraction of such binary features is extremely fast. The naive usage of binary features, however, does not lend itself to significant speedup of localization, since existing indexing approaches, such as hierarchical clustering trees and locality sensitive hashing, are not efficient enough in indexing binary features and matching binary features turns out to be much slower than matching SIFT features. To overcome this, we propose a much more efficient indexing approach for approximate nearest neighbor search of binary features. This approach resorts to randomized trees that are constructed in a supervised training process by exploiting the label information derived from that multiple features correspond to a common 3D point. In the tree construction process, node tests are selected in a way such that trees have uniform leaf sizes and low error rates, which are two desired properties for efficient approximate nearest neighbor search. To further improve the search efficiency, a probabilistic priority search strategy is adopted. Apart from the label information, this strategy also uses non-binary pixel intensity differences available in descriptor extraction. By using the proposed indexing approach, matching binary features is no longer much slower but slightly faster than matching SIFT features. Consequently, the overall localization speed is significantly improved due to the much faster key p- int detection and descriptor extraction. It is empirically demonstrated that the localization speed is improved by an order of magnitude as compared with state-of-the-art methods, while comparable registration rate and localization accuracy are still maintained.\",\n \"Title: A weakly supervised geodesic level set framework for interactive image segmentation\\nYear: 2015\\n\\nAbstract: Interactive image segmentation is growingly useful for selecting objects of interest in images, facilitating spatially localized media manipulation especially on touch screen devices. We present a robust and efficient approach for segmenting image with less and intuitive user interaction. Our approach combines geodesic distance information with the flexibility of level set methods in energy minimization, leveraging the complementary strengths of each to promote accurate boundary placement and strong region connectivity while requiring less user interaction. We harness weakly supervised segment annotation to maximize the user-provided prior knowledge. This leads to a seed generation algorithm which enables image object segmentation without user-provided background seeds. We demonstrate that our approach is less sensitive to seed placement and better at edge localization, whilst requiring less user interaction, compared with the state-of-the-art methods.\",\n \"Title: Using the Similarity Measure between Intuitionistic Fuzzy Sets for the Application on Pattern Recognitions\\nYear: 2015\\n\\nAbstract: AbstractThe measurement of uncertainty is an important topic for the theories dealing with uncertainty. The definition of similarity measure between two IFSs is one of the most interesting topics in IFSs theory. A similarity measure is defined to compare the information carried by IFSs. Many similarity measures have been proposed. A few of them come from the well-known distance measures. In this work, a new similarity measure between IFSs was proposed by the consideration of the information carried by the membership degree, the non-membership degree, and hesitancy degree in intuitionistic fuzzy sets IFSs. To demonstrate the efficiency of the proposed similarity measure, various similarity measures between IFSs were compared with the proposed similarity measure between IFSs by numerical examples. The compared results demonstrated that the new similarity measure is reasonable and has stronger discrimination among them. Finally, the similarity measure was applied to pattern recognition and medical diagnosis. Two illustrative examples were provided to show the effectiveness of the pattern recognition and medical diagnosis.\",\n \"Title: Automated super-voxel based features classification of urban environments by integrating 3D point cloud and image content\\nYear: 2015\\n\\nAbstract: In this paper we present a novel street scene semantic recognition framework, which takes advantage of 3D point cloud captured by a high definition LiDAR laser scanner. An important problem in object recognition is the need for sufficient labeled training data to learn robust classifiers. We show how to significantly reduce the need for manually labeled training data by reduction of scene complexity using non-supervised ground and building segmentation. Our system first automatically segments grounds point cloud. Then, using binary range image processing building facades will be detected. Remained point cloud will grouped into voxels which are then transformed to super voxels. Local 3D features extracted from super voxels are classified by trained boosted decision trees and labeled with semantic classes e.g. tree, pedestrian, car. Given labeled 3D points cloud and 2D image with known viewing camera pose, the proposed association module aligned collections of 3D points to the groups of 2D image pixel to parsing 2D cubic images.\",\n \"Title: Robust interactive image segmentation with weak supervision for mobile touch screen devices\\nYear: 2015\\n\\nAbstract: In this paper, we present a robust and efficient approach for segmenting images with less and intuitive user interaction, particularly targeted for mobile touch screen devices. Our approach combines geodesic distance information with the flexibility of level set methods in energy minimization, leveraging the complementary strengths of each to promote accurate boundary placement and strong region connectivity while requiring less user interaction. To maximize the user-provided prior knowledge, we further propose a weakly supervised seed generation algorithm which enables image object segmentation without user-provided background seeds. Our approach provides a practical solution for visual object cutout on mobile touch screen devices, facilitating various media manipulation applications. We describe such a use case to selectively create oil painting effects on images. We demonstrate that our approach is less sensitive to seed placement and better at edge localization, whilst requiring less user interaction, compared with the state-of-the-art methods.\",\n \"Title: Accurate Robust and Efficient Error Estimation for Decision Trees.\\nYear: 2016\\n\\nAbstract: This paper illustrates a novel approach to the estimation of generalization error of decision tree classifiers. We set out the study of decision tree errors in the context of consistency analysis theory, which proved that the Bayes error can be achieved only if when the number of data samples thrown into each leaf node goes to infinity. For the more challenging and practical case where the sample size is finite or small, a novel sampling error term is introduced in this paper to cope with the small sample problem effectively and efficiently. Extensive experimental results show that the proposed error estimate is superior to the well-known K-fold cross validation methods in terms of robustness and accuracy. Moreover it is orders of magnitudes more efficient than cross validation methods.\",\n \"Title: Global Scale Integral Volumes.\\nYear: 2016\\n\\nAbstract: Integral volume is an important image representation technique, which is useful in many computer vision applications. Processing integral volumes for large scale 3D datasets is challenging due to high memory requirements. The difficulties lie in efficiently computing, storing, querying and updating the integral volume values. In this work, we address the above problems and present a novel solution for processing integral volumes for large scale 3D datasets efficiently. We propose an octree-based method where the worst-case complexity for querying the integral volume of arbitrary regions is (mathcal {O}(log {}n)), here n is the number of nodes in the octree. We evaluate our proposed method on multi-resolution LiDAR point cloud data. Our work can serve as a tool to fast extract features from large scale 3D datasets, which can be beneficial for computer vision applications.\",\n \"Title: Weighted fuzzy track association method based on Dempster-Shafer theory in distributed sensor networks.\\nYear: 2016\\n\\nAbstract: The uncertainty problem in sensor track to local track association is a difficult problem in distributed sensor networks, particularly when there is a big difference of sensors’ tracking performance. To solve this problem, a weighted fuzzy track association (FTA) method based on Dempster–Shafer theory is proposed. In the proposed method, five characteristics of sensor tracks from different sensors are established, and meanwhile their belief functions are defined to determine the corresponding beliefs. Considering the different effects of sensor tracks on track association, the reliabilities of sensor tracks are further presented and their magnitudes can be calculated by the combination belief function defined. Then, these reliabilities are used to reconstruct the fuzzy association degrees by the FTA method. The proposed method has an advantage that it can dynamically allocate the weight of each sensor track in association decision according to its characteristics. The performance of the proposed method is evaluated by using two experiments with simulation data in manoeuvring and uniform situations. It is found to be better than those of other two track association methods in tracking accuracy.\",\n \"Title: Comprehensive Automated 3D Urban Environment Modelling Using Terrestrial Laser Scanning Point Cloud\\nYear: 2016\\n\\nAbstract: In this paper we present a novel street scene modelling framework, which takes advantage of 3D point cloud captured by a high definition LiDAR laser scanner. We propose an automatic and robust approach to detect, segment and classify urban objects from point clouds hence reconstructing a comprehensive 3D urban environment model. Our system first automatically segments grounds point cloud. Then building facades will be detected by using binary range image processing. Remained point cloud will be grouped into voxels and subsequently transformed into super voxels. Local 3D features are extracted from super voxels and classified by trained boosted decision trees with semantic classes e.g. tree, pedestrian, and car. Given labeled point cloud the proposed algorithm reconstructs the realistic model in two phases. Firstly building facades will be rendered by ShadVis algorithm. In the second step we apply a novel and fast method for fitting the solid predefined template mesh models to non-building labeled point cloud. The proposed method is evaluated both quantitatively and qualitatively on a challenging TLS NAVTEQ True databases.\",\n \"Title: Deep Epitome for Unravelling Generalized Hamming Network: A Fuzzy Logic Interpretation of Deep Learning.\\nYear: 2017\\n\\nAbstract: This paper gives a rigorous analysis of trained Generalized Hamming Networks(GHN) proposed by Fan (2017) and discloses an interesting finding about GHNs, i.e., stacked convolution layers in a GHN is equivalent to a single yet wide convolution layer. The revealed equivalence, on the theoretical side, can be regarded as a constructive manifestation of the universal approximation theorem Cybenko(1989); Hornik (1991). In practice, it has profound and multi-fold implications. For network visualization, the constructed deep epitomes at each layer provide a visualization of network internal representation that does not rely on the input data. Moreover, deep epitomes allows the direct extraction of features in just one step, without resorting to regularized optimizations used in existing visualization tools.\",\n \"Title: Urban 3D segmentation and modelling from street view images and LiDAR point clouds.\\nYear: 2017\\n\\nAbstract: 3D urban maps with semantic labels and metric information are not only essential for the next generation robots such autonomous vehicles and city drones, but also help to visualize and augment local environment in mobile user applications. The machine vision challenge is to generate accurate urban maps from existing data with minimal manual annotation. In this work, we propose a novel methodology that takes GPS registered LiDAR (Light Detection And Ranging) point clouds and street view images as inputs and creates semantic labels for the 3D points clouds using a hybrid of rule-based parsing and learning-based labelling that combine point cloud and photometric features. The rule-based parsing boosts segmentation of simple and large structures such as street surfaces and building facades that span almost 75% of the point cloud data. For more complex structures, such as cars, trees and pedestrians, we adopt boosted decision trees that exploit both structure (LiDAR) and photometric (street view) features. We provide qualitative examples of our methodology in 3D visualization where we construct parametric graphical models from labelled data and in 2D image segmentation where 3D labels are back projected to the street view images. In quantitative evaluation we report classification accuracy and computing times and compare results to competing methods with three popular databases: NAVTEQ True, Paris-Rue-Madame and TLS (terrestrial laser scanned) Velodyne.\",\n \"Title: Real-time SLAM relocalization with online learning of binary feature indexing.\\nYear: 2017\\n\\nAbstract: A visual simultaneous localization and mapping (SLAM) system usually contains a relocalization module to recover the camera pose after tracking failure. The core of this module is to establish correspondences between map points and key points in the image, which is typically achieved by local image feature matching. Since recently emerged binary features have orders of magnitudes higher extraction speed than traditional features such as scale invariant feature transform, they can be applied to develop a real-time relocalization module once an efficient method of binary feature matching is provided. In this paper, we propose such a method by indexing binary features with hashing. Being different from the popular locality sensitive hashing, the proposed method constructs the hash keys by an online learning process instead of pure randomness. Specifically, the hash keys are trained with the aim of attaining uniform hash buckets and high collision rates of matched feature pairs, which makes the method more efficient on approximate nearest neighbor search. By distributing the online learning into the simultaneous localization and mapping process, we successfully apply the method to SLAM relocalization. Experiments show that camera poses can be recovered in real time even when there are tens of thousands of landmarks in the map.\",\n \"Title: Memory-Efficient Deep Salient Object Segmentation Networks on Gridized Superpixels\\nYear: 2018\\n\\nAbstract: Computer vision algorithms with pixel-wise labeling tasks, such as semantic segmentation and salient object detection, have gone through a significant accuracy increase with the incorporation of deep learning. Deep segmentation methods slightly modify and fine-tune pre-trained networks that have hundreds of millions of parameters. In this work, we question the need of having such memory demanding networks for a reasonable performance in salient object segmentation. To this end, we propose a way to learn a memory-efficient network from scratch by training it only on salient object detection datasets. Our method encodes images to gridized superpixels that preserve both the object boundaries and the connectivity rules of regular pixels. This representation allows us to use convolutional neural networks that operate on regular grids. By using these encoded images, we train a memory-efficient network using only 0.048% of the number of parameters that a majority of other deep salient object detection networks have. Our method shows comparable accuracy with the state-of-the-art deep salient object detection methods and provides a much more memory-efficient alternative to them. Due to its easy deployment and small size, such a network is preferable for applications in memory limited IoT devices.\",\n \"Title: A Theoretical Investigation of Graph Degree as an Unsupervised Normality Measure.\\nYear: 2018\\n\\nAbstract: For a graph representation of a dataset, a straightforward normality measure for a sample can be its graph degree. Considering a weighted graph, this corresponds to sum of rows in a similarity matrix. The measure is intuitive given the abnormal samples are usually rare and they are dissimilar to the rest of the data. In order to have an in-depth theoretical understanding, in this manuscript, we investigate the graph degree in spectral graph clustering based and kernel based point of views and draw connections to a recent kernel method for the two sample problem. We show that our analyses guide us to choose fully-connected graphs whose edge weights are calculated via universal kernels. We show that a simple graph degree based unsupervised anomaly detection method with the above properties, achieves leading accuracy compared to other unsupervised anomaly detection methods on average over 10 widely used datasets. We also provide an extensive analysis on the effect of the kernel parameter on the methodu0027s accuracy.\",\n \"Title: Depth Masked Discriminative Correlation Filter\\nYear: 2018\\n\\nAbstract: Depth information provides a strong cue for occlusion detection and handling, but has been largely omitted in generic object tracking until recently due to lack of suitable benchmark datasets and applications. In this work, we propose a Depth Masked Discriminative Correlation Filter (DM-DCF) which adopts novel depth segmentation based occlusion detection that stops correlation filter updating and depth masking which adaptively adjusts the spatial support for correlation filter. In Princeton RGBD Tracking Benchmark, our DM-DCF is among the state-of-the-art in overall ranking and the winner on multiple categories. Moreover, since it is based on DCF, \\\"DM-DCF\\\" runs an order of magnitude faster than its competitors making it suitable for time constrained applications.\",\n \"Title: Object Detection in Equirectangular Panorama.\\nYear: 2018\\n\\nAbstract: We introduce a high-resolution equirectangular panorama (aka 360-degree, virtual reality, VR) dataset for object detection and propose a multi-projection variant of the YOLO detector. The main challenges with equirectangular panorama images are i) the lack of annotated training data, ii) high-resolution imagery and iii) severe geometric distortions of objects near the panorama projection poles. In this work, we solve the challenges by I) using training examples available in the “conventional datasets” (ImageNet and COCO), II) employing only low resolution images that require only moderate GPU computing power and memory, and III) our multi-projection YOLO handles projection distortions by making multiple stereographic sub-projections. In our experiments, YOLO outperforms the other state-of-the-art detector, Faster R-CNN, and our multi-projection YOLO achieves the best accuracy with low-resolution input.\",\n \"Title: Deep Epitome for Unravelling Generalized Hamming Network\\nYear: 2018\\n\\nAbstract: This paper gives a rigorous analysis of trained Generalized Hamming Networks (GHN) proposed by [9] and discloses an interesting finding about GHNs, i.e., stacked convolution layers in a GHN is equivalent to a single yet wide convolution layer. The revealed equivalence, on the theoretical side, can be regarded as a constructive manifestation of the universal approximation theorem [7], [16]. In practice, it has profound and multi-fold implications. For network visualization, the constructed deep epitomes at each layer provide a visualization of network internal representation that does not rely on the input data. Moreover, deep epitomes allows the direct extraction of features in just one step, without resorting to regularized optimizations used in existing visualization tools.\",\n \"Title: Automatic Feature Extraction For Wide-Angle And Fish-Eye Camera Calibration\\nYear: 2018\\n\\nAbstract: The increase need to have 360 degree perception of world environment has led to the combination of various cameras in the varieties of narrow and wide-angle field of view. Current state of art methods do not provide an automatic means for fish-eye calibration, which is indispensable in an industrial environment where many lenses are to be calibrated in a relative short time. As automatic feature extraction is the key issue for fully automatic calibration framework, we address this issue to remove any form of human intervention during the calibration process.Unlike state-of-the-art methods, the proposed framework is completely automatic and not prone to detection errors. We evaluate the proposed feature extraction using state-of-the-art generic calibration model with both real and synthetic images.\",\n \"Title: Generic Calibration Of Cameras With Non-Parallel Optical Elements\\nYear: 2018\\n\\nAbstract: Multiple cameras are increasingly prevalent for autonomous driving, and the increase need to have 360 degree perception of world space has led to the combination of various cameras in the varieties of narrow-angle and wide-angle field of view. This has led to issues regarding the quality of these optics as a result of bad and cheap designs. Intrinsic calibration is indispensable for accurate perception of the environment in order for it to make accurate decisions such as camera pose estimation and 3-D reconstruction.In this work, we propose a lens distortion model that has been motivated by unintentional tilt in the optical lens system. The proposed distortion model has been added to current state of art generic method of Kannala to form an extended model. To our knowledge, this is the first time that the idea of tilt distortion has been introduced for wide-angle view and fish-eye cameras. We show improved result of 4 to 13% from experiments.\",\n \"Title: Happy Travelers Take Big Pictures: A Psychological Study with Machine Learning and Big Data.\\nYear: 2017\\n\\nAbstract: In psychology, theory-driven researches are usually conducted with extensive laboratory experiments, yet rarely tested or disproved with big data. In this paper, we make use of 418K travel photos with traveler ratings to test the influential broaden-and-build theory, that suggests positive emotions broaden oneu0027s visual attention. The core hypothesis examined in this study is that positive emotion is associated with a wider attention, hence highly-rated sites would trigger wide-angle photographs. By analyzing travel photos, we find a strong correlation between a preference for wide-angle photos and the high rating of tourist sites on TripAdvisor. We are able to carry out this analysis through the use of deep learning algorithms to classify the photos into wide and narrow angles, and present this study as an exemplar of how big data and deep learning can be used to test laboratory findings in the wild.\",\n \"Title: Proceedings of AAAI 2019 Workshop on Network Interpretability for Deep Learning.\\nYear: 2019\\n\\nAbstract: This is the Proceedings of AAAI 2019 Workshop on Network Interpretability for Deep Learning\",\n \"Title: A Universal Logic Operator for Interpretable Deep Convolution Networks.\\nYear: 2019\\n\\nAbstract: Explaining neural network computation in terms of probabilistic/fuzzy logical operations has attracted much attention due to its simplicity and high interpretability. Different choices of logical operators such as AND, OR and XOR give rise to another dimension for network optimization, and in this paper, we study the open problem of learning a universal logical operator without prescribing to any logical operations manually. Insightful observations along this exploration furnish deep convolution networks with a novel logical interpretation.\",\n \"Title: Digital Passport: A Novel Technological Strategy for Intellectual Property Protection of Convolutional Neural Networks.\\nYear: 2019\\n\\nAbstract: In order to prevent deep neural networks from being infringed by unauthorized parties, we propose a generic solution which embeds a designated digital passport into a network, and subsequently, either paralyzes the network functionalities for unauthorized usages or maintain its functionalities in the presence of a verified passport. Such a desired network behavior is successfully demonstrated in a number of implementation schemes, which provide reliable, preventive and timely protections against tens of thousands of fake-passport deceptions. Extensive experiments also show that the deep neural network performance under unauthorized usages deteriorate significantly (e.g. with 33% to 82% reductions of CIFAR10 classification accuracies), while networks endorsed with valid passports remain intact.\",\n \"Title: DeepIPR: Deep Neural Network Ownership Verification With Passports\\nYear: 2022\\n\\nAbstract: With substantial amount of time, resources and human (team) efforts invested to explore and develop successful deep neural networks (DNN), there emerges an urgent need to protect these inventions from being illegally copied, redistributed, or abused without respecting the intellectual properties of legitimate owners. Following recent progresses along this line, we investigate a number of watermark-based DNN ownership verification methods in the face of ambiguity attacks, which aim to cast doubts on the ownership verification by forging counterfeit watermarks. It is shown that ambiguity attacks pose serious threats to existing DNN watermarking methods. As remedies to the above-mentioned loophole, this paper proposes novel \\n<italic xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">passport</i>\\n-based DNN ownership verification schemes which are both \\n<italic xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">robust to network modifications</i>\\n and \\n<italic xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">resilient to ambiguity attacks</i>\\n. The gist of embedding digital passports is to design and train DNN models in a way such that, the DNN inference performance of an original task will be significantly \\n<italic xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">deteriorated due to forged passports</i>\\n. In other words, genuine passports are not only verified by looking for the predefined signatures, but also reasserted by the \\n<italic xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">unyielding DNN model inference performances</i>\\n. Extensive experimental results justify the effectiveness of the proposed passport-based DNN ownership verification schemes. Code is available at \\n<uri xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">https://github.com/kamwoh/DeepIPR</uri>\\n.\",\n \"Title: ICDAR 2021 Competition on Integrated Circuit Text Spotting and Aesthetic Assessment\\nYear: 2021\\n\\nAbstract: With hundreds of thousands of electronic chip components are being manufactured every day, chip manufacturers have seen an increasing demand in seeking a more efficient and effective way of inspecting the quality of printed texts on chip components. The major problem that deters this area of research is the lacking of realistic text on chips datasets to act as a strong foundation. Hence, a text on chips dataset, ICText is used as the main target for the proposed Robust Reading Challenge on Integrated Circuit Text Spotting and Aesthetic Assessment (RRC-ICText) 2021 to encourage the research on this problem. Throughout the entire competition, we have received a total of 233 submissions from 10 unique teams/individuals. Details of the competition and submission results are presented in this report.\",\n \"Title: L2RS: A Learning-to-Rescore Mechanism for Hybrid Speech Recognition\\nYear: 2021\\n\\nAbstract: ABSTRACTThis paper aims to advance the performance of industrial ASR systems by exploring a more effective method for N-best rescoring, a critical step that greatly affects the final recognition accuracy. Existing rescoring approaches suffer the following issues: (i) limited performance since they optimize an unnecessarily harder problem, namely predicting accurate grammatical legitimacy scores of the N-best hypotheses rather than directly predicting their partial orders regarding a specific acoustic input; (ii) hard to incorporate various information by advanced natural language processing (NLP) models such as BERT to achieve a comprehensive evaluation of each N-best candidate. To relieve the above drawbacks, we propose a simple yet effective mechanism, Learning-to-Rescore (L2RS), to empower ASR systems with state-of-the-art information retrieval (IR) techniques. Specifically, L2RS utilizes a wide range of textual information from the state-of-the-art NLP models and automatically deciding their weights to directly learn the ranking order of each N-best hypothesis with respect to a specific acoustic input. We incorporate various features including BERT sentence embeddings, the topic vectors, and perplexity scores produced by an n-gram language model (LM), topic modeling LM, BERT, and RNNLM to train the rescoring model. Experimental results on a public dataset show that L2RS outperforms not only traditional rescoring methods but also its deep neural network counterparts by a substantial margin of 20.85% in terms of [email protected] The L2RS toolkit has been successfully deployed for many online commercial services in WeBank Co., Ltd, China's leading digital bank. The efficacy and applicability of L2RS are validated by real-life online customer datasets.\",\n \"Title: A Health-Friendly Speaker Verification System Supporting Mask Wearing\\nYear: 2021\\n\\nAbstract: We demonstrate a health-friendly speaker verification system for voice-based identity verification on mobile devices. The system is built upon a speech processing module, a ResNet-based local acoustic feature extractor and a multi head attention-based embedding layer, and is optimized under an additive margin softmax loss for discriminative speaker verification. It is shown that the system achieves superior performance no matter whether there is mask wearing or not. This characteristic is important for speaker verification services operating in regions affected by the raging coronavirus pneumonia. With this demonstrationl, the audience will have an in-depth experience of how the accuracy of bio-metric verification and the personal health are simultaneously ensured. We wish that this demonstration would boost the development of next-generation bio-metric verification technologies.\",\n \"Title: Protect, show, attend and tell: Empowering image captioning models with ownership protection\\nYear: 2022\\n\\nAbstract: •We propose a key-based strategy that provides reliable, preventive and timely IP protection for image captioning task.•We empirically show the effectiveness of our approach against various attacks and prove the ownership of the model.•To the best of our knowledge, we are the first to propose IP protection on image captioning model.\",\n \"Title: You Don't Know My Favorite Color: Preventing Dialogue Representations from Revealing Speakers' Private Personas\\nYear: 2022\\n\\nAbstract: Social chatbots, also known as chit-chat chatbots, evolve rapidly with large pretrained language models. Despite the huge progress, privacy concerns have arisen recently: training data of large language models can be extracted via model inversion attacks. On the other hand, the datasets used for training chatbots contain many private conversations between two individuals. In this work, we further investigate the privacy leakage of the hidden states of chatbots trained by language modeling which has not been well studied yet. We show that speakers' personas can be inferred through a simple neural network with high accuracy. To this end, we propose effective defense objectives to protect persona leakage from hidden states. We conduct extensive experiments to demonstrate that our proposed defense objectives can greatly reduce the attack accuracy from 37.6% to 0.5%. Meanwhile, the proposed objectives preserve language models' powerful generation ability.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":63,"string":"63"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13586,"string":"13,586"}}},{"rowIdx":13587,"cells":{"question_id":{"kind":"string","value":"A L Narasimha Reddy"},"question":{"kind":"string","value":"Generate an abstract for the title \"Scheduling and data distribution in a multiprocessor video server\" using the following items: 1. Multiprocessor video server\n2. Scheduling movies\n3. Communication scheduling\n4. Contention minimization\n5. Simple movie scheduling"},"answer":{"kind":"string","value":"In this paper, we will address the problem of distributing and scheduling movies on a multiprocessor video server. We will also address the issue of scheduling communication over the multiprocessor switch for the playback of the scheduled movies. A solution is proposed in this paper that addresses these three issues at once. The proposed solution minimizes contention for links over the switch. The proposed solution makes movie scheduling very simple - if the first block of the movie is scheduled, the rest of the movie is automatically scheduled. Moreover, if the first block of the movie stream is scheduled without network contention, the proposed solution guarantees that there will be no network contention during the entire duration of playback of that movie."},"correct_answer":{"kind":"string","value":"In this paper, we will address the problem of distributing and scheduling movies on a multiprocessor video server. We will also address the issue of scheduling communication over the multiprocessor switch for the playback of the scheduled movies. A solution is proposed in this paper that addresses these three issues at once. The proposed solution minimizes contention for links over the switch. The proposed solution makes movie scheduling very simple - if the first block of the movie is scheduled, the rest of the movie is automatically scheduled. Moreover, if the first block of the movie stream is scheduled without network contention, the proposed solution guarantees that there will be no network contention during the entire duration of playback of that movie."},"documents":{"kind":"list like","value":["Title: Providing QoS guarantees for disk I/O\nYear: 2000\n\nAbstract: In this paper, we address the problem of providing different levels of performance guarantees or quality ofservice for disk I/O. We classify disk requests into threecategories based on the provided level of service. We propose an integrated scheme that provides different levels ofperformance guarantees in a single system. We propose andevaluate a mechanism for providing deterministic servicefor variable-bit-rate streams at the disk. We will show that,through proper admission control and bandwidth allocation,requests in different categories can be ensured of performance guarantees without getting impacted by requests inother categories. We evaluate the impact of scheduling policy decisions on the provided service. We also quantify theimprovements in stream throughput possible by using statistical guarantees instead of deterministic guarantees in thecontext of the proposed approach.","Title: Exploiting Concurrency to Improve Latency and throughput in a Hybrid Storage System\nYear: 2010\n\nAbstract: This paper considers the problem of how to improve the performance of hybrid storage system employing solid state disks and hard disk drives. We utilize both initial block allocation as well as migration to reach “Wardrop equilibrium”, in which the response times of different devices equalize. We show that such a policy allows adaptive load balancing across devices of different performance. We also show that such a policy exploits parallelism in the storage system effectively to improve throughput and latency simultaneously. We implemented a prototype in Linux and evaluated it in multiple workloads and multiple configurations. The results show that the proposed approach improved both the latency of requests and the throughput significantly, and it adapted to different configurations of the system under different workloads.","Title: Reaching approximate Wardrop equilibrium at reduced costs of link state updates\nYear: 2011\n\nAbstract: Several routing protocols have been proposed to take advantage of the dynamic metrics on links such as link delays, queueing lengths, and available link bandwidths.Wardrop routing, QOS routing are few such example routing protocols. Even when these protocols can be shown to offer convergence properties without oscillations, the protocols have not been widely adopted for a number of reasons. The expected cost of keeping the link metrics updated at various nodes in the network being one of them. In this paper, we study the problem of reducing the cost of propagating dynamic link metrics while keeping the quality of paths within a margin of error. We propose a simple technique of threshold-based metric propagation that is shown, through analysis and simulations, to offer bounded guarantees on path quality while significantly reducing the cost of propagating the dynamic link metric information. Our results indicate that threshold based updates can reduce the number of link updates by up to 90-95% in some cases.","Title: Write amplification due to ECC on flash memory or leave those bit errors alone\nYear: 2012\n\nAbstract: While flash memory is receiving significant attention because of many attractive properties, concerns about write endurance delay the wider deployment of the flash memory. This paper analyzes the effectiveness of protection schemes designed for flash memory, such as ECC and scrubbing. The bit error rate of flash memory is a function of the number of program-erase cycles the cell has gone through, making the reliability dependent on time and workload. Moreover, some of the protection schemes require additional write operations, which degrade flash memory's reliability. These issues make it more complex to reveal the relationship between the protection schemes and flash memory's lifetime. In this paper, a Markov model based analysis of the protection schemes is presented. Our model considers the time varying reliability of flash memory as well as write amplification of various protection schemes such as ECC. Our study shows that write amplification from these various sources can significantly affect the benefits of these schemes in improving the lifetime. Based on the results from our analysis, we propose that bit errors within a page be left uncorrected until a threshold of errors are accumulated. We show that such an approach can significantly improve lifetimes by up to 40%.","Title: TCP-DCR: Making TCP Robust to Non-congestion Events\nYear: 2004\n\nAbstract: In this paper, we propose and evaluate TCP-DCR. TCP-DCR makes simple modifications to the TCP congestion control algorithm to make it more robust to non-congestion events. The key idea here is to delay the congestion response of TCP for a short interval of time tau, thereby creating room for local recovery mechanisms to handle any non-congestion events that may have occurred. If at the end of the delay tau, the event is not handled, then it is treated as a congestion loss. We evaluate TCP-DCR through analysis and simulations. The evaluation is done for three scenarios - a wireless network with channel errors, a wired network with packet reordering and a network with zero non-congestion events. The simulation results show that significant performance improvements can be achieved by using TCP-DCR in the presence of non-congestion events with zero or marginal impact in the absence of non-congestion events. TCP-DCR remains fair to the native implementations of TCP that respond to congestion immediately after receiving three dupacks. TCP-DCR is a simple, effective scheme providing a unified solution to several problems with minimal implementation overhead.","Title: Don't let RAID raid the lifetime of your SSD array\nYear: 2013\n\nAbstract: Parity protection at system level is typically employed to compose reliable storage systems. However, careful consideration is required when SSD based systems employ parity protection. First, additional writes are required for parity updates. Second, parity consumes space on the device, which results in write amplification from less efficient garbage collection at higher space utilization. This paper analyzes the effectiveness of SSD based RAID and discusses the potential benefits and drawbacks in terms of reliability. A Markov model is presented to estimate the lifetime of SSD based RAID systems in different environments. In a single array, our preliminary results show that parity protection provides benefit only with considerably low space utilizations and low data access rates. However, in a large system, RAID improves data lifetime even when we take write amplification into account.","Title: Disjoint Multi-Path Routing and Failure Recovery\nYear: 2010\n\nAbstract: Applications such as Voice over IP and video delivery require continuous network service, requiring fast failure recovery mechanisms. Proactive Fast failure recovery mechanisms have been recently proposed to improve network performance during the failure transients. The proposed mechanisms need extra infrastructural support in the form of routing table entries, extra addresses etc. In this paper, we study if the extra infrastructure support can be exploited to build disjoint paths in those frameworks, while keeping the recovery path lengths close to the primary paths. Our evaluations show that it is possible to extend the proactive recovery mechanisms to provide support for nearly-disjoint paths.","Title: Adaptive marking for assured forwarding service\nYear: 2002\n\nAbstract: This letter looks at the problem of achieving specific quality of service (QoS) goals of individual flows by flexibly managing resources available to an aggregated source. We present an adaptive marker based on a TCP performance model within a Diffserv network, and show that the aggregated marker can maintain state of individual flows at the edge of the network and utilize this state effectively in adaptively marking packets of individual flows to meet their QoS goals.","Title: Design and evaluation of gracefully degradable disk arrays\nYear: 1993\n\nAbstract: : The availability of inexpensive, small, magnetic disks, has made possible thebuilding of a reliable, high-performance disk system by integrating a number of such disks inan array. To achieve high reliability in such systems, equivalent to that of larger disks, parityor other error-correcting codes may be used. In systems where data availability is critical,traditionally dual copy methods have been used. Recently some parity-based schemes havebeen proposed for providing fault tolerance...","Title: A Client Oriented, Ip Level Redirection Mechanism\nYear: 1999\n\nAbstract: This paper introduces a new approach for implementing transparent client access to network services. Ever increasing load on the Internet has made it essential to design services that are fast, reliable, easily manageable, transparent to access, and that can scale gracefully with load. A common way of achieving this has been replicating services across multiple servers and redirecting clients to different servers depending upon various criteria. Existing schemes are either entirely server or network based. This scheme involves the client network layer actively in redirection. The paper describes the redirection protocol in detail and a FreeBSD based implementation of a testbed. The performance of the mechanism is measured by experiments on the testbed and analyzed. The advantages and disadvantages of client based network level redirection are discussed and some useful applications that it enables are described.","Title: NVMFS: A hybrid file system for improving random write in nand-flash SSD\nYear: 2013\n\nAbstract: In this paper, we design a storage system consisting of Nonvolatile DIMMs (as NVRAM) and NAND-flash SSD. We propose a file system NVMFS to exploit the unique characteristics of these devices which simplifies and speeds up file system operations. We use the higher performance NVRAM as both a cache and permanent space for data. Hot data can be permanently stored on NVRAM without writing back to SSD, while relatively cold data can be temporarily cached by NVRAM with another copy on SSD. We also reduce the erase overhead of SSD by reorganizing writes on NVRAM before flushing to SSD. We have implemented a prototype NVMFS within a Linux Kernel and compared with several modern file systems such as ext3, btrfs and NILFS2. We also compared with another hybrid file system Conquest, which originally was designed for NVRAM and HDD. The experimental results show that NVMFS improves IO throughput by an average of 98.9% when segment cleaning is not active, while improves throughput by an average of 19.6% under high disk utilization (over 85%) compared to other file systems. We also show that our file system can reduce the erase operations and overheads at SSD.","Title: An Evaluation of the Effectiveness of Measurement-based Anomaly Detection Techniques\nYear: 2006\n\nAbstract: A number of recent studies have proposed measurement based approaches to network traffic analysis. These techniques treat traffic volume and traffic header data as signals or images in order to make analysis feasible. We use trace-driven experiments and compare the performance of different strategies. Our evaluations on real traces reveal differences in the effectiveness of different traffic header data as potential signals for traffic analysis in terms of their detection rates and false alarm rates. Our results show that address distributions and number of flows are better signals than traffic volume for anomaly detection.","Title: ENDE: An End-to-end Network Delay Emulator Tool for Multimedia Protocol Development\nYear: 2001\n\nAbstract: Multimedia applications and protocols are constantly being developed to run over the Internet. A new protocol or application after being developed has to be tested on the real Internet or simulated on a testbed for debugging and performance evaluation. In this paper, we present a novel tool, ENDE, that can emulate end-to-end delays between two hosts without requiring access to the second host. The tool enables the user to test new multimedia protocols realistically on a single machine. In a delay-observing mode, ENDE can generate accurate traces of one-way delays between two hosts on the network. In a delay-impacting mode, ENDE can be used to simulate the functioning of a protocol or an application as if it were running on the network. We will show that ENDE allows accurate estimation of one-way transit times and hence can be used even when the forward and reverse paths are asymmetric between the two hosts. Experimental results are also presented to show that ENDE is fairly accurate in the delay-impacting mode.","Title: An Evaluation of Multiple-Disk I/O Systems\nYear: 1989\n\nAbstract: Alternative ways of configuring an I/O subsystem with multiple disks to improve the I/O performance are considered. Specifically, the author consider disk synchronization, data declustering/disk striping, and a combination of both these approaches. They evaluate many different organizations that have not been considered before. The effects of block size and other parameters of the system are examined. Two different workloads are considered for the evaluation: a file/transaction system workload and a scientific applications workload. Through simulations it is shown that synchronized organizations perform better than other organizations at very low request rates; that there is a tradeoff in the amount of declustering/synchronization to be used in a system; and that systems with higher parallelism in reading a file perform better in a scientific workload.","Title: ARI: Adaptive LLC-memory traffic management\nYear: 2013\n\nAbstract: Decreasing the traffic from the CPU LLC to main memory is a very important issue in modern systems. Recent work focuses on cache misses, overlooking the impact of writebacks on the total memory traffic, energy consumption, IPC, and so forth. Policies that foster a balanced approach, between reducing write traffic to memory and improving miss rates, can increase overall performance and improve energy efficiency and memory system lifetime for NVM memory technology, such as phase-change memory (PCM). We propose Adaptive Replacement and Insertion (ARI), an adaptive approach to last-level CPU cache management, optimizing the two parameters (miss rate and writeback rate) simultaneously. Our specific focus is to reduce writebacks as much as possible while maintaining or improving the miss rate relative to conventional LRU replacement policy. ARI reduces LLC writebacks by 33%, on average, while also decreasing misses by 4.7%, on average. In a typical system, this boosts IPC by 4.9%, on average, while decreasing energy consumption by 8.9%. These results are achieved with minimal hardware overheads.","Title: Statistical techniques for detecting traffic anomalies through packet header data\nYear: 2008\n\nAbstract: This paper proposes a traffic anomaly detector, operated in postmortem and in real-time, by passively monitoring packet headers of traffic. The frequent attacks on network infrastructure, using various forms of denial of service attacks, have led to an increased need for developing techniques for analyzing network traffic. If efficient analysis tools were available, it could become possible to detect the attacks, anomalies and to take action to contain the attacks appropriately before they have had time to propagate across the network. In this paper, we suggest a technique for traffic anomaly detection based on analyzing correlation of destination IP addresses in outgoing traffic at an egress router. This address correlation data are transformed using discrete wavelet transform for effective detection of anomalies through statistical analysis. Results from trace-driven evaluation suggest that proposed approach could provide an effective means of detecting anomalies close to the source. We also present a multidimensional indicator using the correlation of port numbers and the number of flows as a means of detecting anomalies.","Title: CATS: Characterizing automation of Twitter spammers\nYear: 2013\n\nAbstract: Twitter, with its rising popularity as a micro-blogging website, has inevitably attracted the attention of spammers. Spammers use myriad of techniques to evade security mechanisms and post spam messages, which are either unwelcome advertisements for the victim or lure victims in to clicking malicious URLs embedded in spam tweets. In this paper, we propose several novel features capable of distinguishing spam accounts from legitimate accounts. The features analyze the behavioral and content entropy, bait-techniques, and profile vectors characterizing spammers, which are then fed into supervised learning algorithms to generate models for our tool, CATS. Using our system on two real-world Twitter data sets, we observe a 96% detection rate with about 0.8% false positive rate beating state of the art detection approach. Our analysis reveals detection of more than 90% of spammers with less than five tweets and about half of the spammers detected with only a single tweet. Our feature computation has low latency and resource requirement making fast detection feasible. Additionally, we cluster the unknown spammers to identify and understand the prevalent spam campaigns on Twitter.","Title: A study of I/O system organizations\nYear: 1992\n\nAbstract: With the increasing processing speeds, it has become important to design powerful and efficient I/O systems. In this paper, we look at several design options in designing an I/O system and study their impact on the performance. Specifically, we use trace driven simulations to study a disk system with a nonvolatile cache. Some of the considered design parameters include the cache block size, the fetch size, the cache size and the disk access policy. We show that decoupling the fetch size and the cache block size results in significant performance improvements. A new write-back policy is presented that is shown to offer significant performance benefits. We show that optimal block size in a two-level memory hierarchy is dependent only on the latency, data rate product of the second level as previously conjectured. We also present results showing the effect of a split access operation of a disk read/write head.","Title: A fast rerouting scheme for OSPF/IS-IS networks\nYear: 2004\n\nAbstract: Most current backbone networks use link-state protocol, OSPF or IS-IS, as their intra-domain routing protocol. Link-state protocols perform global routing table update to route around the failures. It usually takes seconds. As real-time applications like VoIP emerge in recent years, there is a requirement for a fast rerouting mechanism to route around failures before all routers on the network update their routing tables. In addition, fast rerouting is more appropriate than global routing table update when failures are transient. We propose such a fast rerouting extension for link-state protocols. In our approach, when a link fails, the affected traffic is rerouted along a pre-computed rerouting path. In case rerouting cannot be done locally, the local router signals the minimal number of upstream routers to setup the rerouting path for rerouting. We propose algorithms that simplify the rerouting operation and the rerouting path setup. With a simple extension to the current link state protocols, our scheme can route around failures faster and involves minimal number of routers for rerouting.","Title: SCMFS: A File System for Storage Class Memory and its Extensions\nYear: 2013\n\nAbstract: Modern computer systems have been built around the assumption that persistent storage is accessed via a slow, block-based interface. However, emerging nonvolatile memory technologies (sometimes referred to as storage class memory (SCM)), are poised to revolutionize storage systems. The SCM devices can be attached directly to the memory bus and offer fast, fine-grained access to persistent storage. In this article, we propose a new file system---SCMFS, which is specially designed for Storage Class Memory. SCMFS is implemented on the virtual address space and utilizes the existing memory management module of the operating system to help mange the file system space. As a result, we largely simplified the file system operations of SCMFS, which allowed us a better exploration of performance gain from SCM. We have implemented a prototype in Linux and evaluated its performance through multiple benchmarks. The experimental results show that SCMFS outperforms other memory resident file systems, tmpfs, ramfs and ext2 on ramdisk, and achieves about 70% of memory bandwidth for file read/write operations.","Title: Umbrella file system: Storage management across heterogeneous devices\nYear: 2009\n\nAbstract: With the advent of and recent developments in Flash storage, device characteristic diversity is becoming both more prevalent and more distinct. In this article, we describe the Umbrella File System (UmbrellaFS), a stackable file system designed to provide flexibility in matching diversity of file access characteristics to diversity of device characteristics through a user or system administrator specified policy. We present the design and results from a prototype implementation of UmbrellaFS on both Linux 2.4 and 2.6. The results show that UmbrellaFS has little overhead for most file system operations while providing an ability better to utilize the differences in Flash and traditional hard drives. With appropriate use of rules, we have shown improvements of up to 44% in certain situations.","Title: An evaluation of storage systems based on network-attached disks\nYear: 1998\n\nAbstract: The emergence of network-attached disks provides the possibility of transferring data between the storage system and the client directly. This offers new possibilities in building a distributed storage system. We examine different storage organizations based on network-attached disks and compare the performance of these systems to a traditional system. Trace driven simulations are used to measure the average response times of the client requests in two different workloads. The results indicate that it is possible to reduce the workload on the file manager and improve performance in some workloads. However, in other workloads, reduced cache hit ratios may offset the gains of distributing the network processing workload","Title: Performance Evaluation of Storage Systems Based on Network-Attached Disks\nYear: 2000\n\nAbstract: The emergence of network-attached disks provides the possibility of transferring data between the storage system and the client directly. This offers new possibilities in building a distributed storage system. In this paper, we examine different storage organizations based on network-attached disks and compare the performance of these systems to a traditional system. Trace-driven simulations are used to measure the average response times of the client requests in two different workloads. We show that the advantages of distributing the server's network processing load to disks can be offset by losses in cache hits in a system based on network-attached disks. The two workloads we considered highlight this impact. We show that it is possible to offload significant load from the server by utilizing network-attached disks and point out that specific applications may be able to better exploit the features of network-attached disks.","Title: I/O issues for hypercubes\nYear: 1989\n\nAbstract: In this paper, we look at several issues concerning the design of a disk system for a multiprocessor such as a hypercube. We propose a methodology for connecting the I/O processors to such a system for efficient I/O access. An analysis is presented to see the effect of I/O communication on the network of the multiprocessor. We evaluate different disk organizations that can be employed within such a system to see which organization has a better performance. Then we consider the problem of mapping specific data structures such as matrices onto the disks such that the data can be accessed efficiently.","Title: Real-time detection and containment of network attacks using QoS regulation\nYear: 2005\n\nAbstract: In this paper, we present a network measurement mechanism that can detect and mitigate attacks and anomalous traffic in real-time using QoS regulation. The detection method rapidly pursues the dynamics of the network on the basis of correlation properties of the network protocols. By observing the proportion occupied by each traffic protocol and correlating it to that of previous states of traffic, it can be possible to determine whether the current traffic is behaving normally. When abnormalities are detected, our mechanism allows aggregated resource regulation of each protocol's traffic. The trace-driven results show that the rate-based regulation of traffic characterized by protocol classes is a feasible vehicle for mitigating the impact of network attacks on end servers.","Title: A Fault Secure Dictionary Machine\nYear: 1987\n\nAbstract: A fault-secure dictionary machine is presented in this paper. The symmetry of the binary tree architecture for a dictionary machine is exploited to obtain fault-secureness with little overhead. The proposed design utilizes only one extra processor and with some other modifications to the structure of the processors, can detect a single failure of a processor or a link. The proposed design keeps two copies of each record and whenever a record is extracted from the machine the two copies are compared to detect if any fault has occurred. The low overhead is a result of observing the fact that at any given time all the processors of the machine need not be active and these potentially-idle processors are used to do redundant processing to enable detecting single faults.","Title: Adaptive marking for aggregated flows\nYear: 2001\n\nAbstract: The differentiated services architecture is receiving wide attention as a framework for providing different levels of service according to a service profile in the Internet. The current architecture allows aggregation of flows sharing a service profile. This paper looks at the problem of achieving specific QoS goals of individual flows by flexibly managing resources available to an aggregated source. We derive a simple analytic model for the relationship between per-session behavior, aggregate packet marking and packet differentiation within a DiffServ network. The paper presents an adaptive marker based on a TCP performance model within a DiffServ network. The paper shows that an aggregated marker can maintain the state of individual flows at the edge of the network and utilize this state effectively in adaptively marking packets of individual flows to meet their QoS goals","Title: A study of analyzing network traffic as images in real-time\nYear: 2005\n\nAbstract: This paper presents NetViewer, a network measurement approach that can simultaneously detect, identify and visualize attacks and anomalous traffic in real-time by passively monitoring packet headers. We propose to represent samples of network packet header data as frames or images. With such a formulation, a series of samples can be seen as a sequence of frames or video. This enables techniques from image processing and video compression to be applied to the packet header data to reveal interesting properties of traffic. We show that \"scene change analysis\" can reveal sudden changes in traffic behavior or anomalies. We also show that \"motion prediction\" techniques can be employed to understand the patterns of some of the attacks. We show that it may be feasible to represent multiple pieces of data as different colors of an image enabling a uniform treatment of multidimensional packet header data. We compare NetViewer with classical detection theory based Neyman-Pearson test and an IDS tool.","Title: Emulating AQM from end hosts\nYear: 2007\n\nAbstract: In this paper, we show that end-host based congestion prediction is more accurate than previously characterized. However, it may not be possible to entirely eliminate the uncertainties in congestion prediction. To address these uncertainties, we propose Probabilistic Early Response TCP (PERT). PERT emulates the behavior of AQM/ECN, in the congestion response function of end-hosts. We present fluid-flow analysis of PERT/RED and PERT/PI, versions of PERT that emulate router-based RED and PI controllers. Our analysis shows that PERT/RED has better stability behavior than router-based RED. We also present results from ns-2 simulations to show the practical feasibility of PERT. The scheme presented here is general and can be used for emulating other AQM algorithms.","Title: Designing ISP-friendly Peer-to-Peer Networks Using Game-based Control\nYear: 2009\n\nAbstract: The rapid growth of peer-to-peer (P2P) networks in the past few years has\nbrought with it increases in transit cost to Internet Service Providers (ISPs),\nas peers exchange large amounts of traffic across ISP boundaries. This ISP\noblivious behavior has resulted in misalignment of incentives between P2P\nnetworks--that seek to maximize user quality--and ISPs--that would seek to\nminimize costs. Can we design a P2P overlay that accounts for both ISP costs as\nwell as quality of service, and attains a desired tradeoff between the two? We\ndesign a system, which we call MultiTrack, that consists of an overlay of\nmultiple \\emph{mTrackers} whose purpose is to align these goals. mTrackers\nsplit demand from users among different ISP domains while trying to minimize\ntheir individual costs (delay plus transit cost) in their ISP domain. We design\nthe signals in this overlay of mTrackers in such a way that potentially\ncompetitive individual optimization goals are aligned across the mTrackers. The\nmTrackers are also capable of doing admission control in order to ensure that\nusers who are from different ISP domains have a fair chance of being admitted\ninto the system, while keeping costs in check. We prove analytically that our\nsystem is stable and achieves maximum utility with minimum cost. Our design\ndecisions and control algorithms are validated by Matlab and ns-2 simulations.","Title: Detecting algorithmically generated malicious domain names\nYear: 2010\n\nAbstract: Recent Botnets such as Conficker, Kraken and Torpig have used DNS based \"domain fluxing\" for command-and-control, where each Bot queries for existence of a series of domain names and the owner has to register only one such domain name. In this paper, we develop a methodology to detect such \"domain fluxes\" in DNS traffic by looking for patterns inherent to domain names that are generated algorithmically, in contrast to those generated by humans. In particular, we look at distribution of alphanumeric characters as well as bigrams in all domains that are mapped to the same set of IP-addresses. We present and compare the performance of several distance metrics, including KL-distance, Edit distance and Jaccard measure. We train by using a good data set of domains obtained via a crawl of domains mapped to all IPv4 address space and modeling bad data sets based on behaviors seen so far and expected. We also apply our methodology to packet traces collected at a Tier-1 ISP and show we can automatically detect domain fluxing as used by Conficker botnet with minimal false positives.","Title: Modeling Network Traffic As Images\nYear: 2005\n\nAbstract: This paper presents a network measurement approach to represent samples of network packet header data as frames or images. With such a formulation, a series of samples can be seen as a sequence of frames or video. This enables techniques from image processing and video compression to be applied to the analysis of packet header data to reveal interesting properties of traffic. We show that traffic images. can reveal sudden changes in traffic behavior or anomalies. Using a combination of visual modeling and trace-driven simulation, we evaluate how the design factors impact the representation of dynamic network traffic. In particular, we study the impact of sampling rate and retained DCT coefficients on the network traffic data representation.","Title: LTCP: improving the performance of TCP in highspeed networks\nYear: 2006\n\nAbstract: In this paper, we propose Layered TCP (LTCP for short), a set of simple modifications to the congestion window response of TCP to make it more scalable in highspeed networks. LTCP modifies the TCP flow to behave as a collection of virtual flows to achieve more efficient bandwidth probing. The number of virtual flows emulated is determined based on the dynamic network conditions by using the concept of virtual layers, such that the convergence properties and RTT-unfairness behavior is maintained similar to that of TCP. In this paper, we provide the intuition and the design for the LTCP protocol modifications and evaluation results based on ns-2 simulations and Linux implementation. Our results show that LTCP has promising convergence properties, is about an order of magnitude faster than TCP in utilizing high bandwidth links, employs few parameters and retains AIMD characteristics","Title: Effectiveness of caching policies for a Web server.\nYear: 1997\n\nAbstract: We look at a number of policies for managing a file system cache in a Web server. Traces from NCSA World Wide Web (WWW) server are used in this study. We study several caching policies for improving the overall performance of such a server and show that request response time can be improved by some of the replacement policies that take size of the request into account. It is shown that least frequently used (LFU) performs well when cache sizes are small (<16 MB) and SpacexAge replacement policy performs consistently better than LRU. We show that previously proposed cache sharing policies improve performance significantly in clustered Web servers","Title: Real-time communication scheduling in a multicomputer video server\nYear: 1999\n\nAbstract: In this paper, we address the problem of scheduling communication over the interconnection network of a distributed-memory multicomputer video server. We show that this problem is closely related to the problem of data distribution and movie scheduling in such a system. A solution is proposed in this paper that addresses these three issues at once. The movies are distri- buted evenly over all nodes of the multicomputer. The proposed solution minimizes the contention for links over the switch. The proposed solution makes movie scheduling very simpleif the first block of the movie is scheduled, the rest of the movie is automatically scheduled. Moreover, if the first block of the movie stream is scheduled without network contention, the proposed solution guarantees that there will be no network contention during the entire duration of playback of that movie. We show that the proposed approach to communication scheduling is optimal in utilizing the network resources. Extensive simulation results are presented to show the effectiveness of the proposed approach. 1999 Academic Press","Title: Detecting algorithmically generated domain-flux attacks with DNS traffic analysis\nYear: 2012\n\nAbstract: Recent botnets such as Conficker, Kraken, and Torpig have used DNS-based \"domain fluxing\" for command-and-control, where each Bot queries for existence of a series of domain names and the owner has to register only one such domain name. In this paper, we develop a methodology to detect such \"domain fluxes\" in DNS traffic by looking for patterns inherent to domain names that are generated algorithmically, in contrast to those generated by humans. In particular, we look at distribution of alphanumeric characters as well as bigrams in all domains that are mapped to the same set of IP addresses. We present and compare the performance of several distance metrics, including K-L distance, Edit distance, and Jaccard measure. We train by using a good dataset of domains obtained via a crawl of domains mapped to all IPv4 address space and modeling bad datasets based on behaviors seen so far and expected. We also apply our methodology to packet traces collected at a Tier-1 ISP and show we can automatically detect domain fluxing as used by Conficker botnet with minimal false positives, in addition to discovering a new botnet within the ISP trace. We also analyze a campus DNS trace to detect another unknown botnet exhibiting advanced domain-name generation technique.","Title: Fault Tolerance Techniques for Systolic Arrays\nYear: 1987\n\nAbstract: First Page of the Article","Title: System support for providing integrated services from networked multimedia storage servers\nYear: 2001\n\nAbstract: In this paper, we describe our experience in building an integrated multimedia storage system, Prism. Our current Linux-based implementation of Prism provides three levels of service: deadline guarantees for periodic applications, best-effort better response times for interactive applications and starvation-free throughput guarantees for aperiodic applications. Prism separates resource allocation from resource scheduling. Resource allocation is controlled across the service classes by a system-wide policy and service class specific admission controllers. Resource scheduling is done at the resources. This separation allows Prism to be deployed even when the storage system is separated on a network from the file system.We report on the important aspects of Prism architecture, innovations required to build Prism on top of Linux and lessons learned during the implementation and testing of Prism. We present experimental results to show that Prism achieves its goals in supporting multiple service classes within a single system. We compare Prism against standard Linux operating system to show the impact of our approach.","Title: Realizing throughput guarantees in a differentiated services network\nYear: 1999\n\nAbstract: This paper discusses techniques for achieving desired throughput guarantees in the Internet that supports a differentiated services framework. The diff-serv framework proposes the use of different drop precedences to achieve service guarantees over the Internet. However, it has been observed that the drop precedences by themselves cannot achieve the desired target rates because of the strong interaction of the transport protocol with packet drops in the network. This paper proposes and evaluates a number of techniques to better achieve the throughput guarantees in such networks. The proposed techniques consider: modifying the transport protocol at the sender; modifying the marking strategies at the marker; and modifying the dropping policies at the router. It is shown that these techniques improve the likelihood of achieving the desired throughput guarantees and also improve the service differentiation","Title: Constructing disjoint paths for failure recovery and multipath routing\nYear: 2012\n\nAbstract: Applications such as Voice over IP and video streaming require continuous network service, requiring fast failure recovery mechanisms. Proactive failure recovery schemes have been recently proposed to improve network performance during the failure transients. These proactive failure recovery schemes need extra infrastructural support in the form of routing table entries, extra addresses etc. In this paper, we study if the extra infrastructure support can be exploited to build disjoint paths in those frameworks, while keeping the lengths of the recovery paths close to those of the primary paths. Our evaluations show that it is possible to extend the proactive failure recovery schemes to provide support for nearly-disjoint paths which can be employed in multipath routing for load balancing and QoS.","Title: Fault tolerance of adaptive routing algorithms in multicomputers\nYear: 1992\n\nAbstract: An evaluation of the effectiveness of adaptive routing techniques in tolerating failures is presented. It is shown that adaptive routing techniques yield gracefully degradable systems for the workloads considered. For medium to large communication granularity and the workloads considered in this study, if adaptive-routing is used, it is shown that the problem completion time does not increase drastically due to failures. When node failures were considered, it was observed that the mismatch of problem communication structure with the physical communication structure did not result in significant loss of performance. Since adaptive routing techniques are warranted for performance reasons, it is argued that making use of this adaptive routing hardware to tolerate failures is a favorable option.","Title: Design, Analysis, and Simulation of I/O Architectures for Hypercube\nYear: 1990\n\nAbstract: Several issues concerning the design of an I/O (input/output) system for a multiprocessor such as a hypercube are examined. A methodology is proposed for connecting the I/O processors to such a system for efficient I/O access. The effect of I/O communication onthe multiprocessor network is analyzed. Different disk organizations that can be employed within such a system are evaluated to see which organization has a better performance. It is observed that parallelism in serving an I/O request plays a dominant role in the scientific workload. The problem of mapping specific data structures such as matrices onto the disks so that the data can be accessed efficiently is considered.","Title: An approach to virtual allocation in storage systems\nYear: 2006\n\nAbstract: This article presents virtual allocation, a scheme for flexible storage allocation. Virtual allocation separates storage allocation from the file system. It employs an allocate-on-write strategy which lets applications fit into the actual usage of storage space, without regard to the configured file system size. This improves flexibility by allowing storage space to be shared across different file systems. This article presents the design of virtual allocation and its evaluation through benchmarks. To illustrate our approach, we implemented a prototype system on PCs running Linux. We present the results from the prototype implementation and its evaluation.","Title: SACRIO: an active buffer management scheme for differentiated service networks\nYear: 2001\n\nAbstract: In this paper, we propose an active resource management approach for D ifferentiated Services networks. The proposed approach, SACRIO, employs caching and localized packet remarking within a router. It is shown that SACRIO is simple and can be implemented transparently within the diff-serv architecture. It is shown that the packet handling cost remains O(1) with SACRIO. SACRIO is shown to be quite effective and scalable through both ns-based simulations and real-world trace-driven simulations.","Title: Evaluation of Data and Request Distribution Policies in Clustered Servers\nYear: 1999\n\nAbstract: The popularity of the WWW has prompted the design of scalable clustered servers. Data and request distribution policies greatly\n impact performance in such servers. In this paper, request response time is used as a measure in evaluating these policies.\n Our results indicate that policies favoring locality seem to perform better than policies favoring load balance. We propose\n and evaluate a dynamic client-based request distribution policy.","Title: Performance of Quantized Congestion Notification in TCP Incast Scenarios of Data Centers\nYear: 2010\n\nAbstract: This paper analyzes the performance of Ethernet layer congestion control mechanism Quantized Congestion Notification (QCN) during data access from clustered servers in data centers. We analyze the reasons why QCN does not perform adequately in these situations and propose several modifications to the protocol to improve its performance in these scenarios. We trace the causes of QCN performance degradation to flow rate variability, and show that adaptive sampling at the switch and adaptive self-increase of flow rates at the rate limiter improve performance in a TCP In cast setup significantly. We compare the performance of QCN against TCP modifications in a heterogeneous environment, and show that modifications to QCN yield better performance.","Title: A Large-Scale Empirical Study of Conficker\nYear: 2012\n\nAbstract: Conficker is the most recent widespread, well-known worm/bot. According to several reports, it has infected about 7 million to 15 million hosts and the victims are still increasing even now. In this paper, we analyze Conficker infections at a large scale, about 25 million victims, and study various interesting aspects about this state-of-the-art malware. By analyzing Conficker, we intend to understand current and new trends in malware propagation, which could be very helpful in predicting future malware trends and providing insights for future malware defense. We observe that Conficker has some very different victim distribution patterns compared to many previous generation worms/botnets, suggesting that new malware spreading models and defense strategies are likely needed. We measure the potential power of Conficker to estimate its effects on the networks/hosts when it performs malicious operations. Furthermore, we intend to determine how well a reputation-based blacklisting approach can perform when faced with new malware threats such as Conficker. We cross-check several DNS blacklists and IP/AS reputation data from Dshield and FIRE and our evaluation shows that unlike a previous study which shows that a blacklist-based approach can detect most bots, these reputation-based approaches did relatively poorly for Conficker. This raises a question of how we can improve and complement existing reputation-based techniques to prepare for future malware defense? Based on this, we look into some insights for defenders. We show that neighborhood watch is a surprisingly effective approach in the case of Conficker. This suggests that security alert sharing/correlation (particularly among neighborhood networks) could be a promising approach and play a more important role for future malware defense.","Title: TCP-DCR: A Novel Protocol for Tolerating Wireless Channel Errors\nYear: 2005\n\nAbstract: This paper presents TCP-DCR, a set of simple modifications to the TCP protocol to improve its robustness to channel errors in wireless networks. TCP-DCR is based on the simple idea of allowing the link-level mechanism to recover the packets lost, due to channel errors, thereby limiting the response of the transport protocol to mostly congestion losses. This is done by delaying the triggering of congestion response algorithms for a small bounded period of time \\tau to allow the link-level retransmissions to recover the loss due to channel errors. If at the end of the delay \\tau the packet is not recovered, then it is treated as a packet lost due to congestion. We analyze TCP-DCR to show that the delay in congestion response does not impact the fairness towards the native implementations of TCP that respond to congestion immediately after receiving three dupacks. We evaluate TCP-DCR through simulations to show that it offers significantly better performance when channel errors contribute more towards packet losses in the network with no or minimal impact on the performance when congestion is the primary cause for packet loss. We also present an analysis to show that protocol evaluation in the wireless networks is significantly influenced by the number of flows in the network.","Title: Impact of bandwidth-delay product and non-responsive flows on the performance of queue management schemes\nYear: 2004\n\nAbstract: In this paper, we study the impact-of bandwidth-delay products and non-responsive flows on queue management schemes. Our focus is to understand the aggregate performance of various classes of traffic under different queue management schemes. Our work is motivated by the expected trends of increasing link capacities and increasing amounts of non-responsive traffic. In this paper, the impact on the performance of RED, RED with ECN enabled and droptail routers are investigated. Our study considers the aggregate bandwidth of different classes of traffic and the delays observed at the router.","Title: Multihoming route control among a group of multihomed stub networks\nYear: 2007\n\nAbstract: In recent years, commercial ''multihoming route control'' devices are used by multihomed stub networks to optimize the routing of their Internet traffic. Previous studies have shown that ''multihoming route control'' can improve the Internet communication performance of the multihomed stub networks significantly. In this work, we study ''multihoming route control'' among a group of multihomed stub networks that may belong to an organization and exchange large amount of data regularly. We show that greedy ''multihoming routing control'' schemes could lead to oscillations. We propose a user-optimal routing based ''multihoming routing control'' scheme that improves routing performance without causing oscillations. The proposed scheme is simpler to implement than optimal routing based approaches. We show through simulations that the proposed scheme achieves performance close to optimal routing and works well in various network conditions.","Title: A method for estimating non-responsive traffic at a router\nYear: 2002\n\nAbstract: In this paper, we propose a scheme for estimating the proportion of the incoming traffic that is not responding to congestion at a router. The idea of the proposed scheme is that if the observed queue length and packet drop probability do not match with the predicted results from the TCP model, then the error must come from the non-responsive traffic; it can then be used for estimating non-responsive traffic. The proposed scheme utilizes queue length history, packet drop history, expected TCP and queue dynamics to estimate the proportion. We show that the proposed scheme is effective over a wide range of traffic scenarios through simulations.","Title: SPIRAL: a client-transparent third-party transfer scheme for network attached disks\nYear: 2003\n\nAbstract: Third-party transfer is a data transfer mechanism where the party initiating the transfer is neither the source nor the sink for the data. In this paper, we present a scheme for supporting third-party transfers on storage systems with network-attached disks (NADs), called SPIRAL. SPIRAL allows NADs to send data directly to clients without going through the server It is transparent to clients and relies only on the linear block interface of the current disks (or the future iSCSI disks). SPIRAL requires no porting of file system or application-level functionality to NADs and requires only simple modifications to server applications. To illustrate our approach, we implemented a prototype system on PCs running Linux. We present experimental results for NFS and HTTP on the prototype to demonstrate the effectiveness of our approach.","Title: Modeling TCP behavior in a differentiated services network\nYear: 2001\n\nAbstract: Abstract—The differentiated services architecture has been proposed for providing different levels of services and has recently received wide attention. A packet in a diff-serv domain is classified into a class of service according to its contract profile and treated differently by its class. While many,studies have addressed issues on the diff-serv architecture (e.g., dropper, marker, classifier and shaper), there have been few attempts to analytically understand a flow’s behavior in a diff-serv network. In this paper, we propose simple models of TCP behavior in a diff-serv network. Our models quantitatively characterize TCP throughput as functions of the contract rate, the packet-drop rate and the round-trip time in either two-drop precedence or threedrop precedence network. We also extend our models to aggregated flows. The models are validated through a number of simulations. Index Terms—AF PHB, differentiated service, TCP modeling.","Title: NetViewer: a network traffic visualization and analysis tool\nYear: 2005\n\nAbstract: The frequent and large-scale network attacks have led to an increased need for developing techniques for analyzing network traffic. If efficient analysis tools were available, it could become possible to detect the attacks, anomalies and to appropriately take action to contain the attacks before they have had time to propagate across the network. This paper describes NetViewer, a network monitoring tool that can simultaneously detect, identify and visualize attacks and anomalous traffic in real-time by passively monitoring packet headers. NetViewer represents the traffic data as images, enabling the application of image/video processing techniques for the analysis of network traffic. NetViewer is released free to the general public. By employing a freely available visualization tool, the users of NetViewer can comprehend the characteristics of the network traffic observed in the aggregate. NetViewer can be employed to detect and identify network anomalies such as DoS/DDoS attacks, worms and flash crowds. NetViewer can also provide information on traffic distributions over IP address/port number domains, utilization of link capacity and effectiveness of Quality of Service policies.","Title: A method for estimating the proportion of nonresponsive traffic at a router\nYear: 2004\n\nAbstract: In this paper, a scheme for estimating the proportion of the incoming traffic that is not responsive to congestion at a router is presented. The idea of the proposed scheme is that if the observed queue length and packet drop probability do not match the predictions from a model of responsive (TCP) traffic, then the error must come from nonresponsive traffic; it can then be used for estimating the proportion of nonresponsive traffic. The proposed scheme is based on the queue length history, packet drop history, and expected TCP and queue dynamics. The effectiveness of the proposed scheme over a wide range of traffic scenarios is corroborated using ns-2-based simulations. Potential applications of the proposed algorithms in traffic engineering and control are discussed.","Title: SCMFS: a file system for storage class memory\nYear: 2011\n\nAbstract: This paper considers the problem of how to implement a file system on Storage Class Memory (SCM), that is directly connected to the memory bus, byte addressable and is also non-volatile. In this paper, we propose a new file system, called SCMFS, which is implemented on the virtual address space. In SCMFS, we utilize the existing memory management module in the operating system to do the block management and keep the space always contiguous for each file. The simplicity of SCMFS not only makes it easy to implement, but also improves the performance. We have implemented a prototype in Linux and evaluated its performance through multiple benchmarks.","Title: Detecting Traffic Anomalies through Aggregate Analysis of Packet Header Data\nYear: 2004\n\nAbstract: If efficient network analysis tools were available, it could become possible to detect the attacks, anomalies and to appropriately take action to contain the attacks. In this paper, we suggest a technique for traffic anomaly detection based on analyzing correlation of destination IP addresses in outgoing traffic at an egress router. This address correlation data are transformed through discrete wavelet transform for effective detection of anomalies through statistical analysis. Our techniques can be employed for postmortem and real-time analysis of outgoing network traffic at a campus edge. Results from trace-driven evaluation suggest that proposed approach could provide an effective means of detecting anomalies close to the network. We also present data analyzing the correlation of port numbers as a means of detecting anomalies.","Title: Identifying Long-Term High-Bandwidth Flows at a Router\nYear: 2001\n\nAbstract: In this paper, we propose a novel approach for identifying long-term high-rate flows at a router. Identifying the long-term high-rate flows allows a router to regulate flows intelligently in times of congestion. We employ a limited amount of state to record the arrivals of packets of individual flows. The state is managed as an LRU cache. The LRU cache self converges to hold only the flows that are high rate. The size of the cache need not depend on the number of flows at the router. We analyze a number of internet packet traces to show the effectiveness of this LRUbased cache. It is shown that this scheme is highly scalable since a few flows contribute a significant fraction of the traffic at a router.","Title: Data organization in a hybrid storage system\nYear: 2012\n\nAbstract: Flash memory based Solid State Devices (SSDs) have become commercially viable and hybrid storage systems incorporating SSDs and magnetic disk drives are being built. There are a number of ways the data can be organized in such hybrid systems. SSDs can be organized at a higher layer as a cache in a transparent hierarchy. SSDs can be organized as peers to magnetic disks in a heterogeneous storage system. There are pros and cons for organizing data in both approaches. This paper argues that a storage system with SSDs and disk drives as peers may deliver the best performance, with maximum flexibility.","Title: Multipath Routing For Reducing Network Energy\nYear: 2012\n\nAbstract: Reducing energy consumption in networks has been recently receiving attention. Networks are typically designed to support peak loads. Energy saving is potentially achieved by appropriately provisioning network capacity, by shutting down unnecessary network elements while the capacity of remaining network elements still meets traffic demand. Traditionally, topology control techniques are employed to reduce the network to an appropriate size. In this paper, we propose techniques to construct an appropriate topology that takes the traffic demands into account. We also study how much multipath routing can contribute to energy saving compared to single/shortest path routing. We show that the proposed techniques reduce energy consumption by an additional 17% over previous schemes.","Title: Gracefully degradable disk arrays\nYear: 1991\n\nAbstract: The problem of designing fault-tolerant disk arrays that are not susceptible to 100% load increases on the functional disks when one of the disks in the system fails is addressed. A technique that combines the advantages of parity schemes and the traditional dual copy methods and offers a wide variety of options in providing fault-tolerance is proposed. A theoretical framework for solving the problem is presented and a number of constructive techniques are proposed. By utilizing the same amount of hardware as the earlier methods but with a better data organization and a different reconstruction technique, the system yields better performance during a failure. Merging two parity groups as a reconfiguration strategy is shown to have a number of benefits, such as reduced hardware overhead and improved reliability. A combination of block designs and the proposed reconfiguration strategy results in a highly reliable disk array with the same or less overhead as the earlier approaches and better performance during a failure.<>","Title: I/O issues in a multimedia system\nYear: 1994\n\nAbstract: In future computer system design, I/O systems will have to support continuous media such as video and audio, whose system demands are different from those of data such as text. Multimedia computing requires us to focus on designing I/O systems that can handle real-time demands. Video- and audio-stream playback and teleconferencing are real-time applications with different I/O demands. We primarily consider playback applications which require guaranteed real-time I/O throughput. In a multimedia server, different service phases of a real-time request are disk, small computer systems interface (SCSI) bus, and processor scheduling. Additional service might be needed if the request must be satisfied across a local area network. We restrict ourselves to the support provided at the server, with special emphasis on two service phases: disk scheduling and SCSI bus contention. When requests have to be satisfied within deadlines, traditional real-time systems use scheduling algorithms such as earliest deadline first (EDF) and least slack time first. However, EDF makes the assumption that disks are preemptable, and the seek-time overheads of its strict real-time scheduling result in poor disk utilization. We can provide the constant data rate necessary for real-time requests in various ways that require trade-offs. We analyze how trade-offs that involve buffer space affect the performance of scheduling policies. We also show that deferred deadlines, which increase buffer requirements, improve system performance significantly.<>","Title: Failure evaluation of disk array organizations\nYear: 1993\n\nAbstract: The authors present an evaluation of some of the disk array organizations proposed in the literature. They evaluate three alternatives for sparing, hot sparing, distributed sparing, and parity sparing, and two options for data layout, regular RAID5 and block designs, and systems based on combinations of these data layout and sparing alternatives. The performance of these organizations is evaluated with different reconstruction strategies. It is shown that parity sparing and distributed sparing have better performance and shorter reconstruction times than hot sparing. It is shown that both block designs as a data layout policy and distributed sparing as a sparing policy reduce the reconstruction time after a failure. The impact of reconstruction strategies is studied, and it is shown that, at higher workloads, choice of reconstruction strategy has a significant impact on the performance of the systems","Title: Integrated QOS management for disk I/O\nYear: 1999\n\nAbstract: In this paper, we address the problem of providing different levels of performance guarantees or quality-of-service (QOS) for disk I/O. We classify disk requests into three categories based on the provided level of service. We propose an integrated scheme that provides different levels of performance guarantees in a single system. The proposed method uses admission control and bandwidth allocation to isolate the different categories of requests and appropriate disk scheduling to achieve the desired performance goals. It is shown that the proposed method achieves the performance goals of individual requests while allowing seek optimizations at the disk. In particular, it is shown that the proposed scheme for integrated QOS management at disk provides significantly better performance than SCAN.","Title: Multi Path PERT\nYear: 2013\n\nAbstract: This paper presents a new multipath delay based algorithm, MPPERT (Multipath Probabilistic Early response TCP), which provides high throughput and efficient load balancing. In all-PERT environment, MPPERT suffers no packet loss and maintains much smaller queue sizes compared to existing MPTCP, making it suitable for real time data transfer. MPPERT is suitable for incremental deployment in a heterogeneous environment. PERT, being a delay based TCP protocol, has continuous information about the state of the bottleneck queue along its path. This information is valuable in enabling MPPERT to detect subflows sharing a common bottleneck and obtain a smaller set of disjoint subflows. This information can even be used to switch from coupled (a set of subflows having interdependent increase/decrease of congestion windows) to uncoupled (independent increase/decrease of congestion windows) subflows, yielding higher throughput when best single-path TCP constraint is relaxed. The ns-2 simulations support MPPERT as a highly competitive multipath approach, suitable for real time data transfer, which is capable of offering higher throughput and improved reliability.","Title: Wavelet Methods for the Detection of Anomalies and their Application to Network Traffic Analysis\nYear: 2006\n\nAbstract: Here we develop an integrated tool for the online detection of network anomalies. We consider statistical change point detection algorithms, for both local changes in the variance and for the detection of jumps, and propose modified versions of these algorithms based on moving window techniques. We investigate performances on simulated data and on network traffic data with several superimposed attacks. All detection methods are based on wavelet packet transforms. Copyright (C) 2006 John Wiley & Sons, Ltd.","Title: Dynamic Memory Pressure Aware Ballooning\nYear: 2015\n\nAbstract: Hardware virtualization is a major component of large scale server and data center deployments due to their facilitation of server consolidation and scalability. Virtualization, however, comes at a high cost in terms of system main memory utilization. Current virtual machine (VM) memory management solutions impose a high performance penalty and are oblivious to the operating regime of the system. Therefore, there is a great need for low-impact VM memory management techniques which are aware of and reactive to current system state, to drive down the overheads of virtualization. We observe that the host machine operates under different memory pressure regimes, as the memory demand from guest VMs changes dynamically at runtime. Adapting to this runtime system state is critical to reduce the performance cost of VM memory management. In this paper, we propose a novel dynamic memory management policy called Memory Pressure Aware (MPA) ballooning. MPA ballooning dynamically allocates memory resources to each VM based on the current memory pressure regime. Moreover, MPA ballooning proactively reacts and adapts to sudden changes in memory demand from guest VMs. MPA ballooning requires neither additional hardware support, nor incurs extra minor page faults in its memory pressure estimation. We show that MPA ballooning provides an 13.2% geomean speed-up versus the current ballooning techniques across a set of application mixes running in guest VMs; often yielding performance nearly identical to that of a non-memory constrained system.","Title: Shared Last-Level Caches and The Case for Longer Timeslices\nYear: 2015\n\nAbstract: Memory performance is important in modern systems. Contention at various levels in memory hierarchy can lead to significant application performance degradation due to interference. Further, modern, large, last-level caches (LLC) have fill times greater than the OS scheduling window. When several threads are running concurrently and timesharing the CPU cores, they may never be able to load their working sets into the cache before being rescheduled, thus permanently stuck in the \"cold-start\" regime. We show that by increasing the system scheduling timeslice length it is possible to amortize the cache cold-start penalty due to the multitasking and improve application performance by 10--15%.","Title: Does RAID Improve Lifetime of SSD Arrays?\nYear: 2016\n\nAbstract: Parity protection at the system level is typically employed to compose reliable storage systems. However, careful consideration is required when SSD-based systems employ parity protection. First, additional writes are required for parity updates. Second, parity consumes space on the device, which results in write amplification from less efficient garbage collection at higher space utilization.\n\nThis article analyzes the effectiveness of SSD-based RAID and discusses the potential benefits and drawbacks in terms of reliability. A Markov model is presented to estimate the lifetime of SSD-based RAID systems in different environments. In a small array, our results show that parity protection provides benefit only with considerably low space utilizations and low data access rates. However, in a large system, RAID improves data lifetime even when we take write amplification into account.","Title: Path confidence based lookahead prefetching.\nYear: 2016\n\nAbstract: Designing prefetchers to maximize system performance often requires a delicate balance between coverage and accuracy. Achieving both high coverage and accuracy is particularly challenging in workloads with complex address patterns, which may require large amounts of history to accurately predict future addresses. This paper describes the Signature Path Prefetcher (SPP), which offers effective solutions for three classic challenges in prefetcher design. First, SPP uses a compressed history based scheme that accurately predicts complex address patterns. Second, unlike other history based algorithms, which miss out on many prefetching opportunities when address patterns make a transition between physical pages, SPP tracks complex patterns across physical page boundaries and continues prefetching as soon as they move to new pages. Finally, SPP uses the confidence it has in its predictions to adaptively throttle itself on a per-prefetch stream basis. In our analysis, we find that SPP improves performance by 27.2% over a no-prefetching baseline, and outperforms the state-of-the-art Best Offset prefetcher by 6.4%. SPP does this with minimal overhead, operating strictly in the physical address space, and without requiring any additional processor core state, such as the PC.","Title: Towards Improving Data Validity of Cyber-Physical Systems through Path Redundancy.\nYear: 2017\n\nAbstract: Cyber-physical systems have shown to be susceptible to cyber-attacks. Incidents such as Stuxnet Attack and Ukraine power outage have shown that attackers are capable of penetrating into industrial control systems, compromising PLCs, and sending false commands to physical devices while reporting normal sensing values. Therefore, one of the critical needs of CPS is to ensure the validity of the sensor values. In this paper, we explore path diversity in SCADA networks and develop Path Redundancy to improve data validity. The proposed solution is shown to be able to effectively prevent data integrity attacks and detect false command attacks from a single compromised path or PLC. We provide detailed analysis on solution design and implement an application of the technique in building automation networks. Our cost-efficient and easy-to-deploy solution improves the resilience of SCADA networks.","Title: Adaptive policies for balancing performance and lifetime of mixed SSD arrays through workload sampling\nYear: 2016\n\nAbstract: Solid-state drives (SSDs) have become promising storage components to serve large I/O demands in modern storage systems. Enterprise class (high-end) SSDs are faster and more resilient than client class (low-end) SSDs but they are expensive to be deployed in large scale storage systems. It is an attractive and practical alternative to exploit the high-end SSDs as a cache and low-end SSDs as main storage. This paper explores how to optimize a mixed SSD array in terms of performance and lifetime. This paper shows that simple integration of different classes of SSDs in traditional caching policies results in poor reliability. This paper also reveals that caching policies with static workload distribution are not always efficient. In this paper, we propose a sampling based adaptive approach that achieves fair workload distribution across the cache and the storage. The proposed algorithm enables fine-grained control of the workload distribution which minimizes latency over lifetime of mixed SSD arrays. We show that our adaptive algorithm is very effective in improving the latency over lifetime metric, on an average, by up to 2.36 times over LRU, across a number of workloads.","Title: Speculative paging for future NVM storage.\nYear: 2017\n\nAbstract: The quest for greater performance and efficiency has driven modern cloud applications towards \"in-memory\" implementations, such as memcached and Apache Spark. Looking forward, however, the costs of DRAM, due to its low area density and high energy consumption, may make this trend unsustainable. Traditionally, OS paging system mechanisms were intended to bridge the gap between expensive, under-provisioned DRAM and inexpensive, dense storage, however, in the past twenty years the latency of storage, relative to DRAM became too great to overcome without significant performance impact. Recent NVM storage devices, such as Intel Optane drives and aggressive, 3D flash SSDs, may dramatically change the picture for OS paging. These newdrives are expected to provide much lower latency compared to the existing flash-based SSDs or traditional HDDs. Unfortunately, even these future NVM drives are still much too slow to replace DRAM, since the access latency of fast NVM storage is expected on the order of tens of microseconds, and they often require block-level access. Unlike traditional HDDs, for which the baseline OS paging policies are designed, these new SSDs place no penalty for \"random\" access and their access latency promises to be significantly less than traditional SSDs, thus arguing for a rearchitecting of the OS paging system. In this paper, we propose SPAN (Speculative PAging for future NVM storage), a software-only, OS swap-based, page management and prefetching scheme designed for emerging NVM storage. Unlike the baseline OS swapping mechanism, which is highly optimized for traditional spinning disks, SPAN leverages the inherent parallelism of NVM devices to proactivley fetch a set of pages from NVM storage to the small and fast main DRAM. In doing so, SPAN yields a speedup of similar to 18% versus swapping into the NVM with the baseline OS (similar to 50% of the performance lost by the baseline OS versus placing the entire working set in DRAM memory). The proposed technique thus enables the utilization of such hybrid systems for memory-hungry applications, lowering the memory cost while keeping the performance comparable to the DRAM-only system.","Title: Securing cyber-physical systems with adaptive commensurate response.\nYear: 2017\n\nAbstract: Cyber-Physical Systems (CPSs) are ubiquitous in power systems, transport and medical systems, and critical infrastructures. Current protocols and applications allow significant changes to a system to take place within a short time or small network footprint, which can be exploited by attackers to cause a great impact on the physical systems. This paper proposes adaptive Commensurate Response (CR) to narrow down the asymmetry between the cost of attacks and their impact through enforcing command footprints to be commensurate with their impact on the system. Such impact is measured by the change of the setpoint (changedriven CR) or the distance between the operating state and the critical state (criticality-driven CR). Change-driven CR is effective for setpoint attacks with big setpoint changes, and criticality-driven CR can be used to combat both setpoint attacks and actuation attacks. Our case study on automobile cruise control demonstrates that CR can effectively improve the system resilience and attack survivability while satisfying QoS requirements.","Title: IoTAegis: A Scalable Framework to Secure the Internet of Things\nYear: 2018\n\nAbstract: The infamous Mirai attack which hijacked nearly half a million Internet connected devices demonstrated the widespread security vulnerabilities of the Internet-of-Things (IoT). This study employs a set of active and passive observation methods to discover the security vulnerabilities of IoT devices within a university campus. We show that (a) the number of non-compute devices dominates the number of compute devices with open ports in a campus network; (b) 58.9% or more devices do not keep up-to-date firmware and 51.3% or more do not have a user defined password; and (c) the number of devices together with the diversity of device ages and vendors make the protection of IoT devices a difficult problem. We further develop IoTAegis framework which offers device-level protection to automatically manage device configurations and security updates. Our solution is shown to be effective, scalable, lightweight, and deployable in different forms and network types.","Title: KVRAID - high performance, write efficient, update friendly erasure coding scheme for KV-SSDs.\nYear: 2021\n\nAbstract: Key-value (KV) stores have been widely deployed in a variety of scale-out enterprise applications such as online retail, big data analytics, social networks, etc. Key-Value SSDs (KVSSDs) provide a key-value interface directly from the device aiming at lowering software overhead and reducing I/O amplification for such applications. In this paper, we present KVRAID, a high performance, write efficient erasure coding management scheme on emerging key-value SSDs. The core innovation of KVRAID is to use logical to physical key conversion to efficiently pack similar size KV objects and dynamically manage the membership of erasure coding groups. Such design enables packing multiple user objects to a single physical object to reduce the object amplification compared to prior works. By applying out-of-place update technique, KVRAID can significantly reduce the I/O amplification compared to the state-of-art designs. Our experiments show that KVRAID outperforms state-of-art software KV-store with block RAID by 28x in terms of insert throughput and reduces CPU utilization, tail latency and write amplification significantly. Compared to state-of-art KV devices erasure coding management, KVRAID reduces object amplification by ~ 2.6x compared to StripeFinder and reduces I/O amplification by ~ 9.6x when compared to KVMD and StripeFinder for update intensive workloads.","Title: Connect Id: Identifying Devices Before They Connect to the Network\nYear: 2021\n\nAbstract: Increasingly, network smart devices are being deployed and operated over the network. The Mirai attack has shown that the vulnerabilities of these devices can be exploited to stage large-scale attacks. In this paper, we explore the potential for identifying a device before it connects to the network such that appropriate security postures can be taken based on the known characteristics of the devi...","Title: Protection against Counterfeiting Attacks in 3D Printing by Streaming Signature-embedded Manufacturing Process Instructions\nYear: 2021\n\nAbstract: ABSTRACTThe emerging Manufacturing-as-a-Service (MaaS) paradigm democratizes manufacturing by connecting people and businesses with manufacturing requests to those with manufacturing resources, via a digital thread. However, the digital thread can be vulnerable to attacks such as counterfeiting, IP theft and sabotage. An approach based on embedding custom anti-counterfeiting signatures in the design files, on-the-fly, is presented to protect against counterfeiting attacks. An experimental study of the effect of geometric and dimensional variations of the signatures in printed components, as well as the effect of the variations on the read rates of the codes are reported. We conduct experiments using a polyjet printer with veroclear material to study the positional and dimensional variations introduced in the embedded signature and the resulting effect on the readability of the design. The results suggest that signatures with spherical units of size φ0.25 mm embedded within 20 mm cubic components can be printed with location precision of ~5% (about 0.1 mm), and dimensional deviation of the order of 0.01 mm. A statistical model is developed to show that these variations pose minimum interference on the readability of the signatures. Embedding of randomized signatures offers security by serving as anti-counterfeit marks in the final part, and makes it harder to reverse engineer and produce counterfeited parts."],"string":"[\n \"Title: Providing QoS guarantees for disk I/O\\nYear: 2000\\n\\nAbstract: In this paper, we address the problem of providing different levels of performance guarantees or quality ofservice for disk I/O. We classify disk requests into threecategories based on the provided level of service. We propose an integrated scheme that provides different levels ofperformance guarantees in a single system. We propose andevaluate a mechanism for providing deterministic servicefor variable-bit-rate streams at the disk. We will show that,through proper admission control and bandwidth allocation,requests in different categories can be ensured of performance guarantees without getting impacted by requests inother categories. We evaluate the impact of scheduling policy decisions on the provided service. We also quantify theimprovements in stream throughput possible by using statistical guarantees instead of deterministic guarantees in thecontext of the proposed approach.\",\n \"Title: Exploiting Concurrency to Improve Latency and throughput in a Hybrid Storage System\\nYear: 2010\\n\\nAbstract: This paper considers the problem of how to improve the performance of hybrid storage system employing solid state disks and hard disk drives. We utilize both initial block allocation as well as migration to reach “Wardrop equilibrium”, in which the response times of different devices equalize. We show that such a policy allows adaptive load balancing across devices of different performance. We also show that such a policy exploits parallelism in the storage system effectively to improve throughput and latency simultaneously. We implemented a prototype in Linux and evaluated it in multiple workloads and multiple configurations. The results show that the proposed approach improved both the latency of requests and the throughput significantly, and it adapted to different configurations of the system under different workloads.\",\n \"Title: Reaching approximate Wardrop equilibrium at reduced costs of link state updates\\nYear: 2011\\n\\nAbstract: Several routing protocols have been proposed to take advantage of the dynamic metrics on links such as link delays, queueing lengths, and available link bandwidths.Wardrop routing, QOS routing are few such example routing protocols. Even when these protocols can be shown to offer convergence properties without oscillations, the protocols have not been widely adopted for a number of reasons. The expected cost of keeping the link metrics updated at various nodes in the network being one of them. In this paper, we study the problem of reducing the cost of propagating dynamic link metrics while keeping the quality of paths within a margin of error. We propose a simple technique of threshold-based metric propagation that is shown, through analysis and simulations, to offer bounded guarantees on path quality while significantly reducing the cost of propagating the dynamic link metric information. Our results indicate that threshold based updates can reduce the number of link updates by up to 90-95% in some cases.\",\n \"Title: Write amplification due to ECC on flash memory or leave those bit errors alone\\nYear: 2012\\n\\nAbstract: While flash memory is receiving significant attention because of many attractive properties, concerns about write endurance delay the wider deployment of the flash memory. This paper analyzes the effectiveness of protection schemes designed for flash memory, such as ECC and scrubbing. The bit error rate of flash memory is a function of the number of program-erase cycles the cell has gone through, making the reliability dependent on time and workload. Moreover, some of the protection schemes require additional write operations, which degrade flash memory's reliability. These issues make it more complex to reveal the relationship between the protection schemes and flash memory's lifetime. In this paper, a Markov model based analysis of the protection schemes is presented. Our model considers the time varying reliability of flash memory as well as write amplification of various protection schemes such as ECC. Our study shows that write amplification from these various sources can significantly affect the benefits of these schemes in improving the lifetime. Based on the results from our analysis, we propose that bit errors within a page be left uncorrected until a threshold of errors are accumulated. We show that such an approach can significantly improve lifetimes by up to 40%.\",\n \"Title: TCP-DCR: Making TCP Robust to Non-congestion Events\\nYear: 2004\\n\\nAbstract: In this paper, we propose and evaluate TCP-DCR. TCP-DCR makes simple modifications to the TCP congestion control algorithm to make it more robust to non-congestion events. The key idea here is to delay the congestion response of TCP for a short interval of time tau, thereby creating room for local recovery mechanisms to handle any non-congestion events that may have occurred. If at the end of the delay tau, the event is not handled, then it is treated as a congestion loss. We evaluate TCP-DCR through analysis and simulations. The evaluation is done for three scenarios - a wireless network with channel errors, a wired network with packet reordering and a network with zero non-congestion events. The simulation results show that significant performance improvements can be achieved by using TCP-DCR in the presence of non-congestion events with zero or marginal impact in the absence of non-congestion events. TCP-DCR remains fair to the native implementations of TCP that respond to congestion immediately after receiving three dupacks. TCP-DCR is a simple, effective scheme providing a unified solution to several problems with minimal implementation overhead.\",\n \"Title: Don't let RAID raid the lifetime of your SSD array\\nYear: 2013\\n\\nAbstract: Parity protection at system level is typically employed to compose reliable storage systems. However, careful consideration is required when SSD based systems employ parity protection. First, additional writes are required for parity updates. Second, parity consumes space on the device, which results in write amplification from less efficient garbage collection at higher space utilization. This paper analyzes the effectiveness of SSD based RAID and discusses the potential benefits and drawbacks in terms of reliability. A Markov model is presented to estimate the lifetime of SSD based RAID systems in different environments. In a single array, our preliminary results show that parity protection provides benefit only with considerably low space utilizations and low data access rates. However, in a large system, RAID improves data lifetime even when we take write amplification into account.\",\n \"Title: Disjoint Multi-Path Routing and Failure Recovery\\nYear: 2010\\n\\nAbstract: Applications such as Voice over IP and video delivery require continuous network service, requiring fast failure recovery mechanisms. Proactive Fast failure recovery mechanisms have been recently proposed to improve network performance during the failure transients. The proposed mechanisms need extra infrastructural support in the form of routing table entries, extra addresses etc. In this paper, we study if the extra infrastructure support can be exploited to build disjoint paths in those frameworks, while keeping the recovery path lengths close to the primary paths. Our evaluations show that it is possible to extend the proactive recovery mechanisms to provide support for nearly-disjoint paths.\",\n \"Title: Adaptive marking for assured forwarding service\\nYear: 2002\\n\\nAbstract: This letter looks at the problem of achieving specific quality of service (QoS) goals of individual flows by flexibly managing resources available to an aggregated source. We present an adaptive marker based on a TCP performance model within a Diffserv network, and show that the aggregated marker can maintain state of individual flows at the edge of the network and utilize this state effectively in adaptively marking packets of individual flows to meet their QoS goals.\",\n \"Title: Design and evaluation of gracefully degradable disk arrays\\nYear: 1993\\n\\nAbstract: : The availability of inexpensive, small, magnetic disks, has made possible thebuilding of a reliable, high-performance disk system by integrating a number of such disks inan array. To achieve high reliability in such systems, equivalent to that of larger disks, parityor other error-correcting codes may be used. In systems where data availability is critical,traditionally dual copy methods have been used. Recently some parity-based schemes havebeen proposed for providing fault tolerance...\",\n \"Title: A Client Oriented, Ip Level Redirection Mechanism\\nYear: 1999\\n\\nAbstract: This paper introduces a new approach for implementing transparent client access to network services. Ever increasing load on the Internet has made it essential to design services that are fast, reliable, easily manageable, transparent to access, and that can scale gracefully with load. A common way of achieving this has been replicating services across multiple servers and redirecting clients to different servers depending upon various criteria. Existing schemes are either entirely server or network based. This scheme involves the client network layer actively in redirection. The paper describes the redirection protocol in detail and a FreeBSD based implementation of a testbed. The performance of the mechanism is measured by experiments on the testbed and analyzed. The advantages and disadvantages of client based network level redirection are discussed and some useful applications that it enables are described.\",\n \"Title: NVMFS: A hybrid file system for improving random write in nand-flash SSD\\nYear: 2013\\n\\nAbstract: In this paper, we design a storage system consisting of Nonvolatile DIMMs (as NVRAM) and NAND-flash SSD. We propose a file system NVMFS to exploit the unique characteristics of these devices which simplifies and speeds up file system operations. We use the higher performance NVRAM as both a cache and permanent space for data. Hot data can be permanently stored on NVRAM without writing back to SSD, while relatively cold data can be temporarily cached by NVRAM with another copy on SSD. We also reduce the erase overhead of SSD by reorganizing writes on NVRAM before flushing to SSD. We have implemented a prototype NVMFS within a Linux Kernel and compared with several modern file systems such as ext3, btrfs and NILFS2. We also compared with another hybrid file system Conquest, which originally was designed for NVRAM and HDD. The experimental results show that NVMFS improves IO throughput by an average of 98.9% when segment cleaning is not active, while improves throughput by an average of 19.6% under high disk utilization (over 85%) compared to other file systems. We also show that our file system can reduce the erase operations and overheads at SSD.\",\n \"Title: An Evaluation of the Effectiveness of Measurement-based Anomaly Detection Techniques\\nYear: 2006\\n\\nAbstract: A number of recent studies have proposed measurement based approaches to network traffic analysis. These techniques treat traffic volume and traffic header data as signals or images in order to make analysis feasible. We use trace-driven experiments and compare the performance of different strategies. Our evaluations on real traces reveal differences in the effectiveness of different traffic header data as potential signals for traffic analysis in terms of their detection rates and false alarm rates. Our results show that address distributions and number of flows are better signals than traffic volume for anomaly detection.\",\n \"Title: ENDE: An End-to-end Network Delay Emulator Tool for Multimedia Protocol Development\\nYear: 2001\\n\\nAbstract: Multimedia applications and protocols are constantly being developed to run over the Internet. A new protocol or application after being developed has to be tested on the real Internet or simulated on a testbed for debugging and performance evaluation. In this paper, we present a novel tool, ENDE, that can emulate end-to-end delays between two hosts without requiring access to the second host. The tool enables the user to test new multimedia protocols realistically on a single machine. In a delay-observing mode, ENDE can generate accurate traces of one-way delays between two hosts on the network. In a delay-impacting mode, ENDE can be used to simulate the functioning of a protocol or an application as if it were running on the network. We will show that ENDE allows accurate estimation of one-way transit times and hence can be used even when the forward and reverse paths are asymmetric between the two hosts. Experimental results are also presented to show that ENDE is fairly accurate in the delay-impacting mode.\",\n \"Title: An Evaluation of Multiple-Disk I/O Systems\\nYear: 1989\\n\\nAbstract: Alternative ways of configuring an I/O subsystem with multiple disks to improve the I/O performance are considered. Specifically, the author consider disk synchronization, data declustering/disk striping, and a combination of both these approaches. They evaluate many different organizations that have not been considered before. The effects of block size and other parameters of the system are examined. Two different workloads are considered for the evaluation: a file/transaction system workload and a scientific applications workload. Through simulations it is shown that synchronized organizations perform better than other organizations at very low request rates; that there is a tradeoff in the amount of declustering/synchronization to be used in a system; and that systems with higher parallelism in reading a file perform better in a scientific workload.\",\n \"Title: ARI: Adaptive LLC-memory traffic management\\nYear: 2013\\n\\nAbstract: Decreasing the traffic from the CPU LLC to main memory is a very important issue in modern systems. Recent work focuses on cache misses, overlooking the impact of writebacks on the total memory traffic, energy consumption, IPC, and so forth. Policies that foster a balanced approach, between reducing write traffic to memory and improving miss rates, can increase overall performance and improve energy efficiency and memory system lifetime for NVM memory technology, such as phase-change memory (PCM). We propose Adaptive Replacement and Insertion (ARI), an adaptive approach to last-level CPU cache management, optimizing the two parameters (miss rate and writeback rate) simultaneously. Our specific focus is to reduce writebacks as much as possible while maintaining or improving the miss rate relative to conventional LRU replacement policy. ARI reduces LLC writebacks by 33%, on average, while also decreasing misses by 4.7%, on average. In a typical system, this boosts IPC by 4.9%, on average, while decreasing energy consumption by 8.9%. These results are achieved with minimal hardware overheads.\",\n \"Title: Statistical techniques for detecting traffic anomalies through packet header data\\nYear: 2008\\n\\nAbstract: This paper proposes a traffic anomaly detector, operated in postmortem and in real-time, by passively monitoring packet headers of traffic. The frequent attacks on network infrastructure, using various forms of denial of service attacks, have led to an increased need for developing techniques for analyzing network traffic. If efficient analysis tools were available, it could become possible to detect the attacks, anomalies and to take action to contain the attacks appropriately before they have had time to propagate across the network. In this paper, we suggest a technique for traffic anomaly detection based on analyzing correlation of destination IP addresses in outgoing traffic at an egress router. This address correlation data are transformed using discrete wavelet transform for effective detection of anomalies through statistical analysis. Results from trace-driven evaluation suggest that proposed approach could provide an effective means of detecting anomalies close to the source. We also present a multidimensional indicator using the correlation of port numbers and the number of flows as a means of detecting anomalies.\",\n \"Title: CATS: Characterizing automation of Twitter spammers\\nYear: 2013\\n\\nAbstract: Twitter, with its rising popularity as a micro-blogging website, has inevitably attracted the attention of spammers. Spammers use myriad of techniques to evade security mechanisms and post spam messages, which are either unwelcome advertisements for the victim or lure victims in to clicking malicious URLs embedded in spam tweets. In this paper, we propose several novel features capable of distinguishing spam accounts from legitimate accounts. The features analyze the behavioral and content entropy, bait-techniques, and profile vectors characterizing spammers, which are then fed into supervised learning algorithms to generate models for our tool, CATS. Using our system on two real-world Twitter data sets, we observe a 96% detection rate with about 0.8% false positive rate beating state of the art detection approach. Our analysis reveals detection of more than 90% of spammers with less than five tweets and about half of the spammers detected with only a single tweet. Our feature computation has low latency and resource requirement making fast detection feasible. Additionally, we cluster the unknown spammers to identify and understand the prevalent spam campaigns on Twitter.\",\n \"Title: A study of I/O system organizations\\nYear: 1992\\n\\nAbstract: With the increasing processing speeds, it has become important to design powerful and efficient I/O systems. In this paper, we look at several design options in designing an I/O system and study their impact on the performance. Specifically, we use trace driven simulations to study a disk system with a nonvolatile cache. Some of the considered design parameters include the cache block size, the fetch size, the cache size and the disk access policy. We show that decoupling the fetch size and the cache block size results in significant performance improvements. A new write-back policy is presented that is shown to offer significant performance benefits. We show that optimal block size in a two-level memory hierarchy is dependent only on the latency, data rate product of the second level as previously conjectured. We also present results showing the effect of a split access operation of a disk read/write head.\",\n \"Title: A fast rerouting scheme for OSPF/IS-IS networks\\nYear: 2004\\n\\nAbstract: Most current backbone networks use link-state protocol, OSPF or IS-IS, as their intra-domain routing protocol. Link-state protocols perform global routing table update to route around the failures. It usually takes seconds. As real-time applications like VoIP emerge in recent years, there is a requirement for a fast rerouting mechanism to route around failures before all routers on the network update their routing tables. In addition, fast rerouting is more appropriate than global routing table update when failures are transient. We propose such a fast rerouting extension for link-state protocols. In our approach, when a link fails, the affected traffic is rerouted along a pre-computed rerouting path. In case rerouting cannot be done locally, the local router signals the minimal number of upstream routers to setup the rerouting path for rerouting. We propose algorithms that simplify the rerouting operation and the rerouting path setup. With a simple extension to the current link state protocols, our scheme can route around failures faster and involves minimal number of routers for rerouting.\",\n \"Title: SCMFS: A File System for Storage Class Memory and its Extensions\\nYear: 2013\\n\\nAbstract: Modern computer systems have been built around the assumption that persistent storage is accessed via a slow, block-based interface. However, emerging nonvolatile memory technologies (sometimes referred to as storage class memory (SCM)), are poised to revolutionize storage systems. The SCM devices can be attached directly to the memory bus and offer fast, fine-grained access to persistent storage. In this article, we propose a new file system---SCMFS, which is specially designed for Storage Class Memory. SCMFS is implemented on the virtual address space and utilizes the existing memory management module of the operating system to help mange the file system space. As a result, we largely simplified the file system operations of SCMFS, which allowed us a better exploration of performance gain from SCM. We have implemented a prototype in Linux and evaluated its performance through multiple benchmarks. The experimental results show that SCMFS outperforms other memory resident file systems, tmpfs, ramfs and ext2 on ramdisk, and achieves about 70% of memory bandwidth for file read/write operations.\",\n \"Title: Umbrella file system: Storage management across heterogeneous devices\\nYear: 2009\\n\\nAbstract: With the advent of and recent developments in Flash storage, device characteristic diversity is becoming both more prevalent and more distinct. In this article, we describe the Umbrella File System (UmbrellaFS), a stackable file system designed to provide flexibility in matching diversity of file access characteristics to diversity of device characteristics through a user or system administrator specified policy. We present the design and results from a prototype implementation of UmbrellaFS on both Linux 2.4 and 2.6. The results show that UmbrellaFS has little overhead for most file system operations while providing an ability better to utilize the differences in Flash and traditional hard drives. With appropriate use of rules, we have shown improvements of up to 44% in certain situations.\",\n \"Title: An evaluation of storage systems based on network-attached disks\\nYear: 1998\\n\\nAbstract: The emergence of network-attached disks provides the possibility of transferring data between the storage system and the client directly. This offers new possibilities in building a distributed storage system. We examine different storage organizations based on network-attached disks and compare the performance of these systems to a traditional system. Trace driven simulations are used to measure the average response times of the client requests in two different workloads. The results indicate that it is possible to reduce the workload on the file manager and improve performance in some workloads. However, in other workloads, reduced cache hit ratios may offset the gains of distributing the network processing workload\",\n \"Title: Performance Evaluation of Storage Systems Based on Network-Attached Disks\\nYear: 2000\\n\\nAbstract: The emergence of network-attached disks provides the possibility of transferring data between the storage system and the client directly. This offers new possibilities in building a distributed storage system. In this paper, we examine different storage organizations based on network-attached disks and compare the performance of these systems to a traditional system. Trace-driven simulations are used to measure the average response times of the client requests in two different workloads. We show that the advantages of distributing the server's network processing load to disks can be offset by losses in cache hits in a system based on network-attached disks. The two workloads we considered highlight this impact. We show that it is possible to offload significant load from the server by utilizing network-attached disks and point out that specific applications may be able to better exploit the features of network-attached disks.\",\n \"Title: I/O issues for hypercubes\\nYear: 1989\\n\\nAbstract: In this paper, we look at several issues concerning the design of a disk system for a multiprocessor such as a hypercube. We propose a methodology for connecting the I/O processors to such a system for efficient I/O access. An analysis is presented to see the effect of I/O communication on the network of the multiprocessor. We evaluate different disk organizations that can be employed within such a system to see which organization has a better performance. Then we consider the problem of mapping specific data structures such as matrices onto the disks such that the data can be accessed efficiently.\",\n \"Title: Real-time detection and containment of network attacks using QoS regulation\\nYear: 2005\\n\\nAbstract: In this paper, we present a network measurement mechanism that can detect and mitigate attacks and anomalous traffic in real-time using QoS regulation. The detection method rapidly pursues the dynamics of the network on the basis of correlation properties of the network protocols. By observing the proportion occupied by each traffic protocol and correlating it to that of previous states of traffic, it can be possible to determine whether the current traffic is behaving normally. When abnormalities are detected, our mechanism allows aggregated resource regulation of each protocol's traffic. The trace-driven results show that the rate-based regulation of traffic characterized by protocol classes is a feasible vehicle for mitigating the impact of network attacks on end servers.\",\n \"Title: A Fault Secure Dictionary Machine\\nYear: 1987\\n\\nAbstract: A fault-secure dictionary machine is presented in this paper. The symmetry of the binary tree architecture for a dictionary machine is exploited to obtain fault-secureness with little overhead. The proposed design utilizes only one extra processor and with some other modifications to the structure of the processors, can detect a single failure of a processor or a link. The proposed design keeps two copies of each record and whenever a record is extracted from the machine the two copies are compared to detect if any fault has occurred. The low overhead is a result of observing the fact that at any given time all the processors of the machine need not be active and these potentially-idle processors are used to do redundant processing to enable detecting single faults.\",\n \"Title: Adaptive marking for aggregated flows\\nYear: 2001\\n\\nAbstract: The differentiated services architecture is receiving wide attention as a framework for providing different levels of service according to a service profile in the Internet. The current architecture allows aggregation of flows sharing a service profile. This paper looks at the problem of achieving specific QoS goals of individual flows by flexibly managing resources available to an aggregated source. We derive a simple analytic model for the relationship between per-session behavior, aggregate packet marking and packet differentiation within a DiffServ network. The paper presents an adaptive marker based on a TCP performance model within a DiffServ network. The paper shows that an aggregated marker can maintain the state of individual flows at the edge of the network and utilize this state effectively in adaptively marking packets of individual flows to meet their QoS goals\",\n \"Title: A study of analyzing network traffic as images in real-time\\nYear: 2005\\n\\nAbstract: This paper presents NetViewer, a network measurement approach that can simultaneously detect, identify and visualize attacks and anomalous traffic in real-time by passively monitoring packet headers. We propose to represent samples of network packet header data as frames or images. With such a formulation, a series of samples can be seen as a sequence of frames or video. This enables techniques from image processing and video compression to be applied to the packet header data to reveal interesting properties of traffic. We show that \\\"scene change analysis\\\" can reveal sudden changes in traffic behavior or anomalies. We also show that \\\"motion prediction\\\" techniques can be employed to understand the patterns of some of the attacks. We show that it may be feasible to represent multiple pieces of data as different colors of an image enabling a uniform treatment of multidimensional packet header data. We compare NetViewer with classical detection theory based Neyman-Pearson test and an IDS tool.\",\n \"Title: Emulating AQM from end hosts\\nYear: 2007\\n\\nAbstract: In this paper, we show that end-host based congestion prediction is more accurate than previously characterized. However, it may not be possible to entirely eliminate the uncertainties in congestion prediction. To address these uncertainties, we propose Probabilistic Early Response TCP (PERT). PERT emulates the behavior of AQM/ECN, in the congestion response function of end-hosts. We present fluid-flow analysis of PERT/RED and PERT/PI, versions of PERT that emulate router-based RED and PI controllers. Our analysis shows that PERT/RED has better stability behavior than router-based RED. We also present results from ns-2 simulations to show the practical feasibility of PERT. The scheme presented here is general and can be used for emulating other AQM algorithms.\",\n \"Title: Designing ISP-friendly Peer-to-Peer Networks Using Game-based Control\\nYear: 2009\\n\\nAbstract: The rapid growth of peer-to-peer (P2P) networks in the past few years has\\nbrought with it increases in transit cost to Internet Service Providers (ISPs),\\nas peers exchange large amounts of traffic across ISP boundaries. This ISP\\noblivious behavior has resulted in misalignment of incentives between P2P\\nnetworks--that seek to maximize user quality--and ISPs--that would seek to\\nminimize costs. Can we design a P2P overlay that accounts for both ISP costs as\\nwell as quality of service, and attains a desired tradeoff between the two? We\\ndesign a system, which we call MultiTrack, that consists of an overlay of\\nmultiple \\\\emph{mTrackers} whose purpose is to align these goals. mTrackers\\nsplit demand from users among different ISP domains while trying to minimize\\ntheir individual costs (delay plus transit cost) in their ISP domain. We design\\nthe signals in this overlay of mTrackers in such a way that potentially\\ncompetitive individual optimization goals are aligned across the mTrackers. The\\nmTrackers are also capable of doing admission control in order to ensure that\\nusers who are from different ISP domains have a fair chance of being admitted\\ninto the system, while keeping costs in check. We prove analytically that our\\nsystem is stable and achieves maximum utility with minimum cost. Our design\\ndecisions and control algorithms are validated by Matlab and ns-2 simulations.\",\n \"Title: Detecting algorithmically generated malicious domain names\\nYear: 2010\\n\\nAbstract: Recent Botnets such as Conficker, Kraken and Torpig have used DNS based \\\"domain fluxing\\\" for command-and-control, where each Bot queries for existence of a series of domain names and the owner has to register only one such domain name. In this paper, we develop a methodology to detect such \\\"domain fluxes\\\" in DNS traffic by looking for patterns inherent to domain names that are generated algorithmically, in contrast to those generated by humans. In particular, we look at distribution of alphanumeric characters as well as bigrams in all domains that are mapped to the same set of IP-addresses. We present and compare the performance of several distance metrics, including KL-distance, Edit distance and Jaccard measure. We train by using a good data set of domains obtained via a crawl of domains mapped to all IPv4 address space and modeling bad data sets based on behaviors seen so far and expected. We also apply our methodology to packet traces collected at a Tier-1 ISP and show we can automatically detect domain fluxing as used by Conficker botnet with minimal false positives.\",\n \"Title: Modeling Network Traffic As Images\\nYear: 2005\\n\\nAbstract: This paper presents a network measurement approach to represent samples of network packet header data as frames or images. With such a formulation, a series of samples can be seen as a sequence of frames or video. This enables techniques from image processing and video compression to be applied to the analysis of packet header data to reveal interesting properties of traffic. We show that traffic images. can reveal sudden changes in traffic behavior or anomalies. Using a combination of visual modeling and trace-driven simulation, we evaluate how the design factors impact the representation of dynamic network traffic. In particular, we study the impact of sampling rate and retained DCT coefficients on the network traffic data representation.\",\n \"Title: LTCP: improving the performance of TCP in highspeed networks\\nYear: 2006\\n\\nAbstract: In this paper, we propose Layered TCP (LTCP for short), a set of simple modifications to the congestion window response of TCP to make it more scalable in highspeed networks. LTCP modifies the TCP flow to behave as a collection of virtual flows to achieve more efficient bandwidth probing. The number of virtual flows emulated is determined based on the dynamic network conditions by using the concept of virtual layers, such that the convergence properties and RTT-unfairness behavior is maintained similar to that of TCP. In this paper, we provide the intuition and the design for the LTCP protocol modifications and evaluation results based on ns-2 simulations and Linux implementation. Our results show that LTCP has promising convergence properties, is about an order of magnitude faster than TCP in utilizing high bandwidth links, employs few parameters and retains AIMD characteristics\",\n \"Title: Effectiveness of caching policies for a Web server.\\nYear: 1997\\n\\nAbstract: We look at a number of policies for managing a file system cache in a Web server. Traces from NCSA World Wide Web (WWW) server are used in this study. We study several caching policies for improving the overall performance of such a server and show that request response time can be improved by some of the replacement policies that take size of the request into account. It is shown that least frequently used (LFU) performs well when cache sizes are small (<16 MB) and SpacexAge replacement policy performs consistently better than LRU. We show that previously proposed cache sharing policies improve performance significantly in clustered Web servers\",\n \"Title: Real-time communication scheduling in a multicomputer video server\\nYear: 1999\\n\\nAbstract: In this paper, we address the problem of scheduling communication over the interconnection network of a distributed-memory multicomputer video server. We show that this problem is closely related to the problem of data distribution and movie scheduling in such a system. A solution is proposed in this paper that addresses these three issues at once. The movies are distri- buted evenly over all nodes of the multicomputer. The proposed solution minimizes the contention for links over the switch. The proposed solution makes movie scheduling very simpleif the first block of the movie is scheduled, the rest of the movie is automatically scheduled. Moreover, if the first block of the movie stream is scheduled without network contention, the proposed solution guarantees that there will be no network contention during the entire duration of playback of that movie. We show that the proposed approach to communication scheduling is optimal in utilizing the network resources. Extensive simulation results are presented to show the effectiveness of the proposed approach. 1999 Academic Press\",\n \"Title: Detecting algorithmically generated domain-flux attacks with DNS traffic analysis\\nYear: 2012\\n\\nAbstract: Recent botnets such as Conficker, Kraken, and Torpig have used DNS-based \\\"domain fluxing\\\" for command-and-control, where each Bot queries for existence of a series of domain names and the owner has to register only one such domain name. In this paper, we develop a methodology to detect such \\\"domain fluxes\\\" in DNS traffic by looking for patterns inherent to domain names that are generated algorithmically, in contrast to those generated by humans. In particular, we look at distribution of alphanumeric characters as well as bigrams in all domains that are mapped to the same set of IP addresses. We present and compare the performance of several distance metrics, including K-L distance, Edit distance, and Jaccard measure. We train by using a good dataset of domains obtained via a crawl of domains mapped to all IPv4 address space and modeling bad datasets based on behaviors seen so far and expected. We also apply our methodology to packet traces collected at a Tier-1 ISP and show we can automatically detect domain fluxing as used by Conficker botnet with minimal false positives, in addition to discovering a new botnet within the ISP trace. We also analyze a campus DNS trace to detect another unknown botnet exhibiting advanced domain-name generation technique.\",\n \"Title: Fault Tolerance Techniques for Systolic Arrays\\nYear: 1987\\n\\nAbstract: First Page of the Article\",\n \"Title: System support for providing integrated services from networked multimedia storage servers\\nYear: 2001\\n\\nAbstract: In this paper, we describe our experience in building an integrated multimedia storage system, Prism. Our current Linux-based implementation of Prism provides three levels of service: deadline guarantees for periodic applications, best-effort better response times for interactive applications and starvation-free throughput guarantees for aperiodic applications. Prism separates resource allocation from resource scheduling. Resource allocation is controlled across the service classes by a system-wide policy and service class specific admission controllers. Resource scheduling is done at the resources. This separation allows Prism to be deployed even when the storage system is separated on a network from the file system.We report on the important aspects of Prism architecture, innovations required to build Prism on top of Linux and lessons learned during the implementation and testing of Prism. We present experimental results to show that Prism achieves its goals in supporting multiple service classes within a single system. We compare Prism against standard Linux operating system to show the impact of our approach.\",\n \"Title: Realizing throughput guarantees in a differentiated services network\\nYear: 1999\\n\\nAbstract: This paper discusses techniques for achieving desired throughput guarantees in the Internet that supports a differentiated services framework. The diff-serv framework proposes the use of different drop precedences to achieve service guarantees over the Internet. However, it has been observed that the drop precedences by themselves cannot achieve the desired target rates because of the strong interaction of the transport protocol with packet drops in the network. This paper proposes and evaluates a number of techniques to better achieve the throughput guarantees in such networks. The proposed techniques consider: modifying the transport protocol at the sender; modifying the marking strategies at the marker; and modifying the dropping policies at the router. It is shown that these techniques improve the likelihood of achieving the desired throughput guarantees and also improve the service differentiation\",\n \"Title: Constructing disjoint paths for failure recovery and multipath routing\\nYear: 2012\\n\\nAbstract: Applications such as Voice over IP and video streaming require continuous network service, requiring fast failure recovery mechanisms. Proactive failure recovery schemes have been recently proposed to improve network performance during the failure transients. These proactive failure recovery schemes need extra infrastructural support in the form of routing table entries, extra addresses etc. In this paper, we study if the extra infrastructure support can be exploited to build disjoint paths in those frameworks, while keeping the lengths of the recovery paths close to those of the primary paths. Our evaluations show that it is possible to extend the proactive failure recovery schemes to provide support for nearly-disjoint paths which can be employed in multipath routing for load balancing and QoS.\",\n \"Title: Fault tolerance of adaptive routing algorithms in multicomputers\\nYear: 1992\\n\\nAbstract: An evaluation of the effectiveness of adaptive routing techniques in tolerating failures is presented. It is shown that adaptive routing techniques yield gracefully degradable systems for the workloads considered. For medium to large communication granularity and the workloads considered in this study, if adaptive-routing is used, it is shown that the problem completion time does not increase drastically due to failures. When node failures were considered, it was observed that the mismatch of problem communication structure with the physical communication structure did not result in significant loss of performance. Since adaptive routing techniques are warranted for performance reasons, it is argued that making use of this adaptive routing hardware to tolerate failures is a favorable option.\",\n \"Title: Design, Analysis, and Simulation of I/O Architectures for Hypercube\\nYear: 1990\\n\\nAbstract: Several issues concerning the design of an I/O (input/output) system for a multiprocessor such as a hypercube are examined. A methodology is proposed for connecting the I/O processors to such a system for efficient I/O access. The effect of I/O communication onthe multiprocessor network is analyzed. Different disk organizations that can be employed within such a system are evaluated to see which organization has a better performance. It is observed that parallelism in serving an I/O request plays a dominant role in the scientific workload. The problem of mapping specific data structures such as matrices onto the disks so that the data can be accessed efficiently is considered.\",\n \"Title: An approach to virtual allocation in storage systems\\nYear: 2006\\n\\nAbstract: This article presents virtual allocation, a scheme for flexible storage allocation. Virtual allocation separates storage allocation from the file system. It employs an allocate-on-write strategy which lets applications fit into the actual usage of storage space, without regard to the configured file system size. This improves flexibility by allowing storage space to be shared across different file systems. This article presents the design of virtual allocation and its evaluation through benchmarks. To illustrate our approach, we implemented a prototype system on PCs running Linux. We present the results from the prototype implementation and its evaluation.\",\n \"Title: SACRIO: an active buffer management scheme for differentiated service networks\\nYear: 2001\\n\\nAbstract: In this paper, we propose an active resource management approach for D ifferentiated Services networks. The proposed approach, SACRIO, employs caching and localized packet remarking within a router. It is shown that SACRIO is simple and can be implemented transparently within the diff-serv architecture. It is shown that the packet handling cost remains O(1) with SACRIO. SACRIO is shown to be quite effective and scalable through both ns-based simulations and real-world trace-driven simulations.\",\n \"Title: Evaluation of Data and Request Distribution Policies in Clustered Servers\\nYear: 1999\\n\\nAbstract: The popularity of the WWW has prompted the design of scalable clustered servers. Data and request distribution policies greatly\\n impact performance in such servers. In this paper, request response time is used as a measure in evaluating these policies.\\n Our results indicate that policies favoring locality seem to perform better than policies favoring load balance. We propose\\n and evaluate a dynamic client-based request distribution policy.\",\n \"Title: Performance of Quantized Congestion Notification in TCP Incast Scenarios of Data Centers\\nYear: 2010\\n\\nAbstract: This paper analyzes the performance of Ethernet layer congestion control mechanism Quantized Congestion Notification (QCN) during data access from clustered servers in data centers. We analyze the reasons why QCN does not perform adequately in these situations and propose several modifications to the protocol to improve its performance in these scenarios. We trace the causes of QCN performance degradation to flow rate variability, and show that adaptive sampling at the switch and adaptive self-increase of flow rates at the rate limiter improve performance in a TCP In cast setup significantly. We compare the performance of QCN against TCP modifications in a heterogeneous environment, and show that modifications to QCN yield better performance.\",\n \"Title: A Large-Scale Empirical Study of Conficker\\nYear: 2012\\n\\nAbstract: Conficker is the most recent widespread, well-known worm/bot. According to several reports, it has infected about 7 million to 15 million hosts and the victims are still increasing even now. In this paper, we analyze Conficker infections at a large scale, about 25 million victims, and study various interesting aspects about this state-of-the-art malware. By analyzing Conficker, we intend to understand current and new trends in malware propagation, which could be very helpful in predicting future malware trends and providing insights for future malware defense. We observe that Conficker has some very different victim distribution patterns compared to many previous generation worms/botnets, suggesting that new malware spreading models and defense strategies are likely needed. We measure the potential power of Conficker to estimate its effects on the networks/hosts when it performs malicious operations. Furthermore, we intend to determine how well a reputation-based blacklisting approach can perform when faced with new malware threats such as Conficker. We cross-check several DNS blacklists and IP/AS reputation data from Dshield and FIRE and our evaluation shows that unlike a previous study which shows that a blacklist-based approach can detect most bots, these reputation-based approaches did relatively poorly for Conficker. This raises a question of how we can improve and complement existing reputation-based techniques to prepare for future malware defense? Based on this, we look into some insights for defenders. We show that neighborhood watch is a surprisingly effective approach in the case of Conficker. This suggests that security alert sharing/correlation (particularly among neighborhood networks) could be a promising approach and play a more important role for future malware defense.\",\n \"Title: TCP-DCR: A Novel Protocol for Tolerating Wireless Channel Errors\\nYear: 2005\\n\\nAbstract: This paper presents TCP-DCR, a set of simple modifications to the TCP protocol to improve its robustness to channel errors in wireless networks. TCP-DCR is based on the simple idea of allowing the link-level mechanism to recover the packets lost, due to channel errors, thereby limiting the response of the transport protocol to mostly congestion losses. This is done by delaying the triggering of congestion response algorithms for a small bounded period of time \\\\tau to allow the link-level retransmissions to recover the loss due to channel errors. If at the end of the delay \\\\tau the packet is not recovered, then it is treated as a packet lost due to congestion. We analyze TCP-DCR to show that the delay in congestion response does not impact the fairness towards the native implementations of TCP that respond to congestion immediately after receiving three dupacks. We evaluate TCP-DCR through simulations to show that it offers significantly better performance when channel errors contribute more towards packet losses in the network with no or minimal impact on the performance when congestion is the primary cause for packet loss. We also present an analysis to show that protocol evaluation in the wireless networks is significantly influenced by the number of flows in the network.\",\n \"Title: Impact of bandwidth-delay product and non-responsive flows on the performance of queue management schemes\\nYear: 2004\\n\\nAbstract: In this paper, we study the impact-of bandwidth-delay products and non-responsive flows on queue management schemes. Our focus is to understand the aggregate performance of various classes of traffic under different queue management schemes. Our work is motivated by the expected trends of increasing link capacities and increasing amounts of non-responsive traffic. In this paper, the impact on the performance of RED, RED with ECN enabled and droptail routers are investigated. Our study considers the aggregate bandwidth of different classes of traffic and the delays observed at the router.\",\n \"Title: Multihoming route control among a group of multihomed stub networks\\nYear: 2007\\n\\nAbstract: In recent years, commercial ''multihoming route control'' devices are used by multihomed stub networks to optimize the routing of their Internet traffic. Previous studies have shown that ''multihoming route control'' can improve the Internet communication performance of the multihomed stub networks significantly. In this work, we study ''multihoming route control'' among a group of multihomed stub networks that may belong to an organization and exchange large amount of data regularly. We show that greedy ''multihoming routing control'' schemes could lead to oscillations. We propose a user-optimal routing based ''multihoming routing control'' scheme that improves routing performance without causing oscillations. The proposed scheme is simpler to implement than optimal routing based approaches. We show through simulations that the proposed scheme achieves performance close to optimal routing and works well in various network conditions.\",\n \"Title: A method for estimating non-responsive traffic at a router\\nYear: 2002\\n\\nAbstract: In this paper, we propose a scheme for estimating the proportion of the incoming traffic that is not responding to congestion at a router. The idea of the proposed scheme is that if the observed queue length and packet drop probability do not match with the predicted results from the TCP model, then the error must come from the non-responsive traffic; it can then be used for estimating non-responsive traffic. The proposed scheme utilizes queue length history, packet drop history, expected TCP and queue dynamics to estimate the proportion. We show that the proposed scheme is effective over a wide range of traffic scenarios through simulations.\",\n \"Title: SPIRAL: a client-transparent third-party transfer scheme for network attached disks\\nYear: 2003\\n\\nAbstract: Third-party transfer is a data transfer mechanism where the party initiating the transfer is neither the source nor the sink for the data. In this paper, we present a scheme for supporting third-party transfers on storage systems with network-attached disks (NADs), called SPIRAL. SPIRAL allows NADs to send data directly to clients without going through the server It is transparent to clients and relies only on the linear block interface of the current disks (or the future iSCSI disks). SPIRAL requires no porting of file system or application-level functionality to NADs and requires only simple modifications to server applications. To illustrate our approach, we implemented a prototype system on PCs running Linux. We present experimental results for NFS and HTTP on the prototype to demonstrate the effectiveness of our approach.\",\n \"Title: Modeling TCP behavior in a differentiated services network\\nYear: 2001\\n\\nAbstract: Abstract—The differentiated services architecture has been proposed for providing different levels of services and has recently received wide attention. A packet in a diff-serv domain is classified into a class of service according to its contract profile and treated differently by its class. While many,studies have addressed issues on the diff-serv architecture (e.g., dropper, marker, classifier and shaper), there have been few attempts to analytically understand a flow’s behavior in a diff-serv network. In this paper, we propose simple models of TCP behavior in a diff-serv network. Our models quantitatively characterize TCP throughput as functions of the contract rate, the packet-drop rate and the round-trip time in either two-drop precedence or threedrop precedence network. We also extend our models to aggregated flows. The models are validated through a number of simulations. Index Terms—AF PHB, differentiated service, TCP modeling.\",\n \"Title: NetViewer: a network traffic visualization and analysis tool\\nYear: 2005\\n\\nAbstract: The frequent and large-scale network attacks have led to an increased need for developing techniques for analyzing network traffic. If efficient analysis tools were available, it could become possible to detect the attacks, anomalies and to appropriately take action to contain the attacks before they have had time to propagate across the network. This paper describes NetViewer, a network monitoring tool that can simultaneously detect, identify and visualize attacks and anomalous traffic in real-time by passively monitoring packet headers. NetViewer represents the traffic data as images, enabling the application of image/video processing techniques for the analysis of network traffic. NetViewer is released free to the general public. By employing a freely available visualization tool, the users of NetViewer can comprehend the characteristics of the network traffic observed in the aggregate. NetViewer can be employed to detect and identify network anomalies such as DoS/DDoS attacks, worms and flash crowds. NetViewer can also provide information on traffic distributions over IP address/port number domains, utilization of link capacity and effectiveness of Quality of Service policies.\",\n \"Title: A method for estimating the proportion of nonresponsive traffic at a router\\nYear: 2004\\n\\nAbstract: In this paper, a scheme for estimating the proportion of the incoming traffic that is not responsive to congestion at a router is presented. The idea of the proposed scheme is that if the observed queue length and packet drop probability do not match the predictions from a model of responsive (TCP) traffic, then the error must come from nonresponsive traffic; it can then be used for estimating the proportion of nonresponsive traffic. The proposed scheme is based on the queue length history, packet drop history, and expected TCP and queue dynamics. The effectiveness of the proposed scheme over a wide range of traffic scenarios is corroborated using ns-2-based simulations. Potential applications of the proposed algorithms in traffic engineering and control are discussed.\",\n \"Title: SCMFS: a file system for storage class memory\\nYear: 2011\\n\\nAbstract: This paper considers the problem of how to implement a file system on Storage Class Memory (SCM), that is directly connected to the memory bus, byte addressable and is also non-volatile. In this paper, we propose a new file system, called SCMFS, which is implemented on the virtual address space. In SCMFS, we utilize the existing memory management module in the operating system to do the block management and keep the space always contiguous for each file. The simplicity of SCMFS not only makes it easy to implement, but also improves the performance. We have implemented a prototype in Linux and evaluated its performance through multiple benchmarks.\",\n \"Title: Detecting Traffic Anomalies through Aggregate Analysis of Packet Header Data\\nYear: 2004\\n\\nAbstract: If efficient network analysis tools were available, it could become possible to detect the attacks, anomalies and to appropriately take action to contain the attacks. In this paper, we suggest a technique for traffic anomaly detection based on analyzing correlation of destination IP addresses in outgoing traffic at an egress router. This address correlation data are transformed through discrete wavelet transform for effective detection of anomalies through statistical analysis. Our techniques can be employed for postmortem and real-time analysis of outgoing network traffic at a campus edge. Results from trace-driven evaluation suggest that proposed approach could provide an effective means of detecting anomalies close to the network. We also present data analyzing the correlation of port numbers as a means of detecting anomalies.\",\n \"Title: Identifying Long-Term High-Bandwidth Flows at a Router\\nYear: 2001\\n\\nAbstract: In this paper, we propose a novel approach for identifying long-term high-rate flows at a router. Identifying the long-term high-rate flows allows a router to regulate flows intelligently in times of congestion. We employ a limited amount of state to record the arrivals of packets of individual flows. The state is managed as an LRU cache. The LRU cache self converges to hold only the flows that are high rate. The size of the cache need not depend on the number of flows at the router. We analyze a number of internet packet traces to show the effectiveness of this LRUbased cache. It is shown that this scheme is highly scalable since a few flows contribute a significant fraction of the traffic at a router.\",\n \"Title: Data organization in a hybrid storage system\\nYear: 2012\\n\\nAbstract: Flash memory based Solid State Devices (SSDs) have become commercially viable and hybrid storage systems incorporating SSDs and magnetic disk drives are being built. There are a number of ways the data can be organized in such hybrid systems. SSDs can be organized at a higher layer as a cache in a transparent hierarchy. SSDs can be organized as peers to magnetic disks in a heterogeneous storage system. There are pros and cons for organizing data in both approaches. This paper argues that a storage system with SSDs and disk drives as peers may deliver the best performance, with maximum flexibility.\",\n \"Title: Multipath Routing For Reducing Network Energy\\nYear: 2012\\n\\nAbstract: Reducing energy consumption in networks has been recently receiving attention. Networks are typically designed to support peak loads. Energy saving is potentially achieved by appropriately provisioning network capacity, by shutting down unnecessary network elements while the capacity of remaining network elements still meets traffic demand. Traditionally, topology control techniques are employed to reduce the network to an appropriate size. In this paper, we propose techniques to construct an appropriate topology that takes the traffic demands into account. We also study how much multipath routing can contribute to energy saving compared to single/shortest path routing. We show that the proposed techniques reduce energy consumption by an additional 17% over previous schemes.\",\n \"Title: Gracefully degradable disk arrays\\nYear: 1991\\n\\nAbstract: The problem of designing fault-tolerant disk arrays that are not susceptible to 100% load increases on the functional disks when one of the disks in the system fails is addressed. A technique that combines the advantages of parity schemes and the traditional dual copy methods and offers a wide variety of options in providing fault-tolerance is proposed. A theoretical framework for solving the problem is presented and a number of constructive techniques are proposed. By utilizing the same amount of hardware as the earlier methods but with a better data organization and a different reconstruction technique, the system yields better performance during a failure. Merging two parity groups as a reconfiguration strategy is shown to have a number of benefits, such as reduced hardware overhead and improved reliability. A combination of block designs and the proposed reconfiguration strategy results in a highly reliable disk array with the same or less overhead as the earlier approaches and better performance during a failure.<>\",\n \"Title: I/O issues in a multimedia system\\nYear: 1994\\n\\nAbstract: In future computer system design, I/O systems will have to support continuous media such as video and audio, whose system demands are different from those of data such as text. Multimedia computing requires us to focus on designing I/O systems that can handle real-time demands. Video- and audio-stream playback and teleconferencing are real-time applications with different I/O demands. We primarily consider playback applications which require guaranteed real-time I/O throughput. In a multimedia server, different service phases of a real-time request are disk, small computer systems interface (SCSI) bus, and processor scheduling. Additional service might be needed if the request must be satisfied across a local area network. We restrict ourselves to the support provided at the server, with special emphasis on two service phases: disk scheduling and SCSI bus contention. When requests have to be satisfied within deadlines, traditional real-time systems use scheduling algorithms such as earliest deadline first (EDF) and least slack time first. However, EDF makes the assumption that disks are preemptable, and the seek-time overheads of its strict real-time scheduling result in poor disk utilization. We can provide the constant data rate necessary for real-time requests in various ways that require trade-offs. We analyze how trade-offs that involve buffer space affect the performance of scheduling policies. We also show that deferred deadlines, which increase buffer requirements, improve system performance significantly.<>\",\n \"Title: Failure evaluation of disk array organizations\\nYear: 1993\\n\\nAbstract: The authors present an evaluation of some of the disk array organizations proposed in the literature. They evaluate three alternatives for sparing, hot sparing, distributed sparing, and parity sparing, and two options for data layout, regular RAID5 and block designs, and systems based on combinations of these data layout and sparing alternatives. The performance of these organizations is evaluated with different reconstruction strategies. It is shown that parity sparing and distributed sparing have better performance and shorter reconstruction times than hot sparing. It is shown that both block designs as a data layout policy and distributed sparing as a sparing policy reduce the reconstruction time after a failure. The impact of reconstruction strategies is studied, and it is shown that, at higher workloads, choice of reconstruction strategy has a significant impact on the performance of the systems\",\n \"Title: Integrated QOS management for disk I/O\\nYear: 1999\\n\\nAbstract: In this paper, we address the problem of providing different levels of performance guarantees or quality-of-service (QOS) for disk I/O. We classify disk requests into three categories based on the provided level of service. We propose an integrated scheme that provides different levels of performance guarantees in a single system. The proposed method uses admission control and bandwidth allocation to isolate the different categories of requests and appropriate disk scheduling to achieve the desired performance goals. It is shown that the proposed method achieves the performance goals of individual requests while allowing seek optimizations at the disk. In particular, it is shown that the proposed scheme for integrated QOS management at disk provides significantly better performance than SCAN.\",\n \"Title: Multi Path PERT\\nYear: 2013\\n\\nAbstract: This paper presents a new multipath delay based algorithm, MPPERT (Multipath Probabilistic Early response TCP), which provides high throughput and efficient load balancing. In all-PERT environment, MPPERT suffers no packet loss and maintains much smaller queue sizes compared to existing MPTCP, making it suitable for real time data transfer. MPPERT is suitable for incremental deployment in a heterogeneous environment. PERT, being a delay based TCP protocol, has continuous information about the state of the bottleneck queue along its path. This information is valuable in enabling MPPERT to detect subflows sharing a common bottleneck and obtain a smaller set of disjoint subflows. This information can even be used to switch from coupled (a set of subflows having interdependent increase/decrease of congestion windows) to uncoupled (independent increase/decrease of congestion windows) subflows, yielding higher throughput when best single-path TCP constraint is relaxed. The ns-2 simulations support MPPERT as a highly competitive multipath approach, suitable for real time data transfer, which is capable of offering higher throughput and improved reliability.\",\n \"Title: Wavelet Methods for the Detection of Anomalies and their Application to Network Traffic Analysis\\nYear: 2006\\n\\nAbstract: Here we develop an integrated tool for the online detection of network anomalies. We consider statistical change point detection algorithms, for both local changes in the variance and for the detection of jumps, and propose modified versions of these algorithms based on moving window techniques. We investigate performances on simulated data and on network traffic data with several superimposed attacks. All detection methods are based on wavelet packet transforms. Copyright (C) 2006 John Wiley & Sons, Ltd.\",\n \"Title: Dynamic Memory Pressure Aware Ballooning\\nYear: 2015\\n\\nAbstract: Hardware virtualization is a major component of large scale server and data center deployments due to their facilitation of server consolidation and scalability. Virtualization, however, comes at a high cost in terms of system main memory utilization. Current virtual machine (VM) memory management solutions impose a high performance penalty and are oblivious to the operating regime of the system. Therefore, there is a great need for low-impact VM memory management techniques which are aware of and reactive to current system state, to drive down the overheads of virtualization. We observe that the host machine operates under different memory pressure regimes, as the memory demand from guest VMs changes dynamically at runtime. Adapting to this runtime system state is critical to reduce the performance cost of VM memory management. In this paper, we propose a novel dynamic memory management policy called Memory Pressure Aware (MPA) ballooning. MPA ballooning dynamically allocates memory resources to each VM based on the current memory pressure regime. Moreover, MPA ballooning proactively reacts and adapts to sudden changes in memory demand from guest VMs. MPA ballooning requires neither additional hardware support, nor incurs extra minor page faults in its memory pressure estimation. We show that MPA ballooning provides an 13.2% geomean speed-up versus the current ballooning techniques across a set of application mixes running in guest VMs; often yielding performance nearly identical to that of a non-memory constrained system.\",\n \"Title: Shared Last-Level Caches and The Case for Longer Timeslices\\nYear: 2015\\n\\nAbstract: Memory performance is important in modern systems. Contention at various levels in memory hierarchy can lead to significant application performance degradation due to interference. Further, modern, large, last-level caches (LLC) have fill times greater than the OS scheduling window. When several threads are running concurrently and timesharing the CPU cores, they may never be able to load their working sets into the cache before being rescheduled, thus permanently stuck in the \\\"cold-start\\\" regime. We show that by increasing the system scheduling timeslice length it is possible to amortize the cache cold-start penalty due to the multitasking and improve application performance by 10--15%.\",\n \"Title: Does RAID Improve Lifetime of SSD Arrays?\\nYear: 2016\\n\\nAbstract: Parity protection at the system level is typically employed to compose reliable storage systems. However, careful consideration is required when SSD-based systems employ parity protection. First, additional writes are required for parity updates. Second, parity consumes space on the device, which results in write amplification from less efficient garbage collection at higher space utilization.\\n\\nThis article analyzes the effectiveness of SSD-based RAID and discusses the potential benefits and drawbacks in terms of reliability. A Markov model is presented to estimate the lifetime of SSD-based RAID systems in different environments. In a small array, our results show that parity protection provides benefit only with considerably low space utilizations and low data access rates. However, in a large system, RAID improves data lifetime even when we take write amplification into account.\",\n \"Title: Path confidence based lookahead prefetching.\\nYear: 2016\\n\\nAbstract: Designing prefetchers to maximize system performance often requires a delicate balance between coverage and accuracy. Achieving both high coverage and accuracy is particularly challenging in workloads with complex address patterns, which may require large amounts of history to accurately predict future addresses. This paper describes the Signature Path Prefetcher (SPP), which offers effective solutions for three classic challenges in prefetcher design. First, SPP uses a compressed history based scheme that accurately predicts complex address patterns. Second, unlike other history based algorithms, which miss out on many prefetching opportunities when address patterns make a transition between physical pages, SPP tracks complex patterns across physical page boundaries and continues prefetching as soon as they move to new pages. Finally, SPP uses the confidence it has in its predictions to adaptively throttle itself on a per-prefetch stream basis. In our analysis, we find that SPP improves performance by 27.2% over a no-prefetching baseline, and outperforms the state-of-the-art Best Offset prefetcher by 6.4%. SPP does this with minimal overhead, operating strictly in the physical address space, and without requiring any additional processor core state, such as the PC.\",\n \"Title: Towards Improving Data Validity of Cyber-Physical Systems through Path Redundancy.\\nYear: 2017\\n\\nAbstract: Cyber-physical systems have shown to be susceptible to cyber-attacks. Incidents such as Stuxnet Attack and Ukraine power outage have shown that attackers are capable of penetrating into industrial control systems, compromising PLCs, and sending false commands to physical devices while reporting normal sensing values. Therefore, one of the critical needs of CPS is to ensure the validity of the sensor values. In this paper, we explore path diversity in SCADA networks and develop Path Redundancy to improve data validity. The proposed solution is shown to be able to effectively prevent data integrity attacks and detect false command attacks from a single compromised path or PLC. We provide detailed analysis on solution design and implement an application of the technique in building automation networks. Our cost-efficient and easy-to-deploy solution improves the resilience of SCADA networks.\",\n \"Title: Adaptive policies for balancing performance and lifetime of mixed SSD arrays through workload sampling\\nYear: 2016\\n\\nAbstract: Solid-state drives (SSDs) have become promising storage components to serve large I/O demands in modern storage systems. Enterprise class (high-end) SSDs are faster and more resilient than client class (low-end) SSDs but they are expensive to be deployed in large scale storage systems. It is an attractive and practical alternative to exploit the high-end SSDs as a cache and low-end SSDs as main storage. This paper explores how to optimize a mixed SSD array in terms of performance and lifetime. This paper shows that simple integration of different classes of SSDs in traditional caching policies results in poor reliability. This paper also reveals that caching policies with static workload distribution are not always efficient. In this paper, we propose a sampling based adaptive approach that achieves fair workload distribution across the cache and the storage. The proposed algorithm enables fine-grained control of the workload distribution which minimizes latency over lifetime of mixed SSD arrays. We show that our adaptive algorithm is very effective in improving the latency over lifetime metric, on an average, by up to 2.36 times over LRU, across a number of workloads.\",\n \"Title: Speculative paging for future NVM storage.\\nYear: 2017\\n\\nAbstract: The quest for greater performance and efficiency has driven modern cloud applications towards \\\"in-memory\\\" implementations, such as memcached and Apache Spark. Looking forward, however, the costs of DRAM, due to its low area density and high energy consumption, may make this trend unsustainable. Traditionally, OS paging system mechanisms were intended to bridge the gap between expensive, under-provisioned DRAM and inexpensive, dense storage, however, in the past twenty years the latency of storage, relative to DRAM became too great to overcome without significant performance impact. Recent NVM storage devices, such as Intel Optane drives and aggressive, 3D flash SSDs, may dramatically change the picture for OS paging. These newdrives are expected to provide much lower latency compared to the existing flash-based SSDs or traditional HDDs. Unfortunately, even these future NVM drives are still much too slow to replace DRAM, since the access latency of fast NVM storage is expected on the order of tens of microseconds, and they often require block-level access. Unlike traditional HDDs, for which the baseline OS paging policies are designed, these new SSDs place no penalty for \\\"random\\\" access and their access latency promises to be significantly less than traditional SSDs, thus arguing for a rearchitecting of the OS paging system. In this paper, we propose SPAN (Speculative PAging for future NVM storage), a software-only, OS swap-based, page management and prefetching scheme designed for emerging NVM storage. Unlike the baseline OS swapping mechanism, which is highly optimized for traditional spinning disks, SPAN leverages the inherent parallelism of NVM devices to proactivley fetch a set of pages from NVM storage to the small and fast main DRAM. In doing so, SPAN yields a speedup of similar to 18% versus swapping into the NVM with the baseline OS (similar to 50% of the performance lost by the baseline OS versus placing the entire working set in DRAM memory). The proposed technique thus enables the utilization of such hybrid systems for memory-hungry applications, lowering the memory cost while keeping the performance comparable to the DRAM-only system.\",\n \"Title: Securing cyber-physical systems with adaptive commensurate response.\\nYear: 2017\\n\\nAbstract: Cyber-Physical Systems (CPSs) are ubiquitous in power systems, transport and medical systems, and critical infrastructures. Current protocols and applications allow significant changes to a system to take place within a short time or small network footprint, which can be exploited by attackers to cause a great impact on the physical systems. This paper proposes adaptive Commensurate Response (CR) to narrow down the asymmetry between the cost of attacks and their impact through enforcing command footprints to be commensurate with their impact on the system. Such impact is measured by the change of the setpoint (changedriven CR) or the distance between the operating state and the critical state (criticality-driven CR). Change-driven CR is effective for setpoint attacks with big setpoint changes, and criticality-driven CR can be used to combat both setpoint attacks and actuation attacks. Our case study on automobile cruise control demonstrates that CR can effectively improve the system resilience and attack survivability while satisfying QoS requirements.\",\n \"Title: IoTAegis: A Scalable Framework to Secure the Internet of Things\\nYear: 2018\\n\\nAbstract: The infamous Mirai attack which hijacked nearly half a million Internet connected devices demonstrated the widespread security vulnerabilities of the Internet-of-Things (IoT). This study employs a set of active and passive observation methods to discover the security vulnerabilities of IoT devices within a university campus. We show that (a) the number of non-compute devices dominates the number of compute devices with open ports in a campus network; (b) 58.9% or more devices do not keep up-to-date firmware and 51.3% or more do not have a user defined password; and (c) the number of devices together with the diversity of device ages and vendors make the protection of IoT devices a difficult problem. We further develop IoTAegis framework which offers device-level protection to automatically manage device configurations and security updates. Our solution is shown to be effective, scalable, lightweight, and deployable in different forms and network types.\",\n \"Title: KVRAID - high performance, write efficient, update friendly erasure coding scheme for KV-SSDs.\\nYear: 2021\\n\\nAbstract: Key-value (KV) stores have been widely deployed in a variety of scale-out enterprise applications such as online retail, big data analytics, social networks, etc. Key-Value SSDs (KVSSDs) provide a key-value interface directly from the device aiming at lowering software overhead and reducing I/O amplification for such applications. In this paper, we present KVRAID, a high performance, write efficient erasure coding management scheme on emerging key-value SSDs. The core innovation of KVRAID is to use logical to physical key conversion to efficiently pack similar size KV objects and dynamically manage the membership of erasure coding groups. Such design enables packing multiple user objects to a single physical object to reduce the object amplification compared to prior works. By applying out-of-place update technique, KVRAID can significantly reduce the I/O amplification compared to the state-of-art designs. Our experiments show that KVRAID outperforms state-of-art software KV-store with block RAID by 28x in terms of insert throughput and reduces CPU utilization, tail latency and write amplification significantly. Compared to state-of-art KV devices erasure coding management, KVRAID reduces object amplification by ~ 2.6x compared to StripeFinder and reduces I/O amplification by ~ 9.6x when compared to KVMD and StripeFinder for update intensive workloads.\",\n \"Title: Connect Id: Identifying Devices Before They Connect to the Network\\nYear: 2021\\n\\nAbstract: Increasingly, network smart devices are being deployed and operated over the network. The Mirai attack has shown that the vulnerabilities of these devices can be exploited to stage large-scale attacks. In this paper, we explore the potential for identifying a device before it connects to the network such that appropriate security postures can be taken based on the known characteristics of the devi...\",\n \"Title: Protection against Counterfeiting Attacks in 3D Printing by Streaming Signature-embedded Manufacturing Process Instructions\\nYear: 2021\\n\\nAbstract: ABSTRACTThe emerging Manufacturing-as-a-Service (MaaS) paradigm democratizes manufacturing by connecting people and businesses with manufacturing requests to those with manufacturing resources, via a digital thread. However, the digital thread can be vulnerable to attacks such as counterfeiting, IP theft and sabotage. An approach based on embedding custom anti-counterfeiting signatures in the design files, on-the-fly, is presented to protect against counterfeiting attacks. An experimental study of the effect of geometric and dimensional variations of the signatures in printed components, as well as the effect of the variations on the read rates of the codes are reported. We conduct experiments using a polyjet printer with veroclear material to study the positional and dimensional variations introduced in the embedded signature and the resulting effect on the readability of the design. The results suggest that signatures with spherical units of size φ0.25 mm embedded within 20 mm cubic components can be printed with location precision of ~5% (about 0.1 mm), and dimensional deviation of the order of 0.01 mm. A statistical model is developed to show that these variations pose minimum interference on the readability of the signatures. Embedding of randomized signatures offers security by serving as anti-counterfeit marks in the final part, and makes it harder to reverse engineer and produce counterfeited parts.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":78,"string":"78"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13587,"string":"13,587"}}},{"rowIdx":13588,"cells":{"question_id":{"kind":"string","value":"Supratik Mukhopadhyay"},"question":{"kind":"string","value":"Generate an abstract for the title \"A Theoretical Analysis of Deep Neural Networks for Texture Classification.\" using the following items: 1. Deep Neural Networks \n2. Texture features \n3. Vapnik-Chervonenkis dimension \n4. Intrinsic dimension \n5. Relative Contrast"},"answer":{"kind":"string","value":"We investigate the use of Deep Neural Networks for the classification of image datasets where texture features are important for generating class-conditional discriminative representations. To this end, we first derive the size of the feature space for some standard textural features extracted from the input dataset and then use the theory of Vapnik-Chervonenkis dimension to show that hand-crafted feature extraction creates low-dimensional representations which help in reducing the overall excess error rate. As a corollary to this analysis, we derive for the first time upper bounds on the VC dimension of Convolutional Neural Network as well as Dropout and Dropconnect networks and the relation between excess error rate of Dropout and Dropconnect networks. The concept of intrinsic dimension is used to validate the intuition that texture-based datasets are inherently higher dimensional as compared to handwritten digits or other object recognition datasets and hence more difficult to be shattered by neural networks. We then derive the mean distance from the centroid to the nearest and farthest sampling points in an n-dimensional manifold and show that the Relative Contrast of the sample data vanishes as dimensionality of the underlying vector space tends to infinity."},"correct_answer":{"kind":"string","value":"We investigate the use of Deep Neural Networks for the classification of image datasets where texture features are important for generating class-conditional discriminative representations. To this end, we first derive the size of the feature space for some standard textural features extracted from the input dataset and then use the theory of Vapnik-Chervonenkis dimension to show that hand-crafted feature extraction creates low-dimensional representations which help in reducing the overall excess error rate. As a corollary to this analysis, we derive for the first time upper bounds on the VC dimension of Convolutional Neural Network as well as Dropout and Dropconnect networks and the relation between excess error rate of Dropout and Dropconnect networks. The concept of intrinsic dimension is used to validate the intuition that texture-based datasets are inherently higher dimensional as compared to handwritten digits or other object recognition datasets and hence more difficult to be shattered by neural networks. We then derive the mean distance from the centroid to the nearest and farthest sampling points in an n-dimensional manifold and show that the Relative Contrast of the sample data vanishes as dimensionality of the underlying vector space tends to infinity."},"documents":{"kind":"list like","value":["Title: Automatic Synthesis And Deployment Of Intensional Kahn Process Networks\nYear: 2009\n\nAbstract: In this paper we introduce and study. theoretically, a clean slate \"formal\" foundational approach for developing and deploying high-assurance distributed embedded systems deployed in mission-critical applications We propose a simple formal distributed asynchronous framework extending Kahn Process Networks with intensional specification. More precisely. we present a model-driven approach based on a platform-independent language and an intensional specification logic that allows us to synthesize distributed agents that can handle interactions with external resources asynchronously. ensure enforcement of information flow and security policies. and have the ability to deal with failures of resources Our approach allows rapid development and automated deployment of formally verified embedded networked systems that provide guarantees that clients' requirements will be met and QoS guarantees will be respected Moreover, It allows modeling (and programming) tellable distributed systems for multi-core hosts Such a capability makes our framework suitable for next generation grid computing systems whew multi-core individual hosts need to be utilized for improving scalability Given an intensional logical specification of a distributed embedded system. that includes Quality of Service (QoS) requirements. a set of software resources and devices available in a network. and then formal interface specifications, a deductive system can automatically generate distributed extended Kahn processes and their deployment information in such a way that the application requirements-including QoS requirements are guaranteed to be met The generated processes use the inputs of the sensors/meters/probes and the management policies of the customer to generate real-time control decisions for manning the system The processes are deployed automatically on a distributed network involving sensors/meters/probes tracking system parameters, actuators controlling devices. and diverse computing and communication elements such as MA's. etc","Title: Counterfactually reasoning about security\nYear: 2011\n\nAbstract: In this short paper, we provide the background to counterfactual logic and give very general suggestions on how we could employ this logic to help us reason about security policies. It seems very appropriate to use this kind of logic to anticipate a change that will compromise the security concerns of a given system before actually applying the changes.","Title: Dynamic Message Sequence Charts\nYear: 2002\n\nAbstract: We introduce a formalism to specify classes of MSCs over an unbounded number of processes. The formalism can describe many interesting behaviours of dynamically changing networks of processes. Moreover, it strictly includes the formalism of Message Sequence Graphs studied in the literature to describe MSCs over a fixed finite set of processes. Our main result is that model-checking of MSCs described in this formalism against a suitable monadic-second order logic is decidable.","Title: Model checking mobile ambients\nYear: 2003\n\nAbstract: We settle the complexity bounds of the model checking problem for the ambient calculus with public names against the ambient logic. We show that if either the calculus contains replication or the logic contains the guarantee operator, the problem is undecidable. In the case of the replication-free calculus and guarantee-free logic we prove that the problem is PSPACE-complete. For the complexity upper bound, we devise a new representation of processes that remains of polynomial size during process execution; this allows us to keep the model checking procedure in polynomial space. Moreover, we prove PSPACE-hardness of the problem for several quite simple fragments of the calculus and the logic; this suggests that there are no interesting fragments with polynomial-time model checking algorithms.","Title: Model Checking Communication Protocols\nYear: 2001\n\nAbstract: Brand and Zafiropulo [1] introduced the model of communicating finite-state machines to represent a distributed system connected with FIFO channels. Several different communication protocols can be specified with this simple model. In this paper we address the problem of automatically validating protocols by verifying properties such as well-formedness and absence of deadlock. Our method is based on a representation of communicating finite-state machines in terms of logic programs. This leads to efficient verification algorithms based on the ground and non-ground semantics of logic programming.","Title: An Algebraic Framework for Abstract Model Checking\nYear: 2002\n\nAbstract: Symbolic forward analysis is a semi-algorithm that in many cases solves the model checking problem for infinite state systems in practice. This semi-algorithm is implemented in manypractical model checking tools like UPPAAL [BLL+96], KRONOS [DT98] and HYTECH [HHWT97]. In most practical experiments, termination of symbolic forward analysis is achieved by employing abstractions resulting in an abstract symbolic forward analysis. This paper presents a unified algebraic framework for deriving and reasoning about abstract symbolic forward analysis procedures for a large class of infinite state systems with variables ranging over a numeric domain. The framework is obtained by lifting notions from classical algebraic theoryof automata to constraints representing sets of states. Our framework provides sufficient conditions under which the derived abstract symbolic forward analysis procedure is always terminating or accurate or both. The class of infinite state systems that we consider here are (possibly non-linear) hybrid systems and (possiblynon-linear) integer-valued systems. The central notions involved are those of constraint transformer monoids and coverings between constraint transformer monoids. We show concrete applications of our framework in deriving abstract symbolic forward analysis algorithms for timed automata and the two process bakeryalgorithm that are both terminating and accurate.","Title: Toward More Precise Radiotherapy Treatment of Lung Tumors\nYear: 2012\n\nAbstract: A computational framework for modeling the respiratory motion of lung tumors provides a 4D parametric representation that tracks, analyzes, and models movement to provide more accurate guidance in the planning and delivery of lung tumor radiotherapy.","Title: A Risk Reduction Framework for Dynamic Workflows\nYear: 2008\n\nAbstract: Workflows tend to fail in real-world scenarios due to the uncertain/unreliable sensory information which sometimes needs to be updated during the execution of workflows. In a logic based framework, these dynamic predicates that can be updated are called non-monotonic predicates (NMPs). In this paper, we focus on reducing the risk of a given workflow due to the NMPs in that workflow. The main idea is to synthesize a backup workflow by augmenting the main workflow without introducing new NMPs. The backup workflow is generated by using expected values of NMPs if necessary instead of given values. The expected values are calculated from the execution history or provided by a domain expert. It is argued that total risk reduces to the square root of the main workflow itself.","Title: RETNA: From Requirements to Testing in a Natural Way\nYear: 2004\n\nAbstract: Most problems in building and refining a system can be traced back to errors in requirements. Poorly organized requirements, most often in natural language are among the major causes of failures of software projects. In this paper, we present a requirements analysis tool called RETNA and the technology behind it. RETNA accepts natural language requirements, classifies them, interacts with the user to refine them, automatically translates natural language requirements to a logical format so that they can be validated and finally generates test cases from the requirements.","Title: An ad-hoc distributed execution environment for multi-agent systems\nYear: 2013\n\nAbstract: We present a distributed execution environment for multi-agent systems running in an intermittently-connected network. We describe the design and implementation of a microkernel-based prototype operating system layer that provides services for managing agent execution. The execution environment provides the user with a consistent view of a single machine, a single file system, and a unified programming model.","Title: Constraint Database Models Characterizing Timed Bisimilarity\nYear: 2001\n\nAbstract: The problem of deciding timed bisimilarity has received increasing attention; it is important for verification of timed systems. Using a characterization of timed bisimilarity in terms of models of constraint databases, we present to our knowledge, the first local, symbolic algorithm for deciding timed bisimilarity; previous algorithms were based on a finite, but prohibitively large, abstraction (the region graph or the full backward stable graph). Our algorithm uses XSB-style tabling with constraints. Our methodology is more general than those followed in the previous approaches in the sense that our algorithm can be used to decide whether two timed systems are alternating timed bisimilar.","Title: Code-Change Impact Analysis Using Counterfactuals\nYear: 2011\n\nAbstract: In this paper we present a framework for what-if analysis of programs based on Lewis' theory of counterfactuals. The framework can be used to statically perform change impact analysis for source code. It enables us to verify assertions about a changed version of the program without actually incorporating the changes. We present a logical calculus that precisely characterizes structural modifications to source code and their impact on the behavior of the program.","Title: Model Checking for Timed Logic Processes\nYear: 2000\n\nAbstract: We apply techniques from logic programming and constraint databases to verify real time systems. We introduce timed logic processes (TLPs) as a fragment of constraint query languages over reals. We establish a formal connection between TLPs and timed automata, and between the procedure of the UPPAAL model checker for restricted temporallogic properties of timed automata and the top-down query evaluation of TLPs (with tabling in the XSB style). This connection yields an alternative implementation of the UPPAAL procedure. Furthermore, we can extend that procedure in order to accommodate more expressive properties.","Title: Optimal-Reachability and Control for Acyclic Weighted Timed Automata\nYear: 2002\n\nAbstract: Abstract: Weighted timed automata extend timed automata with costs on bothlocations and transitions. In this framework we study the optimal reachabilityand the optimal control synthesis problems for the automata withacyclic control graphs. This class of automata is relevant for some practicalproblems such as some static scheduling problems or air-traccontrol problems. We give a nondeterministic polynomial time algorithmto solve the decision version of the considered optimal reachabilityproblem....","Title: New Modified Turbo Decoder for Embedded Local Content in Single-Frequency Networks.\nYear: 2013\n\nAbstract: Current DVB-SH systems rely on hierarchical modulation to insert local content into the complementary ground component. The local content is inserted as a low priority bit stream, while the global content will be transmitted as a high-priority bit stream. At the receiver, the standard approach of extracting the local content from the received signal first requires detection of the global content. Subsequently, one can extract the local content by subtracting the coded and modulated form (restored version) of the detected global content from the received signal. In this paper, we propose a different approach to remove the global content from the received signal. In our new approach, a hybrid signal is designed to represent the turbo-coded global content, which contains decoded bits from the turbo decoder representing systematic bits, and detected bits from the received signal prior to the turbo decoder representing parity bits. This technique would relieve the receiver from performing coding and modulation on the detected global content so as to ensure a less computational burden and a smaller latency than the current technology. The obvious tradeoff between complexity and performance would imply higher bit error rates arising from our proposed system. Therefore, to mitigate the bit-error-rate performance degradation, we propose modifying the turbo decoder so that the systematic bit streams and the parity bit streams will be decoded differently with respect to their corresponding signal-to-noise ratios. Monte Carlo simulation results demonstrate that our proposed new scheme can achieve better computational efficiency and less latency in the slight sacrifice of error performance.","Title: Functional \"AJAX\" in Secure Synchronous Programming\nYear: 2011\n\nAbstract: AJAX (Asynchronous Javascript and XML) is a confederation of technologies aimed at providing improved user interaction with web-based applications. While AJAX provides an improved user experience, it also comes with its baggage of problems. The lack of formal semantics makes AJAX applications difficult to build, debug, understand, and validate. Different component technologies of AJAX (e.g., XMLHttpRequest or Javascript) are browser-sensitive and have different implementations and provide distinct functionalities. Source code is downloaded and run on the clients machines, raising security concerns. In this paper, we present an ``AJAX''-like framework in an event-driven secure synchronous programming environment. Our framework is supported by a formal operational semantics. Applications written in our language can be verified using formal static analysis techniques such as theorem proving. The applications are compiled and run on the SINS (Secure Infrastructure for Networked Systems) infrastructure jointly developed in collaboration with the Naval Research Laboratory.","Title: Towards Formal Verification of a Commercial Wireless Router Firmware\nYear: 2013\n\nAbstract: Formal verification of the trusted computing base of a software system is essential for its deployment in mission-critical environments. Commercial off-the-shelf routers are nowadays being used for managing traffic in high-assurance networks. The specifications for the development of these routers are provided by RFCs that are only described informally in English. It is essential to ensure that a router firmware conforms to its corresponding RFC before it can be deployed for managing mission-critical networks. In this paper, we report the formal verification of the conformance of the open source Netgear WNR3500L wireless router firmware implementation to the RFC 2131 [6] based on which it is designed. The formal verification effort led to the discovery of several possible problems in the implementation that we report in this paper. We have used the Coq proof assistant extensively in this verification effort. The formal verification process demonstrates the usefulness of inductive types and higher-order logic in software certification.","Title: Model-Based Static Source Code Analysis of Java Programs with Applications to Android Security\nYear: 2012\n\nAbstract: We combine static analysis techniques with model- based deductive verification using SMT solvers to provide a framework that, given an analysis aspect of the source code, automatically generates an analyzer capable of inferring information about that aspect. The analyzer is generated by translating the collecting semantics of a program to a \"marked\" formula in first order logic over multiple underlying theories. The \"marking\" can be thought of as a set of holes or contexts corresponding to the \"uninterpreted\" APIs invoked in the program. Just as a program imports packages and uses methods from classes in those packages, we import the semantics of the API invocations as first order logic assertions. These assertions constitute the models used by the analyzer. Logical specification of the desired program behavior (rather its negation) is incorporated as a first order logic formula. An SMT-LIB formula solver treats the combined formula as a \"constraint\" and \"solves\" it. The \"solved form\" can be used to identify logical (security) errors in Java (Android) programs. Security properties of Android are represented as constraints and the analysis aims to show that these constraints are respected.","Title: Preventing Future Oil Spills with Software-Based Event Detection\nYear: 2010\n\nAbstract: Complex event processing systems detect problems in mission-critical, real-time applications and generate intelligent decisions to modulate the system environment.","Title: Deterministic finite automata with recursive calls and DPDAs\nYear: 2003\n\nAbstract: We study deterministic finite automata (DFA) with recursive calls, that is, finite sequences of component DFAs that can call each other recursively. DFAs with recursive calls are akin to recursive state machines and unrestricted hierarchic state machines. We show that they are language equivalent to deterministic pushdown automata (DPDA).","Title: Beyond Region Graphs: Symbolic Forward Analysis of Timed Automata\nYear: 1999\n\nAbstract: Theoretical investigations of infinite-state systems have so far concentrated on decidability results; in the case of timed automata these results are based on region graphs. We investigate the specific procedure that is used practically in order to decide verification problems, namely symbolic forward analysis. This procedure is possibly nonterminating. We present basic concepts and properties that are useful for reasoning about sufficient termination conditions, and then derive some conditions. The central notions here are constraint transformers associated with sequences of automaton edges and zone trees labeled with successor constraints.","Title: Approximation of CFL by Regular Languages for Concurrent Program Verification\nYear: 2010\n\nAbstract: Many problems related to verification of concurrent programs can be reduced to the non-empty intersection problem for context-free languages. Since the latter is an undecidable problem, a practical approach for solving the intersection problem is to convert it to a decidable problem of the non-empty intersection of a context-free language and a regular language. This is done by approximating one of the context-free languages in the intersection from above or from below by a regular language. We give an approximation technique from above by modeling a context-free language L in terms of linear integer inequalities and then obtaining the approximating regular language L' ⊇ L by relaxing the linear inequalities such that each inequality involves at most one variable. Previous approximation techniques focused on approximation below. We also give an alternate technique with a finite-state automata based approach, where we start with an automata M which accepts a suitable finite-subset L0 of L, and then extend M successively based on the pumping property of L till the language accepted by M contains L.","Title: Managing a Cloud for Multi-agent Systems on Ad-Hoc Networks\nYear: 2012\n\nAbstract: We present a novel execution environment for multi-agent systems building on concepts from cloud computing and peer-to-peer networks. The novel environment can provide the computing power of a cloud for multi-agent systems in intermittently connected networks. We present the design and implementation of a prototype operating system for managing the environment. The operating system provides the user with a consistent view of a single machine, a single file system, and a unified programming model while providing elasticity and availability.","Title: Accurate Widenings and Boundedness Properties of Timed Systems\nYear: 2001\n\nAbstract: We propose a symbolic model checking procedure for timed systems that is based on operations on constraints. To accelerate the termination of the model checking procedure, we define history-dependent widening operators, again in terms of constraint-based operations. We show that these widenings are accurate, i.e., they don't lose precision even with respect to the test of boundedness properties.","Title: Automated Situation-Aware Service Composition In Service-Oriented Computing\nYear: 2007\n\nAbstract: Service-based systems have many applications, such as e-business, health care, and homeland security. In these systems, it is necessary to provide users the capability of composing services into workflows providing higher-level functionality. In dynamic service-oriented computing environments, it is desirable that service composition is automated and situation-aware to generate robust and adaptive workflows. In this paper, an automated situation-aware service composition approach is presented This approach is based on the a-logic, a-calculus, and a declarative model for situation awareness (SAW). This approach consists of four major components: (1) analyzing SAW requirements using our SAW model, (2) translating our SAW model representation to a-logic specifications and specifying a control flow graph in a-logic as the service composition goal, (3) automated synthesis of a-calculus terms defining situation-aware workflow agents based on a-logic specifications for SA W requirements and the control flow graph, and (4) compilation of a-calculus terms to executable components.","Title: Code-Change Impact Analysis Using Counterfactuals: Theory And Implementation\nYear: 2013\n\nAbstract: This article shows a novel program analysis framework based on Lewis' theory of counterfactuals. Using this framework we are capable of performing change-impact static analysis on a program's source code. In other words, we are able to prove the properties induced by changes to a given program before applying these changes. Our contribution is two-fold; we show how to use Lewis' logic of counterfactuals to prove that proposed changes to a program preserve its correctness. We report the development of an automated tool based on resolution and theorem proving for performing code change-impact analysis.","Title: Service Composition in a Secure Agent-Based Architecture\nYear: 2005\n\nAbstract: We describe an agent-based situation-aware survivable architecture for the discovery and composition of web services.Our architecture provides for proofs that guarantee the consistency of service level agreements between service providers and their clients.The architecture is based on decentralized situation-aware ambients (SAAs) which are autonomous agents that process information about the current situation, and discover and compose services syntactically to adapt to changing situations. The agents run on the Secure Infrastructure for Networked Systems (SINS) under development at the Center for High Assurance Computer Systems of the Naval Research Laboratory.","Title: A Type-Theoretic Framework for Automated Synthesis of Workflows in Service-Oriented Computing\nYear: 2013\n\nAbstract: We describe a technique for automatic synthesis of workflows in service-oriented systems from a typetheoretic specification of coordination requirements. The technique presented involves formal semantics-based, automatic synthesis of service coordination workflows from the business logic of the clients using type and effect inference. The type and effect deduction, apart from synthesizing the workflow, provides formal proofs that the business requirements of the clients in a service-oriented system are met. The service descriptions and the business requirements are specified as a set of type judgements in an (intuitionistic) type theory. Workflows can be generated from a (type and effect) inference of the requirements from the \"service theory\".","Title: Novel Fast Blind Channel Estimation and Hessian Analysis for Transmitter Identification of Digital Television Signals\nYear: 2014\n\nAbstract: The transmitter identification (Tx-ID) technique has been widely addressed in the modern advanced television systems committee digital television (DTV) standards. Kasami sequences are adopted as the Tx-ID sequences due to the favorable code capacity. However, the conventional cross-correlation based Tx-ID technique cannot combat the realistic multipath channel impairment. Therefore, in this paper, we propose a new least-square Tx-ID approach, which is demonstrated to be very robust when the multipath channels are encountered. We also establish the new Hessian analysis to provide the range for appropriate injection levels accordingly. Based on this new theoretical study, we design new efficient (fast) algorithms to determine the maximum allowable channel length given a particular Kasami Tx-ID sequence. These algorithms and theoretical results can be very useful for the future DTV technologies.","Title: LDPC encoder identification in time-varying flat-fading channels\nYear: 2014\n\nAbstract: This paper tackles the low-density parity-check (LDPC) encoder identification problem encountered in the time-varying flat-fading channels which are modeled as finite-state Markov chains. The in-phase and quadrature-phase components of the channel coefficients are both represented by a number of states. To greatly simplify the computation of the channel observation probabilities, each channel-state region is further quantized to an interior point. Based on our proposed finite-state Markov model, the Viterbi algorithm is thus invoked to blindly estimate the unknown channel-state sequence from each received signal segment. To mitigate the phase ambiguity which is inherent in the channel-state estimation process, the pilot-aided channel estimation method is also proposed here. The LDPC encoder is finally identified in the framework of the log-likelihood ratio of syndrome a posteriori probability. The performance of our proposed LDPC identification scheme is investigated for different normalized Doppler rates and different mechanisms to reconstruct the channel-state information. Monte Carlo simulation results suggest that pilot symbols are necessary for leading to a satisfactory identification performance for time-varying flat-fading channels.","Title: Stability of video rate control algorithms over bandwidth-limited network paths\nYear: 2014\n\nAbstract: The paper considers the stability of AIMD-based rate adaptation algorithm for video transmission over a bandwidth-limited network path. We examine the stability aspects from two angles: first, a quick relief from congestion by a faster reduction of send rate, and second, a larger limit-cycle time with a slow rate reduction in the steady-state (and hence lower rate jitter). The stability assessment also considers the loss of feedback signals from the video receiver that indicate the packet loss on data channels. We provide a management-oriented assessment of the convergence properties of video rate control system. This in turn allows an autonomic adjustment of the AIMD parameters and the system controller models for optimal performance.","Title: Managing shared contexts in distributed multi-player game systems\nYear: 2014\n\nAbstract: In this paper, we consider the impact of a weaker model of eventual consistency on distributed multi-player games. This model is suitable for networks in which hosts can leave and join at anytime, e.g., in an intermittently connected environment. Such a consistency model is provided by the Secure Infrastructure for Networked Systems (SINS) [24], a reliable middleware framework. SINS allows agents to communicate asynchronously through a distributed transactional key-value store using anonymous publish-subscribe. It uses Lamport's Paxos protocol [17] to replicate state. We consider a multi-player maze game as example to illustrate our consistency model and the impact of network losses/delays therein. The framework based on SINS presented herein provides a vehicle for studying the effect of human elements participating in collaborative simulation of a physical world as in war games.","Title: Does your result checker really check?\nYear: 2004\n\nAbstract: A result checker is a program that checks the output of the computation of the observed program for correctness. Introduced originally by Blum, the result-checking paradigm has provided a powerful platform assuring the reliability of software. However, constructing result checkers for most problems requires not only significant domain knowledge but also ingenuity and can be error prone. In this paper we present our experience in validating result checkers using formal methods. We have conducted several case studies in validating result checkers from the commercial LEDA system for combinatorial and geometric computing. In one of our case studies, we detected a logical error in a result checker for a program computing max flow of a graph.","Title: SOLj: A Domain-Specific Language (DSL) for Secure Service-Based Systems\nYear: 2007\n\nAbstract: We present SOLj (Secure Operations Language-Java), an event-driven domain-specific synchronous programming extension of Java for developing secure service-based systems. The language has capabilities for handling service invocations asynchronously, includes strong typing for the enforcement of information flow and security policies, and exception handling mechanisms to deal with failures of components or services (both benign and Byzantine). Applications written in SOLj are formally verifiable using static analysis techniques. SOLj programs may be deployed, con- figured, and run on SINS (Secure Infrastructure for Networked Systems) under development at the Naval Research Laboratory.","Title: A Formal Approach to Developing Reliable Event-Driven Service-Oriented Systems\nYear: 2008\n\nAbstract: In this paper, we present a formal framework for developing distributed service-oriented systems in an event-driven secure synchronous programming environment. More precisely, we present a synchronous programming language called SOL (Secure Operations Language) that has (i) capabilities for handling service invocations asynchronously, (ii) strong typing to ensure enforcement of information flow and security policies, and (iii) the ability to deal with failures (both benign and byzantine) of components. SOL is supported by formal operational semantics. Applications written in our framework can be verified using formal static checking techniques like theorem proving. The framework runs on the top of the SINS (secure infrastructure for networked systems) infrastructure that we have developed.","Title: Cognitive Distributed Networks in Environmental e-Science\nYear: 2008\n\nAbstract: We integrate fundamental principles from formal mathematics, engineering, and information sciences to develop novel distributed sensor-based networks for management of the environment and the natural ecosystem. More precisely, we are developing ENVNET – ENVironmental sensor NETwork, a cognitive wireless sensor network to proactively collect information, monitor, and control the health of our natural environment and ecosystem (that includes, water, soil, air, etc.). It integrates intelligent sensor coordination and data fusion techniques to access, retrieve, process, and communicate with disparate wireless sensors in an ad-hoc manner to deliver dynamic decisions and provide adequate information management. Our approach facilitates large scale deployment and drastically reduces the hardware cost almost by a factor of 10 and removes the main bottleneck in irrigation control arising from wired sensors. Apart from this it provides a smart distributed closed-loop control mechanism with formal reliability guarantees that is rapidly and easily deployable, context-aware, scalable, and dynamically reconfigurable in response to changing requirements.","Title: A Semiautomated Probabilistic Framework for Tree-Cover Delineation From 1-m NAIP Imagery Using a High-Performance Computing Architecture\nYear: 2015\n\nAbstract: Accurate tree-cover estimates are useful in deriving above-ground biomass density estimates from very high resolution (VHR) satellite imagery data. Numerous algorithms have been designed to perform tree-cover delineation in high-to-coarse-resolution satellite imagery, but most of them do not scale to terabytes of data, typical in these VHR data sets. In this paper, we present an automated probabilistic framework for the segmentation and classification of 1-m VHR data as obtained from the National Agriculture Imagery Program (NAIP) for deriving tree-cover estimates for the whole of Continental United States, using a high-performance computing architecture. The results from the classification and segmentation algorithms are then consolidated into a structured prediction framework using a discriminative undirected probabilistic graphical model based on conditional random field, which helps in capturing the higher order contextual dependence relations between neighboring pixels. Once the final probability maps are generated, the framework is updated and retrained by incorporating expert knowledge through the relabeling of misclassified image patches. This leads to a significant improvement in the true positive rates and reduction in false positive rates (FPRs). The tree-cover maps were generated for the state of California, which covers a total of 11 095 NAIP tiles and spans a total geographical area of 163 696 sq. miles. Our framework produced correct detection rates of around 88% for fragmented forests and 74% for urban tree-cover areas, with FPRs lower than 2% for both regions. Comparative studies with the National Land-Cover Data algorithm and the LiDAR high-resolution canopy height model showed the effectiveness of our algorithm for generating accurate high-resolution tree-cover maps.","Title: A graph-based approach to construct target-focused libraries for virtual screening.\nYear: 2016\n\nAbstract: eSynth can successfully reconstruct chemically feasible molecules from molecular fragments. Furthermore, in a procedure mimicking the real application, where one expects to discover novel compounds based on a small set of already developed bioactives, eSynth is capable of generating diverse collections of molecules with the desired activity profiles. Thus, we are very optimistic that our effort will contribute to targeted drug discovery. eSynth is freely available to the academic community at www.brylinski.org/content/molecular-synthesis.Graphical abstractAssuming that organic compounds are composed of sets of rigid fragments connected by flexible linkers, a molecule can be decomposed into its building blocks tracking their atomic connectivity. Here, we developed eSynth, an automated method to synthesize new compounds by reconnecting these building blocks following the connectivity patterns via an exhaustive graph-based search algorithm. eSynth opens up a possibility to rapidly construct virtual screening libraries for targeted drug discovery.","Title: Reasoning about security in sensor networks\nYear: 2015\n\nAbstract: We present a formal framework for reasoning about security concerns in the context of embedded sensor networks. We first provide an agent-based programming model for sensor networks. A logical framework enables reasoning about security, safety, and integrity with respect to usage of resources in this model. Embedded sensor networks often operate in rapidly changing mission-critical environments where both functional and nonfunctional requirements can alter dynamically in an unforeseen manner. The network may need to be reconfigured and reprogrammed in response to changes in its operating conditions. We provide a framework based on counterfactual logic to formally represent changes to the system and perform what-if reasoning about their impact on security and safety even before they have been applied. Copyright © 2015 John Wiley & Sons, Ltd.","Title: OptEx: A Deadline-Aware Cost Optimization Model for Spark\nYear: 2016\n\nAbstract: We present OptEx, a closed-form model of job execution on Apache Spark, a popular parallel processing engine. To the best of our knowledge, OptEx is the first work that analytically models job completion time on Spark. The model can be used to estimate the completion time of a given Spark job on a cloud, with respect to the size of the input dataset, the number of iterations, the number of nodes comprising the underlying cluster. Experimental results demonstrate that OptEx yields a mean relative error of 6% in estimating the job completion time. Furthermore, the model can be applied for estimating the cost optimal cluster composition for running a given Spark job on a cloud under a completion deadline specified in the SLO (i.e.,Service Level Objective). We show experimentally that OptEx is able to correctly estimate the cost optimal cluster composition for running a given Spark job under an SLO deadline with an accuracy of 98%.","Title: OptCon: An Adaptable SLA-Aware Consistency Tuning Framework for Quorum-Based Stores\nYear: 2016\n\nAbstract: Users of distributed datastores that employquorum-based replication are burdened with the choice of asuitable client-centric consistency setting for each storage operation. The above matching choice is difficult to reason about asit requires deliberating about the tradeoff between the latencyand staleness, i.e., how stale (old) the result is. The latencyand staleness for a given operation depend on the client-centricconsistency setting applied, as well as dynamic parameters such asthe current workload and network condition. We present OptCon, a novel machine learning-based predictive framework, that canautomate the choice of client-centric consistency setting underuser-specified latency and staleness thresholds given in the servicelevel agreement (SLA). Under a given SLA, OptCon predictsa client-centric consistency setting that is matching, i.e., it isweak enough to satisfy the latency threshold, while being strongenough to satisfy the staleness threshold. While manually tunedconsistency settings remain fixed unless explicitly reconfigured, OptCon tunes consistency settings on a per-operation basis withrespect to changing workload and network state. Using decisiontree learning, OptCon yields 0.14 cross validation error in predictingmatching consistency settings under latency and stalenessthresholds given in the SLA. We demonstrate experimentally thatOptCon is at least as effective as any manually chosen consistencysettings in adapting to the SLA thresholds for different usecases. We also demonstrate that OptCon adapts to variationsin workload, whereas a given manually chosen fixed consistencysetting satisfies the SLA only for a characteristic workload.","Title: Mutual Visibility with an Optimal Number of Colors.\nYear: 2015\n\nAbstract: We consider the following fundamental Mutual Visibility problem: Given a set of n identical autonomous point robots in arbitrary distinct positions in the Euclidean plane, find a schedule to move them such that within finite time they reach, without collisions, a configuration in which they all see each other. The robots operate following Look-Compute-Move cycles and a robot $$r_i$$ can not see other robot $$r_j$$ if there lies a third robot $$r_l$$ in the line segment connecting the positions of $$r_i$$ and $$r_j$$. Moreover, n is not assumed to be known to the robots. We study this problem in the robots with lights model, where each robot has an externally visible persistent light which can assume colors from a fixed set of colors and the color set is identical to all the robots. This model corresponds to the classical model of oblivious robots when the number of colors $$c=1$$ in the color set. Therefore, we focus here on the objective of minimizing the number of colors required to successfully solve Mutual Visibility. Di Luna et al. [16] presented two algorithms and showed that Mutual Visibility can always be solved without collisions with $$c=3$$ colors for both semi-synchronous and asynchronous robots under both rigid and non-rigid moves. In this paper, we present and analyze an improved algorithm which requires only $$c=2$$ colors and works for both semi-synchronous and asynchronous robots under both rigid and non-rigid moves; this is optimal since any algorithm for Mutual Visibility needs at least 2 colors in the robots with lights model, when n is not known. We employ a non-trivial technique which moves all the interior robots first to the boundary of the initial convex hull and then the robots in the boundary of the hull except the corners to outside of the hull until all the robots eventually become corners, without the need of any third color. Our result is interesting in the sense that asynchronicity and non-rigidity of robot movements have no effect on Mutual Visibility with respect to the number of colors needed to successfully solve it. We also provide an improved solution to the Circle Formation problem.","Title: DeepSat - A Learning framework for Satellite Imagery\nYear: 2015\n\nAbstract: Satellite image classification is a challenging problem that lies at the crossroads of remote sensing, computer vision, and machine learning. Due to the high variability inherent in satellite data, most of the current object classification approaches are not suitable for handling satellite datasets. The progress of satellite image analytics has also been inhibited by the lack of a single labeled high-resolution dataset with multiple class labels. The contributions of this paper are twofold -- (1) first, we present two new satellite datasets called SAT-4 and SAT-6, and (2) then, we propose a classification framework that extracts features from an input image, normalizes them and feeds the normalized feature vectors to a Deep Belief Network for classification. On the SAT-4 dataset, our best network produces a classification accuracy of 97.95% and outperforms three state-of-the-art object recognition algorithms, namely - Deep Belief Networks, Convolutional Neural Networks and Stacked Denoising Autoencoders by ~11%. On SAT-6, it produces a classification accuracy of 93.9% and outperforms the other algorithms by ~15%. Comparative studies with a Random Forest classifier show the advantage of an unsupervised learning approach over traditional supervised learning techniques. A statistical analysis based on Distribution Separability Criterion and Intrinsic Dimensionality Estimation substantiates the effectiveness of our approach in learning better representations for satellite imagery.","Title: The Causal Order is Strict\nYear: 2008\n\nAbstract: The causal order is central to many algorithms in both the message-passing and shared memory models. In the message-passing model, the causal order is clearly a strict, or irreflexive partial, order. However, some shared memory models are defined without reference to time. In those models, the causal order can fail to be a strict order. Existing works have either declared that all non-strict orders are not causal, or assumed that the causal order is strict without proof. We prove that, under a small number of reasonable assumptions about systems, the causal order is strict. In particular, we assume neither a global time model nor that processes issue a single shared memory operation at a time.","Title: Bounds on Mutual Visibility Algorithms.\nYear: 2015\n\nAbstract: We consider the fundamental Mutual Visibility problem for a set of n identical autonomous point robots (n is not known to the robots) that operate following Look-Compute-Move cycles starting from arbitrary distinct positions in the Euclidean plane under obstructed visibility { a robot ri can see robot rj;rj 6 ri; if and only if there is no other robot in the line segment joining their positions. The objective is to determine a schedule to reposition these robots without collisions such that they reach in nite time a conguration where they all see each other. In the recently proposed so-called robots with lights model, Di Luna et al. [15] gave two deterministic algorithms Contain and Shrink for this problem; however, no runtime bounds were given except the proof that they terminate in nite time. In this paper, we rst study the runtime bounds of these algorithms in the fully synchronous setting showing that Contain is tight (( n) rounds) and Shrink needs ( n 2 ) rounds in the worst-case. We then present a new deterministic algorithm, called Modied Shrink, for fully synchronous setting that solves this problem in O(n logn) rounds, improving signicantly over Shrink. We also show that Modied Shrink has the lower bound of ( n) rounds.","Title: MAPTrack - A Probabilistic Real Time Tracking Framework by Integrating Motion, Appearance and Position Models.\nYear: 2015\n\nAbstract: In this paper, we present MAPTrack a robust tracking framework that uses a probabilistic scheme to combine a motion model of an object with that of its appearance and an estimation of its position. The motion of the object is modelled using the Gaussian Mixture Background Subtraction algorithm, the appearance of the tracked object is enumerated using a color histogram and the projected location of the tracked object in the image space/frame sequence is computed by applying a Gaussian to the Region of Interest. Our tracking framework is robust to abrupt changes in lighting conditions, can follow an object through occlusions, and can simultaneously track multiple moving foreground objects of different types (e.g., vehicles, human, etc.) even when they are closely spaced. It is able to start tracks automatically based on a spatio-temporal filtering algorithm. A “dynamic” integration of the framework with optical flow allows us to track videos resulting from significant camera motion. A C++ implementation of the framework has outperformed existing visual tracking algorithms on most videos in the Video Image Retrieval and Analysis Tool (VIRAT), TUD, and the Tracking-Learning-Detection (TLD) datasets.","Title: An Agile Framework for Real-Time Motion Tracking\nYear: 2015\n\nAbstract: We present an agile framework for automated tracking of moving objects in full motion video (FMV). The framework is robust, being able to track multiple foreground objects of different types (e.g., Person, vehicle) having disparate motion characteristics (like speed, uniformity) simultaneously in real time under changing lighting conditions, background, and disparate dynamics of the camera. It is able to start tracks automatically based on a confidence-based spatio-temporal filtering algorithm and is able to follow objects through occlusions. Unlike existing tracking algorithms, with high likelihood, it does not lose or switch tracks while following multiple similar closely-spaced objects. The framework is based on an ensemble of tracking algorithms that are switched automatically for optimal performance based on a performance measure without losing state. Only one of the algorithms, that has the best performance in a particular state is active at any time providing computational advantages over existing ensemble frameworks like boosting. A C++ implementation of the framework has outperformed existing visual tracking algorithms on most videos in the Video Image Retrieval and Analysis Tool (VIRAT: www.viratdata.org) and the Tracking-Learning-Detection data-sets.","Title: Tight analysis of a collisionless robot gathering algorithm\nYear: 2015\n\nAbstract: We consider the fundamental problem of gathering a set of n robots in the Euclidean plane that have a physical extent and hence cannot share their positions with other robots. The objective is to determine a minimum time schedule to gather the robots as close together as possible around a predefined gathering point avoiding collisions. This problem with minimum time objective has applications in many real-world scenarios including fast autonomous coverage formation. Cord-Landwehr et al. (in Proceedings of the International Conference on Current Trends in Theory and Practice of Computer Science, 2011) gave a local greedy algorithm in a fully synchronous setting and proved that, for the discrete version of the problem where robots’ movements are restricted to the positions on an integral grid, their algorithm solves this problem in O(nR) rounds, where R is the distance from the farthest initial robot position to the gathering point. In this article, we improve significantly the round complexity of their algorithm to R + 2 · (n − 1) rounds. This round complexity is obtained in the following modified model: (1) the viewing range of the robots is increased to three hops and (2) robots can additionally move to the diagonally opposite corner to a grid cell in one step—that is, they can traverse the two corresponding grid edges in one time step. We also prove that there are initial configurations of n robots in this problem where at least R+&frac;(n-1)/2 rounds are needed by any local greedy algorithm. Furthermore, we improve the lower bound to R + (n − 1) rounds for the algorithm of Cord-Landwehr et al. These results altogether provide a tight runtime analysis of their algorithm.","Title: Infra: SLO Aware Elastic Auto-scaling in the Cloud for Cost Reduction.\nYear: 2016\n\nAbstract: Enterprises often host applications and services on clusters of virtual machine instances provided by cloud service providers, like Amazon, Rackspace, Microsoft, etc. Users pay a cloud usage cost on the basis of the hourly usage [1] of virtual machine instances composing the cluster. A cluster composition refers to the number of virtual machine instances of each type (from a predefined list of types) comprising a cluster. We present Infra, a cloud provisioning framework that can predict an (a#x03B5; a#x03B4;)-minimum cluster composition required to run a given application workload on a cloud under an SLO (i.e., Service Level Objective) deadline. This paper does not present a new approximation algorithm, instead we provide a tool that applies existing machine learning techniques to predict an (a#x03B5; a#x03B4;)-minimum cluster composition. An (a#x03B5; a#x03B4;)-minimum cluster composition specifies a cluster composition whose cost approximates that of the minimum cluster composition (i.e., the cluster composition that incurs the minimum cloud usage cost that must be incurred in executing a given application under an SLO deadline), the approximation bounds the error to a predefined threshold a#x03B5; with a degree of confidence 100 * (1 - a#x03B4;)%. The degree of confidence 100 * (1 - a#x03B4;)% specifies that the probability of failure in achieving the error threshold a#x03B5; for the above approximation is at most a#x03B4;. For a#x03B5; = 0.1 and a#x03B4; = 0.02, we experimentally demonstrate that an (a#x03B5;a#x03B4;)-minimum cluster composition predicted by Infra successfully approximates the minimum cluster composition, i.e., the accuracy of prediction of minimum cluster composition ranges from 93.1% to 97.99% (the error is bound by the error threshold of 0.1) with a 98% degree of confidence, since 100* (1 - a#x03B4;) = 98%. Auto scaling refers to the process of automatically adding cloud instances to a cluster to adapt to an increase in application workload (increased request rate), and deleting instances from a cluster when there is a decrease in workload (reduced request rate). However, state-of-the-art auto scaling techniques have the following disadvantages: A) they require explicit policy definition for changing the cluster configuration and therefore lack the ability to automatically adapt a cluster with respect to changing workload, B) they do not compute the appropriate size of resources required, and therefore do not result in an optimal cluster composition. Infra provides an auto scaler that automatically adapts a cloud infrastructure to changing application workload, scaling the cluster up/down based on predictions from the Infra provisioning tool.","Title: Novel Fast User-Placement Ushering Algorithms for Indoor Femtocell Networks.\nYear: 2016\n\nAbstract: Nowadays, the sufficient quality-of-service (QoS) provision for mobile applications remains a major challenge in any wireless network. Conventional sufficient QoS provision techniques using resource allocation, data scheduling, and crosslayer optimization have been proposed to tackle this problem. Nevertheless, due to the unpredictable nature of wireless channel conditions, the QoS improvements resulting from the aforementioned approaches are often unsatisfactory. In this paper, we would like to follow our previous concept, namely user-placement ushering (UPU), by use of the user's mobility. A mobile user can be relocated to an optimal, or at least a better place to boost the QoS for a particular application. Novel fast algorithms are devised here to usher (guide) the mobile user to an appropriate spot with sufficient QoS, which is nearby. We address the UPU problem to accommodate realistic complex indoor environments, where obstacles are present. Our simulation results have demonstrated that our new fast UPU algorithms are capable of finding new appropriate locations satisfying the required QoSs for different wireless network applications.","Title: Automated diagnostics for manufacturing machinery based on well-regularized deep neural networks.\nYear: 2017\n\nAbstract: This paper presents a bigdata framework based on regularized deep neural networks for automated diagnostics for manufacturing machinery based on emitted sound, vibration, and magnetic field data. More precisely, we present SpotCheck, a prototype system that uses well-regularized deep neural networks to analyze sound, vibrational, and magnetic emissions of industrial machinery to provide noninvasive machine diagnostics, both for fault detection and to meter the day to day mode of operation of the machinery. It is completely automatic requiring no manual extraction of handcrafted features. It can operate with relatively small amounts of training data, but can take advantage of large volumes of unlabeled data when available, and scale to very large volumes of labeled or unlabeled data to improve performance as more data becomes available after deployment.","Title: StaticGen: Static Generation of UML Sequence Diagrams.\nYear: 2017\n\nAbstract: UML sequence diagrams are visual representations of object interactions in a system and can provide valuable information for program comprehension, debugging, maintenance, and software archeology. Sequence diagrams generated from legacy code are independent of existing documentation that may have eroded. We present a framework for static generation of UML sequence diagrams from object-oriented source code. The framework provides a query refinement system to guide the user to interesting interactions in the source code. Our technique involves constructing a hypergraph representation of the source code, traversing the hypergraph with respect to a user-defined query, and generating the corresponding set of sequence diagrams. We implemented our framework as a tool, StaticGen, analyzing a corpus of 30 Android applications. We provide experimental results demonstrating the efficacy of our technique.","Title: Brief Announcement: Complete Visibility for Oblivious Robots in Linear Time.\nYear: 2017\n\nAbstract: We consider the distributed setting of $N$ autonomous mobile robots that operate in Look-Compute-Move cycles following the well-celebrated classic oblivious robots model. We study the fundamental problem where starting from an arbitrary initial configuration, N autonomous robots reposition themselves to a convex hull formation on the plane where each robot is visible to all others (the Complete Visibility problem). We assume obstructed visibility, where a robot cannot see another robot if a third robot is positioned between them on the straight line connecting them. We provide the first \\\\cO(N) time algorithm for this problem in the fully synchronous setting. Our contribution is a significant improvement over the runtime of the only previously known algorithm for this problem which has a lower bound of \\\\Omega(N^2). Our proposed algorithm is collision-free -- robots do not share positions and their paths do not cross.","Title: Novel electronic scissoring algorithm\nYear: 2017\n\nAbstract: The electronic media, such as digital magazines and newspapers, are becoming more and more popular nowadays. The computer-aided information retrieval of such electronic documents is crucial for any further data-mining applications. In this paper, we proposed a novel electronic scissoring algorithm to automatically extract the image content from an arbitrary electronic document. Our proposed new algorithm contains two steps, namely the information-content extraction and the image-content detection. In the first step, everything other than the background is extracted using a computationally-efficient variance filter. The information content usually consists of text, image, and delimiters. The next step, the entropy measure is employed to distinguish the images from other content. Experiments are carried out using data captured from random web-pages. The receiver-operating characteristic curves have been presented to demonstrate the effectiveness of our proposed new algorithm.","Title: SynAdapt: Automated Synthesis of Adaptive Agents\nYear: 2017\n\nAbstract: Distributed autonomous multi-agent reasoning and classification systems have been thought of to be the basis of intelligence and have wide applications in the space of operational intelligence in closing the loop between sensing, analytics, and actions. This paper targets multi-agent systems that employ rulebased logics (i.e., rules that determine the output/response of an agent depending on the range of the input values) with pre-defined rules to accurately perceive the environment, and provide associated reactions. Such rule-based systems do not perform well in scenarios, where human generated rules cannot adapt to dynamic variations in the data distribution arising due to dynamic changes in the environment, especially if data dimensionality is very high. Examples of such scenarios exist wherever the sensed data arrives from the physical world - such as weather data, physical sensor data, human behaviour controlled data, etc. Clearly, to meet the adaptivity requirements of such scenarios we require the agents to possess adaptive reasoning capability such that they can adapt the underlying rules with respect to the changing environment. Developing such adaptive agents requires the developer to additionally possess considerable expertise of state-of-the-art machine learning techniques, apart from possessing knowledge of the agent's target domain. To address the above issues, we automate the process of development and deployment of adaptive agents. We present the fundamental design concepts behind the development of SynAdapt: a new adaptive meta-learning based multi-agent synthesis framework, that automates the synthesis of adaptive multi-agent systems from high-level user specifications. SynAdapt provides the following key features: a) Automated synthesis and deployment of adaptive agents from high-level user specification, b) Agents synthesised by SynAdapt can select a learning strategy that is particularly suited for given user specifications and input dataset, and c) Agents synthesised by SynAdapt can leverage adaptive ensemble learning techniques to deal with concept drift.","Title: Adaptable SLA-Aware Consistency Tuning for Quorum-Replicated Datastores.\nYear: 2017\n\nAbstract: Users of distributed datastores that employ quorum-based replication are burdened with the choice of a suitable client-centric consistency setting for each storage operation. The above matching choice is difficult to reason about as it requires deliberating about the tradeoff between the latency and staleness, i.e., how stale (old) the result is. The latency and staleness for a given operation dep...","Title: Pixel-Level Reconstruction and Classification for Noisy Handwritten Bangla Characters\nYear: 2018\n\nAbstract: Classification techniques for images of handwritten characters are susceptible to noise. Quadtrees can be an efficient representation for learning from sparse features. In this paper, we improve the effectiveness of probabilistic quadtrees by using a pixel level classifier to extract the character pixels and remove noise from handwritten character images. The pixel level denoiser (a deep belief network) uses the map responses obtained from a pretrained CNN as features for reconstructing the characters eliminating noise. We experimentally demonstrate the effectiveness of our approach by reconstructing and classifying a noisy version of handwritten Bangla Numeral and Basic Character datasets.","Title: Deep neural networks for texture classification-A theoretical analysis.\nYear: 2018\n\nAbstract: We investigate the use of Deep Neural Networks for the classification of image datasets where texture features are important for generating class-conditional discriminative representations. To this end, we first derive the size of the feature space for some standard textural features extracted from the input dataset and then use the theory of Vapnik–Chervonenkis dimension to show that hand-crafted feature extraction creates low-dimensional representations which help in reducing the overall excess error rate. As a corollary to this analysis, we derive for the first time upper bounds on the VC dimension of Convolutional Neural Network as well as Dropout and Dropconnect networks and the relation between excess error rate of Dropout and Dropconnect networks. The concept of intrinsic dimension is used to validate the intuition that texture-based datasets are inherently higher dimensional as compared to handwritten digits or other object recognition datasets and hence more difficult to be shattered by neural networks. We then derive the mean distance from the centroid to the nearest and farthest sampling points in an n-dimensional manifold and show that the Relative Contrast of the sample data vanishes as dimensionality of the underlying vector space tends to infinity.","Title: Consistify - preserving correctness and SLA under weak consistency.\nYear: 2019\n\nAbstract: While executing a client application, comprising a sequence of one or more storage operations (read, write, etc.), replicated datastores provide the option of tuning per-operation client-centric consistency settings, Weaker consistency settings allow client applications to execute with lower observed latency due to less coordination among the replica, improving throughput. To enable a client application running on a geo-replicated datastore to meet stringent SLA (Service Level Agreement) deadlines, i.e., latency thresholds provided in the SLA, users apply weak client-centric consistency settings. Strong consistency settings may result in higher latency due to coordination among greater number of replicas. However, since a weak consistency setting requires only a partial quorum of replicas to coordinate, some of the replicas may be overwritten by conflicting updates from concurrent clients, which, in turn, may result in violation of correctness conditions specified by the developer. Such correctness conditions are provided as postconditions that impose restrictions on the values observed by the clients. Enforcing correctness conditions under weak consistency settings on geo-replicated datastores from the client application code is a tedious and challenging task. The task is even more complicated when considering application-level SLA deadlines. Deadline (latency) is one of the most important SLA parameters for optimizing client-centric performance. To this end, we present Consistify, a novel framework, that acts as an interface layer between client applications and underlying datastores, and enables client applications to execute with weak per-operation consistency settings, while simultaneously respecting: 1) the correctness conditions, specified using logical predicates, and 2) the SLA deadline. Using benchmark workloads, we experimentally demonstrate that Consistify outperforms the state-of-the-art systems, namely QUELA. For several other systems that provide flexible consistency settings, namely Bolt-on causal system, SIEVE, and Cure, we provide a comparison of the throughput and latency observed with Consistify against the numbers reported with those systems in prior papers.","Title: Progressively Growing Generative Adversarial Networks for High Resolution Semantic Segmentation of Satellite Images.\nYear: 2019\n\nAbstract: Machine learning has proven to be useful in classification and segmentation of images. In this paper, we evaluate a training methodology for pixel-wise segmentation on high resolution satellite images using progressive growing of generative adversarial networks. We apply our model to segmenting building rooftops and compare these results to conventional methods for rooftop segmentation. We present our findings using the SpaceNet version 2dataset. Progressive GAN training achieved a test accuracy of 93% compared to 89% for traditional GAN training.","Title: Model-Based static code analysis for MATLAB models\nYear: 2012\n\nAbstract: MATLAB is widely used in scientific, engineering, and numerical computations. Complex systems such as digital signal processors, process control systems, etc. are modeled in MATLAB and analyzed; C implementation of the system can be automatically generated from the validated MATLAB model. We combine static analysis techniques with model-based deductive verification using SMT solvers to provide a framework to analyze MATLAB code. The analyzer is generated by translating the collecting semantics of a MATLAB script to a formula in first order logic over multiple underlying theories. Function calls in a script can be handled by importing SMT assertions obtained by analyzing MATLAB files containing function definitions. Logical specification of the desired program behavior (rather its negation) is incorporated as a first order logic formula. An SMT-LIB formula solver treats the combined formula as a \"constraint\" and \"solves\" it. The \"solved form\" can be used to identify logical errors in the MATLAB model.","Title: Break Down in Order To Build Up: Decomposing Small Molecules for Fragment-Based Drug Design with eMolFrag.\nYear: 2017\n\nAbstract: Constructing high-quality libraries of molecular building blocks is essential for successful fragment-based drug discovery. In this communication, we describe eMolFrag, a new open-source software to decompose organic compounds into nonredundant fragments retaining molecular connectivity information. Given a collection of molecules, eMolFrag generates a set of unique fragments comprising larger moieties, bricks) and smaller linkers connecting bricks. These building blocks can subsequently be used to, construct virtual screening libraries for targeted drug discovery: The robustness and computational performance of eMolFrag is assessed against the Directory of Useful Decoys, Enhanced database conducted in serial and parallel modes with up to 16 computing cores. Further, the application of eMolFrag in de novo drug design is illustrated using the adenosine receptor.","Title: The Complete Visibility Problem for Fat Robots with Lights.\nYear: 2018\n\nAbstract: We consider the distributed setting of N autonomous mobile robots that operate in Look-Compute-Move cycles and communicate with other robots using colored lights (the robots with lights model). We study the fundamental problem of repositioning N autonomous robots on a plane so that each robot is visible to all others (the COMPLETE VISIBILITY problem) on this model. We consider obstructed visibility under which a robot cannot see another robot if a third robot is positioned between then on the straight line connecting them. The literature solves this problem on the robots with lights model for point robots. In this paper, we study this problem, for the very first time, on the robots with lights model for fat robots, i.e., the robots which have physical extents. In particular, we present and analyze an algorithm for the fat robots of unit disc size extents that solves COMPLETE VISIBILITY in O(N) time using 9 colors in the fully synchronous setting avoiding robot collisions.","Title: Why do you take that route?\nYear: 2019\n\nAbstract: The purpose of this paper is to determine whether a particular context factor among the variables that a researcher is interested in causally affects the route choice behavior of drivers. To our knowledge, there is limited literature that consider the effects of various factors on route choice based on causal inference.Yet, collecting data sets that are sensitive to the aforementioned factors are challenging and the existing approaches usually take into account only the general factors motivating drivers route choice behavior. To fill these gaps, we carried out a study using Immersive Virtual Environment (IVE) tools to elicit drivers' route choice behavioral data, covering drivers' network familiarity, educationlevel, financial concern, etc, apart from conventional measurement variables. Having context-aware, high-fidelity properties, IVE data affords the opportunity to incorporate the impacts of human related factors into the route choice causal analysis and advance a more customizable research tool for investigating causal factors on path selection in network routing. This causal analysis provides quantitative evidence to support drivers' diversion decision.","Title: Why Did You Turn On That Light?\nYear: 2021\n\nAbstract: The overarching goal of this paper is to study if certain contextual factors out of the variables that a researcher is interested in has causal influence on human-building interaction for a building under design. To limit the scope of the study, we concentrate on causal analysis of light switch behavior (i.e., an occupant switching on or off a light) in office buildings. Due to the lack of high-fi...","Title: Robustness analysis framework for computations associated with building performance models and immersive virtual experiments\nYear: 2021\n\nAbstract: Building performance models (BPMs) have been used to simulate and analyze building performance during design. While extensive research efforts have made to improve the performance of BPMs, little attention has given to their robustness. Uncertainty is a crucial factor affecting the robustness of BPMs, in which such effect needs to be quantified through a suitable approach. The paper offers a robustness analysis framework for BPMs by using perturbation techniques to simulate uncertainty in input datasets. To investigate the efficacy of the framework, a generative adversarial network (GAN)-based framework was selected as a case study to analyze light switch usages in a single-occupancy office simulated using an immersive virtual environment (IVE). The robustness of the GAN was analyzed by comparing differences between a baseline (i.e., a BPM obtained from the GAN trained on a non-perturbed dataset) and BPMs obtained from the GAN trained on perturbed datasets. Overall, the robustness of the GAN significantly reduced when the training datasets were perturbed by using structured transformation techniques. The GAN remained relatively robust when the training datasets were perturbed by using an additive perturbation. Additionally, the sensitivity of the GAN involves different magnitudes corresponding to different levels of perturbed input datasets. The study suggests that the perturbation analysis is effective for investigating data uncertainty affecting the robustness of BPMs.","Title: SimilarityGAN: Using Similarity to Loosen Structural Constraints in Generative Adversarial Models\nYear: 2021\n\nAbstract: Recently, generative adversarial networks have performed extremely well in image translation. When translating images current models adhere to a strict structural symmetry between the input and output images. This paper, presents a technique for image translation involving a pair of image domains that allows the output image to go beyond the structural symmetry constraints imposed by the input. By..."],"string":"[\n \"Title: Automatic Synthesis And Deployment Of Intensional Kahn Process Networks\\nYear: 2009\\n\\nAbstract: In this paper we introduce and study. theoretically, a clean slate \\\"formal\\\" foundational approach for developing and deploying high-assurance distributed embedded systems deployed in mission-critical applications We propose a simple formal distributed asynchronous framework extending Kahn Process Networks with intensional specification. More precisely. we present a model-driven approach based on a platform-independent language and an intensional specification logic that allows us to synthesize distributed agents that can handle interactions with external resources asynchronously. ensure enforcement of information flow and security policies. and have the ability to deal with failures of resources Our approach allows rapid development and automated deployment of formally verified embedded networked systems that provide guarantees that clients' requirements will be met and QoS guarantees will be respected Moreover, It allows modeling (and programming) tellable distributed systems for multi-core hosts Such a capability makes our framework suitable for next generation grid computing systems whew multi-core individual hosts need to be utilized for improving scalability Given an intensional logical specification of a distributed embedded system. that includes Quality of Service (QoS) requirements. a set of software resources and devices available in a network. and then formal interface specifications, a deductive system can automatically generate distributed extended Kahn processes and their deployment information in such a way that the application requirements-including QoS requirements are guaranteed to be met The generated processes use the inputs of the sensors/meters/probes and the management policies of the customer to generate real-time control decisions for manning the system The processes are deployed automatically on a distributed network involving sensors/meters/probes tracking system parameters, actuators controlling devices. and diverse computing and communication elements such as MA's. etc\",\n \"Title: Counterfactually reasoning about security\\nYear: 2011\\n\\nAbstract: In this short paper, we provide the background to counterfactual logic and give very general suggestions on how we could employ this logic to help us reason about security policies. It seems very appropriate to use this kind of logic to anticipate a change that will compromise the security concerns of a given system before actually applying the changes.\",\n \"Title: Dynamic Message Sequence Charts\\nYear: 2002\\n\\nAbstract: We introduce a formalism to specify classes of MSCs over an unbounded number of processes. The formalism can describe many interesting behaviours of dynamically changing networks of processes. Moreover, it strictly includes the formalism of Message Sequence Graphs studied in the literature to describe MSCs over a fixed finite set of processes. Our main result is that model-checking of MSCs described in this formalism against a suitable monadic-second order logic is decidable.\",\n \"Title: Model checking mobile ambients\\nYear: 2003\\n\\nAbstract: We settle the complexity bounds of the model checking problem for the ambient calculus with public names against the ambient logic. We show that if either the calculus contains replication or the logic contains the guarantee operator, the problem is undecidable. In the case of the replication-free calculus and guarantee-free logic we prove that the problem is PSPACE-complete. For the complexity upper bound, we devise a new representation of processes that remains of polynomial size during process execution; this allows us to keep the model checking procedure in polynomial space. Moreover, we prove PSPACE-hardness of the problem for several quite simple fragments of the calculus and the logic; this suggests that there are no interesting fragments with polynomial-time model checking algorithms.\",\n \"Title: Model Checking Communication Protocols\\nYear: 2001\\n\\nAbstract: Brand and Zafiropulo [1] introduced the model of communicating finite-state machines to represent a distributed system connected with FIFO channels. Several different communication protocols can be specified with this simple model. In this paper we address the problem of automatically validating protocols by verifying properties such as well-formedness and absence of deadlock. Our method is based on a representation of communicating finite-state machines in terms of logic programs. This leads to efficient verification algorithms based on the ground and non-ground semantics of logic programming.\",\n \"Title: An Algebraic Framework for Abstract Model Checking\\nYear: 2002\\n\\nAbstract: Symbolic forward analysis is a semi-algorithm that in many cases solves the model checking problem for infinite state systems in practice. This semi-algorithm is implemented in manypractical model checking tools like UPPAAL [BLL+96], KRONOS [DT98] and HYTECH [HHWT97]. In most practical experiments, termination of symbolic forward analysis is achieved by employing abstractions resulting in an abstract symbolic forward analysis. This paper presents a unified algebraic framework for deriving and reasoning about abstract symbolic forward analysis procedures for a large class of infinite state systems with variables ranging over a numeric domain. The framework is obtained by lifting notions from classical algebraic theoryof automata to constraints representing sets of states. Our framework provides sufficient conditions under which the derived abstract symbolic forward analysis procedure is always terminating or accurate or both. The class of infinite state systems that we consider here are (possibly non-linear) hybrid systems and (possiblynon-linear) integer-valued systems. The central notions involved are those of constraint transformer monoids and coverings between constraint transformer monoids. We show concrete applications of our framework in deriving abstract symbolic forward analysis algorithms for timed automata and the two process bakeryalgorithm that are both terminating and accurate.\",\n \"Title: Toward More Precise Radiotherapy Treatment of Lung Tumors\\nYear: 2012\\n\\nAbstract: A computational framework for modeling the respiratory motion of lung tumors provides a 4D parametric representation that tracks, analyzes, and models movement to provide more accurate guidance in the planning and delivery of lung tumor radiotherapy.\",\n \"Title: A Risk Reduction Framework for Dynamic Workflows\\nYear: 2008\\n\\nAbstract: Workflows tend to fail in real-world scenarios due to the uncertain/unreliable sensory information which sometimes needs to be updated during the execution of workflows. In a logic based framework, these dynamic predicates that can be updated are called non-monotonic predicates (NMPs). In this paper, we focus on reducing the risk of a given workflow due to the NMPs in that workflow. The main idea is to synthesize a backup workflow by augmenting the main workflow without introducing new NMPs. The backup workflow is generated by using expected values of NMPs if necessary instead of given values. The expected values are calculated from the execution history or provided by a domain expert. It is argued that total risk reduces to the square root of the main workflow itself.\",\n \"Title: RETNA: From Requirements to Testing in a Natural Way\\nYear: 2004\\n\\nAbstract: Most problems in building and refining a system can be traced back to errors in requirements. Poorly organized requirements, most often in natural language are among the major causes of failures of software projects. In this paper, we present a requirements analysis tool called RETNA and the technology behind it. RETNA accepts natural language requirements, classifies them, interacts with the user to refine them, automatically translates natural language requirements to a logical format so that they can be validated and finally generates test cases from the requirements.\",\n \"Title: An ad-hoc distributed execution environment for multi-agent systems\\nYear: 2013\\n\\nAbstract: We present a distributed execution environment for multi-agent systems running in an intermittently-connected network. We describe the design and implementation of a microkernel-based prototype operating system layer that provides services for managing agent execution. The execution environment provides the user with a consistent view of a single machine, a single file system, and a unified programming model.\",\n \"Title: Constraint Database Models Characterizing Timed Bisimilarity\\nYear: 2001\\n\\nAbstract: The problem of deciding timed bisimilarity has received increasing attention; it is important for verification of timed systems. Using a characterization of timed bisimilarity in terms of models of constraint databases, we present to our knowledge, the first local, symbolic algorithm for deciding timed bisimilarity; previous algorithms were based on a finite, but prohibitively large, abstraction (the region graph or the full backward stable graph). Our algorithm uses XSB-style tabling with constraints. Our methodology is more general than those followed in the previous approaches in the sense that our algorithm can be used to decide whether two timed systems are alternating timed bisimilar.\",\n \"Title: Code-Change Impact Analysis Using Counterfactuals\\nYear: 2011\\n\\nAbstract: In this paper we present a framework for what-if analysis of programs based on Lewis' theory of counterfactuals. The framework can be used to statically perform change impact analysis for source code. It enables us to verify assertions about a changed version of the program without actually incorporating the changes. We present a logical calculus that precisely characterizes structural modifications to source code and their impact on the behavior of the program.\",\n \"Title: Model Checking for Timed Logic Processes\\nYear: 2000\\n\\nAbstract: We apply techniques from logic programming and constraint databases to verify real time systems. We introduce timed logic processes (TLPs) as a fragment of constraint query languages over reals. We establish a formal connection between TLPs and timed automata, and between the procedure of the UPPAAL model checker for restricted temporallogic properties of timed automata and the top-down query evaluation of TLPs (with tabling in the XSB style). This connection yields an alternative implementation of the UPPAAL procedure. Furthermore, we can extend that procedure in order to accommodate more expressive properties.\",\n \"Title: Optimal-Reachability and Control for Acyclic Weighted Timed Automata\\nYear: 2002\\n\\nAbstract: Abstract: Weighted timed automata extend timed automata with costs on bothlocations and transitions. In this framework we study the optimal reachabilityand the optimal control synthesis problems for the automata withacyclic control graphs. This class of automata is relevant for some practicalproblems such as some static scheduling problems or air-traccontrol problems. We give a nondeterministic polynomial time algorithmto solve the decision version of the considered optimal reachabilityproblem....\",\n \"Title: New Modified Turbo Decoder for Embedded Local Content in Single-Frequency Networks.\\nYear: 2013\\n\\nAbstract: Current DVB-SH systems rely on hierarchical modulation to insert local content into the complementary ground component. The local content is inserted as a low priority bit stream, while the global content will be transmitted as a high-priority bit stream. At the receiver, the standard approach of extracting the local content from the received signal first requires detection of the global content. Subsequently, one can extract the local content by subtracting the coded and modulated form (restored version) of the detected global content from the received signal. In this paper, we propose a different approach to remove the global content from the received signal. In our new approach, a hybrid signal is designed to represent the turbo-coded global content, which contains decoded bits from the turbo decoder representing systematic bits, and detected bits from the received signal prior to the turbo decoder representing parity bits. This technique would relieve the receiver from performing coding and modulation on the detected global content so as to ensure a less computational burden and a smaller latency than the current technology. The obvious tradeoff between complexity and performance would imply higher bit error rates arising from our proposed system. Therefore, to mitigate the bit-error-rate performance degradation, we propose modifying the turbo decoder so that the systematic bit streams and the parity bit streams will be decoded differently with respect to their corresponding signal-to-noise ratios. Monte Carlo simulation results demonstrate that our proposed new scheme can achieve better computational efficiency and less latency in the slight sacrifice of error performance.\",\n \"Title: Functional \\\"AJAX\\\" in Secure Synchronous Programming\\nYear: 2011\\n\\nAbstract: AJAX (Asynchronous Javascript and XML) is a confederation of technologies aimed at providing improved user interaction with web-based applications. While AJAX provides an improved user experience, it also comes with its baggage of problems. The lack of formal semantics makes AJAX applications difficult to build, debug, understand, and validate. Different component technologies of AJAX (e.g., XMLHttpRequest or Javascript) are browser-sensitive and have different implementations and provide distinct functionalities. Source code is downloaded and run on the clients machines, raising security concerns. In this paper, we present an ``AJAX''-like framework in an event-driven secure synchronous programming environment. Our framework is supported by a formal operational semantics. Applications written in our language can be verified using formal static analysis techniques such as theorem proving. The applications are compiled and run on the SINS (Secure Infrastructure for Networked Systems) infrastructure jointly developed in collaboration with the Naval Research Laboratory.\",\n \"Title: Towards Formal Verification of a Commercial Wireless Router Firmware\\nYear: 2013\\n\\nAbstract: Formal verification of the trusted computing base of a software system is essential for its deployment in mission-critical environments. Commercial off-the-shelf routers are nowadays being used for managing traffic in high-assurance networks. The specifications for the development of these routers are provided by RFCs that are only described informally in English. It is essential to ensure that a router firmware conforms to its corresponding RFC before it can be deployed for managing mission-critical networks. In this paper, we report the formal verification of the conformance of the open source Netgear WNR3500L wireless router firmware implementation to the RFC 2131 [6] based on which it is designed. The formal verification effort led to the discovery of several possible problems in the implementation that we report in this paper. We have used the Coq proof assistant extensively in this verification effort. The formal verification process demonstrates the usefulness of inductive types and higher-order logic in software certification.\",\n \"Title: Model-Based Static Source Code Analysis of Java Programs with Applications to Android Security\\nYear: 2012\\n\\nAbstract: We combine static analysis techniques with model- based deductive verification using SMT solvers to provide a framework that, given an analysis aspect of the source code, automatically generates an analyzer capable of inferring information about that aspect. The analyzer is generated by translating the collecting semantics of a program to a \\\"marked\\\" formula in first order logic over multiple underlying theories. The \\\"marking\\\" can be thought of as a set of holes or contexts corresponding to the \\\"uninterpreted\\\" APIs invoked in the program. Just as a program imports packages and uses methods from classes in those packages, we import the semantics of the API invocations as first order logic assertions. These assertions constitute the models used by the analyzer. Logical specification of the desired program behavior (rather its negation) is incorporated as a first order logic formula. An SMT-LIB formula solver treats the combined formula as a \\\"constraint\\\" and \\\"solves\\\" it. The \\\"solved form\\\" can be used to identify logical (security) errors in Java (Android) programs. Security properties of Android are represented as constraints and the analysis aims to show that these constraints are respected.\",\n \"Title: Preventing Future Oil Spills with Software-Based Event Detection\\nYear: 2010\\n\\nAbstract: Complex event processing systems detect problems in mission-critical, real-time applications and generate intelligent decisions to modulate the system environment.\",\n \"Title: Deterministic finite automata with recursive calls and DPDAs\\nYear: 2003\\n\\nAbstract: We study deterministic finite automata (DFA) with recursive calls, that is, finite sequences of component DFAs that can call each other recursively. DFAs with recursive calls are akin to recursive state machines and unrestricted hierarchic state machines. We show that they are language equivalent to deterministic pushdown automata (DPDA).\",\n \"Title: Beyond Region Graphs: Symbolic Forward Analysis of Timed Automata\\nYear: 1999\\n\\nAbstract: Theoretical investigations of infinite-state systems have so far concentrated on decidability results; in the case of timed automata these results are based on region graphs. We investigate the specific procedure that is used practically in order to decide verification problems, namely symbolic forward analysis. This procedure is possibly nonterminating. We present basic concepts and properties that are useful for reasoning about sufficient termination conditions, and then derive some conditions. The central notions here are constraint transformers associated with sequences of automaton edges and zone trees labeled with successor constraints.\",\n \"Title: Approximation of CFL by Regular Languages for Concurrent Program Verification\\nYear: 2010\\n\\nAbstract: Many problems related to verification of concurrent programs can be reduced to the non-empty intersection problem for context-free languages. Since the latter is an undecidable problem, a practical approach for solving the intersection problem is to convert it to a decidable problem of the non-empty intersection of a context-free language and a regular language. This is done by approximating one of the context-free languages in the intersection from above or from below by a regular language. We give an approximation technique from above by modeling a context-free language L in terms of linear integer inequalities and then obtaining the approximating regular language L' ⊇ L by relaxing the linear inequalities such that each inequality involves at most one variable. Previous approximation techniques focused on approximation below. We also give an alternate technique with a finite-state automata based approach, where we start with an automata M which accepts a suitable finite-subset L0 of L, and then extend M successively based on the pumping property of L till the language accepted by M contains L.\",\n \"Title: Managing a Cloud for Multi-agent Systems on Ad-Hoc Networks\\nYear: 2012\\n\\nAbstract: We present a novel execution environment for multi-agent systems building on concepts from cloud computing and peer-to-peer networks. The novel environment can provide the computing power of a cloud for multi-agent systems in intermittently connected networks. We present the design and implementation of a prototype operating system for managing the environment. The operating system provides the user with a consistent view of a single machine, a single file system, and a unified programming model while providing elasticity and availability.\",\n \"Title: Accurate Widenings and Boundedness Properties of Timed Systems\\nYear: 2001\\n\\nAbstract: We propose a symbolic model checking procedure for timed systems that is based on operations on constraints. To accelerate the termination of the model checking procedure, we define history-dependent widening operators, again in terms of constraint-based operations. We show that these widenings are accurate, i.e., they don't lose precision even with respect to the test of boundedness properties.\",\n \"Title: Automated Situation-Aware Service Composition In Service-Oriented Computing\\nYear: 2007\\n\\nAbstract: Service-based systems have many applications, such as e-business, health care, and homeland security. In these systems, it is necessary to provide users the capability of composing services into workflows providing higher-level functionality. In dynamic service-oriented computing environments, it is desirable that service composition is automated and situation-aware to generate robust and adaptive workflows. In this paper, an automated situation-aware service composition approach is presented This approach is based on the a-logic, a-calculus, and a declarative model for situation awareness (SAW). This approach consists of four major components: (1) analyzing SAW requirements using our SAW model, (2) translating our SAW model representation to a-logic specifications and specifying a control flow graph in a-logic as the service composition goal, (3) automated synthesis of a-calculus terms defining situation-aware workflow agents based on a-logic specifications for SA W requirements and the control flow graph, and (4) compilation of a-calculus terms to executable components.\",\n \"Title: Code-Change Impact Analysis Using Counterfactuals: Theory And Implementation\\nYear: 2013\\n\\nAbstract: This article shows a novel program analysis framework based on Lewis' theory of counterfactuals. Using this framework we are capable of performing change-impact static analysis on a program's source code. In other words, we are able to prove the properties induced by changes to a given program before applying these changes. Our contribution is two-fold; we show how to use Lewis' logic of counterfactuals to prove that proposed changes to a program preserve its correctness. We report the development of an automated tool based on resolution and theorem proving for performing code change-impact analysis.\",\n \"Title: Service Composition in a Secure Agent-Based Architecture\\nYear: 2005\\n\\nAbstract: We describe an agent-based situation-aware survivable architecture for the discovery and composition of web services.Our architecture provides for proofs that guarantee the consistency of service level agreements between service providers and their clients.The architecture is based on decentralized situation-aware ambients (SAAs) which are autonomous agents that process information about the current situation, and discover and compose services syntactically to adapt to changing situations. The agents run on the Secure Infrastructure for Networked Systems (SINS) under development at the Center for High Assurance Computer Systems of the Naval Research Laboratory.\",\n \"Title: A Type-Theoretic Framework for Automated Synthesis of Workflows in Service-Oriented Computing\\nYear: 2013\\n\\nAbstract: We describe a technique for automatic synthesis of workflows in service-oriented systems from a typetheoretic specification of coordination requirements. The technique presented involves formal semantics-based, automatic synthesis of service coordination workflows from the business logic of the clients using type and effect inference. The type and effect deduction, apart from synthesizing the workflow, provides formal proofs that the business requirements of the clients in a service-oriented system are met. The service descriptions and the business requirements are specified as a set of type judgements in an (intuitionistic) type theory. Workflows can be generated from a (type and effect) inference of the requirements from the \\\"service theory\\\".\",\n \"Title: Novel Fast Blind Channel Estimation and Hessian Analysis for Transmitter Identification of Digital Television Signals\\nYear: 2014\\n\\nAbstract: The transmitter identification (Tx-ID) technique has been widely addressed in the modern advanced television systems committee digital television (DTV) standards. Kasami sequences are adopted as the Tx-ID sequences due to the favorable code capacity. However, the conventional cross-correlation based Tx-ID technique cannot combat the realistic multipath channel impairment. Therefore, in this paper, we propose a new least-square Tx-ID approach, which is demonstrated to be very robust when the multipath channels are encountered. We also establish the new Hessian analysis to provide the range for appropriate injection levels accordingly. Based on this new theoretical study, we design new efficient (fast) algorithms to determine the maximum allowable channel length given a particular Kasami Tx-ID sequence. These algorithms and theoretical results can be very useful for the future DTV technologies.\",\n \"Title: LDPC encoder identification in time-varying flat-fading channels\\nYear: 2014\\n\\nAbstract: This paper tackles the low-density parity-check (LDPC) encoder identification problem encountered in the time-varying flat-fading channels which are modeled as finite-state Markov chains. The in-phase and quadrature-phase components of the channel coefficients are both represented by a number of states. To greatly simplify the computation of the channel observation probabilities, each channel-state region is further quantized to an interior point. Based on our proposed finite-state Markov model, the Viterbi algorithm is thus invoked to blindly estimate the unknown channel-state sequence from each received signal segment. To mitigate the phase ambiguity which is inherent in the channel-state estimation process, the pilot-aided channel estimation method is also proposed here. The LDPC encoder is finally identified in the framework of the log-likelihood ratio of syndrome a posteriori probability. The performance of our proposed LDPC identification scheme is investigated for different normalized Doppler rates and different mechanisms to reconstruct the channel-state information. Monte Carlo simulation results suggest that pilot symbols are necessary for leading to a satisfactory identification performance for time-varying flat-fading channels.\",\n \"Title: Stability of video rate control algorithms over bandwidth-limited network paths\\nYear: 2014\\n\\nAbstract: The paper considers the stability of AIMD-based rate adaptation algorithm for video transmission over a bandwidth-limited network path. We examine the stability aspects from two angles: first, a quick relief from congestion by a faster reduction of send rate, and second, a larger limit-cycle time with a slow rate reduction in the steady-state (and hence lower rate jitter). The stability assessment also considers the loss of feedback signals from the video receiver that indicate the packet loss on data channels. We provide a management-oriented assessment of the convergence properties of video rate control system. This in turn allows an autonomic adjustment of the AIMD parameters and the system controller models for optimal performance.\",\n \"Title: Managing shared contexts in distributed multi-player game systems\\nYear: 2014\\n\\nAbstract: In this paper, we consider the impact of a weaker model of eventual consistency on distributed multi-player games. This model is suitable for networks in which hosts can leave and join at anytime, e.g., in an intermittently connected environment. Such a consistency model is provided by the Secure Infrastructure for Networked Systems (SINS) [24], a reliable middleware framework. SINS allows agents to communicate asynchronously through a distributed transactional key-value store using anonymous publish-subscribe. It uses Lamport's Paxos protocol [17] to replicate state. We consider a multi-player maze game as example to illustrate our consistency model and the impact of network losses/delays therein. The framework based on SINS presented herein provides a vehicle for studying the effect of human elements participating in collaborative simulation of a physical world as in war games.\",\n \"Title: Does your result checker really check?\\nYear: 2004\\n\\nAbstract: A result checker is a program that checks the output of the computation of the observed program for correctness. Introduced originally by Blum, the result-checking paradigm has provided a powerful platform assuring the reliability of software. However, constructing result checkers for most problems requires not only significant domain knowledge but also ingenuity and can be error prone. In this paper we present our experience in validating result checkers using formal methods. We have conducted several case studies in validating result checkers from the commercial LEDA system for combinatorial and geometric computing. In one of our case studies, we detected a logical error in a result checker for a program computing max flow of a graph.\",\n \"Title: SOLj: A Domain-Specific Language (DSL) for Secure Service-Based Systems\\nYear: 2007\\n\\nAbstract: We present SOLj (Secure Operations Language-Java), an event-driven domain-specific synchronous programming extension of Java for developing secure service-based systems. The language has capabilities for handling service invocations asynchronously, includes strong typing for the enforcement of information flow and security policies, and exception handling mechanisms to deal with failures of components or services (both benign and Byzantine). Applications written in SOLj are formally verifiable using static analysis techniques. SOLj programs may be deployed, con- figured, and run on SINS (Secure Infrastructure for Networked Systems) under development at the Naval Research Laboratory.\",\n \"Title: A Formal Approach to Developing Reliable Event-Driven Service-Oriented Systems\\nYear: 2008\\n\\nAbstract: In this paper, we present a formal framework for developing distributed service-oriented systems in an event-driven secure synchronous programming environment. More precisely, we present a synchronous programming language called SOL (Secure Operations Language) that has (i) capabilities for handling service invocations asynchronously, (ii) strong typing to ensure enforcement of information flow and security policies, and (iii) the ability to deal with failures (both benign and byzantine) of components. SOL is supported by formal operational semantics. Applications written in our framework can be verified using formal static checking techniques like theorem proving. The framework runs on the top of the SINS (secure infrastructure for networked systems) infrastructure that we have developed.\",\n \"Title: Cognitive Distributed Networks in Environmental e-Science\\nYear: 2008\\n\\nAbstract: We integrate fundamental principles from formal mathematics, engineering, and information sciences to develop novel distributed sensor-based networks for management of the environment and the natural ecosystem. More precisely, we are developing ENVNET – ENVironmental sensor NETwork, a cognitive wireless sensor network to proactively collect information, monitor, and control the health of our natural environment and ecosystem (that includes, water, soil, air, etc.). It integrates intelligent sensor coordination and data fusion techniques to access, retrieve, process, and communicate with disparate wireless sensors in an ad-hoc manner to deliver dynamic decisions and provide adequate information management. Our approach facilitates large scale deployment and drastically reduces the hardware cost almost by a factor of 10 and removes the main bottleneck in irrigation control arising from wired sensors. Apart from this it provides a smart distributed closed-loop control mechanism with formal reliability guarantees that is rapidly and easily deployable, context-aware, scalable, and dynamically reconfigurable in response to changing requirements.\",\n \"Title: A Semiautomated Probabilistic Framework for Tree-Cover Delineation From 1-m NAIP Imagery Using a High-Performance Computing Architecture\\nYear: 2015\\n\\nAbstract: Accurate tree-cover estimates are useful in deriving above-ground biomass density estimates from very high resolution (VHR) satellite imagery data. Numerous algorithms have been designed to perform tree-cover delineation in high-to-coarse-resolution satellite imagery, but most of them do not scale to terabytes of data, typical in these VHR data sets. In this paper, we present an automated probabilistic framework for the segmentation and classification of 1-m VHR data as obtained from the National Agriculture Imagery Program (NAIP) for deriving tree-cover estimates for the whole of Continental United States, using a high-performance computing architecture. The results from the classification and segmentation algorithms are then consolidated into a structured prediction framework using a discriminative undirected probabilistic graphical model based on conditional random field, which helps in capturing the higher order contextual dependence relations between neighboring pixels. Once the final probability maps are generated, the framework is updated and retrained by incorporating expert knowledge through the relabeling of misclassified image patches. This leads to a significant improvement in the true positive rates and reduction in false positive rates (FPRs). The tree-cover maps were generated for the state of California, which covers a total of 11 095 NAIP tiles and spans a total geographical area of 163 696 sq. miles. Our framework produced correct detection rates of around 88% for fragmented forests and 74% for urban tree-cover areas, with FPRs lower than 2% for both regions. Comparative studies with the National Land-Cover Data algorithm and the LiDAR high-resolution canopy height model showed the effectiveness of our algorithm for generating accurate high-resolution tree-cover maps.\",\n \"Title: A graph-based approach to construct target-focused libraries for virtual screening.\\nYear: 2016\\n\\nAbstract: eSynth can successfully reconstruct chemically feasible molecules from molecular fragments. Furthermore, in a procedure mimicking the real application, where one expects to discover novel compounds based on a small set of already developed bioactives, eSynth is capable of generating diverse collections of molecules with the desired activity profiles. Thus, we are very optimistic that our effort will contribute to targeted drug discovery. eSynth is freely available to the academic community at www.brylinski.org/content/molecular-synthesis.Graphical abstractAssuming that organic compounds are composed of sets of rigid fragments connected by flexible linkers, a molecule can be decomposed into its building blocks tracking their atomic connectivity. Here, we developed eSynth, an automated method to synthesize new compounds by reconnecting these building blocks following the connectivity patterns via an exhaustive graph-based search algorithm. eSynth opens up a possibility to rapidly construct virtual screening libraries for targeted drug discovery.\",\n \"Title: Reasoning about security in sensor networks\\nYear: 2015\\n\\nAbstract: We present a formal framework for reasoning about security concerns in the context of embedded sensor networks. We first provide an agent-based programming model for sensor networks. A logical framework enables reasoning about security, safety, and integrity with respect to usage of resources in this model. Embedded sensor networks often operate in rapidly changing mission-critical environments where both functional and nonfunctional requirements can alter dynamically in an unforeseen manner. The network may need to be reconfigured and reprogrammed in response to changes in its operating conditions. We provide a framework based on counterfactual logic to formally represent changes to the system and perform what-if reasoning about their impact on security and safety even before they have been applied. Copyright © 2015 John Wiley & Sons, Ltd.\",\n \"Title: OptEx: A Deadline-Aware Cost Optimization Model for Spark\\nYear: 2016\\n\\nAbstract: We present OptEx, a closed-form model of job execution on Apache Spark, a popular parallel processing engine. To the best of our knowledge, OptEx is the first work that analytically models job completion time on Spark. The model can be used to estimate the completion time of a given Spark job on a cloud, with respect to the size of the input dataset, the number of iterations, the number of nodes comprising the underlying cluster. Experimental results demonstrate that OptEx yields a mean relative error of 6% in estimating the job completion time. Furthermore, the model can be applied for estimating the cost optimal cluster composition for running a given Spark job on a cloud under a completion deadline specified in the SLO (i.e.,Service Level Objective). We show experimentally that OptEx is able to correctly estimate the cost optimal cluster composition for running a given Spark job under an SLO deadline with an accuracy of 98%.\",\n \"Title: OptCon: An Adaptable SLA-Aware Consistency Tuning Framework for Quorum-Based Stores\\nYear: 2016\\n\\nAbstract: Users of distributed datastores that employquorum-based replication are burdened with the choice of asuitable client-centric consistency setting for each storage operation. The above matching choice is difficult to reason about asit requires deliberating about the tradeoff between the latencyand staleness, i.e., how stale (old) the result is. The latencyand staleness for a given operation depend on the client-centricconsistency setting applied, as well as dynamic parameters such asthe current workload and network condition. We present OptCon, a novel machine learning-based predictive framework, that canautomate the choice of client-centric consistency setting underuser-specified latency and staleness thresholds given in the servicelevel agreement (SLA). Under a given SLA, OptCon predictsa client-centric consistency setting that is matching, i.e., it isweak enough to satisfy the latency threshold, while being strongenough to satisfy the staleness threshold. While manually tunedconsistency settings remain fixed unless explicitly reconfigured, OptCon tunes consistency settings on a per-operation basis withrespect to changing workload and network state. Using decisiontree learning, OptCon yields 0.14 cross validation error in predictingmatching consistency settings under latency and stalenessthresholds given in the SLA. We demonstrate experimentally thatOptCon is at least as effective as any manually chosen consistencysettings in adapting to the SLA thresholds for different usecases. We also demonstrate that OptCon adapts to variationsin workload, whereas a given manually chosen fixed consistencysetting satisfies the SLA only for a characteristic workload.\",\n \"Title: Mutual Visibility with an Optimal Number of Colors.\\nYear: 2015\\n\\nAbstract: We consider the following fundamental Mutual Visibility problem: Given a set of n identical autonomous point robots in arbitrary distinct positions in the Euclidean plane, find a schedule to move them such that within finite time they reach, without collisions, a configuration in which they all see each other. The robots operate following Look-Compute-Move cycles and a robot $$r_i$$ can not see other robot $$r_j$$ if there lies a third robot $$r_l$$ in the line segment connecting the positions of $$r_i$$ and $$r_j$$. Moreover, n is not assumed to be known to the robots. We study this problem in the robots with lights model, where each robot has an externally visible persistent light which can assume colors from a fixed set of colors and the color set is identical to all the robots. This model corresponds to the classical model of oblivious robots when the number of colors $$c=1$$ in the color set. Therefore, we focus here on the objective of minimizing the number of colors required to successfully solve Mutual Visibility. Di Luna et al. [16] presented two algorithms and showed that Mutual Visibility can always be solved without collisions with $$c=3$$ colors for both semi-synchronous and asynchronous robots under both rigid and non-rigid moves. In this paper, we present and analyze an improved algorithm which requires only $$c=2$$ colors and works for both semi-synchronous and asynchronous robots under both rigid and non-rigid moves; this is optimal since any algorithm for Mutual Visibility needs at least 2 colors in the robots with lights model, when n is not known. We employ a non-trivial technique which moves all the interior robots first to the boundary of the initial convex hull and then the robots in the boundary of the hull except the corners to outside of the hull until all the robots eventually become corners, without the need of any third color. Our result is interesting in the sense that asynchronicity and non-rigidity of robot movements have no effect on Mutual Visibility with respect to the number of colors needed to successfully solve it. We also provide an improved solution to the Circle Formation problem.\",\n \"Title: DeepSat - A Learning framework for Satellite Imagery\\nYear: 2015\\n\\nAbstract: Satellite image classification is a challenging problem that lies at the crossroads of remote sensing, computer vision, and machine learning. Due to the high variability inherent in satellite data, most of the current object classification approaches are not suitable for handling satellite datasets. The progress of satellite image analytics has also been inhibited by the lack of a single labeled high-resolution dataset with multiple class labels. The contributions of this paper are twofold -- (1) first, we present two new satellite datasets called SAT-4 and SAT-6, and (2) then, we propose a classification framework that extracts features from an input image, normalizes them and feeds the normalized feature vectors to a Deep Belief Network for classification. On the SAT-4 dataset, our best network produces a classification accuracy of 97.95% and outperforms three state-of-the-art object recognition algorithms, namely - Deep Belief Networks, Convolutional Neural Networks and Stacked Denoising Autoencoders by ~11%. On SAT-6, it produces a classification accuracy of 93.9% and outperforms the other algorithms by ~15%. Comparative studies with a Random Forest classifier show the advantage of an unsupervised learning approach over traditional supervised learning techniques. A statistical analysis based on Distribution Separability Criterion and Intrinsic Dimensionality Estimation substantiates the effectiveness of our approach in learning better representations for satellite imagery.\",\n \"Title: The Causal Order is Strict\\nYear: 2008\\n\\nAbstract: The causal order is central to many algorithms in both the message-passing and shared memory models. In the message-passing model, the causal order is clearly a strict, or irreflexive partial, order. However, some shared memory models are defined without reference to time. In those models, the causal order can fail to be a strict order. Existing works have either declared that all non-strict orders are not causal, or assumed that the causal order is strict without proof. We prove that, under a small number of reasonable assumptions about systems, the causal order is strict. In particular, we assume neither a global time model nor that processes issue a single shared memory operation at a time.\",\n \"Title: Bounds on Mutual Visibility Algorithms.\\nYear: 2015\\n\\nAbstract: We consider the fundamental Mutual Visibility problem for a set of n identical autonomous point robots (n is not known to the robots) that operate following Look-Compute-Move cycles starting from arbitrary distinct positions in the Euclidean plane under obstructed visibility { a robot ri can see robot rj;rj 6 ri; if and only if there is no other robot in the line segment joining their positions. The objective is to determine a schedule to reposition these robots without collisions such that they reach in nite time a conguration where they all see each other. In the recently proposed so-called robots with lights model, Di Luna et al. [15] gave two deterministic algorithms Contain and Shrink for this problem; however, no runtime bounds were given except the proof that they terminate in nite time. In this paper, we rst study the runtime bounds of these algorithms in the fully synchronous setting showing that Contain is tight (( n) rounds) and Shrink needs ( n 2 ) rounds in the worst-case. We then present a new deterministic algorithm, called Modied Shrink, for fully synchronous setting that solves this problem in O(n logn) rounds, improving signicantly over Shrink. We also show that Modied Shrink has the lower bound of ( n) rounds.\",\n \"Title: MAPTrack - A Probabilistic Real Time Tracking Framework by Integrating Motion, Appearance and Position Models.\\nYear: 2015\\n\\nAbstract: In this paper, we present MAPTrack a robust tracking framework that uses a probabilistic scheme to combine a motion model of an object with that of its appearance and an estimation of its position. The motion of the object is modelled using the Gaussian Mixture Background Subtraction algorithm, the appearance of the tracked object is enumerated using a color histogram and the projected location of the tracked object in the image space/frame sequence is computed by applying a Gaussian to the Region of Interest. Our tracking framework is robust to abrupt changes in lighting conditions, can follow an object through occlusions, and can simultaneously track multiple moving foreground objects of different types (e.g., vehicles, human, etc.) even when they are closely spaced. It is able to start tracks automatically based on a spatio-temporal filtering algorithm. A “dynamic” integration of the framework with optical flow allows us to track videos resulting from significant camera motion. A C++ implementation of the framework has outperformed existing visual tracking algorithms on most videos in the Video Image Retrieval and Analysis Tool (VIRAT), TUD, and the Tracking-Learning-Detection (TLD) datasets.\",\n \"Title: An Agile Framework for Real-Time Motion Tracking\\nYear: 2015\\n\\nAbstract: We present an agile framework for automated tracking of moving objects in full motion video (FMV). The framework is robust, being able to track multiple foreground objects of different types (e.g., Person, vehicle) having disparate motion characteristics (like speed, uniformity) simultaneously in real time under changing lighting conditions, background, and disparate dynamics of the camera. It is able to start tracks automatically based on a confidence-based spatio-temporal filtering algorithm and is able to follow objects through occlusions. Unlike existing tracking algorithms, with high likelihood, it does not lose or switch tracks while following multiple similar closely-spaced objects. The framework is based on an ensemble of tracking algorithms that are switched automatically for optimal performance based on a performance measure without losing state. Only one of the algorithms, that has the best performance in a particular state is active at any time providing computational advantages over existing ensemble frameworks like boosting. A C++ implementation of the framework has outperformed existing visual tracking algorithms on most videos in the Video Image Retrieval and Analysis Tool (VIRAT: www.viratdata.org) and the Tracking-Learning-Detection data-sets.\",\n \"Title: Tight analysis of a collisionless robot gathering algorithm\\nYear: 2015\\n\\nAbstract: We consider the fundamental problem of gathering a set of n robots in the Euclidean plane that have a physical extent and hence cannot share their positions with other robots. The objective is to determine a minimum time schedule to gather the robots as close together as possible around a predefined gathering point avoiding collisions. This problem with minimum time objective has applications in many real-world scenarios including fast autonomous coverage formation. Cord-Landwehr et al. (in Proceedings of the International Conference on Current Trends in Theory and Practice of Computer Science, 2011) gave a local greedy algorithm in a fully synchronous setting and proved that, for the discrete version of the problem where robots’ movements are restricted to the positions on an integral grid, their algorithm solves this problem in O(nR) rounds, where R is the distance from the farthest initial robot position to the gathering point. In this article, we improve significantly the round complexity of their algorithm to R + 2 · (n − 1) rounds. This round complexity is obtained in the following modified model: (1) the viewing range of the robots is increased to three hops and (2) robots can additionally move to the diagonally opposite corner to a grid cell in one step—that is, they can traverse the two corresponding grid edges in one time step. We also prove that there are initial configurations of n robots in this problem where at least R+&frac;(n-1)/2 rounds are needed by any local greedy algorithm. Furthermore, we improve the lower bound to R + (n − 1) rounds for the algorithm of Cord-Landwehr et al. These results altogether provide a tight runtime analysis of their algorithm.\",\n \"Title: Infra: SLO Aware Elastic Auto-scaling in the Cloud for Cost Reduction.\\nYear: 2016\\n\\nAbstract: Enterprises often host applications and services on clusters of virtual machine instances provided by cloud service providers, like Amazon, Rackspace, Microsoft, etc. Users pay a cloud usage cost on the basis of the hourly usage [1] of virtual machine instances composing the cluster. A cluster composition refers to the number of virtual machine instances of each type (from a predefined list of types) comprising a cluster. We present Infra, a cloud provisioning framework that can predict an (a#x03B5; a#x03B4;)-minimum cluster composition required to run a given application workload on a cloud under an SLO (i.e., Service Level Objective) deadline. This paper does not present a new approximation algorithm, instead we provide a tool that applies existing machine learning techniques to predict an (a#x03B5; a#x03B4;)-minimum cluster composition. An (a#x03B5; a#x03B4;)-minimum cluster composition specifies a cluster composition whose cost approximates that of the minimum cluster composition (i.e., the cluster composition that incurs the minimum cloud usage cost that must be incurred in executing a given application under an SLO deadline), the approximation bounds the error to a predefined threshold a#x03B5; with a degree of confidence 100 * (1 - a#x03B4;)%. The degree of confidence 100 * (1 - a#x03B4;)% specifies that the probability of failure in achieving the error threshold a#x03B5; for the above approximation is at most a#x03B4;. For a#x03B5; = 0.1 and a#x03B4; = 0.02, we experimentally demonstrate that an (a#x03B5;a#x03B4;)-minimum cluster composition predicted by Infra successfully approximates the minimum cluster composition, i.e., the accuracy of prediction of minimum cluster composition ranges from 93.1% to 97.99% (the error is bound by the error threshold of 0.1) with a 98% degree of confidence, since 100* (1 - a#x03B4;) = 98%. Auto scaling refers to the process of automatically adding cloud instances to a cluster to adapt to an increase in application workload (increased request rate), and deleting instances from a cluster when there is a decrease in workload (reduced request rate). However, state-of-the-art auto scaling techniques have the following disadvantages: A) they require explicit policy definition for changing the cluster configuration and therefore lack the ability to automatically adapt a cluster with respect to changing workload, B) they do not compute the appropriate size of resources required, and therefore do not result in an optimal cluster composition. Infra provides an auto scaler that automatically adapts a cloud infrastructure to changing application workload, scaling the cluster up/down based on predictions from the Infra provisioning tool.\",\n \"Title: Novel Fast User-Placement Ushering Algorithms for Indoor Femtocell Networks.\\nYear: 2016\\n\\nAbstract: Nowadays, the sufficient quality-of-service (QoS) provision for mobile applications remains a major challenge in any wireless network. Conventional sufficient QoS provision techniques using resource allocation, data scheduling, and crosslayer optimization have been proposed to tackle this problem. Nevertheless, due to the unpredictable nature of wireless channel conditions, the QoS improvements resulting from the aforementioned approaches are often unsatisfactory. In this paper, we would like to follow our previous concept, namely user-placement ushering (UPU), by use of the user's mobility. A mobile user can be relocated to an optimal, or at least a better place to boost the QoS for a particular application. Novel fast algorithms are devised here to usher (guide) the mobile user to an appropriate spot with sufficient QoS, which is nearby. We address the UPU problem to accommodate realistic complex indoor environments, where obstacles are present. Our simulation results have demonstrated that our new fast UPU algorithms are capable of finding new appropriate locations satisfying the required QoSs for different wireless network applications.\",\n \"Title: Automated diagnostics for manufacturing machinery based on well-regularized deep neural networks.\\nYear: 2017\\n\\nAbstract: This paper presents a bigdata framework based on regularized deep neural networks for automated diagnostics for manufacturing machinery based on emitted sound, vibration, and magnetic field data. More precisely, we present SpotCheck, a prototype system that uses well-regularized deep neural networks to analyze sound, vibrational, and magnetic emissions of industrial machinery to provide noninvasive machine diagnostics, both for fault detection and to meter the day to day mode of operation of the machinery. It is completely automatic requiring no manual extraction of handcrafted features. It can operate with relatively small amounts of training data, but can take advantage of large volumes of unlabeled data when available, and scale to very large volumes of labeled or unlabeled data to improve performance as more data becomes available after deployment.\",\n \"Title: StaticGen: Static Generation of UML Sequence Diagrams.\\nYear: 2017\\n\\nAbstract: UML sequence diagrams are visual representations of object interactions in a system and can provide valuable information for program comprehension, debugging, maintenance, and software archeology. Sequence diagrams generated from legacy code are independent of existing documentation that may have eroded. We present a framework for static generation of UML sequence diagrams from object-oriented source code. The framework provides a query refinement system to guide the user to interesting interactions in the source code. Our technique involves constructing a hypergraph representation of the source code, traversing the hypergraph with respect to a user-defined query, and generating the corresponding set of sequence diagrams. We implemented our framework as a tool, StaticGen, analyzing a corpus of 30 Android applications. We provide experimental results demonstrating the efficacy of our technique.\",\n \"Title: Brief Announcement: Complete Visibility for Oblivious Robots in Linear Time.\\nYear: 2017\\n\\nAbstract: We consider the distributed setting of $N$ autonomous mobile robots that operate in Look-Compute-Move cycles following the well-celebrated classic oblivious robots model. We study the fundamental problem where starting from an arbitrary initial configuration, N autonomous robots reposition themselves to a convex hull formation on the plane where each robot is visible to all others (the Complete Visibility problem). We assume obstructed visibility, where a robot cannot see another robot if a third robot is positioned between them on the straight line connecting them. We provide the first \\\\\\\\cO(N) time algorithm for this problem in the fully synchronous setting. Our contribution is a significant improvement over the runtime of the only previously known algorithm for this problem which has a lower bound of \\\\\\\\Omega(N^2). Our proposed algorithm is collision-free -- robots do not share positions and their paths do not cross.\",\n \"Title: Novel electronic scissoring algorithm\\nYear: 2017\\n\\nAbstract: The electronic media, such as digital magazines and newspapers, are becoming more and more popular nowadays. The computer-aided information retrieval of such electronic documents is crucial for any further data-mining applications. In this paper, we proposed a novel electronic scissoring algorithm to automatically extract the image content from an arbitrary electronic document. Our proposed new algorithm contains two steps, namely the information-content extraction and the image-content detection. In the first step, everything other than the background is extracted using a computationally-efficient variance filter. The information content usually consists of text, image, and delimiters. The next step, the entropy measure is employed to distinguish the images from other content. Experiments are carried out using data captured from random web-pages. The receiver-operating characteristic curves have been presented to demonstrate the effectiveness of our proposed new algorithm.\",\n \"Title: SynAdapt: Automated Synthesis of Adaptive Agents\\nYear: 2017\\n\\nAbstract: Distributed autonomous multi-agent reasoning and classification systems have been thought of to be the basis of intelligence and have wide applications in the space of operational intelligence in closing the loop between sensing, analytics, and actions. This paper targets multi-agent systems that employ rulebased logics (i.e., rules that determine the output/response of an agent depending on the range of the input values) with pre-defined rules to accurately perceive the environment, and provide associated reactions. Such rule-based systems do not perform well in scenarios, where human generated rules cannot adapt to dynamic variations in the data distribution arising due to dynamic changes in the environment, especially if data dimensionality is very high. Examples of such scenarios exist wherever the sensed data arrives from the physical world - such as weather data, physical sensor data, human behaviour controlled data, etc. Clearly, to meet the adaptivity requirements of such scenarios we require the agents to possess adaptive reasoning capability such that they can adapt the underlying rules with respect to the changing environment. Developing such adaptive agents requires the developer to additionally possess considerable expertise of state-of-the-art machine learning techniques, apart from possessing knowledge of the agent's target domain. To address the above issues, we automate the process of development and deployment of adaptive agents. We present the fundamental design concepts behind the development of SynAdapt: a new adaptive meta-learning based multi-agent synthesis framework, that automates the synthesis of adaptive multi-agent systems from high-level user specifications. SynAdapt provides the following key features: a) Automated synthesis and deployment of adaptive agents from high-level user specification, b) Agents synthesised by SynAdapt can select a learning strategy that is particularly suited for given user specifications and input dataset, and c) Agents synthesised by SynAdapt can leverage adaptive ensemble learning techniques to deal with concept drift.\",\n \"Title: Adaptable SLA-Aware Consistency Tuning for Quorum-Replicated Datastores.\\nYear: 2017\\n\\nAbstract: Users of distributed datastores that employ quorum-based replication are burdened with the choice of a suitable client-centric consistency setting for each storage operation. The above matching choice is difficult to reason about as it requires deliberating about the tradeoff between the latency and staleness, i.e., how stale (old) the result is. The latency and staleness for a given operation dep...\",\n \"Title: Pixel-Level Reconstruction and Classification for Noisy Handwritten Bangla Characters\\nYear: 2018\\n\\nAbstract: Classification techniques for images of handwritten characters are susceptible to noise. Quadtrees can be an efficient representation for learning from sparse features. In this paper, we improve the effectiveness of probabilistic quadtrees by using a pixel level classifier to extract the character pixels and remove noise from handwritten character images. The pixel level denoiser (a deep belief network) uses the map responses obtained from a pretrained CNN as features for reconstructing the characters eliminating noise. We experimentally demonstrate the effectiveness of our approach by reconstructing and classifying a noisy version of handwritten Bangla Numeral and Basic Character datasets.\",\n \"Title: Deep neural networks for texture classification-A theoretical analysis.\\nYear: 2018\\n\\nAbstract: We investigate the use of Deep Neural Networks for the classification of image datasets where texture features are important for generating class-conditional discriminative representations. To this end, we first derive the size of the feature space for some standard textural features extracted from the input dataset and then use the theory of Vapnik–Chervonenkis dimension to show that hand-crafted feature extraction creates low-dimensional representations which help in reducing the overall excess error rate. As a corollary to this analysis, we derive for the first time upper bounds on the VC dimension of Convolutional Neural Network as well as Dropout and Dropconnect networks and the relation between excess error rate of Dropout and Dropconnect networks. The concept of intrinsic dimension is used to validate the intuition that texture-based datasets are inherently higher dimensional as compared to handwritten digits or other object recognition datasets and hence more difficult to be shattered by neural networks. We then derive the mean distance from the centroid to the nearest and farthest sampling points in an n-dimensional manifold and show that the Relative Contrast of the sample data vanishes as dimensionality of the underlying vector space tends to infinity.\",\n \"Title: Consistify - preserving correctness and SLA under weak consistency.\\nYear: 2019\\n\\nAbstract: While executing a client application, comprising a sequence of one or more storage operations (read, write, etc.), replicated datastores provide the option of tuning per-operation client-centric consistency settings, Weaker consistency settings allow client applications to execute with lower observed latency due to less coordination among the replica, improving throughput. To enable a client application running on a geo-replicated datastore to meet stringent SLA (Service Level Agreement) deadlines, i.e., latency thresholds provided in the SLA, users apply weak client-centric consistency settings. Strong consistency settings may result in higher latency due to coordination among greater number of replicas. However, since a weak consistency setting requires only a partial quorum of replicas to coordinate, some of the replicas may be overwritten by conflicting updates from concurrent clients, which, in turn, may result in violation of correctness conditions specified by the developer. Such correctness conditions are provided as postconditions that impose restrictions on the values observed by the clients. Enforcing correctness conditions under weak consistency settings on geo-replicated datastores from the client application code is a tedious and challenging task. The task is even more complicated when considering application-level SLA deadlines. Deadline (latency) is one of the most important SLA parameters for optimizing client-centric performance. To this end, we present Consistify, a novel framework, that acts as an interface layer between client applications and underlying datastores, and enables client applications to execute with weak per-operation consistency settings, while simultaneously respecting: 1) the correctness conditions, specified using logical predicates, and 2) the SLA deadline. Using benchmark workloads, we experimentally demonstrate that Consistify outperforms the state-of-the-art systems, namely QUELA. For several other systems that provide flexible consistency settings, namely Bolt-on causal system, SIEVE, and Cure, we provide a comparison of the throughput and latency observed with Consistify against the numbers reported with those systems in prior papers.\",\n \"Title: Progressively Growing Generative Adversarial Networks for High Resolution Semantic Segmentation of Satellite Images.\\nYear: 2019\\n\\nAbstract: Machine learning has proven to be useful in classification and segmentation of images. In this paper, we evaluate a training methodology for pixel-wise segmentation on high resolution satellite images using progressive growing of generative adversarial networks. We apply our model to segmenting building rooftops and compare these results to conventional methods for rooftop segmentation. We present our findings using the SpaceNet version 2dataset. Progressive GAN training achieved a test accuracy of 93% compared to 89% for traditional GAN training.\",\n \"Title: Model-Based static code analysis for MATLAB models\\nYear: 2012\\n\\nAbstract: MATLAB is widely used in scientific, engineering, and numerical computations. Complex systems such as digital signal processors, process control systems, etc. are modeled in MATLAB and analyzed; C implementation of the system can be automatically generated from the validated MATLAB model. We combine static analysis techniques with model-based deductive verification using SMT solvers to provide a framework to analyze MATLAB code. The analyzer is generated by translating the collecting semantics of a MATLAB script to a formula in first order logic over multiple underlying theories. Function calls in a script can be handled by importing SMT assertions obtained by analyzing MATLAB files containing function definitions. Logical specification of the desired program behavior (rather its negation) is incorporated as a first order logic formula. An SMT-LIB formula solver treats the combined formula as a \\\"constraint\\\" and \\\"solves\\\" it. The \\\"solved form\\\" can be used to identify logical errors in the MATLAB model.\",\n \"Title: Break Down in Order To Build Up: Decomposing Small Molecules for Fragment-Based Drug Design with eMolFrag.\\nYear: 2017\\n\\nAbstract: Constructing high-quality libraries of molecular building blocks is essential for successful fragment-based drug discovery. In this communication, we describe eMolFrag, a new open-source software to decompose organic compounds into nonredundant fragments retaining molecular connectivity information. Given a collection of molecules, eMolFrag generates a set of unique fragments comprising larger moieties, bricks) and smaller linkers connecting bricks. These building blocks can subsequently be used to, construct virtual screening libraries for targeted drug discovery: The robustness and computational performance of eMolFrag is assessed against the Directory of Useful Decoys, Enhanced database conducted in serial and parallel modes with up to 16 computing cores. Further, the application of eMolFrag in de novo drug design is illustrated using the adenosine receptor.\",\n \"Title: The Complete Visibility Problem for Fat Robots with Lights.\\nYear: 2018\\n\\nAbstract: We consider the distributed setting of N autonomous mobile robots that operate in Look-Compute-Move cycles and communicate with other robots using colored lights (the robots with lights model). We study the fundamental problem of repositioning N autonomous robots on a plane so that each robot is visible to all others (the COMPLETE VISIBILITY problem) on this model. We consider obstructed visibility under which a robot cannot see another robot if a third robot is positioned between then on the straight line connecting them. The literature solves this problem on the robots with lights model for point robots. In this paper, we study this problem, for the very first time, on the robots with lights model for fat robots, i.e., the robots which have physical extents. In particular, we present and analyze an algorithm for the fat robots of unit disc size extents that solves COMPLETE VISIBILITY in O(N) time using 9 colors in the fully synchronous setting avoiding robot collisions.\",\n \"Title: Why do you take that route?\\nYear: 2019\\n\\nAbstract: The purpose of this paper is to determine whether a particular context factor among the variables that a researcher is interested in causally affects the route choice behavior of drivers. To our knowledge, there is limited literature that consider the effects of various factors on route choice based on causal inference.Yet, collecting data sets that are sensitive to the aforementioned factors are challenging and the existing approaches usually take into account only the general factors motivating drivers route choice behavior. To fill these gaps, we carried out a study using Immersive Virtual Environment (IVE) tools to elicit drivers' route choice behavioral data, covering drivers' network familiarity, educationlevel, financial concern, etc, apart from conventional measurement variables. Having context-aware, high-fidelity properties, IVE data affords the opportunity to incorporate the impacts of human related factors into the route choice causal analysis and advance a more customizable research tool for investigating causal factors on path selection in network routing. This causal analysis provides quantitative evidence to support drivers' diversion decision.\",\n \"Title: Why Did You Turn On That Light?\\nYear: 2021\\n\\nAbstract: The overarching goal of this paper is to study if certain contextual factors out of the variables that a researcher is interested in has causal influence on human-building interaction for a building under design. To limit the scope of the study, we concentrate on causal analysis of light switch behavior (i.e., an occupant switching on or off a light) in office buildings. Due to the lack of high-fi...\",\n \"Title: Robustness analysis framework for computations associated with building performance models and immersive virtual experiments\\nYear: 2021\\n\\nAbstract: Building performance models (BPMs) have been used to simulate and analyze building performance during design. While extensive research efforts have made to improve the performance of BPMs, little attention has given to their robustness. Uncertainty is a crucial factor affecting the robustness of BPMs, in which such effect needs to be quantified through a suitable approach. The paper offers a robustness analysis framework for BPMs by using perturbation techniques to simulate uncertainty in input datasets. To investigate the efficacy of the framework, a generative adversarial network (GAN)-based framework was selected as a case study to analyze light switch usages in a single-occupancy office simulated using an immersive virtual environment (IVE). The robustness of the GAN was analyzed by comparing differences between a baseline (i.e., a BPM obtained from the GAN trained on a non-perturbed dataset) and BPMs obtained from the GAN trained on perturbed datasets. Overall, the robustness of the GAN significantly reduced when the training datasets were perturbed by using structured transformation techniques. The GAN remained relatively robust when the training datasets were perturbed by using an additive perturbation. Additionally, the sensitivity of the GAN involves different magnitudes corresponding to different levels of perturbed input datasets. The study suggests that the perturbation analysis is effective for investigating data uncertainty affecting the robustness of BPMs.\",\n \"Title: SimilarityGAN: Using Similarity to Loosen Structural Constraints in Generative Adversarial Models\\nYear: 2021\\n\\nAbstract: Recently, generative adversarial networks have performed extremely well in image translation. When translating images current models adhere to a strict structural symmetry between the input and output images. This paper, presents a technique for image translation involving a pair of image domains that allows the output image to go beyond the structural symmetry constraints imposed by the input. By...\"\n]","isConversational":false},"profile_size":{"kind":"number","value":67,"string":"67"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13588,"string":"13,588"}}},{"rowIdx":13589,"cells":{"question_id":{"kind":"string","value":"Ting Yu Chen"},"question":{"kind":"string","value":"Generate an abstract for the title \"Application of Blockchain Technology to Podcast-Based Enterprise Content Marketing\" using the following items: 1. Blockchain \n2. Podcasts \n3. Advertising \n4. Decentralization \n5. Smart contracts"},"answer":{"kind":"string","value":"This article explores the potential of blockchain technology to enhance podcast content marketing. In the post-pandemic era, more and more enterprises are using podcasts for content marketing. However, the rapid growth of the podcast industry has also raised concerns. This study looks at current podcast advertising methods and uses a 10-step process to examine the need for a blockchain platform. With the decentralization, transparency and traceability of the blockchain and five smart contracts, the new distributed podcast platform improves content marketing metrics and reduces transaction costs. It provides enterprises with a more robust podcast platform, as well as more opportunities for niche creators to showcase and optimize the podcast platform as a useful advertising and marketing tool."},"correct_answer":{"kind":"string","value":"This article explores the potential of blockchain technology to enhance podcast content marketing. In the post-pandemic era, more and more enterprises are using podcasts for content marketing. However, the rapid growth of the podcast industry has also raised concerns. This study looks at current podcast advertising methods and uses a 10-step process to examine the need for a blockchain platform. With the decentralization, transparency and traceability of the blockchain and five smart contracts, the new distributed podcast platform improves content marketing metrics and reduces transaction costs. It provides enterprises with a more robust podcast platform, as well as more opportunities for niche creators to showcase and optimize the podcast platform as a useful advertising and marketing tool."},"documents":{"kind":"list like","value":["Title: Experimental Analysis Of Multi-Attribute Decision-Making Based On Atanassov Intuitionistic Fuzzy Sets: A Discussion Of Anchor Dependency And Accuracy Functions\nYear: 2012\n\nAbstract: This article presents a useful method for relating anchor dependency and accuracy functions to multiple attribute decision-making (MADM) problems in the context of Atanassov intuitionistic fuzzy sets (A-IFSs). Considering anchored judgement with displaced ideals and solution precision with minimal hesitation, several auxiliary optimisation models have proposed to obtain the optimal weights of the attributes and to acquire the corresponding TOPSIS (the technique for order preference by similarity to the ideal solution) index for alternative rankings. Aside from the TOPSIS index, as a decision-maker's personal characteristics and own perception of self may also influence the direction in the axiom of choice, the evaluation of alternatives is conducted based on distances of each alternative from the positive and negative ideal alternatives, respectively. This article originates from Li's [Li, D.-F. (2005), 'Multiattribute Decision Making Models and Methods Using Intuitionistic Fuzzy Sets', Journal of Computer and System Sciences, 70, 73-85] work, which is a seminal study of intuitionistic fuzzy decision analysis using deduced auxiliary programming models, and deems it a benchmark method for comparative studies on anchor dependency and accuracy functions. The feasibility and effectiveness of the proposed methods are illustrated by a numerical example. Finally, a comparative analysis is illustrated with computational experiments on averaging accuracy functions, TOPSIS indices, separation measures from positive and negative ideal alternatives, consistency rates of ranking orders, contradiction rates of the top alternative and average Spearman correlation coefficients.","Title: A note on distances between intuitionistic fuzzy sets and/or interval-valued fuzzy sets based on the Hausdorff metric\nYear: 2007\n\nAbstract: In this note, we point out some errors to Lemma 4 in Grzegorzewski [Distances between intuitionistic fuzzy sets and/or interval-valued fuzzy sets based on the Hausdorff metric, Fuzzy Sets and Systems 148, 319-328] by counterexamples. We find that the inequalities of Euclidean distances and normalized Euclidean distances for intuitionistic fuzzy sets are not valid.","Title: A linear assignment method for multiple-criteria decision analysis with interval type-2 fuzzy sets\nYear: 2013\n\nAbstract: The theory of interval type-2 fuzzy sets is valuable for modeling impressions in multiple-criteria decision analysis problems. Based on signed distances, we develop a new linear assignment method to produce an optimal preference ranking of the alternatives in accordance with a set of criterion-wise rankings and a set of criterion importance within the context of interval type-2 trapezoidal fuzzy numbers. Applying the proposed method to a case involving the selection of a landfill site, we demonstrate that the proposed method is easy to employ and that it produces actionable results that aid the decision-making process. The proposed interval type-2 fuzzy linear assignment method utilizes signed distances, does not require a complicated computation procedure and will be beneficial to decision analysis.","Title: An ELECTRE-based outranking method for multiple criteria group decision making using interval type-2 fuzzy sets\nYear: 2014\n\nAbstract: The aim of this paper is to develop an ELECTRE (ELimination Et Choice Translating REality)-based outranking method for multiple criteria group decision-making within the environment of interval type-2 fuzzy sets. Along with considering the context of interval type-2 trapezoidal fuzzy numbers, this paper employs a hybrid averaging approach with signed distances to construct a collective decision matrix and proposes the use of ELECTRE-based outranking methods to analyze the collective interval type-2 fuzzy data. By applying a signed distance approach, this work identifies the concordance and discordance sets to determine the concordance and discordance indices, respectively, for each pair of alternatives. Based on an aggregate outranking matrix, a decision graph is constructed to determine the partial-preference ordering of the alternatives and the ELECTREcally non-outranked solutions. This paper provides additional approaches at the final selection stage to yield a linear ranking order of the alternatives. The feasibility and applicability of the proposed methods are illustrated with an example that addresses supplier selection, and a comparative analysis is performed with other approaches to validate the effectiveness of the proposed methodology.","Title: Objective weights with intuitionistic fuzzy entropy measures and computational experiment analysis\nYear: 2011\n\nAbstract: It is important to properly assess the weights of attributes when solving multi-attribute decision problems because variations in the weights often influence the rankings of the alternatives. In this paper, we propose an alternative objective weighting method to generate objective weights based on intuitionistic fuzzy (IF) entropy measures, which depend on the nature of a decision matrix in an intuitionistic fuzzy environment. Instead of the traditional fuzzy entropy measure, which is characterized by using the discriminating power to calculate the attribute weights, the proposed approach adopts the IF entropy measure, which emphasizes the credibility of the data. The IF entropy measure applied here is derived from a geometric interpretation of intuitionistic fuzzy sets and incorporates the concept of a ratio of distance measures. We implement four distance measures in the proposed approach and compare them in a computational experiment. The experimental results indicate that different IF entropy measures used in weighting methods can generate distinct objective attribute weights. In particular, when the number of attributes increases, the discrepancy between the IF entropy measures increases.","Title: Identification Of General Fuzzy Measures By Genetic Algorithms Based On Partial Information\nYear: 2000\n\nAbstract: This study develops an identification procedure for general fuzzy measures using genetic algorithms, In view of the difficulty in data collection in practice, the amount of input data is simplified through a sampling procedure concerning attribute subsets, and the corresponding detail design is adapted to the partial information acquired by the procedure. A specially designed genetic algorithm is proposed for better identification, including the development of the initialization procedure, fitness function, and three genetic operations. To show the applicability of the proposed method, this study simulates a set of experimental data that are representative of several typical classes. The experimental analysis indicates that using genetic algorithms to determine general fuzzy measures can obtain satisfactory results under the framework of partial information.","Title: Expanding competence sets for the consumer decision problem\nYear: 2002\n\nAbstract: The framework of competence set analysis provides a new approach to complement the existing models for the consumer decision problem. Since it is essential for the firm to improve the competence set of its product or service to fully address the consumer's truly needed benefits, the effective expansion of competence sets plays an important role in marketing reality. The previous studies regarding competence set expansion have thrown light on the tree expansion processes, but forest learning is more suitable for the acquisition of product benefits than tree learning. Thus, this study looses the assumption of tree learning and conducts forest learning to design an effective expansion program. In addition, this study explores a more general problem involving intermediate attributes, compound benefits, and experiential effects. An algorithm is also provided for effective expansion of competence sets in consumer decision analysis.","Title: A signed-distance-based approach to importance assessment and multi-criteria group decision analysis based on interval type-2 fuzzy set.\nYear: 2013\n\nAbstract: Interval type-2 fuzzy sets are associated with greater imprecision and more ambiguities than ordinary fuzzy sets. This paper presents a signed-distance-based method for determining the objective importance of criteria and handling fuzzy, multiple criteria group decision-making problems in a flexible and intelligent way. These advantages arise from the method’s use of interval type-2 trapezoidal fuzzy numbers to represent alternative ratings and the importance of various criteria. An integrated approach to determine the overall importance of the criteria is also developed using the subjective information provided by decision-makers and the objective information delivered by the decision matrix. In addition, a linear programming model is developed to estimate criterion weights and to extend the proposed multiple criteria decision analysis method. Finally, the feasibility and effectiveness of the proposed methods are illustrated by a group decision-making problem of patient-centered medicine in basilar artery occlusion.","Title: Using competence sets to analyze the consumer decision problem\nYear: 2001\n\nAbstract: This study is intended to provide a different approach to complement the existing consumer decision models (CDMs). Based on the concept of habitual domains and competence sets, we supply a framework for helping a firm in expanding the benefits of its products to fully address the consumer’s needs. According to the features of consumers’ decision making, we use challenging problem types to explore extensive problem solving, fuzzy problem types for limited problem solving, and routine and mixed routine problem types for routine problem solving. In addition, several useful indexes are established using fuzzy measures in this study, including the possibility of successfully appealing to consumers, the degree of consumer satisfaction, the degree of compatibility, and the degree of uniqueness. These indexes can be a decision support for implementing competence set analysis in practical applications. Finally, an empirical study on children’s apparel was conducted to show the applicability and feasibility of our proposed method in practice.","Title: Comparative analysis of SAW and TOPSIS based on interval-valued fuzzy sets: Discussions on score functions and weight constraints\nYear: 2012\n\nAbstract: Interval-valued fuzzy sets involve more uncertainties than ordinary fuzzy sets and can be used to capture imprecise or uncertain decision information in fields that require multiple-criteria decision analysis (MCDA). This paper takes the simple additive weighting (SAW) method and the technique for order preference by similarity to an ideal solution (TOPSIS) as the main structure to deal with interval-valued fuzzy evaluation information. Using an interval-valued fuzzy framework, this paper presents SAW-based and TOPSIS-based MCDA methods and conducts a comparative study through computational experiments. Comprehensive discussions have been made on the influence of score functions and weight constraints, where the score function represents an aggregated effect of positive and negative evaluations in performance ratings and the weight constraint consists of the unbiased condition, positivity bias, and negativity bias. The correlations and contradiction rates obtained in the experiments suggest that evident similarities exist between the interval-valued fuzzy SAW and TOPSIS rankings.","Title: The extended QUALIFLEX method for multiple criteria decision analysis based on interval type-2 fuzzy sets and applications to medical decision making.\nYear: 2013\n\nAbstract: QUALIFLEX, a generalization of Jacquet-Lagreze's permutation method, is a useful outranking method in decision analysis because of its flexibility with respect to cardinal and ordinal information. This paper develops an extended QUALIFLEX method for handling multiple criteria decision-making problems in the context of interval type-2 fuzzy sets. Interval type-2 fuzzy sets contain membership values that are crisp intervals, which are the most widely used of the higher order fuzzy sets because of their relative simplicity. Using the linguistic rating system converted into interval type-2 trapezoidal fuzzy numbers, the extended QUALIFLEX method investigates all possible permutations of the alternatives with respect to the level of concordance of the complete preference order. Based on a signed distance-based approach, this paper proposes the concordance/discordance index, the weighted concordance/discordance index, and the comprehensive concordance/discordance index as evaluative criteria of the chosen hypothesis for ranking the alternatives. The feasibility and applicability of the proposed methods are illustrated by a medical decision-making problem concerning acute inflammatory demyelinating disease, and a comparative analysis with another outranking approach is conducted to validate the effectiveness of the proposed methodology. (C) 2012 Elsevier B.V. All rights reserved.","Title: Data Construction Process And Qualiflex-Based Method For Multiple-Criteria Group Decision Making With Interval-Valued Intuitionistic Fuzzy Sets\nYear: 2013\n\nAbstract: Based on Jacquet-Lagreze's permutation method, QUALIFLEX is an outranking model that investigates all possible permutations of alternatives with respect to the consequences of all criteria. The purpose of this paper is to develop a QUALIFLEX-based method for multiple criteria group decision making within a decision environment of interval-valued intuitionistic fuzzy sets. We conduct a statistical inference approach with finite population correction to construct interval-valued intuitionistic fuzzy numbers. In addition, we incorporate the relative importance of decision makers and fuse individual opinions to form collective ratings using a modified method with weighted interval estimations. In view of diversiform preference types (weak order, strict order, difference order, interval bound, and ratio bound), we represent multiple decision makers' various forms of preference structures and assess criterion weights under incomplete information. By means of score functions, accuracy functions, membershipuncertainty indices, and hesitation-uncertainty indices, a ranking procedure is employed to identify a criterion-wise preference of alternatives. A QUALIFLEX-based model is then established to measure the level of concordance of the complete preference order for handling multiple criteria group decisions. The feasibility of the proposed method is illustrated by a practical problem relating to the selection of a landfill site. As indicated in the application, the proposed method is useful for handling complicated group decision-making problems that involve comprehensive criteria and limited alternatives.","Title: A PROMETHEE-based outranking method for multiple criteria decision analysis with interval type-2 fuzzy sets\nYear: 2014\n\nAbstract: This paper develops new methods based on the preference ranking organization method for enrichment evaluations (PROMETHEE) that use a signed distance-based approach within the environment of interval type-2 fuzzy sets for multiple criteria decision analysis. The theory of interval type-2 fuzzy sets provides an intuitive and computationally feasible way of addressing uncertain and ambiguous information in decision-making fields. Many studies have developed multiple criteria decision analysis methods in the context of interval type-2 fuzzy sets; most of these methods can be characterized as scoring or compromising models. Nevertheless, the extended versions of outranking methods have not been thoroughly investigated. This paper establishes interval type-2 fuzzy PROMETHEE methods for ranking alternative actions among multiple criteria based on the concepts of signed distance-based generalized criteria and comprehensive preference indices. We develop interval type-2 fuzzy PROMETHEE I and interval type-2 fuzzy PROMETHEE II procedures for partial and complete ranking, respectively, of the alternatives. Finally, the feasibility and applicability of the proposed methods are illustrated by a practical problem of landfill site selection. A comparative analysis is also performed with ordinary fuzzy PROMETHEE methods to validate the effectiveness of the proposed methodology.","Title: Using fuzzy measures and habitual domains to analyze the public attitude and apply to the gas taxi policy\nYear: 2002\n\nAbstract: Public acceptance and support are the crucial keys for implementing public policies successfully. Thus, the understanding of public acceptance or rejection towards the policy, as well as the important attributes of concern, could be very helpful to implementing the policy. However, most conventional attitude models could not approximate people's subjective evaluation process exactly by virtue of the additivity and independence assumptions. Additionally, people's decision behavior is deeply affected by their existing habits. Since habitual domains exist in the decision process, if the government can change or extend people's habitual thinking in favor of the public policy, the policy will receive satisfactory acceptance. Therefore, this study uses the habitual domain theory to analyze the public's attitude towards public policies. Furthermore, general fuzzy measures and fuzzy integrals, which require only boundary conditions and monotonicity, are also applied to develop a public attitude analysis model. An empirical study on the compress natural gas (CNG) taxi policy in Taipei City is conducted to show the applicability of the proposed model. The empirical results indicate that there are significant differences between the public's concern and governmental publicity, and some valuable strategies are suggested to the government.","Title: Approaches To Leniency Reduction In Multi-Criteria Decision Making With Interval-Valued Fuzzy Sets And An Experimental Analysis\nYear: 2012\n\nAbstract: The purpose of this paper is to present a useful method for estimating the importance of criteria and reducing the leniency bias in multiple criteria decision analysis based on interval-valued fuzzy sets. Several types of net predispositions are defined to represent an aggregated effect of interval-valued fuzzy evaluations. The suitability function for measuring the overall evaluation of each alternative is then determined based on simple additive weighting (SAW) methods. Because positive or negative leniency may exist when most criteria are assigned unduly high or low ratings, respectively, deviation variables are introduced to mitigate the effects of overestimated and underestimated ratings of criterion importance. Considering the two objectives of maximal weighted suitability and minimal deviation values, an integrated programming model is proposed to compute optimal weights for the criteria and corresponding suitability degrees for alternative rankings. A flexible algorithm using interval-valued fuzzy SAW methods is established by considering both objective and subjective information to compute optimal multiple criteria decisions. The feasibility and effectiveness of the proposed methods are illustrated by a numerical example. Finally, an experimental analysis of interval-valued fuzzy rankings given different conditions for the criterion weights is conducted with discussions on average Spearman correlation coefficients and contradiction rates.","Title: An interval-valued intuitionistic fuzzy LINMAP method with inclusion comparison possibilities and hybrid averaging operations for multiple criteria group decision making\nYear: 2013\n\nAbstract: An interval-valued intuitionistic fuzzy set contains membership and non-membership values that are crisp intervals and represent an extension of the ordinary fuzzy sets that are widely used because of their usefulness in handling imprecise or uncertain information. The linear programming technique for multidimensional analysis of preference (LINMAP) is a representative decision-making method with respect to preference information for given alternatives. In this paper, we present a new linear programming technique with weight assessment, an extended LINMAP method, for addressing multiple criteria group decision-making problems in the interval-valued intuitionistic fuzzy framework. With consideration given to the degrees of relative agreement and the importance weights of multiple decision makers, this paper presents an inclusion-based hybrid averaging operation with an inclusion comparison approach for forming a collective decision environment. The concept of inclusion-based indices that relate to anchor dependency with multiple points of reference is developed as the core of the extended LINMAP method. We also establish a linear programming model to handle the incomplete preference information for alternatives. The optimal weights of the criteria can be determined, and the priority order of the alternatives can be obtained according to the resulting comprehensive inclusion-based indices. The feasibility and applicability of the proposed methods are illustrated with an example addressing graduate admission, and a comparative analysis is performed with another interval-valued intuitionistic fuzzy LINMAP approach to validate the effectiveness of the proposed methodology.","Title: Interval-valued intuitionistic fuzzy QUALIFLEX method with a likelihood-based comparison approach for multiple criteria decision analysis\nYear: 2014\n\nAbstract: QUALIFLEX (i.e., QUALItative FLEXible multiple criteria method) is a useful outranking method used for multiple criteria decision analysis. This paper uses the main structure of QUALIFLEX to develop an interval-valued intuitionistic fuzzy QUALIFLEX outranking method with a likelihood-based comparison approach for handling multiple criteria decision-making problems within a decision environment of interval-valued intuitionistic fuzzy sets. We propose the concept of using the likelihood of fuzzy preference relations to compare interval-valued intuitionistic fuzzy numbers. To address diversiform preference types, we represent the decision-maker's various forms of preference structures and assess the criterion weights using incomplete information. Using a criterion-wise preference of alternatives via comparison of the likelihoods, we develop a new QUALIFLEX-based model to measure the level of concordance of the complete preference order for managing multiple criteria decisions. The feasibility and applicability of the proposed methods are illustrated using a practical example, namely, the selection of a suitable bridge construction method. A comparative analysis with other relevant methods is conducted to validate the effectiveness of the proposed methodology.","Title: A multicriteria group decision-making approach based on interval-valued intuitionistic fuzzy sets: A comparative perspective\nYear: 2011\n\nAbstract: This study develops an approach to tackle multiple criteria group decision-making problems in the context of interval-valued intuitionistic fuzzy sets. Due to conflicting evaluations and insufficient information about the criteria, an interval-valued intuitionistic fuzzy preference relation matrix is employed to determine the relative importance of criteria in terms of pairwise comparisons. The decision matrix, which indicates the degree of alternatives with respect to each criterion, is expressed by interval-valued intuitionistic fuzzy numbers (IVIFNs). In order to integrate interval-valued intuitionistic fuzzy information, some special aggregation operators are created by altering the aggregation operation of IVIFNs. The three families of parametric fuzzy unions and fuzzy intersections are applied in the aggregation operation with comparisons of the ranking results of alternatives. With a linear programming method, the proposed approach uses an optimization model to obtain criterion weights in exact numbers rather than intervals, and subsequently calculates an aggregated IVIFN for each alternative. The score function and accuracy function assist in discriminating between the aggregated IVIFNs, and in generating a final rank of alternatives. Finally, an illustrative supplier selection problem is used to demonstrate how to apply the proposed approach and to observe the computational consequences resulting from various aggregation operators. The results reveal that the parameters of the aggregation operators indeed have an impact on the ranks of alternatives.","Title: A multimeasure approach to optimism and pessimism in multiple criteria decision analysis based on Atanassov fuzzy sets\nYear: 2011\n\nAbstract: The frequently used methods of handling optimism and pessimism in multiple criteria decision analysis are Hurwicz's and Yager's approaches. Despite their wide usage, a critical issue has been raised about unidimensionality. Numerous psychological researches and empirical findings have convincingly supported that optimism and pessimism do not represent opposite poles on a single, bipolar dimension, but they are conceived as two partially independent dimensions instead. The focus of this paper is a two-dimensional approach by adequately employing the degrees of membership and non-membership based on Atanassov fuzzy sets. This study develops optimistic and pessimistic estimations with several fuzzy point operators to draw the influences of optimism and pessimism on multicriteria decision making for the sake of a better fit than the unidimensional model. Given a typology from empirical grounds, the appropriate point operators specific to each type are suggested to identify adaptational outcomes. Base upon a series of new score functions, a useful method for multiple criteria decision analysis is provided to efficiently facilitate decision analysis with a multimeasure approach. Feasibility and effectiveness of the proposed methods are illustrated by a practical example.","Title: An interactive method for multiple criteria group decision analysis based on interval type-2 fuzzy sets and its application to medical decision making\nYear: 2013\n\nAbstract: The theory of interval type-2 fuzzy sets provides an intuitive and computationally feasible way of addressing uncertain and ambiguous information in decision-making fields. The aim of this paper is to develop an interactive method for handling multiple criteria group decision-making problems, in which information about criterion weights is incompletely (imprecisely or partially) known and the criterion values are expressed as interval type-2 trapezoidal fuzzy numbers. With respect to the relative importance of multiple decision-makers and group consensus of fuzzy opinions, a hybrid averaging approach combining weighted averages and ordered weighted averages was employed to construct the collective decision matrix. An integrated programming model was then established based on the concept of signed distance-based closeness coefficients to determine the importance weights of criteria and the priority ranking of alternatives. Subsequently, an interactive procedure was proposed to modify the model according to the decision-makers' feedback on the degree of satisfaction toward undesirable solution results for the sake of gradually improving the integrated model. The feasibility and applicability of the proposed methods are illustrated with a medical decision-making problem of patient-centered medicine concerning basilar artery occlusion. A comparative analysis with other approaches was performed to validate the effectiveness of the proposed methodology.","Title: A multiobjective optimization solver using rank-niche evolution strategy\nYear: 2006\n\nAbstract: A rank-niche evolution strategy (RNES) algorithm has been developed in this paper to solve unconstrained multiobjective optimization problems. A required number of Pareto-optimal solutions can be generated by the algorithm in a single run. In addition to the operations of recombination, mutation and selection used in original evolution strategy (ES), an external elite set which contains a given number of non-dominated elites is updated and trimmed by a clustering technique to maintain a uniformly distributed Pareto front. The fitness function for each individual contains the information of rank and crowding status. The selection operation using this fitness function considers the superiority and distribution simultaneously. Eight test problems illustrated in other papers are used to test RNES. For some test problems the Pareto-optimal solutions obtained by RNES are better than those obtained by GA-based algorithms.","Title: A culture-sensitive agent in Kirman's ant model\nYear: 2011\n\nAbstract: The global financial crisis brought a serious collapse involving a \"systemic\" meltdown. Internet technology and globalization have increased the chances for interaction between countries and people. The global economy has become more complex than ever before. Mark Buchanan [12] indicated that agent-based computer models will prevent another financial crisis and has been particularly influential in contributing insights. There are two reasons why culture-sensitive agent on the financial market has become so important. Therefore, the aim of this article is to establish a culture-sensitive agent and forecast the process of change regarding herding behavior in the financial market. We based our study on the Kirman's Ant Model[4,5] and Hofstede's Natational Culture[11] to establish our culture-sensitive agent based model. Kirman's Ant Model is quite famous and describes financial market herding behavior from the expectations of the future of financial investors. Hofstede's cultural consequence used the staff of IBM in 72 different countries to understand the cultural difference. As a result, this paper focuses on one of the five dimensions of culture from Hofstede: individualism versus collectivism and creates a culture-sensitive agent and predicts the process of change regarding herding behavior in the financial market. To conclude, this study will be of importance in explaining the herding behavior with cultural factors, as well as in providing researchers with a clearer understanding of how herding beliefs of people about different cultures relate to their finance market strategies.","Title: Multidisciplinary design optimization of mechanisms\nYear: 2005\n\nAbstract: The optimization of mechanisms is usually done in the field of mechanism itself. As a result the structural safety of the mechanism is neglected. To ensure the safety and improve the dynamic characteristics of mechanisms, a multidisciplinary design optimization procedure is proposed in this paper to synthesize optimum mechanisms. Two disciplines are involved in the multidisciplinary design optimization. They are the mechanism and the structure. The multi-level decomposition approach is chosen to generate optimum mechanisms. The optimized mechanisms not only satisfy mechanism and structural constraints but also have optimum objective function values in the two disciplines. In order to solve general mechanism design problems two widely used commercial softwares MSC/NASTRAN and MSC/ADAMS are integrated in the procedure to do the structural and the mechanism analysis, respectively. When the structural optimization is performed, a compromised approach is introduced to treat multiple configurations of mechanisms during operation. Two mechanism design problems are given to test the proposed method.","Title: Interval-valued fuzzy TOPSIS method with leniency reduction and an experimental analysis\nYear: 2011\n\nAbstract: The purpose of this paper is to present a new TOPSIS (the technique for order preference by similarity to ideal solution) for estimating the importance of criteria and reducing the leniency bias in multiple-criteria decision analysis based on interval-valued fuzzy sets. Several types of net predispositions are defined to represent an aggregated effect of interval-valued evaluations. The relative closeness of each alternative to the ideal solution is then determined by net predispositions. Because positive or negative leniency may exist when most criteria are assigned unduly high or low ratings, respectively, some deviation variables are introduced to mitigate the effects of overestimated and underestimated ratings on criterion importance. Considering the two objectives of maximal closeness coefficient and minimal deviation values, an integrated programming model is proposed to compute optimal weights for the criteria and corresponding closeness coefficients for alternative rankings. A flexible algorithm using interval-valued fuzzy TOPSIS methods is established by considering both objective and subjective information to compute optimal multiple-criteria decisions. The feasibility and effectiveness of the proposed methods are illustrated by a numerical example. Finally, an experimental analysis of interval-valued fuzzy rankings given different conditions for the criterion weights is conducted with discussions on average Spearman correlation coefficients and contradiction rates.","Title: Interval-valued fuzzy permutation method and experimental analysis on cardinal and ordinal evaluations\nYear: 2009\n\nAbstract: This paper presents interval-valued fuzzy permutation (IVFP) methods for solving multiattribute decision making problems based on interval-valued fuzzy sets. First, we evaluate alternatives according to the achievement levels of attributes, which admits cardinal or ordinal representation. The relative importance of each attribute can also be measured by interval or scalar data. Next, we identify the concordance, midrange concordance, weak concordance, discordance, midrange discordance and weak discordance sets for each ordering. The proposed method consists of testing each possible ranking of the alternatives against all others. The evaluation value of each permutation can be computed either by cardinal weights or by solving programming problems. Then, we choose the permutation with the maximum evaluation value, and the optimal ranking order of alternatives can be obtained. An experimental analysis of IVFP rankings given cardinal and ordinal evaluations is conducted with discussions on consistency rates, contradiction rates, inversion rates, and average Spearman correlation coefficients.","Title: Optimistic and pessimistic decision making with dissonance reduction using interval-valued fuzzy sets\nYear: 2011\n\nAbstract: Interval-valued fuzzy sets have been developed and applied to multiple criteria analysis. However, the influence of optimism and pessimism on subjective judgments and the cognitive dissonance that accompanies the decision making process have not been studied thoroughly. This paper presents a new method to reduce cognitive dissonance and to relate optimism and pessimism in multiple criteria decision analysis in an interval-valued fuzzy decision environment. We utilized optimistic and pessimistic point operators to measure the effects of optimism and pessimism, respectively, and further determined a suitability function through weighted score functions. Considering the two objectives of maximal suitability and dissonance reduction, several optimization models were constructed to obtain the optimal weights for the criteria and to determine the corresponding degree of suitability for alternative rankings. Finally, an empirical study was conducted to validate the feasibility and applicability of the current method. We anticipate that the proposed method can provide insight on the influences of optimism, pessimism, and cognitive dissonance in decision analysis studies.","Title: The ELECTRE multicriteria analysis approach based on Atanassov's intuitionistic fuzzy sets\nYear: 2011\n\nAbstract: In recent decades, intuitionistic fuzzy sets have been applied to many different fields; however, few current studies have used the ELECTRE method to solve multi-criteria decision-making problems with intuitionistic fuzzy information. The purpose of this paper is to develop a new method, the intuitionistic fuzzy ELECTRE method, for solving multi-criteria decision-making problems. Atanassov's intuitionistic fuzzy set (A-IFS) characteristics are simultaneously concerned with the degree of membership, degree of non-membership, and intuitionistic index, and people can use A-IFS to describe uncertain situations in decision-making problems. We use the proposed method to rank all alternatives and determine the best alternative. The proposed method can also use imperfect or insufficient knowledge of data to deal with decision-making problems. Finally, two practical examples are given that illustrate the procedure of the proposed method.","Title: Bivariate models of optimism and pessimism in multi-criteria decision-making based on intuitionistic fuzzy sets\nYear: 2011\n\nAbstract: This paper presents a useful method of relating optimism and pessimism to multiple criteria decision analysis in the context of intuitionistic fuzzy sets based on the unipolar bivariate model. We use eight point operators to estimate the adaptational outcome expectations of optimism and/or pessimism and then determine the net predisposition, the aggregated effect of positive and negative evaluations. A series of new net predispositions for bivariate evaluations are proposed for neutrality; for complete, moderate, and rational optimism; for complete, moderate, and rational pessimism; and for complete and moderate optimism-pessimism. The suitability function, which measures the overall evaluation of each alternative, is then presented. Because positive or negative leniency may exist, such that most of the criteria may be assigned unduly high or low ratings, respectively, we introduce deviation variables to mitigate the effects of such ratings on the apparent importance of various criteria. Based on the two objectives of maximal weighted suitability and minimal deviation values, an integrated programming model is used to compute the optimal weights for the criteria and the corresponding degrees of suitability of the alternative rankings. We establish flexible algorithms that incorporate both objective and subjective information to compute the optimal optimistic and pessimistic decisions. The proposed methods are illustrated and discussed using a numerical example, a multi-criteria supplier selection problem. Finally, an empirical study of job choices is conducted to establish the feasibility and applicability of the current method.","Title: Synchronous Pipeline Circuit Design for an Adaptive Neuro-fuzzy Network\nYear: 2007\n\nAbstract: This paper presents an efficient synchronous pipeline hardware implementation procedure for a neuro-fuzzy (NF) circuit. We decompose the NF circuit into a feedforward circuit and a backpropagation circuit. The concept of pre-calculation to share computation results between the feedforward circuit and backpropagation circuit is introduced to achieve a high throughput rate and low resource usage. A novel pipeline architecture has been adopted to fulfill the concept of pre-calculation. With the unique pipeline architecture, we have successfully enhanced the throughput rate and resource sharing between modules. Particularly, the multiplier usage has been reduced from 7 to 3 and the divider usage from 3 to 1. Finally, we have implemented the NF circuit on FGPA. Our experimental results show a superior performance than that of an asynchronous pipeline design approach and the NF system implemented on MATLAB®.","Title: Determining objective weights with intuitionistic fuzzy entropy measures: A comparative analysis\nYear: 2010\n\nAbstract: In this paper, we propose a new objective weighting method that employs intuitionistic fuzzy (IF) entropy measures to solve multiple-attribute decision-making problems in the context of intuitionistic fuzzy sets. Instead of traditional fuzzy entropy, which uses the probabilistic discrimination of attributes to obtain attribute weights, we utilize the IF entropy to assess objective weights based on the credibility of the input data. We examine various measures for IF entropy with respect to hesitation degree, probability, non-probability, and geometry to calculate the attribute weights. A comparative analysis of different measures to generate attribute rankings is illustrated with both computational experiments as well as analyses of Pearson correlations, Spearman rank correlations, contradiction rates, inversion rates, and consistency rates. The experimental results indicate that ranking the outcomes of attributes not only depends on the type of IF entropy measures but is also affected by the number of attributes and the number of alternatives.","Title: On the improvements of the particle swarm optimization algorithm\nYear: 2010\n\nAbstract: Since a particle swarm optimization (PSO) algorithm uses a coordinated search to find the optimum solution, it has a better chance of finding the global solution. Despite this advantage, it is also observed that some parameters used in PSO may affect the solution significantly. Following this observation, this research tries to tune some of the parameters and to add mechanisms to the PSO algorithm in order to improve its robustness in finding the global solution. The main approaches include using uniform design to ensure uniform distribution of the initial particles in the design space, adding a mutation operation to increase the diversity of particles, decreasing the maximum velocity limitation and the velocity inertia automatically to balance the local and the global search efforts, reducing velocity when constraints are violated, and using Gaussian distribution based local searches to escape local minima. Besides these efforts, an algorithm is also developed to find multiple solutions in a single run. The results show that the overall effect of these approaches can yield better results for most test problems.","Title: A comparative study of optimistic and pessimistic multicriteria decision analysis based on atanassov fuzzy sets\nYear: 2012\n\nAbstract: Following the unipolar bivariate model, this paper presents a useful method of relating optimism and pessimism to multiple criteria decision analysis within the context of Atanassov fuzzy sets. We utilize four point operators to readily estimate the adaptational outcome expectancies generated by optimistic and pessimistic attitudes. We then determine the net predisposition to represent the aggregated effects of positive and negative evaluations. On the basis of the net predispositions, we calculate the suitability function to evaluate each alternative. Accordingly, we propose the optimization model with suitability functions to deal with ill-known grades of importance. We develop algorithms to aid multicriteria decision-making processes under neutral, completely optimistic, moderately optimistic, completely pessimistic, and moderately pessimistic conditions. Furthermore, we present experimental analysis on randomly generated decision problems to investigate the ranking system's consistency and the ranking irregularity issues found in the proposed methods. We establish several valuable test criteria to discuss some types of ranking consistency and irregularities (e.g., ranking consistency, ranking correlations, ranking contradictions, and ranking inversions) that occur when the different methods are used. We designed three computational experiments to compare the ranking orders yielded by different methods and examined several comparison indices, including the consistency rate of the total orders, the contradiction rate of the best choice, and the inversion rate between the better choices and the worse ones. Finally, we provided a second-order regression model to highlight the effects of variant parameter settings on the average Spearman rank correlation coefficients.","Title: A causal model of consumer involvement: A new approach with intuitionistic fuzzy automata\nYear: 2008\n\nAbstract: Involvement has become an important variable in consumer behavior and marketing research for a long time. The purpose of this study is to develop a model that can explain the complicated relationship between antecedents, consequences, and different types of involvement at the same time to understand the internal state of consumers better. Specifically, this model has to be capable of handling the vague interactions between different types of involvement. We develop this integrated model by the using of the intuitionistic fuzzy automata. We choose the cell phone as the stimulus product in this research. There are 19 antecedents, 5 types of involvement, and 10 consequences in the integrated involvement model. According to the results, we find that the intuitionistic fuzzy automata can deal with the interactions between different types of involvement. Second, the intuitionistic fuzzy automata can observe the relations between antecedents, consequences, and 5 types of involvement at the same time. Besides, we can observe the internal state inside consumers through the intuitionistic fuzzy automata, especially the degree of hesitation when they make decisions.","Title: Signed Distanced-Based Topsis Method For Multiple Criteria Decision Analysis Based On Generalized Interval-Valued Fuzzy Numbers\nYear: 2011\n\nAbstract: The theory of interval valued fuzzy sets is very valuable for modeling impressions of decision makers. In addition, it gives ability to quantify the ambiguous nature of subjective judgments in an easy way. In this paper, by extending the technique for order preference by similarity to ideal solution (TOPSIS), it is proposed a useful method based on generalized interval valued trapezoidal fuzzy numbers (GITrFNs) for solving multiple criteria decision analysis (MCDA) problems. In view of complexity in handling sophisticated data of GITrFNs, this paper employs the concept of signed distances to establish a simple and effective MCDA method based on the main structure of TOPSIS. An algorithm based on TOPSIS method is established to determine the priority order of given alternatives by using properties of signed distances. Finally, the feasibility of the proposed method is illustrated by a practical example of supplier selection.","Title: Exploring the effects of intuitionistic fuzzy separation measures on TOPSIS rankings\nYear: 2009\n\nAbstract: The purpose of this study is to extend the TOPSIS method for solving multiple attribute decision analysis problems with intuitionistic fuzzy data. Iintuitionistic fuzzy sets are capable of coping with imprecise information due to the fact that exact data may be difficult to be precisely determined since human judgments are often vague under many conditions. In this paper, a proposed intuitionistic fuzzy version of the TOPSIS method is presented and further deals with a comparative analysis of separation measures. For the sake of the comparison of intuitionistic fuzzy TOPSIS rankings yielded by different separation measures, a simulation experiment of different sizes was generated and examined. The consistency rate, the contradiction rate of the best alternative, and average Spearman correlation coefficients are utilized to conduct a pairwise comparison for all separation measures. The results which are inclusive of one hundred combinations of ten different categories of number for each alternatives and attributes indicate that the preference orders are hardly identical using different separation measures in the intuitionistic fuzzy TOPSIS method. The experimental analysis showed that the different definitions of IFS separations indeed significantly affect the final results by means of the intuitionistic fuzzy TOPSIS method. The comparative results presented in our experimental analysis indicate differentiations in a number of important aspects with some comparative indices.","Title: The interval-valued fuzzy TOPSIS method and experimental analysis\nYear: 2008\n\nAbstract: The purpose of this paper is to extend the TOPSIS method based on interval-valued fuzzy sets in decision analysis. Hwang and Yoon developed the technique for order preference by similarity to ideal solution (TOPSIS) in 1981. TOPSIS has been widely used to rank the preference order of alternatives and determine the optimal choice. Considering the fact that it is difficult to precisely attach the numerical measures to the relative importance of the attributes and to the impacts of the alternatives on these attributes in some cases, therefore, the TOPSIS method has been extended for interval-valued fuzzy data in this paper. In addition, a comprehensive experimental analysis to observe the interval-valued fuzzy TOPSIS results yielded by different distance measures is presented. A comparative analysis of interval-valued fuzzy TOPSIS rankings from each distance measure is illustrated with discussions on consistency rates, contradiction rates, and average Spearman correlation coefficients. Finally, a second-order regression model is provided to highlight the effects of the number of alternatives, the number of attributes, and distance measures on average Spearmen correlation coefficients.","Title: An outcome-oriented approach to multicriteria decision analysis with intuitionistic fuzzy optimistic/pessimistic operators\nYear: 2010\n\nAbstract: The concept of Atanassov's intuitionistic fuzzy sets is a generalization of ordinary fuzzy sets. Under the intuitionistic fuzzy decision environment, this paper develops optimistic and pessimistic point operators to draw the influences of optimism and pessimism on multiple criteria decision analysis. We propose the optimistic and pessimistic score functions based on point operators to measure evaluations of the alternative with respect to each criterion. The suitability function can then be established to determine the degree to which the alternative satisfies the decision maker's requirement. An optimization model with suitability functions is propounded on account of ill-known membership grades to generate optimal weights for criteria. Finally, an outcome-oriented approach is adopted to validate the proposed methods. Several multiattribute evaluation cases in consumer decision making reality are addressed to examine the feasibility and applicability of the current method. From the empirical results of effectiveness examination, the proposed intuitionistic fuzzy multicriteria decision making method with optimistic/pessimistic score functions performed significantly better than the method without consideration of optimism/pessimism. Therefore, we can conclude that the proposed methods by using intuitionistic fuzzy optimistic/pessimistic point operators have desirable potentials in practice.","Title: Nonlinear Assignment-Based Methods For Interval-Valued Intuitionistic Fuzzy Multi-Criteria Decision Analysis With Incomplete Preference Information\nYear: 2012\n\nAbstract: In the context of interval-valued intuitionistic fuzzy sets, this paper develops nonlinear assignment-based methods to manage imprecise and uncertain subjective ratings under incomplete preference structures and thereby determines the optimal ranking order of the alternatives for multiple criteria decision analysis. By comparing each interval-valued intuitionistic fuzzy number's score function, accuracy function, membership uncertainty index, and hesitation uncertainty index, a ranking procedure is employed to identify criterion-wise preference of alternatives. Based on the criterion-wise rankings and a set of known but incomplete information about criterion weights, a nonlinear assignment model is constructed to estimate criterion weights and to order the priority of various alternatives. Considering multiple criteria evaluation problems with preference conflict about criterion importance, an integrated nonlinear programming model is further established with regard to incomplete and inconsistent weight information. These proposed nonlinear assignment-based methods can obtain an aggregate ranking that effectively combines the relative performance of each alternative in each criterion. In addition, this overall ranking most closely agrees with the criterion-wise rankings. Finally, the feasibility of the proposed method is illustrated by a practical example of selecting a suitable bridge construction method.","Title: Multi-Criteria Decision-Making Methods With Optimism And Pessimism Based On Atanassov'S Intuitionistic Fuzzy Sets\nYear: 2012\n\nAbstract: The theory of Atanassov's intuitionistic fuzzy sets (A-IFSs) developed over the last several decades has found useful application in fields requiring multiple-criteria decision analysis. Since the membership-nonmembership pair in A-IFSs belongs to the bivariate unipolarity type, this article describes an approach that relates optimism and pessimism to multi-criteria decision analysis in an intuitionistic fuzzy-decision environment. First, several optimistic and pessimistic point operators were defined to alter the estimation of decision outcomes. Next, based on the core of the estimations, optimistic and pessimistic score functions were developed to evaluate each alternative with respect to each criterion. The suitability function was then established to determine the degree to which each an alternative satisfies the decision maker's requirement. Because the information on multiple criteria corresponding to decision importance is often incomplete, this study included suitability functions in the optimisation models to account for poorly known membership grades. Using a linear equal-weighted summation method, these models were transformed into a single objective optimisation model to generate the optimal weights for criteria. The feasibility and effectiveness of the proposed methods were illustrated through a practical example. Finally, computational experiments with enormous amounts of simulation data were designed to conduct a comparative analysis on the ranking orders yielded by different optimistic/pessimistic point operators.","Title: A comparative analysis of score functions for multiple criteria decision making in intuitionistic fuzzy settings\nYear: 2011\n\nAbstract: The purpose of this paper was to conduct a comparative study of score functions in multiple criteria decision analysis based on intuitionistic fuzzy sets. The concept of score functions has been conceptualized and widely applied to multi-criteria decision-making problems. There are several types of score functions that can identify the mixed results of positive and negative parts in a bi-dimensional framework of intuitionistic fuzzy sets. Considering various perspectives on score functions, the present study adopts an order of preference based on similarity to the ideal solution as the main structure to estimate the importance of different criteria and compute optimal multi-criteria decisions in intuitionistic fuzzy evaluation settings. An experimental analysis is conducted to examine the relationship between the results yielded by different score functions, considering the average Spearman correlation coefficients and contradiction rates. Furthermore, additional discussions clarify the relative differences in the ranking orders obtained from different combinations of numbers of alternatives and criteria as well as different importance conditions.","Title: Efficient Built-In Self-Test Schemes for Video Coding Cores: a Case Study on DCT/IDCT Circuits\nYear: 2006\n\nAbstract: In general, the software-testing time may be measured by two kinds of time scales: calendar time and test-execution time. In this paper, we develop two-dimensional software reliability models with two-time measures and incorporate both of them to assess ...","Title: Identification of general fuzzy measures by genetic algorithmsbased on partial information\nYear: 2000\n\nAbstract: This study develops an identification procedure for general fuzzy measures using genetic algorithms. In view of the difficulty in data collection in practice, the amount of input data is simplified through a sampling procedure concerning attribute subsets, and the corresponding detail design is adapted to the partial information acquired by the procedure. A specially designed genetic algorithm is proposed for better identification, including the development of the initialization procedure, fitness function, and three genetic operations. To show the applicability of the proposed method, this study simulates a set of experimental data that are representative of several typical classes. The experimental analysis indicates that using genetic algorithms to determine general fuzzy measures can obtain satisfactory results under the framework of partial information","Title: Hardware Design of an Adaptive Neuro-fuzzy Network with On-Chip Learning Capability\nYear: 2007\n\nAbstract: This paper aims for the development of the digital circuit of an adaptive neuro-fuzzy network with on-chip learning capability. The on-chip learning capability was realized by a backpropagation learning circuit for optimizing the network parameters. To maximize the throughput of the circuit and minimize its required resources, we proposed to reuse the computational results in both feedforward and backpropagation circuits. This leads to a simpler data flow and the reduction of resource consumption. To verify the effectiveness of the circuit, we implemented the circuit in an FPGA development board and compared the performance with the neuro-fuzzy system written in a MATLAB®code. The experimental results show that the throughput of our neuro-fuzzy circuit significantly outperforms the NF network written in a MATLAB®code with a satisfactory learning performance.","Title: Interval-valued fuzzy multiple criteria decision-making methods based on dual optimistic/pessimistic estimations in averaging operations.\nYear: 2014\n\nAbstract: •Decision-making methods with optimistic/pessimistic averaging operations.•Multiple criteria decision analysis within an interval-valued fuzzy environment.•Algorithms involve changes in overall judgments and in the separate evaluations.•The feasibility is illustrated by an application to medical decision making.•A comparative study is implemented with computational experimental analysis.","Title: The inclusion-based TOPSIS method with interval-valued intuitionistic fuzzy sets for multiple criteria group decision making.\nYear: 2015\n\nAbstract: •An extended TOPSIS method is proposed using an inclusion comparison approach.•Multi-criteria group decision-making problems are addressed in the IVIFS context.•A modified IIHA operation is presented using inclusion-based IIOWA operations.•Inclusion-based closeness coefficients are provided in the new TOPSIS model.•The feasibility is illustrated by a medical group decision-making problem.","Title: Multiple criteria decision analysis using a likelihood-based outranking method based on interval-valued intuitionistic fuzzy sets\nYear: 2014\n\nAbstract: The purpose of this paper is to develop a likelihood-based outranking method for handling multiple criteria decision analysis problems based on interval-valued intuitionistic fuzzy sets. Using the concept of likelihood of the interval-valued intuitionistic fuzzy preference relations, this paper determines certain generalized criteria and the corresponding likelihood-based preference functions. Based on some useful concepts of comprehensive preference indices, concordance indices, counter-likelihood-based preference functions, and discordance indices, this paper conducts a concordance-discordance analysis with concordance and discordance outranking relationships to determine a global Boolean matrix and acquire partial ranking orders of the alternatives. Alternatively, this paper determines complete ranking orders of the alternatives using the concepts of net concordance indices, net discordance indices, and mean outranking values. The feasibility and applicability of the proposed method are illustrated with a practical multiple criteria decision-making application concerning the selection of a suitable bridge construction method. Finally, comparative discussions with different decision-making methods are conducted to verify the effectiveness and advantages of the proposed method in aiding decision making.","Title: A prioritized aggregation operator-based approach to multiple criteria decision making using interval-valued intuitionistic fuzzy sets: A comparative perspective\nYear: 2014\n\nAbstract: Within the environment of interval-valued intuitionistic fuzzy (IVIF) sets, this paper develops a prioritized aggregation operator-based approach to handling multiple criteria decision analysis (MCDA) problems in which there exists a prioritization relationship over evaluative criteria. The prioritization between the criteria is modeled by assessing the IVIF weights that are associated with criteria dependence on the satisfaction of the higher priority criteria. Then, a new IVIF prioritized aggregation operator is presented to aggregate the IVIF ratings of the alternatives with respect to the prioritized criteria. Based on synthetic evaluations given by the IVIF prioritized aggregation operator, the ranking order of the alternatives can be determined according to the overall evaluation values. The feasibility and applicability of the proposed method are illustrated by the practical problem of watershed site selection. Moreover, a comparative analysis with other relevant methods is conducted to validate the effectiveness. The comparative results show that our proposed prioritized aggregation operator-based method is appropriate and effective for managing MCDA problems when there are uncertainties expressed by the IVIF sets.","Title: An interval type-2 fuzzy LINMAP method with approximate ideal solutions for multiple criteria decision analysis.\nYear: 2015\n\nAbstract: The purpose of this paper is to develop a linear programming technique for multidimensional analysis of preference (LINMAP) to address multiple criteria decision analysis problems within the interval type-2 fuzzy environment based on interval type-2 trapezoidal fuzzy numbers. Considering the issue of anchor dependency, we use multiple anchor points in the decision-making process and employ approximate positive-ideal and negative-ideal solutions as the points of reference. Selected useful properties of the approximate ideal solutions are also investigated. In contrast to the classical LINMAP methods, this paper directly generates approximate ideal solutions from the characteristics of all alternatives. Next, this work presents the concept of closeness-based indices using Minkowski distances with approximate ideal solutions to develop a new approach for determining measurements of consistency and inconsistency. Under incomplete preference information on paired comparisons of the alternatives, this paper provides a novel method that uses the concept of comprehensive closeness-based indices to measure the poorness of fit and the goodness of fit. By applying the consistency indices and inconsistency indices, this work formulates an optimization problem that can be solved for the optimal weights of the criteria and thus acquires the best compromise alternative. Additionally, this paper explores the problem of supplier selection and conducts a comparative discussion to validate the effectiveness and applicability of the proposed interval type-2 fuzzy LINMAP method with approximate ideal solutions. Furthermore, the proposed method is applied to address a marketplace decision difficulty (MPDD)-prone decision-making problem to provide additional contributions for practical implications.","Title: Likelihoods of interval type-2 trapezoidal fuzzy preference relations and their application to multiple criteria decision analysis.\nYear: 2015\n\nAbstract: Interval type-2 fuzzy sets are useful and valuable for depicting uncertainty and managing imprecision in decision information. In particular, interval type-2 trapezoidal fuzzy numbers, as a special case of interval type-2 fuzzy sets, can efficiently express qualitative evaluations or assessments. In this work, the concept of the likelihoods of interval type-2 trapezoidal fuzzy preference relations based on lower and upper likelihoods is investigated, and the relevant properties are discussed. This paper focuses on the use of likelihoods in addressing multiple criteria decision analysis problems in which the evaluative ratings of the alternatives and the importance weights of the criteria are expressed as interval type-2 trapezoidal fuzzy numbers. A new likelihood-based decision-making method is developed using the useful concepts of likelihood-based performance indices, likelihood-based comprehensive evaluation values, and signed distance-based evaluation values. A simplified version of the proposed method is also provided to adapt the decision-making context in which the importance weights of the criteria take the form of ordinary numbers. The practical effectiveness of the proposed method is validated with four applications, and several comparative analyses are conducted to verify the advantages of the proposed method over other multiple criteria decision-making methods.","Title: The Inclusion-Based Linmap Method For Multiple Criteria Decision Analysis Within An Interval-Valued Atanassov'S Intuitionistic Fuzzy Environment\nYear: 2014\n\nAbstract: The purpose of this paper is to develop an inclusion-based LINMAP (i.e., Linear Programming Technique for Multidimensional Analysis of Preference) method for multiple criteria decision analysis that is based on interval-valued Atanassov's intuitionistic fuzzy sets. Using the inclusion comparison possibility in the interval-valued Atanassov's intuitionistic fuzzy context, an inclusion-based index of interval-valued Atanassov's intuitionistic fuzzy numbers is proposed that considers positive and negative ideals. An inclusion-based consistency index and an inclusion-based inconsistency index to measure the concordance and discordance, respectively, between paired comparison judgments are suggested. An inclusion-based LINMAP model is constructed using a linear programming technique to determine the optimal criterion weights and obtain the corresponding comprehensive inclusion-based index for each alternative. Then, the priority order of the alternatives can be acquired according to the comprehensive inclusion-based indices. The feasibility of the proposed method is illustrated using a practical problem that relates to the selection of bridge construction methods. A comparative analysis of other relevant decision-making methods is conducted to validate the effectiveness of the developed methodology.","Title: An Interactive Signed Distance Approach For Multiple Criteria Group Decision-Making Based On Simple Additive Weighting Method With Incomplete Preference Information Defined By Interval Type-2 Fuzzy Sets\nYear: 2014\n\nAbstract: Interval type-2 fuzzy sets (T2FSs) with interval membership grades are suitable for dealing with imprecision or uncertainties in many real-world problems. In the Interval type-2 fuzzy context, the aim of this paper is to develop an interactive signed distance-based simple additive weighting (SAW) method for solving multiple criteria group decision-making problems with linguistic ratings and incomplete preference information. This paper first formulates a group decision-making problem with uncertain linguistic variables and their transformation to interval type-2 trapezoidal fuzzy numbers. Concerning the relative importance of multiple decision-makers and group consensus of fuzzy opinions, a procedure using hybrid averages is then employed to construct a collective decision matrix. By an appropriate extension of the classical SAW approach, this paper utilizes the concept of signed distances and establishes an integrated programming model to manage multi-criteria group decisions under the incomplete and inconsistent preference structure. Further, an interactive procedure is established for group decision making. Finally, the feasibility and effectiveness of the proposed methods are illustrated by a collaborative decision-making problem of patient-centered care (PCC).","Title: An interval type-2 fuzzy PROMETHEE method using a likelihood-based outranking comparison approach.\nYear: 2015\n\nAbstract: We develop an interval type-2 fuzzy PROMETHEE method to address MCDA problems.The proposed method uses the approach of likelihood-based outranking comparisons.We present novel likelihood-based preference functions based on outranking indices.We develop two algorithmic procedures to acquire partial and complete rankings.Comparative analysis validates the effectiveness of the proposed method. Based on the preference ranking organization method for enrichment evaluations (PROMETHEE), the purpose of this paper is to develop a new multiple criteria decision-making method that uses the approach of likelihood-based outranking comparisons within the environment of interval type-2 fuzzy sets. Uncertain and imprecise assessment of information often occurs in multiple criteria decision analysis (MCDA). The theory of interval type-2 fuzzy sets is useful and convenient for modeling impressions and quantifying the ambiguous nature of subjective judgments. Using the approach of likelihood-based outranking comparisons, this paper presents an interval type-2 fuzzy PROMETHEE method designed to address MCDA problems based on interval type-2 trapezoidal fuzzy (IT2TrF) numbers. This paper introduces the concepts of lower and upper likelihoods for acquiring the likelihood of an IT2TrF binary relationship and defines a likelihood-based outranking index to develop certain likelihood-based preference functions that correspond to several generalized criteria. The concept of comprehensive preference measures is proposed to determine IT2TrF exiting, entering, and net flows in the valued outranking relationships. In addition, this work establishes the concepts of a comprehensive outranking index, a comprehensive outranked index, and a comprehensive dominance index to induce partial and total preorders for the purpose of acquiring partial ranking and complete ranking, respectively, of the alternative actions. The feasibility and applicability of the proposed method are illustrated with two practical applications to the problem of landfill site selection and a car evaluation problem. Finally, a comparison with other relevant methods is conducted to validate the effectiveness of the proposed method.","Title: A simulated annealing-based permutation method and experimental analysis for multiple criteria decision analysis with interval type-2 fuzzy sets\nYear: 2015\n\nAbstract: Development of a simulated annealing-based permutation decision-making method.Multiple criteria decision analysis within the interval type-2 fuzzy environment.Algorithm for a polynomial time solution in the total completion time problem.Computational experiments are designed to test the solution approach.Comparative analysis verifies the effectiveness and efficiency of the method. The aim of this paper is to develop a simulated annealing-based permutation method for multiple criteria decision analysis within the environment of interval type-2 fuzzy sets. The outranking methodology constitutes one of the most fruitful approaches in multiple criteria decision making and has been applied in numerous real-world problems. The permutation method is a classical outranking model, which generalizes Jacquet-Lagreze's permutation method and is based on a pairwise criterion comparison of the alternatives. Because modeling of the uncertainty in the decision-making process becomes increasingly important, an extension to the interval type-2 fuzzy environment is a useful generalization of the permutation method and is appropriate for handling uncertain and imprecise information in practical decision-making situations. This paper produces a signed-distance-based comparison among the comprehensive rankings of alternatives for concordance and discordance analyses. An integrated nonlinear programming model is constructed for estimation of the criterion weights and the optimal ranking order of the alternatives under incomplete preference information. To enhance the implementation efficiency, a simulated annealing-based permutation method and its meta-heuristic algorithm are developed to produce a polynomial time solution for the total completion time problem. Furthermore, computational experiments with notably large amounts of simulation data are conducted to test the solution approach and validate the correctness of the approximate solution compared with the optimal all-permutation-based result.","Title: An interval-valued intuitionistic fuzzy permutation method with likelihood-based preference functions and its application to multiple criteria decision analysis.\nYear: 2016\n\nAbstract: Developing an IVIF permutation method with likelihood-based preference functions.Using a hybrid approach of the likelihood calculation method and PROMETHEE.Establishing new measurements of concordance/discordance to evaluate permutations.Presenting two algorithms for IVIF importance weights and incomplete information.Verifying the effectiveness using comparative analyses and applications to NPD. This paper presents an interval-valued intuitionistic fuzzy permutation method with likelihood-based preference functions for managing multiple criteria decision analysis based on interval-valued intuitionistic fuzzy sets. First, certain likelihood-based preference functions are proposed using the likelihoods of interval-valued intuitionistic fuzzy preference relationships. Next, selected practical indices of concordance/discordance are established to evaluate all possible permutations of the alternatives. The optimal priority order of the alternatives is determined by comparing all comprehensive concordance/discordance values based on score functions. Furthermore, this paper considers various preference types and develops another interval-valued intuitionistic fuzzy permutation method using programming models to address multiple criteria decision-making problems with incomplete preference information. The feasibility and applicability of the proposed methods are illustrated in the problem of selecting a suitable bridge construction method. Moreover, certain comparative analyses are conducted to verify the advantages of the proposed methods compared with those of other decision-making methods. Finally, the practical effectiveness of the proposed methods is validated with a risk assessment problem in new product development.","Title: An interval type-2 fuzzy technique for order preference by similarity to ideal solutions using a likelihood-based comparison approach for multiple criteria decision analysis\nYear: 2015\n\nAbstract: An interval type-2 fuzzy TOPSIS method is developed to address decision problems.Use of the likelihood-based comparison approach with the approximate ideals.Establishment of likelihood-based closeness coefficients using comparison indices.Multicriteria decision analysis based on interval type-2 trapezoidal fuzzy numbers.Comparative analysis shows the effectiveness and advantages of the proposed method. The technique for order preference by similarity to ideal solutions (TOPSIS) is a well-known compromising method for addressing decision-making problems. In general, incomplete preference information and vague subjective judgments are realistic in practice. Accordingly, the theory of interval type-2 fuzzy sets has received increasing attention in the decision-making field because of its great ability to handle imprecise and ambiguous information in a convenient manner. The purpose of this paper is to develop a novel interval type-2 fuzzy TOPSIS method for multiple criteria decision analysis that is based on interval type-2 trapezoidal fuzzy numbers. This paper introduces the concept of approximate positive-ideal and negative-ideal solutions and presents a simple way to approach the evaluative ratings of ideal solutions using interval type-2 trapezoidal fuzzy numbers. Based on the likelihoods of interval type-2 trapezoidal fuzzy binary relations, this paper proposes certain likelihood-based comparison indices to establish a likelihood-based closeness coefficient of each alternative relative to the approximate ideals. Applying a likelihood-based comparison approach with the approximate ideals, this paper develops the interval type-2 fuzzy TOPSIS procedure to determine the priority ranking orders of the alternatives under consideration of the multiple criteria evaluation/selection. Three practical applications involving landfill site selection, supplier selection, and car evaluation are examined to show the effectiveness and practicability of the proposed method. Furthermore, this paper makes a comparison of the solution results yielded by other interval type-2 fuzzy decision-making methods. The comparative analyses demonstrate that the proposed interval type-2 fuzzy TOPSIS method is easy to implement and produces effective and valid results for solving multiple criteria decision-making problems.","Title: A weight-assessing method with habitual domains\nYear: 1998\n\nAbstract: Weights of criteria are the important factors in decision-making. However, from a behavioral perspective, traditional weighting methods account for too few factors to deal with decisions properly. Based on the behavior mechanism and the theory of habitual domains, this study is undertaken to develop a new weight-assessing model that treats decision-making as a dynamically adjusting process proceeding from the ideal to actual states. The new model is built upon the dynamic analysis for the connectivities between criteria instead of the static analysis of traditional models. Finally, we have studied Taipei City motorcycle users, mode-choice behavior through questionnaire in order to show the applicability of the new model. From the empirical results, it is found that our weight-assessing method has significant application potential in practice.","Title: A likelihood-based QUALIFLEX method with interval type-2 fuzzy sets for multiple criteria decision analysis.\nYear: 2015\n\nAbstract: The aim of this paper is to develop a likelihood-based QUALIFLEX method with interval type-2 fuzzy sets for multiple criteria decision analysis. This paper considers the context of interval type-2 trapezoidal fuzzy numbers and also introduces an extended concept of likelihoods of interval type-2 fuzzy preference relations. The concepts of lower and upper likelihoods are presented to serve as lower and upper bounds, respectively, of the possibility of preference relations and then to determine the likelihoods between interval type-2 trapezoidal fuzzy numbers. This paper uses a likelihood-based comparison between evaluative ratings to propose the concepts of concordance indices and comprehensive concordance indices for evaluating the permutations of alternatives corresponding to criterion-by-criterion consistency. By employing the signed distance-based method or the optimal membership-degree-determination method, the best permutation and its corresponding optimal ranking order of the alternatives can be acquired through a comparison of all of the comprehensive concordance indices. The applicability and effectiveness of the proposed method are illustrated with a medical decision-making problem that addresses the selection of treatment options. Additionally, a comparative analysis is performed with some relevant multiple criteria decision-making methods to demonstrate the advantages of the likelihood-based QUALIFLEX method. Finally, an analysis through computational experiments is conducted to examine the implementation efficiency of the proposed method.","Title: Likelihood-based assignment methods for multiple criteria decision analysis based on interval-valued intuitionistic fuzzy sets\nYear: 2015\n\nAbstract: The aim of this paper is to develop useful likelihood-based assignment methods for addressing multiple criteria decision-making problems within the environment of interval-valued intuitionistic fuzzy sets. Based on the likelihoods of interval-valued intuitionistic fuzzy preference relations, this paper determines the mean likelihoods of outranking relations and presents a mean likelihood determination method for generating a set of criterion-wise rankings of alternatives. By employing the concepts of rank frequency matrices and (ordinary) rank contribution matrices, this paper establishes a likelihood-based linear assignment model for multiple criteria decision analysis in the interval-valued intuitionistic fuzzy context. Additionally, this paper propounds two likelihood-based assignment models for handling incomplete and conflicting certain information of importance weights. These models can transform the criterion-wise ranks into the overall ranks for determining the optimal priority ranking of the alternatives. The feasibility and applicability of the proposed methods are illustrated with a practical problem of selecting a bridge construction method which involves various preference types. Finally, this paper conducts a comparative analysis with previous assignment-based methods in an interval-valued intuitionistic fuzzy setting to validate the effectiveness and advantages of the proposed methods.","Title: IVIF-PROMETHEE outranking methods for multiple criteria decision analysis based on interval-valued intuitionistic fuzzy sets\nYear: 2015\n\nAbstract: This paper attempts to develop an interval-valued intuitionistic fuzzy preference ranking organization method for enrichment evaluations (IVIF-PROMETHEE) and attempts to apply it to multiple criteria decision analysis. The theory of interval-valued intuitionistic fuzzy sets is useful for modeling impressions and quantifying the ambiguous nature of subjective judgments in a convenient manner. PROMETHEE is a well-known and widely used outranking method, but it has not been investigated in depth within the interval-valued intuitionistic fuzzy environment. Based on the concepts of inclusion comparison possibilities, this paper proposes inclusion-based generalized criteria to determine preference functions and global preference indices for acquiring leaving flows, entering flows, and net flows of alternative actions. Using the score functions and accuracy functions of the flows, this paper develops IVIF-PROMETHEE I and IVIF-PROMETHEE II, which are methods for the partial ranking and complete ranking, respectively, of alternatives. The feasibility and applicability of the proposed methods are illustrated through a problem on the selection of bridge construction methods. Finally, a comparative discussion of other decision-making methods is conducted to demonstrate the advantages of the proposed IVIF-PROMETHEE methods.","Title: Digital Artwork Creation Using Water And Sand On A Two-Dimensional Surface\nYear: 2015\n\nAbstract: In this paper, we developed a system which enables users to draw pictures of sand and water on a two-dimensional surface in a virtual environment. By mixing water and sand for artwork creation, we can create different styles of pictures. Our system has three major subsystems: water animation system, sand animation system and rendering system. In the simulation part, we compute the motion of sand and water based on the interactions for water-water, sand-water and sand-sand. In the rendering part, we attempt to balance the rendering quality and the computation cost. We render water using a ray tracing method and the sand is rendered with shadow. Thus, the rendered results of water and sand look like 2.5D. We tested our system and the experimental results showed that the users could create a variety of pictures of sand and water.","Title: A projection-based compromising method for multiple criteria decision analysis with interval-valued intuitionistic fuzzy information.\nYear: 2016\n\nAbstract: Display Omitted Multicriteria decision analysis based on interval-valued intuitionistic fuzzy sets.Development of a projection-based compromising method.Establishment of projection-based compromising indices and comprehensive indices.Determination of the ranking order and the approval status of alternatives.Comparative analyses with other relevant compromising decision-making methods. The purpose of this paper is to develop a projection-based compromising method for addressing multiple criteria decision-making problems based on interval-valued intuitionistic fuzzy sets. The concept of projections considers not only the distance but also the included angle between evaluative ratings of alternative actions with respect to a criterion. In the interval-valued intuitionistic fuzzy context, this paper determines the respective projections of the evaluative ratings of each alternative on the positive-ideal and negative-ideal solutions and explores several essential properties. Next, this paper introduces the concepts of projection-based compromising indices and comprehensive compromising indices and further investigates relevant theorems for supporting the usefulness of these indices. Additionally, this paper proposes the projection-based comparative index and the comprehensive comparative index to serve as benchmark values for the comparison purpose. The improvement percentage of the comprehensive compromising value is acquired to determine the priority order of the alternatives, including the complete ranking order and the approval status for each alternative. The feasibility and the applicability of the proposed method are validated with an application problem of watershed site selection. Finally, several comparative analyses are conducted to verify the effectiveness and advantages of the proposed method over other relevant compromising decision-making methods.","Title: A likelihood-based assignment method for multiple criteria decision analysis with interval type-2 fuzzy information.\nYear: 2017\n\nAbstract: The purpose of this paper was to develop a likelihood-based assignment method based on interval type-2 fuzzy sets and apply it to decision-making problems involving multiple criteria evaluation and the ranking/selection of alternatives. The linear assignment method is a well-known outranking method in the field of multiple criteria decision analysis. The theory of interval type-2 fuzzy sets is useful for addressing the uncertainty and imprecision associated with a subjective environment. In this paper, the key feature of the proposed method is the incorporation of the extended concept of likelihoods of fuzzy preference relations between interval type-2 trapezoidal fuzzy numbers into the main structure of the linear assignment methodology. An effective ranking procedure using the optimal membership degree determination method is proposed to determine criterion-wise preference rankings of the alternatives. The proposed method establishes the novel concepts of an (adjusted) rank frequency matrix and an (adjusted) rank contribution matrix to combine the relative performances of the alternatives in terms of each criterion. Based on a signed distance comparison approach, this paper constructs a likelihood-based assignment model to obtain an aggregate ranking of the alternatives that is in the closest agreement with the criterion-wise preferences of the alternatives. The feasibility and applicability of the proposed method are illustrated with two practical multiple criteria decision-making applications concerning the selection of landfill sites and the selection of treatment options. Finally, a comparative analysis with other relevant methods is conducted to validate the effectiveness and advantages of the current methods in decision aiding.","Title: An Interval-Valued Pythagorean Fuzzy Outranking Method With A Closeness-Based Assignment Model For Multiple Criteria Decision Making\nYear: 2018\n\nAbstract: The concept of interval-valued Pythagorean fuzzy (IVPF) sets is capable of handling imprecise and ambiguous information and managing complex uncertainty in real-world applications. This paper focuses on multiple criteria decision analysis involving IVPF information and proposes a new outranking decision-making method that uses a closeness-based assignment model. In contrast to the existing assignment-based methodology, the uniqueness of this paper is the consideration of uncertain information represented by IVPF values, the determination of criterion-wise precedence rankings based on a closeness-based approach, and the development of a new measure for scalar representation. First, to underlie anchored judgments in subjective decision-making processes, this paper presents a compromising concept of the closeness index with the positive-ideal and negative-ideal IVPF values to identify criterion-wise precedence ranks among alternatives. Next, this paper defines the concept of matrices of precedence frequency and contribution to provide a basis for the proposed assignment model. To overcome the difficulty of lacking nontrivial scalar representations, a useful measure is also developed to appropriately describe IVPF values. Based on a closeness-based assignment approach, a novel outranking decision-making method is proposed to transform the extended criterion-wise ranks into the ultimate priority orders of the alternatives. The proposed method is first implemented in a practical problem of selecting a bridge construction method to demonstrate its feasibility and applicability. Moreover, its practicality and effectiveness are verified through a comparative analysis with relevant assignment-based approaches. Further comparative analyses with newly developed IVPF decision-making methods are conducted for both a risk evaluation problem and an investment problem to examine the advantages of the proposed method and extend the current technique by considering distinct preference information for adapting to the particularities in practice. (C) 2017 Wiley Periodicals, Inc.","Title: Remoteness index-based Pythagorean fuzzy VIKOR methods with a generalized distance measure for multiple criteria decision analysis.\nYear: 2018\n\nAbstract: •Development of remoteness index-based Pythagorean fuzzy VIKOR methods.•Consideration of uncertain information represented by Pythagorean fuzzy sets.•Construction of generalized PF distances and displaced/fixed remoteness indices.•Creation of remoteness-based group utility indices and individual regret indices.•Ultimate ranking using displaced/fixed remoteness-based compromise indices.•Comparative analyses with newly developed methods via five practical applications.","Title: A novel risk evaluation method of technological innovation using an inferior ratio-based assignment model in the face of complex uncertainty.\nYear: 2018\n\nAbstract: •Development of a novel risk evaluation method of technological innovation.•Consideration of highly uncertain information represented by IVPF values.•Construction of IVPF IRs based on the IVPF distance measure.•Establishment of a useful IVPF IR-based assignment model.•Comparative analysis with other methods via a practical application.","Title: Interactive Sand Art Drawing Using Rgb-D Sensor\nYear: 2018\n\nAbstract: We present an interactive system using one RGB-D sensor, which allows a user to use bare hands to perform sand drawing. Our system supports the common sand drawing functions, such as sand erosion, sand spilling, and sand leaking. To use hands to manipulate the virtual sand, we design four key hand gestures. The idea is that the gesture of one hand controls the drawing actions. The motion and gesture of the other hand control the drawing positions. There are three major steps. First, our system adopts a vision-based bare-hand detection method which computes the hand position and recognizes the hand gestures. Second, the drawing positions and the drawing actions are sent to a sand drawing subsystem. Finally, the subsystem performs the sand drawing actions. Experimental results show that our system enables users to draw a rich variety of sand pictures.","Title: An outranking approach using a risk attitudinal assignment model involving Pythagorean fuzzy information and its application to financial decision making.\nYear: 2018\n\nAbstract: •Representation of highly uncertain information using PF/IVPF sets.•Development of risk attitude-based score functions for PF/IVPF values.•Construction of risk attitudinal assignment models in the PF/IVPF context.•Application to financial decision-making problems.•Comparative analysis with other outranking approaches.","Title: An effective correlation-based compromise approach for multiple criteria decision analysis with Pythagorean fuzzy information.\nYear: 2018\n\nAbstract: The aim of this paper is to develop a novel correlation-based compromise approach for addressing multiple criteria decision analysis problems under complex uncertainty based on Pythagorean fuzzy sets. Based on the concepts of information energy and correlations for Pythagorean fuzzy characteristics, this paper proposes useful correlation-based compromise indices and investigates their desirable properties. These compromise indices can be employed to underlie anchored judgments and to reflect a certain balance between the correlations with positive-ideal and negative-ideal points of reference. Moreover, they can fully take into consideration the amount of information associated with higher degrees of uncertainty and can effectively fuse subjective assessments conveyed by Pythagorean fuzziness. From two different perspectives of displaced and fixed ideals, this paper provides two algorithmic procedures of the proposed correlation-based compromise approach for conducting multiple criteria evaluation tasks within Pythagorean fuzzy environments. A real-world problem concerning a financing decision on working capital policies is investigated to show the feasibility and applicability of the developed techniques. The application results, along with a comparative analysis, demonstrate the practicality and effectiveness of the proposed methodology.","Title: A Novel PROMETHEE-Based Outranking Approach for Multiple Criteria Decision Analysis With Pythagorean Fuzzy Information.\nYear: 2018\n\nAbstract: This paper aims to present a novel outranking approach of the preference ranking organization method for enrichment evaluations (PROMETHEEs) based on Pythagorean fuzzy sets for multiple criteria decision analysis. The proposed method utilizes a novel Pythagorean fuzzy precedence index, which is based on the difference of scalar functions under anchored judgments with respect to the superiority/inferiority Pythagorean fuzzy numbers. To appropriately describe the hesitation between indifference and preference in the Pythagorean fuzzy context, this paper introduces useful precedence-based preference functions to establish the precedence relations based on pairwise comparisons. As a multiple criteria measure under Pythagorean fuzzy uncertainty, the concept of overall preference indices is identified at the aggregation stage to exploit certain rules for generating PROMETHEE flows. Next, this paper provides effective Pythagorean fuzzy PROMETHEE I and II ranking procedures to determine partial and complete preorders, respectively, among competing alternatives. The developed Pythagorean fuzzy PROMETHEE-based outranking approach is comparatively validated using a real-world application concerning the selection problem of bridge construction methods. The solution results along with a comparison analysis demonstrate that the proposed methodology outperforms the comparative approach in terms of reasonability and stability.","Title: Multiple Criteria Decision Analysis Using Correlation-Based Precedence Indices Within Pythagorean Fuzzy Uncertain Environments\nYear: 2018\n\nAbstract: The theory of Pythagorean fuzzy sets possesses significant advantages in handling vagueness and complex uncertainty. Additionally, Pythagorean fuzzy information is useful to simulate the ambiguous nature of subjective judgments and measure the fuzziness and imprecision more flexibly. The aim of this research is to develop an effective assignment-based method using a novel concept of correlation-based precedence indices for conducting multiple criteria decision analysis within the Pythagorean fuzzy uncertain environment. Based on the ideas of information energy and correlations, this paper defines a novel concept of correlation-based precedence indices in the Pythagorean fuzzy context and discusses their desirable properties. Next, this paper presents some useful concepts of discordance indicators, weighted discordance indicators, comprehensive discordance indicators, and comprehensive discordance indices to construct a novel assignment model for acquiring a comprehensive ranking of candidate alternatives. As an application of the proposed assignment-based method, a practical example concerning a financing decision of working capital policies is provided to demonstrate its practicality and effectiveness.","Title: A Mixed-Choice-Strategy-Based Consensus Ranking Method for Multiple Criteria Decision Analysis Involving Pythagorean Fuzzy Information.\nYear: 2018\n\nAbstract: The aim of this paper is to develop a novel consensus ranking method that uses a mixed choice strategy for multiple criteria decision analysis (MCDA) under complex uncertainty based on Pythagorean fuzzy (PF) sets. The majority of MCDA methods have focused almost exclusively on \"criterion-specific\" choice tasks that are the tasks in which all alternatives are decomposed into distinct components and evaluated on specific criteria. However, in certain MCDA problems in practical applications, category-based choices tend to be more holistic in nature, especially in affective-like aspects. Therefore, this paper incorporates a mixed choice strategy (i.e., a combination of a category-based strategy and a criterion-specific strategy) into the core structure of the developed MCDA method. Furthermore, this paper utilizes the theory of Pythagorean fuzziness to provide a powerful modeling tool for complex and varied decision-making environments. Employing the developed concepts of a PF precedence index based on PF information and a disagreement indicator based on distances between rankings, this paper proposes a novel consensus ranking method by means of a comprehensive disagreement-based assignment model for addressing a mixed-choice-strategybased MCDA problem in the PF context. As an application of the proposed methodology, a real-world case study of a luxury car selection problem is investigated. The application results, along with a comparative analysis, demonstrate the practicality and effectiveness of the developed approach, which is capable of handling hybrid category-based and criterion-specific choice tasks and managing complex uncertainty in practical situations.","Title: An Interval-Valued Pythagorean Fuzzy Compromise Approach with Correlation-Based Closeness Indices for Multiple-Criteria Decision Analysis of Bridge Construction Methods.\nYear: 2018\n\nAbstract: The purpose of this paper is to develop a novel compromise approach using correlation-based closeness indices for addressing multiple-criteria decision analysis (MCDA) problems of bridge construction methods under complex uncertainty based on interval-valued Pythagorean fuzzy (IVPF) sets. The assessment of bridge construction methods requires the consideration of multiple alternatives and conflicting tangible and intangible criteria in intricate and varied circumstances. The concept of IVPF sets is capable of handling imprecise and ambiguous information and managing complex uncertainty in real-world applications. Inspired by useful ideas concerning information energies, correlations, and correlation coefficients, this paper constructs new concepts of correlation-based closeness indices for IVPF characteristics and investigates their desirable properties. These indices can be utilized to achieve anchored judgments in decision-making processes and to reflect a certain balance between connections with positive and negative ideal points of reference. Moreover, these indices can fully consider the amount of information associated with higher degrees of uncertainty and effectively fuse imprecise and ambiguous evaluative ratings to construct a meaningful comparison approach. By using the correlation-based closeness index, this paper establishes effective algorithmic procedures of the proposed IVPF compromise approach for conducting multiple-criteria evaluation tasks within IVPF environments. The proposed methodology is implemented in a practical problem of selecting a suitable bridge construction method to demonstrate its feasibility and applicability. The practicality and effectiveness of the proposed methodology are verified through a comparative analysis with well-known compromise methods and other relevant nonstandard fuzzy models.","Title: A novel VIKOR method with an application to multiple criteria decision analysis for hospital-based post-acute care within a highly complex uncertain environment\nYear: 2019\n\nAbstract: Post-acute care (PAC) is an interdisciplinary healthcare service used to assist in returning patients to their community after they receive acute medical services. To establish a fully developed PAC service model, the National Health Insurance Administration in Taiwan attempts to recruit professional medical service institutions (consisting of medical centers and regional and district hospitals) in a trial program for improving the rehabilitation quality of patients with cerebrovascular diseases (CVDs) and to reduce the re-hospitalization rate. The evaluation and selection of adequate medical service institutions are critical in the program for enhancing the effectiveness of hospital-based PAC in acute stroke management. However, because of the complexity of the national health insurance system and the healthcare system in Taiwan, the determination of pilot hospitals is a highly complicated and ambiguous multiple criteria decision analysis (MCDA) problem. Focusing on the requirement of generating a set of pilot hospitals for the PAC program and tackling imprecise and uncertain information associated with a complex medical circumstance, the purpose of this paper is to develop a novel compromising decision-making method based on the interval-valued Pythagorean fuzzy (IVPF) set theory and the VlseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) methodology to address the multiple criteria selection problem of pilot hospitals in assisting the establishment of a hospital-based PAC model. Considering the powerfulness of IVPF sets when handling vagueness and complex uncertainty in practical problems, this paper proposes a useful IVPF VIKOR method that is significantly different from the existing VIKOR methodology. The proposed methodology represents a comprehensive integration of high-order uncertainties within a decision environment into the basic VIKOR structure that is designed to lead to both better description and better applicability of MCDA. This paper presents some novel concepts such as remoteness indices and remoteness-based multiple criteria ranking indices and investigates their desirable properties in detail. With the theoretical support of these new concepts, this paper establishes useful remoteness-based group utility indices, individual regret indices, and compromise indices for the assessment of acceptable advantage and acceptable stability. Unlike the current VIKOR-based ranking process, this paper provides a systematic ranking procedure that is capable of improving the efficiency of determining ultimate compromise solutions. Overall, the present study provides several significant contributions, such as structuring the selection problem for hospital-based PAC, extending the IVPF theory and VIKOR to the medical and healthcare fields, simplifying the manipulation procedure in handling IVPF information, constructing valuable concepts of remoteness-based indices, and developing an effective IVPF VIKOR ranking procedure. A real-world example of selecting pilot hospitals in the PAC program for CVDs is provided to illustrate the application of the proposed method and to demonstrate its practicality and effectiveness. Furthermore, other valuable MCDA applications with a comparative analysis are conducted to validate the advantages of the proposed method in a variety of fields.","Title: A novel VIKOR method with an application to multiple criteria decision analysis for hospital-based post-acute care within a highly complex uncertain environment\nYear: 2019\n\nAbstract: Post-acute care (PAC) is an interdisciplinary healthcare service used to assist in returning patients to their community after they receive acute medical services. To establish a fully developed PAC service model, the National Health Insurance Administration in Taiwan attempts to recruit professional medical service institutions (consisting of medical centers and regional and district hospitals) in a trial program for improving the rehabilitation quality of patients with cerebrovascular diseases (CVDs) and to reduce the re-hospitalization rate. The evaluation and selection of adequate medical service institutions are critical in the program for enhancing the effectiveness of hospital-based PAC in acute stroke management. However, because of the complexity of the national health insurance system and the healthcare system in Taiwan, the determination of pilot hospitals is a highly complicated and ambiguous multiple criteria decision analysis (MCDA) problem. Focusing on the requirement of generating a set of pilot hospitals for the PAC program and tackling imprecise and uncertain information associated with a complex medical circumstance, the purpose of this paper is to develop a novel compromising decision-making method based on the interval-valued Pythagorean fuzzy (IVPF) set theory and the VlseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) methodology to address the multiple criteria selection problem of pilot hospitals in assisting the establishment of a hospital-based PAC model. Considering the powerfulness of IVPF sets when handling vagueness and complex uncertainty in practical problems, this paper proposes a useful IVPF VIKOR method that is significantly different from the existing VIKOR methodology. The proposed methodology represents a comprehensive integration of high-order uncertainties within a decision environment into the basic VIKOR structure that is designed to lead to both better description and better applicability of MCDA. This paper presents some novel concepts such as remoteness indices and remoteness-based multiple criteria ranking indices and investigates their desirable properties in detail. With the theoretical support of these new concepts, this paper establishes useful remoteness-based group utility indices, individual regret indices, and compromise indices for the assessment of acceptable advantage and acceptable stability. Unlike the current VIKOR-based ranking process, this paper provides a systematic ranking procedure that is capable of improving the efficiency of determining ultimate compromise solutions. Overall, the present study provides several significant contributions, such as structuring the selection problem for hospital-based PAC, extending the IVPF theory and VIKOR to the medical and healthcare fields, simplifying the manipulation procedure in handling IVPF information, constructing valuable concepts of remoteness-based indices, and developing an effective IVPF VIKOR ranking procedure. A real-world example of selecting pilot hospitals in the PAC program for CVDs is provided to illustrate the application of the proposed method and to demonstrate its practicality and effectiveness. Furthermore, other valuable MCDA applications with a comparative analysis are conducted to validate the advantages of the proposed method in a variety of fields.","Title: A Pythagorean Fuzzy Topsis Method Based On Novel Correlation Measures And Its Application To Multiple Criteria Decision Analysis Of Inpatient Stroke Rehabilitation\nYear: 2019\n\nAbstract: The complex nature of the realistic decision-making process requires the use of Pythagorean fuzzy (PF) sets which have been shown to be a highly promising tool capable of solving highly vague and imprecise problems. Multiple criteria decision analysis (MCDA) methods within the PF environment are very attractive approaches for today's intricate decision environments. With this study, an effective compromise model named as the PF technique for order preference by similarity to ideal solutions (TOPSIS) is proposed based on some novel PF correlation-based concepts to overcome the complexities and ambiguities involved in real-life decision situations. In contrast to the existing distance-based definitions, this paper develops new closeness indices based on an extended concept of PF correlations. This paper employs the proposed PF correlation coefficients to construct two types of closeness measures. A comprehensive concept of PF correlation-based closeness indices can then be established to balance the consequences yielded by the two closeness measures. Based on these useful concepts, an effective PF TOPSIS method is proposed to address MCDA problems involving PF information and determine the ultimate priority orders among competing alternatives. Feasibility and practicability of the developed approach are illustrated by a medical decision-making problem of inpatient stroke rehabilitation. Finally, the proposed methodology is compared with other current methods to further explain its effectiveness. (c) 2019 The Authors. Published by Atlantis Press SARL.","Title: Novel Generalized Distance Measure of Pythagorean Fuzzy Sets and a Compromise Approach for Multiple Criteria Decision Analysis Under Uncertainty.\nYear: 2019\n\nAbstract: This paper aims at developing a novel generalized distance measure of Pythagorean fuzzy (PF) sets and constructing a distance-based compromise approach for multiple criteria decision analysis (MCDA) within PF environments. The theory of Pythagorean fuzziness provides a representative model of nonstandard fuzzy sets; it is valuable for representing complex vague or imprecise information in many practical applications. The distance measure for Pythagorean membership grades is important because it can effectively quantify the separation between PF information. Based on the essential characteristics of PF sets (membership, non-membership, strength, and direction), this paper proposes several distance measures, namely, new Hamming and Euclidean distances and a generalized distance measure that is based on them for Pythagorean membership grades and for PF sets. Moreover, the useful and desirable properties of the proposed PF distance measures are investigated to evaluate their advantages and form a solid theoretical basis. In addition, to evaluate the performance of the proposed distance measures in practice, this paper establishes a PF-distance-based compromise approach for addressing MCDA problems that involve PF information. The effectiveness and practicability of the developed approach are further evaluated through a case study on bridge-superstructure construction methods. According to the application results and comparative analysis, the proposed PF distance measures are accurate and outperform other methods in handling the inherent uncertainties of evaluation information. Furthermore, the PF-distance-based compromise approach can accommodate the much higher degrees of uncertainty in real-life decision scenarios and effectively determine the priority ranking among candidate alternatives for managing complicated MCDA problems.","Title: TCBNN: Error-Correctable Ternary-Coded Binarized Neural Network\nYear: 2021\n\nAbstract: Acceleration and reliability are two critical issues of artificial intelligent circuits and systems. In this paper we propose a novel structure of deep neural network in testing stage that converts the trained weights to optimized ternary-coded binary. Within each shallow layer a constant-shift sub-layer with almostfree cost is inserted to transfer all multipliers to a carry save adder/subtractor....","Title: A T-Spherical Fuzzy Electre Approach For Multiple Criteria Assessment Problem From A Comparative Perspective Of Score Functions\nYear: 2021\n\nAbstract: The theory involving T-spherical fuzziness provides an exceptionally good tool to efficiently manipulate the impreciseness, equivocation, and vagueness inherent in multiple criteria assessment and decision-making processes. By exploiting the notions of score functions and distance measures for complex T-spherical fuzzy information, this paper aims to propound an innovational T-spherical fuzzy ELECTRE (ELimination Et Choice Translating REality) approach to handling intricate and convoluted evaluation problems. Several newly-created score functions are employed from the comparative perspective to constitute a core procedure concerning concordance and discordance determination in the current T-spherical fuzzy ELECTRE method. By the agency of a realistic application, this paper appraises the usefulness and efficacy of available score functions in the advanced ELECTRE mechanism under T-spherical fuzzy uncertainties. This paper incorporates two forms of Minkowski distance measures into the core procedure; moreover, the effectuality of the advocated measure in differentiating T-spherical fuzzy information is validated. The effectiveness outcomes of the evolved method have been investigated through the medium of an investment decision regarding potential company options for extending the business scope. The real-world application also explores the comparative advantages of distinct score functions in tackling multiple criteria decision-making tasks. Finally, this paper puts forward a conclusion and future research directions.","Title: Error Correctable Range-Addressable Lookup for Activation and Quantization in AI Automotive Electronics\nYear: 2021\n\nAbstract: In this paper, we generalize the lightweight-slope piecewise-line range-addressable lookup table for efficient approximate computing in any activation/quantization function, and propose an error-correcting algorithm and circuitry using the AN codes for enhancing reliability. A BLER/SNR simulation proves the SEC/DEC capability of the firstly-presented AN-coded neuron. Comparisons with similar state...","Title: Multiple criteria choice modeling using the grounds of T-spherical fuzzy REGIME analysis\nYear: 2022\n\nAbstract: In an uncertain context of T-spherical fuzzy (T-SF) sets, the purpose of this study is to propound an efficacious multiple criteria choice model, named a T-SF REGIME method, predicated on a dominance analysis and a prioritization analysis. The REGIME methodology is a well-founded evaluation technique suitable for multiple criteria choice analysis owing to it has good applicability to the processing of qualitative information and to complex scenarios. Unlike the existing REGIME techniques, this paper takes advantage of a new Minkowski-type distance measure and an evolved Gaussian preference function to propound a ground-breaking T-SF REGIME model. This paper first concentrates on an evolution of the Minkowski-type distance measure and explores special cases with relevant properties for functional use. Next, this paper concerns an ascertainment of the evolved Gaussian preference function as a part of the proposed T-SF REGIME mechanism. This paper focuses on an advancement of the REGIME structure to T-SF uncertain contexts and unfolds an inventive T-SF REGIME methodology by virtue of superiority identifiers/indices, guide identifiers/indices, and joint generalized indices. The usefulness and strength points of the advanced methodology are scrutinized through a realistic application in the field of solar plant to prioritize the options of photovoltaic cells. The application results and comparisons corroborate the high efficiency and strengths of the evolved methodology.","Title: Decision support modeling for multiple criteria assessments using a likelihood-based consensus ranking method under Pythagorean fuzzy uncertainty\nYear: 2022\n\nAbstract: This paper intends to exploit point operator-oriented likelihood measures to constitute a likelihood-based consensus ranking model aimed at conducting multiple criteria decision making encompassing complex uncertain evaluations with Pythagorean fuzzy sets. This paper takes advantage of Pythagorean fuzzy point operators and the scalar functions of upper and lower estimations to formulate a point operator-oriented likelihood measure for preference intensity. On this basis, this paper propounds the notion of penalty weights to characterize dominated relations for acquiring the measurement of comprehensive disagreement and constituting a likelihood-based consensus ranking model. The primary contributions of this study are fourfold. Firstly, two useful point operators are initiated for upper and lower estimations towards Pythagorean membership grades. Secondly, an effective likelihood measure is exploited for determining outranking relations of Pythagorean fuzzy information. Thirdly, a pragmatic concept of penalty weights is proposed for characterizing the dominated relations among alternatives and measuring degrees of comprehensive disagreement. Fourthly, a functional likelihood-based consensus ranking model is constructed for implementing a multiple criteria evaluation with Pythagorean fuzzy uncertainty. Furthermore, a real-life application relating to a financing problem is presented to provide a justification for the practicability of the proposed methodology. This paper executes an analysis of parameters sensitivity and comparative studies for showing more theoretical insights.","Title: A parametric likelihood measure with beta distributions for Pythagorean fuzzy decision-making\nYear: 2022\n\nAbstract: The objective of this research is to introduce a parametric likelihood measure based on the beta distribution and develop a likelihood-oriented methodology for solving multiple criteria decision analysis (MCDA) problems with Pythagorean fuzzy (PF) sets. With the rapid advancement of PF theory, exploring an effective approach to compare PF information is indispensable in resolving MCDA issues. The beta distribution is one the most commonly used distributions to simulate the theoretical distribution. By changing the parameter values, the beta distribution can generate symmetrical or asymmetrical patterns and various shapes, including flat or steep. Due to its flexibility and adaptability, the beta distribution is able to effectively solve complex real-world problems. To make a major contribution to the technical development of decision support applications, this paper utilizes beta distributions as a parameterization tool to introduce a new parametric likelihood measure for evaluating the outranking relationships among PF information (signified by Pythagorean membership grades). Based on the evolved likelihood-based concepts (e.g., mean outranking indices, weighted outranking grades, and comprehensive outranking measures and indices), this paper proposes a pragmatic PF likelihood-oriented method to prioritize competing alternatives under uncertain and ambiguous Pythagorean fuzzy conditions. To carefully examine the practicality and suitability of the proposed methodology in realistic decision-making environments, this paper utilizes the evolved methods to solve a realistic MCDA problem of selecting pilot hospitals in relation to postacute care. The main results that are generated by the practical application and subsequent experimental analysis and comparative study demonstrate the effectivity and superiority of the developed technique and can be used for practical purposes in flexible and convenient ways. This most important conclusion of this paper is the great aptitude and dominance of the proposed methodology based on the corroboration of the experimental and comparative results of the application. Furthermore, this study has a noticeable originality in the utilization of the generic beta distribution-based approach and the construction of an effective PF likelihood-oriented decision model, which enriches the development of decision-making applications with PF theory.","Title: Likelihood-based agreement measurements with Pythagorean fuzzy paired point operators to enrichment evaluations and priority determination for an uncertain decision-theoretical analysis\nYear: 2022\n\nAbstract: This paper aims to initiate a useful Pythagorean fuzzy likelihood function on grounds of paired point operators and scalar-valued functions, and to contrive several likelihood-based agreement measurements for enrichment evaluations and priority determination within the framework of uncertain multiple criteria analysis. This paper exploits the characterization parameters of Pythagorean membership grades to expound paired point operators and further reveals relevant theoretical benefits. Supported by scalar-valued functions regarding the admissible lower and upper estimations, an innovative likelihood function is propounded for ascertaining the possibility of Pythagorean fuzzy dominance relations. Based on the advanced superiority and inferiority estimations, a number of valuable likelihood-based agreement measurements are unfolded to enrich the assessments, including (dis)agreement measures via rank-wise fittingness, expected (dis)agreement measures as a proxy for satisfaction and hygiene estimations, and overall (dis)agreement measures. An efficacious linear programming model with a hygiene threshold is constructed for prioritizing competing alternatives. A pragmatic decision-making issue related to hospital-based post-acute care is explored to inquire into application outcomes using the established techniques. Additionally, certain comparative analyses are performed to verify the helpfulness and interesting features possessed by the advanced approach.","Title: TAIWAN Online: Test AI with AN Codes Online for Automotive Chips\nYear: 2021\n\nAbstract: Neural networks play a key role in modern AI accelerators, in which acceleration and power reduction have been two known issues. When human lives may be threatened by accidents due to a wrong decision, the reliability bursts out far beyond other issues of the AI accelerators in automotive chips. Although input-side deep layers have been shown to possess considerable self-healing, arithmetic faults in shallow decision layers may still cause unimaginable catastrophe. This motivates us to develop a synthesis system called TAIWAN Online for testing AI with AN codes online. TAIWAN Online takes a trained model with a Keras-like format and the accuracy resolution for predicting a suitable subsystem. For low-resolution datasets, a ternary-coded-binarized neural network called TCBNN is proposed for approximate computing, where AN codes are adopted for arithmetic-weight error correction. While for high-resolution datasets, redundant residue number systems are applied for parallelized acceleration, and AN codes are utilized as AN-RRNS for self-checking in efficient multiple residue-modular redundancies, MMR. Briefly pointing out the key contributions, a k-moduli AN-RRNS can highly reduce the time-area product of MMR decoders from O(k\n<sup xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">4</sup>\n) of state-of-the-art RRNSs to only one. While the TCBNN can have fewer synapses than any regular-weight-quantized BNNs. From experimental results for a neuron-based block, the MTBF can be improved up to 126 times in the proposed infection-rate model."],"string":"[\n \"Title: Experimental Analysis Of Multi-Attribute Decision-Making Based On Atanassov Intuitionistic Fuzzy Sets: A Discussion Of Anchor Dependency And Accuracy Functions\\nYear: 2012\\n\\nAbstract: This article presents a useful method for relating anchor dependency and accuracy functions to multiple attribute decision-making (MADM) problems in the context of Atanassov intuitionistic fuzzy sets (A-IFSs). Considering anchored judgement with displaced ideals and solution precision with minimal hesitation, several auxiliary optimisation models have proposed to obtain the optimal weights of the attributes and to acquire the corresponding TOPSIS (the technique for order preference by similarity to the ideal solution) index for alternative rankings. Aside from the TOPSIS index, as a decision-maker's personal characteristics and own perception of self may also influence the direction in the axiom of choice, the evaluation of alternatives is conducted based on distances of each alternative from the positive and negative ideal alternatives, respectively. This article originates from Li's [Li, D.-F. (2005), 'Multiattribute Decision Making Models and Methods Using Intuitionistic Fuzzy Sets', Journal of Computer and System Sciences, 70, 73-85] work, which is a seminal study of intuitionistic fuzzy decision analysis using deduced auxiliary programming models, and deems it a benchmark method for comparative studies on anchor dependency and accuracy functions. The feasibility and effectiveness of the proposed methods are illustrated by a numerical example. Finally, a comparative analysis is illustrated with computational experiments on averaging accuracy functions, TOPSIS indices, separation measures from positive and negative ideal alternatives, consistency rates of ranking orders, contradiction rates of the top alternative and average Spearman correlation coefficients.\",\n \"Title: A note on distances between intuitionistic fuzzy sets and/or interval-valued fuzzy sets based on the Hausdorff metric\\nYear: 2007\\n\\nAbstract: In this note, we point out some errors to Lemma 4 in Grzegorzewski [Distances between intuitionistic fuzzy sets and/or interval-valued fuzzy sets based on the Hausdorff metric, Fuzzy Sets and Systems 148, 319-328] by counterexamples. We find that the inequalities of Euclidean distances and normalized Euclidean distances for intuitionistic fuzzy sets are not valid.\",\n \"Title: A linear assignment method for multiple-criteria decision analysis with interval type-2 fuzzy sets\\nYear: 2013\\n\\nAbstract: The theory of interval type-2 fuzzy sets is valuable for modeling impressions in multiple-criteria decision analysis problems. Based on signed distances, we develop a new linear assignment method to produce an optimal preference ranking of the alternatives in accordance with a set of criterion-wise rankings and a set of criterion importance within the context of interval type-2 trapezoidal fuzzy numbers. Applying the proposed method to a case involving the selection of a landfill site, we demonstrate that the proposed method is easy to employ and that it produces actionable results that aid the decision-making process. The proposed interval type-2 fuzzy linear assignment method utilizes signed distances, does not require a complicated computation procedure and will be beneficial to decision analysis.\",\n \"Title: An ELECTRE-based outranking method for multiple criteria group decision making using interval type-2 fuzzy sets\\nYear: 2014\\n\\nAbstract: The aim of this paper is to develop an ELECTRE (ELimination Et Choice Translating REality)-based outranking method for multiple criteria group decision-making within the environment of interval type-2 fuzzy sets. Along with considering the context of interval type-2 trapezoidal fuzzy numbers, this paper employs a hybrid averaging approach with signed distances to construct a collective decision matrix and proposes the use of ELECTRE-based outranking methods to analyze the collective interval type-2 fuzzy data. By applying a signed distance approach, this work identifies the concordance and discordance sets to determine the concordance and discordance indices, respectively, for each pair of alternatives. Based on an aggregate outranking matrix, a decision graph is constructed to determine the partial-preference ordering of the alternatives and the ELECTREcally non-outranked solutions. This paper provides additional approaches at the final selection stage to yield a linear ranking order of the alternatives. The feasibility and applicability of the proposed methods are illustrated with an example that addresses supplier selection, and a comparative analysis is performed with other approaches to validate the effectiveness of the proposed methodology.\",\n \"Title: Objective weights with intuitionistic fuzzy entropy measures and computational experiment analysis\\nYear: 2011\\n\\nAbstract: It is important to properly assess the weights of attributes when solving multi-attribute decision problems because variations in the weights often influence the rankings of the alternatives. In this paper, we propose an alternative objective weighting method to generate objective weights based on intuitionistic fuzzy (IF) entropy measures, which depend on the nature of a decision matrix in an intuitionistic fuzzy environment. Instead of the traditional fuzzy entropy measure, which is characterized by using the discriminating power to calculate the attribute weights, the proposed approach adopts the IF entropy measure, which emphasizes the credibility of the data. The IF entropy measure applied here is derived from a geometric interpretation of intuitionistic fuzzy sets and incorporates the concept of a ratio of distance measures. We implement four distance measures in the proposed approach and compare them in a computational experiment. The experimental results indicate that different IF entropy measures used in weighting methods can generate distinct objective attribute weights. In particular, when the number of attributes increases, the discrepancy between the IF entropy measures increases.\",\n \"Title: Identification Of General Fuzzy Measures By Genetic Algorithms Based On Partial Information\\nYear: 2000\\n\\nAbstract: This study develops an identification procedure for general fuzzy measures using genetic algorithms, In view of the difficulty in data collection in practice, the amount of input data is simplified through a sampling procedure concerning attribute subsets, and the corresponding detail design is adapted to the partial information acquired by the procedure. A specially designed genetic algorithm is proposed for better identification, including the development of the initialization procedure, fitness function, and three genetic operations. To show the applicability of the proposed method, this study simulates a set of experimental data that are representative of several typical classes. The experimental analysis indicates that using genetic algorithms to determine general fuzzy measures can obtain satisfactory results under the framework of partial information.\",\n \"Title: Expanding competence sets for the consumer decision problem\\nYear: 2002\\n\\nAbstract: The framework of competence set analysis provides a new approach to complement the existing models for the consumer decision problem. Since it is essential for the firm to improve the competence set of its product or service to fully address the consumer's truly needed benefits, the effective expansion of competence sets plays an important role in marketing reality. The previous studies regarding competence set expansion have thrown light on the tree expansion processes, but forest learning is more suitable for the acquisition of product benefits than tree learning. Thus, this study looses the assumption of tree learning and conducts forest learning to design an effective expansion program. In addition, this study explores a more general problem involving intermediate attributes, compound benefits, and experiential effects. An algorithm is also provided for effective expansion of competence sets in consumer decision analysis.\",\n \"Title: A signed-distance-based approach to importance assessment and multi-criteria group decision analysis based on interval type-2 fuzzy set.\\nYear: 2013\\n\\nAbstract: Interval type-2 fuzzy sets are associated with greater imprecision and more ambiguities than ordinary fuzzy sets. This paper presents a signed-distance-based method for determining the objective importance of criteria and handling fuzzy, multiple criteria group decision-making problems in a flexible and intelligent way. These advantages arise from the method’s use of interval type-2 trapezoidal fuzzy numbers to represent alternative ratings and the importance of various criteria. An integrated approach to determine the overall importance of the criteria is also developed using the subjective information provided by decision-makers and the objective information delivered by the decision matrix. In addition, a linear programming model is developed to estimate criterion weights and to extend the proposed multiple criteria decision analysis method. Finally, the feasibility and effectiveness of the proposed methods are illustrated by a group decision-making problem of patient-centered medicine in basilar artery occlusion.\",\n \"Title: Using competence sets to analyze the consumer decision problem\\nYear: 2001\\n\\nAbstract: This study is intended to provide a different approach to complement the existing consumer decision models (CDMs). Based on the concept of habitual domains and competence sets, we supply a framework for helping a firm in expanding the benefits of its products to fully address the consumer’s needs. According to the features of consumers’ decision making, we use challenging problem types to explore extensive problem solving, fuzzy problem types for limited problem solving, and routine and mixed routine problem types for routine problem solving. In addition, several useful indexes are established using fuzzy measures in this study, including the possibility of successfully appealing to consumers, the degree of consumer satisfaction, the degree of compatibility, and the degree of uniqueness. These indexes can be a decision support for implementing competence set analysis in practical applications. Finally, an empirical study on children’s apparel was conducted to show the applicability and feasibility of our proposed method in practice.\",\n \"Title: Comparative analysis of SAW and TOPSIS based on interval-valued fuzzy sets: Discussions on score functions and weight constraints\\nYear: 2012\\n\\nAbstract: Interval-valued fuzzy sets involve more uncertainties than ordinary fuzzy sets and can be used to capture imprecise or uncertain decision information in fields that require multiple-criteria decision analysis (MCDA). This paper takes the simple additive weighting (SAW) method and the technique for order preference by similarity to an ideal solution (TOPSIS) as the main structure to deal with interval-valued fuzzy evaluation information. Using an interval-valued fuzzy framework, this paper presents SAW-based and TOPSIS-based MCDA methods and conducts a comparative study through computational experiments. Comprehensive discussions have been made on the influence of score functions and weight constraints, where the score function represents an aggregated effect of positive and negative evaluations in performance ratings and the weight constraint consists of the unbiased condition, positivity bias, and negativity bias. The correlations and contradiction rates obtained in the experiments suggest that evident similarities exist between the interval-valued fuzzy SAW and TOPSIS rankings.\",\n \"Title: The extended QUALIFLEX method for multiple criteria decision analysis based on interval type-2 fuzzy sets and applications to medical decision making.\\nYear: 2013\\n\\nAbstract: QUALIFLEX, a generalization of Jacquet-Lagreze's permutation method, is a useful outranking method in decision analysis because of its flexibility with respect to cardinal and ordinal information. This paper develops an extended QUALIFLEX method for handling multiple criteria decision-making problems in the context of interval type-2 fuzzy sets. Interval type-2 fuzzy sets contain membership values that are crisp intervals, which are the most widely used of the higher order fuzzy sets because of their relative simplicity. Using the linguistic rating system converted into interval type-2 trapezoidal fuzzy numbers, the extended QUALIFLEX method investigates all possible permutations of the alternatives with respect to the level of concordance of the complete preference order. Based on a signed distance-based approach, this paper proposes the concordance/discordance index, the weighted concordance/discordance index, and the comprehensive concordance/discordance index as evaluative criteria of the chosen hypothesis for ranking the alternatives. The feasibility and applicability of the proposed methods are illustrated by a medical decision-making problem concerning acute inflammatory demyelinating disease, and a comparative analysis with another outranking approach is conducted to validate the effectiveness of the proposed methodology. (C) 2012 Elsevier B.V. All rights reserved.\",\n \"Title: Data Construction Process And Qualiflex-Based Method For Multiple-Criteria Group Decision Making With Interval-Valued Intuitionistic Fuzzy Sets\\nYear: 2013\\n\\nAbstract: Based on Jacquet-Lagreze's permutation method, QUALIFLEX is an outranking model that investigates all possible permutations of alternatives with respect to the consequences of all criteria. The purpose of this paper is to develop a QUALIFLEX-based method for multiple criteria group decision making within a decision environment of interval-valued intuitionistic fuzzy sets. We conduct a statistical inference approach with finite population correction to construct interval-valued intuitionistic fuzzy numbers. In addition, we incorporate the relative importance of decision makers and fuse individual opinions to form collective ratings using a modified method with weighted interval estimations. In view of diversiform preference types (weak order, strict order, difference order, interval bound, and ratio bound), we represent multiple decision makers' various forms of preference structures and assess criterion weights under incomplete information. By means of score functions, accuracy functions, membershipuncertainty indices, and hesitation-uncertainty indices, a ranking procedure is employed to identify a criterion-wise preference of alternatives. A QUALIFLEX-based model is then established to measure the level of concordance of the complete preference order for handling multiple criteria group decisions. The feasibility of the proposed method is illustrated by a practical problem relating to the selection of a landfill site. As indicated in the application, the proposed method is useful for handling complicated group decision-making problems that involve comprehensive criteria and limited alternatives.\",\n \"Title: A PROMETHEE-based outranking method for multiple criteria decision analysis with interval type-2 fuzzy sets\\nYear: 2014\\n\\nAbstract: This paper develops new methods based on the preference ranking organization method for enrichment evaluations (PROMETHEE) that use a signed distance-based approach within the environment of interval type-2 fuzzy sets for multiple criteria decision analysis. The theory of interval type-2 fuzzy sets provides an intuitive and computationally feasible way of addressing uncertain and ambiguous information in decision-making fields. Many studies have developed multiple criteria decision analysis methods in the context of interval type-2 fuzzy sets; most of these methods can be characterized as scoring or compromising models. Nevertheless, the extended versions of outranking methods have not been thoroughly investigated. This paper establishes interval type-2 fuzzy PROMETHEE methods for ranking alternative actions among multiple criteria based on the concepts of signed distance-based generalized criteria and comprehensive preference indices. We develop interval type-2 fuzzy PROMETHEE I and interval type-2 fuzzy PROMETHEE II procedures for partial and complete ranking, respectively, of the alternatives. Finally, the feasibility and applicability of the proposed methods are illustrated by a practical problem of landfill site selection. A comparative analysis is also performed with ordinary fuzzy PROMETHEE methods to validate the effectiveness of the proposed methodology.\",\n \"Title: Using fuzzy measures and habitual domains to analyze the public attitude and apply to the gas taxi policy\\nYear: 2002\\n\\nAbstract: Public acceptance and support are the crucial keys for implementing public policies successfully. Thus, the understanding of public acceptance or rejection towards the policy, as well as the important attributes of concern, could be very helpful to implementing the policy. However, most conventional attitude models could not approximate people's subjective evaluation process exactly by virtue of the additivity and independence assumptions. Additionally, people's decision behavior is deeply affected by their existing habits. Since habitual domains exist in the decision process, if the government can change or extend people's habitual thinking in favor of the public policy, the policy will receive satisfactory acceptance. Therefore, this study uses the habitual domain theory to analyze the public's attitude towards public policies. Furthermore, general fuzzy measures and fuzzy integrals, which require only boundary conditions and monotonicity, are also applied to develop a public attitude analysis model. An empirical study on the compress natural gas (CNG) taxi policy in Taipei City is conducted to show the applicability of the proposed model. The empirical results indicate that there are significant differences between the public's concern and governmental publicity, and some valuable strategies are suggested to the government.\",\n \"Title: Approaches To Leniency Reduction In Multi-Criteria Decision Making With Interval-Valued Fuzzy Sets And An Experimental Analysis\\nYear: 2012\\n\\nAbstract: The purpose of this paper is to present a useful method for estimating the importance of criteria and reducing the leniency bias in multiple criteria decision analysis based on interval-valued fuzzy sets. Several types of net predispositions are defined to represent an aggregated effect of interval-valued fuzzy evaluations. The suitability function for measuring the overall evaluation of each alternative is then determined based on simple additive weighting (SAW) methods. Because positive or negative leniency may exist when most criteria are assigned unduly high or low ratings, respectively, deviation variables are introduced to mitigate the effects of overestimated and underestimated ratings of criterion importance. Considering the two objectives of maximal weighted suitability and minimal deviation values, an integrated programming model is proposed to compute optimal weights for the criteria and corresponding suitability degrees for alternative rankings. A flexible algorithm using interval-valued fuzzy SAW methods is established by considering both objective and subjective information to compute optimal multiple criteria decisions. The feasibility and effectiveness of the proposed methods are illustrated by a numerical example. Finally, an experimental analysis of interval-valued fuzzy rankings given different conditions for the criterion weights is conducted with discussions on average Spearman correlation coefficients and contradiction rates.\",\n \"Title: An interval-valued intuitionistic fuzzy LINMAP method with inclusion comparison possibilities and hybrid averaging operations for multiple criteria group decision making\\nYear: 2013\\n\\nAbstract: An interval-valued intuitionistic fuzzy set contains membership and non-membership values that are crisp intervals and represent an extension of the ordinary fuzzy sets that are widely used because of their usefulness in handling imprecise or uncertain information. The linear programming technique for multidimensional analysis of preference (LINMAP) is a representative decision-making method with respect to preference information for given alternatives. In this paper, we present a new linear programming technique with weight assessment, an extended LINMAP method, for addressing multiple criteria group decision-making problems in the interval-valued intuitionistic fuzzy framework. With consideration given to the degrees of relative agreement and the importance weights of multiple decision makers, this paper presents an inclusion-based hybrid averaging operation with an inclusion comparison approach for forming a collective decision environment. The concept of inclusion-based indices that relate to anchor dependency with multiple points of reference is developed as the core of the extended LINMAP method. We also establish a linear programming model to handle the incomplete preference information for alternatives. The optimal weights of the criteria can be determined, and the priority order of the alternatives can be obtained according to the resulting comprehensive inclusion-based indices. The feasibility and applicability of the proposed methods are illustrated with an example addressing graduate admission, and a comparative analysis is performed with another interval-valued intuitionistic fuzzy LINMAP approach to validate the effectiveness of the proposed methodology.\",\n \"Title: Interval-valued intuitionistic fuzzy QUALIFLEX method with a likelihood-based comparison approach for multiple criteria decision analysis\\nYear: 2014\\n\\nAbstract: QUALIFLEX (i.e., QUALItative FLEXible multiple criteria method) is a useful outranking method used for multiple criteria decision analysis. This paper uses the main structure of QUALIFLEX to develop an interval-valued intuitionistic fuzzy QUALIFLEX outranking method with a likelihood-based comparison approach for handling multiple criteria decision-making problems within a decision environment of interval-valued intuitionistic fuzzy sets. We propose the concept of using the likelihood of fuzzy preference relations to compare interval-valued intuitionistic fuzzy numbers. To address diversiform preference types, we represent the decision-maker's various forms of preference structures and assess the criterion weights using incomplete information. Using a criterion-wise preference of alternatives via comparison of the likelihoods, we develop a new QUALIFLEX-based model to measure the level of concordance of the complete preference order for managing multiple criteria decisions. The feasibility and applicability of the proposed methods are illustrated using a practical example, namely, the selection of a suitable bridge construction method. A comparative analysis with other relevant methods is conducted to validate the effectiveness of the proposed methodology.\",\n \"Title: A multicriteria group decision-making approach based on interval-valued intuitionistic fuzzy sets: A comparative perspective\\nYear: 2011\\n\\nAbstract: This study develops an approach to tackle multiple criteria group decision-making problems in the context of interval-valued intuitionistic fuzzy sets. Due to conflicting evaluations and insufficient information about the criteria, an interval-valued intuitionistic fuzzy preference relation matrix is employed to determine the relative importance of criteria in terms of pairwise comparisons. The decision matrix, which indicates the degree of alternatives with respect to each criterion, is expressed by interval-valued intuitionistic fuzzy numbers (IVIFNs). In order to integrate interval-valued intuitionistic fuzzy information, some special aggregation operators are created by altering the aggregation operation of IVIFNs. The three families of parametric fuzzy unions and fuzzy intersections are applied in the aggregation operation with comparisons of the ranking results of alternatives. With a linear programming method, the proposed approach uses an optimization model to obtain criterion weights in exact numbers rather than intervals, and subsequently calculates an aggregated IVIFN for each alternative. The score function and accuracy function assist in discriminating between the aggregated IVIFNs, and in generating a final rank of alternatives. Finally, an illustrative supplier selection problem is used to demonstrate how to apply the proposed approach and to observe the computational consequences resulting from various aggregation operators. The results reveal that the parameters of the aggregation operators indeed have an impact on the ranks of alternatives.\",\n \"Title: A multimeasure approach to optimism and pessimism in multiple criteria decision analysis based on Atanassov fuzzy sets\\nYear: 2011\\n\\nAbstract: The frequently used methods of handling optimism and pessimism in multiple criteria decision analysis are Hurwicz's and Yager's approaches. Despite their wide usage, a critical issue has been raised about unidimensionality. Numerous psychological researches and empirical findings have convincingly supported that optimism and pessimism do not represent opposite poles on a single, bipolar dimension, but they are conceived as two partially independent dimensions instead. The focus of this paper is a two-dimensional approach by adequately employing the degrees of membership and non-membership based on Atanassov fuzzy sets. This study develops optimistic and pessimistic estimations with several fuzzy point operators to draw the influences of optimism and pessimism on multicriteria decision making for the sake of a better fit than the unidimensional model. Given a typology from empirical grounds, the appropriate point operators specific to each type are suggested to identify adaptational outcomes. Base upon a series of new score functions, a useful method for multiple criteria decision analysis is provided to efficiently facilitate decision analysis with a multimeasure approach. Feasibility and effectiveness of the proposed methods are illustrated by a practical example.\",\n \"Title: An interactive method for multiple criteria group decision analysis based on interval type-2 fuzzy sets and its application to medical decision making\\nYear: 2013\\n\\nAbstract: The theory of interval type-2 fuzzy sets provides an intuitive and computationally feasible way of addressing uncertain and ambiguous information in decision-making fields. The aim of this paper is to develop an interactive method for handling multiple criteria group decision-making problems, in which information about criterion weights is incompletely (imprecisely or partially) known and the criterion values are expressed as interval type-2 trapezoidal fuzzy numbers. With respect to the relative importance of multiple decision-makers and group consensus of fuzzy opinions, a hybrid averaging approach combining weighted averages and ordered weighted averages was employed to construct the collective decision matrix. An integrated programming model was then established based on the concept of signed distance-based closeness coefficients to determine the importance weights of criteria and the priority ranking of alternatives. Subsequently, an interactive procedure was proposed to modify the model according to the decision-makers' feedback on the degree of satisfaction toward undesirable solution results for the sake of gradually improving the integrated model. The feasibility and applicability of the proposed methods are illustrated with a medical decision-making problem of patient-centered medicine concerning basilar artery occlusion. A comparative analysis with other approaches was performed to validate the effectiveness of the proposed methodology.\",\n \"Title: A multiobjective optimization solver using rank-niche evolution strategy\\nYear: 2006\\n\\nAbstract: A rank-niche evolution strategy (RNES) algorithm has been developed in this paper to solve unconstrained multiobjective optimization problems. A required number of Pareto-optimal solutions can be generated by the algorithm in a single run. In addition to the operations of recombination, mutation and selection used in original evolution strategy (ES), an external elite set which contains a given number of non-dominated elites is updated and trimmed by a clustering technique to maintain a uniformly distributed Pareto front. The fitness function for each individual contains the information of rank and crowding status. The selection operation using this fitness function considers the superiority and distribution simultaneously. Eight test problems illustrated in other papers are used to test RNES. For some test problems the Pareto-optimal solutions obtained by RNES are better than those obtained by GA-based algorithms.\",\n \"Title: A culture-sensitive agent in Kirman's ant model\\nYear: 2011\\n\\nAbstract: The global financial crisis brought a serious collapse involving a \\\"systemic\\\" meltdown. Internet technology and globalization have increased the chances for interaction between countries and people. The global economy has become more complex than ever before. Mark Buchanan [12] indicated that agent-based computer models will prevent another financial crisis and has been particularly influential in contributing insights. There are two reasons why culture-sensitive agent on the financial market has become so important. Therefore, the aim of this article is to establish a culture-sensitive agent and forecast the process of change regarding herding behavior in the financial market. We based our study on the Kirman's Ant Model[4,5] and Hofstede's Natational Culture[11] to establish our culture-sensitive agent based model. Kirman's Ant Model is quite famous and describes financial market herding behavior from the expectations of the future of financial investors. Hofstede's cultural consequence used the staff of IBM in 72 different countries to understand the cultural difference. As a result, this paper focuses on one of the five dimensions of culture from Hofstede: individualism versus collectivism and creates a culture-sensitive agent and predicts the process of change regarding herding behavior in the financial market. To conclude, this study will be of importance in explaining the herding behavior with cultural factors, as well as in providing researchers with a clearer understanding of how herding beliefs of people about different cultures relate to their finance market strategies.\",\n \"Title: Multidisciplinary design optimization of mechanisms\\nYear: 2005\\n\\nAbstract: The optimization of mechanisms is usually done in the field of mechanism itself. As a result the structural safety of the mechanism is neglected. To ensure the safety and improve the dynamic characteristics of mechanisms, a multidisciplinary design optimization procedure is proposed in this paper to synthesize optimum mechanisms. Two disciplines are involved in the multidisciplinary design optimization. They are the mechanism and the structure. The multi-level decomposition approach is chosen to generate optimum mechanisms. The optimized mechanisms not only satisfy mechanism and structural constraints but also have optimum objective function values in the two disciplines. In order to solve general mechanism design problems two widely used commercial softwares MSC/NASTRAN and MSC/ADAMS are integrated in the procedure to do the structural and the mechanism analysis, respectively. When the structural optimization is performed, a compromised approach is introduced to treat multiple configurations of mechanisms during operation. Two mechanism design problems are given to test the proposed method.\",\n \"Title: Interval-valued fuzzy TOPSIS method with leniency reduction and an experimental analysis\\nYear: 2011\\n\\nAbstract: The purpose of this paper is to present a new TOPSIS (the technique for order preference by similarity to ideal solution) for estimating the importance of criteria and reducing the leniency bias in multiple-criteria decision analysis based on interval-valued fuzzy sets. Several types of net predispositions are defined to represent an aggregated effect of interval-valued evaluations. The relative closeness of each alternative to the ideal solution is then determined by net predispositions. Because positive or negative leniency may exist when most criteria are assigned unduly high or low ratings, respectively, some deviation variables are introduced to mitigate the effects of overestimated and underestimated ratings on criterion importance. Considering the two objectives of maximal closeness coefficient and minimal deviation values, an integrated programming model is proposed to compute optimal weights for the criteria and corresponding closeness coefficients for alternative rankings. A flexible algorithm using interval-valued fuzzy TOPSIS methods is established by considering both objective and subjective information to compute optimal multiple-criteria decisions. The feasibility and effectiveness of the proposed methods are illustrated by a numerical example. Finally, an experimental analysis of interval-valued fuzzy rankings given different conditions for the criterion weights is conducted with discussions on average Spearman correlation coefficients and contradiction rates.\",\n \"Title: Interval-valued fuzzy permutation method and experimental analysis on cardinal and ordinal evaluations\\nYear: 2009\\n\\nAbstract: This paper presents interval-valued fuzzy permutation (IVFP) methods for solving multiattribute decision making problems based on interval-valued fuzzy sets. First, we evaluate alternatives according to the achievement levels of attributes, which admits cardinal or ordinal representation. The relative importance of each attribute can also be measured by interval or scalar data. Next, we identify the concordance, midrange concordance, weak concordance, discordance, midrange discordance and weak discordance sets for each ordering. The proposed method consists of testing each possible ranking of the alternatives against all others. The evaluation value of each permutation can be computed either by cardinal weights or by solving programming problems. Then, we choose the permutation with the maximum evaluation value, and the optimal ranking order of alternatives can be obtained. An experimental analysis of IVFP rankings given cardinal and ordinal evaluations is conducted with discussions on consistency rates, contradiction rates, inversion rates, and average Spearman correlation coefficients.\",\n \"Title: Optimistic and pessimistic decision making with dissonance reduction using interval-valued fuzzy sets\\nYear: 2011\\n\\nAbstract: Interval-valued fuzzy sets have been developed and applied to multiple criteria analysis. However, the influence of optimism and pessimism on subjective judgments and the cognitive dissonance that accompanies the decision making process have not been studied thoroughly. This paper presents a new method to reduce cognitive dissonance and to relate optimism and pessimism in multiple criteria decision analysis in an interval-valued fuzzy decision environment. We utilized optimistic and pessimistic point operators to measure the effects of optimism and pessimism, respectively, and further determined a suitability function through weighted score functions. Considering the two objectives of maximal suitability and dissonance reduction, several optimization models were constructed to obtain the optimal weights for the criteria and to determine the corresponding degree of suitability for alternative rankings. Finally, an empirical study was conducted to validate the feasibility and applicability of the current method. We anticipate that the proposed method can provide insight on the influences of optimism, pessimism, and cognitive dissonance in decision analysis studies.\",\n \"Title: The ELECTRE multicriteria analysis approach based on Atanassov's intuitionistic fuzzy sets\\nYear: 2011\\n\\nAbstract: In recent decades, intuitionistic fuzzy sets have been applied to many different fields; however, few current studies have used the ELECTRE method to solve multi-criteria decision-making problems with intuitionistic fuzzy information. The purpose of this paper is to develop a new method, the intuitionistic fuzzy ELECTRE method, for solving multi-criteria decision-making problems. Atanassov's intuitionistic fuzzy set (A-IFS) characteristics are simultaneously concerned with the degree of membership, degree of non-membership, and intuitionistic index, and people can use A-IFS to describe uncertain situations in decision-making problems. We use the proposed method to rank all alternatives and determine the best alternative. The proposed method can also use imperfect or insufficient knowledge of data to deal with decision-making problems. Finally, two practical examples are given that illustrate the procedure of the proposed method.\",\n \"Title: Bivariate models of optimism and pessimism in multi-criteria decision-making based on intuitionistic fuzzy sets\\nYear: 2011\\n\\nAbstract: This paper presents a useful method of relating optimism and pessimism to multiple criteria decision analysis in the context of intuitionistic fuzzy sets based on the unipolar bivariate model. We use eight point operators to estimate the adaptational outcome expectations of optimism and/or pessimism and then determine the net predisposition, the aggregated effect of positive and negative evaluations. A series of new net predispositions for bivariate evaluations are proposed for neutrality; for complete, moderate, and rational optimism; for complete, moderate, and rational pessimism; and for complete and moderate optimism-pessimism. The suitability function, which measures the overall evaluation of each alternative, is then presented. Because positive or negative leniency may exist, such that most of the criteria may be assigned unduly high or low ratings, respectively, we introduce deviation variables to mitigate the effects of such ratings on the apparent importance of various criteria. Based on the two objectives of maximal weighted suitability and minimal deviation values, an integrated programming model is used to compute the optimal weights for the criteria and the corresponding degrees of suitability of the alternative rankings. We establish flexible algorithms that incorporate both objective and subjective information to compute the optimal optimistic and pessimistic decisions. The proposed methods are illustrated and discussed using a numerical example, a multi-criteria supplier selection problem. Finally, an empirical study of job choices is conducted to establish the feasibility and applicability of the current method.\",\n \"Title: Synchronous Pipeline Circuit Design for an Adaptive Neuro-fuzzy Network\\nYear: 2007\\n\\nAbstract: This paper presents an efficient synchronous pipeline hardware implementation procedure for a neuro-fuzzy (NF) circuit. We decompose the NF circuit into a feedforward circuit and a backpropagation circuit. The concept of pre-calculation to share computation results between the feedforward circuit and backpropagation circuit is introduced to achieve a high throughput rate and low resource usage. A novel pipeline architecture has been adopted to fulfill the concept of pre-calculation. With the unique pipeline architecture, we have successfully enhanced the throughput rate and resource sharing between modules. Particularly, the multiplier usage has been reduced from 7 to 3 and the divider usage from 3 to 1. Finally, we have implemented the NF circuit on FGPA. Our experimental results show a superior performance than that of an asynchronous pipeline design approach and the NF system implemented on MATLAB®.\",\n \"Title: Determining objective weights with intuitionistic fuzzy entropy measures: A comparative analysis\\nYear: 2010\\n\\nAbstract: In this paper, we propose a new objective weighting method that employs intuitionistic fuzzy (IF) entropy measures to solve multiple-attribute decision-making problems in the context of intuitionistic fuzzy sets. Instead of traditional fuzzy entropy, which uses the probabilistic discrimination of attributes to obtain attribute weights, we utilize the IF entropy to assess objective weights based on the credibility of the input data. We examine various measures for IF entropy with respect to hesitation degree, probability, non-probability, and geometry to calculate the attribute weights. A comparative analysis of different measures to generate attribute rankings is illustrated with both computational experiments as well as analyses of Pearson correlations, Spearman rank correlations, contradiction rates, inversion rates, and consistency rates. The experimental results indicate that ranking the outcomes of attributes not only depends on the type of IF entropy measures but is also affected by the number of attributes and the number of alternatives.\",\n \"Title: On the improvements of the particle swarm optimization algorithm\\nYear: 2010\\n\\nAbstract: Since a particle swarm optimization (PSO) algorithm uses a coordinated search to find the optimum solution, it has a better chance of finding the global solution. Despite this advantage, it is also observed that some parameters used in PSO may affect the solution significantly. Following this observation, this research tries to tune some of the parameters and to add mechanisms to the PSO algorithm in order to improve its robustness in finding the global solution. The main approaches include using uniform design to ensure uniform distribution of the initial particles in the design space, adding a mutation operation to increase the diversity of particles, decreasing the maximum velocity limitation and the velocity inertia automatically to balance the local and the global search efforts, reducing velocity when constraints are violated, and using Gaussian distribution based local searches to escape local minima. Besides these efforts, an algorithm is also developed to find multiple solutions in a single run. The results show that the overall effect of these approaches can yield better results for most test problems.\",\n \"Title: A comparative study of optimistic and pessimistic multicriteria decision analysis based on atanassov fuzzy sets\\nYear: 2012\\n\\nAbstract: Following the unipolar bivariate model, this paper presents a useful method of relating optimism and pessimism to multiple criteria decision analysis within the context of Atanassov fuzzy sets. We utilize four point operators to readily estimate the adaptational outcome expectancies generated by optimistic and pessimistic attitudes. We then determine the net predisposition to represent the aggregated effects of positive and negative evaluations. On the basis of the net predispositions, we calculate the suitability function to evaluate each alternative. Accordingly, we propose the optimization model with suitability functions to deal with ill-known grades of importance. We develop algorithms to aid multicriteria decision-making processes under neutral, completely optimistic, moderately optimistic, completely pessimistic, and moderately pessimistic conditions. Furthermore, we present experimental analysis on randomly generated decision problems to investigate the ranking system's consistency and the ranking irregularity issues found in the proposed methods. We establish several valuable test criteria to discuss some types of ranking consistency and irregularities (e.g., ranking consistency, ranking correlations, ranking contradictions, and ranking inversions) that occur when the different methods are used. We designed three computational experiments to compare the ranking orders yielded by different methods and examined several comparison indices, including the consistency rate of the total orders, the contradiction rate of the best choice, and the inversion rate between the better choices and the worse ones. Finally, we provided a second-order regression model to highlight the effects of variant parameter settings on the average Spearman rank correlation coefficients.\",\n \"Title: A causal model of consumer involvement: A new approach with intuitionistic fuzzy automata\\nYear: 2008\\n\\nAbstract: Involvement has become an important variable in consumer behavior and marketing research for a long time. The purpose of this study is to develop a model that can explain the complicated relationship between antecedents, consequences, and different types of involvement at the same time to understand the internal state of consumers better. Specifically, this model has to be capable of handling the vague interactions between different types of involvement. We develop this integrated model by the using of the intuitionistic fuzzy automata. We choose the cell phone as the stimulus product in this research. There are 19 antecedents, 5 types of involvement, and 10 consequences in the integrated involvement model. According to the results, we find that the intuitionistic fuzzy automata can deal with the interactions between different types of involvement. Second, the intuitionistic fuzzy automata can observe the relations between antecedents, consequences, and 5 types of involvement at the same time. Besides, we can observe the internal state inside consumers through the intuitionistic fuzzy automata, especially the degree of hesitation when they make decisions.\",\n \"Title: Signed Distanced-Based Topsis Method For Multiple Criteria Decision Analysis Based On Generalized Interval-Valued Fuzzy Numbers\\nYear: 2011\\n\\nAbstract: The theory of interval valued fuzzy sets is very valuable for modeling impressions of decision makers. In addition, it gives ability to quantify the ambiguous nature of subjective judgments in an easy way. In this paper, by extending the technique for order preference by similarity to ideal solution (TOPSIS), it is proposed a useful method based on generalized interval valued trapezoidal fuzzy numbers (GITrFNs) for solving multiple criteria decision analysis (MCDA) problems. In view of complexity in handling sophisticated data of GITrFNs, this paper employs the concept of signed distances to establish a simple and effective MCDA method based on the main structure of TOPSIS. An algorithm based on TOPSIS method is established to determine the priority order of given alternatives by using properties of signed distances. Finally, the feasibility of the proposed method is illustrated by a practical example of supplier selection.\",\n \"Title: Exploring the effects of intuitionistic fuzzy separation measures on TOPSIS rankings\\nYear: 2009\\n\\nAbstract: The purpose of this study is to extend the TOPSIS method for solving multiple attribute decision analysis problems with intuitionistic fuzzy data. Iintuitionistic fuzzy sets are capable of coping with imprecise information due to the fact that exact data may be difficult to be precisely determined since human judgments are often vague under many conditions. In this paper, a proposed intuitionistic fuzzy version of the TOPSIS method is presented and further deals with a comparative analysis of separation measures. For the sake of the comparison of intuitionistic fuzzy TOPSIS rankings yielded by different separation measures, a simulation experiment of different sizes was generated and examined. The consistency rate, the contradiction rate of the best alternative, and average Spearman correlation coefficients are utilized to conduct a pairwise comparison for all separation measures. The results which are inclusive of one hundred combinations of ten different categories of number for each alternatives and attributes indicate that the preference orders are hardly identical using different separation measures in the intuitionistic fuzzy TOPSIS method. The experimental analysis showed that the different definitions of IFS separations indeed significantly affect the final results by means of the intuitionistic fuzzy TOPSIS method. The comparative results presented in our experimental analysis indicate differentiations in a number of important aspects with some comparative indices.\",\n \"Title: The interval-valued fuzzy TOPSIS method and experimental analysis\\nYear: 2008\\n\\nAbstract: The purpose of this paper is to extend the TOPSIS method based on interval-valued fuzzy sets in decision analysis. Hwang and Yoon developed the technique for order preference by similarity to ideal solution (TOPSIS) in 1981. TOPSIS has been widely used to rank the preference order of alternatives and determine the optimal choice. Considering the fact that it is difficult to precisely attach the numerical measures to the relative importance of the attributes and to the impacts of the alternatives on these attributes in some cases, therefore, the TOPSIS method has been extended for interval-valued fuzzy data in this paper. In addition, a comprehensive experimental analysis to observe the interval-valued fuzzy TOPSIS results yielded by different distance measures is presented. A comparative analysis of interval-valued fuzzy TOPSIS rankings from each distance measure is illustrated with discussions on consistency rates, contradiction rates, and average Spearman correlation coefficients. Finally, a second-order regression model is provided to highlight the effects of the number of alternatives, the number of attributes, and distance measures on average Spearmen correlation coefficients.\",\n \"Title: An outcome-oriented approach to multicriteria decision analysis with intuitionistic fuzzy optimistic/pessimistic operators\\nYear: 2010\\n\\nAbstract: The concept of Atanassov's intuitionistic fuzzy sets is a generalization of ordinary fuzzy sets. Under the intuitionistic fuzzy decision environment, this paper develops optimistic and pessimistic point operators to draw the influences of optimism and pessimism on multiple criteria decision analysis. We propose the optimistic and pessimistic score functions based on point operators to measure evaluations of the alternative with respect to each criterion. The suitability function can then be established to determine the degree to which the alternative satisfies the decision maker's requirement. An optimization model with suitability functions is propounded on account of ill-known membership grades to generate optimal weights for criteria. Finally, an outcome-oriented approach is adopted to validate the proposed methods. Several multiattribute evaluation cases in consumer decision making reality are addressed to examine the feasibility and applicability of the current method. From the empirical results of effectiveness examination, the proposed intuitionistic fuzzy multicriteria decision making method with optimistic/pessimistic score functions performed significantly better than the method without consideration of optimism/pessimism. Therefore, we can conclude that the proposed methods by using intuitionistic fuzzy optimistic/pessimistic point operators have desirable potentials in practice.\",\n \"Title: Nonlinear Assignment-Based Methods For Interval-Valued Intuitionistic Fuzzy Multi-Criteria Decision Analysis With Incomplete Preference Information\\nYear: 2012\\n\\nAbstract: In the context of interval-valued intuitionistic fuzzy sets, this paper develops nonlinear assignment-based methods to manage imprecise and uncertain subjective ratings under incomplete preference structures and thereby determines the optimal ranking order of the alternatives for multiple criteria decision analysis. By comparing each interval-valued intuitionistic fuzzy number's score function, accuracy function, membership uncertainty index, and hesitation uncertainty index, a ranking procedure is employed to identify criterion-wise preference of alternatives. Based on the criterion-wise rankings and a set of known but incomplete information about criterion weights, a nonlinear assignment model is constructed to estimate criterion weights and to order the priority of various alternatives. Considering multiple criteria evaluation problems with preference conflict about criterion importance, an integrated nonlinear programming model is further established with regard to incomplete and inconsistent weight information. These proposed nonlinear assignment-based methods can obtain an aggregate ranking that effectively combines the relative performance of each alternative in each criterion. In addition, this overall ranking most closely agrees with the criterion-wise rankings. Finally, the feasibility of the proposed method is illustrated by a practical example of selecting a suitable bridge construction method.\",\n \"Title: Multi-Criteria Decision-Making Methods With Optimism And Pessimism Based On Atanassov'S Intuitionistic Fuzzy Sets\\nYear: 2012\\n\\nAbstract: The theory of Atanassov's intuitionistic fuzzy sets (A-IFSs) developed over the last several decades has found useful application in fields requiring multiple-criteria decision analysis. Since the membership-nonmembership pair in A-IFSs belongs to the bivariate unipolarity type, this article describes an approach that relates optimism and pessimism to multi-criteria decision analysis in an intuitionistic fuzzy-decision environment. First, several optimistic and pessimistic point operators were defined to alter the estimation of decision outcomes. Next, based on the core of the estimations, optimistic and pessimistic score functions were developed to evaluate each alternative with respect to each criterion. The suitability function was then established to determine the degree to which each an alternative satisfies the decision maker's requirement. Because the information on multiple criteria corresponding to decision importance is often incomplete, this study included suitability functions in the optimisation models to account for poorly known membership grades. Using a linear equal-weighted summation method, these models were transformed into a single objective optimisation model to generate the optimal weights for criteria. The feasibility and effectiveness of the proposed methods were illustrated through a practical example. Finally, computational experiments with enormous amounts of simulation data were designed to conduct a comparative analysis on the ranking orders yielded by different optimistic/pessimistic point operators.\",\n \"Title: A comparative analysis of score functions for multiple criteria decision making in intuitionistic fuzzy settings\\nYear: 2011\\n\\nAbstract: The purpose of this paper was to conduct a comparative study of score functions in multiple criteria decision analysis based on intuitionistic fuzzy sets. The concept of score functions has been conceptualized and widely applied to multi-criteria decision-making problems. There are several types of score functions that can identify the mixed results of positive and negative parts in a bi-dimensional framework of intuitionistic fuzzy sets. Considering various perspectives on score functions, the present study adopts an order of preference based on similarity to the ideal solution as the main structure to estimate the importance of different criteria and compute optimal multi-criteria decisions in intuitionistic fuzzy evaluation settings. An experimental analysis is conducted to examine the relationship between the results yielded by different score functions, considering the average Spearman correlation coefficients and contradiction rates. Furthermore, additional discussions clarify the relative differences in the ranking orders obtained from different combinations of numbers of alternatives and criteria as well as different importance conditions.\",\n \"Title: Efficient Built-In Self-Test Schemes for Video Coding Cores: a Case Study on DCT/IDCT Circuits\\nYear: 2006\\n\\nAbstract: In general, the software-testing time may be measured by two kinds of time scales: calendar time and test-execution time. In this paper, we develop two-dimensional software reliability models with two-time measures and incorporate both of them to assess ...\",\n \"Title: Identification of general fuzzy measures by genetic algorithmsbased on partial information\\nYear: 2000\\n\\nAbstract: This study develops an identification procedure for general fuzzy measures using genetic algorithms. In view of the difficulty in data collection in practice, the amount of input data is simplified through a sampling procedure concerning attribute subsets, and the corresponding detail design is adapted to the partial information acquired by the procedure. A specially designed genetic algorithm is proposed for better identification, including the development of the initialization procedure, fitness function, and three genetic operations. To show the applicability of the proposed method, this study simulates a set of experimental data that are representative of several typical classes. The experimental analysis indicates that using genetic algorithms to determine general fuzzy measures can obtain satisfactory results under the framework of partial information\",\n \"Title: Hardware Design of an Adaptive Neuro-fuzzy Network with On-Chip Learning Capability\\nYear: 2007\\n\\nAbstract: This paper aims for the development of the digital circuit of an adaptive neuro-fuzzy network with on-chip learning capability. The on-chip learning capability was realized by a backpropagation learning circuit for optimizing the network parameters. To maximize the throughput of the circuit and minimize its required resources, we proposed to reuse the computational results in both feedforward and backpropagation circuits. This leads to a simpler data flow and the reduction of resource consumption. To verify the effectiveness of the circuit, we implemented the circuit in an FPGA development board and compared the performance with the neuro-fuzzy system written in a MATLAB®code. The experimental results show that the throughput of our neuro-fuzzy circuit significantly outperforms the NF network written in a MATLAB®code with a satisfactory learning performance.\",\n \"Title: Interval-valued fuzzy multiple criteria decision-making methods based on dual optimistic/pessimistic estimations in averaging operations.\\nYear: 2014\\n\\nAbstract: •Decision-making methods with optimistic/pessimistic averaging operations.•Multiple criteria decision analysis within an interval-valued fuzzy environment.•Algorithms involve changes in overall judgments and in the separate evaluations.•The feasibility is illustrated by an application to medical decision making.•A comparative study is implemented with computational experimental analysis.\",\n \"Title: The inclusion-based TOPSIS method with interval-valued intuitionistic fuzzy sets for multiple criteria group decision making.\\nYear: 2015\\n\\nAbstract: •An extended TOPSIS method is proposed using an inclusion comparison approach.•Multi-criteria group decision-making problems are addressed in the IVIFS context.•A modified IIHA operation is presented using inclusion-based IIOWA operations.•Inclusion-based closeness coefficients are provided in the new TOPSIS model.•The feasibility is illustrated by a medical group decision-making problem.\",\n \"Title: Multiple criteria decision analysis using a likelihood-based outranking method based on interval-valued intuitionistic fuzzy sets\\nYear: 2014\\n\\nAbstract: The purpose of this paper is to develop a likelihood-based outranking method for handling multiple criteria decision analysis problems based on interval-valued intuitionistic fuzzy sets. Using the concept of likelihood of the interval-valued intuitionistic fuzzy preference relations, this paper determines certain generalized criteria and the corresponding likelihood-based preference functions. Based on some useful concepts of comprehensive preference indices, concordance indices, counter-likelihood-based preference functions, and discordance indices, this paper conducts a concordance-discordance analysis with concordance and discordance outranking relationships to determine a global Boolean matrix and acquire partial ranking orders of the alternatives. Alternatively, this paper determines complete ranking orders of the alternatives using the concepts of net concordance indices, net discordance indices, and mean outranking values. The feasibility and applicability of the proposed method are illustrated with a practical multiple criteria decision-making application concerning the selection of a suitable bridge construction method. Finally, comparative discussions with different decision-making methods are conducted to verify the effectiveness and advantages of the proposed method in aiding decision making.\",\n \"Title: A prioritized aggregation operator-based approach to multiple criteria decision making using interval-valued intuitionistic fuzzy sets: A comparative perspective\\nYear: 2014\\n\\nAbstract: Within the environment of interval-valued intuitionistic fuzzy (IVIF) sets, this paper develops a prioritized aggregation operator-based approach to handling multiple criteria decision analysis (MCDA) problems in which there exists a prioritization relationship over evaluative criteria. The prioritization between the criteria is modeled by assessing the IVIF weights that are associated with criteria dependence on the satisfaction of the higher priority criteria. Then, a new IVIF prioritized aggregation operator is presented to aggregate the IVIF ratings of the alternatives with respect to the prioritized criteria. Based on synthetic evaluations given by the IVIF prioritized aggregation operator, the ranking order of the alternatives can be determined according to the overall evaluation values. The feasibility and applicability of the proposed method are illustrated by the practical problem of watershed site selection. Moreover, a comparative analysis with other relevant methods is conducted to validate the effectiveness. The comparative results show that our proposed prioritized aggregation operator-based method is appropriate and effective for managing MCDA problems when there are uncertainties expressed by the IVIF sets.\",\n \"Title: An interval type-2 fuzzy LINMAP method with approximate ideal solutions for multiple criteria decision analysis.\\nYear: 2015\\n\\nAbstract: The purpose of this paper is to develop a linear programming technique for multidimensional analysis of preference (LINMAP) to address multiple criteria decision analysis problems within the interval type-2 fuzzy environment based on interval type-2 trapezoidal fuzzy numbers. Considering the issue of anchor dependency, we use multiple anchor points in the decision-making process and employ approximate positive-ideal and negative-ideal solutions as the points of reference. Selected useful properties of the approximate ideal solutions are also investigated. In contrast to the classical LINMAP methods, this paper directly generates approximate ideal solutions from the characteristics of all alternatives. Next, this work presents the concept of closeness-based indices using Minkowski distances with approximate ideal solutions to develop a new approach for determining measurements of consistency and inconsistency. Under incomplete preference information on paired comparisons of the alternatives, this paper provides a novel method that uses the concept of comprehensive closeness-based indices to measure the poorness of fit and the goodness of fit. By applying the consistency indices and inconsistency indices, this work formulates an optimization problem that can be solved for the optimal weights of the criteria and thus acquires the best compromise alternative. Additionally, this paper explores the problem of supplier selection and conducts a comparative discussion to validate the effectiveness and applicability of the proposed interval type-2 fuzzy LINMAP method with approximate ideal solutions. Furthermore, the proposed method is applied to address a marketplace decision difficulty (MPDD)-prone decision-making problem to provide additional contributions for practical implications.\",\n \"Title: Likelihoods of interval type-2 trapezoidal fuzzy preference relations and their application to multiple criteria decision analysis.\\nYear: 2015\\n\\nAbstract: Interval type-2 fuzzy sets are useful and valuable for depicting uncertainty and managing imprecision in decision information. In particular, interval type-2 trapezoidal fuzzy numbers, as a special case of interval type-2 fuzzy sets, can efficiently express qualitative evaluations or assessments. In this work, the concept of the likelihoods of interval type-2 trapezoidal fuzzy preference relations based on lower and upper likelihoods is investigated, and the relevant properties are discussed. This paper focuses on the use of likelihoods in addressing multiple criteria decision analysis problems in which the evaluative ratings of the alternatives and the importance weights of the criteria are expressed as interval type-2 trapezoidal fuzzy numbers. A new likelihood-based decision-making method is developed using the useful concepts of likelihood-based performance indices, likelihood-based comprehensive evaluation values, and signed distance-based evaluation values. A simplified version of the proposed method is also provided to adapt the decision-making context in which the importance weights of the criteria take the form of ordinary numbers. The practical effectiveness of the proposed method is validated with four applications, and several comparative analyses are conducted to verify the advantages of the proposed method over other multiple criteria decision-making methods.\",\n \"Title: The Inclusion-Based Linmap Method For Multiple Criteria Decision Analysis Within An Interval-Valued Atanassov'S Intuitionistic Fuzzy Environment\\nYear: 2014\\n\\nAbstract: The purpose of this paper is to develop an inclusion-based LINMAP (i.e., Linear Programming Technique for Multidimensional Analysis of Preference) method for multiple criteria decision analysis that is based on interval-valued Atanassov's intuitionistic fuzzy sets. Using the inclusion comparison possibility in the interval-valued Atanassov's intuitionistic fuzzy context, an inclusion-based index of interval-valued Atanassov's intuitionistic fuzzy numbers is proposed that considers positive and negative ideals. An inclusion-based consistency index and an inclusion-based inconsistency index to measure the concordance and discordance, respectively, between paired comparison judgments are suggested. An inclusion-based LINMAP model is constructed using a linear programming technique to determine the optimal criterion weights and obtain the corresponding comprehensive inclusion-based index for each alternative. Then, the priority order of the alternatives can be acquired according to the comprehensive inclusion-based indices. The feasibility of the proposed method is illustrated using a practical problem that relates to the selection of bridge construction methods. A comparative analysis of other relevant decision-making methods is conducted to validate the effectiveness of the developed methodology.\",\n \"Title: An Interactive Signed Distance Approach For Multiple Criteria Group Decision-Making Based On Simple Additive Weighting Method With Incomplete Preference Information Defined By Interval Type-2 Fuzzy Sets\\nYear: 2014\\n\\nAbstract: Interval type-2 fuzzy sets (T2FSs) with interval membership grades are suitable for dealing with imprecision or uncertainties in many real-world problems. In the Interval type-2 fuzzy context, the aim of this paper is to develop an interactive signed distance-based simple additive weighting (SAW) method for solving multiple criteria group decision-making problems with linguistic ratings and incomplete preference information. This paper first formulates a group decision-making problem with uncertain linguistic variables and their transformation to interval type-2 trapezoidal fuzzy numbers. Concerning the relative importance of multiple decision-makers and group consensus of fuzzy opinions, a procedure using hybrid averages is then employed to construct a collective decision matrix. By an appropriate extension of the classical SAW approach, this paper utilizes the concept of signed distances and establishes an integrated programming model to manage multi-criteria group decisions under the incomplete and inconsistent preference structure. Further, an interactive procedure is established for group decision making. Finally, the feasibility and effectiveness of the proposed methods are illustrated by a collaborative decision-making problem of patient-centered care (PCC).\",\n \"Title: An interval type-2 fuzzy PROMETHEE method using a likelihood-based outranking comparison approach.\\nYear: 2015\\n\\nAbstract: We develop an interval type-2 fuzzy PROMETHEE method to address MCDA problems.The proposed method uses the approach of likelihood-based outranking comparisons.We present novel likelihood-based preference functions based on outranking indices.We develop two algorithmic procedures to acquire partial and complete rankings.Comparative analysis validates the effectiveness of the proposed method. Based on the preference ranking organization method for enrichment evaluations (PROMETHEE), the purpose of this paper is to develop a new multiple criteria decision-making method that uses the approach of likelihood-based outranking comparisons within the environment of interval type-2 fuzzy sets. Uncertain and imprecise assessment of information often occurs in multiple criteria decision analysis (MCDA). The theory of interval type-2 fuzzy sets is useful and convenient for modeling impressions and quantifying the ambiguous nature of subjective judgments. Using the approach of likelihood-based outranking comparisons, this paper presents an interval type-2 fuzzy PROMETHEE method designed to address MCDA problems based on interval type-2 trapezoidal fuzzy (IT2TrF) numbers. This paper introduces the concepts of lower and upper likelihoods for acquiring the likelihood of an IT2TrF binary relationship and defines a likelihood-based outranking index to develop certain likelihood-based preference functions that correspond to several generalized criteria. The concept of comprehensive preference measures is proposed to determine IT2TrF exiting, entering, and net flows in the valued outranking relationships. In addition, this work establishes the concepts of a comprehensive outranking index, a comprehensive outranked index, and a comprehensive dominance index to induce partial and total preorders for the purpose of acquiring partial ranking and complete ranking, respectively, of the alternative actions. The feasibility and applicability of the proposed method are illustrated with two practical applications to the problem of landfill site selection and a car evaluation problem. Finally, a comparison with other relevant methods is conducted to validate the effectiveness of the proposed method.\",\n \"Title: A simulated annealing-based permutation method and experimental analysis for multiple criteria decision analysis with interval type-2 fuzzy sets\\nYear: 2015\\n\\nAbstract: Development of a simulated annealing-based permutation decision-making method.Multiple criteria decision analysis within the interval type-2 fuzzy environment.Algorithm for a polynomial time solution in the total completion time problem.Computational experiments are designed to test the solution approach.Comparative analysis verifies the effectiveness and efficiency of the method. The aim of this paper is to develop a simulated annealing-based permutation method for multiple criteria decision analysis within the environment of interval type-2 fuzzy sets. The outranking methodology constitutes one of the most fruitful approaches in multiple criteria decision making and has been applied in numerous real-world problems. The permutation method is a classical outranking model, which generalizes Jacquet-Lagreze's permutation method and is based on a pairwise criterion comparison of the alternatives. Because modeling of the uncertainty in the decision-making process becomes increasingly important, an extension to the interval type-2 fuzzy environment is a useful generalization of the permutation method and is appropriate for handling uncertain and imprecise information in practical decision-making situations. This paper produces a signed-distance-based comparison among the comprehensive rankings of alternatives for concordance and discordance analyses. An integrated nonlinear programming model is constructed for estimation of the criterion weights and the optimal ranking order of the alternatives under incomplete preference information. To enhance the implementation efficiency, a simulated annealing-based permutation method and its meta-heuristic algorithm are developed to produce a polynomial time solution for the total completion time problem. Furthermore, computational experiments with notably large amounts of simulation data are conducted to test the solution approach and validate the correctness of the approximate solution compared with the optimal all-permutation-based result.\",\n \"Title: An interval-valued intuitionistic fuzzy permutation method with likelihood-based preference functions and its application to multiple criteria decision analysis.\\nYear: 2016\\n\\nAbstract: Developing an IVIF permutation method with likelihood-based preference functions.Using a hybrid approach of the likelihood calculation method and PROMETHEE.Establishing new measurements of concordance/discordance to evaluate permutations.Presenting two algorithms for IVIF importance weights and incomplete information.Verifying the effectiveness using comparative analyses and applications to NPD. This paper presents an interval-valued intuitionistic fuzzy permutation method with likelihood-based preference functions for managing multiple criteria decision analysis based on interval-valued intuitionistic fuzzy sets. First, certain likelihood-based preference functions are proposed using the likelihoods of interval-valued intuitionistic fuzzy preference relationships. Next, selected practical indices of concordance/discordance are established to evaluate all possible permutations of the alternatives. The optimal priority order of the alternatives is determined by comparing all comprehensive concordance/discordance values based on score functions. Furthermore, this paper considers various preference types and develops another interval-valued intuitionistic fuzzy permutation method using programming models to address multiple criteria decision-making problems with incomplete preference information. The feasibility and applicability of the proposed methods are illustrated in the problem of selecting a suitable bridge construction method. Moreover, certain comparative analyses are conducted to verify the advantages of the proposed methods compared with those of other decision-making methods. Finally, the practical effectiveness of the proposed methods is validated with a risk assessment problem in new product development.\",\n \"Title: An interval type-2 fuzzy technique for order preference by similarity to ideal solutions using a likelihood-based comparison approach for multiple criteria decision analysis\\nYear: 2015\\n\\nAbstract: An interval type-2 fuzzy TOPSIS method is developed to address decision problems.Use of the likelihood-based comparison approach with the approximate ideals.Establishment of likelihood-based closeness coefficients using comparison indices.Multicriteria decision analysis based on interval type-2 trapezoidal fuzzy numbers.Comparative analysis shows the effectiveness and advantages of the proposed method. The technique for order preference by similarity to ideal solutions (TOPSIS) is a well-known compromising method for addressing decision-making problems. In general, incomplete preference information and vague subjective judgments are realistic in practice. Accordingly, the theory of interval type-2 fuzzy sets has received increasing attention in the decision-making field because of its great ability to handle imprecise and ambiguous information in a convenient manner. The purpose of this paper is to develop a novel interval type-2 fuzzy TOPSIS method for multiple criteria decision analysis that is based on interval type-2 trapezoidal fuzzy numbers. This paper introduces the concept of approximate positive-ideal and negative-ideal solutions and presents a simple way to approach the evaluative ratings of ideal solutions using interval type-2 trapezoidal fuzzy numbers. Based on the likelihoods of interval type-2 trapezoidal fuzzy binary relations, this paper proposes certain likelihood-based comparison indices to establish a likelihood-based closeness coefficient of each alternative relative to the approximate ideals. Applying a likelihood-based comparison approach with the approximate ideals, this paper develops the interval type-2 fuzzy TOPSIS procedure to determine the priority ranking orders of the alternatives under consideration of the multiple criteria evaluation/selection. Three practical applications involving landfill site selection, supplier selection, and car evaluation are examined to show the effectiveness and practicability of the proposed method. Furthermore, this paper makes a comparison of the solution results yielded by other interval type-2 fuzzy decision-making methods. The comparative analyses demonstrate that the proposed interval type-2 fuzzy TOPSIS method is easy to implement and produces effective and valid results for solving multiple criteria decision-making problems.\",\n \"Title: A weight-assessing method with habitual domains\\nYear: 1998\\n\\nAbstract: Weights of criteria are the important factors in decision-making. However, from a behavioral perspective, traditional weighting methods account for too few factors to deal with decisions properly. Based on the behavior mechanism and the theory of habitual domains, this study is undertaken to develop a new weight-assessing model that treats decision-making as a dynamically adjusting process proceeding from the ideal to actual states. The new model is built upon the dynamic analysis for the connectivities between criteria instead of the static analysis of traditional models. Finally, we have studied Taipei City motorcycle users, mode-choice behavior through questionnaire in order to show the applicability of the new model. From the empirical results, it is found that our weight-assessing method has significant application potential in practice.\",\n \"Title: A likelihood-based QUALIFLEX method with interval type-2 fuzzy sets for multiple criteria decision analysis.\\nYear: 2015\\n\\nAbstract: The aim of this paper is to develop a likelihood-based QUALIFLEX method with interval type-2 fuzzy sets for multiple criteria decision analysis. This paper considers the context of interval type-2 trapezoidal fuzzy numbers and also introduces an extended concept of likelihoods of interval type-2 fuzzy preference relations. The concepts of lower and upper likelihoods are presented to serve as lower and upper bounds, respectively, of the possibility of preference relations and then to determine the likelihoods between interval type-2 trapezoidal fuzzy numbers. This paper uses a likelihood-based comparison between evaluative ratings to propose the concepts of concordance indices and comprehensive concordance indices for evaluating the permutations of alternatives corresponding to criterion-by-criterion consistency. By employing the signed distance-based method or the optimal membership-degree-determination method, the best permutation and its corresponding optimal ranking order of the alternatives can be acquired through a comparison of all of the comprehensive concordance indices. The applicability and effectiveness of the proposed method are illustrated with a medical decision-making problem that addresses the selection of treatment options. Additionally, a comparative analysis is performed with some relevant multiple criteria decision-making methods to demonstrate the advantages of the likelihood-based QUALIFLEX method. Finally, an analysis through computational experiments is conducted to examine the implementation efficiency of the proposed method.\",\n \"Title: Likelihood-based assignment methods for multiple criteria decision analysis based on interval-valued intuitionistic fuzzy sets\\nYear: 2015\\n\\nAbstract: The aim of this paper is to develop useful likelihood-based assignment methods for addressing multiple criteria decision-making problems within the environment of interval-valued intuitionistic fuzzy sets. Based on the likelihoods of interval-valued intuitionistic fuzzy preference relations, this paper determines the mean likelihoods of outranking relations and presents a mean likelihood determination method for generating a set of criterion-wise rankings of alternatives. By employing the concepts of rank frequency matrices and (ordinary) rank contribution matrices, this paper establishes a likelihood-based linear assignment model for multiple criteria decision analysis in the interval-valued intuitionistic fuzzy context. Additionally, this paper propounds two likelihood-based assignment models for handling incomplete and conflicting certain information of importance weights. These models can transform the criterion-wise ranks into the overall ranks for determining the optimal priority ranking of the alternatives. The feasibility and applicability of the proposed methods are illustrated with a practical problem of selecting a bridge construction method which involves various preference types. Finally, this paper conducts a comparative analysis with previous assignment-based methods in an interval-valued intuitionistic fuzzy setting to validate the effectiveness and advantages of the proposed methods.\",\n \"Title: IVIF-PROMETHEE outranking methods for multiple criteria decision analysis based on interval-valued intuitionistic fuzzy sets\\nYear: 2015\\n\\nAbstract: This paper attempts to develop an interval-valued intuitionistic fuzzy preference ranking organization method for enrichment evaluations (IVIF-PROMETHEE) and attempts to apply it to multiple criteria decision analysis. The theory of interval-valued intuitionistic fuzzy sets is useful for modeling impressions and quantifying the ambiguous nature of subjective judgments in a convenient manner. PROMETHEE is a well-known and widely used outranking method, but it has not been investigated in depth within the interval-valued intuitionistic fuzzy environment. Based on the concepts of inclusion comparison possibilities, this paper proposes inclusion-based generalized criteria to determine preference functions and global preference indices for acquiring leaving flows, entering flows, and net flows of alternative actions. Using the score functions and accuracy functions of the flows, this paper develops IVIF-PROMETHEE I and IVIF-PROMETHEE II, which are methods for the partial ranking and complete ranking, respectively, of alternatives. The feasibility and applicability of the proposed methods are illustrated through a problem on the selection of bridge construction methods. Finally, a comparative discussion of other decision-making methods is conducted to demonstrate the advantages of the proposed IVIF-PROMETHEE methods.\",\n \"Title: Digital Artwork Creation Using Water And Sand On A Two-Dimensional Surface\\nYear: 2015\\n\\nAbstract: In this paper, we developed a system which enables users to draw pictures of sand and water on a two-dimensional surface in a virtual environment. By mixing water and sand for artwork creation, we can create different styles of pictures. Our system has three major subsystems: water animation system, sand animation system and rendering system. In the simulation part, we compute the motion of sand and water based on the interactions for water-water, sand-water and sand-sand. In the rendering part, we attempt to balance the rendering quality and the computation cost. We render water using a ray tracing method and the sand is rendered with shadow. Thus, the rendered results of water and sand look like 2.5D. We tested our system and the experimental results showed that the users could create a variety of pictures of sand and water.\",\n \"Title: A projection-based compromising method for multiple criteria decision analysis with interval-valued intuitionistic fuzzy information.\\nYear: 2016\\n\\nAbstract: Display Omitted Multicriteria decision analysis based on interval-valued intuitionistic fuzzy sets.Development of a projection-based compromising method.Establishment of projection-based compromising indices and comprehensive indices.Determination of the ranking order and the approval status of alternatives.Comparative analyses with other relevant compromising decision-making methods. The purpose of this paper is to develop a projection-based compromising method for addressing multiple criteria decision-making problems based on interval-valued intuitionistic fuzzy sets. The concept of projections considers not only the distance but also the included angle between evaluative ratings of alternative actions with respect to a criterion. In the interval-valued intuitionistic fuzzy context, this paper determines the respective projections of the evaluative ratings of each alternative on the positive-ideal and negative-ideal solutions and explores several essential properties. Next, this paper introduces the concepts of projection-based compromising indices and comprehensive compromising indices and further investigates relevant theorems for supporting the usefulness of these indices. Additionally, this paper proposes the projection-based comparative index and the comprehensive comparative index to serve as benchmark values for the comparison purpose. The improvement percentage of the comprehensive compromising value is acquired to determine the priority order of the alternatives, including the complete ranking order and the approval status for each alternative. The feasibility and the applicability of the proposed method are validated with an application problem of watershed site selection. Finally, several comparative analyses are conducted to verify the effectiveness and advantages of the proposed method over other relevant compromising decision-making methods.\",\n \"Title: A likelihood-based assignment method for multiple criteria decision analysis with interval type-2 fuzzy information.\\nYear: 2017\\n\\nAbstract: The purpose of this paper was to develop a likelihood-based assignment method based on interval type-2 fuzzy sets and apply it to decision-making problems involving multiple criteria evaluation and the ranking/selection of alternatives. The linear assignment method is a well-known outranking method in the field of multiple criteria decision analysis. The theory of interval type-2 fuzzy sets is useful for addressing the uncertainty and imprecision associated with a subjective environment. In this paper, the key feature of the proposed method is the incorporation of the extended concept of likelihoods of fuzzy preference relations between interval type-2 trapezoidal fuzzy numbers into the main structure of the linear assignment methodology. An effective ranking procedure using the optimal membership degree determination method is proposed to determine criterion-wise preference rankings of the alternatives. The proposed method establishes the novel concepts of an (adjusted) rank frequency matrix and an (adjusted) rank contribution matrix to combine the relative performances of the alternatives in terms of each criterion. Based on a signed distance comparison approach, this paper constructs a likelihood-based assignment model to obtain an aggregate ranking of the alternatives that is in the closest agreement with the criterion-wise preferences of the alternatives. The feasibility and applicability of the proposed method are illustrated with two practical multiple criteria decision-making applications concerning the selection of landfill sites and the selection of treatment options. Finally, a comparative analysis with other relevant methods is conducted to validate the effectiveness and advantages of the current methods in decision aiding.\",\n \"Title: An Interval-Valued Pythagorean Fuzzy Outranking Method With A Closeness-Based Assignment Model For Multiple Criteria Decision Making\\nYear: 2018\\n\\nAbstract: The concept of interval-valued Pythagorean fuzzy (IVPF) sets is capable of handling imprecise and ambiguous information and managing complex uncertainty in real-world applications. This paper focuses on multiple criteria decision analysis involving IVPF information and proposes a new outranking decision-making method that uses a closeness-based assignment model. In contrast to the existing assignment-based methodology, the uniqueness of this paper is the consideration of uncertain information represented by IVPF values, the determination of criterion-wise precedence rankings based on a closeness-based approach, and the development of a new measure for scalar representation. First, to underlie anchored judgments in subjective decision-making processes, this paper presents a compromising concept of the closeness index with the positive-ideal and negative-ideal IVPF values to identify criterion-wise precedence ranks among alternatives. Next, this paper defines the concept of matrices of precedence frequency and contribution to provide a basis for the proposed assignment model. To overcome the difficulty of lacking nontrivial scalar representations, a useful measure is also developed to appropriately describe IVPF values. Based on a closeness-based assignment approach, a novel outranking decision-making method is proposed to transform the extended criterion-wise ranks into the ultimate priority orders of the alternatives. The proposed method is first implemented in a practical problem of selecting a bridge construction method to demonstrate its feasibility and applicability. Moreover, its practicality and effectiveness are verified through a comparative analysis with relevant assignment-based approaches. Further comparative analyses with newly developed IVPF decision-making methods are conducted for both a risk evaluation problem and an investment problem to examine the advantages of the proposed method and extend the current technique by considering distinct preference information for adapting to the particularities in practice. (C) 2017 Wiley Periodicals, Inc.\",\n \"Title: Remoteness index-based Pythagorean fuzzy VIKOR methods with a generalized distance measure for multiple criteria decision analysis.\\nYear: 2018\\n\\nAbstract: •Development of remoteness index-based Pythagorean fuzzy VIKOR methods.•Consideration of uncertain information represented by Pythagorean fuzzy sets.•Construction of generalized PF distances and displaced/fixed remoteness indices.•Creation of remoteness-based group utility indices and individual regret indices.•Ultimate ranking using displaced/fixed remoteness-based compromise indices.•Comparative analyses with newly developed methods via five practical applications.\",\n \"Title: A novel risk evaluation method of technological innovation using an inferior ratio-based assignment model in the face of complex uncertainty.\\nYear: 2018\\n\\nAbstract: •Development of a novel risk evaluation method of technological innovation.•Consideration of highly uncertain information represented by IVPF values.•Construction of IVPF IRs based on the IVPF distance measure.•Establishment of a useful IVPF IR-based assignment model.•Comparative analysis with other methods via a practical application.\",\n \"Title: Interactive Sand Art Drawing Using Rgb-D Sensor\\nYear: 2018\\n\\nAbstract: We present an interactive system using one RGB-D sensor, which allows a user to use bare hands to perform sand drawing. Our system supports the common sand drawing functions, such as sand erosion, sand spilling, and sand leaking. To use hands to manipulate the virtual sand, we design four key hand gestures. The idea is that the gesture of one hand controls the drawing actions. The motion and gesture of the other hand control the drawing positions. There are three major steps. First, our system adopts a vision-based bare-hand detection method which computes the hand position and recognizes the hand gestures. Second, the drawing positions and the drawing actions are sent to a sand drawing subsystem. Finally, the subsystem performs the sand drawing actions. Experimental results show that our system enables users to draw a rich variety of sand pictures.\",\n \"Title: An outranking approach using a risk attitudinal assignment model involving Pythagorean fuzzy information and its application to financial decision making.\\nYear: 2018\\n\\nAbstract: •Representation of highly uncertain information using PF/IVPF sets.•Development of risk attitude-based score functions for PF/IVPF values.•Construction of risk attitudinal assignment models in the PF/IVPF context.•Application to financial decision-making problems.•Comparative analysis with other outranking approaches.\",\n \"Title: An effective correlation-based compromise approach for multiple criteria decision analysis with Pythagorean fuzzy information.\\nYear: 2018\\n\\nAbstract: The aim of this paper is to develop a novel correlation-based compromise approach for addressing multiple criteria decision analysis problems under complex uncertainty based on Pythagorean fuzzy sets. Based on the concepts of information energy and correlations for Pythagorean fuzzy characteristics, this paper proposes useful correlation-based compromise indices and investigates their desirable properties. These compromise indices can be employed to underlie anchored judgments and to reflect a certain balance between the correlations with positive-ideal and negative-ideal points of reference. Moreover, they can fully take into consideration the amount of information associated with higher degrees of uncertainty and can effectively fuse subjective assessments conveyed by Pythagorean fuzziness. From two different perspectives of displaced and fixed ideals, this paper provides two algorithmic procedures of the proposed correlation-based compromise approach for conducting multiple criteria evaluation tasks within Pythagorean fuzzy environments. A real-world problem concerning a financing decision on working capital policies is investigated to show the feasibility and applicability of the developed techniques. The application results, along with a comparative analysis, demonstrate the practicality and effectiveness of the proposed methodology.\",\n \"Title: A Novel PROMETHEE-Based Outranking Approach for Multiple Criteria Decision Analysis With Pythagorean Fuzzy Information.\\nYear: 2018\\n\\nAbstract: This paper aims to present a novel outranking approach of the preference ranking organization method for enrichment evaluations (PROMETHEEs) based on Pythagorean fuzzy sets for multiple criteria decision analysis. The proposed method utilizes a novel Pythagorean fuzzy precedence index, which is based on the difference of scalar functions under anchored judgments with respect to the superiority/inferiority Pythagorean fuzzy numbers. To appropriately describe the hesitation between indifference and preference in the Pythagorean fuzzy context, this paper introduces useful precedence-based preference functions to establish the precedence relations based on pairwise comparisons. As a multiple criteria measure under Pythagorean fuzzy uncertainty, the concept of overall preference indices is identified at the aggregation stage to exploit certain rules for generating PROMETHEE flows. Next, this paper provides effective Pythagorean fuzzy PROMETHEE I and II ranking procedures to determine partial and complete preorders, respectively, among competing alternatives. The developed Pythagorean fuzzy PROMETHEE-based outranking approach is comparatively validated using a real-world application concerning the selection problem of bridge construction methods. The solution results along with a comparison analysis demonstrate that the proposed methodology outperforms the comparative approach in terms of reasonability and stability.\",\n \"Title: Multiple Criteria Decision Analysis Using Correlation-Based Precedence Indices Within Pythagorean Fuzzy Uncertain Environments\\nYear: 2018\\n\\nAbstract: The theory of Pythagorean fuzzy sets possesses significant advantages in handling vagueness and complex uncertainty. Additionally, Pythagorean fuzzy information is useful to simulate the ambiguous nature of subjective judgments and measure the fuzziness and imprecision more flexibly. The aim of this research is to develop an effective assignment-based method using a novel concept of correlation-based precedence indices for conducting multiple criteria decision analysis within the Pythagorean fuzzy uncertain environment. Based on the ideas of information energy and correlations, this paper defines a novel concept of correlation-based precedence indices in the Pythagorean fuzzy context and discusses their desirable properties. Next, this paper presents some useful concepts of discordance indicators, weighted discordance indicators, comprehensive discordance indicators, and comprehensive discordance indices to construct a novel assignment model for acquiring a comprehensive ranking of candidate alternatives. As an application of the proposed assignment-based method, a practical example concerning a financing decision of working capital policies is provided to demonstrate its practicality and effectiveness.\",\n \"Title: A Mixed-Choice-Strategy-Based Consensus Ranking Method for Multiple Criteria Decision Analysis Involving Pythagorean Fuzzy Information.\\nYear: 2018\\n\\nAbstract: The aim of this paper is to develop a novel consensus ranking method that uses a mixed choice strategy for multiple criteria decision analysis (MCDA) under complex uncertainty based on Pythagorean fuzzy (PF) sets. The majority of MCDA methods have focused almost exclusively on \\\"criterion-specific\\\" choice tasks that are the tasks in which all alternatives are decomposed into distinct components and evaluated on specific criteria. However, in certain MCDA problems in practical applications, category-based choices tend to be more holistic in nature, especially in affective-like aspects. Therefore, this paper incorporates a mixed choice strategy (i.e., a combination of a category-based strategy and a criterion-specific strategy) into the core structure of the developed MCDA method. Furthermore, this paper utilizes the theory of Pythagorean fuzziness to provide a powerful modeling tool for complex and varied decision-making environments. Employing the developed concepts of a PF precedence index based on PF information and a disagreement indicator based on distances between rankings, this paper proposes a novel consensus ranking method by means of a comprehensive disagreement-based assignment model for addressing a mixed-choice-strategybased MCDA problem in the PF context. As an application of the proposed methodology, a real-world case study of a luxury car selection problem is investigated. The application results, along with a comparative analysis, demonstrate the practicality and effectiveness of the developed approach, which is capable of handling hybrid category-based and criterion-specific choice tasks and managing complex uncertainty in practical situations.\",\n \"Title: An Interval-Valued Pythagorean Fuzzy Compromise Approach with Correlation-Based Closeness Indices for Multiple-Criteria Decision Analysis of Bridge Construction Methods.\\nYear: 2018\\n\\nAbstract: The purpose of this paper is to develop a novel compromise approach using correlation-based closeness indices for addressing multiple-criteria decision analysis (MCDA) problems of bridge construction methods under complex uncertainty based on interval-valued Pythagorean fuzzy (IVPF) sets. The assessment of bridge construction methods requires the consideration of multiple alternatives and conflicting tangible and intangible criteria in intricate and varied circumstances. The concept of IVPF sets is capable of handling imprecise and ambiguous information and managing complex uncertainty in real-world applications. Inspired by useful ideas concerning information energies, correlations, and correlation coefficients, this paper constructs new concepts of correlation-based closeness indices for IVPF characteristics and investigates their desirable properties. These indices can be utilized to achieve anchored judgments in decision-making processes and to reflect a certain balance between connections with positive and negative ideal points of reference. Moreover, these indices can fully consider the amount of information associated with higher degrees of uncertainty and effectively fuse imprecise and ambiguous evaluative ratings to construct a meaningful comparison approach. By using the correlation-based closeness index, this paper establishes effective algorithmic procedures of the proposed IVPF compromise approach for conducting multiple-criteria evaluation tasks within IVPF environments. The proposed methodology is implemented in a practical problem of selecting a suitable bridge construction method to demonstrate its feasibility and applicability. The practicality and effectiveness of the proposed methodology are verified through a comparative analysis with well-known compromise methods and other relevant nonstandard fuzzy models.\",\n \"Title: A novel VIKOR method with an application to multiple criteria decision analysis for hospital-based post-acute care within a highly complex uncertain environment\\nYear: 2019\\n\\nAbstract: Post-acute care (PAC) is an interdisciplinary healthcare service used to assist in returning patients to their community after they receive acute medical services. To establish a fully developed PAC service model, the National Health Insurance Administration in Taiwan attempts to recruit professional medical service institutions (consisting of medical centers and regional and district hospitals) in a trial program for improving the rehabilitation quality of patients with cerebrovascular diseases (CVDs) and to reduce the re-hospitalization rate. The evaluation and selection of adequate medical service institutions are critical in the program for enhancing the effectiveness of hospital-based PAC in acute stroke management. However, because of the complexity of the national health insurance system and the healthcare system in Taiwan, the determination of pilot hospitals is a highly complicated and ambiguous multiple criteria decision analysis (MCDA) problem. Focusing on the requirement of generating a set of pilot hospitals for the PAC program and tackling imprecise and uncertain information associated with a complex medical circumstance, the purpose of this paper is to develop a novel compromising decision-making method based on the interval-valued Pythagorean fuzzy (IVPF) set theory and the VlseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) methodology to address the multiple criteria selection problem of pilot hospitals in assisting the establishment of a hospital-based PAC model. Considering the powerfulness of IVPF sets when handling vagueness and complex uncertainty in practical problems, this paper proposes a useful IVPF VIKOR method that is significantly different from the existing VIKOR methodology. The proposed methodology represents a comprehensive integration of high-order uncertainties within a decision environment into the basic VIKOR structure that is designed to lead to both better description and better applicability of MCDA. This paper presents some novel concepts such as remoteness indices and remoteness-based multiple criteria ranking indices and investigates their desirable properties in detail. With the theoretical support of these new concepts, this paper establishes useful remoteness-based group utility indices, individual regret indices, and compromise indices for the assessment of acceptable advantage and acceptable stability. Unlike the current VIKOR-based ranking process, this paper provides a systematic ranking procedure that is capable of improving the efficiency of determining ultimate compromise solutions. Overall, the present study provides several significant contributions, such as structuring the selection problem for hospital-based PAC, extending the IVPF theory and VIKOR to the medical and healthcare fields, simplifying the manipulation procedure in handling IVPF information, constructing valuable concepts of remoteness-based indices, and developing an effective IVPF VIKOR ranking procedure. A real-world example of selecting pilot hospitals in the PAC program for CVDs is provided to illustrate the application of the proposed method and to demonstrate its practicality and effectiveness. Furthermore, other valuable MCDA applications with a comparative analysis are conducted to validate the advantages of the proposed method in a variety of fields.\",\n \"Title: A novel VIKOR method with an application to multiple criteria decision analysis for hospital-based post-acute care within a highly complex uncertain environment\\nYear: 2019\\n\\nAbstract: Post-acute care (PAC) is an interdisciplinary healthcare service used to assist in returning patients to their community after they receive acute medical services. To establish a fully developed PAC service model, the National Health Insurance Administration in Taiwan attempts to recruit professional medical service institutions (consisting of medical centers and regional and district hospitals) in a trial program for improving the rehabilitation quality of patients with cerebrovascular diseases (CVDs) and to reduce the re-hospitalization rate. The evaluation and selection of adequate medical service institutions are critical in the program for enhancing the effectiveness of hospital-based PAC in acute stroke management. However, because of the complexity of the national health insurance system and the healthcare system in Taiwan, the determination of pilot hospitals is a highly complicated and ambiguous multiple criteria decision analysis (MCDA) problem. Focusing on the requirement of generating a set of pilot hospitals for the PAC program and tackling imprecise and uncertain information associated with a complex medical circumstance, the purpose of this paper is to develop a novel compromising decision-making method based on the interval-valued Pythagorean fuzzy (IVPF) set theory and the VlseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) methodology to address the multiple criteria selection problem of pilot hospitals in assisting the establishment of a hospital-based PAC model. Considering the powerfulness of IVPF sets when handling vagueness and complex uncertainty in practical problems, this paper proposes a useful IVPF VIKOR method that is significantly different from the existing VIKOR methodology. The proposed methodology represents a comprehensive integration of high-order uncertainties within a decision environment into the basic VIKOR structure that is designed to lead to both better description and better applicability of MCDA. This paper presents some novel concepts such as remoteness indices and remoteness-based multiple criteria ranking indices and investigates their desirable properties in detail. With the theoretical support of these new concepts, this paper establishes useful remoteness-based group utility indices, individual regret indices, and compromise indices for the assessment of acceptable advantage and acceptable stability. Unlike the current VIKOR-based ranking process, this paper provides a systematic ranking procedure that is capable of improving the efficiency of determining ultimate compromise solutions. Overall, the present study provides several significant contributions, such as structuring the selection problem for hospital-based PAC, extending the IVPF theory and VIKOR to the medical and healthcare fields, simplifying the manipulation procedure in handling IVPF information, constructing valuable concepts of remoteness-based indices, and developing an effective IVPF VIKOR ranking procedure. A real-world example of selecting pilot hospitals in the PAC program for CVDs is provided to illustrate the application of the proposed method and to demonstrate its practicality and effectiveness. Furthermore, other valuable MCDA applications with a comparative analysis are conducted to validate the advantages of the proposed method in a variety of fields.\",\n \"Title: A Pythagorean Fuzzy Topsis Method Based On Novel Correlation Measures And Its Application To Multiple Criteria Decision Analysis Of Inpatient Stroke Rehabilitation\\nYear: 2019\\n\\nAbstract: The complex nature of the realistic decision-making process requires the use of Pythagorean fuzzy (PF) sets which have been shown to be a highly promising tool capable of solving highly vague and imprecise problems. Multiple criteria decision analysis (MCDA) methods within the PF environment are very attractive approaches for today's intricate decision environments. With this study, an effective compromise model named as the PF technique for order preference by similarity to ideal solutions (TOPSIS) is proposed based on some novel PF correlation-based concepts to overcome the complexities and ambiguities involved in real-life decision situations. In contrast to the existing distance-based definitions, this paper develops new closeness indices based on an extended concept of PF correlations. This paper employs the proposed PF correlation coefficients to construct two types of closeness measures. A comprehensive concept of PF correlation-based closeness indices can then be established to balance the consequences yielded by the two closeness measures. Based on these useful concepts, an effective PF TOPSIS method is proposed to address MCDA problems involving PF information and determine the ultimate priority orders among competing alternatives. Feasibility and practicability of the developed approach are illustrated by a medical decision-making problem of inpatient stroke rehabilitation. Finally, the proposed methodology is compared with other current methods to further explain its effectiveness. (c) 2019 The Authors. Published by Atlantis Press SARL.\",\n \"Title: Novel Generalized Distance Measure of Pythagorean Fuzzy Sets and a Compromise Approach for Multiple Criteria Decision Analysis Under Uncertainty.\\nYear: 2019\\n\\nAbstract: This paper aims at developing a novel generalized distance measure of Pythagorean fuzzy (PF) sets and constructing a distance-based compromise approach for multiple criteria decision analysis (MCDA) within PF environments. The theory of Pythagorean fuzziness provides a representative model of nonstandard fuzzy sets; it is valuable for representing complex vague or imprecise information in many practical applications. The distance measure for Pythagorean membership grades is important because it can effectively quantify the separation between PF information. Based on the essential characteristics of PF sets (membership, non-membership, strength, and direction), this paper proposes several distance measures, namely, new Hamming and Euclidean distances and a generalized distance measure that is based on them for Pythagorean membership grades and for PF sets. Moreover, the useful and desirable properties of the proposed PF distance measures are investigated to evaluate their advantages and form a solid theoretical basis. In addition, to evaluate the performance of the proposed distance measures in practice, this paper establishes a PF-distance-based compromise approach for addressing MCDA problems that involve PF information. The effectiveness and practicability of the developed approach are further evaluated through a case study on bridge-superstructure construction methods. According to the application results and comparative analysis, the proposed PF distance measures are accurate and outperform other methods in handling the inherent uncertainties of evaluation information. Furthermore, the PF-distance-based compromise approach can accommodate the much higher degrees of uncertainty in real-life decision scenarios and effectively determine the priority ranking among candidate alternatives for managing complicated MCDA problems.\",\n \"Title: TCBNN: Error-Correctable Ternary-Coded Binarized Neural Network\\nYear: 2021\\n\\nAbstract: Acceleration and reliability are two critical issues of artificial intelligent circuits and systems. In this paper we propose a novel structure of deep neural network in testing stage that converts the trained weights to optimized ternary-coded binary. Within each shallow layer a constant-shift sub-layer with almostfree cost is inserted to transfer all multipliers to a carry save adder/subtractor....\",\n \"Title: A T-Spherical Fuzzy Electre Approach For Multiple Criteria Assessment Problem From A Comparative Perspective Of Score Functions\\nYear: 2021\\n\\nAbstract: The theory involving T-spherical fuzziness provides an exceptionally good tool to efficiently manipulate the impreciseness, equivocation, and vagueness inherent in multiple criteria assessment and decision-making processes. By exploiting the notions of score functions and distance measures for complex T-spherical fuzzy information, this paper aims to propound an innovational T-spherical fuzzy ELECTRE (ELimination Et Choice Translating REality) approach to handling intricate and convoluted evaluation problems. Several newly-created score functions are employed from the comparative perspective to constitute a core procedure concerning concordance and discordance determination in the current T-spherical fuzzy ELECTRE method. By the agency of a realistic application, this paper appraises the usefulness and efficacy of available score functions in the advanced ELECTRE mechanism under T-spherical fuzzy uncertainties. This paper incorporates two forms of Minkowski distance measures into the core procedure; moreover, the effectuality of the advocated measure in differentiating T-spherical fuzzy information is validated. The effectiveness outcomes of the evolved method have been investigated through the medium of an investment decision regarding potential company options for extending the business scope. The real-world application also explores the comparative advantages of distinct score functions in tackling multiple criteria decision-making tasks. Finally, this paper puts forward a conclusion and future research directions.\",\n \"Title: Error Correctable Range-Addressable Lookup for Activation and Quantization in AI Automotive Electronics\\nYear: 2021\\n\\nAbstract: In this paper, we generalize the lightweight-slope piecewise-line range-addressable lookup table for efficient approximate computing in any activation/quantization function, and propose an error-correcting algorithm and circuitry using the AN codes for enhancing reliability. A BLER/SNR simulation proves the SEC/DEC capability of the firstly-presented AN-coded neuron. Comparisons with similar state...\",\n \"Title: Multiple criteria choice modeling using the grounds of T-spherical fuzzy REGIME analysis\\nYear: 2022\\n\\nAbstract: In an uncertain context of T-spherical fuzzy (T-SF) sets, the purpose of this study is to propound an efficacious multiple criteria choice model, named a T-SF REGIME method, predicated on a dominance analysis and a prioritization analysis. The REGIME methodology is a well-founded evaluation technique suitable for multiple criteria choice analysis owing to it has good applicability to the processing of qualitative information and to complex scenarios. Unlike the existing REGIME techniques, this paper takes advantage of a new Minkowski-type distance measure and an evolved Gaussian preference function to propound a ground-breaking T-SF REGIME model. This paper first concentrates on an evolution of the Minkowski-type distance measure and explores special cases with relevant properties for functional use. Next, this paper concerns an ascertainment of the evolved Gaussian preference function as a part of the proposed T-SF REGIME mechanism. This paper focuses on an advancement of the REGIME structure to T-SF uncertain contexts and unfolds an inventive T-SF REGIME methodology by virtue of superiority identifiers/indices, guide identifiers/indices, and joint generalized indices. The usefulness and strength points of the advanced methodology are scrutinized through a realistic application in the field of solar plant to prioritize the options of photovoltaic cells. The application results and comparisons corroborate the high efficiency and strengths of the evolved methodology.\",\n \"Title: Decision support modeling for multiple criteria assessments using a likelihood-based consensus ranking method under Pythagorean fuzzy uncertainty\\nYear: 2022\\n\\nAbstract: This paper intends to exploit point operator-oriented likelihood measures to constitute a likelihood-based consensus ranking model aimed at conducting multiple criteria decision making encompassing complex uncertain evaluations with Pythagorean fuzzy sets. This paper takes advantage of Pythagorean fuzzy point operators and the scalar functions of upper and lower estimations to formulate a point operator-oriented likelihood measure for preference intensity. On this basis, this paper propounds the notion of penalty weights to characterize dominated relations for acquiring the measurement of comprehensive disagreement and constituting a likelihood-based consensus ranking model. The primary contributions of this study are fourfold. Firstly, two useful point operators are initiated for upper and lower estimations towards Pythagorean membership grades. Secondly, an effective likelihood measure is exploited for determining outranking relations of Pythagorean fuzzy information. Thirdly, a pragmatic concept of penalty weights is proposed for characterizing the dominated relations among alternatives and measuring degrees of comprehensive disagreement. Fourthly, a functional likelihood-based consensus ranking model is constructed for implementing a multiple criteria evaluation with Pythagorean fuzzy uncertainty. Furthermore, a real-life application relating to a financing problem is presented to provide a justification for the practicability of the proposed methodology. This paper executes an analysis of parameters sensitivity and comparative studies for showing more theoretical insights.\",\n \"Title: A parametric likelihood measure with beta distributions for Pythagorean fuzzy decision-making\\nYear: 2022\\n\\nAbstract: The objective of this research is to introduce a parametric likelihood measure based on the beta distribution and develop a likelihood-oriented methodology for solving multiple criteria decision analysis (MCDA) problems with Pythagorean fuzzy (PF) sets. With the rapid advancement of PF theory, exploring an effective approach to compare PF information is indispensable in resolving MCDA issues. The beta distribution is one the most commonly used distributions to simulate the theoretical distribution. By changing the parameter values, the beta distribution can generate symmetrical or asymmetrical patterns and various shapes, including flat or steep. Due to its flexibility and adaptability, the beta distribution is able to effectively solve complex real-world problems. To make a major contribution to the technical development of decision support applications, this paper utilizes beta distributions as a parameterization tool to introduce a new parametric likelihood measure for evaluating the outranking relationships among PF information (signified by Pythagorean membership grades). Based on the evolved likelihood-based concepts (e.g., mean outranking indices, weighted outranking grades, and comprehensive outranking measures and indices), this paper proposes a pragmatic PF likelihood-oriented method to prioritize competing alternatives under uncertain and ambiguous Pythagorean fuzzy conditions. To carefully examine the practicality and suitability of the proposed methodology in realistic decision-making environments, this paper utilizes the evolved methods to solve a realistic MCDA problem of selecting pilot hospitals in relation to postacute care. The main results that are generated by the practical application and subsequent experimental analysis and comparative study demonstrate the effectivity and superiority of the developed technique and can be used for practical purposes in flexible and convenient ways. This most important conclusion of this paper is the great aptitude and dominance of the proposed methodology based on the corroboration of the experimental and comparative results of the application. Furthermore, this study has a noticeable originality in the utilization of the generic beta distribution-based approach and the construction of an effective PF likelihood-oriented decision model, which enriches the development of decision-making applications with PF theory.\",\n \"Title: Likelihood-based agreement measurements with Pythagorean fuzzy paired point operators to enrichment evaluations and priority determination for an uncertain decision-theoretical analysis\\nYear: 2022\\n\\nAbstract: This paper aims to initiate a useful Pythagorean fuzzy likelihood function on grounds of paired point operators and scalar-valued functions, and to contrive several likelihood-based agreement measurements for enrichment evaluations and priority determination within the framework of uncertain multiple criteria analysis. This paper exploits the characterization parameters of Pythagorean membership grades to expound paired point operators and further reveals relevant theoretical benefits. Supported by scalar-valued functions regarding the admissible lower and upper estimations, an innovative likelihood function is propounded for ascertaining the possibility of Pythagorean fuzzy dominance relations. Based on the advanced superiority and inferiority estimations, a number of valuable likelihood-based agreement measurements are unfolded to enrich the assessments, including (dis)agreement measures via rank-wise fittingness, expected (dis)agreement measures as a proxy for satisfaction and hygiene estimations, and overall (dis)agreement measures. An efficacious linear programming model with a hygiene threshold is constructed for prioritizing competing alternatives. A pragmatic decision-making issue related to hospital-based post-acute care is explored to inquire into application outcomes using the established techniques. Additionally, certain comparative analyses are performed to verify the helpfulness and interesting features possessed by the advanced approach.\",\n \"Title: TAIWAN Online: Test AI with AN Codes Online for Automotive Chips\\nYear: 2021\\n\\nAbstract: Neural networks play a key role in modern AI accelerators, in which acceleration and power reduction have been two known issues. When human lives may be threatened by accidents due to a wrong decision, the reliability bursts out far beyond other issues of the AI accelerators in automotive chips. Although input-side deep layers have been shown to possess considerable self-healing, arithmetic faults in shallow decision layers may still cause unimaginable catastrophe. This motivates us to develop a synthesis system called TAIWAN Online for testing AI with AN codes online. TAIWAN Online takes a trained model with a Keras-like format and the accuracy resolution for predicting a suitable subsystem. For low-resolution datasets, a ternary-coded-binarized neural network called TCBNN is proposed for approximate computing, where AN codes are adopted for arithmetic-weight error correction. While for high-resolution datasets, redundant residue number systems are applied for parallelized acceleration, and AN codes are utilized as AN-RRNS for self-checking in efficient multiple residue-modular redundancies, MMR. Briefly pointing out the key contributions, a k-moduli AN-RRNS can highly reduce the time-area product of MMR decoders from O(k\\n<sup xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">4</sup>\\n) of state-of-the-art RRNSs to only one. While the TCBNN can have fewer synapses than any regular-weight-quantized BNNs. From experimental results for a neuron-based block, the MTBF can be improved up to 126 times in the proposed infection-rate model.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":84,"string":"84"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13589,"string":"13,589"}}},{"rowIdx":13590,"cells":{"question_id":{"kind":"string","value":"Udo Hahn"},"question":{"kind":"string","value":"Generate an abstract for the title \"Cross-language MeSH indexing using morpho-semantic normalization.\" using the following items: controlled vocabulary, document retrieval, automated indexing, morphological analysis, MESH thesaurus"},"answer":{"kind":"string","value":"Abstract Using descriptors from a controlled vocabulary to sup- port document,retrieval in large electronically avail- able document,collections is a common,technique in state-of-the-art digital library interfaces. As a man- ual task it is time-consuming and requires some kind of level of expert knowledge. Automated,or semi- automatic approaches could be a promising alterna- tive. We here propose algorithmic solutions for au- tomated indexing, based upon morphological analysis in terms of medical subwords, with the aim to pro- pose MESH thesaurus identifiers as useful document descriptors of biomedical texts."},"correct_answer":{"kind":"string","value":"Abstract Using descriptors from a controlled vocabulary to sup- port document,retrieval in large electronically avail- able document,collections is a common,technique in state-of-the-art digital library interfaces. As a man- ual task it is time-consuming and requires some kind of level of expert knowledge. Automated,or semi- automatic approaches could be a promising alterna- tive. We here propose algorithmic solutions for au- tomated indexing, based upon morphological analysis in terms of medical subwords, with the aim to pro- pose MESH thesaurus identifiers as useful document descriptors of biomedical texts."},"documents":{"kind":"list like","value":["Title: Functional Centering\nYear: 1996\n\nAbstract: Based on empirical evidence from a free word order language (German) we propose a fundamental revision of the principles guiding the ordering of discourse entities in the forward-looking centers within the centering model. We claim that grammatical role criteria should be replaced by indicators of the functional information structure of the utterances, i.e., the distinction between context-bound and unbound discourse elements. This claim is backed up by an empirical evaluation of functional centering.","Title: Bridging Textual Ellipses\nYear: 1996\n\nAbstract: We present a hybrid text understanding methodology for the resolution of textual ellipsis. It integrates language-independent conceptual criteria and language-dependent functional constraints. The methodological framework for text ellipsis resolution is the centering model that has been adapted to constraints reflecting the functional information structure within utterances, i.e., the distinction between context-bound and unbound discourse elements.","Title: Comparatives in Context\nYear: 1997\n\nAbstract: We propose a model of semantic interpretation of comparatives which is based on a mech- anism for semantic copying. Besides common phrasal and clausal forms of comparatives, our model also incorporates the analysis of referential and textual phenomena that inter- act with the interpretation of comparatives, viz. metonymies and omitted complements. In order to allow for efficient processing, guidance from syntactic, semantic, contextual and world knowledge sources is supplied.","Title: Scalable Temporal Reasoning\nYear: 1999\n\nAbstract: We introduce two mechanisms for scaling com- putations in the framework of temporal reasoning. The first one addresses abstraction at the method- ological level. Operators are defined that engender flexible switching between different granularities of temporal representation structures. The second one accounts for abstractions at the interface level of a temporal reasoning engine. Various general- izations of temporal relations are introduced that approximate more fine-grained representations by abstracting away irrelevant details.","Title: Lean Semantic Interpretation\nYear: 1999\n\nAbstract: We introduce two abstraction mechanisms for streamlining the process of semantic interpreta- tion. Configurational descriptions of dependency graphs increase the linguistic generality of interpre- tation schemata, while interfacing them to lexical and conceptual inheritance hierarchies reduces the amount and complexity of semantic specifications.","Title: Iterative Refinement and Quality Checking of Annotation Guidelines - How to Deal Effectively with Semantically Sloppy Named Entity Types, such as Pathological Phenomena.\nYear: 2012\n\nAbstract: We here discuss a methodology for dealing with the annotation of semantically hard to delineate, i.e., sloppy, named entity types. To illustrate sloppiness of entities, we treat an example from the medical domain, namely pathological phenomena. Based on our experience with iterative guideline refinement we propose to carefully characterize the thematic scope of the annotation by positive and negative coding lists and allow for alternative, short vs. long mention span annotations. Short spans account for canonical entity mentions (e.g., standardized disease names), while long spans cover descriptive text snippets which contain entity-specific elaborations (e.g., anatomical locations, observational details, etc.). Using this stratified approach, evidence for increasing annotation performance is provided by kappa-based inter-annotator agreement measurements over several, iterative annotation rounds using continuously refined guidelines. The latter reflects the increasing understanding of the sloppy entity class both from the perspective of guideline writers and users (annotators). Given our data, we have gathered evidence that we can deal with sloppiness in a controlled manner and expect inter-annotator agreement values around 80% for PATHOJEN, the pathological phenomena corpus currently under development in our lab.","Title: Mereotopological reasoning about parts and (w)holes in bio-ontologies\nYear: 2001\n\nAbstract: We here deal with mereotopological properties of parts andassociated wholes, locations and empty spaces (holes), withparticular reference to biological structures. Our considerationslead to a basic ontology which contains 'solid object', 'hole' and'boundary' as mutually disjoint primitives. Formally, we embed therelations 'part-of' and 'location-of' into a parsimoniousdescription logic (ALC) and emulate partonomic and spatialreasoning involving these relations by terminological subsumption.In contrast to common conceptualizations, we do not distinguishbetween solids and the regions they occupy, as well as we allowsolids to have holes as proper parts. In order to support thesemodeling decisions, we discuss various concrete examples from humananatomy.","Title: Quality Control in the Concept Learning Process\nYear: 1998\n\nAbstract: A text understanding system with learning capabilities is presented. New conceptsare acquired by incorporating two kinds of evidence -- knowledge about linguisticconstructions in which unknown lexical items occur and knowledge about structuralpatterns in ontologies such that new concept descriptions can be compared withprior knowledge. On the basis of the quality of evidence gathered this way concepthypotheses are generated, ranked according to plausibility, and the most credibleones are...","Title: Semantic Interpretation of Medical Language - Quantitative Analysis and Qualitative Yield\nYear: 2001\n\nAbstract: We report on results from an empirical analysis of the semantic interpretation of medical free texts. Our approach to semantic interpretation is based on a lean collection of interpretation rules which are triggered by well-defined configurations in dependency graphs in order to compute a conceptual representation of the texts' contents. We evaluate the accuracy of semantic interpretation for three types of syntactic dependency patterns, viz. genitives, auxiliary and modal verb complexes, and prepositional phrases. Besides quantitative considerations, we focus on the heuristic guidance, as provided by patterns underlying the semantic interpretation of prepositional phrases, for monitoring the quality of the medical domain knowledge base.","Title: Content management in the SYNDIKATE system —: how technical documents are automatically transformed to text knowledge bases\nYear: 2000\n\nAbstract: Abstract S YN YN DI IKATE E is a family of natural language,understanding,systems for automatically,acquiring,knowledge,from real-world texts (e.g., information technology test reports, medical finding reports), and for transferring their content to formal representation,structures which,constitute a corresponding,text knowledge,base. We present a general system architecture which integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming,cohesive texts. Properly accounting,for text cohesion,phenomena,is a prerequisite for the soundness and validity of the generated,text representation,structures. It is also crucial for any information,system,application making use of automatically generated text knowledge bases in a reliable way, e.g., by inferentially supported fact retrieval. ” 2000 Published by Elsevier Science B.V. All rights reserved. Keywords: Natural language,processing; Text understanding; Knowledge,acquisition from,texts","Title: You can't beat frequency (unless you use linguistic knowledge): a qualitative evaluation of association measures for collocation and term extraction\nYear: 2006\n\nAbstract: In the past years, a number of lexical association measures have been studied to help extract new scientific terminology or general-language collocations. The implicit assumption of this research was that newly designed term measures involving more sophisticated statistical criteria would outperform simple counts of co-occurrence frequencies. We here explicitly test this assumption. By way of four qualitative criteria, we show that purely statistics-based measures reveal virtually no difference compared with frequency of occurrence counts, while linguistically more informed metrics do reveal such a marked difference.","Title: Quantitative data on referring expressions in biomedical abstracts\nYear: 2007\n\nAbstract: We report on an empirical study that deals with the quantity of different kinds of referring expressions in biomedical abstracts.","Title: MEDSYNDIKATE--a natural language system for the extraction of medical information from findings reports.\nYear: 2002\n\nAbstract: medSynDiKATe is a natural language processor, which automatically acquires medical information from findings reports. In the course of text analysis their contents is transferred to conceptual representation structures, which constitute a corresponding text knowledge base. medSynDiKATe is particularly adapted to deal properly with text structures, such as various forms of anaphoric reference relations spanning several sentences. The strong demands medSynDiKATe poses on the availability of expressive knowledge sources are accounted for by two alternative approaches to acquire medical domain knowledge (semi)automatically. We also present data for the information extraction performance of medSynDiKATe in terms of the semantic interpretation of three major syntactic patterns in medical documents.","Title: Syntactic Simplification and Semantic Enrichment - Trimming Dependency Graphs for Event Extraction.\nYear: 2011\n\nAbstract: In our approach to event extraction, dependency graphs constitute the fundamental data structure for knowledge capture. Two types of trimming operations pave the way to more effective relation extraction. First, we simplify the syntactic representation structures resulting from parsing by pruning informationally irrelevant lexical material from dependency graphs. Second, we enrich informationally relevant lexical material in the simplified dependency graphs with additional semantic meta data at several layers of conceptual granularity. These two aggregation operations on linguistic representation structures are intended to avoid overfitting of machine learning-based classifiers which we use for event extraction (besides manually curated dictionaries). Given this methodological framework, the corresponding JReX system developed by the JulieLab Team from Friedrich-Schiller-Universitat Jena (Germany) scored on 2nd rank among 24 competing teams for Task 1 in the BioNLP09 Shared Task on Event Extraction, with 45.8% recall, 47.5% precision and 46.7% F1-score on all 3,182 events. In more recent experiments, based on slight modifications of JReX and using the same data sets, we were able to achieve 45.9% recall, 57.7% precision, and 51.1% F1-score.","Title: CALBC silver standard corpus.\nYear: 2010\n\nAbstract: The CALBC initiative aims to provide a large-scale biomedical text corpus that contains semantic annotations for named entities of different kinds. The generation of this corpus requires that the annotations from different automatic annotation systems be harmonized. In the first phase, the annotation systems from five participants (EMBL-EBI, EMC Rotterdam, NLM, JULIE Lab Jena, and Linguamatics) were gathered. All annotations were delivered in a common annotation format that included concept identifiers in the boundary assignments and that enabled comparison and alignment of the results. During the harmonization phase, the results produced from those different systems were integrated in a single harmonized corpus (\"silver standard\" corpus) by applying a voting scheme. We give an overview of the processed data and the principles of harmonization--formal boundary reconciliation and semantic matching of named entities. Finally, all submissions of the participants were evaluated against that silver standard corpus. We found that species and disease annotations are better standardized amongst the partners than the annotations of genes and proteins. The raw corpus is now available for additional named entity annotations. Parts of it will be made available later on for a public challenge. We expect that we can improve corpus building activities both in terms of the numbers of named entity classes being covered, as well as the size of the corpus in terms of annotated documents.","Title: Teamwork Support in a Knowledge-Based Information Systems Environment\nYear: 1991\n\nAbstract: Development assistance for interactive database applications (DAIDA) is an experimental environment for the knowledge-assisted development and maintenance of database-intensive information systems from object-oriented requirements and specifications. Within the DAIDA framework, an approach to integrate different tasks encountered in software projects via a conceptual modeling strategy has been developed. Emphasis is put on integrating the semantics of the software development domain with aspects of group work, on social strategies to negotiate problems by argumentation, and on assigning responsibilities for task fulfillment by way of contracting. The implementation of a prototype is demonstrated with a sample session.","Title: Massive Biomedical Term Discovery\nYear: 2005\n\nAbstract: Most technical and scientific terms are comprised of complex, multi-word noun phrases but certainly not all noun phrases are\n technical or scientific terms. The distinction of specific terminology from common non-specific noun phrases can be based\n on the observation that terms reveal a much lesser degree of distributional variation than non-specific noun phrases. We formalize\n the limited paradigmatic modifiability of terms and, subsequently, test the corresponding algorithm on bigram, trigram and\n quadgram noun phrases extracted from a 104-million-word biomedical text corpus. Using an already existing and community-wide\n curated biomedical terminology as an evaluation gold standard, we show that our algorithm significantly outperforms standard\n term identification measures and, therefore, qualifies as a high-performant building block for any terminology identification\n system.","Title: Making Objects More Knowledgeable\nYear: 1997\n\nAbstract: Despite their common roots, the paradigms of object-oriented programmingand object-based knowledge representation techniques could not reallybe joined again. A synthesis of both worlds would allow for the descriptionof complex knowledge-based systems with dynamic behavior in a unifiedframework. We propose a corresponding model that extends object-oriented programminginto the classification-based paradigm of terminological reasoning andconsider a natural language understanding task...","Title: Towards text knowledge engineering\nYear: 1998\n\nAbstract: We introduce a methodology for automating the maintenance of domain-specific taxonomies based on natural language text understanding. A given ontology is incrementally updated as new concepts are acquired from real-world texts. The acquisition process is centered around the linguistic and conceptual \"quality\" of various forms of evidence underlying the generation and refinement of concept hypotheses. On the basis of the quality of evidence, concept hypotheses are ranked according to credibility and the most credible ones are selected for assimilation into the domain knowledge base.","Title: Subword-Based Text Retrieval\nYear: 2003\n\nAbstract: Document retrieval in languages with a rich and complex morphology 驴 particularly in terms of derivation and (single-word) composition 驴 suffers from serious performance degradation with the stemming-only query-term-to-text-word matching paradigm. We propose an alternative approach in which morphologically complex word forms are segmented into relevant subwords (such as stems, prefixes, suffixes), and subwords constitute the basic unit for indexing and retrieval. We evaluate our approach on a large biomedical document collection.","Title: On Text Coherence Parsing\nYear: 1992\n\nAbstract: In this paper global patterns of thematic text organization are considered within the framework of a distributed model of text understanding. Based on the parsing results of prior text cohesion analysis, specialized text grammar modules determine whether some well-defined text macro-organization pattern is computable from the available text representation structures. The model underlying text coherence parsing formalizes hither-to entirely intuitive textlinguistic notions whose origin can be traced back to Danes's work on thematic progression patterns.","Title: ParseTalk about Textual Ellipsis\nYear: 1995\n\nAbstract: A hybrid methodology for the resolution oftext-level ellipsis is presented in this paper.It incorporates conceptual proximity criteriaapplied to ontologically well-engineered domainknowledge bases and an approach tocentering based on functional topic/commentpatterns. We state text grammatical predicatesfor ellipsis and then turn to the proceduralaspects of their evaluation within theframework of an actor-based implementationof a lexically distributed parser.Keywords: text...","Title: PARSETALK about Functional Anaphora\nYear: 1996\n\nAbstract: . We present a hybrid methodology for the resolution of functionalanaphora. It incorporates conceptual proximity criteria applied toontologically well-engineered domain knowledge bases and an approachto centering based on functional topic/comment patterns. We state grammaticalpredicates for functional anaphora and then turn to the proceduralaspects of their evaluation within the framework of an actor-basedimplementation of a lexically distributed parser.1 IntroductionText phenomena,...","Title: Spatial location and its relevance for terminological inferences in bio-ontologies.\nYear: 2007\n\nAbstract: An adequate and expressive ontological representation of biological organisms and their parts requires formal reasoning mechanisms for their relations of physical aggregation and containment.We demonstrate that the proposed formalism allows to deal consistently with \"role propagation along non-taxonomic hierarchies\", a problem which had repeatedly been identified as an intricate reasoning problem in biomedical ontologies.The proposed approach seems to be suitable for the redesign of compositional hierarchies in (bio)medical terminology systems which are embedded into the framework of the OBO (Open Biological Ontologies) Relation Ontology and are using knowledge representation languages developed by the Semantic Web community.","Title: Multi-Task Active Learning for Linguistic Annotations\nYear: 2008\n\nAbstract: We extend the classical single-task active learning (AL) approach. In the multi-task ac- tive learning (MTAL) paradigm, we select ex- amples for several annotation tasks rather than for a single one as usually done in the con- text of AL. We introduce two MTAL meta- protocols, alternating selection and rank com- bination, and propose a method to implement them in practice. We experiment with a two- task annotation scenario that includes named entity and syntactic parse tree annotations on three different corpora. MTAL outperforms random selection and a stronger baseline, one- sided example selection, in which one task is pursued using AL and the selected examples are provided also to the other task.","Title: An integrated model of semantic and conceptual interpretation from dependency structures\nYear: 2000\n\nAbstract: We propose a two-layered model for computing semantic and conceptual interpretations from dependency structures. Abstract interpretation schemata generate semantic interpretations of 'minimal' dependency subgraphs, while production rules whose specification is rooted in ontological categories derive a canonical conceptual interpretation from semantic interpretation structures. Configurational descriptions of dependency graphs increase the linguistic generality of interpretation schemata, while interfacing schemata and productions to lexical and conceptual class hierarchies reduces the amount and complexity of semantic specifications.","Title: A Terminological Qualification Calculus for Preferential Reasoning under Uncertainty\nYear: 1996\n\nAbstract: We introduce a qualitative model of uncertain reasoning and illustrate its application in the framework of a natural language understanding task. Considering uncertain reasoning as a preferential choice problem between alternative hypotheses, the model we provide assigns quality labels to single evidences for or against a hypothesis, combines the generated labels in terms of the overall credibility of a single hypothesis, and, finally, computes a preference order on the entire set of competing hypotheses. This model of quality-based uncertain reasoning is entirely embedded in a terminological logic framework.","Title: Event extraction from trimmed dependency graphs\nYear: 2009\n\nAbstract: We describe the approach to event extraction which the JulieLab Team from FSU Jena (Germany) pursued to solve Task 1 in the \"BioNLP'09 Shared Task on Event Extraction\". We incorporate manually curated dictionaries and machine learning methodologies to sort out associated event triggers and arguments on trimmed dependency graph structures. Trimming combines pruning irrelevant lexical material from a dependency graph and decorating particularly relevant lexical material from that graph with more abstract conceptual class information. Given that methodological framework, the JulieLab Team scored on 2nd rank among 24 competing teams, with 45.8% precision, 47.5% recall and 46.7% F1-score on all 3,182 events.","Title: Concurrent Lexicalized Dependency Parsing: The ParseTalk Model.\nYear: 1994\n\nAbstract: A grammar model for concurrent, object-oriented natural language parsing is introduced. Complete lexical distribution of grammatical knowledge is achieved building upon the head-oriented notions of valency and dependency, while inheritance mechanisms are used to capture lexical generalizations. The underlying concurrent computation model relies upon the actor paradigm. We consider message passing protocols for establishing dependency relations and ambiguity handling.","Title: Paradigmatic modifiability statistics for the extraction of complex multi-word terms\nYear: 2005\n\nAbstract: We here propose a new method which sets apart domain-specific terminology from common non-specific noun phrases. It is based on the observation that terminological multi-word groups reveal a considerably lesser degree of distributional variation than non-specific noun phrases. We define a measure for the observable amount of paradigmatic modifiability of terms and, subsequently, test it on bigram, trigram and quadgram noun phrases extracted from a 104-million-word biomedical text corpus. Using a community-wide curated biomedical terminology system as an evaluation gold standard, we show that our algorithm significantly outperforms a variety of standard term identification measures. We also provide empirical evidence that our methodolgy is essentially domain- and corpus-size-independent.","Title: ParseTalk about Sentence- and Text-Level Anaphora\nYear: 1995\n\nAbstract: We provide a unified account of sentence-level and text-level anaphora within the framework of a dependency-based grammar model. Criteria for anaphora resolution within sentence boundaries rephrase major concepts from GB's binding theory, while those for text-level anaphora incorporate an adapted version of a Grosz-Sidner-style focus model.","Title: A Conceptual Reasoning Approach to Textual Ellipsis\nYear: 1996\n\nAbstract: We present a hybrid text understanding methodology for the resolution of textual ellipsis. It integrates conceptual criteria (based on the well-formedness and conceptual strength of role chains in a terminological knowledge base) and functional constraints reflecting the utterances' information structure (based on the distinction between context-bound and unbound discourse elements). The methodological framework for text ellipsis resolution is the centering model that has been adapted to these constraints.","Title: Towards an upper level ontology for molecular biology.\nYear: 2006\n\nAbstract: There is a growing need for the general-purpose description of the basic conceptual entities in the life sciences. Up until now, upper level models have mainly been purpose-driven, such as the GENIA ontology, originally devised as a vocabulary for corpus annotation. As an alternative,we here present BioTop, a description-logic-based top level ontology for molecular biology, which we consider as an ontologically conscious redesign of the GENIA ontology.","Title: A proposal for a configurable silver standard\nYear: 2010\n\nAbstract: Among the many proposals to promote alternatives to costly to create gold standards, just recently the idea of a fully automatically, and thus cheaply, to set up silver standard has been launched. However, the current construction policy for such a silver standard requires crucial parameters (such as similarity thresholds and agreement cut-offs) to be set a priori, based on extensive testing though, at corpus compile time. Accordingly, such a corpus is static, once it is released. We here propose an alternative policy where silver standards can be dynamically optimized and customized on demand (given a specific goal function) using a gold standard as an oracle.","Title: Automatic generation of hypertext knowledge bases\nYear: 1988\n\nAbstract: A model of knowledge-based text condensation is presented which has been implemented as part of the text analysis system TOPIC. The condensation process transforms the text representation structures resulting from the text parse into a more abstract thematic description of what the text is about, filtering out irrelevant knowledge structures and preserving only the most salient concepts. The topical structure of a text, finally, is represented in a hierarchical text graph which supports variable degrees of abstraction for text summarization as well as content-oriented retrieval of text knowledge. Due to their non-linear organization, text graphs share a lot of similarities with hypertexts. Their contribution to this field incorporates a methodology for the automatic generation of hypertexts from given full-text files, a close coupling of basic hypertext notions (links, nodes) to the formal specifications of a frame representation model, and conceptual navigation and filtering facilities which allow a user-defined level of information granularity when accessing hypertext knowledge bases.","Title: Efficient annotation with the Jena ANnotation Environment (JANE)\nYear: 2007\n\nAbstract: With ever-increasing demands on the diversity of annotations of language data, the need arises to reduce the amount of efforts involved in generating such value-added language resources. We introduce here the Jena ANnotation Environment (Jane), a platform that supports the complete annotation life-cycle and allows for 'focused' annotation based on active learning. The focus we provide yields significant savings in annotation efforts by presenting only informative items to the annotator. We report on our experience with this approach through simulated and real-world annotations in the domain of immunogenetics for NE annotations.","Title: Deep Knowledge Discovery from Natural Language Texts\nYear: 1997\n\nAbstract: We introduce a knowledge-based approach to deep knowledge discovery from real-world natural language texts. Data mining, data interpretation, and data cleaning are all incorporated in cy- cles of quality-based terminological reasoning processes.. The methodology we propose identifies new knowledge items and assimilates them into a continuously updated domain knowl- edge base.","Title: Parts, Locations, and Holes - Formal Reasoning about Anatomical Structures\nYear: 2001\n\nAbstract: We propose an ontology engineering framework for the anatomy domain, focusing on mereotopological properties of parts, locations and empty spaces (holes). We develop and formally describe a basic ontology consisting of the mutually disjoint primitives solid object, hole and boundary. We embed the relations part-of and location-of into a parsimonious description logic (ALC) and emulate advanced reasoning across these relations - such as transitivity at the T-Box level - by taxonomic subsumption. Unlike common conceptualizations we do not distinguish between solids and the regions they occupy, as well as we allow solids to have holes as proper parts. Concrete examples from human anatomy are used to support our claims.","Title: Intelligent Text Analysis for Dynamically Maintaining and Updating Domain Knowledge Bases\nYear: 1997\n\nAbstract: We propose a knowledge-intensive text analysis approach which deals with the continuous assimilation of new concepts into\n domain knowledge bases. Text understanding and knowledge acquisition proceed in tandem on the basis of terminological reasoning.\n Concept learning is considered an evidence-based choice problem the solution of which balances the “quality” of various clues\n from the linguistic structure of the texts and conceptual structures in the knowledge bases.","Title: A qualitative growth model for real world text knowledge bases\nYear: 1997\n\nAbstract: We introduce a knowledge-based approach to the analysis of real-world natural language texts, which addresses the particular needs of dealing with new knowledge items. In order to incrementally expand the underlying conceptual knowledge base, we exploit two kinds of evidence from the text understanding process, viz. qualitative knowledge about linguistic constructions in n atural language texts and that about structural patterns in the emerging text knowledge base. These clues are used to generate concept hypotheses, rank them according to plausibility, and selec t the most credible ones for assimilation into the conceptual knowledge base. We demonstrate the feasibility of our approach and discuss the results of an empirical evaluation in terms of concept learning rate s and learning accuracy.","Title: Building a Large Knowledge Base Semi-Automatically\nYear: 2001\n\nAbstract: We describe a knowledge engineering approach by which conceptual knowledge is extracted from an informal, se- mantically weak medical thesaurus (UMLS) and automat- ically converted into a formally sound description logics system. Our approach consists of four steps: concept definitions are automatically generated from the UMLS source, integrity checking of taxonomic and partonomic hierarchies is performed by the terminological classifier, cycles and inconsistencies are eliminated, and incremental refinement of the evolving knowledge base is performed by a domain expert. We report on experiments with a ter- minological knowledge base composed of 164,000 con- cepts and 76,000 relations.","Title: An empirical assessment of semantic interpretation\nYear: 2000\n\nAbstract: We introduce a framework for semantic interpretation in which dependency structures are mapped to conceptual representations based on a parsimonious set of interpretation schemata. Our focus is on the empirical evaluation of this approach to semantic interpretation, i.e., its quality in terms of recall and precision. Measurements are taken with respect to two real-world domains, viz. information technology test reports and medical finding reports.","Title: Evaluating the impact of alternative dependency graph encodings on solving event extraction tasks\nYear: 2010\n\nAbstract: In state-of-the-art approaches to information extraction (IE), dependency graphs constitute the fundamental data structure for syntactic structuring and subsequent knowledge elicitation from natural language documents. The top-performing systems in the BioNLP 2009 Shared Task on Event Extraction all shared the idea to use dependency structures generated by a variety of parsers --- either directly or in some converted manner --- and optionally modified their output to fit the special needs of IE. As there are systematic differences between various dependency representations being used in this competition, we scrutinize on different encoding styles for dependency information and their possible impact on solving several IE tasks. After assessing more or less established dependency representations such as the Stanford and CoNLL-X dependencies, we will then focus on trimming operations that pave the way to more effective IE. Our evaluation study covers data from a number of constituency- and dependency-based parsers and provides experimental evidence which dependency representations are particularly beneficial for the event extraction task. Based on empirical findings from our study we were able to achieve the performance of 57.2% F-score on the development data set of the BioNLP Shared Task 2009.","Title: Question Answering for Health Professionals and Laypersons.\nYear: 2013\n\nAbstract: and objective Knowledge sharing in the medical domain is limited by availability and language. The availability dimensions are growth of knowledge and communication systems design. The language dimensions are semantic, syntactic and in some cases phonetic. No human can comprehend the total of medical knowledge. Question answering (ICT-) systems would seem to be a good solution to create a “Wizard” or medical “Oracle of Delphi” available to anyone, anywhere, anytime. The “Wizard” how-ever must understand and speak about the questions on the conditions of the questioner. Language mediating the knowledge, no matter if spoken or written (by human or computer), must meet the user on his or hers conditions. Current systems are primarily using terminology resources created by the health professionals. The professional language, even when it is standardized, is a well known obstacle to sharing of knowledge to laypersons. This might be one of the biggest issues in patient empowerment. In this panel we aim at discussing strategies to overcome some of these barriers for medical knowledge sharing between health professionals and laypersons.","Title: Part-whole representation and reasoning in formal biomedical ontologies.\nYear: 2005\n\nAbstract: Biomedical ontologies are typically structured in a biaxial way, reflecting both a taxonomic (is-a) and a partonomic (part-of) hierarchy. Commonly used biomedical terminologies, which incorporate such distinctions excel in terms of broad coverage but lack a rigid formal foundation. The latter, however, is a prerequisite for automated reasoning. For the biomedical domain, it is not only crucial to cope with ontological dependencies between wholes and their parts but also with specific reasoning patterns which underlie the propagation of roles across partonomic hierarchies.We scale down part-whole reasoning to subsumption-based taxonomic reasoning within the formal framework of a parsimonious variant of description logics (viz. ALC).We provide a formal basis for ontological engineering in the domain of biomedicine, as far as part-whole relationships are concerned, by addressing typical reasoning patterns encountered in this domain.","Title: Towards the ontological foundations of symbolic biological theories.\nYear: 2007\n\nAbstract: Support for the symbolic representation of the physical structure of living organisms by an ontologically solid and logically sound foundation as a basis for formal reasoning.A set of canonical relations and attributes necessary for empirically adequate descriptions of biological entities is proposed.It is shown how a broad range of biological organisms and their parts can be represented by cascading theories which are ordered by the dimensions of granularity, development, species, and canonicity.The proposed representation of biological objects is non-redundant and compatible with inter- and intra-species similarities, developmental stages and pathological deviations.","Title: The GeneReg Corpus for Gene Expression Regulation Events - An Overview of the Corpus and its In-Domain and Out-of-Domain Interoperability\nYear: 2010\n\nAbstract: Despite the large variety of corpora in the biomedical domain their annotations differ in many respects, e.g., the coverage of different, highly specialized knowledge domains, varying degrees of granularity of the targeted relations, the specificity of linguistic grounding of relations and named entities referred to in the documents, etc. We here introduce GENEREG (Gene Regulation Corpus), the result of an annotation campaign led by the Jena University Language & Information Engineering (JULIE) Lab. The GENEREG corpus consists of 314 abstracts dealing with the regulation of gene expression in the model organism E. coli. Our emphasis in this paper is on the compatibility and, thus, linkage, of the GENEREG corpus with the alternative GENIA event corpus and with several in-domain and out-of-domain lexical resources, e.g., the SPECIALIST LEXICON, FRAMENET, and WORDNET. The links we established from the GENEREG corpus to these external resources will help improve the performance of the automatic relation extraction engine JREX trained and evaluated on GENEREG.","Title: Semi-supervised active learning for sequence labeling\nYear: 2009\n\nAbstract: While Active Learning (AL) has already been shown to markedly reduce the annotation efforts for many sequence labeling tasks compared to random selection, AL remains unconcerned about the internal structure of the selected sequences (typically, sentences). We propose a semi-supervised AL approach for sequence labeling where only highly uncertain subsequences are presented to human annotators, while all others in the selected sequences are automatically labeled. For the task of entity recognition, our experiments reveal that this approach reduces annotation efforts in terms of manually labeled tokens by up to 60% compared to the standard, fully supervised AL scheme.","Title: Towards a computational paradigm for biomedical structure\nYear: 2004\n\nAbstract: The symbolic representation of the physical struc- ture of living organisms needs an ontologically well-founded and logically sound approach so that formal reasoning can adequately be supported. We describe a set of canonical relations and attrib- utes necessary for the description of biological structures. Based on these epistemological prim- itives, we sketch how a broad range of organisms can be represented by cascading theories which are ordered by various dimensions, such as gran- ularity, development, species and canonicity. We thus aim at a rational reconstruction and non- redundant representation of biological structure notions.","Title: Joint knowledge capture for grammars and ontologies\nYear: 2001\n\nAbstract: We introduce a methodology for automating the maintenance and growth of domain-specific concept taxonomies and grammatical class hierarchies simultaneously, based on knowledge capture from natural language texts. The assimilation process is centered around the linguistic and conceptual `quality' of various forms of evidence underlying the generation, assessment and on-going refinement of lexical and concept hypotheses. On the basis of the strength of evidence, hypotheses are ranked according to plausibility, and the most reasonable ones are selected for assimilation into the given lexical class hierarchy and domain ontology.","Title: Creating knowledge repositories from biomedical reports: the MEDSYNDIKATE text mining system.\nYear: 2002\n\nAbstract: MEDSYNDIKATE is a natural language processor for automatically acquiring knowledge from medical finding reports. The content of these documents is transferred to formal representation structures which constitute a corresponding text knowledge base. The system architecture integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming cohesive texts. The strong demands MEDSYNDIKATE poses to the availability of expressive knowledge sources are accounted for by two alternative approaches to (semi)automatic ontology engineering. We also present data for the knowledge extraction performance of MEDSYNDIKATE for three major syntactic patterns in medical documents.","Title: Learning indexing patterns from one language for the benefit of others\nYear: 2004\n\nAbstract: Using language technology for text analysis and light-weight ontologies as a content-mediating level, we acquire indexing patterns from vast amounts of indexing data for English-language medical documents. This is achieved by statistically relating interlingual representations of these documents (based on text token bigrams) to their associated index terms. From these 'English' indexing patterns, we then induce the associated index terms for German and Portuguese documents when their interlingual representations match those of English documents. Thus, we learn from past English indexing experience and transfer it in an unsupervised way to non-English texts, without ever having seen concrete indexing data for languages other than English.","Title: A reappraisal of sentence and token splitting for life sciences documents.\nYear: 2007\n\nAbstract: Natural language processing of real-world documents requires several low-level tasks such as splitting a piece of text into its constituent sentences, and splitting each sentence into its constituent tokens to be performed by some preprocessor (prior to linguistic analysis). While this task is often considered as unsophisticated clerical work, in the life sciences domain it poses enormous problems due to complex naming conventions. In this paper, we first introduce an annotation framework for sentence and token splitting underlying a newly constructed sentence- and token-tagged biomedical text corpus. This corpus serves as a training environment and test bed for machine-learning based sentence and token splitters using Conditional Random Fields (CRFs). Our evaluation experiments reveal that CRFs with a rich feature set substantially increase sentence and token detection performance.","Title: An integrated, dual learner for grammars and ontologies\nYear: 2002\n\nAbstract: We introduce a dual-use methodology for automating the maintenance and growth of two types of knowledge sources, which are crucial for natural language text understanding-background knowledge of the underlying domain and linguistic knowledge about the lexicon and the grammar of the underlying natural language. A particularity of this approach is that learning occurs simultaneously with the on-going text understanding process. The knowledge assimilation process is centered around the linguistic and conceptual 'quality' of various forms of evidence underlying the generation, assessment and on-going refinement of lexical and concept hypotheses. On the basis of the strength of evidence, hypotheses are ranked according to qualitative plausibility criteria, and the most reasonable ones are selected for assimilation into the already given lexical class hierarchy and domain ontology.","Title: Collocation extraction based on modifiability statistics\nYear: 2004\n\nAbstract: We introduce a new, linguistically grounded measure of collocativity based on the property of limited modifiability and test it on German PP-verb combinations. We show that our measure not only significantly outperforms the standard lexical association measures typically employed for collocation extraction, but also yields a valuable by-product for the creation of collocation databases, viz. possible structural and lexical attributes. Our approach is language-, structure-, and domain-independent because it only requires some shallow syntactic analysis (e.g., a POS-tagger and a phrase chunker).","Title: Collaboratively Annotating Multilingual Parallel Corpora in the Biomedical Domain―some MANTRAs.\nYear: 2014\n\nAbstract: The coverage of multilingual biomedical resources is high for the English language, yet sparse for non-English languages-an observation which holds for seemingly well-resourced, yet still dramatically low-resourced ones such as Spanish, French or German but even more so for really under-resourced ones such as Dutch. We here present experimental results for automatically annotating parallel corpora and simultaneously acquiring new biomedical terminology for these under-resourced non-English languages on the basis of two types of language resources, namely parallel corpora (i.e. full translation equivalents at the document unit level) and (admittedly deficient) multilingual biomedical terminologies, with English as their anchor language. We automatically annotate these parallel corpora with biomedical named entities by an ensemble of named entity taggers and harmonize non-identical annotations the outcome of which is a so-called silver standard corpus. We conclude with an empirical assessment of this approach to automatically identify both known and new terms in multilingual corpora.","Title: A comparison of models for cost-sensitive active learning\nYear: 2010\n\nAbstract: Active Learning (AL) is a selective sampling strategy which has been shown to be particularly cost-efficient by drastically reducing the amount of training data to be manually annotated. For the annotation of natural language data, cost efficiency is usually measured in terms of the number of tokens to be considered. This measure, assuming uniform costs for all tokens involved, is, from a linguistic perspective at least, intrinsically inadequate and should be replaced by a more adequate cost indicator, viz. the time it takes to manually label selected annotation examples. We here propose three different approaches to incorporate costs into the AL selection mechanism and evaluate them on the Muc7T corpus, an extension of the Muc7 newspaper corpus that contains such annotation time information. Our experiments reveal that using a cost-sensitive version of semi-supervised AL, up to 54% of true annotation time can be saved compared to random selection.","Title: The SYNDIKATE text knowledge base generator\nYear: 2001\n\nAbstract: SYNDIKATE comprises a family of text understanding systems for automatically acquiring knowledge from real-world texts, viz. information technology test reports and medical finding reports. Their content is transformed to formal representation structures which constitute corresponding text knowledge bases. SYNDIKATE's architecture integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming cohesive texts. Besides centering-based discourse analysis mechanisms for pronominal, nominal and bridging anaphora, SYNDIKATE is supplied with a learning module for automatically bootstrapping its domain knowledge as text analysis proceeds.","Title: Unsupervised multilingual word sense disambiguation via an interlingua\nYear: 2005\n\nAbstract: We present an unsupervised method for resolving word sense ambiguities in one language by using statistical evidence assembled from other languages. It is crucial for this approach that texts are mapped into a language-independent interlingual representation. We also show that the coverage and accuracy resulting from multilingual sources outperform analyses where only monolingual training data is taken into account.","Title: Cross-lingual alignment of biomedical acronyms and their expansions.\nYear: 2006\n\nAbstract: We propose a method that aligns biomedical acronyms and their definitions across different languages. The approach is based upon a freely available tool for the extraction of abbreviations together with their expansions, and the subsequent normalization of language-specific variants, synonyms, and translations of the extracted acronym definitions. In this step, acronym expansions are mapped onto a language-independent concept-layer on which intra- as well as interlingual comparisons are drawn.","Title: How Knowledge Drives Understanding - Matching Medical Ontologies With The Needs Of Medical Language Processing\nYear: 1999\n\nAbstract: In this article, we introduce a knowledge-based approach to medical text understanding. From an in-depth consideration of deep sentence and text understanding we distill basic requirements for an adequate knowledge representation framework. These requirements are then matched with currently available medical ontologies (thesauri, terminologies, etc.). A fundamental trade-off is recognized between large-scale conceptual coverage on the one hand, and formal mechanisms for integrity preservation and conceptual expressiveness on the ether hand. We discuss various shortcomings of the most wide-spread ontologies to capture medical knowledge in-the-large. As a result, we argue for the need of a formally sound and expressive model along the lines of KL-ONE-style terminological representation systems in the format of description logics. These provide an adequate methodology for designing more sophisticated, flexible medical ontologies serving the needs of 'deep' knowledge applications which are by no means restricted to medical language processing. (C) 1999 Elsevier Science B.V. All rights reserved.","Title: Reducing class imbalance during active learning for named entity annotation\nYear: 2009\n\nAbstract: In lots of natural language processing tasks, the classes to be dealt with often occur heavily imbalanced in the underlying data set and classifiers trained on such skewed data tend to exhibit poor performance for low-frequency classes. We introduce and compare different approaches to reduce class imbalance by design within the context of active learning (AL). Our goal is to compile more balanced data sets up front during annotation time when AL is used as a strategy to acquire training material. We situate our approach in the context of named entity recognition. Our experiments reveal that we can indeed reduce class imbalance and increase the performance of classifiers on minority classes while preserving a good overall performance in terms of macro F-score.","Title: Automatic lexeme acquisition for a multilingual medical subword thesaurus\nYear: 2007\n\nAbstract: Conclusions Our approach constitutes a promising method for the automated creation of new lexicon entries and their linkage to semantic identifiers.","Title: Approximating Learning Curves for Active-Learning-Driven Annotation\nYear: 2008\n\nAbstract: Active learning (AL) is getting more and more popular as a methodology to considerably reduce the annotation effort when building training material for statistical learning methods for various NLP tasks. A crucial issue rarely addressed, however, is when to actually stop the annotation process to profit from the savings in efforts. This question is tightly related to estimating the classifier performance after a certain amount of data has already been annotated. While learning curves are the default means to monitor the progress of the annotation process in terms of classifier performance, this requires a labeled gold standard which - in realistic annotation settings, at least - is often unavailable. We here propose a method for committee-based AL to approximate the progression of the learning curve based on the disagreement among the committee members. This method relies on a separate, unlabeled corpus and is thus well suited for situations where a labeled gold standard is not available or would be too expensive to obtain. Considering named entity recognition as a test case we provide empirical evidence that this approach works well under simulation as well as under real-world annotation conditions.","Title: From GENIA to BIOTOPTowards a Top-Level Ontology for Biology\nYear: 2006\n\nAbstract: The increasing need for advanced ontology-based knowledge manage- ment in the life sciences is generally being acknowledged but, up until now, the development of biological ontologies lacks adherence to foundational principles of ontology design. This is particularly true of so-called upp er-level ontologies such as the GENIA ontology which covers biological continuants and has mainly been devised for corpus annotation in a text mining context. As an alternative, we intro- duce BIOTOP, an upper ontology of physical continuants in the domain of biology, with a coverage similar to the GENIA ontology. We report on design specifications and modeling decisions for BIOTOP which are based upon formal ontology princi- ples. As a major desideratum, these continuants are described in terms of necessary and sufficient conditions. We accomplished this goal for 85 o ut of the 146 existing GENIA classes. We use OWL-DL as a formal knowledge representation language and may thus use a terminological reasoner for classificatio n in order to check and maintain consistency during the ontology engineering phase.","Title: An Experimental Assessment of Direct Versus. Interlingual Translation for Cross-Language Information Retrieval\nYear: 2004\n\nAbstract: We introduce an interlingua-based approach to cross- language information retrieval, in which queries, as well as documents, are mapped onto a language-independent concept layer and retrieval operations are performed at the level of that interlingua. This approach is contrasted with one which oper- ates without such an intermediary concept level. Non-English queries (German ones, in our experiments) are directly trans- lated to English queries which, subsequently, are processed on English documents. We provide an empirical evaluation of both alternatives on a large medical document collection.","Title: Centering in-the-large: computing referential discourse segments\nYear: 1997\n\nAbstract: We specify an algorithm that builds up a hierarchy of referential discourse segments from local centering data. The spatial extension and nesting of these discourse segments constrain the reachability of potential antecedents of an anaphoric expression beyond the local level of adjacent center pairs. Thus, the centering model is scaled up to the level of the global referential structure of discourse. An empirical evaluation of the algorithm is supplied.","Title: Tagging Medical Documents with High Accuracy\nYear: 2004\n\nAbstract: We ran both Brill's rule-based tagger and TNT, a statistical tagger, with a default German newspaper-language model on a medical text corpus. Supplied with limited lexicon resources, TNT outperforms the Brill tagger with state-of-the-art performance figures (close to 97% accuracy). We then trained TNT on a large annotated medical text corpus, with a slightly extended tagset that captures certain medical language particularities, and achieved 98% tagging accuracy. Hence, statistical off-the-shelf POS taggers cannot only be immediately reused for medical NLP, but they also achieve - when trained on medical corpora - a higher performance level than for the newspaper genre.","Title: Incremental Centering And Center Ambiguity\nYear: 1996\n\nAbstract: In this paper, we present a model of anaphor resolution withinthe framework of the centering model. The consideration ofan incremental processing mode introduces the need to managestructural ambiguity at the center level. Hence, the centeringframework is further refined to account for local andglobal parsing ambiguities which propagate up to the level ofcenter representations, yielding moderately adapted data structuresfor the centering algorithm.IntroductionPsycholinguistic studies...","Title: An Empirical Evaluation of a System for Text Knowledge Acquisition\nYear: 1997\n\nAbstract: We introduce a formal model and a corresponding system architecture for the acquisition of new concepts from real-world natural language texts. Our approach is centered around the linguistic and conceptual quality of various forms of evidence underlying the generation and refinement of concept hypotheses. Based on a terminological (meta)reasoning platform, hypotheses are continuously annotated by a stream of linguistic and conceptual evidence, preferentially ranked and, finally, selected according to their overall credibility. We discuss the results of an empirical evaluation study, concentrating on the system's learning rate and learning accuracy.","Title: Semantic Annotations for Biology: a Corpus Development Initiative at the Jena University Language & Information Engineering (JULIE) Lab\nYear: 2008\n\nAbstract: We provide an overview of corpus building efforts at the Jena University Language & Information Engineering (JULIE) Lab, which are focused on life science documents. Special emphasis is laid on semantic annotations in terms of a large amount of biomedical named entities (almost 100 entity types), semantic relations, as well as discourse phenomena, reference relations in particular.","Title: Interlingual Indexing across Different Languages\nYear: 2004\n\nAbstract: We present two methods for automatic indexing, which are based on an interlingual layer of content description. In the first approach, we acquire indexing patterns from English documents by statistically relating interlingual representations of English documents (based on text token bigrams) to their associated index terms. Given such indexing patterns, we then induce the associated index terms when the same interlingual representations turn up for documents of other natural languages (viz. German and Portuguese). Hence, we 'learn' from the past English indexing experience and transfer it in an unsupervised way to non-English languages, without ever having seen any concrete indexing data for languages other than English. In the second approach, documents from the three different languages are heuristically matched with a sophisticated medical thesaurus (the English MESH) after both, documents and the thesaurus, have been transformed into the interlingua. The combination of the statistical and heuristical method in a fully automated indexing system achieves 56% to 68% of the human indexing performance for each of the three languages.","Title: Part-Whole Reasoning in Medical Knowledge Bases Using Description Logics\nYear: 1998\n\nAbstract: The development of powerful, ubiquitous and comprehensive medical ontologies that support formal reasoning on a large scale\n is one of the key requirements for clinical computing. Taxonomic medical knowledge, a major portion of these ontologies, is\n fundamentally characterized by is—a and part-whole relationships between concepts. While reasoning in generalization hierarchies\n is a well-understood process, no fully conclusive mechanism yet exists for part-whole reasoning. We here propose a new representation\n construct for part-whole relations based on the formal framework of description logics, i.e. the well-known concept language\n ALC, and show how part-whole reasoning can naturally be emulated via classification-based reasoning without extending the expressiveness\n of the underlying terminological system.","Title: High-performance tagging on medical texts\nYear: 2004\n\nAbstract: We ran both Brill's rule-based tagger and TNT, a statistical tagger, with a default German newspaper-language model on a medical text corpus. Supplied with limited lexicon resources, TNT outperforms the Brill tagger with state-of-the-art performance figures (close to 97% accuracy). We then trained TNT on a large annotated medical text corpus, with a slightly extended tagset that captures certain medical language particularities, and achieved 98% tagging accuracy. Hence, statistical off-the-shelf POS taggers cannot only be immediately reused for medical NLP, but they also -- when trained on medical corpora -- achieve a higher performance level than for the newspaper genre.","Title: A Search Engine for Morphologically Complex Languages\nYear: 2001\n\nAbstract: Document retrieval on natural languages with a rich morphology -- particularly in terms of derivation and (single-word) composition -- suffers from serious performance degradation with the direct query-term-to-text-word matching paradigm that underlies the vast majority of current search engines. We propose an alternative approach in which morphologically complex word forms, which appear in the query as well as in the documents, are segmented into relevant subwords (such as stems, named entities, acronyms) and are subsequently submitted to the matching procedure. We evaluate our approach with the Alta Vista驴 Search Engine on a large medical document collection.","Title: A text understander that learns\nYear: 1998\n\nAbstract: We introduce an approach to the automatic acquisition of new concepts from natural language texts which is tightly integrated with the underlying text understanding process. The learning model is centered around the 'quality' of different forms of linguistic and conceptual evidence which underlies the incremental generation and refinement of alternative concept hypotheses, each one capturing a different conceptual reading for an unknown lexical item.","Title: Biomedical text retrieval in languages with a complex morphology\nYear: 2002\n\nAbstract: Document retrieval in languages with a rich and complex morphology - particularly in terms of derivation and (single-word) composition - suffers from serious performance degradation with the stemming-only query-term-to-text-word matching paradigm. We propose an alternative approach in which morphologically complex word forms are segmented into relevant subwords (such as stems, named entities, acronyms), and subwords constitute the basic unit for indexing and retrieval. We evaluate our approach on a large biomedical document collection.","Title: Learning from texts - a terminological metareasoning perspective\nYear: 1995\n\nAbstract: We introduce a methodology for concept learning from texts that relies upon second-order reasoning about statements expressed in a (first-order) terminological representation language. This metareasoning approach allows for quality-based evaluation and selection of alternative concept hypotheses.","Title: Entity Recognition In Parallel Multi-Lingual Biomedical Corpora: The Clef-Er Laboratory Overview\nYear: 2013\n\nAbstract: The identification and normalisation of biomedical entities from the scientific literature has a long tradition and a number of challenges have contributed to the development of reliable solutions. Increasingly patient records are processed to align their content with other biomedical data resources, but this approach requires analysing documents in different languages across Europe [1,2].The CLEF-ER challenge has been organized by the Mantra project partners to improve entity recognition (ER) in multilingual documents. Several corpora in different languages, i.e. Medline titles, EMEA documents and patent claims, have been prepared to enable ER in parallel documents. The participants have been ask to annotate entity mentions with concept unique identifiers (CUIs) in the documents of their preferred non-English language.The evaluation determines the number of correctly identified entity mentions against a silver standard (Task A) and the performance measures for the identification of CUIs in the non-English corpora. The participants could make use of the prepared terminological resources for entity normalisation and of the English silver standard corpora (SSCs) as input for concept candidates in the non-English documents.The participants used different approaches including translation techniques and word or phrase alignments apart from lexical lookup and other text mining techniques. The performances for task A and B was lower for the patent corpus in comparison to Medline titles and EMEA documents. In the patent documents, chemical entities were identified at higher performance, whereas the other two document types cover a higher portion of medical terms. The number of novel terms provided from all corpora is currently under investigation.Altogether, the CLEF-ER challenge demonstrates the performances of annotation solutions in different languages against an SSC.","Title: CALBC: Releasing the Final Corpora.\nYear: 2012\n\nAbstract: A number of gold standard corpora for named entity recognition are available to the public. However, the existing gold standard corpora are limited in size and semantic entity types. These usually lead to implementation of trained solutions (1) for a limited number of semantic entity types and (2) lacking in generalization capability. In order to overcome these problems, the CALBC project has aimed to automatically generate large scale corpora annotated with multiple semantic entity types in a community-wide manner based on the consensus of different named entity solutions. The generated corpus is called the silver standard corpus since the corpus generation process does not involve any manual curation. In this publication, we announce the release of the final CALBC corpora which include the silver standard corpus in different versions and several gold standard corpora for the further usage of the biomedical text mining community. The gold standard corpora are utilised to benchmark the methods used in the silver standard corpora generation process and released in a shared format. All the corpora are released in a shared format and accessible at www.calbc.eu.","Title: Crossing Languages in Text Retrieval via an Interlingua\nYear: 2004\n\nAbstract: We introduce an interlingua-based approach to cross-language information retrieval, in which queries, as well as documents, are mapped onto a language-independent concept layer on which retrieval operations are performed. This approach is contrasted with one which directly translates non-English queries (German and Portuguese, in our experiments) to English ones which, subsequently, are processed on English documents. We report on the empirical evaluation of both approaches on a large medical document collection (the OHSUMED corpus).","Title: Learning Grammatical and Taxonomic Knowledge in Tandem\nYear: 2001\n\nAbstract: Abstract. We introduce a methodology for automating the maintenance of domain-specific concept taxonomies and grammatical class hierarchies simultaneously, based on incremental natural language text understanding. The acquisition pro-cess is centered around the linguistic and conceptual `quality'of various forms of evidence underlying the generation, assessment and on-going refinement of lex-ical and concept hypotheses. On the basis of the quality of evidence, hypotheses are ranked according to plausibility, and the most reasonable ones are selected for assimilation into the given lexical class hierarchy and domain taxonomy.","Title: Turning Lead into Gold? Feeding a Formal Knowledge Base with Informal Conceptual Knowledge\nYear: 2002\n\nAbstract: We describe an ontology engineering methodology by which conceptual knowledge is extracted from an informal medical thesaurus (UMLS) and automatically converted into a formal description logics system. Our approach consists of four steps: concept definitions are automatically generated from the UMLS source, integrity checking of taxonomic and partonomic hierarchies is performed by the terminological classifier, cycles and inconsistencies are eliminated, and incremental refinement of the evolving knowledge base is performed by a domain expert. We report on experiments with a knowledge base composed of 164,000 concepts and 76,000 relations.","Title: High-performance gene name normalization with GeNo.\nYear: 2009\n\nAbstract: The recognition and normalization of textual mentions of gene and protein names is both particularly important and challenging. Its importance lies in the fact that they constitute the crucial conceptual entities in biomedicine. Their recognition and normalization remains a challenging task because of widespread gene name ambiguities within species, across species, with common English words and with medical sublanguage terms.We present GeNo, a highly competitive system for gene name normalization, which obtains an F-measure performance of 86.4% (precision: 87.8%, recall: 85.0%) on the BioCreAtIvE-II test set, thus being on a par with the best system on that task. Our system tackles the complex gene normalization problem by employing a carefully crafted suite of symbolic and statistical methods, and by fully relying on publicly available software and data resources, including extensive background knowledge based on semantic profiling. A major goal of our work is to present GeNo's architecture in a lucid and perspicuous way to pave the way to full reproducibility of our results.GeNo, including its underlying resources, will be available from www.julielab.de. It is also currently deployed in the Semedico search engine at www.semedico.org.","Title: An annotation type system for a data-driven NLP pipeline\nYear: 2007\n\nAbstract: We introduce an annotation type system for a data-driven NLP core system. The specifications cover formal document structure and document meta information, as well as the linguistic levels of morphology, syntax and semantics. The type system is embedded in the framework of the Unstructured Information Management Architecture (UIMA).","Title: Large-scale evaluation of a medical cross-language information retrieval system.\nYear: 2007\n\nAbstract: We propose an approach to multilingual medical document retrieval in which complex word forms are segmented according to medically relevant morpho-semantic criteria. At its core lies a multilingual dictionary, it? which entries are equivalence classes of subwords, i.e. semantically minimal units. Using two different standard test collections for the medical domain, we evaluate our approach for six languages covered by our system. http://ebooks.iospress.nl/volume/medinfo-2007","Title: A Quality-Based Terminological Reasoning Model for Text Knowledge Acquisition\nYear: 1996\n\nAbstract: We introduce a methodology for knowledge acquisition and concept learning from texts that relies upon a quality-based model of terminological reasoning. Concept hypotheses which have been derived in the course of the text understanding process are assigned specific quality labels (indicating their significance, reliability, strength). Quality assessment of these hypotheses accounts for conceptual criteria referring to their given knowledge base context as well as linguistic indicators (grammatical constructions, discourse patterns), which led to their generation. We advocate a metareasoning approach which allows for the quality-based evaluation and a bootstrapping-style selection of alternative concept hypotheses as text understanding incrementally proceeds.","Title: Restricted parallelism in object-oriented lexical parsing\nYear: 1996\n\nAbstract: We present an approach to parallel natural language parsing which is based on a concurrent, object-oriented model of computation. A depth-first, yet incomplete parsing algorithm for a dependency grammar is specified and several restrictions on the degree of its parallelization are discussed.","Title: Disclose Models, Hide the Data - How to Make Use of Confidential Corpora without Seeing Sensitive Raw Data.\nYear: 2014\n\nAbstract: Confidential corpora from the medical, enterprise, security or intelligence domains often contain sensitive raw data which lead to severe restrictions as far as the public accessibility and distribution of such language resources are concerned. The enforcement of strict mechanisms of data protection consitutes a serious barrier for progress in language technology (products) in such domains, since these data are extremely rare or even unavailable for scientists and developers not directly involved in the creation and maintenance of such resources. In order to by-pass this problem, we here propose to distribute trained language models which were derived from such resources as a substitute for the original confidential raw data which remain hidden to the outside world. As an example, we exploit the access-protected German-language medical FRAMED corpus from which we generate and distribute models for sentence splitting, tokenization and POS tagging based on software taken from OPENNLP, NLTK and JCORE, our own UIMA-based text analytics pipeline.","Title: Towards Very Large Terminological Knowledge Bases: A Case Study from Medicine\nYear: 2000\n\nAbstract: We describe an ontology engineering methodology by which conceptual knowledge is extracted from an informal medical thesaurus (UMLS) and automatically converted into a formally sound description logics system. Our approach consists of four steps: concept definitions are automatically generated from the UMLS source, integrity checking of taxonomic and partonomic hierarchies is performed by the terminological classifier, cycles and inconsistencies are eliminated, and incremental refinement of the evolving knowledge base is performed by a domain expert. We report on knowledge engineering experiments with a terminological knowledge base composed of 164,000 concepts and 76,000 relations.","Title: Tracking the Evolution of Concepts in Dynamic Worlds\nYear: 1994\n\nAbstract: Technical domains are affected by continuous change as a reflection of technological progress. Correspondingly, concept descriptions of terminological knowledge bases for such domains must be adjusted to these dynamics. In this paper we propose a concept versioning methodology which accounts for the evolutionary adaptation of already established concepts according to the gradual shaping of new standards. The approach is based on the provision of a qualitative progress model for the given domain, measures for the prediction and evaluation of the significance of progress, and a representation and update scheme for concept version management.","Title: Mereological semantics for bio-ontologies\nYear: 2004\n\nAbstract: Biomedical ontologies are typically structured in a biaxial way, reflecting both a taxonomic and a mereological order. Common examples such as the Gene Ontology and the Unified Medical Language System (UMLS) excel in terms of coverage but lack an adequate semantics of the mereological relations they incorporate. This shortcoming is particularly evident as far as the (non-)mandatory existence of parts for their wholes is cOllcerned, on the one hand, and the propagation of properties acmss part-whole hierarchies, on the other hand. We provide a formal specification of the semantic foundations of mereology in the biomedical domain that is closely linked to the paradigm of description logics. In essence, we here propose to emulate mereological reasoning by taxonomic reasoning. In an attempt to capture much of the shared intuition underlying merelogical reasoning in the biomedical domain, we distinguish for each mereologically relevant concept four different classes of parts and wholes which allow for the expression of five different propagation patterns.","Title: Timed annotations: enhancing Muc7 metadata by the time it takes to annotate named entities\nYear: 2009\n\nAbstract: We report on the re-annotation of selected types of named entities from the Muc7 corpus where our focus lies on recording the time it takes to annotate these entities given two basic annotation units -- sentences vs. complex noun phrases. Such information may be helpful to lay the empirical foundations for the development of cost measures for annotation processes based on the investment in time for decision-making per entity mention.","Title: The CALBC Silver Standard Corpus for Biomedical Named Entities - A Study in Harmonizing the Contributions from Four Independent Named Entity Taggers\nYear: 2010\n\nAbstract: The production of gold standard corpora is time-consuming and costly. We propose an alternative: the, silver standard corpus' (SSC), a corpus that has been generated by the harmonisation of the annotations that have been delivered from a selection of annotation systems. The systems have to share the type system for the annotations and the harmonisation solution has use a suitable similarity measure for the pair-wise comparison of the annotations. The annotation systems have been evaluated against the harmonised set (630.324 sentences, 15,956,841 tokens). We can demonstrate that the annotation of proteins and genes shows higher diversity across all used annotation solutions leading to a lower agreement against the harmonised set in comparison to the annotations of diseases and species. An analysis of the most frequent annotations from all systems shows that a high agreement amongst systems leads to the selection of terms that are suitable to be kept in the harmonised set. This is the first large-scale approach to generate an annotated corpus from automated annotation systems. Further research is required to understand, how the annotations from different systems have to be combined to produce the best annotation result for a harmonised corpus.","Title: On the Formal Distinction between Literal and Figurative Language\nYear: 1999\n\nAbstract: The distinction between literal and figurative language (metonymies, metaphors, etc.) is often not made formally explicit, or, if formal criteria exist, insufficient. This poses problems for an adequate computational treatment of these phenomena. The basic criterion for delineating literal from figurative speech we propose is centered around the notion of categorization conflicts that follow from the context of the utterance. In addition, we consider the problem of granularity, which is posed by the dependence of our approach on the underlying ontology.","Title: Concept Versioning: A Methodology for Tracking Evolutionary Concept Drift in Dynamic Concept Systems\nYear: 1994\n\nAbstract: . Technical domains are affected by continuous change as a reflectionof technological progress. Correspondingly, concept descriptions of terminologicalknowledge bases for such domains must be adjusted to thesedynamics. In this paper we propose a concept versioning methodology whichaccounts for the evolutionary adaptation of already established concepts accordingto the gradual shaping of new standards. The approach is based on theprovision of qualitative progress models for the given...","Title: Bootstrapping dictionaries for cross-language information retrieval\nYear: 2005\n\nAbstract: The bottleneck for dictionary-based cross-language information retrieval is the lack of comprehensive dictionaries, in particular for many different languages. We here introduce a methodology by which multilingual dictionaries (for Spanish and Swedish) emerge automatically from simple seed lexicons. These seed lexicons are automatically generated, by cognate mapping, from (previously manually constructed) Portuguese and German as well as English sources. Lexical and semantic hypotheses are then validated and new ones iteratively generated by making use of co-occurrence patterns of hypothesized translation synonyms in parallel corpora. We evaluate these newly derived dictionaries on a large medical document collection within a cross-language retrieval setting.","Title: Text Knowledge Engineering by Qualitative Terminological Learning\nYear: 1997\n\nAbstract: We propose a methodology for enhancing domain knowledge bases through natural language text understanding. The acquisition of previously unknown concepts is based on the assessment of the quality of linguistic and conceptual evidence underlying the generation and refinement of concept hypotheses. Text understanding and concept learning are both grounded on a terminological knowledge representation and reasoning framework.","Title: On \"deep\" knowledge extraction from documents\nYear: 2000\n\nAbstract: SYNDIKATE comprises a family of natural language understanding systems for automatically acquiring knowl- edge from real-world texts (e.g., information technology test reports, medical finding reports), and for transferring their content to formal representation structures which constitute a corresponding text knowledge base. We present a general system architecture which integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming cohesive texts. Properly accounting for text cohesion phenomena is a prerequisite for the soundness and validity of the generated text representation structures. It is also crucial for any information system application making use of automatically generated text knowledge bases in a reliable way.","Title: Context-Based Ambiguity Management for Natural Language Processing\nYear: 2001\n\nAbstract: We introduce a formal context mechanism, embedded into a description logics framework, which allows to uniformly represent and manage different kinds of natural language ambiguities as they occur in the course of text understanding. Multiple lexical, syntactic and semantic interpretations are separated by assigning each of these alternatives a single context space for local reasoning and incremental disambiguation.","Title: Assessment of NER solutions against the first and second CALBC Silver Standard Corpus.\nYear: 2011\n\nAbstract: Competitions in text mining have been used to measure the performance of automatic text processing solutions against a manually annotated gold standard corpus (GSC). The preparation of the GSC is time-consuming and costly and the final corpus consists at the most of a few thousand documents annotated with a limited set of semantic groups. To overcome these shortcomings, the CALBC project partners (PPs) have produced a large-scale annotated biomedical corpus with four different semantic groups through the harmonisation of annotations from automatic text mining solutions, the first version of the Silver Standard Corpus (SSC-I). The four semantic groups are chemical entities and drugs (CHED), genes and proteins (PRGE), diseases and disorders (DISO) and species (SPE). This corpus has been used for the First CALBC Challenge asking the participants to annotate the corpus with their text processing solutions.All four PPs from the CALBC project and in addition, 12 challenge participants (CPs) contributed annotated data sets for an evaluation against the SSC-I. CPs could ignore the training data and deliver the annotations from their genuine annotation system, or could train a machine-learning approach on the provided pre-annotated data. In general, the performances of the annotation solutions were lower for entities from the categories CHED and PRGE in comparison to the identification of entities categorized as DISO and SPE. The best performance over all semantic groups were achieved from two annotation solutions that have been trained on the SSC-I.The data sets from participants were used to generate the harmonised Silver Standard Corpus II (SSC-II), if the participant did not make use of the annotated data set from the SSC-I for training purposes. The performances of the participants' solutions were again measured against the SSC-II. The performances of the annotation solutions showed again better results for DISO and SPE in comparison to CHED and PRGE.The SSC-I delivers a large set of annotations (1,121,705) for a large number of documents (100,000 Medline abstracts). The annotations cover four different semantic groups and are sufficiently homogeneous to be reproduced with a trained classifier leading to an average F-measure of 85%. Benchmarking the annotation solutions against the SSC-II leads to better performance for the CPs' annotation solutions in comparison to the SSC-I.","Title: Finding new terminology in very large corpora\nYear: 2005\n\nAbstract: Most technical and scientific terms are comprised of complex, multi-word noun phrases but certainly not all noun phrases are technical or scientific terms. The distinction of specific terminology from common non-specific noun phrases can be based on the observation that terms reveal a much lesser degree of distributional variation than non-specific noun phrases. We formalize the limited paradigmatic modifiability of terms and, subsequently, test the corresponding algorithm on bigram, trigram and quadgram noun phrases extracted from a 104-million-word biomedical text corpus. Using an already existing and community-wide curated biomedical terminology as an evaluation gold standard, we show that our algorithm significantly outperforms standard term identification measures and, therefore, qualifies as a high-performant building block for any terminology identification system. We also provide empirical evidence that the superiority of our approach, beyond a 10-million-word threshold, is essentially domain- and corpus-size-independent.","Title: BioTop: An upper domain ontology for the life sciences: A description of its current structure, contents and interfaces to OBO ontologies\nYear: 2008\n\nAbstract: In the life sciences, there is an ample need for semantic interoperability of d ata. Thus shared vocabularies are needed for consistently expressing meta data in terms of semantic annotations as well as for querying bibliographic information systems. In the past years, lots of highly specialized, yet also fragmented terminolog ies have evolved. However, they lack principled forms of conceptual interlinkage. In order to provide an ontological basis for a seamless integration of such isolated parts of biological knowledge, we here introduce BIOTOP, an upper domain ontology for molecular biology. We describe its structure and contents, as well as its current interfaces to a selected set of OBO ontologies, which contain more detailed terminological knowledge about specific areas of molecular biology, e.g., cell types, molecular functio ns, biological processes, and chemical compounds.","Title: Text Understanding for Knowledge Base Generation in the SYNDIKATE System\nYear: 1999\n\nAbstract: SynDiKATe is a system for automatically acquiring knowledge from real-world texts and transferring it to formal representation structures which constitute a text knowledge base. We present a system architecture which integrates requirements from the analysis of single sentences. as well as those of referentially linked sentences forming cohesive texts. Properly accounting for text cohesion phenomena is a prerequisite for the completeness and validity of the generated text representation structures and. therefore. also crucial for any information system application making use of automatically generated text knowledge bases in a reliable way, e.g., by inferentially supported fact retrieval.","Title: U-Compare bio-event meta-service: compatible BioNLP event extraction services.\nYear: 2011\n\nAbstract: Background Bio-molecular event extraction from literature is recognized as an important task of bio text mining and, as such, many relevant systems have been developed and made available during the last decade. While such systems provide useful services individually, there is a need for a meta-service to enable comparison and ensemble of such services, offering optimal solutions for various purposes. Results We have integrated nine event extraction systems in the U-Compare framework, making them inter-compatible and interoperable with other U-Compare components. The U-Compare event meta-service provides various meta-level features for comparison and ensemble of multiple event extraction systems. Experimental results show that the performance improvements achieved by the ensemble are significant. ConclusionsWhile individual event extraction systems themselves provide useful features for bio text mining, the U-Compare meta-service is expected to improve the accessibility to the individual systems, and to enable meta-level uses over multiple event extraction systems such as comparison and ensemble. © 2011 Kano et al; licensee BioMed Central Ltd.","Title: \"Tall\", \"good\", \"high\": compared to what?\nYear: 1997\n\nAbstract: In order to rank the performance of machine learning algorithms, many researchers conduct experiments on benchmark data sets. Since most learning algorithms have domain-specific parameters, it is a popular custom to adapt these parameters to obtain a ...","Title: Subword Retrieval on Biomedical Documents\nYear: 2003\n\nAbstract: Document retrieval in languages with a rich and com- plex morphology - particularly in terms of derivation and (single-word) composition - suffers from serious perfor- mance degradation with the stemming-only query-term-to- text-word matching paradigm. We propose an alternative ap- proach in which morphologically complex word forms are segmented into relevant subwords (such as stems, prefixes, suffixes), and subwords constitute the basic unit for index- ing and retrieval. We evaluate our approach on a biomedical document collection.","Title: The Challenges of Automatic Summarization\nYear: 2000\n\nAbstract: Summarization--the art of abstracting key con-tent from one or more information sources--has become an integral part of everyday life. People keep abreast of world affairs by listening to news bites. They base investment decisions on stock market updates. They go to movies largely on the basis of reviews. With summaries, they can make effective decisions in less time.Although summarizing tools are available, with the increasing volume of online information, it is becoming harder to generate meaningful and timely summaries. Researchers are investigating tools and methods that automatically extract or abstract content from information sources.The authors describe how these data summarization methods fall into two categories. Knowledge-poor approaches rely on not having to add new rules for each new application domain or language. Knowledge-rich approaches assume that if you grasp the meaning of the text, you can reduce it more effectively, thus yielding a better summary. They rely on a size-able knowledge base of rules, which must be acquired, maintained, and then adapted to new applications and languages.The authors predict that summarization tools will be key in conquering the vast information universes ahead.","Title: Terminological Meta-Reasoning by Reification and Multiple Contexts\nYear: 1995\n\nAbstract: We introduce a model and a system architecture for second-order reasoning about statements expressed in a (first-order) terminological representation language. This meta-reasoning approach is based on the reification of first-order propositions and the mediation between second- and first-order expressions via translation rules operating on multiple contexts. An application to a concept learning task in a text understanding environment shows how different degrees of credibility can be assigned to alternative concept hypotheses on the basis of such a schema for terminological meta-reasoning.","Title: Gene Regulation Ontology (GRO): Design Principles and Use Cases.\nYear: 2008\n\nAbstract: The Gene Regulation Ontology (GRO) is designed as a novel approach to model complex events that are part of the gene regulatory processes. We introduce the design requirements for such a conceptual model and discuss terminological resources suitable to base its construction on. The ontology defines gene regulation events in terms of ontological classes and imposes constraints on them by specifying the participants involved. The logical structure of the ontology is intended to meet the needs of advanced information extraction and text mining systems which target the identification of event representations in scientific literature. The GRO has just been submitted to the OBO library and is currently under review. It is available at http:// www.ebi.ac.uk/Rebholz-srv/GRO/GRO.html","Title: Generating Semantics for the Life Sciences via Text Analytics\nYear: 2011\n\nAbstract: The life sciences have a strong need for carefully curated, semantically rich fact repositories. Knowledge harvesting from unstructured textual sources is currently performed by highly skilled curators who manually feed semantics into such databases as a result of deep understanding of the documents chosen to populate such repositories. As this is a slow and costly process, we here advocate an automatic approach to the generation of database contents which is based on JREX, a high performance relation extraction system. As a real-life example, we target REGULONDB, the world's largest manually curated reference database for the transcriptional regulation network of E. coli. We investigate in our study the performance of automatic knowledge capture from various literature sources, such as PUBMED abstracts and associated full text articles. Our results show that we can, indeed, automatically re-create a considerable portion of the REGULONDB database by processing the relevant literature sources. Hence, this approach might help curators widen the knowledge acquisition bottleneck in this field.","Title: Parthood as Spatial Inclusion - Evidence from biomedical Conceptualizations\nYear: 2004\n\nAbstract: Modeling complex compositional objects in the field of the life sciences requires to solve intricate ontological problems, especially those related to parts of a whole, space and location. For concrete physical entities, we here stipulate that the distinction between parthood and location can be abstracted away. We outline a logic- based reasoning framework in which taxonomic sub- sumption not only takes care of the propagation of roles across taxonomic hierarchies, but also across nontaxo- nomic, i.e., partonomic or spatial, ones.","Title: Functional centering: grounding referential coherence in information structure\nYear: 1999\n\nAbstract: Considering empirical evidence from a free-word-order language (German) we propose a revision of the principles guiding the ordering of discourse entities in the forward-looking center list within the centering model. We claim that grammatical role criteria should be replaced by criteria that reflect the functional information structure of the utterances. These new criteria are based on the distinction between hearer-old and hearer-new discourse entities. We demonstrate that such a functional model of centering can be successfully applied to the analysis of several forms of referential text phenomena, viz. pronominal, nominal, and functional anaphora. Our methodological and empirical claims are substantiated by two evaluation studies. In the first one, we compare success rates for the resolution of pronominal anaphora that result from a grammatical-role-driven centering algorithm and from a functional centering algorithm. The second study deals with a new cost-based evaluation methodology for the assessment of centering data, one which can be directly derived from and justified by the cognitive load premises of the centering model.","Title: Towards automatic pathway generation from biological full-text publications\nYear: 2011\n\nAbstract: We introduce an approach to the automatic generation of biological pathway diagrams from scientific literature. It is composed of the automatic extraction of single interaction relations which are typically found in the full text (rather than the abstract) of a scientific publication, and their subsequent integration into a complex pathway diagram. Our focus is here on relation extraction from full-text documents. We compare the performance of automatic full-text extraction procedures with a manually generated gold standard in order to validate the extracted data which serve as input for the pathway integration procedure.","Title: Quality-Based Terminological Reasoning for Concept Learning\nYear: 1995\n\nAbstract: We introduce a qualitative methodology for concept learning from texts that relies upon second-order reasoning about statements expressed in a (first-order) terminological representation language. This meta-reasoning approach allows for quality-based evaluation and selection among alternative concept hypotheses.","Title: Making understanders out of parsers: Semantically driven parsing as a key concept for realistic text understanding applications.\nYear: 1989\n\nAbstract: Semantically driven natural language parsers have found wide-spread application as a text processing methodology for knowledge-based information retrieval systems. It is argued that this parsing technique particularly corresponds to the requirements inherent to large-scale text analysis. Unfortunately, this approach suffers from several shortcomings which demand a thorough reformulation of its paradigm. Incorporating principles from conceptual analysis and word expert parsing in a model of lexically distributed text parsing, the focus of the modifications proposed in this article, is on a clean declarative separation of linguistio and other knowledge representation levels, abstraction mechanisms leading to a small collection of specification primitives for the parser, and an attempt to incorporate linguistic generalizations and modularization principles into the design of a semantic text grammar. A sample parse illustrates the operation and linguistic coverage of a lexically distributed text parser based on these theoretical considerations with respect to the semantic analysis of noun groups, simple assertional sentences, nominal anaphora, and textual ellipsis. © 1989 Wiley Periodicals, Inc.","Title: Trading off Completeness for Efficiency - The ParseTalk Performance Grammar Approach to Real-World Text Parsing\nYear: 1996\n\nAbstract: We argue for a performance-based design of natural language grammars and\ntheir associated parsers in order to meet the constraints posed by real-world\nnatural language understanding. This approach incorporates declarative and\nprocedural knowledge about language and language use within an object-oriented\nspecification framework. We discuss several message passing protocols for\nreal-world text parsing and provide reasons for sacrificing completeness of the\nparse in favor of efficiency.","Title: A formal approach to the semantics of a frame data model\nYear: 1983\n\nAbstract: Standard knowledge representation languages are seriously lacking an explicit formal semantic specification. This may cause considerable trouble when applied to large amounts of rapidly changing data. Based on an abstract data type view of knowledge representation languages a formal definition of a frame data model is presented in terms of a denotational semantics approach using a subset of META-IV. After introducing some basic concepts of the model several semantic integrity constraints are outlined which ultimately lead to the formulation of a set of operations in the frame data model.","Title: Die Verdichtung textuellen Wissens zu Information - Vom Wandel methodischer Paradigmen beim automatischen Abstracting\nYear: 2004\n\nAbstract: The automatic summarization of documents (technical magazine articles) has been a central research theme in the information science division at Constance University. We give an overview of the context in which this topic was dealt with, the major methodological results that were achieved, and the text condensation system TOPIC that was built on these principles. We then survey current trends in the field of automatic text summarization which reveal the renaissance of the 'abstracting-as-extracting' approach. We relate this observation to the fundamental paradigmatic shift from natural language understanding to human language technology and discuss its implications for information science.","Title: Textual expertise in word experts: an approach to text parsing based on topic/comment monitoring\nYear: 1984\n\nAbstract: In this paper prototype versions of two word experts for text analysis are dealt with which demonstrate that word experts are a feasible tool for parsing texts on the level of text cohesion as well as text coherence. The analysis is based on two major knowledge sources: context information is modelled in terms of a frame knowledge base, while the co-text keeps record of the linear sequencing of text analysis. The result of text parsing consists of a text graph reflecting the thematic organization of topics in a text.","Title: How feasible and robust is the automatic extraction of gene regulation events?: a cross-method evaluation under lab and real-life conditions\nYear: 2009\n\nAbstract: We explore a rule system and a machine learning (ML) approach to automatically harvest information on gene regulation events (GREs) from biological documents in two different evaluation scenarios -- one uses self-supplied corpora in a clean lab setting, while the other incorporates a standard reference database of curated GREs from RegulonDB, real-life data generated independently from our work. In the lab condition, we test how feasible the automatic extraction of GREs really is and achieve F-scores, under different, not directly comparable test conditions though, for the rule and the ML systems which amount to 34% and 44%, respectively. In the RegulonDB condition, we investigate how robust both methodologies are by comparing them with this routinely used database. Here, the best F-scores for the rule and the ML systems amount to 34% and 19%, respectively.","Title: Concurrent Lexicalized Dependency Parsing: A Behavioral View on ParseTalk Events.\nYear: 1994\n\nAbstract: The behavioral specification of an object-oriented grammar model is considered. The model is based on full lexicalization, head-orientation via valency constraints and dependency relations, inheritance as a means for non-redundant lexicon specification, and concurrency of computation. The computation model relies upon the actors paradigm, with concurrency entering through asynchronous message passing between actors. In particular, we here elaborate on principles of how the global behavior of a lexically distributed grammar and its corresponding parser can be specified in terms of event type networks and event networks, resp.","Title: Towards new information resources for public health--from WordNet to MedicalWordNet.\nYear: 2006\n\nAbstract: In the last two decades, WordNet has evolved as the most comprehensive computational lexicon of general English. In this article, we discuss its potential for supporting the creation of an entirely new kind of information resource for public health, viz. MedicalWordNet. This resource is not to be conceived merely as a lexical extension of the original WordNet to medical terminology; indeed, there is already a considerable degree of overlap between WordNet and the vocabulary of medicine. Instead, we propose a new type of repository, consisting of three large collections of (1) medically relevant word forms, structured along the lines of the existing Princeton WordNet; (2) medically validated propositions, referred to here as medical facts, which will constitute what we shall call MedicalFactNet; and (3) propositions reflecting laypersons' medical beliefs, which will constitute what we shall call the MedicalBeliefNet. We introduce a methodology for setting up the MedicalWordNet. We then turn to the discussion of research challenges that have to be met to build this new type of information resource. We build a database of sentences relevant to the medical domain. The sentences are generated from WordNet via its relations as well as from medical statements broken down into elementary propositions. Two subcorpora of sentences are distinguished, MedicalBeliefNet and MedicalFactNet. The former is rated for assent by laypersons; the latter for correctness by medical experts. The sentence corpora will be valuable for a variety of applications in information retrieval as well as in research in linguistics and psychology with respect to the study of expert and non-expert beliefs and their linguistic expressions. Our work has to meet several considerable challenges. These include accounting for the distinction between medical experts and laypersons, the social issues of expert-layperson communication in different media, the linguistic aspects of encoding medical knowledge, and the reliability, volume, and emergence of medical knowledge. The work described here has been tested in a small pilot experiment and awaits large-scale implementation.","Title: Annotation Time Stamps - Temporal Metadata from the Linguistic Annotation Process\nYear: 2010\n\nAbstract: We describe the re-annotation of selected types of named entities (persons, organizations, locations) from the MUC7 corpus. The focus of this annotation initiative is on recording the time needed for the linguistic process of named entity annotation. Annotation times are measured on two basic annotation units - sentences vs. complex noun phrases. We gathered evidence that decision times are non- uniformly distributed over the annotation units, while they do not substantially deviate among annotators. This data seems to support the hypothesis that annotation times very much depend on the inherent 'hardness' of each single annotation decision. We further show how such time-stamped information can be used for empirically grounded studies of selective sampling techniques, such as Active Learning. We directly compare Active Learning costs on the basis of token-based vs. time-based measurements. The data reveals that Active Learning keeps its competitive advantage over random sampling in both scenarios though the difference is less marked for the time metric than for the token metric.","Title: A Learner with a Sense for Quality\nYear: 1998\n\nAbstract: A text understanding system with learning capabilities is presented. New concepts are acquired by incorporat- ing two kinds of evidence - knowledge about linguistic constructions in which unknown lexical items occur and knowledge about structural patterns in ontologies such that new concept descriptions can be compared with prior knowledge. On the basis of the quality of evidence gath- ered this way concept hypotheses are generated, ranked according to plausibility, and the most credible ones are selected for assimilation into the domain knowledge base.","Title: Representing Natural Kinds by Spatial Inclusion and Containment\nYear: 2004\n\nAbstract: Biomedical terminologies lack a clear-cut understanding of the semantics of part-whole relations. In order to design a conceptually adequate ontology for the biomedical domain, we investigate different forms of ontological dependencies between parts and wholes. We then suggest, for natural kinds at least, a move away from modeling the relations between parts and wholes and advocate instead conceptually simpler though expressive spatial inclusion/containment relations. We outline a multi-facetted encoding pattern for biological concepts, which is centered around these relations.","Title: Computing text constituency: an algorithmic approach to the generation of text graphs\nYear: 1984\n\nAbstract: An algorithm for text summarization (automatic abstracting) is presented which constitutes the text condensation component of TOPIC, a knowledge-based text information system. Based on the results of text parsing knowledge representation structures of text segments are evaluated in order to determine dominant concepts. By means of an interpretation schema dominant concepts are related in terms of thematic units indicative of the topic(s) of the text segment under consideration. The mutual compatibility of topics of adjacent text segments is determined and corresponding text constituents are constructed. Finally, a text graph is generated linking appropriate text constituents on various levels of text constituency. Accordingly, facilities for text-oriented information retrieval will be based on the manipulation of these text graphs.","Title: Small Is Beautiful - Compact Semantics for Medical Language Processing\nYear: 1999\n\nAbstract: We introduce two abstraction mechanisms by which the process of semantic interpretation of medical documents can be simplified and optimized. One relates to the linguistic generality, the other to the inheritance-based specification of semantic rules. The proposed methodology leads to a parsimonious inventory of abstract, simple and domain-independent semantic interpretation schemata whose effectiveness has been evaluated on a medical text corpus.","Title: Distance Constraint Arrays: A Model for Reasoning on Intervals with Qualitative and Quantitative Distances\nYear: 1998\n\nAbstract: We outline a model of one-dimensional reasoning on interval relations with quantitative and qualitative distances. At the core of this model lie constraints on interval boundaries, partial ordering and subsumption relations on interval relations and interval boundary constraints, and the transformation of interval relations to interval boundary constraints and vice versa. By way of subsumption and approximation criteria on distance constraint arrays, a significant level of conceptual abstraction from the underlying interval boundary constraints is realized when new relations are inferred.","Title: A cognitive cost model of annotations based on eye-tracking data\nYear: 2010\n\nAbstract: We report on an experiment to track complex decision points in linguistic metadata annotation where the decision behavior of annotators is observed with an eye-tracking device. As experimental conditions we investigate different forms of textual context and linguistic complexity classes relative to syntax and semantics. Our data renders evidence that annotation performance depends on the semantic and syntactic complexity of the decision points and, more interestingly, indicates that full-scale context is mostly negligible - with the exception of semantic high-complexity cases. We then induce from this observational data a cognitively grounded cost model of linguistic meta-data annotations and compare it with existing non-cognitive models. Our data reveals that the cognitively founded model explains annotation costs (expressed in annotation time) more adequately than non-cognitive ones.","Title: Concurrent, object-oriented natural language parsing: the ParseTalk model\nYear: 1994\n\nAbstract: TheParseTalk model of concurrent, object-oriented natural language parsing is intro- duced. It builds upon the complete lexical distribution of grammatical knowledge and incorporates inheritance mechanisms in order to express generalizations over sets of lex- ical items. The grammar model integrates declarative well-formedness criteria con- straining linguistic relations between heads and modifiers, and procedural specifications of the communication protocol for establishing these relations. The parser's computa- tion model relies upon the actor paradigm, with concurrency entering through asynchro- nous message passing. We consider various extensions of the basic actor model as required for distributed natural language understanding and elaborate on the semantics of the actor computation model in terms of event type networks (a graph representation for actor grammar specifications) and event networks (graphs which represent the actor parser's behavior). Besides theoretical claims, we present an interactive grammar/parser workbench, a graphical development environment with various types of browsers, trac- ers, inspectors and debuggers that has been adapted to the requirements of large-scale grammar engineering in a distributed, object-oriented specification and programming framework.","Title: Message-Passing Protocols for Real-World Parsing -- An Object-Oriented Model and its Preliminary Evaluation\nYear: 1997\n\nAbstract: We argue for a performance-based design of natural language grammars and their associated parsers in order to meet the constraints imposed by real-world NLP. Our approach incorporates declarative and procedural knowledge about language and language use within an object-oriented specification framework. We discuss several message- passing protocols for parsing and provide reasons for sacrificing completeness of the parse in favor of efficiency based on a preliminary empirical evaluation.","Title: Do users appreciate novel interface features for literature search?: a user study in the life sciences domain\nYear: 2009\n\nAbstract: Faced with the challenges to design an easy-to-use, immediately comprehensible and powerful expert-user interface to search very large document collections in the life sciences, we developed several system prototypes. Their main features were faceting of the domain vocabulary for browsing and searching, flexible search-state-dependent drilling of the terminological hierarchy, dynamic query term auto-completions, and highlighting of matched terms (including synonyms and spelling variants). Under lab conditions we then evaluated these features in several task-based scenarios using camera recordings, thinking-aloud protocols and questionnaires. The results reveal that faceting and highlighting were very well received, while autocompletions seemed less important or were misconceptualized as spelling aids.","Title: Cross-language mining for acronyms and their completions from the web\nYear: 2005\n\nAbstract: We propose a method that aligns biomedical acronyms and their long-form definitions across different languages. We use a freely available search and extraction tool by which abbreviations, together with their fully expanded forms, are massively mined from the Web. In a subsequent step, language-specific variants, synonyms, and translations of the extracted acronym definitions are normalized by referring to a language-independent, shared semantic interlingua.","Title: Are morpho-syntactic features more predictive for the resolution of noun phrase coordination ambiguity than lexico-semantic similarity scores?\nYear: 2008\n\nAbstract: Coordinations in noun phrases often pose the problem that elliptified parts have to be reconstructed for proper semantic interpretation. Unfortunately, the detection of coordinated heads and identification of elliptified elements notoriously lead to ambiguous reconstruction alternatives. While linguistic intuition suggests that semantic criteria might play an important, if not superior, role in disambiguating resolution alternatives, our experiments on the reannotated WSJ part of the Penn Treebank indicate that solely morpho-syntactic criteria are more predictive than solely lexico-semantic ones. We also found that the combination of both criteria does not yield any substantial improvement.","Title: Partonomic reasoning as taxonomic reasoning in medicine\nYear: 1999\n\nAbstract: Taxonomic anatomical knowledge, a major portion of medical ontologies, is fundamentally characterized by is-a and part-whole relations between concepts. While taxonomic reasoning in generalization hierarchies is well-understood, no fully conclusive mechanism as yet exists for partonomic reasoning. We here propose a new representation construct for part-whole relations, based on the formal framework of description logics, that allows us to fully reduce partonomic reasoning to classification-based taxonomic reasoning.","Title: Cognate mapping: a heuristic strategy for the semi-supervised acquisition of a Spanish lexicon from a Portuguese seed lexicon\nYear: 2004\n\nAbstract: We deal with the automated acquisition of a Spanish medical subword lexicon from an already existing Portuguese seed lexicon. Using two non-parallel monolingual corpora we determined Spanish lexeme candidates from Portuguese seed lexicon entries by heuristic cognate mapping. We validated the emergent lexical translation hypotheses by determining the similarity of fixed-window context vectors on the basis of Portuguese and Spanish text corpora.","Title: A denotational semantics for joining description logics and object-oriented programming\nYear: 1997\n\nAbstract: . Despite their common roots, the paradigms of object-orientedprogramming and object-centered knowledge representation could not reallybe joined again. While the knowledge representation branch highlights clearsemantics and focuses on subsumption and classification-based inference algorithms,object-oriented programming shines with flexible control structuresand software-engineering issues. Bridging both worlds seems rewarding, as itwould allow for the description of complex...","Title: Mining the pharmacogenomics literature - a survey of the state of the art.\nYear: 2012\n\nAbstract: This article surveys efforts on text mining of the pharmacogenomics literature, mainly from the period 2008 to 2011. Pharmacogenomics (or pharmacogenetics) is the field that studies how human genetic variation impacts drug response. Therefore, publications span the intersection of research in genotypes, phenotypes and pharmacology, a topic that has increasingly become a focus of active research in recent years. This survey covers efforts dealing with the automatic recognition of relevant named entities (e.g. genes, gene variants and proteins, diseases and other pathological phenomena, drugs and other chemicals relevant for medical treatment), as well as various forms of relations between them. A wide range of text genres is considered, such as scientific publications (abstracts, as well as full texts), patent texts and clinical narratives. We also discuss infrastructure and resources needed for advanced text analytics, e.g. document corpora annotated with corresponding semantic metadata (gold standards and training data), biomedical terminologies and ontologies providing domain-specific background knowledge at different levels of formality and specificity, software architectures for building complex and scalable text analytics pipelines and Web services grounded to them, as well as comprehensive ways to disseminate and interact with the typically huge amounts of semiformal knowledge structures extracted by text mining tools. Finally, we consider some of the novel applications that have already been developed in the field of pharmacogenomic text mining and point out perspectives for future research.","Title: MaHCO: an ontology of the major histocompatibility complex for immunoinformatic applications and text mining.\nYear: 2009\n\nAbstract: Motivation: The high level of polymorphism associated with the major histocompatibility complex (MHC) poses a challenge to organizing associated bioinformatic data, particularly in the area of hematopoietic stem cell transplantation. Thus, this area of research has great potential to profit from the ongoing development of biomedical ontologies, which offer structure and definition to MHC-data related communication and portability issues. Results: We introduce the design considerations, methodological foundations and implementational issues underlying MaHCO, an ontology which represents the alleles and encoded molecules of the major histocompatibility complex. Importantly for human immunogenetics, it includes a detailed level of human leukocyte antigen (HLA) classification. We then present an ontology browser, search interfaces for immunogenetic fact and document retrieval, and the specification of an annotation language for semantic metadata, based on MaHCO. These use cases are intended to demonstrate the utility of ontology-driven bioinformatics in the field of immunogenetics.","Title: Incremental Concept Evolution Based on Adaptive Feature Weighting\nYear: 1997\n\nAbstract: We provide a model of concept drift by which concept classes are automatically updated according to the dynamics of change in the underlying learning environment. Instances with a well-defined degree of feature similarity are incrementally grouped together to form so-called concept versions. A comparative empirical evaluation of the versioning system VerGene, which is built on these premises, demonstrates its effectiveness for properly dealing with concept drift phenomena in medium-sized knowledge bases.","Title: On the interaction of metonymies and anaphora\nYear: 1997\n\nAbstract: From the analysis of naturally occurring texts we obtained evidence for the systematic interaction between nominal anaphora and metonymies. This leads us to postulate an integrated model incorporating both phenomena simultaneously. The consideration of discourse constraints for metonymy resolution allows us to challenge the commonly held view that the interpretation of metonymies should proceed from a literal-meaning-first approach. Thus, we argue for an equally balanced treatment of literal and figurative language use.","Title: Exploiting parallel corpora to scale up multilingual biomedical terminologies.\nYear: 2014\n\nAbstract: Creating and maintaining biomedical terminologies for multiple natural languages is a resource-expensive task, typically carried out by human domain experts. We here report on efforts to computationally support this process by treating term acquisition as a machine translation-guided classification problem capitalizing on parallel corpora. We report on experiments for French, German, Spanish and Dutch parts of a UMLS-derived terminology for which we generated 18k, 23k, 19k and 12k new terms and synonyms, respectively. Based on expert assessments of a novel German terminology segment about 80% of the newly acquired terms were judged as bio-medically reasonable and terminologically valid.","Title: Active Learning-based corpus annotation--the PathoJen experience.\nYear: 2012\n\nAbstract: We report on basic design decisions and novel annotation procedures underlying the development of PathoJen, a corpus of Medline abstracts annotated for pathological phenomena, including diseases as a proper subclass. This named entity type is known to be hard to delineate and capture by annotation guidelines. We here propose a two-category encoding schema where we distinguish short from long mention spans, the first covering standardized terminology (e.g. diseases), the latter accounting for less structured descriptive statements about norm-deviant states, as well as criteria and observations that might signal pathologies. The second design decision relates to the way annotation instances are sampled. Here we subscribe to an Active Learning-based approach which is known to save annotation costs without sacrificing annotation quality by means of a sample bias. By design, Active Learning picks up 'hard' to annotate instances for human annotators, whereas 'easier' ones are passed over to the automatic classifier whose models already incorporate and gradually improve with previous annotation experience.","Title: Effective grading of termhood in biomedical literature.\nYear: 2005\n\nAbstract: The ever-increasing amount of textual information in biomedicine calls for effective procedures for automatic terminology extraction which assist biomedical researchers and professionals in gathering and organizing terminological knowledge encoded in text documents. In this study, we propose a new, linguistically grounded measure for automatically identifying multi-word terms from the biomedical literature. Our approach is based on the limited paradigmatic modifiability of terms and is tested on bigram, trigram and quadgram noun phrases extracted from a 104-million-word text corpus comprised of Medline abstracts. Using the UMLS Metathesaurus as a gold standard, we show that our algorithm substantially outperforms the standard term identification measures and, therefore, qualifies as a high-performing building block for any biomedical terminology mining system.","Title: An ontology for major histocompatibility complex (MHC) alleles and molecules.\nYear: 2007\n\nAbstract: We present a formally coherent and consistent multi-species MHC ontology which includes all human MHC alleles and serological groups. The ontology is part of StemNet, a knowledge management system for hematopoietic stem cell transplantation with an integrated semantic search engine. The Owl-encoded MHC ontology contributes to the system in a threefold manner. First, it supports query formulation and query processing as well as mapping onto external terminological resources, second, it eases the interaction with the search engine when navigating through search results, and finally, it provides a formal language for text annotation, a methodological prerequisite for state-of-the-art natural language text processors which are increasingly based on machine learning methods and hence require annotated text corpora.","Title: A knowledge representation view on biomedical structure and function.\nYear: 2002\n\nAbstract: In biomedical ontologies, structural and functional considerations are of outstanding importance, and concepts which belong to these two categories are highly interdependent. At the representational level both axes must be clearly kept separate in order to support disciplined ontology engineering. Furthermore, the biaxial organization of physical structure (both by a taxonomic and partonomic order) entails intricate patterns of inference. We here propose a layered encoding of taxonomic, partonomic and functional aspects of biomedical concepts using description logics.","Title: Grounding Epistemic Modality In Speakers' Judgments\nYear: 2014\n\nAbstract: As an alternative to subjective and informal expert opinions, we here propose to ground natural language expressions for epistemic modality, such as \"possibly\", or \"might indicate\", in empirical data solicited from members of the speakers' community in epistemic judgment tasks. In an online questionnaire study, we asked 86 subjects to rate the degree of certainty of propositions relative to 27 linguistic expressions. Based on a thorough statistical analysis of the responses we received, we are able to quantify the likelihood as well as ranges of certainty associated with these propositions. Moreover, we generalize these data in terms of an empirically motivated, tri-partite category system of expression sets for propositional certainty which can be reused, e.g. for standardizing annotation guidelines.","Title: Automatic lexicon acquisition for a medical cross-language information retrieval system.\nYear: 2005\n\nAbstract: We present a method for the automated acquisition of a multilingual medical lexicon (for Spanish and Swedish) to be used within the framework of a medical cross-language text retrieval system. We incorporate seed lexicons and parallel corpora derived from the UMLS Metathesaurus. The seed lexicons for Spanish and Swedish are automatically generated from (previously manually constructed) Portuguese, German and English sources. Lexical and semantic hypotheses are then validated making iterative use of co-occurrence patterns of hypothesized translation synonyms in the parallel corpora.","Title: Aligning UniProt and MeSH - a case study on human protein terms.\nYear: 2010\n\nAbstract: Terminologies which lack semantic connectivity hamper the effective search in biomedical fact databases and document retrieval systems. We here focus on the integration of two such isolated resources, the term lists from the protein fact database UNIPROT and the indexing vocabulary MESH from the bibliographic database MEDLINE. The generated semantic ties result from string matching and term set inclusion. We investigated the implicit terminological overlap between both resources in the domain of human proteins and evaluated our approach on a sample of 550 randomly selected UNIPROT entries that were manually mapped to their corresponding MESH headings. We achieved 90% precision and 79% recall (applying taxonomy-sensitive metrics). Fortunately, those proteins we were able to map to the MESH are ten times as frequently discussed in the literature as those on which we failed.","Title: How to distinguish parthood from location in bioontologies.\nYear: 2005\n\nAbstract: The pivotal role of the relation part-of in the descrip- tion of living organisms is widely acknowledged. Or- ganisms are open systems, which means that in con- tradistinction to mechanical artifacts they are charac- terized by a continuous ow and exchange of matter. A closer analysis of the spatial relations in biological or- ganisms reveals that the decision as to whether a given particular is part-of a second particular or whether it is only contained-in the second particular is often controversial. We here propose a rule-based approach which allows us to decide on the basis of well-dened criteria which of the two relations holds between two anatomical objects, given that one spatially includes the other. We discuss the advantages and limitations of this approach, using concrete examples from human anatomy.","Title: Really, is medical sublanguage that different? Experimental counter-evidence from tagging medical and newspaper corpora.\nYear: 2004\n\nAbstract: We compare the performance of two part-of-speech taggers trained on a German newspaper corpus for mixed types of medical documents. TnT, a tagger based on a statistical language model, outperforms Brill's rule-based tagger, and supplied with additional lexicon resources matches state-of-the-art performance figures (close to 97% accuracy) on the medical corpus. We explain this unexpected result by focusing on the statistically,significant part-of-speech type overlap between the newspaper training set and the medical test set. At least at that level sub-language differences seem to vanish. Thus, statistical off-the shellf part-of-speech taggers can immediately be reused for medical language processing.","Title: Fostering Multilinguality in the UMLS: A Computational Approach to Terminology Expansion for Multiple Languages.\nYear: 2014\n\nAbstract: We here report on efforts to computationally support the maintenance and extension of multilingual biomedical terminology resources. Our main idea is to treat term acquisition as a classification problem guided by term alignment in parallel multilingual corpora, using termhood information coming from of a named entity recognition system as a novel feature. We report on experiments for Spanish, French, German and Dutch parts of a multilingual UMLS-derived biomedical terminology. These efforts yielded 19k, 18k, 23k and 12k new terms and synonyms, respectively, from which about half relate to concepts without a previously available term label for these non-English languages. Based on expert assessment of a novel German terminology sample, 80% of the newly acquired terms were judged as reasonable additions to the terminology.","Title: Sharing models and tools for processing german clinical texts.\nYear: 2015\n\nAbstract: The automatic processing of non-English clinical documents is massively hampered by the lack of publicly available medical language resources for training, testing and evaluating NLP components. We suggest sharing statistical models derived from access-protected clinical documents as a reasonable substitute and provide solutions for sentence splitting, tokenization and POS tagging of German clinical texts. These three components were trained on the confidential FRAMED corpus, a non-sharable collection of various German-language clinical document types. The models derived therefrom outperform alternative components from OPENNLP and the Stanford POS tagger, also trained on FRAMED.","Title: The extraction of pharmacogenetic and pharmacogenomic relations--a case study using PharmGKB.\nYear: 2012\n\nAbstract: In this paper, we report on adapting the JREX relation extraction engine, originally developed for the elicitation of protein-protein interaction relations, to the domains of pharmacogenctics and pharmacogenomics. We propose an intrinsic and an extrinsic evaluation scenario which is based on knowledge contained in the PHARMGKB knowledge base. Porting JREX yields favorable results in the range of 80% F-score for Gene-Disease, Gene-Drug, and Drug-Disease relations.","Title: Adding Multilingual Terminological Resources to Parallel Corpora for Statistical Machine Translation Deteriorates System Performance: A Negative Result from Experiments in the Biomedical Domain\nYear: 2015\n\nAbstract: Unlike many other domains, biomedicine not only provides a wide range of parallel text corpora to train statistical machine translation SMT systems on, but also offers substantial amounts of 'parallel lexicons' in the form of multilingual terminologies. We included these lexical repositories, together with common parallel text corpora, into a Moses-based SMT system and three commercial systems and performed experiments on four language pairs, three text genres and several corpus sizes to measure the effects of adding the lexical knowledge sources. Much to our surprise, the SMT systems additionally equipped with 'parallel lexicons' underperformed in comparison with those systems trained on parallel text corpora only. This effect could consistently be shown for all systems by Bleu scores, as well as assessments from human judges.","Title: Emotion Analysis as a Regression Problem - Dimensional Models and Their Implications on Emotion Representation and Metrical Evaluation.\nYear: 2016\n\nAbstract: Emotion analysis (EA) and sentiment analysis are closely related tasks differing in the psychological phenomenon they aim to catch. We address fine-grained models for EA which treat the computation of the emotional status of narrative documents as a regression rather than a classification problem, as performed by coarse-grained approaches. We introduce Ekman's Basic Emotions (BE) and Russell and Mehrabian's Valence-Arousal-Dominance (VAD) model-two major schemes of emotion representation following opposing lines of psychological research, i.e., categorical and dimensional models- and discuss problems when BEs are used in a regression approach. We present the first natural language system thoroughly evaluated for fine-grained emotion analysis using the VAD scheme. Although we only employ simple BOW features, we reach correlation values up until r = .65 with human annotations. Furthermore, we show that the prevailing evaluation methodology relying solely on Pearson's correlation coefficient r is deficient which leads us to the introduction of a complementary error-based metric. Due to the lack of comparable (VAD-based) systems, we, finally, introduce a novel method of mapping between VAD and BE emotion representations to create a reasonable basis for comparison. This enables us to evaluate VAD output against human BE judgments and, thus, allows for a more direct comparison with existing BE-based emotion analysis systems. Even with this, admittedly, error-prone transformation step our VAD-based system achieves state-of-the-art performance in three out of six emotion categories, out-performing all existing BE-based systems but one.","Title: CodE Alltag: A German-Language E-Mail Corpus.\nYear: 2016\n\nAbstract: We introduce CODE ALLTAG, a text corpus composed of German-language e-mails. It is divided into two partitions: the first of these portions, CODE ALLTAG XL, consists of a bulk-size collection drawn from an openly accessible e-mail archive (roughly 1.5M e-mails), whereas the second portion, CODE ALLTAG S+d, is much smaller in size (less than thousand e-mails), yet excels with demographic data from each author of an e-mail. CODE ALLTAG, thus, currently constitutes the largest e-mail corpus ever built. In this paper, we describe, for both parts, the solicitation process for gathering e-mails, present descriptive statistical properties of the corpus, and, for CODE ALLTAG S+d, reveal a compilation of demographic features of the donors of e-mails.","Title: UIMA-Based JCoRe 2.0 Goes GitHub and Maven Central ― State-of-the-Art Software Resource Engineering and Distribution of NLP Pipelines.\nYear: 2016\n\nAbstract: We introduce JCORE 2.0, the relaunch of a UIMA-based open software repository for full-scale natural language processing originating from the Jena University Language & Information Engineering (JULIE) Lab. In an attempt to put the new release of JCORE on firm software engineering ground, we uploaded it to GITHUB, a social coding platform, with an underlying source code versioning system and various means to support collaboration for software development and code modification management. In order to automate the builds of complex NLP pipelines and properly represent and track dependencies of the underlying JAVA code, we incorporated MAVEN as part of our software configuration management efforts. In the meantime, we have deployed our artifacts on MAVEN CENTRAL, as well. JCORE 2.0 offers a broad range of text analytics functionality (mostly) for English-language scientific abstracts and full-text articles, especially from the life sciences domain.","Title: Do Enterprises Have Emotions?\nYear: 2016\n\nAbstract: Emotional language of human individuals has been studied for quite a while dealing with opinions and value judgments people have and share with others. In our work, we take a different stance and investigate whether large organizations, such as major industrial players, have and communicate emotions, as well. Such an anthropomorphic perspective has recently been advocated in management and organization studies which consider organizations as social actors. We studied this assumption by analyzing 1,676 annual business and sustainability reports from 90 top-performing enterprises in the United States, Great Britain and Germany. We compared the measurements of emotions in this homogeneous corporate text corpus with those from RCV1, a heterogeneous Reuters newswire corpus. From this, we gathered empirical evidence that business reports compare well with typical emotion-neutral economic news, whereas sustainability reports are much more emotionally loaded, similar to emotion-heavy sports and fashion news from Reuters. Furthermore, our data suggest that these emotions are distinctive and relatively stable over time per organization, thus constituting an emotional profile for enterprises.","Title: JUFIT: A Configurable Rule Engine for Filtering and Generating New Multilingual UMLS Terms.\nYear: 2015\n\nAbstract: We here describe JuFiT, an easily adjustable rule engine which allows to filter non-natural terms (i.e., ones usually not occurring in running citation texts) from the Umls metathesaurus and even adds new terms to the UMLS (by rewriting non-natural terms). Unlike previous attempts (with MetaMap or Casper), JuFiT serves multilingual purposes in that it runs for English, Spanish, French, German and Dutch documents, as well - the most prominent European languages in terms of UMLS coverage. We evaluated JuFiT under a variety of experimental conditions and found evidence that it increases annotation quality for English, and most likely also for German and Spanish.","Title: Selecting and Tailoring Ontologies with JOYCE.\nYear: 2016\n\nAbstract: We present Joyce, a scalable tool for identifying and assembling relevant (pieces of) ontologies from a repository of source ontologies, thus enabling the effective and efficient reuse of formalized domain knowledge. Joyce includes a conceptual filter to identify relevant classes, minimizes unintended redundancies, i.e. concept duplicates, and excludes knowledge considered irrelevant for the specific conceptual design task.","Title: Semedico: A Comprehensive Semantic Search Engine For The Life Sciences\nYear: 2017\n\nAbstract: SEMEDICO is a semantic search engine designed to support literature search in the life sciences by integrating the semantics of the domain at all stages of the search process-from query formulation via query processing up to the presentation of results. SEMEDICO excels with an ad-hoc search approach which directly reflects relevance in terms of information density of entities and relations among them (events) and, a truly unique feature, ranks interaction events by certainty information reflecting the degree of factuality of the encountered event.","Title: Scholarly Information Extraction Is Going to Make a Quantum Leap with PubMed Central (PMC)® - But Moving from Abstracts to Full Texts Seems Harder than Expected.\nYear: 2017\n\nAbstract: With the increasing availability of complete full texts (journal articles), rather than their surrogates (titles, abstracts), as resources for text analytics, entirely new opportunities arise for information extraction and text mining from scholarly publications. Yet, we gathered evidence that a range of problems are encountered for full-text processing when biomedical text analytics simply reuse existing NLP pipelines which were developed on the basis of abstracts (rather than full texts). We conducted experiments with four different relation extraction engines all of which were top performers in previous BioNLP Event Extraction Challenges. We found that abstract-trained engines loose up to 6.6% F-score points when run on full-text data. Hence, the reuse of existing abstract-based NLP software in a full-text scenario is considered harmful because of heavy performance losses. Given the current lack of annotated full-text resources to train on, our study quantifies the price paid for this short cut.","Title: Representation Mapping: A Novel Approach to Generate High-Quality Multi-Lingual Emotion Lexicons.\nYear: 2018\n\nAbstract: In the past years, sentiment analysis has increasingly shifted attention to representational frameworks more expressive than semantic polarity (being positive, negative or neutral). However, these richer formats (like Basic Emotions or Valence-Arousal-Dominance, and variants therefrom), rooted in psychological research, tend to proliferate the number of representation schemes for emotion encoding. Thus, a large amount of representationally incompatible emotion lexicons has been developed by various research groups adopting one or the other emotion representation format. As a consequence, the reusability of these resources decreases as does the comparability of systems using them. In this paper, we propose to solve this dilemma by methods and tools which map different representation formats onto each other for the sake of mutual compatibility and interoperability of language resources. We present the first large-scale investigation of such representation mappings for four typologically diverse languages and find evidence that our approach produces (near-)gold quality emotion lexicons, even in cross-lingual settings. Finally, we use our models to create new lexicons for eight typologically diverse languages.","Title: Emotion Representation Mapping for Automatic Lexicon Construction (Mostly) Performs on Human Level.\nYear: 2018\n\nAbstract: Emotion Representation Mapping (ERM) has the goal to convert existing emotion ratings from one representation format into another one, e.g., mapping Valence-Arousal-Dominance annotations for words or sentences into Ekmanu0027s Basic Emotions and vice versa. ERM can thus not only be considered as an alternative to Word Emotion Induction (WEI) techniques for automatic emotion lexicon construction but may also help mitigate problems that come from the proliferation of emotion representation formats in recent years. We propose a new neural network approach to ERM that not only outperforms the previous state-of-the-art. Equally important, we present a refined evaluation methodology and gather strong evidence that our model yields results which are (almost) as reliable as human annotations, even in cross-lingual settings. Based on these results we generate new emotion ratings for 13 typologically diverse languages and claim that they have near-gold quality, at least.","Title: JeSemE: A Website for Exploring Diachronic Changes in Word Meaning and Emotion.\nYear: 2018\n\nAbstract: We here introduce a substantially extended version of JeSemE, a website for visually exploring computationally derived time-variant information on word meaning and lexical emotion assembled from five large diachronic text corpora. JeSemE is intended as an interactive tool for scholars in the (digital) humanities who are mostly limited to consulting manually compiled dictionaries for such information, if available at all. JeSemE uniquely combines state-of-the-art distributional semantics with a nuanced model of human emotions, two information streams we deem beneficial for a data-driven interpretation of texts in the humanities.","Title: Inducing Affective Lexical Semantics in Historical Language.\nYear: 2018\n\nAbstract: The emotional connotation attached to words undergoes language change. In this paper, we aim at estimating the emotion which is associated with a given word in former language stages of English and German. Emotion is represented following the popular Valence-Arousal-Dominance (VAD) annotation scheme. While being more expressive than polarity alone, existing word polarity induction methods are typically not suited for addressing it. To overcome this limitation, we present adaptations of two popular algorithms to VAD. To measure the effectiveness of these methods in diachronic settings, we present two small gold lexicons for historical word emotions which where annotated by scholars with proficiency in the respective language (stage). In contrast to claims in previous work, we find that limiting the training set to words of supposedly stable emotional meaning does in fact decrease performance, rather than boosting it. This contribution thus elaborates algorithmic and experimental results which form the basis of some of our later publications (Buechel et al., 2017; Buechel and Hahn, 2018; Hellrich et al., 2018) yet had to be skipped there, partly for brevity, partly for a change in focus of our work.","Title: Downsampling Strategies are Crucial for Word Embedding Reliability.\nYear: 2018\n\nAbstract: The reliability of word embeddings algorithms, i.e., their ability to provide consistent computational judgments of word similarity when trained repeatedly on the same data set, has recently raised concerns. We compared the effect of probabilistic and weighting as downsampling strategies. We found the latter to provide superior reliability while being competitive in accuracy when applied to singular value decomposition-based embeddings","Title: Annotation Data Management with JeDIS\nYear: 2018\n\nAbstract: This paper introduces the Jena Document Information System (JeDIS). The focus lies on its capability to partition annotation graphs into modules. Annotation modules are defined in terms of types from the annotation schema. Modules allow easy manipulation of their annotations (deletion or update) and the creation of alternative annotations of individual documents even for annotation formalisms that by design do not support this feature.","Title: 3000PA-Towards a National Reference Corpus of German Clinical Language.\nYear: 2018\n\nAbstract: We introduce 3000PA, a clinical document corpus composed of 3,000 EPRs from three different clinical sites, which will serve as the backbone of a national reference language resource for German clinical NLP. We outline its design principles, results from a medication annotation campaign and the evaluation of a first medication information extraction prototype using a subset of 3000PA.","Title: Constraining the Acquisition of Concepts by the Quality of Heterogeneous Evidence\nYear: 1997\n\nAbstract: A model of concept acquisition is proposed, which combines various forms of linguistic and conceptual evidence that arise in the course of text understanding processes. We use terminological classification for the creation and management of concept hypotheses, for their incremental annotation by assertions which reflect the quality of available evidence, and for their subsequent evaluation and selection.","Title: CDA-Compliant Section Annotation of German-Language Discharge Summaries: Guideline Development, Annotation Campaign, Section Classification.\nYear: 2018\n\nAbstract: We present the outcome of an annotation effort targeting the content-sensitive segmentation of German clinical reports into sections. We recruited an annotation team of up to eight medical students to annotate a clinical text corpus on a sentence-by-sentence basis in four pre-annotation iterations and one final main annotation step. The annotation scheme we came up with adheres to categories developed for clinical documents in the HL7-CDA (Clinical Document Architecture) standard for section headings. Once the scheme became stable, we ran the main annotation campaign on the complete set of roughly 1,000 clinical documents. Due to its reliance on the CDA standard, the annotation scheme allows the integration of legacy and newly produced clinical documents within a common pipeline. We then made direct use of the annotations by training a baseline classifier to automatically identify sections in clinical reports.","Title: A description logics approach to CGPS.\nYear: 2004\n\nAbstract: We develop a formal framework by which clinical guidelines and protocols (CGPs) can be partially represented as a set of terminological concept definitions using standard description logics. There are two benefits in pursuing such an approach. First, it provides a foundation for logic-based CGP fusion and collision detection. Second, it allows for the checking of clinical treatment episodes from the EPR against CGPs.","Title: Coping with Different Types of Ambiguity Using a Uniform Context Handling Mechanism\nYear: 2000\n\nAbstract: We introduce a uniform context mechanism which is able to adequately represent and manage different forms of ambiguities as they occur in the course of text understanding. Different lexical, syntactic and semantic interpretations are clearly separated by assigning each alternative a single context space for local reasoning. The mechanism we propose directly supports the task of disambiguation at all levels of text analysis, since it also incorporates constraints from the discourse context, as text understanding proceeds.","Title: Exploring Diachronic Lexical Semantics With Jeseme\nYear: 2017\n\nAbstract: Recent advances in distributional semantics combined with the availability of large-scale diachronic corpora offer new research avenues for the Digital Humanities. JESEME, the Jena Semantic Explorer, renders assistance to a non-technical audience to investigate diachronic semantic topics. JESEME runs as a website with query options and interactive visualizations of results, as well as a REST API for access to the underlying diachronic data sets.","Title: Modeling Word Emotion in Historical Language: Quantity Beats Supposed Stability in Seed Word Selection.\nYear: 2019\n\nAbstract: To understand historical texts, we must be aware that language -- including the emotional connotation attached to words -- changes over time. In this paper, we aim at estimating the emotion which is associated with a given word in former language stages of English and German. Emotion is represented following the popular Valence-Arousal-Dominance (VAD) annotation scheme. While being more expressive than polarity alone, existing word emotion induction methods are typically not suited for addressing it. To overcome this limitation, we present adaptations of two popular algorithms to VAD. To measure their effectiveness in diachronic settings, we present the first gold standard for historical word emotions, which was created by scholars with proficiency in the respective language stages and covers both English and German. In contrast to claims in previous work, our findings indicate that hand-selecting small sets of seed words with supposedly stable emotional meaning is actually harmful rather than helpful.","Title: GRASCCO - The First Publicly Shareable, Multiply-Alienated German Clinical Text Corpus.\nYear: 2022\n\nAbstract: We describe the creation of GRASCCO, a novel German-language corpus composed of some 60 clinical documents with more than.43,000 tokens. GRASCCO is a synthetic corpus resulting from a series of alienation steps to obfuscate privacy-sensitive information contained in real clinical documents, the true origin of all GRASCCO texts. Therefore, it is publicly shareable without any legal restrictions We also explore whether this corpus still represents common clinical language use by comparison with a real (non-shareable) clinical corpus we developed as a contribution to the Medical Informatics Initiative in Germany (MII) within the SMITH consortium. We find evidence that such a claim can indeed be made."],"string":"[\n \"Title: Functional Centering\\nYear: 1996\\n\\nAbstract: Based on empirical evidence from a free word order language (German) we propose a fundamental revision of the principles guiding the ordering of discourse entities in the forward-looking centers within the centering model. We claim that grammatical role criteria should be replaced by indicators of the functional information structure of the utterances, i.e., the distinction between context-bound and unbound discourse elements. This claim is backed up by an empirical evaluation of functional centering.\",\n \"Title: Bridging Textual Ellipses\\nYear: 1996\\n\\nAbstract: We present a hybrid text understanding methodology for the resolution of textual ellipsis. It integrates language-independent conceptual criteria and language-dependent functional constraints. The methodological framework for text ellipsis resolution is the centering model that has been adapted to constraints reflecting the functional information structure within utterances, i.e., the distinction between context-bound and unbound discourse elements.\",\n \"Title: Comparatives in Context\\nYear: 1997\\n\\nAbstract: We propose a model of semantic interpretation of comparatives which is based on a mech- anism for semantic copying. Besides common phrasal and clausal forms of comparatives, our model also incorporates the analysis of referential and textual phenomena that inter- act with the interpretation of comparatives, viz. metonymies and omitted complements. In order to allow for efficient processing, guidance from syntactic, semantic, contextual and world knowledge sources is supplied.\",\n \"Title: Scalable Temporal Reasoning\\nYear: 1999\\n\\nAbstract: We introduce two mechanisms for scaling com- putations in the framework of temporal reasoning. The first one addresses abstraction at the method- ological level. Operators are defined that engender flexible switching between different granularities of temporal representation structures. The second one accounts for abstractions at the interface level of a temporal reasoning engine. Various general- izations of temporal relations are introduced that approximate more fine-grained representations by abstracting away irrelevant details.\",\n \"Title: Lean Semantic Interpretation\\nYear: 1999\\n\\nAbstract: We introduce two abstraction mechanisms for streamlining the process of semantic interpreta- tion. Configurational descriptions of dependency graphs increase the linguistic generality of interpre- tation schemata, while interfacing them to lexical and conceptual inheritance hierarchies reduces the amount and complexity of semantic specifications.\",\n \"Title: Iterative Refinement and Quality Checking of Annotation Guidelines - How to Deal Effectively with Semantically Sloppy Named Entity Types, such as Pathological Phenomena.\\nYear: 2012\\n\\nAbstract: We here discuss a methodology for dealing with the annotation of semantically hard to delineate, i.e., sloppy, named entity types. To illustrate sloppiness of entities, we treat an example from the medical domain, namely pathological phenomena. Based on our experience with iterative guideline refinement we propose to carefully characterize the thematic scope of the annotation by positive and negative coding lists and allow for alternative, short vs. long mention span annotations. Short spans account for canonical entity mentions (e.g., standardized disease names), while long spans cover descriptive text snippets which contain entity-specific elaborations (e.g., anatomical locations, observational details, etc.). Using this stratified approach, evidence for increasing annotation performance is provided by kappa-based inter-annotator agreement measurements over several, iterative annotation rounds using continuously refined guidelines. The latter reflects the increasing understanding of the sloppy entity class both from the perspective of guideline writers and users (annotators). Given our data, we have gathered evidence that we can deal with sloppiness in a controlled manner and expect inter-annotator agreement values around 80% for PATHOJEN, the pathological phenomena corpus currently under development in our lab.\",\n \"Title: Mereotopological reasoning about parts and (w)holes in bio-ontologies\\nYear: 2001\\n\\nAbstract: We here deal with mereotopological properties of parts andassociated wholes, locations and empty spaces (holes), withparticular reference to biological structures. Our considerationslead to a basic ontology which contains 'solid object', 'hole' and'boundary' as mutually disjoint primitives. Formally, we embed therelations 'part-of' and 'location-of' into a parsimoniousdescription logic (ALC) and emulate partonomic and spatialreasoning involving these relations by terminological subsumption.In contrast to common conceptualizations, we do not distinguishbetween solids and the regions they occupy, as well as we allowsolids to have holes as proper parts. In order to support thesemodeling decisions, we discuss various concrete examples from humananatomy.\",\n \"Title: Quality Control in the Concept Learning Process\\nYear: 1998\\n\\nAbstract: A text understanding system with learning capabilities is presented. New conceptsare acquired by incorporating two kinds of evidence -- knowledge about linguisticconstructions in which unknown lexical items occur and knowledge about structuralpatterns in ontologies such that new concept descriptions can be compared withprior knowledge. On the basis of the quality of evidence gathered this way concepthypotheses are generated, ranked according to plausibility, and the most credibleones are...\",\n \"Title: Semantic Interpretation of Medical Language - Quantitative Analysis and Qualitative Yield\\nYear: 2001\\n\\nAbstract: We report on results from an empirical analysis of the semantic interpretation of medical free texts. Our approach to semantic interpretation is based on a lean collection of interpretation rules which are triggered by well-defined configurations in dependency graphs in order to compute a conceptual representation of the texts' contents. We evaluate the accuracy of semantic interpretation for three types of syntactic dependency patterns, viz. genitives, auxiliary and modal verb complexes, and prepositional phrases. Besides quantitative considerations, we focus on the heuristic guidance, as provided by patterns underlying the semantic interpretation of prepositional phrases, for monitoring the quality of the medical domain knowledge base.\",\n \"Title: Content management in the SYNDIKATE system —: how technical documents are automatically transformed to text knowledge bases\\nYear: 2000\\n\\nAbstract: Abstract S YN YN DI IKATE E is a family of natural language,understanding,systems for automatically,acquiring,knowledge,from real-world texts (e.g., information technology test reports, medical finding reports), and for transferring their content to formal representation,structures which,constitute a corresponding,text knowledge,base. We present a general system architecture which integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming,cohesive texts. Properly accounting,for text cohesion,phenomena,is a prerequisite for the soundness and validity of the generated,text representation,structures. It is also crucial for any information,system,application making use of automatically generated text knowledge bases in a reliable way, e.g., by inferentially supported fact retrieval. ” 2000 Published by Elsevier Science B.V. All rights reserved. Keywords: Natural language,processing; Text understanding; Knowledge,acquisition from,texts\",\n \"Title: You can't beat frequency (unless you use linguistic knowledge): a qualitative evaluation of association measures for collocation and term extraction\\nYear: 2006\\n\\nAbstract: In the past years, a number of lexical association measures have been studied to help extract new scientific terminology or general-language collocations. The implicit assumption of this research was that newly designed term measures involving more sophisticated statistical criteria would outperform simple counts of co-occurrence frequencies. We here explicitly test this assumption. By way of four qualitative criteria, we show that purely statistics-based measures reveal virtually no difference compared with frequency of occurrence counts, while linguistically more informed metrics do reveal such a marked difference.\",\n \"Title: Quantitative data on referring expressions in biomedical abstracts\\nYear: 2007\\n\\nAbstract: We report on an empirical study that deals with the quantity of different kinds of referring expressions in biomedical abstracts.\",\n \"Title: MEDSYNDIKATE--a natural language system for the extraction of medical information from findings reports.\\nYear: 2002\\n\\nAbstract: medSynDiKATe is a natural language processor, which automatically acquires medical information from findings reports. In the course of text analysis their contents is transferred to conceptual representation structures, which constitute a corresponding text knowledge base. medSynDiKATe is particularly adapted to deal properly with text structures, such as various forms of anaphoric reference relations spanning several sentences. The strong demands medSynDiKATe poses on the availability of expressive knowledge sources are accounted for by two alternative approaches to acquire medical domain knowledge (semi)automatically. We also present data for the information extraction performance of medSynDiKATe in terms of the semantic interpretation of three major syntactic patterns in medical documents.\",\n \"Title: Syntactic Simplification and Semantic Enrichment - Trimming Dependency Graphs for Event Extraction.\\nYear: 2011\\n\\nAbstract: In our approach to event extraction, dependency graphs constitute the fundamental data structure for knowledge capture. Two types of trimming operations pave the way to more effective relation extraction. First, we simplify the syntactic representation structures resulting from parsing by pruning informationally irrelevant lexical material from dependency graphs. Second, we enrich informationally relevant lexical material in the simplified dependency graphs with additional semantic meta data at several layers of conceptual granularity. These two aggregation operations on linguistic representation structures are intended to avoid overfitting of machine learning-based classifiers which we use for event extraction (besides manually curated dictionaries). Given this methodological framework, the corresponding JReX system developed by the JulieLab Team from Friedrich-Schiller-Universitat Jena (Germany) scored on 2nd rank among 24 competing teams for Task 1 in the BioNLP09 Shared Task on Event Extraction, with 45.8% recall, 47.5% precision and 46.7% F1-score on all 3,182 events. In more recent experiments, based on slight modifications of JReX and using the same data sets, we were able to achieve 45.9% recall, 57.7% precision, and 51.1% F1-score.\",\n \"Title: CALBC silver standard corpus.\\nYear: 2010\\n\\nAbstract: The CALBC initiative aims to provide a large-scale biomedical text corpus that contains semantic annotations for named entities of different kinds. The generation of this corpus requires that the annotations from different automatic annotation systems be harmonized. In the first phase, the annotation systems from five participants (EMBL-EBI, EMC Rotterdam, NLM, JULIE Lab Jena, and Linguamatics) were gathered. All annotations were delivered in a common annotation format that included concept identifiers in the boundary assignments and that enabled comparison and alignment of the results. During the harmonization phase, the results produced from those different systems were integrated in a single harmonized corpus (\\\"silver standard\\\" corpus) by applying a voting scheme. We give an overview of the processed data and the principles of harmonization--formal boundary reconciliation and semantic matching of named entities. Finally, all submissions of the participants were evaluated against that silver standard corpus. We found that species and disease annotations are better standardized amongst the partners than the annotations of genes and proteins. The raw corpus is now available for additional named entity annotations. Parts of it will be made available later on for a public challenge. We expect that we can improve corpus building activities both in terms of the numbers of named entity classes being covered, as well as the size of the corpus in terms of annotated documents.\",\n \"Title: Teamwork Support in a Knowledge-Based Information Systems Environment\\nYear: 1991\\n\\nAbstract: Development assistance for interactive database applications (DAIDA) is an experimental environment for the knowledge-assisted development and maintenance of database-intensive information systems from object-oriented requirements and specifications. Within the DAIDA framework, an approach to integrate different tasks encountered in software projects via a conceptual modeling strategy has been developed. Emphasis is put on integrating the semantics of the software development domain with aspects of group work, on social strategies to negotiate problems by argumentation, and on assigning responsibilities for task fulfillment by way of contracting. The implementation of a prototype is demonstrated with a sample session.\",\n \"Title: Massive Biomedical Term Discovery\\nYear: 2005\\n\\nAbstract: Most technical and scientific terms are comprised of complex, multi-word noun phrases but certainly not all noun phrases are\\n technical or scientific terms. The distinction of specific terminology from common non-specific noun phrases can be based\\n on the observation that terms reveal a much lesser degree of distributional variation than non-specific noun phrases. We formalize\\n the limited paradigmatic modifiability of terms and, subsequently, test the corresponding algorithm on bigram, trigram and\\n quadgram noun phrases extracted from a 104-million-word biomedical text corpus. Using an already existing and community-wide\\n curated biomedical terminology as an evaluation gold standard, we show that our algorithm significantly outperforms standard\\n term identification measures and, therefore, qualifies as a high-performant building block for any terminology identification\\n system.\",\n \"Title: Making Objects More Knowledgeable\\nYear: 1997\\n\\nAbstract: Despite their common roots, the paradigms of object-oriented programmingand object-based knowledge representation techniques could not reallybe joined again. A synthesis of both worlds would allow for the descriptionof complex knowledge-based systems with dynamic behavior in a unifiedframework. We propose a corresponding model that extends object-oriented programminginto the classification-based paradigm of terminological reasoning andconsider a natural language understanding task...\",\n \"Title: Towards text knowledge engineering\\nYear: 1998\\n\\nAbstract: We introduce a methodology for automating the maintenance of domain-specific taxonomies based on natural language text understanding. A given ontology is incrementally updated as new concepts are acquired from real-world texts. The acquisition process is centered around the linguistic and conceptual \\\"quality\\\" of various forms of evidence underlying the generation and refinement of concept hypotheses. On the basis of the quality of evidence, concept hypotheses are ranked according to credibility and the most credible ones are selected for assimilation into the domain knowledge base.\",\n \"Title: Subword-Based Text Retrieval\\nYear: 2003\\n\\nAbstract: Document retrieval in languages with a rich and complex morphology 驴 particularly in terms of derivation and (single-word) composition 驴 suffers from serious performance degradation with the stemming-only query-term-to-text-word matching paradigm. We propose an alternative approach in which morphologically complex word forms are segmented into relevant subwords (such as stems, prefixes, suffixes), and subwords constitute the basic unit for indexing and retrieval. We evaluate our approach on a large biomedical document collection.\",\n \"Title: On Text Coherence Parsing\\nYear: 1992\\n\\nAbstract: In this paper global patterns of thematic text organization are considered within the framework of a distributed model of text understanding. Based on the parsing results of prior text cohesion analysis, specialized text grammar modules determine whether some well-defined text macro-organization pattern is computable from the available text representation structures. The model underlying text coherence parsing formalizes hither-to entirely intuitive textlinguistic notions whose origin can be traced back to Danes's work on thematic progression patterns.\",\n \"Title: ParseTalk about Textual Ellipsis\\nYear: 1995\\n\\nAbstract: A hybrid methodology for the resolution oftext-level ellipsis is presented in this paper.It incorporates conceptual proximity criteriaapplied to ontologically well-engineered domainknowledge bases and an approach tocentering based on functional topic/commentpatterns. We state text grammatical predicatesfor ellipsis and then turn to the proceduralaspects of their evaluation within theframework of an actor-based implementationof a lexically distributed parser.Keywords: text...\",\n \"Title: PARSETALK about Functional Anaphora\\nYear: 1996\\n\\nAbstract: . We present a hybrid methodology for the resolution of functionalanaphora. It incorporates conceptual proximity criteria applied toontologically well-engineered domain knowledge bases and an approachto centering based on functional topic/comment patterns. We state grammaticalpredicates for functional anaphora and then turn to the proceduralaspects of their evaluation within the framework of an actor-basedimplementation of a lexically distributed parser.1 IntroductionText phenomena,...\",\n \"Title: Spatial location and its relevance for terminological inferences in bio-ontologies.\\nYear: 2007\\n\\nAbstract: An adequate and expressive ontological representation of biological organisms and their parts requires formal reasoning mechanisms for their relations of physical aggregation and containment.We demonstrate that the proposed formalism allows to deal consistently with \\\"role propagation along non-taxonomic hierarchies\\\", a problem which had repeatedly been identified as an intricate reasoning problem in biomedical ontologies.The proposed approach seems to be suitable for the redesign of compositional hierarchies in (bio)medical terminology systems which are embedded into the framework of the OBO (Open Biological Ontologies) Relation Ontology and are using knowledge representation languages developed by the Semantic Web community.\",\n \"Title: Multi-Task Active Learning for Linguistic Annotations\\nYear: 2008\\n\\nAbstract: We extend the classical single-task active learning (AL) approach. In the multi-task ac- tive learning (MTAL) paradigm, we select ex- amples for several annotation tasks rather than for a single one as usually done in the con- text of AL. We introduce two MTAL meta- protocols, alternating selection and rank com- bination, and propose a method to implement them in practice. We experiment with a two- task annotation scenario that includes named entity and syntactic parse tree annotations on three different corpora. MTAL outperforms random selection and a stronger baseline, one- sided example selection, in which one task is pursued using AL and the selected examples are provided also to the other task.\",\n \"Title: An integrated model of semantic and conceptual interpretation from dependency structures\\nYear: 2000\\n\\nAbstract: We propose a two-layered model for computing semantic and conceptual interpretations from dependency structures. Abstract interpretation schemata generate semantic interpretations of 'minimal' dependency subgraphs, while production rules whose specification is rooted in ontological categories derive a canonical conceptual interpretation from semantic interpretation structures. Configurational descriptions of dependency graphs increase the linguistic generality of interpretation schemata, while interfacing schemata and productions to lexical and conceptual class hierarchies reduces the amount and complexity of semantic specifications.\",\n \"Title: A Terminological Qualification Calculus for Preferential Reasoning under Uncertainty\\nYear: 1996\\n\\nAbstract: We introduce a qualitative model of uncertain reasoning and illustrate its application in the framework of a natural language understanding task. Considering uncertain reasoning as a preferential choice problem between alternative hypotheses, the model we provide assigns quality labels to single evidences for or against a hypothesis, combines the generated labels in terms of the overall credibility of a single hypothesis, and, finally, computes a preference order on the entire set of competing hypotheses. This model of quality-based uncertain reasoning is entirely embedded in a terminological logic framework.\",\n \"Title: Event extraction from trimmed dependency graphs\\nYear: 2009\\n\\nAbstract: We describe the approach to event extraction which the JulieLab Team from FSU Jena (Germany) pursued to solve Task 1 in the \\\"BioNLP'09 Shared Task on Event Extraction\\\". We incorporate manually curated dictionaries and machine learning methodologies to sort out associated event triggers and arguments on trimmed dependency graph structures. Trimming combines pruning irrelevant lexical material from a dependency graph and decorating particularly relevant lexical material from that graph with more abstract conceptual class information. Given that methodological framework, the JulieLab Team scored on 2nd rank among 24 competing teams, with 45.8% precision, 47.5% recall and 46.7% F1-score on all 3,182 events.\",\n \"Title: Concurrent Lexicalized Dependency Parsing: The ParseTalk Model.\\nYear: 1994\\n\\nAbstract: A grammar model for concurrent, object-oriented natural language parsing is introduced. Complete lexical distribution of grammatical knowledge is achieved building upon the head-oriented notions of valency and dependency, while inheritance mechanisms are used to capture lexical generalizations. The underlying concurrent computation model relies upon the actor paradigm. We consider message passing protocols for establishing dependency relations and ambiguity handling.\",\n \"Title: Paradigmatic modifiability statistics for the extraction of complex multi-word terms\\nYear: 2005\\n\\nAbstract: We here propose a new method which sets apart domain-specific terminology from common non-specific noun phrases. It is based on the observation that terminological multi-word groups reveal a considerably lesser degree of distributional variation than non-specific noun phrases. We define a measure for the observable amount of paradigmatic modifiability of terms and, subsequently, test it on bigram, trigram and quadgram noun phrases extracted from a 104-million-word biomedical text corpus. Using a community-wide curated biomedical terminology system as an evaluation gold standard, we show that our algorithm significantly outperforms a variety of standard term identification measures. We also provide empirical evidence that our methodolgy is essentially domain- and corpus-size-independent.\",\n \"Title: ParseTalk about Sentence- and Text-Level Anaphora\\nYear: 1995\\n\\nAbstract: We provide a unified account of sentence-level and text-level anaphora within the framework of a dependency-based grammar model. Criteria for anaphora resolution within sentence boundaries rephrase major concepts from GB's binding theory, while those for text-level anaphora incorporate an adapted version of a Grosz-Sidner-style focus model.\",\n \"Title: A Conceptual Reasoning Approach to Textual Ellipsis\\nYear: 1996\\n\\nAbstract: We present a hybrid text understanding methodology for the resolution of textual ellipsis. It integrates conceptual criteria (based on the well-formedness and conceptual strength of role chains in a terminological knowledge base) and functional constraints reflecting the utterances' information structure (based on the distinction between context-bound and unbound discourse elements). The methodological framework for text ellipsis resolution is the centering model that has been adapted to these constraints.\",\n \"Title: Towards an upper level ontology for molecular biology.\\nYear: 2006\\n\\nAbstract: There is a growing need for the general-purpose description of the basic conceptual entities in the life sciences. Up until now, upper level models have mainly been purpose-driven, such as the GENIA ontology, originally devised as a vocabulary for corpus annotation. As an alternative,we here present BioTop, a description-logic-based top level ontology for molecular biology, which we consider as an ontologically conscious redesign of the GENIA ontology.\",\n \"Title: A proposal for a configurable silver standard\\nYear: 2010\\n\\nAbstract: Among the many proposals to promote alternatives to costly to create gold standards, just recently the idea of a fully automatically, and thus cheaply, to set up silver standard has been launched. However, the current construction policy for such a silver standard requires crucial parameters (such as similarity thresholds and agreement cut-offs) to be set a priori, based on extensive testing though, at corpus compile time. Accordingly, such a corpus is static, once it is released. We here propose an alternative policy where silver standards can be dynamically optimized and customized on demand (given a specific goal function) using a gold standard as an oracle.\",\n \"Title: Automatic generation of hypertext knowledge bases\\nYear: 1988\\n\\nAbstract: A model of knowledge-based text condensation is presented which has been implemented as part of the text analysis system TOPIC. The condensation process transforms the text representation structures resulting from the text parse into a more abstract thematic description of what the text is about, filtering out irrelevant knowledge structures and preserving only the most salient concepts. The topical structure of a text, finally, is represented in a hierarchical text graph which supports variable degrees of abstraction for text summarization as well as content-oriented retrieval of text knowledge. Due to their non-linear organization, text graphs share a lot of similarities with hypertexts. Their contribution to this field incorporates a methodology for the automatic generation of hypertexts from given full-text files, a close coupling of basic hypertext notions (links, nodes) to the formal specifications of a frame representation model, and conceptual navigation and filtering facilities which allow a user-defined level of information granularity when accessing hypertext knowledge bases.\",\n \"Title: Efficient annotation with the Jena ANnotation Environment (JANE)\\nYear: 2007\\n\\nAbstract: With ever-increasing demands on the diversity of annotations of language data, the need arises to reduce the amount of efforts involved in generating such value-added language resources. We introduce here the Jena ANnotation Environment (Jane), a platform that supports the complete annotation life-cycle and allows for 'focused' annotation based on active learning. The focus we provide yields significant savings in annotation efforts by presenting only informative items to the annotator. We report on our experience with this approach through simulated and real-world annotations in the domain of immunogenetics for NE annotations.\",\n \"Title: Deep Knowledge Discovery from Natural Language Texts\\nYear: 1997\\n\\nAbstract: We introduce a knowledge-based approach to deep knowledge discovery from real-world natural language texts. Data mining, data interpretation, and data cleaning are all incorporated in cy- cles of quality-based terminological reasoning processes.. The methodology we propose identifies new knowledge items and assimilates them into a continuously updated domain knowl- edge base.\",\n \"Title: Parts, Locations, and Holes - Formal Reasoning about Anatomical Structures\\nYear: 2001\\n\\nAbstract: We propose an ontology engineering framework for the anatomy domain, focusing on mereotopological properties of parts, locations and empty spaces (holes). We develop and formally describe a basic ontology consisting of the mutually disjoint primitives solid object, hole and boundary. We embed the relations part-of and location-of into a parsimonious description logic (ALC) and emulate advanced reasoning across these relations - such as transitivity at the T-Box level - by taxonomic subsumption. Unlike common conceptualizations we do not distinguish between solids and the regions they occupy, as well as we allow solids to have holes as proper parts. Concrete examples from human anatomy are used to support our claims.\",\n \"Title: Intelligent Text Analysis for Dynamically Maintaining and Updating Domain Knowledge Bases\\nYear: 1997\\n\\nAbstract: We propose a knowledge-intensive text analysis approach which deals with the continuous assimilation of new concepts into\\n domain knowledge bases. Text understanding and knowledge acquisition proceed in tandem on the basis of terminological reasoning.\\n Concept learning is considered an evidence-based choice problem the solution of which balances the “quality” of various clues\\n from the linguistic structure of the texts and conceptual structures in the knowledge bases.\",\n \"Title: A qualitative growth model for real world text knowledge bases\\nYear: 1997\\n\\nAbstract: We introduce a knowledge-based approach to the analysis of real-world natural language texts, which addresses the particular needs of dealing with new knowledge items. In order to incrementally expand the underlying conceptual knowledge base, we exploit two kinds of evidence from the text understanding process, viz. qualitative knowledge about linguistic constructions in n atural language texts and that about structural patterns in the emerging text knowledge base. These clues are used to generate concept hypotheses, rank them according to plausibility, and selec t the most credible ones for assimilation into the conceptual knowledge base. We demonstrate the feasibility of our approach and discuss the results of an empirical evaluation in terms of concept learning rate s and learning accuracy.\",\n \"Title: Building a Large Knowledge Base Semi-Automatically\\nYear: 2001\\n\\nAbstract: We describe a knowledge engineering approach by which conceptual knowledge is extracted from an informal, se- mantically weak medical thesaurus (UMLS) and automat- ically converted into a formally sound description logics system. Our approach consists of four steps: concept definitions are automatically generated from the UMLS source, integrity checking of taxonomic and partonomic hierarchies is performed by the terminological classifier, cycles and inconsistencies are eliminated, and incremental refinement of the evolving knowledge base is performed by a domain expert. We report on experiments with a ter- minological knowledge base composed of 164,000 con- cepts and 76,000 relations.\",\n \"Title: An empirical assessment of semantic interpretation\\nYear: 2000\\n\\nAbstract: We introduce a framework for semantic interpretation in which dependency structures are mapped to conceptual representations based on a parsimonious set of interpretation schemata. Our focus is on the empirical evaluation of this approach to semantic interpretation, i.e., its quality in terms of recall and precision. Measurements are taken with respect to two real-world domains, viz. information technology test reports and medical finding reports.\",\n \"Title: Evaluating the impact of alternative dependency graph encodings on solving event extraction tasks\\nYear: 2010\\n\\nAbstract: In state-of-the-art approaches to information extraction (IE), dependency graphs constitute the fundamental data structure for syntactic structuring and subsequent knowledge elicitation from natural language documents. The top-performing systems in the BioNLP 2009 Shared Task on Event Extraction all shared the idea to use dependency structures generated by a variety of parsers --- either directly or in some converted manner --- and optionally modified their output to fit the special needs of IE. As there are systematic differences between various dependency representations being used in this competition, we scrutinize on different encoding styles for dependency information and their possible impact on solving several IE tasks. After assessing more or less established dependency representations such as the Stanford and CoNLL-X dependencies, we will then focus on trimming operations that pave the way to more effective IE. Our evaluation study covers data from a number of constituency- and dependency-based parsers and provides experimental evidence which dependency representations are particularly beneficial for the event extraction task. Based on empirical findings from our study we were able to achieve the performance of 57.2% F-score on the development data set of the BioNLP Shared Task 2009.\",\n \"Title: Question Answering for Health Professionals and Laypersons.\\nYear: 2013\\n\\nAbstract: and objective Knowledge sharing in the medical domain is limited by availability and language. The availability dimensions are growth of knowledge and communication systems design. The language dimensions are semantic, syntactic and in some cases phonetic. No human can comprehend the total of medical knowledge. Question answering (ICT-) systems would seem to be a good solution to create a “Wizard” or medical “Oracle of Delphi” available to anyone, anywhere, anytime. The “Wizard” how-ever must understand and speak about the questions on the conditions of the questioner. Language mediating the knowledge, no matter if spoken or written (by human or computer), must meet the user on his or hers conditions. Current systems are primarily using terminology resources created by the health professionals. The professional language, even when it is standardized, is a well known obstacle to sharing of knowledge to laypersons. This might be one of the biggest issues in patient empowerment. In this panel we aim at discussing strategies to overcome some of these barriers for medical knowledge sharing between health professionals and laypersons.\",\n \"Title: Part-whole representation and reasoning in formal biomedical ontologies.\\nYear: 2005\\n\\nAbstract: Biomedical ontologies are typically structured in a biaxial way, reflecting both a taxonomic (is-a) and a partonomic (part-of) hierarchy. Commonly used biomedical terminologies, which incorporate such distinctions excel in terms of broad coverage but lack a rigid formal foundation. The latter, however, is a prerequisite for automated reasoning. For the biomedical domain, it is not only crucial to cope with ontological dependencies between wholes and their parts but also with specific reasoning patterns which underlie the propagation of roles across partonomic hierarchies.We scale down part-whole reasoning to subsumption-based taxonomic reasoning within the formal framework of a parsimonious variant of description logics (viz. ALC).We provide a formal basis for ontological engineering in the domain of biomedicine, as far as part-whole relationships are concerned, by addressing typical reasoning patterns encountered in this domain.\",\n \"Title: Towards the ontological foundations of symbolic biological theories.\\nYear: 2007\\n\\nAbstract: Support for the symbolic representation of the physical structure of living organisms by an ontologically solid and logically sound foundation as a basis for formal reasoning.A set of canonical relations and attributes necessary for empirically adequate descriptions of biological entities is proposed.It is shown how a broad range of biological organisms and their parts can be represented by cascading theories which are ordered by the dimensions of granularity, development, species, and canonicity.The proposed representation of biological objects is non-redundant and compatible with inter- and intra-species similarities, developmental stages and pathological deviations.\",\n \"Title: The GeneReg Corpus for Gene Expression Regulation Events - An Overview of the Corpus and its In-Domain and Out-of-Domain Interoperability\\nYear: 2010\\n\\nAbstract: Despite the large variety of corpora in the biomedical domain their annotations differ in many respects, e.g., the coverage of different, highly specialized knowledge domains, varying degrees of granularity of the targeted relations, the specificity of linguistic grounding of relations and named entities referred to in the documents, etc. We here introduce GENEREG (Gene Regulation Corpus), the result of an annotation campaign led by the Jena University Language & Information Engineering (JULIE) Lab. The GENEREG corpus consists of 314 abstracts dealing with the regulation of gene expression in the model organism E. coli. Our emphasis in this paper is on the compatibility and, thus, linkage, of the GENEREG corpus with the alternative GENIA event corpus and with several in-domain and out-of-domain lexical resources, e.g., the SPECIALIST LEXICON, FRAMENET, and WORDNET. The links we established from the GENEREG corpus to these external resources will help improve the performance of the automatic relation extraction engine JREX trained and evaluated on GENEREG.\",\n \"Title: Semi-supervised active learning for sequence labeling\\nYear: 2009\\n\\nAbstract: While Active Learning (AL) has already been shown to markedly reduce the annotation efforts for many sequence labeling tasks compared to random selection, AL remains unconcerned about the internal structure of the selected sequences (typically, sentences). We propose a semi-supervised AL approach for sequence labeling where only highly uncertain subsequences are presented to human annotators, while all others in the selected sequences are automatically labeled. For the task of entity recognition, our experiments reveal that this approach reduces annotation efforts in terms of manually labeled tokens by up to 60% compared to the standard, fully supervised AL scheme.\",\n \"Title: Towards a computational paradigm for biomedical structure\\nYear: 2004\\n\\nAbstract: The symbolic representation of the physical struc- ture of living organisms needs an ontologically well-founded and logically sound approach so that formal reasoning can adequately be supported. We describe a set of canonical relations and attrib- utes necessary for the description of biological structures. Based on these epistemological prim- itives, we sketch how a broad range of organisms can be represented by cascading theories which are ordered by various dimensions, such as gran- ularity, development, species and canonicity. We thus aim at a rational reconstruction and non- redundant representation of biological structure notions.\",\n \"Title: Joint knowledge capture for grammars and ontologies\\nYear: 2001\\n\\nAbstract: We introduce a methodology for automating the maintenance and growth of domain-specific concept taxonomies and grammatical class hierarchies simultaneously, based on knowledge capture from natural language texts. The assimilation process is centered around the linguistic and conceptual `quality' of various forms of evidence underlying the generation, assessment and on-going refinement of lexical and concept hypotheses. On the basis of the strength of evidence, hypotheses are ranked according to plausibility, and the most reasonable ones are selected for assimilation into the given lexical class hierarchy and domain ontology.\",\n \"Title: Creating knowledge repositories from biomedical reports: the MEDSYNDIKATE text mining system.\\nYear: 2002\\n\\nAbstract: MEDSYNDIKATE is a natural language processor for automatically acquiring knowledge from medical finding reports. The content of these documents is transferred to formal representation structures which constitute a corresponding text knowledge base. The system architecture integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming cohesive texts. The strong demands MEDSYNDIKATE poses to the availability of expressive knowledge sources are accounted for by two alternative approaches to (semi)automatic ontology engineering. We also present data for the knowledge extraction performance of MEDSYNDIKATE for three major syntactic patterns in medical documents.\",\n \"Title: Learning indexing patterns from one language for the benefit of others\\nYear: 2004\\n\\nAbstract: Using language technology for text analysis and light-weight ontologies as a content-mediating level, we acquire indexing patterns from vast amounts of indexing data for English-language medical documents. This is achieved by statistically relating interlingual representations of these documents (based on text token bigrams) to their associated index terms. From these 'English' indexing patterns, we then induce the associated index terms for German and Portuguese documents when their interlingual representations match those of English documents. Thus, we learn from past English indexing experience and transfer it in an unsupervised way to non-English texts, without ever having seen concrete indexing data for languages other than English.\",\n \"Title: A reappraisal of sentence and token splitting for life sciences documents.\\nYear: 2007\\n\\nAbstract: Natural language processing of real-world documents requires several low-level tasks such as splitting a piece of text into its constituent sentences, and splitting each sentence into its constituent tokens to be performed by some preprocessor (prior to linguistic analysis). While this task is often considered as unsophisticated clerical work, in the life sciences domain it poses enormous problems due to complex naming conventions. In this paper, we first introduce an annotation framework for sentence and token splitting underlying a newly constructed sentence- and token-tagged biomedical text corpus. This corpus serves as a training environment and test bed for machine-learning based sentence and token splitters using Conditional Random Fields (CRFs). Our evaluation experiments reveal that CRFs with a rich feature set substantially increase sentence and token detection performance.\",\n \"Title: An integrated, dual learner for grammars and ontologies\\nYear: 2002\\n\\nAbstract: We introduce a dual-use methodology for automating the maintenance and growth of two types of knowledge sources, which are crucial for natural language text understanding-background knowledge of the underlying domain and linguistic knowledge about the lexicon and the grammar of the underlying natural language. A particularity of this approach is that learning occurs simultaneously with the on-going text understanding process. The knowledge assimilation process is centered around the linguistic and conceptual 'quality' of various forms of evidence underlying the generation, assessment and on-going refinement of lexical and concept hypotheses. On the basis of the strength of evidence, hypotheses are ranked according to qualitative plausibility criteria, and the most reasonable ones are selected for assimilation into the already given lexical class hierarchy and domain ontology.\",\n \"Title: Collocation extraction based on modifiability statistics\\nYear: 2004\\n\\nAbstract: We introduce a new, linguistically grounded measure of collocativity based on the property of limited modifiability and test it on German PP-verb combinations. We show that our measure not only significantly outperforms the standard lexical association measures typically employed for collocation extraction, but also yields a valuable by-product for the creation of collocation databases, viz. possible structural and lexical attributes. Our approach is language-, structure-, and domain-independent because it only requires some shallow syntactic analysis (e.g., a POS-tagger and a phrase chunker).\",\n \"Title: Collaboratively Annotating Multilingual Parallel Corpora in the Biomedical Domain―some MANTRAs.\\nYear: 2014\\n\\nAbstract: The coverage of multilingual biomedical resources is high for the English language, yet sparse for non-English languages-an observation which holds for seemingly well-resourced, yet still dramatically low-resourced ones such as Spanish, French or German but even more so for really under-resourced ones such as Dutch. We here present experimental results for automatically annotating parallel corpora and simultaneously acquiring new biomedical terminology for these under-resourced non-English languages on the basis of two types of language resources, namely parallel corpora (i.e. full translation equivalents at the document unit level) and (admittedly deficient) multilingual biomedical terminologies, with English as their anchor language. We automatically annotate these parallel corpora with biomedical named entities by an ensemble of named entity taggers and harmonize non-identical annotations the outcome of which is a so-called silver standard corpus. We conclude with an empirical assessment of this approach to automatically identify both known and new terms in multilingual corpora.\",\n \"Title: A comparison of models for cost-sensitive active learning\\nYear: 2010\\n\\nAbstract: Active Learning (AL) is a selective sampling strategy which has been shown to be particularly cost-efficient by drastically reducing the amount of training data to be manually annotated. For the annotation of natural language data, cost efficiency is usually measured in terms of the number of tokens to be considered. This measure, assuming uniform costs for all tokens involved, is, from a linguistic perspective at least, intrinsically inadequate and should be replaced by a more adequate cost indicator, viz. the time it takes to manually label selected annotation examples. We here propose three different approaches to incorporate costs into the AL selection mechanism and evaluate them on the Muc7T corpus, an extension of the Muc7 newspaper corpus that contains such annotation time information. Our experiments reveal that using a cost-sensitive version of semi-supervised AL, up to 54% of true annotation time can be saved compared to random selection.\",\n \"Title: The SYNDIKATE text knowledge base generator\\nYear: 2001\\n\\nAbstract: SYNDIKATE comprises a family of text understanding systems for automatically acquiring knowledge from real-world texts, viz. information technology test reports and medical finding reports. Their content is transformed to formal representation structures which constitute corresponding text knowledge bases. SYNDIKATE's architecture integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming cohesive texts. Besides centering-based discourse analysis mechanisms for pronominal, nominal and bridging anaphora, SYNDIKATE is supplied with a learning module for automatically bootstrapping its domain knowledge as text analysis proceeds.\",\n \"Title: Unsupervised multilingual word sense disambiguation via an interlingua\\nYear: 2005\\n\\nAbstract: We present an unsupervised method for resolving word sense ambiguities in one language by using statistical evidence assembled from other languages. It is crucial for this approach that texts are mapped into a language-independent interlingual representation. We also show that the coverage and accuracy resulting from multilingual sources outperform analyses where only monolingual training data is taken into account.\",\n \"Title: Cross-lingual alignment of biomedical acronyms and their expansions.\\nYear: 2006\\n\\nAbstract: We propose a method that aligns biomedical acronyms and their definitions across different languages. The approach is based upon a freely available tool for the extraction of abbreviations together with their expansions, and the subsequent normalization of language-specific variants, synonyms, and translations of the extracted acronym definitions. In this step, acronym expansions are mapped onto a language-independent concept-layer on which intra- as well as interlingual comparisons are drawn.\",\n \"Title: How Knowledge Drives Understanding - Matching Medical Ontologies With The Needs Of Medical Language Processing\\nYear: 1999\\n\\nAbstract: In this article, we introduce a knowledge-based approach to medical text understanding. From an in-depth consideration of deep sentence and text understanding we distill basic requirements for an adequate knowledge representation framework. These requirements are then matched with currently available medical ontologies (thesauri, terminologies, etc.). A fundamental trade-off is recognized between large-scale conceptual coverage on the one hand, and formal mechanisms for integrity preservation and conceptual expressiveness on the ether hand. We discuss various shortcomings of the most wide-spread ontologies to capture medical knowledge in-the-large. As a result, we argue for the need of a formally sound and expressive model along the lines of KL-ONE-style terminological representation systems in the format of description logics. These provide an adequate methodology for designing more sophisticated, flexible medical ontologies serving the needs of 'deep' knowledge applications which are by no means restricted to medical language processing. (C) 1999 Elsevier Science B.V. All rights reserved.\",\n \"Title: Reducing class imbalance during active learning for named entity annotation\\nYear: 2009\\n\\nAbstract: In lots of natural language processing tasks, the classes to be dealt with often occur heavily imbalanced in the underlying data set and classifiers trained on such skewed data tend to exhibit poor performance for low-frequency classes. We introduce and compare different approaches to reduce class imbalance by design within the context of active learning (AL). Our goal is to compile more balanced data sets up front during annotation time when AL is used as a strategy to acquire training material. We situate our approach in the context of named entity recognition. Our experiments reveal that we can indeed reduce class imbalance and increase the performance of classifiers on minority classes while preserving a good overall performance in terms of macro F-score.\",\n \"Title: Automatic lexeme acquisition for a multilingual medical subword thesaurus\\nYear: 2007\\n\\nAbstract: Conclusions Our approach constitutes a promising method for the automated creation of new lexicon entries and their linkage to semantic identifiers.\",\n \"Title: Approximating Learning Curves for Active-Learning-Driven Annotation\\nYear: 2008\\n\\nAbstract: Active learning (AL) is getting more and more popular as a methodology to considerably reduce the annotation effort when building training material for statistical learning methods for various NLP tasks. A crucial issue rarely addressed, however, is when to actually stop the annotation process to profit from the savings in efforts. This question is tightly related to estimating the classifier performance after a certain amount of data has already been annotated. While learning curves are the default means to monitor the progress of the annotation process in terms of classifier performance, this requires a labeled gold standard which - in realistic annotation settings, at least - is often unavailable. We here propose a method for committee-based AL to approximate the progression of the learning curve based on the disagreement among the committee members. This method relies on a separate, unlabeled corpus and is thus well suited for situations where a labeled gold standard is not available or would be too expensive to obtain. Considering named entity recognition as a test case we provide empirical evidence that this approach works well under simulation as well as under real-world annotation conditions.\",\n \"Title: From GENIA to BIOTOPTowards a Top-Level Ontology for Biology\\nYear: 2006\\n\\nAbstract: The increasing need for advanced ontology-based knowledge manage- ment in the life sciences is generally being acknowledged but, up until now, the development of biological ontologies lacks adherence to foundational principles of ontology design. This is particularly true of so-called upp er-level ontologies such as the GENIA ontology which covers biological continuants and has mainly been devised for corpus annotation in a text mining context. As an alternative, we intro- duce BIOTOP, an upper ontology of physical continuants in the domain of biology, with a coverage similar to the GENIA ontology. We report on design specifications and modeling decisions for BIOTOP which are based upon formal ontology princi- ples. As a major desideratum, these continuants are described in terms of necessary and sufficient conditions. We accomplished this goal for 85 o ut of the 146 existing GENIA classes. We use OWL-DL as a formal knowledge representation language and may thus use a terminological reasoner for classificatio n in order to check and maintain consistency during the ontology engineering phase.\",\n \"Title: An Experimental Assessment of Direct Versus. Interlingual Translation for Cross-Language Information Retrieval\\nYear: 2004\\n\\nAbstract: We introduce an interlingua-based approach to cross- language information retrieval, in which queries, as well as documents, are mapped onto a language-independent concept layer and retrieval operations are performed at the level of that interlingua. This approach is contrasted with one which oper- ates without such an intermediary concept level. Non-English queries (German ones, in our experiments) are directly trans- lated to English queries which, subsequently, are processed on English documents. We provide an empirical evaluation of both alternatives on a large medical document collection.\",\n \"Title: Centering in-the-large: computing referential discourse segments\\nYear: 1997\\n\\nAbstract: We specify an algorithm that builds up a hierarchy of referential discourse segments from local centering data. The spatial extension and nesting of these discourse segments constrain the reachability of potential antecedents of an anaphoric expression beyond the local level of adjacent center pairs. Thus, the centering model is scaled up to the level of the global referential structure of discourse. An empirical evaluation of the algorithm is supplied.\",\n \"Title: Tagging Medical Documents with High Accuracy\\nYear: 2004\\n\\nAbstract: We ran both Brill's rule-based tagger and TNT, a statistical tagger, with a default German newspaper-language model on a medical text corpus. Supplied with limited lexicon resources, TNT outperforms the Brill tagger with state-of-the-art performance figures (close to 97% accuracy). We then trained TNT on a large annotated medical text corpus, with a slightly extended tagset that captures certain medical language particularities, and achieved 98% tagging accuracy. Hence, statistical off-the-shelf POS taggers cannot only be immediately reused for medical NLP, but they also achieve - when trained on medical corpora - a higher performance level than for the newspaper genre.\",\n \"Title: Incremental Centering And Center Ambiguity\\nYear: 1996\\n\\nAbstract: In this paper, we present a model of anaphor resolution withinthe framework of the centering model. The consideration ofan incremental processing mode introduces the need to managestructural ambiguity at the center level. Hence, the centeringframework is further refined to account for local andglobal parsing ambiguities which propagate up to the level ofcenter representations, yielding moderately adapted data structuresfor the centering algorithm.IntroductionPsycholinguistic studies...\",\n \"Title: An Empirical Evaluation of a System for Text Knowledge Acquisition\\nYear: 1997\\n\\nAbstract: We introduce a formal model and a corresponding system architecture for the acquisition of new concepts from real-world natural language texts. Our approach is centered around the linguistic and conceptual quality of various forms of evidence underlying the generation and refinement of concept hypotheses. Based on a terminological (meta)reasoning platform, hypotheses are continuously annotated by a stream of linguistic and conceptual evidence, preferentially ranked and, finally, selected according to their overall credibility. We discuss the results of an empirical evaluation study, concentrating on the system's learning rate and learning accuracy.\",\n \"Title: Semantic Annotations for Biology: a Corpus Development Initiative at the Jena University Language & Information Engineering (JULIE) Lab\\nYear: 2008\\n\\nAbstract: We provide an overview of corpus building efforts at the Jena University Language & Information Engineering (JULIE) Lab, which are focused on life science documents. Special emphasis is laid on semantic annotations in terms of a large amount of biomedical named entities (almost 100 entity types), semantic relations, as well as discourse phenomena, reference relations in particular.\",\n \"Title: Interlingual Indexing across Different Languages\\nYear: 2004\\n\\nAbstract: We present two methods for automatic indexing, which are based on an interlingual layer of content description. In the first approach, we acquire indexing patterns from English documents by statistically relating interlingual representations of English documents (based on text token bigrams) to their associated index terms. Given such indexing patterns, we then induce the associated index terms when the same interlingual representations turn up for documents of other natural languages (viz. German and Portuguese). Hence, we 'learn' from the past English indexing experience and transfer it in an unsupervised way to non-English languages, without ever having seen any concrete indexing data for languages other than English. In the second approach, documents from the three different languages are heuristically matched with a sophisticated medical thesaurus (the English MESH) after both, documents and the thesaurus, have been transformed into the interlingua. The combination of the statistical and heuristical method in a fully automated indexing system achieves 56% to 68% of the human indexing performance for each of the three languages.\",\n \"Title: Part-Whole Reasoning in Medical Knowledge Bases Using Description Logics\\nYear: 1998\\n\\nAbstract: The development of powerful, ubiquitous and comprehensive medical ontologies that support formal reasoning on a large scale\\n is one of the key requirements for clinical computing. Taxonomic medical knowledge, a major portion of these ontologies, is\\n fundamentally characterized by is—a and part-whole relationships between concepts. While reasoning in generalization hierarchies\\n is a well-understood process, no fully conclusive mechanism yet exists for part-whole reasoning. We here propose a new representation\\n construct for part-whole relations based on the formal framework of description logics, i.e. the well-known concept language\\n ALC, and show how part-whole reasoning can naturally be emulated via classification-based reasoning without extending the expressiveness\\n of the underlying terminological system.\",\n \"Title: High-performance tagging on medical texts\\nYear: 2004\\n\\nAbstract: We ran both Brill's rule-based tagger and TNT, a statistical tagger, with a default German newspaper-language model on a medical text corpus. Supplied with limited lexicon resources, TNT outperforms the Brill tagger with state-of-the-art performance figures (close to 97% accuracy). We then trained TNT on a large annotated medical text corpus, with a slightly extended tagset that captures certain medical language particularities, and achieved 98% tagging accuracy. Hence, statistical off-the-shelf POS taggers cannot only be immediately reused for medical NLP, but they also -- when trained on medical corpora -- achieve a higher performance level than for the newspaper genre.\",\n \"Title: A Search Engine for Morphologically Complex Languages\\nYear: 2001\\n\\nAbstract: Document retrieval on natural languages with a rich morphology -- particularly in terms of derivation and (single-word) composition -- suffers from serious performance degradation with the direct query-term-to-text-word matching paradigm that underlies the vast majority of current search engines. We propose an alternative approach in which morphologically complex word forms, which appear in the query as well as in the documents, are segmented into relevant subwords (such as stems, named entities, acronyms) and are subsequently submitted to the matching procedure. We evaluate our approach with the Alta Vista驴 Search Engine on a large medical document collection.\",\n \"Title: A text understander that learns\\nYear: 1998\\n\\nAbstract: We introduce an approach to the automatic acquisition of new concepts from natural language texts which is tightly integrated with the underlying text understanding process. The learning model is centered around the 'quality' of different forms of linguistic and conceptual evidence which underlies the incremental generation and refinement of alternative concept hypotheses, each one capturing a different conceptual reading for an unknown lexical item.\",\n \"Title: Biomedical text retrieval in languages with a complex morphology\\nYear: 2002\\n\\nAbstract: Document retrieval in languages with a rich and complex morphology - particularly in terms of derivation and (single-word) composition - suffers from serious performance degradation with the stemming-only query-term-to-text-word matching paradigm. We propose an alternative approach in which morphologically complex word forms are segmented into relevant subwords (such as stems, named entities, acronyms), and subwords constitute the basic unit for indexing and retrieval. We evaluate our approach on a large biomedical document collection.\",\n \"Title: Learning from texts - a terminological metareasoning perspective\\nYear: 1995\\n\\nAbstract: We introduce a methodology for concept learning from texts that relies upon second-order reasoning about statements expressed in a (first-order) terminological representation language. This metareasoning approach allows for quality-based evaluation and selection of alternative concept hypotheses.\",\n \"Title: Entity Recognition In Parallel Multi-Lingual Biomedical Corpora: The Clef-Er Laboratory Overview\\nYear: 2013\\n\\nAbstract: The identification and normalisation of biomedical entities from the scientific literature has a long tradition and a number of challenges have contributed to the development of reliable solutions. Increasingly patient records are processed to align their content with other biomedical data resources, but this approach requires analysing documents in different languages across Europe [1,2].The CLEF-ER challenge has been organized by the Mantra project partners to improve entity recognition (ER) in multilingual documents. Several corpora in different languages, i.e. Medline titles, EMEA documents and patent claims, have been prepared to enable ER in parallel documents. The participants have been ask to annotate entity mentions with concept unique identifiers (CUIs) in the documents of their preferred non-English language.The evaluation determines the number of correctly identified entity mentions against a silver standard (Task A) and the performance measures for the identification of CUIs in the non-English corpora. The participants could make use of the prepared terminological resources for entity normalisation and of the English silver standard corpora (SSCs) as input for concept candidates in the non-English documents.The participants used different approaches including translation techniques and word or phrase alignments apart from lexical lookup and other text mining techniques. The performances for task A and B was lower for the patent corpus in comparison to Medline titles and EMEA documents. In the patent documents, chemical entities were identified at higher performance, whereas the other two document types cover a higher portion of medical terms. The number of novel terms provided from all corpora is currently under investigation.Altogether, the CLEF-ER challenge demonstrates the performances of annotation solutions in different languages against an SSC.\",\n \"Title: CALBC: Releasing the Final Corpora.\\nYear: 2012\\n\\nAbstract: A number of gold standard corpora for named entity recognition are available to the public. However, the existing gold standard corpora are limited in size and semantic entity types. These usually lead to implementation of trained solutions (1) for a limited number of semantic entity types and (2) lacking in generalization capability. In order to overcome these problems, the CALBC project has aimed to automatically generate large scale corpora annotated with multiple semantic entity types in a community-wide manner based on the consensus of different named entity solutions. The generated corpus is called the silver standard corpus since the corpus generation process does not involve any manual curation. In this publication, we announce the release of the final CALBC corpora which include the silver standard corpus in different versions and several gold standard corpora for the further usage of the biomedical text mining community. The gold standard corpora are utilised to benchmark the methods used in the silver standard corpora generation process and released in a shared format. All the corpora are released in a shared format and accessible at www.calbc.eu.\",\n \"Title: Crossing Languages in Text Retrieval via an Interlingua\\nYear: 2004\\n\\nAbstract: We introduce an interlingua-based approach to cross-language information retrieval, in which queries, as well as documents, are mapped onto a language-independent concept layer on which retrieval operations are performed. This approach is contrasted with one which directly translates non-English queries (German and Portuguese, in our experiments) to English ones which, subsequently, are processed on English documents. We report on the empirical evaluation of both approaches on a large medical document collection (the OHSUMED corpus).\",\n \"Title: Learning Grammatical and Taxonomic Knowledge in Tandem\\nYear: 2001\\n\\nAbstract: Abstract. We introduce a methodology for automating the maintenance of domain-specific concept taxonomies and grammatical class hierarchies simultaneously, based on incremental natural language text understanding. The acquisition pro-cess is centered around the linguistic and conceptual `quality'of various forms of evidence underlying the generation, assessment and on-going refinement of lex-ical and concept hypotheses. On the basis of the quality of evidence, hypotheses are ranked according to plausibility, and the most reasonable ones are selected for assimilation into the given lexical class hierarchy and domain taxonomy.\",\n \"Title: Turning Lead into Gold? Feeding a Formal Knowledge Base with Informal Conceptual Knowledge\\nYear: 2002\\n\\nAbstract: We describe an ontology engineering methodology by which conceptual knowledge is extracted from an informal medical thesaurus (UMLS) and automatically converted into a formal description logics system. Our approach consists of four steps: concept definitions are automatically generated from the UMLS source, integrity checking of taxonomic and partonomic hierarchies is performed by the terminological classifier, cycles and inconsistencies are eliminated, and incremental refinement of the evolving knowledge base is performed by a domain expert. We report on experiments with a knowledge base composed of 164,000 concepts and 76,000 relations.\",\n \"Title: High-performance gene name normalization with GeNo.\\nYear: 2009\\n\\nAbstract: The recognition and normalization of textual mentions of gene and protein names is both particularly important and challenging. Its importance lies in the fact that they constitute the crucial conceptual entities in biomedicine. Their recognition and normalization remains a challenging task because of widespread gene name ambiguities within species, across species, with common English words and with medical sublanguage terms.We present GeNo, a highly competitive system for gene name normalization, which obtains an F-measure performance of 86.4% (precision: 87.8%, recall: 85.0%) on the BioCreAtIvE-II test set, thus being on a par with the best system on that task. Our system tackles the complex gene normalization problem by employing a carefully crafted suite of symbolic and statistical methods, and by fully relying on publicly available software and data resources, including extensive background knowledge based on semantic profiling. A major goal of our work is to present GeNo's architecture in a lucid and perspicuous way to pave the way to full reproducibility of our results.GeNo, including its underlying resources, will be available from www.julielab.de. It is also currently deployed in the Semedico search engine at www.semedico.org.\",\n \"Title: An annotation type system for a data-driven NLP pipeline\\nYear: 2007\\n\\nAbstract: We introduce an annotation type system for a data-driven NLP core system. The specifications cover formal document structure and document meta information, as well as the linguistic levels of morphology, syntax and semantics. The type system is embedded in the framework of the Unstructured Information Management Architecture (UIMA).\",\n \"Title: Large-scale evaluation of a medical cross-language information retrieval system.\\nYear: 2007\\n\\nAbstract: We propose an approach to multilingual medical document retrieval in which complex word forms are segmented according to medically relevant morpho-semantic criteria. At its core lies a multilingual dictionary, it? which entries are equivalence classes of subwords, i.e. semantically minimal units. Using two different standard test collections for the medical domain, we evaluate our approach for six languages covered by our system. http://ebooks.iospress.nl/volume/medinfo-2007\",\n \"Title: A Quality-Based Terminological Reasoning Model for Text Knowledge Acquisition\\nYear: 1996\\n\\nAbstract: We introduce a methodology for knowledge acquisition and concept learning from texts that relies upon a quality-based model of terminological reasoning. Concept hypotheses which have been derived in the course of the text understanding process are assigned specific quality labels (indicating their significance, reliability, strength). Quality assessment of these hypotheses accounts for conceptual criteria referring to their given knowledge base context as well as linguistic indicators (grammatical constructions, discourse patterns), which led to their generation. We advocate a metareasoning approach which allows for the quality-based evaluation and a bootstrapping-style selection of alternative concept hypotheses as text understanding incrementally proceeds.\",\n \"Title: Restricted parallelism in object-oriented lexical parsing\\nYear: 1996\\n\\nAbstract: We present an approach to parallel natural language parsing which is based on a concurrent, object-oriented model of computation. A depth-first, yet incomplete parsing algorithm for a dependency grammar is specified and several restrictions on the degree of its parallelization are discussed.\",\n \"Title: Disclose Models, Hide the Data - How to Make Use of Confidential Corpora without Seeing Sensitive Raw Data.\\nYear: 2014\\n\\nAbstract: Confidential corpora from the medical, enterprise, security or intelligence domains often contain sensitive raw data which lead to severe restrictions as far as the public accessibility and distribution of such language resources are concerned. The enforcement of strict mechanisms of data protection consitutes a serious barrier for progress in language technology (products) in such domains, since these data are extremely rare or even unavailable for scientists and developers not directly involved in the creation and maintenance of such resources. In order to by-pass this problem, we here propose to distribute trained language models which were derived from such resources as a substitute for the original confidential raw data which remain hidden to the outside world. As an example, we exploit the access-protected German-language medical FRAMED corpus from which we generate and distribute models for sentence splitting, tokenization and POS tagging based on software taken from OPENNLP, NLTK and JCORE, our own UIMA-based text analytics pipeline.\",\n \"Title: Towards Very Large Terminological Knowledge Bases: A Case Study from Medicine\\nYear: 2000\\n\\nAbstract: We describe an ontology engineering methodology by which conceptual knowledge is extracted from an informal medical thesaurus (UMLS) and automatically converted into a formally sound description logics system. Our approach consists of four steps: concept definitions are automatically generated from the UMLS source, integrity checking of taxonomic and partonomic hierarchies is performed by the terminological classifier, cycles and inconsistencies are eliminated, and incremental refinement of the evolving knowledge base is performed by a domain expert. We report on knowledge engineering experiments with a terminological knowledge base composed of 164,000 concepts and 76,000 relations.\",\n \"Title: Tracking the Evolution of Concepts in Dynamic Worlds\\nYear: 1994\\n\\nAbstract: Technical domains are affected by continuous change as a reflection of technological progress. Correspondingly, concept descriptions of terminological knowledge bases for such domains must be adjusted to these dynamics. In this paper we propose a concept versioning methodology which accounts for the evolutionary adaptation of already established concepts according to the gradual shaping of new standards. The approach is based on the provision of a qualitative progress model for the given domain, measures for the prediction and evaluation of the significance of progress, and a representation and update scheme for concept version management.\",\n \"Title: Mereological semantics for bio-ontologies\\nYear: 2004\\n\\nAbstract: Biomedical ontologies are typically structured in a biaxial way, reflecting both a taxonomic and a mereological order. Common examples such as the Gene Ontology and the Unified Medical Language System (UMLS) excel in terms of coverage but lack an adequate semantics of the mereological relations they incorporate. This shortcoming is particularly evident as far as the (non-)mandatory existence of parts for their wholes is cOllcerned, on the one hand, and the propagation of properties acmss part-whole hierarchies, on the other hand. We provide a formal specification of the semantic foundations of mereology in the biomedical domain that is closely linked to the paradigm of description logics. In essence, we here propose to emulate mereological reasoning by taxonomic reasoning. In an attempt to capture much of the shared intuition underlying merelogical reasoning in the biomedical domain, we distinguish for each mereologically relevant concept four different classes of parts and wholes which allow for the expression of five different propagation patterns.\",\n \"Title: Timed annotations: enhancing Muc7 metadata by the time it takes to annotate named entities\\nYear: 2009\\n\\nAbstract: We report on the re-annotation of selected types of named entities from the Muc7 corpus where our focus lies on recording the time it takes to annotate these entities given two basic annotation units -- sentences vs. complex noun phrases. Such information may be helpful to lay the empirical foundations for the development of cost measures for annotation processes based on the investment in time for decision-making per entity mention.\",\n \"Title: The CALBC Silver Standard Corpus for Biomedical Named Entities - A Study in Harmonizing the Contributions from Four Independent Named Entity Taggers\\nYear: 2010\\n\\nAbstract: The production of gold standard corpora is time-consuming and costly. We propose an alternative: the, silver standard corpus' (SSC), a corpus that has been generated by the harmonisation of the annotations that have been delivered from a selection of annotation systems. The systems have to share the type system for the annotations and the harmonisation solution has use a suitable similarity measure for the pair-wise comparison of the annotations. The annotation systems have been evaluated against the harmonised set (630.324 sentences, 15,956,841 tokens). We can demonstrate that the annotation of proteins and genes shows higher diversity across all used annotation solutions leading to a lower agreement against the harmonised set in comparison to the annotations of diseases and species. An analysis of the most frequent annotations from all systems shows that a high agreement amongst systems leads to the selection of terms that are suitable to be kept in the harmonised set. This is the first large-scale approach to generate an annotated corpus from automated annotation systems. Further research is required to understand, how the annotations from different systems have to be combined to produce the best annotation result for a harmonised corpus.\",\n \"Title: On the Formal Distinction between Literal and Figurative Language\\nYear: 1999\\n\\nAbstract: The distinction between literal and figurative language (metonymies, metaphors, etc.) is often not made formally explicit, or, if formal criteria exist, insufficient. This poses problems for an adequate computational treatment of these phenomena. The basic criterion for delineating literal from figurative speech we propose is centered around the notion of categorization conflicts that follow from the context of the utterance. In addition, we consider the problem of granularity, which is posed by the dependence of our approach on the underlying ontology.\",\n \"Title: Concept Versioning: A Methodology for Tracking Evolutionary Concept Drift in Dynamic Concept Systems\\nYear: 1994\\n\\nAbstract: . Technical domains are affected by continuous change as a reflectionof technological progress. Correspondingly, concept descriptions of terminologicalknowledge bases for such domains must be adjusted to thesedynamics. In this paper we propose a concept versioning methodology whichaccounts for the evolutionary adaptation of already established concepts accordingto the gradual shaping of new standards. The approach is based on theprovision of qualitative progress models for the given...\",\n \"Title: Bootstrapping dictionaries for cross-language information retrieval\\nYear: 2005\\n\\nAbstract: The bottleneck for dictionary-based cross-language information retrieval is the lack of comprehensive dictionaries, in particular for many different languages. We here introduce a methodology by which multilingual dictionaries (for Spanish and Swedish) emerge automatically from simple seed lexicons. These seed lexicons are automatically generated, by cognate mapping, from (previously manually constructed) Portuguese and German as well as English sources. Lexical and semantic hypotheses are then validated and new ones iteratively generated by making use of co-occurrence patterns of hypothesized translation synonyms in parallel corpora. We evaluate these newly derived dictionaries on a large medical document collection within a cross-language retrieval setting.\",\n \"Title: Text Knowledge Engineering by Qualitative Terminological Learning\\nYear: 1997\\n\\nAbstract: We propose a methodology for enhancing domain knowledge bases through natural language text understanding. The acquisition of previously unknown concepts is based on the assessment of the quality of linguistic and conceptual evidence underlying the generation and refinement of concept hypotheses. Text understanding and concept learning are both grounded on a terminological knowledge representation and reasoning framework.\",\n \"Title: On \\\"deep\\\" knowledge extraction from documents\\nYear: 2000\\n\\nAbstract: SYNDIKATE comprises a family of natural language understanding systems for automatically acquiring knowl- edge from real-world texts (e.g., information technology test reports, medical finding reports), and for transferring their content to formal representation structures which constitute a corresponding text knowledge base. We present a general system architecture which integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming cohesive texts. Properly accounting for text cohesion phenomena is a prerequisite for the soundness and validity of the generated text representation structures. It is also crucial for any information system application making use of automatically generated text knowledge bases in a reliable way.\",\n \"Title: Context-Based Ambiguity Management for Natural Language Processing\\nYear: 2001\\n\\nAbstract: We introduce a formal context mechanism, embedded into a description logics framework, which allows to uniformly represent and manage different kinds of natural language ambiguities as they occur in the course of text understanding. Multiple lexical, syntactic and semantic interpretations are separated by assigning each of these alternatives a single context space for local reasoning and incremental disambiguation.\",\n \"Title: Assessment of NER solutions against the first and second CALBC Silver Standard Corpus.\\nYear: 2011\\n\\nAbstract: Competitions in text mining have been used to measure the performance of automatic text processing solutions against a manually annotated gold standard corpus (GSC). The preparation of the GSC is time-consuming and costly and the final corpus consists at the most of a few thousand documents annotated with a limited set of semantic groups. To overcome these shortcomings, the CALBC project partners (PPs) have produced a large-scale annotated biomedical corpus with four different semantic groups through the harmonisation of annotations from automatic text mining solutions, the first version of the Silver Standard Corpus (SSC-I). The four semantic groups are chemical entities and drugs (CHED), genes and proteins (PRGE), diseases and disorders (DISO) and species (SPE). This corpus has been used for the First CALBC Challenge asking the participants to annotate the corpus with their text processing solutions.All four PPs from the CALBC project and in addition, 12 challenge participants (CPs) contributed annotated data sets for an evaluation against the SSC-I. CPs could ignore the training data and deliver the annotations from their genuine annotation system, or could train a machine-learning approach on the provided pre-annotated data. In general, the performances of the annotation solutions were lower for entities from the categories CHED and PRGE in comparison to the identification of entities categorized as DISO and SPE. The best performance over all semantic groups were achieved from two annotation solutions that have been trained on the SSC-I.The data sets from participants were used to generate the harmonised Silver Standard Corpus II (SSC-II), if the participant did not make use of the annotated data set from the SSC-I for training purposes. The performances of the participants' solutions were again measured against the SSC-II. The performances of the annotation solutions showed again better results for DISO and SPE in comparison to CHED and PRGE.The SSC-I delivers a large set of annotations (1,121,705) for a large number of documents (100,000 Medline abstracts). The annotations cover four different semantic groups and are sufficiently homogeneous to be reproduced with a trained classifier leading to an average F-measure of 85%. Benchmarking the annotation solutions against the SSC-II leads to better performance for the CPs' annotation solutions in comparison to the SSC-I.\",\n \"Title: Finding new terminology in very large corpora\\nYear: 2005\\n\\nAbstract: Most technical and scientific terms are comprised of complex, multi-word noun phrases but certainly not all noun phrases are technical or scientific terms. The distinction of specific terminology from common non-specific noun phrases can be based on the observation that terms reveal a much lesser degree of distributional variation than non-specific noun phrases. We formalize the limited paradigmatic modifiability of terms and, subsequently, test the corresponding algorithm on bigram, trigram and quadgram noun phrases extracted from a 104-million-word biomedical text corpus. Using an already existing and community-wide curated biomedical terminology as an evaluation gold standard, we show that our algorithm significantly outperforms standard term identification measures and, therefore, qualifies as a high-performant building block for any terminology identification system. We also provide empirical evidence that the superiority of our approach, beyond a 10-million-word threshold, is essentially domain- and corpus-size-independent.\",\n \"Title: BioTop: An upper domain ontology for the life sciences: A description of its current structure, contents and interfaces to OBO ontologies\\nYear: 2008\\n\\nAbstract: In the life sciences, there is an ample need for semantic interoperability of d ata. Thus shared vocabularies are needed for consistently expressing meta data in terms of semantic annotations as well as for querying bibliographic information systems. In the past years, lots of highly specialized, yet also fragmented terminolog ies have evolved. However, they lack principled forms of conceptual interlinkage. In order to provide an ontological basis for a seamless integration of such isolated parts of biological knowledge, we here introduce BIOTOP, an upper domain ontology for molecular biology. We describe its structure and contents, as well as its current interfaces to a selected set of OBO ontologies, which contain more detailed terminological knowledge about specific areas of molecular biology, e.g., cell types, molecular functio ns, biological processes, and chemical compounds.\",\n \"Title: Text Understanding for Knowledge Base Generation in the SYNDIKATE System\\nYear: 1999\\n\\nAbstract: SynDiKATe is a system for automatically acquiring knowledge from real-world texts and transferring it to formal representation structures which constitute a text knowledge base. We present a system architecture which integrates requirements from the analysis of single sentences. as well as those of referentially linked sentences forming cohesive texts. Properly accounting for text cohesion phenomena is a prerequisite for the completeness and validity of the generated text representation structures and. therefore. also crucial for any information system application making use of automatically generated text knowledge bases in a reliable way, e.g., by inferentially supported fact retrieval.\",\n \"Title: U-Compare bio-event meta-service: compatible BioNLP event extraction services.\\nYear: 2011\\n\\nAbstract: Background Bio-molecular event extraction from literature is recognized as an important task of bio text mining and, as such, many relevant systems have been developed and made available during the last decade. While such systems provide useful services individually, there is a need for a meta-service to enable comparison and ensemble of such services, offering optimal solutions for various purposes. Results We have integrated nine event extraction systems in the U-Compare framework, making them inter-compatible and interoperable with other U-Compare components. The U-Compare event meta-service provides various meta-level features for comparison and ensemble of multiple event extraction systems. Experimental results show that the performance improvements achieved by the ensemble are significant. ConclusionsWhile individual event extraction systems themselves provide useful features for bio text mining, the U-Compare meta-service is expected to improve the accessibility to the individual systems, and to enable meta-level uses over multiple event extraction systems such as comparison and ensemble. © 2011 Kano et al; licensee BioMed Central Ltd.\",\n \"Title: \\\"Tall\\\", \\\"good\\\", \\\"high\\\": compared to what?\\nYear: 1997\\n\\nAbstract: In order to rank the performance of machine learning algorithms, many researchers conduct experiments on benchmark data sets. Since most learning algorithms have domain-specific parameters, it is a popular custom to adapt these parameters to obtain a ...\",\n \"Title: Subword Retrieval on Biomedical Documents\\nYear: 2003\\n\\nAbstract: Document retrieval in languages with a rich and com- plex morphology - particularly in terms of derivation and (single-word) composition - suffers from serious perfor- mance degradation with the stemming-only query-term-to- text-word matching paradigm. We propose an alternative ap- proach in which morphologically complex word forms are segmented into relevant subwords (such as stems, prefixes, suffixes), and subwords constitute the basic unit for index- ing and retrieval. We evaluate our approach on a biomedical document collection.\",\n \"Title: The Challenges of Automatic Summarization\\nYear: 2000\\n\\nAbstract: Summarization--the art of abstracting key con-tent from one or more information sources--has become an integral part of everyday life. People keep abreast of world affairs by listening to news bites. They base investment decisions on stock market updates. They go to movies largely on the basis of reviews. With summaries, they can make effective decisions in less time.Although summarizing tools are available, with the increasing volume of online information, it is becoming harder to generate meaningful and timely summaries. Researchers are investigating tools and methods that automatically extract or abstract content from information sources.The authors describe how these data summarization methods fall into two categories. Knowledge-poor approaches rely on not having to add new rules for each new application domain or language. Knowledge-rich approaches assume that if you grasp the meaning of the text, you can reduce it more effectively, thus yielding a better summary. They rely on a size-able knowledge base of rules, which must be acquired, maintained, and then adapted to new applications and languages.The authors predict that summarization tools will be key in conquering the vast information universes ahead.\",\n \"Title: Terminological Meta-Reasoning by Reification and Multiple Contexts\\nYear: 1995\\n\\nAbstract: We introduce a model and a system architecture for second-order reasoning about statements expressed in a (first-order) terminological representation language. This meta-reasoning approach is based on the reification of first-order propositions and the mediation between second- and first-order expressions via translation rules operating on multiple contexts. An application to a concept learning task in a text understanding environment shows how different degrees of credibility can be assigned to alternative concept hypotheses on the basis of such a schema for terminological meta-reasoning.\",\n \"Title: Gene Regulation Ontology (GRO): Design Principles and Use Cases.\\nYear: 2008\\n\\nAbstract: The Gene Regulation Ontology (GRO) is designed as a novel approach to model complex events that are part of the gene regulatory processes. We introduce the design requirements for such a conceptual model and discuss terminological resources suitable to base its construction on. The ontology defines gene regulation events in terms of ontological classes and imposes constraints on them by specifying the participants involved. The logical structure of the ontology is intended to meet the needs of advanced information extraction and text mining systems which target the identification of event representations in scientific literature. The GRO has just been submitted to the OBO library and is currently under review. It is available at http:// www.ebi.ac.uk/Rebholz-srv/GRO/GRO.html\",\n \"Title: Generating Semantics for the Life Sciences via Text Analytics\\nYear: 2011\\n\\nAbstract: The life sciences have a strong need for carefully curated, semantically rich fact repositories. Knowledge harvesting from unstructured textual sources is currently performed by highly skilled curators who manually feed semantics into such databases as a result of deep understanding of the documents chosen to populate such repositories. As this is a slow and costly process, we here advocate an automatic approach to the generation of database contents which is based on JREX, a high performance relation extraction system. As a real-life example, we target REGULONDB, the world's largest manually curated reference database for the transcriptional regulation network of E. coli. We investigate in our study the performance of automatic knowledge capture from various literature sources, such as PUBMED abstracts and associated full text articles. Our results show that we can, indeed, automatically re-create a considerable portion of the REGULONDB database by processing the relevant literature sources. Hence, this approach might help curators widen the knowledge acquisition bottleneck in this field.\",\n \"Title: Parthood as Spatial Inclusion - Evidence from biomedical Conceptualizations\\nYear: 2004\\n\\nAbstract: Modeling complex compositional objects in the field of the life sciences requires to solve intricate ontological problems, especially those related to parts of a whole, space and location. For concrete physical entities, we here stipulate that the distinction between parthood and location can be abstracted away. We outline a logic- based reasoning framework in which taxonomic sub- sumption not only takes care of the propagation of roles across taxonomic hierarchies, but also across nontaxo- nomic, i.e., partonomic or spatial, ones.\",\n \"Title: Functional centering: grounding referential coherence in information structure\\nYear: 1999\\n\\nAbstract: Considering empirical evidence from a free-word-order language (German) we propose a revision of the principles guiding the ordering of discourse entities in the forward-looking center list within the centering model. We claim that grammatical role criteria should be replaced by criteria that reflect the functional information structure of the utterances. These new criteria are based on the distinction between hearer-old and hearer-new discourse entities. We demonstrate that such a functional model of centering can be successfully applied to the analysis of several forms of referential text phenomena, viz. pronominal, nominal, and functional anaphora. Our methodological and empirical claims are substantiated by two evaluation studies. In the first one, we compare success rates for the resolution of pronominal anaphora that result from a grammatical-role-driven centering algorithm and from a functional centering algorithm. The second study deals with a new cost-based evaluation methodology for the assessment of centering data, one which can be directly derived from and justified by the cognitive load premises of the centering model.\",\n \"Title: Towards automatic pathway generation from biological full-text publications\\nYear: 2011\\n\\nAbstract: We introduce an approach to the automatic generation of biological pathway diagrams from scientific literature. It is composed of the automatic extraction of single interaction relations which are typically found in the full text (rather than the abstract) of a scientific publication, and their subsequent integration into a complex pathway diagram. Our focus is here on relation extraction from full-text documents. We compare the performance of automatic full-text extraction procedures with a manually generated gold standard in order to validate the extracted data which serve as input for the pathway integration procedure.\",\n \"Title: Quality-Based Terminological Reasoning for Concept Learning\\nYear: 1995\\n\\nAbstract: We introduce a qualitative methodology for concept learning from texts that relies upon second-order reasoning about statements expressed in a (first-order) terminological representation language. This meta-reasoning approach allows for quality-based evaluation and selection among alternative concept hypotheses.\",\n \"Title: Making understanders out of parsers: Semantically driven parsing as a key concept for realistic text understanding applications.\\nYear: 1989\\n\\nAbstract: Semantically driven natural language parsers have found wide-spread application as a text processing methodology for knowledge-based information retrieval systems. It is argued that this parsing technique particularly corresponds to the requirements inherent to large-scale text analysis. Unfortunately, this approach suffers from several shortcomings which demand a thorough reformulation of its paradigm. Incorporating principles from conceptual analysis and word expert parsing in a model of lexically distributed text parsing, the focus of the modifications proposed in this article, is on a clean declarative separation of linguistio and other knowledge representation levels, abstraction mechanisms leading to a small collection of specification primitives for the parser, and an attempt to incorporate linguistic generalizations and modularization principles into the design of a semantic text grammar. A sample parse illustrates the operation and linguistic coverage of a lexically distributed text parser based on these theoretical considerations with respect to the semantic analysis of noun groups, simple assertional sentences, nominal anaphora, and textual ellipsis. © 1989 Wiley Periodicals, Inc.\",\n \"Title: Trading off Completeness for Efficiency - The ParseTalk Performance Grammar Approach to Real-World Text Parsing\\nYear: 1996\\n\\nAbstract: We argue for a performance-based design of natural language grammars and\\ntheir associated parsers in order to meet the constraints posed by real-world\\nnatural language understanding. This approach incorporates declarative and\\nprocedural knowledge about language and language use within an object-oriented\\nspecification framework. We discuss several message passing protocols for\\nreal-world text parsing and provide reasons for sacrificing completeness of the\\nparse in favor of efficiency.\",\n \"Title: A formal approach to the semantics of a frame data model\\nYear: 1983\\n\\nAbstract: Standard knowledge representation languages are seriously lacking an explicit formal semantic specification. This may cause considerable trouble when applied to large amounts of rapidly changing data. Based on an abstract data type view of knowledge representation languages a formal definition of a frame data model is presented in terms of a denotational semantics approach using a subset of META-IV. After introducing some basic concepts of the model several semantic integrity constraints are outlined which ultimately lead to the formulation of a set of operations in the frame data model.\",\n \"Title: Die Verdichtung textuellen Wissens zu Information - Vom Wandel methodischer Paradigmen beim automatischen Abstracting\\nYear: 2004\\n\\nAbstract: The automatic summarization of documents (technical magazine articles) has been a central research theme in the information science division at Constance University. We give an overview of the context in which this topic was dealt with, the major methodological results that were achieved, and the text condensation system TOPIC that was built on these principles. We then survey current trends in the field of automatic text summarization which reveal the renaissance of the 'abstracting-as-extracting' approach. We relate this observation to the fundamental paradigmatic shift from natural language understanding to human language technology and discuss its implications for information science.\",\n \"Title: Textual expertise in word experts: an approach to text parsing based on topic/comment monitoring\\nYear: 1984\\n\\nAbstract: In this paper prototype versions of two word experts for text analysis are dealt with which demonstrate that word experts are a feasible tool for parsing texts on the level of text cohesion as well as text coherence. The analysis is based on two major knowledge sources: context information is modelled in terms of a frame knowledge base, while the co-text keeps record of the linear sequencing of text analysis. The result of text parsing consists of a text graph reflecting the thematic organization of topics in a text.\",\n \"Title: How feasible and robust is the automatic extraction of gene regulation events?: a cross-method evaluation under lab and real-life conditions\\nYear: 2009\\n\\nAbstract: We explore a rule system and a machine learning (ML) approach to automatically harvest information on gene regulation events (GREs) from biological documents in two different evaluation scenarios -- one uses self-supplied corpora in a clean lab setting, while the other incorporates a standard reference database of curated GREs from RegulonDB, real-life data generated independently from our work. In the lab condition, we test how feasible the automatic extraction of GREs really is and achieve F-scores, under different, not directly comparable test conditions though, for the rule and the ML systems which amount to 34% and 44%, respectively. In the RegulonDB condition, we investigate how robust both methodologies are by comparing them with this routinely used database. Here, the best F-scores for the rule and the ML systems amount to 34% and 19%, respectively.\",\n \"Title: Concurrent Lexicalized Dependency Parsing: A Behavioral View on ParseTalk Events.\\nYear: 1994\\n\\nAbstract: The behavioral specification of an object-oriented grammar model is considered. The model is based on full lexicalization, head-orientation via valency constraints and dependency relations, inheritance as a means for non-redundant lexicon specification, and concurrency of computation. The computation model relies upon the actors paradigm, with concurrency entering through asynchronous message passing between actors. In particular, we here elaborate on principles of how the global behavior of a lexically distributed grammar and its corresponding parser can be specified in terms of event type networks and event networks, resp.\",\n \"Title: Towards new information resources for public health--from WordNet to MedicalWordNet.\\nYear: 2006\\n\\nAbstract: In the last two decades, WordNet has evolved as the most comprehensive computational lexicon of general English. In this article, we discuss its potential for supporting the creation of an entirely new kind of information resource for public health, viz. MedicalWordNet. This resource is not to be conceived merely as a lexical extension of the original WordNet to medical terminology; indeed, there is already a considerable degree of overlap between WordNet and the vocabulary of medicine. Instead, we propose a new type of repository, consisting of three large collections of (1) medically relevant word forms, structured along the lines of the existing Princeton WordNet; (2) medically validated propositions, referred to here as medical facts, which will constitute what we shall call MedicalFactNet; and (3) propositions reflecting laypersons' medical beliefs, which will constitute what we shall call the MedicalBeliefNet. We introduce a methodology for setting up the MedicalWordNet. We then turn to the discussion of research challenges that have to be met to build this new type of information resource. We build a database of sentences relevant to the medical domain. The sentences are generated from WordNet via its relations as well as from medical statements broken down into elementary propositions. Two subcorpora of sentences are distinguished, MedicalBeliefNet and MedicalFactNet. The former is rated for assent by laypersons; the latter for correctness by medical experts. The sentence corpora will be valuable for a variety of applications in information retrieval as well as in research in linguistics and psychology with respect to the study of expert and non-expert beliefs and their linguistic expressions. Our work has to meet several considerable challenges. These include accounting for the distinction between medical experts and laypersons, the social issues of expert-layperson communication in different media, the linguistic aspects of encoding medical knowledge, and the reliability, volume, and emergence of medical knowledge. The work described here has been tested in a small pilot experiment and awaits large-scale implementation.\",\n \"Title: Annotation Time Stamps - Temporal Metadata from the Linguistic Annotation Process\\nYear: 2010\\n\\nAbstract: We describe the re-annotation of selected types of named entities (persons, organizations, locations) from the MUC7 corpus. The focus of this annotation initiative is on recording the time needed for the linguistic process of named entity annotation. Annotation times are measured on two basic annotation units - sentences vs. complex noun phrases. We gathered evidence that decision times are non- uniformly distributed over the annotation units, while they do not substantially deviate among annotators. This data seems to support the hypothesis that annotation times very much depend on the inherent 'hardness' of each single annotation decision. We further show how such time-stamped information can be used for empirically grounded studies of selective sampling techniques, such as Active Learning. We directly compare Active Learning costs on the basis of token-based vs. time-based measurements. The data reveals that Active Learning keeps its competitive advantage over random sampling in both scenarios though the difference is less marked for the time metric than for the token metric.\",\n \"Title: A Learner with a Sense for Quality\\nYear: 1998\\n\\nAbstract: A text understanding system with learning capabilities is presented. New concepts are acquired by incorporat- ing two kinds of evidence - knowledge about linguistic constructions in which unknown lexical items occur and knowledge about structural patterns in ontologies such that new concept descriptions can be compared with prior knowledge. On the basis of the quality of evidence gath- ered this way concept hypotheses are generated, ranked according to plausibility, and the most credible ones are selected for assimilation into the domain knowledge base.\",\n \"Title: Representing Natural Kinds by Spatial Inclusion and Containment\\nYear: 2004\\n\\nAbstract: Biomedical terminologies lack a clear-cut understanding of the semantics of part-whole relations. In order to design a conceptually adequate ontology for the biomedical domain, we investigate different forms of ontological dependencies between parts and wholes. We then suggest, for natural kinds at least, a move away from modeling the relations between parts and wholes and advocate instead conceptually simpler though expressive spatial inclusion/containment relations. We outline a multi-facetted encoding pattern for biological concepts, which is centered around these relations.\",\n \"Title: Computing text constituency: an algorithmic approach to the generation of text graphs\\nYear: 1984\\n\\nAbstract: An algorithm for text summarization (automatic abstracting) is presented which constitutes the text condensation component of TOPIC, a knowledge-based text information system. Based on the results of text parsing knowledge representation structures of text segments are evaluated in order to determine dominant concepts. By means of an interpretation schema dominant concepts are related in terms of thematic units indicative of the topic(s) of the text segment under consideration. The mutual compatibility of topics of adjacent text segments is determined and corresponding text constituents are constructed. Finally, a text graph is generated linking appropriate text constituents on various levels of text constituency. Accordingly, facilities for text-oriented information retrieval will be based on the manipulation of these text graphs.\",\n \"Title: Small Is Beautiful - Compact Semantics for Medical Language Processing\\nYear: 1999\\n\\nAbstract: We introduce two abstraction mechanisms by which the process of semantic interpretation of medical documents can be simplified and optimized. One relates to the linguistic generality, the other to the inheritance-based specification of semantic rules. The proposed methodology leads to a parsimonious inventory of abstract, simple and domain-independent semantic interpretation schemata whose effectiveness has been evaluated on a medical text corpus.\",\n \"Title: Distance Constraint Arrays: A Model for Reasoning on Intervals with Qualitative and Quantitative Distances\\nYear: 1998\\n\\nAbstract: We outline a model of one-dimensional reasoning on interval relations with quantitative and qualitative distances. At the core of this model lie constraints on interval boundaries, partial ordering and subsumption relations on interval relations and interval boundary constraints, and the transformation of interval relations to interval boundary constraints and vice versa. By way of subsumption and approximation criteria on distance constraint arrays, a significant level of conceptual abstraction from the underlying interval boundary constraints is realized when new relations are inferred.\",\n \"Title: A cognitive cost model of annotations based on eye-tracking data\\nYear: 2010\\n\\nAbstract: We report on an experiment to track complex decision points in linguistic metadata annotation where the decision behavior of annotators is observed with an eye-tracking device. As experimental conditions we investigate different forms of textual context and linguistic complexity classes relative to syntax and semantics. Our data renders evidence that annotation performance depends on the semantic and syntactic complexity of the decision points and, more interestingly, indicates that full-scale context is mostly negligible - with the exception of semantic high-complexity cases. We then induce from this observational data a cognitively grounded cost model of linguistic meta-data annotations and compare it with existing non-cognitive models. Our data reveals that the cognitively founded model explains annotation costs (expressed in annotation time) more adequately than non-cognitive ones.\",\n \"Title: Concurrent, object-oriented natural language parsing: the ParseTalk model\\nYear: 1994\\n\\nAbstract: TheParseTalk model of concurrent, object-oriented natural language parsing is intro- duced. It builds upon the complete lexical distribution of grammatical knowledge and incorporates inheritance mechanisms in order to express generalizations over sets of lex- ical items. The grammar model integrates declarative well-formedness criteria con- straining linguistic relations between heads and modifiers, and procedural specifications of the communication protocol for establishing these relations. The parser's computa- tion model relies upon the actor paradigm, with concurrency entering through asynchro- nous message passing. We consider various extensions of the basic actor model as required for distributed natural language understanding and elaborate on the semantics of the actor computation model in terms of event type networks (a graph representation for actor grammar specifications) and event networks (graphs which represent the actor parser's behavior). Besides theoretical claims, we present an interactive grammar/parser workbench, a graphical development environment with various types of browsers, trac- ers, inspectors and debuggers that has been adapted to the requirements of large-scale grammar engineering in a distributed, object-oriented specification and programming framework.\",\n \"Title: Message-Passing Protocols for Real-World Parsing -- An Object-Oriented Model and its Preliminary Evaluation\\nYear: 1997\\n\\nAbstract: We argue for a performance-based design of natural language grammars and their associated parsers in order to meet the constraints imposed by real-world NLP. Our approach incorporates declarative and procedural knowledge about language and language use within an object-oriented specification framework. We discuss several message- passing protocols for parsing and provide reasons for sacrificing completeness of the parse in favor of efficiency based on a preliminary empirical evaluation.\",\n \"Title: Do users appreciate novel interface features for literature search?: a user study in the life sciences domain\\nYear: 2009\\n\\nAbstract: Faced with the challenges to design an easy-to-use, immediately comprehensible and powerful expert-user interface to search very large document collections in the life sciences, we developed several system prototypes. Their main features were faceting of the domain vocabulary for browsing and searching, flexible search-state-dependent drilling of the terminological hierarchy, dynamic query term auto-completions, and highlighting of matched terms (including synonyms and spelling variants). Under lab conditions we then evaluated these features in several task-based scenarios using camera recordings, thinking-aloud protocols and questionnaires. The results reveal that faceting and highlighting were very well received, while autocompletions seemed less important or were misconceptualized as spelling aids.\",\n \"Title: Cross-language mining for acronyms and their completions from the web\\nYear: 2005\\n\\nAbstract: We propose a method that aligns biomedical acronyms and their long-form definitions across different languages. We use a freely available search and extraction tool by which abbreviations, together with their fully expanded forms, are massively mined from the Web. In a subsequent step, language-specific variants, synonyms, and translations of the extracted acronym definitions are normalized by referring to a language-independent, shared semantic interlingua.\",\n \"Title: Are morpho-syntactic features more predictive for the resolution of noun phrase coordination ambiguity than lexico-semantic similarity scores?\\nYear: 2008\\n\\nAbstract: Coordinations in noun phrases often pose the problem that elliptified parts have to be reconstructed for proper semantic interpretation. Unfortunately, the detection of coordinated heads and identification of elliptified elements notoriously lead to ambiguous reconstruction alternatives. While linguistic intuition suggests that semantic criteria might play an important, if not superior, role in disambiguating resolution alternatives, our experiments on the reannotated WSJ part of the Penn Treebank indicate that solely morpho-syntactic criteria are more predictive than solely lexico-semantic ones. We also found that the combination of both criteria does not yield any substantial improvement.\",\n \"Title: Partonomic reasoning as taxonomic reasoning in medicine\\nYear: 1999\\n\\nAbstract: Taxonomic anatomical knowledge, a major portion of medical ontologies, is fundamentally characterized by is-a and part-whole relations between concepts. While taxonomic reasoning in generalization hierarchies is well-understood, no fully conclusive mechanism as yet exists for partonomic reasoning. We here propose a new representation construct for part-whole relations, based on the formal framework of description logics, that allows us to fully reduce partonomic reasoning to classification-based taxonomic reasoning.\",\n \"Title: Cognate mapping: a heuristic strategy for the semi-supervised acquisition of a Spanish lexicon from a Portuguese seed lexicon\\nYear: 2004\\n\\nAbstract: We deal with the automated acquisition of a Spanish medical subword lexicon from an already existing Portuguese seed lexicon. Using two non-parallel monolingual corpora we determined Spanish lexeme candidates from Portuguese seed lexicon entries by heuristic cognate mapping. We validated the emergent lexical translation hypotheses by determining the similarity of fixed-window context vectors on the basis of Portuguese and Spanish text corpora.\",\n \"Title: A denotational semantics for joining description logics and object-oriented programming\\nYear: 1997\\n\\nAbstract: . Despite their common roots, the paradigms of object-orientedprogramming and object-centered knowledge representation could not reallybe joined again. While the knowledge representation branch highlights clearsemantics and focuses on subsumption and classification-based inference algorithms,object-oriented programming shines with flexible control structuresand software-engineering issues. Bridging both worlds seems rewarding, as itwould allow for the description of complex...\",\n \"Title: Mining the pharmacogenomics literature - a survey of the state of the art.\\nYear: 2012\\n\\nAbstract: This article surveys efforts on text mining of the pharmacogenomics literature, mainly from the period 2008 to 2011. Pharmacogenomics (or pharmacogenetics) is the field that studies how human genetic variation impacts drug response. Therefore, publications span the intersection of research in genotypes, phenotypes and pharmacology, a topic that has increasingly become a focus of active research in recent years. This survey covers efforts dealing with the automatic recognition of relevant named entities (e.g. genes, gene variants and proteins, diseases and other pathological phenomena, drugs and other chemicals relevant for medical treatment), as well as various forms of relations between them. A wide range of text genres is considered, such as scientific publications (abstracts, as well as full texts), patent texts and clinical narratives. We also discuss infrastructure and resources needed for advanced text analytics, e.g. document corpora annotated with corresponding semantic metadata (gold standards and training data), biomedical terminologies and ontologies providing domain-specific background knowledge at different levels of formality and specificity, software architectures for building complex and scalable text analytics pipelines and Web services grounded to them, as well as comprehensive ways to disseminate and interact with the typically huge amounts of semiformal knowledge structures extracted by text mining tools. Finally, we consider some of the novel applications that have already been developed in the field of pharmacogenomic text mining and point out perspectives for future research.\",\n \"Title: MaHCO: an ontology of the major histocompatibility complex for immunoinformatic applications and text mining.\\nYear: 2009\\n\\nAbstract: Motivation: The high level of polymorphism associated with the major histocompatibility complex (MHC) poses a challenge to organizing associated bioinformatic data, particularly in the area of hematopoietic stem cell transplantation. Thus, this area of research has great potential to profit from the ongoing development of biomedical ontologies, which offer structure and definition to MHC-data related communication and portability issues. Results: We introduce the design considerations, methodological foundations and implementational issues underlying MaHCO, an ontology which represents the alleles and encoded molecules of the major histocompatibility complex. Importantly for human immunogenetics, it includes a detailed level of human leukocyte antigen (HLA) classification. We then present an ontology browser, search interfaces for immunogenetic fact and document retrieval, and the specification of an annotation language for semantic metadata, based on MaHCO. These use cases are intended to demonstrate the utility of ontology-driven bioinformatics in the field of immunogenetics.\",\n \"Title: Incremental Concept Evolution Based on Adaptive Feature Weighting\\nYear: 1997\\n\\nAbstract: We provide a model of concept drift by which concept classes are automatically updated according to the dynamics of change in the underlying learning environment. Instances with a well-defined degree of feature similarity are incrementally grouped together to form so-called concept versions. A comparative empirical evaluation of the versioning system VerGene, which is built on these premises, demonstrates its effectiveness for properly dealing with concept drift phenomena in medium-sized knowledge bases.\",\n \"Title: On the interaction of metonymies and anaphora\\nYear: 1997\\n\\nAbstract: From the analysis of naturally occurring texts we obtained evidence for the systematic interaction between nominal anaphora and metonymies. This leads us to postulate an integrated model incorporating both phenomena simultaneously. The consideration of discourse constraints for metonymy resolution allows us to challenge the commonly held view that the interpretation of metonymies should proceed from a literal-meaning-first approach. Thus, we argue for an equally balanced treatment of literal and figurative language use.\",\n \"Title: Exploiting parallel corpora to scale up multilingual biomedical terminologies.\\nYear: 2014\\n\\nAbstract: Creating and maintaining biomedical terminologies for multiple natural languages is a resource-expensive task, typically carried out by human domain experts. We here report on efforts to computationally support this process by treating term acquisition as a machine translation-guided classification problem capitalizing on parallel corpora. We report on experiments for French, German, Spanish and Dutch parts of a UMLS-derived terminology for which we generated 18k, 23k, 19k and 12k new terms and synonyms, respectively. Based on expert assessments of a novel German terminology segment about 80% of the newly acquired terms were judged as bio-medically reasonable and terminologically valid.\",\n \"Title: Active Learning-based corpus annotation--the PathoJen experience.\\nYear: 2012\\n\\nAbstract: We report on basic design decisions and novel annotation procedures underlying the development of PathoJen, a corpus of Medline abstracts annotated for pathological phenomena, including diseases as a proper subclass. This named entity type is known to be hard to delineate and capture by annotation guidelines. We here propose a two-category encoding schema where we distinguish short from long mention spans, the first covering standardized terminology (e.g. diseases), the latter accounting for less structured descriptive statements about norm-deviant states, as well as criteria and observations that might signal pathologies. The second design decision relates to the way annotation instances are sampled. Here we subscribe to an Active Learning-based approach which is known to save annotation costs without sacrificing annotation quality by means of a sample bias. By design, Active Learning picks up 'hard' to annotate instances for human annotators, whereas 'easier' ones are passed over to the automatic classifier whose models already incorporate and gradually improve with previous annotation experience.\",\n \"Title: Effective grading of termhood in biomedical literature.\\nYear: 2005\\n\\nAbstract: The ever-increasing amount of textual information in biomedicine calls for effective procedures for automatic terminology extraction which assist biomedical researchers and professionals in gathering and organizing terminological knowledge encoded in text documents. In this study, we propose a new, linguistically grounded measure for automatically identifying multi-word terms from the biomedical literature. Our approach is based on the limited paradigmatic modifiability of terms and is tested on bigram, trigram and quadgram noun phrases extracted from a 104-million-word text corpus comprised of Medline abstracts. Using the UMLS Metathesaurus as a gold standard, we show that our algorithm substantially outperforms the standard term identification measures and, therefore, qualifies as a high-performing building block for any biomedical terminology mining system.\",\n \"Title: An ontology for major histocompatibility complex (MHC) alleles and molecules.\\nYear: 2007\\n\\nAbstract: We present a formally coherent and consistent multi-species MHC ontology which includes all human MHC alleles and serological groups. The ontology is part of StemNet, a knowledge management system for hematopoietic stem cell transplantation with an integrated semantic search engine. The Owl-encoded MHC ontology contributes to the system in a threefold manner. First, it supports query formulation and query processing as well as mapping onto external terminological resources, second, it eases the interaction with the search engine when navigating through search results, and finally, it provides a formal language for text annotation, a methodological prerequisite for state-of-the-art natural language text processors which are increasingly based on machine learning methods and hence require annotated text corpora.\",\n \"Title: A knowledge representation view on biomedical structure and function.\\nYear: 2002\\n\\nAbstract: In biomedical ontologies, structural and functional considerations are of outstanding importance, and concepts which belong to these two categories are highly interdependent. At the representational level both axes must be clearly kept separate in order to support disciplined ontology engineering. Furthermore, the biaxial organization of physical structure (both by a taxonomic and partonomic order) entails intricate patterns of inference. We here propose a layered encoding of taxonomic, partonomic and functional aspects of biomedical concepts using description logics.\",\n \"Title: Grounding Epistemic Modality In Speakers' Judgments\\nYear: 2014\\n\\nAbstract: As an alternative to subjective and informal expert opinions, we here propose to ground natural language expressions for epistemic modality, such as \\\"possibly\\\", or \\\"might indicate\\\", in empirical data solicited from members of the speakers' community in epistemic judgment tasks. In an online questionnaire study, we asked 86 subjects to rate the degree of certainty of propositions relative to 27 linguistic expressions. Based on a thorough statistical analysis of the responses we received, we are able to quantify the likelihood as well as ranges of certainty associated with these propositions. Moreover, we generalize these data in terms of an empirically motivated, tri-partite category system of expression sets for propositional certainty which can be reused, e.g. for standardizing annotation guidelines.\",\n \"Title: Automatic lexicon acquisition for a medical cross-language information retrieval system.\\nYear: 2005\\n\\nAbstract: We present a method for the automated acquisition of a multilingual medical lexicon (for Spanish and Swedish) to be used within the framework of a medical cross-language text retrieval system. We incorporate seed lexicons and parallel corpora derived from the UMLS Metathesaurus. The seed lexicons for Spanish and Swedish are automatically generated from (previously manually constructed) Portuguese, German and English sources. Lexical and semantic hypotheses are then validated making iterative use of co-occurrence patterns of hypothesized translation synonyms in the parallel corpora.\",\n \"Title: Aligning UniProt and MeSH - a case study on human protein terms.\\nYear: 2010\\n\\nAbstract: Terminologies which lack semantic connectivity hamper the effective search in biomedical fact databases and document retrieval systems. We here focus on the integration of two such isolated resources, the term lists from the protein fact database UNIPROT and the indexing vocabulary MESH from the bibliographic database MEDLINE. The generated semantic ties result from string matching and term set inclusion. We investigated the implicit terminological overlap between both resources in the domain of human proteins and evaluated our approach on a sample of 550 randomly selected UNIPROT entries that were manually mapped to their corresponding MESH headings. We achieved 90% precision and 79% recall (applying taxonomy-sensitive metrics). Fortunately, those proteins we were able to map to the MESH are ten times as frequently discussed in the literature as those on which we failed.\",\n \"Title: How to distinguish parthood from location in bioontologies.\\nYear: 2005\\n\\nAbstract: The pivotal role of the relation part-of in the descrip- tion of living organisms is widely acknowledged. Or- ganisms are open systems, which means that in con- tradistinction to mechanical artifacts they are charac- terized by a continuous ow and exchange of matter. A closer analysis of the spatial relations in biological or- ganisms reveals that the decision as to whether a given particular is part-of a second particular or whether it is only contained-in the second particular is often controversial. We here propose a rule-based approach which allows us to decide on the basis of well-dened criteria which of the two relations holds between two anatomical objects, given that one spatially includes the other. We discuss the advantages and limitations of this approach, using concrete examples from human anatomy.\",\n \"Title: Really, is medical sublanguage that different? Experimental counter-evidence from tagging medical and newspaper corpora.\\nYear: 2004\\n\\nAbstract: We compare the performance of two part-of-speech taggers trained on a German newspaper corpus for mixed types of medical documents. TnT, a tagger based on a statistical language model, outperforms Brill's rule-based tagger, and supplied with additional lexicon resources matches state-of-the-art performance figures (close to 97% accuracy) on the medical corpus. We explain this unexpected result by focusing on the statistically,significant part-of-speech type overlap between the newspaper training set and the medical test set. At least at that level sub-language differences seem to vanish. Thus, statistical off-the shellf part-of-speech taggers can immediately be reused for medical language processing.\",\n \"Title: Fostering Multilinguality in the UMLS: A Computational Approach to Terminology Expansion for Multiple Languages.\\nYear: 2014\\n\\nAbstract: We here report on efforts to computationally support the maintenance and extension of multilingual biomedical terminology resources. Our main idea is to treat term acquisition as a classification problem guided by term alignment in parallel multilingual corpora, using termhood information coming from of a named entity recognition system as a novel feature. We report on experiments for Spanish, French, German and Dutch parts of a multilingual UMLS-derived biomedical terminology. These efforts yielded 19k, 18k, 23k and 12k new terms and synonyms, respectively, from which about half relate to concepts without a previously available term label for these non-English languages. Based on expert assessment of a novel German terminology sample, 80% of the newly acquired terms were judged as reasonable additions to the terminology.\",\n \"Title: Sharing models and tools for processing german clinical texts.\\nYear: 2015\\n\\nAbstract: The automatic processing of non-English clinical documents is massively hampered by the lack of publicly available medical language resources for training, testing and evaluating NLP components. We suggest sharing statistical models derived from access-protected clinical documents as a reasonable substitute and provide solutions for sentence splitting, tokenization and POS tagging of German clinical texts. These three components were trained on the confidential FRAMED corpus, a non-sharable collection of various German-language clinical document types. The models derived therefrom outperform alternative components from OPENNLP and the Stanford POS tagger, also trained on FRAMED.\",\n \"Title: The extraction of pharmacogenetic and pharmacogenomic relations--a case study using PharmGKB.\\nYear: 2012\\n\\nAbstract: In this paper, we report on adapting the JREX relation extraction engine, originally developed for the elicitation of protein-protein interaction relations, to the domains of pharmacogenctics and pharmacogenomics. We propose an intrinsic and an extrinsic evaluation scenario which is based on knowledge contained in the PHARMGKB knowledge base. Porting JREX yields favorable results in the range of 80% F-score for Gene-Disease, Gene-Drug, and Drug-Disease relations.\",\n \"Title: Adding Multilingual Terminological Resources to Parallel Corpora for Statistical Machine Translation Deteriorates System Performance: A Negative Result from Experiments in the Biomedical Domain\\nYear: 2015\\n\\nAbstract: Unlike many other domains, biomedicine not only provides a wide range of parallel text corpora to train statistical machine translation SMT systems on, but also offers substantial amounts of 'parallel lexicons' in the form of multilingual terminologies. We included these lexical repositories, together with common parallel text corpora, into a Moses-based SMT system and three commercial systems and performed experiments on four language pairs, three text genres and several corpus sizes to measure the effects of adding the lexical knowledge sources. Much to our surprise, the SMT systems additionally equipped with 'parallel lexicons' underperformed in comparison with those systems trained on parallel text corpora only. This effect could consistently be shown for all systems by Bleu scores, as well as assessments from human judges.\",\n \"Title: Emotion Analysis as a Regression Problem - Dimensional Models and Their Implications on Emotion Representation and Metrical Evaluation.\\nYear: 2016\\n\\nAbstract: Emotion analysis (EA) and sentiment analysis are closely related tasks differing in the psychological phenomenon they aim to catch. We address fine-grained models for EA which treat the computation of the emotional status of narrative documents as a regression rather than a classification problem, as performed by coarse-grained approaches. We introduce Ekman's Basic Emotions (BE) and Russell and Mehrabian's Valence-Arousal-Dominance (VAD) model-two major schemes of emotion representation following opposing lines of psychological research, i.e., categorical and dimensional models- and discuss problems when BEs are used in a regression approach. We present the first natural language system thoroughly evaluated for fine-grained emotion analysis using the VAD scheme. Although we only employ simple BOW features, we reach correlation values up until r = .65 with human annotations. Furthermore, we show that the prevailing evaluation methodology relying solely on Pearson's correlation coefficient r is deficient which leads us to the introduction of a complementary error-based metric. Due to the lack of comparable (VAD-based) systems, we, finally, introduce a novel method of mapping between VAD and BE emotion representations to create a reasonable basis for comparison. This enables us to evaluate VAD output against human BE judgments and, thus, allows for a more direct comparison with existing BE-based emotion analysis systems. Even with this, admittedly, error-prone transformation step our VAD-based system achieves state-of-the-art performance in three out of six emotion categories, out-performing all existing BE-based systems but one.\",\n \"Title: CodE Alltag: A German-Language E-Mail Corpus.\\nYear: 2016\\n\\nAbstract: We introduce CODE ALLTAG, a text corpus composed of German-language e-mails. It is divided into two partitions: the first of these portions, CODE ALLTAG XL, consists of a bulk-size collection drawn from an openly accessible e-mail archive (roughly 1.5M e-mails), whereas the second portion, CODE ALLTAG S+d, is much smaller in size (less than thousand e-mails), yet excels with demographic data from each author of an e-mail. CODE ALLTAG, thus, currently constitutes the largest e-mail corpus ever built. In this paper, we describe, for both parts, the solicitation process for gathering e-mails, present descriptive statistical properties of the corpus, and, for CODE ALLTAG S+d, reveal a compilation of demographic features of the donors of e-mails.\",\n \"Title: UIMA-Based JCoRe 2.0 Goes GitHub and Maven Central ― State-of-the-Art Software Resource Engineering and Distribution of NLP Pipelines.\\nYear: 2016\\n\\nAbstract: We introduce JCORE 2.0, the relaunch of a UIMA-based open software repository for full-scale natural language processing originating from the Jena University Language & Information Engineering (JULIE) Lab. In an attempt to put the new release of JCORE on firm software engineering ground, we uploaded it to GITHUB, a social coding platform, with an underlying source code versioning system and various means to support collaboration for software development and code modification management. In order to automate the builds of complex NLP pipelines and properly represent and track dependencies of the underlying JAVA code, we incorporated MAVEN as part of our software configuration management efforts. In the meantime, we have deployed our artifacts on MAVEN CENTRAL, as well. JCORE 2.0 offers a broad range of text analytics functionality (mostly) for English-language scientific abstracts and full-text articles, especially from the life sciences domain.\",\n \"Title: Do Enterprises Have Emotions?\\nYear: 2016\\n\\nAbstract: Emotional language of human individuals has been studied for quite a while dealing with opinions and value judgments people have and share with others. In our work, we take a different stance and investigate whether large organizations, such as major industrial players, have and communicate emotions, as well. Such an anthropomorphic perspective has recently been advocated in management and organization studies which consider organizations as social actors. We studied this assumption by analyzing 1,676 annual business and sustainability reports from 90 top-performing enterprises in the United States, Great Britain and Germany. We compared the measurements of emotions in this homogeneous corporate text corpus with those from RCV1, a heterogeneous Reuters newswire corpus. From this, we gathered empirical evidence that business reports compare well with typical emotion-neutral economic news, whereas sustainability reports are much more emotionally loaded, similar to emotion-heavy sports and fashion news from Reuters. Furthermore, our data suggest that these emotions are distinctive and relatively stable over time per organization, thus constituting an emotional profile for enterprises.\",\n \"Title: JUFIT: A Configurable Rule Engine for Filtering and Generating New Multilingual UMLS Terms.\\nYear: 2015\\n\\nAbstract: We here describe JuFiT, an easily adjustable rule engine which allows to filter non-natural terms (i.e., ones usually not occurring in running citation texts) from the Umls metathesaurus and even adds new terms to the UMLS (by rewriting non-natural terms). Unlike previous attempts (with MetaMap or Casper), JuFiT serves multilingual purposes in that it runs for English, Spanish, French, German and Dutch documents, as well - the most prominent European languages in terms of UMLS coverage. We evaluated JuFiT under a variety of experimental conditions and found evidence that it increases annotation quality for English, and most likely also for German and Spanish.\",\n \"Title: Selecting and Tailoring Ontologies with JOYCE.\\nYear: 2016\\n\\nAbstract: We present Joyce, a scalable tool for identifying and assembling relevant (pieces of) ontologies from a repository of source ontologies, thus enabling the effective and efficient reuse of formalized domain knowledge. Joyce includes a conceptual filter to identify relevant classes, minimizes unintended redundancies, i.e. concept duplicates, and excludes knowledge considered irrelevant for the specific conceptual design task.\",\n \"Title: Semedico: A Comprehensive Semantic Search Engine For The Life Sciences\\nYear: 2017\\n\\nAbstract: SEMEDICO is a semantic search engine designed to support literature search in the life sciences by integrating the semantics of the domain at all stages of the search process-from query formulation via query processing up to the presentation of results. SEMEDICO excels with an ad-hoc search approach which directly reflects relevance in terms of information density of entities and relations among them (events) and, a truly unique feature, ranks interaction events by certainty information reflecting the degree of factuality of the encountered event.\",\n \"Title: Scholarly Information Extraction Is Going to Make a Quantum Leap with PubMed Central (PMC)® - But Moving from Abstracts to Full Texts Seems Harder than Expected.\\nYear: 2017\\n\\nAbstract: With the increasing availability of complete full texts (journal articles), rather than their surrogates (titles, abstracts), as resources for text analytics, entirely new opportunities arise for information extraction and text mining from scholarly publications. Yet, we gathered evidence that a range of problems are encountered for full-text processing when biomedical text analytics simply reuse existing NLP pipelines which were developed on the basis of abstracts (rather than full texts). We conducted experiments with four different relation extraction engines all of which were top performers in previous BioNLP Event Extraction Challenges. We found that abstract-trained engines loose up to 6.6% F-score points when run on full-text data. Hence, the reuse of existing abstract-based NLP software in a full-text scenario is considered harmful because of heavy performance losses. Given the current lack of annotated full-text resources to train on, our study quantifies the price paid for this short cut.\",\n \"Title: Representation Mapping: A Novel Approach to Generate High-Quality Multi-Lingual Emotion Lexicons.\\nYear: 2018\\n\\nAbstract: In the past years, sentiment analysis has increasingly shifted attention to representational frameworks more expressive than semantic polarity (being positive, negative or neutral). However, these richer formats (like Basic Emotions or Valence-Arousal-Dominance, and variants therefrom), rooted in psychological research, tend to proliferate the number of representation schemes for emotion encoding. Thus, a large amount of representationally incompatible emotion lexicons has been developed by various research groups adopting one or the other emotion representation format. As a consequence, the reusability of these resources decreases as does the comparability of systems using them. In this paper, we propose to solve this dilemma by methods and tools which map different representation formats onto each other for the sake of mutual compatibility and interoperability of language resources. We present the first large-scale investigation of such representation mappings for four typologically diverse languages and find evidence that our approach produces (near-)gold quality emotion lexicons, even in cross-lingual settings. Finally, we use our models to create new lexicons for eight typologically diverse languages.\",\n \"Title: Emotion Representation Mapping for Automatic Lexicon Construction (Mostly) Performs on Human Level.\\nYear: 2018\\n\\nAbstract: Emotion Representation Mapping (ERM) has the goal to convert existing emotion ratings from one representation format into another one, e.g., mapping Valence-Arousal-Dominance annotations for words or sentences into Ekmanu0027s Basic Emotions and vice versa. ERM can thus not only be considered as an alternative to Word Emotion Induction (WEI) techniques for automatic emotion lexicon construction but may also help mitigate problems that come from the proliferation of emotion representation formats in recent years. We propose a new neural network approach to ERM that not only outperforms the previous state-of-the-art. Equally important, we present a refined evaluation methodology and gather strong evidence that our model yields results which are (almost) as reliable as human annotations, even in cross-lingual settings. Based on these results we generate new emotion ratings for 13 typologically diverse languages and claim that they have near-gold quality, at least.\",\n \"Title: JeSemE: A Website for Exploring Diachronic Changes in Word Meaning and Emotion.\\nYear: 2018\\n\\nAbstract: We here introduce a substantially extended version of JeSemE, a website for visually exploring computationally derived time-variant information on word meaning and lexical emotion assembled from five large diachronic text corpora. JeSemE is intended as an interactive tool for scholars in the (digital) humanities who are mostly limited to consulting manually compiled dictionaries for such information, if available at all. JeSemE uniquely combines state-of-the-art distributional semantics with a nuanced model of human emotions, two information streams we deem beneficial for a data-driven interpretation of texts in the humanities.\",\n \"Title: Inducing Affective Lexical Semantics in Historical Language.\\nYear: 2018\\n\\nAbstract: The emotional connotation attached to words undergoes language change. In this paper, we aim at estimating the emotion which is associated with a given word in former language stages of English and German. Emotion is represented following the popular Valence-Arousal-Dominance (VAD) annotation scheme. While being more expressive than polarity alone, existing word polarity induction methods are typically not suited for addressing it. To overcome this limitation, we present adaptations of two popular algorithms to VAD. To measure the effectiveness of these methods in diachronic settings, we present two small gold lexicons for historical word emotions which where annotated by scholars with proficiency in the respective language (stage). In contrast to claims in previous work, we find that limiting the training set to words of supposedly stable emotional meaning does in fact decrease performance, rather than boosting it. This contribution thus elaborates algorithmic and experimental results which form the basis of some of our later publications (Buechel et al., 2017; Buechel and Hahn, 2018; Hellrich et al., 2018) yet had to be skipped there, partly for brevity, partly for a change in focus of our work.\",\n \"Title: Downsampling Strategies are Crucial for Word Embedding Reliability.\\nYear: 2018\\n\\nAbstract: The reliability of word embeddings algorithms, i.e., their ability to provide consistent computational judgments of word similarity when trained repeatedly on the same data set, has recently raised concerns. We compared the effect of probabilistic and weighting as downsampling strategies. We found the latter to provide superior reliability while being competitive in accuracy when applied to singular value decomposition-based embeddings\",\n \"Title: Annotation Data Management with JeDIS\\nYear: 2018\\n\\nAbstract: This paper introduces the Jena Document Information System (JeDIS). The focus lies on its capability to partition annotation graphs into modules. Annotation modules are defined in terms of types from the annotation schema. Modules allow easy manipulation of their annotations (deletion or update) and the creation of alternative annotations of individual documents even for annotation formalisms that by design do not support this feature.\",\n \"Title: 3000PA-Towards a National Reference Corpus of German Clinical Language.\\nYear: 2018\\n\\nAbstract: We introduce 3000PA, a clinical document corpus composed of 3,000 EPRs from three different clinical sites, which will serve as the backbone of a national reference language resource for German clinical NLP. We outline its design principles, results from a medication annotation campaign and the evaluation of a first medication information extraction prototype using a subset of 3000PA.\",\n \"Title: Constraining the Acquisition of Concepts by the Quality of Heterogeneous Evidence\\nYear: 1997\\n\\nAbstract: A model of concept acquisition is proposed, which combines various forms of linguistic and conceptual evidence that arise in the course of text understanding processes. We use terminological classification for the creation and management of concept hypotheses, for their incremental annotation by assertions which reflect the quality of available evidence, and for their subsequent evaluation and selection.\",\n \"Title: CDA-Compliant Section Annotation of German-Language Discharge Summaries: Guideline Development, Annotation Campaign, Section Classification.\\nYear: 2018\\n\\nAbstract: We present the outcome of an annotation effort targeting the content-sensitive segmentation of German clinical reports into sections. We recruited an annotation team of up to eight medical students to annotate a clinical text corpus on a sentence-by-sentence basis in four pre-annotation iterations and one final main annotation step. The annotation scheme we came up with adheres to categories developed for clinical documents in the HL7-CDA (Clinical Document Architecture) standard for section headings. Once the scheme became stable, we ran the main annotation campaign on the complete set of roughly 1,000 clinical documents. Due to its reliance on the CDA standard, the annotation scheme allows the integration of legacy and newly produced clinical documents within a common pipeline. We then made direct use of the annotations by training a baseline classifier to automatically identify sections in clinical reports.\",\n \"Title: A description logics approach to CGPS.\\nYear: 2004\\n\\nAbstract: We develop a formal framework by which clinical guidelines and protocols (CGPs) can be partially represented as a set of terminological concept definitions using standard description logics. There are two benefits in pursuing such an approach. First, it provides a foundation for logic-based CGP fusion and collision detection. Second, it allows for the checking of clinical treatment episodes from the EPR against CGPs.\",\n \"Title: Coping with Different Types of Ambiguity Using a Uniform Context Handling Mechanism\\nYear: 2000\\n\\nAbstract: We introduce a uniform context mechanism which is able to adequately represent and manage different forms of ambiguities as they occur in the course of text understanding. Different lexical, syntactic and semantic interpretations are clearly separated by assigning each alternative a single context space for local reasoning. The mechanism we propose directly supports the task of disambiguation at all levels of text analysis, since it also incorporates constraints from the discourse context, as text understanding proceeds.\",\n \"Title: Exploring Diachronic Lexical Semantics With Jeseme\\nYear: 2017\\n\\nAbstract: Recent advances in distributional semantics combined with the availability of large-scale diachronic corpora offer new research avenues for the Digital Humanities. JESEME, the Jena Semantic Explorer, renders assistance to a non-technical audience to investigate diachronic semantic topics. JESEME runs as a website with query options and interactive visualizations of results, as well as a REST API for access to the underlying diachronic data sets.\",\n \"Title: Modeling Word Emotion in Historical Language: Quantity Beats Supposed Stability in Seed Word Selection.\\nYear: 2019\\n\\nAbstract: To understand historical texts, we must be aware that language -- including the emotional connotation attached to words -- changes over time. In this paper, we aim at estimating the emotion which is associated with a given word in former language stages of English and German. Emotion is represented following the popular Valence-Arousal-Dominance (VAD) annotation scheme. While being more expressive than polarity alone, existing word emotion induction methods are typically not suited for addressing it. To overcome this limitation, we present adaptations of two popular algorithms to VAD. To measure their effectiveness in diachronic settings, we present the first gold standard for historical word emotions, which was created by scholars with proficiency in the respective language stages and covers both English and German. In contrast to claims in previous work, our findings indicate that hand-selecting small sets of seed words with supposedly stable emotional meaning is actually harmful rather than helpful.\",\n \"Title: GRASCCO - The First Publicly Shareable, Multiply-Alienated German Clinical Text Corpus.\\nYear: 2022\\n\\nAbstract: We describe the creation of GRASCCO, a novel German-language corpus composed of some 60 clinical documents with more than.43,000 tokens. GRASCCO is a synthetic corpus resulting from a series of alienation steps to obfuscate privacy-sensitive information contained in real clinical documents, the true origin of all GRASCCO texts. Therefore, it is publicly shareable without any legal restrictions We also explore whether this corpus still represents common clinical language use by comparison with a real (non-shareable) clinical corpus we developed as a contribution to the Medical Informatics Initiative in Germany (MII) within the SMITH consortium. We find evidence that such a claim can indeed be made.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":178,"string":"178"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13590,"string":"13,590"}}},{"rowIdx":13591,"cells":{"question_id":{"kind":"string","value":"K G Subramanian"},"question":{"kind":"string","value":"Generate an abstract for the title \"Algebraic properties of the shuffle over omega-trajectories\" using the following items: keywords: algebraic properties, relation, ω-words, shuffle, ω-trajectories"},"answer":{"kind":"string","value":"Algebraic properties of a relation on ω-words based on the notion of shuffle over ω-trajectories are obtained."},"correct_answer":{"kind":"string","value":"Algebraic properties of a relation on ω-words based on the notion of shuffle over ω-trajectories are obtained."},"documents":{"kind":"list like","value":["Title: Lyndon trees\nYear: 1992\n\nAbstract: Using deterministic (sample-path) analysis, we generalize andextend fundamental properties of systems with “stationarydeterministic flows” as introduced byGelenbe (1983) and Gelenbeand Finkel ...","Title: A Note on Parallel Splicing on Images\nYear: 2001\n\nAbstract: The concept of splicing on images which is done in parallel is introduced. This is an extension of the operation of splicing on strings extensively studied in the context of DNA computing. Various properties of splicing on images are examined.","Title: Infinite Lyndon words\nYear: 1994\n\nAbstract: We define an infinite Lyndon word as the limit of an increasing sequence of prefix preserving Lyndon words and show that some of the interesting properties of Lyndon words generalize to the infinite case. We construct a queue automaton that recognizes the set of Lyndon words and show that it can be extended to recognize infinite Lyndon words. We discuss certain topological properties of the set of infinite Lyndon words such as homeomorphism with a subspace of the Cantor space.","Title: Hexagonal Pattern Languages\nYear: 2004\n\nAbstract: Hexagonal tiling systems, hexagonal local picture languages and hexagonal recognizable picture languages are defined in this\n paper. Hexagonal Wang tiles and systems are also introduced. It is noticed that the family of hexagonal picture languages\n defined by hexagonal Wang systems coincides with the family of hexagonal picture languages recognized by hexagonal tiling\n system. Similar to hv-domino systems describing rectangular arrays, we define xyz-domino systems and characterize hexagonal\n picture languages using this. Unary hexagonal picture languages are also considered and we analyze some of their properties.","Title: On Parikh Matrices\nYear: 2009\n\nAbstract: Mateescu et al (2000) introduced an interesting new tool, called Parikh matrix, to study in terms of subwords, the numerical properties of words over an alphabet. The Parikh matrix gives more information than the well-known Parikh vector of a word which counts only occurrences of symbols in a word. In this note a property of two words u, v, called \"ratio property\", is introduced. This property is a sufficient condition for the words uv and vu to have the same Parikh matrix. Thus the ratio property gives information on the M-ambiguity of certain words and certain sets of words. In fact certain regular, context-free and context-sensitive languages that have the same set of Parikh matrices are exhibited. In the study of fair words, Cerny (2006) introduced another kind of matrix, called the p-matrix of a word. Here a \"weak-ratio property\" of two words u, v is introduced. This property is a sufficient condition for the words uv and vu to have the same p-matrix. Also the words uv and vu are fair whenever u, v are fair and have the weak ratio property.","Title: Array P Systems and t.Communication\nYear: 2009\n\nAbstract: The two areas of grammar systems and P systems, which have provided interesting computational models in the study of formal string language theory have been in the recent past effectively linked in [4] by incorporating into P systems, a communication mode called t-mode of cooperating distributed grammar systems. On the other hand cooperating array grammar systems [5] and array P systems [1] have been developed in the context of two-dimensional picture description. In this paper, motivated by the study of [4], these two systems are studied by linking them through the t-communication mode, thus bringing out the picture description power of these systems.","Title: Array P systems with permitting features.\nYear: 2014\n\nAbstract: The biologically inspired model, known as P system, has proved to be a general framework for investigating several problems related to computing in different fields. Two-dimensional picture array languages is one such area in which different kinds of P systems have been constructed for picture array generation. Incorporating the feature of permitting symbols in the rules, a new variety of array P system is constructed here for generating picture languages consisting of picture arrays. The advantage of this approach is that there is a reduction in the number of membranes used in the construction, in comparison to the existing array P system models. (C) 2013 Elsevier B.V. All rights reserved.","Title: Cell-Work Ol-Systems With Fins\nYear: 1999\n\nAbstract: Cell-works which are three-dimensional cyclic edge-label controlled OL-systems, were introduced by Lindenmayer. Fracchia and Prusinkiewicz provided cell-work systems using markers to model three-dimensional cellular structures. In this paper, cell-work systems with fins are proposed, generalizing the notion of map with handles. We allow each cell of a cell-work to divide into finitely many cells at any instant of time. This generalization enables us to describe three-dimensional images and fractals. A comparison of various systems with the system introduced in this paper is made.","Title: Array-rewriting P systems\nYear: 2003\n\nAbstract: We consider array languages (sets of picturesconsisting of symbols placed in the lattice points of the 2D grid) and thepossibility to handle them with P systems. After proving binary normal formsfor array matrix grammars (which, even in the case when no appearance checking isused, are known to generate the array languages of arbitrary array grammars), weprove that the P systems with context-free rules (with three membranes and no control on the communication or the use of rules) are computationally universal, able togenerate all computable array languages. Some open problems are also formulated.","Title: Infinite Arrays And Recognizability\nYear: 2000\n\nAbstract: In this paper, the concept of local languages is extended to infinite arrays and omega omega-local languages are defined. ww-recognizable languages of infinite arrays accepted by online tesselation automata are considered. Properties of these languages are studied. The notion of Muller recognizability is extended to infinite arrays. This is related to acceptance of infinite arrays by online tesselation automata.","Title: Selective substitution array grammars\nYear: 1981\n\nAbstract: Selective substitution array grammars are introduced, which abstract the notions of rewriting rules, direct derivation steps, derivations and rewriting in a sequential or parallel way. These array grammars provide for a unified framework for many of the two-dimensional array grammars in the literature. In addition, they point towards new interpretations resulting in possible new array grammars.","Title: Two-dimensional pattern languages.\nYear: 2017\n\nAbstract: We introduce several classes of array languages obtained by generalising Angluin's pattern languages to the two-dimensional case. These classes of two-dimensional pattern languages are compared with respect to their expressive power and their closure properties are investigated.","Title: On Watson-Crick automata\nYear: 2012\n\nAbstract: Motivated by Watson-Crick automata, we introduce here a grammar counterpart called a Watson-Crick regular grammar that has rules as in a regular grammar but involves double stranded strings. The language generated by this grammar consists of strings in the upper strands of the double stranded strings related by a complementarity relation. We obtain some properties of these grammars and also indicate an application to generation of chain code pictures.","Title: Parallel Splicing On Images\nYear: 2004\n\nAbstract: In this paper, splicing on images of rectangular arrays is introduced as a simple and effective extension of the operation of splicing on strings extensively studied in the context of DNA computing. A comparison of the resulting class of splicing array languages with the local two-dimensional array languages and two-dimensional right linear languages is made. Certain closure properties are obtained. Furthermore, the notion of self cross-over of strings is extended to arrays with respect to the splicing of arrays introduced here.","Title: Permitting Features in P Systems Generating Picture Arrays.\nYear: 2012\n\nAbstract: In the area of membrane computing, the biologically inspired model known as P system has proved to be a rich framework for studying several types of problems. Picture array generation is one such problem for which different P systems have been constructed in the literature. Incorporating the feature of permitting symbols in the rules, array P systems are constructed here for generating picture languages consisting of picture arrays. The advantage of this approach is that there is a reduction in the number of membranes used in the construction, in comparison to the existing array P system model.","Title: Tissue-like P Systems with Active Membranes for Picture Generation\nYear: 2003\n\nAbstract: We propose a variant of tissue-like P systems with symport rules and active membranes that generates two-dimensional picture languages. The method is unconventional in that, instead of using the membranes as regions for computation/writing, it uses them to hold elements of pictures. Thus, the picture itself, and actually the whole supporting rectangular grid, is composed of membranes, each one containing (among other symbols) a letter of the picture's alphabet. The method is illustrated by its application to the generation of local and recognizable picture languages.","Title: P Systems Generating Hexagonal Picture Languages\nYear: 2003\n\nAbstract: P systems generating rectangular arrays have been recently introduced in [1,2,3], thus bringing together the two areas of theoretical computer science, namely membrane computing and picture grammars. In this paper, hexagonal arrays on triangular grids are considered and the capabilities of the three approaches [1,2,3] that construct P systems for generation of arrays on rectangular grids, to handle arrays on triangular grids, are demonstrated.","Title: Learning code regular and code linear languages\nYear: 1996\n\nAbstract: In this paper, a subclass of regular languages, called code regular languages is defined. Algorithms for learning these languages are presented in the frame work of identification in the limit. Learning of analogous subclass of linear languages is also examined.","Title: Array Insertion And Deletion P Systems\nYear: 2013\n\nAbstract: We consider the (d-dimensional) array counterpart of string insertion and deletion grammars and use the operations of array insertion and deletion in the framework of P systems where the applicability of the rules depends on the membrane region. In this paper, we especially focus on examples of two-dimensional array insertion and deletion P systems and show that we can already obtain computational completeness using such P systems with a membrane structure of tree height of at most two and only the targets here, in, and out.","Title: Pure 2D picture grammars (P2DPG) and P2DPG with regular control\nYear: 2008\n\nAbstract: In this note a new model of grammatical picture generation is introduced. The model is based on the notion of pure context-free grammars of formal string language theory. The resulting model, called Pure 2D context-free grammar (CFG), generates rectangular picture arrays of symbols. The generative power of this model in comparison to certain other related models is examined. Also we associate a regular control language with a Pure 2D CFG and notice that the generative power increases. Certain closure properties are obtained.","Title: Product Of Parikh Matrices And Commutativity\nYear: 2012\n\nAbstract: The Parikh vector of a word enumerates the symbols of the alphabet that occur in the word. The Parikh matrix of a word which has been recently introduced, is an extension of the notion of Parikh vector and gives more numerical information about the word in terms of certain subwords. Intensive investigation on various theoretical properties of Parikh matrices has taken place. This paper deals with the problem of finding properties of words so that their Parikh matrices commute.","Title: 2D and 3d pictural networks of evolutionary processors\nYear: 2005\n\nAbstract: Networks of Evolutionary Processors constitute an interesting computing model motivated by cell biology. Pictural networks of evolutionary processors have been introduced to handle generation of pictures of rectangular arrays of symbols. Here we introduce an extension of this model to generate pictures that are arrays, not necessarily rectangular, and compare these with array generating Puzzle grammars. We also provide an extension to generate 3D rectangular arrays.","Title: Tile pasting systems for tessellation and tiling patterns\nYear: 2010\n\nAbstract: Pasting System (PS) and Extended Pasting Scheme (EPS) are new techniques to generate tiling patterns by using pasting rules for gluing tiles at their edges. In this paper we introduce a Tabled Pasting System (TPS) based on the technique of suitably grouping the rules of a pasting system into different sets of rules. We compare the generative power of TPS with PS and EPS. We propose theoretical models of P-system, called tile pasting P-system, with active membranes (TPPS−AM) and tissue-like evolution communication tile pasting P system (t−ECTPPS) to generate tiling patterns made of polygon tiles and compare the generative power of TPS with TPPS−AM and t−ECTPPS.","Title: Tree Replacement and Public Key Cryptosystem\nYear: 2002\n\nAbstract: In this paper, a public key cryptosystem based on tree replacement system is proposed.","Title: A note on regular-controlled apical-growth filamentous systems\nYear: 1985\n\nAbstract: The effect of regular control on Filamentous Systems with apical growth is examined. The resulting families of languages are classified with respect to regular, linear, context-free and pure languages.","Title: Recognizing occluded faces by exploiting psychophysically inspired similarity maps\nYear: 2013\n\nAbstract: The presence of occlusions in facial images is inevitable in unconstrained scenarios. However recognizing occluded faces remains a partially solved problem in computer vision. In this contribution we propose a novel Bayesian technique inspired by psychophysical mechanisms relevant to face recognition to address the facial occlusion problem. For some individuals certain facial regions, e.g. features comprising of some of the upper face, might be more discriminative than the rest of the features in the face. For others, it might be the features over the mid face and some of the lower face that are important. The proposed approach in this paper, will allow for such a psychophysical analysis to be factored into the recognition process. We have discovered and modeled similarity mappings that exist in facial domains by means of Bayesian Networks. The model can efficiently learn and exploit these mappings from the facial domain and hence capable of tackling uncertainties caused by occlusions. The proposed technique shows improved recognition rates over state of the art techniques.","Title: PARALLEL CONTEXTUAL ARRAY GRAMMARS AND LANGUAGES\nYear: 2003\n\nAbstract: In this paper a new method of description of pictures of digitized rectangular arrays is introduced based on contextual grammars. Parallel contextual array grammars are defined and their properties are studied.","Title: Efficient Algorithms for Reconstruction of 2D-Arrays from Extended Parikh Images\nYear: 2008\n\nAbstract: The concept of a Parikh matrix or an extended Parikh mapping of words introduced by Mateescu et al (2001) is formulated here for two-dimensional (2D) arrays. A polynomial time algorithm is proposed to reconstruct an unknown 2D-array over { 0,1 } from its image under the extended Parikh mapping along a single direction. On the other hand the problem of reconstructing a 2D-array over { 0,1 } from its image under the extended Parikh mapping along three or more directions is shown to be NP-hard. Also a polynomial time algorithm to reconstruct a 2D-array over {0,1} with a maximum number of ones close to the main diagonal of the array is presented by reducing the problem to Min-cost Max-flow problem.","Title: On the power of permitting features in cooperating context-free array grammar systems\nYear: 2013\n\nAbstract: Among many grammar models for picture description, cooperating context-free distributed array grammar systems utilize the mechanism of cooperation between component array grammars. Here, we endow these grammar systems with the feature of permitting symbols, and examine the generative capacity of the resulting permitting cooperating context-free distributed array grammar systems with certain other kinds of isometric and non-isometric array grammars. An advantage is that the number of components is reduced when this permitting feature is used in a cooperating context-free distributed array grammar system.","Title: A P system model with pure context-free rules for picture array generation\nYear: 2010\n\nAbstract: Syntactic models constitute one of the main areas of mathematical studies on picture array generation. A number of 2D grammar models have been proposed motivated by a variety of applications. A P system is a biologically motivated new computing model, proposed by Paun in the area of membrane computing. It is a rich framework for dealing with different problems, including the problem of handling picture array generation. In this paper, the generative power of the array-rewriting P system with pure 2D context-free rules is investigated by comparing it with other 2D grammar models, thus bringing out the suitability of this P system model for picture array generation.","Title: Inferring Subclasses of Contextual Languages\nYear: 2000\n\nAbstract: In this paper, we show that the languages generated by internal contextual grammars are not inferable from positive data only.\n We define two subclasses of internal contextual languages, namely, k-uniform and strictly internal contextual languages which\n are incomparable classes and provide an algorithm to learn these classes. The algorithm can be used when the rules are applied\n in a parallel mode.","Title: A Variant Of Pure Two-Dimensional Context-Free Grammars Generating Picture Languages\nYear: 2014\n\nAbstract: Considering a large variety of approaches in generating picture languages, the notion of pure two-dimensional context-free grammar (P2DCFG) represents a simple yet expressive non-isometric language generator of picture arrays. In the present paper, we introduce a new variant of P2DCFGs that generates picture arrays in a leftmost way. We concentrate our attention on determining their generative power by comparing it with the power of other picture generators. We also examine the power of these generators that regulate rewriting by control languages.","Title: Array Grammars with Contextual Operations\nYear: 2008\n\nAbstract: In this paper, we consider a class of two-dimensional array grammars, called parallel contextual array grammars, that extend the contextual operations on strings to arrays in a naturalway and generate languages of pictures of rectangular arrays. Several classes of these array grammars and the resulting families of picture languages are considered. Necessary conditions for picture languages to be contained in these classes are obtained and the relations between these families are also established.","Title: Infinite Arrays And Controlled Deterministic Table 0l Array Systems\nYear: 1984\n\nAbstract: Deterministic table 0L array systems with control are considered for the generation of infinite arrays. Rewriting of a rectangular array is done in parallel by a table of rules, the rightmost edge horizontally or the lowermost edge vertically downwards. The application of the tables is controlled by a control set. Cube-free and square-free infinite arrays are obtained as an application of this model. The adherence of the array language of a controlled deterministic table 0L array system is related to the adherence of its control set. The limit language equivalence problem and the adherence equivalence problem are shown to be undecidable for this system.","Title: A Tissue P System Based Solution To Quadratic Assignment Problem\nYear: 2012\n\nAbstract: The quadratic assignment problem (QAP) is one of the fundamental combinatorial optimization problems, which models many real-life problems. However, it is considered as one of the most difficult NP-hard problems, which means that no polynomial-time algorithm is known to solve this intractable problem effectively. Even small instances of QAP may require vast computation time. In this work, a uniform cellular solution to QAP is proposed in the framework of membrane computing by using a family of recognizer tissue P systems with cell division. In the design of the solution, we encode the given instances in binary notations. The paper can be considered as a contribution to the study of considering a binary encoding of the information in P systems.","Title: Cooperating basic puzzle grammar systems\nYear: 2006\n\nAbstract: Nivat et al [3] introduced Context-free Puzzle grammars for generating connected picture arrays in the two-dimensional plane. Basic Puzzle grammars [6] constitute a subclass of these grammars. In this note we consider the Cooperating Array Grammar Systems introduced by Dassow et al [2] with Basic Puzzle grammar rules in the components instead of array grammar rules and examine the picture generating power of the resulting system, called, Cooperating Basic Puzzle Grammar System, in the maximal mode.","Title: Circular DNA and Splicing Systems\nYear: 1992\n\nAbstract: Circular strings representing DNA molecules and certain recombinant behaviour are formalized. Various actions of splicing schemes on linear and circular DNA molecules are examined. It is shown that there is a difference in the regularity result of Culik and Harju [1] between the linear and circular strings. A consequence of this result is that a conjecture of Head [4] that the circular string language of a splicing system under an action on circular strings is regular, when the set of initial circular strings is regular, is disproved.","Title: P systems and picture languages\nYear: 2007\n\nAbstract: Array-rewriting P systems were introduced in [2] linking the two areas of membrane computing and picture grammars. Subsequently a variety of P systems with array objects and different kinds of rewriting has been introduced. Here we discuss a few prominent systems among these, point out their features and indicate possible problems for future study.","Title: P systems with shuffle operation.\nYear: 2010\n\nAbstract: Shuffle operation on trajectories is useful in modeling parallel composition of words and languages. In this work, a new class of P systems with shuffle operation is presented. Such a system has language-objects and shuffle-operation rules in its regions. It can be used as a language generator; specifically it is used to generate languages from sets of languages. Some comparison results are obtained, which show that the power of shuffle operation is enlarged in the framework of P systems. © 2010 IEEE.","Title: A note on regular kolam array grammars generating right triangles\nYear: 1979\n\nAbstract: A group structure is identified among the Regular Kolam Array Grammars that generate the families of Kirsch's right triangles with labelled vertices and their reflections about the leg and/or base of the triangles.","Title: On ambiguity of DTOL systems\nYear: 1987\n\nAbstract: A DTOL system is unambiguous if no two different sequences of morphisms yield the same word from an axiom. A subfamily of DTOL systems with decidable ambiguity problem is exhibited. Four different sufficient conditions for a DTOL system to be unambiguous are formulated. These DTOL systems are very much suitable for the construction of public key cryptosystems based on L systems. We also prove that for DOL systems over a binary alphabet, the ambiguity problem is effectively decidable. This result has useful applications in the construction of public key cryptosystems which encrypt plain-texts over a binary alphabet using a TOL system obtained from an underlying unambiguous DOL system.","Title: Psychophysically inspired bayesian occlusion model to recognize occluded faces\nYear: 2011\n\nAbstract: Face recognition systems robust to major occlusions have wide applications ranging from consumer products with biometric features to surveillance and law enforcement applications. In unconstrained scenarios, faces are often subject to occlusions, apart from common variations such as pose, illumination, scale, orientation and so on. In this paper we propose a novel Bayesian oriented occlusion model inspired by psychophysical mechanisms to recognize faces prone to occlusions amidst other common variations. We have discovered and modeled similarity maps that exist in facial domains by means of Bayesian Networks. The proposed model is capable of efficiently learning and exploiting these maps from the facial domain. Hence it can tackle the occlusion uncertainty reasonably well. Improved recognition rates over state of the art techniques have been observed.","Title: A note on an extension of matrix grammars generating two-dimensional languages\nYear: 1985\n\nAbstract: An extended version of the parallel/sequential table matrix models called extended table matrix grammars (Ex-TMG), generating arrays of symbols, is proposed. The effect of imposing different types of control on the sequence of application of the tables is studied. The notion of assigning “terminal weights” to matrix grammars for describing parquet deformations is extended to Ex-TMGs.","Title: Learning of recognizable picture languages\nYear: 1992\n\nAbstract: Learning of certain classes of two-dimensional picture languages is considered. Linear time algorithms that learn in the limit, from positive data the classes of local picture languages and locally testable picture languages are presented. A crucial step for obtaining the learning algorithm for local picture languages is an explicit construction of a two-dimensional on-line tessellation acceptor for a given local picture language. An efficient algorithm that learns the class of recognizable picture languages from positve data and restricted subset queries, is presented in contrast to the fact that this class is not learnable in the limit from positive data alone.","Title: On Some Classes of Splicing Languages\nYear: 2004\n\nAbstract: We introduce some classes of splicing languages generated with simple and semi-simple splicing rules, in both, the linear and circular cases. We investigate some of their properties.","Title: Tile Pasting P System Model for Pattern Generation\nYear: 2009\n\nAbstract: In the area of membrane computing, a new computability model, called P system, which is a distributed, highly parallel theoretical computing model, was introduced by Păun, inspired from the cell structure and its functioning. There are several application areas of these P systems. Among these one area deals with the problem of picture array generation. Ceterchi et al. began a study on linking the two areas of membrane computing and picture grammars, which were not very much linked before, by relating P systems and array-rewriting grammars generating picture languages and proposing array-rewriting P systems. In this paper we propose a theoretical model of a P System, called Tile pasting P system, for generating two-dimensional patterns that are formed by gluing square tiles and study some of its properties.","Title: Array P System Model with Pure Context-Free Rules\nYear: 2009\n\nAbstract: Motivated by the structure and functioning of the livingcell, a new computing model, called P System, was proposedin the area of Membrane Computing. Among severalareas of application of P systems, one area is concernedwith generation of picture array languages. Ceterchiet al. [1] initially proposed array-rewriting P systemsby extending the notion of string-rewriting P-Systems [8].On the other hand a new theoretical model of grammaticalpicture generation called pure 2D context-free picturegrammar (P2DCFPG) generating rectangular picture arraysof symbols has been introduced by Subramanian et al.[17]. Here we consider P2DCFPG type of rules in arrayrewritingP systems and examine the generating power ofthe resulting array P systems.","Title: A note on inferring uniquely terminating code languages\nYear: 1999\n\nAbstract: A subclass of linear languages, called uniquely terminating code linear languages (utCLL), is introduced, motivated by the study of Makinen (1997), and an inference algorithm for utCLL is provided. The algorithm is extended to a corresponding subclass of equal matrix languages. (C) 1999 Elsevier Science B.V. All rights reserved.","Title: Binary images, M-vectors, and ambiguity\nYear: 2011\n\nAbstract: Mateescu et al (2001) introduced the notion of Parikh matrix of a word as an extension of the well-known concept of Parikh vector of a word. The Parikh matrix provides more numerical information about a word than given by the Parikh vector. Here we introduce the notion of M-vector of a binary word which allows us to have a linear notation in the form of a unique vector representation of the Parikh matrix of the binary word. We then extend this notion of M-vector to a binary image treating it as a binary array over a two-symbol alphabet. This is done by considering the M-vectors of the words in the rows and columns of the array. Among the properties associated with a Parikh matrix, M-ambiguity or simply ambiguity of a word is one which has been investigated extensively in the literature. Here M-ambiguity of a binary array is defined in terms of its M-vector and we obtain conditions for M-ambiguity of a binary array.","Title: P Systems with Shuffle Operation and Catalytic-Like Rules.\nYear: 2012\n\nAbstract: Shuffle operation on trajectories is useful in modeling parallel composition of words and languages. In this work, a new class of P systems with shuffle operation and catalytic-like rules is presented. Such a system has a membrane structure, where language-objects and shuffle-operation rules are placed in its regions. It can be used as a language generator. In this study, we propose a variant P system with shuffle operation on string-language objects. Some comparison results are obtained, which show that the power of shuffle operation is enlarged in the framework of P systems. Moreover, string-language objects are extended to array-language objects, and another variant P system with shuffle operation on picture-language objects is introduced. We also illustrate how to generate picture languages by using this kind of devices.","Title: Two-Dimensional Digitized Picture Arrays And Parikh Matrices\nYear: 2013\n\nAbstract: Parikh matrix mapping or Parikh matrix of a word has been introduced in the literature to count the scattered subwords in the word. Several properties of a Parikh matrix have been extensively investigated. A picture array is a two-dimensional connected digitized rectangular array consisting of a finite number of pixels with each pixel in a cell having a label from a finite alphabet. Here we extend the notion of Parikh matrix of a word to a picture array and associate with it two kinds of Parikh matrices, called row Parikh matrix and column Parikh matrix. Two picture arrays A and B are defined to be M-equivalent if their row Parikh matrices are the same and their column Parikh matrices are the same. This enables to extend the notion of M-ambiguity to a picture array. In the binary and ternary cases, conditions that ensure M-ambiguity are then obtained.","Title: Self Cross-Over Array Languages.\nYear: 2011\n\nAbstract: Self cross over array language consists of arrays that are formed by applying the operation of array splicing in H-array splicing systems to identical copies of picture arrays. This paper compares the family of self cross-over array languages with other two-dimensional language classes. Various properties analogous to H-array splicing systems are proved.","Title: Some results on picture languages\nYear: 1999\n\nAbstract: Two different types of models called puzzle languages and recognizable picture languages describing digitized pictures in a two-dimensional plane have been introduced in the 1990s. We review here these models, reporting the main results already proved. We also give a few new results.","Title: Hyperbolic Tilings And Formal Language Theory\nYear: 2013\n\nAbstract: In this paper, we try to give the appropriate class of languages to which belong various objects associated with tessellations in the hyperbolic plane.","Title: Two-dimensional Picture Grammar models\nYear: 2008\n\nAbstract: A new theoretical model of grammatical picture generation called extended 2D context-free picture grammar (E2DCFPG) generating rectangular picture arrays of symbols is introduced. This model which allows variables in the grammar and uses the squeezing mechanism of forming the picture language over terminal symbols, is an extension of the pure 2D context-free picture grammar (P2DCFPG)[13]. The extended picture grammar model E2DCFPG is shown to have more picture generative power than the P2DCFPG and certain other existing 2D models. Certain closure and other properties of this new model are also examined.","Title: Learning a Subclass of Context-Free Languages\nYear: 1998\n\nAbstract: In this paper, Apical growth Pure context-free grammars (AGPCFG) which are a variation of the pure grammars of Maurer et al\n are introduced. These grammars allow rewriting of active symbols at the ends of a string simultaneously. They also provide\n a variation of another kind of grammars called filamentous systems with apical growth, which are motivated by biological considerations.\n The family of languages generated by AGPCFGs is a subclass of context-free languages. An algorithm for learning this language\n subclass, in the framework of identification in the limit from positive examples, is provided.","Title: Space-filling curves and infinite graphs\nYear: 1982\n\nAbstract: Two-dimensional geometric patterns in the form of the space filling curves of Peano and Hilbert are represented as a string, an array and a graph. The sequence of patterns is such that each curve is a prefix of the succeeding one and hence we get prefix preserving WDOL to generate the corresponding string representation. The limit language is generable by a CDOL system and consists of a single infinite word corresponding to the Peano curve and two infinite words for the Hilbert curve. The property that in the sequence each curve is a prefix of the succeeding one helps in the definition of the infinite arrays and infinite graphs representing the infinite curve patterns and we construct a Context-free Parentheses Kolam array grammar and a coding of a graph DOL system to generate the sequence as well as the limit.","Title: On the generative capacity of compound string and array grammars\nYear: 1982\n\nAbstract: The operation of composition is a device introduced by Abraham [1] to increase the generative capacity of regular matrix grammars. But not all families of grammars have their generative capacity increased by this operation. Păun [5] has proved that composition increases the generative capacity of linear grammars but not of regular, CF, CS or PS grammars. In this note, we show that the generative power of k-linear (k ⩾ 1) grammars is increased by composition. In fact, it is of interest to note that the families of compound linear and compound k-linear languages are equal. Furthermore, the family of compound linear languages is an AFL, properly contained in the family of CFLs. Certain other families of string and array grammars whose generative capacities are increased by the operation of composition are also provided.","Title: Context-Free Systems with a Complementarity Relation\nYear: 2011\n\nAbstract: Motivated by the studies on Watson-Crick finite automata and Watson-Crick L systems, we introduce a generative device known as a Watson-Crick context-free system (WKCFS) making use of a complementarity relation and a context-free type of rewriting on double-stranded strings with the symbols in the string in the lower strand complementary to the symbols of the string in the upper strand. The language generated by a WKCFS is the set of all strings in the upper strands of the double-stranded strings (u, v) generated such that the strings u, v are of the same length and the corresponding symbols in the strings u, v are related by the complementarity relation. We compare the generative power of WKCFS with some families of languages and also certain variants of Watson-Crick finite automata.","Title: Local And Recognizable Hexagonal Picture Languages\nYear: 2005\n\nAbstract: In this paper we consider hexagonal arrays on triangular grids and introduce hexagonal local picture languages and hexagonal tiling systems defining hexagonal recognizable picture languages, motivated by an analogous study of rectangular arrays by Giammarresi and Restivo. We also introduce hexagonal Wang tiles to define hexagonal Wang systems (HWS) as a formalism to describe hexagonal picture languages. It is noticed that the family of hexagonal picture languages defined by hexagonal Wang systems and the family recognized by hexagonal tiling systems coincide. Analogous to hv-domino systems describing rectangular arrays, we define xyz-domino systems and prove that recognizable hexagonal picture languages are characterized as projections of xyz-local picture languages.","Title: Public Key Cryptosystems Based on Free Partially Commutative Monoids and Groups\nYear: 2003\n\nAbstract: A public key cryptosystem based on free partially commutative monoids is constructed. The encryption of A message to create the cryptotext uses a Thue system which is formed from the free partially commutative monoid with the help of a trapdoor morphism. The decidability of the word problem for free partially commutative monoids can be used for decryption. Finding the trapdoor morphism of this system is shown to be NP-haxd. But, a zero - knowledge protocol to convince a verifier that there is such a trapdoor. morphism is provided. A related but different public key cryptosystem based on free partially commutative groups is also proposed.","Title: Subclasses of harmonic mappings defined by convolution\nYear: 2010\n\nAbstract: Two new subclasses of harmonic univalent functions defined by convolution are introduced. The subclasses generate a number of known subclasses of harmonic mappings, and thus provide a unified treatment in the study of these subclasses. Sufficient coefficient conditions are obtained that are shown to be also necessary when the analytic parts of the harmonic functions have negative coefficients. Growth estimates and extreme points are also determined.","Title: Pure 2D picture grammars and languages\nYear: 2009\n\nAbstract: A new syntactic model, called pure two-dimensional (2D) context-free grammar (P2DCFG), is introduced based on the notion of pure context-free string grammar. The rectangular picture generative power of this 2D grammar model is investigated. Certain closure properties are obtained. An analogue of this 2D grammar model called pure 2D hexagonal context-free grammar (P2DHCFG) is also considered to generate hexagonal picture arrays on triangular grids.","Title: Thue Systems and DNA - A Learning Algorithm for a Subclass\nYear: 1993\n\nAbstract: Thue systems are considered to be appropriate for modeling certain kinds of behavior in DNA sequences [6]. We present a polynomial time algorithm for learning a Church-Rosser Thue system when the quotient monoid is finite, making use of membership queries related to the congruential languages, which specify regular languages.","Title: Uniform Solution to Common Algorithmic Problem by P Systems Working in the Minimally Parallel Mode.\nYear: 2015\n\nAbstract: It is known that the Common Algorithmic Problem (CAP) has the nice property that several other NP-complete problems can be reduced to it in linear time. The decision version of this problem is known to be efficiently solved by a family of recognizer P systems with active membranes with three electrical charges working in the maximally parallel way. We here work with a variant of P systems with active membranes without polarizations and present a uniform solution to CAP in the minimally parallel mode.","Title: Morphisms and Weak-Ratio Property.\nYear: 2014\n\nAbstract: The concept of Parikh matrix of a word was introduced by Mateescu et al. [8] to count scattered subwords in a given word over an ordered alphabet. Motivated by a recent study of Atanasiu [3], on this notion using morphisms, we consider, in this paper, general morphisms on words and obtain various properties related to subwords, in the context of a weak-ratio property considered in [12].","Title: Uniform Solution to Common Algorithmic Problem in the Minimally Parallel Mode.\nYear: 2014\n\nAbstract: A P system is a novel computing model introduced by Paun in the area of membrane computing. It is known that the Common Algorithmic Problem (CAP) has a nice property that several other NPcomplete problems can be reduced to it in linear time. The decision version of this problem is known to be efficiently solved with a family of recognizer P systems with active membranes with three electrical charges working in the maximally parallel way. We here work with a variant of a P system with active membranes that does not use polarizations and present a uniform solution to CAP in the minimally parallel mode.","Title: Completeness, accuracy, and computability of National Quality Forum-specified eMeasures.\nYear: 2015\n\nAbstract: Objective To analyze the completeness, computability, and accuracy of specifications for five National Quality Forum-specified (NQF) eMeasures spanning ambulatory, post-discharge, and emergency care within a comprehensive, integrated electronic health record (EHR) environment. Materials and methods To evaluate completeness, we assessed eMeasure logic, data elements, and value sets. To evaluate computability, we assessed the translation of eMeasure algorithms to programmable logic constructs and the availability of EHR data elements to implement specified data criteria, using a de-identified clinical data set from Kaiser Permanente Northwest. To assess accuracy, we compared eMeasure results with those obtained independently by existing audited chart abstraction methods used for external and internal reporting. Results One measure specification was incomplete; missing applicable LOINC codes rendered it non-computable. For three of four computable measures, data availability issues occurred; the literal specification guidance for a data element differed from the physical implementation of the data element in the EHR. In two cases, cross-referencing specified data elements to EHR equivalents allowed variably accurate measure computation. Substantial data availability issues occurred for one of the four computable measures, producing highly inaccurate results. Discussion Existing clinical workflows, documentation, and coding in the EHR were significant barriers to implementing eMeasures as specified. Implementation requires redesigning business or clinical practices and, for one measure, systemic EHR modifications, including clinical text search capabilities. Conclusions Five NQF eMeasures fell short of being machine-consumable specifications. Both clinical domain and technological expertise are required to implement manually intensive steps from data mapping to text mining to EHR-specific eMeasure implementation.","Title: Contextual array grammars and array P systems\nYear: 2015\n\nAbstract: Contextual array grammars, with selectors not having empty cells, are considered. A P system model, called , that makes use of array objects and contextual array rules, is introduced and its generative power for the description of picture arrays is examined. A main result of the paper is that there is a proper infinite hierarchy with respect to the classes of languages described by contextual array P systems. Such a hierarchy holds as well in the case when the selector is also endowed with the #−sensing ability.","Title: Stochastic table arrays\nYear: 1982\n\nAbstract: A probabilistic mechanism is attached to a parallel/sequential model of rectangular arrays with tables. Rules within each table are deterministic in the sense that there is a single right-linear rule for each nonterminal in a table. Applications of the model to syntactic pattern recognition are indicated. An automation is constructed to recognize the class of pictures generated.","Title: Intelligent Evacuation Management Systems: A Review.\nYear: 2016\n\nAbstract: Crowd and evacuation management have been active areas of research and study in the recent past. Various developments continue to take place in the process of efficient evacuation of crowds in mass gatherings. This article is intended to provide a review of intelligent evacuation management systems covering the aspects of crowd monitoring, crowd disaster prediction, evacuation modelling, and evacuation path guidelines. Soft computing approaches play a vital role in the design and deployment of intelligent evacuation applications pertaining to crowd control management. While the review deals with video and nonvideo based aspects of crowd monitoring and crowd disaster prediction, evacuation techniques are reviewed via the theme of soft computing, along with a brief review on the evacuation navigation path. We believe that this review will assist researchers in developing reliable automated evacuation systems that will help in ensuring the safety of the evacuees especially during emergency evacuation scenarios.","Title: The Impact of Bio-Inspired Approaches Toward the Advancement of Face Recognition\nYear: 2015\n\nAbstract: An increased number of bio-inspired face recognition systems have emerged in recent decades owing to their intelligent problem-solving ability, flexibility, scalability, and adaptive nature. Hence, this survey aims to present a detailed overview of bio-inspired approaches pertaining to the advancement of face recognition. Based on a well-classified taxonomy, relevant bio-inspired techniques and their merits and demerits in countering potential problems vital to face recognition are analyzed. A synthesis of various approaches in terms of key governing principles and their associated performance analysis are systematically portrayed. Finally, some intuitive future directions are suggested on how bio-inspired approaches can contribute to the advancement of face biometrics in the years to come.","Title: Picture Array Generation Using Pure 2D Context-Free Grammar Rules.\nYear: 2015\n\nAbstract: Pure two-dimensional context-free grammar P2DCFG is a simple but effective non-isometric 2D grammar model to generate picture arrays. This 2D grammar uses only one kind of symbol as in a pure string grammar and rewrites in parallel all the symbols in a column or row by a set of context-free type rules. P2DCFG and a variant called l﾿/﾿uP2DCFG, which was recently introduced motivated by the \\\"leftmost\\\" rewriting mode in string context-free grammars, have been investigated for different properties. In this paper we introduce another variant of P2DCFG that corresponds to \\\"rightmost\\\" rewriting in string context-free grammars. The resulting grammar is called r﾿/﾿dP2DCFG and rewrites in parallel all the symbols only in the rightmost column or the lowermost row of a picture array by a set of context-free type rules. Unlike the case of string context-free grammars, the picture language families of P2DCFG and the two variants l﾿/﾿uP2DCFG and r﾿/﾿dP2DCFG are mutually incomparable, although they are not disjoint. We also examine the effect of regulating the rewriting in a r﾿/﾿dP2DCFG by suitably adapting two well-known control mechanisms in string grammars, namely, control words and matrix control.","Title: A Membrane Computing Model for Generation of Picture Arrays.\nYear: 2015\n\nAbstract: In the bio-inspired area of membrane computing, P system is a versatile model providing a rich framework for many computational problems. Array P system and its variant with parallel rewriting facilitate the study of picture languages within this area of membrane computing. Here another variant of array P system, called tabled parallel array P system (TPAP), is introduced, by endowing it with the features of parallel rewriting and tables of array-rewriting rules. The generative power of TPAP as well as the ability of this system in describing picture patterns are investigated.","Title: P Systems with Parallel Rewriting for Chain Code Picture Languages\nYear: 2015\n\nAbstract: Chain code pictures are composed of unit lines in the plane, drawn according to a sequence of instructions left, right, up, down codified by words over Sigma = {l, r, u, d}. P systems to generate such languages have been considered in previous work with sequential rewriting in the membranes. We consider here parallel rewriting, with the advantage of reducing the number of membranes. We also consider the problem of generating the finite approximations of space- filling curves, the Hilbert curve and the Peano curve.","Title: A Picture Array Generating Model Based on Flat Splicing Operation.\nYear: 2015\n\nAbstract: The bio-inspired operations of linear and circular splicing respectively on linear and circular strings of symbols have been extensively investigated by many researchers for their theoretical properties. Recently, another kind of splicing of two words, referred to as flat splicing on strings, has been considered. We here extend this operation to flat splicing on picture arrays, thus defining a new model of picture generation, which we call as array flat splicing system (AFS) and obtain some results on the generative power of AFS in comparison with certain well-known picture array defining models.","Title: Non-isometric Contextual Array Grammars with Regular Control and Local Selectors\nYear: 2015\n\nAbstract: We consider the external variant of non-isometric d-dimensional contextual array grammars with regular control together with local selectors allowing for controlling how d-dimensional arrays are evolving by adjoining rectangular (d-1)-dimensional arrays. In the 1-dimensional case, the computational power of these non-isometric contextual array grammars with regular control and local selectors equals the computational power of isometric contextual array grammars with regular control. The string images of the langauges of 1-dimensional arrays generated by these contextual array grammars exactly yield the linear languages. In the more-dimensional case, non-isometric d-dimensional contextual array grammars with regular control and local selectors can simulate the computations of (d - 1)-dimensional array grammars or Turing machines. Hence, for example, the emptiness problem for non-isometric d-dimensional contextual array grammars with regular control and local selectors for d > 1 is undecidable.","Title: Contextual Array Grammars with Matrix and Regular Control.\nYear: 2016\n\nAbstract: We investigate the computational power of d-dimensional contextual array grammars with matrix control and regular control languages. For (dge 2), d-dimensional contextual array grammars are less powerful than matrix contextual array grammars, which themselves are less powerful than contextual array grammars with regular control languages. Yet in the 1-dimensional case, for a one-letter alphabet, the family of 1-dimensional array languages generated by contextual array grammars with regular control languages coincides with the family of regular 1-dimensional array languages, whereas for alphabets with more than one letter, we obtain the array images of the linear languages.","Title: Flat splicing context free array grammar systems generating picture arrays.\nYear: 2016\n\nAbstract: In the area of DNA computing, a biologically motivated operation called splicing on linear words as well as circular words was introduced by Head 1987, 1992. Motivated by this operation, the concept of flat splicing on words was introduced by Berstel et al. 2012. This operation was extended by Subramanian et al. 2015a to picture arrays. Based on flat splicing on picture arrays, a new model of picture generation known as flat splicing regular array grammar system FSRAGS was also proposed by Subramanian et al. 2015b. Here we extend this model by considering a well-known two-phase context-free grammar in the components of the grammar system with the components deriving picture arrays in parallel using the rules and with two different components communicating using the array flat splicing operations. We call the resulting model as flat splicing context-free array grammar system FSCFAGS and obtain certain theoretical results on the generative power of FSCFAGS. As an application we construct a FSCFAGS and a corresponding algorithm to generate a certain class of floor design, also called a kolam pattern.","Title: Matrix Flat Splicing Systems.\nYear: 2016\n\nAbstract: In dealing with the problem of modelling DNA recombination, the operation of splicing on linear and circular strings of symbols was introduced. Inspired by splicing on circular strings, the operation of flat splicing on a pair of strings (u, v) was considered. This operation involves “cutting” u at a specified position and “inserting” v into it, with v having a pre-specified prefix as well as suffix defined by a flat splicing rule. In this work, we consider a well-known technique in formal language theory, known as “matrix of rules”, and introduce matrix of flat splicing rules and thus define matrix flat splicing system (MFSS). Some results on the language generative power of MFSS are provided. An application of MFSS in the generation of chain code pictures is also pointed out.","Title: Array-Rewriting P Systems with Basic Puzzle Grammar Rules and Permitting Features.\nYear: 2016\n\nAbstract: Motivated by the problem of tiling the plane, puzzle grammars were introduced as a mechanism for generating languages of picture arrays in the two-dimensional plane. On the other hand BPG array P system with array objects and basic puzzle grammar (BPG) rules was introduced as a variant of array generating P systems that were developed with a view to link the two areas of membrane computing and picture array grammars. Here we incorporate the feature of permitting symbols in the rules of the BPG array P system, thus introducing permitting array P system with BPG rules (pAP(BPG)). We show that the permitting feature gives more generative power to the BPG array P system with one or two membranes. We also show that a pAP(BPG) with only two membranes under t-communication mode, can generate picture arrays of square frames.","Title: Rewriting P Systems with Flat-Splicing Rules.\nYear: 2016\n\nAbstract: Rewriting P systems, as language generating devices, are one of the earliest classes of P systems with structured strings as objects and the rewriting rules as evolution rules. Flat splicing is an operation on strings, inspired by a splicing operation on circular strings. In this work, we consider a variant of rewriting P systems with only regular or linear rewriting rules and alphabetic flat splicing rules, and the language generative power of rewriting P systems with flat splicing rules in comparison with flat splicing systems and Chomsky hierarchy is investigated.","Title: Contextual array grammars with matrix control, regular control languages, and tissue P systems control.\nYear: 2017\n\nAbstract: We consider d-dimensional contextual array grammars and investigate their computational power when using various control mechanisms – matrices, regular control languages, and tissue P systems, which work like regular control languages, but may end up with a final check for the non-applicability of some rules. For d≥2, d-dimensional contextual array grammars are less powerful than matrix contextual array grammars, which themselves are less powerful than contextual array grammars with regular control languages. The use of tissue P systems with their final non-applicability check even yields some additional computational power.","Title: Non-Isometric Contextual Array Grammars and the Role of Regular Control and Local Selectors.\nYear: 2017\n\nAbstract: We consider the external variant of non-isometric d-dimensional contextual array grammars with regular control together with local selectors allowing for controlling how d-dimensional arrays are evolving by adjoining rectangular (d - 1)-dimensional arrays. In the 1-dimensional case, the computational power of these non-isometric contextual array grammars with regular control and local selectors equals the computational power of isometric contextual array grammars with regular control. The string images of the languages of 1-dimensional arrays generated by these contextual array grammars exactly yield the linear languages. In the more-dimensional case, non-isometric d-dimensional contextual array grammars with regular control and local selectors can simulate the computations of (d - 1)-dimensional array grammars or Turing machines. Hence, for example, the emptiness problem for non-isometric d-dimensional contextual array grammars with regular control and local selectors for d > 1 is undecidable. We also compare the computational power of all variants of non-isometric d-dimensional contextual array grammars that we introduce to each other.","Title: Language generating alphabetic flat splicing P systems.\nYear: 2018\n\nAbstract: An operation on strings, called flat splicing was introduced, inspired by a splicing operation on circular strings considered in the study of modelling of the recombinant behaviour of DNA molecules. A simple kind of flat splicing, called alphabetic flat splicing, allows insertion of a word with a specified start symbol and/or a specified end symbol, between two pre-determined symbols in a given word. In this work, we consider a P system with only alphabetic flat splicing rules as the evolution rules and strings of symbols as objects in its regions. We examine the language generative power of the resulting alphabetic flat splicing P systems (AFS P systems, for short). In particular, we show that AFS P systems with two membranes are more powerful in generative power than AFS P systems with a single membrane. We also construct AFS P systems with at most three membranes to generate languages that do not belong to certain other language classes and show an application to generation of chain code pictures.","Title: Properties Of Parikh Matrices Of Binary Words Obtained By An Extension Of A Restricted Shuffle Operator\nYear: 2018\n\nAbstract: We introduce an extension of the restricted shuffle operator on binary words considered by Atanasiu and Teh (2016). We then derive properties on Parikh matrix equivalence of words over a binary alphabet based on this extended shuffle operator and a weak ratio property of words. We also examine the recently introduced concept of core Parikh matrix equivalence of binary words in the context of the restricted shuffle operator.","Title: Order of weak M-relation and Parikh matrices.\nYear: 2018\n\nAbstract: The classical inference problem on words studies whether a word can be uniquely inferred from some given binomial coefficients. The entries of the Parikh matrix are certain canonical binomial coefficients depending on the ordered alphabet. The wide open injectivity problem asks when two words are M-equivalent, that is, sharing the same Parikh matrix. Two words are weakly M-related if and only if they are M-equivalent with respect to some ordering on the alphabet. In this work, we introduce a function that measures how hard it is for two words to become weakly M-related. We characterize this function and solve various connected combinatorial questions.","Title: Chain Code P System For Generation Of Approximation Patterns Of Sierpinski Curve\nYear: 2018\n\nAbstract: A sequence of approximating geometric patterns that define in the limit, the Sierpinski space filling curve is considered and the problem of generation of the infinite set of these patterns is investigated. A P system model in the bio-inspired area of membrane computing is constructed to generate the language of chain code kind of words that describe the approximating geometric patterns of Sierpinski curve.","Title: Chain Code P System Generating a Variant of the Peano Space-Filling Curve.\nYear: 2018\n\nAbstract: Generation of the finite approximations of the well-known Hilbert and Peano space-filling curves, represented as chain-code words has been studied in an earlier work. The generation was done with parallel chain code P systems with objects as chain code words and rewriting with context-free rules in parallel. Continuing this line of work, finite approximations of a variant of the Peano curve considered by Wunderlich are generated here with parallel chain code P system. We also generate approximating polygons corresponding to the Peano curve with parallel chain code P system.","Title: One-Dimensional Array Grammars And P Systems With Array Insertion And Deletion Rules\nYear: 2013\n\nAbstract: We consider the (one-dimensional) array counterpart of contextual as well as insertion and deletion string grammars and consider the operations of array insertion and deletion in array grammars. First we show that the emptiness problem for P systems with (one-dimensional) insertion rules is undecidable. Then we show computational completeness of P systems using (one-dimensional) array insertion and deletion rules even of norm one only. The main result of the paper exhibits computational completeness of one-dimensional array grammars using array insertion and deletion rules of norm at most two.","Title: Derivation Languages of Splicing P Systems\nYear: 2017\n\nAbstract: Labelled splicing P systems are distributed parallel computing models, where sets of strings that evolve by splicing rules are labelled. In this work, we consider labelled splicing systems with the following modifications: (i) The strings in the membranes are present in arbitrary number of copies; (ii) the rules in the regions are finite in number. Results on the language family generated by the labelled splicing system in comparison with the language families of the Chomsky hierarchy, including recursively enumerable languages, are obtained, by involving only either one or two membranes in the P systems considered.","Title: Some results on Parikh word representable graphs and partitions\nYear: 2019\n\nAbstract: Consequent to the introduction of the concept of Parikh matrix of a word which is based on the notion of subwords of a word, there has been an extensive research and study based on subwords. Parikh word representable graph is one such notion which has been introduced in the recent times. On the other hand connections of partitions of a number with counts of certain subword in a binary word are known. In this paper we introduce the notion of dual of a word and investigate its relationship with conjugate partition. As a result of this study, an expression for the number of nonisomorphic Parikh word representable graphs with a given number of edges, is obtained. Several other properties of Parikh word representable graphs are also derived.","Title: Hilbert words as arrays generated with P systems\nYear: 2021\n\nAbstract: Construction of finite grammars to generate languages of digitized picture patterns, considered as arrays of symbols, has been a problem of interest in two-dimensional formal languages. On the other hand, in the area of membrane computing, P systems were developed for handling the problem of picture array generation, with the rewriting involved being sequential or parallel. We introduce in this paper the array representation for the Hilbert words, the finite approximations of the Hilbert space-filling curve, and we generate them with array-rewriting rules in P systems. The array rewriting is done in parallel, with the P system serving as a control mechanism. A main contribution is the proof of correctness which is done using a linearization procedure. In addition, the advantage of the P system used is that the number of membranes involved is small (only one or two).","Title: Control languages accepted by labeled spiking neural P systems with rules on synapses\nYear: 2021\n\nAbstract: Spiking neural P systems with rules on synapses (RSSNP systems) are a class of computation models which are inspired by the information processing and communication manner of neurons. In this work, we consider labeled RSSNP systems (lRSSNP systems), where each rule is assigned either with a label chosen from an alphabet Sigma or with the empty label lambda. A string over an alphabet is accepted by an lRSSNP system if the string is processed from left to right by the system in the sense that in a step only rules labeled with the processed symbol or with lambda are used (not both), and the system reaches a final configuration in the moment when the whole string is processed. The set of all accepted strings is categorized as the restricted (computations are done with the application of the rules labeled with symbols from the given alphabet) control language and the unrestricted (rules labeled with lambda are allowed in the computations) control language of the system. We study the language accepting power of lRSSNP systems using only standard spiking rules by comparing the family of control languages of lRSSNP systems with the families of languages in the Chomsky hierarchy. It is proved that restricted lRSSNP systems can accept no more than context-sensitive languages, and unrestricted lRSSNP systems can accept recursively enumerable languages. (c) 2021 Elsevier B.V. All rights reserved.","Title: Array P systems and pure 2D context-free grammars with independent mode of rewriting\nYear: 2022\n\nAbstract: Rewriting array P systems for generation of rectangular picture arrays have been considered with the rules in membranes and the application of the rules as in a pure 2D context-free grammar (P2DCFG) and its variants. Here, we introduce in P2DCFG, a different mode of rewriting of an array, which we call as independent mode. We then consider rewriting array P systems involving P2DCFG type of rules but with the independent mode of rewriting. We show that the array generative power is increased in the framework of P systems. This framework also allows for the treatment of the so-called “extended” array grammars."],"string":"[\n \"Title: Lyndon trees\\nYear: 1992\\n\\nAbstract: Using deterministic (sample-path) analysis, we generalize andextend fundamental properties of systems with “stationarydeterministic flows” as introduced byGelenbe (1983) and Gelenbeand Finkel ...\",\n \"Title: A Note on Parallel Splicing on Images\\nYear: 2001\\n\\nAbstract: The concept of splicing on images which is done in parallel is introduced. This is an extension of the operation of splicing on strings extensively studied in the context of DNA computing. Various properties of splicing on images are examined.\",\n \"Title: Infinite Lyndon words\\nYear: 1994\\n\\nAbstract: We define an infinite Lyndon word as the limit of an increasing sequence of prefix preserving Lyndon words and show that some of the interesting properties of Lyndon words generalize to the infinite case. We construct a queue automaton that recognizes the set of Lyndon words and show that it can be extended to recognize infinite Lyndon words. We discuss certain topological properties of the set of infinite Lyndon words such as homeomorphism with a subspace of the Cantor space.\",\n \"Title: Hexagonal Pattern Languages\\nYear: 2004\\n\\nAbstract: Hexagonal tiling systems, hexagonal local picture languages and hexagonal recognizable picture languages are defined in this\\n paper. Hexagonal Wang tiles and systems are also introduced. It is noticed that the family of hexagonal picture languages\\n defined by hexagonal Wang systems coincides with the family of hexagonal picture languages recognized by hexagonal tiling\\n system. Similar to hv-domino systems describing rectangular arrays, we define xyz-domino systems and characterize hexagonal\\n picture languages using this. Unary hexagonal picture languages are also considered and we analyze some of their properties.\",\n \"Title: On Parikh Matrices\\nYear: 2009\\n\\nAbstract: Mateescu et al (2000) introduced an interesting new tool, called Parikh matrix, to study in terms of subwords, the numerical properties of words over an alphabet. The Parikh matrix gives more information than the well-known Parikh vector of a word which counts only occurrences of symbols in a word. In this note a property of two words u, v, called \\\"ratio property\\\", is introduced. This property is a sufficient condition for the words uv and vu to have the same Parikh matrix. Thus the ratio property gives information on the M-ambiguity of certain words and certain sets of words. In fact certain regular, context-free and context-sensitive languages that have the same set of Parikh matrices are exhibited. In the study of fair words, Cerny (2006) introduced another kind of matrix, called the p-matrix of a word. Here a \\\"weak-ratio property\\\" of two words u, v is introduced. This property is a sufficient condition for the words uv and vu to have the same p-matrix. Also the words uv and vu are fair whenever u, v are fair and have the weak ratio property.\",\n \"Title: Array P Systems and t.Communication\\nYear: 2009\\n\\nAbstract: The two areas of grammar systems and P systems, which have provided interesting computational models in the study of formal string language theory have been in the recent past effectively linked in [4] by incorporating into P systems, a communication mode called t-mode of cooperating distributed grammar systems. On the other hand cooperating array grammar systems [5] and array P systems [1] have been developed in the context of two-dimensional picture description. In this paper, motivated by the study of [4], these two systems are studied by linking them through the t-communication mode, thus bringing out the picture description power of these systems.\",\n \"Title: Array P systems with permitting features.\\nYear: 2014\\n\\nAbstract: The biologically inspired model, known as P system, has proved to be a general framework for investigating several problems related to computing in different fields. Two-dimensional picture array languages is one such area in which different kinds of P systems have been constructed for picture array generation. Incorporating the feature of permitting symbols in the rules, a new variety of array P system is constructed here for generating picture languages consisting of picture arrays. The advantage of this approach is that there is a reduction in the number of membranes used in the construction, in comparison to the existing array P system models. (C) 2013 Elsevier B.V. All rights reserved.\",\n \"Title: Cell-Work Ol-Systems With Fins\\nYear: 1999\\n\\nAbstract: Cell-works which are three-dimensional cyclic edge-label controlled OL-systems, were introduced by Lindenmayer. Fracchia and Prusinkiewicz provided cell-work systems using markers to model three-dimensional cellular structures. In this paper, cell-work systems with fins are proposed, generalizing the notion of map with handles. We allow each cell of a cell-work to divide into finitely many cells at any instant of time. This generalization enables us to describe three-dimensional images and fractals. A comparison of various systems with the system introduced in this paper is made.\",\n \"Title: Array-rewriting P systems\\nYear: 2003\\n\\nAbstract: We consider array languages (sets of picturesconsisting of symbols placed in the lattice points of the 2D grid) and thepossibility to handle them with P systems. After proving binary normal formsfor array matrix grammars (which, even in the case when no appearance checking isused, are known to generate the array languages of arbitrary array grammars), weprove that the P systems with context-free rules (with three membranes and no control on the communication or the use of rules) are computationally universal, able togenerate all computable array languages. Some open problems are also formulated.\",\n \"Title: Infinite Arrays And Recognizability\\nYear: 2000\\n\\nAbstract: In this paper, the concept of local languages is extended to infinite arrays and omega omega-local languages are defined. ww-recognizable languages of infinite arrays accepted by online tesselation automata are considered. Properties of these languages are studied. The notion of Muller recognizability is extended to infinite arrays. This is related to acceptance of infinite arrays by online tesselation automata.\",\n \"Title: Selective substitution array grammars\\nYear: 1981\\n\\nAbstract: Selective substitution array grammars are introduced, which abstract the notions of rewriting rules, direct derivation steps, derivations and rewriting in a sequential or parallel way. These array grammars provide for a unified framework for many of the two-dimensional array grammars in the literature. In addition, they point towards new interpretations resulting in possible new array grammars.\",\n \"Title: Two-dimensional pattern languages.\\nYear: 2017\\n\\nAbstract: We introduce several classes of array languages obtained by generalising Angluin's pattern languages to the two-dimensional case. These classes of two-dimensional pattern languages are compared with respect to their expressive power and their closure properties are investigated.\",\n \"Title: On Watson-Crick automata\\nYear: 2012\\n\\nAbstract: Motivated by Watson-Crick automata, we introduce here a grammar counterpart called a Watson-Crick regular grammar that has rules as in a regular grammar but involves double stranded strings. The language generated by this grammar consists of strings in the upper strands of the double stranded strings related by a complementarity relation. We obtain some properties of these grammars and also indicate an application to generation of chain code pictures.\",\n \"Title: Parallel Splicing On Images\\nYear: 2004\\n\\nAbstract: In this paper, splicing on images of rectangular arrays is introduced as a simple and effective extension of the operation of splicing on strings extensively studied in the context of DNA computing. A comparison of the resulting class of splicing array languages with the local two-dimensional array languages and two-dimensional right linear languages is made. Certain closure properties are obtained. Furthermore, the notion of self cross-over of strings is extended to arrays with respect to the splicing of arrays introduced here.\",\n \"Title: Permitting Features in P Systems Generating Picture Arrays.\\nYear: 2012\\n\\nAbstract: In the area of membrane computing, the biologically inspired model known as P system has proved to be a rich framework for studying several types of problems. Picture array generation is one such problem for which different P systems have been constructed in the literature. Incorporating the feature of permitting symbols in the rules, array P systems are constructed here for generating picture languages consisting of picture arrays. The advantage of this approach is that there is a reduction in the number of membranes used in the construction, in comparison to the existing array P system model.\",\n \"Title: Tissue-like P Systems with Active Membranes for Picture Generation\\nYear: 2003\\n\\nAbstract: We propose a variant of tissue-like P systems with symport rules and active membranes that generates two-dimensional picture languages. The method is unconventional in that, instead of using the membranes as regions for computation/writing, it uses them to hold elements of pictures. Thus, the picture itself, and actually the whole supporting rectangular grid, is composed of membranes, each one containing (among other symbols) a letter of the picture's alphabet. The method is illustrated by its application to the generation of local and recognizable picture languages.\",\n \"Title: P Systems Generating Hexagonal Picture Languages\\nYear: 2003\\n\\nAbstract: P systems generating rectangular arrays have been recently introduced in [1,2,3], thus bringing together the two areas of theoretical computer science, namely membrane computing and picture grammars. In this paper, hexagonal arrays on triangular grids are considered and the capabilities of the three approaches [1,2,3] that construct P systems for generation of arrays on rectangular grids, to handle arrays on triangular grids, are demonstrated.\",\n \"Title: Learning code regular and code linear languages\\nYear: 1996\\n\\nAbstract: In this paper, a subclass of regular languages, called code regular languages is defined. Algorithms for learning these languages are presented in the frame work of identification in the limit. Learning of analogous subclass of linear languages is also examined.\",\n \"Title: Array Insertion And Deletion P Systems\\nYear: 2013\\n\\nAbstract: We consider the (d-dimensional) array counterpart of string insertion and deletion grammars and use the operations of array insertion and deletion in the framework of P systems where the applicability of the rules depends on the membrane region. In this paper, we especially focus on examples of two-dimensional array insertion and deletion P systems and show that we can already obtain computational completeness using such P systems with a membrane structure of tree height of at most two and only the targets here, in, and out.\",\n \"Title: Pure 2D picture grammars (P2DPG) and P2DPG with regular control\\nYear: 2008\\n\\nAbstract: In this note a new model of grammatical picture generation is introduced. The model is based on the notion of pure context-free grammars of formal string language theory. The resulting model, called Pure 2D context-free grammar (CFG), generates rectangular picture arrays of symbols. The generative power of this model in comparison to certain other related models is examined. Also we associate a regular control language with a Pure 2D CFG and notice that the generative power increases. Certain closure properties are obtained.\",\n \"Title: Product Of Parikh Matrices And Commutativity\\nYear: 2012\\n\\nAbstract: The Parikh vector of a word enumerates the symbols of the alphabet that occur in the word. The Parikh matrix of a word which has been recently introduced, is an extension of the notion of Parikh vector and gives more numerical information about the word in terms of certain subwords. Intensive investigation on various theoretical properties of Parikh matrices has taken place. This paper deals with the problem of finding properties of words so that their Parikh matrices commute.\",\n \"Title: 2D and 3d pictural networks of evolutionary processors\\nYear: 2005\\n\\nAbstract: Networks of Evolutionary Processors constitute an interesting computing model motivated by cell biology. Pictural networks of evolutionary processors have been introduced to handle generation of pictures of rectangular arrays of symbols. Here we introduce an extension of this model to generate pictures that are arrays, not necessarily rectangular, and compare these with array generating Puzzle grammars. We also provide an extension to generate 3D rectangular arrays.\",\n \"Title: Tile pasting systems for tessellation and tiling patterns\\nYear: 2010\\n\\nAbstract: Pasting System (PS) and Extended Pasting Scheme (EPS) are new techniques to generate tiling patterns by using pasting rules for gluing tiles at their edges. In this paper we introduce a Tabled Pasting System (TPS) based on the technique of suitably grouping the rules of a pasting system into different sets of rules. We compare the generative power of TPS with PS and EPS. We propose theoretical models of P-system, called tile pasting P-system, with active membranes (TPPS−AM) and tissue-like evolution communication tile pasting P system (t−ECTPPS) to generate tiling patterns made of polygon tiles and compare the generative power of TPS with TPPS−AM and t−ECTPPS.\",\n \"Title: Tree Replacement and Public Key Cryptosystem\\nYear: 2002\\n\\nAbstract: In this paper, a public key cryptosystem based on tree replacement system is proposed.\",\n \"Title: A note on regular-controlled apical-growth filamentous systems\\nYear: 1985\\n\\nAbstract: The effect of regular control on Filamentous Systems with apical growth is examined. The resulting families of languages are classified with respect to regular, linear, context-free and pure languages.\",\n \"Title: Recognizing occluded faces by exploiting psychophysically inspired similarity maps\\nYear: 2013\\n\\nAbstract: The presence of occlusions in facial images is inevitable in unconstrained scenarios. However recognizing occluded faces remains a partially solved problem in computer vision. In this contribution we propose a novel Bayesian technique inspired by psychophysical mechanisms relevant to face recognition to address the facial occlusion problem. For some individuals certain facial regions, e.g. features comprising of some of the upper face, might be more discriminative than the rest of the features in the face. For others, it might be the features over the mid face and some of the lower face that are important. The proposed approach in this paper, will allow for such a psychophysical analysis to be factored into the recognition process. We have discovered and modeled similarity mappings that exist in facial domains by means of Bayesian Networks. The model can efficiently learn and exploit these mappings from the facial domain and hence capable of tackling uncertainties caused by occlusions. The proposed technique shows improved recognition rates over state of the art techniques.\",\n \"Title: PARALLEL CONTEXTUAL ARRAY GRAMMARS AND LANGUAGES\\nYear: 2003\\n\\nAbstract: In this paper a new method of description of pictures of digitized rectangular arrays is introduced based on contextual grammars. Parallel contextual array grammars are defined and their properties are studied.\",\n \"Title: Efficient Algorithms for Reconstruction of 2D-Arrays from Extended Parikh Images\\nYear: 2008\\n\\nAbstract: The concept of a Parikh matrix or an extended Parikh mapping of words introduced by Mateescu et al (2001) is formulated here for two-dimensional (2D) arrays. A polynomial time algorithm is proposed to reconstruct an unknown 2D-array over { 0,1 } from its image under the extended Parikh mapping along a single direction. On the other hand the problem of reconstructing a 2D-array over { 0,1 } from its image under the extended Parikh mapping along three or more directions is shown to be NP-hard. Also a polynomial time algorithm to reconstruct a 2D-array over {0,1} with a maximum number of ones close to the main diagonal of the array is presented by reducing the problem to Min-cost Max-flow problem.\",\n \"Title: On the power of permitting features in cooperating context-free array grammar systems\\nYear: 2013\\n\\nAbstract: Among many grammar models for picture description, cooperating context-free distributed array grammar systems utilize the mechanism of cooperation between component array grammars. Here, we endow these grammar systems with the feature of permitting symbols, and examine the generative capacity of the resulting permitting cooperating context-free distributed array grammar systems with certain other kinds of isometric and non-isometric array grammars. An advantage is that the number of components is reduced when this permitting feature is used in a cooperating context-free distributed array grammar system.\",\n \"Title: A P system model with pure context-free rules for picture array generation\\nYear: 2010\\n\\nAbstract: Syntactic models constitute one of the main areas of mathematical studies on picture array generation. A number of 2D grammar models have been proposed motivated by a variety of applications. A P system is a biologically motivated new computing model, proposed by Paun in the area of membrane computing. It is a rich framework for dealing with different problems, including the problem of handling picture array generation. In this paper, the generative power of the array-rewriting P system with pure 2D context-free rules is investigated by comparing it with other 2D grammar models, thus bringing out the suitability of this P system model for picture array generation.\",\n \"Title: Inferring Subclasses of Contextual Languages\\nYear: 2000\\n\\nAbstract: In this paper, we show that the languages generated by internal contextual grammars are not inferable from positive data only.\\n We define two subclasses of internal contextual languages, namely, k-uniform and strictly internal contextual languages which\\n are incomparable classes and provide an algorithm to learn these classes. The algorithm can be used when the rules are applied\\n in a parallel mode.\",\n \"Title: A Variant Of Pure Two-Dimensional Context-Free Grammars Generating Picture Languages\\nYear: 2014\\n\\nAbstract: Considering a large variety of approaches in generating picture languages, the notion of pure two-dimensional context-free grammar (P2DCFG) represents a simple yet expressive non-isometric language generator of picture arrays. In the present paper, we introduce a new variant of P2DCFGs that generates picture arrays in a leftmost way. We concentrate our attention on determining their generative power by comparing it with the power of other picture generators. We also examine the power of these generators that regulate rewriting by control languages.\",\n \"Title: Array Grammars with Contextual Operations\\nYear: 2008\\n\\nAbstract: In this paper, we consider a class of two-dimensional array grammars, called parallel contextual array grammars, that extend the contextual operations on strings to arrays in a naturalway and generate languages of pictures of rectangular arrays. Several classes of these array grammars and the resulting families of picture languages are considered. Necessary conditions for picture languages to be contained in these classes are obtained and the relations between these families are also established.\",\n \"Title: Infinite Arrays And Controlled Deterministic Table 0l Array Systems\\nYear: 1984\\n\\nAbstract: Deterministic table 0L array systems with control are considered for the generation of infinite arrays. Rewriting of a rectangular array is done in parallel by a table of rules, the rightmost edge horizontally or the lowermost edge vertically downwards. The application of the tables is controlled by a control set. Cube-free and square-free infinite arrays are obtained as an application of this model. The adherence of the array language of a controlled deterministic table 0L array system is related to the adherence of its control set. The limit language equivalence problem and the adherence equivalence problem are shown to be undecidable for this system.\",\n \"Title: A Tissue P System Based Solution To Quadratic Assignment Problem\\nYear: 2012\\n\\nAbstract: The quadratic assignment problem (QAP) is one of the fundamental combinatorial optimization problems, which models many real-life problems. However, it is considered as one of the most difficult NP-hard problems, which means that no polynomial-time algorithm is known to solve this intractable problem effectively. Even small instances of QAP may require vast computation time. In this work, a uniform cellular solution to QAP is proposed in the framework of membrane computing by using a family of recognizer tissue P systems with cell division. In the design of the solution, we encode the given instances in binary notations. The paper can be considered as a contribution to the study of considering a binary encoding of the information in P systems.\",\n \"Title: Cooperating basic puzzle grammar systems\\nYear: 2006\\n\\nAbstract: Nivat et al [3] introduced Context-free Puzzle grammars for generating connected picture arrays in the two-dimensional plane. Basic Puzzle grammars [6] constitute a subclass of these grammars. In this note we consider the Cooperating Array Grammar Systems introduced by Dassow et al [2] with Basic Puzzle grammar rules in the components instead of array grammar rules and examine the picture generating power of the resulting system, called, Cooperating Basic Puzzle Grammar System, in the maximal mode.\",\n \"Title: Circular DNA and Splicing Systems\\nYear: 1992\\n\\nAbstract: Circular strings representing DNA molecules and certain recombinant behaviour are formalized. Various actions of splicing schemes on linear and circular DNA molecules are examined. It is shown that there is a difference in the regularity result of Culik and Harju [1] between the linear and circular strings. A consequence of this result is that a conjecture of Head [4] that the circular string language of a splicing system under an action on circular strings is regular, when the set of initial circular strings is regular, is disproved.\",\n \"Title: P systems and picture languages\\nYear: 2007\\n\\nAbstract: Array-rewriting P systems were introduced in [2] linking the two areas of membrane computing and picture grammars. Subsequently a variety of P systems with array objects and different kinds of rewriting has been introduced. Here we discuss a few prominent systems among these, point out their features and indicate possible problems for future study.\",\n \"Title: P systems with shuffle operation.\\nYear: 2010\\n\\nAbstract: Shuffle operation on trajectories is useful in modeling parallel composition of words and languages. In this work, a new class of P systems with shuffle operation is presented. Such a system has language-objects and shuffle-operation rules in its regions. It can be used as a language generator; specifically it is used to generate languages from sets of languages. Some comparison results are obtained, which show that the power of shuffle operation is enlarged in the framework of P systems. © 2010 IEEE.\",\n \"Title: A note on regular kolam array grammars generating right triangles\\nYear: 1979\\n\\nAbstract: A group structure is identified among the Regular Kolam Array Grammars that generate the families of Kirsch's right triangles with labelled vertices and their reflections about the leg and/or base of the triangles.\",\n \"Title: On ambiguity of DTOL systems\\nYear: 1987\\n\\nAbstract: A DTOL system is unambiguous if no two different sequences of morphisms yield the same word from an axiom. A subfamily of DTOL systems with decidable ambiguity problem is exhibited. Four different sufficient conditions for a DTOL system to be unambiguous are formulated. These DTOL systems are very much suitable for the construction of public key cryptosystems based on L systems. We also prove that for DOL systems over a binary alphabet, the ambiguity problem is effectively decidable. This result has useful applications in the construction of public key cryptosystems which encrypt plain-texts over a binary alphabet using a TOL system obtained from an underlying unambiguous DOL system.\",\n \"Title: Psychophysically inspired bayesian occlusion model to recognize occluded faces\\nYear: 2011\\n\\nAbstract: Face recognition systems robust to major occlusions have wide applications ranging from consumer products with biometric features to surveillance and law enforcement applications. In unconstrained scenarios, faces are often subject to occlusions, apart from common variations such as pose, illumination, scale, orientation and so on. In this paper we propose a novel Bayesian oriented occlusion model inspired by psychophysical mechanisms to recognize faces prone to occlusions amidst other common variations. We have discovered and modeled similarity maps that exist in facial domains by means of Bayesian Networks. The proposed model is capable of efficiently learning and exploiting these maps from the facial domain. Hence it can tackle the occlusion uncertainty reasonably well. Improved recognition rates over state of the art techniques have been observed.\",\n \"Title: A note on an extension of matrix grammars generating two-dimensional languages\\nYear: 1985\\n\\nAbstract: An extended version of the parallel/sequential table matrix models called extended table matrix grammars (Ex-TMG), generating arrays of symbols, is proposed. The effect of imposing different types of control on the sequence of application of the tables is studied. The notion of assigning “terminal weights” to matrix grammars for describing parquet deformations is extended to Ex-TMGs.\",\n \"Title: Learning of recognizable picture languages\\nYear: 1992\\n\\nAbstract: Learning of certain classes of two-dimensional picture languages is considered. Linear time algorithms that learn in the limit, from positive data the classes of local picture languages and locally testable picture languages are presented. A crucial step for obtaining the learning algorithm for local picture languages is an explicit construction of a two-dimensional on-line tessellation acceptor for a given local picture language. An efficient algorithm that learns the class of recognizable picture languages from positve data and restricted subset queries, is presented in contrast to the fact that this class is not learnable in the limit from positive data alone.\",\n \"Title: On Some Classes of Splicing Languages\\nYear: 2004\\n\\nAbstract: We introduce some classes of splicing languages generated with simple and semi-simple splicing rules, in both, the linear and circular cases. We investigate some of their properties.\",\n \"Title: Tile Pasting P System Model for Pattern Generation\\nYear: 2009\\n\\nAbstract: In the area of membrane computing, a new computability model, called P system, which is a distributed, highly parallel theoretical computing model, was introduced by Păun, inspired from the cell structure and its functioning. There are several application areas of these P systems. Among these one area deals with the problem of picture array generation. Ceterchi et al. began a study on linking the two areas of membrane computing and picture grammars, which were not very much linked before, by relating P systems and array-rewriting grammars generating picture languages and proposing array-rewriting P systems. In this paper we propose a theoretical model of a P System, called Tile pasting P system, for generating two-dimensional patterns that are formed by gluing square tiles and study some of its properties.\",\n \"Title: Array P System Model with Pure Context-Free Rules\\nYear: 2009\\n\\nAbstract: Motivated by the structure and functioning of the livingcell, a new computing model, called P System, was proposedin the area of Membrane Computing. Among severalareas of application of P systems, one area is concernedwith generation of picture array languages. Ceterchiet al. [1] initially proposed array-rewriting P systemsby extending the notion of string-rewriting P-Systems [8].On the other hand a new theoretical model of grammaticalpicture generation called pure 2D context-free picturegrammar (P2DCFPG) generating rectangular picture arraysof symbols has been introduced by Subramanian et al.[17]. Here we consider P2DCFPG type of rules in arrayrewritingP systems and examine the generating power ofthe resulting array P systems.\",\n \"Title: A note on inferring uniquely terminating code languages\\nYear: 1999\\n\\nAbstract: A subclass of linear languages, called uniquely terminating code linear languages (utCLL), is introduced, motivated by the study of Makinen (1997), and an inference algorithm for utCLL is provided. The algorithm is extended to a corresponding subclass of equal matrix languages. (C) 1999 Elsevier Science B.V. All rights reserved.\",\n \"Title: Binary images, M-vectors, and ambiguity\\nYear: 2011\\n\\nAbstract: Mateescu et al (2001) introduced the notion of Parikh matrix of a word as an extension of the well-known concept of Parikh vector of a word. The Parikh matrix provides more numerical information about a word than given by the Parikh vector. Here we introduce the notion of M-vector of a binary word which allows us to have a linear notation in the form of a unique vector representation of the Parikh matrix of the binary word. We then extend this notion of M-vector to a binary image treating it as a binary array over a two-symbol alphabet. This is done by considering the M-vectors of the words in the rows and columns of the array. Among the properties associated with a Parikh matrix, M-ambiguity or simply ambiguity of a word is one which has been investigated extensively in the literature. Here M-ambiguity of a binary array is defined in terms of its M-vector and we obtain conditions for M-ambiguity of a binary array.\",\n \"Title: P Systems with Shuffle Operation and Catalytic-Like Rules.\\nYear: 2012\\n\\nAbstract: Shuffle operation on trajectories is useful in modeling parallel composition of words and languages. In this work, a new class of P systems with shuffle operation and catalytic-like rules is presented. Such a system has a membrane structure, where language-objects and shuffle-operation rules are placed in its regions. It can be used as a language generator. In this study, we propose a variant P system with shuffle operation on string-language objects. Some comparison results are obtained, which show that the power of shuffle operation is enlarged in the framework of P systems. Moreover, string-language objects are extended to array-language objects, and another variant P system with shuffle operation on picture-language objects is introduced. We also illustrate how to generate picture languages by using this kind of devices.\",\n \"Title: Two-Dimensional Digitized Picture Arrays And Parikh Matrices\\nYear: 2013\\n\\nAbstract: Parikh matrix mapping or Parikh matrix of a word has been introduced in the literature to count the scattered subwords in the word. Several properties of a Parikh matrix have been extensively investigated. A picture array is a two-dimensional connected digitized rectangular array consisting of a finite number of pixels with each pixel in a cell having a label from a finite alphabet. Here we extend the notion of Parikh matrix of a word to a picture array and associate with it two kinds of Parikh matrices, called row Parikh matrix and column Parikh matrix. Two picture arrays A and B are defined to be M-equivalent if their row Parikh matrices are the same and their column Parikh matrices are the same. This enables to extend the notion of M-ambiguity to a picture array. In the binary and ternary cases, conditions that ensure M-ambiguity are then obtained.\",\n \"Title: Self Cross-Over Array Languages.\\nYear: 2011\\n\\nAbstract: Self cross over array language consists of arrays that are formed by applying the operation of array splicing in H-array splicing systems to identical copies of picture arrays. This paper compares the family of self cross-over array languages with other two-dimensional language classes. Various properties analogous to H-array splicing systems are proved.\",\n \"Title: Some results on picture languages\\nYear: 1999\\n\\nAbstract: Two different types of models called puzzle languages and recognizable picture languages describing digitized pictures in a two-dimensional plane have been introduced in the 1990s. We review here these models, reporting the main results already proved. We also give a few new results.\",\n \"Title: Hyperbolic Tilings And Formal Language Theory\\nYear: 2013\\n\\nAbstract: In this paper, we try to give the appropriate class of languages to which belong various objects associated with tessellations in the hyperbolic plane.\",\n \"Title: Two-dimensional Picture Grammar models\\nYear: 2008\\n\\nAbstract: A new theoretical model of grammatical picture generation called extended 2D context-free picture grammar (E2DCFPG) generating rectangular picture arrays of symbols is introduced. This model which allows variables in the grammar and uses the squeezing mechanism of forming the picture language over terminal symbols, is an extension of the pure 2D context-free picture grammar (P2DCFPG)[13]. The extended picture grammar model E2DCFPG is shown to have more picture generative power than the P2DCFPG and certain other existing 2D models. Certain closure and other properties of this new model are also examined.\",\n \"Title: Learning a Subclass of Context-Free Languages\\nYear: 1998\\n\\nAbstract: In this paper, Apical growth Pure context-free grammars (AGPCFG) which are a variation of the pure grammars of Maurer et al\\n are introduced. These grammars allow rewriting of active symbols at the ends of a string simultaneously. They also provide\\n a variation of another kind of grammars called filamentous systems with apical growth, which are motivated by biological considerations.\\n The family of languages generated by AGPCFGs is a subclass of context-free languages. An algorithm for learning this language\\n subclass, in the framework of identification in the limit from positive examples, is provided.\",\n \"Title: Space-filling curves and infinite graphs\\nYear: 1982\\n\\nAbstract: Two-dimensional geometric patterns in the form of the space filling curves of Peano and Hilbert are represented as a string, an array and a graph. The sequence of patterns is such that each curve is a prefix of the succeeding one and hence we get prefix preserving WDOL to generate the corresponding string representation. The limit language is generable by a CDOL system and consists of a single infinite word corresponding to the Peano curve and two infinite words for the Hilbert curve. The property that in the sequence each curve is a prefix of the succeeding one helps in the definition of the infinite arrays and infinite graphs representing the infinite curve patterns and we construct a Context-free Parentheses Kolam array grammar and a coding of a graph DOL system to generate the sequence as well as the limit.\",\n \"Title: On the generative capacity of compound string and array grammars\\nYear: 1982\\n\\nAbstract: The operation of composition is a device introduced by Abraham [1] to increase the generative capacity of regular matrix grammars. But not all families of grammars have their generative capacity increased by this operation. Păun [5] has proved that composition increases the generative capacity of linear grammars but not of regular, CF, CS or PS grammars. In this note, we show that the generative power of k-linear (k ⩾ 1) grammars is increased by composition. In fact, it is of interest to note that the families of compound linear and compound k-linear languages are equal. Furthermore, the family of compound linear languages is an AFL, properly contained in the family of CFLs. Certain other families of string and array grammars whose generative capacities are increased by the operation of composition are also provided.\",\n \"Title: Context-Free Systems with a Complementarity Relation\\nYear: 2011\\n\\nAbstract: Motivated by the studies on Watson-Crick finite automata and Watson-Crick L systems, we introduce a generative device known as a Watson-Crick context-free system (WKCFS) making use of a complementarity relation and a context-free type of rewriting on double-stranded strings with the symbols in the string in the lower strand complementary to the symbols of the string in the upper strand. The language generated by a WKCFS is the set of all strings in the upper strands of the double-stranded strings (u, v) generated such that the strings u, v are of the same length and the corresponding symbols in the strings u, v are related by the complementarity relation. We compare the generative power of WKCFS with some families of languages and also certain variants of Watson-Crick finite automata.\",\n \"Title: Local And Recognizable Hexagonal Picture Languages\\nYear: 2005\\n\\nAbstract: In this paper we consider hexagonal arrays on triangular grids and introduce hexagonal local picture languages and hexagonal tiling systems defining hexagonal recognizable picture languages, motivated by an analogous study of rectangular arrays by Giammarresi and Restivo. We also introduce hexagonal Wang tiles to define hexagonal Wang systems (HWS) as a formalism to describe hexagonal picture languages. It is noticed that the family of hexagonal picture languages defined by hexagonal Wang systems and the family recognized by hexagonal tiling systems coincide. Analogous to hv-domino systems describing rectangular arrays, we define xyz-domino systems and prove that recognizable hexagonal picture languages are characterized as projections of xyz-local picture languages.\",\n \"Title: Public Key Cryptosystems Based on Free Partially Commutative Monoids and Groups\\nYear: 2003\\n\\nAbstract: A public key cryptosystem based on free partially commutative monoids is constructed. The encryption of A message to create the cryptotext uses a Thue system which is formed from the free partially commutative monoid with the help of a trapdoor morphism. The decidability of the word problem for free partially commutative monoids can be used for decryption. Finding the trapdoor morphism of this system is shown to be NP-haxd. But, a zero - knowledge protocol to convince a verifier that there is such a trapdoor. morphism is provided. A related but different public key cryptosystem based on free partially commutative groups is also proposed.\",\n \"Title: Subclasses of harmonic mappings defined by convolution\\nYear: 2010\\n\\nAbstract: Two new subclasses of harmonic univalent functions defined by convolution are introduced. The subclasses generate a number of known subclasses of harmonic mappings, and thus provide a unified treatment in the study of these subclasses. Sufficient coefficient conditions are obtained that are shown to be also necessary when the analytic parts of the harmonic functions have negative coefficients. Growth estimates and extreme points are also determined.\",\n \"Title: Pure 2D picture grammars and languages\\nYear: 2009\\n\\nAbstract: A new syntactic model, called pure two-dimensional (2D) context-free grammar (P2DCFG), is introduced based on the notion of pure context-free string grammar. The rectangular picture generative power of this 2D grammar model is investigated. Certain closure properties are obtained. An analogue of this 2D grammar model called pure 2D hexagonal context-free grammar (P2DHCFG) is also considered to generate hexagonal picture arrays on triangular grids.\",\n \"Title: Thue Systems and DNA - A Learning Algorithm for a Subclass\\nYear: 1993\\n\\nAbstract: Thue systems are considered to be appropriate for modeling certain kinds of behavior in DNA sequences [6]. We present a polynomial time algorithm for learning a Church-Rosser Thue system when the quotient monoid is finite, making use of membership queries related to the congruential languages, which specify regular languages.\",\n \"Title: Uniform Solution to Common Algorithmic Problem by P Systems Working in the Minimally Parallel Mode.\\nYear: 2015\\n\\nAbstract: It is known that the Common Algorithmic Problem (CAP) has the nice property that several other NP-complete problems can be reduced to it in linear time. The decision version of this problem is known to be efficiently solved by a family of recognizer P systems with active membranes with three electrical charges working in the maximally parallel way. We here work with a variant of P systems with active membranes without polarizations and present a uniform solution to CAP in the minimally parallel mode.\",\n \"Title: Morphisms and Weak-Ratio Property.\\nYear: 2014\\n\\nAbstract: The concept of Parikh matrix of a word was introduced by Mateescu et al. [8] to count scattered subwords in a given word over an ordered alphabet. Motivated by a recent study of Atanasiu [3], on this notion using morphisms, we consider, in this paper, general morphisms on words and obtain various properties related to subwords, in the context of a weak-ratio property considered in [12].\",\n \"Title: Uniform Solution to Common Algorithmic Problem in the Minimally Parallel Mode.\\nYear: 2014\\n\\nAbstract: A P system is a novel computing model introduced by Paun in the area of membrane computing. It is known that the Common Algorithmic Problem (CAP) has a nice property that several other NPcomplete problems can be reduced to it in linear time. The decision version of this problem is known to be efficiently solved with a family of recognizer P systems with active membranes with three electrical charges working in the maximally parallel way. We here work with a variant of a P system with active membranes that does not use polarizations and present a uniform solution to CAP in the minimally parallel mode.\",\n \"Title: Completeness, accuracy, and computability of National Quality Forum-specified eMeasures.\\nYear: 2015\\n\\nAbstract: Objective To analyze the completeness, computability, and accuracy of specifications for five National Quality Forum-specified (NQF) eMeasures spanning ambulatory, post-discharge, and emergency care within a comprehensive, integrated electronic health record (EHR) environment. Materials and methods To evaluate completeness, we assessed eMeasure logic, data elements, and value sets. To evaluate computability, we assessed the translation of eMeasure algorithms to programmable logic constructs and the availability of EHR data elements to implement specified data criteria, using a de-identified clinical data set from Kaiser Permanente Northwest. To assess accuracy, we compared eMeasure results with those obtained independently by existing audited chart abstraction methods used for external and internal reporting. Results One measure specification was incomplete; missing applicable LOINC codes rendered it non-computable. For three of four computable measures, data availability issues occurred; the literal specification guidance for a data element differed from the physical implementation of the data element in the EHR. In two cases, cross-referencing specified data elements to EHR equivalents allowed variably accurate measure computation. Substantial data availability issues occurred for one of the four computable measures, producing highly inaccurate results. Discussion Existing clinical workflows, documentation, and coding in the EHR were significant barriers to implementing eMeasures as specified. Implementation requires redesigning business or clinical practices and, for one measure, systemic EHR modifications, including clinical text search capabilities. Conclusions Five NQF eMeasures fell short of being machine-consumable specifications. Both clinical domain and technological expertise are required to implement manually intensive steps from data mapping to text mining to EHR-specific eMeasure implementation.\",\n \"Title: Contextual array grammars and array P systems\\nYear: 2015\\n\\nAbstract: Contextual array grammars, with selectors not having empty cells, are considered. A P system model, called , that makes use of array objects and contextual array rules, is introduced and its generative power for the description of picture arrays is examined. A main result of the paper is that there is a proper infinite hierarchy with respect to the classes of languages described by contextual array P systems. Such a hierarchy holds as well in the case when the selector is also endowed with the #−sensing ability.\",\n \"Title: Stochastic table arrays\\nYear: 1982\\n\\nAbstract: A probabilistic mechanism is attached to a parallel/sequential model of rectangular arrays with tables. Rules within each table are deterministic in the sense that there is a single right-linear rule for each nonterminal in a table. Applications of the model to syntactic pattern recognition are indicated. An automation is constructed to recognize the class of pictures generated.\",\n \"Title: Intelligent Evacuation Management Systems: A Review.\\nYear: 2016\\n\\nAbstract: Crowd and evacuation management have been active areas of research and study in the recent past. Various developments continue to take place in the process of efficient evacuation of crowds in mass gatherings. This article is intended to provide a review of intelligent evacuation management systems covering the aspects of crowd monitoring, crowd disaster prediction, evacuation modelling, and evacuation path guidelines. Soft computing approaches play a vital role in the design and deployment of intelligent evacuation applications pertaining to crowd control management. While the review deals with video and nonvideo based aspects of crowd monitoring and crowd disaster prediction, evacuation techniques are reviewed via the theme of soft computing, along with a brief review on the evacuation navigation path. We believe that this review will assist researchers in developing reliable automated evacuation systems that will help in ensuring the safety of the evacuees especially during emergency evacuation scenarios.\",\n \"Title: The Impact of Bio-Inspired Approaches Toward the Advancement of Face Recognition\\nYear: 2015\\n\\nAbstract: An increased number of bio-inspired face recognition systems have emerged in recent decades owing to their intelligent problem-solving ability, flexibility, scalability, and adaptive nature. Hence, this survey aims to present a detailed overview of bio-inspired approaches pertaining to the advancement of face recognition. Based on a well-classified taxonomy, relevant bio-inspired techniques and their merits and demerits in countering potential problems vital to face recognition are analyzed. A synthesis of various approaches in terms of key governing principles and their associated performance analysis are systematically portrayed. Finally, some intuitive future directions are suggested on how bio-inspired approaches can contribute to the advancement of face biometrics in the years to come.\",\n \"Title: Picture Array Generation Using Pure 2D Context-Free Grammar Rules.\\nYear: 2015\\n\\nAbstract: Pure two-dimensional context-free grammar P2DCFG is a simple but effective non-isometric 2D grammar model to generate picture arrays. This 2D grammar uses only one kind of symbol as in a pure string grammar and rewrites in parallel all the symbols in a column or row by a set of context-free type rules. P2DCFG and a variant called l﾿/﾿uP2DCFG, which was recently introduced motivated by the \\\\\\\"leftmost\\\\\\\" rewriting mode in string context-free grammars, have been investigated for different properties. In this paper we introduce another variant of P2DCFG that corresponds to \\\\\\\"rightmost\\\\\\\" rewriting in string context-free grammars. The resulting grammar is called r﾿/﾿dP2DCFG and rewrites in parallel all the symbols only in the rightmost column or the lowermost row of a picture array by a set of context-free type rules. Unlike the case of string context-free grammars, the picture language families of P2DCFG and the two variants l﾿/﾿uP2DCFG and r﾿/﾿dP2DCFG are mutually incomparable, although they are not disjoint. We also examine the effect of regulating the rewriting in a r﾿/﾿dP2DCFG by suitably adapting two well-known control mechanisms in string grammars, namely, control words and matrix control.\",\n \"Title: A Membrane Computing Model for Generation of Picture Arrays.\\nYear: 2015\\n\\nAbstract: In the bio-inspired area of membrane computing, P system is a versatile model providing a rich framework for many computational problems. Array P system and its variant with parallel rewriting facilitate the study of picture languages within this area of membrane computing. Here another variant of array P system, called tabled parallel array P system (TPAP), is introduced, by endowing it with the features of parallel rewriting and tables of array-rewriting rules. The generative power of TPAP as well as the ability of this system in describing picture patterns are investigated.\",\n \"Title: P Systems with Parallel Rewriting for Chain Code Picture Languages\\nYear: 2015\\n\\nAbstract: Chain code pictures are composed of unit lines in the plane, drawn according to a sequence of instructions left, right, up, down codified by words over Sigma = {l, r, u, d}. P systems to generate such languages have been considered in previous work with sequential rewriting in the membranes. We consider here parallel rewriting, with the advantage of reducing the number of membranes. We also consider the problem of generating the finite approximations of space- filling curves, the Hilbert curve and the Peano curve.\",\n \"Title: A Picture Array Generating Model Based on Flat Splicing Operation.\\nYear: 2015\\n\\nAbstract: The bio-inspired operations of linear and circular splicing respectively on linear and circular strings of symbols have been extensively investigated by many researchers for their theoretical properties. Recently, another kind of splicing of two words, referred to as flat splicing on strings, has been considered. We here extend this operation to flat splicing on picture arrays, thus defining a new model of picture generation, which we call as array flat splicing system (AFS) and obtain some results on the generative power of AFS in comparison with certain well-known picture array defining models.\",\n \"Title: Non-isometric Contextual Array Grammars with Regular Control and Local Selectors\\nYear: 2015\\n\\nAbstract: We consider the external variant of non-isometric d-dimensional contextual array grammars with regular control together with local selectors allowing for controlling how d-dimensional arrays are evolving by adjoining rectangular (d-1)-dimensional arrays. In the 1-dimensional case, the computational power of these non-isometric contextual array grammars with regular control and local selectors equals the computational power of isometric contextual array grammars with regular control. The string images of the langauges of 1-dimensional arrays generated by these contextual array grammars exactly yield the linear languages. In the more-dimensional case, non-isometric d-dimensional contextual array grammars with regular control and local selectors can simulate the computations of (d - 1)-dimensional array grammars or Turing machines. Hence, for example, the emptiness problem for non-isometric d-dimensional contextual array grammars with regular control and local selectors for d > 1 is undecidable.\",\n \"Title: Contextual Array Grammars with Matrix and Regular Control.\\nYear: 2016\\n\\nAbstract: We investigate the computational power of d-dimensional contextual array grammars with matrix control and regular control languages. For (dge 2), d-dimensional contextual array grammars are less powerful than matrix contextual array grammars, which themselves are less powerful than contextual array grammars with regular control languages. Yet in the 1-dimensional case, for a one-letter alphabet, the family of 1-dimensional array languages generated by contextual array grammars with regular control languages coincides with the family of regular 1-dimensional array languages, whereas for alphabets with more than one letter, we obtain the array images of the linear languages.\",\n \"Title: Flat splicing context free array grammar systems generating picture arrays.\\nYear: 2016\\n\\nAbstract: In the area of DNA computing, a biologically motivated operation called splicing on linear words as well as circular words was introduced by Head 1987, 1992. Motivated by this operation, the concept of flat splicing on words was introduced by Berstel et al. 2012. This operation was extended by Subramanian et al. 2015a to picture arrays. Based on flat splicing on picture arrays, a new model of picture generation known as flat splicing regular array grammar system FSRAGS was also proposed by Subramanian et al. 2015b. Here we extend this model by considering a well-known two-phase context-free grammar in the components of the grammar system with the components deriving picture arrays in parallel using the rules and with two different components communicating using the array flat splicing operations. We call the resulting model as flat splicing context-free array grammar system FSCFAGS and obtain certain theoretical results on the generative power of FSCFAGS. As an application we construct a FSCFAGS and a corresponding algorithm to generate a certain class of floor design, also called a kolam pattern.\",\n \"Title: Matrix Flat Splicing Systems.\\nYear: 2016\\n\\nAbstract: In dealing with the problem of modelling DNA recombination, the operation of splicing on linear and circular strings of symbols was introduced. Inspired by splicing on circular strings, the operation of flat splicing on a pair of strings (u, v) was considered. This operation involves “cutting” u at a specified position and “inserting” v into it, with v having a pre-specified prefix as well as suffix defined by a flat splicing rule. In this work, we consider a well-known technique in formal language theory, known as “matrix of rules”, and introduce matrix of flat splicing rules and thus define matrix flat splicing system (MFSS). Some results on the language generative power of MFSS are provided. An application of MFSS in the generation of chain code pictures is also pointed out.\",\n \"Title: Array-Rewriting P Systems with Basic Puzzle Grammar Rules and Permitting Features.\\nYear: 2016\\n\\nAbstract: Motivated by the problem of tiling the plane, puzzle grammars were introduced as a mechanism for generating languages of picture arrays in the two-dimensional plane. On the other hand BPG array P system with array objects and basic puzzle grammar (BPG) rules was introduced as a variant of array generating P systems that were developed with a view to link the two areas of membrane computing and picture array grammars. Here we incorporate the feature of permitting symbols in the rules of the BPG array P system, thus introducing permitting array P system with BPG rules (pAP(BPG)). We show that the permitting feature gives more generative power to the BPG array P system with one or two membranes. We also show that a pAP(BPG) with only two membranes under t-communication mode, can generate picture arrays of square frames.\",\n \"Title: Rewriting P Systems with Flat-Splicing Rules.\\nYear: 2016\\n\\nAbstract: Rewriting P systems, as language generating devices, are one of the earliest classes of P systems with structured strings as objects and the rewriting rules as evolution rules. Flat splicing is an operation on strings, inspired by a splicing operation on circular strings. In this work, we consider a variant of rewriting P systems with only regular or linear rewriting rules and alphabetic flat splicing rules, and the language generative power of rewriting P systems with flat splicing rules in comparison with flat splicing systems and Chomsky hierarchy is investigated.\",\n \"Title: Contextual array grammars with matrix control, regular control languages, and tissue P systems control.\\nYear: 2017\\n\\nAbstract: We consider d-dimensional contextual array grammars and investigate their computational power when using various control mechanisms – matrices, regular control languages, and tissue P systems, which work like regular control languages, but may end up with a final check for the non-applicability of some rules. For d≥2, d-dimensional contextual array grammars are less powerful than matrix contextual array grammars, which themselves are less powerful than contextual array grammars with regular control languages. The use of tissue P systems with their final non-applicability check even yields some additional computational power.\",\n \"Title: Non-Isometric Contextual Array Grammars and the Role of Regular Control and Local Selectors.\\nYear: 2017\\n\\nAbstract: We consider the external variant of non-isometric d-dimensional contextual array grammars with regular control together with local selectors allowing for controlling how d-dimensional arrays are evolving by adjoining rectangular (d - 1)-dimensional arrays. In the 1-dimensional case, the computational power of these non-isometric contextual array grammars with regular control and local selectors equals the computational power of isometric contextual array grammars with regular control. The string images of the languages of 1-dimensional arrays generated by these contextual array grammars exactly yield the linear languages. In the more-dimensional case, non-isometric d-dimensional contextual array grammars with regular control and local selectors can simulate the computations of (d - 1)-dimensional array grammars or Turing machines. Hence, for example, the emptiness problem for non-isometric d-dimensional contextual array grammars with regular control and local selectors for d > 1 is undecidable. We also compare the computational power of all variants of non-isometric d-dimensional contextual array grammars that we introduce to each other.\",\n \"Title: Language generating alphabetic flat splicing P systems.\\nYear: 2018\\n\\nAbstract: An operation on strings, called flat splicing was introduced, inspired by a splicing operation on circular strings considered in the study of modelling of the recombinant behaviour of DNA molecules. A simple kind of flat splicing, called alphabetic flat splicing, allows insertion of a word with a specified start symbol and/or a specified end symbol, between two pre-determined symbols in a given word. In this work, we consider a P system with only alphabetic flat splicing rules as the evolution rules and strings of symbols as objects in its regions. We examine the language generative power of the resulting alphabetic flat splicing P systems (AFS P systems, for short). In particular, we show that AFS P systems with two membranes are more powerful in generative power than AFS P systems with a single membrane. We also construct AFS P systems with at most three membranes to generate languages that do not belong to certain other language classes and show an application to generation of chain code pictures.\",\n \"Title: Properties Of Parikh Matrices Of Binary Words Obtained By An Extension Of A Restricted Shuffle Operator\\nYear: 2018\\n\\nAbstract: We introduce an extension of the restricted shuffle operator on binary words considered by Atanasiu and Teh (2016). We then derive properties on Parikh matrix equivalence of words over a binary alphabet based on this extended shuffle operator and a weak ratio property of words. We also examine the recently introduced concept of core Parikh matrix equivalence of binary words in the context of the restricted shuffle operator.\",\n \"Title: Order of weak M-relation and Parikh matrices.\\nYear: 2018\\n\\nAbstract: The classical inference problem on words studies whether a word can be uniquely inferred from some given binomial coefficients. The entries of the Parikh matrix are certain canonical binomial coefficients depending on the ordered alphabet. The wide open injectivity problem asks when two words are M-equivalent, that is, sharing the same Parikh matrix. Two words are weakly M-related if and only if they are M-equivalent with respect to some ordering on the alphabet. In this work, we introduce a function that measures how hard it is for two words to become weakly M-related. We characterize this function and solve various connected combinatorial questions.\",\n \"Title: Chain Code P System For Generation Of Approximation Patterns Of Sierpinski Curve\\nYear: 2018\\n\\nAbstract: A sequence of approximating geometric patterns that define in the limit, the Sierpinski space filling curve is considered and the problem of generation of the infinite set of these patterns is investigated. A P system model in the bio-inspired area of membrane computing is constructed to generate the language of chain code kind of words that describe the approximating geometric patterns of Sierpinski curve.\",\n \"Title: Chain Code P System Generating a Variant of the Peano Space-Filling Curve.\\nYear: 2018\\n\\nAbstract: Generation of the finite approximations of the well-known Hilbert and Peano space-filling curves, represented as chain-code words has been studied in an earlier work. The generation was done with parallel chain code P systems with objects as chain code words and rewriting with context-free rules in parallel. Continuing this line of work, finite approximations of a variant of the Peano curve considered by Wunderlich are generated here with parallel chain code P system. We also generate approximating polygons corresponding to the Peano curve with parallel chain code P system.\",\n \"Title: One-Dimensional Array Grammars And P Systems With Array Insertion And Deletion Rules\\nYear: 2013\\n\\nAbstract: We consider the (one-dimensional) array counterpart of contextual as well as insertion and deletion string grammars and consider the operations of array insertion and deletion in array grammars. First we show that the emptiness problem for P systems with (one-dimensional) insertion rules is undecidable. Then we show computational completeness of P systems using (one-dimensional) array insertion and deletion rules even of norm one only. The main result of the paper exhibits computational completeness of one-dimensional array grammars using array insertion and deletion rules of norm at most two.\",\n \"Title: Derivation Languages of Splicing P Systems\\nYear: 2017\\n\\nAbstract: Labelled splicing P systems are distributed parallel computing models, where sets of strings that evolve by splicing rules are labelled. In this work, we consider labelled splicing systems with the following modifications: (i) The strings in the membranes are present in arbitrary number of copies; (ii) the rules in the regions are finite in number. Results on the language family generated by the labelled splicing system in comparison with the language families of the Chomsky hierarchy, including recursively enumerable languages, are obtained, by involving only either one or two membranes in the P systems considered.\",\n \"Title: Some results on Parikh word representable graphs and partitions\\nYear: 2019\\n\\nAbstract: Consequent to the introduction of the concept of Parikh matrix of a word which is based on the notion of subwords of a word, there has been an extensive research and study based on subwords. Parikh word representable graph is one such notion which has been introduced in the recent times. On the other hand connections of partitions of a number with counts of certain subword in a binary word are known. In this paper we introduce the notion of dual of a word and investigate its relationship with conjugate partition. As a result of this study, an expression for the number of nonisomorphic Parikh word representable graphs with a given number of edges, is obtained. Several other properties of Parikh word representable graphs are also derived.\",\n \"Title: Hilbert words as arrays generated with P systems\\nYear: 2021\\n\\nAbstract: Construction of finite grammars to generate languages of digitized picture patterns, considered as arrays of symbols, has been a problem of interest in two-dimensional formal languages. On the other hand, in the area of membrane computing, P systems were developed for handling the problem of picture array generation, with the rewriting involved being sequential or parallel. We introduce in this paper the array representation for the Hilbert words, the finite approximations of the Hilbert space-filling curve, and we generate them with array-rewriting rules in P systems. The array rewriting is done in parallel, with the P system serving as a control mechanism. A main contribution is the proof of correctness which is done using a linearization procedure. In addition, the advantage of the P system used is that the number of membranes involved is small (only one or two).\",\n \"Title: Control languages accepted by labeled spiking neural P systems with rules on synapses\\nYear: 2021\\n\\nAbstract: Spiking neural P systems with rules on synapses (RSSNP systems) are a class of computation models which are inspired by the information processing and communication manner of neurons. In this work, we consider labeled RSSNP systems (lRSSNP systems), where each rule is assigned either with a label chosen from an alphabet Sigma or with the empty label lambda. A string over an alphabet is accepted by an lRSSNP system if the string is processed from left to right by the system in the sense that in a step only rules labeled with the processed symbol or with lambda are used (not both), and the system reaches a final configuration in the moment when the whole string is processed. The set of all accepted strings is categorized as the restricted (computations are done with the application of the rules labeled with symbols from the given alphabet) control language and the unrestricted (rules labeled with lambda are allowed in the computations) control language of the system. We study the language accepting power of lRSSNP systems using only standard spiking rules by comparing the family of control languages of lRSSNP systems with the families of languages in the Chomsky hierarchy. It is proved that restricted lRSSNP systems can accept no more than context-sensitive languages, and unrestricted lRSSNP systems can accept recursively enumerable languages. (c) 2021 Elsevier B.V. All rights reserved.\",\n \"Title: Array P systems and pure 2D context-free grammars with independent mode of rewriting\\nYear: 2022\\n\\nAbstract: Rewriting array P systems for generation of rectangular picture arrays have been considered with the rules in membranes and the application of the rules as in a pure 2D context-free grammar (P2DCFG) and its variants. Here, we introduce in P2DCFG, a different mode of rewriting of an array, which we call as independent mode. We then consider rewriting array P systems involving P2DCFG type of rules but with the independent mode of rewriting. We show that the array generative power is increased in the framework of P systems. This framework also allows for the treatment of the so-called “extended” array grammars.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":95,"string":"95"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13591,"string":"13,591"}}},{"rowIdx":13592,"cells":{"question_id":{"kind":"string","value":"Mahmoud Al Ayyoub"},"question":{"kind":"string","value":"Generate an abstract for the title \"QWERTY Keyboard? }.?BZQ is Better!\" using the following items: algorithm, placement optimization, keyboard, QWERTY, Dvorak Simplified."},"answer":{"kind":"string","value":"In this work, we provide a Genetic-based algorithm that is used to quickly find a placement for a set of objects within a given layout such that access to these objects is optimized. The given layout describes the free locations of the objects and the object handles and the access is done through a corpus of object requests. The proposed algorithm optimizes the placement of the objects by searching through a small fraction of the search space. As a case study, we use the algorithm to find a better placement for the keyboard characters than QWERTY and Dvorak Simplified characters placements. The algorithm finds a placement that is better than both QWERTY and Dvorak Simplified by 32.68% and 15.79% respectively on the training set, and 32.71% and 15.84% respectively on the testing set. This result is achieved after searching through only 500K possible solutions, which is about 1.23 × 10-19percent only of the total search space. Both training and testing sets are extracted randomly from TED2013 v1.1 English corpus. Moreover, we release the dataset, code and experimental results on our GitHub repository."},"correct_answer":{"kind":"string","value":"In this work, we provide a Genetic-based algorithm that is used to quickly find a placement for a set of objects within a given layout such that access to these objects is optimized. The given layout describes the free locations of the objects and the object handles and the access is done through a corpus of object requests. The proposed algorithm optimizes the placement of the objects by searching through a small fraction of the search space. As a case study, we use the algorithm to find a better placement for the keyboard characters than QWERTY and Dvorak Simplified characters placements. The algorithm finds a placement that is better than both QWERTY and Dvorak Simplified by 32.68% and 15.79% respectively on the training set, and 32.71% and 15.84% respectively on the testing set. This result is achieved after searching through only 500K possible solutions, which is about 1.23 × 10-19percent only of the total search space. Both training and testing sets are extracted randomly from TED2013 v1.1 English corpus. Moreover, we release the dataset, code and experimental results on our GitHub repository."},"documents":{"kind":"list like","value":["Title: An analytical study of Arabic sentiments: Maktoob case study\nYear: 2013\n\nAbstract: The problem of automatically extracting opinions and emotions from textual data have gained a lot of interest recently. Unfortunately, most studies on Sentiment Analysis (SA) focus mainly on the English language, whereas studies considering other important and wide-spread languages such as Arabic are few. Moreover, publicly-available Arabic datasets are seldom found on the Web. In this work, a labeled dataset of Arabic reviews/comments is collected from a social networking website (Yahoo!-Maktoob). A detailed analysis of different aspects of the collected dataset such as the reviews' length, the numbers of likes/dislikes, the polarity distribution and the languages used is presented. Finally, the dataset is used to test popular classifiers commonly used for SA.","Title: Automatic Arabic text categorization: A comprehensive comparative study\nYear: 2015\n\nAbstract: Text categorization or classification (TC) is concerned with placing text documents in their proper category according to their contents. Owing to the various applications of TC and the large volume of text documents uploaded on the Internet daily, the need for such an automated method stems from the difficulty and tedium of performing such a process manually. The usefulness of TC is manifested in different fields and needs. For instance, the ability to automatically classify an article or an email into its right class (Arts, Economics, Politics, Sports, etc.) would be appreciated by individual users as well as companies. This paper is concerned with TC of Arabic articles. It contains a comparison of the five best known algorithms for TC. It also studies the effects of utilizing different Arabic stemmers (light and root-based stemmers) on the effectiveness of these classifiers. Furthermore, a comparison between different data mining software tools (Weka and RapidMiner) is presented. The results illustrate the good accuracy provided by the SVM classifier, especially when used with the light10 stemmer. This outcome can be used in future as a baseline to compare with other unexplored classifiers and Arabic stemmers.","Title: CloudExp: A comprehensive cloud computing experimental framework.\nYear: 2014\n\nAbstract: Cloud computing is an emerging and fast-growing computing paradigm that has gained great interest from both industry and academia. Consequently, many researchers are actively involved in cloud computing research projects. One major challenge facing cloud computing researchers is the lack of a comprehensive cloud computing experimental tool to use in their studies. This paper introduces CloudExp, a modeling and simulation environment for cloud computing. CloudExp can be used to evaluate a wide spectrum of cloud components such as processing elements, data centers, storage, networking, Service Level Agreement (SLA) constraints, web-based applications, Service Oriented Architecture (SOA), virtualization, management and automation, and Business Process Management (BPM). Moreover, CloudExp introduces the Rain workload generator which emulates real workloads in cloud environments. Also, MapReduce processing model is integrated in CloudExp in order to handle the processing of big data problems.","Title: SDStorage: A Software Defined Storage Experimental Framework\nYear: 2015\n\nAbstract: With the rapid growth of data centers and the unprecedented increase in storage demands, the traditional storage control techniques are considered unsuitable to deal with this large volume of data in an efficient manner. The Software Defined Storage (SDStore) comes as a solution for this issue by abstracting the storage control operations from the storage devices and set it inside a centralized controller in the software layer. Building a real SDStore system without any simulation and emulation is considered an expensive solution and may have a lot of risks. Thus, there is a need to simulate such systems before the real-life implementation and deployment. In this paper we present SDStorage, an experimental framework to provide a novel virtualized test bed environment for SDStore systems. The main idea of SDStorage is based on the Mininet Software Defined Network (SDN) Open Flow simulator and is built over of it. The main components of Mininet, which are the host, the switch and the controller, are customized to serve the needs of SDStore simulation environments.","Title: To Build or Not to Build? Addressing the Expansion Strategies of Cloud Providers\nYear: 2014\n\nAbstract: With the increasing popularity gained by cloud computing systems, cloud providers are facing rapidly increasing traffic loads, which requires them to have proper expansion strategies for their datacenters. Expanding the capacities of existing datacenters or building new ones requires many factors to be considered such as power resources availability, prices (of power, land, etc.), carbon tax, free cooling options, and renewable energy. In this paper, we address the problem of deciding the best expansion strategy for a given cloud provider by deciding whether it is beneficial for the cloud provider to build new datacenters or to simply expand the datacenters it currently has.","Title: Compression-based arabic text classification\nYear: 2014\n\nAbstract: Text classification (TC) is one of the fundamental problems in text mining. Plenty of works exist on TC with interesting approaches and excellent results; however, most of these works follow a word-based approach for feature extraction. In this work, we are interested in an alternative (byte-based or character-based) approach known as compression-based TC (CTC). CTC has been used for some languages such as English and Portuguese and it is shown to have certain advantages/ disadvantages compared with word-based approaches. This work applies CTC on the Arabic language with the purpose of investigating whether these advantages/disadvantages exists for the Arabic language as well. The results are encouraging as they show the viability of using CTC for Arabic TC.","Title: Truthful Spectrum Auctions With Approximate Social-Welfare or Revenue\nYear: 2014\n\nAbstract: In cellular networks, a recent trend in research is to make spectrum access dynamic in the spatial and temporal dimensions for the sake of efficient utilization of spectrum. In one such model, the spectrum is divided into channels and periodically allocated to competing base stations using an auction-based market mechanism. An “efficient” auction mechanism is essential to the success of such a dynamic spectrum access model. A key objective in designing an auction mechanism is “truthfulness.” Combining this objective with an optimization of some social choice function (such as the social-welfare or the generated revenue) is highly desirable. In this paper, we design polynomial-time spectrum auction mechanisms that are truthful and yield an allocation with O(1)-approximate social-welfare or revenue. Our mechanisms generalize to general interference models. To the best of our knowledge, ours is the first work to design polynomial-time truthful spectrum auction mechanisms with a constant-factor approximation of either the expected revenue or the social-welfare. We demonstrate the performance of our designed mechanism through simulations.","Title: Collaboration Networks of Levantine Biomedical Researchers\nYear: 2014\n\nAbstract: Social networks and social network analysis are one of the hottest topics in Academia as well as in the industry nowadays. A collaboration (or a co-authorship) network, is a type of social networks in which the actors are scholars and researchers and a tie between two actors represents the fact that they have co-authored at least one paper together. The objective of this work is to study scientific collaborations patterns in order to construct and analyze a collaboration network of authors in the Levant region. The focus is on publications in the biomedical field. Even though collaboration networks have been constructed for several fields and in different regions, the Middle East and North Africa (MENA) region in general is still largely understudied, to the best of our knowledge. Constructing the network is challenging due to many reasons including the lack of sufficient data. Analyzing the constructed network reveal that there is an increase in both biomedical researchers and published papers that matches the spur of growth that the Levant-based academic institutions enjoyed in the early 2000s. Among many things, the analysis reveal that local researchers are shifting from working individually or with non-local collaborators to working within relatively large groups of local collaborators.","Title: A new broadcast scheme for sensor networks\nYear: 2014\n\nAbstract: The future of the Internet envisions connecting all objects around us in a proactive way in what is known as the Internet of Things (IoT). Wireless sensor networks (WSN) represent a prominent example of IoT. Broadcasting is one of most important procedures in wireless networks. However, broadcast may have immense effect on the network's performance in what is know as the broadcast storm problem. In this work, we propose a new broadcast scheme to address the broadcast storm problem. The proposed scheme is an amalgamation of three popular broadcast schemes: the probabilistic scheme, the distance-based scheme and the counter-based scheme. The proposed scheme is implemented and tested on the QualNet simulation environment where it is integrated into the Ad-hoc On-Demand Distance Vector (AODV) routing protocol. The results show that the percentage of saved rebroadcasts can reach more than 50% in some cases.","Title: The analysis of large-scale climate data: Jordan case study\nYear: 2014\n\nAbstract: The analysis of large-scale data for the purpose of extracting patterns is applicable to several research fields. In this paper, a dataset of climate related historical data from Jordan is collected. The main focus is to study evolution criteria of weather attributes in Jordan over the evaluated period of time and how it is connected to climate change. Results showed that humidity and dew point weather attributes are going to face a significant increase in the future. Such an increase is expected to have a direct, as well as an indirect impact on human life.","Title: Joint Routing, Channel Assignment, and Scheduling for Throughput Maximization in General Interference Models\nYear: 2010\n\nAbstract: Throughput optimization in wireless networks with multiple channels and multiple radio interfaces per node is a challenging problem. For general traffic models (given a set of source-destination pairs), optimization of throughput entails design of ??efficient?? routes between the given source-destination pairs, in conjunction with (1) assignment of channels to interfaces and communication links and (2) scheduling of noninterfering links for simultaneous transmission. Prior work has looked at restricted versions of the above problem. In this paper, we design approximation algorithms for the joint routing, channel assignment, and link scheduling problem in wireless networks with general interference models. The unique contributions of our work include addressing the above joint problem in the context of physical interference model and single-path routing (wherein, traffic between a source-destination pair is restricted to a single path). To the best of our knowledge, ours is the first work to address the throughput maximization problem in such general contexts. For each setting, we design approximation algorithms with provable performance guarantees. We demonstrate the effectiveness of our algorithms in general contexts through simulations.","Title: Truthful spectrum auctions with approximate revenue\nYear: 2011\n\nAbstract: In cellular networks, a recent trend is to make spectrum access dynamic in the spatial and temporal dimensions, for the sake of efficient utilization of spectrum. In such a model, the spectrum is divided into channels and periodically allocated to competing base stations using an auction-based market mechanism. An “efficient” auction mechanism is essential to the success of such a dynamic spectrum access model. Two of the key objectives in designing an auction mechanism are “truthfulness” and revenue maximization. In this article, we design a polynomial-time spectrum auction mechanism that is truthful and yields an allocation with O(1)-approximate expected revenue, in the Bayesian setting. Our mechanism generalizes to general interference models. To the best of our knowledge, ours is the first work to design a polynomial-time truthful spectrum auction mechanism with a performance guarantee on the expected revenue. We demonstrate the performance of our designed mechanism through simulations.","Title: An extensive study of the Bag-of-Words approach for gender identification of Arabic articles\nYear: 2014\n\nAbstract: The prevalent use of Online Social Networks (OSN) and the anonymity and lack of accountability they inherent from being online give rise to many problems related to finding the connection between the massive amount of text data on OSN and the people who actually wrote them. Analyzing text data for such purposes is called authorship analysis. This work is focused on one specific type of authorship analysis, which is identifying the author's gender. Gender identification has various applications from marketing to security. The focus of this work is on Arabic articles. The problem is basically a classification problem and the current approaches differ in the way they compute the features of each document. However, they all agree on following some “stylometric features” approach. Unlike these works, ours treat this problem as a variation of the Text Classification (TC) problem and follow the Bag-Of-Words (BOW) approach for feature selection. We perform an extensive set of experiments on the feature selection and classification phase and the results show that such an approach yield surprisingly high results.","Title: An Extensive Study Of Authorship Authentication Of Arabic Articles\nYear: 2017\n\nAbstract: Purpose - The authorship authentication (AA) problem is concerned with correctly attributing a text document to its corresponding author. Historically, this problem has been the focus of various studies focusing on the intuitive idea that each author has a unique style that can be captured using stylometric features (SF). Another approach to this problem, known as the bag-of-words (BOW) approach, uses keywords occurrences/frequencies in each document to identify its author. Unlike the first one, this approach is more language-independent. This paper aims to study and compare both approaches focusing on the Arabic language which is still largely understudied despite its importance.Design/methodology/approach - Being a supervised learning problem, the authors start by collecting a very large data set of Arabic documents to be used for training and testing purposes. For the SF approach, they compute hundreds of SF, whereas, for the BOW approach, the popular term frequency-inverse document frequency technique is used. Both approaches are compared under various settings.Findings - The results show that the SF approach, which is much cheaper to train, can generate more accurate results under most settings.Practical implications - Numerous advantages of efficiently solving the AA problem are obtained in different fields of academia as well as the industry including literature, security, forensics, electronic markets and trading, etc. Another practical implication of this work is the public release of its sources. Specifically, some of the SF can be very useful for other problems such as sentiment analysis.Originality/value - This is the first study of its kind to compare the SF and BOW approaches for authorship analysis of Arabic articles. Moreover, many of the computed SF are novel, while other features are inspired by the literature. As SF are language-dependent and most existing papers focus on English, extra effort must be invested to adapt such features to Arabic text.","Title: Multi-agent based dynamic resource provisioning and monitoring for cloud computing systems infrastructure\nYear: 2015\n\nAbstract: The cloud computing paradigm provides a shared pool of resources and services with different models delivered to the customers through the Internet via an on-demand dynamically-scalable form charged using a pay-per-use model. The main problem we tackle in this paper is to optimize the resource provisioning task by shortening the completion time for the customers' tasks while minimizing the associated cost. This study presents the dynamic resources provisioning and monitoring (DRPM) system, a multi-agent system to manage the cloud provider's resources while taking into account the customers' quality of service requirements as determined by the service-level agreement (SLA). Moreover, DRPM includes a new virtual machine selection algorithm called the host fault detection algorithm. The proposed DRPM system is evaluated using the CloudSim tool. The results show that using the DRPM system increases resource utilization and decreases power consumption while avoiding SLA violations.","Title: Software Defined Cognitive Radio Network Framework: Design and Evaluation\nYear: 2015\n\nAbstract: Software defined networking SDN provides a novel network resource management framework that overcomes several challenges related to network resources management. On the other hand, Cognitive Radio CR technology is a promising paradigm for addressing the spectrum scarcity problem through efficient dynamic spectrum access DSA. In this paper, the authors introduce a virtualization based SDN resource management framework for cognitive radio networks CRNs. The framework uses the concept of multilayer hypervisors for efficient resources allocation. It also introduces a semi-decentralized control scheme that allows the CRN Base Station BS to delegate some of the management responsibilities to the network users. The main objective of the proposed framework is to reduce the CR users' reliance on the CRN BS and physical network resources while improving the network performance by reducing the control overhead.","Title: Towards improving resource management in cloud systems using a multi-agent framework.\nYear: 2016\n\nAbstract: With the goal of efficient sharing of resources and services, the cloud computing paradigm has gained a lot of interest recently. This work focuses on improving the resource utilisation by optimising the resource provisioning through a multi-agent framework in which different agents are responsible for different tasks including the monitoring of customers (behaviour, resource usage patterns and QoS requirements as stated in the SLA) and available resources based on customeru0027s requests. Moreover, we introduce the concept of TaskFlow which allows a more elastic resources provisioning to match the customer real usage of the resources. The proposed system is implemented and tested on the CloudSim simulator and the results show it increases resource utilisation and decreases power consumption while avoiding SLA violations. The results also show that the introduction of the concept of TaskFlow into our proposed system leads to more resource saving but with a higher risk of SLA violations.","Title: Evaluating map reduce tasks scheduling algorithms over cloud computing infrastructure\nYear: 2015\n\nAbstract: Efficiently scheduling MapReduce tasks is considered as one of the major challenges that face MapReduce frameworks. Many algorithms were introduced to tackle this issue. Most of these algorithms are focusing on the data locality property for tasks scheduling. The data locality may cause less physical resources utilization in non-virtualized clusters and more power consumption. Virtualized clusters provide a viable solution to support both data locality and better cluster resources utilization. In this paper, we evaluate the major MapReduce scheduling algorithms such as FIFO, Matchmaking, Delay, and multithreading locality MTL on virtualized infrastructure. Two major factors are used to test the evaluated algorithms: the simulation time and the energy consumption. The evaluated schedulers are compared, and the results show the superiority and the preference of the MTL scheduler over the other existing schedulers. Also, we present a comparison study between virtualized and non-virtualized clusters for MapReduce tasks scheduling. Copyright © 2015 John Wiley & Sons, Ltd.","Title: SDIoT: a software defined based internet of things framework\nYear: 2015\n\nAbstract: The internet of things (IoT) represent the current and future state of the Internet. The large number of things (objects), which are connected to the Internet, produce a huge amount of data that needs a lot of effort and processing operations to transfer it to useful information. Moreover, the organization and control of this large volume of data requires novel ideas in the design and management of the IoT network to accelerate and enhance its performance. The software defined systems is a new paradigm that appeared recently to hide all complexity in traditional system architecture by abstracting all the controls and management operations from the underling devices (things in the IoT) and setting them inside a middleware layer, a software layer. In this work, a comprehensive software defined based framework model is proposed to simplify the IoT management process and provide a vital solution for the challenges in the traditional IoT architecture to forward, store, and secure the produced data from the IoT objects by integrating the software defined network, software defined storage, and software defined security into one software defined based control model.","Title: A Prototype For A Standard Arabic Sentiment Analysis Corpus\nYear: 2016\n\nAbstract: The researchers in the field of Arabic Sentiment Analysis (SA) need a relatively big standard corpus to conduct their studies. There are a number of existing datasets; however, they suffer from certain limitations such as the small number of reviews or topics they contain, the restriction to Modern Standard Arabic (MSA), etc., Moreover, most of them are in-house datasets that are not publicly available. Therefore, this study aims to establish a flexible and relatively big standard Arabic SA corpus that can he considered as a foundation to build larger Arabic corpora. In addition to MSA, this corpus contains reviews written in the five main Arabic dialects (Egyptian, Levantine, Arabian Peninsula, Mesopotamian, and Maghrebi group). Furthermore, this corpus has other five types of reviews (English, mixed MSA English, French, mixed MSA and Emoticons, and mixed Egyptian and Emoticons). This corpus is released for free to be used by researchers in this field, where it is characterized by its flexibility in allowing the users to add, remove, and revise its contents. The total number of topics and reviews of this initial version are 250 and 1,442, respectively. The collected topics are distributed equally among five domains (classes): Economy, Food-Life style, Religion, Sport, and Technology, where each domain has 50 topics. This corpus is built manually to ensure the highest quality to the researchers in this field.","Title: Towards Improving the Lexicon-Based Approach for Arabic Sentiment Analysis\nYear: 2014\n\nAbstract: The emergence of the Web 2.0 technology generated a massive amount of raw data by enabling Internet users to post their opinions on the web. Processing this raw data to extract useful information can be a very challenging task. An example of important information that can be automatically extracted from the users' posts is their opinions on different issues. This problem of Sentiment Analysis SA has been studied well on the English language and two main approaches have been devised: corpus-based and lexicon-based. This work focuses on the later approach due to its various challenges and high potential. The discussions in this paper take the reader through the detailed steps of building the main two components of the lexicon-based SA approach: the lexicon and the SA tool. The experiments show that significant efforts are still needed to reach a satisfactory level of accuracy for the lexicon-based Arabic SA. Nonetheless, they do provide an interesting guide for the researchers in their on-going efforts to improve lexicon-based SA.","Title: A GPU-based implementations of the fuzzy C-means algorithms for medical image segmentation\nYear: 2015\n\nAbstract: Fuzzy clustering is one of the most popular techniques in medical image segmentation. The fuzzy C-means (FCM) algorithm has been widely used as it provides better performance and more information than other algorithms. As the data set becomes large, the serial implementation of the FCM algorithm becomes too slow to accomplish the clustering task within acceptable time. Hence, a parallel implementation [for example, using today’s fast graphics processing unit (GPU)] is needed. In this paper, we implement brFCM algorithm, a faster variant of the FCM algorithm, on two different GPU cards, Tesla M2070 and Tesla K20m. We compare our brFCM GPU-based implementation with its CPU-based sequential implementation. Moreover, we compare brFCM with the traditional version of the FCM algorithm. The experiments used lung CT and knee MRI images for clustering. The results show that our implementation has a significant improvement over the traditional CPU sequential implementation. GPU parallel brFCM is 2.24 times faster than its CPU implementation, and 23.43 times faster than a GPU parallel implementation of the traditional FCM.","Title: Optimizing expansion strategies for ultrascale cloud computing data centers\nYear: 2015\n\nAbstract: With the increasing popularity gained by cloud computing systems over the past few years, cloud providers have built several ultrascale data centers at a variety of geographical locations, each including hundreds of thousands of computing servers. Since cloud providers are facing rapidly increasing traffic loads, they must have proper expansion strategies for their ultrascale data centers. The decision of expanding the capacities of existing data centers or building new ones over a certain period requires considering many factors, such as high power consumption, availability of resources, prices (of power, land, etc.), carbon tax, free cooling options, and availability of local renewable power generation. While a rich volume of recent research works focused on reducing the operational cost (OPEX) of the data centers, there exists no prior work, to the best of our knowledge, on investigating the trade-off between minimizing the OPEX of the data centers and maximizing their revenue from the services they offer while respecting the service level agreement (SLA) with their customers. In this study, we model this optimization problem using mixed integer-linear programming. Our proposed model is unique compared to the published works in many aspects such as its ability to handle realistic scenarios in which both data centers’ resources (servers) and user generated traffic are heterogeneous. To evaluate the proposed model and the impact of different parameters on it’s performance, several simulation experiments are conducted.","Title: Software defined cloud: Survey, system and evaluation.\nYear: 2016\n\nAbstract: Next generation cloud systems will require a paradigm shift in how they are constructed and managed. Conventional control and management platforms are facing considerable challenges regarding flexibility, dependability and security that next generation systems must handle. The cloud computing technology has already contributed in alleviating a number of the problems associated with resource allocation, utilization and management. However, many of the elements of a well-designed cloud environment remain “stiff” and hard to modify and adapt in an integrated fashion. This includes the underlying networking topologies, many aspects of the user control over IaaS, PaaS or SaaS layers, construction of XaaS services, provenance and meta-data collection, to mention but few. In many situations the problem may be due to inadequacy of service abstraction. Software Defined Systems (SDSys) is a concept that help abstract the actual hardware at different layers with software components; one classical example of this abstractions are hypervisors. Such abstraction provides an opportunity for system administrators to construct and manage their systems, more easily, through flexible software layers. SDSys is an umbrella for different software defined subsystems including Software Defined Networking (SDN), Software Defined Storage (SDStorage), Software Defined Servers (Virtualization), Software Defined Data Centers (SDDC), Software Defined Security (SDSec) etc. and ultimately Software Defined Clouds (SDCloud). Individual solutions and seamless integration of these different abstractions remains in many respects a challenge. In this paper, the authors introduce Software Defined Cloud (SDCloud), a novel software defined cloud management framework that integrates different software defined cloud components to handle complexities associated with cloud computing systems. The first part of paper presents, for the first time, an extensive state of the art critical review of different components of software defined systems, constructing the proposed SDCloud. The second part of the paper proposes the novel concept of SDCloud, which is implemented and evaluated for its feasibility, flexibility and potential superiority.","Title: A hierarchical optimization model for energy data flow in smart grid power systems\nYear: 2015\n\nAbstract: Environmental concerns and high prices of fossil fuels increase the feasibility of using renewable energy sources in smart grid. Smart grid technologies are currently being developed to provide efficient and clean power systems. Communication in smart grid allows different components to collaborate and exchange information. Traditionally, the utility company uses a central management unit to schedule energy generation, distribution, and consumption. Using centralized management in a very large scale smart grid forms a single point of failure and leads to serious scalability issues in terms of information delivery and processing. In this paper, a three-level hierarchical optimization approach is proposed to solve scalability, computational overhead, and minimize daily electricity cost through maximizing the used percentage of renewable energy. At level one, a single home or a group of homes are combined to form an optimized power entity (OPE) that satisfies its load demand from its own renewable energy sources (RESs). At level two, a group of OPEs satisfies energy requirements of all OPEs within the group. At level three, excess in renewable energy from different groups along with the energy from the grid is used to fulfill unsatisfied demands and the remaining energy are sent to storage devices. HighlightsWe examine the three scopes of smart grid systems.We write mathematical models to cover the different aspects in SG.We use linear optimization to solve the equations and yield optimal solutions.Hierarchical method achieves better distribution and cooperative scenario.","Title: Using Aspect-Based Sentiment Analysis to Evaluate Arabic News Affect on Readers\nYear: 2015\n\nAbstract: The rapid increase in digital information has raised great challenges especially when it comes to automated content analysis. The adoption of social media as a communication channel for political views demands automated methods for posts' tone analysis, sentiment analysis, and emotional affect. This paper proposes a novel approach of using aspect-based sentiment analysis in evaluating Arabic news posts affect on readers. The approach adopts several phases of text processing, features selection, and text classification. Two widely used classifiers, namely Conditional Random Fields (CRF) and J48, are tested. Experimentation results show that J48 outperforms CRF in aspect terms extraction whereas CRF is slightly better in aspect terms polarity identification.","Title: Using Big Data Analytics for Authorship Authentication of Arabic Tweets\nYear: 2015\n\nAbstract: Authorship authentication of a certain text is concerned with correctly attributing it to its author based on its contents. It is a very important problem with deep root in history as many classical texts have doubtful attributions. The information age and ubiquitous use of the Internet is further complicating this problem and adding more dimensions to it. We are interested in the modern version of this problem where the text whose authorship needs authentication is an online text found in online social networks. Specifically, we are interested in the authorship authentication of tweets. This is not the only challenging aspect we consider here. Another challenging aspect is the language of the tweets. Most current works and existing tools support English. We chose to focus on the very important, yet largely understudied, Arabic language. Finally, we add another challenging aspect to the problem at hand by addressing it at a very large scale. We present our effort to employ big data analytics to address the authorship authentication problem of Arabic tweets. We start by crawling a dataset of more than 53K tweets distributed across 20 authors. We then use preprocessing steps to clean the data and prepare it for analysis. The next step is to compute the feature vectors of each tweet. We use the Bag-Of-Words (BOW) approach and compute the weights using the Term Frequency-Inverse Document Frequency (TF-IDF). Then, we feed the dataset to a Naive Bayes classifier implemented on a parallel and distributed computing framework known as Hadoop. To the best of our knowledge, none of the previous works on authorship authentication of Arabic text addressed the unique challenges associated with (1) tweets and (2) large-scale datasets. This makes our work unique on many levels. The results show that the testing accuracy is not very high (61.6%), which is expected in the very challenging setting that we consider.","Title: SDDC: A Software Defined Datacenter Experimental Framework\nYear: 2015\n\nAbstract: The rapid growth and the distributed sites of the datacenters increase the complexity of control and management processes. A new paradigm which is called Software Defined Systems (SDSys) comes as a solution to reduce the overhead of datacenters management by abstracting all the control functionalities from the hardware devices and setting it inside a software layer. These functionalities are responsible for organizing and controlling the main blocks of the datacenter; network, storage, compute and security. The Software Defined Datacenter (SDD) integrates the software defined concepts into all of these main blocks. Transferring these concepts into workable systems requires checking several performance aspects and testing its correctness before building real systems. In this paper we introduce a novel experimental framework (SDDC) to provide a novel virtualized testbed environment for SDD systems. This work builds on the Mininet simulator, where its core components, the host, the switch and the controller, are customized to build the proposed experimental simulation framework for SDD. This simulator lets the users develop and test their own SDD solutions, and at the same time gives the researchers an experimentation tool for benchmarking purposes. The developed simulator could also be used as an educational tool to train students and novice researchers.","Title: Power Optimization of Large Scale Mobile Cloud Computing Systems\nYear: 2015\n\nAbstract: The main objective of deploying large scale Mobile Cloud Computing (MCC) systems using the cloudlet based infrastructure is to reduce the total power consumption and the network delay while satisfying the Service Level Agreement (SLA). However, this objective is challenging to achieve due to the limited capabilities of the cloudlet system which limit the number of available services on each cloudlet system. This will force different cloudlets to cooperate in order to meet the users demands for different types of services. Moreover, the cooperation will require routing of user requests from the local cloudlet to a remote cloudlet that provide the requested service using a backbone network. This cooperation will increase the total power consumption per request which will include both the user to the local cloudlet power consumption and the power consumption for routing the request to a remote cloudlet. In this paper, we address the problem of optimizing the power consumption for large scale cooperative cloudlets deployments. The evaluation of the proposed model under different realistic scenarios shows that it can be used to accurately optimize power consumption in large scale MCC systems.","Title: SDSecurity: A Software Defined Security experimental framework\nYear: 2015\n\nAbstract: The emerging Software Defined Systems (SDSys) is a recent paradigm, which has been introduced to reduce the overhead in the control and management operations of complex computing systems. The main concept behind this technology is around isolating the data plane from the control plane. Traditional security mechanisms are facing more challenges in providing sufficient levels of protection and efficiency. SDSys for security has been proposed to address these challenges. Software Defined Security (SDSec) provides a flexible and centralized security solution by abstracting the security mechanisms from the hardware layer to a software layer. In this paper we present a novel experimental framework to provide a novel virtualized testbed environment for SDSec systems. This work builds on the Mininet simulator, where its core components, the host, switch and the controller, are customized to build the proposed experimental simulation framework for SDSec. To the best of the authors' knowledge, this is the first experimental framework and simulator for SDSec solutions. The developed simulator, will not only support the development and testing of SDSecurity solutions, it will also serve as an experimentation tool for researchers and for benchmarking purposes. The developed simulator could also be used as an educational tool to train students and novice researchers.","Title: Virtualization-based Cognitive Radio Networks.\nYear: 2016\n\nAbstract: Integrate virtualization (multilayer hypervisors) into CRNs to improve performance.Allow BSs to delegate some of their management responsibilities to CR users.Reduces control overhead which leads to reduced delay and improved throughput.Show how to handle multi-cell CRNs (requires addressing self-coexistence problem).Significant improvements gained in terms of blocking probability and throughput. The emerging network virtualization technique is considered as a promising technology that enables the deployment of multiple virtual networks over a single physical network. These virtual networks are allowed to share the set of available resources in order to provide different services to their intended users. While several previous studies have focused on wired network virtualization, the field of wireless network virtualization is not well investigated. One of the promising wireless technologies is the Cognitive Radio (CR) technology that aims to handle the spectrum scarcity problem through efficient Dynamic Spectrum Access (DSA). In this paper, we propose to incorporate virtualization concepts into CR Networks (CRNs) to improve their performance. We start by explaining how the concept of multilayer hypervisors can be used within a CRN cell to manage its resources more efficiently by allowing the CR Base Station (BS) to delegate some of its management responsibilities to the CR users. By reducing the CRN users' reliance on the CRN BS, the amount of control messages can be decreased leading to reduced delay and improved throughput. Moreover, the proposed framework allows CRNs to better utilize its resources and support higher traffic loads which is in accordance with the recent technological advances that enable the Customer-Premises Equipments (CPEs) of potential CR users (such as smart phone users) to concurrently run multiple applications each generating its own traffic. We then show how our framework can be extended to handle multi-cell CRNs. Such an extension requires addressing the self-coexistence problem. To this end, we use a traffic load aware channel distribution algorithm. Through simulations, we show that our proposed framework can significantly enhance the CRN performance in terms of blocking probability and network throughput with different primary user level of activities.","Title: An Aspect-Based Sentiment Analysis Approach to Evaluating Arabic News Affect on Readers.\nYear: 2016\n\nAbstract: Great challenges arise due to the rapid growth of online data. The widespread use of online social networks (OSN) have enabled the generation of massive amounts of raw data where users post their own material. One interesting example of user generated data is their political views and opinions. The ability to crawl OSN and automatically analyze their political content is of undeniable importance. However, this requires automated methods for posts' tone analysis, sentiment analysis, and emotional affect. The purpose of this paper is to evaluate Arabic news posts affect on readers using a novel approach of aspect-based sentiment analysis (ABSA). There are many tasks typically associated with ABSA such as the extraction and polarity identification of aspect terms and categories. The focus of this work is on the tasks related to aspect terms. A typical approach to address these tasks goes through several stages of text pre-processing, features extraction and classification. This paper follows this approach and makes use of widely used features and classifiers. The features considered include Part of Speech (POS) tagging, Named Entity Recognition (NER), and N-Grams. As for the considered classifiers, they are: Conditional Random Fields (CRF), Decision Tree (J48), Naive Bayes and K-Nearest Neighbor (IBk). A set of experiments are conducted to compare the considered classifiers against each other and against a baseline classifier that is very common for ABSA. The results show that the extracted features allow all of the four considered classifiers to significantly outperform the baseline classifier. They also show that J48 performs the best for the task of aspect terms extraction whereas CRF and Naive Bayes are slightly better in aspect terms polarity identification.","Title: Sdmec: Software Defined System For Mobile Edge Computing\nYear: 2016\n\nAbstract: Mobile Edge Computing (MEC) promises a paradigm shift in enabling efficient Mobile Cloud Computing (MCC) services by providing storage and processing capacity within the access range of the mobile devices. In MEC, Mobile Edge (ME) servers are placed at the edge of the mobile networks eliminating the need to offload compute/storage-intensive tasks of the mobile devices to the core of the network (the centralized cloud data center). This reduces the network latency and enhances the quality of service provided for the mobile end users. Different applications can benefit from the large scale deployments of ME servers such as smart grid applications, content delivery networks, content sharing, traffic management, and E-health applications. This promising paradigm comes with its own downside related to the management complexity of large scale deployments that offers hundreds of applications to millions of users. In this paper, we introduce a software defined based framework to enable efficient MCC services through the integration of different software defined system components with the MEC system.","Title: Cloud Support Data Management Infrastructure For Upcoming Smart Cities\nYear: 2016\n\nAbstract: This paper presents a novel large scale data management model for future smart cities. The system is exploiting the emerging cloud computing services availability over the globe by processing the collected data at the edge. To handle the large scale requirements in smart city, we introduce a Mobile Edge Computing (MEC) framework in order to increase the reliability of the deployed applications. MEC is a promising framework to reduce the cloud core utilization as well as to provide applications that require low latency to mobile end users. As near as possible to the location of the end user, MEC relates to the mobile network applications and data stream acceleration through caching and/or compressing of relevant data at the edge of the mobile network. Although MEC is not yet deployed in real-life systems, recent studies discussed some technical details and related concepts. This paper studies the definition of MEC and similar concepts with respect to typical application scenarios, like in smart cities, and provides scope and limitations that may be encountered when implementing and deploying MEC. Then, a MEC framework is used to explore the data management infrastructure for upcoming smart cities. (C) 2016 The Authors. Published by Elsevier B.V.","Title: GA-Based Resource Augmentation Negotation for Energy-Optimised Mobile Ad-hoc Cloud\nYear: 2016\n\nAbstract: Mobile and cloud computing are two areas which are rapidly expanding in terms of use case and functionality. An ad-hoc mobile cloud can be constructed upon local devices, where devices work together and share resources, aspects such as energy and costs may be minimised. This paper builds on previous work which proposed an end-to-end system whereby user applications may be profiled for their resource consumption locally and then if augmentation is required, they may negotiate with a local mobile ad-hoc cloud for optimum energy and resource utilisation. The contribution of this paper focuses upon a novel and adapted Genetic Algorithm approach to resource negotiation in order to further enable energy-optimised resource augmentation for Mobile Ad-hoc clouds.","Title: Cloud support for large scale e-healthcare systems.\nYear: 2016\n\nAbstract: Rapid development of wearable devices and mobile cloud computing technologies has led to new opportunities for large scale e-healthcare systems. In these systems, individuals’ health information are remotely detected using wearable sensors and forwarded through wireless devices to a dedicated computing system for processing and evaluation where a set of specialists namely, hospitals, healthcare agencies and physicians will take care of such health information. Real-time or semi-real time health information are used for online monitoring of patients at home. This in fact enables the doctors and specialists to provide immediate medical treatments. Large scale e-healthcare systems aim at extending the monitoring coverage from individuals to include a crowd of people who live in communities, cities, or even up to a whole country. In this paper, we propose a large scale e-healthcare monitoring system that targets a crowd of individuals in a wide geographical area. The system is efficiently integrating many emerging technologies such as mobile computing, edge computing, wearable sensors, cloud computing, big data techniques, and decision support systems. It can offer remote monitoring of patients anytime and anywhere in a timely manner. The system also features some unique functions that are of great importance for patients’ health as well as for societies, cities, and countries. These unique features are characterized by taking long-term, proactive, and intelligent decisions for expected risks that might arise by detecting abnormal health patterns shown after analyzing huge amounts of patients’ data. Furthermore, it is using a set of supportive information to enhance the decision support system outcome. A rigorous set of evaluation experiments are conducted and presented to validate the efficiency of the proposed model. The obtained results show that the proposed model is scalable by handling a large number of monitored individuals with minimal overhead. Moreover, exploiting the cloud-based system reduces both the resources consumption and the delay overhead for each individual patient.","Title: Power Optimization of Large Scale Mobile Cloud System Using Cooperative Cloudlets\nYear: 2016\n\nAbstract: Reducing the total power consumption and network delay are among the most interesting issues facing large scale Mobile Cloud Computing (MCC) systems and their ability to satisfy the Service Level Agreement (SLA). Such systems utilize cloudlet based infrastructure to support off-loading some of user's computationally heavy tasks to the cloudlets. However, the limited capabilities of the cloudlet system (in terms of the ability of serve different request type and the ability to serve users in large geographical regions) represent serious challenges to achieve those objectives. To cover the users demand for different types of services and in wide geographical regions, cloudlets cooperate among each others by passing user requests from one cloudlets to another. By adapting this cooperation, the total power consumption per request will be increased so that it includes the power consumption between the user and the local cloudlet and the power consumption of passing the request to a remote cloudlet. In this paper, we consider two types of cloudlets: local cloudlets and global cloudlets. The global cloudlets are a special kind of local cloudlets but with higher capabilities. The user can connect only to the local cloudlet and sends all its traffics to it. If the local cloudlet cannot serve the desired request, then the request is moved to other local cloudlet. If no local cloudlet can serve the request, then it is moved to a global cloudlet in which it can serve all service types. We optimize the power consumption for large scale cooperative cloudlets and evaluate the proposed model under two realistic scenarios. The result prove that the proposed model can be used to optimize power consumption in large scale MCC systems.","Title: Using Dynamic Parallelism to Speed Up Clustering-Based Community Detection in Social Networks\nYear: 2016\n\nAbstract: Social Network Analysis (SNA) has been gaining a lot of attention lately. One of the common steps in SNA is community detection. SNA literature has many interesting algorithms for community detection. One of the popular ones was proposed by Newman and it is mainly revolved around using a clustering algorithm. Three phases are iteratively applied in this algorithm in order to find the \"best\" community structure. These phases are: spectral mapping, clustering and modularity computation. Despite its effectiveness, this method suffers greatly in terms of running time when dealing with largescale networks. A parallel implementation using GPUs is one of the feasible solutions to address this problem. Moreover, due to the iterative nature of this algorithm, dynamic parallelism lends itself as a very appealing solution. Dynamic parallelism is a novel parallel programming technique that refers to the ability to launch new grids from the GPU. In this work, we present three implementation of the clustering-based community detection algorithm. In addition to the sequential implementation, we present two implementations: a Hybrid CPU-GPU (HCG) one and a Dynamic Parallel (DP) one. We test our parallel implementations on benchmark datasets to show the speed-up of each parallel implementation compared with the sequential one. The results show that the DP implementation achieves good speed-ups reaching up to 4.45X, however, the speed-ups achieved by HCG are almost twice as much.","Title: SDCache: Software Defined Data Caching Control for Cloud Services\nYear: 2016\n\nAbstract: The complexity of traditional control approaches typically features slow response times to system and component changes and runs the risk of accessing outdated information. Dynamic environments require fast control approaches to deal with variability in a reliable manner. We adopt the Software Defined System (SDSys) paradigm to handle this issue. The concept behind this fast emerging theme revolves around abstracting all controls as modules within a software control layer. In this paper, we exploit this concept to propose a new system architecture for data caching. We further develop and emulate a testbed environment to evaluate the proposed models by extending the elements of Mininet. Several metrics and multiple scenarios have been considered to capture processing costs and network congestion so as to reflect the merits of this approach. Our experiments validate that the proposed solution based on SDSys renders possible a simple programmable data caching control mechanism.","Title: SemEval-2016 Task 5: Aspect Based Sentiment Analysis.\nYear: 2016\n\nAbstract: This paper describes the SemEval 2016 shared task on Aspect Based Sentiment Analysis (ABSA), a continuation of the respective tasks of 2014 and 2015. In its third year, the task provided 19 training and 20 testing datasets for 8 languages and 7 domains, as well as a common evaluation procedure. From these datasets, 25 were for sentence-level and 14 for text-level ABSA; the latter was introduced for the first time as a subtask in SemEval. The task attracted 245 submissions from 29 teams.","Title: A Lexicon Based Approach For Classifying Arabic Multi-Labeled Text\nYear: 2016\n\nAbstract: Purpose - Multi-label Text Classification (MTC) is one of the most recent research trends in data mining and information retrieval domains because of many reasons such as the rapid growth of online data and the increasing tendency of internet users to be more comfortable with assigning multiple labels/tags to describe documents, emails, posts, etc. The dimensionality of labels makes MTC more difficult and challenging compared with traditional single-labeled text classification (TC). Because it is a natural extension of TC, several ways are proposed to benefit from the rich literature of TC through what is called problem transformation (PT) methods. Basically, PT methods transform the multi-label data into a single-label one that is suitable for traditional single-label classification algorithms. Another approach is to design novel classification algorithms customized for MTC. Over the past decade, several works have appeared on both approaches focusing mainly on the English language. This work aims to present an elaborate study of MTC of Arabic articles.Design/methodology/approach - This paper presents a novel lexicon-based method for MTC, where the keywords that are most associated with each label are extracted from the training data along with a threshold that can later be used to determine whether each test document belongs to a certain label.Findings - The experiments show that the presented approach outperforms the currently available approaches. Specifically, the results of our experiments show that the best accuracy obtained from existing approaches is only 18 per cent, whereas the accuracy of the presented lexicon-based approach can reach an accuracy level of 31 per cent.Originality/value - Although there exist some tools that can be customized to address the MTC problem for Arabic text, their accuracies are very low when applied to Arabic articles. This paper presents a novel method for MTC. The experiments show that the presented approach outperforms the currently available approaches.","Title: Accelerating compute intensive medical imaging segmentation algorithms using hybrid CPU-GPU implementations.\nYear: 2017\n\nAbstract: Medical image processing is one of the most famous image processing fields in this era. This fame comes because of the big revolution in information technology that is used to diagnose many illnesses and saves patients lives. There are many image processing techniques used in this field, such as image reconstructing, image segmentation and many more. Image segmentation is a mandatory step in many image processing based diagnosis procedures. Many segmentation algorithms use clustering approach. In this paper, we focus on Fuzzy C-Means based segmentation algorithms because of the segmentation accuracy they provide. In many cases, these algorithms need long execution times. In this paper, we accelerate the execution time of these algorithms using Graphics Process Unit (GPU) capabilities. We achieve performance enhancement by up to 8.9x without compromising the segmentation accuracy.","Title: MedGraph: a graph-based representation and computation to handle large sets of images.\nYear: 2017\n\nAbstract: In order to process and analyze very large volumes of images, efficient representation and structuring techniques are required. Since, current computing machines can provide large memory size, trading off reasonable amount of memory in order to achieve efficient and parallelizable representation of images is preferable. In this paper, we propose a new structure to represent and store images based on in-memory graph concept. Our method of computation provides a faster execution time than the traditional array-based representation. Each pixel of an image is represented as a one node in the graph. In addition, nodes have pointers to other neighboring nodes (pixels). The structure represents an image as a grid of connected linked lists and each grid is connected to other grids. Using our method, an image can be represented in one of three different representations which are: octal linked list, quadratic linked list, and dual linked list representations. We provide experiments and evaluations using the dual linked list representation as it requires less memory space. We apply our methodology for medical images as a proof of the concept to find a region of interest in an image. We have collected and used real medical images to build and process the graph which we call MedGraph. Our experimental results show that the proposed MedGraph technique improves the searching time for finding a region of interest when compared to the traditional representation. It is worth mentioning here that MedGraph is a generic representation strategy that can be applied to any type of images, although this paper uses medical images as a proof of the concept.","Title: The Internet of Energy: Smart Sensor Networks and Big Data Management for Smart Grid.\nYear: 2015\n\nAbstract: Smart sensor networks provide numerous opportunities for smart grid applications including power monitoring, demand-side energy management, coordination of distributed storage, and integration of renewable energy generators. Because of their low cost and ease-of-deployment, smart sensor networks are likely to be used on a large scale in future of smart power grids. The result is a huge volume of different variety of data sets. Processing and analyzing these data reveals deeper insights that can help expert to improve the operation of power grid to achieve better performance. The technology to collect massive amounts of data is available today, but managing the data efficiently and extracting the most useful information out of it remains a challenge. This paper discusses and provides recommendations and practices to be used in the future of smart grid and Internet of things. We explore the different applications of smart sensor networks in the domain of smart power grid. Also we discuss the techniques used to manage big data generated by sensors and meters for application processing.","Title: Leveraging Software-Defined-Networking for Energy Optimisation in Mobile-Cloud-Computing.\nYear: 2016\n\nAbstract: Both mobile and cloud computing are two areas which are rapidly expanding in terms of use case and functionality. Whilst mobile computing enables a variety of feature rich functionality for users in a non-fixed location, cloud computing is revolu-tionising the way in which computing resources are being provisioned, used and optimised for both service providers and end users. These two fields are being combined in order to provide greater functionality for mobile devices by off loading resources to the cloud. However, the advantages of this only hold true if the device resources are truly optimised. This paper examines the effect of network simulation on understanding methods for optimising device life and suggests a number of ways in which Software-Defined-Networking (SDN) may be leveraged in order to determine the exact nature of optimisations available through this combination.","Title: Using the Panama Papers to explore the financial networks of the Middle East\nYear: 2016\n\nAbstract: In what has been described as the WikiLeaks of the financial world, the release of millions of documents (known as the “Panama Papers”) have placed at the center of global media attention the elaborate ways used by some of the elite to hide their financial assets leading to serious allegation of financial corruption. In this work, we explore the information contained in these documents using social network analytics. Due to the large size of the network constructed from the Panama Papers, we limit our attention to a specific region, which is the Middle East. The analysis reveals that while the constructed network enjoys some typical characteristics, there are many interesting observations and properties worth discussing. Specifically, using the extracted network consisting of 37,442 nodes and 79,544 edges, our social network analysis finding show that, perhaps surprisingly, the nodes or the social network are not necessarily directly correlation with perceived financial influence.","Title: Software-defined systems support for secure cloud computing based on data classification.\nYear: 2017\n\nAbstract: The newly emerged Software-Defined Systems (SDSs) promised to reduce computing systems management complexity. This can be achieved by separating the control plane from the data plane. On the other hand, cloud computing usage proliferation creates new challenges for managing user data efficiently. Such challenges include data ownership, data access policies, data privacy and integrity, and the availability of storage space. Using encryption-based solution is effective but very costly. Other solutions that used data classification based on the data priority can mitigate the high cost problem for data encryption. However, managing such data classification systems is very complex. In this paper, we are introducing a Software-Defined System-based solution for deploying efficient data classification framework. Our results show the superiority of the proposed model.","Title: Paraphrase identification and semantic text similarity analysis in Arabic news tweets using lexical, syntactic, and semantic features.\nYear: 2017\n\nAbstract: The rapid growth in digital information has raised considerable challenges in particular when it comes to automated content analysis. Social media such as twitter share a lot of its users information about their events, opinions, personalities, etc. Paraphrase Identification (PI) is concerned with recognizing whether two texts have the same/similar meaning, whereas the Semantic Text Similarity (STS) is concerned with the degree of that similarity. This research proposes a state-of-the-art approach for paraphrase identification and semantic text similarity analysis in Arabic news tweets. The approach adopts several phases of text processing, features extraction and text classification. Lexical, syntactic, and semantic features are extracted to overcome the weakness and limitations of the current technologies in solving these tasks for the Arabic language. Maximum Entropy (MaxEnt) and Support Vector Regression (SVR) classifiers are trained using these features and are evaluated using a dataset prepared for this research. The experimentation results show that the approach achieves good results in comparison to the baseline results.","Title: Feature extraction and selection for Arabic tweets authorship authentication.\nYear: 2017\n\nAbstract: In tweet authentication, we are concerned with correctly attributing a tweet to its true author based on its textual content. The more general problem of authenticating long documents has been studied before and the most common approach relies on the intuitive idea that each author has a unique style that can be captured using stylometric features (SF). Inspired by the success of modern automatic document classification problem, some researchers followed the Bag-Of-Words (BOW) approach for authenticating long documents. In this work, we consider both approaches and their application on authenticating tweets, which represent additional challenges due to the limitation in their sizes. We focus on the Arabic language due to its importance and the scarcity of works related on it. We create different sets of features from both approaches and compare the performance of different classifiers using them. We experiment with various feature selection techniques in order to extract the most discriminating features. To the best of our knowledge, this is the first study of its kind to combine these different sets of features for authorship analysis of Arabic tweets. The results show that combining all the feature sets we compute yields the best results.","Title: Framework for Affective News Analysis of Arabic News: 2014 Gaza Attacks Case Study.\nYear: 2017\n\nAbstract: This paper aims at fostering the domain of Arabic affective news analysis through providing: (a) a benchmark annotated Arabic dataset of news for affective news analysis, (b) an aspect-based sentiment analysis (ABSA) approach for evaluating the sentimental affect of Arabic news posts on the reader, and (c) a baseline approach with a common evaluation framework to compare future research results with the baseline ones.","Title: Accelerating compute-intensive image segmentation algorithms using GPUs.\nYear: 2017\n\nAbstract: Image segmentation is an important process that facilitates image analysis such as in object detection. Because of its importance, many different algorithms were proposed in the last decade to enhance image segmentation techniques. Clustering algorithms are among the most popular in image segmentation. The proposed algorithms differ in their accuracy and computational efficiency. This paper studies the most famous and new clustering algorithms and provides an analysis on their feasibility for parallel implementation. We have studied four algorithms which are: fuzzy C-mean, type-2 fuzzy C-mean, interval type-2 fuzzy C-mean, and modified interval type-2 fuzzy C-mean. We have implemented them in a sequential (CPU only) and a parallel hybrid CPU---GPU version. Speedup gains of 6$$\\\\times $$× to 20$$\\\\times $$× were achieved in the parallel implementation over the sequential implementation. We detail in this paper our discoveries on the portions of the algorithms that are highly parallel so as to help the image processing community, especially if these algorithms are to be used in real-time processing where efficient computation is critical.","Title: Cloudlet solution for digital forensic investigation of multiple cases of multiple devices\nYear: 2017\n\nAbstract: Multiple device ownership exponentially increases the volume and variety of data, with detrimental implications to digital forensic investigations. Several authors have proposed data reduction approaches in attempts to enhance the data acquisition and processing phases of the investigation process. Other works have aimed to take advantage of cloud computing's seemingly unlimited resources to leverage investigations. However, such approaches inadvertently affect the credibility of forensic evidence and its admissibility in a court of law, and degrade the efficiency of forensic processes. In this paper, we propose a novel approach which leverages current processes by focusing on augmenting computational and latency capabilities. To achieve this, we motivate a cloudlet-based digital forensic (DF) approach to complement existing cloud computing systems. Based on their proximity to end-devices and remote DF investigation teams, our proposed solution effectively tackles low latency challenges present with the cloud alternative. In addition, configuring the cloudlet solution as the sole custodian of data counters ensures that investigators remain in control of their data, and hence can maintain a comprehensive evidence trail. Finally, have also proposed a cloudlet-based DF resource optimization approach to facilitate upward and downward scaling of resources to cope with a variety of data sizes, multiple devices, and concurrent multiple cases.","Title: Collision-free anycast transmission scheduling in UWSNs\nYear: 2017\n\nAbstract: Underwater Wireless Sensor Network (UWSNs) have been proposed to monitor underwater regions such as seas and oceans. A typical example of an UWSN consists of a set of underwater sensor nodes and a set of sink nodes that are deployed at the sea surface. Using acoustic transmission, the sensors send their collected data to at least one of the sinks in what is known as the anycast transmission problem. One of the major problems with underwater nodes is their limited power, which means that transmission collisions are very costly. In this paper, we propose a collision-free anycast transmission scheduling algorithm for UWSNs. The proposed algorithm is a location-based routing algorithm that schedules transmissions between sensor nodes and sink nodes in order to minimize the energy consumption and possible collisions. We perform a set of experiments using the NS2 Aqua-Sim simulator to show that the proposed algorithm can provide up to 26% in energy savings compared with a simple greedy algorithm.","Title: A Fog Computing Based System for Selective Forwarding Detection in Mobile Wireless Sensor Networks\nYear: 2016\n\nAbstract: Intrusions detection is one of the major issues that worry organizations in wireless sensor networks (WSNs). Many researchers have dealt with this problem and have proposed many methods for detecting different kinds of intrusions such as selective forwarding, which is a serious attack that may obstruct communications in WSNs. However, as the applications of mobile computing, vehicular networks, and internet of things (IoT) are spreading immensely, selective forwarding detection in Mobile Wireless Sensor Networks (MWSNs) has become a key demand. This paper introduces the problem of selective forwarding in MWSNs, and discusses how available techniques for mitigation this problem in WSNs are not applicable in handling the problem in MWSNs due to sensors mobility. Therefore, the paper proposes a model that provides a global monitoring capability for tracing moving sensors and detecting malicious ones. The model leverages the infrastructure of Fog Computing to achieve this purpose. Furthermore, the paper provides a complete algorithm, a comprehensive discussion and experiments that show the correctness and importance of the proposed approach.","Title: On The Road To The Internet Of Biometric Things: A Survey Of Fingerprint Acquisition Technologies And Fingerprint Databases\nYear: 2016\n\nAbstract: Since security vulnerabilities represent one of the grand challenges of the Internet of Things (IoT), researchers have proposed what is known as the Internet of Biometric Things (IoBT), which mixes traditional biometric technologies with context-aware authentication techniques. One of the most famous biometric technologies is electronic fingerprint recognition, which acquire fingerprints using different technologies, some of which are more suitable for IoBT than others. Moreover, different fingerprint databases have been built to study the impact of different factors on the accuracy of different fingerprint recognition algorithms and systems. In this paper, we survey the available fingerprint acquisition technologies and the available fingerprint databases. We also identify the advantages and disadvantages of each technology and database.","Title: A Compression-Based Technique To Classify Metamorphic Malware\nYear: 2016\n\nAbstract: Metamorphic malware are able to change their appearance to evade detection by traditional anti-malware software. One of the ways to help mitigate the threat of new metamorphic malware is to determine their origins, i.e., the families to which they belong. This type of metamorphic malware analysis is not typically handled by commercial software. Moreover, existing works rely on analyzing the op-code sequences extracted from the Assembly files of the malware. Very few papers have tried to perform analysis on the binary files of the malware. However, they focused on the simple binary problem of differentiating between a certain malware family and benign files. In this work, we address the more difficult problem of determining the origin of a new metamorphic malware by measuring its similarity to hundreds of variants taken from 13 families of real malware. To address this problem, we use a compression-based classification approach. We experiment with two such approaches: AMDL and BCN. The results showed that AMDL performed no better than a random guess (11% accuracy for AMDL and 18% for the random baseline). On the other hand, BCN performed really well with 67% accuracy.","Title: Parallel Implementation Of Fcm-Based Volume Segmentation Of 3d Images\nYear: 2016\n\nAbstract: Parallel programming has many benefits that can help developers and researchers to improve the performance of some algorithms to become more efficient in real life. This is especially true for systems involving medical images. Image segmentation for volume extraction is a famous segmentation process that takes long time to finish execution. In this paper, we consider a new version of the Fuzzy C-Means (FCM) segmentation algorithm (known as IT2FPCM) and provide a parallel implementation of it that is 12X time faster than the sequential implementation. The considered algorithm is based on Interval Type-2 FCM and combines fuzzy and possibilistic ideas in order to obtain higher accuracy. We conduct our experiments using two different machines and the results show that the improvement gains for both machines 11X and12X, respectively.","Title: Building A Standard Dataset For Arabic Sentiment Analysis: Identifying Potential Annotation Pitfalls\nYear: 2016\n\nAbstract: Sentiment Analysis (SA) is one of the hottest research fields nowadays. It is concerned with identifying the sentiment conveyed in a piece of text. The current efforts in SA require the existence of standard datasets for training/testing purposes. Such datasets already exist for some languages such as English. Unfortunately, the same cannot be said about other languages such as Arabic. Currently existing Arabic SA datasets are restricted (in their domain, size, dialects covered, etc.) and/or have limited availability. Moreover, the annotation process did not receive the proper attention it deserves. Some of the existing datasets relied on the author's point of view for annotation, while others employed annotators, but did not take into account the personal variations between the annotators and how would that affect their agreement. This study presents our efforts to build a standard Arabic dataset with the above concerns in mind. The constructed dataset is intended for generic use as it contains reviews from different domains written in Modern Standard Arabic (MSA) as well as several dialects. As for the annotation process, it is given high attention by studying the inter-annotator agreements and investigating the potential factors affecting them.","Title: Exploiting Gpus To Accelerate Clustering Algorithms\nYear: 2016\n\nAbstract: Big data is a main problem for data mining methods. Fortunately, the rapid advances in affordable high performance computing platforms such as the Graphics Processing Unit (GPU) have helped researchers in reducing the execution time of many algorithms including data mining algorithms. This paper discusses the utilization of the parallelism capabilities of the GPU to improve the the performance of two common clustering algorithms, which are K-Means (KM) and Fuzzy C-Means (FCM) algorithms. Two main parallelism approaches are presented: pure and hybrid. These different versions are tested under different settings including two different GPU-equipped machines (a laptop and a server). The results show excellent improvement gains of the hybrid implementations compared with the pure parallel and sequential ones. On the laptop, the best gains of the hybrid implementations compared with the sequential ones are 11.3X for KM and 10.9X for FCM. As for the server, the best gains are 13.5X for KM and 16.3X for FCM. Moreover, the paper explores the usage of a recent memory management technique for GPU called Unified Memory (UM). The results show a decrease in the performance gain of the hybrid implementations that is equal to 44% for hybrid version of KM and 61% for FCM. On the other hand, the use of UM does introduce a small advantage for the pure parallel implementation.","Title: Authorship Attribution Of Arabic Tweets\nYear: 2016\n\nAbstract: In tweet authentication, we are concerned with correctly attributing a tweet to its true author based on its textual content. The more general problem of authenticating long documents has been studied before and the most common approach relies on the intuitive idea that each author has a unique style that can be captured using stylometric features (SF). Inspired by the success of modern automatic document classification problem, some researchers followed the Bag-Of-Words (BOW) approach for authenticating long documents. In this work, we consider both approaches and their application on authenticating tweets, which represent additional challenges due to the limitation in their sizes. We focus on the Arabic language due to its importance and the scarcity of works related on it. We create different sets of features from both approaches and compare the performance of different classifiers using them. To the best of our knowledge, this is the first study of its kind to combine these different sets of features for authorship analysis of Arabic tweets. The results show that combining all the feature sets we compute yields the best results.","Title: Improved Hierarchical Classifiers For Multi-Way Sentiment Analysis\nYear: 2017\n\nAbstract: Sentiment Analysis (SA) is field in computational linguistics concerned with determining the sentiment conveyed in a piece of text towards certain entities (such as people, organizations, products, services, events, etc.) using NLP tools. The considered sentiments can be as simple as positive vs. negative. A more fine-grained approach known as Multi-Way Sentiment Analysis (MWSA) is based on ranking systems, such as the 5-star ranking system. In such systems, rankings close to each other can be confusing; thus, some researchers have suggested that using Hierarchical Classifiers (HCs) can yield better results compared with traditional Flat Classifier (FCs). Unlike FCs, which try to address the entire classification problem at once, HCs employ some kind of tree structures where the nodes are simple \"core\" classifiers customized to address a subset of the classification problem. This study aims to explore extensively the use of HCs to address MWSA by studying six different hierarchies. We compare these hierarchies using four well-known core classifiers (SVM, Decision Tree, Naive Bayes, and KNN) using many measures such as Precision, Recall, F1, Accuracy and Mean Square Error (MSE). The experiments are conducted on the Large Arabic Book Reviews (LABR) dataset, which consists of 63K book reviews in Arabic. The results show that using some of the proposed HCs yield significant improvements in accuracy. Specifically, while the best Accuracy and MSE for FC are 45.77% and 1.61, respective-ly, the best accuracy and MSE for an HC are 72.64% and 0.53, respectively. Also, the results show that, in general, KNN(k-nearest neighbors) benefitted the most from using hierarchical classification.","Title: A novel framework for software defined based secure storage systems.\nYear: 2017\n\nAbstract: The Software Defined Systems (SDSys) paradigm has been introduced recently as a solution to reduce the overhead in the control and management operations of complex computing systems and to maintain a high level of security and protection. The main concept behind this technology is around isolating the data plane from the control plane. Building a Software Defined System in a real life environment is considered an expensive solution and may have a lot of risks. Thus, there is a need to simulate such systems before the real-life implementation and deployment. In this paper we present a novel experimental framework as a virtualized testbed environment for software defined based secure storage systems. Its also covers some related issues for large scale data storage and sharing such as deduplication. This work builds on the Mininet simulator, where its core components, the host, switch and the controller, are customized to build the proposed experimental simulation framework. The developed emulator, will not only support the development and testing of SD-based secure storage solutions, it will also serve as an experimentation tool for researchers and for benchmarking purposes. The developed simulator/emulator could also be used as an educational tool to train students and novice researchers.","Title: Using Logistic Regression To Improve Virtual Machines Management In Cloud Computing Systems\nYear: 2017\n\nAbstract: Cloud computing (CC) is a computing model that enables its customers to access a shared pool of resources (e.g., storage, network, servers, etc.) through the Internet with a pay per-use pricing model. Different service models are employed in CC including the Platform-as-a-Service (PaaS) model, in which the costumers request a certain set of resources and the cloud service providers provide these resources in the form of a virtual machine (VM) running on one of the thousands of hosting servers or physical machines (PMs) of a data center. Where to \"place\" VMs, how to \"execute\" them and whether there is a need to \"move/migrate\" them are important decisions that affect the overall resource utilization and power consumption in the hosting data center. VM consolidation is a technique of migrating or consolidating VMs to PMs in order to prevent the PMs from being overloaded or reduce the number of active PMs and increase their utilization. Consolidation techniques measure PM utilization to decide whether to consolidate the VMs running on it or migrate some of them to another PM. This study aims to optimize resource utilization and energy efficiency in cloud data centers by proposing a new Logistic Regression based host overloading prediction technique that can be used by any VM consolidation technique. The new algorithm have been evaluated using a dynamic workload using the CloudSim simulator. The simulation results show that the proposed algorithm outperforms all other known host status prediction techniques.","Title: Enhanced 3D segmentation techniques for reconstructed 3D medical volumes: Robust and Accurate Intelligent System.\nYear: 2017\n\nAbstract: Medical images play an important role in treating a large number of ailments as they are integral and even indispensable to the diagnosis process of such ailments. Medical images come from different acquisition systems (such as PET, CT, MRI) and, in many situations, automated processing of these images can greatly aid physicians and make their jobs easier. In medical imaging and its applications, 2D segmentation (with its different approaches such as FCM, k-means, MRFM and NN) is the first step which is used to extract ROI. This helps in extracting ROI in each slice (2D medical image) separately regardless of its relation to the next and the previous slices. In this paper, a 3D model of FCM segmentation techniques is proposed to enhance the segmentation process and take in mind the overall 3D-Volume as one testing data.","Title: Software-Defined System Support for Enabling Ubiquitous Mobile Edge Computing.\nYear: 2017\n\nAbstract: Mobile Edge Computing (MEC) enables ubiquitous and efficient cloud services to mobile users which facilitates mobile cloud computing (MCC) more easily by providing storage and processing capacity within the access range of the mobile devices. To achieve the goals of MEC, Mobile Edge (ME) servers are co-placed with the mobile network base station (at the edge of the mobile network). This eliminates the need to move computation and storage intensive tasks from a mobile device to a centralized cloud server. This in turn reduces the network communication load and delay and it also enhances the quality of service provided for the mobile end users. Applications, such as smart grid applications, content delivery networks, crowed sourcing, traffic management and E-health, will greatly benefit from such deployment. Unfortunately, a large-scale deployment of ME servers that is needed to make hundreds of applications available to millions of users, comes with great management complexity. The emerging technique of Software-Defined Systems (SDSys) abstracts the management complexities of many systems at different layers by utilizing software components. In this paper, we build a software-defined based framework that enable efficient and ubiquitous MCC services by integrating different SDSys components with the MEC system. The integrated framework is implemented and it is evaluated for its feasibility, flexibility and potential superiority.","Title: Delay-aware power optimization model for mobile edge computing systems.\nYear: 2017\n\nAbstract: Reducing the total power consumption and network delay are among the most interesting issues facing large-scale Mobile Cloud Computing (MCC) systems and their ability to satisfy the Service Level Agreement (SLA). Such systems utilize cloud computing infrastructure to support offloading some of user’s computationally heavy tasks to the cloud’s datacenters. However, the delay incurred by such offloading process lead the use of servers (called cloudlets) placed in the physical proximity of the users, creating what is known as Mobile Edge Computing (MEC). The cloudlet-based infrastructure has its challenges such as the limited capabilities of the cloudlet system (in terms of the ability to serve different request types from users in vast geographical regions). To cover the users demand for different types of services and in vast geographical regions, cloudlets cooperate among each other by passing user requests from one cloudlet to another. This cooperation affects both power consumption and delay. In this work, we present a mixed integer linear programming (MILP) optimization model for MEC systems with these two issues in mind. Specifically, we consider two types of cloudlets: local cloudlets and global cloudlets, which have higher capabilities. A user connects to a local cloudlet and sends all of its traffics to it. If the local cloudlet cannot serve the desired request, then the request is moved to another local cloudlet. If no local cloudlet can serve the request, then it is moved to a global cloudlet which can serve all service types. The process of routing requests through the hierarchical network of cloudlets increases power consumption and delay. Our model minimizes power consumption while incurring an acceptable amount of delay. We evaluate it under several realistic scenarios to show that it can indeed be used for power optimization of large-scale MEC systems without violating delay constraints.","Title: An extended study of collaboration networks of Levantine biomedical researchers.\nYear: 2018\n\nAbstract: A collaboration (or a co-authorship) network (ColNet) consists of researchers as nodes with the edges connecting two researchers if they have coauthored at least one paper together. In this work, we study the ColNet of biomedical researchers in the Levant region. In addition to performing the standard measures of social network analysis on the constructed network, we propose new metrics (for the stability of research groups) to help us explore the patterns of highly productive researchers. Analyzing the constructed network reveals that local researchers are shifting from working individually or with non-local collaborators to working within relatively large groups of local collaborators. Moreover, we find that highly productive researchers enjoy a certain balance between maintaining stable relationship with their previous collaborators and engaging in new projects with new collaborators.","Title: Accelerating 3D medical volume segmentation using GPUs.\nYear: 2018\n\nAbstract: Medical images have an undeniably integral role in the process of diagnosing and treating of a very large number of ailments. Processing such images (for different purposes) can significantly improve the efficiency and effectiveness of this process. The first step in many medical image processing applications is segmentation, which is used to extract the Region of Interest (ROI) from a given image. Due to its effectiveness, a very popular segmentation algorithm is the Fuzzy C-Means (FCM) algorithm. However, FCM takes a long processing time especially for 3D model. This problem can be solved by utilizing parallel programming using Graphics Processing Unit (GPU). In this paper, a hybrid parallel implementation of FCM for extracting volume object from medical DICOM files has been proposed. The proposed algorithm improves the performance 5× compared with the sequential version.","Title: Building an Image Database for Studying Image Retargeting\nYear: 2017\n\nAbstract: Modern electronic devices(such as TVs, laptops, and mobile devices) come with a huge variety in screen sizes, resolutions, and aspect ratios. Image retargeting is a technique to retarget or (resize) an image to better utilize the viewing device screen and to protect the main content of the image. Different retargeting techniques have been proposed in the literature that mainly utilizes one of the following main techniques: cropping, seam carving, and scale and stretch. The current problem of image retargeting is that it is very hard to determine the best technique to use on an image to get a target dimension. To apply techniques such as machine learning to determine the best technique to perform image retargeting, an annotated image set is needed to perform the training step. In this work, we build and annotate an image set that is suitable to develop such advance retargeting techniques. We build a dataset that include 500 original images. We apply 4 different retargeting techniques to get two different sizes. The resulting image set contains 4000 images annotated by three people. We also analyze the annotation results to get useful remarks from the annotators perceptual point of view.","Title: Fine-Grained Emotion Analysis of Arabic Tweets: A Multi-target Multi-label Approach\nYear: 2018\n\nAbstract: Emotion Analysis (EA) is the task of determining the emotion of a given piece of text. This is an important task with many applications especially when applied to tweets. However, existing work take a rather coarse-grained approach by assuming that each tweet has a single emotion and that this emotion has no intensity. In this work, we take a fine-grained approach by considering cases where a single tweet may have several emotions (multi-label) each with possibly different intensity (multi-target). Moreover, unlike existing work, which consider languages such as English and Chinese, we focus on the Arabic language, a severely under-studied language despite its importance. We build the first dataset (to the best of our knowledge) of Arabic tweets annotated for emotion analysis as a multi-label multi-target problem. Two human experts participated in the annotation process and Cohen's Kappa measure was used to determine their concordance.","Title: Studying the controversy in online crowds' interactions.\nYear: 2018\n\nAbstract: Detecting controversy in OSN can be very beneficial in understanding the behavior of the online crowds and the dynamics of their interactions, which leads to better understanding of what moves them and how they can be influenced. In this study, we present a hybrid approach that benefits from structural and content-based approaches to detect controversies in Twitteru0027s trending topics. Unlike many previous works, we do not limit ourselves to a certain domain. However, we do focus on social content written in Arabic. To the best of our knowledge, very limited attention has been invested to undertake this approach in studying controversy in general topics in this region of the world. We collect tweets on different trending topics from different domains. We apply several approaches for controversy detection and compare their outcomes to determine which one is the best.","Title: Leveraging fog computing and software defined systems for selective forwarding attacks detection in mobile wireless sensor networks\nYear: 2018\n\nAbstract: AbstractAbstractSelective forwarding is a major problem in wireless sensor networks (WSNs). The nature of sensor environments and the sensitivity of collected measurements in some fields such as war fields increase the need to prevent, detect, or mitigate the problem. One of the most used countermeasures for such problem is the use of voting system based on watchdogs' votes. However, this approach is not applicable in the case of mobile sensors. Mobile WSNs (MWSNs) is growing immensely due to the exposure of applications of mobile computing, vehicular networks, and Internet of things. This exposure has shed light on the security of using mobile sensors and raises the need to set appropriate methods for securing MWSNs against many attacks such as selective forwarding attacks. This paper introduces the problem of selective forwarding in MWSNs and discusses how the voting system used for mitigation; this problem in WSNs is not applicable in handling the problem in MWSNs due to sensors mobility. Therefore, the paper proposes a model that provides a global monitoring capability for tracing moving sensors and detecting malicious ones. The model leverages the infrastructure of fog computing to achieve this purpose. In addition, the paper suggests using software defined systems to be used along with the proposed model, which generalize the model to be used to secure MWSNs against other types of attacks easily and flexibly. The paper provides a complete algorithm, a comprehensive discussion and experiments that show the correctness and importance of the proposed approach.A model for detecting selective forwarding attacks in mobile wireless sensor networks. The model leverages the infrastructure of fog computing and its close proximity from sensor layers to provide a global intrusion detection systems that can trace and monitor mobile wireless sensors in their movement and detect malicious ones. View Figure","Title: GPU parallelization of sequence segmentation using information theoretic models.\nYear: 2018\n\nAbstract: Sequence segmentation has gained popularity in bioinformatics and particularly in studying DNA sequences. Information theoretic models have been used in providing accurate solutions in the segmentation of DNA sequences. Existing dynamic programming approaches provide optimal solution to the segmentation problem. However, their quadratic time complexity prohibits their applicability to long sequences. In this paper, we propose a parallel approach to improve the performance of a quasilinear sequence segmentation algorithm. The target segmentation technique is a divide-and-conquer recursive algorithm that is based on information theory principles and models. We present three parallel implementations that aim at reducing the segmentation time. The first implementation uses the multithreading capabilities of CPUs. The second one is a hybrid implementation that utilizes the synergy between the CPU and the multithreading power of GPUs. The third implementation is a variation of the hybrid approach where it utilizes the concept of unified memory between the CPU and the GPU instead of the standard memory copy approach. We demonstrate the applicability of the parallel implementations by testing them on real DNA sequences and randomly generated sequences with different lengths and different number of unique elements. The results show that the hybrid CPU-GPU approach outperforms the sequential implementation with a speedup of up to 5.9X while the CPU parallel implementation provides a poor speedup of only 1.7X.","Title: Automatic categorization of Arabic articles based on their political orientation.\nYear: 2018\n\nAbstract: The ability to automatically determine the political orientation of an article can be of great benefit in many areas from academia to security. However, this problem has been largely understudied for Arabic texts in the literature. The contribution of this work lies in two aspects. First, collecting and manually labeling a corpus of articles and comments from different political orientations in the Arab world and making different versions of it. Second, studying the performance of various feature reduction methods and various classifiers on these synthesized datasets. The two most popular feature extraction approaches for such a problem were compared, namely the Traditional Text Categorization (TC) approach and the Stylometric Features approach (SF). Although the experimental results show the superiority of the TC approach over the SF approach, the results also indicate that the latter approach can be significantly improved by adding new and more discriminating features. The experimental results also show that the feature selection techniques reduce the accuracies of the considered classifiers under the TC and SF approaches in general. The only exception is the Partition Membership (PM) technique which has an opposite effect. The highest accuracies are obtained when PM feature selection method is used with the Support Vector Machine (SVM) classifier.","Title: Cloud Data Centers Revenue Maximization Using Server Consolidation: Modeling And Evaluation\nYear: 2018\n\nAbstract: Cloud based data centers (DC) offering of virtually unlimited processing and storage capabilities faced a critical downside related to the enormous power consumption requirements. Powering and cooling DC has considerable share of the total DC operational costs in addition to the carbon tax. Server consolidation has been proposed as one viable solution that guarantees efficient power usage while maintaining quality of service requirements for the DC's services. This approach is simply about turning on and off the DC resources based on the incoming workload. In this paper, we are extending our previous work on expansion strategies for cloud services providers by considering the server consolidation technique in order to maximize the DC revenues while maintaining their service quality and reducing their carbon footprints.","Title: Collusion attacks mitigation in internet of things: a fog based model.\nYear: 2018\n\nAbstract: Collusion attacks are among the major security concerns nowadays due to the growth exposure in networks and communications. Internet of Things (IoT) environments are an attractive target for such type attacks. This paper discusses the problem of collusion attacks in IoT environments and how mobility of IoT devices increases the difficulty of detecting such types of attacks. It demonstrates how approaches used in detection collusion attacks in WSNs are not applicable in IoT environments. To this end, the paper introduces a model based on Fog Computing infrastructure to keep track of IoT devices and detect collusion attackers. The model uses fog computing layer for realtime monitoring and detection of collusion attacks in IoT environments. Moreover, the model uses a software defined systems layer to add a degree of flexibility for configuring Fog nodes to enable them to detect various types of collusion attacks. The paper provides algorithms, theorems, lemmas and mathematical proofs of the proposed model. Furthermore, the it highlights the possible overhead on fog nodes and network when applying the proposed model, and claims that fog layer infrastructure can provide the required resources for the scalability of the model. The experiments show how the proposed model can keep track of malicious nodes while moving from one cluster to other clusters in IoT environments in contrary to the models used in WSNs. Moreover, the experiments show that the proposed model can bear the computation overhead effectevilly, and reduces the power consumption of aggregator nodes in comparison to the models used in WSNs.","Title: Deep learning for Arabic NLP: A survey.\nYear: 2018\n\nAbstract: The recent advances in deep learning (DL) have caused breakthroughs in many fields such as computer vision, natural language processing (NLP) and speech processing. Many DL based approaches have been shown to produce state-of-the-art results on various tasks that are of great importance to online social networks (OSN) and social computing such as sentiment analysis (SA) and pharmacovigilance. NLP tasks are becoming very prominent in OSN and DL is offering researchers and practitioners exciting new directions to address these tasks. In this paper, we provide a survey of the published papers on using DL techniques for NLP. We focus on the Arabic language due to its importance, the scarcity of resources on it and the challenges associated with working on it. We notice that DL has yet to receive the attention it deserves from the Arabic NLP (ANLP) community compared with the attention it is getting for other languages despite the vast adoption of social networks in the Arab world. The majority of the early works on using DL for ANLP focused on OCR-related problems while the more recent ones are more diverse with the increasing interest in applying DL to SA, machine translation, diacritization, etc. This survey should serve as a guide for the young and growing ANLP community in order to help bridge the huge gap between ANLP literature and the much richer and more mature English NLP literature.","Title: Resilient service provisioning in cloud based data centers.\nYear: 2018\n\nAbstract: Cloud service providers usually have several geo-distributed data centers to meet these increasing demands. On peak times, these data centers receive unpredictable amount of workload that might go beyond their capacity and service providers need to distribute this workload with the objective of maximizing their revenue. This objective is hard to be guaranteed given the possible incidents that may cause cloud services interruption such as power outages, security breaches, natural disasters, etc. Since cloud providers are facing serious threats to the continuation of their services offering which will cause a serious decrease in their revenues, they must have proper strategies for handling such incidents. While a rich volume of recent research works focused on optimizing cloud data centers operations in order to reduce their operational cost (OPEX), little attention has been paid to such serious incidents and how they would affect service provisioning. In this study, we use mixed integer-linear programming to model the problem of maximizing revenue of a given workload by creating/expanding/renting data centers. The model takes into account a wide range of incidents affecting the users traffic and/or the data centers service capacities. Moreover, it handles many issues such as different types of costs (capital, operational and recovery), power consumption, carbon footprint, SLA related issues such as constraints (e.g., delay) and penalties, etc. To the best of our knowledge, this formation is unique among the body of existing works. Finally, to evaluate the proposed model and the impact of different parameters on its performance, several simulation experiments are conducted.","Title: Deep Recurrent neural network vs. support vector machine for aspect-based sentiment analysis of Arabic hotels' reviews.\nYear: 2018\n\nAbstract: •In this research, state-of-the-art approaches based on supervised machine learning are presented to tackle aspect-based sentiment analysis (ABSA) challenges of Arabic Hotels’ reviews.•Two approaches of deep recurrent neural network (RNN) and support vector machine (SVM) are implemented and trained along with lexical, word, syntactic, morphological, and semantic features.•The proposed approaches are evaluated using a reference dataset of Arabic Hotels’ reviews annotated using an ABSA framework presented in the Semantic Evaluation workshop 2016 (SemEval-ABSA16).•Evaluation results show that the SVM approach outperforms the other deep RNN approach in the research investigated tasks (T1: aspect category identification (E#A allocation), T2: aspect opinion target expression (OTE) extraction, and T3: aspect sentiment polarity identification).•Whereas, when focusing on the execution time required for training and testing the models, the deep RNN was faster especially for the second task.","Title: Cooperative mobile edge computing system for VANET-based software-defined content delivery.\nYear: 2018\n\nAbstract: Next-generation smart cities and Internet of Things (IoT) are getting more mature in terms of services and infrastructure requirements. Multiple smart vehicle applications are being conceived these days, including road traffic, road safety and infotainment, all of which are suffering from the WAN-latency problem. In this paper, we propose a Vehicular Adhoc Network (VANET)-based Software-Defined Edge Computing infrastructure supporting content delivery services among connected vehicles. The proposed approach leverages network base stations to embed mobile edge computing (MEC) services closer to the vehicles. Our approach can enable the delivery of more competitive services with reduced-latency by utilizing cooperative MEC search strategy for vehicle to infrastructure (V2I) communications as well as utilizing vehicle-level caching for vehicle to vehicle (V2V) communications between peers. The framework prototype has been implemented as a clean extension of the Mininet-WiFi emulator. Preliminary results serve as validation of the proposed framework and point out the potential benefits of the approach in mitigating WAN-latency in VANET.","Title: A Novel Recommender System Based on Apriori Algorithm for Requirements Engineering\nYear: 2018\n\nAbstract: Even though requirement gathering is an important step in the construction of any project, it imposes tedious work on the side of the system administrator. Recently, with the advent of data mining methods, many opportunities for improvements on the requirement gathering process have become available, one of the which is the use of recommender systems. Recommendation systems for requirements engineering can be used to provide the right information at the right time to requirements engineers. In this work, we propose a novel efficient recommender system based on Apriori algorithm for user requirements. Such recommender system would improve the accuracy of the obtained requirements and produce more comprehensive results. Furthermore, it would provide interesting information that can be used by various parties. Experimental work showed that our recommender system is efficient in term of execution time and can be widely implemented. In details, the system needed 11-21 seconds to execute when the number of users was 2000-4000.","Title: Determine Bipolar Disorder Level from Patient Interviews Using Bi-LSTM and Feature Fusion\nYear: 2018\n\nAbstract: Patients with Bipolar Disorder (BD) suffer from a brain disorder that cause them to change mood without reasons and prevent them from performing ordinary daily tasks. In this work, we classify patients with BD into one of its three levels: remission, hypo-mania, and mania, based solely on audio-visual recordings of structured interviews with these patients by the use of different deep learning techniques coupled with feature fusion and concatenation techniques along with a simple sliding window procedure. The results of our approach are promising and open up the door for many contributions and improvements in the future.","Title: A Deep Learning Approach for Amazon EC2 Spot Price Prediction\nYear: 2018\n\nAbstract: Spot Instances (SI) represent one of the ways cloud service providers use to deal with idle resources in off-peak periods, where these resources are being auctioned at low prices to customers with limited budgets in a dynamic manner. However, SI are poorly utilized due to issues like out-of-bid failures and bidding complexity. Thus, effective SI price models are of great importance to customers in order to plan their bidding strategies. This paper proposes a deep learning approach for Amazon EC2 SI price prediction, which is a time-series analysis (TSA) problem. The proposed Long Short-Term Memory (LSTM) approach is compared with a well-known classical (i.e., non deep learning) approach for TSA, which is AutoRegressive Integrated Moving Average (ARIMA), using different accuracy measures commonly used in TSA. The results show the superiority of the LSTM approach compared with the ARIMA approach in many aspects.","Title: The Impact of the Number of Eigen-Faces on the Face Recognition Accuracy using Different Distance Measures\nYear: 2018\n\nAbstract: The embedded and real-time systems are the main motivation for this research where the computations are critical to be reduced as much as possible. Face recognition method using eigen-faces yields good accuracy if enough eigen-faces are considered in the classification process. The more eigen-faces used, the more computation power is needed. In this paper, the main goal is to investigate the trade-off between the used number of eigen-faces and the accuracy and the needed computation power of face recognition. Three different distance measures are studied. Namely: Euclidean, block-city, and chess board distances are used. It is concluded that there is some optimum number of eigen-faces that provides the highest recognition rate and acceptable execution time. Moreover, the best number of eigen-faces highly depends on the selected distance measure.","Title: Impact of digital fingerprint image quality on the fingerprint recognition accuracy.\nYear: 2019\n\nAbstract: Despite the large body of work on fingerprint identification systems, most of it focused on using specialized devices. Due to the high price of such devices, some researchers directed their attention to digital cameras as an alternative source for fingerprints images. However, such sources introduce new challenges related to image quality. Specifically, most digital cameras compress captured images before storing them leading to potential losses of information. This study comes to address the need to determine the optimum ratio of the fingerprint image compression to ensure the fingerprint identification system’s high accuracy. This study is conducted using a large in-house dataset of raw images. Therefore, all fingerprint information is stored in order to determine the compression ratio accurately. The results proved that the used software functioned perfectly until a compression ratio of (30–40%) of the raw images; any higher ratio would negatively affect the accuracy of the used system.","Title: Truthful Multi-Unit Multi-Broker Auction Mechanism with Approximate Social-Welfare in Ad Hoc Cognitive Radio Networks.\nYear: 2019\n\nAbstract: In this paper, we consider the cooperative relay problem in an ad hoc cognitive radio (CR) network consisting of two types of CR nodes: continuous power nodes (CPN) and limited/intermittent power nodes (LPN). All CR nodes are associated with a Secondary Base Station (SBS). To improve energy efficiency and extend the lifetime of LPNs, we allow LPNs to cooperate with neighboring \"idle\" CPNs to deliver/relay their packets to the SBS while providing incentives to relay CPNs. A crucial challenge in enabling such type of cooperative relay lies in resolving the competition between LPNs on the potential relay opportunities. To regulate such a competitive environment, we develop an auction-based market model consisting of a set of brokers (idle CPNs) and a set of available frequency channels. The brokers iteratively issue short-term dynamic spectrum leases of these channels to competing LPNs. The SBS receives the LPN's bids and computes the output of the auction such that the maximum social-welfare (SW) is achieved by effectively allocating the available channels to the CR nodes. Computing a solution with maximum SW is, in general, NP-hard (even for one broker), but can be solved optimally using a pseudo polynomial time algorithm. Thus, we devise a polynomial time auction mechanism with near-optimal SW. Our mechanism enforces the very valuable property of strategy-proofness to mitigate the market manipulation problem resulting from untruthful behavior of the bidders (enforcing truthful bidders behavior). Simulation results are provided, which verify the effectiveness of our mechanism and demonstrate the significant gain achieved over a reference greedy solution.","Title: An enhanced framework for aspect-based sentiment analysis of Hotels' reviews: Arabic reviews case study\nYear: 2016\n\nAbstract: This research proposes a framework for aspect-based sentiment analysis (ABSA) of Hotels' reviews. The proposed framework consists of: (a) a reference human annotated Arabic dataset to support ABSA tasks such as aspect category identification, opinion target expression extraction, and opinion sentiment polarity. The dataset was annotated on both sentence-level and text-level, (b) baseline approach where a Support Vector Machine (SVM) was trained as part of the ABSA tasks, (c) baseline experiments and results, and (d) a common evaluation technique to provide a unified evaluation of future research working on the same dataset and ABSA tasks.","Title: A GPU-based breast cancer detection system using Single Pass Fuzzy C-Means clustering algorithm\nYear: 2016\n\nAbstract: Breast cancer is the most lethal type of cancer affecting women in the world. To improve the life quality of women, an early detection of this malignancy is always promising because this cancer is one of the cancers that can be managed and treated easily if it is detected earlier. Mammography is the standard screening method to diagnose breast cancer. At these days, there are a lot of researches that have been done to improve this screening method by benefiting from the enormous growth of technology. Graphics Processing Unit (GPU) is a parallel processor that can divide the complex computations tasks into subtasks and run them concurrently. Many medical imaging modalities offloaded their processing to this processor to help in improving the speed of healthcare systems in order to diagnose the illnesses in real time. This research introduces an acceleration method for the segmentation of the mammography images based on the GPU. In order to provide a better detection for the cancerous tumor, we use a modified version of the most common algorithm for image segmentation, which is the Single Pass Fuzzy C-Means (FCM) algorithm. The approach will be applied to a set of mammogram images to distinguish between malignant and benign cases. Additionally, the system is implemented on GPU parallel processor as well as the traditional CPU in order to compare the performance of both implementations. The performance results are compared according to the execution time and the speedup metrics. The proposed implementation on GPU provides a fine speedup compared to its serial implementation on CPU.","Title: Author gender identification from Arabic text.\nYear: 2017\n\nAbstract: The Gender Identification (GI) problem is concerned with determining the gender of a given text’s author. It has a wide range of academic/commercial applications in various fields including literature, security, forensics, electronic markets and trading, etc. To address this problem, researchers have proposed that the writing styles of authors of the same gender share certain aspects, which can be captured by certain stylometric features (SF). Another approach to address this problem focuses mainly on keywords occurrences in each document. This is known as the Bag-Of-Words (BOW) approach. In this work, we study and compare both approaches and focus on the Arabic language for which this problem is still largely understudied despite its importance. To the best of our knowledge, no previous work has considered these approaches for the GI problem of Arabic text. The comparison is carried out under different settings and the results show that the SF approach, which is much cheaper to train, can generate more accurate results under most settings. In fact, the best accuracy levels obtained by the SF and BOW approaches on our in-house dataset are 80.4% and 73.9%, respectively.","Title: Improving the performance of the needleman-wunsch algorithm using parallelization and vectorization techniques\nYear: 2019\n\nAbstract: The Needleman-Wunsch (NW) is a dynamic programming algorithm used in the pairwise global alignment of two biological sequences. In this paper, three sets of parallel implementations of the NW algorithm are presented using a mixture of specialized software and hardware solutions: POSIX Threads-based, SIMD Extensions-based and a GPU-based implementations. The three implementations aim at improving the performance of the NW algorithm on large scale input without affecting its accuracy. Our experiments show that the GPU-based implementation is the best implementation as it achieves performance 72.5X faster than the sequential implementation, whereas the best performance achieved by the POSIX threads and the SIMD techniques are 2X and 18.2X faster than the sequential implementation, respectively.","Title: Improving fuzzy C-mean-based community detection in social networks using dynamic parallelism\nYear: 2019\n\nAbstract: In Social Network Analysis (SNA), a common algorithm for community detection iteratively applies three phases: spectral mapping, clustering (using either the Fuzzy C-Means or the K-Means algorithms) and modularity computation. Despite its effectiveness, this method is not very efficient. A feasible solution to this problem is to use Graphics Processing Units. Moreover, due to the iterative nature of this algorithm, the emerging dynamic parallelism technology lends itself as a very appealing solution. In this work, we present different novel GPU implementations of both versions of the algorithm: Hybrid CPU-GPU, Dynamic Parallel and Hybrid Nested Parallel. These novel implementations differ in how much they rely on CPU and whether they utilize dynamic parallelism or not. We perform an extensive set of experiments to compare these implementations under different settings. The results show that the Hybrid Nested Parallel implementation provide about two orders of magnitude of speedup.","Title: An experimental framework for future smart cities using data fusion and software defined systems: The case of environmental monitoring for smart healthcare\nYear: 2020\n\nAbstract: With the evolution of the concept of Internet into the Internet-of-Things (IoT), in which sensors, devices, and everyday objects (things), are interconnected through the Internet, a large number of interconnected things are expected to produce massive amounts of data, which needs efficient processing in order to provide various services. IoT is of little worth without the ability to process, manage, and control a large amount of heterogeneous data in a timely and efficient manner. The concept of Data Fusion (DF) is based on combining heterogeneous multi-source data in order to produce high quality information, and reduce network traffic. On the other hand, the highly dynamic and heterogeneous nature of IoT requires very flexible and efficient management. The recent software defined systems paradigm is an appealing solution to reduce complexity and overhead in managing complex systems such as IoT by separating the control plane from the data plane. In this work, we present a novel experimental framework for IoT based environmental monitoring applications using concepts from DF and software defined systems (SDS). It is built on top of the software defined networking platform where the core components (the host, switch and the controller) are expanded to support other software defined systems components (such as software defined storage and security) and enable the applications of different DF techniques in IoT environments. We conduct several experiments to highlight features of the presented framework.","Title: Using long short-term memory deep neural networks for aspect-based sentiment analysis of Arabic reviews\nYear: 2019\n\nAbstract: This paper proposes a state-of-the-art research for aspect-based sentiment analysis of Arabic Hotels’ reviews using two implementations of long short-term memory (LSTM) neural networks. The first one is (a) a character-level bidirectional LSTM along with conditional random field classifier (Bi-LSTM-CRF) for aspect opinion target expressions (OTEs) extraction, and the second one is (b) an aspect-based LSTM for aspect sentiment polarity classification in which the aspect-OTEs are considered as attention expressions to support the sentiment polarity identification. Proposed approaches are evaluated using a reference dataset of Arabic Hotels’ reviews. Results show that our approaches outperform baseline research on both tasks with an enhancement of 39% for the task of aspect-OTEs extraction and 6% for the aspect sentiment polarity classification task.","Title: Using Deep Learning For Automatically Determining Correct Application Of Basic Quranic Recitation Rules\nYear: 2018\n\nAbstract: Quranic Recitation Rules (Ahkam Al-Tajweed) are the articulation rules that should be applied properly when reciting the Holy Quran. Most of the current automatic Quran recitation systems focus on the basic aspects of recitation, which are concerned with the correct pronunciation of words and neglect the other Ahkam Al-Tajweed that are related to the rhythmic and melodious way of recitation such as where to stop and how to \"stretch\" or \"merge\" certain letters. The only existing works on the latter parts are limited in terms of the rules they consider or the parts of Quran they cover. This paper comes to fill these gaps. It addresses the problem of identifying the correct usage of Ahkam Al-Tajweed in the entire Quran. Specifically, we focus on eight Ahkam Al-Tajweed faced by early learners of recitation. In the first part of our work, we used traditional audio processing techniques for feature extraction (such as Linear predictive Code (LPC), Mel-Frequency Cepstral Coefficient (MFCC), Wavelet Packet Decomposition (WPD) and Markov Model based Spectral Peak Location (HMM-SPL)) and classification (such as k-Nearest Neighbors (KNN), Support Vector Machines (SVM), and Random Forest (RF)) on an in-house dataset of thousands of audio recordings covering all occurrences of the rules under consideration in the entire Holy Quran by different reciters of both genders. In this part, we show how to improve the classification accuracy to surpass 97.7% by incorporating deep learning techniques. Specifically, this result is obtained by incorporating most traditional features with ones extracted using Convolutional Deep Belief Network (CDBN) while the classification is performed using SVM.","Title: Energy efficient dynamic resource management in cloud computing based on logistic regression model and median absolute deviation\nYear: 2018\n\nAbstract: •Optimize resource utilization and energy efficiency in cloud data centers.•Proposing host overloading detection algorithm based on logistic regression model and median absolute derivation.•Promote the VM consolidation technique.","Title: Parallel implementation for 3D medical volume fuzzy segmentation\nYear: 2020\n\nAbstract: •This research introduces the hybrid optimization of both CPU and GPU for image processing.•A novel 3D medical volume segmentation is proposed.•Modified FCM versions are used to increase the accuracy of volume segmentation.•Volumetric neighborhood has been considered to validate the actual 3D FCM.","Title: Collaboration networks of arab biomedical researchers\nYear: 2019\n\nAbstract: Social networks (SN) consist of a set of actors and connections between them. A collaboration network (ColNet) is a special type of SN, in which the actors represent researchers and the link between them indicate that they have co-authored at least one paper. ColNet analysis reveals how researchers interact and behave. A wide range of applications can be based on such studies. The current works on ColNet usually focus on a specific domain/discipline, country/geographical region or time interval. In our study, we focus on one of the understudied regions (the Arab world), and present a novel study on the ColNet of researchers in this region. The domain of interest in our study is biomedicine. We construct, analyze, and study ColNet of biomedical researchers in the Arab world. We divide the region of interest (the Arab world) into four geographical regions and look into the evolution of ColNet of each region separately over time. Our analysis reveals that there is an increase in the number of both authors and publications over time, and that authors tend to work in increasingly larger groups rather than working individually, which is consistent with what is assumed about the nature of research in this field. Our analysis also reveals that a researcher’s productivity is correlated with the amount of change in his/her circle of collaborators over time. For example, researchers working in stable or fixed groups and researchers who have completely different research group every few years are not necessarily the most productive ones.","Title: A comprehensive survey of arabic sentiment analysis\nYear: 2019\n\nAbstract: Sentiment analysis (SA) is a continuing field of research that lies at the intersection of many fields such as data mining, natural language processing and machine learning. It is concerned with the automatic extraction of opinions conveyed in a certain text. Due to its vast applications, many studies have been conducted in the area of SA especially on English texts, while other languages such as Arabic received less attention. This survey presents a comprehensive overview of the works done so far on Arabic SA (ASA). The survey groups published papers based on the SA-related problems they address and tries to identify the gaps in the current literature laying foundation for future studies in this field.","Title: Toward a new approach for sorting extremely large data files in the big data era\nYear: 2019\n\nAbstract: The extensive amount of data and contents generated today will require a paradigm shift in processing and management techniques for these data. One of the important data processing operations is the data sorting. Using multiple passes in external merge sort has a great influence on speeding up the sorting of extremely large data files. Since in large files, the swapping time is dominant in many applications, algorithms that minimize the swapping operations are normally superior to those which only focus on CPU time optimizations. In sorting extremely large files, external algorithms, such as the merge sort, are normally used. It is shown that using multiple passes over the data set, as proposed in our algorithm, has resulted in a great improvement in the number of swaps, thus, reducing the overall sorting time. Moreover, the proposed technique is suitable to be used with the emerging parallelization techniques such as GPUs. The reported results show the superiority of the proposed technique for “CPU only” and hybrid CPU–GPU implementations.","Title: Plant Classification in the Wild: A Transfer Learning Approach\nYear: 2018\n\nAbstract: Datasets specialized in wildlife usually contain imbalanced classes of natural wild images such as, for instance, plant images, which are acquired from the surrounding environment with natural scene background. Deep neural networks have proven their efficiency in classifying such datasets. However, such an approach requires a workaround to approximately balance the classes in order to prevent the occurrence of overfitting during the training phase of the neural network. Many approaches exist to overcome this problem includes over-sampling, undersampling, generating synthetic samples, data augmentation, etc. The iNaturalist species classification and detection dataset represents a good example of vastly imbalanced datasets. It contains 13 superclasses. This work focuses on the Plantae superclass and builds a Convolutional Neural Network to distinguish a subset of the subclasses of Plantae. Our model benefits from cutting-edge techniques such as transfer learning and data augmentation to obtain a reasonably high level of accuracy (78.76%).","Title: Enhancing Aspect-Based Sentiment Analysis of Arabic Hotels' reviews using morphological, syntactic and semantic features.\nYear: 2019\n\nAbstract: This research presents an enhanced approach for Aspect-Based Sentiment Analysis (ABSA) of Hotels’ Arabic reviews using supervised machine learning. The proposed approach employs a state-of-the-art research of training a set of classifiers with morphological, syntactic, and semantic features to address the research tasks namely: (a) T1:Aspect Category Identification, (b) T2:Opinion Target Expression (OTE) Extraction, and (c) T3: Sentiment Polarity Identification. Employed classifiers include Naïve Bayes, Bayes Networks, Decision Tree, K-Nearest Neighbor (K-NN), and Support-Vector Machine (SVM).The approach was evaluated using a reference dataset based on Semantic Evaluation 2016 workshop (SemEval-2016: Task-5). Results show that the supervised learning approach outperforms related work evaluated using the same dataset. More precisely, evaluation results show that all classifiers in the proposed approach outperform the baseline approach, and the overall enhancement for the best performing classifier (SVM) is around 53% for T1, around 59% for T2, and around 19% in T3.","Title: On the Use of Emojis to Train Emotion Classifiers.\nYear: 2019\n\nAbstract: Nowadays, the automatic detection of emotions is employed by many applications in different fields like security informatics, e-learning, humor detection, targeted advertising, etc. Many of these applications focus on social media and treat this problem as a classification problem, which requires preparing training data. The typical method for annotating the training data by human experts is considered time consuming, labor intensive and sometimes prone to error. Moreover, such an approach is not easily extensible to new domains/languages since such extensions require annotating new training data. In this study, we propose a distant supervised learning approach where the training sentences are automatically annotated based on the emojis they have. Such training data would be very cheap to produce compared with the manually created training data, thus, much larger training data can be easily obtained. On the other hand, this training data would naturally have lower quality as it may contain some errors in the annotation. Nonetheless, we experimentally show that training classifiers on cheap, large and possibly erroneous data annotated using this approach leads to more accurate results compared with training the same classifiers on the more expensive, much smaller and error-free manually annotated training data. Our experiments are conducted on an in-house dataset of emotional Arabic tweets and the classifiers we consider are: Support Vector Machine (SVM), Multinomial Naive Bayes (MNB) and Random Forest (RF). In addition to experimenting with single classifiers, we also consider using an ensemble of classifiers. The results show that using an automatically annotated training data (that is only one order of magnitude larger than the manually annotated one) gives better results in almost all settings considered.","Title: A detailed comparative study of open source deep learning frameworks.\nYear: 2019\n\nAbstract: Deep Learning (DL) is one of the hottest trends in machine learning as DL approaches produced results superior to the state-of-the-art in problematic areas such as image processing and natural language processing (NLP). To foster the growth of DL, several open source frameworks appeared providing implementations of the most common DL algorithms. These frameworks vary in the algorithms they support and in the quality of their implementations. The purpose of this work is to provide a qualitative and quantitative comparison among three of the most popular and most comprehensive DL frameworks (namely Googleu0027s TensorFlow, University of Montrealu0027s Theano and Microsoftu0027s CNTK). The ultimate goal of this work is to help end users make an informed decision about the best DL framework that suits their needs and resources. To ensure that our study is as comprehensive as possible, we conduct several experiments using multiple benchmark datasets from different fields (image processing, NLP, etc.) and measure the performance of the frameworksu0027 implementations of different DL algorithms. For most of our experiments, we find out that CNTKu0027s implementations are superior to the other ones under consideration.","Title: Arabic Text Diacritization Using Deep Neural Networks\nYear: 2019\n\nAbstract: Diacritization of Arabic text is both an interesting and a challenging problem at the same time with various applications ranging from speech synthesis to helping students learning the Arabic language. Like many other tasks or problems in Arabic language processing, the weak efforts invested into this problem and the lack of available (open-source) resources hinder the progress towards solving this problem. This work provides a critical review for the currently existing systems, measures and resources for Arabic text diacritization. Moreover, it introduces a much-needed free-for-all cleaned dataset that can be easily used to benchmark any work on Arabic diacritization. Extracted from the Tashkeela Corpus, the dataset consists of 55K lines containing about 2.3M words. After constructing the dataset, existing tools and systems are tested on it. The results of the experiments show that the neural Shakkala system significantly outperforms traditional rule-based approaches and other closed-source tools with a Diacritic Error Rate (DER) of 2.88% compared with 13.78%, which the best DER for the non-neural approach (obtained by the Mishkal tool).","Title: Recognizing Emotion from Speech Based on Age and Gender Using Hierarchical Models\nYear: 2019\n\nAbstract: Age and gender are two factors that affect the physiologic and acoustic features of human voice. In fact, most of the speech emotion recognition applications use these voice features as a foundation to complete the classification task. Significant improvements have been made for voice emotion recognition; and several studies have addressed the age and gender identification from speech topics. We studied the effect of age and gender on the emotion recognition applications. In our work, we built hierarchical classification models to investigate the importance of identifying the age and gender before identifying the emotional label. We compared the performance of four different models and presented the relationship between the age \\ gender and the emotion recognition accuracy. Our results showed that using a separated emotion model for each of gender and age category gives a higher accuracy compared with using one classifier for all the data.","Title: Sarcasm Detection and Quantification in Arabic Tweets\nYear: 2021\n\nAbstract: The role of predicting sarcasm in the text is known as automatic sarcasm detection. Given the prevalence and challenges of sarcasm in sentiment-bearing text, this is a critical phase in most sentiment analysis tasks. With the increasing popularity and usage of different social media platforms among users around the world, people are using sarcasm more and more in their day-to-day conversations, social media posts and tweets, and it is considered as a way for people to express their sentiment about some certain topics or issues. As a result of the increasing popularity, researchers started to focus their research endeavors on detecting sarcasm from a text in different languages especially the English language. However, the task of sarcasm detection is a challenging task due to the nature of sarcastic texts; which can be relative and significantly differs from one person to another depending on the topic, region, the user's mentality and other factors. In addition to the aforementioned challenges, sarcasm detection in the Arabic language has its own challenges due to the complexity of the Arabic language, such as being morphologically rich, with many dialects that significantly vary between each other, while also being lowly resourced when compared to English. In recent years, only few research attempts started tackling the task of sarcasm detection in Arabic, including creating and collecting corpora, organizing workshops and establishing baseline models. This paper intends to create a new humanly annotated Arabic corpus for sarcasm detection collected from tweets, and implementing a new approach for sarcasm detection and quantification in Arabic tweets. The annotation technique followed in this paper is unique in sarcasm detection and the proposed approach tackles the problem as a regression problem instead of classification; i.e., the model attempts to predict the level of sarcasm instead of binary classification (sarcastic vs. non-sarcastic) for the purpose of tackling the complex and user-dependent nature of the sarcastic text. The humanly annotated dataset will be available to the public for any usage.","Title: Visual Question Answering In The Medical Domain Based On Deep Learning Approaches: A Comprehensive Study\nYear: 2021\n\nAbstract: Visual Question Answering (VQA) in the medical domain has attracted more attention from research communities in the last few years due to its various applications. This paper investigates several deep learning approaches in building a medical VQA system based on ImageCLEF's VQA-Med dataset, which consists of about 4K images with about 15K question-answer pairs. Due to the wide variety of the im-ages and questions included in this dataset, the proposed model is a hierarchical one consisting of many sub-models, each tailored to handle certain questions. For that, a special model is built to classify the questions into four categories, where each category is handled by a separate sub-model. At their core, all of these models consist of pre-trained Convolution Neural Networks (CNN). In order to get the best results, extensive experiments are performed and various techniques are employed including Data Aug-mentation (DA), Multi-Task Learning (MTL), Global Average Pooling (GAP), Ensembling, and Sequence to Sequence (Seq2Seq) models. Overall, the final model achieves 60.8 accuracy and 63.4 BLEU score, which are competitive with the state-of-the-art results despite using less demanding and simpler sub-models. (c) 2021 Elsevier B.V. All rights reserved.","Title: Neural Arabic Text Diacritization: State-of-the-Art Results and a Novel Approach for Arabic NLP Downstream Tasks\nYear: 2022\n\nAbstract: AbstractIn this work, we present several deep learning models for the automatic diacritization of Arabic text. Our models are built using two main approaches, viz. Feed-Forward Neural Network (FFNN) and Recurrent Neural Network (RNN), with several enhancements such as 100-hot encoding, embeddings, Conditional Random Field (CRF), and Block-Normalized Gradient (BNG). The models are tested on the only freely available benchmark dataset and the results show that our models are either better or on par with other models even those requiring human-crafted language-dependent post-processing steps, unlike ours. Moreover, we show how diacritics in Arabic can be used to enhance the models of downstream NLP tasks such as Machine Translation (MT) and Sentiment Analysis (SA) by proposing novel Translation over Diacritization (ToD) and Sentiment over Diacritization (SoD) approaches.","Title: Deep Learning-Based Disk Herniation Computer Aided Diagnosis System From MRI Axial Scans\nYear: 2022\n\nAbstract: Computer-aided diagnosis (CAD) systems have been the focus of many researchers in both computer and medical fields. In this paper, we build two convolutional neural network (CNN) based CAD systems for diagnosing lumbar disk herniation from Magnetic Resonance Imaging (MRI) axial scans. The first one is a disk herniation detection CAD system which is a binary CAD system that determines whether the case image contains disk herniation or not. The second system is a disk herniation type classification CAD system which can determine the type of the disk herniation in the image if one exists. To train and test the proposed systems, an image set is built and annotated with the help of a radiologist. In order to get rid of the \"noisy\" parts of the input images and reduce their complexity, we experiment with different ROI extraction methods. The image set is also preprocessed and enlarged using augmentation techniques to make it suitable to be used with CNN. There are many novel aspects of this work. First, the problems of disk herniation detection and recognition from axial scans are not well-studied in the literature. Second, we use deep learning techniques which produces ground-breaking results in many image processing tasks, but are yet to reach their full potential with medical image processing in general. Finally, we explore the use of many innovative techniques such as data augmentation and transfer learning, which greatly improves the accuracy of our models. The results of our systems are impressive with a 95.65% accuracy for the detection problem and a 91.38% accuracy for the recognition problem.","Title: An Adaptive Black-box Defense against Trojan Attacks on Text Data\nYear: 2021\n\nAbstract: Trojan backdoor is a poisoning attack against Neural Network (NN) classifiers in which adversaries try to exploit the (highly desirable) model reuse property to implant Trojans into model parameters for backdoor breaches through a poisoned training process. Most of the proposed defenses against Trojan attacks assume a white-box setup, in which the defender either has access to the inner state of N...","Title: Plant classification in the wild: Energy evaluation for deep learning models\nYear: 2022\n\nAbstract: Having a system that can take an image of a natural scene and accurately classify the plants in it is of undeniable importance. However, the complexities of dealing with natural scene images and the vast diversity of plants in the wild make designing such a classifier a challenging task. Deep Learning (DL) lends itself as viable solution to tackle such complex problem. However, advanced in DL architectures and software (including DL frameworks) come with a high cost in terms of energy consumption especially when employing Graphics Processing Units (GPU). As data expands rapidly, the need to create energy-aware models increases in order to reduce energy consumption and move towards “Greener AI”. Since the problem of designing energy-aware architectures for plant classification has not been studied significantly in the literature, our work comes to start bridging this gap by focusing not only on the models’ performance, but also on their energy usage on both CPU and GPU platforms. We consider different state-of-the-art Convolutional Neural Networks (CNN) architectures and train them on two famous challenging plants datasets: iNaturalist and Herbarium. Our experiments are meant to highlight the trade-off between accuracy and energy consumption. For examples, the results show that while GPU-bound models can be about 40% faster in terms of training time than simple models running on CPU, the latter’s energy consumption is only two thirds of the former. We hope that such findings will encourage the community to reduce its reliance on accuracy measures to compare different architectures and start taking other factors into account such as power consumption, simplicity, etc.","Title: Social Network Analysis of the Panama Papers Concentrating on the MENA Region\nYear: 2022\n\nAbstract: The release of millions of financial documents, which has been known as the 'WikiLeaks' of the financial world (a.k.a. 'Panama Papers'), has dragged global attention in how highly structured means applied by some of the elite to conceal their financial assets. Consequently, significant financial corruption allegations were raised. We concentrate on a somewhat overlooked region, the Middle East and North Africa (MENA) region. This study aims to use social network analytics to study the information contained in these documents. We are checking the major players in the MENA's trends and patterns to determine if it matches the known economic powers. The analysis reveals that while the constructed network enjoys some typical characteristics, many interesting observations and properties are worth discussing. Specifically, using the extracted network consisting of 62 987 nodes and 84 692 edges, our social network analysis finding shows that, perhaps surprisingly, the nodes or the social network are not necessarily directly correlated with perceived economic influence.","Title: Using K-means Clustering Ensemble to Improve the Performance in Recommender Systems\nYear: 2022\n\nAbstract: Collaborative filtering methods are often utilized in the industry of recommender systems. They work by identifying users with similar tastes and recommending items for each active user. Besides, clustering techniques are extensively utilized to create systems based on collaborative filtering recommendation in the context of big data. Nevertheless, the cluster ensemble has emerged in last years as a powerful technique that can replace single clustering algorithms in enhancing the performance of recommendation and prediction. This paper presents a k-means clustering ensemble-based method to improve the performance in recommender systems. The proposed system incorporates the Cosine Similarity and the Pearson Correlation Coefficient as similarity metrics to form clusters. Moreover, it uses the HyperGraph Partitioning Algorithm (HGPA) to combine the results of the k-means clustering technique. The recommendation algorithm constructs the recommendations based on the clusters obtained earlier by the HGPA ensemble clustering. To this end, it finds the nearest cluster for each active user and selects its top N items. Finally, it recommends these top items to the user‘s favorite list. The experiments on two well-known datasets demonstrate that cluster ensembles by HGPA outperform the baseline methods."],"string":"[\n \"Title: An analytical study of Arabic sentiments: Maktoob case study\\nYear: 2013\\n\\nAbstract: The problem of automatically extracting opinions and emotions from textual data have gained a lot of interest recently. Unfortunately, most studies on Sentiment Analysis (SA) focus mainly on the English language, whereas studies considering other important and wide-spread languages such as Arabic are few. Moreover, publicly-available Arabic datasets are seldom found on the Web. In this work, a labeled dataset of Arabic reviews/comments is collected from a social networking website (Yahoo!-Maktoob). A detailed analysis of different aspects of the collected dataset such as the reviews' length, the numbers of likes/dislikes, the polarity distribution and the languages used is presented. Finally, the dataset is used to test popular classifiers commonly used for SA.\",\n \"Title: Automatic Arabic text categorization: A comprehensive comparative study\\nYear: 2015\\n\\nAbstract: Text categorization or classification (TC) is concerned with placing text documents in their proper category according to their contents. Owing to the various applications of TC and the large volume of text documents uploaded on the Internet daily, the need for such an automated method stems from the difficulty and tedium of performing such a process manually. The usefulness of TC is manifested in different fields and needs. For instance, the ability to automatically classify an article or an email into its right class (Arts, Economics, Politics, Sports, etc.) would be appreciated by individual users as well as companies. This paper is concerned with TC of Arabic articles. It contains a comparison of the five best known algorithms for TC. It also studies the effects of utilizing different Arabic stemmers (light and root-based stemmers) on the effectiveness of these classifiers. Furthermore, a comparison between different data mining software tools (Weka and RapidMiner) is presented. The results illustrate the good accuracy provided by the SVM classifier, especially when used with the light10 stemmer. This outcome can be used in future as a baseline to compare with other unexplored classifiers and Arabic stemmers.\",\n \"Title: CloudExp: A comprehensive cloud computing experimental framework.\\nYear: 2014\\n\\nAbstract: Cloud computing is an emerging and fast-growing computing paradigm that has gained great interest from both industry and academia. Consequently, many researchers are actively involved in cloud computing research projects. One major challenge facing cloud computing researchers is the lack of a comprehensive cloud computing experimental tool to use in their studies. This paper introduces CloudExp, a modeling and simulation environment for cloud computing. CloudExp can be used to evaluate a wide spectrum of cloud components such as processing elements, data centers, storage, networking, Service Level Agreement (SLA) constraints, web-based applications, Service Oriented Architecture (SOA), virtualization, management and automation, and Business Process Management (BPM). Moreover, CloudExp introduces the Rain workload generator which emulates real workloads in cloud environments. Also, MapReduce processing model is integrated in CloudExp in order to handle the processing of big data problems.\",\n \"Title: SDStorage: A Software Defined Storage Experimental Framework\\nYear: 2015\\n\\nAbstract: With the rapid growth of data centers and the unprecedented increase in storage demands, the traditional storage control techniques are considered unsuitable to deal with this large volume of data in an efficient manner. The Software Defined Storage (SDStore) comes as a solution for this issue by abstracting the storage control operations from the storage devices and set it inside a centralized controller in the software layer. Building a real SDStore system without any simulation and emulation is considered an expensive solution and may have a lot of risks. Thus, there is a need to simulate such systems before the real-life implementation and deployment. In this paper we present SDStorage, an experimental framework to provide a novel virtualized test bed environment for SDStore systems. The main idea of SDStorage is based on the Mininet Software Defined Network (SDN) Open Flow simulator and is built over of it. The main components of Mininet, which are the host, the switch and the controller, are customized to serve the needs of SDStore simulation environments.\",\n \"Title: To Build or Not to Build? Addressing the Expansion Strategies of Cloud Providers\\nYear: 2014\\n\\nAbstract: With the increasing popularity gained by cloud computing systems, cloud providers are facing rapidly increasing traffic loads, which requires them to have proper expansion strategies for their datacenters. Expanding the capacities of existing datacenters or building new ones requires many factors to be considered such as power resources availability, prices (of power, land, etc.), carbon tax, free cooling options, and renewable energy. In this paper, we address the problem of deciding the best expansion strategy for a given cloud provider by deciding whether it is beneficial for the cloud provider to build new datacenters or to simply expand the datacenters it currently has.\",\n \"Title: Compression-based arabic text classification\\nYear: 2014\\n\\nAbstract: Text classification (TC) is one of the fundamental problems in text mining. Plenty of works exist on TC with interesting approaches and excellent results; however, most of these works follow a word-based approach for feature extraction. In this work, we are interested in an alternative (byte-based or character-based) approach known as compression-based TC (CTC). CTC has been used for some languages such as English and Portuguese and it is shown to have certain advantages/ disadvantages compared with word-based approaches. This work applies CTC on the Arabic language with the purpose of investigating whether these advantages/disadvantages exists for the Arabic language as well. The results are encouraging as they show the viability of using CTC for Arabic TC.\",\n \"Title: Truthful Spectrum Auctions With Approximate Social-Welfare or Revenue\\nYear: 2014\\n\\nAbstract: In cellular networks, a recent trend in research is to make spectrum access dynamic in the spatial and temporal dimensions for the sake of efficient utilization of spectrum. In one such model, the spectrum is divided into channels and periodically allocated to competing base stations using an auction-based market mechanism. An “efficient” auction mechanism is essential to the success of such a dynamic spectrum access model. A key objective in designing an auction mechanism is “truthfulness.” Combining this objective with an optimization of some social choice function (such as the social-welfare or the generated revenue) is highly desirable. In this paper, we design polynomial-time spectrum auction mechanisms that are truthful and yield an allocation with O(1)-approximate social-welfare or revenue. Our mechanisms generalize to general interference models. To the best of our knowledge, ours is the first work to design polynomial-time truthful spectrum auction mechanisms with a constant-factor approximation of either the expected revenue or the social-welfare. We demonstrate the performance of our designed mechanism through simulations.\",\n \"Title: Collaboration Networks of Levantine Biomedical Researchers\\nYear: 2014\\n\\nAbstract: Social networks and social network analysis are one of the hottest topics in Academia as well as in the industry nowadays. A collaboration (or a co-authorship) network, is a type of social networks in which the actors are scholars and researchers and a tie between two actors represents the fact that they have co-authored at least one paper together. The objective of this work is to study scientific collaborations patterns in order to construct and analyze a collaboration network of authors in the Levant region. The focus is on publications in the biomedical field. Even though collaboration networks have been constructed for several fields and in different regions, the Middle East and North Africa (MENA) region in general is still largely understudied, to the best of our knowledge. Constructing the network is challenging due to many reasons including the lack of sufficient data. Analyzing the constructed network reveal that there is an increase in both biomedical researchers and published papers that matches the spur of growth that the Levant-based academic institutions enjoyed in the early 2000s. Among many things, the analysis reveal that local researchers are shifting from working individually or with non-local collaborators to working within relatively large groups of local collaborators.\",\n \"Title: A new broadcast scheme for sensor networks\\nYear: 2014\\n\\nAbstract: The future of the Internet envisions connecting all objects around us in a proactive way in what is known as the Internet of Things (IoT). Wireless sensor networks (WSN) represent a prominent example of IoT. Broadcasting is one of most important procedures in wireless networks. However, broadcast may have immense effect on the network's performance in what is know as the broadcast storm problem. In this work, we propose a new broadcast scheme to address the broadcast storm problem. The proposed scheme is an amalgamation of three popular broadcast schemes: the probabilistic scheme, the distance-based scheme and the counter-based scheme. The proposed scheme is implemented and tested on the QualNet simulation environment where it is integrated into the Ad-hoc On-Demand Distance Vector (AODV) routing protocol. The results show that the percentage of saved rebroadcasts can reach more than 50% in some cases.\",\n \"Title: The analysis of large-scale climate data: Jordan case study\\nYear: 2014\\n\\nAbstract: The analysis of large-scale data for the purpose of extracting patterns is applicable to several research fields. In this paper, a dataset of climate related historical data from Jordan is collected. The main focus is to study evolution criteria of weather attributes in Jordan over the evaluated period of time and how it is connected to climate change. Results showed that humidity and dew point weather attributes are going to face a significant increase in the future. Such an increase is expected to have a direct, as well as an indirect impact on human life.\",\n \"Title: Joint Routing, Channel Assignment, and Scheduling for Throughput Maximization in General Interference Models\\nYear: 2010\\n\\nAbstract: Throughput optimization in wireless networks with multiple channels and multiple radio interfaces per node is a challenging problem. For general traffic models (given a set of source-destination pairs), optimization of throughput entails design of ??efficient?? routes between the given source-destination pairs, in conjunction with (1) assignment of channels to interfaces and communication links and (2) scheduling of noninterfering links for simultaneous transmission. Prior work has looked at restricted versions of the above problem. In this paper, we design approximation algorithms for the joint routing, channel assignment, and link scheduling problem in wireless networks with general interference models. The unique contributions of our work include addressing the above joint problem in the context of physical interference model and single-path routing (wherein, traffic between a source-destination pair is restricted to a single path). To the best of our knowledge, ours is the first work to address the throughput maximization problem in such general contexts. For each setting, we design approximation algorithms with provable performance guarantees. We demonstrate the effectiveness of our algorithms in general contexts through simulations.\",\n \"Title: Truthful spectrum auctions with approximate revenue\\nYear: 2011\\n\\nAbstract: In cellular networks, a recent trend is to make spectrum access dynamic in the spatial and temporal dimensions, for the sake of efficient utilization of spectrum. In such a model, the spectrum is divided into channels and periodically allocated to competing base stations using an auction-based market mechanism. An “efficient” auction mechanism is essential to the success of such a dynamic spectrum access model. Two of the key objectives in designing an auction mechanism are “truthfulness” and revenue maximization. In this article, we design a polynomial-time spectrum auction mechanism that is truthful and yields an allocation with O(1)-approximate expected revenue, in the Bayesian setting. Our mechanism generalizes to general interference models. To the best of our knowledge, ours is the first work to design a polynomial-time truthful spectrum auction mechanism with a performance guarantee on the expected revenue. We demonstrate the performance of our designed mechanism through simulations.\",\n \"Title: An extensive study of the Bag-of-Words approach for gender identification of Arabic articles\\nYear: 2014\\n\\nAbstract: The prevalent use of Online Social Networks (OSN) and the anonymity and lack of accountability they inherent from being online give rise to many problems related to finding the connection between the massive amount of text data on OSN and the people who actually wrote them. Analyzing text data for such purposes is called authorship analysis. This work is focused on one specific type of authorship analysis, which is identifying the author's gender. Gender identification has various applications from marketing to security. The focus of this work is on Arabic articles. The problem is basically a classification problem and the current approaches differ in the way they compute the features of each document. However, they all agree on following some “stylometric features” approach. Unlike these works, ours treat this problem as a variation of the Text Classification (TC) problem and follow the Bag-Of-Words (BOW) approach for feature selection. We perform an extensive set of experiments on the feature selection and classification phase and the results show that such an approach yield surprisingly high results.\",\n \"Title: An Extensive Study Of Authorship Authentication Of Arabic Articles\\nYear: 2017\\n\\nAbstract: Purpose - The authorship authentication (AA) problem is concerned with correctly attributing a text document to its corresponding author. Historically, this problem has been the focus of various studies focusing on the intuitive idea that each author has a unique style that can be captured using stylometric features (SF). Another approach to this problem, known as the bag-of-words (BOW) approach, uses keywords occurrences/frequencies in each document to identify its author. Unlike the first one, this approach is more language-independent. This paper aims to study and compare both approaches focusing on the Arabic language which is still largely understudied despite its importance.Design/methodology/approach - Being a supervised learning problem, the authors start by collecting a very large data set of Arabic documents to be used for training and testing purposes. For the SF approach, they compute hundreds of SF, whereas, for the BOW approach, the popular term frequency-inverse document frequency technique is used. Both approaches are compared under various settings.Findings - The results show that the SF approach, which is much cheaper to train, can generate more accurate results under most settings.Practical implications - Numerous advantages of efficiently solving the AA problem are obtained in different fields of academia as well as the industry including literature, security, forensics, electronic markets and trading, etc. Another practical implication of this work is the public release of its sources. Specifically, some of the SF can be very useful for other problems such as sentiment analysis.Originality/value - This is the first study of its kind to compare the SF and BOW approaches for authorship analysis of Arabic articles. Moreover, many of the computed SF are novel, while other features are inspired by the literature. As SF are language-dependent and most existing papers focus on English, extra effort must be invested to adapt such features to Arabic text.\",\n \"Title: Multi-agent based dynamic resource provisioning and monitoring for cloud computing systems infrastructure\\nYear: 2015\\n\\nAbstract: The cloud computing paradigm provides a shared pool of resources and services with different models delivered to the customers through the Internet via an on-demand dynamically-scalable form charged using a pay-per-use model. The main problem we tackle in this paper is to optimize the resource provisioning task by shortening the completion time for the customers' tasks while minimizing the associated cost. This study presents the dynamic resources provisioning and monitoring (DRPM) system, a multi-agent system to manage the cloud provider's resources while taking into account the customers' quality of service requirements as determined by the service-level agreement (SLA). Moreover, DRPM includes a new virtual machine selection algorithm called the host fault detection algorithm. The proposed DRPM system is evaluated using the CloudSim tool. The results show that using the DRPM system increases resource utilization and decreases power consumption while avoiding SLA violations.\",\n \"Title: Software Defined Cognitive Radio Network Framework: Design and Evaluation\\nYear: 2015\\n\\nAbstract: Software defined networking SDN provides a novel network resource management framework that overcomes several challenges related to network resources management. On the other hand, Cognitive Radio CR technology is a promising paradigm for addressing the spectrum scarcity problem through efficient dynamic spectrum access DSA. In this paper, the authors introduce a virtualization based SDN resource management framework for cognitive radio networks CRNs. The framework uses the concept of multilayer hypervisors for efficient resources allocation. It also introduces a semi-decentralized control scheme that allows the CRN Base Station BS to delegate some of the management responsibilities to the network users. The main objective of the proposed framework is to reduce the CR users' reliance on the CRN BS and physical network resources while improving the network performance by reducing the control overhead.\",\n \"Title: Towards improving resource management in cloud systems using a multi-agent framework.\\nYear: 2016\\n\\nAbstract: With the goal of efficient sharing of resources and services, the cloud computing paradigm has gained a lot of interest recently. This work focuses on improving the resource utilisation by optimising the resource provisioning through a multi-agent framework in which different agents are responsible for different tasks including the monitoring of customers (behaviour, resource usage patterns and QoS requirements as stated in the SLA) and available resources based on customeru0027s requests. Moreover, we introduce the concept of TaskFlow which allows a more elastic resources provisioning to match the customer real usage of the resources. The proposed system is implemented and tested on the CloudSim simulator and the results show it increases resource utilisation and decreases power consumption while avoiding SLA violations. The results also show that the introduction of the concept of TaskFlow into our proposed system leads to more resource saving but with a higher risk of SLA violations.\",\n \"Title: Evaluating map reduce tasks scheduling algorithms over cloud computing infrastructure\\nYear: 2015\\n\\nAbstract: Efficiently scheduling MapReduce tasks is considered as one of the major challenges that face MapReduce frameworks. Many algorithms were introduced to tackle this issue. Most of these algorithms are focusing on the data locality property for tasks scheduling. The data locality may cause less physical resources utilization in non-virtualized clusters and more power consumption. Virtualized clusters provide a viable solution to support both data locality and better cluster resources utilization. In this paper, we evaluate the major MapReduce scheduling algorithms such as FIFO, Matchmaking, Delay, and multithreading locality MTL on virtualized infrastructure. Two major factors are used to test the evaluated algorithms: the simulation time and the energy consumption. The evaluated schedulers are compared, and the results show the superiority and the preference of the MTL scheduler over the other existing schedulers. Also, we present a comparison study between virtualized and non-virtualized clusters for MapReduce tasks scheduling. Copyright © 2015 John Wiley & Sons, Ltd.\",\n \"Title: SDIoT: a software defined based internet of things framework\\nYear: 2015\\n\\nAbstract: The internet of things (IoT) represent the current and future state of the Internet. The large number of things (objects), which are connected to the Internet, produce a huge amount of data that needs a lot of effort and processing operations to transfer it to useful information. Moreover, the organization and control of this large volume of data requires novel ideas in the design and management of the IoT network to accelerate and enhance its performance. The software defined systems is a new paradigm that appeared recently to hide all complexity in traditional system architecture by abstracting all the controls and management operations from the underling devices (things in the IoT) and setting them inside a middleware layer, a software layer. In this work, a comprehensive software defined based framework model is proposed to simplify the IoT management process and provide a vital solution for the challenges in the traditional IoT architecture to forward, store, and secure the produced data from the IoT objects by integrating the software defined network, software defined storage, and software defined security into one software defined based control model.\",\n \"Title: A Prototype For A Standard Arabic Sentiment Analysis Corpus\\nYear: 2016\\n\\nAbstract: The researchers in the field of Arabic Sentiment Analysis (SA) need a relatively big standard corpus to conduct their studies. There are a number of existing datasets; however, they suffer from certain limitations such as the small number of reviews or topics they contain, the restriction to Modern Standard Arabic (MSA), etc., Moreover, most of them are in-house datasets that are not publicly available. Therefore, this study aims to establish a flexible and relatively big standard Arabic SA corpus that can he considered as a foundation to build larger Arabic corpora. In addition to MSA, this corpus contains reviews written in the five main Arabic dialects (Egyptian, Levantine, Arabian Peninsula, Mesopotamian, and Maghrebi group). Furthermore, this corpus has other five types of reviews (English, mixed MSA English, French, mixed MSA and Emoticons, and mixed Egyptian and Emoticons). This corpus is released for free to be used by researchers in this field, where it is characterized by its flexibility in allowing the users to add, remove, and revise its contents. The total number of topics and reviews of this initial version are 250 and 1,442, respectively. The collected topics are distributed equally among five domains (classes): Economy, Food-Life style, Religion, Sport, and Technology, where each domain has 50 topics. This corpus is built manually to ensure the highest quality to the researchers in this field.\",\n \"Title: Towards Improving the Lexicon-Based Approach for Arabic Sentiment Analysis\\nYear: 2014\\n\\nAbstract: The emergence of the Web 2.0 technology generated a massive amount of raw data by enabling Internet users to post their opinions on the web. Processing this raw data to extract useful information can be a very challenging task. An example of important information that can be automatically extracted from the users' posts is their opinions on different issues. This problem of Sentiment Analysis SA has been studied well on the English language and two main approaches have been devised: corpus-based and lexicon-based. This work focuses on the later approach due to its various challenges and high potential. The discussions in this paper take the reader through the detailed steps of building the main two components of the lexicon-based SA approach: the lexicon and the SA tool. The experiments show that significant efforts are still needed to reach a satisfactory level of accuracy for the lexicon-based Arabic SA. Nonetheless, they do provide an interesting guide for the researchers in their on-going efforts to improve lexicon-based SA.\",\n \"Title: A GPU-based implementations of the fuzzy C-means algorithms for medical image segmentation\\nYear: 2015\\n\\nAbstract: Fuzzy clustering is one of the most popular techniques in medical image segmentation. The fuzzy C-means (FCM) algorithm has been widely used as it provides better performance and more information than other algorithms. As the data set becomes large, the serial implementation of the FCM algorithm becomes too slow to accomplish the clustering task within acceptable time. Hence, a parallel implementation [for example, using today’s fast graphics processing unit (GPU)] is needed. In this paper, we implement brFCM algorithm, a faster variant of the FCM algorithm, on two different GPU cards, Tesla M2070 and Tesla K20m. We compare our brFCM GPU-based implementation with its CPU-based sequential implementation. Moreover, we compare brFCM with the traditional version of the FCM algorithm. The experiments used lung CT and knee MRI images for clustering. The results show that our implementation has a significant improvement over the traditional CPU sequential implementation. GPU parallel brFCM is 2.24 times faster than its CPU implementation, and 23.43 times faster than a GPU parallel implementation of the traditional FCM.\",\n \"Title: Optimizing expansion strategies for ultrascale cloud computing data centers\\nYear: 2015\\n\\nAbstract: With the increasing popularity gained by cloud computing systems over the past few years, cloud providers have built several ultrascale data centers at a variety of geographical locations, each including hundreds of thousands of computing servers. Since cloud providers are facing rapidly increasing traffic loads, they must have proper expansion strategies for their ultrascale data centers. The decision of expanding the capacities of existing data centers or building new ones over a certain period requires considering many factors, such as high power consumption, availability of resources, prices (of power, land, etc.), carbon tax, free cooling options, and availability of local renewable power generation. While a rich volume of recent research works focused on reducing the operational cost (OPEX) of the data centers, there exists no prior work, to the best of our knowledge, on investigating the trade-off between minimizing the OPEX of the data centers and maximizing their revenue from the services they offer while respecting the service level agreement (SLA) with their customers. In this study, we model this optimization problem using mixed integer-linear programming. Our proposed model is unique compared to the published works in many aspects such as its ability to handle realistic scenarios in which both data centers’ resources (servers) and user generated traffic are heterogeneous. To evaluate the proposed model and the impact of different parameters on it’s performance, several simulation experiments are conducted.\",\n \"Title: Software defined cloud: Survey, system and evaluation.\\nYear: 2016\\n\\nAbstract: Next generation cloud systems will require a paradigm shift in how they are constructed and managed. Conventional control and management platforms are facing considerable challenges regarding flexibility, dependability and security that next generation systems must handle. The cloud computing technology has already contributed in alleviating a number of the problems associated with resource allocation, utilization and management. However, many of the elements of a well-designed cloud environment remain “stiff” and hard to modify and adapt in an integrated fashion. This includes the underlying networking topologies, many aspects of the user control over IaaS, PaaS or SaaS layers, construction of XaaS services, provenance and meta-data collection, to mention but few. In many situations the problem may be due to inadequacy of service abstraction. Software Defined Systems (SDSys) is a concept that help abstract the actual hardware at different layers with software components; one classical example of this abstractions are hypervisors. Such abstraction provides an opportunity for system administrators to construct and manage their systems, more easily, through flexible software layers. SDSys is an umbrella for different software defined subsystems including Software Defined Networking (SDN), Software Defined Storage (SDStorage), Software Defined Servers (Virtualization), Software Defined Data Centers (SDDC), Software Defined Security (SDSec) etc. and ultimately Software Defined Clouds (SDCloud). Individual solutions and seamless integration of these different abstractions remains in many respects a challenge. In this paper, the authors introduce Software Defined Cloud (SDCloud), a novel software defined cloud management framework that integrates different software defined cloud components to handle complexities associated with cloud computing systems. The first part of paper presents, for the first time, an extensive state of the art critical review of different components of software defined systems, constructing the proposed SDCloud. The second part of the paper proposes the novel concept of SDCloud, which is implemented and evaluated for its feasibility, flexibility and potential superiority.\",\n \"Title: A hierarchical optimization model for energy data flow in smart grid power systems\\nYear: 2015\\n\\nAbstract: Environmental concerns and high prices of fossil fuels increase the feasibility of using renewable energy sources in smart grid. Smart grid technologies are currently being developed to provide efficient and clean power systems. Communication in smart grid allows different components to collaborate and exchange information. Traditionally, the utility company uses a central management unit to schedule energy generation, distribution, and consumption. Using centralized management in a very large scale smart grid forms a single point of failure and leads to serious scalability issues in terms of information delivery and processing. In this paper, a three-level hierarchical optimization approach is proposed to solve scalability, computational overhead, and minimize daily electricity cost through maximizing the used percentage of renewable energy. At level one, a single home or a group of homes are combined to form an optimized power entity (OPE) that satisfies its load demand from its own renewable energy sources (RESs). At level two, a group of OPEs satisfies energy requirements of all OPEs within the group. At level three, excess in renewable energy from different groups along with the energy from the grid is used to fulfill unsatisfied demands and the remaining energy are sent to storage devices. HighlightsWe examine the three scopes of smart grid systems.We write mathematical models to cover the different aspects in SG.We use linear optimization to solve the equations and yield optimal solutions.Hierarchical method achieves better distribution and cooperative scenario.\",\n \"Title: Using Aspect-Based Sentiment Analysis to Evaluate Arabic News Affect on Readers\\nYear: 2015\\n\\nAbstract: The rapid increase in digital information has raised great challenges especially when it comes to automated content analysis. The adoption of social media as a communication channel for political views demands automated methods for posts' tone analysis, sentiment analysis, and emotional affect. This paper proposes a novel approach of using aspect-based sentiment analysis in evaluating Arabic news posts affect on readers. The approach adopts several phases of text processing, features selection, and text classification. Two widely used classifiers, namely Conditional Random Fields (CRF) and J48, are tested. Experimentation results show that J48 outperforms CRF in aspect terms extraction whereas CRF is slightly better in aspect terms polarity identification.\",\n \"Title: Using Big Data Analytics for Authorship Authentication of Arabic Tweets\\nYear: 2015\\n\\nAbstract: Authorship authentication of a certain text is concerned with correctly attributing it to its author based on its contents. It is a very important problem with deep root in history as many classical texts have doubtful attributions. The information age and ubiquitous use of the Internet is further complicating this problem and adding more dimensions to it. We are interested in the modern version of this problem where the text whose authorship needs authentication is an online text found in online social networks. Specifically, we are interested in the authorship authentication of tweets. This is not the only challenging aspect we consider here. Another challenging aspect is the language of the tweets. Most current works and existing tools support English. We chose to focus on the very important, yet largely understudied, Arabic language. Finally, we add another challenging aspect to the problem at hand by addressing it at a very large scale. We present our effort to employ big data analytics to address the authorship authentication problem of Arabic tweets. We start by crawling a dataset of more than 53K tweets distributed across 20 authors. We then use preprocessing steps to clean the data and prepare it for analysis. The next step is to compute the feature vectors of each tweet. We use the Bag-Of-Words (BOW) approach and compute the weights using the Term Frequency-Inverse Document Frequency (TF-IDF). Then, we feed the dataset to a Naive Bayes classifier implemented on a parallel and distributed computing framework known as Hadoop. To the best of our knowledge, none of the previous works on authorship authentication of Arabic text addressed the unique challenges associated with (1) tweets and (2) large-scale datasets. This makes our work unique on many levels. The results show that the testing accuracy is not very high (61.6%), which is expected in the very challenging setting that we consider.\",\n \"Title: SDDC: A Software Defined Datacenter Experimental Framework\\nYear: 2015\\n\\nAbstract: The rapid growth and the distributed sites of the datacenters increase the complexity of control and management processes. A new paradigm which is called Software Defined Systems (SDSys) comes as a solution to reduce the overhead of datacenters management by abstracting all the control functionalities from the hardware devices and setting it inside a software layer. These functionalities are responsible for organizing and controlling the main blocks of the datacenter; network, storage, compute and security. The Software Defined Datacenter (SDD) integrates the software defined concepts into all of these main blocks. Transferring these concepts into workable systems requires checking several performance aspects and testing its correctness before building real systems. In this paper we introduce a novel experimental framework (SDDC) to provide a novel virtualized testbed environment for SDD systems. This work builds on the Mininet simulator, where its core components, the host, the switch and the controller, are customized to build the proposed experimental simulation framework for SDD. This simulator lets the users develop and test their own SDD solutions, and at the same time gives the researchers an experimentation tool for benchmarking purposes. The developed simulator could also be used as an educational tool to train students and novice researchers.\",\n \"Title: Power Optimization of Large Scale Mobile Cloud Computing Systems\\nYear: 2015\\n\\nAbstract: The main objective of deploying large scale Mobile Cloud Computing (MCC) systems using the cloudlet based infrastructure is to reduce the total power consumption and the network delay while satisfying the Service Level Agreement (SLA). However, this objective is challenging to achieve due to the limited capabilities of the cloudlet system which limit the number of available services on each cloudlet system. This will force different cloudlets to cooperate in order to meet the users demands for different types of services. Moreover, the cooperation will require routing of user requests from the local cloudlet to a remote cloudlet that provide the requested service using a backbone network. This cooperation will increase the total power consumption per request which will include both the user to the local cloudlet power consumption and the power consumption for routing the request to a remote cloudlet. In this paper, we address the problem of optimizing the power consumption for large scale cooperative cloudlets deployments. The evaluation of the proposed model under different realistic scenarios shows that it can be used to accurately optimize power consumption in large scale MCC systems.\",\n \"Title: SDSecurity: A Software Defined Security experimental framework\\nYear: 2015\\n\\nAbstract: The emerging Software Defined Systems (SDSys) is a recent paradigm, which has been introduced to reduce the overhead in the control and management operations of complex computing systems. The main concept behind this technology is around isolating the data plane from the control plane. Traditional security mechanisms are facing more challenges in providing sufficient levels of protection and efficiency. SDSys for security has been proposed to address these challenges. Software Defined Security (SDSec) provides a flexible and centralized security solution by abstracting the security mechanisms from the hardware layer to a software layer. In this paper we present a novel experimental framework to provide a novel virtualized testbed environment for SDSec systems. This work builds on the Mininet simulator, where its core components, the host, switch and the controller, are customized to build the proposed experimental simulation framework for SDSec. To the best of the authors' knowledge, this is the first experimental framework and simulator for SDSec solutions. The developed simulator, will not only support the development and testing of SDSecurity solutions, it will also serve as an experimentation tool for researchers and for benchmarking purposes. The developed simulator could also be used as an educational tool to train students and novice researchers.\",\n \"Title: Virtualization-based Cognitive Radio Networks.\\nYear: 2016\\n\\nAbstract: Integrate virtualization (multilayer hypervisors) into CRNs to improve performance.Allow BSs to delegate some of their management responsibilities to CR users.Reduces control overhead which leads to reduced delay and improved throughput.Show how to handle multi-cell CRNs (requires addressing self-coexistence problem).Significant improvements gained in terms of blocking probability and throughput. The emerging network virtualization technique is considered as a promising technology that enables the deployment of multiple virtual networks over a single physical network. These virtual networks are allowed to share the set of available resources in order to provide different services to their intended users. While several previous studies have focused on wired network virtualization, the field of wireless network virtualization is not well investigated. One of the promising wireless technologies is the Cognitive Radio (CR) technology that aims to handle the spectrum scarcity problem through efficient Dynamic Spectrum Access (DSA). In this paper, we propose to incorporate virtualization concepts into CR Networks (CRNs) to improve their performance. We start by explaining how the concept of multilayer hypervisors can be used within a CRN cell to manage its resources more efficiently by allowing the CR Base Station (BS) to delegate some of its management responsibilities to the CR users. By reducing the CRN users' reliance on the CRN BS, the amount of control messages can be decreased leading to reduced delay and improved throughput. Moreover, the proposed framework allows CRNs to better utilize its resources and support higher traffic loads which is in accordance with the recent technological advances that enable the Customer-Premises Equipments (CPEs) of potential CR users (such as smart phone users) to concurrently run multiple applications each generating its own traffic. We then show how our framework can be extended to handle multi-cell CRNs. Such an extension requires addressing the self-coexistence problem. To this end, we use a traffic load aware channel distribution algorithm. Through simulations, we show that our proposed framework can significantly enhance the CRN performance in terms of blocking probability and network throughput with different primary user level of activities.\",\n \"Title: An Aspect-Based Sentiment Analysis Approach to Evaluating Arabic News Affect on Readers.\\nYear: 2016\\n\\nAbstract: Great challenges arise due to the rapid growth of online data. The widespread use of online social networks (OSN) have enabled the generation of massive amounts of raw data where users post their own material. One interesting example of user generated data is their political views and opinions. The ability to crawl OSN and automatically analyze their political content is of undeniable importance. However, this requires automated methods for posts' tone analysis, sentiment analysis, and emotional affect. The purpose of this paper is to evaluate Arabic news posts affect on readers using a novel approach of aspect-based sentiment analysis (ABSA). There are many tasks typically associated with ABSA such as the extraction and polarity identification of aspect terms and categories. The focus of this work is on the tasks related to aspect terms. A typical approach to address these tasks goes through several stages of text pre-processing, features extraction and classification. This paper follows this approach and makes use of widely used features and classifiers. The features considered include Part of Speech (POS) tagging, Named Entity Recognition (NER), and N-Grams. As for the considered classifiers, they are: Conditional Random Fields (CRF), Decision Tree (J48), Naive Bayes and K-Nearest Neighbor (IBk). A set of experiments are conducted to compare the considered classifiers against each other and against a baseline classifier that is very common for ABSA. The results show that the extracted features allow all of the four considered classifiers to significantly outperform the baseline classifier. They also show that J48 performs the best for the task of aspect terms extraction whereas CRF and Naive Bayes are slightly better in aspect terms polarity identification.\",\n \"Title: Sdmec: Software Defined System For Mobile Edge Computing\\nYear: 2016\\n\\nAbstract: Mobile Edge Computing (MEC) promises a paradigm shift in enabling efficient Mobile Cloud Computing (MCC) services by providing storage and processing capacity within the access range of the mobile devices. In MEC, Mobile Edge (ME) servers are placed at the edge of the mobile networks eliminating the need to offload compute/storage-intensive tasks of the mobile devices to the core of the network (the centralized cloud data center). This reduces the network latency and enhances the quality of service provided for the mobile end users. Different applications can benefit from the large scale deployments of ME servers such as smart grid applications, content delivery networks, content sharing, traffic management, and E-health applications. This promising paradigm comes with its own downside related to the management complexity of large scale deployments that offers hundreds of applications to millions of users. In this paper, we introduce a software defined based framework to enable efficient MCC services through the integration of different software defined system components with the MEC system.\",\n \"Title: Cloud Support Data Management Infrastructure For Upcoming Smart Cities\\nYear: 2016\\n\\nAbstract: This paper presents a novel large scale data management model for future smart cities. The system is exploiting the emerging cloud computing services availability over the globe by processing the collected data at the edge. To handle the large scale requirements in smart city, we introduce a Mobile Edge Computing (MEC) framework in order to increase the reliability of the deployed applications. MEC is a promising framework to reduce the cloud core utilization as well as to provide applications that require low latency to mobile end users. As near as possible to the location of the end user, MEC relates to the mobile network applications and data stream acceleration through caching and/or compressing of relevant data at the edge of the mobile network. Although MEC is not yet deployed in real-life systems, recent studies discussed some technical details and related concepts. This paper studies the definition of MEC and similar concepts with respect to typical application scenarios, like in smart cities, and provides scope and limitations that may be encountered when implementing and deploying MEC. Then, a MEC framework is used to explore the data management infrastructure for upcoming smart cities. (C) 2016 The Authors. Published by Elsevier B.V.\",\n \"Title: GA-Based Resource Augmentation Negotation for Energy-Optimised Mobile Ad-hoc Cloud\\nYear: 2016\\n\\nAbstract: Mobile and cloud computing are two areas which are rapidly expanding in terms of use case and functionality. An ad-hoc mobile cloud can be constructed upon local devices, where devices work together and share resources, aspects such as energy and costs may be minimised. This paper builds on previous work which proposed an end-to-end system whereby user applications may be profiled for their resource consumption locally and then if augmentation is required, they may negotiate with a local mobile ad-hoc cloud for optimum energy and resource utilisation. The contribution of this paper focuses upon a novel and adapted Genetic Algorithm approach to resource negotiation in order to further enable energy-optimised resource augmentation for Mobile Ad-hoc clouds.\",\n \"Title: Cloud support for large scale e-healthcare systems.\\nYear: 2016\\n\\nAbstract: Rapid development of wearable devices and mobile cloud computing technologies has led to new opportunities for large scale e-healthcare systems. In these systems, individuals’ health information are remotely detected using wearable sensors and forwarded through wireless devices to a dedicated computing system for processing and evaluation where a set of specialists namely, hospitals, healthcare agencies and physicians will take care of such health information. Real-time or semi-real time health information are used for online monitoring of patients at home. This in fact enables the doctors and specialists to provide immediate medical treatments. Large scale e-healthcare systems aim at extending the monitoring coverage from individuals to include a crowd of people who live in communities, cities, or even up to a whole country. In this paper, we propose a large scale e-healthcare monitoring system that targets a crowd of individuals in a wide geographical area. The system is efficiently integrating many emerging technologies such as mobile computing, edge computing, wearable sensors, cloud computing, big data techniques, and decision support systems. It can offer remote monitoring of patients anytime and anywhere in a timely manner. The system also features some unique functions that are of great importance for patients’ health as well as for societies, cities, and countries. These unique features are characterized by taking long-term, proactive, and intelligent decisions for expected risks that might arise by detecting abnormal health patterns shown after analyzing huge amounts of patients’ data. Furthermore, it is using a set of supportive information to enhance the decision support system outcome. A rigorous set of evaluation experiments are conducted and presented to validate the efficiency of the proposed model. The obtained results show that the proposed model is scalable by handling a large number of monitored individuals with minimal overhead. Moreover, exploiting the cloud-based system reduces both the resources consumption and the delay overhead for each individual patient.\",\n \"Title: Power Optimization of Large Scale Mobile Cloud System Using Cooperative Cloudlets\\nYear: 2016\\n\\nAbstract: Reducing the total power consumption and network delay are among the most interesting issues facing large scale Mobile Cloud Computing (MCC) systems and their ability to satisfy the Service Level Agreement (SLA). Such systems utilize cloudlet based infrastructure to support off-loading some of user's computationally heavy tasks to the cloudlets. However, the limited capabilities of the cloudlet system (in terms of the ability of serve different request type and the ability to serve users in large geographical regions) represent serious challenges to achieve those objectives. To cover the users demand for different types of services and in wide geographical regions, cloudlets cooperate among each others by passing user requests from one cloudlets to another. By adapting this cooperation, the total power consumption per request will be increased so that it includes the power consumption between the user and the local cloudlet and the power consumption of passing the request to a remote cloudlet. In this paper, we consider two types of cloudlets: local cloudlets and global cloudlets. The global cloudlets are a special kind of local cloudlets but with higher capabilities. The user can connect only to the local cloudlet and sends all its traffics to it. If the local cloudlet cannot serve the desired request, then the request is moved to other local cloudlet. If no local cloudlet can serve the request, then it is moved to a global cloudlet in which it can serve all service types. We optimize the power consumption for large scale cooperative cloudlets and evaluate the proposed model under two realistic scenarios. The result prove that the proposed model can be used to optimize power consumption in large scale MCC systems.\",\n \"Title: Using Dynamic Parallelism to Speed Up Clustering-Based Community Detection in Social Networks\\nYear: 2016\\n\\nAbstract: Social Network Analysis (SNA) has been gaining a lot of attention lately. One of the common steps in SNA is community detection. SNA literature has many interesting algorithms for community detection. One of the popular ones was proposed by Newman and it is mainly revolved around using a clustering algorithm. Three phases are iteratively applied in this algorithm in order to find the \\\"best\\\" community structure. These phases are: spectral mapping, clustering and modularity computation. Despite its effectiveness, this method suffers greatly in terms of running time when dealing with largescale networks. A parallel implementation using GPUs is one of the feasible solutions to address this problem. Moreover, due to the iterative nature of this algorithm, dynamic parallelism lends itself as a very appealing solution. Dynamic parallelism is a novel parallel programming technique that refers to the ability to launch new grids from the GPU. In this work, we present three implementation of the clustering-based community detection algorithm. In addition to the sequential implementation, we present two implementations: a Hybrid CPU-GPU (HCG) one and a Dynamic Parallel (DP) one. We test our parallel implementations on benchmark datasets to show the speed-up of each parallel implementation compared with the sequential one. The results show that the DP implementation achieves good speed-ups reaching up to 4.45X, however, the speed-ups achieved by HCG are almost twice as much.\",\n \"Title: SDCache: Software Defined Data Caching Control for Cloud Services\\nYear: 2016\\n\\nAbstract: The complexity of traditional control approaches typically features slow response times to system and component changes and runs the risk of accessing outdated information. Dynamic environments require fast control approaches to deal with variability in a reliable manner. We adopt the Software Defined System (SDSys) paradigm to handle this issue. The concept behind this fast emerging theme revolves around abstracting all controls as modules within a software control layer. In this paper, we exploit this concept to propose a new system architecture for data caching. We further develop and emulate a testbed environment to evaluate the proposed models by extending the elements of Mininet. Several metrics and multiple scenarios have been considered to capture processing costs and network congestion so as to reflect the merits of this approach. Our experiments validate that the proposed solution based on SDSys renders possible a simple programmable data caching control mechanism.\",\n \"Title: SemEval-2016 Task 5: Aspect Based Sentiment Analysis.\\nYear: 2016\\n\\nAbstract: This paper describes the SemEval 2016 shared task on Aspect Based Sentiment Analysis (ABSA), a continuation of the respective tasks of 2014 and 2015. In its third year, the task provided 19 training and 20 testing datasets for 8 languages and 7 domains, as well as a common evaluation procedure. From these datasets, 25 were for sentence-level and 14 for text-level ABSA; the latter was introduced for the first time as a subtask in SemEval. The task attracted 245 submissions from 29 teams.\",\n \"Title: A Lexicon Based Approach For Classifying Arabic Multi-Labeled Text\\nYear: 2016\\n\\nAbstract: Purpose - Multi-label Text Classification (MTC) is one of the most recent research trends in data mining and information retrieval domains because of many reasons such as the rapid growth of online data and the increasing tendency of internet users to be more comfortable with assigning multiple labels/tags to describe documents, emails, posts, etc. The dimensionality of labels makes MTC more difficult and challenging compared with traditional single-labeled text classification (TC). Because it is a natural extension of TC, several ways are proposed to benefit from the rich literature of TC through what is called problem transformation (PT) methods. Basically, PT methods transform the multi-label data into a single-label one that is suitable for traditional single-label classification algorithms. Another approach is to design novel classification algorithms customized for MTC. Over the past decade, several works have appeared on both approaches focusing mainly on the English language. This work aims to present an elaborate study of MTC of Arabic articles.Design/methodology/approach - This paper presents a novel lexicon-based method for MTC, where the keywords that are most associated with each label are extracted from the training data along with a threshold that can later be used to determine whether each test document belongs to a certain label.Findings - The experiments show that the presented approach outperforms the currently available approaches. Specifically, the results of our experiments show that the best accuracy obtained from existing approaches is only 18 per cent, whereas the accuracy of the presented lexicon-based approach can reach an accuracy level of 31 per cent.Originality/value - Although there exist some tools that can be customized to address the MTC problem for Arabic text, their accuracies are very low when applied to Arabic articles. This paper presents a novel method for MTC. The experiments show that the presented approach outperforms the currently available approaches.\",\n \"Title: Accelerating compute intensive medical imaging segmentation algorithms using hybrid CPU-GPU implementations.\\nYear: 2017\\n\\nAbstract: Medical image processing is one of the most famous image processing fields in this era. This fame comes because of the big revolution in information technology that is used to diagnose many illnesses and saves patients lives. There are many image processing techniques used in this field, such as image reconstructing, image segmentation and many more. Image segmentation is a mandatory step in many image processing based diagnosis procedures. Many segmentation algorithms use clustering approach. In this paper, we focus on Fuzzy C-Means based segmentation algorithms because of the segmentation accuracy they provide. In many cases, these algorithms need long execution times. In this paper, we accelerate the execution time of these algorithms using Graphics Process Unit (GPU) capabilities. We achieve performance enhancement by up to 8.9x without compromising the segmentation accuracy.\",\n \"Title: MedGraph: a graph-based representation and computation to handle large sets of images.\\nYear: 2017\\n\\nAbstract: In order to process and analyze very large volumes of images, efficient representation and structuring techniques are required. Since, current computing machines can provide large memory size, trading off reasonable amount of memory in order to achieve efficient and parallelizable representation of images is preferable. In this paper, we propose a new structure to represent and store images based on in-memory graph concept. Our method of computation provides a faster execution time than the traditional array-based representation. Each pixel of an image is represented as a one node in the graph. In addition, nodes have pointers to other neighboring nodes (pixels). The structure represents an image as a grid of connected linked lists and each grid is connected to other grids. Using our method, an image can be represented in one of three different representations which are: octal linked list, quadratic linked list, and dual linked list representations. We provide experiments and evaluations using the dual linked list representation as it requires less memory space. We apply our methodology for medical images as a proof of the concept to find a region of interest in an image. We have collected and used real medical images to build and process the graph which we call MedGraph. Our experimental results show that the proposed MedGraph technique improves the searching time for finding a region of interest when compared to the traditional representation. It is worth mentioning here that MedGraph is a generic representation strategy that can be applied to any type of images, although this paper uses medical images as a proof of the concept.\",\n \"Title: The Internet of Energy: Smart Sensor Networks and Big Data Management for Smart Grid.\\nYear: 2015\\n\\nAbstract: Smart sensor networks provide numerous opportunities for smart grid applications including power monitoring, demand-side energy management, coordination of distributed storage, and integration of renewable energy generators. Because of their low cost and ease-of-deployment, smart sensor networks are likely to be used on a large scale in future of smart power grids. The result is a huge volume of different variety of data sets. Processing and analyzing these data reveals deeper insights that can help expert to improve the operation of power grid to achieve better performance. The technology to collect massive amounts of data is available today, but managing the data efficiently and extracting the most useful information out of it remains a challenge. This paper discusses and provides recommendations and practices to be used in the future of smart grid and Internet of things. We explore the different applications of smart sensor networks in the domain of smart power grid. Also we discuss the techniques used to manage big data generated by sensors and meters for application processing.\",\n \"Title: Leveraging Software-Defined-Networking for Energy Optimisation in Mobile-Cloud-Computing.\\nYear: 2016\\n\\nAbstract: Both mobile and cloud computing are two areas which are rapidly expanding in terms of use case and functionality. Whilst mobile computing enables a variety of feature rich functionality for users in a non-fixed location, cloud computing is revolu-tionising the way in which computing resources are being provisioned, used and optimised for both service providers and end users. These two fields are being combined in order to provide greater functionality for mobile devices by off loading resources to the cloud. However, the advantages of this only hold true if the device resources are truly optimised. This paper examines the effect of network simulation on understanding methods for optimising device life and suggests a number of ways in which Software-Defined-Networking (SDN) may be leveraged in order to determine the exact nature of optimisations available through this combination.\",\n \"Title: Using the Panama Papers to explore the financial networks of the Middle East\\nYear: 2016\\n\\nAbstract: In what has been described as the WikiLeaks of the financial world, the release of millions of documents (known as the “Panama Papers”) have placed at the center of global media attention the elaborate ways used by some of the elite to hide their financial assets leading to serious allegation of financial corruption. In this work, we explore the information contained in these documents using social network analytics. Due to the large size of the network constructed from the Panama Papers, we limit our attention to a specific region, which is the Middle East. The analysis reveals that while the constructed network enjoys some typical characteristics, there are many interesting observations and properties worth discussing. Specifically, using the extracted network consisting of 37,442 nodes and 79,544 edges, our social network analysis finding show that, perhaps surprisingly, the nodes or the social network are not necessarily directly correlation with perceived financial influence.\",\n \"Title: Software-defined systems support for secure cloud computing based on data classification.\\nYear: 2017\\n\\nAbstract: The newly emerged Software-Defined Systems (SDSs) promised to reduce computing systems management complexity. This can be achieved by separating the control plane from the data plane. On the other hand, cloud computing usage proliferation creates new challenges for managing user data efficiently. Such challenges include data ownership, data access policies, data privacy and integrity, and the availability of storage space. Using encryption-based solution is effective but very costly. Other solutions that used data classification based on the data priority can mitigate the high cost problem for data encryption. However, managing such data classification systems is very complex. In this paper, we are introducing a Software-Defined System-based solution for deploying efficient data classification framework. Our results show the superiority of the proposed model.\",\n \"Title: Paraphrase identification and semantic text similarity analysis in Arabic news tweets using lexical, syntactic, and semantic features.\\nYear: 2017\\n\\nAbstract: The rapid growth in digital information has raised considerable challenges in particular when it comes to automated content analysis. Social media such as twitter share a lot of its users information about their events, opinions, personalities, etc. Paraphrase Identification (PI) is concerned with recognizing whether two texts have the same/similar meaning, whereas the Semantic Text Similarity (STS) is concerned with the degree of that similarity. This research proposes a state-of-the-art approach for paraphrase identification and semantic text similarity analysis in Arabic news tweets. The approach adopts several phases of text processing, features extraction and text classification. Lexical, syntactic, and semantic features are extracted to overcome the weakness and limitations of the current technologies in solving these tasks for the Arabic language. Maximum Entropy (MaxEnt) and Support Vector Regression (SVR) classifiers are trained using these features and are evaluated using a dataset prepared for this research. The experimentation results show that the approach achieves good results in comparison to the baseline results.\",\n \"Title: Feature extraction and selection for Arabic tweets authorship authentication.\\nYear: 2017\\n\\nAbstract: In tweet authentication, we are concerned with correctly attributing a tweet to its true author based on its textual content. The more general problem of authenticating long documents has been studied before and the most common approach relies on the intuitive idea that each author has a unique style that can be captured using stylometric features (SF). Inspired by the success of modern automatic document classification problem, some researchers followed the Bag-Of-Words (BOW) approach for authenticating long documents. In this work, we consider both approaches and their application on authenticating tweets, which represent additional challenges due to the limitation in their sizes. We focus on the Arabic language due to its importance and the scarcity of works related on it. We create different sets of features from both approaches and compare the performance of different classifiers using them. We experiment with various feature selection techniques in order to extract the most discriminating features. To the best of our knowledge, this is the first study of its kind to combine these different sets of features for authorship analysis of Arabic tweets. The results show that combining all the feature sets we compute yields the best results.\",\n \"Title: Framework for Affective News Analysis of Arabic News: 2014 Gaza Attacks Case Study.\\nYear: 2017\\n\\nAbstract: This paper aims at fostering the domain of Arabic affective news analysis through providing: (a) a benchmark annotated Arabic dataset of news for affective news analysis, (b) an aspect-based sentiment analysis (ABSA) approach for evaluating the sentimental affect of Arabic news posts on the reader, and (c) a baseline approach with a common evaluation framework to compare future research results with the baseline ones.\",\n \"Title: Accelerating compute-intensive image segmentation algorithms using GPUs.\\nYear: 2017\\n\\nAbstract: Image segmentation is an important process that facilitates image analysis such as in object detection. Because of its importance, many different algorithms were proposed in the last decade to enhance image segmentation techniques. Clustering algorithms are among the most popular in image segmentation. The proposed algorithms differ in their accuracy and computational efficiency. This paper studies the most famous and new clustering algorithms and provides an analysis on their feasibility for parallel implementation. We have studied four algorithms which are: fuzzy C-mean, type-2 fuzzy C-mean, interval type-2 fuzzy C-mean, and modified interval type-2 fuzzy C-mean. We have implemented them in a sequential (CPU only) and a parallel hybrid CPU---GPU version. Speedup gains of 6$$\\\\\\\\times $$× to 20$$\\\\\\\\times $$× were achieved in the parallel implementation over the sequential implementation. We detail in this paper our discoveries on the portions of the algorithms that are highly parallel so as to help the image processing community, especially if these algorithms are to be used in real-time processing where efficient computation is critical.\",\n \"Title: Cloudlet solution for digital forensic investigation of multiple cases of multiple devices\\nYear: 2017\\n\\nAbstract: Multiple device ownership exponentially increases the volume and variety of data, with detrimental implications to digital forensic investigations. Several authors have proposed data reduction approaches in attempts to enhance the data acquisition and processing phases of the investigation process. Other works have aimed to take advantage of cloud computing's seemingly unlimited resources to leverage investigations. However, such approaches inadvertently affect the credibility of forensic evidence and its admissibility in a court of law, and degrade the efficiency of forensic processes. In this paper, we propose a novel approach which leverages current processes by focusing on augmenting computational and latency capabilities. To achieve this, we motivate a cloudlet-based digital forensic (DF) approach to complement existing cloud computing systems. Based on their proximity to end-devices and remote DF investigation teams, our proposed solution effectively tackles low latency challenges present with the cloud alternative. In addition, configuring the cloudlet solution as the sole custodian of data counters ensures that investigators remain in control of their data, and hence can maintain a comprehensive evidence trail. Finally, have also proposed a cloudlet-based DF resource optimization approach to facilitate upward and downward scaling of resources to cope with a variety of data sizes, multiple devices, and concurrent multiple cases.\",\n \"Title: Collision-free anycast transmission scheduling in UWSNs\\nYear: 2017\\n\\nAbstract: Underwater Wireless Sensor Network (UWSNs) have been proposed to monitor underwater regions such as seas and oceans. A typical example of an UWSN consists of a set of underwater sensor nodes and a set of sink nodes that are deployed at the sea surface. Using acoustic transmission, the sensors send their collected data to at least one of the sinks in what is known as the anycast transmission problem. One of the major problems with underwater nodes is their limited power, which means that transmission collisions are very costly. In this paper, we propose a collision-free anycast transmission scheduling algorithm for UWSNs. The proposed algorithm is a location-based routing algorithm that schedules transmissions between sensor nodes and sink nodes in order to minimize the energy consumption and possible collisions. We perform a set of experiments using the NS2 Aqua-Sim simulator to show that the proposed algorithm can provide up to 26% in energy savings compared with a simple greedy algorithm.\",\n \"Title: A Fog Computing Based System for Selective Forwarding Detection in Mobile Wireless Sensor Networks\\nYear: 2016\\n\\nAbstract: Intrusions detection is one of the major issues that worry organizations in wireless sensor networks (WSNs). Many researchers have dealt with this problem and have proposed many methods for detecting different kinds of intrusions such as selective forwarding, which is a serious attack that may obstruct communications in WSNs. However, as the applications of mobile computing, vehicular networks, and internet of things (IoT) are spreading immensely, selective forwarding detection in Mobile Wireless Sensor Networks (MWSNs) has become a key demand. This paper introduces the problem of selective forwarding in MWSNs, and discusses how available techniques for mitigation this problem in WSNs are not applicable in handling the problem in MWSNs due to sensors mobility. Therefore, the paper proposes a model that provides a global monitoring capability for tracing moving sensors and detecting malicious ones. The model leverages the infrastructure of Fog Computing to achieve this purpose. Furthermore, the paper provides a complete algorithm, a comprehensive discussion and experiments that show the correctness and importance of the proposed approach.\",\n \"Title: On The Road To The Internet Of Biometric Things: A Survey Of Fingerprint Acquisition Technologies And Fingerprint Databases\\nYear: 2016\\n\\nAbstract: Since security vulnerabilities represent one of the grand challenges of the Internet of Things (IoT), researchers have proposed what is known as the Internet of Biometric Things (IoBT), which mixes traditional biometric technologies with context-aware authentication techniques. One of the most famous biometric technologies is electronic fingerprint recognition, which acquire fingerprints using different technologies, some of which are more suitable for IoBT than others. Moreover, different fingerprint databases have been built to study the impact of different factors on the accuracy of different fingerprint recognition algorithms and systems. In this paper, we survey the available fingerprint acquisition technologies and the available fingerprint databases. We also identify the advantages and disadvantages of each technology and database.\",\n \"Title: A Compression-Based Technique To Classify Metamorphic Malware\\nYear: 2016\\n\\nAbstract: Metamorphic malware are able to change their appearance to evade detection by traditional anti-malware software. One of the ways to help mitigate the threat of new metamorphic malware is to determine their origins, i.e., the families to which they belong. This type of metamorphic malware analysis is not typically handled by commercial software. Moreover, existing works rely on analyzing the op-code sequences extracted from the Assembly files of the malware. Very few papers have tried to perform analysis on the binary files of the malware. However, they focused on the simple binary problem of differentiating between a certain malware family and benign files. In this work, we address the more difficult problem of determining the origin of a new metamorphic malware by measuring its similarity to hundreds of variants taken from 13 families of real malware. To address this problem, we use a compression-based classification approach. We experiment with two such approaches: AMDL and BCN. The results showed that AMDL performed no better than a random guess (11% accuracy for AMDL and 18% for the random baseline). On the other hand, BCN performed really well with 67% accuracy.\",\n \"Title: Parallel Implementation Of Fcm-Based Volume Segmentation Of 3d Images\\nYear: 2016\\n\\nAbstract: Parallel programming has many benefits that can help developers and researchers to improve the performance of some algorithms to become more efficient in real life. This is especially true for systems involving medical images. Image segmentation for volume extraction is a famous segmentation process that takes long time to finish execution. In this paper, we consider a new version of the Fuzzy C-Means (FCM) segmentation algorithm (known as IT2FPCM) and provide a parallel implementation of it that is 12X time faster than the sequential implementation. The considered algorithm is based on Interval Type-2 FCM and combines fuzzy and possibilistic ideas in order to obtain higher accuracy. We conduct our experiments using two different machines and the results show that the improvement gains for both machines 11X and12X, respectively.\",\n \"Title: Building A Standard Dataset For Arabic Sentiment Analysis: Identifying Potential Annotation Pitfalls\\nYear: 2016\\n\\nAbstract: Sentiment Analysis (SA) is one of the hottest research fields nowadays. It is concerned with identifying the sentiment conveyed in a piece of text. The current efforts in SA require the existence of standard datasets for training/testing purposes. Such datasets already exist for some languages such as English. Unfortunately, the same cannot be said about other languages such as Arabic. Currently existing Arabic SA datasets are restricted (in their domain, size, dialects covered, etc.) and/or have limited availability. Moreover, the annotation process did not receive the proper attention it deserves. Some of the existing datasets relied on the author's point of view for annotation, while others employed annotators, but did not take into account the personal variations between the annotators and how would that affect their agreement. This study presents our efforts to build a standard Arabic dataset with the above concerns in mind. The constructed dataset is intended for generic use as it contains reviews from different domains written in Modern Standard Arabic (MSA) as well as several dialects. As for the annotation process, it is given high attention by studying the inter-annotator agreements and investigating the potential factors affecting them.\",\n \"Title: Exploiting Gpus To Accelerate Clustering Algorithms\\nYear: 2016\\n\\nAbstract: Big data is a main problem for data mining methods. Fortunately, the rapid advances in affordable high performance computing platforms such as the Graphics Processing Unit (GPU) have helped researchers in reducing the execution time of many algorithms including data mining algorithms. This paper discusses the utilization of the parallelism capabilities of the GPU to improve the the performance of two common clustering algorithms, which are K-Means (KM) and Fuzzy C-Means (FCM) algorithms. Two main parallelism approaches are presented: pure and hybrid. These different versions are tested under different settings including two different GPU-equipped machines (a laptop and a server). The results show excellent improvement gains of the hybrid implementations compared with the pure parallel and sequential ones. On the laptop, the best gains of the hybrid implementations compared with the sequential ones are 11.3X for KM and 10.9X for FCM. As for the server, the best gains are 13.5X for KM and 16.3X for FCM. Moreover, the paper explores the usage of a recent memory management technique for GPU called Unified Memory (UM). The results show a decrease in the performance gain of the hybrid implementations that is equal to 44% for hybrid version of KM and 61% for FCM. On the other hand, the use of UM does introduce a small advantage for the pure parallel implementation.\",\n \"Title: Authorship Attribution Of Arabic Tweets\\nYear: 2016\\n\\nAbstract: In tweet authentication, we are concerned with correctly attributing a tweet to its true author based on its textual content. The more general problem of authenticating long documents has been studied before and the most common approach relies on the intuitive idea that each author has a unique style that can be captured using stylometric features (SF). Inspired by the success of modern automatic document classification problem, some researchers followed the Bag-Of-Words (BOW) approach for authenticating long documents. In this work, we consider both approaches and their application on authenticating tweets, which represent additional challenges due to the limitation in their sizes. We focus on the Arabic language due to its importance and the scarcity of works related on it. We create different sets of features from both approaches and compare the performance of different classifiers using them. To the best of our knowledge, this is the first study of its kind to combine these different sets of features for authorship analysis of Arabic tweets. The results show that combining all the feature sets we compute yields the best results.\",\n \"Title: Improved Hierarchical Classifiers For Multi-Way Sentiment Analysis\\nYear: 2017\\n\\nAbstract: Sentiment Analysis (SA) is field in computational linguistics concerned with determining the sentiment conveyed in a piece of text towards certain entities (such as people, organizations, products, services, events, etc.) using NLP tools. The considered sentiments can be as simple as positive vs. negative. A more fine-grained approach known as Multi-Way Sentiment Analysis (MWSA) is based on ranking systems, such as the 5-star ranking system. In such systems, rankings close to each other can be confusing; thus, some researchers have suggested that using Hierarchical Classifiers (HCs) can yield better results compared with traditional Flat Classifier (FCs). Unlike FCs, which try to address the entire classification problem at once, HCs employ some kind of tree structures where the nodes are simple \\\"core\\\" classifiers customized to address a subset of the classification problem. This study aims to explore extensively the use of HCs to address MWSA by studying six different hierarchies. We compare these hierarchies using four well-known core classifiers (SVM, Decision Tree, Naive Bayes, and KNN) using many measures such as Precision, Recall, F1, Accuracy and Mean Square Error (MSE). The experiments are conducted on the Large Arabic Book Reviews (LABR) dataset, which consists of 63K book reviews in Arabic. The results show that using some of the proposed HCs yield significant improvements in accuracy. Specifically, while the best Accuracy and MSE for FC are 45.77% and 1.61, respective-ly, the best accuracy and MSE for an HC are 72.64% and 0.53, respectively. Also, the results show that, in general, KNN(k-nearest neighbors) benefitted the most from using hierarchical classification.\",\n \"Title: A novel framework for software defined based secure storage systems.\\nYear: 2017\\n\\nAbstract: The Software Defined Systems (SDSys) paradigm has been introduced recently as a solution to reduce the overhead in the control and management operations of complex computing systems and to maintain a high level of security and protection. The main concept behind this technology is around isolating the data plane from the control plane. Building a Software Defined System in a real life environment is considered an expensive solution and may have a lot of risks. Thus, there is a need to simulate such systems before the real-life implementation and deployment. In this paper we present a novel experimental framework as a virtualized testbed environment for software defined based secure storage systems. Its also covers some related issues for large scale data storage and sharing such as deduplication. This work builds on the Mininet simulator, where its core components, the host, switch and the controller, are customized to build the proposed experimental simulation framework. The developed emulator, will not only support the development and testing of SD-based secure storage solutions, it will also serve as an experimentation tool for researchers and for benchmarking purposes. The developed simulator/emulator could also be used as an educational tool to train students and novice researchers.\",\n \"Title: Using Logistic Regression To Improve Virtual Machines Management In Cloud Computing Systems\\nYear: 2017\\n\\nAbstract: Cloud computing (CC) is a computing model that enables its customers to access a shared pool of resources (e.g., storage, network, servers, etc.) through the Internet with a pay per-use pricing model. Different service models are employed in CC including the Platform-as-a-Service (PaaS) model, in which the costumers request a certain set of resources and the cloud service providers provide these resources in the form of a virtual machine (VM) running on one of the thousands of hosting servers or physical machines (PMs) of a data center. Where to \\\"place\\\" VMs, how to \\\"execute\\\" them and whether there is a need to \\\"move/migrate\\\" them are important decisions that affect the overall resource utilization and power consumption in the hosting data center. VM consolidation is a technique of migrating or consolidating VMs to PMs in order to prevent the PMs from being overloaded or reduce the number of active PMs and increase their utilization. Consolidation techniques measure PM utilization to decide whether to consolidate the VMs running on it or migrate some of them to another PM. This study aims to optimize resource utilization and energy efficiency in cloud data centers by proposing a new Logistic Regression based host overloading prediction technique that can be used by any VM consolidation technique. The new algorithm have been evaluated using a dynamic workload using the CloudSim simulator. The simulation results show that the proposed algorithm outperforms all other known host status prediction techniques.\",\n \"Title: Enhanced 3D segmentation techniques for reconstructed 3D medical volumes: Robust and Accurate Intelligent System.\\nYear: 2017\\n\\nAbstract: Medical images play an important role in treating a large number of ailments as they are integral and even indispensable to the diagnosis process of such ailments. Medical images come from different acquisition systems (such as PET, CT, MRI) and, in many situations, automated processing of these images can greatly aid physicians and make their jobs easier. In medical imaging and its applications, 2D segmentation (with its different approaches such as FCM, k-means, MRFM and NN) is the first step which is used to extract ROI. This helps in extracting ROI in each slice (2D medical image) separately regardless of its relation to the next and the previous slices. In this paper, a 3D model of FCM segmentation techniques is proposed to enhance the segmentation process and take in mind the overall 3D-Volume as one testing data.\",\n \"Title: Software-Defined System Support for Enabling Ubiquitous Mobile Edge Computing.\\nYear: 2017\\n\\nAbstract: Mobile Edge Computing (MEC) enables ubiquitous and efficient cloud services to mobile users which facilitates mobile cloud computing (MCC) more easily by providing storage and processing capacity within the access range of the mobile devices. To achieve the goals of MEC, Mobile Edge (ME) servers are co-placed with the mobile network base station (at the edge of the mobile network). This eliminates the need to move computation and storage intensive tasks from a mobile device to a centralized cloud server. This in turn reduces the network communication load and delay and it also enhances the quality of service provided for the mobile end users. Applications, such as smart grid applications, content delivery networks, crowed sourcing, traffic management and E-health, will greatly benefit from such deployment. Unfortunately, a large-scale deployment of ME servers that is needed to make hundreds of applications available to millions of users, comes with great management complexity. The emerging technique of Software-Defined Systems (SDSys) abstracts the management complexities of many systems at different layers by utilizing software components. In this paper, we build a software-defined based framework that enable efficient and ubiquitous MCC services by integrating different SDSys components with the MEC system. The integrated framework is implemented and it is evaluated for its feasibility, flexibility and potential superiority.\",\n \"Title: Delay-aware power optimization model for mobile edge computing systems.\\nYear: 2017\\n\\nAbstract: Reducing the total power consumption and network delay are among the most interesting issues facing large-scale Mobile Cloud Computing (MCC) systems and their ability to satisfy the Service Level Agreement (SLA). Such systems utilize cloud computing infrastructure to support offloading some of user’s computationally heavy tasks to the cloud’s datacenters. However, the delay incurred by such offloading process lead the use of servers (called cloudlets) placed in the physical proximity of the users, creating what is known as Mobile Edge Computing (MEC). The cloudlet-based infrastructure has its challenges such as the limited capabilities of the cloudlet system (in terms of the ability to serve different request types from users in vast geographical regions). To cover the users demand for different types of services and in vast geographical regions, cloudlets cooperate among each other by passing user requests from one cloudlet to another. This cooperation affects both power consumption and delay. In this work, we present a mixed integer linear programming (MILP) optimization model for MEC systems with these two issues in mind. Specifically, we consider two types of cloudlets: local cloudlets and global cloudlets, which have higher capabilities. A user connects to a local cloudlet and sends all of its traffics to it. If the local cloudlet cannot serve the desired request, then the request is moved to another local cloudlet. If no local cloudlet can serve the request, then it is moved to a global cloudlet which can serve all service types. The process of routing requests through the hierarchical network of cloudlets increases power consumption and delay. Our model minimizes power consumption while incurring an acceptable amount of delay. We evaluate it under several realistic scenarios to show that it can indeed be used for power optimization of large-scale MEC systems without violating delay constraints.\",\n \"Title: An extended study of collaboration networks of Levantine biomedical researchers.\\nYear: 2018\\n\\nAbstract: A collaboration (or a co-authorship) network (ColNet) consists of researchers as nodes with the edges connecting two researchers if they have coauthored at least one paper together. In this work, we study the ColNet of biomedical researchers in the Levant region. In addition to performing the standard measures of social network analysis on the constructed network, we propose new metrics (for the stability of research groups) to help us explore the patterns of highly productive researchers. Analyzing the constructed network reveals that local researchers are shifting from working individually or with non-local collaborators to working within relatively large groups of local collaborators. Moreover, we find that highly productive researchers enjoy a certain balance between maintaining stable relationship with their previous collaborators and engaging in new projects with new collaborators.\",\n \"Title: Accelerating 3D medical volume segmentation using GPUs.\\nYear: 2018\\n\\nAbstract: Medical images have an undeniably integral role in the process of diagnosing and treating of a very large number of ailments. Processing such images (for different purposes) can significantly improve the efficiency and effectiveness of this process. The first step in many medical image processing applications is segmentation, which is used to extract the Region of Interest (ROI) from a given image. Due to its effectiveness, a very popular segmentation algorithm is the Fuzzy C-Means (FCM) algorithm. However, FCM takes a long processing time especially for 3D model. This problem can be solved by utilizing parallel programming using Graphics Processing Unit (GPU). In this paper, a hybrid parallel implementation of FCM for extracting volume object from medical DICOM files has been proposed. The proposed algorithm improves the performance 5× compared with the sequential version.\",\n \"Title: Building an Image Database for Studying Image Retargeting\\nYear: 2017\\n\\nAbstract: Modern electronic devices(such as TVs, laptops, and mobile devices) come with a huge variety in screen sizes, resolutions, and aspect ratios. Image retargeting is a technique to retarget or (resize) an image to better utilize the viewing device screen and to protect the main content of the image. Different retargeting techniques have been proposed in the literature that mainly utilizes one of the following main techniques: cropping, seam carving, and scale and stretch. The current problem of image retargeting is that it is very hard to determine the best technique to use on an image to get a target dimension. To apply techniques such as machine learning to determine the best technique to perform image retargeting, an annotated image set is needed to perform the training step. In this work, we build and annotate an image set that is suitable to develop such advance retargeting techniques. We build a dataset that include 500 original images. We apply 4 different retargeting techniques to get two different sizes. The resulting image set contains 4000 images annotated by three people. We also analyze the annotation results to get useful remarks from the annotators perceptual point of view.\",\n \"Title: Fine-Grained Emotion Analysis of Arabic Tweets: A Multi-target Multi-label Approach\\nYear: 2018\\n\\nAbstract: Emotion Analysis (EA) is the task of determining the emotion of a given piece of text. This is an important task with many applications especially when applied to tweets. However, existing work take a rather coarse-grained approach by assuming that each tweet has a single emotion and that this emotion has no intensity. In this work, we take a fine-grained approach by considering cases where a single tweet may have several emotions (multi-label) each with possibly different intensity (multi-target). Moreover, unlike existing work, which consider languages such as English and Chinese, we focus on the Arabic language, a severely under-studied language despite its importance. We build the first dataset (to the best of our knowledge) of Arabic tweets annotated for emotion analysis as a multi-label multi-target problem. Two human experts participated in the annotation process and Cohen's Kappa measure was used to determine their concordance.\",\n \"Title: Studying the controversy in online crowds' interactions.\\nYear: 2018\\n\\nAbstract: Detecting controversy in OSN can be very beneficial in understanding the behavior of the online crowds and the dynamics of their interactions, which leads to better understanding of what moves them and how they can be influenced. In this study, we present a hybrid approach that benefits from structural and content-based approaches to detect controversies in Twitteru0027s trending topics. Unlike many previous works, we do not limit ourselves to a certain domain. However, we do focus on social content written in Arabic. To the best of our knowledge, very limited attention has been invested to undertake this approach in studying controversy in general topics in this region of the world. We collect tweets on different trending topics from different domains. We apply several approaches for controversy detection and compare their outcomes to determine which one is the best.\",\n \"Title: Leveraging fog computing and software defined systems for selective forwarding attacks detection in mobile wireless sensor networks\\nYear: 2018\\n\\nAbstract: AbstractAbstractSelective forwarding is a major problem in wireless sensor networks (WSNs). The nature of sensor environments and the sensitivity of collected measurements in some fields such as war fields increase the need to prevent, detect, or mitigate the problem. One of the most used countermeasures for such problem is the use of voting system based on watchdogs' votes. However, this approach is not applicable in the case of mobile sensors. Mobile WSNs (MWSNs) is growing immensely due to the exposure of applications of mobile computing, vehicular networks, and Internet of things. This exposure has shed light on the security of using mobile sensors and raises the need to set appropriate methods for securing MWSNs against many attacks such as selective forwarding attacks. This paper introduces the problem of selective forwarding in MWSNs and discusses how the voting system used for mitigation; this problem in WSNs is not applicable in handling the problem in MWSNs due to sensors mobility. Therefore, the paper proposes a model that provides a global monitoring capability for tracing moving sensors and detecting malicious ones. The model leverages the infrastructure of fog computing to achieve this purpose. In addition, the paper suggests using software defined systems to be used along with the proposed model, which generalize the model to be used to secure MWSNs against other types of attacks easily and flexibly. The paper provides a complete algorithm, a comprehensive discussion and experiments that show the correctness and importance of the proposed approach.A model for detecting selective forwarding attacks in mobile wireless sensor networks. The model leverages the infrastructure of fog computing and its close proximity from sensor layers to provide a global intrusion detection systems that can trace and monitor mobile wireless sensors in their movement and detect malicious ones. View Figure\",\n \"Title: GPU parallelization of sequence segmentation using information theoretic models.\\nYear: 2018\\n\\nAbstract: Sequence segmentation has gained popularity in bioinformatics and particularly in studying DNA sequences. Information theoretic models have been used in providing accurate solutions in the segmentation of DNA sequences. Existing dynamic programming approaches provide optimal solution to the segmentation problem. However, their quadratic time complexity prohibits their applicability to long sequences. In this paper, we propose a parallel approach to improve the performance of a quasilinear sequence segmentation algorithm. The target segmentation technique is a divide-and-conquer recursive algorithm that is based on information theory principles and models. We present three parallel implementations that aim at reducing the segmentation time. The first implementation uses the multithreading capabilities of CPUs. The second one is a hybrid implementation that utilizes the synergy between the CPU and the multithreading power of GPUs. The third implementation is a variation of the hybrid approach where it utilizes the concept of unified memory between the CPU and the GPU instead of the standard memory copy approach. We demonstrate the applicability of the parallel implementations by testing them on real DNA sequences and randomly generated sequences with different lengths and different number of unique elements. The results show that the hybrid CPU-GPU approach outperforms the sequential implementation with a speedup of up to 5.9X while the CPU parallel implementation provides a poor speedup of only 1.7X.\",\n \"Title: Automatic categorization of Arabic articles based on their political orientation.\\nYear: 2018\\n\\nAbstract: The ability to automatically determine the political orientation of an article can be of great benefit in many areas from academia to security. However, this problem has been largely understudied for Arabic texts in the literature. The contribution of this work lies in two aspects. First, collecting and manually labeling a corpus of articles and comments from different political orientations in the Arab world and making different versions of it. Second, studying the performance of various feature reduction methods and various classifiers on these synthesized datasets. The two most popular feature extraction approaches for such a problem were compared, namely the Traditional Text Categorization (TC) approach and the Stylometric Features approach (SF). Although the experimental results show the superiority of the TC approach over the SF approach, the results also indicate that the latter approach can be significantly improved by adding new and more discriminating features. The experimental results also show that the feature selection techniques reduce the accuracies of the considered classifiers under the TC and SF approaches in general. The only exception is the Partition Membership (PM) technique which has an opposite effect. The highest accuracies are obtained when PM feature selection method is used with the Support Vector Machine (SVM) classifier.\",\n \"Title: Cloud Data Centers Revenue Maximization Using Server Consolidation: Modeling And Evaluation\\nYear: 2018\\n\\nAbstract: Cloud based data centers (DC) offering of virtually unlimited processing and storage capabilities faced a critical downside related to the enormous power consumption requirements. Powering and cooling DC has considerable share of the total DC operational costs in addition to the carbon tax. Server consolidation has been proposed as one viable solution that guarantees efficient power usage while maintaining quality of service requirements for the DC's services. This approach is simply about turning on and off the DC resources based on the incoming workload. In this paper, we are extending our previous work on expansion strategies for cloud services providers by considering the server consolidation technique in order to maximize the DC revenues while maintaining their service quality and reducing their carbon footprints.\",\n \"Title: Collusion attacks mitigation in internet of things: a fog based model.\\nYear: 2018\\n\\nAbstract: Collusion attacks are among the major security concerns nowadays due to the growth exposure in networks and communications. Internet of Things (IoT) environments are an attractive target for such type attacks. This paper discusses the problem of collusion attacks in IoT environments and how mobility of IoT devices increases the difficulty of detecting such types of attacks. It demonstrates how approaches used in detection collusion attacks in WSNs are not applicable in IoT environments. To this end, the paper introduces a model based on Fog Computing infrastructure to keep track of IoT devices and detect collusion attackers. The model uses fog computing layer for realtime monitoring and detection of collusion attacks in IoT environments. Moreover, the model uses a software defined systems layer to add a degree of flexibility for configuring Fog nodes to enable them to detect various types of collusion attacks. The paper provides algorithms, theorems, lemmas and mathematical proofs of the proposed model. Furthermore, the it highlights the possible overhead on fog nodes and network when applying the proposed model, and claims that fog layer infrastructure can provide the required resources for the scalability of the model. The experiments show how the proposed model can keep track of malicious nodes while moving from one cluster to other clusters in IoT environments in contrary to the models used in WSNs. Moreover, the experiments show that the proposed model can bear the computation overhead effectevilly, and reduces the power consumption of aggregator nodes in comparison to the models used in WSNs.\",\n \"Title: Deep learning for Arabic NLP: A survey.\\nYear: 2018\\n\\nAbstract: The recent advances in deep learning (DL) have caused breakthroughs in many fields such as computer vision, natural language processing (NLP) and speech processing. Many DL based approaches have been shown to produce state-of-the-art results on various tasks that are of great importance to online social networks (OSN) and social computing such as sentiment analysis (SA) and pharmacovigilance. NLP tasks are becoming very prominent in OSN and DL is offering researchers and practitioners exciting new directions to address these tasks. In this paper, we provide a survey of the published papers on using DL techniques for NLP. We focus on the Arabic language due to its importance, the scarcity of resources on it and the challenges associated with working on it. We notice that DL has yet to receive the attention it deserves from the Arabic NLP (ANLP) community compared with the attention it is getting for other languages despite the vast adoption of social networks in the Arab world. The majority of the early works on using DL for ANLP focused on OCR-related problems while the more recent ones are more diverse with the increasing interest in applying DL to SA, machine translation, diacritization, etc. This survey should serve as a guide for the young and growing ANLP community in order to help bridge the huge gap between ANLP literature and the much richer and more mature English NLP literature.\",\n \"Title: Resilient service provisioning in cloud based data centers.\\nYear: 2018\\n\\nAbstract: Cloud service providers usually have several geo-distributed data centers to meet these increasing demands. On peak times, these data centers receive unpredictable amount of workload that might go beyond their capacity and service providers need to distribute this workload with the objective of maximizing their revenue. This objective is hard to be guaranteed given the possible incidents that may cause cloud services interruption such as power outages, security breaches, natural disasters, etc. Since cloud providers are facing serious threats to the continuation of their services offering which will cause a serious decrease in their revenues, they must have proper strategies for handling such incidents. While a rich volume of recent research works focused on optimizing cloud data centers operations in order to reduce their operational cost (OPEX), little attention has been paid to such serious incidents and how they would affect service provisioning. In this study, we use mixed integer-linear programming to model the problem of maximizing revenue of a given workload by creating/expanding/renting data centers. The model takes into account a wide range of incidents affecting the users traffic and/or the data centers service capacities. Moreover, it handles many issues such as different types of costs (capital, operational and recovery), power consumption, carbon footprint, SLA related issues such as constraints (e.g., delay) and penalties, etc. To the best of our knowledge, this formation is unique among the body of existing works. Finally, to evaluate the proposed model and the impact of different parameters on its performance, several simulation experiments are conducted.\",\n \"Title: Deep Recurrent neural network vs. support vector machine for aspect-based sentiment analysis of Arabic hotels' reviews.\\nYear: 2018\\n\\nAbstract: •In this research, state-of-the-art approaches based on supervised machine learning are presented to tackle aspect-based sentiment analysis (ABSA) challenges of Arabic Hotels’ reviews.•Two approaches of deep recurrent neural network (RNN) and support vector machine (SVM) are implemented and trained along with lexical, word, syntactic, morphological, and semantic features.•The proposed approaches are evaluated using a reference dataset of Arabic Hotels’ reviews annotated using an ABSA framework presented in the Semantic Evaluation workshop 2016 (SemEval-ABSA16).•Evaluation results show that the SVM approach outperforms the other deep RNN approach in the research investigated tasks (T1: aspect category identification (E#A allocation), T2: aspect opinion target expression (OTE) extraction, and T3: aspect sentiment polarity identification).•Whereas, when focusing on the execution time required for training and testing the models, the deep RNN was faster especially for the second task.\",\n \"Title: Cooperative mobile edge computing system for VANET-based software-defined content delivery.\\nYear: 2018\\n\\nAbstract: Next-generation smart cities and Internet of Things (IoT) are getting more mature in terms of services and infrastructure requirements. Multiple smart vehicle applications are being conceived these days, including road traffic, road safety and infotainment, all of which are suffering from the WAN-latency problem. In this paper, we propose a Vehicular Adhoc Network (VANET)-based Software-Defined Edge Computing infrastructure supporting content delivery services among connected vehicles. The proposed approach leverages network base stations to embed mobile edge computing (MEC) services closer to the vehicles. Our approach can enable the delivery of more competitive services with reduced-latency by utilizing cooperative MEC search strategy for vehicle to infrastructure (V2I) communications as well as utilizing vehicle-level caching for vehicle to vehicle (V2V) communications between peers. The framework prototype has been implemented as a clean extension of the Mininet-WiFi emulator. Preliminary results serve as validation of the proposed framework and point out the potential benefits of the approach in mitigating WAN-latency in VANET.\",\n \"Title: A Novel Recommender System Based on Apriori Algorithm for Requirements Engineering\\nYear: 2018\\n\\nAbstract: Even though requirement gathering is an important step in the construction of any project, it imposes tedious work on the side of the system administrator. Recently, with the advent of data mining methods, many opportunities for improvements on the requirement gathering process have become available, one of the which is the use of recommender systems. Recommendation systems for requirements engineering can be used to provide the right information at the right time to requirements engineers. In this work, we propose a novel efficient recommender system based on Apriori algorithm for user requirements. Such recommender system would improve the accuracy of the obtained requirements and produce more comprehensive results. Furthermore, it would provide interesting information that can be used by various parties. Experimental work showed that our recommender system is efficient in term of execution time and can be widely implemented. In details, the system needed 11-21 seconds to execute when the number of users was 2000-4000.\",\n \"Title: Determine Bipolar Disorder Level from Patient Interviews Using Bi-LSTM and Feature Fusion\\nYear: 2018\\n\\nAbstract: Patients with Bipolar Disorder (BD) suffer from a brain disorder that cause them to change mood without reasons and prevent them from performing ordinary daily tasks. In this work, we classify patients with BD into one of its three levels: remission, hypo-mania, and mania, based solely on audio-visual recordings of structured interviews with these patients by the use of different deep learning techniques coupled with feature fusion and concatenation techniques along with a simple sliding window procedure. The results of our approach are promising and open up the door for many contributions and improvements in the future.\",\n \"Title: A Deep Learning Approach for Amazon EC2 Spot Price Prediction\\nYear: 2018\\n\\nAbstract: Spot Instances (SI) represent one of the ways cloud service providers use to deal with idle resources in off-peak periods, where these resources are being auctioned at low prices to customers with limited budgets in a dynamic manner. However, SI are poorly utilized due to issues like out-of-bid failures and bidding complexity. Thus, effective SI price models are of great importance to customers in order to plan their bidding strategies. This paper proposes a deep learning approach for Amazon EC2 SI price prediction, which is a time-series analysis (TSA) problem. The proposed Long Short-Term Memory (LSTM) approach is compared with a well-known classical (i.e., non deep learning) approach for TSA, which is AutoRegressive Integrated Moving Average (ARIMA), using different accuracy measures commonly used in TSA. The results show the superiority of the LSTM approach compared with the ARIMA approach in many aspects.\",\n \"Title: The Impact of the Number of Eigen-Faces on the Face Recognition Accuracy using Different Distance Measures\\nYear: 2018\\n\\nAbstract: The embedded and real-time systems are the main motivation for this research where the computations are critical to be reduced as much as possible. Face recognition method using eigen-faces yields good accuracy if enough eigen-faces are considered in the classification process. The more eigen-faces used, the more computation power is needed. In this paper, the main goal is to investigate the trade-off between the used number of eigen-faces and the accuracy and the needed computation power of face recognition. Three different distance measures are studied. Namely: Euclidean, block-city, and chess board distances are used. It is concluded that there is some optimum number of eigen-faces that provides the highest recognition rate and acceptable execution time. Moreover, the best number of eigen-faces highly depends on the selected distance measure.\",\n \"Title: Impact of digital fingerprint image quality on the fingerprint recognition accuracy.\\nYear: 2019\\n\\nAbstract: Despite the large body of work on fingerprint identification systems, most of it focused on using specialized devices. Due to the high price of such devices, some researchers directed their attention to digital cameras as an alternative source for fingerprints images. However, such sources introduce new challenges related to image quality. Specifically, most digital cameras compress captured images before storing them leading to potential losses of information. This study comes to address the need to determine the optimum ratio of the fingerprint image compression to ensure the fingerprint identification system’s high accuracy. This study is conducted using a large in-house dataset of raw images. Therefore, all fingerprint information is stored in order to determine the compression ratio accurately. The results proved that the used software functioned perfectly until a compression ratio of (30–40%) of the raw images; any higher ratio would negatively affect the accuracy of the used system.\",\n \"Title: Truthful Multi-Unit Multi-Broker Auction Mechanism with Approximate Social-Welfare in Ad Hoc Cognitive Radio Networks.\\nYear: 2019\\n\\nAbstract: In this paper, we consider the cooperative relay problem in an ad hoc cognitive radio (CR) network consisting of two types of CR nodes: continuous power nodes (CPN) and limited/intermittent power nodes (LPN). All CR nodes are associated with a Secondary Base Station (SBS). To improve energy efficiency and extend the lifetime of LPNs, we allow LPNs to cooperate with neighboring \\\"idle\\\" CPNs to deliver/relay their packets to the SBS while providing incentives to relay CPNs. A crucial challenge in enabling such type of cooperative relay lies in resolving the competition between LPNs on the potential relay opportunities. To regulate such a competitive environment, we develop an auction-based market model consisting of a set of brokers (idle CPNs) and a set of available frequency channels. The brokers iteratively issue short-term dynamic spectrum leases of these channels to competing LPNs. The SBS receives the LPN's bids and computes the output of the auction such that the maximum social-welfare (SW) is achieved by effectively allocating the available channels to the CR nodes. Computing a solution with maximum SW is, in general, NP-hard (even for one broker), but can be solved optimally using a pseudo polynomial time algorithm. Thus, we devise a polynomial time auction mechanism with near-optimal SW. Our mechanism enforces the very valuable property of strategy-proofness to mitigate the market manipulation problem resulting from untruthful behavior of the bidders (enforcing truthful bidders behavior). Simulation results are provided, which verify the effectiveness of our mechanism and demonstrate the significant gain achieved over a reference greedy solution.\",\n \"Title: An enhanced framework for aspect-based sentiment analysis of Hotels' reviews: Arabic reviews case study\\nYear: 2016\\n\\nAbstract: This research proposes a framework for aspect-based sentiment analysis (ABSA) of Hotels' reviews. The proposed framework consists of: (a) a reference human annotated Arabic dataset to support ABSA tasks such as aspect category identification, opinion target expression extraction, and opinion sentiment polarity. The dataset was annotated on both sentence-level and text-level, (b) baseline approach where a Support Vector Machine (SVM) was trained as part of the ABSA tasks, (c) baseline experiments and results, and (d) a common evaluation technique to provide a unified evaluation of future research working on the same dataset and ABSA tasks.\",\n \"Title: A GPU-based breast cancer detection system using Single Pass Fuzzy C-Means clustering algorithm\\nYear: 2016\\n\\nAbstract: Breast cancer is the most lethal type of cancer affecting women in the world. To improve the life quality of women, an early detection of this malignancy is always promising because this cancer is one of the cancers that can be managed and treated easily if it is detected earlier. Mammography is the standard screening method to diagnose breast cancer. At these days, there are a lot of researches that have been done to improve this screening method by benefiting from the enormous growth of technology. Graphics Processing Unit (GPU) is a parallel processor that can divide the complex computations tasks into subtasks and run them concurrently. Many medical imaging modalities offloaded their processing to this processor to help in improving the speed of healthcare systems in order to diagnose the illnesses in real time. This research introduces an acceleration method for the segmentation of the mammography images based on the GPU. In order to provide a better detection for the cancerous tumor, we use a modified version of the most common algorithm for image segmentation, which is the Single Pass Fuzzy C-Means (FCM) algorithm. The approach will be applied to a set of mammogram images to distinguish between malignant and benign cases. Additionally, the system is implemented on GPU parallel processor as well as the traditional CPU in order to compare the performance of both implementations. The performance results are compared according to the execution time and the speedup metrics. The proposed implementation on GPU provides a fine speedup compared to its serial implementation on CPU.\",\n \"Title: Author gender identification from Arabic text.\\nYear: 2017\\n\\nAbstract: The Gender Identification (GI) problem is concerned with determining the gender of a given text’s author. It has a wide range of academic/commercial applications in various fields including literature, security, forensics, electronic markets and trading, etc. To address this problem, researchers have proposed that the writing styles of authors of the same gender share certain aspects, which can be captured by certain stylometric features (SF). Another approach to address this problem focuses mainly on keywords occurrences in each document. This is known as the Bag-Of-Words (BOW) approach. In this work, we study and compare both approaches and focus on the Arabic language for which this problem is still largely understudied despite its importance. To the best of our knowledge, no previous work has considered these approaches for the GI problem of Arabic text. The comparison is carried out under different settings and the results show that the SF approach, which is much cheaper to train, can generate more accurate results under most settings. In fact, the best accuracy levels obtained by the SF and BOW approaches on our in-house dataset are 80.4% and 73.9%, respectively.\",\n \"Title: Improving the performance of the needleman-wunsch algorithm using parallelization and vectorization techniques\\nYear: 2019\\n\\nAbstract: The Needleman-Wunsch (NW) is a dynamic programming algorithm used in the pairwise global alignment of two biological sequences. In this paper, three sets of parallel implementations of the NW algorithm are presented using a mixture of specialized software and hardware solutions: POSIX Threads-based, SIMD Extensions-based and a GPU-based implementations. The three implementations aim at improving the performance of the NW algorithm on large scale input without affecting its accuracy. Our experiments show that the GPU-based implementation is the best implementation as it achieves performance 72.5X faster than the sequential implementation, whereas the best performance achieved by the POSIX threads and the SIMD techniques are 2X and 18.2X faster than the sequential implementation, respectively.\",\n \"Title: Improving fuzzy C-mean-based community detection in social networks using dynamic parallelism\\nYear: 2019\\n\\nAbstract: In Social Network Analysis (SNA), a common algorithm for community detection iteratively applies three phases: spectral mapping, clustering (using either the Fuzzy C-Means or the K-Means algorithms) and modularity computation. Despite its effectiveness, this method is not very efficient. A feasible solution to this problem is to use Graphics Processing Units. Moreover, due to the iterative nature of this algorithm, the emerging dynamic parallelism technology lends itself as a very appealing solution. In this work, we present different novel GPU implementations of both versions of the algorithm: Hybrid CPU-GPU, Dynamic Parallel and Hybrid Nested Parallel. These novel implementations differ in how much they rely on CPU and whether they utilize dynamic parallelism or not. We perform an extensive set of experiments to compare these implementations under different settings. The results show that the Hybrid Nested Parallel implementation provide about two orders of magnitude of speedup.\",\n \"Title: An experimental framework for future smart cities using data fusion and software defined systems: The case of environmental monitoring for smart healthcare\\nYear: 2020\\n\\nAbstract: With the evolution of the concept of Internet into the Internet-of-Things (IoT), in which sensors, devices, and everyday objects (things), are interconnected through the Internet, a large number of interconnected things are expected to produce massive amounts of data, which needs efficient processing in order to provide various services. IoT is of little worth without the ability to process, manage, and control a large amount of heterogeneous data in a timely and efficient manner. The concept of Data Fusion (DF) is based on combining heterogeneous multi-source data in order to produce high quality information, and reduce network traffic. On the other hand, the highly dynamic and heterogeneous nature of IoT requires very flexible and efficient management. The recent software defined systems paradigm is an appealing solution to reduce complexity and overhead in managing complex systems such as IoT by separating the control plane from the data plane. In this work, we present a novel experimental framework for IoT based environmental monitoring applications using concepts from DF and software defined systems (SDS). It is built on top of the software defined networking platform where the core components (the host, switch and the controller) are expanded to support other software defined systems components (such as software defined storage and security) and enable the applications of different DF techniques in IoT environments. We conduct several experiments to highlight features of the presented framework.\",\n \"Title: Using long short-term memory deep neural networks for aspect-based sentiment analysis of Arabic reviews\\nYear: 2019\\n\\nAbstract: This paper proposes a state-of-the-art research for aspect-based sentiment analysis of Arabic Hotels’ reviews using two implementations of long short-term memory (LSTM) neural networks. The first one is (a) a character-level bidirectional LSTM along with conditional random field classifier (Bi-LSTM-CRF) for aspect opinion target expressions (OTEs) extraction, and the second one is (b) an aspect-based LSTM for aspect sentiment polarity classification in which the aspect-OTEs are considered as attention expressions to support the sentiment polarity identification. Proposed approaches are evaluated using a reference dataset of Arabic Hotels’ reviews. Results show that our approaches outperform baseline research on both tasks with an enhancement of 39% for the task of aspect-OTEs extraction and 6% for the aspect sentiment polarity classification task.\",\n \"Title: Using Deep Learning For Automatically Determining Correct Application Of Basic Quranic Recitation Rules\\nYear: 2018\\n\\nAbstract: Quranic Recitation Rules (Ahkam Al-Tajweed) are the articulation rules that should be applied properly when reciting the Holy Quran. Most of the current automatic Quran recitation systems focus on the basic aspects of recitation, which are concerned with the correct pronunciation of words and neglect the other Ahkam Al-Tajweed that are related to the rhythmic and melodious way of recitation such as where to stop and how to \\\"stretch\\\" or \\\"merge\\\" certain letters. The only existing works on the latter parts are limited in terms of the rules they consider or the parts of Quran they cover. This paper comes to fill these gaps. It addresses the problem of identifying the correct usage of Ahkam Al-Tajweed in the entire Quran. Specifically, we focus on eight Ahkam Al-Tajweed faced by early learners of recitation. In the first part of our work, we used traditional audio processing techniques for feature extraction (such as Linear predictive Code (LPC), Mel-Frequency Cepstral Coefficient (MFCC), Wavelet Packet Decomposition (WPD) and Markov Model based Spectral Peak Location (HMM-SPL)) and classification (such as k-Nearest Neighbors (KNN), Support Vector Machines (SVM), and Random Forest (RF)) on an in-house dataset of thousands of audio recordings covering all occurrences of the rules under consideration in the entire Holy Quran by different reciters of both genders. In this part, we show how to improve the classification accuracy to surpass 97.7% by incorporating deep learning techniques. Specifically, this result is obtained by incorporating most traditional features with ones extracted using Convolutional Deep Belief Network (CDBN) while the classification is performed using SVM.\",\n \"Title: Energy efficient dynamic resource management in cloud computing based on logistic regression model and median absolute deviation\\nYear: 2018\\n\\nAbstract: •Optimize resource utilization and energy efficiency in cloud data centers.•Proposing host overloading detection algorithm based on logistic regression model and median absolute derivation.•Promote the VM consolidation technique.\",\n \"Title: Parallel implementation for 3D medical volume fuzzy segmentation\\nYear: 2020\\n\\nAbstract: •This research introduces the hybrid optimization of both CPU and GPU for image processing.•A novel 3D medical volume segmentation is proposed.•Modified FCM versions are used to increase the accuracy of volume segmentation.•Volumetric neighborhood has been considered to validate the actual 3D FCM.\",\n \"Title: Collaboration networks of arab biomedical researchers\\nYear: 2019\\n\\nAbstract: Social networks (SN) consist of a set of actors and connections between them. A collaboration network (ColNet) is a special type of SN, in which the actors represent researchers and the link between them indicate that they have co-authored at least one paper. ColNet analysis reveals how researchers interact and behave. A wide range of applications can be based on such studies. The current works on ColNet usually focus on a specific domain/discipline, country/geographical region or time interval. In our study, we focus on one of the understudied regions (the Arab world), and present a novel study on the ColNet of researchers in this region. The domain of interest in our study is biomedicine. We construct, analyze, and study ColNet of biomedical researchers in the Arab world. We divide the region of interest (the Arab world) into four geographical regions and look into the evolution of ColNet of each region separately over time. Our analysis reveals that there is an increase in the number of both authors and publications over time, and that authors tend to work in increasingly larger groups rather than working individually, which is consistent with what is assumed about the nature of research in this field. Our analysis also reveals that a researcher’s productivity is correlated with the amount of change in his/her circle of collaborators over time. For example, researchers working in stable or fixed groups and researchers who have completely different research group every few years are not necessarily the most productive ones.\",\n \"Title: A comprehensive survey of arabic sentiment analysis\\nYear: 2019\\n\\nAbstract: Sentiment analysis (SA) is a continuing field of research that lies at the intersection of many fields such as data mining, natural language processing and machine learning. It is concerned with the automatic extraction of opinions conveyed in a certain text. Due to its vast applications, many studies have been conducted in the area of SA especially on English texts, while other languages such as Arabic received less attention. This survey presents a comprehensive overview of the works done so far on Arabic SA (ASA). The survey groups published papers based on the SA-related problems they address and tries to identify the gaps in the current literature laying foundation for future studies in this field.\",\n \"Title: Toward a new approach for sorting extremely large data files in the big data era\\nYear: 2019\\n\\nAbstract: The extensive amount of data and contents generated today will require a paradigm shift in processing and management techniques for these data. One of the important data processing operations is the data sorting. Using multiple passes in external merge sort has a great influence on speeding up the sorting of extremely large data files. Since in large files, the swapping time is dominant in many applications, algorithms that minimize the swapping operations are normally superior to those which only focus on CPU time optimizations. In sorting extremely large files, external algorithms, such as the merge sort, are normally used. It is shown that using multiple passes over the data set, as proposed in our algorithm, has resulted in a great improvement in the number of swaps, thus, reducing the overall sorting time. Moreover, the proposed technique is suitable to be used with the emerging parallelization techniques such as GPUs. The reported results show the superiority of the proposed technique for “CPU only” and hybrid CPU–GPU implementations.\",\n \"Title: Plant Classification in the Wild: A Transfer Learning Approach\\nYear: 2018\\n\\nAbstract: Datasets specialized in wildlife usually contain imbalanced classes of natural wild images such as, for instance, plant images, which are acquired from the surrounding environment with natural scene background. Deep neural networks have proven their efficiency in classifying such datasets. However, such an approach requires a workaround to approximately balance the classes in order to prevent the occurrence of overfitting during the training phase of the neural network. Many approaches exist to overcome this problem includes over-sampling, undersampling, generating synthetic samples, data augmentation, etc. The iNaturalist species classification and detection dataset represents a good example of vastly imbalanced datasets. It contains 13 superclasses. This work focuses on the Plantae superclass and builds a Convolutional Neural Network to distinguish a subset of the subclasses of Plantae. Our model benefits from cutting-edge techniques such as transfer learning and data augmentation to obtain a reasonably high level of accuracy (78.76%).\",\n \"Title: Enhancing Aspect-Based Sentiment Analysis of Arabic Hotels' reviews using morphological, syntactic and semantic features.\\nYear: 2019\\n\\nAbstract: This research presents an enhanced approach for Aspect-Based Sentiment Analysis (ABSA) of Hotels’ Arabic reviews using supervised machine learning. The proposed approach employs a state-of-the-art research of training a set of classifiers with morphological, syntactic, and semantic features to address the research tasks namely: (a) T1:Aspect Category Identification, (b) T2:Opinion Target Expression (OTE) Extraction, and (c) T3: Sentiment Polarity Identification. Employed classifiers include Naïve Bayes, Bayes Networks, Decision Tree, K-Nearest Neighbor (K-NN), and Support-Vector Machine (SVM).The approach was evaluated using a reference dataset based on Semantic Evaluation 2016 workshop (SemEval-2016: Task-5). Results show that the supervised learning approach outperforms related work evaluated using the same dataset. More precisely, evaluation results show that all classifiers in the proposed approach outperform the baseline approach, and the overall enhancement for the best performing classifier (SVM) is around 53% for T1, around 59% for T2, and around 19% in T3.\",\n \"Title: On the Use of Emojis to Train Emotion Classifiers.\\nYear: 2019\\n\\nAbstract: Nowadays, the automatic detection of emotions is employed by many applications in different fields like security informatics, e-learning, humor detection, targeted advertising, etc. Many of these applications focus on social media and treat this problem as a classification problem, which requires preparing training data. The typical method for annotating the training data by human experts is considered time consuming, labor intensive and sometimes prone to error. Moreover, such an approach is not easily extensible to new domains/languages since such extensions require annotating new training data. In this study, we propose a distant supervised learning approach where the training sentences are automatically annotated based on the emojis they have. Such training data would be very cheap to produce compared with the manually created training data, thus, much larger training data can be easily obtained. On the other hand, this training data would naturally have lower quality as it may contain some errors in the annotation. Nonetheless, we experimentally show that training classifiers on cheap, large and possibly erroneous data annotated using this approach leads to more accurate results compared with training the same classifiers on the more expensive, much smaller and error-free manually annotated training data. Our experiments are conducted on an in-house dataset of emotional Arabic tweets and the classifiers we consider are: Support Vector Machine (SVM), Multinomial Naive Bayes (MNB) and Random Forest (RF). In addition to experimenting with single classifiers, we also consider using an ensemble of classifiers. The results show that using an automatically annotated training data (that is only one order of magnitude larger than the manually annotated one) gives better results in almost all settings considered.\",\n \"Title: A detailed comparative study of open source deep learning frameworks.\\nYear: 2019\\n\\nAbstract: Deep Learning (DL) is one of the hottest trends in machine learning as DL approaches produced results superior to the state-of-the-art in problematic areas such as image processing and natural language processing (NLP). To foster the growth of DL, several open source frameworks appeared providing implementations of the most common DL algorithms. These frameworks vary in the algorithms they support and in the quality of their implementations. The purpose of this work is to provide a qualitative and quantitative comparison among three of the most popular and most comprehensive DL frameworks (namely Googleu0027s TensorFlow, University of Montrealu0027s Theano and Microsoftu0027s CNTK). The ultimate goal of this work is to help end users make an informed decision about the best DL framework that suits their needs and resources. To ensure that our study is as comprehensive as possible, we conduct several experiments using multiple benchmark datasets from different fields (image processing, NLP, etc.) and measure the performance of the frameworksu0027 implementations of different DL algorithms. For most of our experiments, we find out that CNTKu0027s implementations are superior to the other ones under consideration.\",\n \"Title: Arabic Text Diacritization Using Deep Neural Networks\\nYear: 2019\\n\\nAbstract: Diacritization of Arabic text is both an interesting and a challenging problem at the same time with various applications ranging from speech synthesis to helping students learning the Arabic language. Like many other tasks or problems in Arabic language processing, the weak efforts invested into this problem and the lack of available (open-source) resources hinder the progress towards solving this problem. This work provides a critical review for the currently existing systems, measures and resources for Arabic text diacritization. Moreover, it introduces a much-needed free-for-all cleaned dataset that can be easily used to benchmark any work on Arabic diacritization. Extracted from the Tashkeela Corpus, the dataset consists of 55K lines containing about 2.3M words. After constructing the dataset, existing tools and systems are tested on it. The results of the experiments show that the neural Shakkala system significantly outperforms traditional rule-based approaches and other closed-source tools with a Diacritic Error Rate (DER) of 2.88% compared with 13.78%, which the best DER for the non-neural approach (obtained by the Mishkal tool).\",\n \"Title: Recognizing Emotion from Speech Based on Age and Gender Using Hierarchical Models\\nYear: 2019\\n\\nAbstract: Age and gender are two factors that affect the physiologic and acoustic features of human voice. In fact, most of the speech emotion recognition applications use these voice features as a foundation to complete the classification task. Significant improvements have been made for voice emotion recognition; and several studies have addressed the age and gender identification from speech topics. We studied the effect of age and gender on the emotion recognition applications. In our work, we built hierarchical classification models to investigate the importance of identifying the age and gender before identifying the emotional label. We compared the performance of four different models and presented the relationship between the age \\\\ gender and the emotion recognition accuracy. Our results showed that using a separated emotion model for each of gender and age category gives a higher accuracy compared with using one classifier for all the data.\",\n \"Title: Sarcasm Detection and Quantification in Arabic Tweets\\nYear: 2021\\n\\nAbstract: The role of predicting sarcasm in the text is known as automatic sarcasm detection. Given the prevalence and challenges of sarcasm in sentiment-bearing text, this is a critical phase in most sentiment analysis tasks. With the increasing popularity and usage of different social media platforms among users around the world, people are using sarcasm more and more in their day-to-day conversations, social media posts and tweets, and it is considered as a way for people to express their sentiment about some certain topics or issues. As a result of the increasing popularity, researchers started to focus their research endeavors on detecting sarcasm from a text in different languages especially the English language. However, the task of sarcasm detection is a challenging task due to the nature of sarcastic texts; which can be relative and significantly differs from one person to another depending on the topic, region, the user's mentality and other factors. In addition to the aforementioned challenges, sarcasm detection in the Arabic language has its own challenges due to the complexity of the Arabic language, such as being morphologically rich, with many dialects that significantly vary between each other, while also being lowly resourced when compared to English. In recent years, only few research attempts started tackling the task of sarcasm detection in Arabic, including creating and collecting corpora, organizing workshops and establishing baseline models. This paper intends to create a new humanly annotated Arabic corpus for sarcasm detection collected from tweets, and implementing a new approach for sarcasm detection and quantification in Arabic tweets. The annotation technique followed in this paper is unique in sarcasm detection and the proposed approach tackles the problem as a regression problem instead of classification; i.e., the model attempts to predict the level of sarcasm instead of binary classification (sarcastic vs. non-sarcastic) for the purpose of tackling the complex and user-dependent nature of the sarcastic text. The humanly annotated dataset will be available to the public for any usage.\",\n \"Title: Visual Question Answering In The Medical Domain Based On Deep Learning Approaches: A Comprehensive Study\\nYear: 2021\\n\\nAbstract: Visual Question Answering (VQA) in the medical domain has attracted more attention from research communities in the last few years due to its various applications. This paper investigates several deep learning approaches in building a medical VQA system based on ImageCLEF's VQA-Med dataset, which consists of about 4K images with about 15K question-answer pairs. Due to the wide variety of the im-ages and questions included in this dataset, the proposed model is a hierarchical one consisting of many sub-models, each tailored to handle certain questions. For that, a special model is built to classify the questions into four categories, where each category is handled by a separate sub-model. At their core, all of these models consist of pre-trained Convolution Neural Networks (CNN). In order to get the best results, extensive experiments are performed and various techniques are employed including Data Aug-mentation (DA), Multi-Task Learning (MTL), Global Average Pooling (GAP), Ensembling, and Sequence to Sequence (Seq2Seq) models. Overall, the final model achieves 60.8 accuracy and 63.4 BLEU score, which are competitive with the state-of-the-art results despite using less demanding and simpler sub-models. (c) 2021 Elsevier B.V. All rights reserved.\",\n \"Title: Neural Arabic Text Diacritization: State-of-the-Art Results and a Novel Approach for Arabic NLP Downstream Tasks\\nYear: 2022\\n\\nAbstract: AbstractIn this work, we present several deep learning models for the automatic diacritization of Arabic text. Our models are built using two main approaches, viz. Feed-Forward Neural Network (FFNN) and Recurrent Neural Network (RNN), with several enhancements such as 100-hot encoding, embeddings, Conditional Random Field (CRF), and Block-Normalized Gradient (BNG). The models are tested on the only freely available benchmark dataset and the results show that our models are either better or on par with other models even those requiring human-crafted language-dependent post-processing steps, unlike ours. Moreover, we show how diacritics in Arabic can be used to enhance the models of downstream NLP tasks such as Machine Translation (MT) and Sentiment Analysis (SA) by proposing novel Translation over Diacritization (ToD) and Sentiment over Diacritization (SoD) approaches.\",\n \"Title: Deep Learning-Based Disk Herniation Computer Aided Diagnosis System From MRI Axial Scans\\nYear: 2022\\n\\nAbstract: Computer-aided diagnosis (CAD) systems have been the focus of many researchers in both computer and medical fields. In this paper, we build two convolutional neural network (CNN) based CAD systems for diagnosing lumbar disk herniation from Magnetic Resonance Imaging (MRI) axial scans. The first one is a disk herniation detection CAD system which is a binary CAD system that determines whether the case image contains disk herniation or not. The second system is a disk herniation type classification CAD system which can determine the type of the disk herniation in the image if one exists. To train and test the proposed systems, an image set is built and annotated with the help of a radiologist. In order to get rid of the \\\"noisy\\\" parts of the input images and reduce their complexity, we experiment with different ROI extraction methods. The image set is also preprocessed and enlarged using augmentation techniques to make it suitable to be used with CNN. There are many novel aspects of this work. First, the problems of disk herniation detection and recognition from axial scans are not well-studied in the literature. Second, we use deep learning techniques which produces ground-breaking results in many image processing tasks, but are yet to reach their full potential with medical image processing in general. Finally, we explore the use of many innovative techniques such as data augmentation and transfer learning, which greatly improves the accuracy of our models. The results of our systems are impressive with a 95.65% accuracy for the detection problem and a 91.38% accuracy for the recognition problem.\",\n \"Title: An Adaptive Black-box Defense against Trojan Attacks on Text Data\\nYear: 2021\\n\\nAbstract: Trojan backdoor is a poisoning attack against Neural Network (NN) classifiers in which adversaries try to exploit the (highly desirable) model reuse property to implant Trojans into model parameters for backdoor breaches through a poisoned training process. Most of the proposed defenses against Trojan attacks assume a white-box setup, in which the defender either has access to the inner state of N...\",\n \"Title: Plant classification in the wild: Energy evaluation for deep learning models\\nYear: 2022\\n\\nAbstract: Having a system that can take an image of a natural scene and accurately classify the plants in it is of undeniable importance. However, the complexities of dealing with natural scene images and the vast diversity of plants in the wild make designing such a classifier a challenging task. Deep Learning (DL) lends itself as viable solution to tackle such complex problem. However, advanced in DL architectures and software (including DL frameworks) come with a high cost in terms of energy consumption especially when employing Graphics Processing Units (GPU). As data expands rapidly, the need to create energy-aware models increases in order to reduce energy consumption and move towards “Greener AI”. Since the problem of designing energy-aware architectures for plant classification has not been studied significantly in the literature, our work comes to start bridging this gap by focusing not only on the models’ performance, but also on their energy usage on both CPU and GPU platforms. We consider different state-of-the-art Convolutional Neural Networks (CNN) architectures and train them on two famous challenging plants datasets: iNaturalist and Herbarium. Our experiments are meant to highlight the trade-off between accuracy and energy consumption. For examples, the results show that while GPU-bound models can be about 40% faster in terms of training time than simple models running on CPU, the latter’s energy consumption is only two thirds of the former. We hope that such findings will encourage the community to reduce its reliance on accuracy measures to compare different architectures and start taking other factors into account such as power consumption, simplicity, etc.\",\n \"Title: Social Network Analysis of the Panama Papers Concentrating on the MENA Region\\nYear: 2022\\n\\nAbstract: The release of millions of financial documents, which has been known as the 'WikiLeaks' of the financial world (a.k.a. 'Panama Papers'), has dragged global attention in how highly structured means applied by some of the elite to conceal their financial assets. Consequently, significant financial corruption allegations were raised. We concentrate on a somewhat overlooked region, the Middle East and North Africa (MENA) region. This study aims to use social network analytics to study the information contained in these documents. We are checking the major players in the MENA's trends and patterns to determine if it matches the known economic powers. The analysis reveals that while the constructed network enjoys some typical characteristics, many interesting observations and properties are worth discussing. Specifically, using the extracted network consisting of 62 987 nodes and 84 692 edges, our social network analysis finding shows that, perhaps surprisingly, the nodes or the social network are not necessarily directly correlated with perceived economic influence.\",\n \"Title: Using K-means Clustering Ensemble to Improve the Performance in Recommender Systems\\nYear: 2022\\n\\nAbstract: Collaborative filtering methods are often utilized in the industry of recommender systems. They work by identifying users with similar tastes and recommending items for each active user. Besides, clustering techniques are extensively utilized to create systems based on collaborative filtering recommendation in the context of big data. Nevertheless, the cluster ensemble has emerged in last years as a powerful technique that can replace single clustering algorithms in enhancing the performance of recommendation and prediction. This paper presents a k-means clustering ensemble-based method to improve the performance in recommender systems. The proposed system incorporates the Cosine Similarity and the Pearson Correlation Coefficient as similarity metrics to form clusters. Moreover, it uses the HyperGraph Partitioning Algorithm (HGPA) to combine the results of the k-means clustering technique. The recommendation algorithm constructs the recommendations based on the clusters obtained earlier by the HGPA ensemble clustering. To this end, it finds the nearest cluster for each active user and selects its top N items. Finally, it recommends these top items to the user‘s favorite list. The experiments on two well-known datasets demonstrate that cluster ensembles by HGPA outperform the baseline methods.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":113,"string":"113"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13592,"string":"13,592"}}},{"rowIdx":13593,"cells":{"question_id":{"kind":"string","value":"Matthieu Crussi re"},"question":{"kind":"string","value":"Generate an abstract for the title \"Quasi-Optimal Tone Reservation PAPR Reduction Algorithm for Next Generation Broadcasting Systems: A Performance/Complexity/Latency Tradeoff With Testbed Implementation.\" using the following items: High PAPR, multicarrier communication, DVB-T2, power efficiency, tone reservation."},"answer":{"kind":"string","value":"High peak-to-average power ratio (PAPR) of the transmitted signal is a serious issue in multicarrier communication systems, such as second generation terrestrial digital video broadcasting (DVB-T2) systems. These large fluctuations prevent feeding the high power amplifier at an operating point near its non-linear saturation region thereby lowering its power efficiency. In recent years, tone reserv..."},"correct_answer":{"kind":"string","value":"High peak-to-average power ratio (PAPR) of the transmitted signal is a serious issue in multicarrier communication systems, such as second generation terrestrial digital video broadcasting (DVB-T2) systems. These large fluctuations prevent feeding the high power amplifier at an operating point near its non-linear saturation region thereby lowering its power efficiency. In recent years, tone reserv..."},"documents":{"kind":"list like","value":["Title: Enhanced mobile digital video broadcasting with distributed space-time coding\nYear: 2012\n\nAbstract: This paper investigates the distributed space-time (ST) coding proposals for the future Digital Video Broadcasting-Next Generation Handheld (DVB-NGH) standard. We first theoretically show that the distributed MIMO scheme is the best broadcasting scenario in terms of channel capacity. Consequently we evaluate the performance of several ST coding proposals for DVB-NGH with practical system specifications and channel conditions. Simulation results demonstrate that the 3D code is the best ST coding solution for broadcasting in the distributed MIMO scenario.","Title: A Combined Time and Frequency Algorithm for Improved Channel Estimation in TDS-OFDM\nYear: 2010\n\nAbstract: In contrast to the classical cyclic prefix (CP)-OFDM, the time domain synchronous (TDS)-OFDM employs a known pseudo noise (PN) sequence as guard interval (GI). Conventional channel estimation methods for TDS-OFDM are only based on the PN sequence and consequently suffer from intersymbol interference. This paper proposes a novel two-stage channel estimation method which combines the estimation results from the PN sequence and, most importantly, the estimation results obtained from the OFDM data symbols. A simple feedback loop that excludes the channel decoder is employed for the OFDM data based estimation. The MMSE criteria is used to obtain the best combination results and an iterative process is proposed to progressively refine the estimation. Both MSE and BER simulations are carried out to prove the effectiveness of the proposed algorithm in the DTMB system which is based on TDS-OFDM signalling.","Title: Improved channel estimation methods based on PN sequence for TDS-OFDM\nYear: 2012\n\nAbstract: An accurate channel estimation is crucial for the novel time domain synchronous orthogonal frequency-division multiplexing (TDS-OFDM) scheme in which pseudo noise (PN) sequences serve as both guard intervals (GI) for OFDM data symbols and training sequences for synchronization/channel estimation. This paper studies the channel estimation method based on the cross-correlation of PN sequences. A theoretical analysis of this estimator is conducted and several improved estimators are then proposed to reduce the estimation error floor encountered by the PN-correlation-based estimator. It is shown through mathematical derivations and simulations that the new estimators approach or even achieve the Cramér-Rao bound.","Title: Special Issue on MC-SS Loading algorithms for adaptive SS-MC-MA systems over wireline channels: comparison with DMT\nYear: 2006\n\nAbstract: In this paper, we propose to combine adaptive loading principles with the spread-spectrum multicarrier multiple access (SS-MC-MA) scheme. Such an approach has particular interests in the context of powerline communications (PLC), where the transmitter has not only to exploit robust transmission techniques, but has also to adapt the waveform to the channel response. We introduce finite-granularity loading algorithms that dynamically handle the configuration of the system under power spectral density constraints. The presented algorithms assign subcarriers, spreading codes, bits and energy to each user in order to maximise either the data rate or the noise margin at a given target symbol error rate. These algorithms can actually be viewed as a widening of the classical waterfilling approach in the case of an hybrid spread-spectrum multicarrier system. Simulation results of the new scheme are presented for different measured PLC channels and are compared with those of the classical discrete multitone modulation (DMT) approach. It is shown that the adaptive SS-MC-MA scheme performs significantly better than DMT, due to its natural energy Gathering capability. Adaptive SS-MC-MA then leads to a more efficient bits and energies distribution and constitutes a simple solution to reduce the quantification loss induced by the use of finite order modulations. Copyright (c) 2006 AEIT.","Title: Performance of time reversal precoding technique for MISO-OFDM systems.\nYear: 2013\n\nAbstract: Time reversal (TR) is considered as a promising technique for green and multi-user communications thanks to its time and space focusing properties. TR can be viewed as a precoding scheme which can be combined with orthogonal frequency division multiplexing (OFDM) and easily carried out in a multiple transmit antenna context. This paper analyzes the performance of TR for a multiple-input single-output (MISO) OFDM system and provides a comparison with maximum ratio transmission (MRT) and equal gain transmission (EGT) precoding techniques. The analytical performance of the three precoding techniques is derived by computing the capacity and the bit error rate (BER) as a function of the transmit signal-to-noise ratio (SNR). First, the capacity analysis highlights the ability of the TR system to provide higher bit rates than the MRT system at low SNRs, while the capacity of the MRT system is the highest at high SNRs. From the obtained BER analytical expressions, the diversity exploitation of each system is discussed. In particular, it is shown that the TR system only exploits half the available diversity, while the systems using EGT or MRT exploit the full diversity. Hence, contrary to what is expected from the theoretical capacity analysis, the TR system is shown to underperform the other precoding schemes in terms of BER. To overcome such a drawback, the combination of TR with classical adaptive modulation techniques is studied, allowing the achievable throughput to be increased without destroying the focusing properties of TR. It is then observed that TR takes advantage of adaptive modulations and outperforms the other schemes at low SNRs. In this study, analytical results and closed-form expressions of capacity and BER performance are provided and confirmed through Monte Carlo simulations.","Title: Adaptive Linear Precoded Dmt As An Efficient Resource Allocation Scheme For Power-Line Communications\nYear: 2006\n\nAbstract: In this paper, we propose to apply adaptive loading principles to linear precoded digital multitone (LP-DMT). This new approach can especially be exploited in the powerline context, since it requires the knowledge of the channel at the transmitter. We first show that maximal waveform capacity is achieved when orthogonal matrices are used to linearly precode the multitone signal. A practical loading algorithm based on tone clustering is then developed to handle the configuration of the system. This algorithm assigns tones, size of precoding matrices, bits and energies in order to maximize the total throughput of the system. The optimization procedure is led under power spectral density (PSD) limitations and finite order modulations constraint. Through simulations over power line channels, it is shown that the new adaptive LP-DMT system takes advantage of the carrier merging effect offered by the precoding function and then outperforms the non-precoded DMT system.","Title: Achieving Low-Complexity Maximum-Likelihood Detection for the 3D MIMO Code.\nYear: 2014\n\nAbstract: The 3D Multiple-input multiple-output (MIMO) code is a robust and efficient space-time block code (STBC) for the distributed MIMO broadcasting but suffers from high maximum-likelihood (ML) decoding complexity. In this paper, we first analyze some properties of the 3D MIMO code to show that the 3D MIMO code is fast decodable. It is proven that the ML decoding performance can be achieved with a complexity of O(M4.5) instead of O(M8) in quasi-static channel with M-ary square QAM modulations. Consequently, we propose a simplified ML decoder exploiting the unique properties of the 3D MIMO code. Simulation results show that the proposed simplified ML decoder can achieve much lower processing time latency compared to the classical sphere decoder with Schnorr-Euchner enumeration.","Title: Efficient Mimo-Ofdm Schemes For Future Terrestrial Digital Tv With Unequal Received Powers\nYear: 2008\n\nAbstract: This article investigates the effect of equal and unequal received powers on the performances of different MIMO-OFOM schemes for terrestrial digital TV More precisely, we focus on three types of non-orthogonal schemes: the BLAST scheme, the Linear Dispersion (LD) code and the Golden code, and we compare their performances to that of Alamouti scheme. Using two receiving antennas, we show that for moderate attenuation on the second antenna and high spectral efficiency, Golden code outperforms other schemes. However, Alamouti scheme presents the best performance for low spectral efficiency and equal received powers or when one antenna is dramatically damaged When three antennas are used, we show that Golden code offers the highest robustness to power unbalance at the receiving side.","Title: Channel Shortening For Bit Rate Maximization In Dmt Communication Systems\nYear: 2010\n\nAbstract: This paper considers the problem of impulse response shortening for discrete multitone (DMT) transceivers. The proposed channel shortening filter shortens the effective channel impulse response (CIR) while concentrating energy in a predetermined window. This approach is based on convolution operation and filter coefficients decomposition which over constraint the effective CIR and allows achieving better performance. Simulation results show that the proposed algorithm outperforms the classical channel shortening approachs in terms of achievable bit rate and computational complexity.","Title: Efficient 3D space time space block code for future terrestrial digital TV\nYear: 2008\n\nAbstract: This article introduces a 3D space-time-space block code for future terrestrial digital TV in single frequency networks. The proposed 3D code is based on a double layer structure designed for inter-cell and intra-cell space time coded transmissions. We show that this new structure is particularly efficient for SFN environments regardless of the location of the receiver. It is then suitable for fixed, portable and mobile receptions.","Title: Bit error rate prediction of coded MIMO-OFDM systems\nYear: 2008\n\nAbstract: Bit error rate (BER) prediction over channel realisations has emerged as an active research area. In this paper, we give analytical signal to interference and noise ratio (SINR) evaluation of MIMO-OFDM systems using an iterative receiver. Using this analytical SINR expression, we propose an accurate BER prediction method based on effective exponential SINR mapping (EESM) method. We show by simulations that our method is independent of the channel realisation and of the MIMO scheme. It is only dependent on the modulation and coding scheme.","Title: Time reversal applied to large MISO-OFDM systems.\nYear: 2013\n\nAbstract: In this paper we study the Time Reversal (TR) technique in a large MISO-OFDM system. This technique has been proven to provide good beamforming gains at a price of a channel estimation availability at the transmitter side. Moreover, the greater the number of transmit antennas, the better the spatial and temporal focusing gains, which makes TR a good candidate for large MISO systems while keeping a low complexity. The capacity of the system and the bit error rate (BER) performances are first derived and simulated and compared with systems applying the maximum ratio transmission technique. The time focusing gain of the TR technique is then exploited to show that the equivalent channel is shortened when the number of antennas increases, allowing a reduction of the guard interval and an increase of the spectral efficiency of the system. Finally, the robustness of the system to channel estimation errors is evaluated by adding simple channel estimation techniques in the system and evaluating the BER performance.","Title: A coded bit-loading linear precoded discrete multitone solution for power line communication\nYear: 2008\n\nAbstract: Linear precoded discrete multitone modulation (LP-DMT) system has been already proved advantageous with adaptive resource allocation algorithm in a power line communication (PLC) context. In this paper, we investigate the bit and energy allocation algorithm of an adaptive LP-DMT system taking into account the channel coding scheme. A coded adaptive LP-DMT system is presented in the PLC context with a loading algorithm which accommodates the channel coding gains in bit and energy calculations. The performance of a concatenated channel coding scheme, consisting of an inner Weipsilas 4-dimensional 16-states trellis code and an outer Reed-Solomon code, in combination with the proposed algorithm is analyzed. Simulation results are presented for a fixed target bit error rate in a multicarrier scenario under power spectral density constraint. Using a multipath model of PLC channel, it is shown that the proposed coded adaptive LP-DMT system performs better than classical coded discrete multitone.","Title: Reduced-complexity maximum-likelihood decoding for 3D MIMO code\nYear: 2014\n\nAbstract: The 3D MIMO code is a robust and efficient spacetime coding scheme for the distributed MIMO broadcasting. However, it suffers from the high computational complexity if the optimal maximum-likelihood (ML) decoding is used. In this paper we first investigate the unique properties of the 3D MIMO code and consequently propose a simplified decoding algorithm without sacrificing the ML optimality. Analysis shows that the decoding complexity is reduced from O(M8) to O(M4.5) in quasi-static channels when M-ary square QAM constellation is used. Moreover, we propose an efficient implementation of the simplified ML decoder which achieves a much lower decoding time delay compared to the classical sphere decoder with Schnorr-Euchner enumeration.","Title: Dynamic Cross-Layer Spectrum Allocation for Multi-Band High-Rate UWB Systems\nYear: 2009\n\nAbstract: In this paper, we investigate a new approach for the spectrum allocation in UWB systems. This approach consists in a cross-layer scheme that takes into consideration the different users channel quality and quality of service (QoS) requirements. The new scheme is based on the WiMedia solution proposed for multiband OFDM UWB systems. The main objective is to propose a low-complexity solution for the spectrum allocation that can manage all the users constraints. Thus, we study the optimal solution of the spectrum allocation and formulate it as a convex optimization problem. Then, we show that our proposed scheme reduces significantly the complexity of the optimal solution. Moreover, we show through simulations, that the new approach and the optimal solution have close performance in term of error rate and they outperform WiMedia solution proposed for UWB systems.","Title: A Novel Data-Aided Channel Estimation with Reduced Complexity for TDS-OFDM Systems\nYear: 2012\n\nAbstract: In contrast to the classical cyclic prefix (CP)-OFDM, the time domain synchronous (TDS)-OFDM employs a known pseudo noise (PN) sequence as guard interval (GI). Conventional channel estimation methods for TDS-OFDM are based on the exploitation of the PN sequence and consequently suffer from intersymbol interference (ISI). This paper proposes a novel data-aided channel estimation method which combin...","Title: Multi-user cross-layer allocation design for LP-OFDM high data rate UWB systems\nYear: 2008\n\nAbstract: In this paper, we investigate a cross-layer design for the packet scheduling and the resource allocation in UWB systems. This design considers the combination of queuing and channel state information (CSI) which provides QoS support for multimedia applications in UWB. For the physical layer, the use of a linear precoded orthogonal division multiplexing (LP-OFDM) waveform is proposed because of its significant performance increase compared to the WiMedia proposal. For the medium access control layer, scheduling is performed in order to differentiate between the different users and to satisfy their quality of service constraints. This cross-layer approach optimizes the system spectral efficiency and solves the problem in the WiMedia solution of cohabitation of more than three users sharing the three sub-bands of the same channel. Simulation results show that the proposed scheme leads to a considerable improvement in resource allocation and can guarantee the required quality of service.","Title: A fast decodable full-rate STBC with high coding gain for 4 × 2 MIMO systems.\nYear: 2013\n\nAbstract: In this work, a new fast-decodable space-time block code (STBC) is proposed. The code is full-rate and full-diversity for 4 × 2 multiple-input multiple-output (MIMO) transmission. Due to the unique structure of the codeword, the proposed code requires a much lower computational complexity to provide maximum-likelihood (ML) decoding performance. It is shown that the ML decoding complexity is only O(M4,5) when M-ary square QAM constellation is used. Finally, the proposed code has highest minimum determinant among the fast-decodable STBCs known in the literature. Simulation results prove that the proposed code provides the best bit error rate (BER) performance among the state-of-the-art STBCs.","Title: Multiuser Service Differentiated Spectrum Allocation Scheme for High Rate UWB Systems\nYear: 2010\n\nAbstract: In this paper, we propose a multiuser spectrum allocation scheme for high-rate UWB systems under QoS requirements. This scheme comes as a solution to the coexistence of multiple users sharing the three sub-bands of the same channel as defined in the WiMedia solution adopted for multiband UWB systems. Indeed, WiMedia solution does not allow more than three users to coexist in the same channel. Based on a constrained multiuser optimization problem, the proposed allocation algorithm allows multiple users to access the medium following a mixed sub-band assignment and priority-based scheduling approach in order to ensure an efficient differentiated spectrum sharing. The resulting time-frequency scheduling algorithm relies on the combination of two main metrics available at PHY and MAC levels: the channel quality of each user provided by the exploitation of the effective SINR method, and the QoS constraint represented by a simple weighting parameter that differentiates between two service classes. Simulation results show the efficiency of the proposed scheme and how it guarantees a good performance level for users having strict QoS requirements.","Title: On the hybrid use of unicast/broadcast networks under energy criterion\nYear: 2012\n\nAbstract: Mobile network operators are confronted to an exponential growth of their traffic. One of the causes of this growth is the delivery of video services, including mobile television. Using a broadcast component like MBMS or DVB-H for the provision of the most popular mobile television channels can minimize their impact on the mobile network traffic. Such a component can also be investigated for other types of popular services and open the door to hybrid broadcast/unicast networks. The key factor for the efficiency of this approach is the choice of the services to be transmitted through the broadcast component. In this article, we introduce an energy criterion to make this choice. We use a simplified model of an hybrid network combining a DVB-T2 broadcast component to a LTE unicast component. Through the statistical study of the reception conditions of the users in this network, we evaluate the power gain brought by the use of the broadcast component for the delivery of a service as a function of the number of users that are using this service.","Title: Coded Adaptive Linear Precoded Discrete Multitone Over Plc Channel\nYear: 2008\n\nAbstract: Discrete multitone modulation (DMT) systems exploit the capabilities of orthogonal subcarriers to cope efficiently with narrowband interference, high frequency attenuations and multipath fadings with the help of simple equalization filters. Adaptive linear precoded discrete multitone (LP-DMT) system is based on classical DMT, combined with a linear precoding component. In this paper, we investigate the bit and energy allocation algorithm of an adaptive LP-DMT system taking into account the channel coding scheme. A coded adaptive LP-DMT system is presented in the power line communication (PLC) context with a loading algorithm which accommodates the channel coding gains in bit and energy calculations. The performance of a concatenated channel coding scheme, consisting of an inner Wei's 4-dimensional 16-states trellis code and an outer Reed-Solomon code, in combination with the proposed algorithm is analyzed. Theoretical coding gains are derived and simulation results are presented for a fixed target bit error rate in a multicarrier scenario under power spectral density constraint. Using a multipath model of PLC channel, it is shown that the proposed coded adaptive LP-DMT system performs better than coded DMT and can achieve higher throughput for PLC applications.","Title: Cross-layer resource allocation scheme for multi-band high rate UWB systems\nYear: 2009\n\nAbstract: In this paper, we investigate the use of a cross-layer allocation mechanism for the high-rate ultra-wideband (UWB) systems. The aim of this paper is twofold. First, through the cross-layer approach that provides a new service differentiation approach to the fully distributed UWB systems, we support traffic with quality of service (QoS) guarantee in a multi-user context. Second, we exploit the effective SINR method that represents the characteristics of multiple sub-carrier SINRs in the multi-band WiMedia solution proposed for UWB systems, in order to provide the channel state information needed for the multi-user sub-band allocation. This new approach improves the system performance and optimizes the spectrum utilization with a low cost data exchange between the different users while guaranteeing the required QoS. In addition, this new approach solves the problem of the cohabitation of more than three users in the same WiMedia channel.","Title: On The Influence Of Carrier Frequency Offset And Sampling Frequency Offset In Mimo-Ofdm Systems For Future Digital Tv\nYear: 2008\n\nAbstract: This paper investigates the impact of carrier frequency offset (CFO) and sampling frequency offset (SFO) on the performance of different MIMO-OFDM schemes with high spectral efficiency for next generation of terrestrial digital TV. We analyze particularly orthogonal Alamouti scheme, and non-orthogonal (NO) schemes like VBLAST, linear dispersion (LD) code and Golden code. This analysis gives a global view on the best suitable MIMO-OFDM scheme with respect to CFO and SFO. We show that for high spectral efficiency, Alamouti is more sensitive to CFO and SFO. Moreover, we show that all studied MIMO-OFDM schemes are sensitive to CFO when it is greater than 1% of inter-carrier spacing. Their sensitivity to the SFO is less than to their sensitivity to CFO.","Title: Time-frequency spectrum sharing scheme for next generation OFDM systems\nYear: 2010\n\nAbstract: Spectrum sharing and multiple-access techniques are challenging matters for future wireless communication systems. Orthogonal frequency division multiplexing (OFDM) has been considered as the most important physical layer candidate for the next generation wireless systems. Alternatively, multiband OFDM (MB-OFDM) approach has shown some novelty and efficiency for high-rate (HR) WPAN applications and so has been selected for future HR ultra wideband (UWB) systems. Based on an optimization problem formulation, we jointly consider resource allocation and scheduling to define a novel dynamic timefrequency spectrum sharing mechanism based on the MB-OFDM approach. Viewed as a cross-layer solution, the proposed MB-OFDM multiple-access (MB-OFDM-MA) scheme can achieve an efficient multiuser spectrum sharing under QoS constraints. While the optimization formulation results in a high-complexity time-frequency sharing solution, we propose a low-complexity suboptimal solution able to jointly provide fairness provision with the QoS support. Compared with traditional orthogonal frequency divisionmultiple access (OFDMA), the simple time-frequency MB-OFDM-MA approach ensures a high level of fairness among the users while satisfying high-priority users having strict QoS requirements.","Title: Resource Allocation with Adaptive Spread Spectrum OFDM Using 2D Spreading for Power Line Communications\nYear: 2007\n\nAbstract: Bit-loading techniques based on orthogonal frequency division mutiplexing (OFDM) are frequently used over wireline channels. In the power line context, channel state information can reasonably be obtained at both transmitter and receiver sides, and adaptive loading can advantageously be carried out. In this paper, we propose to apply loading principles to an spread spectrum OFDM (SS-OFDM) waveform which is a multicarrier system using 2D spreading in the time and frequency domains. The presented algorithm handles the subcarriers, spreading codes, bits and energies assignment in order to maximize the data rate and the range of the communication system. The optimization is realized at a target symbol error rate and under spectral mask constraint as usually imposed. The analytical study shows that the merging principle realized by the spreading code improves the rate and the range of the discrete multitone (DMT) system in single and multiuser contexts. Simulations have been run over measured power line communication (PLC) channel responses and highlight that the proposed system is all the more interesting than the received signal-to-noise ratio (SNR) is low.","Title: Interference-Aware Prioritized Spectrum Scheduling For Mb-Ofdm Systems\nYear: 2010\n\nAbstract: In this paper, we investigate the spectrum scheduling problem for next generation multiband orthogonal frequency division multiplexing (MB-OFDM) ultra-wideband (UWB) systems under QoS and interference constraints. Based on a constrained optimization problem formulation, we propose a cross-layer scheduling solution that takes into account the unlicensed UWB users channel and QoS conditions as well as the interference impact on primary users sharing the same bandwidth. The cross-layer scheduling scheme is based on a priority approach that gives advantage to users having strict QoS requirements while avoiding or reducing the interference effect on primary users. Simulation results demonstrate the efficiency of the proposed scheme in terms of QoS support and interference reduction.","Title: Spectrum sharing optimization model for multiuser linear-precoded OFDM UWB systems\nYear: 2012\n\nAbstract: In this paper, we investigate spectrum sharing and multiple access problems for the high-rate ultra-wideband (UWB) systems. Based on the use of the linear precoded orthogonal frequency division multiplexing (LP-OFDM) proposed as an evolution of the well known multi-band OFDM (MB-OFDM) solution supported by the WiMedia Alliance for future high-rate UWB systems, the objective of this paper is to study a multiuser optimization spectrum sharing scheme for LP-OFDM UWB systems. This optimization study leads to a novel multiple access solution for the high-rate UWB systems. Simulation results demonstrate the efficiency of the use of the LP-OFDM transmission technique in the multiuser spectrum sharing scheme. Besides, the novel LP-OFDM multiuser spectrum sharing solution shows its capacity to provide a good tradeoff between the QoS support of the high-priority users and the fairness between the competing users.","Title: Robustness of MIMO-OFDM schemes for future digital TV to carrier frequency offset\nYear: 2008\n\nAbstract: This paper investigates the impact of carrier frequency offset (CFO) on the performance of different MIMO-OFDM schemes with high spectral efficiency for next generation of terrestrial digital TV. We show that all studied MIMO-OFDM schemes are sensitive to CFO when it is greater than 1% of inter- carrier spacing. We show also that the Alamouti scheme is the most sensitive MIMO scheme to CFO.","Title: System Level Evaluation Of Innovative Coded Mimo-Ofdm Systems For Broadcasting Digital Tv\nYear: 2008\n\nAbstract: Single-frequency networks (SFNs) for broadcasting digital TV is a topic of theoretical and practical interest for future broadcasting systems. Although progress has been made in the characterization of its description, there are still considerable gaps in its deployment with MIMO technique. The contribution of this paper is multifold. First, we investigate the possibility of applying a space-time (ST) encoder between the antennas of two sites in SFN. Then, we introduce a 3D space-time-space block code for future terrestrial digital TV in SFN architecture. The proposed 3D code is based on a double-layer structure designed for intercell and intracell space time-coded transmissions. Eventually, we propose to adapt a technique called effective exponential signal-to-noise ratio (SNR) mapping (EESM) to predict the bit error rate (BER) at the output of the channel decoder in the MIMO systems. The EESM technique as well as the simulations results will be used to doubly check the efficiency of our 3D code. This efficiency is obtained for equal and unequal received powers whatever is the location of the receiver by adequately combining ST codes. The 3D code is then a very promising candidate for SFN architecture with MIMO transmission. Copyright (C) 2008 Y. Nasser et al.","Title: Resource Allocation in High Data Rate Mesh WPAN: A Survey Paper\nYear: 2014\n\nAbstract: IEEE 802.15.3 High data rate wireless personal area networks (HDR WPANs) have been developed to communicate with devices within 10 m at high speed. A mesh network made up of a parent piconet and several child piconets can support multi-hop communications. Wireless mesh networks (WMNs) have been expected to be the ultimate solution for the next decade wireless networking, showing rapid progress and many new inspiring applications. The international standardization organizations formed working groups to address the problem of standardization for WMNs. These groups were the IEEE 802.15.5 (mesh extensions for WPANs),the IEEE 802.11s (mesh extensions for WLANs), and the IEEE 802.16a (mesh extensions for WiMAX). The IEEE 802.15.5 standard is the standard which defines specifications for including multi-hop functionality in the legacy 802.15.3 and 802.15.4 low data rate (LDR) WPAN networks. The impetus for a WPAN to operate in a mesh topology is to increase the network coverage without increasing the transmit power, to increase the route reliability via route redundancy, self-configuration, and efficient use of device battery life. In the case of meshed WPANs, multiple WPAN clusters compete for channel time in a shared superframe. Therefore, it is essential to determine the channel time requirements of each cluster with a certain number of devices and to determine how these clusters can compete to the shared channel time. In this paper, we investigate the different resource allocation mechanisms related to the meshed HDR WPANs for the 802.15.3 and the 802.15.5 standards. We introduce the single hop and the multi-hop IEEE 802.15.3 WPAN architectures. This is followed by the introduction of the IEEE 802.15.5 standard that provides the mesh capabilities for extending the coverage area of HDR WPANs. The current on-going research issues for resource allocation, including beacon interference, reservation collision etc., in both meshed 802.15.3 and 802.15.5 are alluded to.","Title: Analysis And Performance Comparison Of Dvb-T And Dtmb Systems For Terrestrial Digital Tv\nYear: 2008\n\nAbstract: Orthogonal frequency-division multiplexing (OFDM) is the most popular transmission technology in digital terrestrial broadcasting (DTTB), adopted by many DTTB standards. In this paper, the bit error rate (BER) performance of two DTTB systems, namely cyclic prefix OFDM (CP-OFDM) based DVB-T and time domain synchronous OFDM (TDS-OFDM) based DTMB, is evaluated in different channel conditions. Spectrum utilization and power efficiency are also discussed to demonstrate the transmission overhead of both systems. Simulation results show that the performances of the two systems are much close. Given the same ratio of guard interval (GI), the DVB-T outperforms DTMB in terms of signal to noise ratio (SNR) in Gaussian and Ricean channels, while DTMB behaves better performance in Rayleigh channel in higher code rates and higher orders of constellation thanks to its efficient channel coding and interleaving scheme.","Title: Cross-layer multiple-access optimization scheme for linear precoded OFDM UWB systems.\nYear: 2013\n\nAbstract: With the growing demand for new wireless applications accompanied with high expectations for better QoS fulfillment especially for multimedia and real-time applications, the performance of the radio resource management in a multiuser context is ensured by the ability to provide an efficient and optimized spectrum sharing scheme that should respect the wireless channel conditions and satisfy the different users’ demands. From the physical layer perspective, metrics such as spectrum efficiency and minimum BER are the most important criteria to be considered. On the other hand, from a user perspective, QoS as well as fairness among the competing users are the main metrics because they determine how much end-users are satisfied and how efficient the available resources are shared among the existing users. Based on the use of the linear precoded orthogonal frequency division multiplexing (LP-OFDM) solution proposed as an evolution of the well-known multiband OFDM (MB-OFDM) solution supported by the WiMedia Alliance for future high-rate ultra-wideband (UWB) systems, the objective of this paper is twofold. First, we study a multiuser optimization spectrum sharing scheme for LP-OFDM systems. Second, based on the optimization study, we define a novel multiple-access solution which jointly considers the frequency resource allocation and the time scheduling for the high-rate LP-OFDM UWB systems. Simulation results demonstrate the efficiency of the use of the LP-OFDM transmission technique in the multiuser spectrum sharing scheme. Besides, the novel multiuser time–frequency sharing scheme shows its capacity to provide a high performance level for high-priority users.","Title: Spectral efficiency optimization in overlapping channels using TR-MISO systems\nYear: 2013\n\nAbstract: Traditionally, cellular radio communication systems define a spectrum mask to avoid inter-channel interference between services operated in adjacent frequency bands. In this paper, we intend to optimize the spectrum efficiency of wireless systems by frequency band reused, which is performed by partial channel overlapping of adjacent bands. The impact of interchannel interference is studied for both single and multicarrier modulations. Then time reversal (TR) with multiple-input-single-output (MISO) is applied to depress the inter-channel interference thanks to the TR focusing properties. The spectral efficiency of the system is evaluated according to the overlap ratio τ and the number of transmitting antennas. The optimal τmax is identified and, as a result, channel overlapping is proved to be interesting at low SNR regime in a range depending on the number of antennas.","Title: Distributed MIMO coding scheme with low decoding complexity for future mobile TV broadcasting\nYear: 2012\n\nAbstract: A novel distributed space-time block code (STBC) for the next generation mobile TV broadcasting is proposed. The new code provides efficient performance within a wide range of power imbalance showing strong adaptivity to the single frequency network broadcasting deployments. The new code outperforms existing STBCs with equivalent decoding complexity and approaches those with much higher complexities.","Title: Adaptive Self-Learning Resource Allocation Scheme for Unlicensed Users in High-Rate UWB Systems\nYear: 2011\n\nAbstract: In this paper, we investigate a dynamic sub-band allocation for spectrum sharing in the multiuser OFDM-UWB systems while respecting the system constraints and the distributed MAC architecture. The contribution of this work is twofold: first we propose a dynamic spectrum allocation scheme based on an analytical study by deriving a multiuser optimization problem to find the optimal allocation solution. Based on an adaptive and low-complexity approach, the proposed scheme shares the available resources among unlicensed UWB users while taking into consideration the QoS requirements and the channel conditions. Then we extend the proposed scheme to allow the coexistence of a maximum number of secondary users in one channel while maintaining all the constraints support. This leads to a time-frequency allocation approach that exploits information laying in different layers, allowing a cross-layer design. This new approach improves the system performance compared to WiMedia solution and offers a considerable gain for users that have strict QoS requirements.","Title: Efficient space code block code MIMO channel estimation for future mobile video broadcasting\nYear: 2009\n\nAbstract: In this paper, we propose a novel and simple MIMO channel estimation technique based on 2D spread pilots. Taking advantage of the simplicity and flexibility of 2D spread pilot channel estimation, we easily combine the Alamouti code with the Walsh-Hadamard codes to obtain a space code block code (SCBC) MIMO scheme. We demonstrate that our proposed system is much more efficient in high mobility scenario than the classical STBC scheme. The performance evaluated over a realistic channel model shows the efficiency of this technique which turns out to be a promising MIMO channel estimation technique for the future mobile video broadcasting systems.","Title: Effect of relaying on coverage in realistic deployment scenarios\nYear: 2012\n\nAbstract: In this paper, we investigate the impact of fixed relay deployment on coverage in a wireless infrastructure. We consider a realistic scenario with OFDM based transmission which takes into account some practical constraints and assumptions (path-loss models, power, etc...). We analyze and compare scenarios with relays which operate in both half-duplex and full-duplex modes. Relays are assumed to perform decode-and-forward strategy. We show that for half-duplex scenarios, multiple relays can be deployed to extend the coverage and to compensate the loss of the half-duplex constraint where SFN-like transmission is used at the relays, i.e, the relays share the same subcarriers. Full-duplex mode outperforms the half-duplex mode in terms of coverage on the expense of additional complexity at the receiver side and offers a more convenient solution when the relay transmission power is constrained. The main contribution of this paper is giving guidelines for relay deployment in practical environments, and providing insight into benefits relays can offer in terms of coverage extension in realistic deployment scenarios.","Title: Optimization of the energy efficiency of a hybrid broadcast/unicast network\nYear: 2013\n\nAbstract: In the last few years, mobile broadband communications operators have been confronted to an exponential growth of the traffic over their networks. Recently, the use of a broadcast component as a solution to face this growth has received a particular attention. In [1], a model for a hybrid network is introduced. This network consists in the inter-operation of a LTE unicast component and a DVB-T2 broadcast component. The introduced model allows the evaluation of the potential energy gain brought by such a network. In this paper, using an enhanced version of the model previously defined, we propose a new operating mode for the hybrid network that can improve the energy efficiency of the system.","Title: A Joint Channel and Carrier Frequency Offset Estimation Based on Spread Pilot for Future Broadcasting Systems\nYear: 2009\n\nAbstract: In this paper, we propose a novel and simple joint channel frequency response (CFR) and carrier frequency offset (CFO) estimation based on 2D spread pilots for digital video broadcasting (DVB) systems. The advantage of the proposed algorithm compared to DVB-T/H standard, is to estimate the CFR and the CFO using the same pilot symbols. The performance evaluated over a realistic channel model, in a mobility scenario, in the presence of CFO, shows the efficiency of this technique which turns out to be a promising joint CFR and CFO estimation technique for the future DVB systems.","Title: Cross-Layer Resource Allocation Scheme Under Heterogeneous Constraints for Next Generation High Rate WPAN\nYear: 2010\n\nAbstract: In the next generation wireless networks, the growing demand for new wireless applications is accompanied with high expectations for better quality of service (QoS) fulfillment especially for multimedia applications. Furthermore, the coexistence of future unlicensed users with existing licensed users is becoming a challenging task in the next generation communication systems to overcome the underutilization of the spectrum. A QoS and interference aware resource allocation is thus of special interest in order to respond to the heterogeneous constraints of the next generation networks. In this work, we address the issue of resource allocation under heterogeneous constraints for unlicensed multi- band ultra-wideband (UWB) systems in the context of Future Home Networks, i.e. the wireless personal area network (WPAN). The problem is first studied analytically using a heterogeneous constrained optimization problem formulation. After studying the characteristics of the optimal solution, we propose a low-complexity suboptimal algorithm based on a cross-layer approach that combines information provided by the PHY and MAC layers. While the PHY layer is responsible for providing the channel quality of the unlicensed UWB users as well as their interference power that they cause on licensed users, the MAC layer is responsible for classifying the unlicensed users using a two-class based approach that guarantees for multimedia services a high-priority level compared to other services. Combined in an efficient and simple way, the PHY and MAC information present the key elements of the aimed resource allocation. Simulation results demonstrate that the proposed scheme provides a good tradeoff between the QoS satisfaction of the unlicensed applications with hard QoS requirements and the limitation of the interference affecting the licensed users.","Title: A Unified Broadcast Layer for Horizon 2020 Delivery of Multimedia Services.\nYear: 2014\n\nAbstract: It is expected that, in the coming years, the video data traffic carried by mobile networks will increase drastically. Among the different ways of facing this mobile data tsunami, the cooperation of several complementary access technologies offers promising prospects. In the context of such a cooperative approach, this paper proposes the definition of a unified broadcast layer, close to the 3GPP LTE technology, and therefore liable to be easily integrated in mobile devices and infrastructure equipment. The general technical requirements for the design of such a common physical layer are detailed and a practical example based on the 3GPP LTE/E-MBMS and DVB-T2 standards is described. The verification of the proposed approach is performed through simulation results and validation tests carried out on a hardware platform.","Title: An Ofdm-Cdma Scheme For High Data Rate Uwb Applications\nYear: 2007\n\nAbstract: In this paper, we investigate a new waveform for UWB systems obtained by the combination of Orthogonal Frequency Division Multiplex (OFDM) and Coded Division Multiple Access (CDMA). The proposed system, called Spread Spectrum - Multi-Carrier - Multiple Access (SS-MC-MA) turns out to be a judicious solution to combat frequency selectivity and narrowband interferers, and to manage the coexistence of several users and piconets. It is shown that the addition of a degree of freedom brought by the spreading component of SS-MC-MA allows to optimize jointly the assignment of the number of used codes and coding rates in order to make the system more robust. Through simulations, it is demonstrated that the new system can outperform Multi-Band OFDM Alliance (MBOA) for low data rates and is able to provide wider range of rates.","Title: 3D MIMO Scheme for Broadcasting Future Digital TV in Single Frequency Networks\nYear: 2008\n\nAbstract: A three-dimensional space-time-space block code for future terrestrial digital TV systems is introduced. The code is based on a double-layer structure designed for inter-cell and intra-cell transmissions in single-frequency networks. Without increasing the complexity of the receiver, the proposed code is very efficient to cope with equal and unequal received powers in single-frequency network scenarios.","Title: Optimization of linear precoded OFDM for high-data-rate UWB systems\nYear: 2008\n\nAbstract: We investigate the use of a linear precoded orthogonal frequency division multiplexing (LP-OFDM) waveform for high-data-rate ultra-wideband (UWB) systems. This waveform applied for the first time to UWB applications is an evolution of the multiband OFDM (MB-OFDM) solution supported by WiMedia-MBOA (MultiBand OFDM Alliance). The aim of this paper is twofold. Firstly, an analytical study of the LP-OFDM waveform allows to find how to efficiently precode an OFDM signal in order to improve the robustness of the system. Secondly, a global system study is led to highlight the benefits of adding a precoding function to an OFDM signal in the UWB context. Different system choices and parameterization strategies are thus proposed. In both analytical and global system studies, the LP component is optimized without channel state information (CSI) at the transmitter side, as in the MBOA solution. To go further, the MBOA constraints are relaxed, and an additional optimization is performed, with a CSI at the transmitter. The analytical and simulation results show that the joint use of linear precoding and OFDM leads to a significant performance increase compared to the MBOA solution. This improvement is due to the precoding scheme that provides better exploitation of the channel diversity.","Title: Admission control and resource allocation strategies for IEEE 802.15.5.\nYear: 2014\n\nAbstract: In this paper, a novel two-tier mechanism for the access control and resource allocation of a two-hop high data rate IEEE 802.15.5 network is proposed. One of the main contributions in this proposal is the use of a dynamic superframe size in the IEEE 802.15.5. Additionally, the ideas of using a superframe utilization threshold and a channel time allocation period (CTAP) utilization threshold are incorporated, with the aim of allowing more devices to access the shared superframe. Using these thresholds, two new hop-1 resource allocation algorithms are described; CTAP utilization threshold-based resource allocation algorithm (CTRA) and superframe utilization threshold-based resource allocation algorithm (STRA). These algorithms control the access of the hop-1 Mesh PicoNet Coordinators (MPNCs) and Mesh DEVices (MDEVs) to the channel time resources in the superframe. Furthermore, an algorithm for the distribution of the hop-1 MPNCs' resources among the real-time (RT) and non-real-time (NRT) MDEVs at hop-2 is introduced. In this mechanism, a new flag metric is applied to provide higher priorities to MPNCs over MDEVs in the hop-1 algorithms and to the RT MDEVs over the NRT MDEVs in the hop-2 algorithm. Simulation results show the superiority of the proposed mechanism over others by providing a higher satisfaction factor and higher fairness among the competing devices at both hop-1 and hop-2 of the IEEE 802.15.5.","Title: Analysis of Decoding Failures of DVB-S2 LDPC Codes\nYear: 2014\n\nAbstract: The DVB-S2 standard has been revealed a reliable standard for data transmission thanks to the LDPC codes. The waterfall of the bit-error rate is very abrupt to reach low values of order 10^(-10) in the error floor. In this paper, we propose a novel investigation on the iterative decoding of these codes by the Belief Propagation algorithm on Gaussian channels. Introducing definitions of persistent and created errors, we analyze them according to the degrees of the Tanner graph nodes and the noise power of the channel. We show that the two-degree bits used to accelerate the encoding process of the DVB-S2 are responsible for most decoding errors. This result could help design DVB-S2 codes in a more appropriate way, for instance avoiding the use of BCH codes to tackle the error floor.","Title: Partial channel overlay in moderate-scale MIMO systems using WH precoded OFDM\nYear: 2014\n\nAbstract: In this paper, we propose a multi-user orthogonal frequency-division multiplexing (OFDM) transmission scheme with partial channel reusing between users, suitable for multiple-input multiple-output (MIMO) systems with not-so-large antenna arrays. The spectrum efficiency is improved at the price of adding multiuser interference (MUI) which can be overcome by the multiplexing gain of MIMO. In addition, we propose to employ the spread spectrum OFDM (SS-OFDM) modulation with the Walsh-Hadamard (WH) codebook for such a multi-user system to further mitigate the MUI. The bit error rate (BER) performance of the proposed scheme in the additive white Gaussian noise (AWGN) channels is derived analytically and then verified through Monte-Carlo simulations. Moreover, the proposed SS-OFDM system with partial bandwidth reuse is compared with the OFDM-based solution in a moderate-scale MIMO system with equal gain combining (EGC) precoding. It is shown that the proposed SS-OFDM MIMO system outperforms the classical OFDM approach in frequency-selective fading channels when the number of transmit antennas remains moderate.","Title: Distributed MIMO schemes for the future digital video broadcasting\nYear: 2013\n\nAbstract: This paper studies the application of distributed multiple-input multiple-output (MIMO), i.e. MIMO transmission over several geographically separated but cooperated transmitters, for future TV broadcasting systems. It is first shown that distributed MIMO is promising for the future broadcasting systems from a channel capacity perspective. Several STBCs that can be applied in the distributed MIMO broadcasting scenarios are then discussed. Through performance comparison and complexity analyses with realistic system settings and channel model, it can be concluded that simple STBCs are efficient for low data rate applications, while the sophisticated ones are more suitable to deliver high data rate services.","Title: EVM Derivation for Multicarrier Signals: Joint Impact of Non-Linear Amplification and Predistortion\nYear: 2015\n\nAbstract: Orthogonal Frequency Division Multiplexing (OFDM) systems are extremely sensitive to nonlinear distortions brought by the Power Amplifier (PA) because of the inherent high peak-to-average power ratio (PAPR) of the signal. In this paper, we analyze the impact of the non-linearities on the output signal, assuming that the transmitter is equipped with a linearization stage before the PA. Therefore, the expression of the Error Vector Magnitude (EVM) of the output signal is analytically derived with or without the use of the linearization process. We provide analytical expressions of the EVM in the form of a series expansion that depends on the transition factors of the PA and predistortion characteristics. Simulations results confirm the accuracy of our analytical expressions, even when limiting the series expansion to an order of 4.","Title: Tone Reservation Based PAPR Reduction Technique with Individual Carrier Power Allocation for Multiple Peaks Reduction\nYear: 2015\n\nAbstract: Multiple algorithms have been proposed to deal with the high Peak to Average Power Ratio (PAPR) of Orthogonal Frequency Division Multiplexing (OFDM) signals. The second generation of Digital Video Broadcasting for Terrestrial transmission (DVB-T2) has adopted the Tone Reservation (TR) technique. It is based on an iterative process where, at each iteration, a predefined kernel is used to reduce one peak in time domain. In this paper, we first assess the performance of the power control schemes adopted by TR algorithms. Then, we analyze their complexity and show that their kernels are complex to generate, hence practically, difficult to implement. Secondly, we propose a novel kernel signal, simple to implement yet compatible with the DVB-T2 standard to reduce the PAPR value. The proposed kernel is defined to deal with the reduction of multiple peaks at a time. We show that the proposed PAPR reduction outperforms the TR algorithm adopted in DVB-T2 in terms of CCDF and Modulation Error Rate (MER) and then offers a better performance-complexity tradeoff.","Title: Towards the Next Generation Video Broadcasting: Improved Performance Using Distributed MIMO\nYear: 2015\n\nAbstract: Distributed multiple-input multiple-output (MIMO) is a promising technique for the future terrestrial TV broadcasting systems due to its higher efficiency over the traditional single-antenna transmission scheme. In this article, we investigate the application of distributed MIMO technique in the future broadcasting systems. We first show the advantage of the distributed MIMO from the channel capacity perspective. Then different space-time block codes (STBCs) that can be applied in the distributed MIMO broadcasting scenarios are discussed and compared with real system configurations and channel conditions. It is shown that the simple and sophisticated STBCs have mutually complementary application scenarios. Particularly, short 3D MIMO code is a good performance versus decoding complexity compromise among different application scenarios.","Title: Resource allocation mechanism in IEEE 802.15.3 parent/child model\nYear: 2015\n\nAbstract: In this paper, a novel mechanism for access control and resource allocation in a parent/child IEEE 802.15.3 network is proposed. The main objective is twofold; first the idea of using a dynamic superframe size that ensures the efficient use of the superframe is implemented; second, two approaches for resource allocation among child PicoNet Coordinators (PNCs) and DEVices (DEVs) based on the dynamic superframe idea are proposed. These approaches, referred to as “” and “”, are studied in comparison with a basic approach. They share the concepts of the dynamic superframe size, the prioritization of child PNCs over DEVs and the admission control strategy. They are then differentiated by applying the superframe utilization threshold concept in the “” and the addition of service differentiation in the “”. The superframe utilization threshold allows more devices to access the shared superframe, whereas the service differentiation guarantees the quality of service (QoS) requirements for real-time traffic requests. The proposed mechanism seeks to apply a compromise between the devices satisfaction, the QoS requirements and the requests rejection rate to induce fairness and service differentiation. Simulation results show that the proposed algorithms provide high satisfaction for both child PNCs and DEVs with greater satisfaction for the child PNCs. The superiority of using the dynamic superframe size over the fixed superframe size is also shown by the efficient use of the superframe and the minimized delay between the channel time allocations of the same device in the subsequent superframes.","Title: BER Analysis for Equal Gain Transmission in Downlink Multiuser MIMO Systems\nYear: 2015\n\nAbstract: In this letter, we derive the bit error rate (BER) performance of a multiuser multiple-input multiple-output (MU-MIMO) system using equal gain transmission (EGT) precoding. Considering binary phase shift keying (BPSK) modulation, we analyze and propose a statistical model based on the Nakagami distribution for the interference terms involved in the received signals. The precision of our model is compared with the common Gaussian approximation and validated through simulation results. A closed-form BER expression is derived as a function of the number of transmit antennas for arbitrary E b /N 0 .","Title: Impact Of Clipping On Evm Of The Predistorted Non-Linear Amplified Multicarrier Signals\nYear: 2015\n\nAbstract: In this paper, we analyse the impact of clipping used as a peak-to-average power ratio reduction technique for multicarrier signals, assuming that a linearization stage is implemented before the power amplifier. Therefore, the expression of the Error Vector Magnitude (EVM) of the output signal is analytically derived. We mainly focus on the effect of clipping in both cases with and without the use of the predistortion stage. We provide analytical expressions of the EVM in the form of a series expansion that depends on the clipping threshold, the transition factors of the power amplifier, the predistortion characteristics and the PAPR of the input signal. We show through simulations that the proposed expressions present very good accuracy, even when limiting the series expansion to an order of 4. Finally, a simple scenario is proposed to demonstrate the benefit of our EVM expressions in controlling the average transmitted power with the aim to guarantee a high quality of service.","Title: Global throughput maximization of a hybrid unicast-broadcast network for linear services\nYear: 2015\n\nAbstract: An exponential growth of the mobile data traffic is expected in future networks. The reason of this growth is related to the increasing popularity of linear services such as mobile TV, live and sports events, which may lead to the delivery of the same contents to a large audience. Recent studies have shown that network cooperation is a promising candidate to deal with such an issue. This paper investigates, from a planning perspective, the optimization of a hybrid unicast/broadcast network for linear services. We introduce an analytical approach with the aim of stating an optimization problem on the hybrid network configuration, namely in terms of the sharing between the unicast and broadcast modes to deliver a particular service. We show by simulation the existence of an optimal operation point which leads to the maximization of the overall throughput of the hybrid network under the constraint of a minimum service success rate.","Title: Modeling Bp Decoding Error Events With The Ldpc Codes Of The Dvb-S2 Standard\nYear: 2014\n\nAbstract: In this paper, we focus on the behavior of the Belief Propagation (BP) algorithm when decoding the Low-Density Parity-Check (LDPC) code of rate 3/4 in the DVB-S2 standard. By studying the topological structure of its Tanner graph, we raise properties inherent to the degree distribution that turns out to be strongly correlated with the decoding failures. The irregularity of the degrees seriously damages the performance, visible by an abrupt error floor. We accordingly propose a novel model of the error events based on the degree distribution which helps simulate typical error events and observing the flaws of the DVB-S2 code. This work could be used as a good basis to design future codes with a better decoding behavior, especially in the error floor region.","Title: Power control optimization for tone reservation based PAPR reduction algorithms\nYear: 2014\n\nAbstract: Orthogonal Frequency Division Multiplexing (OFDM) systems are characterized by a high Peak to Average Power Ratio (PAPR), which results in performance degradation due to High Power Amplifiers' nonlinearities. Multiple algorithms have been proposed to deal with this problem, among them the Tone Reservation (TR) technique adopted in the second generation of Digital Video Broadcasting for Terrestrial transmission (DVB-T2). In this paper we analyze the power control schemes used in TR algorithms by assessing both their performance and complexity. We then propose two novel solutions to improve the performance without requiring additional complexity. The first is based on an enhanced selection of the peaks to be processed by the TR algorithm, and the second is based on a dynamic computation of the threshold used. We show that the proposed techniques outperform the TR algorithm adopted in DVB-T2 by 0.77 dB in terms of CCDF gain at 10\n<sup xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">-3</sup>\n and by 0.33 dB in terms of Input Back Off (IBO) gain for an MER of -40dB.","Title: Admission control and resource allocation in IEEE 802.15.5 HDR WPAN\nYear: 2014\n\nAbstract: In this paper, we propose a novel admission control and centralized channel time allocation algorithm for two-hop communications based on the IEEE 802.15.5 specifications. The first contribution is the introduction of a dynamic superframe size for efficient resource utilization. The second contribution adopts the use of a threshold concept and a flag metric, which jointly provide satisfaction and fairness among the requesting devices. The performance of the proposed algorithm is evaluated and the results show that the proposed approach achieves good performance in terms of satisfaction and fairness.","Title: SDRAN-based user association and resource allocation in heterogeneous wireless networks\nYear: 2016\n\nAbstract: Integrating different types of wireless access technologies in heterogeneous networks provides flexible choices for users to handover to a more preferable network that satisfies their needs. In this paper, we discuss the recent 802.21.1 Software Defined Radio Access Network (SDRAN) architecture that couples the Software Defined Networking (SDN) mechanisms by the IEEE 802.21 Media Independent Handover Services. Based on this architecture, we propose a centralized handover algorithm that manages user association and downlink capacity resource allocation. The algorithm is formulated as a Generalized Assignment Problem (GAP) which maximizes the overall profit of mobile nodes which is studied in terms of power consumption and Received Signal Strength (RSS). Comprehensive simulation has been conducted; It is shown that the formulated algorithm reduces the overall consumed power and increases the average relative received signal strength while demanding an extra number of HOs.","Title: Optimal operating point of the power amplifier with respect to the EVM for TV broadcasting applications\nYear: 2016\n\nAbstract: The high Power Amplifier (PA) dominates the power consumption in DVB transceiver systems and requests focusing the power efficiency improvements on this device; especially when multi-carrier modulations are used which is characterized by high Peak-to-Average Power Ratio (PAPR). In this paper, first, we provide analytical expression of the Error Vector Magnitude (EVM) of the amplified signal when polynomial predistortion and clipping techniques are activated. Secondly, a theoretical analysis of the trade-off between the power amplifier linearity and efficiency is investigated considering the performance of the clipping and predistortion techniques. Therefore, an analytical expression which gives the optimal Input Back-Off (IBO) maximizing the PA efficiency, with respect to an EVM constraint, is provided. It is demonstrated that significant power efficiency improvement can be achieved using our analytical expressions.","Title: Optimal capacity of hybrid unicast-broadcast networks for mobile TV services\nYear: 2016\n\nAbstract: This paper studies the optimal broadcast coverage radius in a hybrid network from the perspective of network planning. The hybrid network is composed of a unicast and broadcast systems which cooperate to deliver linear services on portable and mobile devices. First we define a hybrid approach in which the coverage radius of the broadcast network is adjusted according to the distribution of the users in the service area. Then, by applying convex optimization methods, we derive a closed-form expression of the optimal broadcast coverage radius that maximizes the average capacity per user in the hybrid network. The results show that the closed-form expression of the broadcast coverage radius is in good agreement with the simulation results, which validates the theoretical approach.","Title: Blind repetitions for cellular-IoT: Performance analysis of combination mechanisms\nYear: 2016\n\nAbstract: As evident from the increasing number of technical studies on machine type communications, Internet of Things (IoT) has become a hot topic. Among all the initiatives attempting to build end-to-end connectivity solutions, the 3rd Generation Partnership Project (3GPP) is actively working on the integration of IoT features into the existing 2G and 4G network specifications. In the framework of the adaptation of the GSM (Global System for Mobile communication) and EDGE (Enhanced Data rates for GSM Evolution) systems to IoT applications, several solutions have been proposed at the PHY/MAC level, all sharing a common trait: enhancing the coverage using blind repetition mechanisms. The question of how to recombine the repeated symbols at the receiver side has then to be raised. In this work, we analyze the performance of state-of-the-art combination approaches that are carried out at different levels of a classic GSM receiver and we propose a new combination scheme. To mitigate the drawbacks of each individual solution, two hybrid strategies are proposed involving these different levels of combination. Performance comparison is led for the different cases considering constant as well as time variant channels, leading to a discussion about the most efficient strategy to follow, depending on the targeted application. To the best of our knowledge, this is the first work dealing with the comparison of combination mechanisms including hybrid combining solutions.","Title: Quantifying the Memory Effects of Power Amplifiers: EVM Closed-Form Derivations of Multicarrier Signals.\nYear: 2017\n\nAbstract: This letter deals with the analysis of the memory effects of the power amplifier (PA) on multicarrier signals. We analytically derive the expressions of the error vector magnitude (EVM) of amplified signals jointly considering a memory polynomial PA model and clipping effects. We provide analytical EVM expressions that depend on the PA coefficients, PA order, and memory depth as well as the clippi...","Title: A low latency algorithm for efficient PAPR reduction for DVB-T2 and ATSC 3.0 broadcast\nYear: 2017\n\nAbstract: In this paper, the problem of peak-to-average power ratio (PAPR) reduction of orthogonal frequency division multiplexing (OFDM) signals is investigated in the context of second generation of digital video broadcasting for terrestrial transmission (DVB-T2) and the American digital video broadcasting standard ATSC3.0. As a multicarrier modulation technique is characterized by a high PAPR of the transmitted signal and OFDM is prone to non-linear (NL) effects of power amplifiers. DVB-T2 and ATSC3.0 have adopted a gradient-based tone reservation (TR) PAPR reduction technique which is based on an iterative process where, at each iteration, a predefined kernel is used to reduce one peak in time domain. Recently, a new TR PAPR reduction technique termed individual carrier allocation for multiple peaks (ICMP), based on a novel kernel definition has been proposed. However, it suffers from latency issues in ATSC3.0 and higher modes of DVB-T2. So, we propose another novel TR technique, grouped ICMP (GICMP). The simulation results show that GICMP offers better performance than the gradient-based DVB-T2 algorithm. Furthermore, it not only yields the same performance as ICMP but also has less latency.","Title: Time reversal based beamforming for wireless communications: A theoretical analysis from moderate-scale to large-scale antenna systems\nYear: 2015\n\nAbstract: This paper aims at presenting a comprehensive analysis of time-reversal (TR) as a precoding scheme for wireless communications. As exposed in the paper, TR can be viewed as a particular spatial filtering scheme with some strong commonalities with more conventional beamforming strategies used in large-scale antenna systems (LSAS). Our goal is to compare the behavior and the performance of the TR-based beamforming with that of the well-known Maximum Ratio Transmission (MRT). Therefore, we develop a generic system model for TR-based beamforming systems. Then we lead a theoretical performance analysis relying on ergodic capacity computations and based on statistical models of the equivalent channels obtained with both RT and MRT precoding schemes.","Title: Space Shift Keying Modulations for Low Complexity Internet-of-Things Devices.\nYear: 2017\n\nAbstract: The new MIMO schemes commonly referred to as spatial modulations (SM) use the index of the transmit (resp. the receive) antenna to transmit additional data and hereby increase the spectral efficiency of digital communication systems. Most of proposed SM schemes allow for simultaneously transmitting two types of symbols, the classical IQ-symbols of a M-order modulation, and a spatial information related to the index of the transmit or targeted receive antenna(s). In this paper, we propose a complete uplink and downlink transmission system between a Base Station and a Connected Device for Internet-of-Things application based on SM. We thus concentrate on SM schemes only conveying spatial information for both downlink (from a base station to a connected device) and uplink (from a connected device to a base station) transmissions and leading to low complexity at the device. For the downlink we propose the use of an existing receive SM, the ERASK that is an extension of a very simple receive spatial modulation known as RASK (Receive Antenna Shift Keying) which aims at increasing the spectral efficiency. For the uplink, we firstly propose a new scheme which analogy with ERASK is highlighted. Both ERASK and ESSK can be efficiently used in the global system while keeping very low complexity at the device side. All complexity is reported at the base station with pre-processing function used to concentrate the signal energy towards the device targeted antenna(s). The channel estimation is also only needed at the base station for pre-processing of the ERASK and detection process of the ESSK.","Title: Near optimal linear-service oriented resource allocation strategy for LTE networks\nYear: 2017\n\nAbstract: Extensive research efforts are being attracted towards resource allocation in LTE networks. However, from the resource allocation perspectives, linear services such as Digital TV programs have been scarcely studied. In fact, for end-users requesting linear services, the maximization of the the number of users served is much important goal than maximizing the overall capacity of the network. In this paper, linear services oriented resource allocation strategy is proposed. The goal of the proposed algorithm is to maximize the success rate of the users, depending on a minimum required quality of service (QoS), measured in terms of user capacity constraint. The proposed method was tested and compared to the optimal branch and Cut solution and has shown good robustness against network and user parameters.","Title: Application of AL-FEC to HbbTV DVB-T2 systems for the provision of VoD services in portable and fixed TV reception\nYear: 2017\n\nAbstract: Interactive TV services have grown significantly in recent years thanks to the standardization of several approaches such as Hybrid broadcast and broadband TV (HbbTV) technology. HbbTV applied to portable and fixed DVB-T2 reception, allows TV channels broadcast via terrestrial channels to publish interactive applications in addition to their TV programs. Video-On-Demand (VoD) provided via Push-VoD service, represents one of these applications. However the provision of HbbTV Push-VoD services in portable and fixed DVB-T2 reception is challenged by severe propagation conditions of the static multipath scenarios. In this paper, we provide a first reference addressing the question of how HbbTV DVB-T2 systems enables a reliable delivery of Push-VoD services in portable and fixed broadcast reception, in absence of broadband networks. We analyze the coexistence of HbbTV standard and DVB-T2 standard. We investigate the solution based on the LDPC-Staircase AL-FEC scheme for improving the portable and fixed broadcast reception of Push-VoD service in HbbTV DVB-T2 systems. Simulation results show that LDPC-Staircase AL-FEC code enables to enhance the delivery of HbbTV Push-VoD services over DVB-T2 systems with portable and fixed TV reception.","Title: EVM derivation of multicarrier signals to determine the operating point of the power amplifier considering clipping and predistortion.\nYear: 2016\n\nAbstract: The high peak-to-average power ratio (PAPR) remains a major drawback of multicarrier modulations. Hence, the nonlinear characteristics of power amplifiers (PA) results in strong distortion and low power efficiency when multicarrier modulations are used. In this paper, the impact of the nonlinearities on the amplified multicarrier signal is analyzed, considering both PAPR limitation using clipping and PA linearization. Therefore, we derive the expression of the error vector magnitude (EVM) of the amplified signal with and without the use of polynomial predistortion and/or clipping techniques. We provide analytical EVM expressions that depend on the PA and predistortion characteristics, as well as the PAPR and the average power of both input and clipped signals. These expressions are general formulas which allow to measure in-band distortion at the PA output. The simulation results show that the proposed expressions present perfect accuracy. Moreover, the trade-off between the PA linearity and efficiency is investigated considering the performance of the clipping and predistortion techniques. An analytical expression which gives the optimal input back-off (IBO) maximizing the PA efficiency with respect to any EVM constraint is provided. Finally, the predistortion complexity is discussed aiming at reducing it with respect to an EVM constraint.","Title: Mobile TV directed resource allocation scheme for LTE networks\nYear: 2017\n\nAbstract: Resource allocation in LTE networks has attracted extensive research efforts. However, linear services (such as Digital TV programs) have been barely considered from the resource allocation perspectives. Indeed, when a linear service is offered to the end-users, the network should satisfy the maximum number of users served rather than maximizing the sum rate capacity. In this paper, The linear services oriented resource allocation strategy is discussed. This approach aims at maximizing the users' success rate, based on a minimum required quality of service (QoS) measured in terms of minimum throughput. Here, the focus will particularly on both Poisson Process and grid models as well their comparison. The method was tested in different scenarios and under different conditions to prove its robustness against user density, user requirements and reuse factor.","Title: MIMO Channel Hardening: A Physical Model based Analysis.\nYear: 2018\n\nAbstract: In a multiple-input-multiple-output (MIMO) communication system, the multipath fading is averagedover radio links. This well-known channel hardening phenomenon plays a central role inthe design of massive MIMO systems. The aim of this paper is to study channel hardening using aphysical channel model in which the influences of propagation rays and antenna array topologiesare highlighted. A measure of channel hardening is derived through the coefficient of variationof the channel gain. Our analyses and closed form results based on the used physicalmodel are consistent with those of the literature relying on more abstract Rayleigh fadingmodels, but offer further insights on the relationship with channel characteristics.","Title: Extended Receive Spatial Modulation Mimo Scheme For Higher Spectral Efficiency\nYear: 2018\n\nAbstract: Spatial modulation (SM) schemes use the index of the transmit (resp. the receive) antenna to transmit additional data, and therefore allow for a new exploitation of the space dimension to increase the spectral efficiency of MIMO systems. The baseline of the proposed study relies on the so-called extended receive antenna shift keying (ERASK) concept consisting in transmitting the spatial bits by focusing the transmit energy towards subsets of receive antennas. With ERASK, the subsets of targeted antennas can be of any size leading to an amount of additional spatial bits directly equal to the number of receive antennas. However, an ON/OFF keying is used to carry the spatial symbols meaning that a zero power level is associated for non targeted antennas. Consequently, the transmission of additional IQ symbols in ERASK is not compatible with every spatial symbol configurations. In this paper, we propose a novel scheme referred to as Extended RSM (ERSM) that sends additional IQ modulated signals in ERASK, in order to increase the overall spectral efficiency. The key idea is to associate a non zero power levels for all spatial symbols. Two power levels are used to transmit the spatial and the IQ symbols, and an optimal ratio of the power is calculated. The theoretical performance of the proposed ERSM scheme is derived and then validated through simulations.","Title: Angular Based Beamforming and Power Allocation Framework in a Multi-User Millimeter-Wave Massive MIMO System\nYear: 2018\n\nAbstract: Millimeter-Wave (mmWave) systems offer extremely large Band-Width (BW) and highly Line-of-Sight (LOS) dominant channels. Many beamforming and power allocation techniques have recently emerged to leverage such channel characteristics in order to enhance the sum capacity and the coverage for multi-user transmissions. However, most of the existing algorithms are practically limited due to full channel knowledge and high complexity requirements. In this paper we introduce a novel, low complexity, angular based beamforming and power allocation framework, that requires the knowledge of the main contributive Directions of Arrival/Departure (DoAs/DoDs) of the propagation channel only. Exploiting the fact that propagation conditions are highly driven by the geometrical structure of the channel in mmWave scenarios, our method relies on the estimation of the leakage caused by each User Equipment (UE) on all the other UEs, approximated from the DoAs contributions. We prove with the simulation results that this approach, in addition to be practically plausible, can enhance the network Spectral Efficiency (SE) alongside with maintaining acceptable data rates for the cell edge UEs.","Title: Efficiency of the TR Technique and the POPS Algorithm for Waveform Optimization in MISO-OFDM Systems\nYear: 2018\n\nAbstract: In this paper, we extend the joint use of the Time Reversal (TR) precoding technique and the Ping-Pong Optimized Pulse Shaping (POPS) waveform optimization algorithm to design optimal transmit and receive waveforms for Multiple Input Single Output-Orthogonal Frequency Division Multiplexing (MISO-OFDM) systems. The design process banks on a full characterization of the scattering function of the TR-MISO channel, followed by the implementation of the POPS approach. Numerical analysis reveal a significant gain, brought by the optimized waveforms in TR-MISO-OFDM systems, in terms of Signal to Interference plus Noise Ratio (SINR), compared to the optimized waveforms in TR-SISO-OFDM systems and the PHYDYAS waveform in TR-MISO-OFDM systems.","Title: Theoretical Performance of Coherent and Incoherent Detection for Zero-Forcing Receive Antenna Shift Keying.\nYear: 2018\n\nAbstract: Receive antenna shift keying (RASK) is a multiple-input multiple-output transmission scheme that applies the spatial modulation concept at the receiver. Zero-forcing (ZF) precoding can be used at the transmitter side to steer the signal toward the targeted antenna. Different detection schemes can be implemented in order to detect the RASK information. Moreover, for the sake of low complexity and low power consumption, decreasing the number of RF chains by the use of switched antennas can also be implemented. In this paper, we lead a theoretical analysis of the transmission of spatial symbols in RASK, using the ZF precoding. Two detection schemes relying on maximum likelihood (ML) detection are presented and proved to be implemented as a simple single-tap detector. One, called coherent ML (CML), needs one RF chain per receive antenna and the other one, named incoherent ML (IML), is based on an envelope detector and needs only one RF chain for the entire antenna array. Theoretical expressions of the bit error rate (BER) are derived for each scheme and validated through simulation. The CML detector is demonstrated to obviously outperform the IML detector in terms of the BER but at the cost of higher complexity and power consumption. We then proposed to use switches to reduce the number of RF chains, thus trading off some performance against complexity. At some point, IML is shown to outperform CML, depending on the number of switches, RF chains, and system configuration. This turning point is analytically calculated and validated through simulations, and so can be used for further system comparison studies.","Title: Very Efficient Tone Reservation PAPR Reduction Fully Compatible With ATSC 3.0 Standard: Performance and Practical Implementation Analysis.\nYear: 2018\n\nAbstract: A common issue in any multicarrier communication system such as the American digital video broadcasting (ATSC 3.0) standard is the high peaks of the transmitted signal. This disadvantage constrains the high power amplifiers to be deployed in their linear region which lowers their power efficiency. To overcome this issue, various techniques aiming at reducing the signal fluctuations have been investigated. Recently, Tone Reservation (TR)-based algorithms have been studied and proposed for the second generation of digital video broadcasting (DVB-T2) and ATSC 3.0 standards. The algorithm is essentially based on an iterative gradient approach to cancel one peak of the time-domain signal at each iteration. In this paper, we propose a novel TR-based algorithm implementable and compatible with the ATSC 3.0 standard and named as grouped carrier peak windowing (GCPW). This algorithm is based on a new kernel definition targeting the cancellation of multiple signal peaks at a time which highly reduces the total number of iterations, therefore reducing the system global latency and being more suited for implementation in today's ATSC 3.0 transmitters. Taking into account hardware resources requirements, we propose a new method to select the highest signal peaks to be considered in the PAPR reduction process. Hence, the system latency, complexity, and memory resources are reduced and better performance than the ATSC 3.0 gradient-based algorithm can be offered. The implementation of the GCPW algorithm in fixed-point architectures is also studied and optimized in this paper. We demonstrate through simulation results that the proposed algorithm offers very good performance/latency/complexity/memory tradeoff in both floating and fixed point implementations.","Title: A Novel Statistical Model Of Ofdm Signals With Optimal Tone Reservation Papr Reduction For Evm Theoretical Analysis\nYear: 2018\n\nAbstract: In this paper, we study the statistical distribution of the OFDM time-domain signal amplitude to which the optimal tone reservation (TR) PAPR reduction solution is applied. This study comes as a major step towards the derivation of the analytical expression of the error vector magnitude (EVM) metric on OFDM waveforms when non-linear amplification is used. We demonstrate through simulations that this distribution has a bimodal behavior and can be tightly modeled as a superposition of two probability density functions whose expressions are given and whose parameters are obtained through proper estimation. From such analysis, we provide the complete distribution expression of the PAPR-reduced OFDM signal, which can be further exploited for the derivation of performance bounds of the TR PAPR reduction strategy.","Title: Leakage Based Users Selection For Hybrid Beamforming In Millimeterwave Mimo\nYear: 2018\n\nAbstract: Millimeter-Wave (mmWave) channels are characterized as being sparse and Line of Sight (LoS) dominated. Applying hybrid beamforming in mmWave systems guarantees Multi-User (MU) transmission with low hardware requirements. Zero Forcing (ZF) precoding can be utilized at the digital layer of the hybrid beamformer to mitigate the Inter User Interference (IUI). However, due to the sparse nature of the mmWave channels, a ZF precoder may provide limited performance due to the highly correlated MU channel matrix. The aim of this paper is hence to address the interference mitigation problem in the analog domain before handling it at the ZF digital level. Signal to Leakage and Noise Ration (SLNR) based users selection algorithm is introduced for maximizing the Spectral Efficiency (SE) of the system with low complexity requirements. We finally highlight by simulations the potential gains achieved by the introduced interference aware users selection approach.","Title: Analytical Study of the Probability of Coverage in Hybrid Broadcast-Unicast Networks\nYear: 2018\n\nAbstract: Linear services such as mobile TV are witnessing an increasing demand, creating challenges for conventional delivery methods in both broadcast and broadband networks. To overcome this problem, a hybrid delivery approach combining both networks has recently emerged. Using stochastic geometry and a Poisson point process distribution of the users and access points, this paper discusses a hybrid broadcast/unicast network model where an analytical approach was undertaken, deriving a new expression for the probability of coverage. The model is then used to assess the feasibility of the proposed hybrid scheme and then extended to numerically find the optimal operation point in terms of key design parameters including user density in the network.","Title: In-Band and Out-Of-Band Distortions Optimization for ATSC 3.0 Transmission: A Novel TR PAPR Reduction Algorithm\nYear: 2018\n\nAbstract: This paper presents a novel Tone Reservation (TR) Peak to Average Power Ratio (PAPR) algorithm for the American digital video broadcasting (ATSC 3.0) standard. As any multicarrier communication systems, ATSC 3.0 transmitters suffer from high peaks of the transmitted signal. The high signal fluctuations constitute a major disadvantage of these systems. Indeed, they constraint the high power amplifiers to operate in their linear region which leads to poor spectral efficiency. To reduce the amplitude of the signal peaks, a TR gradient based algorithm has been proposed as an option within the ATSC 3.0 standard. Although TR-based algorithms have been intensively investigated these last years, very few today's transmitters implement these algorithms as they do not offer a high performance/complexity trade-off. This paper presents a novel TR-based algorithm, named as Grouped Carrier Peak Windowing (GCPW), which offers higher performance than the standard TR gradient-based algorithm while being implementable and compatible with the ATSC 3.0 standard. We also show through simulations that the proposed algorithm highly outperforms the gradient-based standard algorithm in terms of both in-band and out-of-band distortions. Results show that the novel algorithm offers very promising performance in terms of both modulation error rate and shoulder attenuation.","Title: MIMO Channel Hardening for Ray-based Models\nYear: 2018\n\nAbstract: In a multiple-input-multiple-output (MIMO) communication system, the multipath fading tends to vanish with increasing number of radio links. This well-known channel hardening phenomenon plays a central role in the design of massive MIMO systems. It is quantified by the coefficient of variation of the channel gain. The aim of this paper is to study channel hardening using a physical channel model in which the influences of propagation rays and antenna array topologies are highlighted. Our analyses and closed form results extend the hardening properties beyond the classical Rayleigh fading models and offer further insights on the relationship with channel characteristics.","Title: Context-aware and priority-based user association and resource allocation in heterogeneous wireless networks.\nYear: 2019\n\nAbstract: Heterogeneous wireless networks (HWNs) are usually characterized by the integration of cellular networks and wireless local area networks (WLANs) to meet user requirements and enhance system capacity. This paper proposes a user association and downlink resource allocation algorithm in HWNs with users having different priorities. The proposed solution employs contextual information related to the preferences of the users, their requested data rate, and the characteristics of networks. The user preference is translated through a profit function that is based on the received signal quality and the power consumption at mobile terminals (MTs). Accordingly, an optimization problem is formulated to maximize the overall user satisfaction for each priority level. The formulated problem throws firm restrictions to prevent low-priority users from allocating resources utilized by other users with higher priorities. We then propose a novel heuristic approach with polynomial-time complexity to approximate the formulated problem. Furthermore, the system architecture is discussed and a new solution management strategy is proposed to limit the complexity of the algorithm. Simulation results show that the proposed approximation method maintains the nearest performance to the optimal solution.","Title: Analytical Derivation and Optimization of a Hybrid Unicast-Broadcast Network for Linear Services.\nYear: 2016\n\nAbstract: This paper provides a theoretical analysis of the performance of an hybrid network composed of a broadband system and a broadcast system which cooperate to deliver linear services on portable and mobile devices. First, we propose a global system model for the hybrid network in which the coverage area of the broadcast system is adjusted with respect to the spatial distribution of the users. Then based on the number of users requesting the services, the model parameters are optimized to increase the capacity of the hybrid network while guaranteeing a given quality of service. By applying convex optimization techniques, we derive a closed-form expression of the optimal coverage radius of the broadcast system that maximizes the capacity of the hybrid network. Next, based on this theoretical model, we analyze the performances of the optimized hybrid network. The results show that, depending on the number of users requesting the service, significant improvements of the capacity and quality of service are achieved with the hybrid network. Finally, regarding the planning of a cooperative network, this paper can be used to determine the optimal planning parameters of the broadcasting system for the delivery of linear services.","Title: HbbTV based Push-VOD services over DVB networks: Analysis and AL-FEC code application\nYear: 2016\n\nAbstract: Hybrid broadcast broadband TV (HbbTV) is an European initiative for providing viewers a richer experience of enhanced and interactive hybrid TV, through the launching of innovative hybrid services on both broadcast and broadband networks, allowing the viewer to interact with the current TV program. The latest release, HbbTV standard v2.0, provides additional features to HbbTV applications. Among the most important advancement of this version was the addition of the Push-VOD service. In this paper we focus on the broadcast method principle of Push-VOD services presented in the HbbTV v2.0. We investigate various technical aspects that have remained open questions in the standard to provide a first reference on the depth analysis of HbbTV based Push-VOD services. In addition, we propose and analyse a first solution for a reliable Push-VOD service delivery within file delivery protocol. This solution is based on the LDPC-Staircase AL-FEC codes using the hybrid (IT/ML) decoding method. Simulation results show that the merging of the repetition and the AL-FEC coding allows to inherit the benefits of both repetition and coding. A reliable file delivery is assured with low recovering complexity and download time.","Title: Performance analysis of the maximum ratio transmission preprocessing for extended receive antenna shift keying\nYear: 2017\n\nAbstract: The Extended Receive Antenna Shift Keying (ERASK) scheme is a MIMO scheme based on the Receive Spatial Modulation concept, invented to increase the overall spectral efficiency, by exploiting all possible combinations of receive antenna indexes. In this paper, we evaluated the ERASK scheme using the Maximum Ratio Transmission (MRT) preprocessing (MRT-ERASK), using the real amplitude threshold detector and compare it with the ERASK scheme using the Zero Forcing (ZF) preprocessing (ZF-ERASK). Analytical derivations of the received signal of the MRT-ERASK show that a complex inter-antenna interference is added to other antennas depending on the transmitted spatial symbol. The Bit Error Rate performance is also derived analytically. Simulation results over MIMO Rayleigh channel are provided to compare both systems, showing that ZF-ERASK outperforms MRT-ERASK but at the expense of a higher implementation complexity for ZF-ERASK. On the other hand, increasing the number of transmit antennas of a MRT-ERASK improves its performance getting closer to the performance ZF-ERASK. Therefore, the higher the number of transmit antennas, the nearer the performance of both systems, and the more suitable the MRT-ERASK to be implemented.","Title: A Low Complexity PAPR Reduction Algorithm for ATSC 3.0 Transmitters\nYear: 2018\n\nAbstract: Tone Reservation (TR) is one of the common algorithms proposed for peak-to-average power ratio (PAPR) reduction in next generation American digital video broadcasting (ATSC 3.0) transmitters. Recently, a quasi-optimal TR technique with a novel kernel definition named as grouped individual carrier multiple peaks (GICMP) scheme has been proposed. In GICMP, firstly, the reserved tones are partitioned into G groups, which also equals to the number of iterations. Then, a comb-like kernel is constructed with each group and it is adjusted to cancel a given set of peaks of data signal at each iteration. This scheme not only offers good performance/complexity trade-off but also compatible with ATSC 3.0 standard. Based on same kernel definition, we have proposed a new method, named as grouped carrier peak windowing (GCPW), which requires less hardware resources, compared to GICMP. Instead of choosing a set of peaks, this new method introduces a peak window to consider a given number of signal samples that exceed the predetermined threshold. This new method aims of reducing first peaks falling within the window, while leaving out higher peaks that may occur later. The simulation results show that the proposed low-complexity GCPW has less hardware memory requirements and with a good choice of threshold, window size and groups, offers quasi-optimal PAPR reduction performance same as that of GICMP.","Title: Mesh Hdr Wpan Resource Allocation Optimization Approaches\nYear: 2019\n\nAbstract: In the multihop IEEE 802.15.5 networks, all the devices compete to the resources of a shared superframe. In order to distribute these resources in a fair and satisfactory manner among the competing devices, we propose a distributed optimization framework for resource allocation scheme in an IEEE 802.15.5 hop-1. For this purpose, we introduce in this paper a suite of optimization problems for the hop-1 IEEE 802.15.5 resource allocation to optimize fairness and satisfaction without exceeding the superframe size and respecting the demanded channel time size sent by the requesting devices. Simulation results studied and compared the satisfaction factor and fairness index of the different proposed optimization problems. Consequently, a trade-off between satisfaction and fairness should be conducted for choosing the optimal solution.","Title: A Systematic Beam Broadening Method for Large Phased Arrays\nYear: 2021\n\nAbstract: Hybrid beamforming architecture is a key enabler for implementing Massive Multiple-Input Multiple-Output (mMIMO) solutions in next generation wireless communication systems. With such an approach, precoder and combiner designs not only depend on digital weights but also on modulus constraint analog weights controlled by phase shifters. This becomes a critical issue since the calculation of weights becomes a nonconvex problem that needs extensive search or complex iterative methods. In particular, beam broadening capability is of key interest for mMIMO systems to find a trade-off between the array gain and the beamwidth. Indeed, a high gain narrow beam is not always beneficial as far as mobility scenarios or control channel transmissions are concerned. In this paper, a non-iterative phase-only beam broadening method is proposed, relying on a quadratic phase excitation pattern applied to linear and planar arrays. We establish the key laws driving the beamwidth and steering angle adaption. We show that our technique can be applied systematically and is especially well-suited for large antenna arrays.","Title: Compression of Clipped OFDM IQ Samples for Cloud Radio Access Network\nYear: 2021\n\nAbstract: In this paper, we analyze the impact of reducing the Peak to Average Power Ratio (PAPR) of a time domain Orthogonal Frequency Division Multiplexing (OFDM) signal before its compression, as required for the transmission of IQ samples in a Centralized Radio Access Network architecture. The analysis is performed considering a clipping function for PAPR reduction. First, the distribution of the clipped OFDM IQ samples is obtained which in turn allows to derive the threshold and codebook levels of a non-uniform quantizer optimized for clipped IQ samples and the number of representation bits for each quantized levels in the entropy coding stage. Then, we provide a closed-form expression for the Modulation Error Ratio of the compressed and clipped OFDM signal. The simulation results show that the Gaussian quantizer applied to a clipped signal is not quite robust compared to a quantizer that has been optimized w.r.t. the distribution of clipped IQ samples, especially in the low-rate quantization region.","Title: Deep Learning for Location Based Beamforming with Nlos Channels\nYear: 2022\n\nAbstract: Massive MIMO systems are highly efficient but critically rely on accurate channel state information (CSI) at the base station in order to determine appropriate precoders. CSI acquisition requires sending pilot symbols which induce an important overhead. In this paper, a method whose objective is to determine an appropriate precoder from the knowledge of the user's location only is proposed. Such a way to determine precoders is known as location based beamforming. It allows to reduce or even eliminate the need for pilot symbols, depending on how the location is obtained. the proposed method learns a direct mapping from location to precoder in a supervised way. It involves a neural network with a specific structure based on random Fourier features allowing to learn functions containing high spatial frequencies. It is assessed empirically and yields promising results on realistic synthetic channels. As opposed to previously proposed methods, it allows to handle both line-of-sight (LOS) and non-line-of-sight (NLOS) channels.","Title: Analog Eigen-Beamforming for mmWave Systems: Performance under practical constraints\nYear: 2021\n\nAbstract: Many millimeter wave systems embed phased array antennas to overcome the large path loss that is experienced at such high frequencies. Digital Fourier Transform based beamforming (DFT-BF) is often implemented as a low-complexity and very limited feedback technique. Although DFT-BF is optimal in pure Line-of-Sight scenarios, it is substantially outperformed in richer scattering environments by more advanced approaches such as Eigenvector-based beamforming (Eigen-BF). The latter is harder to implement in practical scenarios due to its computational complexity and amount of feedback. In this paper, we propose several solutions to alleviate those drawbacks and analyze various performance-complexity trade-offs in the perspective of a practical usage of Eigen-BF solutions.","Title: Leveraging triplet loss and nonlinear dimensionality reduction for on-the-fly channel charting\nYear: 2022\n\nAbstract: Channel charting is an unsupervised learning method that aims at mapping wireless channels to a so-called chart, preserving as much as possible spatial neighborhoods. In this paper, a model-based deep learning approach to this problem is proposed. It builds on a physically motivated distance measure to structure and initialize a neural network that is subsequently trained using a triplet loss function. The proposed structure exhibits a low number of parameters and clever initialization leads to fast training. These two features make the proposed approach amenable to on-the-fly channel charting. The method is empirically assessed on realistic synthetic channels, yielding encouraging results.","Title: Analysis of Compressing PAPR-Reduced OFDM IQ Samples for Cloud Radio Access Network\nYear: 2022\n\nAbstract: A common problem of the virtualized cloud radio access network architecture (C-RAN) is the compression of the time-domain IQ samples before transmission over the fronthaul link. Considering a multicarrier waveform such as OFDM, whose IQ samples follow a quasi-Gaussian distribution, the conventional Gaussian quantizer may be used as the optimal solution to the compression problem. However, since the high peak-to-average power ratio (PAPR) of OFDM signals remains a serious problem, various techniques may be employed to reduce the time-domain fluctuations of the IQ samples in the OFDM, resulting in a change in its distribution. The latter fact makes the Gaussian quantizer suboptimal. The literature lacks a performance analysis of the conventional OFDM-based compression techniques when the PAPR of the OFDM signal is reduced. Therefore, in this paper, we study for the first time the impact of reducing the PAPR of the OFDM signal before compression in the C-RAN architecture through rate-distortion analysis. We consider clipping and tone reservation PAPR reduction algorithms. The former is the simplest PAPR reduction approach, while the latter is one of the most effective algorithms used in broadcasting standards such as DVB-T2 and ATSC 3.0. We first derive the distribution of the PAPR-reduced OFDM IQ samples. This is used to optimize the thresholds and codebook levels of a non-uniform scalar quantizer and the number of quantization bits allocated for each quantized level in the entropy coding stage, along with the MER performance analysis. The simulation results show that the conventional Gaussian-based compression techniques applied to a PAPR-reduced signal is not very robust to the statistical changes in the signal unless the signal distribution at the input of the Gaussian quantizer is not significantly affected. However, a significant gain is obtained when the quantizer is optimized with respect to the true distribution of the PAPR-reduced IQ samples.","Title: An Efficient Analog Eigen-Beamforming Procedure for Wideband mmWave MIMO-OFDM Systems\nYear: 2022\n\nAbstract: Millimeter wave communication systems define a new paradigm for wireless communications. At such high frequencies, multi-input multi-output (MIMO) processing is done on the radio-frequency signals to limit the hardware complexity and the energy consumption. Consequently, the channel experienced by the receiver directly depends on the transmitter and receiver antenna array responses, making the determination of the analog precoder and combiner an intertwined problem. In this paper, we propose a procedure capable of calculating analog weights at the transmitter and at the receiver, which allows the maximization of the wideband spectral efficiency for single stream transmission single user MIMO (SU-MIMO). While solely relying on partial channel state information (CSI), our procedure has close performance to full CSI-dependent schemes."],"string":"[\n \"Title: Enhanced mobile digital video broadcasting with distributed space-time coding\\nYear: 2012\\n\\nAbstract: This paper investigates the distributed space-time (ST) coding proposals for the future Digital Video Broadcasting-Next Generation Handheld (DVB-NGH) standard. We first theoretically show that the distributed MIMO scheme is the best broadcasting scenario in terms of channel capacity. Consequently we evaluate the performance of several ST coding proposals for DVB-NGH with practical system specifications and channel conditions. Simulation results demonstrate that the 3D code is the best ST coding solution for broadcasting in the distributed MIMO scenario.\",\n \"Title: A Combined Time and Frequency Algorithm for Improved Channel Estimation in TDS-OFDM\\nYear: 2010\\n\\nAbstract: In contrast to the classical cyclic prefix (CP)-OFDM, the time domain synchronous (TDS)-OFDM employs a known pseudo noise (PN) sequence as guard interval (GI). Conventional channel estimation methods for TDS-OFDM are only based on the PN sequence and consequently suffer from intersymbol interference. This paper proposes a novel two-stage channel estimation method which combines the estimation results from the PN sequence and, most importantly, the estimation results obtained from the OFDM data symbols. A simple feedback loop that excludes the channel decoder is employed for the OFDM data based estimation. The MMSE criteria is used to obtain the best combination results and an iterative process is proposed to progressively refine the estimation. Both MSE and BER simulations are carried out to prove the effectiveness of the proposed algorithm in the DTMB system which is based on TDS-OFDM signalling.\",\n \"Title: Improved channel estimation methods based on PN sequence for TDS-OFDM\\nYear: 2012\\n\\nAbstract: An accurate channel estimation is crucial for the novel time domain synchronous orthogonal frequency-division multiplexing (TDS-OFDM) scheme in which pseudo noise (PN) sequences serve as both guard intervals (GI) for OFDM data symbols and training sequences for synchronization/channel estimation. This paper studies the channel estimation method based on the cross-correlation of PN sequences. A theoretical analysis of this estimator is conducted and several improved estimators are then proposed to reduce the estimation error floor encountered by the PN-correlation-based estimator. It is shown through mathematical derivations and simulations that the new estimators approach or even achieve the Cramér-Rao bound.\",\n \"Title: Special Issue on MC-SS Loading algorithms for adaptive SS-MC-MA systems over wireline channels: comparison with DMT\\nYear: 2006\\n\\nAbstract: In this paper, we propose to combine adaptive loading principles with the spread-spectrum multicarrier multiple access (SS-MC-MA) scheme. Such an approach has particular interests in the context of powerline communications (PLC), where the transmitter has not only to exploit robust transmission techniques, but has also to adapt the waveform to the channel response. We introduce finite-granularity loading algorithms that dynamically handle the configuration of the system under power spectral density constraints. The presented algorithms assign subcarriers, spreading codes, bits and energy to each user in order to maximise either the data rate or the noise margin at a given target symbol error rate. These algorithms can actually be viewed as a widening of the classical waterfilling approach in the case of an hybrid spread-spectrum multicarrier system. Simulation results of the new scheme are presented for different measured PLC channels and are compared with those of the classical discrete multitone modulation (DMT) approach. It is shown that the adaptive SS-MC-MA scheme performs significantly better than DMT, due to its natural energy Gathering capability. Adaptive SS-MC-MA then leads to a more efficient bits and energies distribution and constitutes a simple solution to reduce the quantification loss induced by the use of finite order modulations. Copyright (c) 2006 AEIT.\",\n \"Title: Performance of time reversal precoding technique for MISO-OFDM systems.\\nYear: 2013\\n\\nAbstract: Time reversal (TR) is considered as a promising technique for green and multi-user communications thanks to its time and space focusing properties. TR can be viewed as a precoding scheme which can be combined with orthogonal frequency division multiplexing (OFDM) and easily carried out in a multiple transmit antenna context. This paper analyzes the performance of TR for a multiple-input single-output (MISO) OFDM system and provides a comparison with maximum ratio transmission (MRT) and equal gain transmission (EGT) precoding techniques. The analytical performance of the three precoding techniques is derived by computing the capacity and the bit error rate (BER) as a function of the transmit signal-to-noise ratio (SNR). First, the capacity analysis highlights the ability of the TR system to provide higher bit rates than the MRT system at low SNRs, while the capacity of the MRT system is the highest at high SNRs. From the obtained BER analytical expressions, the diversity exploitation of each system is discussed. In particular, it is shown that the TR system only exploits half the available diversity, while the systems using EGT or MRT exploit the full diversity. Hence, contrary to what is expected from the theoretical capacity analysis, the TR system is shown to underperform the other precoding schemes in terms of BER. To overcome such a drawback, the combination of TR with classical adaptive modulation techniques is studied, allowing the achievable throughput to be increased without destroying the focusing properties of TR. It is then observed that TR takes advantage of adaptive modulations and outperforms the other schemes at low SNRs. In this study, analytical results and closed-form expressions of capacity and BER performance are provided and confirmed through Monte Carlo simulations.\",\n \"Title: Adaptive Linear Precoded Dmt As An Efficient Resource Allocation Scheme For Power-Line Communications\\nYear: 2006\\n\\nAbstract: In this paper, we propose to apply adaptive loading principles to linear precoded digital multitone (LP-DMT). This new approach can especially be exploited in the powerline context, since it requires the knowledge of the channel at the transmitter. We first show that maximal waveform capacity is achieved when orthogonal matrices are used to linearly precode the multitone signal. A practical loading algorithm based on tone clustering is then developed to handle the configuration of the system. This algorithm assigns tones, size of precoding matrices, bits and energies in order to maximize the total throughput of the system. The optimization procedure is led under power spectral density (PSD) limitations and finite order modulations constraint. Through simulations over power line channels, it is shown that the new adaptive LP-DMT system takes advantage of the carrier merging effect offered by the precoding function and then outperforms the non-precoded DMT system.\",\n \"Title: Achieving Low-Complexity Maximum-Likelihood Detection for the 3D MIMO Code.\\nYear: 2014\\n\\nAbstract: The 3D Multiple-input multiple-output (MIMO) code is a robust and efficient space-time block code (STBC) for the distributed MIMO broadcasting but suffers from high maximum-likelihood (ML) decoding complexity. In this paper, we first analyze some properties of the 3D MIMO code to show that the 3D MIMO code is fast decodable. It is proven that the ML decoding performance can be achieved with a complexity of O(M4.5) instead of O(M8) in quasi-static channel with M-ary square QAM modulations. Consequently, we propose a simplified ML decoder exploiting the unique properties of the 3D MIMO code. Simulation results show that the proposed simplified ML decoder can achieve much lower processing time latency compared to the classical sphere decoder with Schnorr-Euchner enumeration.\",\n \"Title: Efficient Mimo-Ofdm Schemes For Future Terrestrial Digital Tv With Unequal Received Powers\\nYear: 2008\\n\\nAbstract: This article investigates the effect of equal and unequal received powers on the performances of different MIMO-OFOM schemes for terrestrial digital TV More precisely, we focus on three types of non-orthogonal schemes: the BLAST scheme, the Linear Dispersion (LD) code and the Golden code, and we compare their performances to that of Alamouti scheme. Using two receiving antennas, we show that for moderate attenuation on the second antenna and high spectral efficiency, Golden code outperforms other schemes. However, Alamouti scheme presents the best performance for low spectral efficiency and equal received powers or when one antenna is dramatically damaged When three antennas are used, we show that Golden code offers the highest robustness to power unbalance at the receiving side.\",\n \"Title: Channel Shortening For Bit Rate Maximization In Dmt Communication Systems\\nYear: 2010\\n\\nAbstract: This paper considers the problem of impulse response shortening for discrete multitone (DMT) transceivers. The proposed channel shortening filter shortens the effective channel impulse response (CIR) while concentrating energy in a predetermined window. This approach is based on convolution operation and filter coefficients decomposition which over constraint the effective CIR and allows achieving better performance. Simulation results show that the proposed algorithm outperforms the classical channel shortening approachs in terms of achievable bit rate and computational complexity.\",\n \"Title: Efficient 3D space time space block code for future terrestrial digital TV\\nYear: 2008\\n\\nAbstract: This article introduces a 3D space-time-space block code for future terrestrial digital TV in single frequency networks. The proposed 3D code is based on a double layer structure designed for inter-cell and intra-cell space time coded transmissions. We show that this new structure is particularly efficient for SFN environments regardless of the location of the receiver. It is then suitable for fixed, portable and mobile receptions.\",\n \"Title: Bit error rate prediction of coded MIMO-OFDM systems\\nYear: 2008\\n\\nAbstract: Bit error rate (BER) prediction over channel realisations has emerged as an active research area. In this paper, we give analytical signal to interference and noise ratio (SINR) evaluation of MIMO-OFDM systems using an iterative receiver. Using this analytical SINR expression, we propose an accurate BER prediction method based on effective exponential SINR mapping (EESM) method. We show by simulations that our method is independent of the channel realisation and of the MIMO scheme. It is only dependent on the modulation and coding scheme.\",\n \"Title: Time reversal applied to large MISO-OFDM systems.\\nYear: 2013\\n\\nAbstract: In this paper we study the Time Reversal (TR) technique in a large MISO-OFDM system. This technique has been proven to provide good beamforming gains at a price of a channel estimation availability at the transmitter side. Moreover, the greater the number of transmit antennas, the better the spatial and temporal focusing gains, which makes TR a good candidate for large MISO systems while keeping a low complexity. The capacity of the system and the bit error rate (BER) performances are first derived and simulated and compared with systems applying the maximum ratio transmission technique. The time focusing gain of the TR technique is then exploited to show that the equivalent channel is shortened when the number of antennas increases, allowing a reduction of the guard interval and an increase of the spectral efficiency of the system. Finally, the robustness of the system to channel estimation errors is evaluated by adding simple channel estimation techniques in the system and evaluating the BER performance.\",\n \"Title: A coded bit-loading linear precoded discrete multitone solution for power line communication\\nYear: 2008\\n\\nAbstract: Linear precoded discrete multitone modulation (LP-DMT) system has been already proved advantageous with adaptive resource allocation algorithm in a power line communication (PLC) context. In this paper, we investigate the bit and energy allocation algorithm of an adaptive LP-DMT system taking into account the channel coding scheme. A coded adaptive LP-DMT system is presented in the PLC context with a loading algorithm which accommodates the channel coding gains in bit and energy calculations. The performance of a concatenated channel coding scheme, consisting of an inner Weipsilas 4-dimensional 16-states trellis code and an outer Reed-Solomon code, in combination with the proposed algorithm is analyzed. Simulation results are presented for a fixed target bit error rate in a multicarrier scenario under power spectral density constraint. Using a multipath model of PLC channel, it is shown that the proposed coded adaptive LP-DMT system performs better than classical coded discrete multitone.\",\n \"Title: Reduced-complexity maximum-likelihood decoding for 3D MIMO code\\nYear: 2014\\n\\nAbstract: The 3D MIMO code is a robust and efficient spacetime coding scheme for the distributed MIMO broadcasting. However, it suffers from the high computational complexity if the optimal maximum-likelihood (ML) decoding is used. In this paper we first investigate the unique properties of the 3D MIMO code and consequently propose a simplified decoding algorithm without sacrificing the ML optimality. Analysis shows that the decoding complexity is reduced from O(M8) to O(M4.5) in quasi-static channels when M-ary square QAM constellation is used. Moreover, we propose an efficient implementation of the simplified ML decoder which achieves a much lower decoding time delay compared to the classical sphere decoder with Schnorr-Euchner enumeration.\",\n \"Title: Dynamic Cross-Layer Spectrum Allocation for Multi-Band High-Rate UWB Systems\\nYear: 2009\\n\\nAbstract: In this paper, we investigate a new approach for the spectrum allocation in UWB systems. This approach consists in a cross-layer scheme that takes into consideration the different users channel quality and quality of service (QoS) requirements. The new scheme is based on the WiMedia solution proposed for multiband OFDM UWB systems. The main objective is to propose a low-complexity solution for the spectrum allocation that can manage all the users constraints. Thus, we study the optimal solution of the spectrum allocation and formulate it as a convex optimization problem. Then, we show that our proposed scheme reduces significantly the complexity of the optimal solution. Moreover, we show through simulations, that the new approach and the optimal solution have close performance in term of error rate and they outperform WiMedia solution proposed for UWB systems.\",\n \"Title: A Novel Data-Aided Channel Estimation with Reduced Complexity for TDS-OFDM Systems\\nYear: 2012\\n\\nAbstract: In contrast to the classical cyclic prefix (CP)-OFDM, the time domain synchronous (TDS)-OFDM employs a known pseudo noise (PN) sequence as guard interval (GI). Conventional channel estimation methods for TDS-OFDM are based on the exploitation of the PN sequence and consequently suffer from intersymbol interference (ISI). This paper proposes a novel data-aided channel estimation method which combin...\",\n \"Title: Multi-user cross-layer allocation design for LP-OFDM high data rate UWB systems\\nYear: 2008\\n\\nAbstract: In this paper, we investigate a cross-layer design for the packet scheduling and the resource allocation in UWB systems. This design considers the combination of queuing and channel state information (CSI) which provides QoS support for multimedia applications in UWB. For the physical layer, the use of a linear precoded orthogonal division multiplexing (LP-OFDM) waveform is proposed because of its significant performance increase compared to the WiMedia proposal. For the medium access control layer, scheduling is performed in order to differentiate between the different users and to satisfy their quality of service constraints. This cross-layer approach optimizes the system spectral efficiency and solves the problem in the WiMedia solution of cohabitation of more than three users sharing the three sub-bands of the same channel. Simulation results show that the proposed scheme leads to a considerable improvement in resource allocation and can guarantee the required quality of service.\",\n \"Title: A fast decodable full-rate STBC with high coding gain for 4 × 2 MIMO systems.\\nYear: 2013\\n\\nAbstract: In this work, a new fast-decodable space-time block code (STBC) is proposed. The code is full-rate and full-diversity for 4 × 2 multiple-input multiple-output (MIMO) transmission. Due to the unique structure of the codeword, the proposed code requires a much lower computational complexity to provide maximum-likelihood (ML) decoding performance. It is shown that the ML decoding complexity is only O(M4,5) when M-ary square QAM constellation is used. Finally, the proposed code has highest minimum determinant among the fast-decodable STBCs known in the literature. Simulation results prove that the proposed code provides the best bit error rate (BER) performance among the state-of-the-art STBCs.\",\n \"Title: Multiuser Service Differentiated Spectrum Allocation Scheme for High Rate UWB Systems\\nYear: 2010\\n\\nAbstract: In this paper, we propose a multiuser spectrum allocation scheme for high-rate UWB systems under QoS requirements. This scheme comes as a solution to the coexistence of multiple users sharing the three sub-bands of the same channel as defined in the WiMedia solution adopted for multiband UWB systems. Indeed, WiMedia solution does not allow more than three users to coexist in the same channel. Based on a constrained multiuser optimization problem, the proposed allocation algorithm allows multiple users to access the medium following a mixed sub-band assignment and priority-based scheduling approach in order to ensure an efficient differentiated spectrum sharing. The resulting time-frequency scheduling algorithm relies on the combination of two main metrics available at PHY and MAC levels: the channel quality of each user provided by the exploitation of the effective SINR method, and the QoS constraint represented by a simple weighting parameter that differentiates between two service classes. Simulation results show the efficiency of the proposed scheme and how it guarantees a good performance level for users having strict QoS requirements.\",\n \"Title: On the hybrid use of unicast/broadcast networks under energy criterion\\nYear: 2012\\n\\nAbstract: Mobile network operators are confronted to an exponential growth of their traffic. One of the causes of this growth is the delivery of video services, including mobile television. Using a broadcast component like MBMS or DVB-H for the provision of the most popular mobile television channels can minimize their impact on the mobile network traffic. Such a component can also be investigated for other types of popular services and open the door to hybrid broadcast/unicast networks. The key factor for the efficiency of this approach is the choice of the services to be transmitted through the broadcast component. In this article, we introduce an energy criterion to make this choice. We use a simplified model of an hybrid network combining a DVB-T2 broadcast component to a LTE unicast component. Through the statistical study of the reception conditions of the users in this network, we evaluate the power gain brought by the use of the broadcast component for the delivery of a service as a function of the number of users that are using this service.\",\n \"Title: Coded Adaptive Linear Precoded Discrete Multitone Over Plc Channel\\nYear: 2008\\n\\nAbstract: Discrete multitone modulation (DMT) systems exploit the capabilities of orthogonal subcarriers to cope efficiently with narrowband interference, high frequency attenuations and multipath fadings with the help of simple equalization filters. Adaptive linear precoded discrete multitone (LP-DMT) system is based on classical DMT, combined with a linear precoding component. In this paper, we investigate the bit and energy allocation algorithm of an adaptive LP-DMT system taking into account the channel coding scheme. A coded adaptive LP-DMT system is presented in the power line communication (PLC) context with a loading algorithm which accommodates the channel coding gains in bit and energy calculations. The performance of a concatenated channel coding scheme, consisting of an inner Wei's 4-dimensional 16-states trellis code and an outer Reed-Solomon code, in combination with the proposed algorithm is analyzed. Theoretical coding gains are derived and simulation results are presented for a fixed target bit error rate in a multicarrier scenario under power spectral density constraint. Using a multipath model of PLC channel, it is shown that the proposed coded adaptive LP-DMT system performs better than coded DMT and can achieve higher throughput for PLC applications.\",\n \"Title: Cross-layer resource allocation scheme for multi-band high rate UWB systems\\nYear: 2009\\n\\nAbstract: In this paper, we investigate the use of a cross-layer allocation mechanism for the high-rate ultra-wideband (UWB) systems. The aim of this paper is twofold. First, through the cross-layer approach that provides a new service differentiation approach to the fully distributed UWB systems, we support traffic with quality of service (QoS) guarantee in a multi-user context. Second, we exploit the effective SINR method that represents the characteristics of multiple sub-carrier SINRs in the multi-band WiMedia solution proposed for UWB systems, in order to provide the channel state information needed for the multi-user sub-band allocation. This new approach improves the system performance and optimizes the spectrum utilization with a low cost data exchange between the different users while guaranteeing the required QoS. In addition, this new approach solves the problem of the cohabitation of more than three users in the same WiMedia channel.\",\n \"Title: On The Influence Of Carrier Frequency Offset And Sampling Frequency Offset In Mimo-Ofdm Systems For Future Digital Tv\\nYear: 2008\\n\\nAbstract: This paper investigates the impact of carrier frequency offset (CFO) and sampling frequency offset (SFO) on the performance of different MIMO-OFDM schemes with high spectral efficiency for next generation of terrestrial digital TV. We analyze particularly orthogonal Alamouti scheme, and non-orthogonal (NO) schemes like VBLAST, linear dispersion (LD) code and Golden code. This analysis gives a global view on the best suitable MIMO-OFDM scheme with respect to CFO and SFO. We show that for high spectral efficiency, Alamouti is more sensitive to CFO and SFO. Moreover, we show that all studied MIMO-OFDM schemes are sensitive to CFO when it is greater than 1% of inter-carrier spacing. Their sensitivity to the SFO is less than to their sensitivity to CFO.\",\n \"Title: Time-frequency spectrum sharing scheme for next generation OFDM systems\\nYear: 2010\\n\\nAbstract: Spectrum sharing and multiple-access techniques are challenging matters for future wireless communication systems. Orthogonal frequency division multiplexing (OFDM) has been considered as the most important physical layer candidate for the next generation wireless systems. Alternatively, multiband OFDM (MB-OFDM) approach has shown some novelty and efficiency for high-rate (HR) WPAN applications and so has been selected for future HR ultra wideband (UWB) systems. Based on an optimization problem formulation, we jointly consider resource allocation and scheduling to define a novel dynamic timefrequency spectrum sharing mechanism based on the MB-OFDM approach. Viewed as a cross-layer solution, the proposed MB-OFDM multiple-access (MB-OFDM-MA) scheme can achieve an efficient multiuser spectrum sharing under QoS constraints. While the optimization formulation results in a high-complexity time-frequency sharing solution, we propose a low-complexity suboptimal solution able to jointly provide fairness provision with the QoS support. Compared with traditional orthogonal frequency divisionmultiple access (OFDMA), the simple time-frequency MB-OFDM-MA approach ensures a high level of fairness among the users while satisfying high-priority users having strict QoS requirements.\",\n \"Title: Resource Allocation with Adaptive Spread Spectrum OFDM Using 2D Spreading for Power Line Communications\\nYear: 2007\\n\\nAbstract: Bit-loading techniques based on orthogonal frequency division mutiplexing (OFDM) are frequently used over wireline channels. In the power line context, channel state information can reasonably be obtained at both transmitter and receiver sides, and adaptive loading can advantageously be carried out. In this paper, we propose to apply loading principles to an spread spectrum OFDM (SS-OFDM) waveform which is a multicarrier system using 2D spreading in the time and frequency domains. The presented algorithm handles the subcarriers, spreading codes, bits and energies assignment in order to maximize the data rate and the range of the communication system. The optimization is realized at a target symbol error rate and under spectral mask constraint as usually imposed. The analytical study shows that the merging principle realized by the spreading code improves the rate and the range of the discrete multitone (DMT) system in single and multiuser contexts. Simulations have been run over measured power line communication (PLC) channel responses and highlight that the proposed system is all the more interesting than the received signal-to-noise ratio (SNR) is low.\",\n \"Title: Interference-Aware Prioritized Spectrum Scheduling For Mb-Ofdm Systems\\nYear: 2010\\n\\nAbstract: In this paper, we investigate the spectrum scheduling problem for next generation multiband orthogonal frequency division multiplexing (MB-OFDM) ultra-wideband (UWB) systems under QoS and interference constraints. Based on a constrained optimization problem formulation, we propose a cross-layer scheduling solution that takes into account the unlicensed UWB users channel and QoS conditions as well as the interference impact on primary users sharing the same bandwidth. The cross-layer scheduling scheme is based on a priority approach that gives advantage to users having strict QoS requirements while avoiding or reducing the interference effect on primary users. Simulation results demonstrate the efficiency of the proposed scheme in terms of QoS support and interference reduction.\",\n \"Title: Spectrum sharing optimization model for multiuser linear-precoded OFDM UWB systems\\nYear: 2012\\n\\nAbstract: In this paper, we investigate spectrum sharing and multiple access problems for the high-rate ultra-wideband (UWB) systems. Based on the use of the linear precoded orthogonal frequency division multiplexing (LP-OFDM) proposed as an evolution of the well known multi-band OFDM (MB-OFDM) solution supported by the WiMedia Alliance for future high-rate UWB systems, the objective of this paper is to study a multiuser optimization spectrum sharing scheme for LP-OFDM UWB systems. This optimization study leads to a novel multiple access solution for the high-rate UWB systems. Simulation results demonstrate the efficiency of the use of the LP-OFDM transmission technique in the multiuser spectrum sharing scheme. Besides, the novel LP-OFDM multiuser spectrum sharing solution shows its capacity to provide a good tradeoff between the QoS support of the high-priority users and the fairness between the competing users.\",\n \"Title: Robustness of MIMO-OFDM schemes for future digital TV to carrier frequency offset\\nYear: 2008\\n\\nAbstract: This paper investigates the impact of carrier frequency offset (CFO) on the performance of different MIMO-OFDM schemes with high spectral efficiency for next generation of terrestrial digital TV. We show that all studied MIMO-OFDM schemes are sensitive to CFO when it is greater than 1% of inter- carrier spacing. We show also that the Alamouti scheme is the most sensitive MIMO scheme to CFO.\",\n \"Title: System Level Evaluation Of Innovative Coded Mimo-Ofdm Systems For Broadcasting Digital Tv\\nYear: 2008\\n\\nAbstract: Single-frequency networks (SFNs) for broadcasting digital TV is a topic of theoretical and practical interest for future broadcasting systems. Although progress has been made in the characterization of its description, there are still considerable gaps in its deployment with MIMO technique. The contribution of this paper is multifold. First, we investigate the possibility of applying a space-time (ST) encoder between the antennas of two sites in SFN. Then, we introduce a 3D space-time-space block code for future terrestrial digital TV in SFN architecture. The proposed 3D code is based on a double-layer structure designed for intercell and intracell space time-coded transmissions. Eventually, we propose to adapt a technique called effective exponential signal-to-noise ratio (SNR) mapping (EESM) to predict the bit error rate (BER) at the output of the channel decoder in the MIMO systems. The EESM technique as well as the simulations results will be used to doubly check the efficiency of our 3D code. This efficiency is obtained for equal and unequal received powers whatever is the location of the receiver by adequately combining ST codes. The 3D code is then a very promising candidate for SFN architecture with MIMO transmission. Copyright (C) 2008 Y. Nasser et al.\",\n \"Title: Resource Allocation in High Data Rate Mesh WPAN: A Survey Paper\\nYear: 2014\\n\\nAbstract: IEEE 802.15.3 High data rate wireless personal area networks (HDR WPANs) have been developed to communicate with devices within 10 m at high speed. A mesh network made up of a parent piconet and several child piconets can support multi-hop communications. Wireless mesh networks (WMNs) have been expected to be the ultimate solution for the next decade wireless networking, showing rapid progress and many new inspiring applications. The international standardization organizations formed working groups to address the problem of standardization for WMNs. These groups were the IEEE 802.15.5 (mesh extensions for WPANs),the IEEE 802.11s (mesh extensions for WLANs), and the IEEE 802.16a (mesh extensions for WiMAX). The IEEE 802.15.5 standard is the standard which defines specifications for including multi-hop functionality in the legacy 802.15.3 and 802.15.4 low data rate (LDR) WPAN networks. The impetus for a WPAN to operate in a mesh topology is to increase the network coverage without increasing the transmit power, to increase the route reliability via route redundancy, self-configuration, and efficient use of device battery life. In the case of meshed WPANs, multiple WPAN clusters compete for channel time in a shared superframe. Therefore, it is essential to determine the channel time requirements of each cluster with a certain number of devices and to determine how these clusters can compete to the shared channel time. In this paper, we investigate the different resource allocation mechanisms related to the meshed HDR WPANs for the 802.15.3 and the 802.15.5 standards. We introduce the single hop and the multi-hop IEEE 802.15.3 WPAN architectures. This is followed by the introduction of the IEEE 802.15.5 standard that provides the mesh capabilities for extending the coverage area of HDR WPANs. The current on-going research issues for resource allocation, including beacon interference, reservation collision etc., in both meshed 802.15.3 and 802.15.5 are alluded to.\",\n \"Title: Analysis And Performance Comparison Of Dvb-T And Dtmb Systems For Terrestrial Digital Tv\\nYear: 2008\\n\\nAbstract: Orthogonal frequency-division multiplexing (OFDM) is the most popular transmission technology in digital terrestrial broadcasting (DTTB), adopted by many DTTB standards. In this paper, the bit error rate (BER) performance of two DTTB systems, namely cyclic prefix OFDM (CP-OFDM) based DVB-T and time domain synchronous OFDM (TDS-OFDM) based DTMB, is evaluated in different channel conditions. Spectrum utilization and power efficiency are also discussed to demonstrate the transmission overhead of both systems. Simulation results show that the performances of the two systems are much close. Given the same ratio of guard interval (GI), the DVB-T outperforms DTMB in terms of signal to noise ratio (SNR) in Gaussian and Ricean channels, while DTMB behaves better performance in Rayleigh channel in higher code rates and higher orders of constellation thanks to its efficient channel coding and interleaving scheme.\",\n \"Title: Cross-layer multiple-access optimization scheme for linear precoded OFDM UWB systems.\\nYear: 2013\\n\\nAbstract: With the growing demand for new wireless applications accompanied with high expectations for better QoS fulfillment especially for multimedia and real-time applications, the performance of the radio resource management in a multiuser context is ensured by the ability to provide an efficient and optimized spectrum sharing scheme that should respect the wireless channel conditions and satisfy the different users’ demands. From the physical layer perspective, metrics such as spectrum efficiency and minimum BER are the most important criteria to be considered. On the other hand, from a user perspective, QoS as well as fairness among the competing users are the main metrics because they determine how much end-users are satisfied and how efficient the available resources are shared among the existing users. Based on the use of the linear precoded orthogonal frequency division multiplexing (LP-OFDM) solution proposed as an evolution of the well-known multiband OFDM (MB-OFDM) solution supported by the WiMedia Alliance for future high-rate ultra-wideband (UWB) systems, the objective of this paper is twofold. First, we study a multiuser optimization spectrum sharing scheme for LP-OFDM systems. Second, based on the optimization study, we define a novel multiple-access solution which jointly considers the frequency resource allocation and the time scheduling for the high-rate LP-OFDM UWB systems. Simulation results demonstrate the efficiency of the use of the LP-OFDM transmission technique in the multiuser spectrum sharing scheme. Besides, the novel multiuser time–frequency sharing scheme shows its capacity to provide a high performance level for high-priority users.\",\n \"Title: Spectral efficiency optimization in overlapping channels using TR-MISO systems\\nYear: 2013\\n\\nAbstract: Traditionally, cellular radio communication systems define a spectrum mask to avoid inter-channel interference between services operated in adjacent frequency bands. In this paper, we intend to optimize the spectrum efficiency of wireless systems by frequency band reused, which is performed by partial channel overlapping of adjacent bands. The impact of interchannel interference is studied for both single and multicarrier modulations. Then time reversal (TR) with multiple-input-single-output (MISO) is applied to depress the inter-channel interference thanks to the TR focusing properties. The spectral efficiency of the system is evaluated according to the overlap ratio τ and the number of transmitting antennas. The optimal τmax is identified and, as a result, channel overlapping is proved to be interesting at low SNR regime in a range depending on the number of antennas.\",\n \"Title: Distributed MIMO coding scheme with low decoding complexity for future mobile TV broadcasting\\nYear: 2012\\n\\nAbstract: A novel distributed space-time block code (STBC) for the next generation mobile TV broadcasting is proposed. The new code provides efficient performance within a wide range of power imbalance showing strong adaptivity to the single frequency network broadcasting deployments. The new code outperforms existing STBCs with equivalent decoding complexity and approaches those with much higher complexities.\",\n \"Title: Adaptive Self-Learning Resource Allocation Scheme for Unlicensed Users in High-Rate UWB Systems\\nYear: 2011\\n\\nAbstract: In this paper, we investigate a dynamic sub-band allocation for spectrum sharing in the multiuser OFDM-UWB systems while respecting the system constraints and the distributed MAC architecture. The contribution of this work is twofold: first we propose a dynamic spectrum allocation scheme based on an analytical study by deriving a multiuser optimization problem to find the optimal allocation solution. Based on an adaptive and low-complexity approach, the proposed scheme shares the available resources among unlicensed UWB users while taking into consideration the QoS requirements and the channel conditions. Then we extend the proposed scheme to allow the coexistence of a maximum number of secondary users in one channel while maintaining all the constraints support. This leads to a time-frequency allocation approach that exploits information laying in different layers, allowing a cross-layer design. This new approach improves the system performance compared to WiMedia solution and offers a considerable gain for users that have strict QoS requirements.\",\n \"Title: Efficient space code block code MIMO channel estimation for future mobile video broadcasting\\nYear: 2009\\n\\nAbstract: In this paper, we propose a novel and simple MIMO channel estimation technique based on 2D spread pilots. Taking advantage of the simplicity and flexibility of 2D spread pilot channel estimation, we easily combine the Alamouti code with the Walsh-Hadamard codes to obtain a space code block code (SCBC) MIMO scheme. We demonstrate that our proposed system is much more efficient in high mobility scenario than the classical STBC scheme. The performance evaluated over a realistic channel model shows the efficiency of this technique which turns out to be a promising MIMO channel estimation technique for the future mobile video broadcasting systems.\",\n \"Title: Effect of relaying on coverage in realistic deployment scenarios\\nYear: 2012\\n\\nAbstract: In this paper, we investigate the impact of fixed relay deployment on coverage in a wireless infrastructure. We consider a realistic scenario with OFDM based transmission which takes into account some practical constraints and assumptions (path-loss models, power, etc...). We analyze and compare scenarios with relays which operate in both half-duplex and full-duplex modes. Relays are assumed to perform decode-and-forward strategy. We show that for half-duplex scenarios, multiple relays can be deployed to extend the coverage and to compensate the loss of the half-duplex constraint where SFN-like transmission is used at the relays, i.e, the relays share the same subcarriers. Full-duplex mode outperforms the half-duplex mode in terms of coverage on the expense of additional complexity at the receiver side and offers a more convenient solution when the relay transmission power is constrained. The main contribution of this paper is giving guidelines for relay deployment in practical environments, and providing insight into benefits relays can offer in terms of coverage extension in realistic deployment scenarios.\",\n \"Title: Optimization of the energy efficiency of a hybrid broadcast/unicast network\\nYear: 2013\\n\\nAbstract: In the last few years, mobile broadband communications operators have been confronted to an exponential growth of the traffic over their networks. Recently, the use of a broadcast component as a solution to face this growth has received a particular attention. In [1], a model for a hybrid network is introduced. This network consists in the inter-operation of a LTE unicast component and a DVB-T2 broadcast component. The introduced model allows the evaluation of the potential energy gain brought by such a network. In this paper, using an enhanced version of the model previously defined, we propose a new operating mode for the hybrid network that can improve the energy efficiency of the system.\",\n \"Title: A Joint Channel and Carrier Frequency Offset Estimation Based on Spread Pilot for Future Broadcasting Systems\\nYear: 2009\\n\\nAbstract: In this paper, we propose a novel and simple joint channel frequency response (CFR) and carrier frequency offset (CFO) estimation based on 2D spread pilots for digital video broadcasting (DVB) systems. The advantage of the proposed algorithm compared to DVB-T/H standard, is to estimate the CFR and the CFO using the same pilot symbols. The performance evaluated over a realistic channel model, in a mobility scenario, in the presence of CFO, shows the efficiency of this technique which turns out to be a promising joint CFR and CFO estimation technique for the future DVB systems.\",\n \"Title: Cross-Layer Resource Allocation Scheme Under Heterogeneous Constraints for Next Generation High Rate WPAN\\nYear: 2010\\n\\nAbstract: In the next generation wireless networks, the growing demand for new wireless applications is accompanied with high expectations for better quality of service (QoS) fulfillment especially for multimedia applications. Furthermore, the coexistence of future unlicensed users with existing licensed users is becoming a challenging task in the next generation communication systems to overcome the underutilization of the spectrum. A QoS and interference aware resource allocation is thus of special interest in order to respond to the heterogeneous constraints of the next generation networks. In this work, we address the issue of resource allocation under heterogeneous constraints for unlicensed multi- band ultra-wideband (UWB) systems in the context of Future Home Networks, i.e. the wireless personal area network (WPAN). The problem is first studied analytically using a heterogeneous constrained optimization problem formulation. After studying the characteristics of the optimal solution, we propose a low-complexity suboptimal algorithm based on a cross-layer approach that combines information provided by the PHY and MAC layers. While the PHY layer is responsible for providing the channel quality of the unlicensed UWB users as well as their interference power that they cause on licensed users, the MAC layer is responsible for classifying the unlicensed users using a two-class based approach that guarantees for multimedia services a high-priority level compared to other services. Combined in an efficient and simple way, the PHY and MAC information present the key elements of the aimed resource allocation. Simulation results demonstrate that the proposed scheme provides a good tradeoff between the QoS satisfaction of the unlicensed applications with hard QoS requirements and the limitation of the interference affecting the licensed users.\",\n \"Title: A Unified Broadcast Layer for Horizon 2020 Delivery of Multimedia Services.\\nYear: 2014\\n\\nAbstract: It is expected that, in the coming years, the video data traffic carried by mobile networks will increase drastically. Among the different ways of facing this mobile data tsunami, the cooperation of several complementary access technologies offers promising prospects. In the context of such a cooperative approach, this paper proposes the definition of a unified broadcast layer, close to the 3GPP LTE technology, and therefore liable to be easily integrated in mobile devices and infrastructure equipment. The general technical requirements for the design of such a common physical layer are detailed and a practical example based on the 3GPP LTE/E-MBMS and DVB-T2 standards is described. The verification of the proposed approach is performed through simulation results and validation tests carried out on a hardware platform.\",\n \"Title: An Ofdm-Cdma Scheme For High Data Rate Uwb Applications\\nYear: 2007\\n\\nAbstract: In this paper, we investigate a new waveform for UWB systems obtained by the combination of Orthogonal Frequency Division Multiplex (OFDM) and Coded Division Multiple Access (CDMA). The proposed system, called Spread Spectrum - Multi-Carrier - Multiple Access (SS-MC-MA) turns out to be a judicious solution to combat frequency selectivity and narrowband interferers, and to manage the coexistence of several users and piconets. It is shown that the addition of a degree of freedom brought by the spreading component of SS-MC-MA allows to optimize jointly the assignment of the number of used codes and coding rates in order to make the system more robust. Through simulations, it is demonstrated that the new system can outperform Multi-Band OFDM Alliance (MBOA) for low data rates and is able to provide wider range of rates.\",\n \"Title: 3D MIMO Scheme for Broadcasting Future Digital TV in Single Frequency Networks\\nYear: 2008\\n\\nAbstract: A three-dimensional space-time-space block code for future terrestrial digital TV systems is introduced. The code is based on a double-layer structure designed for inter-cell and intra-cell transmissions in single-frequency networks. Without increasing the complexity of the receiver, the proposed code is very efficient to cope with equal and unequal received powers in single-frequency network scenarios.\",\n \"Title: Optimization of linear precoded OFDM for high-data-rate UWB systems\\nYear: 2008\\n\\nAbstract: We investigate the use of a linear precoded orthogonal frequency division multiplexing (LP-OFDM) waveform for high-data-rate ultra-wideband (UWB) systems. This waveform applied for the first time to UWB applications is an evolution of the multiband OFDM (MB-OFDM) solution supported by WiMedia-MBOA (MultiBand OFDM Alliance). The aim of this paper is twofold. Firstly, an analytical study of the LP-OFDM waveform allows to find how to efficiently precode an OFDM signal in order to improve the robustness of the system. Secondly, a global system study is led to highlight the benefits of adding a precoding function to an OFDM signal in the UWB context. Different system choices and parameterization strategies are thus proposed. In both analytical and global system studies, the LP component is optimized without channel state information (CSI) at the transmitter side, as in the MBOA solution. To go further, the MBOA constraints are relaxed, and an additional optimization is performed, with a CSI at the transmitter. The analytical and simulation results show that the joint use of linear precoding and OFDM leads to a significant performance increase compared to the MBOA solution. This improvement is due to the precoding scheme that provides better exploitation of the channel diversity.\",\n \"Title: Admission control and resource allocation strategies for IEEE 802.15.5.\\nYear: 2014\\n\\nAbstract: In this paper, a novel two-tier mechanism for the access control and resource allocation of a two-hop high data rate IEEE 802.15.5 network is proposed. One of the main contributions in this proposal is the use of a dynamic superframe size in the IEEE 802.15.5. Additionally, the ideas of using a superframe utilization threshold and a channel time allocation period (CTAP) utilization threshold are incorporated, with the aim of allowing more devices to access the shared superframe. Using these thresholds, two new hop-1 resource allocation algorithms are described; CTAP utilization threshold-based resource allocation algorithm (CTRA) and superframe utilization threshold-based resource allocation algorithm (STRA). These algorithms control the access of the hop-1 Mesh PicoNet Coordinators (MPNCs) and Mesh DEVices (MDEVs) to the channel time resources in the superframe. Furthermore, an algorithm for the distribution of the hop-1 MPNCs' resources among the real-time (RT) and non-real-time (NRT) MDEVs at hop-2 is introduced. In this mechanism, a new flag metric is applied to provide higher priorities to MPNCs over MDEVs in the hop-1 algorithms and to the RT MDEVs over the NRT MDEVs in the hop-2 algorithm. Simulation results show the superiority of the proposed mechanism over others by providing a higher satisfaction factor and higher fairness among the competing devices at both hop-1 and hop-2 of the IEEE 802.15.5.\",\n \"Title: Analysis of Decoding Failures of DVB-S2 LDPC Codes\\nYear: 2014\\n\\nAbstract: The DVB-S2 standard has been revealed a reliable standard for data transmission thanks to the LDPC codes. The waterfall of the bit-error rate is very abrupt to reach low values of order 10^(-10) in the error floor. In this paper, we propose a novel investigation on the iterative decoding of these codes by the Belief Propagation algorithm on Gaussian channels. Introducing definitions of persistent and created errors, we analyze them according to the degrees of the Tanner graph nodes and the noise power of the channel. We show that the two-degree bits used to accelerate the encoding process of the DVB-S2 are responsible for most decoding errors. This result could help design DVB-S2 codes in a more appropriate way, for instance avoiding the use of BCH codes to tackle the error floor.\",\n \"Title: Partial channel overlay in moderate-scale MIMO systems using WH precoded OFDM\\nYear: 2014\\n\\nAbstract: In this paper, we propose a multi-user orthogonal frequency-division multiplexing (OFDM) transmission scheme with partial channel reusing between users, suitable for multiple-input multiple-output (MIMO) systems with not-so-large antenna arrays. The spectrum efficiency is improved at the price of adding multiuser interference (MUI) which can be overcome by the multiplexing gain of MIMO. In addition, we propose to employ the spread spectrum OFDM (SS-OFDM) modulation with the Walsh-Hadamard (WH) codebook for such a multi-user system to further mitigate the MUI. The bit error rate (BER) performance of the proposed scheme in the additive white Gaussian noise (AWGN) channels is derived analytically and then verified through Monte-Carlo simulations. Moreover, the proposed SS-OFDM system with partial bandwidth reuse is compared with the OFDM-based solution in a moderate-scale MIMO system with equal gain combining (EGC) precoding. It is shown that the proposed SS-OFDM MIMO system outperforms the classical OFDM approach in frequency-selective fading channels when the number of transmit antennas remains moderate.\",\n \"Title: Distributed MIMO schemes for the future digital video broadcasting\\nYear: 2013\\n\\nAbstract: This paper studies the application of distributed multiple-input multiple-output (MIMO), i.e. MIMO transmission over several geographically separated but cooperated transmitters, for future TV broadcasting systems. It is first shown that distributed MIMO is promising for the future broadcasting systems from a channel capacity perspective. Several STBCs that can be applied in the distributed MIMO broadcasting scenarios are then discussed. Through performance comparison and complexity analyses with realistic system settings and channel model, it can be concluded that simple STBCs are efficient for low data rate applications, while the sophisticated ones are more suitable to deliver high data rate services.\",\n \"Title: EVM Derivation for Multicarrier Signals: Joint Impact of Non-Linear Amplification and Predistortion\\nYear: 2015\\n\\nAbstract: Orthogonal Frequency Division Multiplexing (OFDM) systems are extremely sensitive to nonlinear distortions brought by the Power Amplifier (PA) because of the inherent high peak-to-average power ratio (PAPR) of the signal. In this paper, we analyze the impact of the non-linearities on the output signal, assuming that the transmitter is equipped with a linearization stage before the PA. Therefore, the expression of the Error Vector Magnitude (EVM) of the output signal is analytically derived with or without the use of the linearization process. We provide analytical expressions of the EVM in the form of a series expansion that depends on the transition factors of the PA and predistortion characteristics. Simulations results confirm the accuracy of our analytical expressions, even when limiting the series expansion to an order of 4.\",\n \"Title: Tone Reservation Based PAPR Reduction Technique with Individual Carrier Power Allocation for Multiple Peaks Reduction\\nYear: 2015\\n\\nAbstract: Multiple algorithms have been proposed to deal with the high Peak to Average Power Ratio (PAPR) of Orthogonal Frequency Division Multiplexing (OFDM) signals. The second generation of Digital Video Broadcasting for Terrestrial transmission (DVB-T2) has adopted the Tone Reservation (TR) technique. It is based on an iterative process where, at each iteration, a predefined kernel is used to reduce one peak in time domain. In this paper, we first assess the performance of the power control schemes adopted by TR algorithms. Then, we analyze their complexity and show that their kernels are complex to generate, hence practically, difficult to implement. Secondly, we propose a novel kernel signal, simple to implement yet compatible with the DVB-T2 standard to reduce the PAPR value. The proposed kernel is defined to deal with the reduction of multiple peaks at a time. We show that the proposed PAPR reduction outperforms the TR algorithm adopted in DVB-T2 in terms of CCDF and Modulation Error Rate (MER) and then offers a better performance-complexity tradeoff.\",\n \"Title: Towards the Next Generation Video Broadcasting: Improved Performance Using Distributed MIMO\\nYear: 2015\\n\\nAbstract: Distributed multiple-input multiple-output (MIMO) is a promising technique for the future terrestrial TV broadcasting systems due to its higher efficiency over the traditional single-antenna transmission scheme. In this article, we investigate the application of distributed MIMO technique in the future broadcasting systems. We first show the advantage of the distributed MIMO from the channel capacity perspective. Then different space-time block codes (STBCs) that can be applied in the distributed MIMO broadcasting scenarios are discussed and compared with real system configurations and channel conditions. It is shown that the simple and sophisticated STBCs have mutually complementary application scenarios. Particularly, short 3D MIMO code is a good performance versus decoding complexity compromise among different application scenarios.\",\n \"Title: Resource allocation mechanism in IEEE 802.15.3 parent/child model\\nYear: 2015\\n\\nAbstract: In this paper, a novel mechanism for access control and resource allocation in a parent/child IEEE 802.15.3 network is proposed. The main objective is twofold; first the idea of using a dynamic superframe size that ensures the efficient use of the superframe is implemented; second, two approaches for resource allocation among child PicoNet Coordinators (PNCs) and DEVices (DEVs) based on the dynamic superframe idea are proposed. These approaches, referred to as “” and “”, are studied in comparison with a basic approach. They share the concepts of the dynamic superframe size, the prioritization of child PNCs over DEVs and the admission control strategy. They are then differentiated by applying the superframe utilization threshold concept in the “” and the addition of service differentiation in the “”. The superframe utilization threshold allows more devices to access the shared superframe, whereas the service differentiation guarantees the quality of service (QoS) requirements for real-time traffic requests. The proposed mechanism seeks to apply a compromise between the devices satisfaction, the QoS requirements and the requests rejection rate to induce fairness and service differentiation. Simulation results show that the proposed algorithms provide high satisfaction for both child PNCs and DEVs with greater satisfaction for the child PNCs. The superiority of using the dynamic superframe size over the fixed superframe size is also shown by the efficient use of the superframe and the minimized delay between the channel time allocations of the same device in the subsequent superframes.\",\n \"Title: BER Analysis for Equal Gain Transmission in Downlink Multiuser MIMO Systems\\nYear: 2015\\n\\nAbstract: In this letter, we derive the bit error rate (BER) performance of a multiuser multiple-input multiple-output (MU-MIMO) system using equal gain transmission (EGT) precoding. Considering binary phase shift keying (BPSK) modulation, we analyze and propose a statistical model based on the Nakagami distribution for the interference terms involved in the received signals. The precision of our model is compared with the common Gaussian approximation and validated through simulation results. A closed-form BER expression is derived as a function of the number of transmit antennas for arbitrary E b /N 0 .\",\n \"Title: Impact Of Clipping On Evm Of The Predistorted Non-Linear Amplified Multicarrier Signals\\nYear: 2015\\n\\nAbstract: In this paper, we analyse the impact of clipping used as a peak-to-average power ratio reduction technique for multicarrier signals, assuming that a linearization stage is implemented before the power amplifier. Therefore, the expression of the Error Vector Magnitude (EVM) of the output signal is analytically derived. We mainly focus on the effect of clipping in both cases with and without the use of the predistortion stage. We provide analytical expressions of the EVM in the form of a series expansion that depends on the clipping threshold, the transition factors of the power amplifier, the predistortion characteristics and the PAPR of the input signal. We show through simulations that the proposed expressions present very good accuracy, even when limiting the series expansion to an order of 4. Finally, a simple scenario is proposed to demonstrate the benefit of our EVM expressions in controlling the average transmitted power with the aim to guarantee a high quality of service.\",\n \"Title: Global throughput maximization of a hybrid unicast-broadcast network for linear services\\nYear: 2015\\n\\nAbstract: An exponential growth of the mobile data traffic is expected in future networks. The reason of this growth is related to the increasing popularity of linear services such as mobile TV, live and sports events, which may lead to the delivery of the same contents to a large audience. Recent studies have shown that network cooperation is a promising candidate to deal with such an issue. This paper investigates, from a planning perspective, the optimization of a hybrid unicast/broadcast network for linear services. We introduce an analytical approach with the aim of stating an optimization problem on the hybrid network configuration, namely in terms of the sharing between the unicast and broadcast modes to deliver a particular service. We show by simulation the existence of an optimal operation point which leads to the maximization of the overall throughput of the hybrid network under the constraint of a minimum service success rate.\",\n \"Title: Modeling Bp Decoding Error Events With The Ldpc Codes Of The Dvb-S2 Standard\\nYear: 2014\\n\\nAbstract: In this paper, we focus on the behavior of the Belief Propagation (BP) algorithm when decoding the Low-Density Parity-Check (LDPC) code of rate 3/4 in the DVB-S2 standard. By studying the topological structure of its Tanner graph, we raise properties inherent to the degree distribution that turns out to be strongly correlated with the decoding failures. The irregularity of the degrees seriously damages the performance, visible by an abrupt error floor. We accordingly propose a novel model of the error events based on the degree distribution which helps simulate typical error events and observing the flaws of the DVB-S2 code. This work could be used as a good basis to design future codes with a better decoding behavior, especially in the error floor region.\",\n \"Title: Power control optimization for tone reservation based PAPR reduction algorithms\\nYear: 2014\\n\\nAbstract: Orthogonal Frequency Division Multiplexing (OFDM) systems are characterized by a high Peak to Average Power Ratio (PAPR), which results in performance degradation due to High Power Amplifiers' nonlinearities. Multiple algorithms have been proposed to deal with this problem, among them the Tone Reservation (TR) technique adopted in the second generation of Digital Video Broadcasting for Terrestrial transmission (DVB-T2). In this paper we analyze the power control schemes used in TR algorithms by assessing both their performance and complexity. We then propose two novel solutions to improve the performance without requiring additional complexity. The first is based on an enhanced selection of the peaks to be processed by the TR algorithm, and the second is based on a dynamic computation of the threshold used. We show that the proposed techniques outperform the TR algorithm adopted in DVB-T2 by 0.77 dB in terms of CCDF gain at 10\\n<sup xmlns:mml=\\\"http://www.w3.org/1998/Math/MathML\\\" xmlns:xlink=\\\"http://www.w3.org/1999/xlink\\\">-3</sup>\\n and by 0.33 dB in terms of Input Back Off (IBO) gain for an MER of -40dB.\",\n \"Title: Admission control and resource allocation in IEEE 802.15.5 HDR WPAN\\nYear: 2014\\n\\nAbstract: In this paper, we propose a novel admission control and centralized channel time allocation algorithm for two-hop communications based on the IEEE 802.15.5 specifications. The first contribution is the introduction of a dynamic superframe size for efficient resource utilization. The second contribution adopts the use of a threshold concept and a flag metric, which jointly provide satisfaction and fairness among the requesting devices. The performance of the proposed algorithm is evaluated and the results show that the proposed approach achieves good performance in terms of satisfaction and fairness.\",\n \"Title: SDRAN-based user association and resource allocation in heterogeneous wireless networks\\nYear: 2016\\n\\nAbstract: Integrating different types of wireless access technologies in heterogeneous networks provides flexible choices for users to handover to a more preferable network that satisfies their needs. In this paper, we discuss the recent 802.21.1 Software Defined Radio Access Network (SDRAN) architecture that couples the Software Defined Networking (SDN) mechanisms by the IEEE 802.21 Media Independent Handover Services. Based on this architecture, we propose a centralized handover algorithm that manages user association and downlink capacity resource allocation. The algorithm is formulated as a Generalized Assignment Problem (GAP) which maximizes the overall profit of mobile nodes which is studied in terms of power consumption and Received Signal Strength (RSS). Comprehensive simulation has been conducted; It is shown that the formulated algorithm reduces the overall consumed power and increases the average relative received signal strength while demanding an extra number of HOs.\",\n \"Title: Optimal operating point of the power amplifier with respect to the EVM for TV broadcasting applications\\nYear: 2016\\n\\nAbstract: The high Power Amplifier (PA) dominates the power consumption in DVB transceiver systems and requests focusing the power efficiency improvements on this device; especially when multi-carrier modulations are used which is characterized by high Peak-to-Average Power Ratio (PAPR). In this paper, first, we provide analytical expression of the Error Vector Magnitude (EVM) of the amplified signal when polynomial predistortion and clipping techniques are activated. Secondly, a theoretical analysis of the trade-off between the power amplifier linearity and efficiency is investigated considering the performance of the clipping and predistortion techniques. Therefore, an analytical expression which gives the optimal Input Back-Off (IBO) maximizing the PA efficiency, with respect to an EVM constraint, is provided. It is demonstrated that significant power efficiency improvement can be achieved using our analytical expressions.\",\n \"Title: Optimal capacity of hybrid unicast-broadcast networks for mobile TV services\\nYear: 2016\\n\\nAbstract: This paper studies the optimal broadcast coverage radius in a hybrid network from the perspective of network planning. The hybrid network is composed of a unicast and broadcast systems which cooperate to deliver linear services on portable and mobile devices. First we define a hybrid approach in which the coverage radius of the broadcast network is adjusted according to the distribution of the users in the service area. Then, by applying convex optimization methods, we derive a closed-form expression of the optimal broadcast coverage radius that maximizes the average capacity per user in the hybrid network. The results show that the closed-form expression of the broadcast coverage radius is in good agreement with the simulation results, which validates the theoretical approach.\",\n \"Title: Blind repetitions for cellular-IoT: Performance analysis of combination mechanisms\\nYear: 2016\\n\\nAbstract: As evident from the increasing number of technical studies on machine type communications, Internet of Things (IoT) has become a hot topic. Among all the initiatives attempting to build end-to-end connectivity solutions, the 3rd Generation Partnership Project (3GPP) is actively working on the integration of IoT features into the existing 2G and 4G network specifications. In the framework of the adaptation of the GSM (Global System for Mobile communication) and EDGE (Enhanced Data rates for GSM Evolution) systems to IoT applications, several solutions have been proposed at the PHY/MAC level, all sharing a common trait: enhancing the coverage using blind repetition mechanisms. The question of how to recombine the repeated symbols at the receiver side has then to be raised. In this work, we analyze the performance of state-of-the-art combination approaches that are carried out at different levels of a classic GSM receiver and we propose a new combination scheme. To mitigate the drawbacks of each individual solution, two hybrid strategies are proposed involving these different levels of combination. Performance comparison is led for the different cases considering constant as well as time variant channels, leading to a discussion about the most efficient strategy to follow, depending on the targeted application. To the best of our knowledge, this is the first work dealing with the comparison of combination mechanisms including hybrid combining solutions.\",\n \"Title: Quantifying the Memory Effects of Power Amplifiers: EVM Closed-Form Derivations of Multicarrier Signals.\\nYear: 2017\\n\\nAbstract: This letter deals with the analysis of the memory effects of the power amplifier (PA) on multicarrier signals. We analytically derive the expressions of the error vector magnitude (EVM) of amplified signals jointly considering a memory polynomial PA model and clipping effects. We provide analytical EVM expressions that depend on the PA coefficients, PA order, and memory depth as well as the clippi...\",\n \"Title: A low latency algorithm for efficient PAPR reduction for DVB-T2 and ATSC 3.0 broadcast\\nYear: 2017\\n\\nAbstract: In this paper, the problem of peak-to-average power ratio (PAPR) reduction of orthogonal frequency division multiplexing (OFDM) signals is investigated in the context of second generation of digital video broadcasting for terrestrial transmission (DVB-T2) and the American digital video broadcasting standard ATSC3.0. As a multicarrier modulation technique is characterized by a high PAPR of the transmitted signal and OFDM is prone to non-linear (NL) effects of power amplifiers. DVB-T2 and ATSC3.0 have adopted a gradient-based tone reservation (TR) PAPR reduction technique which is based on an iterative process where, at each iteration, a predefined kernel is used to reduce one peak in time domain. Recently, a new TR PAPR reduction technique termed individual carrier allocation for multiple peaks (ICMP), based on a novel kernel definition has been proposed. However, it suffers from latency issues in ATSC3.0 and higher modes of DVB-T2. So, we propose another novel TR technique, grouped ICMP (GICMP). The simulation results show that GICMP offers better performance than the gradient-based DVB-T2 algorithm. Furthermore, it not only yields the same performance as ICMP but also has less latency.\",\n \"Title: Time reversal based beamforming for wireless communications: A theoretical analysis from moderate-scale to large-scale antenna systems\\nYear: 2015\\n\\nAbstract: This paper aims at presenting a comprehensive analysis of time-reversal (TR) as a precoding scheme for wireless communications. As exposed in the paper, TR can be viewed as a particular spatial filtering scheme with some strong commonalities with more conventional beamforming strategies used in large-scale antenna systems (LSAS). Our goal is to compare the behavior and the performance of the TR-based beamforming with that of the well-known Maximum Ratio Transmission (MRT). Therefore, we develop a generic system model for TR-based beamforming systems. Then we lead a theoretical performance analysis relying on ergodic capacity computations and based on statistical models of the equivalent channels obtained with both RT and MRT precoding schemes.\",\n \"Title: Space Shift Keying Modulations for Low Complexity Internet-of-Things Devices.\\nYear: 2017\\n\\nAbstract: The new MIMO schemes commonly referred to as spatial modulations (SM) use the index of the transmit (resp. the receive) antenna to transmit additional data and hereby increase the spectral efficiency of digital communication systems. Most of proposed SM schemes allow for simultaneously transmitting two types of symbols, the classical IQ-symbols of a M-order modulation, and a spatial information related to the index of the transmit or targeted receive antenna(s). In this paper, we propose a complete uplink and downlink transmission system between a Base Station and a Connected Device for Internet-of-Things application based on SM. We thus concentrate on SM schemes only conveying spatial information for both downlink (from a base station to a connected device) and uplink (from a connected device to a base station) transmissions and leading to low complexity at the device. For the downlink we propose the use of an existing receive SM, the ERASK that is an extension of a very simple receive spatial modulation known as RASK (Receive Antenna Shift Keying) which aims at increasing the spectral efficiency. For the uplink, we firstly propose a new scheme which analogy with ERASK is highlighted. Both ERASK and ESSK can be efficiently used in the global system while keeping very low complexity at the device side. All complexity is reported at the base station with pre-processing function used to concentrate the signal energy towards the device targeted antenna(s). The channel estimation is also only needed at the base station for pre-processing of the ERASK and detection process of the ESSK.\",\n \"Title: Near optimal linear-service oriented resource allocation strategy for LTE networks\\nYear: 2017\\n\\nAbstract: Extensive research efforts are being attracted towards resource allocation in LTE networks. However, from the resource allocation perspectives, linear services such as Digital TV programs have been scarcely studied. In fact, for end-users requesting linear services, the maximization of the the number of users served is much important goal than maximizing the overall capacity of the network. In this paper, linear services oriented resource allocation strategy is proposed. The goal of the proposed algorithm is to maximize the success rate of the users, depending on a minimum required quality of service (QoS), measured in terms of user capacity constraint. The proposed method was tested and compared to the optimal branch and Cut solution and has shown good robustness against network and user parameters.\",\n \"Title: Application of AL-FEC to HbbTV DVB-T2 systems for the provision of VoD services in portable and fixed TV reception\\nYear: 2017\\n\\nAbstract: Interactive TV services have grown significantly in recent years thanks to the standardization of several approaches such as Hybrid broadcast and broadband TV (HbbTV) technology. HbbTV applied to portable and fixed DVB-T2 reception, allows TV channels broadcast via terrestrial channels to publish interactive applications in addition to their TV programs. Video-On-Demand (VoD) provided via Push-VoD service, represents one of these applications. However the provision of HbbTV Push-VoD services in portable and fixed DVB-T2 reception is challenged by severe propagation conditions of the static multipath scenarios. In this paper, we provide a first reference addressing the question of how HbbTV DVB-T2 systems enables a reliable delivery of Push-VoD services in portable and fixed broadcast reception, in absence of broadband networks. We analyze the coexistence of HbbTV standard and DVB-T2 standard. We investigate the solution based on the LDPC-Staircase AL-FEC scheme for improving the portable and fixed broadcast reception of Push-VoD service in HbbTV DVB-T2 systems. Simulation results show that LDPC-Staircase AL-FEC code enables to enhance the delivery of HbbTV Push-VoD services over DVB-T2 systems with portable and fixed TV reception.\",\n \"Title: EVM derivation of multicarrier signals to determine the operating point of the power amplifier considering clipping and predistortion.\\nYear: 2016\\n\\nAbstract: The high peak-to-average power ratio (PAPR) remains a major drawback of multicarrier modulations. Hence, the nonlinear characteristics of power amplifiers (PA) results in strong distortion and low power efficiency when multicarrier modulations are used. In this paper, the impact of the nonlinearities on the amplified multicarrier signal is analyzed, considering both PAPR limitation using clipping and PA linearization. Therefore, we derive the expression of the error vector magnitude (EVM) of the amplified signal with and without the use of polynomial predistortion and/or clipping techniques. We provide analytical EVM expressions that depend on the PA and predistortion characteristics, as well as the PAPR and the average power of both input and clipped signals. These expressions are general formulas which allow to measure in-band distortion at the PA output. The simulation results show that the proposed expressions present perfect accuracy. Moreover, the trade-off between the PA linearity and efficiency is investigated considering the performance of the clipping and predistortion techniques. An analytical expression which gives the optimal input back-off (IBO) maximizing the PA efficiency with respect to any EVM constraint is provided. Finally, the predistortion complexity is discussed aiming at reducing it with respect to an EVM constraint.\",\n \"Title: Mobile TV directed resource allocation scheme for LTE networks\\nYear: 2017\\n\\nAbstract: Resource allocation in LTE networks has attracted extensive research efforts. However, linear services (such as Digital TV programs) have been barely considered from the resource allocation perspectives. Indeed, when a linear service is offered to the end-users, the network should satisfy the maximum number of users served rather than maximizing the sum rate capacity. In this paper, The linear services oriented resource allocation strategy is discussed. This approach aims at maximizing the users' success rate, based on a minimum required quality of service (QoS) measured in terms of minimum throughput. Here, the focus will particularly on both Poisson Process and grid models as well their comparison. The method was tested in different scenarios and under different conditions to prove its robustness against user density, user requirements and reuse factor.\",\n \"Title: MIMO Channel Hardening: A Physical Model based Analysis.\\nYear: 2018\\n\\nAbstract: In a multiple-input-multiple-output (MIMO) communication system, the multipath fading is averagedover radio links. This well-known channel hardening phenomenon plays a central role inthe design of massive MIMO systems. The aim of this paper is to study channel hardening using aphysical channel model in which the influences of propagation rays and antenna array topologiesare highlighted. A measure of channel hardening is derived through the coefficient of variationof the channel gain. Our analyses and closed form results based on the used physicalmodel are consistent with those of the literature relying on more abstract Rayleigh fadingmodels, but offer further insights on the relationship with channel characteristics.\",\n \"Title: Extended Receive Spatial Modulation Mimo Scheme For Higher Spectral Efficiency\\nYear: 2018\\n\\nAbstract: Spatial modulation (SM) schemes use the index of the transmit (resp. the receive) antenna to transmit additional data, and therefore allow for a new exploitation of the space dimension to increase the spectral efficiency of MIMO systems. The baseline of the proposed study relies on the so-called extended receive antenna shift keying (ERASK) concept consisting in transmitting the spatial bits by focusing the transmit energy towards subsets of receive antennas. With ERASK, the subsets of targeted antennas can be of any size leading to an amount of additional spatial bits directly equal to the number of receive antennas. However, an ON/OFF keying is used to carry the spatial symbols meaning that a zero power level is associated for non targeted antennas. Consequently, the transmission of additional IQ symbols in ERASK is not compatible with every spatial symbol configurations. In this paper, we propose a novel scheme referred to as Extended RSM (ERSM) that sends additional IQ modulated signals in ERASK, in order to increase the overall spectral efficiency. The key idea is to associate a non zero power levels for all spatial symbols. Two power levels are used to transmit the spatial and the IQ symbols, and an optimal ratio of the power is calculated. The theoretical performance of the proposed ERSM scheme is derived and then validated through simulations.\",\n \"Title: Angular Based Beamforming and Power Allocation Framework in a Multi-User Millimeter-Wave Massive MIMO System\\nYear: 2018\\n\\nAbstract: Millimeter-Wave (mmWave) systems offer extremely large Band-Width (BW) and highly Line-of-Sight (LOS) dominant channels. Many beamforming and power allocation techniques have recently emerged to leverage such channel characteristics in order to enhance the sum capacity and the coverage for multi-user transmissions. However, most of the existing algorithms are practically limited due to full channel knowledge and high complexity requirements. In this paper we introduce a novel, low complexity, angular based beamforming and power allocation framework, that requires the knowledge of the main contributive Directions of Arrival/Departure (DoAs/DoDs) of the propagation channel only. Exploiting the fact that propagation conditions are highly driven by the geometrical structure of the channel in mmWave scenarios, our method relies on the estimation of the leakage caused by each User Equipment (UE) on all the other UEs, approximated from the DoAs contributions. We prove with the simulation results that this approach, in addition to be practically plausible, can enhance the network Spectral Efficiency (SE) alongside with maintaining acceptable data rates for the cell edge UEs.\",\n \"Title: Efficiency of the TR Technique and the POPS Algorithm for Waveform Optimization in MISO-OFDM Systems\\nYear: 2018\\n\\nAbstract: In this paper, we extend the joint use of the Time Reversal (TR) precoding technique and the Ping-Pong Optimized Pulse Shaping (POPS) waveform optimization algorithm to design optimal transmit and receive waveforms for Multiple Input Single Output-Orthogonal Frequency Division Multiplexing (MISO-OFDM) systems. The design process banks on a full characterization of the scattering function of the TR-MISO channel, followed by the implementation of the POPS approach. Numerical analysis reveal a significant gain, brought by the optimized waveforms in TR-MISO-OFDM systems, in terms of Signal to Interference plus Noise Ratio (SINR), compared to the optimized waveforms in TR-SISO-OFDM systems and the PHYDYAS waveform in TR-MISO-OFDM systems.\",\n \"Title: Theoretical Performance of Coherent and Incoherent Detection for Zero-Forcing Receive Antenna Shift Keying.\\nYear: 2018\\n\\nAbstract: Receive antenna shift keying (RASK) is a multiple-input multiple-output transmission scheme that applies the spatial modulation concept at the receiver. Zero-forcing (ZF) precoding can be used at the transmitter side to steer the signal toward the targeted antenna. Different detection schemes can be implemented in order to detect the RASK information. Moreover, for the sake of low complexity and low power consumption, decreasing the number of RF chains by the use of switched antennas can also be implemented. In this paper, we lead a theoretical analysis of the transmission of spatial symbols in RASK, using the ZF precoding. Two detection schemes relying on maximum likelihood (ML) detection are presented and proved to be implemented as a simple single-tap detector. One, called coherent ML (CML), needs one RF chain per receive antenna and the other one, named incoherent ML (IML), is based on an envelope detector and needs only one RF chain for the entire antenna array. Theoretical expressions of the bit error rate (BER) are derived for each scheme and validated through simulation. The CML detector is demonstrated to obviously outperform the IML detector in terms of the BER but at the cost of higher complexity and power consumption. We then proposed to use switches to reduce the number of RF chains, thus trading off some performance against complexity. At some point, IML is shown to outperform CML, depending on the number of switches, RF chains, and system configuration. This turning point is analytically calculated and validated through simulations, and so can be used for further system comparison studies.\",\n \"Title: Very Efficient Tone Reservation PAPR Reduction Fully Compatible With ATSC 3.0 Standard: Performance and Practical Implementation Analysis.\\nYear: 2018\\n\\nAbstract: A common issue in any multicarrier communication system such as the American digital video broadcasting (ATSC 3.0) standard is the high peaks of the transmitted signal. This disadvantage constrains the high power amplifiers to be deployed in their linear region which lowers their power efficiency. To overcome this issue, various techniques aiming at reducing the signal fluctuations have been investigated. Recently, Tone Reservation (TR)-based algorithms have been studied and proposed for the second generation of digital video broadcasting (DVB-T2) and ATSC 3.0 standards. The algorithm is essentially based on an iterative gradient approach to cancel one peak of the time-domain signal at each iteration. In this paper, we propose a novel TR-based algorithm implementable and compatible with the ATSC 3.0 standard and named as grouped carrier peak windowing (GCPW). This algorithm is based on a new kernel definition targeting the cancellation of multiple signal peaks at a time which highly reduces the total number of iterations, therefore reducing the system global latency and being more suited for implementation in today's ATSC 3.0 transmitters. Taking into account hardware resources requirements, we propose a new method to select the highest signal peaks to be considered in the PAPR reduction process. Hence, the system latency, complexity, and memory resources are reduced and better performance than the ATSC 3.0 gradient-based algorithm can be offered. The implementation of the GCPW algorithm in fixed-point architectures is also studied and optimized in this paper. We demonstrate through simulation results that the proposed algorithm offers very good performance/latency/complexity/memory tradeoff in both floating and fixed point implementations.\",\n \"Title: A Novel Statistical Model Of Ofdm Signals With Optimal Tone Reservation Papr Reduction For Evm Theoretical Analysis\\nYear: 2018\\n\\nAbstract: In this paper, we study the statistical distribution of the OFDM time-domain signal amplitude to which the optimal tone reservation (TR) PAPR reduction solution is applied. This study comes as a major step towards the derivation of the analytical expression of the error vector magnitude (EVM) metric on OFDM waveforms when non-linear amplification is used. We demonstrate through simulations that this distribution has a bimodal behavior and can be tightly modeled as a superposition of two probability density functions whose expressions are given and whose parameters are obtained through proper estimation. From such analysis, we provide the complete distribution expression of the PAPR-reduced OFDM signal, which can be further exploited for the derivation of performance bounds of the TR PAPR reduction strategy.\",\n \"Title: Leakage Based Users Selection For Hybrid Beamforming In Millimeterwave Mimo\\nYear: 2018\\n\\nAbstract: Millimeter-Wave (mmWave) channels are characterized as being sparse and Line of Sight (LoS) dominated. Applying hybrid beamforming in mmWave systems guarantees Multi-User (MU) transmission with low hardware requirements. Zero Forcing (ZF) precoding can be utilized at the digital layer of the hybrid beamformer to mitigate the Inter User Interference (IUI). However, due to the sparse nature of the mmWave channels, a ZF precoder may provide limited performance due to the highly correlated MU channel matrix. The aim of this paper is hence to address the interference mitigation problem in the analog domain before handling it at the ZF digital level. Signal to Leakage and Noise Ration (SLNR) based users selection algorithm is introduced for maximizing the Spectral Efficiency (SE) of the system with low complexity requirements. We finally highlight by simulations the potential gains achieved by the introduced interference aware users selection approach.\",\n \"Title: Analytical Study of the Probability of Coverage in Hybrid Broadcast-Unicast Networks\\nYear: 2018\\n\\nAbstract: Linear services such as mobile TV are witnessing an increasing demand, creating challenges for conventional delivery methods in both broadcast and broadband networks. To overcome this problem, a hybrid delivery approach combining both networks has recently emerged. Using stochastic geometry and a Poisson point process distribution of the users and access points, this paper discusses a hybrid broadcast/unicast network model where an analytical approach was undertaken, deriving a new expression for the probability of coverage. The model is then used to assess the feasibility of the proposed hybrid scheme and then extended to numerically find the optimal operation point in terms of key design parameters including user density in the network.\",\n \"Title: In-Band and Out-Of-Band Distortions Optimization for ATSC 3.0 Transmission: A Novel TR PAPR Reduction Algorithm\\nYear: 2018\\n\\nAbstract: This paper presents a novel Tone Reservation (TR) Peak to Average Power Ratio (PAPR) algorithm for the American digital video broadcasting (ATSC 3.0) standard. As any multicarrier communication systems, ATSC 3.0 transmitters suffer from high peaks of the transmitted signal. The high signal fluctuations constitute a major disadvantage of these systems. Indeed, they constraint the high power amplifiers to operate in their linear region which leads to poor spectral efficiency. To reduce the amplitude of the signal peaks, a TR gradient based algorithm has been proposed as an option within the ATSC 3.0 standard. Although TR-based algorithms have been intensively investigated these last years, very few today's transmitters implement these algorithms as they do not offer a high performance/complexity trade-off. This paper presents a novel TR-based algorithm, named as Grouped Carrier Peak Windowing (GCPW), which offers higher performance than the standard TR gradient-based algorithm while being implementable and compatible with the ATSC 3.0 standard. We also show through simulations that the proposed algorithm highly outperforms the gradient-based standard algorithm in terms of both in-band and out-of-band distortions. Results show that the novel algorithm offers very promising performance in terms of both modulation error rate and shoulder attenuation.\",\n \"Title: MIMO Channel Hardening for Ray-based Models\\nYear: 2018\\n\\nAbstract: In a multiple-input-multiple-output (MIMO) communication system, the multipath fading tends to vanish with increasing number of radio links. This well-known channel hardening phenomenon plays a central role in the design of massive MIMO systems. It is quantified by the coefficient of variation of the channel gain. The aim of this paper is to study channel hardening using a physical channel model in which the influences of propagation rays and antenna array topologies are highlighted. Our analyses and closed form results extend the hardening properties beyond the classical Rayleigh fading models and offer further insights on the relationship with channel characteristics.\",\n \"Title: Context-aware and priority-based user association and resource allocation in heterogeneous wireless networks.\\nYear: 2019\\n\\nAbstract: Heterogeneous wireless networks (HWNs) are usually characterized by the integration of cellular networks and wireless local area networks (WLANs) to meet user requirements and enhance system capacity. This paper proposes a user association and downlink resource allocation algorithm in HWNs with users having different priorities. The proposed solution employs contextual information related to the preferences of the users, their requested data rate, and the characteristics of networks. The user preference is translated through a profit function that is based on the received signal quality and the power consumption at mobile terminals (MTs). Accordingly, an optimization problem is formulated to maximize the overall user satisfaction for each priority level. The formulated problem throws firm restrictions to prevent low-priority users from allocating resources utilized by other users with higher priorities. We then propose a novel heuristic approach with polynomial-time complexity to approximate the formulated problem. Furthermore, the system architecture is discussed and a new solution management strategy is proposed to limit the complexity of the algorithm. Simulation results show that the proposed approximation method maintains the nearest performance to the optimal solution.\",\n \"Title: Analytical Derivation and Optimization of a Hybrid Unicast-Broadcast Network for Linear Services.\\nYear: 2016\\n\\nAbstract: This paper provides a theoretical analysis of the performance of an hybrid network composed of a broadband system and a broadcast system which cooperate to deliver linear services on portable and mobile devices. First, we propose a global system model for the hybrid network in which the coverage area of the broadcast system is adjusted with respect to the spatial distribution of the users. Then based on the number of users requesting the services, the model parameters are optimized to increase the capacity of the hybrid network while guaranteeing a given quality of service. By applying convex optimization techniques, we derive a closed-form expression of the optimal coverage radius of the broadcast system that maximizes the capacity of the hybrid network. Next, based on this theoretical model, we analyze the performances of the optimized hybrid network. The results show that, depending on the number of users requesting the service, significant improvements of the capacity and quality of service are achieved with the hybrid network. Finally, regarding the planning of a cooperative network, this paper can be used to determine the optimal planning parameters of the broadcasting system for the delivery of linear services.\",\n \"Title: HbbTV based Push-VOD services over DVB networks: Analysis and AL-FEC code application\\nYear: 2016\\n\\nAbstract: Hybrid broadcast broadband TV (HbbTV) is an European initiative for providing viewers a richer experience of enhanced and interactive hybrid TV, through the launching of innovative hybrid services on both broadcast and broadband networks, allowing the viewer to interact with the current TV program. The latest release, HbbTV standard v2.0, provides additional features to HbbTV applications. Among the most important advancement of this version was the addition of the Push-VOD service. In this paper we focus on the broadcast method principle of Push-VOD services presented in the HbbTV v2.0. We investigate various technical aspects that have remained open questions in the standard to provide a first reference on the depth analysis of HbbTV based Push-VOD services. In addition, we propose and analyse a first solution for a reliable Push-VOD service delivery within file delivery protocol. This solution is based on the LDPC-Staircase AL-FEC codes using the hybrid (IT/ML) decoding method. Simulation results show that the merging of the repetition and the AL-FEC coding allows to inherit the benefits of both repetition and coding. A reliable file delivery is assured with low recovering complexity and download time.\",\n \"Title: Performance analysis of the maximum ratio transmission preprocessing for extended receive antenna shift keying\\nYear: 2017\\n\\nAbstract: The Extended Receive Antenna Shift Keying (ERASK) scheme is a MIMO scheme based on the Receive Spatial Modulation concept, invented to increase the overall spectral efficiency, by exploiting all possible combinations of receive antenna indexes. In this paper, we evaluated the ERASK scheme using the Maximum Ratio Transmission (MRT) preprocessing (MRT-ERASK), using the real amplitude threshold detector and compare it with the ERASK scheme using the Zero Forcing (ZF) preprocessing (ZF-ERASK). Analytical derivations of the received signal of the MRT-ERASK show that a complex inter-antenna interference is added to other antennas depending on the transmitted spatial symbol. The Bit Error Rate performance is also derived analytically. Simulation results over MIMO Rayleigh channel are provided to compare both systems, showing that ZF-ERASK outperforms MRT-ERASK but at the expense of a higher implementation complexity for ZF-ERASK. On the other hand, increasing the number of transmit antennas of a MRT-ERASK improves its performance getting closer to the performance ZF-ERASK. Therefore, the higher the number of transmit antennas, the nearer the performance of both systems, and the more suitable the MRT-ERASK to be implemented.\",\n \"Title: A Low Complexity PAPR Reduction Algorithm for ATSC 3.0 Transmitters\\nYear: 2018\\n\\nAbstract: Tone Reservation (TR) is one of the common algorithms proposed for peak-to-average power ratio (PAPR) reduction in next generation American digital video broadcasting (ATSC 3.0) transmitters. Recently, a quasi-optimal TR technique with a novel kernel definition named as grouped individual carrier multiple peaks (GICMP) scheme has been proposed. In GICMP, firstly, the reserved tones are partitioned into G groups, which also equals to the number of iterations. Then, a comb-like kernel is constructed with each group and it is adjusted to cancel a given set of peaks of data signal at each iteration. This scheme not only offers good performance/complexity trade-off but also compatible with ATSC 3.0 standard. Based on same kernel definition, we have proposed a new method, named as grouped carrier peak windowing (GCPW), which requires less hardware resources, compared to GICMP. Instead of choosing a set of peaks, this new method introduces a peak window to consider a given number of signal samples that exceed the predetermined threshold. This new method aims of reducing first peaks falling within the window, while leaving out higher peaks that may occur later. The simulation results show that the proposed low-complexity GCPW has less hardware memory requirements and with a good choice of threshold, window size and groups, offers quasi-optimal PAPR reduction performance same as that of GICMP.\",\n \"Title: Mesh Hdr Wpan Resource Allocation Optimization Approaches\\nYear: 2019\\n\\nAbstract: In the multihop IEEE 802.15.5 networks, all the devices compete to the resources of a shared superframe. In order to distribute these resources in a fair and satisfactory manner among the competing devices, we propose a distributed optimization framework for resource allocation scheme in an IEEE 802.15.5 hop-1. For this purpose, we introduce in this paper a suite of optimization problems for the hop-1 IEEE 802.15.5 resource allocation to optimize fairness and satisfaction without exceeding the superframe size and respecting the demanded channel time size sent by the requesting devices. Simulation results studied and compared the satisfaction factor and fairness index of the different proposed optimization problems. Consequently, a trade-off between satisfaction and fairness should be conducted for choosing the optimal solution.\",\n \"Title: A Systematic Beam Broadening Method for Large Phased Arrays\\nYear: 2021\\n\\nAbstract: Hybrid beamforming architecture is a key enabler for implementing Massive Multiple-Input Multiple-Output (mMIMO) solutions in next generation wireless communication systems. With such an approach, precoder and combiner designs not only depend on digital weights but also on modulus constraint analog weights controlled by phase shifters. This becomes a critical issue since the calculation of weights becomes a nonconvex problem that needs extensive search or complex iterative methods. In particular, beam broadening capability is of key interest for mMIMO systems to find a trade-off between the array gain and the beamwidth. Indeed, a high gain narrow beam is not always beneficial as far as mobility scenarios or control channel transmissions are concerned. In this paper, a non-iterative phase-only beam broadening method is proposed, relying on a quadratic phase excitation pattern applied to linear and planar arrays. We establish the key laws driving the beamwidth and steering angle adaption. We show that our technique can be applied systematically and is especially well-suited for large antenna arrays.\",\n \"Title: Compression of Clipped OFDM IQ Samples for Cloud Radio Access Network\\nYear: 2021\\n\\nAbstract: In this paper, we analyze the impact of reducing the Peak to Average Power Ratio (PAPR) of a time domain Orthogonal Frequency Division Multiplexing (OFDM) signal before its compression, as required for the transmission of IQ samples in a Centralized Radio Access Network architecture. The analysis is performed considering a clipping function for PAPR reduction. First, the distribution of the clipped OFDM IQ samples is obtained which in turn allows to derive the threshold and codebook levels of a non-uniform quantizer optimized for clipped IQ samples and the number of representation bits for each quantized levels in the entropy coding stage. Then, we provide a closed-form expression for the Modulation Error Ratio of the compressed and clipped OFDM signal. The simulation results show that the Gaussian quantizer applied to a clipped signal is not quite robust compared to a quantizer that has been optimized w.r.t. the distribution of clipped IQ samples, especially in the low-rate quantization region.\",\n \"Title: Deep Learning for Location Based Beamforming with Nlos Channels\\nYear: 2022\\n\\nAbstract: Massive MIMO systems are highly efficient but critically rely on accurate channel state information (CSI) at the base station in order to determine appropriate precoders. CSI acquisition requires sending pilot symbols which induce an important overhead. In this paper, a method whose objective is to determine an appropriate precoder from the knowledge of the user's location only is proposed. Such a way to determine precoders is known as location based beamforming. It allows to reduce or even eliminate the need for pilot symbols, depending on how the location is obtained. the proposed method learns a direct mapping from location to precoder in a supervised way. It involves a neural network with a specific structure based on random Fourier features allowing to learn functions containing high spatial frequencies. It is assessed empirically and yields promising results on realistic synthetic channels. As opposed to previously proposed methods, it allows to handle both line-of-sight (LOS) and non-line-of-sight (NLOS) channels.\",\n \"Title: Analog Eigen-Beamforming for mmWave Systems: Performance under practical constraints\\nYear: 2021\\n\\nAbstract: Many millimeter wave systems embed phased array antennas to overcome the large path loss that is experienced at such high frequencies. Digital Fourier Transform based beamforming (DFT-BF) is often implemented as a low-complexity and very limited feedback technique. Although DFT-BF is optimal in pure Line-of-Sight scenarios, it is substantially outperformed in richer scattering environments by more advanced approaches such as Eigenvector-based beamforming (Eigen-BF). The latter is harder to implement in practical scenarios due to its computational complexity and amount of feedback. In this paper, we propose several solutions to alleviate those drawbacks and analyze various performance-complexity trade-offs in the perspective of a practical usage of Eigen-BF solutions.\",\n \"Title: Leveraging triplet loss and nonlinear dimensionality reduction for on-the-fly channel charting\\nYear: 2022\\n\\nAbstract: Channel charting is an unsupervised learning method that aims at mapping wireless channels to a so-called chart, preserving as much as possible spatial neighborhoods. In this paper, a model-based deep learning approach to this problem is proposed. It builds on a physically motivated distance measure to structure and initialize a neural network that is subsequently trained using a triplet loss function. The proposed structure exhibits a low number of parameters and clever initialization leads to fast training. These two features make the proposed approach amenable to on-the-fly channel charting. The method is empirically assessed on realistic synthetic channels, yielding encouraging results.\",\n \"Title: Analysis of Compressing PAPR-Reduced OFDM IQ Samples for Cloud Radio Access Network\\nYear: 2022\\n\\nAbstract: A common problem of the virtualized cloud radio access network architecture (C-RAN) is the compression of the time-domain IQ samples before transmission over the fronthaul link. Considering a multicarrier waveform such as OFDM, whose IQ samples follow a quasi-Gaussian distribution, the conventional Gaussian quantizer may be used as the optimal solution to the compression problem. However, since the high peak-to-average power ratio (PAPR) of OFDM signals remains a serious problem, various techniques may be employed to reduce the time-domain fluctuations of the IQ samples in the OFDM, resulting in a change in its distribution. The latter fact makes the Gaussian quantizer suboptimal. The literature lacks a performance analysis of the conventional OFDM-based compression techniques when the PAPR of the OFDM signal is reduced. Therefore, in this paper, we study for the first time the impact of reducing the PAPR of the OFDM signal before compression in the C-RAN architecture through rate-distortion analysis. We consider clipping and tone reservation PAPR reduction algorithms. The former is the simplest PAPR reduction approach, while the latter is one of the most effective algorithms used in broadcasting standards such as DVB-T2 and ATSC 3.0. We first derive the distribution of the PAPR-reduced OFDM IQ samples. This is used to optimize the thresholds and codebook levels of a non-uniform scalar quantizer and the number of quantization bits allocated for each quantized level in the entropy coding stage, along with the MER performance analysis. The simulation results show that the conventional Gaussian-based compression techniques applied to a PAPR-reduced signal is not very robust to the statistical changes in the signal unless the signal distribution at the input of the Gaussian quantizer is not significantly affected. However, a significant gain is obtained when the quantizer is optimized with respect to the true distribution of the PAPR-reduced IQ samples.\",\n \"Title: An Efficient Analog Eigen-Beamforming Procedure for Wideband mmWave MIMO-OFDM Systems\\nYear: 2022\\n\\nAbstract: Millimeter wave communication systems define a new paradigm for wireless communications. At such high frequencies, multi-input multi-output (MIMO) processing is done on the radio-frequency signals to limit the hardware complexity and the energy consumption. Consequently, the channel experienced by the receiver directly depends on the transmitter and receiver antenna array responses, making the determination of the analog precoder and combiner an intertwined problem. In this paper, we propose a procedure capable of calculating analog weights at the transmitter and at the receiver, which allows the maximization of the wideband spectral efficiency for single stream transmission single user MIMO (SU-MIMO). While solely relying on partial channel state information (CSI), our procedure has close performance to full CSI-dependent schemes.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":94,"string":"94"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13593,"string":"13,593"}}},{"rowIdx":13594,"cells":{"question_id":{"kind":"string","value":"Michele Marchesi"},"question":{"kind":"string","value":"Generate an abstract for the title \"A machine learning approach for text categorization of fixing-issue commits on CVS\" using the following items: Data mining, CVS repositories, \"fixing-issue\" commits, Machine Learning techniques, classification."},"answer":{"kind":"string","value":"We studied data mining from CVS repositories of two large OO projects, Eclipse and Netbeans, focusing on \"fixing-issue\" commits. We highlight common characteristics of issue reporting, and problems related to the identification of these messages, and compare static traditional approaches, like Knowledge Engineering, to dynamic approaches based on Machine Learning techniques. We compare for the first time performances of Machine Learning (ML) techniques to automatic classify \"fixing-issues\" among message commits. Our study calculates precision and recall of different Machine Learning Classifiers for the correct classification of issue-reporting commits. Our results show that some ML classifiers can correctly classify up to 99.9% of such commits."},"correct_answer":{"kind":"string","value":"We studied data mining from CVS repositories of two large OO projects, Eclipse and Netbeans, focusing on \"fixing-issue\" commits. We highlight common characteristics of issue reporting, and problems related to the identification of these messages, and compare static traditional approaches, like Knowledge Engineering, to dynamic approaches based on Machine Learning techniques. We compare for the first time performances of Machine Learning (ML) techniques to automatic classify \"fixing-issues\" among message commits. Our study calculates precision and recall of different Machine Learning Classifiers for the correct classification of issue-reporting commits. Our results show that some ML classifiers can correctly classify up to 99.9% of such commits."},"documents":{"kind":"list like","value":["Title: Studying the evolution of quality metrics in an agile/distributed project\nYear: 2006\n\nAbstract: This paper analyzes the development of a project initiated by a co-located agile team that subsequently evolved into a distributed context. The project, named JAPS (Java Agile Portal System) [1], has been monitored on a regular basis since it started in January 2005, collecting both process and product metrics. Product metrics have been calculated by checking out the source code history from the CVS repository. By analyzing the evolution of these metrics, it has been possible to evaluate how the distribution of the team has impacted the source code quality.","Title: Micro Pattern Fault-Proneness\nYear: 2012\n\nAbstract: One of the goals of Software Engineering is to reduce, or at least to try to control, the defectiveness of software systems during the development phase. The aim of our study is to analyze the relationship between micro patterns (introduced by Gil and Maman) and faults in a software system. Micro patterns are similar to design patterns, but their characteristic is that they can be identified automatically, and are at a lower level of abstraction with respect to design patterns. Our study aims to show, through empirical studies of open source software systems, which categories of micro patterns are more correlated to faults. Gil and Maman demonstrated, and subsequent studies confirmed, that 75% of the classes of a software system are covered by micro patterns. In our study we also analyze the relationship between faults and the remaining 25% of classes that do not match with any micro pattern. We found that these classes are more likely to be fault-prone than the others. We also studied the correlation among all the micro patterns of the catalog, in order to verify the existence of relationships between them.","Title: Simulating Kanban and Scrum vs. Waterfall with System Dynamics\nYear: 2011\n\nAbstract: Nowadays, Scrum is the most used Agile Methodology, while the Lean-Kanban approach is perhaps the fastest growing AM. On the other hand, traditional, waterfall-like approaches are still very used in real-life software projects, due to the ease of up-front planning and budgeting, that however are seldom matched upon project completion. In our opinion, more effort is needed to study and model the inner structure and behavior of these approaches, highlighting positive and negative feedback loops that are strategic to understand their features, and to decide on their adoption. In this paper we analyze the dynamic behavior of the adoption of Kanban and Scrum, versus a traditional software development process such as the Waterfall approach. We use a system dynamics model, based on the relationships between system variables, to assess the relative benefits of the studied approaches. The model is simulated using a commercial tool. The proposed model visualizes the relationships among these software development processes, and can be used to study their relative advantages and disadvantages.","Title: An analysis of SNA metrics on the Java Qualitas Corpus\nYear: 2011\n\nAbstract: We computed the software graphs of 96 systems of the Java Qualitas Corpus, parsing the source code and identifying the dependencies among classes. We analyzed 12 software metrics on these 96 graphs, nine borrowed from Social Network Analysis (SNA), and three more traditional software metrics, such as Loc, Fan-in and Fan-out. We analyzed their correlations at system level, and studied the correlation statistics at data-set level. Our results show that these correlations are independent from the specific software system and are general properties of Java software systems. We show how the metrics can be partitioned in groups for almost the whole Java Qualitas Corpus, and that such grouping can provide insights on the topology of software networks. For two systems, Eclipse and Netbeans, we computed also the number of bugs, identifying the bugs affecting each class, and finding that some SNA metrics are highly correlated with bugs, while others are strongly anticorrelated. This suggests that practitioners and software engineers might take advantage of such metrics to keep control of software quality.","Title: Effort Estimation of Web Applications through Web CMF Objects\nYear: 2012\n\nAbstract: Several Content Management Frameworks (CMF), either open source or with a commercial license, are available to help the publication of huge amounts of information and to develop Web applications. Unfortunately, developing Web applications through a CMF is not exempt from cost and time overruns, as in traditional software projects, and currently there is no estimation model able to adequately measure the effort of Web application development. This work presents a new methodology for estimat ing the effort of Web applications developed with a Content Management Framework (CMF). We present the new key elements for analysis and planning, needed to define every important step in developing a Web application through a CMF. Using those elements, it is possible to estimate the effort needed to build such an ap plication in man-days. We also present the experimental validation of the proposed methodology per formed on a 9-project dataset, provided by three different Italian software companies.","Title: A Neural Network Diagnosis Approach for Analog Circuits\nYear: 1999\n\nAbstract: This paper presents a neural network system for the diagnosis of analogcircuits and shows how the performance of such a system can be affected by the choice of differenttechniques used by its submodules. In particular we discuss the influenceof feature extraction techniques such as Fourier Transforms, Wavelets and Principal ComponentAnalysis. The system uses several different power supplies and as manyneural networks “in parallel”. Two different algorithms that canbe used to combine the candidate sets produced by each network arealso presented. The system is capable of diagnosing multiplefaults even if trained on single ones.","Title: A revised web objects method to estimate web application development effort.\nYear: 2011\n\nAbstract: We present a study of the effectiveness of estimating web application development effort using Function Points and Web Objects methods, and a method we propose - the Revised Web Objects (RWO). RWO is an upgrading of WO method, aimed to account for new web development styles and technologies. It also introduces an up-front classification of web applications according to their size, scope and technology, to further refine their effort estimation. These methods were applied to a data-set of 24 projects obtained by Datasiel spa, a mid-sized Italian company, focused on web application projects, showing that RWO performs statistically better than WO, and roughly in the same way as FP.","Title: An empirical study of social networks metrics in object-oriented software\nYear: 2010\n\nAbstract: We study the application to object-oriented software of new metrics, derived from Social Network Analysis. Social Networks metrics, as for instance, the EGO metrics, allow to identify the role of each single node in the information flow through the network, being related to software modules and their dependencies. These metrics are compared with other traditional software metrics, like the Chidamber-Kemerer suite, and software graph metrics. We examine the empirical distributions of all the metrics, bugs included, across the software modules of several releases of two large Java systems, Eclipse and Netbeans. We provide analytical distribution functions suitable for describing and studying the observed distributions. We study also correlations among metrics and bugs. We found that the empirical distributions systematically show fat-tails for all the metrics. Moreover, the various metric distributions look very similar and consistent across all system releases and are also very similar in both the studied systems. These features appear to be typical properties of these software metrics.","Title: A model of the dynamics of the market of COTS software, in the absence of new entrants\nYear: 2007\n\nAbstract: This work proposes a simple model to simulate the dynamics of the market related to COTS software products, when there are no new entrants. The model takes in explicit account the interactions between products. The model uses a set of random variables, as the initial condition, to simulate the time evolution. The results of the simulations predict the emergence of monopolies, when the interactions between products are significant, and of open markets, when they are limited. The model is validated, by comparing the results of the simulation with the well established observations related to software markets and by comparing the resulting distribution of COTS software products with real data on the software market, which was provided by PC Data Inc.","Title: System performance analyses through object-oriented fault and coupling prisms\nYear: 2014\n\nAbstract: A fundamental aspect of a system's performance over time is the number of faults it generates. The relationship between the software engineering concept of \\\"coupling\\\" (i.e., the degree of inter-connectedness of a system's components) and faults is still a research question attracting attention and a relationship with strong implications for performance; excessive coupling is generally acknowledged to contribute to fault-proneness. In this paper, we explore the relationship between faults and coupling. Two releases from each of three open-source Eclipse projects (six releases in total) were used as an empirical basis and coupling and fault data extracted from those systems. A contrasting coupling profile between fault-free and fault-prone classes was observed and this result was statistically supported. Object-oriented (OO) classes with low values of fan-in (incoming coupling) and fan-out (outgoing coupling) appeared to support fault-free classes, while classes with high fan-out supported relatively fault-prone classes. We also considered size as an influence on fault-proneness. The study thus emphasizes the importance of minimizing coupling where possible (and particularly that of fan-out); failing to control coupling may store up problems for later in a system's life; equally, controlling class size should be a concomitant goal.","Title: Empirical Analysis on the Satisfaction of IT Employees Comparing XP Practices with Other Software Development Methodologies\nYear: 2004\n\nAbstract: Job satisfaction has been studied by economists and psychologists. We believe this factor is very important in that it influences the effectiveness of the software development process. This paper reports the first results of a comparative analytic study on the job satisfaction of developers that use XP practices and others that do not use XP practices. By determining the factors that are highly valued by developers, the research can provide insight in currently practised software development processes, and help make changes that increase their strategic value.","Title: An Analysis of Bug Distribution in Object Oriented Systems\nYear: 2009\n\nAbstract: We introduced a new approach to describe Java software as graph, where nodes\nrepresent a Java file - called compilation unit (CU) - and an edges represent a\nrelations between them. The software system is characterized by the degree\ndistribution of the graph properties, like in-or-out links, as well as by the\ndistribution of Chidamber and Kemerer metrics computed on its CUs. Every CU can\nbe related to one or more bugs during its life. We find a relationship among\nthe software system and the bugs hitting its nodes. We found that the\ndistribution of some metrics, and the number of bugs per CU, exhibit a\npower-law behavior in their tails, as well as the number of CUs influenced by a\nspecific bug. We examine the evolution of software metrics across different\nreleases to understand how relationships among CUs metrics and CUs faultness\nchange with time.","Title: An Empirical Study Of Software Metrics For Assessing The Phases Of An Agile Project\nYear: 2012\n\nAbstract: We present an analysis of the evolution of a Web application project developed with object-oriented technology and an agile process. During the development we systematically performed measurements on the source code, using software metrics that have been proved to be correlated with software quality, such as the Chidamber and Kemerer suite and Lines of Code metrics. We also computed metrics derived from the class dependency graph, including metrics derived from Social Network Analysis. The application development evolved through phases, characterized by a different level of adoption of some key agile practices - namely pair programming, test-based development and refactoring. The evolution of the metrics of the system, and their behavior related to the agile practices adoption level, is presented and discussed. We show that, in the reported case study, a few metrics are enough to characterize with high significance the various phases of the project. Consequently, software quality, as measured using these metrics, seems directly related to agile practices adoption.","Title: Studying Lean-Kanban Approach Using Software Process Simulation\nYear: 2011\n\nAbstract: We developed an event-driven simulator of the Kanban process a WIP limited pull system visualized by the Kanban board. WIP (work in process) represent the capacity in the activity to perform features simoultaneously. The simulator is fully object-oriented, and its design model reflects the objects of the Lean software development domain. We used this simulator to assess comparatively WIP-limited and unlimited processes. We also studied the optimum values of the working item limits in the activities, using a paradigmatic case of 4 activities and 100 work items. The cost function used is equal to the total time needed to complete the project, plus a weighted sum of the limits themselves. We performed an exhaustive search on all the admissible values of the solution; finding sensible optimal values, and a non-trivial behavior of the cost function in the optimization space. This demonstrates the feasibility and usefulness of the approach.","Title: 2010 ICSE workshop on emerging trends in software metrics\nYear: 2010\n\nAbstract: This paper reports on the 2010 ICSE Workshop on Emerging Trends in Software Metrics (WETSoM 2010) held on Tuesday 4 May 2010 in Cape Town, South Africa, as part of the ICSE workshop series. The goal of this workshop was to bring together researchers and practitioners to discuss the progress of software metrics.","Title: Agile practices in a large organization: the experience of poste Italiane\nYear: 2007\n\nAbstract: In this paper we show how agile practices have been used at the Poste Italiane for building a monitoring system of its complex IT infrastructure. The system, called Datamart, is built upon the existing monitoring infrastructure. A testing framework has been developed for performing assertion checking either on existing legacy modules or on the new functionalities. This framework is currently used, and is able to process data coming from 100,000 distributed computers, enabling and improving their centralized control.","Title: An analysis of anti-micro-patterns effects on fault-proneness in large Java systems\nYear: 2012\n\nAbstract: Micro patterns are similar to design patterns, but are at a lower level of abstraction, closer to the implementation. Anti patterns are micro patterns not respecting the prescriptions of good Object Oriented programming practices. In this paper, we use the definitions introduced by Arcelli and Maggioni [3] in order to study the evolution of five particular micro patterns (anti patterns) in different releases of the Eclipse and NetBeans systems, and the correlations between anti patterns and faults. Our analysis confirms previous findings regarding the high coverage of micro patterns onto the system classes, and show that anti patterns not only represent bad Object Oriented programming practices, but may also be associated to the production of lower quality software, since they present a fault proneness significantly enhanced.","Title: An Empirical Study of Refactoring in the Context of FanIn and FanOut Coupling\nYear: 2011\n\nAbstract: The aim of refactoring is to reduce software complexity and hence simplify the maintenance process. In this paper, we explore the impact of refactorings on \"FanIn\" and \"FanOut\" coupling metrics through extraction of refactoring data from multiple releases of five Java open-source systems, We first considered how a single refactoring modified these metric values, what happened when refactorings had been applied to a single class in unison and finally, what influence a set of refactorings had on the shape of Fan In and Fan Out distributions. Results indicated that, on average, refactored classes tended to have larger FanIn and Fan Out values when compared with non-refactored classes. Where evidence of multiple (different) refactorings applied to the same class was found, the net effect (in terms of FanIn and Fan Out coupling values) was negligible.","Title: A case study of the use of Open Source CMS in Public Administrations\nYear: 2012\n\nAbstract: The use of Open Source Content Management System (CMS) on the Web is increasing thanks to the flexibility provided by these systems and the semplification of the management operations. The use of CMS is simple for non programming experts. Many Public Administrations have adopted Open Source CMS for their websites, greatly reducing operating costs and ensuring scalability, advanced performance and privacy to the users. A special opportunity for a case study are the Italian municipalities. In fact the Italian Government, since January 2011, has required all municipalities to develop a website publishing all the public acts and everything related to the citizen-municipality communication (information regarding office hours, salaries of public employees, news and so on). Our study aims to quantify through empirical studies, the use of Open Source CMS in Italian municipalities websites in response to government regulations. Our results show that the number of Italian Public Administrations that use Open Source CMS is increasing, and our case study highlights how the adoption of Open Source CMS has increased about 5% in a year, leading to the 21.2% the percentage of Open Source CMS, with respect to the 15.8% registered in 2011.","Title: Assessing traditional and new metrics for object-oriented systems.\nYear: 2010\n\nAbstract: We present an extensive analysis of software metrics for 111 object-oriented systems written in Java. For each system, we considered 18 traditional metrics such as LOC and Chidamber and Kemerer metrics, as well as metrics derived from complex network theory and social network analysis. These metrics were computed at class level. We also considered two metrics at system level, namely the total number of classes and interfaces, and the fractal dimension. We discuss the distribution of these metrics, and their correlation, both at class and at system level. We found that most metrics follow a leptokurtotic distribution. Only a couple of metrics have patent normal behavior while three others are very irregular, and even bimodal. The statistics gathered allow us to study and discuss the variability of metrics along different systems, and to devise a roadmap for further research.","Title: Communication flow in open source projects: an analysis of developers' mailing lists\nYear: 2007\n\nAbstract: Team work is defined by different factors: team size, members role, interactions and communication. In Agile Methodologies communication plays a fundamental role. Some research studies have found that the involvement of team members in a project also depends on the information provided. Communication also depends on the tool employed. The aim of this paper is to analyze communication among team members of the 9 most active projects on Source-forge provided of a developers' mailing list. In particular we tried to investigate how and why members post to developers' mailing list and the use that different members make of these mailing lists.","Title: Study of the Competition between Proprietary Software Firms and Free/Libre Open Source Software Firms Using a Simulation Model.\nYear: 2011\n\nAbstract: In recent years, a very important structural change in the software industry took place, with an increasing number of firms that got involved in Free/Libre Open Source Software (FLOSS) development communities. FLOSS communities and products have been studied as complementary to proprietary software companies and products. In this paper we propone a business model for the software market, and in particular we analyze the competition between proprietary software firms and FLOSS firms. Our software market is a system where each agent is independent of each other in the choice about buying or selling software products or services. The proposed work aims to analyze the influence of FLOSS firms producing both software and services in vertical software markets, nowadays mostly dominated by large proprietary firms. The findings show that FLOSS firms are able to compete with proprietary firms, though in the end a monopoly or oligopoly of the latters emerges. The ousted FLOSS firms, however, survive longer than proprietary ones, when these are not able to compete in the market.","Title: Developing a Tool Supporting XP Process\nYear: 2003\n\nAbstract: We present the development of XPSwiki, a tool supporting the XP practices for requirement gathering and project management - user stories and the Planning Game. XPSwiki has been developed in Smalltalk using Squeak Swiki technology, and is accessed through the Internet in a user friendly, agile way. It keeps track of multiple projects, each with its releases, iterations, user stories, acceptance tests, and tasks. XPSwiki allows project tracking, and automates documentation generation and metrics collection. It is presently in use in real software development environments.","Title: Micro Patterns in Agile Software.\nYear: 2013\n\nAbstract: In this paper we present a study on micro patterns in different releases of two software systems developed with Object Oriented technologies and Agile process. Micro patterns are design decisions in code that can be easily automatically recognised. Gil and Maman introduced the concept to support providing objective assessment of design decisions [1]. They catalogued 27 micro patterns that capture a variety of programming practices in Java. Micro patterns can be a useful metrics in order to measure the quality of software by showing that certain categories of micro patterns are more fault prone than others, and that the classes that do not correspond to any category of micro patterns are more likely to be faulty. In our study we present some empirical results on two case studies of systems developed with Agile methodologies, and compare them to previous results obtained for non Agile systems. In particular we have verified that the distribution of micro patterns in a software system developed using Agile methodologies does not differ from the distribution studied in other systems, and that the micro patterns fault-proneness is about the same. We also analyzed how the distribution of micro patterns changes in different releases of the same software system. We demonstrate that there is a relationship between the number of faults and the classes that do not match with any micro patterns. We found that these classes are more likely to be fault-prone than the others even in software developed with Agile methodologies.","Title: Workshop on Emerging Trends in Software Metrics (WETSoM 2010)\nYear: 2011\n\nAbstract: The Workshop on Emerging Trends in Software Metrics aims at bringing together researchers and practitioners to discuss the progress of software metrics. The motivation for this workshop is the low impact that software metrics has on current software development. The goals of this workshop are to critically examine the evidence for the effectiveness of existing metrics and to identify new directions for development of software metrics.","Title: The Distribution of Program Sizes and Its Implications: An Eclipse Case Study\nYear: 2009\n\nAbstract: A large software system is often composed of many inter-related programs of\ndifferent sizes. Using the public Eclipse dataset, we replicate our previous\nstudy on the distribution of program sizes. Our results confirm that the\nprogram sizes follow the lognormal distribution. We also investigate the\nimplications of the program size distribution on size estimation and quality\npredication. We find that the nature of size distribution can be used to\nestimate the size of a large Java system. We also find that a small percentage\nof largest programs account for a large percentage of defects, and the number\nof defects across programs follows the Weibull distribution when the programs\nare ranked by their sizes. Our results show that the distribution of program\nsizes is an important property for understanding large and complex software\nsystems.","Title: XPSwiki: an agile tool supporting the planning game\nYear: 2003\n\nAbstract: We present XPSwiki, a tool supporting the XP practices for requirement gathering and project management - user stories and the Planning Game. XPSwiki has been developed in Smalltalk using Squeak Swiki technology. It is accessed through the Internet in a user friendly agile way. XPSwiki keeps track of multiple projects, each with its releases, iterations, user stories, acceptance tests, and engineering tasks. It also manages teams of developers, who sign stories and tasks, and who pair-program them with another team member. XPSwiki allows project tracking and documentation, and is at present in use among real software firms.","Title: Simulation of software maintenance process, with and without a work-in-process limit.\nYear: 2013\n\nAbstract: A software maintenance process is important for reducing maintenance effort and for improving software quality. In recent years, the Lean-Kanban approach has been widely applied in software practice including software maintenance. This approach minimizes Work-in-Progress (WIP), which is the number of items that are worked on by the team at any given time, thus improving the maintenance process. In this paper, we describe our simulation studies, which show that the Lean-Kanban approach can indeed help reduce the average time needed to complete maintenance requests. We develop a process simulator that can simulate both existing maintenance processes that do not use a WIP limit and that adopt it. We perform two case studies using real maintenance data collected from a Microsoft project and from a Chinese software firm. The main results of our study are twofold. First, we demonstrate that it is possible to effectively model and simulate, using actors and events, a maintenance process where a flow of issues is processed through a sequence of activities, correctly reproducing key statistics of real data. Second, our results confirm that the WIP-limited process could be useful to increase the efficiency of software maintenance, as reported in previous industrial practices. Copyright (c) 2013 John Wiley & Sons, Ltd.","Title: A Ontology-based Process Modelling for XP\nYear: 2003\n\nAbstract: In this paper we describe the Extreme Programming Ontology(XPO), a formal model specifying the main conceptsused in the Extreme Programming methodology and theirproperties. XPO's modular structure was developed usingthe usual normative top down approach to software engineeringprocess modeling. It relies on a set of core componentsrooted in three main concepts: Organisational Role,Product and Phase. Besides being useful for indexing relevantdocuments and XP artifacts such as user stories andWiki pages, XPO is aimed at being a sound basis for non-intrusiveanalysis of agile processes, mining process dataabout programmers activity and repositories content in orderto extract new concepts potentially identifying criticalfactors in agile software development. Extension to XPOare also discussed, including other Agile Methodologiesand more general Software Engineering concepts.","Title: Distributed Scrum in research project management\nYear: 2007\n\nAbstract: Can research projects be agile? In this paper we describe our proposal of applying Scrum for the management of an European research project aimed at developing an agent-based software platform for European economic policy design. The use of an agile, adaptive methodology is justified because successful research projects are complex, unstable processes, that should be continuously adapted along their way. We describe in detail the roles, artifacts and practices of the proposed process, and the first steps of its adoption.","Title: An Agile Development Process and Its Assessment Using Quantitative Object-Oriented Metrics\nYear: 2008\n\nAbstract: The development of a Web application using agile practices is presented, characterized by empirical software measurements that have been performed throughout the duration of the project. The project is the specialization of jAPS, an open source Java Web portal generation framework, for building a Research Register management system. The agile principles and practices used in the project are described and discussed. During the various phases of the project some key agile practices, such as pair programming, test-based development and refactoring, were used at different levels of adoption. The evolution of some object-oriented metrics of the system, and their behavior related to the agile practices adoption level is presented and discussed, showing that software quality, as measured using standard 00 metrics, looks directly related to agile practices adoption.","Title: A Comparative Study of Scrum and Kanban Approaches on a Real Case Study Using Simulation.\nYear: 2012\n\nAbstract: We present the application of software process modeling and simulation using an agent-based approach to a real case study of software maintenance. The original process used PSP/TSP; it spent a large amount of time estimating in advance maintenance requests, and needed to be greatly improved. To this purpose, a Kanban system was successfully implemented, that demonstrated to be able to substantially improve the process without giving up PSP/TSP. We customized the simulator and, using input data with the same characteristics of the real ones, we were able to obtain results very similar to that of the processes of the case study, in particular of the original process. We also simulated, using the same input data, the possible application of the Scrum process to the same data, showing results comparable to the Kanban process.","Title: Refactoring and its Relationship with Fan-in and Fan-out: An Empirical Study\nYear: 2012\n\nAbstract: Many of the seventy-two refactorings originally proposed by Fowler have the aim of reducing coupling in classes and thereby aiding programmer maintainability. While we might assume that classes with high levels of coupling would be the targets for refactoring more often than any other type of class, no empirical evidence exists to support this assumption. In this paper, we explore that open question through the extraction of fifty-two of Fowler's catalog of refactorings drawn from versions of four open-source systems. We compare the coupling characteristics of each set of refactored classes with the corresponding set of non-refactored classes (those for which no refactorings were applied across versions), we seek to inform one overriding research question: is there a significant difference between these two sets of classes in terms of their level of coupling? Results showed that highly-coupled classes were more prone to refactoring, particularly through a set of 'core' refactorings. However, wide variations were found across systems for our measures of coupling namely, fan-in and fan-out. Specific individual refactorings were also explored to gain an understanding of why these differences may have occurred.","Title: Entropy Of Some Ck Metrics To Assess Object-Oriented Software Quality\nYear: 2013\n\nAbstract: The term \"software entropy\" refers to the tendency for software, over time, to become difficult and costly to maintain. A software system that undergoes continuous change, such as having new functionality added to its original design, will eventually become more complex and can become disorganized as it grows, losing its original design structure.A recent study show that software degradation may be measured using the WMC expressed in terms of Shannon entropy. In this paper we extended the empirical analyses also to RFC and CBO since these CK metrics have been shown to be correlated with fault-proneness of OO classes.We analyzed various releases of the publicly available Eclipse and Netbeans software systems, calculating the entropy of some CK metrics for every release analyzed. The validity is shown through a direct measure of software quality such as the number of detected defects. Our results display a very good correlation between the entropy of CBO and RFC and the number of bugs for Eclipse and Netbeans.Complexity and quality metrics are in general computed on every system module while the entropy is just a scalar number that characterizes a whole system, this result suggests that the entropy of some CK metrics could be considered as a global quality metric for large software systems. Our results need, however, to be confirmed for other large software systems.","Title: The fractal dimension metric and its use to assess object-oriented software quality.\nYear: 2011\n\nAbstract: We present a study were software systems are considered as complex networks which have a self-similar structure under a length-scale transformation. On such complex software networks we computed a self-similar coefficient, also known as fractal dimension, using \"the box counting method\". We analyzed various releases of the publically available Eclipse software systems, calculating the fractal dimension for twenty sub-projects, randomly chosen, for every release, as well as for each release as a whole. Our results display an overall consistency among the sub-projects and among all the analyzed releases. We found a very good correlation between the fractal dimension and the number of bugs for Eclipse and for twenty sub-projects. Since the fractal dimension is just a scalar number that characterizes a whole system, while complexity and quality metrics are in general computed on every system module, this result suggests that the fractal dimension could be considered as a global quality metric for large software systems. Our results need however to be confirmed for other large software systems.","Title: Study of the Evolution of an Agile Project Featuring a Web Application Using Software Metrics\nYear: 2008\n\nAbstract: We present an agile process used for the development of a Web application written in Java, devised by choosing a set of proven agile practices taken by existing popular agile methodologies. During the project, we regularly measured the software using Chidamber and Kemerer object-oriented metrics suite, and other metrics. The application development evolved through phases, characterized by a different level of adoption of some key agile practices --- such as pair programming, test-based development and refactoring. The evolution of the OO metrics of the system, and their behavior related to the agile practices adoption level is presented and discussed, showing that soft ware quality, as measured using standard OO metrics, looks directly related to agile practices adoption.","Title: Power-Laws in a Large Object-Oriented Software System\nYear: 2007\n\nAbstract: We present a comprehensive study of an implementation of the Smalltalk object oriented system, one of the first and purest object-oriented programming environment, searching for scaling laws in its properties. We study ten system properties, including the distributions of variable and method names, inheritance hierarchies, class and method sizes, system architecture graph. We systematically found Pareto - or sometimes log-normal - distributions in these properties. This denotes that the programming activity, even when modeled from a statistical perspective, can in no way be simply modeled as a random addition of independent increments with finite variance, but exhibits strong organic dependencies on what has been already developed. We compare our results with similar ones obtained for large Java systems, reported in the literature or computed by ourselves for those properties never studied before, showing that the behavior found is similar in all studied object oriented systems. We show how the Yule process is able to stochastically model the generation of several of the power-laws found, identifying the process parameters and comparing theoretical and empirical tail indexes. Lastly, we discuss how the distributions found are related to existing object-oriented metrics, like Chidamber and Kemerer's, and how they could provide a starting point for measuring the quality of a whole system, versus that of single classes. In fact, the usual evaluation of systems based on mean and standard deviation of metrics can be misleading. It is more interesting to measure differences in the shape and coefficients of the data?s statistical distributions.","Title: On the influence of maintenance activity types on the issue resolution time\nYear: 2014\n\nAbstract: The ISO/IEC 14764 standard specifies four types of software maintenance activities spanning the different motivations that software engineers have while performing changes to an existing software system. Undoubtedly, this classification has helped in organizing the workflow within software projects, however for planning purposes the relative time differences for the respective tasks remains largely unexplored. In this empirical study, we investigate the influence of the maintenance type on issue resolution time. From GitHub's issue repository, we analyze more than 14000 issue reports taken from 34 open source projects and classify them as corrective, adaptive, perfective or preventive maintenance. Based on this data, we show that the issue resolution time depends on the maintenance type. Moreover, we propose a statistical model to describe the distribution of the issue resolution time for each type of maintenance activity. Finally, we demonstrate the usefulness of this model for scheduling the maintenance workload.","Title: Are Some Refactorings Attached to Fault-Prone Classes and Others to Fault-Free Classes?\nYear: 2014\n\nAbstract: A topical and relevant issue in the area of refactoring is the nature and characteristics of classes to which refactorings are applied. In particular, if we scrutinise the total set of refactorings applied to the classes of a system over different releases, which refactorings are applied to fault-prone classes and which to fault-free classes? In this paper, we explore that facet of refactoring. Refactorings applied between six releases of three Eclipse packages are used as a basis of the study and the Ref-Finder tool used to extract up to sixty-five different refactorings. Interestingly, results showed that refactorings applied to highly fault-prone classes differed significantly from those applied to fault-free classes, in particular related to the 'rename method' refactoring; a corresponding trend for the 'move method' and 'move field' refactoring was found in 'fault-free' classes over the period while the add and remove parameter refactorings tended to remain constant. The research offers an insight into refactoring behaviour in light of faults (or no faults).","Title: Applications of simulated annealing for the design of special digital filters\nYear: 1992\n\nAbstract: The authors describe the salient features of using a simulated annealing (SA) algorithm in the context of designing digital filters with coefficient values expressed as the sum of power of two. A procedure for linear phase digital filter design, using this algorithm, is presented and tested, yielding results as good as those for known optimal methods. The algorithm is then applied to the design of Nyquist filters, optimizing at the same time both frequency response and intersymbol interference, and to the design of cascade form finite-impulse-response (FIR) filters. The drawback of using SA is that the computation time is on the order of 1-2 h for each filter design, on the Sun 3/60. However, this was more than compensated by the versatility of the new algorithm, which can be used to design filters with multiple constraints","Title: Entropy of the degree distribution and object-oriented software quality\nYear: 2012\n\nAbstract: The entropy of degree distribution has been considered from many authors as a measure of a network's heterogeneity and consequently of the resilience to random failures. In this paper we propose the entropy of degree distribution as a new measure of software quality.\n\nWe present a study were software systems are considered as complex networks which are characterized by heterogeneous distribution of links. On such complex software networks we computed the entropy of degree distribution. We analyzed various releases of the publically available Eclipse and Netbeans software systems, calculating the entropy of degree distribution for every release analyzed. Our results display a good correlation between the entropy of degree distribution and the number of bugs for Eclipse and Netbeans.\n\nComplexity and quality metrics are in general computed on every system module while the entropy is just a scalar number that characterizes a whole system, this result suggests that the entropy of degree distribution could be considered as a global quality metric for large software systems. Our results need however to be confirmed for other large software systems.","Title: Size estimation of Web applications through Web CMF Object\nYear: 2012\n\nAbstract: This work outlines a new methodology for estimating the size of Web applications developed with a Content Management Framework (CMF). The reason for proposing - through this work - a new methodology for size estimation is the realization of the inadequacy of the RWO method, which we had recently developed, in estimating the effort of the latest Web applications. The size metric used in the RWO method was found not to be well suited for Web applications developed through a CMF. In this work, we present the new key elements for analysis and planning, needed to define every important step in developing a Web application through a CMF. Using those elements, it is possible to obtain the size of such an application. We also present the experimental validation performed on a 7-project dataset, provided by an Italian software company.","Title: Measuring and Understanding the Effectiveness of JIRA Developers Communities\nYear: 2015\n\nAbstract: Tools for project management and issues/bugs tracking are becoming useful for governing the development process of Open Source software. Such tools simplify the communications process among developers and ensure the scalability of a project. The more information developers are able to exchange, the clearer are the goals, and the higher is the number of developers keen on joining and actively collaborating on a project. In this paper we present a preliminary empirical analysis of the communities-structure of developers in JIRA by analyzing 7 popular projects hosted in the repository. We analyze how these communities perform in terms of issue-resolution time of any given issue. The main contributions of this work are the confirmation of the existence of communities in developer networks, and the empirical finding that the issue resolution-time of any given issue is not correlated with the dimension of a developer community.","Title: Software development: do good manners matter?\nYear: 2015\n\nAbstract: A successful software project is the result of a complex process involving, above all, people. Developers are the key factors for the success of a software development process, not merely as executors of tasks, but as protagonists and core of the whole development process. This paper investigates social aspects among developers working on software projects developed with the support of Agile tools. We studied 22 open-source software projects developed using the Agile board of the JIRA repository. All comments committed by developers involved in the projects were analyzed and we explored whether the politeness of comments affected the number of developers involved and the time required to fix any given issue. Our results showed that the level of politeness in the communication process among developers does have an effect on the time required to fix issues and, in the majority of the analysed projects, it had a positive correlation with attractiveness of the project to both active and potential developers. The more polite developers were, the less time it took to fix an issue.","Title: Simulation of the competition among traditional and on-demand software vendors\nYear: 2016\n\nAbstract: AbstractIn this work, we propose a simulation model to study the software market, and in particular the competition among on-premise and on-demand vendors of customer relationship management software. In this market two kinds of vendors, on-premise and on-demand vendors, and corporate customers interact. Vendors have an initial capital and a specific number of developers. They invest a fraction of their capital in their product quality and try to sell their products making a profit. Corporate customers buy software products for their employees, trying to minimize their cost and maximize their utility. The proposed model is able to reproduce the real trends of the market as regards both the price of software and the market shares of the vendors, despite there is relatively little empirical evidence available about the various mechanisms at play, and few experimental data. Our model could be used as a tool to forecast future market trends, or to plan business policies of investment and pricing. Any firm could calibrate this model depending on its past business trends in order to obtain simple guidelines to follow in the planning of business winning strategies.","Title: Web framework points: an effort estimation methodology for Web application development using a content management framework\nYear: 2015\n\nAbstract: Web applications are among the most popular and relevant kinds of application. Most Web applications are developed using a content management framework CMF. CMF helps to accelerate the publication of large amounts of information and the development of Web applications. However, developing Web applications through CMF is not exempt from cost and time overruns, as in traditional software projects. Currently, there is no estimation model able to adequately measure the effort of developing a Web application.","Title: A Curated Benchmark Collection of Python Systems for Empirical Studies on Software Engineering\nYear: 2015\n\nAbstract: The aim of this paper is to present a dataset of metrics associated to the first release of a curated collection of Python software systems. We describe the dataset along with the adopted criteria and the issues we faced while building such corpus. This dataset can enhance the reliability of empirical studies, enabling their reproducibility, reducing their cost, and it can foster further research on Python software.","Title: The JIRA Repository Dataset: Understanding Social Aspects of Software Development\nYear: 2015\n\nAbstract: Issue tracking systems store valuable data for testing hypotheses concerning maintenance, building statistical prediction models and recently investigating developers \\\"affectiveness\\\". In particular, the Jira Issue Tracking System is a proprietary tracking system that has gained a tremendous popularity in the last years and offers unique features like the project management system and the Jira agile kanban board. This paper presents a dataset extracted from the Jira ITS of four popular open source ecosystems (as well as the tools and infrastructure used for extraction) the Apache Software Foundation, Spring, JBoss and CodeHaus communities. Our dataset hosts more than 1K projects, containing more than 700K issue reports and more than 2 million issue comments. Using this data, we have been able to deeply study the communication process among developers, and how this aspect affects the development process. Furthermore, comments posted by developers contain not only technical information, but also valuable information about sentiments and emotions. Since sentiment analysis and human aspects in software engineering are gaining more and more importance in the last years, with this repository we would like to encourage further studies in this direction.","Title: 15 Years of Application of Statistical Physics Methods to the Study of Software Systems\nYear: 2015\n\nAbstract: Despite the fact that software development is a design optimization process with nothing random, large software systems exhibit statistical properties that can be studied with the approach of statistical physics. In this keynote talk, I shall recall the history of the application of statistical physics methods to the study of software systems, since its inception at the beginning of 2000.","Title: The Predictor Impact of Web Search Media on Bitcoin Trading Volumes.\nYear: 2015\n\nAbstract: In the last decade, Web 2.0 services have been widely used as communication media. Due to the huge amount of available information, searching has become dominant in the use of Internet. Millions of users daily interact with search engines, producing valuable sources of interesting data regarding several aspects of the world. Search queries prove to be a useful source of information in financial applications, where the frequency of searches of terms related to the digital currency can be a good measure of interest in it. Bitcoin, a decentralized electronic currency, represents a radical change in financial systems, attracting a large number of users and a lot of media attention. In this work we studied the existing relationship between Bitcoinâs trading volumes and the queries volumes of Google search engine. We achieved significant cross correlation values, demonstrating search volumes power to anticipate trading volumes of Bitcoin currency.","Title: Bitcoin Spread Prediction Using Social and Web Search Media.\nYear: 2015\n\nAbstract: In the last decade, Web 2.0 services such as blogs, tweets, forums, chats, email etc. have been widely used as communication media, with very good results. Sharing knowledge is an important part of learning and enhancing skills. Furthermore, emotions may affect decisionmaking and individual behavior. Bitcoin, a decentralized electronic currency system, represents a radical change in financial systems, attracting a large number of users and a lot of media attention. In this work, we investigated if the spread of the Bitcoin’s price is related to the volumes of tweets or Web Search media results. We compared trends of price with Google Trends data, volume of tweets and particularly with those that express a positive sentiment. We found significant cross correlation values, especially between Bitcoin price and Google Trends data, arguing our initial idea based on studies about trends in stock and goods market.","Title: Understanding approval rating of agile project management tools using Twitter\nYear: 2015\n\nAbstract: The role of managing a software project can be extremely complicated, requiring many teams and organizational resources. Many people are acknowledging that Agile development is helpful to business, with an high increase over the last years in the number of people who believe that Agile helps companies to complete projects faster. Since there is a multitude of available tools that supports Agile methods, in this position paper we tried to understand which tools are most looked for using Web Search media, most mentioned and most appreciated through Social media. Furthermore, we applied automated Sentiment Analysis on shared short messages of users on Twitter, one of the most popular social networks, in order to analyze automatically developers' opinions, sentiments, evaluations and attitudes.","Title: A Preliminary Study on Mobile Apps Call Graphs through a Complex Network Approach.\nYear: 2015\n\nAbstract: Mobile apps are certainly a significant part of thesoftware industry market, and their centrality in our economicand social life is becoming more and more relevant as time goesby. In this preliminary work we propose a complex networkapproach, never used before in this field, to study mobileapps from a software engineering perspective. We analyzed thenetworks associated to 5 popular open-source android apps, and found that they share some properties with other kinds ofsoftware systems. We employed complex networks to obtain adeeper insight both on internal and external dependencies, aimingat studying the relationship between mobile apps dependenciesand software quality. Our goal is to better understand thenature of the existing relationship between mobile apps and theunderlying hosting operating system thus providing developerswith valuable information.","Title: Re-visiting the 'Maintainability Index' Metric from an Object-Oriented Perspective\nYear: 2015\n\nAbstract: The Maintainability Index (MI) metric was proposed in the early nineties to gauge ease of code maintenance. A high value of MI indicates well-constructed code which is easy to maintain and a low MI, the opposite. The metric has been criticized in the past and more recently for its applicability to code and validity more generally. Very few studies however have explored whether the MI correlates strongly with class features such as coupling, defects or size - features of object-oriented (OO) classes which are acknowledged to be surrogates of maintenance complexity. We explore the relationship between the MI and these four perspectives. Three releases of two Eclipse projects were used as the empirical basis of the study and the JHawk tool was used to extract class-based metrics. Significant correlations were found between class features and the MI for all class features except fan-in which measures incoming coupling and determined externally by 'using' classes, however, further exploration of the data revealed that the application of the MI to an OO paradigm was not particularly successful, due primarily to the influence of class size. For OO systems, we would thus warn against the use of MI as a measure of maintenance.","Title: The evolution of knowledge in the refactoring research field\nYear: 2015\n\nAbstract: Refactoring is certainly one of the most widespread practices used by developers to improve software quality. During the last two decades, it has been the subject of an increasing number of research studies. But how much do we know about the trends, and the appearance of new topics, in the research field of refactoring? What have researchers found about its application since the time of its first introduction? In this work we provide a preliminary analysis of the state of the art and the evolution of the research on refactoring. We attempt to represent the actual body of knowledge in this field through the analysis of its cognitive structure, leveraging science mapping methodology to focus on the most relevant concepts in this research area. We model the body of knowledge by mining bibliographic databases and by retrieving the co-occurrence of keywords. We have found that some different general themes can be recognized, but not all of them have the same role and equally catalyzed researchers' interest. In addition, we provide a preliminary analysis on the trends and directions as well as the hot topics we identified in the refactoring research field.","Title: Are bullies more productive?: empirical study of affectiveness vs. issue fixing time\nYear: 2015\n\nAbstract: Human Affectiveness, i.e., the emotional state of a person, plays a crucial role in many domains where it can make or break a team's ability to produce successful products. Software development is a collaborative activity as well, yet there is little information on how affectiveness impacts software productivity. As a first measure of this impact, this paper analyzes the relation between sentiment, emotions and politeness of developers in more than 560K Jira comments with the time to fix a Jira issue. We found that the happier developers are (expressing emotions such as JOY and LOVE in their comments), the shorter the issue fixing time is likely to be. In contrast, negative emotions such as SADNESS, are linked with longer issue fixing time. Politeness plays a more complex role and we empirically analyze its impact on developers' productivity.","Title: How diverse is your team? Investigating gender and nationality diversity in GitHub teams.\nYear: 2017\n\nAbstract: Building an effective team of developers is a complex task faced by both software companies and open source communities. The problem of forming a “dream”","Title: Measuring high and low priority defects on traditional and mobile open source software.\nYear: 2016\n\nAbstract: Software defects are the major cause for system failures. To effectively design tools and provide support for detecting and recovering from software failures, requires a deep understanding of defect features. In this paper we present an analysis of defect characteristics in two different open source software development domains: Mobile and Traditional. Our attention is focused on measuring the differences between High-Priority and Low-Priority defects. High or Low priority of a given defect is decided by a developer when creating a bug report for an issue tracking system. We sampled hundreds of real world bugs in hundreds of large and representative open-source projects. We used natural language text classification techniques to automatically analyse roughly 700,000 bug reports from the Bugzilla, Jira and Google Issues issue tracking systems. Results show that there are differences between High-Priority and Low-Priority defects classification in Mobile and Traditional development domains.","Title: A statistical comparison of Java and Python software metric properties.\nYear: 2016\n\nAbstract: This paper presents a statistical analysis of 20 opens ource object-oriented systems with the purpose of detecting differences in metrics distribution between Java and Python projects. We selected ten Java projects from the Java Qualitas Corpus and ten projects written in Python. For each system, we considered 10 class-level software metrics. We performed a best fit procedure on the empirical distributions through the log-normal distribution and the double Pareto distribution to identify differences between the two languages. Even though the statistical distributions for projects written in Java and Python may appear the same for lower values of the metric, performing the procedure with the double Pareto distribution for the Number of Local Methods metric reveals that major differences can be noticed along the queue of the distributions. On the contrary, the same analysis performed with the Number of Statements metric reveals that only the initial portion of the double Pareto distribution shows differences between the two languages. In addition, the dispersion parameter associated to the log-normal distribution fit for the total Number Of Methods can be used for distinguishing Java projects from Python projects.","Title: A case study on the relationship between code ownership and refactoring activities in a Java software system.\nYear: 2016\n\nAbstract: Refactoring, the activity of changing source code design without affecting its external behavior, is a widely used practice among developers, since it is considered to positively affect the quality of software systems. However, there are some \\\"human factors\\\" to be considered while performing refactoring, including developers knowledge of systems architecture. Recent studies showed how much \\\"people\\\" metrics, such as code ownership, might affect software quality as well. In this preliminary study we investigated the relationship between code ownership and refactoring activity performed by developers. This study can provide useful insights on who performs refactoring and help team leaders to properly manage human resources during software development.","Title: Arsonists or Firefighters? Affectiveness in Agile Software Development.\nYear: 2016\n\nAbstract: In this paper, we present an analysis of more than 500 K comments from open-source repositories of software systems developed using agile methodologies. Our aim is to empirically determine how developers interact with each other under certain psychological conditions generated by politeness, sentiment and emotion expressed within developers’ comments. Developers involved in an open-source projects do not usually know each other; they mainly communicate through mailing lists, chat, and tools such as issue tracking systems. The way in which they communicate affects the development process and the productivity of the people involved in the project. We evaluated politeness, sentiment and emotions of comments posted by agile developers and studied the communication flow to understand how they interacted in the presence of impolite and negative comments (and vice versa). Our analysis shows that “firefighters” prevail. When in presence of impolite or negative comments, the probability of the next comment being impolite or negative is 13 % and 25 %, respectively; ANGER however, has a probability of 40 % of being followed by a further ANGER comment. The result could help managers take control the development phases of a system, since social aspects can seriously affect a developer’s productivity. In a distributed agile environment this may have a particular resonance.","Title: The emotional side of software developers in JIRA.\nYear: 2016\n\nAbstract: Issue tracking systems store valuable data for testing hypotheses concerning maintenance, building statistical prediction models and (recently) investigating developer affectiveness. For the latter, issue tracking systems can be mined to explore developers emotions, sentiments and politeness---affects for short. However, research on affect detection in software artefacts is still in its early stage due to the lack of manually validated data and tools.\n\nIn this paper, we contribute to the research of affects on software artefacts by providing a labeling of emotions present on issue comments.\n\nWe manually labeled 2,000 issue comments and 4,000 sentences written by developers with emotions such as love, joy, surprise, anger, sadness and fear. Labeled comments and sentences are linked to software artefacts reported in our previously published dataset (containing more than 1K projects, more than 700K issue reports and more than 2 million issue comments). The enriched dataset presented in this paper allows the investigation of the role of affects in software development.","Title: How Do Python Programs Use Inheritance? A Replication Study\nYear: 2015\n\nAbstract: In this work we present an empirical study on the use of inheritance in a curated corpus of Python systems. Replicating a study preformed on Java, we analyzed a collection of 51 software systems written in Python, and investigated how inheritance is effectively used by Python developers in practice through a convenient set of inheritance metrics. Our results suggest that on average fewer classes inherit from other classes than in Java, but more classes are inherited from. We also see a sort of symmetry relating the number of ancestors and the number of descendants in each system.","Title: Software Startups - A Research Agenda.\nYear: 2016\n\nAbstract: Software startup companies develop innovative, software-intensive products within limited time frames and with few resources, searching for sustainable and scalable business models. Software startups are quite distinct from traditional mature software companies, but also from micro-, small-, and medium-sized enterprises, introducing new challenges relevant for software engineering research. This paper's research agenda focuses on software engineering in startups, identifying, in particular, 70+ research questions in the areas of supporting startup engineering activities, startup evolution models and patterns, ecosystems and innovation hubs, human aspects in software startups, applying startup concepts in non-startup environments, and methodologies and theories for startup research. We connect and motivate this research agenda with past studies in software startup research, while pointing out possible future directions. While all authors of this research agenda have their main background in Software Engineering or Computer Science, their interest in software startups broadens the perspective to the challenges, but also to the opportunities that emerge from multi-disciplinary research. Our audience is therefore primarily software engineering researchers, even though we aim at stimulating collaborations and research that crosses disciplinary boundaries. We believe that with this research agenda we cover a wide spectrum of the software startup industry current needs.","Title: Estimating Story Points from Issue Reports.\nYear: 2016\n\nAbstract: Estimating the effort of software engineering tasks is notoriously hard but essential for project planning. The agile community often adopts issue reports to describe tasks, and story points to estimate task effort. In this paper, we propose a machine learning classifier for estimating the story points required to address an issue. Through empirical evaluation on one industrial project and eight open source projects, we demonstrate that such classifier is feasible. We show that ---after an initial training on over 300 issue reports--- the classifier estimates a new issue in less than 15 seconds with a mean magnitude of relative error between 0.16 and 0.61. In addition, issue type, summary, description, and related components prove to be project dependent features pivotal for story point estimation.","Title: Assessment Of Approaches For The Analysis Of Refactoring Activity On Software Repositories An Empirical Study\nYear: 2016\n\nAbstract: Refactoring is the practice of changing source code without altering its external behavior. It is widely used since it is acknowledged to have a positive effect on software quality. However, different studies on the impact of Refactoring on software quality led to contrasting results. This might be due also to the fact that research works on Refactoring rely on different approaches to collect information about the refactoring activity on software repositories. Having a reliable approach is fundamental to draw solid conclusions. In this paper we empirically analyzed two of the most popular approaches to detect the refactoring activity on software repositories, finding that they return different outcomes. These results should be taken into account by researchers while designing the experimental settings of their studies.","Title: Blockchain-oriented Software Engineering: Challenges and New Directions.\nYear: 2017\n\nAbstract: In this work, we acknowledge the need for software engineers to devise specialized tools and techniques for blockchain-oriented software development. Ensuring effective testing activities, enhancing collaboration in large teams, and facilitating the development of smart contracts all appear as key factors in the future of blockchain-oriented software development.","Title: Is Bitcoin's Market Predictable? Analysis of Web Search and Social Media.\nYear: 2015\n\nAbstract: In recent years, Internet has completely changed the way real life works. In particular, it has been possible to witness the online emergence of web 2.0 services that have been widely used as communication media. On one hand, services such as blogs, tweets, forums, chats, email have gained wide popularity. On the other hand, due to the huge amount of available information, searching has become dominant in the use of Internet. Millions of users daily interact with search engines, producing valuable sources of interesting data regarding several aspects of the world. Bitcoin, a decentralized electronic currency, represents a radical change in financial systems, attracting a large number of users and a lot of media attention. In this work we studied whether Bitcoin’s trading volume is related to the web search and social volumes about Bitcoin. We investigated whether public sentiment, expressed in large-scale collections of daily Twitter posts, can be used to predict the Bitcoin market too. We achieved significant cross correlation outcomes, demonstrating the search and social volumes power to anticipate trading volumes of Bitcoin currency.","Title: Measuring Affectiveness and Effectiveness in Software Systems.\nYear: 2017\n\nAbstract: The summary presented in this paper highlights the results obtained in a four-years project aiming at analyzing the development process of software artifacts from two points of view: Effectiveness and Affectiveness. The first attribute is meant to analyze the productivity of the Open Source Communities by measuring the time required to resolve an issue, while the latter provides a novel approach for studying the development process by analyzing the affectiveness ex-pressed by developers in their comments posted during the issue resolution phase. Affectivenes is obtained by measuring Sentiment, Politeness and Emotions. All the study presented in this summary are based on Jira, one of the most used software repositories.","Title: Connecting the Dots: Measuring Effectiveness and Affectiveness in Software Systems.\nYear: 2017\n\nAbstract: The summary presented in this paper highlights the results obtained in a four-year project aiming at analyzing the development process of software artifacts from two points of view: Effectiveness and Affectiveness. The first attribute is meant to analyze the productivity of the Open Source Communities by measuring the time required to resolve an issue, while the latter provides a novel approach for studying the development process by analyzing the affectiveness expressed by developers in their comments posted during the issue resolution phase. Affectivenes is obtained by measuring Sentiment, Politeness and","Title: On the randomness and seasonality of affective metrics for software development.\nYear: 2017\n\nAbstract: The role of emotions and human aspects of software development is gaining increasing attention from both academics and developers. Software development is a complex activity, but since software is everywhere and it is necessary to achieve the highest quality possible. High technical skills, knowledge and competence of a software developer are not the only factors which influence the final product. Good attitudes, communication skills and good manners are equally as important. In this paper we focused our attention on understanding how developer sentiment and emotions evolved over time during the development process of 10 systems, considering affectiveness as time series. We studied seasonality, randomness and correlations. Results showed that there was not a significant correlation among joy and love, sadness and anger, sentiment and joy. Regarding seasonality and randomness, results showed that in the majority of the cases, the analysed time series are seasonal and not random.","Title: Using measurement and simulation for understanding distributed development processes in the cloud.\nYear: 2017\n\nAbstract: Organizations increasingly develop software in a distributed manner. The Cloud provides an environment to create and maintain software-based products and services. Currently, it is widely unknown which software processes are suited for Cloud-based development and what their effects in specific contexts are. This paper presents a process simulation to study distributed development in the Cloud. We contribute a simulation model, which helps analyzing different project parameters and their impact on projects carried out in the Cloud. The simulator helps reproducing activities, developers, issues and events in the project, and it generates statistics, e.g., on throughput, total time, and lead and cycle time. The aim of this simulation model is thus to analyze the tradeoffs regarding throughput, total time, project size, and team size. Furthermore, the modified simulation model aims to help project managers select the most suitable planning alternative. Based on observed projects in Finland and Spain, we simulated a distributed project using artificial and real data. Particularly, we studied the variables project size, team size, throughput, and total project duration. A comparison of the real project data with the results obtained from the simulation shows the simulation producing results close to the real data, and we could successfully replicate a distributed software project. By improving the understanding of distributed development processes, our simulation model thus supports project managers in their decision-making.","Title: Software Quality And Community Structure In Java Software Networks\nYear: 2017\n\nAbstract: We present a study of 600 Java software networks with the aim of characterizing the relationship among their defectiveness and community metrics. We analyze the community structure of such networks, defined as their topological division into subnetworks of densely connected nodes. A high density of connections represents a higher level of cooperation between classes, so a well-defined division in communities could indicate that the software system has been designed in a modular fashion and all its functionalities are well separated. We show how the community structure can be an indicator of well-written, high quality code by retrieving the communities of the analyzed systems and by ranking their division in communities through the built-in metric called modularity. We found that the software systems with highest modularity possess the majority of bugs, and tested whether this result is related to some confounding effect. We found two power laws relating the maximum defect density with two different metrics: the number of detected communities inside a software network and the clustering coeffcient. We finally found a linear correlation between clustering coefficient and number of communities. Our results can be used to make predictive hypotheses about software defectiveness of future releases of the analyzed systems.","Title: Banking on Blockchain: Costs Savings Thanks to the Blockchain Technology.\nYear: 2017\n\nAbstract: This paper looks at the challenges and opportunities of implementing blockchain technology across banking, providing food for thought about the potentialities of this disruptive technology. The blockchain technology can optimize the global financial infrastructure, achieving sustainable development, using more efficient systems than at present. In fact, many banks are currently focusing on blockchain technology to promote economic growth and accelerate the development of green technologies. In order to understand the potential of blockchain technology to support the financial system, we studied the actual performance of the Bitcoin system, also highlighting its major limitations, such as the significant energy consumption due to the high computing power required, and the high cost of hardware. We estimated the electrical power and the hash rate of the Bitcoin network, over time, and, in order to evaluate the efficiency of the Bitcoin system in its actual operation, we defined three quantities: \"economic efficiency\", \"operational efficiency\", and \"efficient service\". The obtained results show that by overcoming the disadvantages of the Bitcoin system, and therefore of blockchain technology, we could be able to handle financial processes in a more efficient way than under the current system.","Title: A Petri Nets Model for Blockchain Analysis.\nYear: 2018\n\nAbstract: A Blockchain is a global shared infrastructure where cryptocurrency transactions among addresses are recorded, validated and made publicly available in a peer-to-peer network. To date, the best known and important cryptocurrency is the bitcoin. In this paper, we focus on this cryptocurrency and in particular on the modeling of the Bitcoin Blockchain by using the Petri Nets formalism. The proposed ...","Title: Smart Contracts Software Metrics: a First Study.\nYear: 2018\n\nAbstract: Smart contracts (SC) are software codes which reside and run over a blockchain. The code can be written in different languages with the common purpose of implementing various kinds of transactions onto the hosting blockchain, They are ruled by the blockchain infrastructure and work in order to satisfy conditions typical of traditional contracts. The software code must satisfy constrains strongly context dependent which are quite different from traditional software code. In particular, since the bytecode is uploaded in the hosting blockchain, size, computational resources, interaction between different parts of software are all limited and even if the specific software languages implement more or less the same constructs of traditional languages there is not the same freedom as in normal software development. SC software is expected to reflect these constrains on SC software metrics which should display metric values characteristic of the domain and different from more traditional software metrics. We tested this hypothesis on the code of more than twelve thousands SC written in Solidity and uploaded on the Ethereum blockchain. We downloaded the SC from a public repository and computed the statistics of a set of software metrics related to SC and compared them to the metrics extracted from more traditional software projects. Our results show that generally Smart Contracts metrics have ranges more restricted than the corresponding metrics in traditional software systems. Some of the stylized facts, like power law in the tail of the distribution of some metrics, are only approximate but the lines of code follow a log normal distribution which reminds of the same behavior already found in traditional software systems.","Title: Why blockchain is important for software developers, and why software engineering is important for blockchain software (Keynote)\nYear: 2018\n\nAbstract: In the past few years, cryptocurrencies and blockchain applications has been one of the most rapidly emerging fields of computer science, leading to a strong demand of software applications. Several new projects have been emerging almost daily, with an impetus that was not seen since the days of the dawn of the Internet. However, the need of being timely on the market and the lack of experience in a brand new field led to epic disasters, such as those of DAO in 2016 and of Parity Ethereum wallet in 2017. Also, there have been several hacks successfully performed on cryptocurrency exchanges, the biggest being those of MtGox in 2014 (350 million US$), Bitfinex in 2016 (72 million US$), and Coincheck in 2017 (400 million US$). The application of sound SE practices to Blockchain software development, both for Smart Contract and generic Blockchain software, might be crucial to the success of this new field. Here the issues are the need for specific analysis and design methods, quality control through testing and metrics, security assessment and overall development process. At the same time, Blockchain development offers new opportunities, such as the certification of empirical data used for experiment; the ability to design processes where developers are paid upon completion of their tasks through Blockchain tokens, after acceptance tests performed using Smart Contracts; and more sound techniques enabling pay-per-use software, again using tokens.","Title: Smart contracts vulnerabilities: a call for blockchain software engineering?\nYear: 2018\n\nAbstract: Smart Contracts have gained tremendous popularity in the past few years, to the point that billions of US Dollars are currently exchanged every day through such technology. However, since the release of the Frontier network of Ethereum in 2015, there have been many cases in which the execution of Smart Contracts managing Ether coins has led to problems or conflicts. Compared to traditional Software Engineering, a discipline of Smart Contract and Blockchain programming, with standardized best practices that can help solve the mentioned problems and conflicts, is not yet sufficiently developed. Furthermore, Smart Contracts rely on a non-standard software life-cycle, according to which, for instance, delivered applications can hardly be updated or bugs resolved by releasing a new version of the software. In this paper we advocate the need for a discipline of Blockchain Software Engineering, addressing the issues posed by smart contract programming and other applications running on blockchains.We analyse a case of study where a bug discovered in a Smart Contract library, and perhaps \"unsafe\" programming, allowed an attack on Parity, a wallet application, causing the freezing of about 500K Ethers (about 150M USD, in November 2017). In this study we analyze the source code of Parity and the library, and discuss how recognised best practices could mitigate, if adopted and adapted, such detrimental software misbehavior. We also reflect on the specificity of Smart Contract software development, which makes some of the existing approaches insufficient, and call for the definition of a specific Blockchain Software Engineering.","Title: The ICO phenomenon and its relationships with ethereum smart contract environment\nYear: 2018\n\nAbstract: Initial Coin Offerings (ICO) are public offers of new cryptocurrencies in exchange of existing ones, aimed to finance projects in the blockchain development arena. In the last 8 months of 2017, the total amount gathered by ICOs exceeded 4 billion US$, and overcame the venture capital funnelled toward high tech initiatives in the same period. A high percentage of ICOs is managed through Smart Contracts running on Ethereum blockchain, and in particular to ERC-20 Token Standard Contract. In this work we examine 1387 ICOs, published on December 31, 2017 on icobench.com website, gathering information relevant to the assessment of their quality and software development management, including data on their development teams. We also study, at the same date, the financial data of 450 ICO tokens available on coinmarketcap.com website, among which 355 tokens are managed on Ethereum blochain. We define success criteria for the ICOs, based on the funds actually gathered, and on the behavior of the price of the related tokens, finding the factors that most likely influence the ICO success likeliness.","Title: Analysing Developers Affectiveness through Markov chain Models.\nYear: 2018\n\nAbstract: In this paper, we present an analysis of more than 500K comments from open-source repositories of software systems. Our aim is to empirically determine how developers interact with each other under certain psychological conditions generated by politeness, sentiment and emotion expressed in developersu0027 comments. Developers involved in open-source projects do not usually know each other; they mainly communicate through mailing lists, chat rooms, and tools such as issue tracking systems. The way in which they communicate affects the development process and the productivity of the people involved in the project. We evaluated politeness, sentiment, and emotions of comments posted by developers and studied the communication flow to understand how they interacted in the presence of impolite and negative comments (and vice versa). Our analysis shows that when in presence of impolite or negative comments, the probability of the next comment being impolite or negative is 14% and 25%, respectively; anger, however, has a probability of 40% of being followed by a further anger comment. The result could help managers take control the development phases of a system since social aspects can seriously affect a developeru0027s productivity. In a distributed environment this may have a particular resonance.","Title: ICOs Overview: Should Investors Choose an ICO Developed with the Lean Startup Methodology?\nYear: 2018\n\nAbstract: An Initial Coin Offering (ICO) is an innovative way to raise funds and launch a startup. It is also an opportunity to take part in a project, or in a DAO (Decentralized Autonomous Organization). The use of ICO is a global phenomenon that involves many nations and several business categories: ICOs collected over 5.2 billion dollars only in 2017. The success of an ICO is based on the credibility and innovativeness of project proposals. This fund-raising tool contains however some critical issues, such as the use of tokens that have no intrinsic value and do not generate direct liquidity, and the role of investors in the management of the startup. We analyzed if the Lean Startup methodology is helpful to face this critical aspects and we examined some ICOs in which the proposing team states explicitly that a lean startup approach is used.","Title: Fast wavelet transform assisted predictors of streaming time series.\nYear: 2018\n\nAbstract: We explore the shift variance of the decimated, convolutional Discrete Wavelet Transform, also known as Fast Wavelet Transform. We prove a novel theorem improving the FWT algorithm and implement a new prediction method suitable to the multiresolution analysis of streaming univariate datasets using compactly supported Daubechies Wavelets. An effective real value forecast is obtained synthesizing the one step ahead crystal and performing its inverse DWT, using an integrated group of estimating machines. We call Wa.R.P. (Wavelet transform Reduced Predictor) the new prediction method. A case study, testing a cryptocurrency exchange price series, shows that the proposed system can outperform the benchmark methods in terms of forecasting accuracy achieved. This result is confirmed by further tests performed on other time series. Developed in C++, Standard 2014 conformant, the code implementing the FWT and the novel Shift Variance Theorem is available to research purposes and to build efficient industrial applications.","Title: On measuring affects of github issues' commenters.\nYear: 2018\n\nAbstract: In this study, we analyzed issues and comments on GitHub projects and built collaboration networks dividing contributors into two categories: users and commenters. We identified as commenters those users who only post comments without posting any issues nor committing changes in the source code. Since previous studies showed that there is a link between a positive environment (regarding affectiveness) and productivity, our goal was to investigate commenters' contribution to the project concerning affectiveness.\n\nWe analyzed more than 370K comments from 100K issues of 25K contributors from 3 open source projects. We then calculated and compared the affectiveness of the issues' comments written by users and commenters in terms of sentiment, politeness, and emotions. We provide empirical evidence that commenters are less polite, less positive and in general they express a lower level of emotions in their comments than users. Our results also confirm that GitHub's contributors consist of different groups which behave differently, and this provides useful information for future studies in the field.","Title: Using simulation for understanding and reproducing distributed software development processes in the cloud.\nYear: 2018\n\nAbstract: Context: Organizations increasingly develop software in a distributed manner. The Cloud provides an environment to create and maintain software-based products and services. Currently, it is unknown which software processes are suited for Cloud-based development and what their effects in specific contexts are.","Title: Re-visiting a Test Taxonomy with Refactoring and Defect-fix Data.\nYear: 2018\n\nAbstract: In a previous empirical study by Bavota et al., multiple releases of three open-source systems reported the extent to which refactorings induced defect-fixes. In a much earlier study, van Deursen and Moonen (vDu0026M) provided a test taxonomy in which Fowleru0027s seventy-two refactorings were categorized according to the post refactoring test burden of each (i.e., the changes required to unit tests after each refactoring had been undertaken). A refactoring was categorized as u0027Type Bu0027 if it required no change to the original tests and u0027Type Eu0027 if significant changes were necessary. In this paper, we investigate nine refactorings spread across vDu0026Mu0027s taxonomy and the corresponding defect-fix data provided by Bavota et al., to explore the relationship between defect-fixes due to refactoring and vDu0026Mu0027s taxonomy. Results showed that, in contrast to our intuition, the most defect-fix prone refactorings were of Types C and D and not, as we thought, of Type E. The u0027Extract methodu0027 refactoring stood out as particularly u0027defect-fixu0027 inducing, suggesting that while it may solve one problem (i.e., in decomposing an excessively long method), it may well introduce other problems and required defect-fixes as a by-product.","Title: Mining Communication Patterns in Software Development: A GitHub Analysis.\nYear: 2018\n\nAbstract: Background: Studies related to human factors in software engineering are providing insightful information on the emotional state of contributors and the impact this has on the code. The open source software development paradigm involves different roles, and previous studies about emotions in software development have not taken into account what different roles might play when people express their feelings. Aim: We present an analysis of issues and commits on five GitHub projects distinguishing contributors between users and developers, and between one-commit and multi-commit developers. Method: We analyzed more than 650K comments from 130K issues of 64K contributors. We calculated emotions (love, joy, anger, sadness) and politeness of the comments related to the issues of the considered projects and introduced the definition of contributor fan-in and fan-out. Results: Results show that users and developers communicate differently as well as multi-commit developers and one-commit developers do. Conclusions: We provide empirical evidence that one-commit developers are more active and more polite in posting comments. Multi-commit developers are less active in posting comments, and while commenting, they are less polite than when commented.","Title: Initial Coin Offerings and Agile Practices.\nYear: 2018\n\nAbstract: An ICO (Initial Coin Offering) is an innovative way to fund projects based on blockchain. The funding is based on the selling of tokens by means of decentralized applications called smart contracts written in Solidity, a programming language specific for Ethereum blockchain. The ICOs work in a volatile context and it is crucial that the team is capable of handling constant changes. The Agile methods, proven practices enabling to develop software in presence of changing requirements, could be a means for managing uncertainty. The main goals of this work are to understand software engineering activities related to ICOs, recognize the ICOs developed using Agile methods, and make a comparison between ICOs and Agile ICOs. In addition, we perform a deeper analysis of Agile ICOs concerning project planning, software development, and code features. Our work shows that the roles of the people involved in an ICO can be compared to the typical roles of the SCRUM methodology. The majority of Agile ICOs use tool of testing before storing smart contract on blockchain. Finally, the application of volumetric and complexity software metrics shows that the files of Agile ICOs is on average shorter and less complex than in other smart contracts.","Title: A Blockchain Based System to Ensure Transparency and Reliability in Food Supply Chain.\nYear: 2018\n\nAbstract: We propose a blockchain oriented platform to secure storage origin provenance for food data. By exploiting the blockchain distributed and immutable nature the proposed system ensures the supply chain transparency with a view to encourage local region by promoting the smart food tourism and by increasing local economy. Thanks to the decentralized application platforms that makes us able to develop smart contracts, we define and implement a system that works inside the blockchain and guarantees transparency, reliability to all actors of the food supply chain. Food, in fact, is the most direct way to get in touch with a place. The touristic activities related to wine and food consumption and sale in fact influence the choice of a destination and may encourage the purchase of typical food also once tourists are back home to the country of origin. Touristic destinations must therefore be equipped with innovative tools that, in a context of Smart Tourism, guarantee the originality of the products and their traceability.","Title: An Agent Based Model to Analyze the Bitcoin Mining Activity and a Comparison with the Gold Mining Industry.\nYear: 2019\n\nAbstract: In this paper, we present an analysis of the mining process of two popular assets, Bitcoin and gold. The analysis highlights that Bitcoin, more specifically its underlying technology, is a safe haven that allows facing the modern environmental challenges better than gold. Our analysis emphasizes that crypto-currencies systems have a social and economic impact much smaller than that of the traditional financial systems. We present an analysis of the several stages needed to produce an ounce of gold and an artificial agent-based market model simulating the Bitcoin mining process and allowing the quantification of Bitcoin mining costs. In this market model, miners validate the Bitcoin transactions using the proof of work as the consensus mechanism, get a reward in Bitcoins, sell a fraction of them to cover their expenses, and stay competitive in the market by buying and divesting hardware units and adjusting their expenses by turning off/on their machines according to the signals provided by a technical analysis indicator, the so-called relative strength index.","Title: Blockchain applications for agile methodologies.\nYear: 2018\n\nAbstract: We present an application of Blockchain technology and Smart Contracts to the management of Agile projects, using Scrum or Lean-Kanban processes. In our application the duties of the Product Owner for certifying the correctness of the outcomes are delegated to one or more Smart Contracts deployed on the Ethereum Blockchain and written in Solidity. An agreement with the Customer can also allow the Smart Contracts to automatically enable payments, to introduce penalties or grants on the basis of the outcome. Product Owner duties and work can thus be relieved allowing to allocate resources on more profitable and productive tasks. Other possibilities are examined as well.","Title: Angry-builds - an empirical study of affect metrics and builds success on github ecosystem.\nYear: 2018\n\nAbstract: Automatic and repeatable builds are an established software engineering practices for achieving continuous integration and continuous delivery processes. The building phase of modern software systems is an important part of the development process such that dedicated roles as \"Release Engineer\" are more and more required. Software development is a collaborative activity, and when multiple developers work on the same project, they will be changing a shared master development branch at overlapping intervals. This overlap occurs because developers create parallel branches for working and then merge these branches when features are completed. Continuous integration, CI, is a workflow strategy which helps ensure everyoneâĂ&Zacutes changes will integrate with the current version of the project. This activity allows developers to catch bugs and reduce merge conflicts. Improving the building process leads to higher productivity and therefore shorter time to market, but understanding or measuring such a delicate phase is a big challenge. Open Source Communities provide valuable empirical data such as GitHub an Travis CI. These repositories represent a golden mine containing important data which can help researchers understanding the process behind the manufacturing of a software artifact. By analyzing Travis CI logs, we can directly connect a particular build with the development process behind it, not only regarding code changes but also regarding human activities, such as discussions about the implementation of a specific feature or bug resolution. Thanks to this information we can analyze the social activities of the build process enabling us to apply the same approach used for the development process.","Title: An Agent-Based Artificial Market Model for Studying the Bitcoin Trading.\nYear: 2019\n\nAbstract: The objective of this paper is to simulate the trading of the currency pair BTC/USD, investigating through the theory of the genetic algorithms the best sets of trading strategies, simulating through a realistic order book the bitcoin price formation, and reproducing a bitcoin price series that exhibits some stylized facts found in real-time price series. In this artificial market model two kinds of agents, Chartists and Random traders, perform trading. Chartists trade through the application of trading rules. Specifically, a part of Chartists trades applying the best sets of trading rules selected by a genetic algorithm that simulates a trading system, based on four technical analysis indicators, searching for parameters of each indicator that guarantee the highest profits in the training period; the remaining part trades applying trading rules choosing their parameters in a random way. On the contrary random trader's trade without applying any trading strategy, issuing in a random way sell or buy orders. Results show that the best sets of rules found to guarantee the highest profits both in the training and in the testing periods, and perform well also in the artificial market model where the Chartists who adopt the best sets of trading rules are able to achieve higher profits.","Title: A Massive Analysis of Ethereum Smart Contracts Empirical Study and Code Metrics.\nYear: 2019\n\nAbstract: In this work, we perform a comprehensive empirical study of smart contracts deployed on the ethereum blockchain. The objective of the analysis is to provide empirical results on smart contracts features, smart contract transactions within the blockchain, the role of the development community, and the source code characteristics. We collected a set of more than 10 000 smart contracts source codes and a dataset of meta-data regarding their interaction with the blockchain from etherscan.io . We examined the collected data computing different statistics on naming policies, smart contract ether balance, number of smart contract transactions, functions, and other quantities characterizing the use and purpose of smart contracts. We found that the number of transactions and the balances follow power-law distributions and the software code metrics display, on average, values lower than corresponding metrics in standard software but have high variances. Focusing the attention on the 20 smart contracts with the topmost number of transactions, we found that most of them represent financial smart contracts and some of them have peculiar software development stories behind them. The results show that blockchain software is rapidly changing and evolving and it is no longer devoted only to cryptovalues applications but to general purpose computation.","Title: Empirical analysis of affect of merged issues on GitHub\nYear: 2019\n\nAbstract: Pull-request based workflows are popular trends of modern software development platform such as GitHub. A pull-request notifies other developers that new changes are proposed, a code review process follows the pull-request that may be merged in the main branch if other developers accept the changes. Many factors influence the acceptance of pull-requests. Since open source software is based on collaboration, it is essential to discover how the affect expressed by developer discussing pull-request issues, namely how they collaborate, influences the acceptance of the pull-request proposed. In this study we analysed the relations with the affect expressed in pull-request issues' comments and whether an issue is merged in the main branch or not. We focused on pull-request issues and we found that issues with higher level anger, sadness, arousal and valence are less likely to be merged while issues with higher level of valence, joy are more likely to be merged. Positive affect indicates a good collaboration environment, and our finding shows that this healthy collaboration is likely to increase the acceptance of pull-requests.","Title: A user-oriented model for Oracles’ Gas price prediction\nYear: 2022\n\nAbstract: The Ethereum blockchain is a distributed database of transactions, where the Gas Oracles suggest the users the Gas price’s categories to get a transaction recorded. The paper explores the idea that the Gas Oracles are based on a data-centered model which does not provide users with a reliable prediction. We present an empirical study to test the reliability of the existing Gas Oracles from both the points of view of the Gas price predictions and the existing categories.","Title: Blockchain and Self Sovereign Identity to Support Quality in the Food Supply Chain\nYear: 2021\n\nAbstract: This work presents how a digital identity management system can support food supply chains in guaranteeing the quality of the products marketed and the compliance of the several supply-chain's nodes to standards and technical regulations. Specific goal of this work is to present a system that provides full visibility of process/food certifications, which nowadays are issued by accredited and approved certification bodies (issuers) and delivered and stored in paper version by the several participants (holders) of the supply chain. The system is designed and implemented by combining the latest most innovative and disruptive technologies in the market-Self Sovereign Identity system, Blockchain, and Inter Planetary File System. The crucial aspects that it aims to hit are the storage and access of food/process certifications, and the proper eligibility verification of these certifications exploiting the concepts of the Self Sovereign Identity-based models. The proposed system, realized by using standards that are WWW Consortium-compatible and the Ethereum Blockchain, ensures eligibility, transparency, and traceability of the certifications along a food supply chain, and could be an innovation model/idea that the companies that adopt the Open Innovation paradigm might want to pursue.","Title: Bitcoin as a Safe Haven during COVID-19 Disease\nYear: 2022\n\nAbstract: In this paper, we investigate the role of Bitcoin as a safe haven against the stock market losses during the spread of COVID-19. The performed analysis was based on a regression model with dummy variables defined around some crucial dates of the pandemic and on the dynamic conditional correlations. To try to model the real dynamics of the markets, we studied the safe-haven properties of Bitcoin against thirteen of the major stock market indexes losses using daily data spanning from 1 July 2019 until 20 February 2021. A similar analysis was also performed for Ether. Results show that this pandemic impacts on the Bitcoin status as safe haven, but we are still far from being able to define Bitcoin as a safe haven.","Title: Automatic Generation of Ethereum-Based Smart Contracts for Agri-Food Traceability System\nYear: 2022\n\nAbstract: There is a growing demand for transparency along the agri-food chain, both from customers and governments. The adoption of blockchain technology to enable secure traceability for the management of the agri-food chain, provide information such as the provenance of a food product and prevent food fraud, is emerging rapidly, due to the inherent trust and inalterability provided by this technology. However, developing the right smart contracts for these use cases is even more of a challenge than it is for those used in other fields. Several management systems for the agri-food chain based on blockchain technology and smart contract have been proposed, all however ad-hoc for a specific product or production process and difficult to generalize. In this paper, we propose a new approach to easily customize and compose general Ethereum-based smart contracts designed for the agri-food industrial domain, to be able to reuse the code and modules and automate the process to shorten development times, while keeping it safe and reliable. Starting from the definition of the real production process, we aim to automatically generate both the smart contracts to manage the system and the user interfaces to interact with them, thus producing a system that works semi-automatically. Additionally, we describe a honey production case study to show how our approach works. Future work will first extend the scope of the approach to other supply chains, furthermore, while the current platform used is Ethereum, in the future our approach will be easily extended to other blockchain platforms.","Title: A System Proposal for Information Management in Building Sector Based on BIM, SSI, IoT and Blockchain\nYear: 2022\n\nAbstract: This work presents a Self Sovereign Identity based system proposal to show how Blockchain, Building Information Modeling, Internet of Thing devices, and Self Sovereign Identity concepts can support the process of building digitalization, guaranteeing the compliance standards and technical regulations. The proposal ensures eligibility, transparency and traceability of all information produced by stakeholders, or generated by IoT devices appropriately placed, during the entire life cycle of a building artifact. By exploiting the concepts of the Self Sovereign Identity, our proposal allows the identification of all involved stakeholders, the storage off-chain of all information, and that on-chain of the sole data necessary for the information notarization and certification, adopting multi-signature approval mechanisms where appropriate. In addition it allows the eligibility verification of the certificated information, providing also useful information for facility management. It is proposed as an innovative system and companies that adopt the Open Innovation paradigm might want to pursue it. The model proposal is designed exploiting the Veramo platform, hence the Ethereum Blockchain, and all the recommendations about Self Sovereign Identity systems given by the European Blockchain Partnership, and by the World Wide Web Consortium."],"string":"[\n \"Title: Studying the evolution of quality metrics in an agile/distributed project\\nYear: 2006\\n\\nAbstract: This paper analyzes the development of a project initiated by a co-located agile team that subsequently evolved into a distributed context. The project, named JAPS (Java Agile Portal System) [1], has been monitored on a regular basis since it started in January 2005, collecting both process and product metrics. Product metrics have been calculated by checking out the source code history from the CVS repository. By analyzing the evolution of these metrics, it has been possible to evaluate how the distribution of the team has impacted the source code quality.\",\n \"Title: Micro Pattern Fault-Proneness\\nYear: 2012\\n\\nAbstract: One of the goals of Software Engineering is to reduce, or at least to try to control, the defectiveness of software systems during the development phase. The aim of our study is to analyze the relationship between micro patterns (introduced by Gil and Maman) and faults in a software system. Micro patterns are similar to design patterns, but their characteristic is that they can be identified automatically, and are at a lower level of abstraction with respect to design patterns. Our study aims to show, through empirical studies of open source software systems, which categories of micro patterns are more correlated to faults. Gil and Maman demonstrated, and subsequent studies confirmed, that 75% of the classes of a software system are covered by micro patterns. In our study we also analyze the relationship between faults and the remaining 25% of classes that do not match with any micro pattern. We found that these classes are more likely to be fault-prone than the others. We also studied the correlation among all the micro patterns of the catalog, in order to verify the existence of relationships between them.\",\n \"Title: Simulating Kanban and Scrum vs. Waterfall with System Dynamics\\nYear: 2011\\n\\nAbstract: Nowadays, Scrum is the most used Agile Methodology, while the Lean-Kanban approach is perhaps the fastest growing AM. On the other hand, traditional, waterfall-like approaches are still very used in real-life software projects, due to the ease of up-front planning and budgeting, that however are seldom matched upon project completion. In our opinion, more effort is needed to study and model the inner structure and behavior of these approaches, highlighting positive and negative feedback loops that are strategic to understand their features, and to decide on their adoption. In this paper we analyze the dynamic behavior of the adoption of Kanban and Scrum, versus a traditional software development process such as the Waterfall approach. We use a system dynamics model, based on the relationships between system variables, to assess the relative benefits of the studied approaches. The model is simulated using a commercial tool. The proposed model visualizes the relationships among these software development processes, and can be used to study their relative advantages and disadvantages.\",\n \"Title: An analysis of SNA metrics on the Java Qualitas Corpus\\nYear: 2011\\n\\nAbstract: We computed the software graphs of 96 systems of the Java Qualitas Corpus, parsing the source code and identifying the dependencies among classes. We analyzed 12 software metrics on these 96 graphs, nine borrowed from Social Network Analysis (SNA), and three more traditional software metrics, such as Loc, Fan-in and Fan-out. We analyzed their correlations at system level, and studied the correlation statistics at data-set level. Our results show that these correlations are independent from the specific software system and are general properties of Java software systems. We show how the metrics can be partitioned in groups for almost the whole Java Qualitas Corpus, and that such grouping can provide insights on the topology of software networks. For two systems, Eclipse and Netbeans, we computed also the number of bugs, identifying the bugs affecting each class, and finding that some SNA metrics are highly correlated with bugs, while others are strongly anticorrelated. This suggests that practitioners and software engineers might take advantage of such metrics to keep control of software quality.\",\n \"Title: Effort Estimation of Web Applications through Web CMF Objects\\nYear: 2012\\n\\nAbstract: Several Content Management Frameworks (CMF), either open source or with a commercial license, are available to help the publication of huge amounts of information and to develop Web applications. Unfortunately, developing Web applications through a CMF is not exempt from cost and time overruns, as in traditional software projects, and currently there is no estimation model able to adequately measure the effort of Web application development. This work presents a new methodology for estimat ing the effort of Web applications developed with a Content Management Framework (CMF). We present the new key elements for analysis and planning, needed to define every important step in developing a Web application through a CMF. Using those elements, it is possible to estimate the effort needed to build such an ap plication in man-days. We also present the experimental validation of the proposed methodology per formed on a 9-project dataset, provided by three different Italian software companies.\",\n \"Title: A Neural Network Diagnosis Approach for Analog Circuits\\nYear: 1999\\n\\nAbstract: This paper presents a neural network system for the diagnosis of analogcircuits and shows how the performance of such a system can be affected by the choice of differenttechniques used by its submodules. In particular we discuss the influenceof feature extraction techniques such as Fourier Transforms, Wavelets and Principal ComponentAnalysis. The system uses several different power supplies and as manyneural networks “in parallel”. Two different algorithms that canbe used to combine the candidate sets produced by each network arealso presented. The system is capable of diagnosing multiplefaults even if trained on single ones.\",\n \"Title: A revised web objects method to estimate web application development effort.\\nYear: 2011\\n\\nAbstract: We present a study of the effectiveness of estimating web application development effort using Function Points and Web Objects methods, and a method we propose - the Revised Web Objects (RWO). RWO is an upgrading of WO method, aimed to account for new web development styles and technologies. It also introduces an up-front classification of web applications according to their size, scope and technology, to further refine their effort estimation. These methods were applied to a data-set of 24 projects obtained by Datasiel spa, a mid-sized Italian company, focused on web application projects, showing that RWO performs statistically better than WO, and roughly in the same way as FP.\",\n \"Title: An empirical study of social networks metrics in object-oriented software\\nYear: 2010\\n\\nAbstract: We study the application to object-oriented software of new metrics, derived from Social Network Analysis. Social Networks metrics, as for instance, the EGO metrics, allow to identify the role of each single node in the information flow through the network, being related to software modules and their dependencies. These metrics are compared with other traditional software metrics, like the Chidamber-Kemerer suite, and software graph metrics. We examine the empirical distributions of all the metrics, bugs included, across the software modules of several releases of two large Java systems, Eclipse and Netbeans. We provide analytical distribution functions suitable for describing and studying the observed distributions. We study also correlations among metrics and bugs. We found that the empirical distributions systematically show fat-tails for all the metrics. Moreover, the various metric distributions look very similar and consistent across all system releases and are also very similar in both the studied systems. These features appear to be typical properties of these software metrics.\",\n \"Title: A model of the dynamics of the market of COTS software, in the absence of new entrants\\nYear: 2007\\n\\nAbstract: This work proposes a simple model to simulate the dynamics of the market related to COTS software products, when there are no new entrants. The model takes in explicit account the interactions between products. The model uses a set of random variables, as the initial condition, to simulate the time evolution. The results of the simulations predict the emergence of monopolies, when the interactions between products are significant, and of open markets, when they are limited. The model is validated, by comparing the results of the simulation with the well established observations related to software markets and by comparing the resulting distribution of COTS software products with real data on the software market, which was provided by PC Data Inc.\",\n \"Title: System performance analyses through object-oriented fault and coupling prisms\\nYear: 2014\\n\\nAbstract: A fundamental aspect of a system's performance over time is the number of faults it generates. The relationship between the software engineering concept of \\\\\\\"coupling\\\\\\\" (i.e., the degree of inter-connectedness of a system's components) and faults is still a research question attracting attention and a relationship with strong implications for performance; excessive coupling is generally acknowledged to contribute to fault-proneness. In this paper, we explore the relationship between faults and coupling. Two releases from each of three open-source Eclipse projects (six releases in total) were used as an empirical basis and coupling and fault data extracted from those systems. A contrasting coupling profile between fault-free and fault-prone classes was observed and this result was statistically supported. Object-oriented (OO) classes with low values of fan-in (incoming coupling) and fan-out (outgoing coupling) appeared to support fault-free classes, while classes with high fan-out supported relatively fault-prone classes. We also considered size as an influence on fault-proneness. The study thus emphasizes the importance of minimizing coupling where possible (and particularly that of fan-out); failing to control coupling may store up problems for later in a system's life; equally, controlling class size should be a concomitant goal.\",\n \"Title: Empirical Analysis on the Satisfaction of IT Employees Comparing XP Practices with Other Software Development Methodologies\\nYear: 2004\\n\\nAbstract: Job satisfaction has been studied by economists and psychologists. We believe this factor is very important in that it influences the effectiveness of the software development process. This paper reports the first results of a comparative analytic study on the job satisfaction of developers that use XP practices and others that do not use XP practices. By determining the factors that are highly valued by developers, the research can provide insight in currently practised software development processes, and help make changes that increase their strategic value.\",\n \"Title: An Analysis of Bug Distribution in Object Oriented Systems\\nYear: 2009\\n\\nAbstract: We introduced a new approach to describe Java software as graph, where nodes\\nrepresent a Java file - called compilation unit (CU) - and an edges represent a\\nrelations between them. The software system is characterized by the degree\\ndistribution of the graph properties, like in-or-out links, as well as by the\\ndistribution of Chidamber and Kemerer metrics computed on its CUs. Every CU can\\nbe related to one or more bugs during its life. We find a relationship among\\nthe software system and the bugs hitting its nodes. We found that the\\ndistribution of some metrics, and the number of bugs per CU, exhibit a\\npower-law behavior in their tails, as well as the number of CUs influenced by a\\nspecific bug. We examine the evolution of software metrics across different\\nreleases to understand how relationships among CUs metrics and CUs faultness\\nchange with time.\",\n \"Title: An Empirical Study Of Software Metrics For Assessing The Phases Of An Agile Project\\nYear: 2012\\n\\nAbstract: We present an analysis of the evolution of a Web application project developed with object-oriented technology and an agile process. During the development we systematically performed measurements on the source code, using software metrics that have been proved to be correlated with software quality, such as the Chidamber and Kemerer suite and Lines of Code metrics. We also computed metrics derived from the class dependency graph, including metrics derived from Social Network Analysis. The application development evolved through phases, characterized by a different level of adoption of some key agile practices - namely pair programming, test-based development and refactoring. The evolution of the metrics of the system, and their behavior related to the agile practices adoption level, is presented and discussed. We show that, in the reported case study, a few metrics are enough to characterize with high significance the various phases of the project. Consequently, software quality, as measured using these metrics, seems directly related to agile practices adoption.\",\n \"Title: Studying Lean-Kanban Approach Using Software Process Simulation\\nYear: 2011\\n\\nAbstract: We developed an event-driven simulator of the Kanban process a WIP limited pull system visualized by the Kanban board. WIP (work in process) represent the capacity in the activity to perform features simoultaneously. The simulator is fully object-oriented, and its design model reflects the objects of the Lean software development domain. We used this simulator to assess comparatively WIP-limited and unlimited processes. We also studied the optimum values of the working item limits in the activities, using a paradigmatic case of 4 activities and 100 work items. The cost function used is equal to the total time needed to complete the project, plus a weighted sum of the limits themselves. We performed an exhaustive search on all the admissible values of the solution; finding sensible optimal values, and a non-trivial behavior of the cost function in the optimization space. This demonstrates the feasibility and usefulness of the approach.\",\n \"Title: 2010 ICSE workshop on emerging trends in software metrics\\nYear: 2010\\n\\nAbstract: This paper reports on the 2010 ICSE Workshop on Emerging Trends in Software Metrics (WETSoM 2010) held on Tuesday 4 May 2010 in Cape Town, South Africa, as part of the ICSE workshop series. The goal of this workshop was to bring together researchers and practitioners to discuss the progress of software metrics.\",\n \"Title: Agile practices in a large organization: the experience of poste Italiane\\nYear: 2007\\n\\nAbstract: In this paper we show how agile practices have been used at the Poste Italiane for building a monitoring system of its complex IT infrastructure. The system, called Datamart, is built upon the existing monitoring infrastructure. A testing framework has been developed for performing assertion checking either on existing legacy modules or on the new functionalities. This framework is currently used, and is able to process data coming from 100,000 distributed computers, enabling and improving their centralized control.\",\n \"Title: An analysis of anti-micro-patterns effects on fault-proneness in large Java systems\\nYear: 2012\\n\\nAbstract: Micro patterns are similar to design patterns, but are at a lower level of abstraction, closer to the implementation. Anti patterns are micro patterns not respecting the prescriptions of good Object Oriented programming practices. In this paper, we use the definitions introduced by Arcelli and Maggioni [3] in order to study the evolution of five particular micro patterns (anti patterns) in different releases of the Eclipse and NetBeans systems, and the correlations between anti patterns and faults. Our analysis confirms previous findings regarding the high coverage of micro patterns onto the system classes, and show that anti patterns not only represent bad Object Oriented programming practices, but may also be associated to the production of lower quality software, since they present a fault proneness significantly enhanced.\",\n \"Title: An Empirical Study of Refactoring in the Context of FanIn and FanOut Coupling\\nYear: 2011\\n\\nAbstract: The aim of refactoring is to reduce software complexity and hence simplify the maintenance process. In this paper, we explore the impact of refactorings on \\\"FanIn\\\" and \\\"FanOut\\\" coupling metrics through extraction of refactoring data from multiple releases of five Java open-source systems, We first considered how a single refactoring modified these metric values, what happened when refactorings had been applied to a single class in unison and finally, what influence a set of refactorings had on the shape of Fan In and Fan Out distributions. Results indicated that, on average, refactored classes tended to have larger FanIn and Fan Out values when compared with non-refactored classes. Where evidence of multiple (different) refactorings applied to the same class was found, the net effect (in terms of FanIn and Fan Out coupling values) was negligible.\",\n \"Title: A case study of the use of Open Source CMS in Public Administrations\\nYear: 2012\\n\\nAbstract: The use of Open Source Content Management System (CMS) on the Web is increasing thanks to the flexibility provided by these systems and the semplification of the management operations. The use of CMS is simple for non programming experts. Many Public Administrations have adopted Open Source CMS for their websites, greatly reducing operating costs and ensuring scalability, advanced performance and privacy to the users. A special opportunity for a case study are the Italian municipalities. In fact the Italian Government, since January 2011, has required all municipalities to develop a website publishing all the public acts and everything related to the citizen-municipality communication (information regarding office hours, salaries of public employees, news and so on). Our study aims to quantify through empirical studies, the use of Open Source CMS in Italian municipalities websites in response to government regulations. Our results show that the number of Italian Public Administrations that use Open Source CMS is increasing, and our case study highlights how the adoption of Open Source CMS has increased about 5% in a year, leading to the 21.2% the percentage of Open Source CMS, with respect to the 15.8% registered in 2011.\",\n \"Title: Assessing traditional and new metrics for object-oriented systems.\\nYear: 2010\\n\\nAbstract: We present an extensive analysis of software metrics for 111 object-oriented systems written in Java. For each system, we considered 18 traditional metrics such as LOC and Chidamber and Kemerer metrics, as well as metrics derived from complex network theory and social network analysis. These metrics were computed at class level. We also considered two metrics at system level, namely the total number of classes and interfaces, and the fractal dimension. We discuss the distribution of these metrics, and their correlation, both at class and at system level. We found that most metrics follow a leptokurtotic distribution. Only a couple of metrics have patent normal behavior while three others are very irregular, and even bimodal. The statistics gathered allow us to study and discuss the variability of metrics along different systems, and to devise a roadmap for further research.\",\n \"Title: Communication flow in open source projects: an analysis of developers' mailing lists\\nYear: 2007\\n\\nAbstract: Team work is defined by different factors: team size, members role, interactions and communication. In Agile Methodologies communication plays a fundamental role. Some research studies have found that the involvement of team members in a project also depends on the information provided. Communication also depends on the tool employed. The aim of this paper is to analyze communication among team members of the 9 most active projects on Source-forge provided of a developers' mailing list. In particular we tried to investigate how and why members post to developers' mailing list and the use that different members make of these mailing lists.\",\n \"Title: Study of the Competition between Proprietary Software Firms and Free/Libre Open Source Software Firms Using a Simulation Model.\\nYear: 2011\\n\\nAbstract: In recent years, a very important structural change in the software industry took place, with an increasing number of firms that got involved in Free/Libre Open Source Software (FLOSS) development communities. FLOSS communities and products have been studied as complementary to proprietary software companies and products. In this paper we propone a business model for the software market, and in particular we analyze the competition between proprietary software firms and FLOSS firms. Our software market is a system where each agent is independent of each other in the choice about buying or selling software products or services. The proposed work aims to analyze the influence of FLOSS firms producing both software and services in vertical software markets, nowadays mostly dominated by large proprietary firms. The findings show that FLOSS firms are able to compete with proprietary firms, though in the end a monopoly or oligopoly of the latters emerges. The ousted FLOSS firms, however, survive longer than proprietary ones, when these are not able to compete in the market.\",\n \"Title: Developing a Tool Supporting XP Process\\nYear: 2003\\n\\nAbstract: We present the development of XPSwiki, a tool supporting the XP practices for requirement gathering and project management - user stories and the Planning Game. XPSwiki has been developed in Smalltalk using Squeak Swiki technology, and is accessed through the Internet in a user friendly, agile way. It keeps track of multiple projects, each with its releases, iterations, user stories, acceptance tests, and tasks. XPSwiki allows project tracking, and automates documentation generation and metrics collection. It is presently in use in real software development environments.\",\n \"Title: Micro Patterns in Agile Software.\\nYear: 2013\\n\\nAbstract: In this paper we present a study on micro patterns in different releases of two software systems developed with Object Oriented technologies and Agile process. Micro patterns are design decisions in code that can be easily automatically recognised. Gil and Maman introduced the concept to support providing objective assessment of design decisions [1]. They catalogued 27 micro patterns that capture a variety of programming practices in Java. Micro patterns can be a useful metrics in order to measure the quality of software by showing that certain categories of micro patterns are more fault prone than others, and that the classes that do not correspond to any category of micro patterns are more likely to be faulty. In our study we present some empirical results on two case studies of systems developed with Agile methodologies, and compare them to previous results obtained for non Agile systems. In particular we have verified that the distribution of micro patterns in a software system developed using Agile methodologies does not differ from the distribution studied in other systems, and that the micro patterns fault-proneness is about the same. We also analyzed how the distribution of micro patterns changes in different releases of the same software system. We demonstrate that there is a relationship between the number of faults and the classes that do not match with any micro patterns. We found that these classes are more likely to be fault-prone than the others even in software developed with Agile methodologies.\",\n \"Title: Workshop on Emerging Trends in Software Metrics (WETSoM 2010)\\nYear: 2011\\n\\nAbstract: The Workshop on Emerging Trends in Software Metrics aims at bringing together researchers and practitioners to discuss the progress of software metrics. The motivation for this workshop is the low impact that software metrics has on current software development. The goals of this workshop are to critically examine the evidence for the effectiveness of existing metrics and to identify new directions for development of software metrics.\",\n \"Title: The Distribution of Program Sizes and Its Implications: An Eclipse Case Study\\nYear: 2009\\n\\nAbstract: A large software system is often composed of many inter-related programs of\\ndifferent sizes. Using the public Eclipse dataset, we replicate our previous\\nstudy on the distribution of program sizes. Our results confirm that the\\nprogram sizes follow the lognormal distribution. We also investigate the\\nimplications of the program size distribution on size estimation and quality\\npredication. We find that the nature of size distribution can be used to\\nestimate the size of a large Java system. We also find that a small percentage\\nof largest programs account for a large percentage of defects, and the number\\nof defects across programs follows the Weibull distribution when the programs\\nare ranked by their sizes. Our results show that the distribution of program\\nsizes is an important property for understanding large and complex software\\nsystems.\",\n \"Title: XPSwiki: an agile tool supporting the planning game\\nYear: 2003\\n\\nAbstract: We present XPSwiki, a tool supporting the XP practices for requirement gathering and project management - user stories and the Planning Game. XPSwiki has been developed in Smalltalk using Squeak Swiki technology. It is accessed through the Internet in a user friendly agile way. XPSwiki keeps track of multiple projects, each with its releases, iterations, user stories, acceptance tests, and engineering tasks. It also manages teams of developers, who sign stories and tasks, and who pair-program them with another team member. XPSwiki allows project tracking and documentation, and is at present in use among real software firms.\",\n \"Title: Simulation of software maintenance process, with and without a work-in-process limit.\\nYear: 2013\\n\\nAbstract: A software maintenance process is important for reducing maintenance effort and for improving software quality. In recent years, the Lean-Kanban approach has been widely applied in software practice including software maintenance. This approach minimizes Work-in-Progress (WIP), which is the number of items that are worked on by the team at any given time, thus improving the maintenance process. In this paper, we describe our simulation studies, which show that the Lean-Kanban approach can indeed help reduce the average time needed to complete maintenance requests. We develop a process simulator that can simulate both existing maintenance processes that do not use a WIP limit and that adopt it. We perform two case studies using real maintenance data collected from a Microsoft project and from a Chinese software firm. The main results of our study are twofold. First, we demonstrate that it is possible to effectively model and simulate, using actors and events, a maintenance process where a flow of issues is processed through a sequence of activities, correctly reproducing key statistics of real data. Second, our results confirm that the WIP-limited process could be useful to increase the efficiency of software maintenance, as reported in previous industrial practices. Copyright (c) 2013 John Wiley & Sons, Ltd.\",\n \"Title: A Ontology-based Process Modelling for XP\\nYear: 2003\\n\\nAbstract: In this paper we describe the Extreme Programming Ontology(XPO), a formal model specifying the main conceptsused in the Extreme Programming methodology and theirproperties. XPO's modular structure was developed usingthe usual normative top down approach to software engineeringprocess modeling. It relies on a set of core componentsrooted in three main concepts: Organisational Role,Product and Phase. Besides being useful for indexing relevantdocuments and XP artifacts such as user stories andWiki pages, XPO is aimed at being a sound basis for non-intrusiveanalysis of agile processes, mining process dataabout programmers activity and repositories content in orderto extract new concepts potentially identifying criticalfactors in agile software development. Extension to XPOare also discussed, including other Agile Methodologiesand more general Software Engineering concepts.\",\n \"Title: Distributed Scrum in research project management\\nYear: 2007\\n\\nAbstract: Can research projects be agile? In this paper we describe our proposal of applying Scrum for the management of an European research project aimed at developing an agent-based software platform for European economic policy design. The use of an agile, adaptive methodology is justified because successful research projects are complex, unstable processes, that should be continuously adapted along their way. We describe in detail the roles, artifacts and practices of the proposed process, and the first steps of its adoption.\",\n \"Title: An Agile Development Process and Its Assessment Using Quantitative Object-Oriented Metrics\\nYear: 2008\\n\\nAbstract: The development of a Web application using agile practices is presented, characterized by empirical software measurements that have been performed throughout the duration of the project. The project is the specialization of jAPS, an open source Java Web portal generation framework, for building a Research Register management system. The agile principles and practices used in the project are described and discussed. During the various phases of the project some key agile practices, such as pair programming, test-based development and refactoring, were used at different levels of adoption. The evolution of some object-oriented metrics of the system, and their behavior related to the agile practices adoption level is presented and discussed, showing that software quality, as measured using standard 00 metrics, looks directly related to agile practices adoption.\",\n \"Title: A Comparative Study of Scrum and Kanban Approaches on a Real Case Study Using Simulation.\\nYear: 2012\\n\\nAbstract: We present the application of software process modeling and simulation using an agent-based approach to a real case study of software maintenance. The original process used PSP/TSP; it spent a large amount of time estimating in advance maintenance requests, and needed to be greatly improved. To this purpose, a Kanban system was successfully implemented, that demonstrated to be able to substantially improve the process without giving up PSP/TSP. We customized the simulator and, using input data with the same characteristics of the real ones, we were able to obtain results very similar to that of the processes of the case study, in particular of the original process. We also simulated, using the same input data, the possible application of the Scrum process to the same data, showing results comparable to the Kanban process.\",\n \"Title: Refactoring and its Relationship with Fan-in and Fan-out: An Empirical Study\\nYear: 2012\\n\\nAbstract: Many of the seventy-two refactorings originally proposed by Fowler have the aim of reducing coupling in classes and thereby aiding programmer maintainability. While we might assume that classes with high levels of coupling would be the targets for refactoring more often than any other type of class, no empirical evidence exists to support this assumption. In this paper, we explore that open question through the extraction of fifty-two of Fowler's catalog of refactorings drawn from versions of four open-source systems. We compare the coupling characteristics of each set of refactored classes with the corresponding set of non-refactored classes (those for which no refactorings were applied across versions), we seek to inform one overriding research question: is there a significant difference between these two sets of classes in terms of their level of coupling? Results showed that highly-coupled classes were more prone to refactoring, particularly through a set of 'core' refactorings. However, wide variations were found across systems for our measures of coupling namely, fan-in and fan-out. Specific individual refactorings were also explored to gain an understanding of why these differences may have occurred.\",\n \"Title: Entropy Of Some Ck Metrics To Assess Object-Oriented Software Quality\\nYear: 2013\\n\\nAbstract: The term \\\"software entropy\\\" refers to the tendency for software, over time, to become difficult and costly to maintain. A software system that undergoes continuous change, such as having new functionality added to its original design, will eventually become more complex and can become disorganized as it grows, losing its original design structure.A recent study show that software degradation may be measured using the WMC expressed in terms of Shannon entropy. In this paper we extended the empirical analyses also to RFC and CBO since these CK metrics have been shown to be correlated with fault-proneness of OO classes.We analyzed various releases of the publicly available Eclipse and Netbeans software systems, calculating the entropy of some CK metrics for every release analyzed. The validity is shown through a direct measure of software quality such as the number of detected defects. Our results display a very good correlation between the entropy of CBO and RFC and the number of bugs for Eclipse and Netbeans.Complexity and quality metrics are in general computed on every system module while the entropy is just a scalar number that characterizes a whole system, this result suggests that the entropy of some CK metrics could be considered as a global quality metric for large software systems. Our results need, however, to be confirmed for other large software systems.\",\n \"Title: The fractal dimension metric and its use to assess object-oriented software quality.\\nYear: 2011\\n\\nAbstract: We present a study were software systems are considered as complex networks which have a self-similar structure under a length-scale transformation. On such complex software networks we computed a self-similar coefficient, also known as fractal dimension, using \\\"the box counting method\\\". We analyzed various releases of the publically available Eclipse software systems, calculating the fractal dimension for twenty sub-projects, randomly chosen, for every release, as well as for each release as a whole. Our results display an overall consistency among the sub-projects and among all the analyzed releases. We found a very good correlation between the fractal dimension and the number of bugs for Eclipse and for twenty sub-projects. Since the fractal dimension is just a scalar number that characterizes a whole system, while complexity and quality metrics are in general computed on every system module, this result suggests that the fractal dimension could be considered as a global quality metric for large software systems. Our results need however to be confirmed for other large software systems.\",\n \"Title: Study of the Evolution of an Agile Project Featuring a Web Application Using Software Metrics\\nYear: 2008\\n\\nAbstract: We present an agile process used for the development of a Web application written in Java, devised by choosing a set of proven agile practices taken by existing popular agile methodologies. During the project, we regularly measured the software using Chidamber and Kemerer object-oriented metrics suite, and other metrics. The application development evolved through phases, characterized by a different level of adoption of some key agile practices --- such as pair programming, test-based development and refactoring. The evolution of the OO metrics of the system, and their behavior related to the agile practices adoption level is presented and discussed, showing that soft ware quality, as measured using standard OO metrics, looks directly related to agile practices adoption.\",\n \"Title: Power-Laws in a Large Object-Oriented Software System\\nYear: 2007\\n\\nAbstract: We present a comprehensive study of an implementation of the Smalltalk object oriented system, one of the first and purest object-oriented programming environment, searching for scaling laws in its properties. We study ten system properties, including the distributions of variable and method names, inheritance hierarchies, class and method sizes, system architecture graph. We systematically found Pareto - or sometimes log-normal - distributions in these properties. This denotes that the programming activity, even when modeled from a statistical perspective, can in no way be simply modeled as a random addition of independent increments with finite variance, but exhibits strong organic dependencies on what has been already developed. We compare our results with similar ones obtained for large Java systems, reported in the literature or computed by ourselves for those properties never studied before, showing that the behavior found is similar in all studied object oriented systems. We show how the Yule process is able to stochastically model the generation of several of the power-laws found, identifying the process parameters and comparing theoretical and empirical tail indexes. Lastly, we discuss how the distributions found are related to existing object-oriented metrics, like Chidamber and Kemerer's, and how they could provide a starting point for measuring the quality of a whole system, versus that of single classes. In fact, the usual evaluation of systems based on mean and standard deviation of metrics can be misleading. It is more interesting to measure differences in the shape and coefficients of the data?s statistical distributions.\",\n \"Title: On the influence of maintenance activity types on the issue resolution time\\nYear: 2014\\n\\nAbstract: The ISO/IEC 14764 standard specifies four types of software maintenance activities spanning the different motivations that software engineers have while performing changes to an existing software system. Undoubtedly, this classification has helped in organizing the workflow within software projects, however for planning purposes the relative time differences for the respective tasks remains largely unexplored. In this empirical study, we investigate the influence of the maintenance type on issue resolution time. From GitHub's issue repository, we analyze more than 14000 issue reports taken from 34 open source projects and classify them as corrective, adaptive, perfective or preventive maintenance. Based on this data, we show that the issue resolution time depends on the maintenance type. Moreover, we propose a statistical model to describe the distribution of the issue resolution time for each type of maintenance activity. Finally, we demonstrate the usefulness of this model for scheduling the maintenance workload.\",\n \"Title: Are Some Refactorings Attached to Fault-Prone Classes and Others to Fault-Free Classes?\\nYear: 2014\\n\\nAbstract: A topical and relevant issue in the area of refactoring is the nature and characteristics of classes to which refactorings are applied. In particular, if we scrutinise the total set of refactorings applied to the classes of a system over different releases, which refactorings are applied to fault-prone classes and which to fault-free classes? In this paper, we explore that facet of refactoring. Refactorings applied between six releases of three Eclipse packages are used as a basis of the study and the Ref-Finder tool used to extract up to sixty-five different refactorings. Interestingly, results showed that refactorings applied to highly fault-prone classes differed significantly from those applied to fault-free classes, in particular related to the 'rename method' refactoring; a corresponding trend for the 'move method' and 'move field' refactoring was found in 'fault-free' classes over the period while the add and remove parameter refactorings tended to remain constant. The research offers an insight into refactoring behaviour in light of faults (or no faults).\",\n \"Title: Applications of simulated annealing for the design of special digital filters\\nYear: 1992\\n\\nAbstract: The authors describe the salient features of using a simulated annealing (SA) algorithm in the context of designing digital filters with coefficient values expressed as the sum of power of two. A procedure for linear phase digital filter design, using this algorithm, is presented and tested, yielding results as good as those for known optimal methods. The algorithm is then applied to the design of Nyquist filters, optimizing at the same time both frequency response and intersymbol interference, and to the design of cascade form finite-impulse-response (FIR) filters. The drawback of using SA is that the computation time is on the order of 1-2 h for each filter design, on the Sun 3/60. However, this was more than compensated by the versatility of the new algorithm, which can be used to design filters with multiple constraints\",\n \"Title: Entropy of the degree distribution and object-oriented software quality\\nYear: 2012\\n\\nAbstract: The entropy of degree distribution has been considered from many authors as a measure of a network's heterogeneity and consequently of the resilience to random failures. In this paper we propose the entropy of degree distribution as a new measure of software quality.\\n\\nWe present a study were software systems are considered as complex networks which are characterized by heterogeneous distribution of links. On such complex software networks we computed the entropy of degree distribution. We analyzed various releases of the publically available Eclipse and Netbeans software systems, calculating the entropy of degree distribution for every release analyzed. Our results display a good correlation between the entropy of degree distribution and the number of bugs for Eclipse and Netbeans.\\n\\nComplexity and quality metrics are in general computed on every system module while the entropy is just a scalar number that characterizes a whole system, this result suggests that the entropy of degree distribution could be considered as a global quality metric for large software systems. Our results need however to be confirmed for other large software systems.\",\n \"Title: Size estimation of Web applications through Web CMF Object\\nYear: 2012\\n\\nAbstract: This work outlines a new methodology for estimating the size of Web applications developed with a Content Management Framework (CMF). The reason for proposing - through this work - a new methodology for size estimation is the realization of the inadequacy of the RWO method, which we had recently developed, in estimating the effort of the latest Web applications. The size metric used in the RWO method was found not to be well suited for Web applications developed through a CMF. In this work, we present the new key elements for analysis and planning, needed to define every important step in developing a Web application through a CMF. Using those elements, it is possible to obtain the size of such an application. We also present the experimental validation performed on a 7-project dataset, provided by an Italian software company.\",\n \"Title: Measuring and Understanding the Effectiveness of JIRA Developers Communities\\nYear: 2015\\n\\nAbstract: Tools for project management and issues/bugs tracking are becoming useful for governing the development process of Open Source software. Such tools simplify the communications process among developers and ensure the scalability of a project. The more information developers are able to exchange, the clearer are the goals, and the higher is the number of developers keen on joining and actively collaborating on a project. In this paper we present a preliminary empirical analysis of the communities-structure of developers in JIRA by analyzing 7 popular projects hosted in the repository. We analyze how these communities perform in terms of issue-resolution time of any given issue. The main contributions of this work are the confirmation of the existence of communities in developer networks, and the empirical finding that the issue resolution-time of any given issue is not correlated with the dimension of a developer community.\",\n \"Title: Software development: do good manners matter?\\nYear: 2015\\n\\nAbstract: A successful software project is the result of a complex process involving, above all, people. Developers are the key factors for the success of a software development process, not merely as executors of tasks, but as protagonists and core of the whole development process. This paper investigates social aspects among developers working on software projects developed with the support of Agile tools. We studied 22 open-source software projects developed using the Agile board of the JIRA repository. All comments committed by developers involved in the projects were analyzed and we explored whether the politeness of comments affected the number of developers involved and the time required to fix any given issue. Our results showed that the level of politeness in the communication process among developers does have an effect on the time required to fix issues and, in the majority of the analysed projects, it had a positive correlation with attractiveness of the project to both active and potential developers. The more polite developers were, the less time it took to fix an issue.\",\n \"Title: Simulation of the competition among traditional and on-demand software vendors\\nYear: 2016\\n\\nAbstract: AbstractIn this work, we propose a simulation model to study the software market, and in particular the competition among on-premise and on-demand vendors of customer relationship management software. In this market two kinds of vendors, on-premise and on-demand vendors, and corporate customers interact. Vendors have an initial capital and a specific number of developers. They invest a fraction of their capital in their product quality and try to sell their products making a profit. Corporate customers buy software products for their employees, trying to minimize their cost and maximize their utility. The proposed model is able to reproduce the real trends of the market as regards both the price of software and the market shares of the vendors, despite there is relatively little empirical evidence available about the various mechanisms at play, and few experimental data. Our model could be used as a tool to forecast future market trends, or to plan business policies of investment and pricing. Any firm could calibrate this model depending on its past business trends in order to obtain simple guidelines to follow in the planning of business winning strategies.\",\n \"Title: Web framework points: an effort estimation methodology for Web application development using a content management framework\\nYear: 2015\\n\\nAbstract: Web applications are among the most popular and relevant kinds of application. Most Web applications are developed using a content management framework CMF. CMF helps to accelerate the publication of large amounts of information and the development of Web applications. However, developing Web applications through CMF is not exempt from cost and time overruns, as in traditional software projects. Currently, there is no estimation model able to adequately measure the effort of developing a Web application.\",\n \"Title: A Curated Benchmark Collection of Python Systems for Empirical Studies on Software Engineering\\nYear: 2015\\n\\nAbstract: The aim of this paper is to present a dataset of metrics associated to the first release of a curated collection of Python software systems. We describe the dataset along with the adopted criteria and the issues we faced while building such corpus. This dataset can enhance the reliability of empirical studies, enabling their reproducibility, reducing their cost, and it can foster further research on Python software.\",\n \"Title: The JIRA Repository Dataset: Understanding Social Aspects of Software Development\\nYear: 2015\\n\\nAbstract: Issue tracking systems store valuable data for testing hypotheses concerning maintenance, building statistical prediction models and recently investigating developers \\\\\\\"affectiveness\\\\\\\". In particular, the Jira Issue Tracking System is a proprietary tracking system that has gained a tremendous popularity in the last years and offers unique features like the project management system and the Jira agile kanban board. This paper presents a dataset extracted from the Jira ITS of four popular open source ecosystems (as well as the tools and infrastructure used for extraction) the Apache Software Foundation, Spring, JBoss and CodeHaus communities. Our dataset hosts more than 1K projects, containing more than 700K issue reports and more than 2 million issue comments. Using this data, we have been able to deeply study the communication process among developers, and how this aspect affects the development process. Furthermore, comments posted by developers contain not only technical information, but also valuable information about sentiments and emotions. Since sentiment analysis and human aspects in software engineering are gaining more and more importance in the last years, with this repository we would like to encourage further studies in this direction.\",\n \"Title: 15 Years of Application of Statistical Physics Methods to the Study of Software Systems\\nYear: 2015\\n\\nAbstract: Despite the fact that software development is a design optimization process with nothing random, large software systems exhibit statistical properties that can be studied with the approach of statistical physics. In this keynote talk, I shall recall the history of the application of statistical physics methods to the study of software systems, since its inception at the beginning of 2000.\",\n \"Title: The Predictor Impact of Web Search Media on Bitcoin Trading Volumes.\\nYear: 2015\\n\\nAbstract: In the last decade, Web 2.0 services have been widely used as communication media. Due to the huge amount of available information, searching has become dominant in the use of Internet. Millions of users daily interact with search engines, producing valuable sources of interesting data regarding several aspects of the world. Search queries prove to be a useful source of information in financial applications, where the frequency of searches of terms related to the digital currency can be a good measure of interest in it. Bitcoin, a decentralized electronic currency, represents a radical change in financial systems, attracting a large number of users and a lot of media attention. In this work we studied the existing relationship between Bitcoinâs trading volumes and the queries volumes of Google search engine. We achieved significant cross correlation values, demonstrating search volumes power to anticipate trading volumes of Bitcoin currency.\",\n \"Title: Bitcoin Spread Prediction Using Social and Web Search Media.\\nYear: 2015\\n\\nAbstract: In the last decade, Web 2.0 services such as blogs, tweets, forums, chats, email etc. have been widely used as communication media, with very good results. Sharing knowledge is an important part of learning and enhancing skills. Furthermore, emotions may affect decisionmaking and individual behavior. Bitcoin, a decentralized electronic currency system, represents a radical change in financial systems, attracting a large number of users and a lot of media attention. In this work, we investigated if the spread of the Bitcoin’s price is related to the volumes of tweets or Web Search media results. We compared trends of price with Google Trends data, volume of tweets and particularly with those that express a positive sentiment. We found significant cross correlation values, especially between Bitcoin price and Google Trends data, arguing our initial idea based on studies about trends in stock and goods market.\",\n \"Title: Understanding approval rating of agile project management tools using Twitter\\nYear: 2015\\n\\nAbstract: The role of managing a software project can be extremely complicated, requiring many teams and organizational resources. Many people are acknowledging that Agile development is helpful to business, with an high increase over the last years in the number of people who believe that Agile helps companies to complete projects faster. Since there is a multitude of available tools that supports Agile methods, in this position paper we tried to understand which tools are most looked for using Web Search media, most mentioned and most appreciated through Social media. Furthermore, we applied automated Sentiment Analysis on shared short messages of users on Twitter, one of the most popular social networks, in order to analyze automatically developers' opinions, sentiments, evaluations and attitudes.\",\n \"Title: A Preliminary Study on Mobile Apps Call Graphs through a Complex Network Approach.\\nYear: 2015\\n\\nAbstract: Mobile apps are certainly a significant part of thesoftware industry market, and their centrality in our economicand social life is becoming more and more relevant as time goesby. In this preliminary work we propose a complex networkapproach, never used before in this field, to study mobileapps from a software engineering perspective. We analyzed thenetworks associated to 5 popular open-source android apps, and found that they share some properties with other kinds ofsoftware systems. We employed complex networks to obtain adeeper insight both on internal and external dependencies, aimingat studying the relationship between mobile apps dependenciesand software quality. Our goal is to better understand thenature of the existing relationship between mobile apps and theunderlying hosting operating system thus providing developerswith valuable information.\",\n \"Title: Re-visiting the 'Maintainability Index' Metric from an Object-Oriented Perspective\\nYear: 2015\\n\\nAbstract: The Maintainability Index (MI) metric was proposed in the early nineties to gauge ease of code maintenance. A high value of MI indicates well-constructed code which is easy to maintain and a low MI, the opposite. The metric has been criticized in the past and more recently for its applicability to code and validity more generally. Very few studies however have explored whether the MI correlates strongly with class features such as coupling, defects or size - features of object-oriented (OO) classes which are acknowledged to be surrogates of maintenance complexity. We explore the relationship between the MI and these four perspectives. Three releases of two Eclipse projects were used as the empirical basis of the study and the JHawk tool was used to extract class-based metrics. Significant correlations were found between class features and the MI for all class features except fan-in which measures incoming coupling and determined externally by 'using' classes, however, further exploration of the data revealed that the application of the MI to an OO paradigm was not particularly successful, due primarily to the influence of class size. For OO systems, we would thus warn against the use of MI as a measure of maintenance.\",\n \"Title: The evolution of knowledge in the refactoring research field\\nYear: 2015\\n\\nAbstract: Refactoring is certainly one of the most widespread practices used by developers to improve software quality. During the last two decades, it has been the subject of an increasing number of research studies. But how much do we know about the trends, and the appearance of new topics, in the research field of refactoring? What have researchers found about its application since the time of its first introduction? In this work we provide a preliminary analysis of the state of the art and the evolution of the research on refactoring. We attempt to represent the actual body of knowledge in this field through the analysis of its cognitive structure, leveraging science mapping methodology to focus on the most relevant concepts in this research area. We model the body of knowledge by mining bibliographic databases and by retrieving the co-occurrence of keywords. We have found that some different general themes can be recognized, but not all of them have the same role and equally catalyzed researchers' interest. In addition, we provide a preliminary analysis on the trends and directions as well as the hot topics we identified in the refactoring research field.\",\n \"Title: Are bullies more productive?: empirical study of affectiveness vs. issue fixing time\\nYear: 2015\\n\\nAbstract: Human Affectiveness, i.e., the emotional state of a person, plays a crucial role in many domains where it can make or break a team's ability to produce successful products. Software development is a collaborative activity as well, yet there is little information on how affectiveness impacts software productivity. As a first measure of this impact, this paper analyzes the relation between sentiment, emotions and politeness of developers in more than 560K Jira comments with the time to fix a Jira issue. We found that the happier developers are (expressing emotions such as JOY and LOVE in their comments), the shorter the issue fixing time is likely to be. In contrast, negative emotions such as SADNESS, are linked with longer issue fixing time. Politeness plays a more complex role and we empirically analyze its impact on developers' productivity.\",\n \"Title: How diverse is your team? Investigating gender and nationality diversity in GitHub teams.\\nYear: 2017\\n\\nAbstract: Building an effective team of developers is a complex task faced by both software companies and open source communities. The problem of forming a “dream”\",\n \"Title: Measuring high and low priority defects on traditional and mobile open source software.\\nYear: 2016\\n\\nAbstract: Software defects are the major cause for system failures. To effectively design tools and provide support for detecting and recovering from software failures, requires a deep understanding of defect features. In this paper we present an analysis of defect characteristics in two different open source software development domains: Mobile and Traditional. Our attention is focused on measuring the differences between High-Priority and Low-Priority defects. High or Low priority of a given defect is decided by a developer when creating a bug report for an issue tracking system. We sampled hundreds of real world bugs in hundreds of large and representative open-source projects. We used natural language text classification techniques to automatically analyse roughly 700,000 bug reports from the Bugzilla, Jira and Google Issues issue tracking systems. Results show that there are differences between High-Priority and Low-Priority defects classification in Mobile and Traditional development domains.\",\n \"Title: A statistical comparison of Java and Python software metric properties.\\nYear: 2016\\n\\nAbstract: This paper presents a statistical analysis of 20 opens ource object-oriented systems with the purpose of detecting differences in metrics distribution between Java and Python projects. We selected ten Java projects from the Java Qualitas Corpus and ten projects written in Python. For each system, we considered 10 class-level software metrics. We performed a best fit procedure on the empirical distributions through the log-normal distribution and the double Pareto distribution to identify differences between the two languages. Even though the statistical distributions for projects written in Java and Python may appear the same for lower values of the metric, performing the procedure with the double Pareto distribution for the Number of Local Methods metric reveals that major differences can be noticed along the queue of the distributions. On the contrary, the same analysis performed with the Number of Statements metric reveals that only the initial portion of the double Pareto distribution shows differences between the two languages. In addition, the dispersion parameter associated to the log-normal distribution fit for the total Number Of Methods can be used for distinguishing Java projects from Python projects.\",\n \"Title: A case study on the relationship between code ownership and refactoring activities in a Java software system.\\nYear: 2016\\n\\nAbstract: Refactoring, the activity of changing source code design without affecting its external behavior, is a widely used practice among developers, since it is considered to positively affect the quality of software systems. However, there are some \\\\\\\"human factors\\\\\\\" to be considered while performing refactoring, including developers knowledge of systems architecture. Recent studies showed how much \\\\\\\"people\\\\\\\" metrics, such as code ownership, might affect software quality as well. In this preliminary study we investigated the relationship between code ownership and refactoring activity performed by developers. This study can provide useful insights on who performs refactoring and help team leaders to properly manage human resources during software development.\",\n \"Title: Arsonists or Firefighters? Affectiveness in Agile Software Development.\\nYear: 2016\\n\\nAbstract: In this paper, we present an analysis of more than 500 K comments from open-source repositories of software systems developed using agile methodologies. Our aim is to empirically determine how developers interact with each other under certain psychological conditions generated by politeness, sentiment and emotion expressed within developers’ comments. Developers involved in an open-source projects do not usually know each other; they mainly communicate through mailing lists, chat, and tools such as issue tracking systems. The way in which they communicate affects the development process and the productivity of the people involved in the project. We evaluated politeness, sentiment and emotions of comments posted by agile developers and studied the communication flow to understand how they interacted in the presence of impolite and negative comments (and vice versa). Our analysis shows that “firefighters” prevail. When in presence of impolite or negative comments, the probability of the next comment being impolite or negative is 13 % and 25 %, respectively; ANGER however, has a probability of 40 % of being followed by a further ANGER comment. The result could help managers take control the development phases of a system, since social aspects can seriously affect a developer’s productivity. In a distributed agile environment this may have a particular resonance.\",\n \"Title: The emotional side of software developers in JIRA.\\nYear: 2016\\n\\nAbstract: Issue tracking systems store valuable data for testing hypotheses concerning maintenance, building statistical prediction models and (recently) investigating developer affectiveness. For the latter, issue tracking systems can be mined to explore developers emotions, sentiments and politeness---affects for short. However, research on affect detection in software artefacts is still in its early stage due to the lack of manually validated data and tools.\\n\\nIn this paper, we contribute to the research of affects on software artefacts by providing a labeling of emotions present on issue comments.\\n\\nWe manually labeled 2,000 issue comments and 4,000 sentences written by developers with emotions such as love, joy, surprise, anger, sadness and fear. Labeled comments and sentences are linked to software artefacts reported in our previously published dataset (containing more than 1K projects, more than 700K issue reports and more than 2 million issue comments). The enriched dataset presented in this paper allows the investigation of the role of affects in software development.\",\n \"Title: How Do Python Programs Use Inheritance? A Replication Study\\nYear: 2015\\n\\nAbstract: In this work we present an empirical study on the use of inheritance in a curated corpus of Python systems. Replicating a study preformed on Java, we analyzed a collection of 51 software systems written in Python, and investigated how inheritance is effectively used by Python developers in practice through a convenient set of inheritance metrics. Our results suggest that on average fewer classes inherit from other classes than in Java, but more classes are inherited from. We also see a sort of symmetry relating the number of ancestors and the number of descendants in each system.\",\n \"Title: Software Startups - A Research Agenda.\\nYear: 2016\\n\\nAbstract: Software startup companies develop innovative, software-intensive products within limited time frames and with few resources, searching for sustainable and scalable business models. Software startups are quite distinct from traditional mature software companies, but also from micro-, small-, and medium-sized enterprises, introducing new challenges relevant for software engineering research. This paper's research agenda focuses on software engineering in startups, identifying, in particular, 70+ research questions in the areas of supporting startup engineering activities, startup evolution models and patterns, ecosystems and innovation hubs, human aspects in software startups, applying startup concepts in non-startup environments, and methodologies and theories for startup research. We connect and motivate this research agenda with past studies in software startup research, while pointing out possible future directions. While all authors of this research agenda have their main background in Software Engineering or Computer Science, their interest in software startups broadens the perspective to the challenges, but also to the opportunities that emerge from multi-disciplinary research. Our audience is therefore primarily software engineering researchers, even though we aim at stimulating collaborations and research that crosses disciplinary boundaries. We believe that with this research agenda we cover a wide spectrum of the software startup industry current needs.\",\n \"Title: Estimating Story Points from Issue Reports.\\nYear: 2016\\n\\nAbstract: Estimating the effort of software engineering tasks is notoriously hard but essential for project planning. The agile community often adopts issue reports to describe tasks, and story points to estimate task effort. In this paper, we propose a machine learning classifier for estimating the story points required to address an issue. Through empirical evaluation on one industrial project and eight open source projects, we demonstrate that such classifier is feasible. We show that ---after an initial training on over 300 issue reports--- the classifier estimates a new issue in less than 15 seconds with a mean magnitude of relative error between 0.16 and 0.61. In addition, issue type, summary, description, and related components prove to be project dependent features pivotal for story point estimation.\",\n \"Title: Assessment Of Approaches For The Analysis Of Refactoring Activity On Software Repositories An Empirical Study\\nYear: 2016\\n\\nAbstract: Refactoring is the practice of changing source code without altering its external behavior. It is widely used since it is acknowledged to have a positive effect on software quality. However, different studies on the impact of Refactoring on software quality led to contrasting results. This might be due also to the fact that research works on Refactoring rely on different approaches to collect information about the refactoring activity on software repositories. Having a reliable approach is fundamental to draw solid conclusions. In this paper we empirically analyzed two of the most popular approaches to detect the refactoring activity on software repositories, finding that they return different outcomes. These results should be taken into account by researchers while designing the experimental settings of their studies.\",\n \"Title: Blockchain-oriented Software Engineering: Challenges and New Directions.\\nYear: 2017\\n\\nAbstract: In this work, we acknowledge the need for software engineers to devise specialized tools and techniques for blockchain-oriented software development. Ensuring effective testing activities, enhancing collaboration in large teams, and facilitating the development of smart contracts all appear as key factors in the future of blockchain-oriented software development.\",\n \"Title: Is Bitcoin's Market Predictable? Analysis of Web Search and Social Media.\\nYear: 2015\\n\\nAbstract: In recent years, Internet has completely changed the way real life works. In particular, it has been possible to witness the online emergence of web 2.0 services that have been widely used as communication media. On one hand, services such as blogs, tweets, forums, chats, email have gained wide popularity. On the other hand, due to the huge amount of available information, searching has become dominant in the use of Internet. Millions of users daily interact with search engines, producing valuable sources of interesting data regarding several aspects of the world. Bitcoin, a decentralized electronic currency, represents a radical change in financial systems, attracting a large number of users and a lot of media attention. In this work we studied whether Bitcoin’s trading volume is related to the web search and social volumes about Bitcoin. We investigated whether public sentiment, expressed in large-scale collections of daily Twitter posts, can be used to predict the Bitcoin market too. We achieved significant cross correlation outcomes, demonstrating the search and social volumes power to anticipate trading volumes of Bitcoin currency.\",\n \"Title: Measuring Affectiveness and Effectiveness in Software Systems.\\nYear: 2017\\n\\nAbstract: The summary presented in this paper highlights the results obtained in a four-years project aiming at analyzing the development process of software artifacts from two points of view: Effectiveness and Affectiveness. The first attribute is meant to analyze the productivity of the Open Source Communities by measuring the time required to resolve an issue, while the latter provides a novel approach for studying the development process by analyzing the affectiveness ex-pressed by developers in their comments posted during the issue resolution phase. Affectivenes is obtained by measuring Sentiment, Politeness and Emotions. All the study presented in this summary are based on Jira, one of the most used software repositories.\",\n \"Title: Connecting the Dots: Measuring Effectiveness and Affectiveness in Software Systems.\\nYear: 2017\\n\\nAbstract: The summary presented in this paper highlights the results obtained in a four-year project aiming at analyzing the development process of software artifacts from two points of view: Effectiveness and Affectiveness. The first attribute is meant to analyze the productivity of the Open Source Communities by measuring the time required to resolve an issue, while the latter provides a novel approach for studying the development process by analyzing the affectiveness expressed by developers in their comments posted during the issue resolution phase. Affectivenes is obtained by measuring Sentiment, Politeness and\",\n \"Title: On the randomness and seasonality of affective metrics for software development.\\nYear: 2017\\n\\nAbstract: The role of emotions and human aspects of software development is gaining increasing attention from both academics and developers. Software development is a complex activity, but since software is everywhere and it is necessary to achieve the highest quality possible. High technical skills, knowledge and competence of a software developer are not the only factors which influence the final product. Good attitudes, communication skills and good manners are equally as important. In this paper we focused our attention on understanding how developer sentiment and emotions evolved over time during the development process of 10 systems, considering affectiveness as time series. We studied seasonality, randomness and correlations. Results showed that there was not a significant correlation among joy and love, sadness and anger, sentiment and joy. Regarding seasonality and randomness, results showed that in the majority of the cases, the analysed time series are seasonal and not random.\",\n \"Title: Using measurement and simulation for understanding distributed development processes in the cloud.\\nYear: 2017\\n\\nAbstract: Organizations increasingly develop software in a distributed manner. The Cloud provides an environment to create and maintain software-based products and services. Currently, it is widely unknown which software processes are suited for Cloud-based development and what their effects in specific contexts are. This paper presents a process simulation to study distributed development in the Cloud. We contribute a simulation model, which helps analyzing different project parameters and their impact on projects carried out in the Cloud. The simulator helps reproducing activities, developers, issues and events in the project, and it generates statistics, e.g., on throughput, total time, and lead and cycle time. The aim of this simulation model is thus to analyze the tradeoffs regarding throughput, total time, project size, and team size. Furthermore, the modified simulation model aims to help project managers select the most suitable planning alternative. Based on observed projects in Finland and Spain, we simulated a distributed project using artificial and real data. Particularly, we studied the variables project size, team size, throughput, and total project duration. A comparison of the real project data with the results obtained from the simulation shows the simulation producing results close to the real data, and we could successfully replicate a distributed software project. By improving the understanding of distributed development processes, our simulation model thus supports project managers in their decision-making.\",\n \"Title: Software Quality And Community Structure In Java Software Networks\\nYear: 2017\\n\\nAbstract: We present a study of 600 Java software networks with the aim of characterizing the relationship among their defectiveness and community metrics. We analyze the community structure of such networks, defined as their topological division into subnetworks of densely connected nodes. A high density of connections represents a higher level of cooperation between classes, so a well-defined division in communities could indicate that the software system has been designed in a modular fashion and all its functionalities are well separated. We show how the community structure can be an indicator of well-written, high quality code by retrieving the communities of the analyzed systems and by ranking their division in communities through the built-in metric called modularity. We found that the software systems with highest modularity possess the majority of bugs, and tested whether this result is related to some confounding effect. We found two power laws relating the maximum defect density with two different metrics: the number of detected communities inside a software network and the clustering coeffcient. We finally found a linear correlation between clustering coefficient and number of communities. Our results can be used to make predictive hypotheses about software defectiveness of future releases of the analyzed systems.\",\n \"Title: Banking on Blockchain: Costs Savings Thanks to the Blockchain Technology.\\nYear: 2017\\n\\nAbstract: This paper looks at the challenges and opportunities of implementing blockchain technology across banking, providing food for thought about the potentialities of this disruptive technology. The blockchain technology can optimize the global financial infrastructure, achieving sustainable development, using more efficient systems than at present. In fact, many banks are currently focusing on blockchain technology to promote economic growth and accelerate the development of green technologies. In order to understand the potential of blockchain technology to support the financial system, we studied the actual performance of the Bitcoin system, also highlighting its major limitations, such as the significant energy consumption due to the high computing power required, and the high cost of hardware. We estimated the electrical power and the hash rate of the Bitcoin network, over time, and, in order to evaluate the efficiency of the Bitcoin system in its actual operation, we defined three quantities: \\\"economic efficiency\\\", \\\"operational efficiency\\\", and \\\"efficient service\\\". The obtained results show that by overcoming the disadvantages of the Bitcoin system, and therefore of blockchain technology, we could be able to handle financial processes in a more efficient way than under the current system.\",\n \"Title: A Petri Nets Model for Blockchain Analysis.\\nYear: 2018\\n\\nAbstract: A Blockchain is a global shared infrastructure where cryptocurrency transactions among addresses are recorded, validated and made publicly available in a peer-to-peer network. To date, the best known and important cryptocurrency is the bitcoin. In this paper, we focus on this cryptocurrency and in particular on the modeling of the Bitcoin Blockchain by using the Petri Nets formalism. The proposed ...\",\n \"Title: Smart Contracts Software Metrics: a First Study.\\nYear: 2018\\n\\nAbstract: Smart contracts (SC) are software codes which reside and run over a blockchain. The code can be written in different languages with the common purpose of implementing various kinds of transactions onto the hosting blockchain, They are ruled by the blockchain infrastructure and work in order to satisfy conditions typical of traditional contracts. The software code must satisfy constrains strongly context dependent which are quite different from traditional software code. In particular, since the bytecode is uploaded in the hosting blockchain, size, computational resources, interaction between different parts of software are all limited and even if the specific software languages implement more or less the same constructs of traditional languages there is not the same freedom as in normal software development. SC software is expected to reflect these constrains on SC software metrics which should display metric values characteristic of the domain and different from more traditional software metrics. We tested this hypothesis on the code of more than twelve thousands SC written in Solidity and uploaded on the Ethereum blockchain. We downloaded the SC from a public repository and computed the statistics of a set of software metrics related to SC and compared them to the metrics extracted from more traditional software projects. Our results show that generally Smart Contracts metrics have ranges more restricted than the corresponding metrics in traditional software systems. Some of the stylized facts, like power law in the tail of the distribution of some metrics, are only approximate but the lines of code follow a log normal distribution which reminds of the same behavior already found in traditional software systems.\",\n \"Title: Why blockchain is important for software developers, and why software engineering is important for blockchain software (Keynote)\\nYear: 2018\\n\\nAbstract: In the past few years, cryptocurrencies and blockchain applications has been one of the most rapidly emerging fields of computer science, leading to a strong demand of software applications. Several new projects have been emerging almost daily, with an impetus that was not seen since the days of the dawn of the Internet. However, the need of being timely on the market and the lack of experience in a brand new field led to epic disasters, such as those of DAO in 2016 and of Parity Ethereum wallet in 2017. Also, there have been several hacks successfully performed on cryptocurrency exchanges, the biggest being those of MtGox in 2014 (350 million US$), Bitfinex in 2016 (72 million US$), and Coincheck in 2017 (400 million US$). The application of sound SE practices to Blockchain software development, both for Smart Contract and generic Blockchain software, might be crucial to the success of this new field. Here the issues are the need for specific analysis and design methods, quality control through testing and metrics, security assessment and overall development process. At the same time, Blockchain development offers new opportunities, such as the certification of empirical data used for experiment; the ability to design processes where developers are paid upon completion of their tasks through Blockchain tokens, after acceptance tests performed using Smart Contracts; and more sound techniques enabling pay-per-use software, again using tokens.\",\n \"Title: Smart contracts vulnerabilities: a call for blockchain software engineering?\\nYear: 2018\\n\\nAbstract: Smart Contracts have gained tremendous popularity in the past few years, to the point that billions of US Dollars are currently exchanged every day through such technology. However, since the release of the Frontier network of Ethereum in 2015, there have been many cases in which the execution of Smart Contracts managing Ether coins has led to problems or conflicts. Compared to traditional Software Engineering, a discipline of Smart Contract and Blockchain programming, with standardized best practices that can help solve the mentioned problems and conflicts, is not yet sufficiently developed. Furthermore, Smart Contracts rely on a non-standard software life-cycle, according to which, for instance, delivered applications can hardly be updated or bugs resolved by releasing a new version of the software. In this paper we advocate the need for a discipline of Blockchain Software Engineering, addressing the issues posed by smart contract programming and other applications running on blockchains.We analyse a case of study where a bug discovered in a Smart Contract library, and perhaps \\\"unsafe\\\" programming, allowed an attack on Parity, a wallet application, causing the freezing of about 500K Ethers (about 150M USD, in November 2017). In this study we analyze the source code of Parity and the library, and discuss how recognised best practices could mitigate, if adopted and adapted, such detrimental software misbehavior. We also reflect on the specificity of Smart Contract software development, which makes some of the existing approaches insufficient, and call for the definition of a specific Blockchain Software Engineering.\",\n \"Title: The ICO phenomenon and its relationships with ethereum smart contract environment\\nYear: 2018\\n\\nAbstract: Initial Coin Offerings (ICO) are public offers of new cryptocurrencies in exchange of existing ones, aimed to finance projects in the blockchain development arena. In the last 8 months of 2017, the total amount gathered by ICOs exceeded 4 billion US$, and overcame the venture capital funnelled toward high tech initiatives in the same period. A high percentage of ICOs is managed through Smart Contracts running on Ethereum blockchain, and in particular to ERC-20 Token Standard Contract. In this work we examine 1387 ICOs, published on December 31, 2017 on icobench.com website, gathering information relevant to the assessment of their quality and software development management, including data on their development teams. We also study, at the same date, the financial data of 450 ICO tokens available on coinmarketcap.com website, among which 355 tokens are managed on Ethereum blochain. We define success criteria for the ICOs, based on the funds actually gathered, and on the behavior of the price of the related tokens, finding the factors that most likely influence the ICO success likeliness.\",\n \"Title: Analysing Developers Affectiveness through Markov chain Models.\\nYear: 2018\\n\\nAbstract: In this paper, we present an analysis of more than 500K comments from open-source repositories of software systems. Our aim is to empirically determine how developers interact with each other under certain psychological conditions generated by politeness, sentiment and emotion expressed in developersu0027 comments. Developers involved in open-source projects do not usually know each other; they mainly communicate through mailing lists, chat rooms, and tools such as issue tracking systems. The way in which they communicate affects the development process and the productivity of the people involved in the project. We evaluated politeness, sentiment, and emotions of comments posted by developers and studied the communication flow to understand how they interacted in the presence of impolite and negative comments (and vice versa). Our analysis shows that when in presence of impolite or negative comments, the probability of the next comment being impolite or negative is 14% and 25%, respectively; anger, however, has a probability of 40% of being followed by a further anger comment. The result could help managers take control the development phases of a system since social aspects can seriously affect a developeru0027s productivity. In a distributed environment this may have a particular resonance.\",\n \"Title: ICOs Overview: Should Investors Choose an ICO Developed with the Lean Startup Methodology?\\nYear: 2018\\n\\nAbstract: An Initial Coin Offering (ICO) is an innovative way to raise funds and launch a startup. It is also an opportunity to take part in a project, or in a DAO (Decentralized Autonomous Organization). The use of ICO is a global phenomenon that involves many nations and several business categories: ICOs collected over 5.2 billion dollars only in 2017. The success of an ICO is based on the credibility and innovativeness of project proposals. This fund-raising tool contains however some critical issues, such as the use of tokens that have no intrinsic value and do not generate direct liquidity, and the role of investors in the management of the startup. We analyzed if the Lean Startup methodology is helpful to face this critical aspects and we examined some ICOs in which the proposing team states explicitly that a lean startup approach is used.\",\n \"Title: Fast wavelet transform assisted predictors of streaming time series.\\nYear: 2018\\n\\nAbstract: We explore the shift variance of the decimated, convolutional Discrete Wavelet Transform, also known as Fast Wavelet Transform. We prove a novel theorem improving the FWT algorithm and implement a new prediction method suitable to the multiresolution analysis of streaming univariate datasets using compactly supported Daubechies Wavelets. An effective real value forecast is obtained synthesizing the one step ahead crystal and performing its inverse DWT, using an integrated group of estimating machines. We call Wa.R.P. (Wavelet transform Reduced Predictor) the new prediction method. A case study, testing a cryptocurrency exchange price series, shows that the proposed system can outperform the benchmark methods in terms of forecasting accuracy achieved. This result is confirmed by further tests performed on other time series. Developed in C++, Standard 2014 conformant, the code implementing the FWT and the novel Shift Variance Theorem is available to research purposes and to build efficient industrial applications.\",\n \"Title: On measuring affects of github issues' commenters.\\nYear: 2018\\n\\nAbstract: In this study, we analyzed issues and comments on GitHub projects and built collaboration networks dividing contributors into two categories: users and commenters. We identified as commenters those users who only post comments without posting any issues nor committing changes in the source code. Since previous studies showed that there is a link between a positive environment (regarding affectiveness) and productivity, our goal was to investigate commenters' contribution to the project concerning affectiveness.\\n\\nWe analyzed more than 370K comments from 100K issues of 25K contributors from 3 open source projects. We then calculated and compared the affectiveness of the issues' comments written by users and commenters in terms of sentiment, politeness, and emotions. We provide empirical evidence that commenters are less polite, less positive and in general they express a lower level of emotions in their comments than users. Our results also confirm that GitHub's contributors consist of different groups which behave differently, and this provides useful information for future studies in the field.\",\n \"Title: Using simulation for understanding and reproducing distributed software development processes in the cloud.\\nYear: 2018\\n\\nAbstract: Context: Organizations increasingly develop software in a distributed manner. The Cloud provides an environment to create and maintain software-based products and services. Currently, it is unknown which software processes are suited for Cloud-based development and what their effects in specific contexts are.\",\n \"Title: Re-visiting a Test Taxonomy with Refactoring and Defect-fix Data.\\nYear: 2018\\n\\nAbstract: In a previous empirical study by Bavota et al., multiple releases of three open-source systems reported the extent to which refactorings induced defect-fixes. In a much earlier study, van Deursen and Moonen (vDu0026M) provided a test taxonomy in which Fowleru0027s seventy-two refactorings were categorized according to the post refactoring test burden of each (i.e., the changes required to unit tests after each refactoring had been undertaken). A refactoring was categorized as u0027Type Bu0027 if it required no change to the original tests and u0027Type Eu0027 if significant changes were necessary. In this paper, we investigate nine refactorings spread across vDu0026Mu0027s taxonomy and the corresponding defect-fix data provided by Bavota et al., to explore the relationship between defect-fixes due to refactoring and vDu0026Mu0027s taxonomy. Results showed that, in contrast to our intuition, the most defect-fix prone refactorings were of Types C and D and not, as we thought, of Type E. The u0027Extract methodu0027 refactoring stood out as particularly u0027defect-fixu0027 inducing, suggesting that while it may solve one problem (i.e., in decomposing an excessively long method), it may well introduce other problems and required defect-fixes as a by-product.\",\n \"Title: Mining Communication Patterns in Software Development: A GitHub Analysis.\\nYear: 2018\\n\\nAbstract: Background: Studies related to human factors in software engineering are providing insightful information on the emotional state of contributors and the impact this has on the code. The open source software development paradigm involves different roles, and previous studies about emotions in software development have not taken into account what different roles might play when people express their feelings. Aim: We present an analysis of issues and commits on five GitHub projects distinguishing contributors between users and developers, and between one-commit and multi-commit developers. Method: We analyzed more than 650K comments from 130K issues of 64K contributors. We calculated emotions (love, joy, anger, sadness) and politeness of the comments related to the issues of the considered projects and introduced the definition of contributor fan-in and fan-out. Results: Results show that users and developers communicate differently as well as multi-commit developers and one-commit developers do. Conclusions: We provide empirical evidence that one-commit developers are more active and more polite in posting comments. Multi-commit developers are less active in posting comments, and while commenting, they are less polite than when commented.\",\n \"Title: Initial Coin Offerings and Agile Practices.\\nYear: 2018\\n\\nAbstract: An ICO (Initial Coin Offering) is an innovative way to fund projects based on blockchain. The funding is based on the selling of tokens by means of decentralized applications called smart contracts written in Solidity, a programming language specific for Ethereum blockchain. The ICOs work in a volatile context and it is crucial that the team is capable of handling constant changes. The Agile methods, proven practices enabling to develop software in presence of changing requirements, could be a means for managing uncertainty. The main goals of this work are to understand software engineering activities related to ICOs, recognize the ICOs developed using Agile methods, and make a comparison between ICOs and Agile ICOs. In addition, we perform a deeper analysis of Agile ICOs concerning project planning, software development, and code features. Our work shows that the roles of the people involved in an ICO can be compared to the typical roles of the SCRUM methodology. The majority of Agile ICOs use tool of testing before storing smart contract on blockchain. Finally, the application of volumetric and complexity software metrics shows that the files of Agile ICOs is on average shorter and less complex than in other smart contracts.\",\n \"Title: A Blockchain Based System to Ensure Transparency and Reliability in Food Supply Chain.\\nYear: 2018\\n\\nAbstract: We propose a blockchain oriented platform to secure storage origin provenance for food data. By exploiting the blockchain distributed and immutable nature the proposed system ensures the supply chain transparency with a view to encourage local region by promoting the smart food tourism and by increasing local economy. Thanks to the decentralized application platforms that makes us able to develop smart contracts, we define and implement a system that works inside the blockchain and guarantees transparency, reliability to all actors of the food supply chain. Food, in fact, is the most direct way to get in touch with a place. The touristic activities related to wine and food consumption and sale in fact influence the choice of a destination and may encourage the purchase of typical food also once tourists are back home to the country of origin. Touristic destinations must therefore be equipped with innovative tools that, in a context of Smart Tourism, guarantee the originality of the products and their traceability.\",\n \"Title: An Agent Based Model to Analyze the Bitcoin Mining Activity and a Comparison with the Gold Mining Industry.\\nYear: 2019\\n\\nAbstract: In this paper, we present an analysis of the mining process of two popular assets, Bitcoin and gold. The analysis highlights that Bitcoin, more specifically its underlying technology, is a safe haven that allows facing the modern environmental challenges better than gold. Our analysis emphasizes that crypto-currencies systems have a social and economic impact much smaller than that of the traditional financial systems. We present an analysis of the several stages needed to produce an ounce of gold and an artificial agent-based market model simulating the Bitcoin mining process and allowing the quantification of Bitcoin mining costs. In this market model, miners validate the Bitcoin transactions using the proof of work as the consensus mechanism, get a reward in Bitcoins, sell a fraction of them to cover their expenses, and stay competitive in the market by buying and divesting hardware units and adjusting their expenses by turning off/on their machines according to the signals provided by a technical analysis indicator, the so-called relative strength index.\",\n \"Title: Blockchain applications for agile methodologies.\\nYear: 2018\\n\\nAbstract: We present an application of Blockchain technology and Smart Contracts to the management of Agile projects, using Scrum or Lean-Kanban processes. In our application the duties of the Product Owner for certifying the correctness of the outcomes are delegated to one or more Smart Contracts deployed on the Ethereum Blockchain and written in Solidity. An agreement with the Customer can also allow the Smart Contracts to automatically enable payments, to introduce penalties or grants on the basis of the outcome. Product Owner duties and work can thus be relieved allowing to allocate resources on more profitable and productive tasks. Other possibilities are examined as well.\",\n \"Title: Angry-builds - an empirical study of affect metrics and builds success on github ecosystem.\\nYear: 2018\\n\\nAbstract: Automatic and repeatable builds are an established software engineering practices for achieving continuous integration and continuous delivery processes. The building phase of modern software systems is an important part of the development process such that dedicated roles as \\\"Release Engineer\\\" are more and more required. Software development is a collaborative activity, and when multiple developers work on the same project, they will be changing a shared master development branch at overlapping intervals. This overlap occurs because developers create parallel branches for working and then merge these branches when features are completed. Continuous integration, CI, is a workflow strategy which helps ensure everyoneâĂ&Zacutes changes will integrate with the current version of the project. This activity allows developers to catch bugs and reduce merge conflicts. Improving the building process leads to higher productivity and therefore shorter time to market, but understanding or measuring such a delicate phase is a big challenge. Open Source Communities provide valuable empirical data such as GitHub an Travis CI. These repositories represent a golden mine containing important data which can help researchers understanding the process behind the manufacturing of a software artifact. By analyzing Travis CI logs, we can directly connect a particular build with the development process behind it, not only regarding code changes but also regarding human activities, such as discussions about the implementation of a specific feature or bug resolution. Thanks to this information we can analyze the social activities of the build process enabling us to apply the same approach used for the development process.\",\n \"Title: An Agent-Based Artificial Market Model for Studying the Bitcoin Trading.\\nYear: 2019\\n\\nAbstract: The objective of this paper is to simulate the trading of the currency pair BTC/USD, investigating through the theory of the genetic algorithms the best sets of trading strategies, simulating through a realistic order book the bitcoin price formation, and reproducing a bitcoin price series that exhibits some stylized facts found in real-time price series. In this artificial market model two kinds of agents, Chartists and Random traders, perform trading. Chartists trade through the application of trading rules. Specifically, a part of Chartists trades applying the best sets of trading rules selected by a genetic algorithm that simulates a trading system, based on four technical analysis indicators, searching for parameters of each indicator that guarantee the highest profits in the training period; the remaining part trades applying trading rules choosing their parameters in a random way. On the contrary random trader's trade without applying any trading strategy, issuing in a random way sell or buy orders. Results show that the best sets of rules found to guarantee the highest profits both in the training and in the testing periods, and perform well also in the artificial market model where the Chartists who adopt the best sets of trading rules are able to achieve higher profits.\",\n \"Title: A Massive Analysis of Ethereum Smart Contracts Empirical Study and Code Metrics.\\nYear: 2019\\n\\nAbstract: In this work, we perform a comprehensive empirical study of smart contracts deployed on the ethereum blockchain. The objective of the analysis is to provide empirical results on smart contracts features, smart contract transactions within the blockchain, the role of the development community, and the source code characteristics. We collected a set of more than 10 000 smart contracts source codes and a dataset of meta-data regarding their interaction with the blockchain from etherscan.io . We examined the collected data computing different statistics on naming policies, smart contract ether balance, number of smart contract transactions, functions, and other quantities characterizing the use and purpose of smart contracts. We found that the number of transactions and the balances follow power-law distributions and the software code metrics display, on average, values lower than corresponding metrics in standard software but have high variances. Focusing the attention on the 20 smart contracts with the topmost number of transactions, we found that most of them represent financial smart contracts and some of them have peculiar software development stories behind them. The results show that blockchain software is rapidly changing and evolving and it is no longer devoted only to cryptovalues applications but to general purpose computation.\",\n \"Title: Empirical analysis of affect of merged issues on GitHub\\nYear: 2019\\n\\nAbstract: Pull-request based workflows are popular trends of modern software development platform such as GitHub. A pull-request notifies other developers that new changes are proposed, a code review process follows the pull-request that may be merged in the main branch if other developers accept the changes. Many factors influence the acceptance of pull-requests. Since open source software is based on collaboration, it is essential to discover how the affect expressed by developer discussing pull-request issues, namely how they collaborate, influences the acceptance of the pull-request proposed. In this study we analysed the relations with the affect expressed in pull-request issues' comments and whether an issue is merged in the main branch or not. We focused on pull-request issues and we found that issues with higher level anger, sadness, arousal and valence are less likely to be merged while issues with higher level of valence, joy are more likely to be merged. Positive affect indicates a good collaboration environment, and our finding shows that this healthy collaboration is likely to increase the acceptance of pull-requests.\",\n \"Title: A user-oriented model for Oracles’ Gas price prediction\\nYear: 2022\\n\\nAbstract: The Ethereum blockchain is a distributed database of transactions, where the Gas Oracles suggest the users the Gas price’s categories to get a transaction recorded. The paper explores the idea that the Gas Oracles are based on a data-centered model which does not provide users with a reliable prediction. We present an empirical study to test the reliability of the existing Gas Oracles from both the points of view of the Gas price predictions and the existing categories.\",\n \"Title: Blockchain and Self Sovereign Identity to Support Quality in the Food Supply Chain\\nYear: 2021\\n\\nAbstract: This work presents how a digital identity management system can support food supply chains in guaranteeing the quality of the products marketed and the compliance of the several supply-chain's nodes to standards and technical regulations. Specific goal of this work is to present a system that provides full visibility of process/food certifications, which nowadays are issued by accredited and approved certification bodies (issuers) and delivered and stored in paper version by the several participants (holders) of the supply chain. The system is designed and implemented by combining the latest most innovative and disruptive technologies in the market-Self Sovereign Identity system, Blockchain, and Inter Planetary File System. The crucial aspects that it aims to hit are the storage and access of food/process certifications, and the proper eligibility verification of these certifications exploiting the concepts of the Self Sovereign Identity-based models. The proposed system, realized by using standards that are WWW Consortium-compatible and the Ethereum Blockchain, ensures eligibility, transparency, and traceability of the certifications along a food supply chain, and could be an innovation model/idea that the companies that adopt the Open Innovation paradigm might want to pursue.\",\n \"Title: Bitcoin as a Safe Haven during COVID-19 Disease\\nYear: 2022\\n\\nAbstract: In this paper, we investigate the role of Bitcoin as a safe haven against the stock market losses during the spread of COVID-19. The performed analysis was based on a regression model with dummy variables defined around some crucial dates of the pandemic and on the dynamic conditional correlations. To try to model the real dynamics of the markets, we studied the safe-haven properties of Bitcoin against thirteen of the major stock market indexes losses using daily data spanning from 1 July 2019 until 20 February 2021. A similar analysis was also performed for Ether. Results show that this pandemic impacts on the Bitcoin status as safe haven, but we are still far from being able to define Bitcoin as a safe haven.\",\n \"Title: Automatic Generation of Ethereum-Based Smart Contracts for Agri-Food Traceability System\\nYear: 2022\\n\\nAbstract: There is a growing demand for transparency along the agri-food chain, both from customers and governments. The adoption of blockchain technology to enable secure traceability for the management of the agri-food chain, provide information such as the provenance of a food product and prevent food fraud, is emerging rapidly, due to the inherent trust and inalterability provided by this technology. However, developing the right smart contracts for these use cases is even more of a challenge than it is for those used in other fields. Several management systems for the agri-food chain based on blockchain technology and smart contract have been proposed, all however ad-hoc for a specific product or production process and difficult to generalize. In this paper, we propose a new approach to easily customize and compose general Ethereum-based smart contracts designed for the agri-food industrial domain, to be able to reuse the code and modules and automate the process to shorten development times, while keeping it safe and reliable. Starting from the definition of the real production process, we aim to automatically generate both the smart contracts to manage the system and the user interfaces to interact with them, thus producing a system that works semi-automatically. Additionally, we describe a honey production case study to show how our approach works. Future work will first extend the scope of the approach to other supply chains, furthermore, while the current platform used is Ethereum, in the future our approach will be easily extended to other blockchain platforms.\",\n \"Title: A System Proposal for Information Management in Building Sector Based on BIM, SSI, IoT and Blockchain\\nYear: 2022\\n\\nAbstract: This work presents a Self Sovereign Identity based system proposal to show how Blockchain, Building Information Modeling, Internet of Thing devices, and Self Sovereign Identity concepts can support the process of building digitalization, guaranteeing the compliance standards and technical regulations. The proposal ensures eligibility, transparency and traceability of all information produced by stakeholders, or generated by IoT devices appropriately placed, during the entire life cycle of a building artifact. By exploiting the concepts of the Self Sovereign Identity, our proposal allows the identification of all involved stakeholders, the storage off-chain of all information, and that on-chain of the sole data necessary for the information notarization and certification, adopting multi-signature approval mechanisms where appropriate. In addition it allows the eligibility verification of the certificated information, providing also useful information for facility management. It is proposed as an innovative system and companies that adopt the Open Innovation paradigm might want to pursue it. The model proposal is designed exploiting the Veramo platform, hence the Ethereum Blockchain, and all the recommendations about Self Sovereign Identity systems given by the European Blockchain Partnership, and by the World Wide Web Consortium.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":99,"string":"99"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13594,"string":"13,594"}}},{"rowIdx":13595,"cells":{"question_id":{"kind":"string","value":"Changming Sun"},"question":{"kind":"string","value":"Generate an abstract for the title \"Junction detection for linear structures based on Hessian, correlation and shape information\" using the following items: Junctions, detection, linear structures, orientation, accuracy."},"answer":{"kind":"string","value":"Junctions have been demonstrated to be important features in many visual tasks such as image registration, matching, and segmentation, as they can provide reliable local information. This paper presents a method for detecting junctions in 2D images with linear structures as well as providing the number of branches and branch orientations. The candidate junction points are selected through a new measurement which combines Hessian information and correlation matrix. Then the locations of the junction centers are refined and the branches of the junctions are found using the intensity information of a stick-shaped window at a number of orientations and the correlation value between the intensity of a local region and a Gaussian-shaped multi-scale stick template. The multi-scale template is used here to detect the structures with various widths. We present the results of our algorithm on images of different types and compare our algorithm with three other methods. The results have shown that the proposed approach can detect junctions more accurately."},"correct_answer":{"kind":"string","value":"Junctions have been demonstrated to be important features in many visual tasks such as image registration, matching, and segmentation, as they can provide reliable local information. This paper presents a method for detecting junctions in 2D images with linear structures as well as providing the number of branches and branch orientations. The candidate junction points are selected through a new measurement which combines Hessian information and correlation matrix. Then the locations of the junction centers are refined and the branches of the junctions are found using the intensity information of a stick-shaped window at a number of orientations and the correlation value between the intensity of a local region and a Gaussian-shaped multi-scale stick template. The multi-scale template is used here to detect the structures with various widths. We present the results of our algorithm on images of different types and compare our algorithm with three other methods. The results have shown that the proposed approach can detect junctions more accurately."},"documents":{"kind":"list like","value":["Title: Embedded Voxel Colouring\nYear: 2003\n\nAbstract: The reconstruction of a complex scene from multiple im- ages is a fundamental problem in the field of computer vision. Volu- metric methods have proven to be a strong alternative to traditional correspondence-based methods due to their flexible visibility models. In this paper we analyse existing methods for volumetric reconstruction and identify three key properties of voxel colouring algorithms: a water-tight surface model, a monotonic carving order, and causality .W e present a new Voxel Colouring algorithm which embeds all reconstructions of a scene into a single output. While modelling exact visibility for arbitrary camera locations, Embedded Voxel Colouring removes the need for a priori threshold selection present in previous work. An efficient imple- mentation is given along with results demonstrating the advantages of posteriori threshold selection.","Title: Stereoscopic flatbed scanner\nYear: 2009\n\nAbstract: 3-D scanning has become increasingly popular in a wide range of applications. We present a prototype 3-D/stereoscopic scanning system based on a cheap, readily available flatbed scanner. Stereoscopic imaging is achieved by modifying the optical path of the ordinary flatbed scanner. The results of 3-D imaging using our prototype system are demonstrated, and a number of design alternatives are discussed. (C) 2009 SPIE and IS&T. [DOI: 10.1117/1.3059582]","Title: Circular shortest paths by branch and bound\nYear: 2003\n\nAbstract: Shortest path algorithms are used for a large variety of optimisation problems in network and transportation analysis. They are also used in image analysis for object segmentation, disparity estimation, path finding and crack detection. Sometimes the topology of the problem demands that the path be circular. Such circular path constraints occur in polar object segmentation, disparity estimation for panoramic stereo images and in shortest paths around a cylinder. In this paper we present a new efficient algorithm for circular shortest path determination on a u-by-v trellis in O(u1.6v) average time. We impose a binary search tree on the set of path endpoints and use a best-first Branch and Bound search technique to efficiently obtain the global minimum circular path. The typical running time of our circular shortest path algorithm on a 256×256 image is in the order of 0.1s on a 1GHz Dell P3 workstation under the Linux operating system. Applications to crack detection and object segmentation are presented.","Title: Mosaicing of Microscope Images\nYear: 2005\n\nAbstract: Image mosaicing has found a number of applications such as panoramic imaging, digital terrain mapping, ophthalmology, and virtual microscopy. In this paper, we present an automated mosaicing technique for generating mosaics from microscope images. We carry out robust image feature matching and global geometric and radiometric parameter estimation. The images are transformed using the estimated parameters and mosaiced together, producing accurate registration of overlapping regions without visible seams.","Title: Priority-Based Path Growing for Linear Feature Detection\nYear: 2007\n\nAbstract: Linear feature detection is a very important step in image analysis, computer vision, and pattern recognition. It has a wide range of applications. In this paper, we present a new algorithm for linear feature detection using priority- based path growing from automatically selected seeds. The use of path growing in the neighbourhood of automatically selected seeds ensures that all local information along a linear feature is taken into account. This enables tracing of weakly contrasted linear features in an image. We illustrate our method with examples.","Title: Automating the quantification of membrane proteins under confocal microscopy.\nYear: 2012\n\nAbstract: In a prior contribution, we described a semi-automated system that allowed a user to quantify the relative abundance of fluorescently labeled membrane proteins in confocal microscopy images. Here, we describe a step change in assay automation, enabled by explicitly casting the problem in terms of mathematical graphs. This permitted to include extensive and relevant image information in the tracing process; something not possible when using the marginally faster traditional algorithms. These improvements bring us closer to an industrial strength membrane tracing assay suitable in a drug discovery and development environment.","Title: Membrane boundary extraction using circular multiple paths\nYear: 2009\n\nAbstract: Membrane proteins represent over 50% of known drug targets. Accordingly, several widely used assays in the high content analysis area rely on quantitative measures of the translocation of proteins between intracellular organelles and the cell surface. In order to increase the sensitivity of these assays, one needs to measure the signal specifically along the membrane, requiring a precise segmentation of this compartment. Manual tracing of membrane boundary is very time-consuming and confronts us with issues of objectivity and reproducibility. In this paper, we present an approach based on a circular multiple paths technique on transformed images that enables us to segment the membrane compartment accurately and rapidly. We have presented three approaches for image transformation. The circular property of the multiple paths ensures that we are obtaining closed contours for the membrane boundary. The position of the multiple paths provides the edges of the membrane boundary. The effectiveness of our algorithm is illustrated using cells expressing epitope-tagged membrane proteins.","Title: Moving average algorithms for diamond, hexagon, and general polygonal shaped window operations\nYear: 2006\n\nAbstract: This paper presents fast moving window algorithms for calculating local statistics in a diamond, hexagon, and general polygonal shaped windows of an image which is important for real-time applications. The algorithms for a diamond shaped window requires only seven or eight additions and subtractions per pixel. A fast sparse algorithm only needs four additions and subtractions for a sparse diamond shaped window. A number of other shapes of diamond windows such as skewed or parallelogram shaped diamond, long diamond, and lozenged diamond shaped, are also investigated. Similar algorithms are also developed for hexagon shaped windows. The computation for a hexagon window only needs eight additions and subtractions for each pixel. Fast algorithms for general polygonal shaped windows are also developed. The computation cost of all these algorithms is independent of the window size. A variety of synthetic and real images have been tested.","Title: Cross image inference scheme for stereo matching\nYear: 2012\n\nAbstract: In this paper, we propose a new interconnected Markov Random Field (MRF) or iMRF model for the stereo matching problem. Comparing with the standard MRF, our model takes into account the consistency between the label of a pixel in one image and the labels of its possible matching points in the other image. Inspired by the turbo decoding scheme, we formulate this consistency by a cross image reference term which is iteratively updated in our matching framework. The proposed iMRF model represents the matching problem better than the standard MRF and gives better results even without using any other information from segmentation prior or occlusion detection. We incorporate segmentation information and the coarse-to-fine scheme into our model to further improve the matching performance.","Title: Clustered Nuclei Splitting Using Curvature Information\nYear: 2011\n\nAbstract: Automated splitting of clustered nuclei from images of tissue sections is essential to many biomedical studies. Many existing image segmentation methods tend to produce over-segmented or under-segmented results for clustered nuclei images. In this paper, a new curvature information based image segmentation algorithm is proposed. Through combining curvature information with a distance map, our algorithm can extract correct markers corresponding to each nucleus. Afterwards, marker based watershed segmentation is used to segment the clustered nuclei. The algorithm is tested on both synthetic and real images. Experimental results show that our algorithm is accurate and robust to noise in segmentation of clustered nuclei.","Title: Fast optical flow using 3D shortest path techniques\nYear: 2002\n\nAbstract: Optical flow or image motion estimation is important in the area of computer vision. This paper presents a fast and reliable optical flow algorithm which produces a dense optical flow map by using fast cross correlation and 3D shortest path techniques. Fast correlation is achieved by using the box-filtering technique which is invariant to the size of the correlation window. The motion for each scanline or each column of the input image is obtained from the correlation coefficient volume by finding the best 3D path using dynamic programming techniques rather than simply choosing the position that gives the maximum cross correlation coefficient. Sub-pixel accuracy is achieved by fitting the local correlation coefficients to a quadratic surface. Typical running time for a 256×256 image is in the order of a few seconds on a 85MHz computer. A variety of synthetic and real images have been tested, and good results have been obtained.","Title: 3D Symmetry Detection Using The Extended Gaussian Image\nYear: 1997\n\nAbstract: Symmetry detection is important in the area of computer vision. A 3D symmetry detection algorithm is presented in this correspondence. The symmetry detection problem is converted to the correlation of the Gaussian image. Once the Gaussian image of the object has been obtained, the algorithm is independent of the input format. The algorithm can handle different kinds of images or objects. Simulated and real images have been tested in a variety of formats, and the results show that the symmetry can be determined using the Gaussian image.","Title: Iterative infrared ship target segmentation based on multiple features.\nYear: 2014\n\nAbstract: This paper presents an efficient method for ship target segmentation in infrared (IR) images. It consists of mainly two procedures: iterative image segmentation and ship target selection. First, based on the intensity distribution of an IR image, we design a global background subtraction filter (GBSF) to suppress the background, and an adaptive row mean subtraction filter (ARMSF) to enhance the target. After iteratively applying these two filters, we can obtain a proper threshold for image segmentation. Second, based on the geometric properties of the ship target, we construct four shape features and a selection criterion to identify the real target and remove the non-target regions. Experimental results demonstrate that the proposed method can effectively segment ship targets from different backgrounds in IR images. The advantage of the proposed method over the others in the previous literatures is validated in both visual and quantitative comparisons, especially for IR images with low contrast and uneven intensities.","Title: Automatic segmentation of fiducial marks using attribute-based mathematical morphology\nYear: 2001\n\nAbstract: This paper described a fast and automatic method for segmenting fiducial marks in an image taken with a metric camera. These marks could be at the four corners, four middle sides, or inside an image (reseau marks). The segmentation was realized by using attribute-based mathematical morphology techniques. The attributes that we used in the morphology processing step were the area, aspect ratio, and orientation of the best fitted ellipse of an object. The algorithm took about 0.25 s in the automatic segmentation stage. The positions of these automatically segmented fiducial marks were then refined to subpixel accuracy. Dozens of real images with different shapes of fiducial marks were tested and reliable results were obtained. Accuracy comparisons were made using our method and previously published methods. (C) 2001 SPIE and IS&T.","Title: Segmentation of clustered nuclei based on curvature weighting\nYear: 2012\n\nAbstract: Cluster of nuclei are frequently observed in thick tissue section images. It is very important to segment overlapping nuclei in many biomedical applications. Many existing methods tend to produce under segmented results when there is a high overlap rate. In this paper, we present a curvature weighting based algorithm which weights each pixel using the curvature information of its nearby boundaries to extract markers, each of which represents an object, from input images. Then we use marker-controlled watershed to obtain the final segmentation. Test results using both synthetic and real cell images are presented in the paper.","Title: Tree structural watershed for stereo matching\nYear: 2012\n\nAbstract: We present a new method for dense stereo matching based on a tree structural cost volume watershed (TSCVW) and a region combination (RC) process. Given a cost volume as the data cost and an initial segmentation result, the proposed TSCVW method reliably estimates the disparities in a segment by using energy optimization to control plane segmentation and plane fitting. Then the disparities in the incorrectly fitted and occluded regions are refined using our RC process. Experimental results show that our method is very robust to different initial segmentation results and the shape of a segment. The comparison between our algorithm and the current state-of-the-art algorithms on the Middlebury website shows that our algorithm is very competitive.","Title: Multiple paths extraction in images using a constrained expanded trellis.\nYear: 2005\n\nAbstract: Single shortest path extraction algorithms have been used in a number of areas such as network flow and image analysis. In image analysis, shortest path techniques can be used for object boundary detection, crack detection, or stereo disparity estimation. Sometimes one needs to find multiple paths as opposed to a single path in a network or an image where the paths must satisfy certain constraints. In this paper, we propose a new algorithm to extract multiple paths simultaneously within an image using a constrained expanded trellis (CET) for feature extraction and object segmentation. We also give a number of application examples for our multiple paths extraction algorithm.","Title: Multi-Image 2D-PAGE Feature Detection\nYear: 2005\n\nAbstract: This paper proposes an algorithm for decomposing a multi-class classification problem into a set of two-class classification problems. The algorithm divides a set of input pattern vectors in each class into subsets according to the distribution of the ...","Title: Fast Stereo Matching by Iterated Dynamic Programming and Quadtree Subregioning\nYear: 2004\n\nAbstract: The application of energy minimisation methods for stereo matching has been demonstrated to produce high quality disparity maps. However the ma- jority of these methods are known to be computationally expensive, requir- ing minutes or even hours of computation. We propose a fast minimisation scheme that produces strongly competitive results for significantly reduced computation, requiring only a few seconds of computation. In this paper, we present our iterated dynamic programming algorithm along with a quadtree subregioning process for fast stereo matching.","Title: De-interlacing of video images using a shortest path technique\nYear: 2001\n\nAbstract: This paper presents a fast algorithm for de-interlacing of video images using a shortest path technique. The algorithm applies dynamic programming techniques to find a shortest path in a cost matrix. The motion information obtained from this shortest path is used to re-align the fields of a video image. By using the shortest path via dynamic programming, the motion information estimated is more reliable than simply performing a search in a local neighbourhood. A variety of real images have been tested, and good results have been obtained","Title: Detection of Nuclear Buds Based on Ellipse Fitting\nYear: 2010\n\nAbstract: Assays of micronucleus are extensively used in genotoxicity testing and in monitoring of human exposure to genotoxic materials. As nuclear buds could be a new source of micronuclei formed in interphase, the assay of contents of nuclear bud in normal and comparison group is needed. In this paper, we proposed a new automatic nuclear buds detection algorithm based on ellipse fitting. Experimental results show that our algorithm is effective and efficient. We believe that this is the first report on automated nuclear bud detection for DNA damage scoring.","Title: Touching Cells Splitting by Using Concave Points and Ellipse Fitting\nYear: 2008\n\nAbstract: A novel touching cells splitting algorithm by using concave points and ellipse fitting is proposed to split circle-like or ellipse-like touching cells in this paper. The algorithm is divided into two parts. The first part is contour pre-processing, whose purpose is to find the concave points of the contour and separate the contour into different segments using the concave points. The second part is ellipse processing, whose purpose is to process the different segments of the contour as a possible single cell by using the properties of the fitted ellipse. The second part is the main contribution of this paper which demonstrates a new way to use ellipse fitting to split the binary contour of the touching cells. Experimental results show that the proposed algorithm is efficient.","Title: Closed-form stereo image rectification\nYear: 2012\n\nAbstract: Stereo image rectification transforms a pair of images into a new pair such that the epipolar lines in the transformed images have the same direction as the image rows and matching epipolar lines in different images have the same row index to enable an efficient and reliable dense stereo matching. In this paper we propose a first closed-form algorithm to automatically rectify stereo images just using the fundamental matrix. The algorithm involves only direct and purely geometric transformation processes. There is no iterative or optimization processes for the rectification parameter estimation in our method. Real images have been used for testing purposes, and convincing results have been obtained from our algorithm. Comparison with the state-of-the-art method shows that our method produces better rectification results.","Title: Fast algorithm for local statistics calculation for N-dimensional images\nYear: 2001\n\nAbstract: This paper presents a fast algorithm for calculating these local statistics in awindow of an N-dimensional image. The new algorithm, which is called the plunger method,recursively reduces the dimensions of the input N-dimensional image to achieve fast computation.The speed of the algorithm is independent of the window size. Another advantage of thealgorithm is that it calculates the local statistics in one pass. Real image tests have beenperformed","Title: Circular shortest path in images\nYear: 2003\n\nAbstract: Shortest path algorithms have been used in a number of applications such as crack detection, road or linear feature extraction in images. There are applications where the starting and ending positions of the shortest path need to be constrained. In this paper, we present several new algorithms for the extraction of a circular shortest path in an image such that the starting and ending positions coincide. The new algorithms we developed include multiple search algorithm, image patching algorithm, multiple backtracking algorithm, the combination of image patching and multiple back-tracking algorithm, and approximate algorithm. The typical running time of our circular shortest path extraction algorithm on a 256×256 image is about 0.3s on a rather slow 85MHz Sun SPARC computer. A variety of real images for crack detection in borehole data and object boundary extraction have been tested and good results have been obtained.","Title: Linear Feature Detection on GPUs\nYear: 2010\n\nAbstract: The acceleration of an existing linear feature detection algorithm for 2D images using GPUs is discussed. The two most time consuming components of this process are implemented on the GPU, namely, linear feature detection using dual-peak directional non-maximum suppression, and a gap filling process that joins disconnected feature masks to rectify false negatives. Multiple steps or image filters in each component are combined into a single GPU kernel to minimise data transfers to off-chip GPU RAM, and issues relating to on-chip memory utilisation, caching, and memory coalescing are considered. The presented algorithm is useful for applications needing to analyse complex linear structures, and examples are given for dense neurite images from the biotech domain.","Title: Thickness measurement and crease detection of wheat grains using stereo vision\nYear: 2007\n\nAbstract: Wheat grain quality assessment is important in meeting market requirements. The thickness of grains can be used for the measurement of the mass proportion of grains that pass through a sieve. This measure is known as ''screenings''. The determination of the presence or absence of the grain crease aids the detection of a stain called blackpoint, which is usually most evident on the non-crease side of the grain. In this paper we investigate the use of stereo vision techniques for measuring the thickness and detecting the presence or absence of the crease of a sample of wheat grains placed on a tray with dimples.","Title: Diamond, Hexagon, and General Polygonal Shaped Window Smoothing\nYear: 2003\n\nAbstract: This paper presents fast recursive or moving windows algo- rithms for calculating local means in a diamond, hexagon and general polygonal shaped windows of an image. The algorithms for diamond shaped window require only seven or eight additions and subtractions per pixel. A number of other shapes of diamond windows such as skewed or parallelogram shaped diamond, long diamond, and lozenged diamond shaped, are also investigated. Similar algorithms are also developed for hexagon shaped windows. The computation for hexagon window only needs eight additions and subtractions for each pixel. Fast algorithms for general polygonal shaped windows are also developed. The computa- tion costs of all these algorithms are independent of the window size. A variety of synthetic and real images have been tested.","Title: Automated Feature Weighting in Fuzzy Declustering-based Vector Quantization\nYear: 2010\n\nAbstract: Feature weighting plays an important role in improving the performance of clustering technique. We propose an automated feature weighting in fuzzy declustering-based vector quantization (FDVQ), namely AFDVQ algorithm, for enhancing effectiveness and efficiency in classification. The proposed AFDVQ imposes weights on the modified fuzzy c-means (FCM) so that it can automatically calculate feature weights based on their degrees of importance rather than treating them equally. Moreover, the extension of FDVQ and AFDVQ algorithms based on generalized improved fuzzy partitions (GIFP), known as GIFP-FDVQ and GIFP-AFDVQ respectively, are proposed. The experimental results on real data (original and noisy data) and modified data (biased and noisy-biased data) have demonstrated that the proposed algorithms outperformed standard algorithms in classifying clusters especially for biased data.","Title: 2-Point linear algorithm for camera translation vector estimation with known rotation\nYear: 2000\n\nAbstract: This paper introduces a simple method for obtaining the pure 3-D translational motion parameter (without rotation) and the translation parameter with a known rotation of a camera. Once the translation vector is available, the 3-D structure of an object can also be obtained. The minimum number of matching points in the two images can be as few as two, and the equations to be solved are linear. Synthetic and real image tests has been performed, and good results have been obtained.","Title: Applications of heterogeneous computing in computational and simulation science\nYear: 2013\n\nAbstract: This paper presents a summary and overview of heterogeneous algorithms and applications developed by the Commonwealth Scientific and Industrial Research Organisation CSIRO for solving practical and challenging science problems such as: 1 high-content analysis of brain cell images for medical research and drug discovery; 2 genetic analysis of complex experimental designs for crop breeding; 3 solving computational fluid dynamics problems; 4 deconvolving 3D images from microscopy and medical imaging; 5 reconstructing large 3D computed tomography CT images from medical and materials science; 6 quantifying uncertainty in complex environmental models. Applications discussed utilise GPUs and multicore CPUs on a scale ranging from single desktop workstations through to large GPU clusters. Results demonstrate that both types of system can successfully accelerate a variety of practical science applications. We have seen significant gains in productivity and opportunity as a result of leveraging GPUs, tackling computational problems in which execution time was previously infeasible.","Title: Automated Detection of Nucleoplasmic Bridges for DNA Damage Scoring in Binucleated Cells\nYear: 2010\n\nAbstract: Quantification of DNA damage, which may be caused by radiation or exposure to chemicals, is very important and can be very time consuming and subject to variability if carried out visually. The quantification of scoring DNA damage includes biomarkers such as micronuclei, nucleoplasmic bridges, and nuclear buds as scored in cytokinesis-blocked binucleated cells. In this paper, we present a new algorithm based on a shortest path technique that enables us to detect the nucleoplasmic bridges joining two nuclei in cell images of binucleated cells. The effectiveness of our algorithm is illustrated using a set of cell images. We believe that this is the first time that a feasible automated nucleoplasmic bridge detection system has been reported.","Title: Embedded Voxel Colouring with Adaptive Threshold Selection Using Globally Minimal Surfaces\nYear: 2012\n\nAbstract: Image-based 3D reconstruction remains a competitive field of research as state-of-the-art algorithms continue to improve. This paper presents a voxel-based algorithm that adapts the earliest space-carving methods and utilises a minimal surface technique to obtain a cleaner result. Embedded Voxel Colouring is built in two stages: (a) progressive voxel carving is used to build a volume of embedded surfaces and (b) the volume is processed to obtain a surface that maximises photo-consistency data in the volume. This algorithm combines the strengths of classical carving techniques with those of minimal surface approaches. We require only a single pass through the voxel volume, this significantly reduces computation time and is the key to the speed of our approach. We also specify three requirements for volumetric reconstruction: monotonic carving order, causality of carving and water-tightness. Experimental results are presented that demonstrate the strengths of this approach.","Title: Uncalibrated three-view image rectification\nYear: 2003\n\nAbstract: Image rectification is a process of transforming a set of images into a new set such that the epipolar lines in the transformed images have the same direction as the image rows or columns to enable an efficient and reliable stereo matching. Previous algorithms for stereo image rectification either work for two view uncalibrated or two/three view calibrated situations. In this paper we propose several novel techniques to rectify uncalibrated trinocular images using the trilinear tensor or projective invariants or fundamental matrices obtained from a triplet of images. Our new methods include: a rotation and skew method, an affine transformation method, and a vanishing points method. Real images have been used for testing purposes, and accurate results have been obtained.","Title: Splitting touching cells based on concave points and ellipse fitting\nYear: 2009\n\nAbstract: A new touching cells splitting algorithm based on concave points and ellipse fitting is proposed in this paper. The algorithm includes two parts: contour pre-processing and ellipse processing. The purpose of contour pre-processing is to smooth fluctuations of the contour, find concave points of the contour and divide the contour into different segments via the concave points. The purpose of ellipse processing is to process the different segments of the contour into possible single cells by using the properties of the fitted ellipses. Because concave points divide the whole contour of touching cells into different segments and different segments of one single cell have similar properties, the ellipse processing can separate the touching cells through ellipse fitting. This paper demonstrates a new way of using ellipse fitting to split the binary contour of touching cells. Experimental results show that our algorithm is efficient.","Title: An Energy Minimisation Approach to Stereo-Temporal Dense Reconstruction\nYear: 2004\n\nAbstract: We propose a novel energy minimisation framework for the dense reconstruction of stereo image sequences that incorporates data delity as well as spatial and temporal regularity. An iterated dynamic programming scheme is proposed to minimise the energy function. We also present an efficient implementation of the minimisation scheme by introducing morphological decomposition techniques to solve the dynamic programming subproblem. Our proposed method is capable of reconstructing dynamic scenes with complex motion. Results are presented demonstrating the strength of our proposed algorithm.","Title: Symmetry detection using gradient information\nYear: 1995\n\nAbstract: Symmetry detection is important in the area of computer vision. A simple and fast symmetry detection algorithm has been developed in this paper. The algorithm employs only the original image and the gradient information. The direction of the symmetry axis is obtained from the gradient orientation histogram; and the position of this symmetry axis is decided either by the center of gravity or by the profile of the image projection along the direction of the symmetry axis. This method works directly on the grey-scale image and does not require any prior segmentation of the input image. Both simulated and real images have been tested and the results are very convincing.","Title: Boundary extraction of linear features using dual paths through gradient profiles\nYear: 2008\n\nAbstract: An algorithm for automated extraction of linear feature boundaries in 2D images is presented. From a marker set approximating the medial axis, we generate 1D gradient profiles orthogonal to linear features. The algorithm uses dual shortest paths through an image generated from gradient profiles to extract boundaries. The algorithm offers an alternative to a watershed type approach and performs well on images with noise and areas of low contrast. We present the results of our algorithm on microscopy images of neurite outgrowth and other images containing linear features.","Title: Iterated dynamic programming and quadtree subregioning for fast stereo matching\nYear: 2008\n\nAbstract: The application of energy minimisation methods for stereo matching has been demonstrated to produce high quality disparity maps. However, the majority of these methods are known to be computationally expensive requiring minutes of computation. In this paper, we propose a fast minimisation scheme that produces high quality stereo reconstructions for significantly reduced running time, requiring only a few seconds of computation. The minimisation scheme is carried out using our iterated dynamic programming algorithm, which iterates over entire rows and columns for fast stereo matching. A quadtree subregioning process is also used for efficient computation of a matching cost volume where iterated dynamic programming operates on.","Title: Fast linear feature detection using multiple directional non-maximum suppression.\nYear: 2006\n\nAbstract: The capacity to detect linear features is central to image analysis, computer vision and pattern recognition and has practical applications in areas such as neurite outgrowth detection, retinal vessel extraction, skin hair removal, plant root analysis and road detection. Linear feature detection often represents the starting point for image segmentation and image interpretation. In this paper, we present a new algorithm for linear feature detection using multiple directional non-maximum suppression with symmetry checking and gap linking. Given its low computational complexity, the algorithm is very fast. We show in several examples that it performs very well in terms of both sensitivity and continuity of detected linear features.","Title: Fast panoramic stereo matching using cylindrical maximum surfaces.\nYear: 2004\n\nAbstract: This paper presents a fast panoramic stereo matching algorithm using a cylindrical maximum surface technique. The disparity for a pair of panoramic images is found in a cylindrical shaped correlation coefficient volume by obtaining the maximum surface rather than simply choosing a position that gives the maximum correlation coefficient value. The use of our cylindrical maximum surface technique ensures that the disparities obtained at the left and the right columns of the panoramic stereo images are properly constrained. Typical running time for a pair of 1324 x 120 images is about 0.33 s on a 1.7-GHz PC. A variety of real images have been tested, and good results have been obtained.","Title: Fast Stereo Matching Using Rectangular Subregioning and 3D Maximum-Surface Techniques\nYear: 2002\n\nAbstract: This paper presents a fast and reliable stereo matching algorithm which produces a dense disparity map by using fast cross correlation, rectangular subregioning (RSR) and 3D maximum-surface techniques in a coarse-to-fine scheme. Fast correlation is achieved by using the box-filtering technique whose speed is invariant to the size of the correlation window and by segmenting the stereo images into rectangular subimages at different levels of the pyramid. By working with rectangular subimages, not only can the speed of the correlation be further increased, the intermediate memory storage requirement can also be reduced. The disparity map for the stereo images is found in the 3D correlation coefficient volume by obtaining the global 3D maximum-surface rather than simply choosing the position that gives the local maximum correlation coefficient value for each pixel. The 3D maximum-surface is obtained using our new two-stage dynamic programming (TSDP) technique. There are two original contributions in this paper: (1) development of the RSR technique for fast similarity measure; and (2) development of the TSDP technique for efficiently obtaining 3D maximum-surface in a 3D volume. Typical running time of our algorithm implemented in the C language on a 512 × 512 image is in the order of a few seconds on a 500 MHz PC. A variety of synthetic and real images have been tested, and good results have been obtained.","Title: Skew and slant correction for document images using gradient direction\nYear: 1997\n\nAbstract: A fast algorithm is presented for skew and slant correction in printed document images. The algorithm employs only the gradient information. The skew angle is obtained by searching for a peak in the histogram of the gradient orientation of the input grey-level image. The skewness of the document is corrected by a rotation at such an angle. The slant of characters can also be detected using the same technique, and can be corrected by a shear operation. A second method for character slant correction by fitting parallelograms to the connected components is also described. Document images with different contents (tables, figures, and photos) have been tested for skew correction and the algorithm gives accurate results on all the test images, and the algorithm is very easy to implement","Title: Linear feature enhancement based on morphological operation and Gabor function\nYear: 2012\n\nAbstract: In many image processing tasks, feature enhancement is an important preprocessing step. It is the process of adjusting images so that the results are more suitable for analysis. Good enhancement results of linear features such as vessels or neurites can be inputs for segmentation or tracking. In this paper, we propose a novel linear feature enhancement method based on morphological operation and Gabor function, which can enhance and smooth linear features. We use and improve the Hessian matrix to calculate the orientation information for our directional morphological operation and Gabor smoothing. An approach for junction enhancement in each branch is also proposed here. We present the results of our algorithm on images of different types. The obtained outputs show that the proposed approach can achieve very good results.","Title: Trinocular Stereo Image Rectification In Closed-Form Only Using Fundamental Matrices\nYear: 2013\n\nAbstract: Trinocular stereo image rectification is a process to transform a set of three images into a new set so that the epipolar lines in the transformed images have the same direction as the image row or column and matching epipolar lines in different images have the same row or column indices to enable an efficient and reliable dense stereo matching. In this paper we propose a new closed-form method to rectify three-view stereo images just using fundamental matrices. The algorithm involves only direct and purely geometric transformation processes. No iteration or optimization process is involved in our method. Real images have been used for testing purposes, and accurate results have been obtained using our algorithm.","Title: Stereo matching using cost volume watershed and region merging.\nYear: 2014\n\nAbstract: Segment based disparity estimation methods have been proposed in many different ways. Most of these studies are built upon the hypothesis that no large disparity jump exists within a segment. When this hypothesis does not hold, it is difficult for these methods to estimate disparities correctly. Therefore, these methods work well only when the images are initially over segmented but do not work well for under segmented cases. To solve this problem, we present a new segment based stereo matching method which consists of two algorithms: a cost volume watershed algorithm (CVW) and a region merging (RM) algorithm. For incorrectly under segmented regions where pixels on different objects are grouped into one segment, the CVW algorithm regroups the pixels on different objects into different segments and provides disparity estimation to the pixels in different segments accordingly. For unreliable and occluded regions, we merge them into neighboring reliable segments for robust disparity estimation. The comparison between our method and the current state-of-the-art methods shows that our method is very competitive and is robust particularly when the images are initially under segmented.","Title: A novel method for dendritic spines detection based on directional morphological filter and shortest path.\nYear: 2014\n\nAbstract: Dendritic spines are tiny membranous protrusions from neuron's dendrites. They play a very important role in the nervous system. A number of mental diseases such as Alzheimer's disease and mental retardation are revealed to have close relations with spine morphologies or spine number changes. Spines have various shapes, and spine images are often not of good quality; hence it is very challenging to detect spines in neuron images. This paper presents a novel pipeline to detect dendritic spines in 2D maximum intensity projection (MIP) images and a new dendrite backbone extraction method is developed in the pipeline. The strategy for the backbone extraction approach is that it iteratively refines the extraction result based on directional morphological filtering and improved Hessian filtering until a satisfactory extraction result is obtained. A shortest path method is applied along a backbone to extract the boundary of the dendrites. Spines are then segmented from the dendrites outside the extracted boundary. Touching spines will be split using a marker-controlled watershed algorithm. We present the results of our algorithm on real images and compare our algorithm with two other spine detection methods. The results show that the proposed approach can detect dendrites and spines more accurately. Measurements and classification of spines are also made in this paper.","Title: A new method for linear feature and junction enhancement in 2D images based on morphological operation, oriented anisotropic Gaussian function and Hessian information.\nYear: 2014\n\nAbstract: Feature enhancement is an important preprocessing step in many image processing tasks. It is the process of adjusting image intensities so that the enhanced results are more suitable for analysis. Good enhancement results for linear structures such as vessels or neurites can be used as inputs for segmentation and other operations. In this paper, a novel linear feature enhancement filter – an adaptive multi-scale morpho-Gaussian filter – which can enhance and smooth linear features is proposed based on morphological operation, anisotropic Gaussian function and Hessian information. This filter can enhance and smooth along the local orientation of the linear structures and the Hessian measurement is used to further enhance the linear features. We utilize the Hessian matrix to calculate the orientation information for our directional morphological operation and the oriented anisotropic Gaussian smoothing. We also propose a novel method for junction enhancement, which can solve the problem of junction suppression. We decompose the junctions and enhance along each linear structure within a junction region. We present the test results of our algorithm on images of different types and compare our method with three existing methods. The experimental results show that the proposed approach can achieve better results.","Title: Multipoint Filtering with Local Polynomial Approximation and Range Guidance\nYear: 2014\n\nAbstract: This paper presents a novel guided image filtering method using multipoint local polynomial approximation (LPA) with range guidance. In our method, the LPA is extended from a pointwise model into a multipoint model for reliable filtering and better preserving image spatial variation which usually contains the essential information in the input image. In addition, we develop a scheme with constant computational complexity (invariant to the size of filtering kernel) for generating a spatial adaptive support region around a point. By using the hybrid of the local polynomial model and color/intensity based range guidance, the proposed method not only preserves edges but also does a much better job in preserving spatial variation than existing popular filtering methods. Our method proves to be effective in a number of applications: depth image upsampling, joint image denoising, details enhancement, and image abstraction. Experimental results show that our method produces better results than state-of-the-art methods and it is also computationally efficient.","Title: Infrared Ship Target Image Smoothing Based on Adaptive Mean Shift\nYear: 2014\n\nAbstract: Infrared (IR) image denoising is important for IR image analysis. In this paper, we propose a method based on adaptive range bandwidth mean shift for IR ship target image smoothing, aiming to effectively suppress noise as well as preserve important target structures. First, local image properties, including the mean value and standard deviation, are combined to build a salient region map, and a thresholding method is applied to obtain a binary mask on the target region. Then, we develop an adaptive range bandwidth mean shift method for image denoising. By associating the range bandwidth of the mean shift with local region saliency, we can adjust the bandwidth adaptively, thus to smooth the background region while preserving important target structures. Experimental results show that this method works well for IR ship target images with different backgrounds. It demonstrates superior performance for image denoising and target preserving, compared with some existing image denoising methods.","Title: Soft Cost Aggregation With Multi-Resolution Fusion\nYear: 2014\n\nAbstract: This paper presents a simple and effective cost volume aggregation framework for addressing pixels labeling problem. Our idea is based on the observation that incorrect labelings are greatly reduced in cost volume aggregation results from low resolutions. However, image details may be lost in the low resolution results. To take advantage of the results from low resolution for reducing these incorrect labelings while preserving details, we propose a multi-resolution cost aggregation method (MultiAgg) by using a soft fusion scheme based on min-convolution. We implement our MultiAgg in applications on stereo matching and interactive image segmentation. Experimental results show that our method significantly outperforms conventional cost aggregation methods in labeling accuracy. Moreover, although MultiAgg is a simple and straight-forward method, it produces results which are close to or even better than those from iterative methods based on global optimization.","Title: A Modular Learning Approach for Fish Counting and Measurement Using Stereo Baited Remote Underwater Video\nYear: 2014\n\nAbstract: An approach is suggested for automating fish identification and measurement using stereo Baited Remote Underwater Video footage. Simple methods for identifying fish are not sufficient for measurement, since the snout and tail points must be found, and the stereo data should be incorporated to find a true measurement. We present a modular framework that ties together various approaches in order to develop a generalised system for automated fish detection. A method is also suggested for using machine learning to improve identification. Experimental results indicate the suitability of our approach.","Title: Dendritic spines detection based on directional morphological filter and shortest path.\nYear: 2012\n\nAbstract: Dendritic spines are membranous protrusions from neuron's dendrites. They play a very important role in the nervous system. They are very small and have various shapes; hence it is very challenging to detect them in neuron images. This paper presents a novel method for detecting dendritc spines in 2D images. A new dendrite backbone or centerline extraction method is introduced herein which is based on an iterative process between smoothing and extraction. The proposed method iteratively refines the extraction result using both directional morphological filtering and improved Hessian filtering until a satisfactory extraction result is obtained. A shortest path method is applied along a backbone to extract the boundary of the dendrites. Spines are then segmented from the dendrites outside the extracted boundary. The proposed algorithm has been tested with many images and good results are achieved.","Title: Tubule detection in testis images using boundary weighting and circular shortest path.\nYear: 2013\n\nAbstract: In studies of germ cell transplantation, measureing tubule diameters and counting cells from different populations using antibodies as markers are very important. Manual measurement of tubule sizes and cell counts is a tedious and sanity grinding work. In this paper, we propose a new boundary weighting based tubule detection method. We first enhance the linear features of the input image and detect the approximate centers of tubules. Next, a boundary weighting transform is applied to the polar transformed image of each tubule region and a circular shortest path is used for the boundary detection. Then, ellipse fitting is carried out for tubule selection and measurement. The algorithm has been tested on a dataset consisting of 20 images, each having about 20 tubules. Experiments show that the detection results of our algorithm are very close to the results obtained manually.","Title: Finding splitting lines for touching cell nuclei with a shortest path algorithm.\nYear: 2015\n\nAbstract: A shortest path-based algorithm is proposed in this paper to find splitting lines for touching cell nuclei. First, an initial splitting line is obtained through the distance transform of a marker image and the watershed algorithm. The initial splitting line is then separated into different line segments as necessary, and the endpoint positions of these line segments are adjusted to the concave points on the contour. Finally, a shortest path algorithm is used to find the accurate splitting line between the starting-point and the end-point, and the final split can be achieved by the contour of the touching cell nuclei and the splitting lines. Comparisons of experimental results show that the proposed algorithm is effective for segmentation of different types of touching cell nuclei.","Title: Edge-Aware Filtering with Local Polynomial Approximation and Rectangle-Based Weighting.\nYear: 2016\n\nAbstract: This paper presents a novel method for performing guided image filtering using local polynomial approximation (LPA) with range guidance. In our method, the LPA is introduced into a multipoint framework for reliable model regression and better preservation on image spatial variation which usually contains the essential information in the input image. In addition, we develop a weighting scheme which...","Title: Stereo Vision Technologies For China'S Lunar Rover Exploration Mission\nYear: 2016\n\nAbstract: In December 2013, China's Chang'e-3 system successfully landed on the moon and started carrying out rover exploration. Stereo vision technologies have played important roles in terrain mapping and rover localization from beginning to end in this mission. This article mainly discusses two stereo vision implementations. First, using stereo images, we propose a novel solution for terrain reconstruction that combines both geometry and appearance information within a joint model. This model could provide accurate terrain maps for the rover navigation and virtual reality simulations of the mission. Then, we present a binocular visual odometry algorithm which leverages an image-network-based bundle adjustment to obtain rover's position and orientation from cross-site images. In addition, both field experiments and onboard results are presented to verify the effectiveness of the proposed algorithms. These technologies have contributed substantially to the Chang'e-3 mission and help to lay a foundation for future autonomous missions.","Title: Orientation-guided geodesic weighting for PatchMatch-based stereo matching\nYear: 2016\n\nAbstract: We propose an orientation-guided geodesic weighting (OGGW) strategy for local stereo matching.We propose a method of cost volume filtering combining a multipoint LPA method with our OGGW strategy.We propose a PatchMatch filter with curved surface fitting (PMF-CS) to obtain a disparity map with sub-pixel accuracy. Recently, PatchMatch-based methods for local stereo matching are experiencing great progress with the use of compact and over-segmented regions that have similar intensities or colors. Using patches as support regions, this paper proposes an orientation-guided geodesic weighting (OGGW) strategy to search for an approximate shortest path from a support pixel in the patch to a pixel of interest along a guided orientation. The OGGW is computed by accumulating intensity differences or color dissimilarities between connected pixels along the path. After obtaining matching cost updates by model fitting, the OGGW is used for weighted averaging on the updated costs to obtain a filtered cost volume. In addition, a new filtering method that combines the PatchMatch filter with curved surface fitting (PMF-CS) is presented in this paper. Curved surface fitting along with outliers removal is carried out to seek for a reliable regression model for estimating the disparities on a patch and to achieve a disparity map with sub-pixel accuracy. We conduct a number of experiments to evaluate the performances of OGGW and PMF-CS on cost volume filtering and disparity estimation. Experimental results show that our algorithm produces accurate stereo matching results and outperforms the current state-of-the-art PatchMatch-based methods.","Title: Guided image completion by confidence propagation\nYear: 2016\n\nAbstract: This paper presents a new guided image completion method which fills any missing values by considering information from a guidance image. We develop a confidence propagation scheme that allows the filling process to be carried out simultaneously without the need of deciding the filling order. We conduct experiments in several applications where the problem can be formulated into a guided image completion problem, such as interactive segmentation and colorization. The experimental results show that our method provides good results and succeeds in situations where conventional methods fail. In addition, as all building blocks are parallel processes, our method is much suitable for hardware acceleration. HighlightsWe proposed an image completion method based on a new formulation.Filling process was carried out simultaneously without deciding the filling order.Good results were obtained in a variety of applications.","Title: Feature matching in stereo images encouraging uniform spatial distribution\nYear: 2015\n\nAbstract: Finding feature correspondences between a pair of stereo images is a key step in computer vision for 3D reconstruction and object recognition. In practice, a larger number of correct correspondences and a higher percentage of correct matches are beneficial. Previous researches show that the spatial distribution of correspondences are also very important especially for fundamental matrix estimation. So far, no existing feature matching method considers the spatial distribution of correspondences. In our research, we develop a new algorithm to find good correspondences in all the three aspects mentioned, i.e., larger number of correspondences, higher percentage of correct correspondences, and better spatial distribution of correspondences. Our method consists of two processes: an adaptive disparity smoothing filter to remove false correspondences based on the disparities of neighboring correspondences and a matching exploration algorithm to find more correspondences according to the spatial distribution of correspondences so that the correspondences are as uniformly distributed as possible in the images. To find correspondences correctly and efficiently, we incorporate the cheirality constraint under an epipole polar transformation together with the epipolar constraint to predict the potential location of matching point. Experiments demonstrate that our method performs very well on both the number of correct correspondences and the percentage of correct correspondences; and the obtained correspondences are also well distributed over the image space. HighlightsWe propose a feature matching method incorporating epipolar geometry estimation.Uniformly distributed correspondences improve epipolar geometry estimation.We propose an adaptive weighted median filter for removing false correspondences.Good results are obtained on both epipolar geometry estimation and correspondences.","Title: Stereo matching based on multi-direction polynomial model.\nYear: 2016\n\nAbstract: This paper presents a segmentation based stereo matching algorithm. For the purposes of both preserving the shape of object surfaces and being robust to under segmentations, we introduce a new scene formulation where the reference image is divided into overlapping lines. The disparity value and the index of pixels on lines are modeled by polynomial functions. Polynomial functions are propagated among lines to obtain smooth surfaces via solving energy minimizing problems. Finally, the disparity of pixels is estimated from the disparity fields provided by lines. Because lines in multiple directions implicitly segment different objects in an under segmentation region, our method is robust for under segmented regions where it is usually difficult for conventional region based methods to produce satisfactory results. Experimental results demonstrate that the proposed method has an outstanding performance compared with the current state-of-the-art methods. The scene representation method in this work is also a powerful approach to surface based scene representations.","Title: Infrared ship target segmentation through integration of multiple feature maps.\nYear: 2016\n\nAbstract: We investigate the issue of ship target segmentation in infrared (IR) images, and propose an efficient method based on feature map integration. It consists of mainly two procedures: salient region detection based on multiple feature map integration and salient region segmentation based on locally adaptive thresholding. Firstly, a saliency map is constructed by integrating multiple features of IR ship targets, including gray level intensity, local contrast, salient linear structures, and edge strength. Secondly, we propose an adaptive thresholding method to segment each local salient region, and a target selection procedure based on shape features is used to remove background and obtain the true target. Experimental results show that the proposed method performs well for IR ship target segmentation. The advantage of the proposed method is demonstrated in both visual and quantitative comparisons, especially for IR images with a bright background or a ship target close to port. Proposing a saliency map detection method through multiple feature map integration.Proposing an adaptive thresholding method for local salient region segmentation.Achieving a better performance for both visual and quantitative evaluations.","Title: Automated Image Analysis on Insect Soups\nYear: 2016\n\nAbstract: Insect soup is a term which refers to the bulk ethanol samples stored in many insect collections which contain unsorted material and augment the dried and pinned collections. These samples often contain rare or unique specimens, and it would be beneficial to researchers or even the general public to have an easy way to access them to view and annotate their contents. In this paper we propose an image analysis approach for analysing insect soup images and making their content readily available through the web. The analysis includes insect segmentation, scale bar segmentation, measurements, and insect sorting. The image analysis results will be used for online access of insect images for insect classification, research, species discovery, or educational purposes.","Title: A Local Scale Selection Scheme for Multiscale Area Integral Invariants\nYear: 2016\n\nAbstract: Area integral invariant (AII) is a functional obtained by performing integral operations on the closed planar contour of a shape via the convolution with disc kernels. This shape descriptor is insensitive to noise and robust with respect to occlusions. AII intrinsically introduces the notion of scale using the size of kernel radius. However how to select an optimal scale remains unresolved. In this paper, we propose a local scale selection scheme for generating multiscale area integral invariants. For the same scale level, the disc kernel size is not fixed and varies with the contour point where the disc is centered. This scheme also provides a scale assignment for emphasising the features extraction at finer scales. The strong discriminative power of the multiscale area integral invariant derived from the proposed scale selection scheme has been validated through experiments on very challenging leaf image retrievals.","Title: Robust Multi-Frame Adaptive Optics Image Restoration Algorithm Using Maximum Likelihood Estimation with Poisson Statistics.\nYear: 2017\n\nAbstract: An adaptive optics (AO) system provides real-time compensation for atmospheric turbulence. However, an AO image is usually of poor contrast because of the nature of the imaging process, meaning that the image contains information coming from both out-of-focus and in-focus planes of the object, which also brings about a loss in quality. In this paper, we present a robust multi-frame adaptive optics image restoration algorithm via maximum likelihood estimation. Our proposed algorithm uses a maximum likelihood method with image regularization as the basic principle, and constructs the joint log likelihood function for multi-frame AO images based on a Poisson distribution model. To begin with, a frame selection method based on image variance is applied to the observed multi-frame AO images to select images with better quality to improve the convergence of a blind deconvolution algorithm. Then, by combining the imaging conditions and the AO system properties, a point spread function estimation model is built. Finally, we develop our iterative solutions for AO image restoration addressing the joint deconvolution issue. We conduct a number of experiments to evaluate the performances of our proposed algorithm. Experimental results show that our algorithm produces accurate AO image restoration results and outperforms the current state-of-the-art blind deconvolution methods.","Title: Fog Density Estimation and Image Defogging Based on Surrogate Modeling for Optical Depth.\nYear: 2017\n\nAbstract: In order to estimate fog density correctly and to remove fog from foggy images appropriately, a surrogate model for optical depth is presented in this paper. We comprehensively investigate various fog-relevant features and propose a novel feature based on the hue, saturation, and value color space, which correlate well with the perception of fog density. We use a surrogate-based method to learn a ...","Title: Multi-Modal Ship Target Image Smoothing Based on Adaptive Mean Shift.\nYear: 2018\n\nAbstract: In this paper, we propose an adaptive image smoothing method for infrared (IR) and visual ship target images, aiming to effectively suppress noise as well as preserve important target structures, thus benefiting image segmentation. First, by analyzing the specific features of ship target images, a block based method combining local region mean and standard deviation is developed to highlight ship target regions. It is helpful to distinguish the ship target region from the background. Then, by associating the range bandwidth with local image properties of the ship target region and the background region, we develop an adaptive range bandwidth mean shift filtering method for IR and visual ship target image smoothing. With this proposed method, we can obtain a small bandwidth for ship target region and a large one for the background region. Therefore, we can effectively smooth the background while preserving the details of the targets. Experimental results show that this method works well for IR and visual ship target images with different backgrounds. The method demonstrates superior performance for image smoothing and target preservation compared with the four well-known edge-preserving denoising methods, including the anisotropic diffusion filtering, the bilateral filtering, the propagation filtering, and the mean shift filtering with fixed range bandwidth.","Title: Matching Pursuit Based On Kernel Non-Second Order Minimization\nYear: 2018\n\nAbstract: The orthogonal matching pursuit (OMP) is an important sparse approximation algorithm to recover sparse signals from compressed measurements. However, most MP algorithms are based on the mean square error(MSE) to minimize the recovery error, which is suboptimal when there are outliers. In this paper, we present a new robust OMP algorithm based on kernel non-second order statistics (KNS-OMP), which not only takes advantages of the outlier resistance ability of correntropy but also further extends the second order statistics based correntropy to a non-second order similarity measurement to improve its robustness. The resulted framework is more accurate than the second order ones in reducing the effect of outliers. Experimental results on synthetic and real data show that the proposed method achieves better performances compared with existing methods.","Title: Disparity-based Error Concealment for Stereoscopic Images with Superpixel Segmentation.\nYear: 2018\n\nAbstract: To solve the problem of transmission errors in stereoscopic images, this paper proposes a novel error concealment (EC) method using superpixel segmentation and adaptive disparity selection (SSADS). Our algorithm consists of two steps. The first step is disparity estimation for each pixel in a reference image. In this step, the numbers of superpixel segmentation labels of stereoscopic images are used as a new constraint for disparity matching to reduce the effect of mismatching. The second step is disparity selection for a lost block. In this step, a strategy based on boundary smoothness is proposed to adaptively select the optimal disparity which is used for error concealment. Experimental results demonstrate that compared with other methods, the proposed method has significant advantages in both objective and subjective quality assessment.","Title: RA-UNet: A hybrid deep attention-aware network to extract liver and tumor in CT scans.\nYear: 2018\n\nAbstract: Automatic extraction of liver and tumor from CT volumes is a challenging task due to their heterogeneous and diffusive shapes. Recently, 2D and 3D convolutional neural networks have become popular in medical image segmentation tasks because of the utilization of large labeled datasets to learn hierarchical features. However, 3D networks have some drawbacks due to their high cost on computational resources. In this paper, we propose a 3D hybrid residual attention-aware segmentation method, named RA-UNet, to precisely extract the liver volume of interests (VOI) and segment tumors from the liver VOI. The proposed network has a basic architecture as a 3D U-Net which extracts contextual information combining low-level feature maps with high-level ones. Attention modules are stacked so that the attention-aware features change adaptively as the network goes very deep and this is made possible by residual learning. This is the first work that an attention residual mechanism is used to process medical volumetric images. We evaluated our framework on the public MICCAI 2017 Liver Tumor Segmentation dataset and the 3DIRCADb dataset. The results show that our architecture outperforms other state-of-the-art methods. We also extend our RA-UNet to brain tumor segmentation on the BraTS2018 and BraTS2017 datasets, and the results indicate that RA-UNet achieves good performance on a brain tumor segmentation task as well.","Title: Design of a Clinical Decision Support System for Predicting Erectile Dysfunction in Men Using NHIRD Dataset.\nYear: 2019\n\nAbstract: Erectile dysfunction (ED) affects millions of men worldwide. Men with ED generally complain failure to attain or maintain an adequate erection during sexual activity. The prevalence of ED is strongly correlated with age, affecting about 40% of men at age 40 and nearly 70% at age 70. A variety of chronic diseases, including diabetes, ischemic heart disease, congestive heart failure, hypertension, depression, chronic renal failure, obstructive sleep apnea, prostate disease, gout, and sleep disorder, were reported to be associated with ED. In this study, data retrieved from a subset of the National Health Insurance Research Database (NHIRD) of Taiwan were used for designing the clinical decision support system (CDSS) for predicting ED incidences in men. The positive cases were male patients aged 20-65 who were diagnosed with ED between Jan. 2000 and Dec. 2010 confirmed by at least 3 outpatient visits or at least one inpatient visit, while the negative cases were randomly selected from the database without a history of ED and were frequency (1:1), age, and index year matched with the ED patients. Data of a total of 2,832 ED patients and 2,832 non-ED patients, each consisting of 41 features including index age, 10 comorbidities, and 30 other comorbidity-related variables, were retrieved for designing the predictive models. Integrated genetic algorithm (GA) and support vector machine (SVM) was adopted to design the CDSSs with 2 experiments of independent training and testing (ITT) conducted to verify their effectiveness. In the 1st ITT experiment, data extracted from Jan. 2000 till Dec. 2005 (61.51%, 1,742 positive cases and 1,742 negative cases) were used for training and validating and the data retrieved from Jan. 2006 till Dec. 2010 were used for testing (38.49%); whereas in the 2nd ITT experiment, data in the training set (77.78%) were extracted from Jan. 2000 till Dec. 2007 and those in the testing set (22.22%) were retrieved afterward. Tenfold cross validation and 3 different objective functions were adopted for obtaining the optimal models with best predictive performance in the training phase. The testing results show that the CDSSs achieved a predictive performance with accuracy, sensitivity, specificity, g-mean, and area under ROC curve (AUC) of 74.72-76.65%, 72.33-83.76%, 69.54-77.10%, 0.7468-0.7632, and 0.766-0.817, respectively. In conclusion, the CDSSs designed based on cost-sensitive objective functions as well as salient comorbidity-related features achieve satisfactory predictive performance for predicting ED incidences.","Title: Image dehazing using adaptive bi-channel priors on superpixels.\nYear: 2017\n\nAbstract: •We propose an adaptive bi-channel priors (ABiCP) method for image dehazing.•We propose a method for determining the white and black pixels in haze images based on the HSV space.•We introduce the superpixels in image dehazing to reduce computational time and improve the quality of restored images.","Title: Cell Segmentation Based on FOPSO Combined with Shape Information Improved Intuitionistic FCM.\nYear: 2019\n\nAbstract: Fuzzy c-means (FCM) clustering algorithms have been proved to be effective image segmentation techniques. However, FCM clustering algorithms are sensitive to noises and initialization. They cannot effectively segment cell images with inhomogeneous gray value distributions and complex touching cells. Aiming to overcome these disadvantages, this paper proposes a cell image segmentation algorithm usi...","Title: Discrete Curvature Representations for Noise Robust Image Corner Detection.\nYear: 2019\n\nAbstract: Image corner detection is very important in the fields of image analysis and computer vision. Curvature calculation techniques are used in many contour-based corner detectors. We identify that existing calculation of curvature is sensitive to local variation and noise in the discrete domain and does not perform well when corners are closely located. In this paper, discrete curvature representations of single and double corner models are investigated and obtained. A number of model properties have been discovered which help us detect corners on contours. It is shown that the proposed method has a high corner resolution (the ability to accurately detect neighbouring corners) and a corresponding corner resolution constant is also derived. Meanwhile, this method is less sensitive to any local variations and noise on the contour; and false corner detection is less likely to occur. The proposed detector is compared with seven state-of-the-art detectors. Three test images with ground truths are used to assess the detection capability and localization accuracy of these methods in noise-free and cases with different noise levels. Twenty-four images with various scenes without ground truths are used to evaluate their repeatability under affine transformation, JPEG compression, and noise degradations. The experimental results show that our proposed detector attains a better overall performance.","Title: Robust Retinal Image Enhancement via Dual-Tree Complex Wavelet Transform and Morphology-Based Method.\nYear: 2019\n\nAbstract: Retinal image processing is very important in the field of clinical medicine. As the first step in retinal image processing, image enhancement is essential. Because the details of a retinal image are complex and difficult to enhance, we present a robust retinal image enhancement algorithm via a dual-tree complex wavelet transform (DTCWT) and morphology-based method in this paper. To begin with, we utilize the pre-processing method to the captured retinal images. Then, the DTCWT is applied to decompose the gray retinal image to obtain high-pass subbands and low-pass subbands. Then, a Contourlet-based enhancement method is applied to the high-pass subbands. For the low-pass subbands, we improve the morphology top-hat transform by adding dynamic multi-scale parameters to achieve an equivalent percentage enhancement and at the same time achieve multi-scale transforms in multiple directions. Finally, we develop the inverse DTCWT method to obtain the enhanced retinal image after processing the low-frequency subimages and high-frequency subimages. We compare this approach with enhancement based on the adaptive unsharp masking, histogram equalization, and multi-scale retinex. We present the test results of our algorithm on 440 retinal images from the DRIVE and the STARE databases. The experimental results show that the proposed approach can achieve better results, and might be helpful for vessel segmentation.","Title: Knowledge Adaptation for Efficient Semantic Segmentation\nYear: 2019\n\nAbstract: Both accuracy and efficiency are of significant importance to the task of semantic segmentation. Existing deep FCNs suffer from heavy computations due to a series of high-resolution feature maps for preserving the detailed knowledge in dense estimation. Although reducing the feature map resolution (i.e., applying a large overall stride) via subsampling operations (e.g., polling and convolution striding) can instantly increase the efficiency, it dramatically decreases the estimation accuracy. To tackle this dilemma, we propose a knowledge distillation method tailored for semantic segmentation to improve the performance of the compact FCNs with large overall stride. To handle the inconsistency between the features of the student and teacher network, we optimize the feature similarity in a transferred latent domain formulated by utilizing a pre-trained autoencoder. Moreover, an affinity distillation module is proposed to capture the long-range dependency by calculating the non local interactions across the whole image. To validate the effectiveness of our proposed method, extensive experiments have been conducted on three popular benchmarks: Pascal VOC, Cityscapes and Pascal Context. Built upon a highly competitive baseline, our proposed method can improve the performance of a student network by 2.5% (mIOU boosts from 70.2 to 72.7 on the cityscapes test set) and can train a better compact model with only 8% float operations (FLOPS) of a model that achieves comparable performances.","Title: Chord Bunch Walks for Recognizing Naturally Self-overlapped and Compound Leaves.\nYear: 2019\n\nAbstract: Effectively describing and recognizing leaf shapes under arbitrary variations, particularly from a large database, remains an unsolved problem. In this research, we attempted a new strategy of describing leaf shapes by walking and measuring along a bunch of chords that pass through the shape. A novel chord bunch walks (CBW) descriptor is developed through the chord walking behaviour that effectively integrates the shape image function over the walked chord to reflect both the contour features and the inner properties of the shape. For each contour point, the chord bunch groups multiple pairs of chords to build a hierarchical framework for a coarse-to-fine description that can effectively characterize not only the subtle differences among leaf margin patterns but also the interior part of the shape contour formed inside a self-overlapped or compound leaf. Instead of using optimal correspondence based matching, a Log-Min distance that encourages one-to-one correspondences is proposed for efficient and effective CBW matching. The proposed CBW shape analysis method is invariant to rotation, scaling, translation, and mirror transforms. Five experiments, including image retrieval of compound leaves, image retrieval of naturally self-overlapped leaves, and retrieval of mixed leaves on three large scale datasets, are conducted. The proposed method achieved large accuracy increases with low computational costs over the state-of-the-art benchmarks, which indicates the research potential along this direction.","Title: Disentangled Representation Learning and Enhancement Network for Single Image De-Raining\nYear: 2021\n\nAbstract: ABSTRACTIn this paper, we present a disentangled representation learning and enhancement network (DRLE-Net) to address the challenging single image de-raining problems, i.e., raindrop and rain streak removal. Specifically, the DRLE-Net is formulated as a multi-task learning framework, and an elegant knowledge transfer strategy is designed to train the encoder of DRLE-Net to embed a rainy image into two separated latent spaces representing the task (clean image reconstruction in this paper) relevant and irrelevant variations respectively, such that only the essential task-relevant factors will be used by the decoder of DRLE-Net to generate high-quality de-raining results. Furthermore, visual attention information is modeled and fed into the disentangled representation learning network to enhance the task-relevant factor learning. To facilitate the optimization of the hierarchical network, a new adversarial loss formulation is proposed and used together with the reconstruction loss to train the proposed DRLE-Net. Extensive experiments are carried out for removing raindrops or rainstreaks from both synthetic and real rainy images, and DRLE-Net is demonstrated to produce significantly better results than state-of-the-art models.","Title: A Novel Decision Mechanism for Image Edge Detection.\nYear: 2021\n\nAbstract: Edge detection plays an important role in image processing and computer vision tasks such as image matching and image segmentation. In this paper, we argue that edges in image should always be detected even under different image rotation transformations or noise conditions. Then a novel edge detection method is presented for improving the robustness of image edge detection. Firstly, we perform the following operations on the input image to obtain a set of images: random destruction operation, random rotation operation, and randomly addition of Gaussian noise with zero mean and random variance. Secondly, multi-directional Gabor filters with multiple scales are used as a tool to smooth the set of images and obtain candidate edges from the set of images. Thirdly, a novel edge decision mechanism is designed with a 0.95 confidence interval for selecting true edges from the candidate edges. Finally, two edge detection evaluation criteria (i.e., the aggregate test receiver-operating-characteristic and the Pratt’s figure of merit) are utilized to evaluate the proposed edge detection method against five state-of-the-art methods. The experimental results show that our proposed method outperforms all the other tested methods.","Title: Deep learning based stereo cost aggregation on a small dataset\nYear: 2021\n\nAbstract: Deep learning (DL) has been used in many computer vision tasks including stereo matching. However, DL is data hungry, and a large number of highly accurate real-world training images for stereo matching is too expensive to acquire in practice. The majority of studies rely on large simulated datasets during training, which inevitably results in domain shift problems that are commonly compensated by...","Title: Efficient corner detection based on corner enhancement filters\nYear: 2022\n\nAbstract: Multi-scale analysis based corner detection algorithms yield impressive performance, however, they are not efficient and not suitable for real-time computer vision tasks. The classical corner detection algorithms including FAST and Harris are computationally efficient, but their detection accuracy and repeatability are insufficient. This paper describes a novel fast corner detector with a simple architecture and high parallel computing characteristics. In order to simplify the corner detection architecture and improve its parallel computing performance, a new type of filter is proposed that can enhance corners and suppress edges and noise simultaneously. Then a novel efficient corner detector is proposed, which can be adapted to achieve real-time detection in hardware. Experimental results show that, with a very low computational cost and simple architecture, the proposed detector can achieve or even exceed the detection accuracy of multi-scale analysis based detectors. Its repeatability is similar to multi-scale analysis based detectors and clearly higher than other types of corner detectors. Therefore, it is potentially useful as an efficient corner detector for computer vision applications especially for portable real-time tasks. (c) 2022 Elsevier Inc. All rights reserved.","Title: Improving Monocular Visual Odometry Using Learned Depth\nYear: 2022\n\nAbstract: Monocular visual odometry (VO) is an important task in robotics and computer vision. Thus far, how to build accurate and robust monocular VO systems that can work well in diverse scenarios remains largely unsolved. In this article, we propose a framework to exploit monocular depth estimation for improving VO. The core of our framework is a monocular depth estimation module with a strong generalization capability for diverse scenes. It consists of two separate working modes to assist the localization and mapping. With a single monocular image input, the depth estimation module predicts a relative depth to help the localization module on improving the accuracy. With a sparse depth map and an RGB image input, the depth estimation module can generate accurate scale-consistent depth for dense mapping. Compared with current learning-based VO methods, our method demonstrates a stronger generalization ability to diverse scenes. More significantly, our framework is able to boost the performances of existing geometry-based VO methods by a large margin.","Title: Recent advances on image edge detection: A comprehensive review\nYear: 2022\n\nAbstract: •The recent edge detection methods are classified and introduced in detail.•The existing widely used datasets and evaluation criteria for edge detection methods are reviewed.•Future research directions for edge detection are elaborated.","Title: EOCSA: Predicting prognosis of Epithelial ovarian cancer with whole slide histopathological images\nYear: 2022\n\nAbstract: Ovarian cancer is one of the most serious cancers that threaten women around the world. Epithelial ovarian cancer (EOC), as the most commonly seen subtype of ovarian cancer, has rather high mortality rate and poor prognosis among various gynecological cancers. Survival analysis outcome is able to provide treatment advices to doctors. In recent years, with the development of medical imaging technology, survival prediction approaches based on pathological images have been proposed. In this study, we designed a deep framework named EOCSA which analyzes the prognosis of EOC patients based on pathological whole slide images (WSIs). Specifically, we first randomly extracted patches from WSIs and grouped them into multiple clusters. Next, we developed a survival prediction model, named DeepConvAttentionSurv (DCAS), which was able to extract patch-level features, removed less discriminative clusters and predicted the EOC survival precisely. Particularly, channel attention, spatial attention, and neuron attention mechanisms were used to improve the performance of feature extraction. Then patient-level features were generated from our weight calculation method and the survival time was finally estimated using LASSO-Cox model. The proposed EOCSA is efficient and effective in predicting prognosis of EOC and the DCAS ensures more informative and discriminative features can be extracted. As far as we know, our work is the first to analyze the survival of EOC based on WSIs and deep neural network technologies. The experimental results demonstrate that our proposed framework has achieved state-of-the-art performance of 0.980 C-index. The implementation of the approach can be found at https://github.com/RanSuLab/EOCprognosis.","Title: Semi-supervised Histological Image Segmentation via Hierarchical Consistency Enforcement\nYear: 2022\n\nAbstract: Acquiring pixel-level annotations for histological image segmentation is time- and labor- consuming. Semi-supervised learning enables learning from the unlabeled and limited amount of labeled data. A challenging issue is the inconsistent and uncertain predictions on unlabeled data. To enforce invariant predictions over the perturbations applied to the hidden feature space, we propose a Mean-Teacher based hierarchical consistency enforcement (HCE) framework and a novel hierarchical consistency loss (HC-loss) with learnable and selfguided mechanisms. Specifically, the HCE takes the perturbed versions of the hierarchical features from the encoder as input to the auxiliary decoders, and encourages the predictions of the auxiliary decoders and the main decoder to be consistent. The HC-loss facilitates the teacher model to generate reliable guidance and enhances the consistency among all the decoders of the student model. The proposed method is simple, yet effective, which can easily be extended to other frameworks. The quantitative and qualitative experimental results indicate the effectiveness of the hierarchical consistency enforcement on the MoNuSeg and CRAG datasets."],"string":"[\n \"Title: Embedded Voxel Colouring\\nYear: 2003\\n\\nAbstract: The reconstruction of a complex scene from multiple im- ages is a fundamental problem in the field of computer vision. Volu- metric methods have proven to be a strong alternative to traditional correspondence-based methods due to their flexible visibility models. In this paper we analyse existing methods for volumetric reconstruction and identify three key properties of voxel colouring algorithms: a water-tight surface model, a monotonic carving order, and causality .W e present a new Voxel Colouring algorithm which embeds all reconstructions of a scene into a single output. While modelling exact visibility for arbitrary camera locations, Embedded Voxel Colouring removes the need for a priori threshold selection present in previous work. An efficient imple- mentation is given along with results demonstrating the advantages of posteriori threshold selection.\",\n \"Title: Stereoscopic flatbed scanner\\nYear: 2009\\n\\nAbstract: 3-D scanning has become increasingly popular in a wide range of applications. We present a prototype 3-D/stereoscopic scanning system based on a cheap, readily available flatbed scanner. Stereoscopic imaging is achieved by modifying the optical path of the ordinary flatbed scanner. The results of 3-D imaging using our prototype system are demonstrated, and a number of design alternatives are discussed. (C) 2009 SPIE and IS&T. [DOI: 10.1117/1.3059582]\",\n \"Title: Circular shortest paths by branch and bound\\nYear: 2003\\n\\nAbstract: Shortest path algorithms are used for a large variety of optimisation problems in network and transportation analysis. They are also used in image analysis for object segmentation, disparity estimation, path finding and crack detection. Sometimes the topology of the problem demands that the path be circular. Such circular path constraints occur in polar object segmentation, disparity estimation for panoramic stereo images and in shortest paths around a cylinder. In this paper we present a new efficient algorithm for circular shortest path determination on a u-by-v trellis in O(u1.6v) average time. We impose a binary search tree on the set of path endpoints and use a best-first Branch and Bound search technique to efficiently obtain the global minimum circular path. The typical running time of our circular shortest path algorithm on a 256×256 image is in the order of 0.1s on a 1GHz Dell P3 workstation under the Linux operating system. Applications to crack detection and object segmentation are presented.\",\n \"Title: Mosaicing of Microscope Images\\nYear: 2005\\n\\nAbstract: Image mosaicing has found a number of applications such as panoramic imaging, digital terrain mapping, ophthalmology, and virtual microscopy. In this paper, we present an automated mosaicing technique for generating mosaics from microscope images. We carry out robust image feature matching and global geometric and radiometric parameter estimation. The images are transformed using the estimated parameters and mosaiced together, producing accurate registration of overlapping regions without visible seams.\",\n \"Title: Priority-Based Path Growing for Linear Feature Detection\\nYear: 2007\\n\\nAbstract: Linear feature detection is a very important step in image analysis, computer vision, and pattern recognition. It has a wide range of applications. In this paper, we present a new algorithm for linear feature detection using priority- based path growing from automatically selected seeds. The use of path growing in the neighbourhood of automatically selected seeds ensures that all local information along a linear feature is taken into account. This enables tracing of weakly contrasted linear features in an image. We illustrate our method with examples.\",\n \"Title: Automating the quantification of membrane proteins under confocal microscopy.\\nYear: 2012\\n\\nAbstract: In a prior contribution, we described a semi-automated system that allowed a user to quantify the relative abundance of fluorescently labeled membrane proteins in confocal microscopy images. Here, we describe a step change in assay automation, enabled by explicitly casting the problem in terms of mathematical graphs. This permitted to include extensive and relevant image information in the tracing process; something not possible when using the marginally faster traditional algorithms. These improvements bring us closer to an industrial strength membrane tracing assay suitable in a drug discovery and development environment.\",\n \"Title: Membrane boundary extraction using circular multiple paths\\nYear: 2009\\n\\nAbstract: Membrane proteins represent over 50% of known drug targets. Accordingly, several widely used assays in the high content analysis area rely on quantitative measures of the translocation of proteins between intracellular organelles and the cell surface. In order to increase the sensitivity of these assays, one needs to measure the signal specifically along the membrane, requiring a precise segmentation of this compartment. Manual tracing of membrane boundary is very time-consuming and confronts us with issues of objectivity and reproducibility. In this paper, we present an approach based on a circular multiple paths technique on transformed images that enables us to segment the membrane compartment accurately and rapidly. We have presented three approaches for image transformation. The circular property of the multiple paths ensures that we are obtaining closed contours for the membrane boundary. The position of the multiple paths provides the edges of the membrane boundary. The effectiveness of our algorithm is illustrated using cells expressing epitope-tagged membrane proteins.\",\n \"Title: Moving average algorithms for diamond, hexagon, and general polygonal shaped window operations\\nYear: 2006\\n\\nAbstract: This paper presents fast moving window algorithms for calculating local statistics in a diamond, hexagon, and general polygonal shaped windows of an image which is important for real-time applications. The algorithms for a diamond shaped window requires only seven or eight additions and subtractions per pixel. A fast sparse algorithm only needs four additions and subtractions for a sparse diamond shaped window. A number of other shapes of diamond windows such as skewed or parallelogram shaped diamond, long diamond, and lozenged diamond shaped, are also investigated. Similar algorithms are also developed for hexagon shaped windows. The computation for a hexagon window only needs eight additions and subtractions for each pixel. Fast algorithms for general polygonal shaped windows are also developed. The computation cost of all these algorithms is independent of the window size. A variety of synthetic and real images have been tested.\",\n \"Title: Cross image inference scheme for stereo matching\\nYear: 2012\\n\\nAbstract: In this paper, we propose a new interconnected Markov Random Field (MRF) or iMRF model for the stereo matching problem. Comparing with the standard MRF, our model takes into account the consistency between the label of a pixel in one image and the labels of its possible matching points in the other image. Inspired by the turbo decoding scheme, we formulate this consistency by a cross image reference term which is iteratively updated in our matching framework. The proposed iMRF model represents the matching problem better than the standard MRF and gives better results even without using any other information from segmentation prior or occlusion detection. We incorporate segmentation information and the coarse-to-fine scheme into our model to further improve the matching performance.\",\n \"Title: Clustered Nuclei Splitting Using Curvature Information\\nYear: 2011\\n\\nAbstract: Automated splitting of clustered nuclei from images of tissue sections is essential to many biomedical studies. Many existing image segmentation methods tend to produce over-segmented or under-segmented results for clustered nuclei images. In this paper, a new curvature information based image segmentation algorithm is proposed. Through combining curvature information with a distance map, our algorithm can extract correct markers corresponding to each nucleus. Afterwards, marker based watershed segmentation is used to segment the clustered nuclei. The algorithm is tested on both synthetic and real images. Experimental results show that our algorithm is accurate and robust to noise in segmentation of clustered nuclei.\",\n \"Title: Fast optical flow using 3D shortest path techniques\\nYear: 2002\\n\\nAbstract: Optical flow or image motion estimation is important in the area of computer vision. This paper presents a fast and reliable optical flow algorithm which produces a dense optical flow map by using fast cross correlation and 3D shortest path techniques. Fast correlation is achieved by using the box-filtering technique which is invariant to the size of the correlation window. The motion for each scanline or each column of the input image is obtained from the correlation coefficient volume by finding the best 3D path using dynamic programming techniques rather than simply choosing the position that gives the maximum cross correlation coefficient. Sub-pixel accuracy is achieved by fitting the local correlation coefficients to a quadratic surface. Typical running time for a 256×256 image is in the order of a few seconds on a 85MHz computer. A variety of synthetic and real images have been tested, and good results have been obtained.\",\n \"Title: 3D Symmetry Detection Using The Extended Gaussian Image\\nYear: 1997\\n\\nAbstract: Symmetry detection is important in the area of computer vision. A 3D symmetry detection algorithm is presented in this correspondence. The symmetry detection problem is converted to the correlation of the Gaussian image. Once the Gaussian image of the object has been obtained, the algorithm is independent of the input format. The algorithm can handle different kinds of images or objects. Simulated and real images have been tested in a variety of formats, and the results show that the symmetry can be determined using the Gaussian image.\",\n \"Title: Iterative infrared ship target segmentation based on multiple features.\\nYear: 2014\\n\\nAbstract: This paper presents an efficient method for ship target segmentation in infrared (IR) images. It consists of mainly two procedures: iterative image segmentation and ship target selection. First, based on the intensity distribution of an IR image, we design a global background subtraction filter (GBSF) to suppress the background, and an adaptive row mean subtraction filter (ARMSF) to enhance the target. After iteratively applying these two filters, we can obtain a proper threshold for image segmentation. Second, based on the geometric properties of the ship target, we construct four shape features and a selection criterion to identify the real target and remove the non-target regions. Experimental results demonstrate that the proposed method can effectively segment ship targets from different backgrounds in IR images. The advantage of the proposed method over the others in the previous literatures is validated in both visual and quantitative comparisons, especially for IR images with low contrast and uneven intensities.\",\n \"Title: Automatic segmentation of fiducial marks using attribute-based mathematical morphology\\nYear: 2001\\n\\nAbstract: This paper described a fast and automatic method for segmenting fiducial marks in an image taken with a metric camera. These marks could be at the four corners, four middle sides, or inside an image (reseau marks). The segmentation was realized by using attribute-based mathematical morphology techniques. The attributes that we used in the morphology processing step were the area, aspect ratio, and orientation of the best fitted ellipse of an object. The algorithm took about 0.25 s in the automatic segmentation stage. The positions of these automatically segmented fiducial marks were then refined to subpixel accuracy. Dozens of real images with different shapes of fiducial marks were tested and reliable results were obtained. Accuracy comparisons were made using our method and previously published methods. (C) 2001 SPIE and IS&T.\",\n \"Title: Segmentation of clustered nuclei based on curvature weighting\\nYear: 2012\\n\\nAbstract: Cluster of nuclei are frequently observed in thick tissue section images. It is very important to segment overlapping nuclei in many biomedical applications. Many existing methods tend to produce under segmented results when there is a high overlap rate. In this paper, we present a curvature weighting based algorithm which weights each pixel using the curvature information of its nearby boundaries to extract markers, each of which represents an object, from input images. Then we use marker-controlled watershed to obtain the final segmentation. Test results using both synthetic and real cell images are presented in the paper.\",\n \"Title: Tree structural watershed for stereo matching\\nYear: 2012\\n\\nAbstract: We present a new method for dense stereo matching based on a tree structural cost volume watershed (TSCVW) and a region combination (RC) process. Given a cost volume as the data cost and an initial segmentation result, the proposed TSCVW method reliably estimates the disparities in a segment by using energy optimization to control plane segmentation and plane fitting. Then the disparities in the incorrectly fitted and occluded regions are refined using our RC process. Experimental results show that our method is very robust to different initial segmentation results and the shape of a segment. The comparison between our algorithm and the current state-of-the-art algorithms on the Middlebury website shows that our algorithm is very competitive.\",\n \"Title: Multiple paths extraction in images using a constrained expanded trellis.\\nYear: 2005\\n\\nAbstract: Single shortest path extraction algorithms have been used in a number of areas such as network flow and image analysis. In image analysis, shortest path techniques can be used for object boundary detection, crack detection, or stereo disparity estimation. Sometimes one needs to find multiple paths as opposed to a single path in a network or an image where the paths must satisfy certain constraints. In this paper, we propose a new algorithm to extract multiple paths simultaneously within an image using a constrained expanded trellis (CET) for feature extraction and object segmentation. We also give a number of application examples for our multiple paths extraction algorithm.\",\n \"Title: Multi-Image 2D-PAGE Feature Detection\\nYear: 2005\\n\\nAbstract: This paper proposes an algorithm for decomposing a multi-class classification problem into a set of two-class classification problems. The algorithm divides a set of input pattern vectors in each class into subsets according to the distribution of the ...\",\n \"Title: Fast Stereo Matching by Iterated Dynamic Programming and Quadtree Subregioning\\nYear: 2004\\n\\nAbstract: The application of energy minimisation methods for stereo matching has been demonstrated to produce high quality disparity maps. However the ma- jority of these methods are known to be computationally expensive, requir- ing minutes or even hours of computation. We propose a fast minimisation scheme that produces strongly competitive results for significantly reduced computation, requiring only a few seconds of computation. In this paper, we present our iterated dynamic programming algorithm along with a quadtree subregioning process for fast stereo matching.\",\n \"Title: De-interlacing of video images using a shortest path technique\\nYear: 2001\\n\\nAbstract: This paper presents a fast algorithm for de-interlacing of video images using a shortest path technique. The algorithm applies dynamic programming techniques to find a shortest path in a cost matrix. The motion information obtained from this shortest path is used to re-align the fields of a video image. By using the shortest path via dynamic programming, the motion information estimated is more reliable than simply performing a search in a local neighbourhood. A variety of real images have been tested, and good results have been obtained\",\n \"Title: Detection of Nuclear Buds Based on Ellipse Fitting\\nYear: 2010\\n\\nAbstract: Assays of micronucleus are extensively used in genotoxicity testing and in monitoring of human exposure to genotoxic materials. As nuclear buds could be a new source of micronuclei formed in interphase, the assay of contents of nuclear bud in normal and comparison group is needed. In this paper, we proposed a new automatic nuclear buds detection algorithm based on ellipse fitting. Experimental results show that our algorithm is effective and efficient. We believe that this is the first report on automated nuclear bud detection for DNA damage scoring.\",\n \"Title: Touching Cells Splitting by Using Concave Points and Ellipse Fitting\\nYear: 2008\\n\\nAbstract: A novel touching cells splitting algorithm by using concave points and ellipse fitting is proposed to split circle-like or ellipse-like touching cells in this paper. The algorithm is divided into two parts. The first part is contour pre-processing, whose purpose is to find the concave points of the contour and separate the contour into different segments using the concave points. The second part is ellipse processing, whose purpose is to process the different segments of the contour as a possible single cell by using the properties of the fitted ellipse. The second part is the main contribution of this paper which demonstrates a new way to use ellipse fitting to split the binary contour of the touching cells. Experimental results show that the proposed algorithm is efficient.\",\n \"Title: Closed-form stereo image rectification\\nYear: 2012\\n\\nAbstract: Stereo image rectification transforms a pair of images into a new pair such that the epipolar lines in the transformed images have the same direction as the image rows and matching epipolar lines in different images have the same row index to enable an efficient and reliable dense stereo matching. In this paper we propose a first closed-form algorithm to automatically rectify stereo images just using the fundamental matrix. The algorithm involves only direct and purely geometric transformation processes. There is no iterative or optimization processes for the rectification parameter estimation in our method. Real images have been used for testing purposes, and convincing results have been obtained from our algorithm. Comparison with the state-of-the-art method shows that our method produces better rectification results.\",\n \"Title: Fast algorithm for local statistics calculation for N-dimensional images\\nYear: 2001\\n\\nAbstract: This paper presents a fast algorithm for calculating these local statistics in awindow of an N-dimensional image. The new algorithm, which is called the plunger method,recursively reduces the dimensions of the input N-dimensional image to achieve fast computation.The speed of the algorithm is independent of the window size. Another advantage of thealgorithm is that it calculates the local statistics in one pass. Real image tests have beenperformed\",\n \"Title: Circular shortest path in images\\nYear: 2003\\n\\nAbstract: Shortest path algorithms have been used in a number of applications such as crack detection, road or linear feature extraction in images. There are applications where the starting and ending positions of the shortest path need to be constrained. In this paper, we present several new algorithms for the extraction of a circular shortest path in an image such that the starting and ending positions coincide. The new algorithms we developed include multiple search algorithm, image patching algorithm, multiple backtracking algorithm, the combination of image patching and multiple back-tracking algorithm, and approximate algorithm. The typical running time of our circular shortest path extraction algorithm on a 256×256 image is about 0.3s on a rather slow 85MHz Sun SPARC computer. A variety of real images for crack detection in borehole data and object boundary extraction have been tested and good results have been obtained.\",\n \"Title: Linear Feature Detection on GPUs\\nYear: 2010\\n\\nAbstract: The acceleration of an existing linear feature detection algorithm for 2D images using GPUs is discussed. The two most time consuming components of this process are implemented on the GPU, namely, linear feature detection using dual-peak directional non-maximum suppression, and a gap filling process that joins disconnected feature masks to rectify false negatives. Multiple steps or image filters in each component are combined into a single GPU kernel to minimise data transfers to off-chip GPU RAM, and issues relating to on-chip memory utilisation, caching, and memory coalescing are considered. The presented algorithm is useful for applications needing to analyse complex linear structures, and examples are given for dense neurite images from the biotech domain.\",\n \"Title: Thickness measurement and crease detection of wheat grains using stereo vision\\nYear: 2007\\n\\nAbstract: Wheat grain quality assessment is important in meeting market requirements. The thickness of grains can be used for the measurement of the mass proportion of grains that pass through a sieve. This measure is known as ''screenings''. The determination of the presence or absence of the grain crease aids the detection of a stain called blackpoint, which is usually most evident on the non-crease side of the grain. In this paper we investigate the use of stereo vision techniques for measuring the thickness and detecting the presence or absence of the crease of a sample of wheat grains placed on a tray with dimples.\",\n \"Title: Diamond, Hexagon, and General Polygonal Shaped Window Smoothing\\nYear: 2003\\n\\nAbstract: This paper presents fast recursive or moving windows algo- rithms for calculating local means in a diamond, hexagon and general polygonal shaped windows of an image. The algorithms for diamond shaped window require only seven or eight additions and subtractions per pixel. A number of other shapes of diamond windows such as skewed or parallelogram shaped diamond, long diamond, and lozenged diamond shaped, are also investigated. Similar algorithms are also developed for hexagon shaped windows. The computation for hexagon window only needs eight additions and subtractions for each pixel. Fast algorithms for general polygonal shaped windows are also developed. The computa- tion costs of all these algorithms are independent of the window size. A variety of synthetic and real images have been tested.\",\n \"Title: Automated Feature Weighting in Fuzzy Declustering-based Vector Quantization\\nYear: 2010\\n\\nAbstract: Feature weighting plays an important role in improving the performance of clustering technique. We propose an automated feature weighting in fuzzy declustering-based vector quantization (FDVQ), namely AFDVQ algorithm, for enhancing effectiveness and efficiency in classification. The proposed AFDVQ imposes weights on the modified fuzzy c-means (FCM) so that it can automatically calculate feature weights based on their degrees of importance rather than treating them equally. Moreover, the extension of FDVQ and AFDVQ algorithms based on generalized improved fuzzy partitions (GIFP), known as GIFP-FDVQ and GIFP-AFDVQ respectively, are proposed. The experimental results on real data (original and noisy data) and modified data (biased and noisy-biased data) have demonstrated that the proposed algorithms outperformed standard algorithms in classifying clusters especially for biased data.\",\n \"Title: 2-Point linear algorithm for camera translation vector estimation with known rotation\\nYear: 2000\\n\\nAbstract: This paper introduces a simple method for obtaining the pure 3-D translational motion parameter (without rotation) and the translation parameter with a known rotation of a camera. Once the translation vector is available, the 3-D structure of an object can also be obtained. The minimum number of matching points in the two images can be as few as two, and the equations to be solved are linear. Synthetic and real image tests has been performed, and good results have been obtained.\",\n \"Title: Applications of heterogeneous computing in computational and simulation science\\nYear: 2013\\n\\nAbstract: This paper presents a summary and overview of heterogeneous algorithms and applications developed by the Commonwealth Scientific and Industrial Research Organisation CSIRO for solving practical and challenging science problems such as: 1 high-content analysis of brain cell images for medical research and drug discovery; 2 genetic analysis of complex experimental designs for crop breeding; 3 solving computational fluid dynamics problems; 4 deconvolving 3D images from microscopy and medical imaging; 5 reconstructing large 3D computed tomography CT images from medical and materials science; 6 quantifying uncertainty in complex environmental models. Applications discussed utilise GPUs and multicore CPUs on a scale ranging from single desktop workstations through to large GPU clusters. Results demonstrate that both types of system can successfully accelerate a variety of practical science applications. We have seen significant gains in productivity and opportunity as a result of leveraging GPUs, tackling computational problems in which execution time was previously infeasible.\",\n \"Title: Automated Detection of Nucleoplasmic Bridges for DNA Damage Scoring in Binucleated Cells\\nYear: 2010\\n\\nAbstract: Quantification of DNA damage, which may be caused by radiation or exposure to chemicals, is very important and can be very time consuming and subject to variability if carried out visually. The quantification of scoring DNA damage includes biomarkers such as micronuclei, nucleoplasmic bridges, and nuclear buds as scored in cytokinesis-blocked binucleated cells. In this paper, we present a new algorithm based on a shortest path technique that enables us to detect the nucleoplasmic bridges joining two nuclei in cell images of binucleated cells. The effectiveness of our algorithm is illustrated using a set of cell images. We believe that this is the first time that a feasible automated nucleoplasmic bridge detection system has been reported.\",\n \"Title: Embedded Voxel Colouring with Adaptive Threshold Selection Using Globally Minimal Surfaces\\nYear: 2012\\n\\nAbstract: Image-based 3D reconstruction remains a competitive field of research as state-of-the-art algorithms continue to improve. This paper presents a voxel-based algorithm that adapts the earliest space-carving methods and utilises a minimal surface technique to obtain a cleaner result. Embedded Voxel Colouring is built in two stages: (a) progressive voxel carving is used to build a volume of embedded surfaces and (b) the volume is processed to obtain a surface that maximises photo-consistency data in the volume. This algorithm combines the strengths of classical carving techniques with those of minimal surface approaches. We require only a single pass through the voxel volume, this significantly reduces computation time and is the key to the speed of our approach. We also specify three requirements for volumetric reconstruction: monotonic carving order, causality of carving and water-tightness. Experimental results are presented that demonstrate the strengths of this approach.\",\n \"Title: Uncalibrated three-view image rectification\\nYear: 2003\\n\\nAbstract: Image rectification is a process of transforming a set of images into a new set such that the epipolar lines in the transformed images have the same direction as the image rows or columns to enable an efficient and reliable stereo matching. Previous algorithms for stereo image rectification either work for two view uncalibrated or two/three view calibrated situations. In this paper we propose several novel techniques to rectify uncalibrated trinocular images using the trilinear tensor or projective invariants or fundamental matrices obtained from a triplet of images. Our new methods include: a rotation and skew method, an affine transformation method, and a vanishing points method. Real images have been used for testing purposes, and accurate results have been obtained.\",\n \"Title: Splitting touching cells based on concave points and ellipse fitting\\nYear: 2009\\n\\nAbstract: A new touching cells splitting algorithm based on concave points and ellipse fitting is proposed in this paper. The algorithm includes two parts: contour pre-processing and ellipse processing. The purpose of contour pre-processing is to smooth fluctuations of the contour, find concave points of the contour and divide the contour into different segments via the concave points. The purpose of ellipse processing is to process the different segments of the contour into possible single cells by using the properties of the fitted ellipses. Because concave points divide the whole contour of touching cells into different segments and different segments of one single cell have similar properties, the ellipse processing can separate the touching cells through ellipse fitting. This paper demonstrates a new way of using ellipse fitting to split the binary contour of touching cells. Experimental results show that our algorithm is efficient.\",\n \"Title: An Energy Minimisation Approach to Stereo-Temporal Dense Reconstruction\\nYear: 2004\\n\\nAbstract: We propose a novel energy minimisation framework for the dense reconstruction of stereo image sequences that incorporates data delity as well as spatial and temporal regularity. An iterated dynamic programming scheme is proposed to minimise the energy function. We also present an efficient implementation of the minimisation scheme by introducing morphological decomposition techniques to solve the dynamic programming subproblem. Our proposed method is capable of reconstructing dynamic scenes with complex motion. Results are presented demonstrating the strength of our proposed algorithm.\",\n \"Title: Symmetry detection using gradient information\\nYear: 1995\\n\\nAbstract: Symmetry detection is important in the area of computer vision. A simple and fast symmetry detection algorithm has been developed in this paper. The algorithm employs only the original image and the gradient information. The direction of the symmetry axis is obtained from the gradient orientation histogram; and the position of this symmetry axis is decided either by the center of gravity or by the profile of the image projection along the direction of the symmetry axis. This method works directly on the grey-scale image and does not require any prior segmentation of the input image. Both simulated and real images have been tested and the results are very convincing.\",\n \"Title: Boundary extraction of linear features using dual paths through gradient profiles\\nYear: 2008\\n\\nAbstract: An algorithm for automated extraction of linear feature boundaries in 2D images is presented. From a marker set approximating the medial axis, we generate 1D gradient profiles orthogonal to linear features. The algorithm uses dual shortest paths through an image generated from gradient profiles to extract boundaries. The algorithm offers an alternative to a watershed type approach and performs well on images with noise and areas of low contrast. We present the results of our algorithm on microscopy images of neurite outgrowth and other images containing linear features.\",\n \"Title: Iterated dynamic programming and quadtree subregioning for fast stereo matching\\nYear: 2008\\n\\nAbstract: The application of energy minimisation methods for stereo matching has been demonstrated to produce high quality disparity maps. However, the majority of these methods are known to be computationally expensive requiring minutes of computation. In this paper, we propose a fast minimisation scheme that produces high quality stereo reconstructions for significantly reduced running time, requiring only a few seconds of computation. The minimisation scheme is carried out using our iterated dynamic programming algorithm, which iterates over entire rows and columns for fast stereo matching. A quadtree subregioning process is also used for efficient computation of a matching cost volume where iterated dynamic programming operates on.\",\n \"Title: Fast linear feature detection using multiple directional non-maximum suppression.\\nYear: 2006\\n\\nAbstract: The capacity to detect linear features is central to image analysis, computer vision and pattern recognition and has practical applications in areas such as neurite outgrowth detection, retinal vessel extraction, skin hair removal, plant root analysis and road detection. Linear feature detection often represents the starting point for image segmentation and image interpretation. In this paper, we present a new algorithm for linear feature detection using multiple directional non-maximum suppression with symmetry checking and gap linking. Given its low computational complexity, the algorithm is very fast. We show in several examples that it performs very well in terms of both sensitivity and continuity of detected linear features.\",\n \"Title: Fast panoramic stereo matching using cylindrical maximum surfaces.\\nYear: 2004\\n\\nAbstract: This paper presents a fast panoramic stereo matching algorithm using a cylindrical maximum surface technique. The disparity for a pair of panoramic images is found in a cylindrical shaped correlation coefficient volume by obtaining the maximum surface rather than simply choosing a position that gives the maximum correlation coefficient value. The use of our cylindrical maximum surface technique ensures that the disparities obtained at the left and the right columns of the panoramic stereo images are properly constrained. Typical running time for a pair of 1324 x 120 images is about 0.33 s on a 1.7-GHz PC. A variety of real images have been tested, and good results have been obtained.\",\n \"Title: Fast Stereo Matching Using Rectangular Subregioning and 3D Maximum-Surface Techniques\\nYear: 2002\\n\\nAbstract: This paper presents a fast and reliable stereo matching algorithm which produces a dense disparity map by using fast cross correlation, rectangular subregioning (RSR) and 3D maximum-surface techniques in a coarse-to-fine scheme. Fast correlation is achieved by using the box-filtering technique whose speed is invariant to the size of the correlation window and by segmenting the stereo images into rectangular subimages at different levels of the pyramid. By working with rectangular subimages, not only can the speed of the correlation be further increased, the intermediate memory storage requirement can also be reduced. The disparity map for the stereo images is found in the 3D correlation coefficient volume by obtaining the global 3D maximum-surface rather than simply choosing the position that gives the local maximum correlation coefficient value for each pixel. The 3D maximum-surface is obtained using our new two-stage dynamic programming (TSDP) technique. There are two original contributions in this paper: (1) development of the RSR technique for fast similarity measure; and (2) development of the TSDP technique for efficiently obtaining 3D maximum-surface in a 3D volume. Typical running time of our algorithm implemented in the C language on a 512 × 512 image is in the order of a few seconds on a 500 MHz PC. A variety of synthetic and real images have been tested, and good results have been obtained.\",\n \"Title: Skew and slant correction for document images using gradient direction\\nYear: 1997\\n\\nAbstract: A fast algorithm is presented for skew and slant correction in printed document images. The algorithm employs only the gradient information. The skew angle is obtained by searching for a peak in the histogram of the gradient orientation of the input grey-level image. The skewness of the document is corrected by a rotation at such an angle. The slant of characters can also be detected using the same technique, and can be corrected by a shear operation. A second method for character slant correction by fitting parallelograms to the connected components is also described. Document images with different contents (tables, figures, and photos) have been tested for skew correction and the algorithm gives accurate results on all the test images, and the algorithm is very easy to implement\",\n \"Title: Linear feature enhancement based on morphological operation and Gabor function\\nYear: 2012\\n\\nAbstract: In many image processing tasks, feature enhancement is an important preprocessing step. It is the process of adjusting images so that the results are more suitable for analysis. Good enhancement results of linear features such as vessels or neurites can be inputs for segmentation or tracking. In this paper, we propose a novel linear feature enhancement method based on morphological operation and Gabor function, which can enhance and smooth linear features. We use and improve the Hessian matrix to calculate the orientation information for our directional morphological operation and Gabor smoothing. An approach for junction enhancement in each branch is also proposed here. We present the results of our algorithm on images of different types. The obtained outputs show that the proposed approach can achieve very good results.\",\n \"Title: Trinocular Stereo Image Rectification In Closed-Form Only Using Fundamental Matrices\\nYear: 2013\\n\\nAbstract: Trinocular stereo image rectification is a process to transform a set of three images into a new set so that the epipolar lines in the transformed images have the same direction as the image row or column and matching epipolar lines in different images have the same row or column indices to enable an efficient and reliable dense stereo matching. In this paper we propose a new closed-form method to rectify three-view stereo images just using fundamental matrices. The algorithm involves only direct and purely geometric transformation processes. No iteration or optimization process is involved in our method. Real images have been used for testing purposes, and accurate results have been obtained using our algorithm.\",\n \"Title: Stereo matching using cost volume watershed and region merging.\\nYear: 2014\\n\\nAbstract: Segment based disparity estimation methods have been proposed in many different ways. Most of these studies are built upon the hypothesis that no large disparity jump exists within a segment. When this hypothesis does not hold, it is difficult for these methods to estimate disparities correctly. Therefore, these methods work well only when the images are initially over segmented but do not work well for under segmented cases. To solve this problem, we present a new segment based stereo matching method which consists of two algorithms: a cost volume watershed algorithm (CVW) and a region merging (RM) algorithm. For incorrectly under segmented regions where pixels on different objects are grouped into one segment, the CVW algorithm regroups the pixels on different objects into different segments and provides disparity estimation to the pixels in different segments accordingly. For unreliable and occluded regions, we merge them into neighboring reliable segments for robust disparity estimation. The comparison between our method and the current state-of-the-art methods shows that our method is very competitive and is robust particularly when the images are initially under segmented.\",\n \"Title: A novel method for dendritic spines detection based on directional morphological filter and shortest path.\\nYear: 2014\\n\\nAbstract: Dendritic spines are tiny membranous protrusions from neuron's dendrites. They play a very important role in the nervous system. A number of mental diseases such as Alzheimer's disease and mental retardation are revealed to have close relations with spine morphologies or spine number changes. Spines have various shapes, and spine images are often not of good quality; hence it is very challenging to detect spines in neuron images. This paper presents a novel pipeline to detect dendritic spines in 2D maximum intensity projection (MIP) images and a new dendrite backbone extraction method is developed in the pipeline. The strategy for the backbone extraction approach is that it iteratively refines the extraction result based on directional morphological filtering and improved Hessian filtering until a satisfactory extraction result is obtained. A shortest path method is applied along a backbone to extract the boundary of the dendrites. Spines are then segmented from the dendrites outside the extracted boundary. Touching spines will be split using a marker-controlled watershed algorithm. We present the results of our algorithm on real images and compare our algorithm with two other spine detection methods. The results show that the proposed approach can detect dendrites and spines more accurately. Measurements and classification of spines are also made in this paper.\",\n \"Title: A new method for linear feature and junction enhancement in 2D images based on morphological operation, oriented anisotropic Gaussian function and Hessian information.\\nYear: 2014\\n\\nAbstract: Feature enhancement is an important preprocessing step in many image processing tasks. It is the process of adjusting image intensities so that the enhanced results are more suitable for analysis. Good enhancement results for linear structures such as vessels or neurites can be used as inputs for segmentation and other operations. In this paper, a novel linear feature enhancement filter – an adaptive multi-scale morpho-Gaussian filter – which can enhance and smooth linear features is proposed based on morphological operation, anisotropic Gaussian function and Hessian information. This filter can enhance and smooth along the local orientation of the linear structures and the Hessian measurement is used to further enhance the linear features. We utilize the Hessian matrix to calculate the orientation information for our directional morphological operation and the oriented anisotropic Gaussian smoothing. We also propose a novel method for junction enhancement, which can solve the problem of junction suppression. We decompose the junctions and enhance along each linear structure within a junction region. We present the test results of our algorithm on images of different types and compare our method with three existing methods. The experimental results show that the proposed approach can achieve better results.\",\n \"Title: Multipoint Filtering with Local Polynomial Approximation and Range Guidance\\nYear: 2014\\n\\nAbstract: This paper presents a novel guided image filtering method using multipoint local polynomial approximation (LPA) with range guidance. In our method, the LPA is extended from a pointwise model into a multipoint model for reliable filtering and better preserving image spatial variation which usually contains the essential information in the input image. In addition, we develop a scheme with constant computational complexity (invariant to the size of filtering kernel) for generating a spatial adaptive support region around a point. By using the hybrid of the local polynomial model and color/intensity based range guidance, the proposed method not only preserves edges but also does a much better job in preserving spatial variation than existing popular filtering methods. Our method proves to be effective in a number of applications: depth image upsampling, joint image denoising, details enhancement, and image abstraction. Experimental results show that our method produces better results than state-of-the-art methods and it is also computationally efficient.\",\n \"Title: Infrared Ship Target Image Smoothing Based on Adaptive Mean Shift\\nYear: 2014\\n\\nAbstract: Infrared (IR) image denoising is important for IR image analysis. In this paper, we propose a method based on adaptive range bandwidth mean shift for IR ship target image smoothing, aiming to effectively suppress noise as well as preserve important target structures. First, local image properties, including the mean value and standard deviation, are combined to build a salient region map, and a thresholding method is applied to obtain a binary mask on the target region. Then, we develop an adaptive range bandwidth mean shift method for image denoising. By associating the range bandwidth of the mean shift with local region saliency, we can adjust the bandwidth adaptively, thus to smooth the background region while preserving important target structures. Experimental results show that this method works well for IR ship target images with different backgrounds. It demonstrates superior performance for image denoising and target preserving, compared with some existing image denoising methods.\",\n \"Title: Soft Cost Aggregation With Multi-Resolution Fusion\\nYear: 2014\\n\\nAbstract: This paper presents a simple and effective cost volume aggregation framework for addressing pixels labeling problem. Our idea is based on the observation that incorrect labelings are greatly reduced in cost volume aggregation results from low resolutions. However, image details may be lost in the low resolution results. To take advantage of the results from low resolution for reducing these incorrect labelings while preserving details, we propose a multi-resolution cost aggregation method (MultiAgg) by using a soft fusion scheme based on min-convolution. We implement our MultiAgg in applications on stereo matching and interactive image segmentation. Experimental results show that our method significantly outperforms conventional cost aggregation methods in labeling accuracy. Moreover, although MultiAgg is a simple and straight-forward method, it produces results which are close to or even better than those from iterative methods based on global optimization.\",\n \"Title: A Modular Learning Approach for Fish Counting and Measurement Using Stereo Baited Remote Underwater Video\\nYear: 2014\\n\\nAbstract: An approach is suggested for automating fish identification and measurement using stereo Baited Remote Underwater Video footage. Simple methods for identifying fish are not sufficient for measurement, since the snout and tail points must be found, and the stereo data should be incorporated to find a true measurement. We present a modular framework that ties together various approaches in order to develop a generalised system for automated fish detection. A method is also suggested for using machine learning to improve identification. Experimental results indicate the suitability of our approach.\",\n \"Title: Dendritic spines detection based on directional morphological filter and shortest path.\\nYear: 2012\\n\\nAbstract: Dendritic spines are membranous protrusions from neuron's dendrites. They play a very important role in the nervous system. They are very small and have various shapes; hence it is very challenging to detect them in neuron images. This paper presents a novel method for detecting dendritc spines in 2D images. A new dendrite backbone or centerline extraction method is introduced herein which is based on an iterative process between smoothing and extraction. The proposed method iteratively refines the extraction result using both directional morphological filtering and improved Hessian filtering until a satisfactory extraction result is obtained. A shortest path method is applied along a backbone to extract the boundary of the dendrites. Spines are then segmented from the dendrites outside the extracted boundary. The proposed algorithm has been tested with many images and good results are achieved.\",\n \"Title: Tubule detection in testis images using boundary weighting and circular shortest path.\\nYear: 2013\\n\\nAbstract: In studies of germ cell transplantation, measureing tubule diameters and counting cells from different populations using antibodies as markers are very important. Manual measurement of tubule sizes and cell counts is a tedious and sanity grinding work. In this paper, we propose a new boundary weighting based tubule detection method. We first enhance the linear features of the input image and detect the approximate centers of tubules. Next, a boundary weighting transform is applied to the polar transformed image of each tubule region and a circular shortest path is used for the boundary detection. Then, ellipse fitting is carried out for tubule selection and measurement. The algorithm has been tested on a dataset consisting of 20 images, each having about 20 tubules. Experiments show that the detection results of our algorithm are very close to the results obtained manually.\",\n \"Title: Finding splitting lines for touching cell nuclei with a shortest path algorithm.\\nYear: 2015\\n\\nAbstract: A shortest path-based algorithm is proposed in this paper to find splitting lines for touching cell nuclei. First, an initial splitting line is obtained through the distance transform of a marker image and the watershed algorithm. The initial splitting line is then separated into different line segments as necessary, and the endpoint positions of these line segments are adjusted to the concave points on the contour. Finally, a shortest path algorithm is used to find the accurate splitting line between the starting-point and the end-point, and the final split can be achieved by the contour of the touching cell nuclei and the splitting lines. Comparisons of experimental results show that the proposed algorithm is effective for segmentation of different types of touching cell nuclei.\",\n \"Title: Edge-Aware Filtering with Local Polynomial Approximation and Rectangle-Based Weighting.\\nYear: 2016\\n\\nAbstract: This paper presents a novel method for performing guided image filtering using local polynomial approximation (LPA) with range guidance. In our method, the LPA is introduced into a multipoint framework for reliable model regression and better preservation on image spatial variation which usually contains the essential information in the input image. In addition, we develop a weighting scheme which...\",\n \"Title: Stereo Vision Technologies For China'S Lunar Rover Exploration Mission\\nYear: 2016\\n\\nAbstract: In December 2013, China's Chang'e-3 system successfully landed on the moon and started carrying out rover exploration. Stereo vision technologies have played important roles in terrain mapping and rover localization from beginning to end in this mission. This article mainly discusses two stereo vision implementations. First, using stereo images, we propose a novel solution for terrain reconstruction that combines both geometry and appearance information within a joint model. This model could provide accurate terrain maps for the rover navigation and virtual reality simulations of the mission. Then, we present a binocular visual odometry algorithm which leverages an image-network-based bundle adjustment to obtain rover's position and orientation from cross-site images. In addition, both field experiments and onboard results are presented to verify the effectiveness of the proposed algorithms. These technologies have contributed substantially to the Chang'e-3 mission and help to lay a foundation for future autonomous missions.\",\n \"Title: Orientation-guided geodesic weighting for PatchMatch-based stereo matching\\nYear: 2016\\n\\nAbstract: We propose an orientation-guided geodesic weighting (OGGW) strategy for local stereo matching.We propose a method of cost volume filtering combining a multipoint LPA method with our OGGW strategy.We propose a PatchMatch filter with curved surface fitting (PMF-CS) to obtain a disparity map with sub-pixel accuracy. Recently, PatchMatch-based methods for local stereo matching are experiencing great progress with the use of compact and over-segmented regions that have similar intensities or colors. Using patches as support regions, this paper proposes an orientation-guided geodesic weighting (OGGW) strategy to search for an approximate shortest path from a support pixel in the patch to a pixel of interest along a guided orientation. The OGGW is computed by accumulating intensity differences or color dissimilarities between connected pixels along the path. After obtaining matching cost updates by model fitting, the OGGW is used for weighted averaging on the updated costs to obtain a filtered cost volume. In addition, a new filtering method that combines the PatchMatch filter with curved surface fitting (PMF-CS) is presented in this paper. Curved surface fitting along with outliers removal is carried out to seek for a reliable regression model for estimating the disparities on a patch and to achieve a disparity map with sub-pixel accuracy. We conduct a number of experiments to evaluate the performances of OGGW and PMF-CS on cost volume filtering and disparity estimation. Experimental results show that our algorithm produces accurate stereo matching results and outperforms the current state-of-the-art PatchMatch-based methods.\",\n \"Title: Guided image completion by confidence propagation\\nYear: 2016\\n\\nAbstract: This paper presents a new guided image completion method which fills any missing values by considering information from a guidance image. We develop a confidence propagation scheme that allows the filling process to be carried out simultaneously without the need of deciding the filling order. We conduct experiments in several applications where the problem can be formulated into a guided image completion problem, such as interactive segmentation and colorization. The experimental results show that our method provides good results and succeeds in situations where conventional methods fail. In addition, as all building blocks are parallel processes, our method is much suitable for hardware acceleration. HighlightsWe proposed an image completion method based on a new formulation.Filling process was carried out simultaneously without deciding the filling order.Good results were obtained in a variety of applications.\",\n \"Title: Feature matching in stereo images encouraging uniform spatial distribution\\nYear: 2015\\n\\nAbstract: Finding feature correspondences between a pair of stereo images is a key step in computer vision for 3D reconstruction and object recognition. In practice, a larger number of correct correspondences and a higher percentage of correct matches are beneficial. Previous researches show that the spatial distribution of correspondences are also very important especially for fundamental matrix estimation. So far, no existing feature matching method considers the spatial distribution of correspondences. In our research, we develop a new algorithm to find good correspondences in all the three aspects mentioned, i.e., larger number of correspondences, higher percentage of correct correspondences, and better spatial distribution of correspondences. Our method consists of two processes: an adaptive disparity smoothing filter to remove false correspondences based on the disparities of neighboring correspondences and a matching exploration algorithm to find more correspondences according to the spatial distribution of correspondences so that the correspondences are as uniformly distributed as possible in the images. To find correspondences correctly and efficiently, we incorporate the cheirality constraint under an epipole polar transformation together with the epipolar constraint to predict the potential location of matching point. Experiments demonstrate that our method performs very well on both the number of correct correspondences and the percentage of correct correspondences; and the obtained correspondences are also well distributed over the image space. HighlightsWe propose a feature matching method incorporating epipolar geometry estimation.Uniformly distributed correspondences improve epipolar geometry estimation.We propose an adaptive weighted median filter for removing false correspondences.Good results are obtained on both epipolar geometry estimation and correspondences.\",\n \"Title: Stereo matching based on multi-direction polynomial model.\\nYear: 2016\\n\\nAbstract: This paper presents a segmentation based stereo matching algorithm. For the purposes of both preserving the shape of object surfaces and being robust to under segmentations, we introduce a new scene formulation where the reference image is divided into overlapping lines. The disparity value and the index of pixels on lines are modeled by polynomial functions. Polynomial functions are propagated among lines to obtain smooth surfaces via solving energy minimizing problems. Finally, the disparity of pixels is estimated from the disparity fields provided by lines. Because lines in multiple directions implicitly segment different objects in an under segmentation region, our method is robust for under segmented regions where it is usually difficult for conventional region based methods to produce satisfactory results. Experimental results demonstrate that the proposed method has an outstanding performance compared with the current state-of-the-art methods. The scene representation method in this work is also a powerful approach to surface based scene representations.\",\n \"Title: Infrared ship target segmentation through integration of multiple feature maps.\\nYear: 2016\\n\\nAbstract: We investigate the issue of ship target segmentation in infrared (IR) images, and propose an efficient method based on feature map integration. It consists of mainly two procedures: salient region detection based on multiple feature map integration and salient region segmentation based on locally adaptive thresholding. Firstly, a saliency map is constructed by integrating multiple features of IR ship targets, including gray level intensity, local contrast, salient linear structures, and edge strength. Secondly, we propose an adaptive thresholding method to segment each local salient region, and a target selection procedure based on shape features is used to remove background and obtain the true target. Experimental results show that the proposed method performs well for IR ship target segmentation. The advantage of the proposed method is demonstrated in both visual and quantitative comparisons, especially for IR images with a bright background or a ship target close to port. Proposing a saliency map detection method through multiple feature map integration.Proposing an adaptive thresholding method for local salient region segmentation.Achieving a better performance for both visual and quantitative evaluations.\",\n \"Title: Automated Image Analysis on Insect Soups\\nYear: 2016\\n\\nAbstract: Insect soup is a term which refers to the bulk ethanol samples stored in many insect collections which contain unsorted material and augment the dried and pinned collections. These samples often contain rare or unique specimens, and it would be beneficial to researchers or even the general public to have an easy way to access them to view and annotate their contents. In this paper we propose an image analysis approach for analysing insect soup images and making their content readily available through the web. The analysis includes insect segmentation, scale bar segmentation, measurements, and insect sorting. The image analysis results will be used for online access of insect images for insect classification, research, species discovery, or educational purposes.\",\n \"Title: A Local Scale Selection Scheme for Multiscale Area Integral Invariants\\nYear: 2016\\n\\nAbstract: Area integral invariant (AII) is a functional obtained by performing integral operations on the closed planar contour of a shape via the convolution with disc kernels. This shape descriptor is insensitive to noise and robust with respect to occlusions. AII intrinsically introduces the notion of scale using the size of kernel radius. However how to select an optimal scale remains unresolved. In this paper, we propose a local scale selection scheme for generating multiscale area integral invariants. For the same scale level, the disc kernel size is not fixed and varies with the contour point where the disc is centered. This scheme also provides a scale assignment for emphasising the features extraction at finer scales. The strong discriminative power of the multiscale area integral invariant derived from the proposed scale selection scheme has been validated through experiments on very challenging leaf image retrievals.\",\n \"Title: Robust Multi-Frame Adaptive Optics Image Restoration Algorithm Using Maximum Likelihood Estimation with Poisson Statistics.\\nYear: 2017\\n\\nAbstract: An adaptive optics (AO) system provides real-time compensation for atmospheric turbulence. However, an AO image is usually of poor contrast because of the nature of the imaging process, meaning that the image contains information coming from both out-of-focus and in-focus planes of the object, which also brings about a loss in quality. In this paper, we present a robust multi-frame adaptive optics image restoration algorithm via maximum likelihood estimation. Our proposed algorithm uses a maximum likelihood method with image regularization as the basic principle, and constructs the joint log likelihood function for multi-frame AO images based on a Poisson distribution model. To begin with, a frame selection method based on image variance is applied to the observed multi-frame AO images to select images with better quality to improve the convergence of a blind deconvolution algorithm. Then, by combining the imaging conditions and the AO system properties, a point spread function estimation model is built. Finally, we develop our iterative solutions for AO image restoration addressing the joint deconvolution issue. We conduct a number of experiments to evaluate the performances of our proposed algorithm. Experimental results show that our algorithm produces accurate AO image restoration results and outperforms the current state-of-the-art blind deconvolution methods.\",\n \"Title: Fog Density Estimation and Image Defogging Based on Surrogate Modeling for Optical Depth.\\nYear: 2017\\n\\nAbstract: In order to estimate fog density correctly and to remove fog from foggy images appropriately, a surrogate model for optical depth is presented in this paper. We comprehensively investigate various fog-relevant features and propose a novel feature based on the hue, saturation, and value color space, which correlate well with the perception of fog density. We use a surrogate-based method to learn a ...\",\n \"Title: Multi-Modal Ship Target Image Smoothing Based on Adaptive Mean Shift.\\nYear: 2018\\n\\nAbstract: In this paper, we propose an adaptive image smoothing method for infrared (IR) and visual ship target images, aiming to effectively suppress noise as well as preserve important target structures, thus benefiting image segmentation. First, by analyzing the specific features of ship target images, a block based method combining local region mean and standard deviation is developed to highlight ship target regions. It is helpful to distinguish the ship target region from the background. Then, by associating the range bandwidth with local image properties of the ship target region and the background region, we develop an adaptive range bandwidth mean shift filtering method for IR and visual ship target image smoothing. With this proposed method, we can obtain a small bandwidth for ship target region and a large one for the background region. Therefore, we can effectively smooth the background while preserving the details of the targets. Experimental results show that this method works well for IR and visual ship target images with different backgrounds. The method demonstrates superior performance for image smoothing and target preservation compared with the four well-known edge-preserving denoising methods, including the anisotropic diffusion filtering, the bilateral filtering, the propagation filtering, and the mean shift filtering with fixed range bandwidth.\",\n \"Title: Matching Pursuit Based On Kernel Non-Second Order Minimization\\nYear: 2018\\n\\nAbstract: The orthogonal matching pursuit (OMP) is an important sparse approximation algorithm to recover sparse signals from compressed measurements. However, most MP algorithms are based on the mean square error(MSE) to minimize the recovery error, which is suboptimal when there are outliers. In this paper, we present a new robust OMP algorithm based on kernel non-second order statistics (KNS-OMP), which not only takes advantages of the outlier resistance ability of correntropy but also further extends the second order statistics based correntropy to a non-second order similarity measurement to improve its robustness. The resulted framework is more accurate than the second order ones in reducing the effect of outliers. Experimental results on synthetic and real data show that the proposed method achieves better performances compared with existing methods.\",\n \"Title: Disparity-based Error Concealment for Stereoscopic Images with Superpixel Segmentation.\\nYear: 2018\\n\\nAbstract: To solve the problem of transmission errors in stereoscopic images, this paper proposes a novel error concealment (EC) method using superpixel segmentation and adaptive disparity selection (SSADS). Our algorithm consists of two steps. The first step is disparity estimation for each pixel in a reference image. In this step, the numbers of superpixel segmentation labels of stereoscopic images are used as a new constraint for disparity matching to reduce the effect of mismatching. The second step is disparity selection for a lost block. In this step, a strategy based on boundary smoothness is proposed to adaptively select the optimal disparity which is used for error concealment. Experimental results demonstrate that compared with other methods, the proposed method has significant advantages in both objective and subjective quality assessment.\",\n \"Title: RA-UNet: A hybrid deep attention-aware network to extract liver and tumor in CT scans.\\nYear: 2018\\n\\nAbstract: Automatic extraction of liver and tumor from CT volumes is a challenging task due to their heterogeneous and diffusive shapes. Recently, 2D and 3D convolutional neural networks have become popular in medical image segmentation tasks because of the utilization of large labeled datasets to learn hierarchical features. However, 3D networks have some drawbacks due to their high cost on computational resources. In this paper, we propose a 3D hybrid residual attention-aware segmentation method, named RA-UNet, to precisely extract the liver volume of interests (VOI) and segment tumors from the liver VOI. The proposed network has a basic architecture as a 3D U-Net which extracts contextual information combining low-level feature maps with high-level ones. Attention modules are stacked so that the attention-aware features change adaptively as the network goes very deep and this is made possible by residual learning. This is the first work that an attention residual mechanism is used to process medical volumetric images. We evaluated our framework on the public MICCAI 2017 Liver Tumor Segmentation dataset and the 3DIRCADb dataset. The results show that our architecture outperforms other state-of-the-art methods. We also extend our RA-UNet to brain tumor segmentation on the BraTS2018 and BraTS2017 datasets, and the results indicate that RA-UNet achieves good performance on a brain tumor segmentation task as well.\",\n \"Title: Design of a Clinical Decision Support System for Predicting Erectile Dysfunction in Men Using NHIRD Dataset.\\nYear: 2019\\n\\nAbstract: Erectile dysfunction (ED) affects millions of men worldwide. Men with ED generally complain failure to attain or maintain an adequate erection during sexual activity. The prevalence of ED is strongly correlated with age, affecting about 40% of men at age 40 and nearly 70% at age 70. A variety of chronic diseases, including diabetes, ischemic heart disease, congestive heart failure, hypertension, depression, chronic renal failure, obstructive sleep apnea, prostate disease, gout, and sleep disorder, were reported to be associated with ED. In this study, data retrieved from a subset of the National Health Insurance Research Database (NHIRD) of Taiwan were used for designing the clinical decision support system (CDSS) for predicting ED incidences in men. The positive cases were male patients aged 20-65 who were diagnosed with ED between Jan. 2000 and Dec. 2010 confirmed by at least 3 outpatient visits or at least one inpatient visit, while the negative cases were randomly selected from the database without a history of ED and were frequency (1:1), age, and index year matched with the ED patients. Data of a total of 2,832 ED patients and 2,832 non-ED patients, each consisting of 41 features including index age, 10 comorbidities, and 30 other comorbidity-related variables, were retrieved for designing the predictive models. Integrated genetic algorithm (GA) and support vector machine (SVM) was adopted to design the CDSSs with 2 experiments of independent training and testing (ITT) conducted to verify their effectiveness. In the 1st ITT experiment, data extracted from Jan. 2000 till Dec. 2005 (61.51%, 1,742 positive cases and 1,742 negative cases) were used for training and validating and the data retrieved from Jan. 2006 till Dec. 2010 were used for testing (38.49%); whereas in the 2nd ITT experiment, data in the training set (77.78%) were extracted from Jan. 2000 till Dec. 2007 and those in the testing set (22.22%) were retrieved afterward. Tenfold cross validation and 3 different objective functions were adopted for obtaining the optimal models with best predictive performance in the training phase. The testing results show that the CDSSs achieved a predictive performance with accuracy, sensitivity, specificity, g-mean, and area under ROC curve (AUC) of 74.72-76.65%, 72.33-83.76%, 69.54-77.10%, 0.7468-0.7632, and 0.766-0.817, respectively. In conclusion, the CDSSs designed based on cost-sensitive objective functions as well as salient comorbidity-related features achieve satisfactory predictive performance for predicting ED incidences.\",\n \"Title: Image dehazing using adaptive bi-channel priors on superpixels.\\nYear: 2017\\n\\nAbstract: •We propose an adaptive bi-channel priors (ABiCP) method for image dehazing.•We propose a method for determining the white and black pixels in haze images based on the HSV space.•We introduce the superpixels in image dehazing to reduce computational time and improve the quality of restored images.\",\n \"Title: Cell Segmentation Based on FOPSO Combined with Shape Information Improved Intuitionistic FCM.\\nYear: 2019\\n\\nAbstract: Fuzzy c-means (FCM) clustering algorithms have been proved to be effective image segmentation techniques. However, FCM clustering algorithms are sensitive to noises and initialization. They cannot effectively segment cell images with inhomogeneous gray value distributions and complex touching cells. Aiming to overcome these disadvantages, this paper proposes a cell image segmentation algorithm usi...\",\n \"Title: Discrete Curvature Representations for Noise Robust Image Corner Detection.\\nYear: 2019\\n\\nAbstract: Image corner detection is very important in the fields of image analysis and computer vision. Curvature calculation techniques are used in many contour-based corner detectors. We identify that existing calculation of curvature is sensitive to local variation and noise in the discrete domain and does not perform well when corners are closely located. In this paper, discrete curvature representations of single and double corner models are investigated and obtained. A number of model properties have been discovered which help us detect corners on contours. It is shown that the proposed method has a high corner resolution (the ability to accurately detect neighbouring corners) and a corresponding corner resolution constant is also derived. Meanwhile, this method is less sensitive to any local variations and noise on the contour; and false corner detection is less likely to occur. The proposed detector is compared with seven state-of-the-art detectors. Three test images with ground truths are used to assess the detection capability and localization accuracy of these methods in noise-free and cases with different noise levels. Twenty-four images with various scenes without ground truths are used to evaluate their repeatability under affine transformation, JPEG compression, and noise degradations. The experimental results show that our proposed detector attains a better overall performance.\",\n \"Title: Robust Retinal Image Enhancement via Dual-Tree Complex Wavelet Transform and Morphology-Based Method.\\nYear: 2019\\n\\nAbstract: Retinal image processing is very important in the field of clinical medicine. As the first step in retinal image processing, image enhancement is essential. Because the details of a retinal image are complex and difficult to enhance, we present a robust retinal image enhancement algorithm via a dual-tree complex wavelet transform (DTCWT) and morphology-based method in this paper. To begin with, we utilize the pre-processing method to the captured retinal images. Then, the DTCWT is applied to decompose the gray retinal image to obtain high-pass subbands and low-pass subbands. Then, a Contourlet-based enhancement method is applied to the high-pass subbands. For the low-pass subbands, we improve the morphology top-hat transform by adding dynamic multi-scale parameters to achieve an equivalent percentage enhancement and at the same time achieve multi-scale transforms in multiple directions. Finally, we develop the inverse DTCWT method to obtain the enhanced retinal image after processing the low-frequency subimages and high-frequency subimages. We compare this approach with enhancement based on the adaptive unsharp masking, histogram equalization, and multi-scale retinex. We present the test results of our algorithm on 440 retinal images from the DRIVE and the STARE databases. The experimental results show that the proposed approach can achieve better results, and might be helpful for vessel segmentation.\",\n \"Title: Knowledge Adaptation for Efficient Semantic Segmentation\\nYear: 2019\\n\\nAbstract: Both accuracy and efficiency are of significant importance to the task of semantic segmentation. Existing deep FCNs suffer from heavy computations due to a series of high-resolution feature maps for preserving the detailed knowledge in dense estimation. Although reducing the feature map resolution (i.e., applying a large overall stride) via subsampling operations (e.g., polling and convolution striding) can instantly increase the efficiency, it dramatically decreases the estimation accuracy. To tackle this dilemma, we propose a knowledge distillation method tailored for semantic segmentation to improve the performance of the compact FCNs with large overall stride. To handle the inconsistency between the features of the student and teacher network, we optimize the feature similarity in a transferred latent domain formulated by utilizing a pre-trained autoencoder. Moreover, an affinity distillation module is proposed to capture the long-range dependency by calculating the non local interactions across the whole image. To validate the effectiveness of our proposed method, extensive experiments have been conducted on three popular benchmarks: Pascal VOC, Cityscapes and Pascal Context. Built upon a highly competitive baseline, our proposed method can improve the performance of a student network by 2.5% (mIOU boosts from 70.2 to 72.7 on the cityscapes test set) and can train a better compact model with only 8% float operations (FLOPS) of a model that achieves comparable performances.\",\n \"Title: Chord Bunch Walks for Recognizing Naturally Self-overlapped and Compound Leaves.\\nYear: 2019\\n\\nAbstract: Effectively describing and recognizing leaf shapes under arbitrary variations, particularly from a large database, remains an unsolved problem. In this research, we attempted a new strategy of describing leaf shapes by walking and measuring along a bunch of chords that pass through the shape. A novel chord bunch walks (CBW) descriptor is developed through the chord walking behaviour that effectively integrates the shape image function over the walked chord to reflect both the contour features and the inner properties of the shape. For each contour point, the chord bunch groups multiple pairs of chords to build a hierarchical framework for a coarse-to-fine description that can effectively characterize not only the subtle differences among leaf margin patterns but also the interior part of the shape contour formed inside a self-overlapped or compound leaf. Instead of using optimal correspondence based matching, a Log-Min distance that encourages one-to-one correspondences is proposed for efficient and effective CBW matching. The proposed CBW shape analysis method is invariant to rotation, scaling, translation, and mirror transforms. Five experiments, including image retrieval of compound leaves, image retrieval of naturally self-overlapped leaves, and retrieval of mixed leaves on three large scale datasets, are conducted. The proposed method achieved large accuracy increases with low computational costs over the state-of-the-art benchmarks, which indicates the research potential along this direction.\",\n \"Title: Disentangled Representation Learning and Enhancement Network for Single Image De-Raining\\nYear: 2021\\n\\nAbstract: ABSTRACTIn this paper, we present a disentangled representation learning and enhancement network (DRLE-Net) to address the challenging single image de-raining problems, i.e., raindrop and rain streak removal. Specifically, the DRLE-Net is formulated as a multi-task learning framework, and an elegant knowledge transfer strategy is designed to train the encoder of DRLE-Net to embed a rainy image into two separated latent spaces representing the task (clean image reconstruction in this paper) relevant and irrelevant variations respectively, such that only the essential task-relevant factors will be used by the decoder of DRLE-Net to generate high-quality de-raining results. Furthermore, visual attention information is modeled and fed into the disentangled representation learning network to enhance the task-relevant factor learning. To facilitate the optimization of the hierarchical network, a new adversarial loss formulation is proposed and used together with the reconstruction loss to train the proposed DRLE-Net. Extensive experiments are carried out for removing raindrops or rainstreaks from both synthetic and real rainy images, and DRLE-Net is demonstrated to produce significantly better results than state-of-the-art models.\",\n \"Title: A Novel Decision Mechanism for Image Edge Detection.\\nYear: 2021\\n\\nAbstract: Edge detection plays an important role in image processing and computer vision tasks such as image matching and image segmentation. In this paper, we argue that edges in image should always be detected even under different image rotation transformations or noise conditions. Then a novel edge detection method is presented for improving the robustness of image edge detection. Firstly, we perform the following operations on the input image to obtain a set of images: random destruction operation, random rotation operation, and randomly addition of Gaussian noise with zero mean and random variance. Secondly, multi-directional Gabor filters with multiple scales are used as a tool to smooth the set of images and obtain candidate edges from the set of images. Thirdly, a novel edge decision mechanism is designed with a 0.95 confidence interval for selecting true edges from the candidate edges. Finally, two edge detection evaluation criteria (i.e., the aggregate test receiver-operating-characteristic and the Pratt’s figure of merit) are utilized to evaluate the proposed edge detection method against five state-of-the-art methods. The experimental results show that our proposed method outperforms all the other tested methods.\",\n \"Title: Deep learning based stereo cost aggregation on a small dataset\\nYear: 2021\\n\\nAbstract: Deep learning (DL) has been used in many computer vision tasks including stereo matching. However, DL is data hungry, and a large number of highly accurate real-world training images for stereo matching is too expensive to acquire in practice. The majority of studies rely on large simulated datasets during training, which inevitably results in domain shift problems that are commonly compensated by...\",\n \"Title: Efficient corner detection based on corner enhancement filters\\nYear: 2022\\n\\nAbstract: Multi-scale analysis based corner detection algorithms yield impressive performance, however, they are not efficient and not suitable for real-time computer vision tasks. The classical corner detection algorithms including FAST and Harris are computationally efficient, but their detection accuracy and repeatability are insufficient. This paper describes a novel fast corner detector with a simple architecture and high parallel computing characteristics. In order to simplify the corner detection architecture and improve its parallel computing performance, a new type of filter is proposed that can enhance corners and suppress edges and noise simultaneously. Then a novel efficient corner detector is proposed, which can be adapted to achieve real-time detection in hardware. Experimental results show that, with a very low computational cost and simple architecture, the proposed detector can achieve or even exceed the detection accuracy of multi-scale analysis based detectors. Its repeatability is similar to multi-scale analysis based detectors and clearly higher than other types of corner detectors. Therefore, it is potentially useful as an efficient corner detector for computer vision applications especially for portable real-time tasks. (c) 2022 Elsevier Inc. All rights reserved.\",\n \"Title: Improving Monocular Visual Odometry Using Learned Depth\\nYear: 2022\\n\\nAbstract: Monocular visual odometry (VO) is an important task in robotics and computer vision. Thus far, how to build accurate and robust monocular VO systems that can work well in diverse scenarios remains largely unsolved. In this article, we propose a framework to exploit monocular depth estimation for improving VO. The core of our framework is a monocular depth estimation module with a strong generalization capability for diverse scenes. It consists of two separate working modes to assist the localization and mapping. With a single monocular image input, the depth estimation module predicts a relative depth to help the localization module on improving the accuracy. With a sparse depth map and an RGB image input, the depth estimation module can generate accurate scale-consistent depth for dense mapping. Compared with current learning-based VO methods, our method demonstrates a stronger generalization ability to diverse scenes. More significantly, our framework is able to boost the performances of existing geometry-based VO methods by a large margin.\",\n \"Title: Recent advances on image edge detection: A comprehensive review\\nYear: 2022\\n\\nAbstract: •The recent edge detection methods are classified and introduced in detail.•The existing widely used datasets and evaluation criteria for edge detection methods are reviewed.•Future research directions for edge detection are elaborated.\",\n \"Title: EOCSA: Predicting prognosis of Epithelial ovarian cancer with whole slide histopathological images\\nYear: 2022\\n\\nAbstract: Ovarian cancer is one of the most serious cancers that threaten women around the world. Epithelial ovarian cancer (EOC), as the most commonly seen subtype of ovarian cancer, has rather high mortality rate and poor prognosis among various gynecological cancers. Survival analysis outcome is able to provide treatment advices to doctors. In recent years, with the development of medical imaging technology, survival prediction approaches based on pathological images have been proposed. In this study, we designed a deep framework named EOCSA which analyzes the prognosis of EOC patients based on pathological whole slide images (WSIs). Specifically, we first randomly extracted patches from WSIs and grouped them into multiple clusters. Next, we developed a survival prediction model, named DeepConvAttentionSurv (DCAS), which was able to extract patch-level features, removed less discriminative clusters and predicted the EOC survival precisely. Particularly, channel attention, spatial attention, and neuron attention mechanisms were used to improve the performance of feature extraction. Then patient-level features were generated from our weight calculation method and the survival time was finally estimated using LASSO-Cox model. The proposed EOCSA is efficient and effective in predicting prognosis of EOC and the DCAS ensures more informative and discriminative features can be extracted. As far as we know, our work is the first to analyze the survival of EOC based on WSIs and deep neural network technologies. The experimental results demonstrate that our proposed framework has achieved state-of-the-art performance of 0.980 C-index. The implementation of the approach can be found at https://github.com/RanSuLab/EOCprognosis.\",\n \"Title: Semi-supervised Histological Image Segmentation via Hierarchical Consistency Enforcement\\nYear: 2022\\n\\nAbstract: Acquiring pixel-level annotations for histological image segmentation is time- and labor- consuming. Semi-supervised learning enables learning from the unlabeled and limited amount of labeled data. A challenging issue is the inconsistent and uncertain predictions on unlabeled data. To enforce invariant predictions over the perturbations applied to the hidden feature space, we propose a Mean-Teacher based hierarchical consistency enforcement (HCE) framework and a novel hierarchical consistency loss (HC-loss) with learnable and selfguided mechanisms. Specifically, the HCE takes the perturbed versions of the hierarchical features from the encoder as input to the auxiliary decoders, and encourages the predictions of the auxiliary decoders and the main decoder to be consistent. The HC-loss facilitates the teacher model to generate reliable guidance and enhances the consistency among all the decoders of the student model. The proposed method is simple, yet effective, which can easily be extended to other frameworks. The quantitative and qualitative experimental results indicate the effectiveness of the hierarchical consistency enforcement on the MoNuSeg and CRAG datasets.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":85,"string":"85"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13595,"string":"13,595"}}},{"rowIdx":13596,"cells":{"question_id":{"kind":"string","value":"Stefan Kiefer"},"question":{"kind":"string","value":"Generate an abstract for the title \"On the Sequential Probability Ratio Test in Hidden Markov Models\" using the following items: SPRT, Hidden Markov Models, Execution Time, Lyapunov Exponents, Complexity Results."},"answer":{"kind":"string","value":"We consider the Sequential Probability Ratio Test applied to Hidden Markov Models. Given two Hidden Markov Models and a sequence of observations generated by one of them, the Sequential Probability Ratio Test attempts to decide which model produced the sequence. We show relationships between the execution time of such an algorithm and Lyapunov exponents of random matrix systems. Further, we give complexity results about the execution time taken by the Sequential Probability Ratio Test."},"correct_answer":{"kind":"string","value":"We consider the Sequential Probability Ratio Test applied to Hidden Markov Models. Given two Hidden Markov Models and a sequence of observations generated by one of them, the Sequential Probability Ratio Test attempts to decide which model produced the sequence. We show relationships between the execution time of such an algorithm and Lyapunov exponents of random matrix systems. Further, we give complexity results about the execution time taken by the Sequential Probability Ratio Test."},"documents":{"kind":"list like","value":["Title: Newtonian program analysis\nYear: 2010\n\nAbstract: This article presents a novel generic technique for solving dataflow equations in interprocedural dataflow analysis. The technique is obtained by generalizing Newton's method for computing a zero of a differentiable function to ω-continuous semirings. Complete semilattices, the common program analysis framework, are a special class of ω-continuous semirings. We show that our generalized method always converges to the solution, and requires at most as many iterations as current methods based on Kleene's fixed-point theorem. We also show that, contrary to Kleene's method, Newton's method always terminates for arbitrary idempotent and commutative semirings. More precisely, in the latter setting the number of iterations required to solve a system of n equations is at most n.","Title: Newton's Method for omega-Continuous Semirings\nYear: 2008\n\nAbstract: Fixed point equations X = f (X) over omega-continuous semirings are a natural mathematical foundation of interprocedural program analysis. Generic algorithms for solving these equations are based on Kleene's theorem, which states that the sequence 0, f (0), f (f (0)),... converges to the least fixed point. However, this approach is often inefficient. We report on recent work in which we extend Newton's method, the well-known technique from numerical mathematics, to arbitrary omega-continuous semirings, and analyze its convergence speed in the real semiring.","Title: Abstraction refinement with craig interpolation and symbolic pushdown systems\nYear: 2006\n\nAbstract: Counterexample-guided abstraction refinement (CEGAR) has proven to be a powerful method for software model-checking. In this paper, we investigate this concept in the context of sequential (possibly recursive) programs whose statements are given as BDDs. We examine how Craig interpolants can be computed efficiently in this case and propose a new, special type of interpolants. Moreover, we show how to treat multiple counterexamples in one refinement cycle. We have implemented this approach within the model-checker Moped and report on experiments.","Title: Zero-reachability in probabilistic multi-counter automata\nYear: 2014\n\nAbstract: We study the qualitative and quantitative zero-reachability problem in probabilistic multi-counter systems. We identify the undecidable variants of the problems, and then we concentrate on the remaining two cases. In the first case, when we are interested in the probability of all runs that visit zero in some counter, we show that the qualitative zero-reachability is decidable in time which is polynomial in the size of a given pMC and doubly exponential in the number of counters. Further, we show that the probability of all zero-reaching runs can be effectively approximated up to an arbitrarily small given error ε > 0 in time which is polynomial in log(ε), exponential in the size of a given pMC, and doubly exponential in the number of counters. In the second case, we are interested in the probability of all runs that visit zero in some counter different from the last counter. Here we show that the qualitative zero-reachability is decidable and SquareRootSum-hard, and the probability of all zero-reaching runs can be effectively approximated up to an arbitrarily small given error ε > 0 (these results apply to pMC satisfying a suitable technical condition that can be verified in polynomial time). The proof techniques invented in the second case allow to construct counterexamples for some classical results about ergodicity in stochastic Petri nets.","Title: Analyzing probabilistic pushdown automata\nYear: 2013\n\nAbstract: The paper gives a summary of the existing results about algorithmic analysis of probabilistic pushdown automata and their subclasses.","Title: Analysis of Probabilistic Basic Parallel Processes.\nYear: 2014\n\nAbstract: Basic Parallel Processes (BPPs) are a well-known subclass of Petri Nets. They are the simplest common model of concurrent programs that allows unbounded spawning of processes. In the probabilistic version of BPPs, every process generates other processes according to a probability distribution. We study the decidability and complexity of fundamental qualitative problems over probabilistic BPPs -- in particular reachability with probability 1 of different classes of target sets (e.g. upward-closed sets). Our results concern both the Markov-chain model, where processes are scheduled randomly, and the MDP model, where processes are picked by a scheduler.","Title: Solving Monotone Polynomial Equations\nYear: 2008\n\nAbstract: We survey some recent results on iterative methods for approximating the least solution of a system of monotone fixed-point polynomial equations.","Title: Computing the Least Fixed Point of Positive Polynomial Systems\nYear: 2010\n\nAbstract: We consider equation systems of the form $X_1=f_1(X_1,\\dots,X_n)$, $\\dots$, $X_n = f_n(X_1,\\dots,X_n)$, where $f_1,\\dots,f_n$ are polynomials with positive real coefficients. In vector form we denote such an equation system by ${\\bf X}={\\bf f}({\\bf X})$ and call ${\\bf f}$ a system of positive polynomials (SPP). Equation systems of this kind appear naturally in the analysis of stochastic models like stochastic context-free grammars (with numerous applications to natural language processing and computational biology), probabilistic programs with procedures, web-surfing models with back buttons, and branching processes. The least nonnegative solution $\\mu{\\bf f}$ of an SPP equation ${\\bf X}={\\bf f}({\\bf X})$ is of central interest for these models. Etessami and Yannakakis [J. ACM, 56 (2009), pp. 1-66] have suggested a particular version of Newton's method to approximate $\\mu{\\bf f}$. We extend a result of Etessami and Yannakakis and show that Newton's method starting at ${\\bf 0}$ always converges to $\\mu{\\bf f}$. We obtain lower bounds on the convergence speed of the method. For so-called strongly connected SPPs we prove the existence of a threshold $k_{{\\bf f}}\\in\\mathbb{N}$ such that for every $i\\geq0$ the $(k_{{\\bf f}}+i)$th iteration of Newton's method has at least $i$ valid bits of $\\mu{\\bf f}$. The proof yields an explicit bound for $k_{{\\bf f}}$ depending only on syntactic parameters of ${\\bf f}$. We further show that for arbitrary SPP equations, Newton's method still converges linearly: there exists a threshold $k_{{\\bf f}}$ and an $\\alpha_{{\\bf f}}0$ such that for every $i\\geq0$ the $(k_{{\\bf f}}+\\alpha_{{\\bf f}}\\cdot i)$th iteration of Newton's method has at least $i$ valid bits of $\\mu{\\bf f}$. The proof yields an explicit bound for $\\alpha_{{\\bf f}}$; the bound is exponential in the number of equations in ${\\bf X}={\\bf f}({\\bf X})$, but we also show that it is essentially optimal. The proof does not yield any bound for $k_{{\\bf f}}$, but only proves its existence. Constructing a bound for $k_{{\\bf f}}$ is still an open problem. Finally, we also provide a geometric interpretation of Newton's method for SPPs.","Title: APEX: an analyzer for open probabilistic programs\nYear: 2012\n\nAbstract: We present APEX, a tool for analysing probabilistic programs that are open, i.e. where variables or even functions can be left unspecified. APEX transforms a program into an automaton that captures the program's probabilistic behaviour under all instantiations of the unspecified components. The translation is compositional and effectively leverages state reduction techniques. APEX can then further analyse the produced automata; in particular, it can check two automata for equivalence which translates to equivalence of the corresponding programs under all environments. In this way, APEX can verify a broad range of anonymity and termination properties of randomised protocols and other open programs, sometimes with an exponential speed-up over competing state-of-the-art approaches.","Title: Runtime analysis of probabilistic programs with unbounded recursion\nYear: 2010\n\nAbstract: We study the runtime in probabilistic programs with unbounded recursion. As underlying formal model for such programs we use probabilistic pushdown automata (pPDAs) which exactly correspond to recursive Markov chains. We show that every pPDA can be transformed into a stateless pPDA (called \\\"pBPA\\\") whose runtime and further properties are closely related to those of the original pPDA. This result substantially simplifies the analysis of runtime and other pPDA properties. We prove that for every pPDA the probability of performing a long run decreases exponentially in the length of the run, if and only if the expected runtime in the pPDA is finite. If the expectation is infinite, then the probability decreases \\\"polynomially\\\". We show that these bounds are asymptotically tight. Our tail bounds on the runtime are generic, i.e., applicable to any probabilistic program with unbounded recursion. We study the runtime in probabilistic programs with unbounded recursion.A transformation to make probabilistic PDAs stateless is proposed.We give bounds on the probability of long runs of a recursive program.","Title: Convergence Thresholds of Newton's Method for Monotone Polynomial Equations\nYear: 2008\n\nAbstract: Monotone systems of polynomial equations (MSPEs) axe systems of fixed-point equations X-1 = f(1) (X-1, . . . , X-n), . . . , X-n = f (X-1, . . . , X-n) where each fi is a polynomial with positive real coefficients. The question of computing the least non-negative solution of a given MSPE X = f(X) arises naturally in the analysis of stochastic models such as stochastic context-free grammars, probabilistic pushdown automata, and barkbutton processes. Etessami! and Yannakakis have recently adapted Newton's iterative method to MSPEs. In a previous paper we have proved the existence of a threshold k(f) for strongly connected MSPEs, such that after kf iterations of Newton's method each new iteration computes at least 1 new bit of the solution. However, the proof was purely existential. In this paper we give an upper bound for kf as a function of the minimal component of the least fixed-point mu f of f(X). Using this result we show that kf is at most single exponential resp. linear for strongly connected MSPEs derived from probabilistic pushdown automata resp. from back-button processes. Further, we prove the existence of a threshold for arbitrary MSPEs after which each new iteration computes at least 1/w2(h) new bits of the solution, where w and h are the width and height of the DAG of strongly connected components.","Title: Efficient Analysis of Probabilistic Programs with an Unbounded Counter.\nYear: 2011\n\nAbstract: We show that a subclass of infinite-state probabilistic programs that can be modeled by probabilistic one-counter automata (pOC) admits an efficient quantitative analysis. In particular, we show that the expected termination time can be approximated up to an arbitrarily small relative error with polynomially many arithmetic operations, and the same holds for the probability of all runs that satisfy a given ω-regular property. Further, our results establish a powerful link between pOC and martingale theory, which leads to fundamental observations about quantitative properties of runs in pOC. In particular, we provide a \"divergence gap theorem\", which bounds a positive non-termination probability in pOC away from zero.","Title: Derivation Tree Analysis for Accelerated Fixed-Point Computation\nYear: 2011\n\nAbstract: We show that for several classes of idempotent semirings the least fixed-point of a polynomial system of equations X = f(X) is equal to the least fixed-point of a linear system obtained by \"linearizing\" the polynomials of f in a certain way. Our proofs rely on derivation tree analysis, a proof prin ciple that combines methods from algebra, calculus, and formal language theory, and was first used in (5) to show that Newton's method over commutative and idem po- tent semirings converges in a linear number of steps. Our results lead to efficient generic algorithms for computing the least fixed-point. We use these algor ithms to derive several consequences, including an O(N3) algorithm for computing the throughput of a context-free grammar (obtained by speeding up the O(N4) algo- rithm of (2)), and a generalization of Courcelle's result stating that the do wnward- closed image of a context-free language is regular (3).","Title: Stability and Complexity of Minimising Probabilistic Automata.\nYear: 2014\n\nAbstract: We consider the state-minimisation problem for weighted and probabilistic automata. We provide a numerically stable polynomial-time minimisation algorithm for weighted automata, with guaranteed bounds on the numerical error when run with floating-point arithmetic. Our algorithm can also be used for \"lossy\" minimisation with bounded error. We show an application in image compression. In the second part of the paper we study the complexity of the minimisation problem for probabilistic automata. We prove that the problem is NP-hard and in PSPACE, improving a recent EXPTIME-result.","Title: On the convergence of Newton's method for monotone systems of polynomial equations\nYear: 2007\n\nAbstract: Monotone systems of polynomial equations (MSPEs) are systems of fixed-point equations X1 = f1(X1, ..., Xn), ..., Xn = fn(X1, ..., Xn) where each fi is a polynomial with positive real coefficients. The question of computing the least non-negative solution of a given MSPE X = f(X) arises naturally in the analysis of stochastic context-free grammars, recursive Markov chains, and probabilistic pushdown automata. While the Kleene sequence f(0), f(f(0)), ... always converges to the least solution mu.f, if it exists, the number of iterations needed to compute the first i bits of mu.f may grow exponentially in i.Etessami and Yannakakis have recently adapted Newton's iterative method to MSPEs and proved that the Newton sequence converges at least as fast as the Kleene sequence and exponentially faster in many cases.They conjecture that, given an MSPE of size m, the number of Newton iterations needed to obtain i accurate bits of mu.f grows polynomially in i and m. In this paper we show that the number of iterations grows linearly in i for strongly connected MSPEs and may grow exponentially in m for general MSPEs.","Title: On the total variation distance of labelled Markov chains\nYear: 2014\n\nAbstract: Labelled Markov chains (LMCs) are widely used in probabilistic verification, speech recognition, computational biology, and many other fields. Checking two LMCs for equivalence is a classical problem subject to extensive studies, while the total variation distance provides a natural measure for the \\\"inequivalence\\\" of two LMCs: it is the maximum difference between probabilities that the LMCs assign to the same event. In this paper we develop a theory of the total variation distance between two LMCs, with emphasis on the algorithmic aspects: (1) we provide a polynomial-time algorithm for determining whether two LMCs have distance 1, i.e., whether they can almost always be distinguished; (2) we provide an algorithm for approximating the distance with arbitrary precision; and (3) we show that the threshold problem, i.e., whether the distance exceeds a given threshold, is NP-hard and hard for the square-root-sum problem. We also make a connection between the total variation distance and Bernoulli convolutions.","Title: BPA bisimilarity is EXPTIME-hard\nYear: 2013\n\nAbstract: Given a basic process algebra (BPA) and two stack symbols, the BPA bisimilarity problem asks whether the two stack symbols are bisimilar. We show that this problem is EXPTIME-hard.","Title: Space-efficient scheduling of stochastically generated tasks\nYear: 2010\n\nAbstract: We study the problem of scheduling tasks for execution by a processor when the tasks can stochastically generate new tasks. Tasks can be of different types, and each type has a fixed, known probability of generating other tasks. We present results on the random variable S^@s modeling the maximal space needed by the processor to store the currently active tasks when acting under the scheduler @s. We obtain tail bounds for the distribution of S^@s for both offline and online schedulers, and investigate the expected value E[S^@s].","Title: Bisimilarity of Probabilistic Pushdown Automata\nYear: 2012\n\nAbstract: We study the bisimilarity problem for probabilistic pushdown automata (pPDA) and subclasses thereof. Our definition of pPDA allows both probabilistic and non-deterministic branching, generalising the classical notion of pushdown automata (without epsilon-transitions). Our first contribution is a general construction that reduces checking bisimilarity of probabilistic transition systems to checking bisimilarity of non-deterministic transition systems. This construction directly yields decidability of bisimilarity for pPDA, as well as an elementary upper bound for the bisimilarity problem on the subclass of probabilistic basic process algebras, i.e., single-state pPDA. We further show that, with careful analysis, the general reduction can be used to prove an EXPTIME upper bound for bisimilarity of probabilistic visibly pushdown automata. Here we also provide a matching lower bound, establishing EXPTIME-completeness. Finally we prove that deciding bisimilarity of probabilistic one-counter automata, another subclass of pPDA, is PSPACE-complete. Here we use a more specialised argument to obtain optimal complexity bounds.","Title: Three tokens in Herman’s algorithm\nYear: 2012\n\nAbstract: Herman’s algorithm is a synchronous randomized protocol for achieving self-stabilization in a token ring consisting of N processes. The interaction of tokens makes the dynamics of the protocol very difficult to analyze. In this paper we study the distribution of the time to stabilization, assuming that there are three tokens in the initial configuration. We show for arbitrary N and for an arbitrary timeout t that the probability of stabilization within time t is minimized by choosing as the initial three-token configuration the configuration in which the tokens are placed equidistantly on the ring. Our result strengthens a corollary of a theorem of McIver and Morgan (Inf. Process Lett. 94(2): 79–84, 2005), which states that the expected stabilization time is minimized by the equidistant configuration.","Title: On probabilistic parallel programs with process creation and synchronisation\nYear: 2011\n\nAbstract: We initiate the study of probabilistic parallel programs with dynamic process creation and synchronisation. To this end, we introduce probabilistic split-join systems (pSJSs), a model for parallel programs, generalising both probabilistic pushdown systems (a model for sequential probabilistic procedural programs which is equivalent to recursive Markov chains) and stochastic branching processes (a classical mathematical model with applications in various areas such as biology, physics, and language processing). Our pSJS model allows for a possibly recursive spawning of parallel processes; the spawned processes can synchronise and return values. We study the basic performance measures of pSJSs, especially the distribution and expectation of space, work and time. Our results extend and improve previously known results on the subsumed models. We also show how to do performance analysis in practice, and present two case studies illustrating the modelling power of pSJSs.","Title: Bisimilarity of Pushdown Systems is Nonelementary\nYear: 2012\n\nAbstract: Given two pushdown systems, the bisimilarity problem asks whether they are bisimilar. While this problem is known to be decidable our main result states that it is nonelementary, improving EXPTIME-hardness, which was the previously best known lower bound for this problem. Our lower bound result holds for normed pushdown systems as well.","Title: An Extension of Newton's Method to omega -Continuous Semirings\nYear: 2007\n\nAbstract: Fixed point equations x = F (x) over !-continuous semi- rings are a natural mathematical foundation of interprocedural program analysis. Equations over the semiring of the real numbers can be solved numerically using Newton's method. We generalize the method to any !-continuous semiring and show that it converges faster to the least xed point than the Kleene sequence 0; F (0); F (F (0)); : : : We prove that the Newton approximants in the semiring of languages coincide with nite- index approximations studied by several authors in the 1960s. Finally, we apply our results to the analysis of stochastic context-free grammars.","Title: Proving termination of probabilistic programs using patterns\nYear: 2012\n\nAbstract: Proving programs terminating is a fundamental computer science challenge. Recent research has produced powerful tools that can check a wide range of programs for termination. The analog for probabilistic programs, namely termination with probability one (\"almost-sure termination\"), is an equally important property for randomized algorithms and probabilistic protocols. We suggest a novel algorithm for proving almost-sure termination of probabilistic programs. Our algorithm exploits the power of state-of-the-art model checkers and termination provers for nonprobabilistic programs: it calls such tools within a refinement loop and thereby iteratively constructs a \"terminating pattern\", which is a set of terminating runs with probability one. We report on various case studies illustrating the effectiveness of our algorithm. As a further application, our algorithm can improve lower bounds on reachability probabilities.","Title: Interprocedural Dataflow Analysis over Weight Domains with Infinite Descending Chains\nYear: 2009\n\nAbstract: We study generalized fixed-point equations over idempotent semirings and provide an efficient algorithm for the detection whether a sequence of Kleene's iterations stabilizes after a finite number of steps. Previously known approaches considered only bounded semirings where there are no infinite descending chains. The main novelty of our work is that we deal with semirings without the boundedness restriction. Our study is motivated by several applications from interprocedural dataflow analysis. We demonstrate how the reachability problem for weighted pushdown automata can be reduced to solving equations in the framework mentioned above and we describe a few applications to demonstrate its usability.","Title: Model checking stochastic branching processes\nYear: 2012\n\nAbstract: Stochastic branching processes are a classical model for describing random trees, which have applications in numerous fields including biology, physics, and natural language processing. In particular, they have recently been proposed to describe parallel programs with stochastic process creation. In this paper, we consider the problem of model checking stochastic branching process. Given a branching process and a deterministic parity tree automaton, we are interested in computing the probability that the generated random tree is accepted by the automaton. We show that this probability can be compared with any rational number in PSPACE, and with 0 and 1 in polynomial time. In a second part, we suggest a tree extension of the logic PCTL, and develop a PSPACE algorithm for model checking a branching process against a formula of this logic. We also show that the qualitative fragment of this logic can be model checked in polynomial time.","Title: Language equivalence for probabilistic automata\nYear: 2011\n\nAbstract: In this paper, we propose a new randomised algorithm for deciding language equivalence for probabilistic automata. This algorithm\n is based on polynomial identity testing and thus returns an answer with an error probability that can be made arbitrarily\n small. We implemented our algorithm, as well as deterministic algorithms of Tzeng and Doyen et al., optimised for running\n time whilst adequately handling issues of numerical stability. We conducted extensive benchmarking experiments, including\n the verification of randomised anonymity protocols, the outcome of which establishes that the randomised algorithm significantly\n outperforms the deterministic ones in a majority of our test cases. Finally, we also provide fine-grained analytical bounds\n on the complexity of these algorithms, accounting for the differences in performance.","Title: On stabilization in Herman's algorithm\nYear: 2011\n\nAbstract: Herman's algorithm is a synchronous randomized protocol for achieving self-stabilization in a token ring consisting of N processes. The interaction of tokens makes the dynamics of the protocol very difficult to analyze. In this paper we study the expected time to stabilization in terms of the initial configuration. It is straightforward that the algorithm achieves stabilization almost surely from any initial configuration, and it is known that the worst-case expected time to stabilization (with respect to the initial configuration) is Θ(N2). Our first contribution is to give an upper bound of 0.64N2 on the expected stabilization time, improving on previous upper bounds and reducing the gap with the best existing lower bound. We also introduce an asynchronous version of the protocol, showing a similar O(N2) convergence bound in this case. Assuming that errors arise from the corruption of some number k of bits, where k is fixed independently of the size of the ring, we show that the expected time to stabilization is O(N). This reveals a hitherto unknown and highly desirable property of Herman's algorithm: it recovers quickly from bounded errors. We also show that if the initial configuration arises by resetting each bit independently and uniformly at random, then stabilization is significantly faster than in the worst case.","Title: On the complexity of the equivalence problem for probabilistic automata\nYear: 2012\n\nAbstract: Deciding equivalence of probabilistic automata is a key problem for establishing various behavioural and anonymity properties of probabilistic systems. In recent experiments a randomised equivalence test based on polynomial identity testing outperformed deterministic algorithms. In this paper we show that polynomial identity testing yields efficient algorithms for various generalisations of the equivalence problem. First, we provide a randomized NC procedure that also outputs a counterexample trace in case of inequivalence. Second, we consider equivalence of probabilistic cost automata. In these automata transitions are labelled with integer costs and each word is associated with a distribution on costs, corresponding to the cumulative costs of the accepting runs on that word. Two automata are equivalent if they induce the same cost distributions on each input word. We show that equivalence can be checked in randomised polynomial time. Finally we show that the equivalence problem for probabilistic visibly pushdown automata is logspace equivalent to the problem of whether a polynomial represented by an arithmetic circuit is identically zero.","Title: On the Complexity of Equivalence and Minimisation for Q-weighted Automata\nYear: 2013\n\nAbstract: This paper is concerned with the computational complexity of equivalence and minimisation for automata with transition weights in the ring Q of rational numbers. We use polynomial identity testing and the Isolation Lemma to obtain complexity bounds, focussing on the class NC of problems within P solvable in polylogarithmic parallel time. For finite Q-weighted automata, we give a randomised NC procedure that either outputs that two automata are equivalent or returns a word on which they differ. We also give an NC procedure for deciding whether a given automaton is minimal, as well as a randomised NC procedure that minimises an automaton. We consider probabilistic automata with rewards, similar to Markov Decision Processes. For these automata we consider two notions of equivalence: expectation equivalence and distribution equivalence. The former requires that two automata have the same expected reward on each input word, while the latter requires that each input word induce the same distribution on rewards in each automaton. For both notions we give algorithms for deciding equivalence by reduction to equivalence of Q-weighted automata. Finally we show that the equivalence problem for Q-weighted visibly pushdown automata is logspace equivalent to the polynomial identity testing problem.","Title: Computing Least Fixed Points of Probabilistic Systems of Polynomials\nYear: 2009\n\nAbstract: We study systems of equations of the form X-1 = f(1)(X-1,..., X-n),..., X-n = f(n)(X-1,..., X-n) where each f(i) is a polynomial with nonnegative coefficients that add up to 1. The least nonnegative solution, say mu, of such equation systems is central to problems from various areas, like physics, biology, computational linguistics and probabilistic program verification. We give a simple and strongly polynomial algorithm to decide whether mu = (1,..., 1) holds. Furthermore, we present an algorithm that computes reliable sequences of lower and upper bounds on mu, converging linearly to mu. Our algorithm has these features despite using inexact arithmetic for efficiency. We report on experiments that show the performance of our algorithms.","Title: Stabilization of Branching Queueing Networks\nYear: 2011\n\nAbstract: Queueing networks are gaining attraction for the performance analysis of parallel computer systems. A Jackson network is a set of interconnected servers, where the completion of a job at server i may result in the creation of a new job for server j. We propose to extend Jackson networks by \"branching\" and by \"control\" features. Both extensions are new and substantially expand the modelling power of Jackson networks. On the other hand, the extensions raise computational questions, particularly concerning the stability of the networks, i.e, the ergodicity of the underlying Markov chain. We show for our extended model that it is decidable in polynomial time if there exists a controller that achieves stability. Moreover, if such a controller exists, one can efficiently compute a static randomized controller which stabilizes the network in a very strong sense; in particular, all moments of the queue sizes are finite.","Title: Parikh's theorem: A simple and direct automaton construction\nYear: 2011\n\nAbstract: Parikh's theorem states that the Parikh image of a context-free language is semilinear or, equivalently, that every context-free language has the same Parikh image as some regular language. We present a very simple construction that, given a context-free grammar, produces a finite automaton recognizing such a regular language.","Title: On fixed point equations over commutative semirings\nYear: 2007\n\nAbstract: Fixed point equations x = f (x) over ω-continuous semirings can be seen as the mathematical foundation of interprocedural program analysis. The sequence 0, f (0), f2(0), . . . converges to the least fixed point µf . The convergence can be accelerated if the underlying semiring is commutative. We show that accelerations in the literature, namely Newton's method for the arithmetic semiring [4] and an acceleration for commutative Kleene algebras due to Hopkins and Kozen [5], are instances of a general algorithm for arbitrary commutative ω-continuous semirings. In a second contribution, we improve the O(3n) bound of [5] and show that their acceleration reaches µf after n iterations, where n is the number of equations. Finally, we apply the Hopkins-Kozen acceleration to itself and study the resulting hierarchy of increasingly fast accelerations.","Title: On the Memory Consumption of Probabilistic Pushdown Automata\nYear: 2009\n\nAbstract: We investigate the problem of evaluating memory consumption for systems modelled by probabilistic pushdown automata (pPDA). The space needed by a run of a pPDA is the maxi- mal height reached by the stack during the run. The problem is motivated by the investigation of depth-first computations that play an important role for space-efficient schedulings of multithreaded programs. We study the computation of both the distribution of the memory consumption and its expectation. For the distribution, we show that a naive method incurs an exponential blow-up, and that it can be avoided using linear equation systems. We also suggest a possibly even faster approximation method. Given # > 0, these methods allow to compute bounds on the memory consumption that are exceeded with a probability of at most #. For the expected memory consumption, we show that whether it is infinite can be decided in polyno- mial time for stateless pPDA (pBPA) and in polynomial space for pPDA. We also provide an iterative method for approximating the expectation. We show how to compute error bounds of our approx- imation method and analyze its convergence speed. We prove that our method converges linearly, i.e., the number of accurate bits of the approximation is a linear function of the number of iterations.","Title: A strongly polynomial algorithm for criticality of branching processes and consistency of stochastic context-free grammars.\nYear: 2013\n\nAbstract: We provide a strongly polynomial algorithm for determining whether a given multi-type branching process is subcritical, critical, or supercritical. The same algorithm also decides consistency of stochastic context-free grammars.","Title: Approximative Methods for Monotone Systems of Min-Max-Polynomial Equations\nYear: 2008\n\nAbstract: On the Internet, there are advertisements (ads) of different kinds: image, text, video and other specially marked objects that are distinct from the underlying content of the page. There is an industry behind the management of such ads, and they face ...","Title: Language equivalence of probabilistic pushdown automata\nYear: 2014\n\nAbstract: We study the language equivalence problem for probabilistic pushdown automata (pPDA) and their subclasses. We show that the problem is interreducible with the multiplicity equivalence problem for context-free grammars, the decidability of which has been open for several decades. Interreducibility also holds for pPDA with one control state. In contrast, for the case of a one-letter input alphabet we show that pPDA language equivalence (and hence multiplicity equivalence of context-free grammars) is in PSPACE and at least as hard as the polynomial identity testing problem.","Title: Minimisation of Multiplicity Tree Automata.\nYear: 2014\n\nAbstract: We consider the problem of minimising the number of states in a multiplicity tree automaton over the field of rational numbers. We give a minimisation algorithm that runs in polynomial time assuming unit-cost arithmetic. We also show that a polynomial bound in the standard Turing model would require a breakthrough in the complexity of polynomial identity testing by proving that the latter problem is logspace equivalent to the decision version of minimisation. The developed techniques also improve the state of the art in multiplicity word automata: we give an NC algorithm for minimising multiplicity word automata. Finally, we consider the minimal consistency problem: does there exist an automaton with n states that is consistent with a given finite sample of weight-labelled words or trees? We show that this decision problem is complete for the existential theory of the rationals, both for words and for trees of a fixed alphabet rank.","Title: Proving the Herman-Protocol Conjecture.\nYear: 2015\n\nAbstract: Herman's self-stabilisation algorithm, introduced 25 years ago, is a well-studied synchronous randomised protocol for enabling a ring of $N$ processes collectively holding any odd number of tokens to reach a stable state in which a single token remains. Determining the worst-case expected time to stabilisation is the central outstanding open problem about this protocol. It is known that there is a constant $h$ such that any initial configuration has expected stabilisation time at most $h N^2$. Ten years ago, McIver and Morgan established a lower bound of $4/27 \\approx 0.148$ for $h$, achieved with three equally-spaced tokens, and conjectured this to be the optimal value of $h$. A series of papers over the last decade gradually reduced the upper bound on $h$, with the present record (achieved in 2014) standing at approximately $0.156$. In this paper, we prove McIver and Morgan's conjecture and establish that $h = 4/27$ is indeed optimal.","Title: Tree Buffers.\nYear: 2015\n\nAbstract: In runtime verification, the central problem is to decide if a given program execution violates a given property. In online runtime verification, a monitor observes a program's execution as it happens. If the program being observed has hard real-time constraints, then the monitor inherits them. In the presence of hard real-time constraints it becomes a challenge to maintain enough information to produce error traces, should a property violation be observed. In this paper we introduce a data structure, called tree buffer, that solves this problem in the context of automata-based monitors: If the monitor itself respects hard real-time constraints, then enriching it by tree buffers makes it possible to provide error traces, which are essential for diagnosing defects. We show that tree buffers are also useful in other application domains. For example, they can be used to implement functionality of capturing groups in regular expressions. We prove optimal asymptotic bounds for our data structure, and validate them using empirical data from two sources: regular expression searching through Wikipedia, and runtime verification of execution traces obtained from the DaCapo test suite.","Title: The Odds of Staying on Budget.\nYear: 2014\n\nAbstract: Given Markov chains and Markov decision processes MDPs whose transitions are labelled with non-negative integer costs, we study the computational complexity of deciding whether the probability of paths whose accumulated cost satisfies a Boolean combination of inequalities exceeds a given threshold. For acyclic Markov chains, we show that this problem is PP-complete, whereas it is hard for the PosSLP problem and in PSpace for general Markov chains. Moreover, for acyclic and general MDPs, we prove PSpace- and EXP-completeness, respectively. Our results have direct implications on the complexity of computing reward quantiles in succinctly represented stochastic systems.","Title: Bisimilarity of Pushdown Automata is Nonelementary\nYear: 2013\n\nAbstract: Given two pushdown automata, the bisimilarity problem asks whether the infinite transition systems they induce are bisimilar. While this problem is known to be decidable our main result states that it is nonelementary, improving EXPTIME-hardness, which was the best previously known lower bound for this problem. Our lower bound result holds for normed pushdown automata as well.","Title: The complexity of the Kth largest subset problem and related problems.\nYear: 2016\n\nAbstract: •The Kth largest subset problem is shown PP-complete.•The Kth largest m-tuple problem is shown PP-complete.•Further problems whose complexity was open are also shown PP-complete.","Title: Efficient Quantile Computation in Markov Chains via Counting Problems for Parikh Images.\nYear: 2016\n\nAbstract: A cost Markov chain is a Markov chain whose transitions are labelled with non-negative integer costs. A fundamental problem on this model, with applications in the verification of stochastic systems, is to compute information about the distribution of the total cost accumulated in a run. This includes the probability of large total costs, the median cost, and other quantiles. While expectations can be computed in polynomial time, previous work has demonstrated that the computation of cost quantiles is harder but can be done in PSPACE. In this paper we show that cost quantiles in cost Markov chains can be computed in the counting hierarchy, thus providing evidence that computing those quantiles is likely not PSPACE-hard. We obtain this result by exhibiting a tight link to a problem in formal language theory: counting the number of words that are both accepted by a given automaton and have a given Parikh image. Motivated by this link, we comprehensively investigate the complexity of the latter problem. Among other techniques, we rely on the so-called BEST theorem for efficiently computing the number of Eulerian circuits in a directed graph.","Title: Trace Refinement in Labelled Markov Decision Processes.\nYear: 2016\n\nAbstract: Given two labelled Markov decision processes (MDPs), the trace-refinement problem asks whether for all strategies of the first MDP there exists a strategy of the second MDP such that the induced labelled Markov chains are trace-equivalent. We show that this problem is decidable in polynomial time if the second MDP is a Markov chain. The algorithm is based on new results on a particular notion of bisimulation between distributions over the states. However, we show that the general trace-refinement problem is undecidable, even if the first MDP is a Markov chain. Decidability of those problems was stated as open in 2008. We further study the decidability and complexity of the trace-refinement problem provided that the strategies are restricted to be memoryless.","Title: Nonnegative Matrix Factorization Requires Irrationality\nYear: 2016\n\nAbstract: Nonnegative matrix factorization (NMF) is the problem of decomposing a given nonnegative is n x m so matrix M into a product of a nonnegative is n x d matrix W and a nonnegative d x m to matrix H. A longstanding open question, posed by Cohen and Rothblum in 1993, is whether a rational matrix M always has an NMF of minimal inner dimension d whose factors W and H are also rational. We answer this question negatively, by exhibiting a matrix for which W and H require irrational entries.","Title: On rationality of nonnegative matrix factorization\nYear: 2016\n\nAbstract: Nonnegative matrix factorization (NMF) is the problem of decomposing a given nonnegative n × m matrix M into a product of a nonnegative n × d matrix W and a nonnegative d × m matrix H. NMF has a wide variety of applications, including bioinformatics, chemometrics, communication complexity, machine learning, polyhedral combinatorics, among many others. A longstanding open question, posed by Cohen and Rothblum in 1993, is whether every rational matrix M has an NMF with minimal d whose factors W and H are also rational. We answer this question negatively, by exhibiting a matrix M for which W and H require irrational entries. As an application of this result, we show that state minimization of labeled Markov chains can require the introduction of irrational transition probabilities. We complement these irrationality results with an NP-complete version of NMF for which rational numbers suffice.","Title: Markov Chains And Unambiguous Buchi Automata\nYear: 2016\n\nAbstract: Unambiguous automata, i.e., nondeterministic automata with the restriction of having at most one accepting run over a word, have the potential to be used instead of deterministic automata in settings where nondeterministic automata can not be applied in general. In this paper, we provide a polynomially time-bounded algorithm for probabilistic model checking of discrete-time Markov chains against unambiguous Buchi automata specifications and report on our implementation and experiments.","Title: On strong determinacy of countable stochastic games\nYear: 2017\n\nAbstract: We study 2-player turn-based perfect-information stochastic games with countably infinite state space. The players aim at maximizing/minimizing the probability of a given event (i.e., measurable set of infinite plays), such as reachability, Büchi, ω-regular or more general objectives. These games are known to be weakly determined, i.e., they have value. However, strong determinacy of threshold objectives (given by an event ε and a threshold c ∈ [0,1]) was open in many cases: is it always the case that the maximizer or the minimizer has a winning strategy, i.e., one that enforces, against all strategies of the other player, that ε is satisfied with probability ≥ c (resp. <; c)? We show that almost-sure objectives (where c = 1) are strongly determined. This vastly generalizes a previous result on finite games with almost-sure tail objectives. On the other hand we show that ≥ 1/2 (co-)Biichi objectives are not strongly determined, not even if the game is finitely branching. Moreover, for almost-sure reachability and almost-sure Biichi objectives in finitely branching games, we strengthen strong determinacy by showing that one of the players must have a memory less deterministic (MD) winning strategy.","Title: Counting Problems for Parikh Images.\nYear: 2017\n\nAbstract: Given finite-state automata (or context-free grammars) A,B over the same alphabet and a Parikh vector p, we study the complexity of deciding whether the number of words in the language of A with Parikh image p is greater than the number of such words in the language of B. Recently, this problem turned out to be tightly related to the cost problem for weighted Markov chains. We classify the complexity according to whether A and B are deterministic, the size of the alphabet, and the encoding of p (binary or unary).","Title: Game Characterization of Probabilistic Bisimilarity, and Applications to Pushdown Automata.\nYear: 2018\n\nAbstract: We study the bisimilarity problem for probabilistic pushdown automata (pPDA) and subclasses thereof. Our definition of pPDA allows both probabilistic and non-deterministic branching, generalising the classical notion of pushdown automata (without epsilon-transitions). We first show a general characterization of probabilistic bisimilarity in terms of two-player games, which naturally reduces checking bisimilarity of probabilistic labelled transition systems to checking bisimilarity of standard (non-deterministic) labelled transition systems. This reduction can be easily implemented in the framework of pPDA, allowing to use known results for standard (non-probabilistic) PDA and their subclasses. A direct use of the reduction incurs an exponential increase of complexity, which does not matter in deriving decidability of bisimilarity for pPDA due to the non-elementary complexity of the problem. In the cases of probabilistic one-counter automata (pOCA), of probabilistic visibly pushdown automata (pvPDA), and of probabilistic basic process algebras (i.e., single-state pPDA) we show that an implicit use of the reduction can avoid the complexity increase; we thus get PSPACE, EXPTIME, and 2-EXPTIME upper bounds, respectively, like for the respective non-probabilistic versions. The bisimilarity problems for OCA and vPDA are known to have matching lower bounds (thus being PSPACE-complete and EXPTIME-complete, respectively); we show that these lower bounds also hold for fully probabilistic versions that do not use non-determinism.","Title: On Computing the Total Variation Distance of Hidden Markov Models.\nYear: 2018\n\nAbstract: We prove results on the decidability and complexity of computing the total variation distance (equivalently, the $L_1$-distance) of hidden Markov models (equivalently, labelled Markov chains). This distance measures the difference between the distributions on words that two hidden Markov models induce. The main results are: (1) it is undecidable whether the distance is greater than a given threshold; (2) approximation is #P-hard and in PSPACE.","Title: On the Complexity of Iterative Tropical Computation with Applications to Markov Decision Processes.\nYear: 2018\n\nAbstract: We study the complexity of evaluating powered functions implemented by straight-line programs (SLPs) over the tropical semiring (i.e., with max and + operations). In this problem, a given (max,+)-SLP with the same number of input and output wires is composed with H copies of itself, where H is given in binary. The problem of evaluating powered SLPs is intimately connected with iterative arithmetic computations that arise in algorithmic decision making and operations research. Specifically, it is essentially equivalent to finding optimal strategies in finite-horizon Markov Decision Processes (MDPs). We show that evaluating powered SLPs and finding optimal strategies in finite-horizon MDPs are both EXPTIME-complete problems. This resolves an open problem that goes back to the seminal 1987 paper on the complexity of MDPs by Papadimitriou and Tsitsiklis.","Title: Selective Monitoring.\nYear: 2018\n\nAbstract: We study selective monitors for labelled Markov chains. Monitors observe the outputs that are generated by a Markov chain during its run, with the goal of identifying runs as correct or faulty. A monitor is selective if it skips observations in order to reduce monitoring overhead. We are interested in monitors that minimize the expected number of observations. We establish an undecidability result for selectively monitoring general Markov chains. On the other hand, we show for non-hidden Markov chains (where any output identifies the state the Markov chain is in) that simple optimal monitors exist and can be computed efficiently, based on DFA language equivalence. These monitors do not depend on the precise transition probabilities in the Markov chain. We report on experiments where we compute these monitors for several open-source Java projects.","Title: On the Complexity of Value Iteration.\nYear: 2019\n\nAbstract: Value iteration is a fundamental algorithm for solving Markov Decision Processes (MDPs). It computes the maximal $n$-step payoff by iterating $n$ times a recurrence equation which is naturally associated to the MDP. At the same time, value iteration provides a policy for the MDP that is optimal on a given finite horizon $n$. In this paper, we settle the computational complexity of value iteration. We show that, given a horizon $n$ in binary and an MDP, computing an optimal policy is EXP-complete, thus resolving an open problem that goes back to the seminal 1987 paper on the complexity of MDPs by Papadimitriou and Tsitsiklis. As a stepping stone, we show that it is EXP-complete to compute the $n$-fold iteration (with $n$ in binary) of a function given by a straight-line program over the integers with $max$ and $+$ as operators.","Title: On Semigroups of Two-Dimensional Upper-Triangular Integer Matrices.\nYear: 2019\n\nAbstract: We analyze algorithmic problems in finitely generated semigroups of two-dimensional upper-triangular integer matrices. These semigroup problems are tightly connected with problems about compositions of affine functions over one variable. Building on a variety of techniques from recent years, we obtain a number of complexity results, including NP-completeness of the mortality problem for matrices with determinants +1 and 0.","Title: Efficient Analysis of Unambiguous Automata Using Matrix Semigroup Techniques.\nYear: 2019\n\nAbstract: We introduce a novel technique to analyse unambiguous B\\\"uchi automata quantitatively, and apply this to the model checking problem. It is based on linear-algebra arguments that originate from the analysis of matrix semigroups with constant spectral radius. This method can replace a combinatorial procedure that dominates the computational complexity of the existing procedure by Baier et al. We analyse the complexity in detail, showing that, in terms of the set $Q$ of states of the automaton, the new algorithm runs in time $O(|Q|^4)$, improving on an efficient implementation of the combinatorial algorithm by a factor of $|Q|$.","Title: Linear-Time Model Checking Branching Processes.\nYear: 2021\n\nAbstract: (Multi-type) branching processes are a natural and well-studied model for generating random infinite trees. Branching processes feature both nondeterministic and probabilistic branching, generalizing both transition systems and Markov chains (but not generally Markov decision processes). We study the complexity of model checking branching processes against linear-time omega-regular specifications: is it the case almost surely that every branch of a tree randomly generated by the branching process satisfies the omega-regular specification? The main result is that for LTL specifications this problem is in PSPACE, subsuming classical results for transition systems and Markov chains, respectively. The underlying general model-checking algorithm is based on the automata-theoretic approach, using unambiguous B\\\"uchi automata.","Title: An ETH-Tight Algorithm for Multi-Team Formation.\nYear: 2021\n\nAbstract: We propose polynomial-time algorithms to minimise labelled Markov chains whose transition probabilities are not known exactly, have been perturbed, or can only be obtained by sampling. Our algorithms are based on a new notion of an approximate bisimulation quotient, obtained by lumping together states that are exactly bisimilar in a slightly perturbed system. We present experiments that show that our algorithms are able to recover the structure of the bisimulation quotient of the unperturbed system."],"string":"[\n \"Title: Newtonian program analysis\\nYear: 2010\\n\\nAbstract: This article presents a novel generic technique for solving dataflow equations in interprocedural dataflow analysis. The technique is obtained by generalizing Newton's method for computing a zero of a differentiable function to ω-continuous semirings. Complete semilattices, the common program analysis framework, are a special class of ω-continuous semirings. We show that our generalized method always converges to the solution, and requires at most as many iterations as current methods based on Kleene's fixed-point theorem. We also show that, contrary to Kleene's method, Newton's method always terminates for arbitrary idempotent and commutative semirings. More precisely, in the latter setting the number of iterations required to solve a system of n equations is at most n.\",\n \"Title: Newton's Method for omega-Continuous Semirings\\nYear: 2008\\n\\nAbstract: Fixed point equations X = f (X) over omega-continuous semirings are a natural mathematical foundation of interprocedural program analysis. Generic algorithms for solving these equations are based on Kleene's theorem, which states that the sequence 0, f (0), f (f (0)),... converges to the least fixed point. However, this approach is often inefficient. We report on recent work in which we extend Newton's method, the well-known technique from numerical mathematics, to arbitrary omega-continuous semirings, and analyze its convergence speed in the real semiring.\",\n \"Title: Abstraction refinement with craig interpolation and symbolic pushdown systems\\nYear: 2006\\n\\nAbstract: Counterexample-guided abstraction refinement (CEGAR) has proven to be a powerful method for software model-checking. In this paper, we investigate this concept in the context of sequential (possibly recursive) programs whose statements are given as BDDs. We examine how Craig interpolants can be computed efficiently in this case and propose a new, special type of interpolants. Moreover, we show how to treat multiple counterexamples in one refinement cycle. We have implemented this approach within the model-checker Moped and report on experiments.\",\n \"Title: Zero-reachability in probabilistic multi-counter automata\\nYear: 2014\\n\\nAbstract: We study the qualitative and quantitative zero-reachability problem in probabilistic multi-counter systems. We identify the undecidable variants of the problems, and then we concentrate on the remaining two cases. In the first case, when we are interested in the probability of all runs that visit zero in some counter, we show that the qualitative zero-reachability is decidable in time which is polynomial in the size of a given pMC and doubly exponential in the number of counters. Further, we show that the probability of all zero-reaching runs can be effectively approximated up to an arbitrarily small given error ε > 0 in time which is polynomial in log(ε), exponential in the size of a given pMC, and doubly exponential in the number of counters. In the second case, we are interested in the probability of all runs that visit zero in some counter different from the last counter. Here we show that the qualitative zero-reachability is decidable and SquareRootSum-hard, and the probability of all zero-reaching runs can be effectively approximated up to an arbitrarily small given error ε > 0 (these results apply to pMC satisfying a suitable technical condition that can be verified in polynomial time). The proof techniques invented in the second case allow to construct counterexamples for some classical results about ergodicity in stochastic Petri nets.\",\n \"Title: Analyzing probabilistic pushdown automata\\nYear: 2013\\n\\nAbstract: The paper gives a summary of the existing results about algorithmic analysis of probabilistic pushdown automata and their subclasses.\",\n \"Title: Analysis of Probabilistic Basic Parallel Processes.\\nYear: 2014\\n\\nAbstract: Basic Parallel Processes (BPPs) are a well-known subclass of Petri Nets. They are the simplest common model of concurrent programs that allows unbounded spawning of processes. In the probabilistic version of BPPs, every process generates other processes according to a probability distribution. We study the decidability and complexity of fundamental qualitative problems over probabilistic BPPs -- in particular reachability with probability 1 of different classes of target sets (e.g. upward-closed sets). Our results concern both the Markov-chain model, where processes are scheduled randomly, and the MDP model, where processes are picked by a scheduler.\",\n \"Title: Solving Monotone Polynomial Equations\\nYear: 2008\\n\\nAbstract: We survey some recent results on iterative methods for approximating the least solution of a system of monotone fixed-point polynomial equations.\",\n \"Title: Computing the Least Fixed Point of Positive Polynomial Systems\\nYear: 2010\\n\\nAbstract: We consider equation systems of the form $X_1=f_1(X_1,\\\\dots,X_n)$, $\\\\dots$, $X_n = f_n(X_1,\\\\dots,X_n)$, where $f_1,\\\\dots,f_n$ are polynomials with positive real coefficients. In vector form we denote such an equation system by ${\\\\bf X}={\\\\bf f}({\\\\bf X})$ and call ${\\\\bf f}$ a system of positive polynomials (SPP). Equation systems of this kind appear naturally in the analysis of stochastic models like stochastic context-free grammars (with numerous applications to natural language processing and computational biology), probabilistic programs with procedures, web-surfing models with back buttons, and branching processes. The least nonnegative solution $\\\\mu{\\\\bf f}$ of an SPP equation ${\\\\bf X}={\\\\bf f}({\\\\bf X})$ is of central interest for these models. Etessami and Yannakakis [J. ACM, 56 (2009), pp. 1-66] have suggested a particular version of Newton's method to approximate $\\\\mu{\\\\bf f}$. We extend a result of Etessami and Yannakakis and show that Newton's method starting at ${\\\\bf 0}$ always converges to $\\\\mu{\\\\bf f}$. We obtain lower bounds on the convergence speed of the method. For so-called strongly connected SPPs we prove the existence of a threshold $k_{{\\\\bf f}}\\\\in\\\\mathbb{N}$ such that for every $i\\\\geq0$ the $(k_{{\\\\bf f}}+i)$th iteration of Newton's method has at least $i$ valid bits of $\\\\mu{\\\\bf f}$. The proof yields an explicit bound for $k_{{\\\\bf f}}$ depending only on syntactic parameters of ${\\\\bf f}$. We further show that for arbitrary SPP equations, Newton's method still converges linearly: there exists a threshold $k_{{\\\\bf f}}$ and an $\\\\alpha_{{\\\\bf f}}0$ such that for every $i\\\\geq0$ the $(k_{{\\\\bf f}}+\\\\alpha_{{\\\\bf f}}\\\\cdot i)$th iteration of Newton's method has at least $i$ valid bits of $\\\\mu{\\\\bf f}$. The proof yields an explicit bound for $\\\\alpha_{{\\\\bf f}}$; the bound is exponential in the number of equations in ${\\\\bf X}={\\\\bf f}({\\\\bf X})$, but we also show that it is essentially optimal. The proof does not yield any bound for $k_{{\\\\bf f}}$, but only proves its existence. Constructing a bound for $k_{{\\\\bf f}}$ is still an open problem. Finally, we also provide a geometric interpretation of Newton's method for SPPs.\",\n \"Title: APEX: an analyzer for open probabilistic programs\\nYear: 2012\\n\\nAbstract: We present APEX, a tool for analysing probabilistic programs that are open, i.e. where variables or even functions can be left unspecified. APEX transforms a program into an automaton that captures the program's probabilistic behaviour under all instantiations of the unspecified components. The translation is compositional and effectively leverages state reduction techniques. APEX can then further analyse the produced automata; in particular, it can check two automata for equivalence which translates to equivalence of the corresponding programs under all environments. In this way, APEX can verify a broad range of anonymity and termination properties of randomised protocols and other open programs, sometimes with an exponential speed-up over competing state-of-the-art approaches.\",\n \"Title: Runtime analysis of probabilistic programs with unbounded recursion\\nYear: 2010\\n\\nAbstract: We study the runtime in probabilistic programs with unbounded recursion. As underlying formal model for such programs we use probabilistic pushdown automata (pPDAs) which exactly correspond to recursive Markov chains. We show that every pPDA can be transformed into a stateless pPDA (called \\\\\\\"pBPA\\\\\\\") whose runtime and further properties are closely related to those of the original pPDA. This result substantially simplifies the analysis of runtime and other pPDA properties. We prove that for every pPDA the probability of performing a long run decreases exponentially in the length of the run, if and only if the expected runtime in the pPDA is finite. If the expectation is infinite, then the probability decreases \\\\\\\"polynomially\\\\\\\". We show that these bounds are asymptotically tight. Our tail bounds on the runtime are generic, i.e., applicable to any probabilistic program with unbounded recursion. We study the runtime in probabilistic programs with unbounded recursion.A transformation to make probabilistic PDAs stateless is proposed.We give bounds on the probability of long runs of a recursive program.\",\n \"Title: Convergence Thresholds of Newton's Method for Monotone Polynomial Equations\\nYear: 2008\\n\\nAbstract: Monotone systems of polynomial equations (MSPEs) axe systems of fixed-point equations X-1 = f(1) (X-1, . . . , X-n), . . . , X-n = f (X-1, . . . , X-n) where each fi is a polynomial with positive real coefficients. The question of computing the least non-negative solution of a given MSPE X = f(X) arises naturally in the analysis of stochastic models such as stochastic context-free grammars, probabilistic pushdown automata, and barkbutton processes. Etessami! and Yannakakis have recently adapted Newton's iterative method to MSPEs. In a previous paper we have proved the existence of a threshold k(f) for strongly connected MSPEs, such that after kf iterations of Newton's method each new iteration computes at least 1 new bit of the solution. However, the proof was purely existential. In this paper we give an upper bound for kf as a function of the minimal component of the least fixed-point mu f of f(X). Using this result we show that kf is at most single exponential resp. linear for strongly connected MSPEs derived from probabilistic pushdown automata resp. from back-button processes. Further, we prove the existence of a threshold for arbitrary MSPEs after which each new iteration computes at least 1/w2(h) new bits of the solution, where w and h are the width and height of the DAG of strongly connected components.\",\n \"Title: Efficient Analysis of Probabilistic Programs with an Unbounded Counter.\\nYear: 2011\\n\\nAbstract: We show that a subclass of infinite-state probabilistic programs that can be modeled by probabilistic one-counter automata (pOC) admits an efficient quantitative analysis. In particular, we show that the expected termination time can be approximated up to an arbitrarily small relative error with polynomially many arithmetic operations, and the same holds for the probability of all runs that satisfy a given ω-regular property. Further, our results establish a powerful link between pOC and martingale theory, which leads to fundamental observations about quantitative properties of runs in pOC. In particular, we provide a \\\"divergence gap theorem\\\", which bounds a positive non-termination probability in pOC away from zero.\",\n \"Title: Derivation Tree Analysis for Accelerated Fixed-Point Computation\\nYear: 2011\\n\\nAbstract: We show that for several classes of idempotent semirings the least fixed-point of a polynomial system of equations X = f(X) is equal to the least fixed-point of a linear system obtained by \\\"linearizing\\\" the polynomials of f in a certain way. Our proofs rely on derivation tree analysis, a proof prin ciple that combines methods from algebra, calculus, and formal language theory, and was first used in (5) to show that Newton's method over commutative and idem po- tent semirings converges in a linear number of steps. Our results lead to efficient generic algorithms for computing the least fixed-point. We use these algor ithms to derive several consequences, including an O(N3) algorithm for computing the throughput of a context-free grammar (obtained by speeding up the O(N4) algo- rithm of (2)), and a generalization of Courcelle's result stating that the do wnward- closed image of a context-free language is regular (3).\",\n \"Title: Stability and Complexity of Minimising Probabilistic Automata.\\nYear: 2014\\n\\nAbstract: We consider the state-minimisation problem for weighted and probabilistic automata. We provide a numerically stable polynomial-time minimisation algorithm for weighted automata, with guaranteed bounds on the numerical error when run with floating-point arithmetic. Our algorithm can also be used for \\\"lossy\\\" minimisation with bounded error. We show an application in image compression. In the second part of the paper we study the complexity of the minimisation problem for probabilistic automata. We prove that the problem is NP-hard and in PSPACE, improving a recent EXPTIME-result.\",\n \"Title: On the convergence of Newton's method for monotone systems of polynomial equations\\nYear: 2007\\n\\nAbstract: Monotone systems of polynomial equations (MSPEs) are systems of fixed-point equations X1 = f1(X1, ..., Xn), ..., Xn = fn(X1, ..., Xn) where each fi is a polynomial with positive real coefficients. The question of computing the least non-negative solution of a given MSPE X = f(X) arises naturally in the analysis of stochastic context-free grammars, recursive Markov chains, and probabilistic pushdown automata. While the Kleene sequence f(0), f(f(0)), ... always converges to the least solution mu.f, if it exists, the number of iterations needed to compute the first i bits of mu.f may grow exponentially in i.Etessami and Yannakakis have recently adapted Newton's iterative method to MSPEs and proved that the Newton sequence converges at least as fast as the Kleene sequence and exponentially faster in many cases.They conjecture that, given an MSPE of size m, the number of Newton iterations needed to obtain i accurate bits of mu.f grows polynomially in i and m. In this paper we show that the number of iterations grows linearly in i for strongly connected MSPEs and may grow exponentially in m for general MSPEs.\",\n \"Title: On the total variation distance of labelled Markov chains\\nYear: 2014\\n\\nAbstract: Labelled Markov chains (LMCs) are widely used in probabilistic verification, speech recognition, computational biology, and many other fields. Checking two LMCs for equivalence is a classical problem subject to extensive studies, while the total variation distance provides a natural measure for the \\\\\\\"inequivalence\\\\\\\" of two LMCs: it is the maximum difference between probabilities that the LMCs assign to the same event. In this paper we develop a theory of the total variation distance between two LMCs, with emphasis on the algorithmic aspects: (1) we provide a polynomial-time algorithm for determining whether two LMCs have distance 1, i.e., whether they can almost always be distinguished; (2) we provide an algorithm for approximating the distance with arbitrary precision; and (3) we show that the threshold problem, i.e., whether the distance exceeds a given threshold, is NP-hard and hard for the square-root-sum problem. We also make a connection between the total variation distance and Bernoulli convolutions.\",\n \"Title: BPA bisimilarity is EXPTIME-hard\\nYear: 2013\\n\\nAbstract: Given a basic process algebra (BPA) and two stack symbols, the BPA bisimilarity problem asks whether the two stack symbols are bisimilar. We show that this problem is EXPTIME-hard.\",\n \"Title: Space-efficient scheduling of stochastically generated tasks\\nYear: 2010\\n\\nAbstract: We study the problem of scheduling tasks for execution by a processor when the tasks can stochastically generate new tasks. Tasks can be of different types, and each type has a fixed, known probability of generating other tasks. We present results on the random variable S^@s modeling the maximal space needed by the processor to store the currently active tasks when acting under the scheduler @s. We obtain tail bounds for the distribution of S^@s for both offline and online schedulers, and investigate the expected value E[S^@s].\",\n \"Title: Bisimilarity of Probabilistic Pushdown Automata\\nYear: 2012\\n\\nAbstract: We study the bisimilarity problem for probabilistic pushdown automata (pPDA) and subclasses thereof. Our definition of pPDA allows both probabilistic and non-deterministic branching, generalising the classical notion of pushdown automata (without epsilon-transitions). Our first contribution is a general construction that reduces checking bisimilarity of probabilistic transition systems to checking bisimilarity of non-deterministic transition systems. This construction directly yields decidability of bisimilarity for pPDA, as well as an elementary upper bound for the bisimilarity problem on the subclass of probabilistic basic process algebras, i.e., single-state pPDA. We further show that, with careful analysis, the general reduction can be used to prove an EXPTIME upper bound for bisimilarity of probabilistic visibly pushdown automata. Here we also provide a matching lower bound, establishing EXPTIME-completeness. Finally we prove that deciding bisimilarity of probabilistic one-counter automata, another subclass of pPDA, is PSPACE-complete. Here we use a more specialised argument to obtain optimal complexity bounds.\",\n \"Title: Three tokens in Herman’s algorithm\\nYear: 2012\\n\\nAbstract: Herman’s algorithm is a synchronous randomized protocol for achieving self-stabilization in a token ring consisting of N processes. The interaction of tokens makes the dynamics of the protocol very difficult to analyze. In this paper we study the distribution of the time to stabilization, assuming that there are three tokens in the initial configuration. We show for arbitrary N and for an arbitrary timeout t that the probability of stabilization within time t is minimized by choosing as the initial three-token configuration the configuration in which the tokens are placed equidistantly on the ring. Our result strengthens a corollary of a theorem of McIver and Morgan (Inf. Process Lett. 94(2): 79–84, 2005), which states that the expected stabilization time is minimized by the equidistant configuration.\",\n \"Title: On probabilistic parallel programs with process creation and synchronisation\\nYear: 2011\\n\\nAbstract: We initiate the study of probabilistic parallel programs with dynamic process creation and synchronisation. To this end, we introduce probabilistic split-join systems (pSJSs), a model for parallel programs, generalising both probabilistic pushdown systems (a model for sequential probabilistic procedural programs which is equivalent to recursive Markov chains) and stochastic branching processes (a classical mathematical model with applications in various areas such as biology, physics, and language processing). Our pSJS model allows for a possibly recursive spawning of parallel processes; the spawned processes can synchronise and return values. We study the basic performance measures of pSJSs, especially the distribution and expectation of space, work and time. Our results extend and improve previously known results on the subsumed models. We also show how to do performance analysis in practice, and present two case studies illustrating the modelling power of pSJSs.\",\n \"Title: Bisimilarity of Pushdown Systems is Nonelementary\\nYear: 2012\\n\\nAbstract: Given two pushdown systems, the bisimilarity problem asks whether they are bisimilar. While this problem is known to be decidable our main result states that it is nonelementary, improving EXPTIME-hardness, which was the previously best known lower bound for this problem. Our lower bound result holds for normed pushdown systems as well.\",\n \"Title: An Extension of Newton's Method to omega -Continuous Semirings\\nYear: 2007\\n\\nAbstract: Fixed point equations x = F (x) over !-continuous semi- rings are a natural mathematical foundation of interprocedural program analysis. Equations over the semiring of the real numbers can be solved numerically using Newton's method. We generalize the method to any !-continuous semiring and show that it converges faster to the least xed point than the Kleene sequence 0; F (0); F (F (0)); : : : We prove that the Newton approximants in the semiring of languages coincide with nite- index approximations studied by several authors in the 1960s. Finally, we apply our results to the analysis of stochastic context-free grammars.\",\n \"Title: Proving termination of probabilistic programs using patterns\\nYear: 2012\\n\\nAbstract: Proving programs terminating is a fundamental computer science challenge. Recent research has produced powerful tools that can check a wide range of programs for termination. The analog for probabilistic programs, namely termination with probability one (\\\"almost-sure termination\\\"), is an equally important property for randomized algorithms and probabilistic protocols. We suggest a novel algorithm for proving almost-sure termination of probabilistic programs. Our algorithm exploits the power of state-of-the-art model checkers and termination provers for nonprobabilistic programs: it calls such tools within a refinement loop and thereby iteratively constructs a \\\"terminating pattern\\\", which is a set of terminating runs with probability one. We report on various case studies illustrating the effectiveness of our algorithm. As a further application, our algorithm can improve lower bounds on reachability probabilities.\",\n \"Title: Interprocedural Dataflow Analysis over Weight Domains with Infinite Descending Chains\\nYear: 2009\\n\\nAbstract: We study generalized fixed-point equations over idempotent semirings and provide an efficient algorithm for the detection whether a sequence of Kleene's iterations stabilizes after a finite number of steps. Previously known approaches considered only bounded semirings where there are no infinite descending chains. The main novelty of our work is that we deal with semirings without the boundedness restriction. Our study is motivated by several applications from interprocedural dataflow analysis. We demonstrate how the reachability problem for weighted pushdown automata can be reduced to solving equations in the framework mentioned above and we describe a few applications to demonstrate its usability.\",\n \"Title: Model checking stochastic branching processes\\nYear: 2012\\n\\nAbstract: Stochastic branching processes are a classical model for describing random trees, which have applications in numerous fields including biology, physics, and natural language processing. In particular, they have recently been proposed to describe parallel programs with stochastic process creation. In this paper, we consider the problem of model checking stochastic branching process. Given a branching process and a deterministic parity tree automaton, we are interested in computing the probability that the generated random tree is accepted by the automaton. We show that this probability can be compared with any rational number in PSPACE, and with 0 and 1 in polynomial time. In a second part, we suggest a tree extension of the logic PCTL, and develop a PSPACE algorithm for model checking a branching process against a formula of this logic. We also show that the qualitative fragment of this logic can be model checked in polynomial time.\",\n \"Title: Language equivalence for probabilistic automata\\nYear: 2011\\n\\nAbstract: In this paper, we propose a new randomised algorithm for deciding language equivalence for probabilistic automata. This algorithm\\n is based on polynomial identity testing and thus returns an answer with an error probability that can be made arbitrarily\\n small. We implemented our algorithm, as well as deterministic algorithms of Tzeng and Doyen et al., optimised for running\\n time whilst adequately handling issues of numerical stability. We conducted extensive benchmarking experiments, including\\n the verification of randomised anonymity protocols, the outcome of which establishes that the randomised algorithm significantly\\n outperforms the deterministic ones in a majority of our test cases. Finally, we also provide fine-grained analytical bounds\\n on the complexity of these algorithms, accounting for the differences in performance.\",\n \"Title: On stabilization in Herman's algorithm\\nYear: 2011\\n\\nAbstract: Herman's algorithm is a synchronous randomized protocol for achieving self-stabilization in a token ring consisting of N processes. The interaction of tokens makes the dynamics of the protocol very difficult to analyze. In this paper we study the expected time to stabilization in terms of the initial configuration. It is straightforward that the algorithm achieves stabilization almost surely from any initial configuration, and it is known that the worst-case expected time to stabilization (with respect to the initial configuration) is Θ(N2). Our first contribution is to give an upper bound of 0.64N2 on the expected stabilization time, improving on previous upper bounds and reducing the gap with the best existing lower bound. We also introduce an asynchronous version of the protocol, showing a similar O(N2) convergence bound in this case. Assuming that errors arise from the corruption of some number k of bits, where k is fixed independently of the size of the ring, we show that the expected time to stabilization is O(N). This reveals a hitherto unknown and highly desirable property of Herman's algorithm: it recovers quickly from bounded errors. We also show that if the initial configuration arises by resetting each bit independently and uniformly at random, then stabilization is significantly faster than in the worst case.\",\n \"Title: On the complexity of the equivalence problem for probabilistic automata\\nYear: 2012\\n\\nAbstract: Deciding equivalence of probabilistic automata is a key problem for establishing various behavioural and anonymity properties of probabilistic systems. In recent experiments a randomised equivalence test based on polynomial identity testing outperformed deterministic algorithms. In this paper we show that polynomial identity testing yields efficient algorithms for various generalisations of the equivalence problem. First, we provide a randomized NC procedure that also outputs a counterexample trace in case of inequivalence. Second, we consider equivalence of probabilistic cost automata. In these automata transitions are labelled with integer costs and each word is associated with a distribution on costs, corresponding to the cumulative costs of the accepting runs on that word. Two automata are equivalent if they induce the same cost distributions on each input word. We show that equivalence can be checked in randomised polynomial time. Finally we show that the equivalence problem for probabilistic visibly pushdown automata is logspace equivalent to the problem of whether a polynomial represented by an arithmetic circuit is identically zero.\",\n \"Title: On the Complexity of Equivalence and Minimisation for Q-weighted Automata\\nYear: 2013\\n\\nAbstract: This paper is concerned with the computational complexity of equivalence and minimisation for automata with transition weights in the ring Q of rational numbers. We use polynomial identity testing and the Isolation Lemma to obtain complexity bounds, focussing on the class NC of problems within P solvable in polylogarithmic parallel time. For finite Q-weighted automata, we give a randomised NC procedure that either outputs that two automata are equivalent or returns a word on which they differ. We also give an NC procedure for deciding whether a given automaton is minimal, as well as a randomised NC procedure that minimises an automaton. We consider probabilistic automata with rewards, similar to Markov Decision Processes. For these automata we consider two notions of equivalence: expectation equivalence and distribution equivalence. The former requires that two automata have the same expected reward on each input word, while the latter requires that each input word induce the same distribution on rewards in each automaton. For both notions we give algorithms for deciding equivalence by reduction to equivalence of Q-weighted automata. Finally we show that the equivalence problem for Q-weighted visibly pushdown automata is logspace equivalent to the polynomial identity testing problem.\",\n \"Title: Computing Least Fixed Points of Probabilistic Systems of Polynomials\\nYear: 2009\\n\\nAbstract: We study systems of equations of the form X-1 = f(1)(X-1,..., X-n),..., X-n = f(n)(X-1,..., X-n) where each f(i) is a polynomial with nonnegative coefficients that add up to 1. The least nonnegative solution, say mu, of such equation systems is central to problems from various areas, like physics, biology, computational linguistics and probabilistic program verification. We give a simple and strongly polynomial algorithm to decide whether mu = (1,..., 1) holds. Furthermore, we present an algorithm that computes reliable sequences of lower and upper bounds on mu, converging linearly to mu. Our algorithm has these features despite using inexact arithmetic for efficiency. We report on experiments that show the performance of our algorithms.\",\n \"Title: Stabilization of Branching Queueing Networks\\nYear: 2011\\n\\nAbstract: Queueing networks are gaining attraction for the performance analysis of parallel computer systems. A Jackson network is a set of interconnected servers, where the completion of a job at server i may result in the creation of a new job for server j. We propose to extend Jackson networks by \\\"branching\\\" and by \\\"control\\\" features. Both extensions are new and substantially expand the modelling power of Jackson networks. On the other hand, the extensions raise computational questions, particularly concerning the stability of the networks, i.e, the ergodicity of the underlying Markov chain. We show for our extended model that it is decidable in polynomial time if there exists a controller that achieves stability. Moreover, if such a controller exists, one can efficiently compute a static randomized controller which stabilizes the network in a very strong sense; in particular, all moments of the queue sizes are finite.\",\n \"Title: Parikh's theorem: A simple and direct automaton construction\\nYear: 2011\\n\\nAbstract: Parikh's theorem states that the Parikh image of a context-free language is semilinear or, equivalently, that every context-free language has the same Parikh image as some regular language. We present a very simple construction that, given a context-free grammar, produces a finite automaton recognizing such a regular language.\",\n \"Title: On fixed point equations over commutative semirings\\nYear: 2007\\n\\nAbstract: Fixed point equations x = f (x) over ω-continuous semirings can be seen as the mathematical foundation of interprocedural program analysis. The sequence 0, f (0), f2(0), . . . converges to the least fixed point µf . The convergence can be accelerated if the underlying semiring is commutative. We show that accelerations in the literature, namely Newton's method for the arithmetic semiring [4] and an acceleration for commutative Kleene algebras due to Hopkins and Kozen [5], are instances of a general algorithm for arbitrary commutative ω-continuous semirings. In a second contribution, we improve the O(3n) bound of [5] and show that their acceleration reaches µf after n iterations, where n is the number of equations. Finally, we apply the Hopkins-Kozen acceleration to itself and study the resulting hierarchy of increasingly fast accelerations.\",\n \"Title: On the Memory Consumption of Probabilistic Pushdown Automata\\nYear: 2009\\n\\nAbstract: We investigate the problem of evaluating memory consumption for systems modelled by probabilistic pushdown automata (pPDA). The space needed by a run of a pPDA is the maxi- mal height reached by the stack during the run. The problem is motivated by the investigation of depth-first computations that play an important role for space-efficient schedulings of multithreaded programs. We study the computation of both the distribution of the memory consumption and its expectation. For the distribution, we show that a naive method incurs an exponential blow-up, and that it can be avoided using linear equation systems. We also suggest a possibly even faster approximation method. Given # > 0, these methods allow to compute bounds on the memory consumption that are exceeded with a probability of at most #. For the expected memory consumption, we show that whether it is infinite can be decided in polyno- mial time for stateless pPDA (pBPA) and in polynomial space for pPDA. We also provide an iterative method for approximating the expectation. We show how to compute error bounds of our approx- imation method and analyze its convergence speed. We prove that our method converges linearly, i.e., the number of accurate bits of the approximation is a linear function of the number of iterations.\",\n \"Title: A strongly polynomial algorithm for criticality of branching processes and consistency of stochastic context-free grammars.\\nYear: 2013\\n\\nAbstract: We provide a strongly polynomial algorithm for determining whether a given multi-type branching process is subcritical, critical, or supercritical. The same algorithm also decides consistency of stochastic context-free grammars.\",\n \"Title: Approximative Methods for Monotone Systems of Min-Max-Polynomial Equations\\nYear: 2008\\n\\nAbstract: On the Internet, there are advertisements (ads) of different kinds: image, text, video and other specially marked objects that are distinct from the underlying content of the page. There is an industry behind the management of such ads, and they face ...\",\n \"Title: Language equivalence of probabilistic pushdown automata\\nYear: 2014\\n\\nAbstract: We study the language equivalence problem for probabilistic pushdown automata (pPDA) and their subclasses. We show that the problem is interreducible with the multiplicity equivalence problem for context-free grammars, the decidability of which has been open for several decades. Interreducibility also holds for pPDA with one control state. In contrast, for the case of a one-letter input alphabet we show that pPDA language equivalence (and hence multiplicity equivalence of context-free grammars) is in PSPACE and at least as hard as the polynomial identity testing problem.\",\n \"Title: Minimisation of Multiplicity Tree Automata.\\nYear: 2014\\n\\nAbstract: We consider the problem of minimising the number of states in a multiplicity tree automaton over the field of rational numbers. We give a minimisation algorithm that runs in polynomial time assuming unit-cost arithmetic. We also show that a polynomial bound in the standard Turing model would require a breakthrough in the complexity of polynomial identity testing by proving that the latter problem is logspace equivalent to the decision version of minimisation. The developed techniques also improve the state of the art in multiplicity word automata: we give an NC algorithm for minimising multiplicity word automata. Finally, we consider the minimal consistency problem: does there exist an automaton with n states that is consistent with a given finite sample of weight-labelled words or trees? We show that this decision problem is complete for the existential theory of the rationals, both for words and for trees of a fixed alphabet rank.\",\n \"Title: Proving the Herman-Protocol Conjecture.\\nYear: 2015\\n\\nAbstract: Herman's self-stabilisation algorithm, introduced 25 years ago, is a well-studied synchronous randomised protocol for enabling a ring of $N$ processes collectively holding any odd number of tokens to reach a stable state in which a single token remains. Determining the worst-case expected time to stabilisation is the central outstanding open problem about this protocol. It is known that there is a constant $h$ such that any initial configuration has expected stabilisation time at most $h N^2$. Ten years ago, McIver and Morgan established a lower bound of $4/27 \\\\approx 0.148$ for $h$, achieved with three equally-spaced tokens, and conjectured this to be the optimal value of $h$. A series of papers over the last decade gradually reduced the upper bound on $h$, with the present record (achieved in 2014) standing at approximately $0.156$. In this paper, we prove McIver and Morgan's conjecture and establish that $h = 4/27$ is indeed optimal.\",\n \"Title: Tree Buffers.\\nYear: 2015\\n\\nAbstract: In runtime verification, the central problem is to decide if a given program execution violates a given property. In online runtime verification, a monitor observes a program's execution as it happens. If the program being observed has hard real-time constraints, then the monitor inherits them. In the presence of hard real-time constraints it becomes a challenge to maintain enough information to produce error traces, should a property violation be observed. In this paper we introduce a data structure, called tree buffer, that solves this problem in the context of automata-based monitors: If the monitor itself respects hard real-time constraints, then enriching it by tree buffers makes it possible to provide error traces, which are essential for diagnosing defects. We show that tree buffers are also useful in other application domains. For example, they can be used to implement functionality of capturing groups in regular expressions. We prove optimal asymptotic bounds for our data structure, and validate them using empirical data from two sources: regular expression searching through Wikipedia, and runtime verification of execution traces obtained from the DaCapo test suite.\",\n \"Title: The Odds of Staying on Budget.\\nYear: 2014\\n\\nAbstract: Given Markov chains and Markov decision processes MDPs whose transitions are labelled with non-negative integer costs, we study the computational complexity of deciding whether the probability of paths whose accumulated cost satisfies a Boolean combination of inequalities exceeds a given threshold. For acyclic Markov chains, we show that this problem is PP-complete, whereas it is hard for the PosSLP problem and in PSpace for general Markov chains. Moreover, for acyclic and general MDPs, we prove PSpace- and EXP-completeness, respectively. Our results have direct implications on the complexity of computing reward quantiles in succinctly represented stochastic systems.\",\n \"Title: Bisimilarity of Pushdown Automata is Nonelementary\\nYear: 2013\\n\\nAbstract: Given two pushdown automata, the bisimilarity problem asks whether the infinite transition systems they induce are bisimilar. While this problem is known to be decidable our main result states that it is nonelementary, improving EXPTIME-hardness, which was the best previously known lower bound for this problem. Our lower bound result holds for normed pushdown automata as well.\",\n \"Title: The complexity of the Kth largest subset problem and related problems.\\nYear: 2016\\n\\nAbstract: •The Kth largest subset problem is shown PP-complete.•The Kth largest m-tuple problem is shown PP-complete.•Further problems whose complexity was open are also shown PP-complete.\",\n \"Title: Efficient Quantile Computation in Markov Chains via Counting Problems for Parikh Images.\\nYear: 2016\\n\\nAbstract: A cost Markov chain is a Markov chain whose transitions are labelled with non-negative integer costs. A fundamental problem on this model, with applications in the verification of stochastic systems, is to compute information about the distribution of the total cost accumulated in a run. This includes the probability of large total costs, the median cost, and other quantiles. While expectations can be computed in polynomial time, previous work has demonstrated that the computation of cost quantiles is harder but can be done in PSPACE. In this paper we show that cost quantiles in cost Markov chains can be computed in the counting hierarchy, thus providing evidence that computing those quantiles is likely not PSPACE-hard. We obtain this result by exhibiting a tight link to a problem in formal language theory: counting the number of words that are both accepted by a given automaton and have a given Parikh image. Motivated by this link, we comprehensively investigate the complexity of the latter problem. Among other techniques, we rely on the so-called BEST theorem for efficiently computing the number of Eulerian circuits in a directed graph.\",\n \"Title: Trace Refinement in Labelled Markov Decision Processes.\\nYear: 2016\\n\\nAbstract: Given two labelled Markov decision processes (MDPs), the trace-refinement problem asks whether for all strategies of the first MDP there exists a strategy of the second MDP such that the induced labelled Markov chains are trace-equivalent. We show that this problem is decidable in polynomial time if the second MDP is a Markov chain. The algorithm is based on new results on a particular notion of bisimulation between distributions over the states. However, we show that the general trace-refinement problem is undecidable, even if the first MDP is a Markov chain. Decidability of those problems was stated as open in 2008. We further study the decidability and complexity of the trace-refinement problem provided that the strategies are restricted to be memoryless.\",\n \"Title: Nonnegative Matrix Factorization Requires Irrationality\\nYear: 2016\\n\\nAbstract: Nonnegative matrix factorization (NMF) is the problem of decomposing a given nonnegative is n x m so matrix M into a product of a nonnegative is n x d matrix W and a nonnegative d x m to matrix H. A longstanding open question, posed by Cohen and Rothblum in 1993, is whether a rational matrix M always has an NMF of minimal inner dimension d whose factors W and H are also rational. We answer this question negatively, by exhibiting a matrix for which W and H require irrational entries.\",\n \"Title: On rationality of nonnegative matrix factorization\\nYear: 2016\\n\\nAbstract: Nonnegative matrix factorization (NMF) is the problem of decomposing a given nonnegative n × m matrix M into a product of a nonnegative n × d matrix W and a nonnegative d × m matrix H. NMF has a wide variety of applications, including bioinformatics, chemometrics, communication complexity, machine learning, polyhedral combinatorics, among many others. A longstanding open question, posed by Cohen and Rothblum in 1993, is whether every rational matrix M has an NMF with minimal d whose factors W and H are also rational. We answer this question negatively, by exhibiting a matrix M for which W and H require irrational entries. As an application of this result, we show that state minimization of labeled Markov chains can require the introduction of irrational transition probabilities. We complement these irrationality results with an NP-complete version of NMF for which rational numbers suffice.\",\n \"Title: Markov Chains And Unambiguous Buchi Automata\\nYear: 2016\\n\\nAbstract: Unambiguous automata, i.e., nondeterministic automata with the restriction of having at most one accepting run over a word, have the potential to be used instead of deterministic automata in settings where nondeterministic automata can not be applied in general. In this paper, we provide a polynomially time-bounded algorithm for probabilistic model checking of discrete-time Markov chains against unambiguous Buchi automata specifications and report on our implementation and experiments.\",\n \"Title: On strong determinacy of countable stochastic games\\nYear: 2017\\n\\nAbstract: We study 2-player turn-based perfect-information stochastic games with countably infinite state space. The players aim at maximizing/minimizing the probability of a given event (i.e., measurable set of infinite plays), such as reachability, Büchi, ω-regular or more general objectives. These games are known to be weakly determined, i.e., they have value. However, strong determinacy of threshold objectives (given by an event ε and a threshold c ∈ [0,1]) was open in many cases: is it always the case that the maximizer or the minimizer has a winning strategy, i.e., one that enforces, against all strategies of the other player, that ε is satisfied with probability ≥ c (resp. <; c)? We show that almost-sure objectives (where c = 1) are strongly determined. This vastly generalizes a previous result on finite games with almost-sure tail objectives. On the other hand we show that ≥ 1/2 (co-)Biichi objectives are not strongly determined, not even if the game is finitely branching. Moreover, for almost-sure reachability and almost-sure Biichi objectives in finitely branching games, we strengthen strong determinacy by showing that one of the players must have a memory less deterministic (MD) winning strategy.\",\n \"Title: Counting Problems for Parikh Images.\\nYear: 2017\\n\\nAbstract: Given finite-state automata (or context-free grammars) A,B over the same alphabet and a Parikh vector p, we study the complexity of deciding whether the number of words in the language of A with Parikh image p is greater than the number of such words in the language of B. Recently, this problem turned out to be tightly related to the cost problem for weighted Markov chains. We classify the complexity according to whether A and B are deterministic, the size of the alphabet, and the encoding of p (binary or unary).\",\n \"Title: Game Characterization of Probabilistic Bisimilarity, and Applications to Pushdown Automata.\\nYear: 2018\\n\\nAbstract: We study the bisimilarity problem for probabilistic pushdown automata (pPDA) and subclasses thereof. Our definition of pPDA allows both probabilistic and non-deterministic branching, generalising the classical notion of pushdown automata (without epsilon-transitions). We first show a general characterization of probabilistic bisimilarity in terms of two-player games, which naturally reduces checking bisimilarity of probabilistic labelled transition systems to checking bisimilarity of standard (non-deterministic) labelled transition systems. This reduction can be easily implemented in the framework of pPDA, allowing to use known results for standard (non-probabilistic) PDA and their subclasses. A direct use of the reduction incurs an exponential increase of complexity, which does not matter in deriving decidability of bisimilarity for pPDA due to the non-elementary complexity of the problem. In the cases of probabilistic one-counter automata (pOCA), of probabilistic visibly pushdown automata (pvPDA), and of probabilistic basic process algebras (i.e., single-state pPDA) we show that an implicit use of the reduction can avoid the complexity increase; we thus get PSPACE, EXPTIME, and 2-EXPTIME upper bounds, respectively, like for the respective non-probabilistic versions. The bisimilarity problems for OCA and vPDA are known to have matching lower bounds (thus being PSPACE-complete and EXPTIME-complete, respectively); we show that these lower bounds also hold for fully probabilistic versions that do not use non-determinism.\",\n \"Title: On Computing the Total Variation Distance of Hidden Markov Models.\\nYear: 2018\\n\\nAbstract: We prove results on the decidability and complexity of computing the total variation distance (equivalently, the $L_1$-distance) of hidden Markov models (equivalently, labelled Markov chains). This distance measures the difference between the distributions on words that two hidden Markov models induce. The main results are: (1) it is undecidable whether the distance is greater than a given threshold; (2) approximation is #P-hard and in PSPACE.\",\n \"Title: On the Complexity of Iterative Tropical Computation with Applications to Markov Decision Processes.\\nYear: 2018\\n\\nAbstract: We study the complexity of evaluating powered functions implemented by straight-line programs (SLPs) over the tropical semiring (i.e., with max and + operations). In this problem, a given (max,+)-SLP with the same number of input and output wires is composed with H copies of itself, where H is given in binary. The problem of evaluating powered SLPs is intimately connected with iterative arithmetic computations that arise in algorithmic decision making and operations research. Specifically, it is essentially equivalent to finding optimal strategies in finite-horizon Markov Decision Processes (MDPs). We show that evaluating powered SLPs and finding optimal strategies in finite-horizon MDPs are both EXPTIME-complete problems. This resolves an open problem that goes back to the seminal 1987 paper on the complexity of MDPs by Papadimitriou and Tsitsiklis.\",\n \"Title: Selective Monitoring.\\nYear: 2018\\n\\nAbstract: We study selective monitors for labelled Markov chains. Monitors observe the outputs that are generated by a Markov chain during its run, with the goal of identifying runs as correct or faulty. A monitor is selective if it skips observations in order to reduce monitoring overhead. We are interested in monitors that minimize the expected number of observations. We establish an undecidability result for selectively monitoring general Markov chains. On the other hand, we show for non-hidden Markov chains (where any output identifies the state the Markov chain is in) that simple optimal monitors exist and can be computed efficiently, based on DFA language equivalence. These monitors do not depend on the precise transition probabilities in the Markov chain. We report on experiments where we compute these monitors for several open-source Java projects.\",\n \"Title: On the Complexity of Value Iteration.\\nYear: 2019\\n\\nAbstract: Value iteration is a fundamental algorithm for solving Markov Decision Processes (MDPs). It computes the maximal $n$-step payoff by iterating $n$ times a recurrence equation which is naturally associated to the MDP. At the same time, value iteration provides a policy for the MDP that is optimal on a given finite horizon $n$. In this paper, we settle the computational complexity of value iteration. We show that, given a horizon $n$ in binary and an MDP, computing an optimal policy is EXP-complete, thus resolving an open problem that goes back to the seminal 1987 paper on the complexity of MDPs by Papadimitriou and Tsitsiklis. As a stepping stone, we show that it is EXP-complete to compute the $n$-fold iteration (with $n$ in binary) of a function given by a straight-line program over the integers with $max$ and $+$ as operators.\",\n \"Title: On Semigroups of Two-Dimensional Upper-Triangular Integer Matrices.\\nYear: 2019\\n\\nAbstract: We analyze algorithmic problems in finitely generated semigroups of two-dimensional upper-triangular integer matrices. These semigroup problems are tightly connected with problems about compositions of affine functions over one variable. Building on a variety of techniques from recent years, we obtain a number of complexity results, including NP-completeness of the mortality problem for matrices with determinants +1 and 0.\",\n \"Title: Efficient Analysis of Unambiguous Automata Using Matrix Semigroup Techniques.\\nYear: 2019\\n\\nAbstract: We introduce a novel technique to analyse unambiguous B\\\\\\\"uchi automata quantitatively, and apply this to the model checking problem. It is based on linear-algebra arguments that originate from the analysis of matrix semigroups with constant spectral radius. This method can replace a combinatorial procedure that dominates the computational complexity of the existing procedure by Baier et al. We analyse the complexity in detail, showing that, in terms of the set $Q$ of states of the automaton, the new algorithm runs in time $O(|Q|^4)$, improving on an efficient implementation of the combinatorial algorithm by a factor of $|Q|$.\",\n \"Title: Linear-Time Model Checking Branching Processes.\\nYear: 2021\\n\\nAbstract: (Multi-type) branching processes are a natural and well-studied model for generating random infinite trees. Branching processes feature both nondeterministic and probabilistic branching, generalizing both transition systems and Markov chains (but not generally Markov decision processes). We study the complexity of model checking branching processes against linear-time omega-regular specifications: is it the case almost surely that every branch of a tree randomly generated by the branching process satisfies the omega-regular specification? The main result is that for LTL specifications this problem is in PSPACE, subsuming classical results for transition systems and Markov chains, respectively. The underlying general model-checking algorithm is based on the automata-theoretic approach, using unambiguous B\\\\\\\"uchi automata.\",\n \"Title: An ETH-Tight Algorithm for Multi-Team Formation.\\nYear: 2021\\n\\nAbstract: We propose polynomial-time algorithms to minimise labelled Markov chains whose transition probabilities are not known exactly, have been perturbed, or can only be obtained by sampling. Our algorithms are based on a new notion of an approximate bisimulation quotient, obtained by lumping together states that are exactly bisimilar in a slightly perturbed system. We present experiments that show that our algorithms are able to recover the structure of the bisimulation quotient of the unperturbed system.\"\n]","isConversational":false},"profile_size":{"kind":"number","value":60,"string":"60"},"task":{"kind":"string","value":"longlamp_2"},"split_kind":{"kind":"string","value":"user_based"},"id":{"kind":"number","value":13596,"string":"13,596"}}},{"rowIdx":13597,"cells":{"question_id":{"kind":"string","value":"Zhe Li"},"question":{"kind":"string","value":"Generate an abstract for the title \"SC-DCNN: Highly-Scalable Deep Convolutional Neural Network using Stochastic Computing.\" using the following items: 1. Wearable devices and IoTs\n2. DCNN accelerators\n3. Stochastic Computing (SC)\n4. SC-DCNN design framework\n5. Area and power-efficient DC"},"answer":{"kind":"string","value":"With the recent advance of wearable devices and Internet of Things (IoTs), it becomes attractive to implement the Deep Convolutional Neural Networks (DCNNs) in embedded and portable systems. Currently, executing the software-based DCNNs requires high-performance servers, restricting the widespread deployment on embedded and mobile IoT devices. To overcome this obstacle, considerable research efforts have been made to develop highly-parallel and specialized DCNN accelerators using GPGPUs, FPGAs or ASICs. Stochastic Computing (SC), which uses a bit-stream to represent a number within [-1, 1] by counting the number of ones in the bit-stream, has high potential for implementing DCNNs with high scalability and ultra-low hardware footprint. Since multiplications and additions can be calculated using AND gates and multiplexers in SC, significant reductions in power (energy) and hardware footprint can be achieved compared to the conventional binary arithmetic implementations. The tremendous savings in power (energy) and hardware resources allow immense design space for enhancing scalability and robustness for hardware DCNNs. This paper presents SC-DCNN, the first comprehensive design and optimization framework of SC-based DCNNs, using a bottom-up approach. We first present the designs of function blocks that perform the basic operations in DCNN, including inner product, pooling, and activation function. Then we propose four designs of feature extraction blocks, which are in charge of extracting features from input feature maps, by connecting different basic function blocks with joint optimization. Moreover, the efficient weight storage methods are proposed to reduce the area and power (energy) consumption. Putting all together, with feature extraction blocks carefully selected, SC-DCNN is holistically optimized to minimize area and power (energy) consumption while maintaining high network accuracy. Experimental results demonstrate that the LeNet5 implemented in SC-DCNN consumes only 17 mm2 area and 1.53 W power, achieves throughput of 781250 images/s, area efficiency of 45946 images/s/mm2, and energy efficiency of 510734 images/J."},"correct_answer":{"kind":"string","value":"With the recent advance of wearable devices and Internet of Things (IoTs), it becomes attractive to implement the Deep Convolutional Neural Networks (DCNNs) in embedded and portable systems. Currently, executing the software-based DCNNs requires high-performance servers, restricting the widespread deployment on embedded and mobile IoT devices. To overcome this obstacle, considerable research efforts have been made to develop highly-parallel and specialized DCNN accelerators using GPGPUs, FPGAs or ASICs. Stochastic Computing (SC), which uses a bit-stream to represent a number within [-1, 1] by counting the number of ones in the bit-stream, has high potential for implementing DCNNs with high scalability and ultra-low hardware footprint. Since multiplications and additions can be calculated using AND gates and multiplexers in SC, significant reductions in power (energy) and hardware footprint can be achieved compared to the conventional binary arithmetic implementations. The tremendous savings in power (energy) and hardware resources allow immense design space for enhancing scalability and robustness for hardware DCNNs. This paper presents SC-DCNN, the first comprehensive design and optimization framework of SC-based DCNNs, using a bottom-up approach. We first present the designs of function blocks that perform the basic operations in DCNN, including inner product, pooling, and activation function. Then we propose four designs of feature extraction blocks, which are in charge of extracting features from input feature maps, by connecting different basic function blocks with joint optimization. Moreover, the efficient weight storage methods are proposed to reduce the area and power (energy) consumption. Putting all together, with feature extraction blocks carefully selected, SC-DCNN is holistically optimized to minimize area and power (energy) consumption while maintaining high network accuracy. Experimental results demonstrate that the LeNet5 implemented in SC-DCNN consumes only 17 mm2 area and 1.53 W power, achieves throughput of 781250 images/s, area efficiency of 45946 images/s/mm2, and energy efficiency of 510734 images/J."},"documents":{"kind":"list like","value":["Title: Towards a hybrid optimization model for elemental content analysis in EDXRF.\nYear: 2012\n\nAbstract: This paper presents a hybrid optimization model for predicting the elemental contents such as Ti, V and Fe in energy dispersive X-ray fluorescence (EDXRF) based on least square support vector machine (LS-SVM) and particle swarm optimization (PSO) methods. The model used PSO to optimize LS-SVM parameters. In order to assess the capability and effectiveness of the proposed model, several measurement methods such as SVM model and BP neural network model were compared. The results indicate that the proposed model is feasible for quantitative analysis of elemental contents in nondestructive nuclear measurement applications. © 2012 IEEE.","Title: Optimal network locality in distributed virtualized data-centers\nYear: 2011\n\nAbstract: Cost efficiency is a key aspect in deploying distributed service in networks within decentralized service delivery architectures. In this paper, we address this aspect from an optimization and algorithmic standpoint. The research deals with the placement of service components to network sites, where the performance metric is the cost for acquiring components between the sites. The resulting optimization problem, which we refer to as the k-Component Multi-site Placement Problem, is applicable to service distribution in a wide range of communication networking scenarios. We provide a theoretical analysis of the problem's computational complexity, and develop an integer programming model for providing reference results for performance benchmarking. On the algorithmic side, we present four approaches: an algorithm with approximation guarantee and three heuristics algorithms. The first heuristic is derived from graph theory on domatic partition. The second heuristic, built on intuition, admits distributed computation. The third heuristic emphasizes on fairness in cost distribution among the sites. We report simulation results for sets of networks where cost is represented by round-trip time (RTT) originating from real measurements. For small networks, the integer model is used to study algorithm performance in terms of optimality. Large networks are used to compare the algorithms relatively to each other. Among the algorithms, the heuristic based on intuition has close-to-optimal performance, and the fairness heuristic achieves a good balance between single-site cost and the overall one. In addition, the experiments demonstrate the significance of optimization for cost reduction in comparison to a the random allocation strategy.","Title: Multiple Sources Data Fusion Strategies Based on Multi-class Support Vector Machine\nYear: 2008\n\nAbstract: Data fusion from multiple sources is very important and useful. Some strategies for multiple sources data fusion based on Multi-class Support Vector Machine (MSVM) are proposed in this paper. The features (independent components) of data from multiple sources are extracted for fusion. The Dempster-Shafer theory (DS theory) and Bayesian theory are used to combine the probabilistic outputs of MSVMs. Then the outputs of DS theory combination are classified by a MSVM to get the final decision of the classification. Finally, these strategies are evaluated by three data sets and the results show that DS theory can improve the accuracy obviously, and the strategies based on MSVM and DS theory is very fit for solving problems with small data sets.","Title: View-adaptive manipulative action recognition for robot companions\nYear: 2007\n\nAbstract: This paper puts forward an approach for a mobile robot to recognize the human's manipulative actions from different single camera views. While most of the related work in action recognition assume a fixed static camera view that is the same for training and testing, such kind of constraints do not apply for mobile robot companions. We propose a recognition scheme that is able to generalize an action model, that has been learned from a very few data items observed from a single camera view, to variant view points and different settings. We tackle the problem of compensating the view dependence of 2D motion models on three different levels. Firstly, we pre-segment the trajectories based on an object vicinity that depends on the camera tilt and object detections. Secondly, an interactive feature vector is designed that represents the relative movements between the human hand and the objects. Thirdly, we propose an adaptive HMM-based matching process that is based on a particle filter and includes a dynamically adjusted scaling parameter that models the systematic error of the view dependency. Finally, we use a two-layered approach for task recognition which decouples the task knowledge from the view dependent primitive recognition. The results of experiments in an office environment show the applicability of this approach.","Title: Blind Detection of Digital Forgery Image Based on the Local Entropy of the Gradient\nYear: 2008\n\nAbstract: A novel method based on the local entropy of the gradient is proposed to detect the forged digital images. The method can discover some traces of artificial feather operation, which is necessary to create a smooth transition between a forged image region and its surroundings. The local entropy of the gradient is used to determine the forged region, and the credibility is computed to show the reality level of the image. Results of experiments on several forged images demonstrate the effectiveness of the algorithm.","Title: Pinning dynamic systems of networks with Markovian switching couplings and controller-node set.\nYear: 2014\n\nAbstract: In this paper, we study pinning control problem of coupled dynamical systems with stochastically switching couplings and stochastically selected controller–node set. Here, the coupling matrices and the controller–node sets change with time, induced by a continuous-time Markov chain. By constructing Lyapunov functions, we establish tractable sufficient conditions for exponential stability of the coupled system. Two scenarios are considered here. First, we prove that if each subsystem in the switching system, i.e. with the fixed coupling, can be stabilized by the fixed pinning controller–node set, and in addition, the Markovian switching is sufficiently slow, then the time-varying dynamical system is stabilized. Second, we conclude that if the system with the average coupling and pinning gains can be stabilized and the switching is sufficiently fast, the time-varying system is stabilized. Two numerical examples are provided to demonstrate the validity of these theoretical results, including a switching dynamical system between several stable subsystems, and a dynamical system with mobile nodes and spatial pinning control towards the nodes when these nodes are being in a pre-designed region.","Title: Spatio-temporal variation of NDVI in Chinese coastal zone during recent 20 years\nYear: 2005\n\nAbstract: Monthly digital NOAA/AVHRR NDVI images with 8km spatial resolution, multi-temporal land use data and monthly average air temperature and precipitation data recorded in standard observatories from 1981 to 2000 are obtained in Chinese coastal zone. The relationship between Normalized Difference Vegetation Index (NDVI) and land use factors and climate factors are explored through spatio-temporal variation analysis. At First, NDVI, monthly average air temperature and precipitation data are normalized to the range of 0.01 to 0.99, then 20 typical areas based on the latitude from north to south in Chinese coastal zone are selected, which is used to analyze the correlation between NDVI and air temperature and precipitation, Then, NDVI spatial and temporal variations are achieved for Chinese coastal area combined with regional land use data. Based on these results, it is pointed out that, in recent 20 years, NDVI annul variation in Chinese coastal zone has a stronger relationship with land use factors; moreover, NDVI spatial variation has obvious strap regularity: NDVI in North China is highly sensitive to precipitation, while it is highly sensitive to air temperature in South China. © 2005 IEEE.","Title: Research of localization approach for the new comer in wireless sensor networks\nYear: 2009\n\nAbstract: When some sensor nodes of wireless sensor networks (WSN) can not work forever because of long-term work or failure caused\n by attack, a few new comers need to be put into the network. For the application of the new comer in WSN, an accurate and\n effective localization algorithm based on received signal strength indicator (RSSI) is proposed. Through a few necessary nodes’\n participation and the collaboration between the new comer and its one-hop and two-hop neighbor nodes, the accurate localization\n of the new comer is achieved. Simulation results show that the localization accuracy is about 17% of sensor node’s radio frequency\n (RF) transmission range, when the measurement error is 10% and the standard deviation for Gauss error of original sensor nodes’\n coordinate is about 20% of sensor node’s RF transmission range. Simulation results also verify nice stability and adaptability\n of the new comer’s location algorithm.","Title: Evaluation Of Lauge-Hansen Classification For Ankle Fractures And Follow-Up Management\nYear: 2013\n\nAbstract: Objective: To evaluate Lauge-Hansen classification on the diagnosis and treatment of ankle fractures. Methods: From April 2008 to June 2011, and following ethical approval of our research, patients with ankle fractures admitted at our second orthopaedics department were enrolled. All patients were followed up for a minimum of 8 months (8 similar to 46 with a mean of 27 months). Trauma mechanism and American Orthopaedic Foot and Ankle Society (AOFAS) scores were analyzed. The position and grade of ankle fracture injury were accessed for making treatment strategies for patients. Results: Thirty-two patients were recruited fulfilling the inclusion criteria and were available for follow-up. After treatment, none of the cases had postoperative infection, cut aneous necrosis, failure in internal fixation or nerve injury. At 12-month follow-up, AOFAS scores were 97.7 at average, among which 30 cases were excellent and 2 cases were good. Conclusion: The Lauge-Hansen classification contributes fully to diagnose ankle fractures, so that surgeons can make correct treatment strategies and achieve good treatment effect. Besides, perfect follow-up records, genial attitude, good communication skill and flexible follow-up time are the key to finish evaluation as schedule.","Title: Fast Coding Unit Depth Decision Algorithm for Interframe Coding in HEVC\nYear: 2013\n\nAbstract: As the next generation standard of video coding, the High Efficiency Video Coding (HEVC) achieves significantly better coding efficiency than all existing video coding standards. A Coding Unit (CU) quad tree concept is introduced to HEVC to improve the coding efficiency. Each CU node in quad tree will be traversed by depth first search process to find the best Coding Tree Unit (CTU) partition. Although this quad tree search process can obtain the best CTU partition, it is very time consuming, especially in interframe coding. To alleviate the encoder computation load in interframe coding, a fast CU depth decision method is proposed by reducing the depth search range. Based on the depth information correlation between spatio-temporal adjacent CTUs and the current CTU, some depths can be adaptively excluded from the depth search process in advance. Experimental results show that the proposed scheme provides almost 30% encoder time savings on average compared to the default encoding scheme in HM8.0 with only 0.38% bit rate increment in coding performance.","Title: A PSO-SVM based model for alpha particle activity prediction inside decommissioned channels\nYear: 2012\n\nAbstract: This paper presents a hybrid Support Vector Machine (SVM) and Particle Swarm Optimization (PSO) model for predicting alpha particles emitting contamination on the internal surfaces of decommissioned channels. Six measuring parameters (channel diameter, channel length, distance to radioactive source, radioactive strength, wind speed and flux) and one ionizing value have been obtained via experiments. These parameters show complex linear and nonlinear relationships to measuring results. The model used PSO to optimize SVM parameters. The comparison of computational results of the hybrid approach with normal BP networks confirms its clear advantage for dealing with this complex nonlinear prediction.","Title: An Improved Similarity-Based Fast Coding Unit Depth Decision Algorithm for Inter-frame Coding in HEVC\nYear: 2014\n\nAbstract: The emerging High Efficiency Video Coding (HEVC) aims to achieve significantly improved compression performance with respect to the state-of-the-art H.264/AVC high profile. This better compression performance is achieved at a much higher computational complexity, which makes it difficult for real-time video systems. To reduce the encoding complexity, this paper presents an improved similarity-based fast coding unit depth decision algorithm for inter-frame coding. Firstly, a fast and precise depth information acquisition method is proposed to improve the accuracy of depth prediction. Secondly, CTUs of the medium similarity degree is further divided into two similarity degree categories, and the complexity of coding the coding tree units of these two categories is reduced by different coding unit depth decision strategies. Experimental results show that proposed algorithm can save on average 35.72% of the encoding time with negligible loss on the rate-distortion performance compared with HM11.0 and consistently outperforms the state-of-the-art schemes.","Title: Fast Logo Detection and Recognition in Document Images\nYear: 2010\n\nAbstract: The scientific significance of automatic logo detection and recognition is more and more growing because of the increasing requirements of intelligent document image analysis and retrieval. In this paper, we introduce a system architecture which is aiming at segmentation-free and layout-independent logo detection and recognition. Along with the unique logo feature design, a novel way to ensure the geometrical relationships among the features, and different optimizations in the recognition process, this system can achieve improvements concerning both the recognition performance and the running time. The experimental results on several sets of real-word documents demonstrate the effectiveness of our approach.","Title: On the existence of certain error formulas for a special class of ideal projectors.\nYear: 2011\n\nAbstract: In this paper, we focus on a special class of ideal projectors. With the aid of algebraic geometry, we prove that for this special class of ideal projectors, there exist “good” error formulas as defined by C. de Boor. Furthermore, we completely analyze the properties of the interpolation conditions matched by this special class of ideal projectors, and show that the ranges of this special class of ideal projectors are the minimal degree interpolation spaces with regard to their associated interpolation conditions.","Title: The Nature and Classification of Unlabelled Neurons in the Use of Kohonen's Self-Organizing Map for Supervised Classification\nYear: 2006\n\nAbstract: Kohonen's Self-Organizing Map is a neural network procedure in which a layer of neurons is initialized with random weights, and subsequently organized by inspection of the data to be classified. The organization procedure uses progressive adjustment of weights based on data characteristics and lateral interaction such that neurons with similar weights will tend to spatially cluster in the neuron layer. When the SOM is associated with a supervised classification, a majority voting technique is usually used to associate these neurons with training data classes. This technique, however, cannot guarantee that every neuron in the output layer will be labeled, and thus causes unclassified pixels in the final map. This problem is similar to but fundamentally different from the problem of dead units that arises in unsupervised SOM classification (neurons which are never organized by the input data). Although dead unit problems have been widely discussed and some possible solutions proposed, few studies are associated with land cover classification. In this paper we specifically address the problem and nature of unlabelled neurons in the use of SOM for supervised classification. Through a case study it is shown that unlabelled neurons are associated with unknown image classes and, most particularly, mixed pixels. It is also shown that an auxiliary algorithm proposed here for assigning classes to unlabelled neurons performs with the same success as that experienced with Maximum Likelihood.","Title: Opportunistic decision making and complexity in emergency care.\nYear: 2011\n\nAbstract: In critical care environments such as the emergency department (ED), many activities and decisions are not planned. In this study, we developed a new methodology for systematically studying what are these unplanned activities and decisions. This methodology expands the traditional naturalistic decision making (NDM) frameworks by explicitly identifying the role of environmental factors in decision making. We focused on decisions made by ED physicians as they transitioned between tasks. Through ethnographic data collection, we developed a taxonomy of decision types. The empirical data provide important insight to the complexity of the ED environment by highlighting adaptive behavior in this intricate milieu. Our results show that half of decisions in the ED we studied are not planned, rather decisions are opportunistic decision (34%) or influenced by interruptions or distractions (21%). What impacts these unplanned decisions have on the quality, safety, and efficiency in the ED environment are important research topics for future investigation.","Title: Unsupervised text segmentation using LDA and MCMC\nYear: 2012\n\nAbstract: In this paper, we propose a data driven approach to text segmentation, while most of the existing unsupervised methods determine segmentation boundaries by empirically exploring similarity measurement between adjacent units (e.g. sentences). Firstly, we train a latent Dirichlet allocation (LDA) model with the large scale Wikipedia Corpus to avoid the problem of vocabulary mismatch, which makes our approach domain-independent. Secondly, each segment unit is represented with a distribution of the topics, instead of a set of word tokens. Finally, a text input is modeled as a sequence of segment units and Markov Chain Monte Carlo technique is employed to decide the appropriate boundaries. The major advantage of using MCMC is its ability to detect both strong and weak boundaries. Experimental results demonstrate that our proposed approach achieve promising results on a widely used benchmark dataset when compared with the state-of-the-art methods.","Title: Efficient Nonuniform Fourier Reconstruction for Spaceborne/Airborne Bistatic SAR\nYear: 2014\n\nAbstract: In this letter, an efficient imaging algorithm is proposed for reconstructing the spaceborne/airborne bistatic synthetic aperture radar (SA-BSAR) data acquired with a high squint angle. By utilizing the Taylor expansion and the chain rule for derivatives of composite function, the expression of the SA-BSAR ideal matching filter is converted to a 2-D nonuniform discrete Fourier transform (NUDFT). Due to neither input nor output data of the NUDFT being equispaced, nonuniform fast Fourier transform of type 3 (NUFFT-3) is implemented to speed up the SA-BSAR image reconstruction. The computation complexity of the proposed imaging algorithm is O(MNlogMN), where MN is the number of the image pixels. Simulation results demonstrate the validity of the proposed algorithm.","Title: Computing The K-Error Linear Complexity Of Q-Ary Sequences With Period 2p(N)\nYear: 2012\n\nAbstract: The linear complexity and its stability of periodic sequences are of fundamental importance as measure indexes on the security of stream ciphers and the k-error linear complexity reveals the stability of the linear complexity properly. Recently, Zhou designed an algorithm for computing the k-error linear complexity of 2p(n) periodic sequences over GF(q). In this paper, we develop a genetic algorithm to confirm that one can't get the real k-error linear complexity for some sequenes by the Zhou's algorithm. Analysis indicates that the Zhou's algorithm is unreasonable in some steps. The corrected algorithm is presented. Such algorithm will increase the amount of computation, but is necessary to get the real k-error linear complexity. Here p and q are odd prime, and q is a primitive root (mod p(2)).","Title: Load balancing strategy based on network load capacity\nYear: 2013\n\nAbstract: This paper presents an economical extensible model and a new load balancing strategy based on network load capacity, so that the integrity of evidence in high-speed network forensics can be guaranteed. This algorithm gets the capture ability of evidence capture by dynamic feedback and forecast mechanism. Taking one session as a distribution unit, it distributes the network, it distributes the network packets to the host with the maximal-load capacity. The result shows the stretchability of this system can satisfy the current huge network-flow.","Title: Financial contagion analysis based on hybrid nonlinear mutual prediction algorithm and fuzzy neural networks\nYear: 2008\n\nAbstract: The contagion of the financial crisis became more and more evident since 1990s. The traditional works are mostly based on the linear methods, which have limitation to investigate the nonlinear features of the financial contagion. The hybrid nonlinear mutual prediction algorithm and fuzzy neural networks method is used in this paper to work on it. The searching of the dynamical interdependence is the first and important step of the financial contagion study. The basis is the multi-series nonlinear mutual prediction method, which means that mutual predictability of two time series is the evidence of the dynamical interdependence. The hybrid nonlinear mutual prediction and fuzzy neural networks are used to detect the dynamical interdependence of the stock indices time series of China, Hong Kong and USA. The results show that there exists strong mutual predictability in each pair of these economies, which means that dynamical interdependence is detectable in these economies. © 2008 IEEE.","Title: A calibration and validation procedure for microscopic simulation model\nYear: 2010\n\nAbstract: Simulation modeling is an increasingly popular and effective tool for analyzing transportation problems that are not amendable to study by other means. For any simulation study, model calibration and validation is a crucial step to obtaining any results from analysis. This paper proposed a systematic, practical procedure for microscopic simulation model calibration and validation. The validity of the proposed procedure was demonstrated visa a case study containing a freeway in Guangdong Province, China using a widely used microscopic traffic simulation model, PARAMICS. The simulation results compared against multiple days of field data to determine the performance of the calibrated model. It was found that the calibrated parameters using the proposed procedure generated performance measures that were representative of the field conditions while the simulation results of the default parameters were significantly different from the field data.","Title: Universal approximation of TS fuzzy systems constructed dynamically-MISO cases\nYear: 2010\n\nAbstract: Study on approximation capabilities of fuzzy systems has been an important and hot issue for many years. Many existing studies on fuzzy systems based on standard affine TS fuzzy models suffer seriously from curse of dimensionality though such models are relatively simple to construct; whereas other investigations based on hierarchical fuzzy models have to face the difficulty of relatively more complexity. Through utilizing the dynamic partition along all dimensions, this paper proposes a novel dynamically constructive method in MISO cases to construct fuzzy systems based on standard affine TS fuzzy models, which have simple structures and also can ease the serious curse of dimensionality greatly. Correspondingly, new sufficient conditions for a affine TS fuzzy system as an universal approximator are obtained. The situation in 2ISO cases is derived first which is then extended to general MISO cases. Theoretical analyses comparing to some typical methods and several numerical examples all confirm that the dynamic method in this paper can reduce the fuzzy rules number dramatically and therefore can ease the serious curse of dimensionality to some extent. Some conclusions and discussions on further work are also given.","Title: Great East Japan Earthquake emergency evolution and contingency decision based on system engineering approach\nYear: 2015\n\nAbstract: In recent years, unconventional emergencies that have generated widespread concern in the society have frequently occurred. In 2011, the Great East Japan Earthquake and the nuclear crisis as a result of this disaster have gained tremendous attention worldwide. The present paper extracted unexpected incident attributes by the approach of multi-case study to describe the Great East Japan Earthquake structurally from the event type, the key attributes, the secondary attributes, the environment attributes and the hazard assessment attributes. On the basis of the structural description, we adopted a system engineering approach, an extended event graph, to describe and analyse this earthquake and its event evolution as well as associated contingency decisions. Some pertinent countermeasures are also presented for reference of similar emergency event evolution in the future.","Title: A Fast Radial Scanned Near-Field 3-D SAR Imaging System and the Reconstruction Method\nYear: 2015\n\nAbstract: This paper presents a near-field 3-D synthetic aperture radar (SAR) imaging system for which the 2-D aperture is radially scanned. Compared to the current SAR imaging systems, the proposed system has several advantages such as quick data collection, full 360° inspection of target, and simple image formation processing. However, in radial scan, the samples do not fall on a Cartesian grid, which prevents us from using the fast Fourier transform (FFT) to form SAR image without calling for interpolation. In this paper, the 2-D nonuniform FFT (NUFFT) is used for dealing with the problem. After 2-D NUFFT of the radial sampled data, the 3-D reflectivity image can be efficiently reconstructed by using the 3-D version of the range migration algorithm. The Stolt mapping has been implemented implicitly by another 1-D NUFFT to reduce the artifacts caused by the conventional interpolation processing. In addition, to alleviate the data sampling burden, a compressed 2-D slow-time sampling strategy is also discussed. Finally, the proposed Rad-SAR system and the imaging method are demonstrated using near-field wideband simulation data.","Title: A heuristic verification of the degree of the approximate GCD of two univariate polynomials\nYear: 2014\n\nAbstract: We consider the problem of computing verified real interval perturbations of the coefficients of two univariate polynomials such that there exist corresponding perturbed polynomials which have an exact greatest common divisor (GCD) of a given degree k. Based on the certification of the rank deficiency of a submatrix of the Bezout matrix of two univariate polynomials, we propose an algorithm to compute verified real perturbations. Numerical experiments show the performance of our algorithm.","Title: Adaptive KPCA Modeling of Nonlinear Systems\nYear: 2015\n\nAbstract: This paper proposes an adaptive algorithm for kernel principal component analysis (KPCA). Compared to existing work: (i) the proposed algorithm does not rely on assumptions, (ii) combines the up- and downdating step to become a single operation, (iii) the adaptation of the eigendecompsition can, computationally, reduce to O(N) and (iv) the proposed algorithm is more accurate. To demonstrate these benefits, the proposed adaptive KPCA, or AKPCA, algorithm is contrasted with existing work in terms of accuracy and efficiency. The article finally presents an application to an industrial data set showing that the adaptive algorithm allows modeling time-varying and non-stationary process behavior.","Title: Exposing Image Forgery by Detecting Traces of Feather Operation.\nYear: 2015\n\nAbstract: Powerful digital image editing tools make it very easy to produce a perfect image forgery. The feather operation is necessary when tampering an image by copy-paste operation because it can help the boundary of pasted object to blend smoothly and unobtrusively with its surroundings. We propose a blind technique capable of detecting traces of feather operation to expose image forgeries. We model the feather operation, and the pixels of feather region will present similarity in their gradient phase angle and feather radius. An effectual scheme is designed to estimate each feather region pixel's gradient phase angle and feather radius, and the pixel's similarity to its neighbor pixels is defined and used to distinguish the feathered pixels from un-feathered pixels. The degree of image credibility is defined, and it is more acceptable to evaluate the reality of one image than just using a decision of YES or NO. Results of experiments on several forgeries demonstrate the effectiveness of the technique. A blind technique of detecting traces of feather operation image forgeries.Modeling influence of feather operation by gradient phase angle and radius.An approach to expose image forgeries region by traces of feather operation.The degree of credibility is defined to evaluate the reality of one image.","Title: M-Algorithm-Based Optimal Detectors for Spatial Modulation.\nYear: 2015\n\nAbstract: A novel M-algorithm-based detector, namely M- algorithm to maximum-likelihood (M-ML), for spatial modulation (SM) receiver is proposed recently, however, due to the miscalculation of metrics, bit-error-rate (BER) performance of M-ML is far away from optimal actually. In this paper, a modified M-ML detector, which calculates accumulated metrics of each layer and can achieve optimal BER performance, is proposed; Furthermore, an M-algorithm-based constellation- reduction (MCR) method specially for SM in high correlated environment is proposed, MCR is capable of combining with maximum-likelihood (ML) and M-ML to further reduce complexity. The BER performance of M-ML and MCR is simulated, simulation results show that the proposed detectors can achieve almost the same BER performance to ML detector while reducing the computational complexity significantly. Index Terms—Spatial modulation, M-algorithm, correlated channel, maximum-likelihood","Title: A Fast Low Power Window-Opening Logic For High Speed Sar Adc\nYear: 2014\n\nAbstract: A new window-opening low-power area-efficient switching logic for high speed successive approximation register (SAR) analog-to-digital converter (ADC) is proposed in this paper. Unlike conventional SAR logic based on the shift register, the window-opening scheme minimizes the delay by putting the comparator results almost directly to DAC, and utilizes domino-based structure to reduce the capacitive load for comparator. According to pre-layout simulation in 65 nm CMOS technology, a 10 bit 100MS/s SAR ADC with the new logic achieves a logic delay of 73 ps including DAC buffer delay, which is much lower than most SAR ADC.","Title: Calibration for split capacitor DAC in SAR ADC\nYear: 2013\n\nAbstract: The nonlinearity of split capacitor DAC (CDAC) in SAR ADC is mainly caused by the mismatch and parasitic effect of the bridge capacitor. A tunable capacitor array for compensation can be a solution. This paper first analyzes the nonlinearity of the split capacitor structure, and then proves that the linear tuning method results in linearity improvements and finally, an improved tunable structure is proposed by calculating the minimal step which keeps the tuning error within 0.5LSB. The new implementation achieves smaller area and less power consumption during the calibration, while maintaining the same circuit complexity. Behavioral Simulations based on a 5b-5b split DAC show that the proposed calibration further improves SNDR and SFDR by 2.2dB and 1.6dB, respectively compared with the prototype of compensation capacitor array.","Title: Hyper Mirror System Based Collaborative Language Learning between Foreign Language Learners and Native Speakers\nYear: 2014\n\nAbstract: This study verifies the validity of collaborative language learning between foreign language learners and native speakers in distance environment. In order to provide a high presence experience in remote video conferencing system, the environment used in our experiment is based on the Hyper Mirror (HM) video conferencing system in which students communicate with an experience of being somewhat similar to stay with partners side by side. In the current study, two separate coding schemes are separately used to analyze the students' learning outcomes and their learning process in HM system environment. Experimental results demonstrate the effectiveness of our combination of collaborative language learning and HM system for the foreign language learners.","Title: Why Don't Companies Borrow? Evidence from China's Listed Companies.\nYear: 2013\n\nAbstract: In order to find out why some companies adopt an extreme capital structure of zero debt, we use the data of China's listed companies to analyze the features of those zero debt companies. Our empirical results show that some factors influence the choice of zero debt policy. Our analysis explains what kind of company may adopt zero debt, and also contribute to the research of low leverage.","Title: Impact of asymmetric transmission power on operator competition in shared spectrum\nYear: 2012\n\nAbstract: To maintain affordable access for the rapidly increasing mobile traffic, base station deployment has to be tailored to hot-spot areas and primarily indoors where facility owners, e.g., shopping malls or hotels, mostly provide wireless service. Since such local access providers (LAPs) do not have access to exclusive spectrum, one proposed option is sharing spectrum with other nearby LAPs, e.g. unlicensed or secondary spectrum. Due to limited or no coordination between the LAPs, they selfishly access the spectrum, causing harmful interference to the neighboring networks. Especially by increasing transmission power, one operator may attempt to improve its own throughput at the expense of its neighbors. In this paper, we explore the impact of power asymmetry on competition between LAPs. We model the competition between two networks with different maximum power constraints as a network-wide power control game. By analyzing the pure Nash equilibria, we find that a lower power (LP) network becomes more aggressive to overcome the inter-network interference. Due to the aggressive behavior, sharing spectrum can out-perform fixed spectrum split even for the LP network, provided that the power asymmetry is below a certain limit. On the other hand, a higher power (HP) network is mainly affected by its own “self-interference” so that it has little incentive to employ complicated inter-operator interference management schemes. In addition, we demonstrate that the power asymmetry limit strongly depends on the inter-network propagation conditions, e.g., inter-building distance or building penetration loss.","Title: A Novel SOH Prediction Framework for the Lithium-ion Battery Using Echo State Network.\nYear: 2014\n\nAbstract: Li-ion batteries provide lightweight, high energy density power sources for a variety of devices. Therefore, monitoring battery health in an effective way could increase the reliability and stability of the prediction system. So in this paper, we present a novel prediction framework based on Echo State Network to realize the prediction for battery state of health by training and testing battery impedance values and capacity values. To evaluate the proposed prediction approach, we have executed experiments with lithium-ion battery. Experimental results prove its effectiveness and confirm the estimation system can be effectively applied to the battery health state prediction. Moreover, the prediction system can run multiple data sets at a time to make the estimation process more efficient. Therefore, we can choose a battery which meets the requirement through the comparison between different batteries' prediction results.","Title: A neural network based distributed intrusion detection system on cloud platform\nYear: 2012\n\nAbstract: Intrusion detection system (IDS) is an important component to maintain network security. Also, as the cloud platform is quickly evolving and becoming more popular in our everyday life, it is useful and necessary to build an effective IDS for the cloud. However, existing intrusion detection techniques will be likely to face challenges when deployed on the cloud platform. The pre-determined IDS architecture may lead to overloading of a part of the cloud due to the extra detection overhead. This paper proposes a neural network based IDS which is a distributed system with an adaptive architecture so as to make full use of the available resources without overloading any single machine in the cloud. Moreover, with the machine learning ability from the neural network, the proposed IDS can detect new types of attacks with fairly accurate results. Evaluation of the proposed IDS with the KDD dataset on a physical cloud testbed shows that it is a promising approach to detecting attacks in the cloud infrastructure.","Title: Low-cost photolithographic fabrication of nanowires and microfilters for advanced bioassay devices.\nYear: 2015\n\nAbstract: Integrated microfluidic devices with nanosized array electrodes and microfiltration capabilities can greatly increase sensitivity and enhance automation in immunoassay devices. In this contribution, we utilize the edge-patterning method of thin aluminum (Al) films in order to form nano- to micron-sized gaps. Evaporation of high work-function metals (i.e., Au, Ag, etc.) on these gaps, followed by Al lift-off, enables the formation of electrical uniform nanowires from low-cost, plastic-based, photomasks. By replacing Al with chromium (Cr), the formation of high resolution, custom-made photomasks that are ideal for low-cost fabrication of a plurality of array devices were realized. To demonstrate the feasibility of such Cr photomasks, SU-8 micro-pillar masters were formed and replicated into PDMS to produce micron-sized filters with 3-4 mu m gaps and an aspect ratio of 3. These microfilters were capable of retaining 6 mu m beads within a localized site, while allowing solvent flow. The combination of nanowire arrays and micro-pillar filtration opens new perspectives for rapid R&D screening of various microfluidic-based immunoassay geometries, where analyte pre-concentration and highly sensitive, electrochemical detection can be readily co-localized.","Title: Spectrally-efficient direct-detection WDM transmission systems\nYear: 2015\n\nAbstract: Short- and medium-haul links in metro, access and inter-data centre networks require low cost direct-detection WDM transceivers offering dispersion tolerance, energy efficiency and high information spectral density. In this paper, we describe recent work aiming to increase the reach and spectral efficiency of such systems, using the single-sideband Nyquist subcarrier modulation (N-SCM) signal format.","Title: Frequency-Domain Backprojection Algorithm for Synthetic Aperture Radar Imaging\nYear: 2015\n\nAbstract: A new frequency-domain backprojection method for processing synthetic aperture radar (SAR) data is presented. It forms SAR images by backprojecting the spectrum of the radar echo into the image wavenumber domain and is capable of fully correcting wavefront curvature and handling a general aperture geometry. From each reconstructed spectral pixel, information about the entire scene can be obtained, allowing the formation of interim lower resolution images during the processing. In addition, the method allows for an arbitrary sampling of the image spectrum. It forms SAR images by backprojecting the spectrum of the radar echo into the image wavenumber domain. Processing of simulated and experimental ultrawideband/widebeam SAR raw data demonstrates the efficacy of this new method.","Title: Bayexer: an accurate and fast Bayesian Demultiplexer for Illumina sequences.\nYear: 2015\n\nAbstract: Demultiplexing is used after high-throughput sequencing to in silico assign reads to the samples of origin based on the sequenced reads of the indices. Existing demultiplexing tools based on the similarity between the read index and the reference index sequences may fail to provide satisfactory results on low-quality datasets. We developed Bayexer, a Bayesian demultiplexing algorithm for Illumina sequencers. Bayexer uses the information extracted directly from the contaminant sequences of the targeting reads as the training dataset for a naive Bayes classifier to assign reads. According to our evaluation, Bayexer provides higher capability, accuracy and speed on various real datasets than other tools.","Title: An immersed boundary-lattice Boltzmann method for single- and multi-component fluid flows\nYear: 2016\n\nAbstract: The paper presents a numerical method to simulate single- and multi-component fluid flows around moving/deformable solid boundaries, based on the coupling of Immersed Boundary (IB) and Lattice Boltzmann (LB) methods. The fluid domain is simulated with LB method using the single relaxation time BGK model, in which an interparticle potential model is applied for multi-component fluid flows. The IB-related force is directly calculated with the interpolated definition of the fluid macroscopic velocity on the Lagrangian points that define the immersed solid boundary. The present IB–LB method can better ensure the no-slip solid boundary condition, thanks to an improved spreading operator. The proposed method is validated through several 2D/3D single- and multi-component fluid test cases with a particular emphasis on wetting conditions on solid wall. Finally, a 3D two-fluid application case is given to show the feasibility of modeling the fluid transport via a cluster of beating cilia.","Title: On the Relationship between Model Coverage and Code Coverage Using MATLAB's Simulink\nYear: 2015\n\nAbstract: Software Testing is an approach to ensuring the quality of software systems. Testing of safety-critical systems often requires conformance to certain code coverage criteria, including for example, in aviation, Modified Condition/Decision Coverage (MC/DC). In some situations, however, access to the actual code may be restricted with black Box approaches, and testers may only be able to use models of the system, such as those in MATLAB's Simulink. Without access to the code, exact code coverage measurement may not be possible. This paper presents a method of identifying and using the Simulink model's constraints to generate test cases which can achieve high coverage of the actual source code. A case study confirming the relationship between the model's coverage and the code coverage is also presented.","Title: Energy-efficient caching for Video-on-Demand in Fixed-Mobile Convergent networks\nYear: 2015\n\nAbstract: The success of novel bandwidth-consuming multimedia services such as Video-on-Demand (VoD) is leading to a tremendous growth of the Internet traffic. Content caching can help to mitigate such uncontrolled growth by storing video content closer to the users in core, metro and access network nodes. So far, metro and especially access networks supporting mobile and fixed users have evolved independently, leveraging logically (and often also physically) separate infrastructures; this means that mobile users cannot access caches placed in the fixed access network (and vice-versa), even if they are geographically close to them, and energy consumption implications of such undesired effect must be investigated. We define an optimization problem modeling an energy-efficient placement of caches in core, metro and fixed/mobile access nodes of the network. Then, we show how the evolution towards a Fixed-Mobile Converged metro/access network, where fixed and mobile users can share caches, can reduce the energy consumed for VoD content delivery.","Title: A Model DrivenTesting Solution for Embedded System with Simulink/Stateflow Model\nYear: 2015\n\nAbstract: The Simulink/Stateflow is a commonly used approach in modeling embedded systems. This paper presents a model driven testing method for embedded systems using Simulink/Stateflow model, where a test model of a system constructed with Simulink/Stateflow is to be converted to hierarchical interface automata. Dependencies within hierarchical interface automata and test sequences are acquired through analyzing model state coverage. And then a solver tool, named Yices, is used to solve linear constraints in test sequences for the purpose of selecting test cases that conform to the constraints. The impact of states of test model is studied using test coverage criteria and the validity of the obtained test cases will also be evaluated.","Title: Cloud Service Recommendation: State of the Art and Research Challenges\nYear: 2015\n\nAbstract: Cloud computing is an attractive platform which offers on-demand resources as services. When many cloud services are available, some may have similar or same functionalities. So cloud service recommendation, which can help users to select the services based on their preferences, become an important technique for cloud services. In this paper, we review the relevant technologies that can perform cloud service recommendation. First, we introduce the web service selection and recommendation technologies. Next, we describe QoS measurement for cloud service and different recommendation methods focusing on Collaborative filtering. Third, we discuss the research challenges and opportunities for cloud service recommendation.","Title: Local Variational Bayesian Inference Using Niche Differential Evolution For Brain Magnetic Resonance Image Segmentation\nYear: 2015\n\nAbstract: Brain magnetic resonance (MR) image segmentation is pivotal for quantitative brain analyses, in which statistical models are most commonly used. However, in spite of its computational effectiveness, these models are less capable of handling the intensity non-uniformity (INU) and partial volume effect (PVE), and hence may produce less accurate results. In this paper, a novel brain MR image segmentation algorithm is proposed. To address the INU and PVE, voxel values in each small volume are characterized by a local variational Bayes (LVB) model, which is inferred by the niche differential evolution (NDE) technique to avoid local optima. A probabilistic brain atlas is constructed for each image to incorporate the anatomical prior into the segmentation process. The proposed NDE-LVB algorithm has been compared to the variational expectation-maximization based and genetic algorithm based segmentation algorithms and the segmentation routine in the widely used statistical parametric mapping package on both synthetic and clinical brain MR images. Our results suggest that the NDE-LVB algorithm can differentiate major brain tissue types more effectively and produce more accurate segmentation results.","Title: Transparent Handover Using WiFi Network Prediction for Mobile Video Streaming.\nYear: 2015\n\nAbstract: Loosely controlled Wi-Fi networks, such as deployed at festivals and events, typically do not provide a full coverage of the terrain. As a result, the user experience of a mobile user watching a video stream is rather erratic, because of frequent disconnections and re-connections. However, these periods of disconnections are generally rather short (10 - 50 seconds for a coverage area of around 70%). To ensure a continuous experience one needs to graciously handle these \"disconnection gaps\" until a client connects to the subsequent access point. In this paper we will introduce a disconnection prediction (DP) mechanism and pre-fetcher (PF) to successfully initiate a transparent handover ensuring a continuous user experience for video streaming. The DP predicts disconnections and re-connections based on a collaboratively generated connectivity map. This information is used to activate the pre-fetcher, which will start to download video material to the users device, that will be needed during the disconnected period. When the mobile user physically disconnects from the network, the video will continue to play as the requested data has been buffered on the mobile device. When the user re-enters a connected area, the wireless network connection is restored and operation is resumed as normal.","Title: COAST: A Connected Open plAtform for Smart objecTs\nYear: 2015\n\nAbstract: Quick reactions and rescues are critical requirements in disaster and emergency management. However, the decision on how to react should be taken based on the correct judgment on the disaster situation. Nowadays ICT plays a pivotal role in the decision-making in disasters management. Particularly, the IoT technology is indispensable for information collection and analysis. Despite the rapid development in the ICT domain, the real-time analysis and response in certain emergency is still impossible. This is mainly due to the lack of computation power of thin smart objects and the vertical isolation among IoT applications. In this work, we will introduce the Connected Open plAtform for Smart objecTs (COAST) that gives smart objects, such as robots, drones and smartphones, the capability to autonomously negotiate and deploy additional resources on itself, on other objects or in the cloud. The overall goal of the COAST platform is to give the capability to smart object applications to self-adapt to the runtime context by providing an interoperable platform for real-time discovery, negotiation and on-the-fly allocation of additional resources. These resources can be provided by remote sensors, by on-demand deployment of additional - possibly 3rd party - services on the same or another smart object, or by outsourcing computation to edge cloud infrastructure tailored for high-performance and real-time computation. COAST provides the necessary platform services supporting run-time adaptation, monitoring, big data analytics. To allow the secure deployment of 3rd party application components on smart objects, the COAST provides a novel virtualization layer that supports high-performance execution through hardware accelerators. Expected performance will be reported to highlight the benefits of introducing such a platform to the IoT services in disaster management.","Title: MagicBox: An intelligent caching-enabled mobile router/server\nYear: 2015\n\nAbstract: Information Centric Network (ICN) has been recently studied to support future content-oriented services by designing new network architectures and protocols. In this paper, we demonstrate our implementation of a mobile router/server built on a ICN platform. This demo called MagicBox enables content caching and prefetching for personal and professional uses. Particularly, we detail how our system will be showcased in big events where Internet access and media content can be efficiently shared.","Title: ICN based shared caching in future converged fixed and mobile network\nYear: 2015\n\nAbstract: The explosion of mobile multimedia and Internet-of-things (IoT) services implies strong requirements for seamless switching among various types of networks. Thus, to offer true ubiquitous Internet connection, a Fixed and Mobile Converged (FMC) network architecture is essential for the future 5G network. Such a convergent network can not only improve the utilization of network resources, but also inspire new add-on services for FMC network operators. In this paper, we introduce a shared caching overlay based on Information Centric Networking (ICN). It is deployed on top of the FMC network and controlled by the FMC network operator to offer Caching as a Service (CaaS) to Over-The-Top (OTT) service providers and virtual network operators. Business analysis and performance evaluation will highlight the benefits of deploying such a controlled Shared Caching System (SCS) over an FMC network.","Title: Random Time Delay Effect on Out-of-Sequence Measurements.\nYear: 2016\n\nAbstract: The communication links of multi-agent systems (MASs) generate random time delays, which significantly affect accuracy and real-time control. A Markov chain is established at the input end and output end of a communication link to express this time delay and is introduced to asynchronous data fusion at an intelligent location. The influence of random time delays and data fusion on the positioning accuracy and real-time performance is investigated. A distributed fusion simulation system is established based on MASs and employed to simulate the random delay in the fusion of multi-sensor data. Different fusion algorithms are tested by changing the upper boundary of the time delay. The test results indicate the existence of a boundary between the data fusion and the time delay. The fusion accuracies on both sides of the boundary significantly differ. This difference is helpful for selecting sensors and fusion algorithms for different sampling frequencies in multi-agent collaborative control systems, improving the positioning accuracy of the system and effectively controlling computation costs.","Title: An algorithm for the discretization of an ideal projector.\nYear: 2016\n\nAbstract: Ideal interpolation is a generalization of the univariate Hermite interpolation. It is well known that every univariate Hermite interpolant is a pointwise limit of some Lagrange interpolants. However, a counterexample provided by Shekhtman Boris shows that, for more than two variables, there exist ideal interpolants that are not the limit of any Lagrange interpolants. So it is natural to consider: Given an ideal interpolant, how to find a sequence of Lagrange interpolants (if any) that converge to it. The authors call this problem the discretization for ideal interpolation. This paper presents an algorithm to solve the discretization problem. If the algorithm returns “True”, the authors get a set of pairwise distinct points such that the corresponding Lagrange interpolants converge to the given ideal interpolant.","Title: Dscnn: Hardware-Oriented Optimization For Stochastic Computing Based Deep Convolutional Neural Networks\nYear: 2016\n\nAbstract: Deep Convolutional Neural Networks (DCNN), a branch of Deep Neural Networks which use the deep graph with multiple processing layers, enables the convolutional model to finely abstract the high-level features behind an image. Large-scale applications using DCNN mainly operate in high-performance server clusters, GPUs or FPGA clusters; it is restricted to extend the applications onto mobile/wearable devices and Internet-of-Things (IoT) entities due to high power/energy consumption. Stochastic Computing is a promising method to overcome this shortcoming used in specific hardware-based systems. Many complex arithmetic operations can be implemented with very simple hardware logic in the SC framework, which alleviates the extensive computation complexity. The exploration of network-wise optimization and the revision of network structure with respect to stochastic computing based hardware design have not been discussed in previous work. In this paper, we investigate Deep Stochastic Convolutional Neural Network (DSCNN) for DCNN using stochastic computing. The essential calculation components using SC are designed and evaluated. We propose a joint optimization method to collaborate components guaranteeing a high calculation accuracy in each stage of the network. The structure of original DSCNN is revised to accommodate SC hardware design's simplicity. Experimental Results show that as opposed to software inspired feature extraction block in DSCNN, an optimized hardware oriented feature extraction block achieves as higher as 59.27% calculation precision. And the optimized DSCNN can achieve only 3.48% network test error rate compared to 27.83% for baseline DSCNN using software inspired feature extraction block.","Title: A dynamic equivalent method of wind farm considering wind farm dispersity and wind turbine difference.\nYear: 2016\n\nAbstract: With large scale wind farm integration, the effect of wind farm dynamic equivalent model accuracy is increasing on power system dynamic and transient stability, so the traditional equivalent and multi-group method can't meet requirement of some studies. Therefore, due to the deficiency of traditional wind farm equivalent method, a new multi-group method is proposed in this paper. Wind farms dispersity and wind turbines difference have been both considered in this method. When grouping wind farm turbines, based on that, a modified wind farm dynamic equivalent model is proposed using coherency method. For the contrastive analysis, three traditional equivalent models are built too, and simulated research of single, two and three phases short circuit is proceeded. Using relative error index of active power, reactive power and bus voltage, dynamic characteristics contrastive analysis of four equivalent models including the proposed model is preceded too. The superiority of this proposed method is verified by simulation results, and dynamic characteristics of the proposed model located on the parallel point are more close to that of real wind farm.","Title: Noncircular Measurement and Mitigation of I/Q Imbalance for OFDM-Based WLAN Transmitters.\nYear: 2017\n\nAbstract: In future high-speed communication networks, the in-phase/quadrature (I/Q) imbalance mitigation and oscillator drift compensation is a key issue in the design of orthogonal frequency division multiplexing (OFDM)-based wireless LAN (WLAN) transmitters. To this end, we propose a two-stage I/Q imbalance measurement method, where by virtue of the WLAN standard-compliant training sequences, a coarse I/...","Title: Template Development for Web-based Simulation Analysis of Wire Drawing Dies.\nYear: 2016\n\nAbstract: Based on the \\\"abstract modeling\\\" and \\\"tag technique\\\", a template which can achieve automatic simulation, modeling and calculation for the drawing process of wire drawing dies was developed in the present work. The template, featured by process automation, has a wider application scope than the customized development, and will greatly improve the efficiency of those simulation analyses involving many repetitive operations. The template has been integrated to the Web, through which users can download tag files. The next steps for them are to add tags to the simulation model and submit simulation parameters. Then computing will be executed at the cloud background. By virtue of this mode, the users no longer need to pay much for software and hardware resources, and it will be more accessible to the wire drawing die simulation analysis. Thus, the analysis based on computer aided engineering (CAE) can also be applied by the users unskilled in CAE software.","Title: Toward Automatic Review of Building Energy Efficiency Based on Building Information Modeling\nYear: 2016\n\nAbstract: Building energy consumption increases with the acceleration process of urban constructions. The city faces the problem of energy shortage, which leads to the building energy efficiency becoming an urgent need for the global environment. BIM (Building Information Modeling) technology has been widely used in the construction industry. The current building energy consumption calculation depending on that the BIM model is imported to different energy consumption software for energy consumption analysis. There are few methods to review building energy efficiency requirements automatically. In view of this, this paper discusses the method of review of building energy consumption in Revit platform. The feasibility of the method is proved through an office building construction. The method can automatically review whether design results of the building meet the building energy efficiency requirements in the design process and can shorten the review time.","Title: Discovery of Novel Phosphodiesterase-2A Inhibitors by Structure-Based Virtual Screening, Structural Optimization, and Bioassay.\nYear: 2017\n\nAbstract: Phosphodiesterase-2A (PDE2A) is a potential therapeutic target for treatment of Alzheimer's disease and pulmonary hypertension. However, most of the current PDE2A inhibitors have moderate selectivity over other PDEs. In the present study, we described the discovery of novel PDE2A inhibitors by structure-based virtual screening combining pharmacophore model screening, molecular docking, molecular dynamics simulations, and bioassay validation, Nine hits out of 30 molecules from the SPECS database (a hit rate of 30%) inhibited PDE2A with affinity less than 50 Optimization of compound AQ-390/10779040 (IC50 = 4.6 mu M) from the virtual screening, which holds a novel scaffold of benzo[cd]indol-2(1H)-one among PDE inhibitors, leads to discovery of a new compound LHB-8 with a significant improvement of inhibition (IC50 = 570 nM). The modeling studies demonstrated' that LHB-8 formed an extra hydrogen bond with Asp808 and a hydrophobic interaction with Thr768, in addition to the common interactions with Gln859 and Phe862 of PDE2A. The novel scaffolds discovered in the present study can be used for rational design of PDE2A inhibitors with high affinity.","Title: Robust Passivity Control for 2-D Uncertain Markovian Jump Linear Discrete-Time Systems.\nYear: 2017\n\nAbstract: This paper discusses the problem of robust controller design for two-dimensional (2-D) Markovian jump linear systems. The problem is demonstrated using Fornasini Marchesini local state-space models, which are affected by uncertainties. The transition-mode probability matrix is homogenous and known. It is assumed that the mode information is available for the controller design and implementation. Then, a mode-dependent state-feedback controller is proposed. By substituting the controller into the 2-D system, a stochastic closed-loop system is obtained, because the stochastic variable, external disturbance, and uncertainties are all included in the closed-loop system. Based on the analysis results, an approach to design the controller and its gains is proposed, and the gains are calculated by solving linear matrix inequalities. In section V, a 2-D case is used to verify the performance of the controller.","Title: SEP-Nets: Small and Effective Pattern Networks.\nYear: 2017\n\nAbstract: While going deeper has been witnessed to improve the performance of convolutional neural networks (CNN), going smaller for CNN has received increasing attention recently due to its attractiveness for mobile/embedded applications. It remains an active and important topic how to design a small network while retaining the performance of large and deep CNNs (e.g., Inception Nets, ResNets). Albeit there are already intensive studies on compressing the size of CNNs, the considerable drop of performance is still a key concern in many designs. This paper addresses this concern with several new contributions. First, we propose a simple yet powerful method for compressing the size of deep CNNs based on parameter binarization. The striking difference from most previous work on parameter binarization/quantization lies at different treatments of $1times 1$ convolutions and $ktimes k$ convolutions ($ku003e1$), where we only binarize $ktimes k$ convolutions into binary patterns. The resulting networks are referred to as pattern networks. By doing this, we show that previous deep CNNs such as GoogLeNet and Inception-type Nets can be compressed dramatically with marginal drop in performance. Second, in light of the different functionalities of $1times 1$ (data projection/transformation) and $ktimes k$ convolutions (pattern extraction), we propose a new block structure codenamed the pattern residual block that adds transformed feature maps generated by $1times 1$ convolutions to the pattern feature maps generated by $ktimes k$ convolutions, based on which we design a small network with $sim 1$ million parameters. Combining with our parameter binarization, we achieve better performance on ImageNet than using similar sized networks including recently released Google MobileNets.","Title: A Hierarchical Framework of Cloud Resource Allocation and Power Management Using Deep Reinforcement Learning\nYear: 2017\n\nAbstract: Automatic decision-making approaches, such as reinforcement learning (RL), have been applied to (partially) solve the resource allocation problem adaptively in the cloud computing system. However, a complete cloud resource allocation framework exhibits high dimensions in state and action spaces, which prohibit the usefulness of traditional RL techniques. In addition, high power consumption has become one of the critical concerns in design and control of cloud computing systems, which degrades system reliability and increases cooling cost. An effective dynamic power management (DPM) policy should minimize power consumption while maintaining performance degradation within an acceptable level. Thus, a joint virtual machine (VM) resource allocation and power management framework is critical to the overall cloud computing system. Moreover, novel solution framework is necessary to address the even higher dimensions in state and action spaces. In this paper, we propose a novel hierarchical framework for solving the overall resource allocation and power management problem in cloud computing systems. The proposed hierarchical framework comprises a global tier for VM resource allocation to the servers and a local tier for distributed power management of local servers. The emerging deep reinforcement learning (DRL) technique, which can deal with complicated control problems with large state space, is adopted to solve the global tier problem. Furthermore, an autoencoder and a novel weight sharing structure are adopted to handle the high-dimensional state space and accelerate the convergence speed. On the other hand, the local tier of distributed server power managements comprises an LSTM based workload predictor and a model-free RL based power manager, operating in a distributed manner. Experiment results using actual Google cluster traces show that our proposed hierarchical framework significantly saves the power consumption and energy usage than the baseline while achieving no severe latency degradation. Meanwhile, the proposed framework can achieve the best trade-off between latency and power/energy consumption in a server cluster.","Title: A novel computer-aided diagnosis system for breast MRI based on feature selection and ensemble learning.\nYear: 2017\n\nAbstract: Breast cancer is a common cancer among women. With the development of modern medical science and information technology, medical imaging techniques have an increasingly important role in the early detection and diagnosis of breast cancer. In this paper, we propose an automated computer-aided diagnosis (CADx) framework for magnetic resonance imaging (MRI). The scheme consists of an ensemble of several machine learning-based techniques, including ensemble under-sampling (EUS) for imbalanced data processing, the Relief algorithm for feature selection, the subspace method for providing data diversity, and Adaboost for improving the performance of base classifiers. We extracted morphological, various texture, and Gabor features. To clarify the feature subsets' physical meaning, subspaces are built by combining morphological features with each kind of texture or Gabor feature. We tested our proposal using a manually segmented Region of Interest (ROI) data set, which contains 438 images of malignant tumors and 1898 images of normal tissues or benign tumors. Our proposal achieves an area under the ROC curve (AUC) value of 0.9617, which outperforms most other state-of-the-art breast MRI CADx systems. Compared with other methods, our proposal significantly reduces the false-positive classification rate. HighlightsThe dimensionality of the features we use is larger than most state-of-the-art methods. Various features including morphological, gabor, and several types of texture features are extracted to make a comprehensive characterization of breast masses.We select the optimal feature subset from the original feature set with Relief according to their type, which helps diminish the redundant and irrelevant features as well as take the physical meaning of features into consideration.We propose a novel ensemble learning framework based on the combination of EUS, subspace and Adaboost, which helps alleviate the data imbalance problem and improve the overall classification accuracy of the whole CADx system.","Title: Correlation-based fault-tolerant controller design for MIMO systems under the Youla parameterisation.\nYear: 2017\n\nAbstract: This paper develops a correlation-based method into the Youla parameterisation structure for a fault-tolerant controller design strategy. By tuning the Youla parameters with the proposed correlation-based algorithm, a number of conditional faults described by the dual Youla parameters are attenuated. The traditional fault-tolerant control FTC schemes under the Youla parameterisation often require the gradient information of the defined cost function for minimisation, which is either tedious or even unfeasible with unknown fault model. However, the proposed correlation-based FTC algorithm in this paper can compensate the faults via system data without the explicit fault model or the cost function gradient information. It is also proved that the algorithm convergence can be achieved without identifying the unknown fault model. For illustration, a simulation example with corresponding comparisons are presented to show the effectiveness of the proposed method in the end.","Title: DroneMyo: Proactive Control of Unmanned Aerial Vehicle Based on Wearable Devices\nYear: 2015\n\nAbstract: Wearable devices promise a full immersion in the usage environment and most of these devices are mainly passive information receivers. We believe that integration of proactive wearable devices in general can bring more advantages for more natural user interactions. This paper explores proactive control of Unmanned Aerial Vehicle (UAV) based on a smart phone and a smart wristband. We have conducted indoor and outdoor trial flights and gained encouraging results as well as valuable ideas for future research.","Title: Structural design optimization for deep convolutional neural networks using stochastic computing.\nYear: 2017\n\nAbstract: Deep Convolutional Neural Networks (DCNNs) have been demonstrated as effective models for understanding image content. The computation behind DCNNs highly relies on the capability of hardware resources due to the deep structure. DCNNs have been implemented on different large-scale computing platforms. However, there is a trend that DCNNs have been embedded into light-weight local systems, which requires low power/energy consumptions and small hardware footprints. Stochastic Computing (SC) radically simplifies the hardware implementation of arithmetic units and has the potential to satisfy the small low-power needs of DCNNs. Local connectivities and down-sampling operations have made DCNNs more complex to be implemented using SC. In this paper, eight feature extraction designs for DCNNs using SC in two groups are explored and optimized in detail from the perspective of calculation precision, where we permute two SC implementations for inner-product calculation, two down-sampling schemes, and two structures of DCNN neurons. We evaluate the network in aspects of network accuracy and hardware performance for each DCNN using one feature extraction design out of eight. Through exploration and optimization, the accuracies of SC-based DCNNs are guaranteed compared with software implementations on CPU/GPU/binary-based ASIC synthesis, while area, power, and energy are significantly reduced by up to 776X, 190X, and 32835X.","Title: Softmax Regression Design for Stochastic Computing Based Deep Convolutional Neural Networks.\nYear: 2017\n\nAbstract: Recently, Deep Convolutional Neural Networks (DCNNs) have made tremendous advances, achieving close to or even better accuracy than human-level perception in various tasks. Stochastic Computing (SC), as an alternate to the conventional binary computing paradigm, has the potential to enable massively parallel and highly scalable hardware implementations of DCNNs. In this paper, we design and optimize the SC based Softmax Regression function. Experiment results show that compared with a binary SR, the proposed SC-SR under longer bit stream can reach the same level of accuracy with the improvement of 295X, 62X, 2617X in terms of power, area and energy, respectively. Binary SR is suggested for future DCNNs with short bit stream length input whereas SC-SR is recommended for longer bit stream.","Title: Research on particle swarm optimization of variable parameter.\nYear: 2016\n\nAbstract: Aimed at particle swarm optimization, since there are a fewer adjustable parameters, when solving the multi-dimensional function, it is easy to meet premature convergence problem, so an improved particle swarm optimization of variable parameters is proposed. According to particle movement characteristics, the formula of particle velocity updating is improved to make all integrated into the corresponding weight factor; through weight factor, the particle optimization perfornance is adjusted. Three standard test functions are used for test, with comparison with other algorithms, and the simulation results show that by setting different weight factors, the proposed algorithm has better optimization precision and ability to execute, and the better result can be achieved when solving the multi-dimensional finction.","Title: Research on Directional AF Cooperative Communication System Based on Outage Probability.\nYear: 2016\n\nAbstract: In order to improve outage property of cooperative communication system, directional antenna technology could be introduced. The paper proposes single relay AF cooperative communication protocol based on directional antenna and deduces the outage probability of system in fully directional mode theoretically so as to deduce the lower bound. With the minimum system outage probability to be the target, the paper analyzes OPA (Optimum Power Allocation) and makes detailed analysis and comparison of the system property of the established model by combining with specific simulation value. The simulation result shows that on the condition of high SNR (Signal to Noise Ratio) or greater difference of channel coefficient between two terminals and relay, the lower bound and accurate value are approximated and the outage property of fully directional cooperative communication is superior to that of traditional mode. Under the situation of limited power, OPA can enhance the outage property of system effectively.","Title: A Two-Stage Approach for Learning a Sparse Model with Sharp Excess Risk Analysis.\nYear: 2017\n\nAbstract: This paper aims to provide a sharp excess risk guarantee for learning a sparse linear model without any assumptions about the strong convexity of the expected loss and the sparsity of the optimal solution in hindsight. Given a target level for the excess risk, an interesting question to ask is how many examples and how large the support set of the solution are enough for learning a good model with the target excess risk. To answer these questions, we present a two-stage algorithm that (i) in the first stage an epoch based stochastic optimization algorithm is exploited with an established O(1/epsilon) bound on the sample complexity; and (ii) in the second stage a distribution dependent randomized sparsification is presented with an O(1/epsilon) bound on the sparsity (referred to as support complexity) of the resulting model. Compared to previous works, our contributions lie at (i) we reduce the order of the sample complexity from O(1/epsilon(2)) to O(1/epsilon) without the strong convexity assumption; and (ii) we reduce the constant in O(1/epsilon) for the sparsity by exploring the distribution dependent sampling.","Title: Directional AF Cooperative Communication System Based on Outage Probability\nYear: 2017\n\nAbstract: In order to improve outage property of cooperative communication system, directional antenna technology could be introduced. The paper proposes single relay AF cooperative communication protocol based on directional antenna and deduces the outage probability of system in fully directional mode theoretically so as to deduce the lower bound. With the minimum system outage probability to be the target, the paper analyzes OPA Optimum Power Allocation and makes detailed analysis and comparison of the system property of the established model by combining with specific simulation value. The simulation result shows that on the condition of high SNR Signal to Noise Ratio or greater difference of channel coefficient between two terminals and relay, the lower bound and accurate value are approximated and the outage property of fully directional cooperative communication is superior to that of traditional mode. Under the situation of limited power, OPA can enhance the outage property of system effectively.","Title: Adaptive Ensemble Undersampling-Boost: A novel learning framework for imbalanced data.\nYear: 2017\n\nAbstract: •We propose a framework for imbalanced classification tasks with ensemble learning.•We propose a weight modification method to give each classifier an adaptive weight.•We introduce OTSU algorithm to determine the optimal threshold adaptively.•We make comparisons between our proposals and several state-of-the-art algorithms.","Title: Simulation of 2-DOF Articulated Robot Control Based on Adaptive Fuzzy Sliding Mode Control.\nYear: 2017\n\nAbstract: About the control of articulated robot, researchers put out an adaptive fuzzy sliding mode control algorithm by analyzing the classical sliding mode algorithm. There is an adaptive single input and output fuzzy to calculate the control gain. Meanwhile, researchers designed a adaptive law which is based on Lyapunov theory, and simulated the adaptive fuzzy sliding mode control in Simulink. Simulink results show that when the chattering exiting becomes weaker, the function of system is stronger. Since the adaptive algorithm joined, fuzzy sliding mode control can be different with the change of system’s state and adjusted automatically. Steady-state convergence is constant, adaptive fuzzy sliding mode control algorithm still with a good robustness under the condition that articulated robot’s parameters uncertain and external interference.","Title: A Single, Continuously Applied Control Policy for Modeling Reaching Movements with and without Perturbation.\nYear: 2018\n\nAbstract: It has been debated whether kinematic features, such as the number of peaks or decomposed submovements in a velocity profile, indicate the number of discrete motor impulses or result from a continuous control process. The debate is particularly relevant for tasks involving target perturbation, which can alter movement kinematics. To simulate such tasks, finite-horizon models require two preset mov...","Title: CirCNN: Accelerating and Compressing Deep Neural Networks Using Block-Circulant Weight Matrices\nYear: 2017\n\nAbstract: Large-scale deep neural networks (DNNs) are both compute and memory intensive. As the size of DNNs continues to grow, it is critical to improve the energy efficiency and performance while maintaining accuracy. For DNNs, the model size is an important factor affecting performance, scalability and energy efficiency. Weight pruning achieves good compression ratios but suffers from three drawbacks: 1) the irregular network structure after pruning, which affects performance and throughput; 2) the increased training complexity; and 3) the lack of rigirous guarantee of compression ratio and inference accuracy. To overcome these limitations, this paper proposes CIRCNN, a principled approach to represent weights and process neural networks using block-circulant matrices. CIRCNN utilizes the Fast Fourier Transform (FFT)-based fast multiplication, simultaneously reducing the computational complexity (both in inference and training) from O(n\n<sup xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">2</sup>\n) to O(n log n) and the storage complexity from O(n\n<sup xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">2</sup>\n) to O(n), with negligible accuracy loss. Compared to other approaches, CIRCNN is distinct due to its mathematical rigor: the DNNs based on CIRCNN can converge to the same “effectiveness” as DNNs without compression. We propose the CIRCNN architecture, a universal DNN inference engine that can be implemented in various hardware/software platforms with configurable network architecture (e.g., layer type, size, scales, etc.). In CIRCNN architecture: 1) Due to the recursive property, FFT can be used as the key computing kernel, which ensures universal and small-footprint implementations. 2) The compressed but regular network structure avoids the pitfalls of the network pruning and facilitates high performance and throughput with highly pipelined and parallel design. To demonstrate the performance and energy efficiency, we test CmCNN in FPGA, ASIC and embedded processors. Our results show that CIRCNN architecture achieves very high energy efficiency and=performance with a small hardware footprint. Based on the FPGA implementation and ASIC synthesis results, CIRCNN achieves 6 - 102X energy efficiency improvements compared with the best state-of-the-art results.","Title: Brain voxel classification in magnetic resonance images using niche differential evolution based Bayesian inference of variational mixture of Gaussians.\nYear: 2017\n\nAbstract: Classification of brain voxels into gray matter, white matter, and cerebrospinal fluid (CSF) using magnetic resonance imaging (MRI) is pivotal for quantitative brain analyses. In spite of its computational effectiveness, the most commonly used statistical classification models are less capable of handling the intensity non-uniformity (INU) and partial volume effect (PVE), and hence may produce less accurate results. In this paper, we propose a novel approach, namely the VMG-NDE algorithm, to improve brain voxel classification in MRI images by considering all effects simultaneously. There are four planks in this algorithm, including (1) using variational mixture of Gaussians (VMG) model to characterize the variation of voxel values caused by PVE, (2) training a cohort of local VMG models on small data volumes extracted from the image to reduce the impact of INU, (3) employing the niche differential evolution (NDE) to infer each local VMG model, aiming to avoid falling into local optima, and (4) constructing a probabilistic brain atlas for each study and using it to incorporate the anatomy prior into the classification process. After training local VMG models, we classify each brain voxel using a linear combination of the predictions generated by all those models. This algorithm has been evaluated against the variational expectation-maximization based and genetic algorithm based segmentation algorithms and the segmentation routines in the widely used statistical parametric mapping (SPM) package, expectation-maximization segmentation (EMS) package and FSL package on both synthetic and clinical T1-weighted brain MRI studies. Our results suggest that the proposed algorithm can differentiate major brain tissue types more effectively and produce improved brain voxel classification accuracy.","Title: On the Closest Vector Problem for Lattices Constructed from Polynomials and Their Cryptographic Applications.\nYear: 2017\n\nAbstract: In this paper, we propose new classes of trapdoor functions to solve the closest vector problem in lattices. Specifically, we construct lattices based on properties of polynomials for which the closest vector problem is hard to solve unless some trapdoor information is revealed. We thoroughly analyze the security of our proposed functions using state-of-the-art attacks and results on lattice reductions. Finally, we describe how our functions can be used to design quantum-safe encryption schemes with reasonable public key sizes. In particular, our scheme can offer around $106$ bits of security with a public key size of around $6.4$ $texttt{KB}$. Our encryption schemes are efficient with respect to key generation, encryption and decryption.","Title: A Simple Nonlinearity-Tailored Probabilistic Shaping Distribution For Square Qam\nYear: 2017\n\nAbstract: A new probabilistic shaping distribution that outperforms Maxwell-Boltzmann is studied for the nonlinear fiber channel. Additional gains of 0.1 bit/symbol MI or 0.2 dB SNR for both DP-256QAM and DP-1024QAM are reported after 200 km nonlinear fiber transmission.","Title: Energy-efficient, high-performance, highly-compressed deep neural network design using block-circulant matrices.\nYear: 2017\n\nAbstract: Deep neural networks (DNNs) have emerged as the most powerful machine learning technique in numerous artificial intelligent applications. However, the large sizes of DNNs make themselves both computation and memory intensive, thereby limiting the hardware performance of dedicated DNN accelerators. In this paper, we propose a holistic framework for energy-efficient high-performance highly-compressed DNN hardware design. First, we propose block-circulant matrix-based DNN training and inference schemes, which theoretically guarantee Big-O complexity reduction in both computational cost (from O(n2) to O(n log n)) and storage requirement (from O(n2) to O(n)) of DNNs. Second, we dedicatedly optimize the hardware architecture, especially on the key fast Fourier transform (FFT) module, to improve the overall performance in terms of energy efficiency, computation performance and resource cost. Third, we propose a design flow to perform hardware-software co-optimization with the purpose of achieving good balance between test accuracy and hardware performance of DNNs. Based on the proposed design flow, two block-circulant matrix-based DNNs on two different datasets are implemented and evaluated on FPGA. The fixed-point quantization and the proposed block-circulant matrix-based inference scheme enables the network to achieve as high as 3.5 TOPS computation performance and 3.69 TOPS/W energy efficiency while the memory is saved by 108X ∼ 116X with negligible accuracy degradation.","Title: Trajectory Optimization-Based Auxiliary Power Unit Control Strategy for an Extended Range Electric Vehicle.\nYear: 2017\n\nAbstract: In the pursuit of fuel economy improvement and emission reduction, a trajectory optimization-based engine multioperating-point control strategy is proposed in this paper for the auxiliary power unit (APU) in an extended range electric vehicle. This strategy design deeply considers the transient process of the operating point switching that somehow influences the engine fuel economy and emission, a...","Title: Evaluate the Malignancy of Pulmonary Nodules Using the 3D Deep Leaky Noisy-or Network.\nYear: 2017\n\nAbstract: Automatic diagnosing lung cancer from computed tomography scans involves two steps: detect all suspicious lesions (pulmonary nodules) and evaluate the whole-lung/pulmonary malignancy. Currently, there are many studies about the first step, but few about the second step. Since the existence of nodule does not definitely indicate cancer, and the morphology of nodule has a complicated relationship with cancer, the diagnosis of lung cancer demands careful investigations on every suspicious nodule and integration of information of all nodules. We propose a 3-D deep neural network to solve this problem. The model consists of two modules. The first one is a 3-D region proposal network for nodule detection, which outputs all suspicious nodules for a subject. The second one selects the top five nodules based on the detection confidence, evaluates their cancer probabilities, and combines them with a leaky noisy-or gate to obtain the probability of lung cancer for the subject. The two modules share the same backbone network, a modified U-net. The overfitting caused by the shortage of the training data is alleviated by training the two modules alternately. The proposed model won the first place in the Data Science Bowl 2017 competition.","Title: Designing reconfigurable large-scale deep learning systems using stochastic computing\nYear: 2016\n\nAbstract: Deep Learning, as an important branch of machine learning and neural network, is playing an increasingly important role in a number of fields like computer vision, natural language processing, etc. However, large-scale deep learning systems mainly operate in high-performance server clusters, thus restricting the application extensions to personal or mobile devices. The solution proposed in this paper is taking advantage of the fantastic features of stochastic computing methods. Stochastic computing is a type of data representation and processing technique, which uses a binary bit stream to represent a probability number (by counting the number of ones in this bit stream). In the stochastic computing area, some key arithmetic operations such as additions or multiplications can be implemented with very simple components like AND gates or multiplexers, respectively. Thus it provides an immense design space for integrating a large amount of neurons and enabling fully parallel and scalable hardware implementations of large-scale deep learning systems. In this paper, we present a reconfigurable large-scale deep learning system based on stochastic computing technologies, including the design of the neuron, the convolution function, the back-propagation function and some other basic operations. And the network-on-chip technique is also proposed in this paper to achieve the goal of implementing a large-scale hardware system. Our experiments validate the functionality of reconfigurable deep learning systems using stochastic computing, and demonstrate that when the bit streams are set to be 8192 bits, classification of MNIST digits by stochastic computing can perform as low error rate as that by normal arithmetic operations.","Title: SIMD acceleration for HEVC encoding on DSP.\nYear: 2017\n\nAbstract: As the new generation video coding standard, High Efficient Video Coding (HEVC) significantly improves the video compression efficiency, which is however at the cost of a far more computational payload than the capacity of real-time video applications and general purpose processors. In this paper, we focus on the SIMD-based fast implementation of the HEVC encoder over modern TI Digital Signal Processors (DSPs). We first test the DSP-based HEVC encoder and indentify the most time-consuming encoding modules. Then SIMD instructions are exploited to improve the parallel computing capacity of these modules and thus speed up the encoder. The experimental results show that the proposed implementations can significantly improve the encoding speed of the DSP-based HEVC encoder, with a speedup ratio of 8.38-87.32 over the original C-based encoder and 1.59-6.56 over o3-optimization enabled encoder.","Title: Adaptive fuzzy sliding mode control algorithm simulation for 2-DOF articulated robot.\nYear: 2017\n\nAbstract: About articulated robotu0027s control, chattering phenomenon existed in conventional sliding mode algorithm. Therefore a robotic adaptive fuzzy sliding mode control algorithm is also proposed. This kind of algorithm is designed as an adaptive single input and output fuzzy system to calculate the control gain. Meanwhile, the mathematical proof is carried out on systemu0027s stability and astringency by Lyapunov theoryu0027s adaptive law. The adaptive fuzzy sliding mode control is simulated in Simulink, the simulation results show that when the chattering becomes weaker, the function of the system is promoted. Fuzzy sliding mode controller can be changed and adjusted with the transformation of systemu0027s state. Steady-state convergence is constant, adaptive fuzzy sliding mode control algorithm still with a good robustness under the condition that articulated robotu0027s parameters uncertain and external interference.","Title: Towards Ultra-High Performance and Energy Efficiency of Deep Learning Systems: An Algorithm-Hardware Co-Optimization Framework.\nYear: 2018\n\nAbstract: Hardware accelerations of deep learning systems have been extensively investigated in industry and academia. The aim of this paper is to achieve ultra-high energy efficiency and performance for hardware implementations of deep neural networks (DNNs). An algorithm-hardware co-optimization framework is developed, which is applicable to different DNN types, sizes, and application scenarios. The algorithm part adopts the general block-circulant matrices to achieve a fine-grained tradeoff of accuracy and compression ratio. It applies to both fully-connected and convolutional layers and contains a mathematically rigorous proof of the effectiveness of the method. The proposed algorithm reduces computational complexity per layer from O(n(2)) to O(n log n) and storage complexity from O(n(2)) to O(n), both for training and inference. The hardware part consists of highly efficient Field Programmable Gate Array (FPGA)-based implementations using effective reconfiguration, batch processing, deep pipelining, resource re-using, and hierarchical control. Experimental results demonstrate that the proposed framework achieves at least 152X speedup and 71X energy efficiency gain compared with IBM TrueNorth processor under the same test accuracy. It achieves at least 31X energy efficiency gain compared with the reference FPGA-based work.","Title: Ser Performance Optimisation Of Af Cooperative Communication System Based On Directional Antenna\nYear: 2018\n\nAbstract: Aiming at the cooperative communication system with directional antenna, this paper has studied the SER (Symbol Error Rate) performance under AF (Amplify-and-Forward protocol). The model of AF cooperative communication system using directional antenna was firstly established to deduce the closed-form expression of SER in this model as well as the upper limit of SER. Then, the OPA (Optimum Power Allocation) was also analysed on the purpose of minimising SER. Combining specific simulation numerical values, SER performance of established model was thoroughly researched in this paper. Simulation results demonstrate that system SER obviously decreases by adopting cooperative communication system with directional transmitting and directional receiving. Each node's directional gain, channel quality and power allocation method all have great influence on system's overall performance. And the OPA is also proved to be superior to EPA (Equal Power Allocation).","Title: A Framework in CRM Customer Lifecycle: Identify Downward Trend and Potential Issues Detection.\nYear: 2018\n\nAbstract: Customer retention is one of the primary goals in the area of customer relationship management. A mass of work exists in which machine learning models or business rules are established to predict churn. However, targeting users at an early stage when they start to show a downward trend is a better strategy. In downward trend prediction, the reasons why customers show a downward trend is of great interest in the industry as it helps the business to understand the pain points that customers suffer and to take early action to prevent them from churning. A commonly used method is to collect feedback from customers by either aggressively reaching out to them or by passively hearing from them. However, it is believed that there are a large number of customers who have unpleasant experiences and never speak out. In the literature, there is limited research work that provides a comprehensive and scientific approach to identify these suffers. In this study, we propose a novel two-part framework: developing the downward prediction process and establishing the methodology to identify the reasons why customers are in the downward trend. In the first prediction part, we focus on predicting the downward trend, which is an earlier stage of the customer lifecycle compared to churn. In the second part, we propose an approach to figuring out the cause (of the downward trend) based on a causal inference method and semi-supervised learning. The proposed approach is capable of identifying potential silent sufferers. We take bad shopping experiences as inputs to develop the framework and validate it via a marketing A/B test in the real world. The test readout demonstrates the effectiveness of the framework by driving 88.5% incremental lift in purchase volume.","Title: Efficient Recurrent Neural Networks using Structured Matrices in FPGAs.\nYear: 2018\n\nAbstract: Recurrent Neural Networks (RNNs) are becoming increasingly important for time series-related applications which require efficient and real-time implementations. The recent pruning based work ESE suffers from degradation of performance/energy efficiency due to the irregular network structure after pruning. We propose block-circulant matrices for weight matrix representation in RNNs, thereby achieving simultaneous model compression and acceleration. We aim to implement RNNs in FPGA with highest performance and energy efficiency, with certain accuracy requirement (negligible accuracy degradation). Experimental results on actual FPGA deployments shows that the proposed framework achieves a maximum energy efficiency improvement of 35.7$times$ compared with ESE.","Title: Learning Topics Using Semantic Locality\nYear: 2018\n\nAbstract: The topic modeling discovers the latent topic probability of the given text documents. To generate the more meaningful topic that better represents the given document, we proposed a new feature extraction technique which can be used in the data preprocessing stage. The method consists of three steps. First, it generates the word/word-pair from every single document. Second, it applies a two-way TF-IDF algorithm to word/word-pair for semantic filtering. Third, it uses the K-means algorithm to merge the word pairs that have the similar semantic meaning.Experiments are carried out on the Open Movie Database (OMDb), Reuters Dataset and 20NewsGroup Dataset. The mean Average Precision score is used as the evaluation metric. Comparing our results with other state-of-the-art topic models, such as Latent Dirichlet allocation and traditional Restricted Boltzmann Machines. Our proposed data preprocessing can improve the generated topic accuracy by up to 12.99%.","Title: Design of Low-Power WiNoC with Congestion-Aware Wireless Node.\nYear: 2018\n\nAbstract: As the key component of wireless network-on-chip (WiNoC), wireless router is required to handle a large number of data packets which could cause network congestion. The congestion not only reduces the network performance, but also results in additional power consumption. In this paper, a low-power WiNoC with congestion-aware wireless node is proposed. Firstly, flit counter unit and address resolution unit are added in wireless interface (WI), which can sense the congestion information and destination address information of each wireless node dynamically. Secondly, through the proposed congestion judgment algorithm, the congestion judgment unit in global network can judge the priority of each wireless node pair and set the highest-priority wireless node pair to use the channel resources preferentially. The mechanism can effectively alleviate the congestion of wireless nodes and ensure the stability of system performance. In addition, a sleep mechanism is utilized to switch off WI which fails in the wireless channel competition. Experimental results show that the proposed scheme can effectively improve the performance in terms of packets' transmission latency and network throughput.","Title: Error Formulas for Lagrange Projectors Determined by Cartesian Sets.\nYear: 2018\n\nAbstract: This paper studies error formulas for Lagrange projectors determined by Cartesian sets. Cartesian sets are properly subgrids of tensor product grids. Given interpolated functions with all order continuous partial derivatives, the authors directly construct the good error formulas for Lagrange projectors determined by Cartesian sets. Owing to the special algebraic structure, such a good error formula is useful for error estimate.","Title: Diaphragm-Free Fiber-Optic Fabry-Perot Interferometric Gas Pressure Sensor for High Temperature Application.\nYear: 2018\n\nAbstract: A diaphragm-free fiber-optic Fabry-Perot (FP) interferometric gas pressure sensor is designed and experimentally verified in this paper. The FP cavity was fabricated by inserting a well-cut fiber Bragg grating (FBG) and hollow silica tube (HST) from both sides into a silica casing. The FP cavity length between the ends of the SMF and HST changes with the gas density. Using temperature decoupling method to improve the accuracy of the pressure sensor in high temperature environments. An experimental system for measuring the pressure under different temperatures was established to verify the performance of the sensor. The pressure sensitivity of the FP gas pressure sensor is 4.28 nm/MPa with a high linear pressure response over the range of 0.1-0.7 MPa, and the temperature sensitivity is 14.8 pm/degrees C under the range of 20-800 degrees C. The sensor has less than 1.5% non-linearity at different temperatures by using temperature decoupling method. The simple fabrication and low-cost will help sensor to maintain the excellent features required by pressure measurement in high temperature applications.","Title: VBI-MRF model for image segmentation.\nYear: 2018\n\nAbstract: In statistical image segmentation, the distribution of pixel values is usually assumed to be Gaussian and the optimal result is believed to be the one that has maximum a posteriori (MAP) probability. In spite of its prevalence and computational efficiency, the Gaussian assumption, however, is not always strictly followed, and hence may lead to less accurate results. Although the variational Bayes inference (VBI), in which statistical model parameters are also assumed to be random variables, has been widely used, it can hardly handle the spatial information embedded in pixels. In this paper, we incorporate spatial smoothness constraints on pixels labels interpreted by the Markov random field (MRF) model into the VBI process, and thus propose a novel statistical model called VBI-MRF for image segmentation. We evaluated our algorithm against the variational expectation-maximization (VEM) algorithm and the hidden Markov random field (HMRF) model and MAP-MRF model based algorithms on both noise-corrupted synthetic images and mosaics of natural texture. Our pilot results suggest that the proposed algorithm can segment images more accurately than other three methods and is capable of producing robust image segmentation.","Title: A subspace predictive control method for partially unknown systems with parameter learning event-triggered law.\nYear: 2018\n\nAbstract: This paper proposes a novel event-triggered subspace predictive control (SPC) method for a class of linear discrete-time partially unknown systems. Without the complete system parameter information, the design parameters of the event-triggered law are first derived via system data by the reinforcement learning method. The proposed event-triggered law depends on the defined input error and the state-dependent threshold. The receding horizon principle in the typical predictive control methods is substituted by the event-triggered law, which can ensure the stability of the predictive input with optimality. The proposed method can considerably reduce the data computation and transmission load of the conventional SPC methods. The simulation results illustrate the effect and the satisfactory performance of the proposed method.","Title: An Augmented Nonlinear LMS for Digital Self-Interference Cancellation in Full-Duplex Direct-Conversion Transceivers.\nYear: 2018\n\nAbstract: In future full-duplex communications, the cancellation of self-interference (SI) arising from hardware nonidealities will play an important role in the design of mobile-scale devices. To this end, we introduce an optimal digital SI cancellation solution for shared-antenna-based direct-conversion transceivers. To establish that the underlying widely linear signal model is not adequate for strong tr...","Title: Towards Budget-Driven Hardware Optimization for Deep Convolutional Neural Networks Using Stochastic Computing\nYear: 2018\n\nAbstract: Recently, Deep Convolutional Neural Network (DCNN) has achieved tremendous success in many machine learning applications. Nevertheless, the deep structure has brought significant increases in computation complexity. Largescale deep learning systems mainly operate in high-performance server clusters, thus restricting the application extensions to personal or mobile devices. Previous works on GPU and/or FPGA acceleration for DCNNs show increasing speedup, but ignore other constraints, such as area, power, and energy. Stochastic Computing (SC), as a unique data representation and processing technique, has the potential to enable the design of fully parallel and scalable hardware implementations of large-scale deep learning systems. This paper proposed an automatic design allocation algorithm driven by budget requirement considering overall accuracy performance. This systematic method enables the automatic design of a DCNN where all design parameters are jointly optimized. Experimental results demonstrate that proposed algorithm can achieve a joint optimization of all design parameters given the comprehensive budget of a DCNN.","Title: Verified error bounds for eigenvalues of geometric multiplicity and corresponding invariant subspaces\nYear: 2018\n\nAbstract: In this paper, we generalize the algorithm described by Rump and Graillat to compute verified and narrow error bounds such that a slightly perturbed matrix is guaranteed to have an eigenvalue with geometric multiplicity q within computed error bounds. The corresponding invariant subspace can be directly obtained by our algorithm. Our verification method is based on border matrix technique. We demonstrate the performance of our algorithm for matrices of dimension up to hundreds with non-defective and defective eigenvalues.","Title: A Unified Analysis of Stochastic Momentum Methods for Deep Learning.\nYear: 2018\n\nAbstract: Stochastic momentum methods have been widely adopted in training deep neural networks. However, their theoretical analysis of convergence of the training objective and the generalization error for prediction is still under-explored. This paper aims to bridge the gap between practice and theory by analyzing the stochastic gradient (SG) method, and the stochastic momentum methods including two famous variants, i.e., the stochastic heavy-ball (SHB) method and the stochastic variant of Nesterovu0027s accelerated gradient (SNAG) method. We propose a framework that unifies the three variants. We then derive the convergence rates of the norm of gradient for the non-convex optimization problem, and analyze the generalization performance through the uniform stability approach. Particularly, the convergence analysis of the training objective exhibits that SHB and SNAG have no advantage over SG. However, the stability analysis shows that the momentum term can improve the stability of the learned model and hence improve the generalization performance. These theoretical insights verify the common wisdom and are also corroborated by our empirical analysis on deep learning.","Title: An Aggressive Genetic Programming Approach for Searching Neural Network Structure Under Computational Constraints.\nYear: 2018\n\nAbstract: Recently, there emerged revived interests of designing automatic programs (e.g., using genetic/evolutionary algorithms) to optimize the structure of Convolutional Neural Networks (CNNs) for a specific task. The challenge in designing such programs lies in how to balance between large search space of the network structures and high computational costs. Existing works either impose strong restrictions on the search space or use enormous computing resources. In this paper, we study how to design a genetic programming approach for optimizing the structure of a CNN for a given task under limited computational resources yet without imposing strong restrictions on the search space. To reduce the computational costs, we propose two general strategies that are observed to be helpful: (i) aggressively selecting strongest individuals for survival and reproduction, and killing weaker individuals at a very early age; (ii) increasing mutation frequency to encourage diversity and faster evolution. The combined strategy with additional optimization techniques allows us to explore a large search space but with affordable computational costs. Our results on standard benchmark datasets (MNIST, SVHN, CIFAR-10, CIFAR-100) are competitive to similar approaches with significantly reduced computational costs.","Title: Application of Federal Kalman Filter with Neural Networks in the Velocity and Attitude Matching of Transfer Alignment.\nYear: 2018\n\nAbstract: The centralized Kalman filter is always applied in the velocity and attitude matching of Transfer Alignment (TA). But the centralized Kalman has many disadvantages, such as large amount of calculation, poor real-time performance, and low reliability. In the paper, the federal Kalman filter (FKF) based on neural networks is used in the velocity and attitude matching of TA, the Kalman filter is adjusted by the neural networks in the two subfilters, the federal filter is used to fuse the information of the two subfilters, and the global suboptimal state estimation is obtained. The result of simulation shows that the federal Kalman filter based on neural networks is better in estimating the initial attitude misalignment angle of inertial navigation system (INS) when the system dynamic model and noise statistics characteristics of inertial navigation system are unclear, and the estimation error is smaller and the accuracy is higher.","Title: A New Variable-Oriented Propagation Scheme for Constraint Satisfaction Problem.\nYear: 2018\n\nAbstract: The coarse-grained arc consistency algorithms maintain a list of elements that need to be revised. It is well known that the performance of such algorithms is affected by the way in which the revisions are performed. In this paper, we propose a new variable-oriented propagation scheme that splits the propagation procedure into two independent phases. We show how it reduces the number of revisions and list operations. In experiments, we apply different revision ordering heuristics to this new variable-oriented scheme and compare them with the most efficient existing propagation schemes. The results from various structured and random problems show that the proposed propagation scheme reduces the number of revisions and offers speed-ups.","Title: Multiscale Comparative Evaluation of the GPM IMERG v5 and TRMM 3B42 v7 Precipitation Products from 2015 to 2017 over a Climate Transition Area of China.\nYear: 2018\n\nAbstract: The performance of the latest released Integrated Multi-satellitE Retrievals for GPM mission (IMERG) version 5 (IMERG v5) and the TRMM Multisatellite Precipitation Analysis 3B42 version 7 (3B42 v7) are evaluated and compared at multiple temporal scales over a semi-humid to humid climate transition area (Huaihe River basin) from 2015 to 2017. The impacts of rainfall rate, latitude and elevation on precipitation detection skills are also investigated. Results indicate that both satellite estimates showed a high Pearson correlation coefficient (r, above 0.89) with gauge observations, and an overestimation of precipitation at monthly and annual scales. Mean daily precipitation of IMERG v5 and 3B42 v7 display a consistent spatial pattern, and both characterize the observed precipitation distribution well, but 3B42 v7 tends to markedly overestimate precipitation over water bodies. Both satellite precipitation products overestimate rainfalls with intensity ranging from 0.5 to 25 mm/day, but tend to underestimate light (0-0.5 mm/day) and heavy (>25 mm/day) rainfalls, especially for torrential rains (above 100 mm/day). Regarding each gauge station, the IMERG v5 has larger mean r (0.36 for GPM, 0.33 for TRMM) and lower mean relative root mean square error (RRMSE, 1.73 for GPM, 1.88 for TRMM) than those of 3B42 v7. The higher probability of detection (POD), critical success index (CSI) and lower false alarm ratio (FAR) of IMERG v5 than those of 3B42 v7 at different rainfall rates indicates that IMERG v5 in general performs better in detecting the observed precipitations. This study provides a better understanding of the spatiotemporal distribution of accuracy of IMERG v5 and 3B42 v7 precipitation and the influencing factors, which is of great significance to hydrological applications.","Title: Widely Linear CLMS Based Cancelation of Nonlinear Self -Interference in Full-Duplex Direct-Conversion Transceivers.\nYear: 2018\n\nAbstract: An augmented nonlinear complex LMS (ANCLMS) algorithm is proposed to adaptively mitigate both the linear and nonlinear self-interference (SI) components in a full-duplex direct-conversion transceiver (DCT). A data prewhitening scheme, which exploits the known SI signal distributions, is also adopted to accelerate the convergence. Theoretical mean and mean square performance evaluations of the proposed SI canceller are performed and fully support the proposed approach. Computer simulations on wireless local area network (WLAN) standard compliant waveforms in practical full-duplex (FD) direct-conversion transceiver settings support the analysis.","Title: Towards the next generation of multi-criteria recommender systems.\nYear: 2018\n\nAbstract: This paper presents the motivation, concepts, ideas and research questions underlying a PhD research project in the domain of recommender systems, and more specifically on multi-criteria recommendation. While we build on the existing work in this direction, we aim at introducing recommendation frameworks that do not only optimize for different criteria simultaneously, but also exploit their interrelations. For this aim, we will address three multi-criteria recommendation challenges, namely multi-modal user and item modeling, package recommendation, and user-centric recommendation. For realizing these frameworks, and in particular, for learning interactions and interrelations in the criteria space, we will rely on the state-of-the-art deep learning systems, and in particular the Generative Adversarial Networks (GANs). In addition, a novel evaluation strategy for multi-criteria recommendation targeting the maximization of the user's satisfaction will also be devised.","Title: A data-driven covert attack strategy in the closed-loop cyber-physical systems.\nYear: 2018\n\nAbstract: In this paper, a data-driven covert attack strategy is proposed for a class of closed-loop cyber-physical systems. Without the parameters of the system plant and the nominal controller, the attacker can only use the intercepted input and output data of the nominal system. The injected input attack signals are designed based on the subspace predictive control method, which can deviate the real outputs to the expected attack references in a future time horizon. Meanwhile, by injecting the designed output attack signals for compensation, the attack cannot be detected by the anomaly detector. The simulation results of an irrigation canal system illustrate the effect of the proposed strategy with satisfactory performances.","Title: A Fast Modular Inversion FPGA Implementation over GF(2m) using Modified x2n Unit.\nYear: 2018\n\nAbstract: Modular inversion is an important step in Elliptic Curve Cryptography (ECC). Itoh-Tsujiiu0027s algorithm (ITA) is a commonly applied modular inversion algorithm. This paper proposes a novel architecture of modular inversion implementation based on ITA. The modular multiplier and 2 n units are modified into non-iterative logic to increase clock frequency. Least-clock-cycle-ITA and High-speed-ITA architectures are proposed to minimize latency in different situations. Both architectures are implemented on FPGA Virtex-5. LCC-ITA completes modular inversion in 9 clock cycles with maximum clock frequency 126.1MHz, while HS-ITA completes two modular inversions in 20 clock cycles with maximum clock frequency 177.6MHz. The performance of the proposed architectures is 56% and 134% higher than the best work before.","Title: CSA-DE/EDA: A Clonal Selection Algorithm Using Differential Evolution and Estimation of Distribution Algorithm.\nYear: 2018\n\nAbstract: The clonal selection algorithm (CSA), which describes the basic features of an immune response to an antigenic stimulus, has drawn a lot of research attention in the bio-inspired computing community, due to its highly-adaptive and easy-to-implement nature. However, despite many successful applications, this optimization technique still suffers from limited ability to explore the solution space. In this paper, we incorporate the differential evolution (DE) and estimation of distribution algorithm (EDA) into CSA, and thus propose a novel bio-inspired computing algorithm called CSA-DE/EDA. In this algorithm, the hypermutaion and receptor editing processes are implemented based on DE and EDA, which provide improved local and global search ability, respectively. We have applied this algorithm to brain image segmentation. Our comparative experimental results suggest that the proposed CSA-DE/EDA algorithm outperforms several bio-inspired computing techniques on the segmentation problem.","Title: Generative Adversarial Active Learning for Unsupervised Outlier Detection.\nYear: 2018\n\nAbstract: Outlier detection is an important topic in machine learning and has been used in a wide range of applications. In this paper, we approach outlier detection as a binary-classification issue by sampling potential outliers from a uniform reference distribution. However, due to the sparsity of data in high-dimensional space, a limited number of potential outliers may fail to provide sufficient information to assist the classifier in describing a boundary that can separate outliers from normal data effectively. To address this, we propose a novel Single-Objective Generative Adversarial Active Learning (SO-GAAL) method for outlier detection, which can directly generate informative potential outliers based on the mini-max game between a generator and a discriminator. Moreover, to prevent the generator from falling into the mode collapsing problem, the stop node of training should be determined when SO-GAAL is able to provide sufficient information. But without any prior information, it is extremely difficult for SO-GAAL. Therefore, we expand the network structure of SO-GAAL from a single generator to multiple generators with different objectives (MO-GAAL), which can generate a reasonable reference distribution for the whole dataset. We empirically compare the proposed approach with several state-of-the-art outlier detection methods on both synthetic and real-world datasets. The results show that MO-GAAL outperforms its competitors in the majority of cases, especially for datasets with various cluster types or high irrelevant variable ratio. The experiment codes are available at: \n<uri xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">https://github.com/leibinghe/GAAL-based-outlier-detection</uri>\n.","Title: Fast similarity search on keyword-induced point groups.\nYear: 2018\n\nAbstract: Location-based social media (e.g., Twitter, Foursquare) have been generating massive amount of geo-textual data. In this paper, we represent the spatial distribution of a keyword by the group of locations tagged with such keyword. Given a query keyword, our problem is to find k keywords with the most similar distribution of locations. Such query finds applications in targeted marketing and recommendation. The performance of existing solutions degrade when different point groups have significant overlapping, which happens rather frequently in real data. We propose efficient techniques to process similarity search on point groups. Experimental results on Twitter data demonstrate that our solution is faster than the state-of-the-art by up to 6 times.","Title: Co-Estimation of State of Charge and State of Health for Lithium-Ion Batteries Based on Fractional-Order Calculus.\nYear: 2018\n\nAbstract: Lithium-ion batteries have emerged as the state-of-the-art energy storage for portable electronics, electrified vehicles, and smart grids. An enabling Battery Management System holds the key for efficient and reliable system operation, in which State-of-Charge (SOC) estimation and State-of-Health (SOH) monitoring are of particular importance. In this paper, an SOC and SOH co-estimation scheme is p...","Title: E-RNN: Design Optimization for Efficient Recurrent Neural Networks in FPGAs.\nYear: 2019\n\nAbstract: Recurrent Neural Networks (RNNs) are becoming increasingly important for time series-related applications which require efficient and real-time implementations. The two major types are Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks. It is a challenging task to have real-time, efficient, and accurate hardware RNN implementations because of the high sensitivity to imprecision accumulation and the requirement of special activation function implementations. A key limitation of the prior works is the lack of a systematic design optimization framework of RNN model and hardware implementations, especially when the block size (or compression ratio) should be jointly optimized with RNN type, layer size, etc. In this paper, we adopt the block-circulant matrix-based framework, and present the Efficient RNN (E-RNN) framework for FPGA implementations of the Automatic Speech Recognition (ASR) application. The overall goal is to improve performance/energy efficiency under accuracy requirement. We use the alternating direction method of multipliers (ADMM) technique for more accurate block-circulant training, and present two design explorations providing guidance on block size and reducing RNN training trials. Based on the two observations, we decompose E-RNN in two phases: Phase I on determining RNN model to reduce computation and storage subject to accuracy requirement, and Phase II on hardware implementations given RNN model, including processing element design/optimization, quantization, activation implementation, etc. Experimental results on actual FPGA deployments show that E-RNN achieves a maximum energy efficiency improvement of 37.4$times$ compared with ESE, and more than 2$times$ compared with C-LSTM, under the same accuracy.","Title: A deep learning driven method for fault classification and degradation assessment in mechanical equipment.\nYear: 2019\n\nAbstract: •Deep learning method helps improve performance of degradation assessment.•Wavelet packet decomposition was leveraged to represent the features in both time and frequency domain.•Vibration-based condition monitoring techniques are discussed.•Comprehensive comparison was provided to validate the superiority of deep neural network in the experiment.•The superiority of deep learning in current industrial applications is interpreted.","Title: Rotation-Free Online Handwritten Chinese Character Recognition using Two-Stage Convolutional Neural Network\nYear: 2018\n\nAbstract: Handwritten Chinese character recognition (HCCR) is a major research area in pattern recognition. Although recent years have seen tremendous improvement in the use of deep neural networks in handwriting recognition, unconstrained handwriting recognition remains an open problem. Considering the problem of rotated handwritten Chinese character recognition, we propose a two-stage convolutional neural network combined with path signature features to achieve high-accuracy rotation-free HCCR. For handwritten Chinese characters whose rotation angle is up to ±45°, the recognition rate can reach 97.38% on the ICDAR-2013 online HCCR competition dataset, which is comparable to the accuracy achieved by several state-of-the-art methods for non-rotated characters, thus showing the effectiveness of the proposed method.","Title: The knowledge flow analysis on multimedia information using evolutionary game model.\nYear: 2019\n\nAbstract: Knowledge flow of multimedia information can trigger the formation of alliances and clusters. Our study focuses on the conditions when alliances or clusters emerge subsequent to the perception of knowledge flow. Based on several fundamental assumptions, we build an evolutionary game model for quantitative calculation and further conclude several propositions via geometrical analysis. The findings show that when originators participate in games without perception to the outflowing of knowledge via multimedia, the similarity, complementarity and spillage of knowledge all facilitate alliances formation after spillovers, and when originators participate in games with perception to the outflowing information, alliance formation is still positively related to the similarity and complementarity of knowledge, while the effect of spillage depends on initial conditions. This study not only analyzes the multimedia information from knowledge spillover perspective, but also introduces the evolutionary game model into the exploration of multimedia information flow, thus it provides novel guidance for the further research.","Title: Electromechanical Coupling Mechanism and Control Strategy for In-Wheel-Motor-Driven Electric Vehicles.\nYear: 2019\n\nAbstract: This paper presents a control methodology for solving the vibration issues emerged in in-wheel motor electric vehicles (IWM-EVs). Unlike existing techniques and methods, the proposed investigation focuses on the electromechanical coupling effects between the subsystems in IWM-EV, which were considered as another negative effects bought by power integration. To this aim, an integrated model which d...","Title: A Novel Four-Point Model Based Unit-Norm Constrained Least Squares Method For Single-Tone Frequency Estimation\nYear: 2019\n\nAbstract: A novel time-series relationship among four consecutive real-valued single-tone sinusoid samples is proposed based on their linear prediction property. In order to achieve unbiased frequency estimates for a real sinusoid in white noise, based on the proposed four-point time-series relationship, a constrained least squares cost function is minimized based on the unit-norm principle. Closed-form expressions for the variance and the asymptotic expression for the variance of the proposed frequency estimator are derived, facilitating a theoretical performance comparison with the existing three-point counterpart, called as the reformed Pisarenko harmonic decomposer (RPHD). The region of performance advantage of the proposed four-point based constrained least squares frequency estimator over the RPHD is also discussed. Computer simulations are conducted to support our theoretical development and to compare the proposed estimator performance with the RPHD as well as the Cramer-Rao lower bound (CRLB).","Title: Parameters Design Of A Sliding Mode Controller With A Hysteresis Band For Buck Converters In Continuous And Discontinuous Conduction Modes\nYear: 2016\n\nAbstract: In this paper, a parameter design method of a sliding mode controller with a hysteresis band is proposed for buck converter application, which is based on the proper analysis of the controller parameters' influences on existence condition, conduction mode and transient performance. This work discovers the fact that the trajectory of the closed loop system will end at an undesired limit cycle or equilibrium in discontinuous conduction mode, and establishes the mathematical relationship between the system overshoot and the sliding coefficient a. Next, in order to keep operating in continuous conduction mode and have a superior steady-state and transient performance, the upper limit of the hysteresis band border delta, the upper and lower limit of a are given accordingly. Finally, The effectiveness of the given design method is validated by simulation and experiment for steady-state and transient responses.","Title: Real-Time Image Transmission Algorithm In Wsn With Limited Bandwidth\nYear: 2018\n\nAbstract: In this paper, a real-time image transmission algorithm in WSN with limited bandwidth networks is studied. Firstly, a simple and effective monitoring network architecture is established, which allows multiple video monitoring nodes to access the network, and the data transmission is controlled by the synchronization mechanism without collision. Then, the image data is compressed locally at the monitoring nodes (over 85%), so that the image of each node can meet the needs of real-time data transmission, and the overall power consumption of the system is greatly reduced. Finally, based on NVIDIA TX1, four test nodes are constructed to test the algorithm cumulatively, which verifies the effectiveness of the system framework and compression algorithm.","Title: Infrared-Based Short-Distance Fso Sensor Network System\nYear: 2018\n\nAbstract: Energy consumption is one of the major problems in the development of Wireless Sensor Network (WSN). Under certain application conditions, Free Space Optical (FSO) communication can effectively reduce system energy consumption. Combining the need of the short-distance communication in the cooperative control of the mobile robot, this paper focuses on the study for the infrared FSO communication method and devices. Eight infrared-transmitting and receiving modules with the angular spacing 45 were deployed to realize the omnidirectional communication. The systematic measurement shows that the designed communication system can effectively carry out data transmission within a distance of 14 meters when the bit error ratio (BER) is 0%. Through the link budget analysis, the theoretical transmission distance of the design system is about 24 meters, only accounting for 60% of the communication module data manual, and the BER rises rapidly as the distance increases. The analysis of the oscilloscope observation results indicates that due to the attenuation caused by each medium (mainly moisture, dust, etc.) in the propagation path, the BER rapidly increases when the distance increases. Therefore, based on the test results, this system is applicable for constructing a short-distance wireless sensor network to meet the needs of cooperative control of mobile robots, without involving the complicated pointing process.","Title: Performance Improvement Of Electronic Dispersion Post-Compensation In Direct Detection Systems Using Dsp-Based Receiver Linearization\nYear: 2017\n\nAbstract: Significant improvements in the performance of electronic dispersion post-compensation enabled by beating interference compensation were experimentally demonstrated in a 112 Gb/s/lambda WDM Nyquist-subcarrier modulation direct-detection system in transmission over distances up to 240 km.","Title: Normalization and dropout for stochastic computing-based deep convolutional neural networks\nYear: 2019\n\nAbstract: Recently, Deep Convolutional Neural Network (DCNN) has been recognized as the most effective model for pattern recognition and classification tasks. With the fast growing Internet of Things (IoTs) and wearable devices, it becomes attractive to implement DCNNs in embedded and portable systems. However, novel computing paradigms are urgently required to deploy DCNNs that have huge power consumptions and complex topologies in systems with limited area and power supply. Recent works have demonstrated that Stochastic Computing (SC) can radically simplify the hardware implementation of arithmetic units and has the potential to bring the success of DCNNs to embedded systems. This paper introduces normalization and dropout, which are essential techniques for the state-of-the-art DCNNs, to the existing SC-based DCNN frameworks. In this work, the feature extraction block of DCNNs is implemented using an approximate parallel counter, a near-max pooling block and an SC-based rectified linear activation unit. A novel SC-based normalization design is proposed, which includes a square and summation unit, an activation unit and a division unit. The dropout technique is integrated into the training phase and the learned weights are adjusted during the hardware implementation. Experimental results on AlexNet with the ImageNet dataset show that the SC-based DCNN with the proposed normalization and dropout techniques achieves 3.26% top-1 accuracy improvement and 3.05% top-5 accuracy improvement compared with the SC-based DCNN without these two essential techniques, confirming the effectiveness of our normalization and dropout designs.","Title: A data-driven method based on deep belief networks for backlash error prediction in machining centers\nYear: 2020\n\nAbstract: Backlash error occurs in a machining center may lead to a series of changes in the geometry of the components and subsequently deteriorate the overall performance of the equipment. Due to the uncertainty of mechanical wear between kinematic pairs, it is challenging to predict backlash error through physical models directly. An alternative method is to leverage data-driven models to map the degradation. This paper proposes a data-driven method for backlash error predication through Deep Belief Network (DBN). The proposed method focuses on the assessment of both current and future geometric errors for backlash error prediction and subsequent maintenance in machining centers. During the process of prognosis, a DBN via stacking Restricted Boltzmann Machines is constructed for backlash error prediction. Energy-based models enable DBN to mine information hidden behind highly coupled inputs, which makes DBN a feasible method for fault diagnosis and prognosis when the target condition is beyond the historical data. In the experiment, to confirm the effectiveness of deep learning for backlash error prediction, similar popular regression methods, including Support Vector Machine Regression and Back Propagation Neural Network, are employed to present a comprehensive comparison in both diagnosis and prognosis. The experimental results show that the performances of all these regression methods are acceptable in the diagnostic stage. In the prognostic stage, DBN demonstrates its superiority and significantly outperforms the other models for backlash error prediction in machining centers.","Title: Multi-hypothesis-Based Error Concealment for Whole Frame Loss in HEVC.\nYear: 2018\n\nAbstract: In video transmissions over wired and wireless networks, packet losses occurs frequently due to channel errors and/or network latency, which may result in the whole video frame losses and severe video quality degradation. To address this problem, many error concealment algorithms have been proposed in order to reduce the effect of channel errors. However, most of these algorithms only focus on concealment of a missing block rather than a missing frame, which is far more difficult than the former since intra-frame information are no longer available in concealing the error. In this paper, we propose a practical multi-hypothesis based error concealment algorithm to recover the whole missing frame. This algorithm first reconstructs a motion vector field for each coding tree unit in the lost video frame according to the motion vector relationship between adjacent coding unit levels, and then the lost frame is reconstructed based on the theory of multi-hypothesis using each motion vector in the reconstructed motion vector field. The experimental results show that the proposed algorithm can effectively reconstruct the lost frames and significantly improve the video quality under frame loss.","Title: Radio Frequency Fingerprint Identification Method in Wireless Communication\nYear: 2017\n\nAbstract: The Radio frequency fingerprinting (RFF) generation mechanism is analyzed in this paper. It is proved to be a secure means for network security access. At the same time, the method of RFF extraction is also given. The characteristics of RFF are analyzed theoretically. Then, a high-precision fingerprint feature identification method based on Kalman filter is proposed. The results of the experiments show that the proposed system can work effectively in the environment where the signal-to-noise ratio (SNR) is higher than 10 dB, and the achieved identification rate is higher than 90%.","Title: C-LSTM: Enabling Efficient LSTM using Structured Compression Techniques on FPGAs.\nYear: 2018\n\nAbstract: Recently, significant accuracy improvement has been achieved for acoustic recognition systems by increasing the model size of Long Short-Term Memory (LSTM) networks. Unfortunately, the ever-increasing size of LSTM model leads to inefficient designs on FPGAs due to the limited on-chip resources. The previous work proposes to use a pruning based compression technique to reduce the model size and thus speedups the inference on FPGAs. However, the random nature of the pruning technique transforms the dense matrices of the model to highly unstructured sparse ones, which leads to unbalanced computation and irregular memory accesses and thus hurts the overall performance and energy efficiency.\n\n In contrast, we propose to use a structured compression technique which could not only reduce the LSTM model size but also eliminate the irregularities of computation and memory accesses. This approach employs block-circulant instead of sparse matrices to compress weight matrices and reduces the storage requirement from $\\mathcalO (k^2)$ to $\\mathcalO (k)$. Fast Fourier Transform algorithm is utilized to further accelerate the inference by reducing the computational complexity from $\\mathcalO (k^2)$ to $\\mathcalO (k\\textlog k)$. The datapath and activation functions are quantized as 16-bit to improve the resource utilization. More importantly, we propose a comprehensive framework called C-LSTM to automatically optimize and implement a wide range of LSTM variants on FPGAs. According to the experimental results, C-LSTM achieves up to 18.8X and 33.5X gains for performance and energy efficiency compared with the state-of-the-art LSTM implementation under the same experimental setup, and the accuracy degradation is very small.","Title: C3PO: Database and Benchmark for Early-stage Malicious Activity Detection in 3D Printing.\nYear: 2018\n\nAbstract: Increasing malicious users have sought practices to leverage 3D printing technology to produce unlawful tools in criminal activities. Current regulations are inadequate to deal with the rapid growth of 3D printers. It is of vital importance to enable 3D printers to identify the objects to be printed, so that the manufacturing procedure of an illegal weapon can be terminated at the early stage. Deep learning yields significant rises in performance in the object recognition tasks. However, the lack of large-scale databases in 3D printing domain stalls the advancement of automatic illegal weapon recognition. This paper presents a new 3D printing image database, namely C3PO, which compromises two subsets for the different system working scenarios. We extract images from the numerical control programming code files of 22 3D models, and then categorize the images into 10 distinct labels. The first set consists of 62,200 images which represent the object projections on the three planes in a Cartesian coordinate system. And the second sets consists of sequences of total 671,677 images to simulate the camerasu0027 captures of the printed objects. Importantly, we demonstrate that the weapons can be recognized in either scenario using deep learning based approaches using our proposed database. % We also use the trained deep models to build a prototype of object-aware 3D printer. The quantitative results are promising, and the future exploration of the database and the crime prevention in 3D printing are demanding tasks.","Title: Performance Of Digital Back-Propagation In Kramers-Kronig Direct-Detection Receivers\nYear: 2018\n\nAbstract: We report the first investigation of fiber nonlinearity compensation by digital backpropagation in Kramers-Kronig direct-detection receivers. Improvements in performance of 112 Gb/s/WDM transmission over single-span SSMF links of up to 160 km are demonstrated.","Title: EIGEN: Ecologically-Inspired GENetic Approach for Neural Network Structure Searching.\nYear: 2018\n\nAbstract: Designing the structure of neural networks is considered one of the most challenging tasks in deep learning. Recently, a few approaches have been proposed to automatically search for the optimal structure of neural networks, however, they suffer from either prohibitive computation cost (e.g., 256 Hours on 250 GPU in [1]) or unsatisfactory performance compared to those of hand-crafted neural networks. In this paper, we propose an Ecologically-Inspired GENetic approach for neural network structure search (EIGEN), that includes succession, mimicry and gene duplication. Specifically, we first use primary succession to rapidly evolve a community of poor initialized neural network structures into a more diverse community, followed by a secondary succession stage for fine-grained searching based on the networks from the primary succession. Extinction is applied in both stages to reduce computational cost. Mimicry is employed during the entire evolution process to help the inferior networks imitate the behavior of a superior network and gene duplication is utilized to duplicate the learned blocks of novel structures, both of which help to find the better network structures. Extensive experimental results show that our proposed approach can achieve the similar or better performance compared to the existing genetic approaches with dramatically reduced computation cost. For example, the network discovered by our approach on CIFAR-100 dataset achieves 78.1% test accuracy under 120 GPU hours, compared to 77.0% test accuracy in more than 65, 536 GPU hours in [1].","Title: HEIF: Highly Efficient Stochastic Computing based Inference Framework for Deep Neural Networks\nYear: 2019\n\nAbstract: Deep convolutional neural networks (DCNNs) are one of the most promising deep learning techniques and have been recognized as the dominant approach for almost all recognition and detection tasks. The computation of DCNNs is memory intensive due to large feature maps and neuron connections, and the performance highly depends on the capability of hardware resources. With the recent trend of wearable devices and Internet of Things, it becomes desirable to integrate the DCNNs onto embedded and portable devices that require low power and energy consumptions and small hardware footprints. Recently stochastic computing (SC)-DCNN demonstrated that SC as a low-cost substitute to binary-based computing radically simplifies the hardware implementation of arithmetic units and has the potential to satisfy the stringent power requirements in embedded devices. In SC, many arithmetic operations that are resource-consuming in binary designs can be implemented with very simple hardware logic, alleviating the extensive computational complexity. It offers a colossal design space for integration and optimization due to its reduced area and soft error resiliency. In this paper, we present HEIF, a highly efficient SC-based inference framework of the large-scale DCNNs, with broad applications including (but not limited to) \n<italic xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">LeNet-5 and AlexNet</italic>\n, that achieves high energy efficiency and low area/hardware cost. Compared to SC-DCNN, HEIF features: 1) the first (to the best of our knowledge) SC-based rectified linear unit activation function to catch up with the recent advances in software models and mitigate degradation in application-level accuracy; 2) the redesigned approximate parallel counter and optimized stochastic multiplication using transmission gates and inverse mirror adders; and 3) the new optimization of weight storage using clustering. Most importantly, to achieve maximum energy efficiency while maintaining acceptable accuracy, HEIF considers holistic optimizations on cascade connection of function blocks in DCNN, pipelining technique, and bit-stream length reduction. Experimental results show that in large-scale applications HEIF outperforms previous SC-DCNN by the throughput of \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"> <tex-math notation=\"LaTeX\">$4.1\\times $ </tex-math></inline-formula>\n, by area efficiency of up to \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"> <tex-math notation=\"LaTeX\">$6.5\\times $ </tex-math></inline-formula>\n, and achieves up to \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"> <tex-math notation=\"LaTeX\">${5.6\\times }$ </tex-math></inline-formula>\n energy improvement.","Title: Dual-mode inventory management under a chance credit constraint\nYear: 2019\n\nAbstract: We study a dual-mode inventory management problem of a high-value component where the customer demand and the regular transportation lead time are stochastic, and the review periods of the two modes are different. The manufacturer is subject to a chance credit constraint that bounds the working capital. To solve the resulting chance-constrained stochastic optimization problem, we develop a hybrid simulation optimization algorithm that combines the modified nested partitions method as the global search framework, a feasibility detection procedure for chance constraint verification, and a \\(\\hbox {KN}{++}\\) procedure as the final “cleanup” procedure to ensure solution quality. We are then able to analyze the impact of the chance credit constraint on the inventory policies and operational cost. Our numerical study shows that the effects of the reduction in mean or variance of the regular transportation lead time depend on whether the chance credit constraint is loose or tight. We show in this way that this tightness may lead to different mechanisms dominating the observed behavior. Further, we show that substantially extending the deterministic credit limit is less effective than having a slight increase in the probability parameter of the chance credit constraint.","Title: Phased-Array Radar System Simulator (PASIM): Development and Simulation Result Assessment.\nYear: 2019\n\nAbstract: In this paper, a system-specific phased-array radar system simulator was developed, based on a time-domain modeling and simulation method, mainly for system performance evaluation of the future Spectrum-Efficient National Surveillance Radar (SENSR). The goal of the simulation study was to establish a complete data quality prediction method based on specific radar hardware and electronics designs. The distributed weather targets were modeled using a covariance matrix-based method. The data quality analysis was conducted using Next-Generation Radar (NEXRAD) Level-II data as a basis, in which the impact of various pulse compression waveforms and channel electronic instability on weather radar data quality was evaluated. Two typical weather scenarios were employed to assess the simulator's performance, including a tornado case and a convective precipitation case. Also, modeling of some demonstration systems was evaluated, including a generic weather radar, a planar polarimetric phased-array radar, and a cylindrical polarimetric phased-array radar. Corresponding error statistics were provided to help multifunction phased-array radar (MPAR) designers perform trade-off studies.","Title: A cost-effective nonlinear self-interference canceller in full-duplex direct-conversion transceivers.\nYear: 2019\n\nAbstract: •A dual channel nonlinear complex least-mean-square (DC-NCLMS) canceller is proposed, where a convenient factorisation of the mean square error (MSE) function is used in the ANCLMS based SI canceller, followed by two independent optimisation processes.•The recently introduced complex dual channel (CDC) estimation framework, which exploits the duality between complex-valued signal representations and their bivariate-real counterparts, is employed to reduce computational complexity of augmented CLMS (ACLMS).•Rigourous mean and mean square convergence and steady state performance evaluations are conducted, to ensure a second order optimality and low computational complexity of the proposed canceller in the context of widely nonlinear modelled SI cancellation for FD DCTs.•To further speed up convergence of the proposed DC-NCLMS SI canceller, a data prewhitening scheme is also investigated.","Title: Highly Paralleled Low-Cost Embedded HEVC Video Encoder on TI KeyStone Multicore DSP\nYear: 2019\n\nAbstract: Although HEVC, the emerging video coding standard, has doubled the coding performance of its predecessor H.264/AVC, its significantly increased computational complexity imposes great obstacles for HEVC encoders to be employed in real-time applications with embedded processors, such as digital signal processors (DSPs). In this paper, a TI Keystone multicore TMS320C6678 DSP-based highly paralleled low-cost fast HEVC encoding solution is well designed and implemented. First, the overall structure of HEVC encoder with CTU-level parallelism is re-designed to well support the encoding parallelism, with full consideration of the hardware characteristics. Second, a low-delay and low-memory multicore data transmission mechanism is proposed to reduce the latency of data access between internal L2 memory and external DDR3. Third, the encoding bottlenecks, i.e., the most time-consuming encoding modules, are identified and optimized for acceleration with TI powerful C6000 SIMD instructions. Experimental results show that our proposed HEVC encoder on TI TMS320C6678 DSPs can significantly improve the real-time capacity with tolerable performance loss, 0.93 dB performance loss under on average 465.50 times speedup as compared to CPU-based HM reference software, more specifically, which makes it desirable in power-constrained real-time video applications.","Title: A Widely Linear MMSE Anti-Collision Method for Multi-Antenna RFID Readers.\nYear: 2019\n\nAbstract: In a multi-tag radio frequency identification (RFID) system, signals sent from different tags are likely to interfere with others, leading to a degradation in the transmission quality. To solve this problem, a novel widely linear minimum-mean-square-error (WLMMSE) anti-collision method is proposed by taking into account the improper second-order statistics of backscattered tag signals, which resul...","Title: Passquerade - Improving Error Correction of Text Passwords on Mobile Devices by using Graphic Filters for Password Masking.\nYear: 2019\n\nAbstract: Entering text passwords on mobile devices is a significant challenge. Current systems either display passwords in plain text: making them visible to bystanders, or replace characters with asterisks shortly after they are typed: making editing them harder. This work presents a novel approach to mask text passwords by distorting them using graphical filters. Distorted passwords are difficult to observe by attackers because they cannot mentally reverse the distortions. Yet passwords remain readable by their owners because humans can recognize visually distorted versions of content they saw before. We present results of an online questionnaire and a user study where we compared Color-halftone, Crystallize, Blurring, and Mosaic filters to Plain text and Asterisks when 1) entering, 2) editing, and 3) shoulder surfing one-word passwords, random character passwords, and passphrases. Rigorous analysis shows that Color-halftone and Crystallize filters significantly improve editing speed, editing accuracy and observation resistance compared to current approaches.","Title: 112 Gb/S/Lambda Wdm Direct-Detection Nyquist-Scm Transmission At 3.15 (B/S)/Hz Over 240 Km Ssmf Enabled By Novel Beating Interference Compensation\nYear: 2017\n\nAbstract: We experimentally demonstrate 112Gb/s/channel 35GHz-spaced WDM direct-detection SSB Nyquist-SCM transmission over a record distance of 240km SSMF using a novel beating interference compensation method, which offers a 7.6dB required OSNR improvement, and 200% reach enhancement.","Title: Speed-Oriented Architecture For Binary Field Point Multiplication On Elliptic Curves\nYear: 2019\n\nAbstract: This paper introduces a novel speed-oriented architecture of point multiplication in elliptic curve cryptography. A balanced full-precision multiplier is proposed to shorten latency, and a new modular inversion architecture is integrated to reduce the total number of clock cycles in point multiplication. A modified Montgomery Ladder algorithm that takes three clock cycles to calculate one input bit is proposed to best utilize hardware resources. A mixed-pipeline technique is used to balance the delay of different paths and increase frequency. The proposed architecture is implemented on GF(2(163)) and GF(2(571)), based on Xilinx Virtex-5 and Virtex-7 FPGA. For GF(2 (163)), the design reaches 211 MHz, with 29309 LUTs, and 547 clock cycles or 2.6 mu s latency on Virtex-5; 320.5 MHz, with 28911 LUTs and 1.7 mu s latency on Virtex-7. For GF(2(571)), the design reaches 186 MHz, with 286400 LUTs, and 1813 clock cycles or 9.6 mu s latency on Virtex-5; 267 MHz, 290001 LUTs and 6.79 mu s latency on Virtex-7. The proposed design achieves the lowest latency among all existing works, and its performance is also among the top. Furthermore, it is demonstrated that the proposed architecture maintains a high speed for larger binary fields, making it more suitable to be implemented in large-bit-length platforms with a higher security level. Since the multiplier and its segments work in different bit-length and refer to different fields, the proposed architecture can also be upgraded to a reconfigurable design to support multiple-field point multiplication in the future.","Title: CircConv: A Structured Convolution with Low Complexity.\nYear: 2019\n\nAbstract: Deep neural networks (DNNs), especially deep convolutional neural networks (CNNs), have emerged as the powerful technique in various machine learning applications. However, the large model sizes of DNNs yield high demands on computation resource and weight storage, thereby limiting the practical deployment of DNNs. To overcome these limitations, this paper proposes to impose the circulant structure to the construction of convolutional layers, and hence leads to circulant convolutional layers (CircConvs) and circulant CNNs. The circulant structure and models can be either trained from scratch or re-trained from a pre-trained non-circulant model, thereby making it very flexible for different training environments. Through extensive experiments, such strong structure-imposing approach is proved to be able to substantially reduce the number of parameters of convolutional layers and enable significant saving of computational cost by using fast multiplication of the circulant tensor.","Title: Structural information integration for predicting damages in bridges\nYear: 2019\n\nAbstract: Industrial Information Integration Engineering interacts with each of the twelve engineering disciplines defined by the U.S. National Academy of Engineering. Civil Engineering is one of the twelve engineering sections. Integration of information can be a very valuable method for civil engineering. In the management of the highway bridges, bridge design and operation parameters and inspection results are recorded according to national and state guides and the data is coded and archived in what is known as the National Bridge Inventory (NBI) database. The integration of the information in the NBI database can be a very valuable method for managing of bridges. The major challenges of the information in such a database are unbalance, complexity, subjectivity, and incompleteness, which make developing a damage prediction model from such information difficult. Particularly, the prediction and management of bridge abutment distress is difficult since their generation mechanism is not clear. The purpose of this study was to integrate the information of the inspections database and establish models to predict abutment distress. Input variables for the networks were selected from knowledge-based evaluations, statistical analyses, and trial and error. Four neural network models were developed and evaluated: the multilayer perceptron and support vector machine models predicted abutment condition better than a supervised self-organizing map and radial basis function networks.","Title: A High Gain, 808Mhz GBW Four-Stage OTA in 65 nm CMOS\nYear: 2019\n\nAbstract: A four-stage operational transconductance amplifier (OTA) with a novel compensation structure combining multipath G(m)-C compensation and no capacitor feed-forward compensation is proposed in this paper. Based on the small-signal model, stability analysis and design consideration are carried out to demonstrate the stability of the compensation technique. To verify the effectiveness of the compensation scheme, the proposed OTA which drives a 2 pF capacitance, is simulated in TSMC 65 nm 1.2 V CMOS process, achieving 808 MHz gain-bandwidth, 119 dB DC gain, 585 V/mu s slew rate (SR) and 6 ns 1% settling time. The circuit is operated at the single supply voltage of 1.2 V with power consumption of 2.17 mW and the layout area is 0.011 mm(2).","Title: Design of moving target detection and tracking system based on cortex-A7 and openCV\nYear: 2018\n\nAbstract: Applications based on real-time image processing are affected by communication pressure, random time lag, data packet loss and other problems. The main reason is image transmission and server message feedback. This can be resolved by terminal processing. But it requires the processing terminal to be able to complete the requirements of more complex image processing tasks under limited processing power. A Camshit based optimization local algorithm considering the characteristics of the platform is proposed. It can run the algorithm in-local without transferring images to the server. And Applications such as behavioral analysis are supported because the system still has the spare. Through algorithm analysis and experiment, we can prove the validity and real-time of this algorithm based on the hardware conditions. Firstly, the upper limit of hardware condition for engineering realization is studied. And provides predictable performance basis for the implementation of low-real-time requirements in operational performance on cortex Al systems. Second, the maximum processing speed of 41 FPS means more additional cost space or performance redundancy. Finally, real-time recognition of personnel behavior is implemented in the current hardware environment. Provides an important reference for other behavioral analysis researchers. In this paper, the work of hardware carrier and algorithm is preliminarily completed, which can provide cutting for further optimization research and engineering realization.","Title: Neural network controller for nanopositioning of a smooth impact drive mechanism\nYear: 2019\n\nAbstract: In this paper, neural network theory is used to improve the positioning accuracy of smooth impact drive mechanisms (SIDMs), by designing a displacement controller that consists of a neural network identification (NNI) and a neural network controller (NNC). The dynamics of the SIDM are described by the NNI, which consists of an input layer, hidden layer, and output layer. The parameters of the NNI are adjusted using back propagation. The NNC is designed as a proportional-derivative (PD) controller, which is used to accurately control the displacement of the SIDM. The PD parameters are adjusted with an adaptive adjustment algorithm. A prototype of the SIDM was fabricated and an experimental control system was built that consists of a laser displacement sensor, power amplifier, data acquisition board, and SIDM prototype. The experimental results show that nanoscale positioning accuracy can be obtained. The control system can maintain steady operation, even if the output load mass is changed.","Title: Eigen: Ecologically-Inspired Genetic Approach For Neural Network Structure Searching From Scratch\nYear: 2019\n\nAbstract: Designing the structure of neural networks is considered one of the most challenging tasks in deep learning, especially when there is few prior knowledge about the task domain. In this paper, we propose an Ecologically-Inspired GENetic (EIGEN) approach that uses the concept of succession, extinction, mimicry, and gene duplication to search neural network structure from scratch with poorly initialized simple network and few constraints forced during the evolution, as we assume no prior knowledge about the task domain. Specifically, we first use primary succession to rapidly evolve a population of poorly initialized neural network structures into a more diverse population, followed by a secondary succession stage for fine-grained searching based on the networks from the primary succession. Extinction is applied in both stages to reduce computational cost. Mimicry is employed during the entire evolution process to help the inferior networks imitate the behavior of a superior network and gene duplication is utilized to duplicate the learned blocks of novel structures, both of which help to find better network structures. Experimental results show that our proposed approach can achieve similar or better performance compared to the existing genetic approaches with dramatically reduced computation cost. For example, the network discovered by our approach on CIFAR-100 dataset achieves 78.1% test accuracy under 120 GPU hours, compared to 77.0% test accuracy in more than 65, 536 GPU hours in [36].","Title: CPCA: An efficient wireless routing algorithm in WiNoC for cross path congestion awareness\nYear: 2019\n\nAbstract: Wireless network-on-chip (WiNoC) is a new paradigm to mitigate the long-distance transmission latency for conventional wired network-on-chip. The wireless routers in WiNoC have to handle a large number of packets which could cause data congestion, thus reducing the network performance. In this paper, we propose a novel wireless routing algorithm, called CPCA, which exploits the cross path congestion information as hints to route the packets. Under CPCA, the whole network is partitioned into sub-networks. In each subnet, the congestion information of the wireless router is propagated along the cross path. As a result, the routers in the same dimension can get the congestion degree of wireless router within the subnet. Based on the congestion information, CPCA can compute the suitable path for packets routing, which can prominently avoid the congestion aggravation in the wireless router. Experimental results show that our proposed method can effectively improve performance in terms of packets transmission latency and network throughput.","Title: Using Rfid Technology To Introduce Properties Of Lms\nYear: 2019\n\nAbstract: This paper presents a new course design to enhance students' interest in our graduate curriculum of morden digital signal processing (MDSP). The familiar radio frequency identification (RFID) technology is introduced to reveal the fundamentals of the least-mean-square (LMS) algorithm. Students are encouraged to encode their customised data into their respective RFID tags, and to use the software-defined radio (SDR) platform to monitor the inventory process integrated within RFID communication systems. After acquiring the digital signal from the SDR hardware, they need to perform adaptive channel estimation and signal detection to identify the customised data. By applying appropriate mathematical tools in the Matlab programming environment to solve the signal detection problem in two different communication scenarios, students reported to have had a deeper understanding of the signal processing concepts of adaptive filtering techniques in a self-directed manner.","Title: Hybrid Data Fusion DBN for Intelligent Fault Diagnosis of Vehicle Reducers.\nYear: 2019\n\nAbstract: Given its importance, fault diagnosis has attracted considerable attention in the literature, and several machine learning methods have been proposed to discover the characteristics of different aspects in fault diagnosis. In this paper, we propose a Hybrid Deep Belief Network (HDBN) learning model that integrates data in different ways for intelligent fault diagnosis in motor drive systems, such as a vehicle drive system. In particular, we propose three data fusion methods: data union, data join, and data hybrid, based on detailed data fusion research. Additionally, the significance of the fusion is explained from the energy perspective of the signal. In particular, the appropriate fusion methods and data structures suitable for model training requirements can help improve the accuracy of fault diagnosis. Moreover, mixed-precision training is used as a special fusion method to further improve the performance of the model. Experiments with the datasets obtained from the simulation platform demonstrate the superiority of our proposed model over the state-of-the-art methods.","Title: CNV_IFTV: An Isolation Forest and Total Variation-Based Detection of CNVs from Short-Read Sequencing Data\nYear: 2021\n\nAbstract: AbstractAccurate detection of copy number variations (CNVs) from short-read sequencing data is challenging due to the uneven distribution of reads and the unbalanced amplitudes of gains and losses. The direct use of read depths to measure CNVs tends to limit performance. Thus, robust computational approaches equipped with appropriate statistics are required to detect CNV regions and boundaries. This study proposes a new method called CNV_IFTV to address this need. CNV_IFTV assigns an anomaly score to each genome bin through a collection of isolation trees. The trees are trained based on isolation forest algorithm through conducting subsampling from measured read depths. With the anomaly scores, CNV_IFTV uses a total variation model to smooth adjacent bins, leading to a denoised score profile. Finally, a statistical model is established to test the denoised scores for calling CNVs. CNV_IFTV is tested on both simulated and real data in comparison to several peer methods. The results indicate that the proposed method outperforms the peer methods. CNV_IFTV is a reliable tool for detecting CNVs from short-read sequencing data even for low-level coverage and tumor purity. The detection results on tumor samples can aid to evaluate known cancer genes and to predict target drugs for disease diagnosis.","Title: Optimal CTU-level bit allocation in HEVC for low bit-rate applications\nYear: 2019\n\nAbstract: The coding efficiency of High Efficiency Video Coding (HEVC) outperforms all the past video coding standards. But for low bit-rate video applications, fewer bits cannot guarantee the reconstructed quality of each coding frame. Rate control is used to get better quality with fewer bits, while the bit allocation is the key of rate control. This paper proposes an optimal CTU-level bit allocation based on region classification in HEVC for low bit-rate applications. The motion regions are extracted in each frame from the CTU-level. The bits allocated to the CTUs belong to the motion regions will be calculated according to the motion vectors of CTUs and the overall frames to improve the quality of these CTUs. Finally, the R-lambda model rate control adopted in HM16.0 will be used to calculate QP of each CTU. Experimental results demonstrate that the proposed bit allocation approach can achieve an average improvement of 0.32 dB in objective quality compared with the conventional rate control adapted in HM16.0, and the subjective quality improves obviously and the accuracy of rate control is also good.","Title: Multi-Satellite Non-Cooperative Communication Based On Transform Domain Communication System\nYear: 2019\n\nAbstract: Cognitive communication based on the unoccupied channels of transparent transponders in commercial satellites is a promising method to solve the shortage of satellite spectrum resource. However, most opportunity spectrum access systems cannot obtain real-time channel information from a commercial satellite. Although some systems can achieve non-cooperative communication, they usually utilize single satellite, which limits their overall performance. In this paper, to take full use of unoccupied satellite spectrum without disturbing primary users, a non-cooperative communication method by using satellites of different types is proposed based on transform domain communication system. Compared with existing methods, it is compatible with different transparent transponder parameters, such as frequency, bandwidth, primary user type, and delay. Besides, it also can achieve transmitting rate adaptation without signaling overhead.","Title: EEG-Based Maritime Object Detection for IoT-Driven Surveillance Systems in Smart Ocean\nYear: 2020\n\nAbstract: Automated maritime object detection is a significant research challenge in intelligent marine surveillance systems for the Internet of Things (IoT) and smart ocean applications. In particular, ship detection is recognized as one of the core research issues of these IoT-driven intelligent marine surveillance systems. Traditional methods based on machine learning have made some achievements in detection tasks for specific objects. However, the ship objects are relatively small, and they are usually not accurately detected. In this article, we propose an electroencephalography (EEG)-based maritime object detection algorithm for IoT-driven surveillance systems in the smart ocean. For this purpose, we conduct experiments to record the EEG signals of subjects when they are watching the maritime image scenes. With the feature analysis of EEG signals, the event-related potential (ERP) components associated with detecting objects are induced, such as the \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"> <tex-math notation=\"LaTeX\">$P3$ </tex-math></inline-formula>\n and \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"> <tex-math notation=\"LaTeX\">$N2$ </tex-math></inline-formula>\n components. Employing classification based on linear discriminant analysis (LDA), the area under curve (AUC) of the receiver operating characteristic (ROC) is used to evaluate the detection accuracy. We use this novel method to determine and identify essential objects and areas from IoT devices, such as digital camera imaging sensors. Our proposed method can not only help to detect small objects accurately using fewer samples but can also be used to reduce the data volume needed to be stored and transmitted in IoT-driven marine surveillance systems.","Title: Selective partial-update augmented complex-valued LMS algorithm and its performance analysis\nYear: 2021\n\nAbstract: AbstractHighlights •This paper proposes a selective partial-update augmented complex-valued LMS algorithm.•The performance of the proposed algorithm is analyzed to predict its stochastic behavior.•The values of two scale factors are derived by using multiple integral, which play an important role in the performance analysis. AbstractAdaptive filtering applications such as acoustic echo cancellation may require a large number of filter weights in order to estimate the unknown system with a long response, leading to high computational complexity. Using partial-update techniques is an effective method to alleviate this issue, and two partial-update augmented complex-valued LMS (PU-ACLMS) algorithms have been recently proposed. However, their convergence rates severely deteriorate due to using a stochastic or sequential weight update manner. This paper proposes a new PU-ACLMS, which selects the weights corresponding to the M largest moduli of input samples to update, with M less than the number of all weights. The mean and mean-square performance of the proposed algorithm is then analyzed to predict its stochastic behavior. In order to achieve this goal, the values of two scale factors are derived by using multiple integrals, which play an important role in our performance analysis. Finally, the performance of the proposed algorithm is compared with that of the sequential PU-ACLMS and stochastic PU-ACLMS, and the theoretical findings are verified by extensive simulations.","Title: SynSig2Vec: Forgery-Free Learning of Dynamic Signature Representations by Sigma Lognormal-Based Synthesis and 1D CNN\nYear: 2022\n\nAbstract: Handwritten signature verification is a challenging task because signatures of a writer may be skillfully imitated by a forger. As skilled forgeries are generally difficult to acquire for training, in this paper, we propose a deep learning-based dynamic signature verification framework, SynSig2Vec, to address the skilled forgery attack without training with any skilled forgeries. Specifically, SynSig2Vec consists of a novel learning-by-synthesis method for training and a 1D convolutional neural network model, called Sig2Vec, for signature representation extraction. The learning-by-synthesis method first applies the Sigma Lognormal model to synthesize signatures with different distortion levels for genuine template signatures, and then learns to rank these synthesized samples in a learnable representation space based on average precision optimization. The representation space is achieved by the proposed Sig2Vec model, which is designed to extract fixed-length representations from dynamic signatures of arbitrary lengths. Through this training method, the Sig2Vec model can extract extremely effective signature representations for verification. Our SynSig2Vec framework requires only genuine signatures for training, yet achieves state-of-the-art performance on the largest dynamic signature database to date, DeepSignDB, in both skilled forgery and random forgery scenarios. Source codes of SynSig2Vec will be available at \n<uri xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">https://github.com/LaiSongxuan/SynSig2Vec</uri>\n.","Title: Implicit Feature Alignment: Learn to Convert Text Recognizer to Text Spotter\nYear: 2021\n\nAbstract: Text recognition is a popular research subject with many associated challenges. Despite the considerable progress made in recent years, the text recognition task itself is still constrained to solve the problem of reading cropped line text images and serves as a subtask of optical character recognition (OCR) systems. As a result, the final text recognition result is limited by the performance of the text detector. In this paper, we propose a simple, elegant and effective paradigm called Implicit Feature Alignment (IFA), which can be easily integrated into current text recognizers, resulting in a novel inference mechanism called IFA-inference. This enables an ordinary text recognizer to process multi-line text such that text detection can be completely freed. Specifically, we integrate IFA into the two most prevailing text recognition streams (attention-based and CTC-based) and propose attention-guided dense prediction (ADP) and Extended CTC (ExCTC). Furthermore, the Wasserstein-based Hollow Aggregation Cross-Entropy (WH-ACE) is proposed to suppress negative predictions to assist in training ADP and ExCTC. We experimentally demonstrate that IFA achieves state-of-the-art performance on end-to-end document recognition tasks while maintaining the fastest speed, and ADP and ExCTC complement each other on the perspective of different application scenarios. Code will be available at https://github.com/Wang-Tianwei/Implicit-feature-alignment.","Title: Lightweight Multi-person Total Motion Capture Using Sparse Multi-view Cameras.\nYear: 2021\n\nAbstract: Multi-person total motion capture is extremely challenging when it comes to handle severe occlusions, different reconstruction granularities from body to face and hands, drastically changing observation scales and fast body movements. To overcome these challenges above, we contribute a lightweight total motion capture system for multi-person interactive scenarios using only sparse multi-view cameras. By contributing a novel hand and face bootstrapping algorithm, our method is capable of efficient localization and accurate association of the hands and faces even on severe occluded occasions. We leverage both pose regression and keypoints detection methods and further propose a unified two-stage parametric fitting method for achieving pixel-aligned accuracy. Moreover, for extremely self-occluded poses and close interactions, a novel feedback mechanism is proposed to propagate the pixel-aligned reconstructions into the next frame for more accurate association. Overall, we propose the first light-weight total capture system and achieves fast, robust and accurate multi-person total motion capture performance. The results and experiments show that our method achieves more accurate results than existing methods under sparse-view setups.","Title: Fast augmentation algorithms for network kernel density visualization\nYear: 2021\n\nAbstract: AbstractNetwork kernel density visualization, or NKDV, has been extensively used to visualize spatial data points in various domains, including traffic accident hotspot detection, crime hotspot detection, disease outbreak detection, and business and urban planning. Due to a wide range of applications for NKDV, some geographical software, e.g., ArcGIS, can also support this operation. However, computing NKDV is very time-consuming. Although NKDV has been used for more than a decade in different domains, existing algorithms are not scalable to million-sized datasets. To address this issue, we propose three efficient methods in this paper, namely aggregate distance augmentation (ADA), interval augmentation (IA), and hybrid augmentation (HA), which can significantly reduce the time complexity for computing NKDV. In our experiments, ADA, IA and HA can achieve at least 5x to 10x speedup, compared with the state-of-the-art solutions.","Title: Cooperative computation offloading and resource allocation for delay minimization in mobile edge computing\nYear: 2021\n\nAbstract: Mobile edge computing (MEC) is a promising paradigm, which brings computation resources in proximity to mobile devices and allows the tasks of mobile devices to be offloaded to MEC servers with low latency. The joint problem of cooperative computation task offloading and resource allocation is a challenging issue. The joint problem of cooperative computation task offloading scheme and resource assignment in MEC is investigated in this paper, where the vertical cooperation among mobile devices, mobile edge server nodes and mobile cloud server nodes is considered, and the horizontal computation cooperation between edge nodes is considered as well. A computation offloading decision, cooperative selection, power allocation and CPU cycle frequency assignment problem is formulated. The objective is to minimize the latency while guaranteeing the constraint of transmission power, energy consumption and CPU cycle frequency. The formulated latency optimization problem is a nonconvex mixed-integer problem in general, which has binary variables and continuous variables. In order to solve the formulated problem. A joint iterative algorithm based on the Lagrangian dual decomposition, ShengJin Formula method, and monotonic optimization method is proposed. The CPU cycle frequence allocation is handled by the ShengJin Formula method due to the cubic equation of one variable about the CPU frequence allocation. The transmission power assignment is handled by the monotonic optimization method. In the algorithm convergence with different number of tasks, the proposed algorithm can quickly and effectively reach the convergence state and getting the minimum task execution delay. Numerical results demonstrate that the proposed algorithm outperforms the Full MEC, Full Local and Full Cloud three schemes in terms of execution latency.","Title: Integrating Remote-Sensing And Assimilation Data To Improve Air Temperature On Hot Weather In East China\nYear: 2021\n\nAbstract: Land-surface characteristics (LSCs) and land-soil moisture conditions can modulate energy partition at the land surface, impact near-surface atmosphere conditions, and further affect land-atmosphere interactions. This study investigates the effect of land-surface-characteristic parameters (LSCPs) including albedo, leaf-area index (LAI), and soil moisture (SM) on hot weather by in East China using the numerical model. Simulations using the Weather Research and Forecasting (WRF) Model were conducted for a hot weather event with a high spatial resolution of 1 km in domain 3 by using ERA-Interim forcing fields on 20 July 2017 until 16:00 UTC on 25 July 2017. The satellite-based albedo and LAI, and assimilation-based soil-moisture data of high temporal-spatial resolution, which are more accurate to match fine weather forecasts and high-resolution simulations, were used to update the default LSCPs. A control simulation with the default LSCPs (WRF_CTL), a main sensitivity simulation with the updated LSCP albedo, LAI and SM (WRF_CHAR), and a series of other sensitivity simulations with one or two updated LSCPs were performed. Results show that WRF_CTL could reproduce the spatial distribution of hot weather, but overestimated air temperature (Ta) and maximal air temperature (Tamax) with a warming bias of 1.05 and 1.32 degrees C, respectively. However, the WRF_CHAR simulation reduced the warming bias, and improved the simulated Ta and Tamax with reducing relative biases of 33.08% and 29.24%, respectively. Compared to the WRF_CTL, WRF_CHAR presented a negative sensible heat-flux difference, positive latent heat flux, and net radiation difference of the area average. LSCPs modulated the partition of available land-surface energy and then changed the air temperature. On the basis of statistical-correlation analysis, the soil moisture of the top 10 cm is the main factor to improve warming bias on hot weather in East China.","Title: Robust and Accurate 3D Self-Portraits in Seconds\nYear: 2022\n\nAbstract: In this paper, we propose an efficient method for robust and accurate 3D self-portraits using a single RGBD camera. Our method can generate detailed and realistic 3D self-portraits in seconds and shows the ability to handle subjects wearing extremely loose clothes. To achieve highly efficient and robust reconstruction, we propose PIFusion, which combines learning-based 3D recovery with volumetric non-rigid fusion to generate accurate sparse partial scans of the subject. Meanwhile, a non-rigid volumetric deformation method is proposed to continuously refine the learned shape prior. Moreover, a lightweight bundle adjustment algorithm is proposed to guarantee that all the partial scans can not only “loop” with each other but also remain consistent with the selected live key observations. Finally, to further generate realistic portraits, we propose non-rigid texture optimization to improve the texture quality. Additionally, we also contribute a benchmark for single-view 3D self-portrait reconstruction, an evaluation dataset that contains 10 single-view RGBD sequences of a self-rotating performer wearing various clothes and the corresponding ground-truth 3D models in the first frame of each sequence. The results and experiments based on this dataset show that the proposed method outperforms state-of-the-art methods on accuracy, efficiency, and generality.","Title: A Layer-Wise Distribution Analysis Of The Wlmmse-Sic Mimo Receiver For Rectilinear Or Quasi-Rectilinear Signals\nYear: 2021\n\nAbstract: In this brief, we investigate the widely linear minimum mean square error (WLMMSE) successive interference cancellation (SIC) detector in multiple-input-multiple-output systems with rectilinear or quasi rectilinear transmit signals. To this end, layer-wise approximate probability density functions (PDFs) of the post-detection signal-to-interference-plus-noise ratio are first derived. It is next shown that the so derived PDFs enable explicit and concise performance evaluations in terms of the symbol error rate and the outage probability. For rigor, both the cases with and without inter-layer error propagation are addressed, and closed-form solutions are provided. Validity of theoretical analysis is demonstrated via Monte Carlo simulations. (c) 2021 Elsevier B.V. All rights reserved.","Title: Cross Validation of GOES-16 and NOAA Multi-Radar Multi-Sensor (MRMS) QPE over the Continental United States\nYear: 2021\n\nAbstract: The Geostationary Operational Environmental Satellite-R (GOES-R) series provides new opportunities for continuous observation of precipitation at large scales with a high resolution. An operational quantitative precipitation estimation (QPE) product has been produced based on multi-channel measurements from the Advanced Baseline Imager (ABI) aboard the GOES-16 (formerly known as GOES-R). This paper presents a comprehensive evaluation of this GOES-16 QPE product against a ground reference QPE product from the National Oceanic and Atmospheric Administration (NOAA) Multi-Radar Multi-Sensor (MRMS) system over the continental United States (CONUS) during the warm seasons of 2018 and 2019. For the first time, the accuracy of GOES-16 QPE product was quantified using the gauge-corrected MRMS (GC-MRMS) QPE product, and a number of evaluation metrics were applied to adequately resolve the associated errors. The results indicated that precipitation occurrence and intensity estimated by the GOES-16 QPE agreed with GC-MRMS fairly well over the eastern United States (e.g., the probability of detection was close to 1.0, and the Pearson's correlation coefficient was 0.80 during September 2019), while the discrepancies were noticeable over the western United States (e.g., the Pearson's correlation coefficient was 0.64 for the same month). The performance of GOES-16 QPE was downgraded over the western United States, in part due to the limitations of the GOES-16 rainfall retrieval algorithm over complex terrains, and in part because of the poor radar coverage analyzed by the MRMS system. In addition, it was found that the GOES-16 QPE product significantly overestimated rainfall induced by the mesoscale convective systems in the midwestern United States, which must be addressed in the future development of GOES satellite rainfall retrieval algorithms.</p>","Title: Concolic Execution of NMap Scripts for Honeyfarm Generation\nYear: 2021\n\nAbstract: ABSTRACTAttackers rely upon a vast array of tools for automating attacksagainst vulnerable servers and services. It is often the case thatwhen vulnerabilities are disclosed, scripts for detecting and exploit-ing them in tools such asNmapandMetasploitare released soonafter, leading to the immediate identification and compromise ofvulnerable systems. Honeypots, honeynets, tarpits, and other decep-tive techniques can be used to slow attackers down, however, such approaches have difficulty keeping up with the sheer number of vulnerabilities being discovered and attacking scripts that are being released. To address this issue, this paper describes an approach for applying concolic execution on attacking scripts in Nmap in order to automatically generate lightweight fake versions of the vulnerable services that can fool the scripts. By doing so in an automated and scalable manner, the approach can enable rapid deployment of custom honeyfarms that leverage the results of concolic execution to trick an attacker's script into returning a result chosen by the honeyfarm, making the script unreliable for the use by the attacker.","Title: The verification of multiplicity support of a defective eigenvalue of a real matrix\nYear: 2022\n\nAbstract: Computing a defective eigenvalue is an ill-posed problem if components of the matrix are approximate data. Using the definition of multiplicity support of a defective eigenvalue introduced by Zeng, we consider the verification about the sensitivity and computation of a defective eigenvalue of a real matrix. We discuss how to construct a slightly perturbed interval matrix which is guaranteed to possess a real matrix with computed defective eigenvalue of computed multiplicity support. Furthermore, we also obtain an interval matrix which is guaranteed to possess a real matrix. The columns of the real matrix span the corresponding eigenvector space. (c) 2021 Elsevier Inc. All rights reserved.","Title: Focal and Global Knowledge Distillation for Detectors.\nYear: 2022\n\nAbstract: Knowledge distillation has been applied to image classification successfully. However, object detection is much more sophisticated and most knowledge distillation methods have failed on it. In this paper, we point out that in object detection, the features of the teacher and student vary greatly in different areas, especially in the foreground and background. If we distill them equally, the uneven differences between feature maps will negatively affect the distillation. Thus, we propose Focal and Global Distillation (FGD). Focal distillation separates the foreground and background, forcing the student to focus on the teacher's critical pixels and channels. Global distillation rebuilds the relation between different pixels and transfers it from teachers to students, compensating for missing global information in focal distillation. As our method only needs to calculate the loss on the feature map, FGD can be applied to various detectors. We experiment on various detectors with different backbones and the results show that the student detector achieves excellent mAP improvement. For example, ResNet-50 based RetinaNet, Faster RCNN, RepPoints and Mask RCNN with our distillation method achieve 40.7%, 42.0%, 42.0% and 42.1% mAP on COCO2017, which are 3.3, 3.6, 3.4 and 2.9 higher than the baseline, respectively. Our codes are available at https://github.com/yzd-v/FGD.","Title: Research on the Data Literacy Index System of Pre-service Teachers Under the Background of Big Data\nYear: 2021\n\nAbstract: The arrival of the big data era has brought new challenges to the teaching methods and professional development of the teachers in the field of education. As a group that is about to become a teacher, pre-service teachers are also one of the necessary qualities of their way to the teacher's road. Based on the literature analysis and case study, this paper constructs a preliminary index system of the data literacy of the pre-service teachers, and carries out the pre-investigation. The research data is analyzed by using the SPSS25.0 software, and the index system is repaired according to the analysis results. In the end, the index system of the data literacy of the pre-service teachers is divided into three levels: attitude and awareness, knowledge and skill, morality and law. There are 8 secondary indicators, a total of 19 projects.","Title: A partitioned framework for coupling LBM and FEM through an implicit IBM allowing non-conforming time-steps: Application to fluid-structure interaction in biomechanics\nYear: 2022\n\nAbstract: This paper presents a partitioned framework for the numerical simulation of fluid-structure interactions by coupling the lattice Boltzmann method (LBM) and the finite element method (FEM). The two numerical methods LBM and FEM are coupled with an implicit immersed boundary method (IBM) in a strong way, which ensures exactly the no-slip condition and the continuities of velocity and stress at the fluid-solid interface and each instant in time. In the proposed partitioned coupling procedure, the coupling system of equations are first established and then condensed to the interface. By solving the condensed coupling system of equations, the interface force field is obtained and sent to both solvers to accomplish time integrations in each sub-domain. In addition, two strategies based on linear interpolation in time are proposed to handle the cases with non-conforming time-steps in the fluid and solid sub-domains. Through several 2D and 3D numerical test-cases on the mechanical heart valve, the fluid-induced vibration of a deformable solid beam, the flapping flag, the proposed coupling framework is validated with good agreements with references. Finally, a test-case on the interaction between the blood flow and the aortic valve is carried out, showing the applicability of the present framework in realistic biomechanical applications.","Title: Construction of Smart City Street Landscape Big Data-Driven Intelligent System Based on Industry 4.0\nYear: 2021\n\nAbstract: With the application of engineering management in smart city construction under Industry 4.0, the intelligent design of urban street landscape has attracted extensive attention. Affected by the low intelligent level of traditional landscape design, the existing urban landscape composite system has difficulty in meeting the needs of smart city construction. Therefore, this paper proposes the construction of street landscape big data-driven intelligent decision support system based on Industry 4.0. Based on the complex network theory, this paper analyzes the structure, links, nodes, driving forces, and functional requirements of urban street landscape and then puts forward the construction content and implementation method of urban street landscape intelligent decision support system. The system consists of four aspects: intelligent infrastructure, service, protection and maintenance, and management and evaluation system. Its implementation not only reflects the cooperation and effective application of intelligent technology in each stage of street landscape construction, but also provides reference for the application of engineering management in other fields under Industry 4.0.","Title: FAFNet: A False Alarm Filter Algorithm for License Plate Detection Based on Deep Neural Network\nYear: 2021\n\nAbstract: The license plate detection technology has been widely applied in our daily life, but it encounters many challenges when performing license plate detection tasks in special scenarios. In this paper, a license plate detection algorithm is proposed for the problem of license plate detection, and an efficient false alarm filter algorithm, namely the FAFNet (False-Alarm Filter Network) is proposed for solving the problem of false alarms in license plate location scenarios in China. At first, this paper adopted the YOLOv5 target detection algorithm to detect license plates, and used the FAFNet to re-identify the images to avoid false detection. FAFNet is a lightweight convolutional neural network (CNN) that can solve the false alarm problem of real-time license plate recognition on embedded devices, and its performance is good. Next, this paper proposed a model generalization method for the purpose of making the proposed FAFNet be applicable to the license plate false alarm scenarios in other countries without the need to re-train the model. Then, this paper built a large-scale false alarm filter dataset, all samples in the dataset came from the industries and contained a variety of complex real-life scenarios. At last, experiments were conducted and the results showed that, the proposed FAFNet can achieve high-accuracy false alarm filtering and can run in real-time on embedded devices.","Title: Computing PUR of Zero-Dimensional Ideals of Breadth at Most One\nYear: 2021\n\nAbstract: In this paper, for a zero-dimensional polynomial ideal I, the authors prove that k[x(1), x(2), MIDLINE HORIZONTAL ELLIPSIS, x(n)]/I is cyclic if and only if the breadth of I is 0 or 1. Furthermore, the authors present a new algorithm to compute polynomial univariate representation (PUR) of such an ideal.","Title: Cooperative jamming in downlink satellite network with hardware impairments\nYear: 2021\n\nAbstract: In this article, a land mobile satellite communication network with an external jammer to fight against the eavesdropper is investigated, where the hardware impairments are introduced to make the analyses more practical. It is worth noting that three jamming schemes, that is, the Gaussian noise jamming, the friendly jamming, and the scheme without jamming, are taken into account to explore the effect of the jammer comprehensively. The secrecy metrics such as the connectivity outage probability (COP), the security outage probability, and the secrecy throughput (ST), are derived for the passive eavesdropping scenario, while the ergodic secrecy capacity is analyzed for the active eavesdropping scenario. Moreover, to get intuitive insight from the closed-form expressions, the asymptotic ST and the asymptotic COP under the condition of high signal-to-noise ratio are also considered. Finally, all the derivations are validated by Monte Carlo simulations, and the effects of different parameters on the performance of the network are analyzed.","Title: Route recommendation based on temporal-spatial metric\nYear: 2022\n\nAbstract: The acquisition and analysis of large amounts of mobile trajectory data promote the rapid development of mobile recommender systems, which can essentially recommend a series of positions to the mobile users. In this paper, we concentrate on the mobile sequential recommendation problem by considering both the distance and time factors, the objective of which is to recommend an optimal driving route consisting of a sequence of pick-up locations for the drivers. Firstly, we formulate this problem and develop a potential travel cost function to comprehensively evaluate each candidate route. Then we propose an efficient and effective Route recommendation based on Temporal-Spatial metric scheme called RTS, which adopts the simulated annealing algorithm and can achieve real-time route recommendation without severely sacrificing the recommendation performance. The extensive experiments based on both the real-world and synthetic data are conducted, the results of which demonstrate the efficiency and effectiveness of the proposed RTS scheme.","Title: VRConvMF: Visual Recurrent Convolutional Matrix Factorization for Movie Recommendation\nYear: 2022\n\nAbstract: Sparsity of user-to-item rating data becomes one of challenging issues in the recommender systems, which severely deteriorates the recommendation performance. Fortunately, context-aware recommender systems can alleviate the sparsity problem by making use of some auxiliary information, such as the information of both the users and items. In particular, the visual information of items, such as the movie poster, can be considered as the supplement for item description documents, which helps to obtain more item features. In this paper, we focus on movie recommender system and propose a probabilistic matrix factorization based recommendation scheme called \n<italic xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">visual recurrent convolutional matrix factorization</i>\n (VRConvMF), which utilizes the textual and multi-level visual features extracted from the descriptive texts and posters respectively. We implement the proposed VRConvMF and conduct extensive experiments on three commonly used real world datasets to validate its effectiveness. The experimental results illustrate that the proposed VRConvMF outperforms the existing schemes.","Title: Pedestrian Positioning Based on Dual Inertial Sensors and Foot Geometric Constraints\nYear: 2022\n\nAbstract: Foot-mounted inertial navigation with zero velocity update (ZUPT) is commonly used in pedestrian positioning. To suppress the accumulating position error owing to the low-cost inertial sensor, classical dual-foot pedestrian navigation method provides the compensation using the two feet's relative step length, whose values can be set an upper bound constraint or measured by ranging sensors. However, it suffers from low improvement and poor wearability. In this study, the two sensors are distributed on the toe and ankle of the same leg. The sensor-to-sensor position is unchangeable with the lower limb segments. This makes the uncoupled two subsystems be related. Through gait analysis, the step is divided into the stance and swing phases. For each phase, positions, velocities, and the related characteristics of the toe and the ankle are analyzed and modeled, respectively. A Kalman filter is constructed to fuse the position and velocity observations according to the geometric constraints and the foot vector. The Mahony filter by fusing the inertial sensor and magnetometer is adopted for the attitude improvement. Several experiments at low and high walking velocities on the plane and upstairs are carried out to verify the proposed method. The results show that compared with the conventional single-inertial measurement unit (IMU) ZUPT and dual-IMU two-foot constraint methods, the position and heading errors have been well suppressed.","Title: Evaluation of Logistics Service Quality: Sentiment Analysis of Comment Text Based on Multi-Level Graph Neural Network\nYear: 2021\n\nAbstract: The world is severely impacted by the coronavirus (COVID19). During the epidemic, logistics service, an often-overlooked pillar of the modern society, steps into the spotlight. However, the service capability is inevitably weakened by the epidemic. The fatigued service providers are increasingly unable to meet the high expectations of users, who therefore leave harsh comments on logistics services. It is important for managers to find information that helps to improve management, out of the biased and angry comments. Text sentiment analysis is a fundamental work in natural language processing (NLP). In recent years, graph neural network (GNN) has achieved excellent performance in various NLP tasks. Nevertheless, GNN only considers the adjacent words, as it updates graph nodes. The model thereby emphasizes local features over global features, and misses the intent of the comment text. This paper constructs a triple graph neural network (TGNN) to serve the sentiment analysis of service texts. Firstly, the corresponding node connection windows were applied on different network layers to consider both local and global features. Next, the graph attention network (GAT) was adopted as the message delivery mechanism to fuse the features of all word nodes in the graph. Experimental results show that, the TGNN can evaluate the comment texts on logistics service quality more accurately than the other models.","Title: On the Distribution of SINR for Widely Linear MMSE MIMO Systems With Rectilinear or Quasi-Rectilinear Signals\nYear: 2022\n\nAbstract: Although the widely linear least mean square error (WLMMSE) receiver has been an appealing option for multiple-input-multiple-output (MIMO) wireless systems, a statistical understanding on its pose-detection signal-to-interference-plus-noise ratio (SINR) in detail is still missing. To this end, we consider a WLMMSE MIMO transmission system with rectilinear or quasi-rectilinear (QR) signals over the uncorrelated Rayleigh fading channel and investigate the statistical properties of its SINR for an arbitrary antenna configuration with \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$N_t$</tex-math></inline-formula>\n transmit antennas and \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$N_r$</tex-math></inline-formula>\n receive ones. We first derive an analytic probability density function (PDF) of the SINR in terms of the confluent hypergeometric function of the second kind, for WLMMSE MIMO systems with an arbitrary \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$N_r$</tex-math></inline-formula>\n and \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$N_t=2, 3$</tex-math></inline-formula>\n. For a more general case in practice, i.e., \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$N_t>3$</tex-math></inline-formula>\n, we resort to the moment generating function to obtain an approximate but closed form PDF under some mild conditions, which, as expected, is more Gaussian-like as \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$2N_r-N_t$</tex-math></inline-formula>\n increases. The so-derived PDFs are able to provide key insights into the WLMMSE MIMO receiver in terms of the outage probability, the symbol error rate, and the diversity gain, all presented in closed form. In particular, its diversity gain and the gain improvement over the conventional LMMSE one are explicitly quantified as \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$N_r-(N_t-1)/2$</tex-math></inline-formula>\n and \n<inline-formula xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math notation=\"LaTeX\">$(N_t-1)/2$</tex-math></inline-formula>\n, respectively. Finally, Monte Carlo simulations support the analysis.","Title: A DFDD Based Detector for Space-Time Block Coded Differential Spatial Modulation Under Time-Selective Channels\nYear: 2022\n\nAbstract: Space-time block coded differential spatial modulation (STBC-DSM) is a recently proposed spectral efficient multiple-input-multiple-output (MIMO) transmission technique. Existing detection algorithms for STBC-DSM are designed for quasi-static channels, which limits the application of STBC-DSM in fast fading channels. In this letter, we propose a novel decision-feedback differential detection (DFDD) based detector for STBC-DSM system under fast-fading channels. The proposed detector mitigates the impact of channel vector changes by using the DFDD metric which contains multiple previous detection results. Simulation results show that the proposed detector is more tolerable in time-selective channels than other existing STBC-DSM detectors.","Title: An improved mean-square performance analysis of the diffusion least stochastic entropy algorithm\nYear: 2022\n\nAbstract: The diffusion least stochastic entropy (DLSE) algorithm yields a better performance than the diffusion complex-valued least mean-square (DCLMS) algorithm for second-order noncircular noise. The existing theoretical steady-state mean-square deviation (MSD) shows a low accuracy due to the use of improper estimations and approximations. Besides, its transient MSD has not been thoroughly studied. This paper provides an improved performance analysis. We first improve the method of estimating the circularity coefficient (CC) of the error signal, which is required for the update equation of the DLSE. This improvement leads to a better match between the theoretical and simulated MSD and makes the transient MSD analysis mathematically tractable. The MSD analysis is performed by using the Gaussian moment factorizing theorem with some statistical assumptions. Besides, another key performance measure, i.e., the excess mean-square error (EMSE) of the DLSE is also analyzed. Simulation results validate the theoretical findings of the DLSE algorithm for second-order noncircular input signals and noise.","Title: Hierarchical Amplitude-Aware Permutation Entropy-Based Fault Feature Extraction Method for Rolling Bearings\nYear: 2022\n\nAbstract: In order to detect the incipient fault of rolling bearings and to effectively identify fault characteristics, based on amplitude-aware permutation entropy (AAPE), an enhanced method named hierarchical amplitude-aware permutation entropy (HAAPE) is proposed in this paper to solve complex time series in a new dynamic change analysis. Firstly, hierarchical analysis and AAPE are combined to excavate multilevel fault information, both low-frequency and high-frequency components of the abnormal bearing vibration signal. Secondly, from the experimental analysis, it is found that HAAPE is sensitive to the early failure of rolling bearings, which makes it suitable to evaluate the performance degradation of a bearing in its run-to-failure life cycle. Finally, a fault feature selection strategy based on HAAPE is put forward to select the bearing fault characteristics after the application of the least common multiple in singular value decomposition (LCM-SVD) method to the fault vibration signal. Moreover, several other entropy-based methods are also introduced for a comparative analysis of the experimental data, and the results demonstrate that HAAPE can extract fault features more effectively and with a higher accuracy.","Title: Masked Generative Distillation.\nYear: 2022\n\nAbstract: Knowledge distillation has been applied to various tasks successfully. The current distillation algorithm usually improves students’ performance by imitating the output of the teacher. This paper shows that teachers can also improve students’ representation power by guiding students’ feature recovery. From this point of view, we propose Masked Generative Distillation (MGD), which is simple: we mask random pixels of the student’s feature and force it to generate the teacher’s full feature through a simple block. MGD is a truly general feature-based distillation method, which can be utilized on various tasks, including image classification, object detection, semantic segmentation and instance segmentation. We experiment on different models with extensive datasets and the results show that all the students achieve excellent improvements. Notably, we boost ResNet-18 from 69.90% to 71.69% ImageNet top-1 accuracy, RetinaNet with ResNet-50 backbone from 37.4 to 41.0 Boundingbox mAP, SOLO based on ResNet-50 from 33.1 to 36.2 Mask mAP and DeepLabV3 based on ResNet-18 from 73.20 to 76.02 mIoU. Our codes are available at https://github.com/yzd-v/MGD.","Title: RFBFN: A Relation-First Blank Filling Network for Joint Relational Triple Extraction.\nYear: 2022\n\nAbstract: Joint relational triple extraction from unstructured text is an important task in information extraction. However, most existing works either ignore the semantic information of relations or predict subjects and objects sequentially. To address the issues, we introduce a new blank filling paradigm for the task, and propose a relation-first blank filling network (RFBFN). Specifically, we first detect potential relations maintained in the text to aid the following entity pair extraction. Then, we transform relations into relation templates with blanks which contain the fine-grained semantic representation of the relations. Finally, corresponding subjects and objects are extracted simultaneously by filling the blanks. We evaluate the proposed model on public benchmark datasets. Experimental results show our model outperforms current state-of-the-art methods. The source code of our work is available at: https://github.com/lizhe2016/RFBFN.","Title: Automatic Synthesis Technology of Music Teaching Melodies Based on Recurrent Neural Network\nYear: 2021\n\nAbstract: Computer music creation boasts broad application prospects. It generally relies on artificial intelligence (AI) and machine learning (ML) to generate the music score that matches the original mono-symbol score model or memorize/recognize the rhythms and beats of the music. However, there are very few music melody synthesis models based on artificial neural networks (ANNs). Some ANN-based models cannot adapt to the transposition invariance of original rhythm training set. To overcome the defect, this paper tries to develop an automatic synthesis technology of music teaching melodies based on recurrent neural network (RNN). Firstly, a strategy was proposed to extract the acoustic features from music melody. Next, the sequence-sequence model was adopted to synthetize general music melodies. After that, an RNN was established to synthetize music melody with singing melody, such as to find the suitable singing segments for the music melody in teaching scenario. The RNN can synthetize music melody with a short delay solely based on static acoustic features, eliminating the need for dynamic features. The proposed model was proved valid through experiments.","Title: CrimeTensor: Fine-Scale Crime Prediction via Tensor Learning with Spatiotemporal Consistency\nYear: 2022\n\nAbstract: AbstractCrime poses a major threat to human life and property, which has been recognized as one of the most crucial problems in our society. Predicting the number of crime incidents in each region of a city before they happen is of great importance to fight against crime. There has been a great deal of research focused on crime prediction, ranging from introducing diversified data sources to exploring various prediction models. However, most of the existing approaches fail to offer fine-scale prediction results and take little notice of the intricate spatial-temporal-categorical correlations contained in crime incidents. In this article, we propose a tailor-made framework called CrimeTensor to predict the number of crime incidents belonging to different categories within each target region via tensor learning with spatiotemporal consistency. In particular, we model the crime data as a tensor and present an objective function which tries to take full advantage of the spatial, temporal, and categorical correlations contained in crime incidents. Moreover, a well-designed optimization algorithm which transforms the objective into a compact form and then applies CP decomposition to find the optimal solution is elaborated to solve the objective function. Furthermore, we develop an enhanced framework which takes a set of pre-selected regions to conduct prediction so as to further improve the computational efficiency of the optimization algorithm. Finally, extensive experiments are performed on both proprietary and public datasets and our framework significantly outperforms all the baselines in terms of each evaluation metric.","Title: S(2)ENet: Spatial-Spectral Cross-Modal Enhancement Network for Classification of Hyperspectral and LiDAR Data\nYear: 2022\n\nAbstract: The effective utilization of multimodal data (e.g., hyperspectral and light detection and ranging (LiDAR) data) has profound implications for further development of the remote sensing (RS) field. Many studies have explored how to effectively fuse features from multiple modalities; however, few of them focus on information interactions that can effectively promote the complementary semantic content of multisource data before fusion. In this letter, we propose a spatial-spectral enhancement module ((SEM)-E-2) for cross-modal information interaction in deep neural networks. Specifically, (SEM)-E-2 consists of SpAtial Enhancement Module (SAEM) for enhancing spatial representation of hyperspectral data by LiDAR features and SpEctral Enhancement Module (SEEM) for enhancing spectral representation of LiDAR data by hyperspectral features. A series of experiments and ablation studies on the Houston2013 dataset show that (SEM)-E-2 can effectively facilitate the interaction and understanding between multimodal data. Our source code is available at https://github.com/likyoo/Multimodal-Remote-Sensing-Toolkit, contributing to the RS community.","Title: Shuffle-invariant Network for Action Recognition in Videos\nYear: 2022\n\nAbstract: AbstractThe local key features in video are important for improving the accuracy of human action recognition. However, most end-to-end methods focus on global feature learning from videos, while few works consider the enhancement of the local information in a feature. In this article, we discuss how to automatically enhance the ability to discriminate the local information in an action feature and improve the accuracy of action recognition. To address these problems, we assume that the critical level of each region for the action recognition task is different and will not change with the region location shuffle. We therefore propose a novel action recognition method called the shuffle-invariant network. In the proposed method, the shuffled video is generated by regular region cutting and random confusion to enhance the input data. The proposed network adopts the multitask framework, which includes one feature backbone network and three task branches: local critical feature shuffle-invariant learning, adversarial learning, and an action classification network. To enhance the local features, the feature response of each region is predicted by a local critical feature learning network. To train this network, an L1-based critical feature shuffle-invariant loss is defined to ensure that the ordered feature response list of these regions remains unchanged after region location shuffle. Then, the adversarial learning is applied to eliminate the noise caused by the region shuffle. Finally, the action classification network combines these two tasks to jointly guide the training of the feature backbone network and obtain more effective action features. In the testing phase, only the action classification network is applied to identify the action category of the input video. We verify the proposed method on the HMDB51 and UCF101 action datasets. Several ablation experiments are constructed to verify the effectiveness of each module. The experimental results show that our approach achieves the state-of-the-art performance.","Title: A Fast and Accurate Method of Power Line Intelligent Inspection Based on Edge Computing\nYear: 2022\n\nAbstract: The combination of power intelligent inspection and edge computing plays a crucial role in the construction of power Internet of Things and transparent power grid. However, as a consequence of the low computing power of edge devices, the detection speed of the model running on edge devices tends to be slow. Accordingly, a fast and accurate method of power line edge intelligent inspection is proposed in this article. First, for key component detection of power lines, RepVGG, diverse branch block (DBB), efficient channel attention (ECA), and improved spatial pyramid pooling (SPP) are utilized to improve YOLOv5, giving rise to RepYOLO. In addition, to solve the problem of low accuracy in pin defect detection, a two-stage cascaded method is proposed. The first stage localizes various connection fittings in the input image, and the second stage recognizes normal pins and missing pins in these connection fittings. Finally, C++ language combined with TensorRT is employed to optimize and accelerate the model on the NVIDIA Jetson Xavier NX embedded platform, fulfilling efficient power line edge intelligent inspection. Experimental results show that the TensorRT optimized RepYOLO algorithm is four times the inference speed of YOLOv5 with a 1.2% increase in accuracy. Moreover, TensorRT optimized two-stage cascaded method can achieve accurate and real-time pin defect detection on edge devices.","Title: Design and optimisation of a green manufacturing-recycling network considering heavy metal pollutants - an electronic assembly case\nYear: 2022\n\nAbstract: This paper presents a green manufacturing-recycling network design approach for multi-echelon electronic assembly process under a complex product mix scenario which incorporates trade-off between economic cost and heavy metal pollutants. The approach is developed by formulating the green manufacturing-recycling network into a closed-loop logistics bi-objective optimisation model, in order to alleviate negative environmental influence by applying proper cleaner production technologies. The environmental influence is quantified by pollution equivalent numbers which are integrated in the model. Moreover, the volume of the heavy metal pollutants is optimised by selecting the cleaner production level and thus is jointly coordinated with the operation cost. To solve the bi-objective optimisation model, an enhanced global criterion approach is presented to improve the effectiveness on dealing with non-convexity of the Pareto-optimal frontier with computational efficient solutions. The proposed model is implemented on a case study to verify its flexibility to handle the closed-loop manufacturing-recycling network design of different sizes, as well as to obtain trade-off between operation cost and environmental influence by the heavy metal pollutants under various scenarios.","Title: Research on the Error of Global Positioning System Based on Time Series Analysis\nYear: 2022\n\nAbstract: Due to the poor dynamic positioning precision of the Global Positioning System (GPS), Time Series Analysis (TSA) and Kalman filter technology are used to construct the positioning error of GPS. According to the statistical characteristics of the autocorrelation function and partial autocorrelation function of sample data, the Autoregressive (AR) model which is based on a Kalman filter is determined, and the error model of GPS is combined with a Kalman filter to eliminate the random error in GPS dynamic positioning data. The least square method is used for model parameter estimation and adaptability tests, and the experimental results show that the absolute value of the maximum error of longitude and latitude, the mean square error of longitude and latitude and average absolute error of longitude and latitude are all reduced, and the dynamic positioning precision after correction has been significantly improved.","Title: 3-D path planning system for autonomous vehicle considering the rollover and path length\nYear: 2022\n\nAbstract: •A 3-D path planning system considering the rollover and path length is proposed.•A novel method for evaluating the roll angle is developed based on the terrain.•2-D and 3-D path length are also measured synthetically.•The weights of roll angle and path length are optimized by a fuzzy-basedoptimizer.","Title: AvatarCap: Animatable Avatar Conditioned Monocular Human Volumetric Capture.\nYear: 2022\n\nAbstract: To address the ill-posed problem caused by partial observations in monocular human volumetric capture, we present AvatarCap, a novel framework that introduces animatable avatars into the capture pipeline for high-fidelity reconstruction in both visible and invisible regions. Our method firstly creates an animatable avatar for the subject from a small number (~20) of 3D scans as a prior. Then given a monocular RGB video of this subject, our method integrates information from both the image observation and the avatar prior, and accordingly recon-structs high-fidelity 3D textured models with dynamic details regardless of the visibility. To learn an effective avatar for volumetric capture from only few samples, we propose GeoTexAvatar, which leverages both geometry and texture supervisions to constrain the pose-dependent dynamics in a decomposed implicit manner. An avatar-conditioned volumetric capture method that involves a canonical normal fusion and a reconstruction network is further proposed to integrate both image observations and avatar dynamics for high-fidelity reconstruction in both observed and invisible regions. Overall, our method enables monocular human volumetric capture with detailed and pose-dependent dynamics, and the experiments show that our method outperforms state of the art. Code is available at https://github.com/lizhe00/AvatarCap.","Title: Multiperiodic Repetitive Control for Functional Electrical Stimulation-Based Wrist Tremor Suppression\nYear: 2022\n\nAbstract: Intention tremor refers to the rhythmic and involuntary contraction and relaxation of muscles with movement toward a target, which is a common sequela of multiple sclerosis and usually occurs in the distal joints of the upper limb. Functional electrical stimulation (FES) is feasible for tremor suppression because of its fewer side effects, low cost, and portability. Most existing FES-based design methods assume that tremor is a single-frequency signal, though it is multifrequency in reality. The idealized simplification will limit the performance of tremor suppression. To address the problem, this article proposes an FES-based multiperiodic repetitive control (MP-RC) scheme to suppress multiple frequency wrist tremors. First, a nonlinear wrist musculoskeletal model with a Hammerstein structure is established. Then, a control strategy combining the model inverse linearization control and MP-RC is proposed for tremor suppression. A frequency-modified inverse RC algorithm and a gradient-based RC algorithm are developed to regulate the FES level. Finally, comparative experiments on four unimpaired participants and an intention tremor patient are conducted to validate the effectiveness of the proposed control schemes. Experimental results show that the MP-RC scheme can suppress tremors by up to 90.52%. Compared with the traditional filter-based feedback controller and the single-periodic repetitive controller, the proposed multiperiodic repetitive controller can achieve an average of 26% and 16% improvement, respectively, in tremor suppression, demonstrating the advantages of the proposed design.","Title: Spatiotemporal Evolution of the Urban Thermal Environment Effect and Its Influencing Factors: A Case Study of Beijing, China\nYear: 2022\n\nAbstract: Rapid urbanization has led to significant changes in land surface temperature (LST), which in turn affect the urban thermal environment effect and the health of residents. Exploring the causes of the urban thermal environment effect will provide guidance for promoting sustainable urban development. The spatiotemporal evolution of the urban thermal environment effect within the sixth ring road of Beijing was analyzed by inversion of remote sensing data to obtain the LST in 2004, 2009, 2014, and 2019. In addition, based on multivariate spatial data, we applied the standard deviation ellipse (SDE), spatial principal component analysis (PCA), and other methods to analyze and identify the relationships between the urban thermal environment effect and its influencing factors. The results show that from 2004 to 2019, the spatial distribution of urban development and LST within the sixth ring road of Beijing were closely related, the heat island area showed a small increasing trend, and differences in the thermal environment effect between different administrative regions in different periods were obvious. The main factors affecting the urban thermal environment effect were urban construction intensity, vegetation and water bodies, socioeconomic activities, and geomorphology. It is noteworthy that human factors had a greater impact than natural factors. Among them, the positive effect of the normalized difference impervious surface index (NDBBI) and the negative effect of the fractional vegetation cover (FVC) were the most prominent. This study provides theoretical support for mitigating the urban thermal environment effect and promoting sustainable urban development.","Title: Influencing factors and multi-scenario prediction of China's ecological footprint based on the STIRPAT model\nYear: 2022\n\nAbstract: Focusing on the spatial and temporal pattern, evolution law, influencing factors and prediction of regional ecological footprint (EF) is conducive to promoting sustainable development of regional populations, resources, and the environment. Firstly, this study used the EF model to analyze the spatial and temporal patterns and dynamic evolution characteristics of the total ecological footprint (TEF) and the relative contributions of the biological ecological footprint (BEF), energy ecological footprint (EEF), and pollution ecological footprint (PEF) in China from 2000 to 2019. Secondly, the impact of socioeconomic factors on China's TEF was analyzed based on the STIRPAT model. Finally, the future development trend of China's TEF was analyzed by multi-scenario prediction. The results demonstrate that: (1) from 2000 to 2019, the TEF levels of China form three gradient spaces, (2) the BEF is the biggest contributor to EF in most of China's provinces, (3) total energy consumption is the most important positive factor for China's TEF while proportion of tertiary industry in the three industries is the most important negative factor, and (4) maintaining low growth in total energy consumption and high growth in proportion of tertiary industry in the three industries is crucial for limiting the future growth of China's TEF","Title: Prompting for Multi-Modal Tracking\nYear: 2022\n\nAbstract: ABSTRACTMulti-modal tracking gains attention due to its ability to be more accurate and robust in complex scenarios compared to traditional RGB-based tracking. Its key lies in how to fuse multi-modal data and reduce the gap between modalities. However, multi-modal tracking still severely suffers from data deficiency, thus resulting in the insufficient learning of fusion modules. Instead of building such a fusion module, in this paper, we provide a new perspective on multi-modal tracking by attaching importance to the multi-modal visual prompts. We design a novel multi-modal prompt tracker (ProTrack), which can transfer the multi-modal inputs to a single modality by the prompt paradigm. By best employing the tracking ability of pre-trained RGB trackers learning at scale, our ProTrack can achieve high-performance multi-modal tracking by only altering the inputs, even without any extra training on multi-modal data. Extensive experiments on 5 benchmark datasets demonstrate the effectiveness of the proposed ProTrack.","Title: An enhanced CNN-LSTM remaining useful life prediction model for aircraft engine with attention mechanism\nYear: 2022\n\nAbstract: Remaining useful life (RUL) prediction is one of the key technologies of aircraft prognosis and health management (PHM) which could provide better maintenance decisions. In order to improve the accuracy of aircraft engine RUL prediction under real flight conditions and better meet the needs of PHM system, we put forward an improved CNN-LSTM model based on the convolutional block attention module (CBAM). First, the features of aircraft engine operation data are extracted by multi -layer CNN network, and then the attention mechanism is processed by CBAM in channel and spatial dimensions to find key variables related to RUL. Finally, the hidden relationship between features and service time is learned by LSTM and the predicted RUL is output. Experiments were conducted using C-MPASS dataset. Experimental results indicate that our prediction model has feasibility. Compared with other state-of-the-art methods, the RMSE of our method decreased by 17.4%, and the score of the prediction model was improved by 25.9%.","Title: Differences in Spectrum Width Estimates Between Electronic Scans and Mechanical Scans With a Phased Array Weather Radar\nYear: 2022\n\nAbstract: This letter presents differences in spectrum width estimates between electronic scans and mechanical scans with a cylindrical polarimetric phased array radar (CPPAR). Theoretical formulation of spectrum width estimates considering the transverse wind, shear, and turbulence within the radar’s resolution volume is described. Cross correlation function between two resolution volumes is derived based on wave scattering by randomly distributed scatterers. Spectrum width estimates of ground clutter and weather scatterers with the CPPAR in electronic scans and mechanical scans are presented and compared. It is shown that instead of antenna rotation, the transverse wind, shear, and turbulence within the radar’s resolution volume, which were previously ignored for narrow beams, contribute the most to the differences in spectrum width estimates between mechanical scans and electronic scans for the CPPAR with a wide beam.","Title: Late feature supplement network for early action prediction\nYear: 2022\n\nAbstract: Early action prediction is a new hotspot in the field of computer vision. To improve the accuracy of early action prediction, a new end-to-end late feature supplement-based early action prediction network is proposed in this work. Different from the existing methods that use the model transfer strategy, a feature transfer strategy is defined in this work. Specifically, the features of the late clip are regarded as labels, and a feature transfer model is built to achieve mapping from the features of the early clip to the late features. After feature transfer, the generated late feature is fused into the early feature to form the final video feature. Finally, the final video feature is applied to action classification. The proposed method is evaluated on the action classification task of the early clip. The experimental results show that compared with the existing methods, the proposed method has better performance at different observation ratios. The ablation study verifies that the proposed feature transfer strategy can significantly improve the accuracy of early action prediction.","Title: Deep transfer learning for failure prediction across failure types\nYear: 2022\n\nAbstract: With the increasing development of artificial intelligence (AI) technologies, deep learning-driven approaches have been widely applied to predicate different machinery failures. One key challenge of failure prediction is to collect sufficient data, especially data of various failure types, to train the data-driven models. Existing studies focus on using transfer learning to transfer knowledge across machines or domains, but not across failure types. In this study, we hypothesise that knowledge about failure among similar failure types is transferable. Should the hypothesis hold, companies may no longer require a large amount of all types of failure data for predictive maintenance. This will increase the companies’ overall implementation feasibility and productivity gains. We tested our hypothesis on knowledge transferability for failure prediction in an experiment performed on rotating machinery with vibration signals. During the experiment, we first calibrated the performance of the trained deep neural network in each impending failure type. Then, we leveraged the architecture and hyperparameters of the neural network model trained from one type of failure as the pre-trained model for knowledge transfer. The pre-trained model is fine-tuned with data from another type of failure of the same machine. After that, we compared the performance of the neural network model to predict the second type of failure before and after knowledge transfer. Results showed that transferring knowledge obtained from one type of failure could vastly improve the performance of predicting another type of failure, which may not have sufficient data to train a good prediction model. This result implies that predictive analytics can apply parameter-based deep transfer learning (TL) to address the challenge of insufficient data on all types of machine failures for failure prediction.","Title: An affine combination of two augmented CLMS adaptive filters for processing noncircular Gaussian signals\nYear: 2022\n\nAbstract: •We propose affine combining of two ACLMS filters and present its optimal mixing parameter, based on which an explicit expression describing the steady state mean behavior of the optimal mixing parameter is quantified for doubly white noncircular Gaussian input data. The so established physical insight shows that in steady state, the optimal mixing parameter is, statistically, negative and monotonically increases against the input SO noncircularity coefficient.•Upon applying the optimal mixing parameter, we perform the MSE analysis of the proposed affine-combined ACLMS filters. The so derived explicit MSE expression not only theoretically presents the universality of the combination scheme, that is, smaller or equal steady state MSE than either of the individual ACLMS subfilters, but also demonstrates that the overall MSE in steady state is approximately a monotonic increasing function of the input SO noncircularity coefficient.•Considering that the optimal affine combiner is unacquirable in practical applications, we provide an error power based scheme to iterate the affine combiner. Numerical examples verify that the proposed iterative combiner converges to the optimal one. The overall MSE curves respectively obtained via the iterative and optimal combiners are shown to be in close agreement.","Title: PAW: Data Partitioning Meets Workload Variance\nYear: 2022\n\nAbstract: In distributed storage systems (e.g., HDFS, Amazon S3, Databricks), partitioning is applied on a dataset in order to enhance performance and availability. Recently, partitioning methods have been designed to optimize the query performance of partitions with respect to the historical query workload. Never-theless, in practice, future query workloads may deviate from the historical query workload, thus deteriorating the performance of existing partitioning methods. To fill this research gap, we model the variance of future query workloads from the historical query workload, then exploit this characteristic to produce partitions that perform well for future query workloads. In addition, we explore the space of irregular shaped partition regions to further optimize the query performance. Experimental results on TPC-H and real datasets show that our proposal is up to 70x more efficient than the state-of-the-art method.","Title: Naturalistic Driving Scenario Recognition with Multimodal Data\nYear: 2022\n\nAbstract: Driving Scenario recognition is one fundamental technology of automated driving systems or advanced driver assistance systems. A common practice of driving scenario recognition is to conduct classification tasks with the data collected by in-vehicle data acquisition system or driving simulator. In most existing works, visual data were used since the relevant information of driving scenarios is usually inferable from their visual appearance. However, the non-visual information, e.g. physiological state of the driver, also provide complementary information for the scenario recognition task especially when the visual appearance is insufficient to differentiate similar naturalistic scenarios in some cases. In this paper, we propose a hybrid driving scenario recognition model with multimodal input. The model consists of a convolutional neural network based visual data sub-model, a stacked autoencoder based physiological data sub-model, and a fusion sub-model that combines the extracted features from both visual and physiological sub-models. Besides, a post-processing is adopted to correct the recognition results of some ambiguous scenarios. Experimental results on the trip data collected in the naturalistic driving context demonstrated the effectiveness of the proposed method.","Title: High-Speed Long-Haul Multi-Channel Nonlinear Optical Communication Systems Influenced by Equalization Enhanced Phase Noise\nYear: 2022\n\nAbstract: In this work, the performance of high-speed long- haul nonlinear Nyquist-spaced multi-channel coherent optical fiber communication systems utilizing electronic dispersion compensation and digital nonlinearity compensation is explored taking into consideration the enhanced equalization phase noise. The analytical model has also been developed to estimate the system performance under different transmission scenarios.","Title: P&GGD: A Joint-Way Model Optimization Strategy Based on Filter Pruning and Filter Grafting For Tea Leaves Classification\nYear: 2022\n\nAbstract: In the automatic processing of tea, the category of fresh tea leaves determines the processing parameters for the automatic control systems. The automatic classification of fresh tea leaves is a necessity. Previous works used deep and wide architectures of CNNs to classify the tea leaves, which have a limitation on the model deployment. Despite the filter pruning achieve superior performance about network compression and acceleration, the pruning-based approaches have to trade accuracy for low computational cost and model size. we propose a novel joint-way model optimization strategy, named Filter Pruning and Grafting via Geometric Distance(P&GGD), to solve the above problems. Specifically, we utilize the filter pruning method to compress models and uses the filter grafting method to boost accuracy. To graft pruned models effectively, we develop the grafting criteria based on the geometric distance to measure the information of layers. Extensive experiments validate our approach on the fresh-tea-leaves dataset and two image classification benchmarks. For example, with the fresh-tea-leaves dataset, we achieve 96.296% top-1 accuracy, with more than 45% FLOPs-reduction, even outperforms the non-compress MobileNetV2 by 2.395% accuracy. Our proposed method can provide more accurate, more compact, and faster models. Thus, it can promote the deployment of the model at the edge of the automatic production line."],"string":"[\n \"Title: Towards a hybrid optimization model for elemental content analysis in EDXRF.\\nYear: 2012\\n\\nAbstract: This paper presents a hybrid optimization model for predicting the elemental contents such as Ti, V and Fe in energy dispersive X-ray fluorescence (EDXRF) based on least square support vector machine (LS-SVM) and particle swarm optimization (PSO) methods. The model used PSO to optimize LS-SVM parameters. In order to assess the capability and effectiveness of the proposed model, several measurement methods such as SVM model and BP neural network model were compared. The results indicate that the proposed model is feasible for quantitative analysis of elemental contents in nondestructive nuclear measurement applications. © 2012 IEEE.\",\n \"Title: Optimal network locality in distributed virtualized data-centers\\nYear: 2011\\n\\nAbstract: Cost efficiency is a key aspect in deploying distributed service in networks within decentralized service delivery architectures. In this paper, we address this aspect from an optimization and algorithmic standpoint. The research deals with the placement of service components to network sites, where the performance metric is the cost for acquiring components between the sites. The resulting optimization problem, which we refer to as the k-Component Multi-site Placement Problem, is applicable to service distribution in a wide range of communication networking scenarios. We provide a theoretical analysis of the problem's computational complexity, and develop an integer programming model for providing reference results for performance benchmarking. On the algorithmic side, we present four approaches: an algorithm with approximation guarantee and three heuristics algorithms. The first heuristic is derived from graph theory on domatic partition. The second heuristic, built on intuition, admits distributed computation. The third heuristic emphasizes on fairness in cost distribution among the sites. We report simulation results for sets of networks where cost is represented by round-trip time (RTT) originating from real measurements. For small networks, the integer model is used to study algorithm performance in terms of optimality. Large networks are used to compare the algorithms relatively to each other. Among the algorithms, the heuristic based on intuition has close-to-optimal performance, and the fairness heuristic achieves a good balance between single-site cost and the overall one. In addition, the experiments demonstrate the significance of optimization for cost reduction in comparison to a the random allocation strategy.\",\n \"Title: Multiple Sources Data Fusion Strategies Based on Multi-class Support Vector Machine\\nYear: 2008\\n\\nAbstract: Data fusion from multiple sources is very important and useful. Some strategies for multiple sources data fusion based on Multi-class Support Vector Machine (MSVM) are proposed in this paper. The features (independent components) of data from multiple sources are extracted for fusion. The Dempster-Shafer theory (DS theory) and Bayesian theory are used to combine the probabilistic outputs of MSVMs. Then the outputs of DS theory combination are classified by a MSVM to get the final decision of the classification. Finally, these strategies are evaluated by three data sets and the results show that DS theory can improve the accuracy obviously, and the strategies based on MSVM and DS theory is very fit for solving problems with small data sets.\",\n \"Title: View-adaptive manipulative action recognition for robot companions\\nYear: 2007\\n\\nAbstract: This paper puts forward an approach for a mobile robot to recognize the human's manipulative actions from different single camera views. While most of the related work in action recognition assume a fixed static camera view that is the same for training and testing, such kind of constraints do not apply for mobile robot companions. We propose a recognition scheme that is able to generalize an action model, that has been learned from a very few data items observed from a single camera view, to variant view points and different settings. We tackle the problem of compensating the view dependence of 2D motion models on three different levels. Firstly, we pre-segment the trajectories based on an object vicinity that depends on the camera tilt and object detections. Secondly, an interactive feature vector is designed that represents the relative movements between the human hand and the objects. Thirdly, we propose an adaptive HMM-based matching process that is based on a particle filter and includes a dynamically adjusted scaling parameter that models the systematic error of the view dependency. Finally, we use a two-layered approach for task recognition which decouples the task knowledge from the view dependent primitive recognition. The results of experiments in an office environment show the applicability of this approach.\",\n \"Title: Blind Detection of Digital Forgery Image Based on the Local Entropy of the Gradient\\nYear: 2008\\n\\nAbstract: A novel method based on the local entropy of the gradient is proposed to detect the forged digital images. The method can discover some traces of artificial feather operation, which is necessary to create a smooth transition between a forged image region and its surroundings. The local entropy of the gradient is used to determine the forged region, and the credibility is computed to show the reality level of the image. Results of experiments on several forged images demonstrate the effectiveness of the algorithm.\",\n \"Title: Pinning dynamic systems of networks with Markovian switching couplings and controller-node set.\\nYear: 2014\\n\\nAbstract: In this paper, we study pinning control problem of coupled dynamical systems with stochastically switching couplings and stochastically selected controller–node set. Here, the coupling matrices and the controller–node sets change with time, induced by a continuous-time Markov chain. By constructing Lyapunov functions, we establish tractable sufficient conditions for exponential stability of the coupled system. Two scenarios are considered here. First, we prove that if each subsystem in the switching system, i.e. with the fixed coupling, can be stabilized by the fixed pinning controller–node set, and in addition, the Markovian switching is sufficiently slow, then the time-varying dynamical system is stabilized. Second, we conclude that if the system with the average coupling and pinning gains can be stabilized and the switching is sufficiently fast, the time-varying system is stabilized. Two numerical examples are provided to demonstrate the validity of these theoretical results, including a switching dynamical system between several stable subsystems, and a dynamical system with mobile nodes and spatial pinning control towards the nodes when these nodes are being in a pre-designed region.\",\n \"Title: Spatio-temporal variation of NDVI in Chinese coastal zone during recent 20 years\\nYear: 2005\\n\\nAbstract: Monthly digital NOAA/AVHRR NDVI images with 8km spatial resolution, multi-temporal land use data and monthly average air temperature and precipitation data recorded in standard observatories from 1981 to 2000 are obtained in Chinese coastal zone. The relationship between Normalized Difference Vegetation Index (NDVI) and land use factors and climate factors are explored through spatio-temporal variation analysis. At First, NDVI, monthly average air temperature and precipitation data are normalized to the range of 0.01 to 0.99, then 20 typical areas based on the latitude from north to south in Chinese coastal zone are selected, which is used to analyze the correlation between NDVI and air temperature and precipitation, Then, NDVI spatial and temporal variations are achieved for Chinese coastal area combined with regional land use data. Based on these results, it is pointed out that, in recent 20 years, NDVI annul variation in Chinese coastal zone has a stronger relationship with land use factors; moreover, NDVI spatial variation has obvious strap regularity: NDVI in North China is highly sensitive to precipitation, while it is highly sensitive to air temperature in South China. © 2005 IEEE.\",\n \"Title: Research of localization approach for the new comer in wireless sensor networks\\nYear: 2009\\n\\nAbstract: When some sensor nodes of wireless sensor networks (WSN) can not work forever because of long-term work or failure caused\\n by attack, a few new comers need to be put into the network. For the application of the new comer in WSN, an accurate and\\n effective localization algorithm based on received signal strength indicator (RSSI) is proposed. Through a few necessary nodes’\\n participation and the collaboration between the new comer and its one-hop and two-hop neighbor nodes, the accurate localization\\n of the new comer is achieved. Simulation results show that the localization accuracy is about 17% of sensor node’s radio frequency\\n (RF) transmission range, when the measurement error is 10% and the standard deviation for Gauss error of original sensor nodes’\\n coordinate is about 20% of sensor node’s RF transmission range. Simulation results also verify nice stability and adaptability\\n of the new comer’s location algorithm.\",\n \"Title: Evaluation Of Lauge-Hansen Classification For Ankle Fractures And Follow-Up Management\\nYear: 2013\\n\\nAbstract: Objective: To evaluate Lauge-Hansen classification on the diagnosis and treatment of ankle fractures. Methods: From April 2008 to June 2011, and following ethical approval of our research, patients with ankle fractures admitted at our second orthopaedics department were enrolled. All patients were followed up for a minimum of 8 months (8 similar to 46 with a mean of 27 months). Trauma mechanism and American Orthopaedic Foot and Ankle Society (AOFAS) scores were analyzed. The position and grade of ankle fracture injury were accessed for making treatment strategies for patients. Results: Thirty-two patients were recruited fulfilling the inclusion criteria and were available for follow-up. After treatment, none of the cases had postoperative infection, cut aneous necrosis, failure in internal fixation or nerve injury. At 12-month follow-up, AOFAS scores were 97.7 at average, among which 30 cases were excellent and 2 cases were good. Conclusion: The Lauge-Hansen classification contributes fully to diagnose ankle fractures, so that surgeons can make correct treatment strategies and achieve good treatment effect. Besides, perfect follow-up records, genial attitude, good communication skill and flexible follow-up time are the key to finish evaluation as schedule.\",\n \"